literature review on malaria in zambia

• Parasitic Diseases
• Protozoan Infections
• Parasitology

Current Status and Challenges of Malaria in Zambia

• Theoretical and Natural Science 3(1):499-504
• 3(1):499-504
• This person is not on ResearchGate, or hasn't claimed this research yet.

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations

• Ashok Kumar Pannu

• Peter O. Oluoch
• D Chikamata
• E Kherlopian
• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

• Open access
• Published: 30 June 2021

Malaria prevention knowledge, attitudes, and practices in Zambezia Province, Mozambique

• Liliana de Sousa Pinto   ORCID: orcid.org/0000-0001-9089-4375 1 ,
• Jorge A. H. Arroz 1 ,
• Maria do Rosário O. Martins 2 ,
• Zulmira Hartz 2 ,
• Nuria Negrao 3 ,
• Victor Muchanga 4 ,
• Amadeu Cossa 4 &
• Rose Zulliger 5  

Malaria Journal volume  20 , Article number:  293 ( 2021 ) Cite this article

9169 Accesses

9 Citations

1 Altmetric

Metrics details

In Mozambique, socio-economic and cultural factors influence the wide adoption of disease preventive measures that are relevant for malaria control strategies to promote early recognition of disease, prompt seeking of medical care, sleeping under insecticide-treated nets (ITNs), and taking intermittent preventive treatment for pregnant women. However, there is a critical information gap regarding previous and ongoing malaria social and behavioural change (SBC) interventions. The aim of this study is to assess the knowledge, attitudes, practices of beneficiaries of SBC interventions.

A descriptive cross-sectional survey was undertaken in 2018 in two rural districts of Zambezia Province, Mozambique. A structured questionnaire was administered to 773 randomly selected households. Respondents were the adult heads of the households. Descriptive statistics were done.

The main results show that 96.4% of respondents recalled hearing about malaria in the previous 6 months, 90.0% had knowledge of malaria prevention, and 70.0% of preventive measures. Of the 97.7% respondents that had received ITNs through a mass ITN distribution campaign, 81.7% had slept under an ITN the night before the survey. In terms of source of health information, 70.5% mentioned the role of community volunteers in dissemination of malaria prevention messages, 76.1% of respondents considered worship places (churches and mosques) to be the main places where they heard key malaria prevention messages, and 79.1% asserted that community dialogue sessions helped them better understand how to prevent malaria.

Conclusions

Results show that volunteers/activists/teachers played an important role in dissemination of key malaria prevention messages, which brought the following successes: community actors are recognized and people have knowledge of malaria transmission, signs and symptoms, preventive measures, and where to get treatment. There is, however, room for improvement on SBC messaging regarding some malaria symptoms (anaemia and convulsions) and operational research is needed to ascertain the drivers of malaria prevalence and inform the SBC approach.

In 2018, there were an estimated 228 million cases of malaria globally, the vast majority (93%) in the African region [ 1 ]; Mozambique is one of the six countries to account for more than half of all malaria cases. It is important to understand the factors that contribute to such a high disease burden in the country.

The World Health Organization’s Global Technical Strategy for Malaria 2016–2030 is comprised of three major pillars with two supporting elements: (i) innovation and research, and, (ii) a strong enabling environment [ 2 ]. These supporting elements are aligned with the Mozambican National Malaria Control Programme (NMCP) for the elimination of malaria through implementation of research to optimize the impact and cost-effectiveness of new and existing tools, interventions and strategies, strong political and financial commitments, multi-sectorial approaches, stewardship of the health system, and capacity building development [ 3 ].

The NMCP prioritizes planning, implementation, monitoring, and evaluation based on an evidence-based, multi-cultural, and gender equality approach and an interpersonal communication and mass media (radio) approach [ 3 ]. However, there is a critical information gap about outcomes and impact of social and behavioural change (SBC) interventions in Mozambique [ 3 ]. SBC interventions are widely used in malaria prevention and control programmes to promote appropriate care-seeking and provision and utilization of insecticide-treated nets (ITNs) and indoor residual spraying (IRS). These interventions play an important role in increasing knowledge and creating awareness and the demand for prevention and treatment programmes [ 4 ]. Human behaviour is an important factor contributing to disease burden [ 5 ]. It is important to do formative research into various aspects that influence human behaviour, such as individual preferences, community characteristics, leadership practices, and quality of available goods and services, to determine their impact and to design effective SBC strategies and interventions [ 5 ]. Behavioural research is also useful for evaluation of SBC interventions.

In Mozambique, the goal of research on SBC interventions is to identify knowledge, attitudes, practices (KAP), and behaviours of communities, in order to define key strategies, target groups, cultural barriers, and community beliefs for improving malaria health outcomes through the adoption of positive health behaviours [ 3 , 6 ]. This finding is similar to a study conducted in rural Tanzania which demonstrated that more research on malaria knowledge and beliefs of the community is necessary to obtain and maintain community engagement and participation in malaria control activities [ 7 ]. The importance of obtaining this knowledge was underlined in a study conducted in Southeastern Iran that showed that strategies for the control of malaria can be effective, useful and valuable if prior studies are taken to explore and understand people’s KAP [ 8 ]. In addition, and encouragingly, a study from rural Uganda indicated that communities with knowledge can influence practices in households and support control of the disease [ 9 ]. In areas with high burden of disease it is important to have a clear understanding of the community to design good SBC interventions.

The present study was conducted with the aim of assessing the KAP of beneficiaries of malaria prevention SBC interventions in rural Mozambique.

Study area and design

This study was a cross-sectional survey carried out in November and December 2018 in Namacurra and Nicoadala districts of Zambézia Province. The districts were selected based on: (i) high malaria incidence; (ii) accessibility; (iii) population size similarities; (iv) geographic location; and, (v) experience with SBC interventions. The estimated population of Namacurra and Nicoadala is 390,410 and 270,825 inhabitants, respectively [ 10 ]. Both are rural districts with more than 60% of the population being illiterate and living in low social and economic conditions. The main public health problems are: malaria, HIV and diarrhoeal diseases [ 11 ]. In 2017, the incidence of malaria (per 1000 inhabitants) in Namacurra was 272 and in Nicoadala was 506 [ 12 ]. Both districts have targeted SBC and vector control interventions. SBC interventions included training of volunteers, local religious and community leaders, members of community structures, schoolteachers, and community health workers. These are all community actors that play an important role in dissemination of malaria prevention key messages. Additionally, different communication channels used to reach community members by these actors include meetings with health committee councils and health facilities, dissemination of standardized malaria SBC messages through community radios, door to door visits, sermons at worship places (mosque and church), community dialogues, focus groups discussions with adults, and dissemination of information, education and communication materials. Health committee councils, focal groups discussions (men to men, women), and community dialogues are part of the Mozambique National Health Promotion Strategy to promote and protect individual, family and community health by promoting positive health behaviours. Previously validated community dialogues are also used [ 13 ]. Community dialogues to promote healthy living habits are based on a set of ‘Life Stories’ prepared and used in a series of sessions to stimulate dialogue between people living in the same geographic area (neighbourhoods or communities). Additionally, women and men are given tools that enable them to reflect on how gender norms and social roles work in their lives, and the skills to begin a process of changing those norms, beliefs and roles that are considered harmful to health and the environment, and to the social well-being of people and communities, while reinforcing those that are perceived as positive and to be maintained [ 13 ].

Many of these efforts, particularly promotion on net use, were connected to the ITN universal coverage campaign (UCC) distribution, which covered the population of each district in 2017, and through ITN distribution in antenatal care (ANC) services to pregnant women. UCC in 2017 distributed 161,591 ITNs in Namacurra (100% of target) and 125,161 ITNs in Nicoadala (86% of target) [ 14 ].

All localities of these districts were selected for the study. Within each locality, household sample size was calculated by dividing the total sample size of the district by the number of existing localities. Households, the sampling units used in this study, were selected using a systematic random sampling method, after determining the total number in each locality.

Sample size

Sample size was calculated based on the equation:

where: n = sample size; Z = 1.96 (assuming a level of confidence of 95%); p = proportion = 0.5; d = error = 0.05. A total of 768 households were required for the study, 384 per district. The households were divided between localities with sample size equal to the proportion of households per locality. Additionally, five households were added during data collection, resulting in a study population of 773 households.

Selection of households

In each locality, the households were selected based on the following strategy: first, a household list (sampling frame) was developed and a number was assigned to each household; then, the sample interval (number of households divided by sample size) was computed and a random start number was chosen; finally, from this first random number, households were systematically selected using the sampling interval until the calculated sample size was met.

Data collection and measurement

A structured, close-ended questionnaire was pre-tested and administered by previously trained local interviewers. The first section of the questionnaire included standardized socio-demographic questions based on the Malaria Indicator Survey 2018 and the following parts of the questionnaire assessed the head of the household malaria KAP/behaviours, and information channels. The questionnaire was designed in Portuguese, and the interviews were conducted in the local language, Enlowe . The questionnaire was pre-tested in a district similar to the study districts. The head of the household was defined as the primary decision-maker in the family and the household and as an individual living in the household and having meals from a common cooking facility [ 15 ]. A responsible adult, 18-years or older, was appointed to participate in the interview in the absence of the head of the household.

The variables selected for this study were: place of residence, age, gender, level of education, number of people that live in the household, information channels, malaria KAP.

Household inclusion criteria

The inclusion criteria used to select the households for the study were: (i) households from the selected districts; (ii) household members living in the district from 2011 to 2017 (this period covers the SBC interventions funded by different malaria donors); (iii) interviewee at least 18 years old (head of the household), regardless of gender; (iv) community located in the study districts in which SBC interventions were performed by local community actors (volunteers from community structures, schoolteacher facilitators, activists, and faith leaders); (v) mosquito net mass distribution campaigns; and, (vi) presence of community radios.

Outcomes of interest

The measured outcomes were: (i) percentage of people who remember hearing or seeing a message about malaria in the previous 6 months; (ii) percentage of people with favourable attitudes towards ITNs, malaria-related practices (use of ITNs, taking anti-malarials) and services (timely demand for health, institutional or community services when noticing signs and symptoms of malaria); (iii) percentage of people who believe the majority of their friends and communities practice the behaviours (using ITNs and seeking counselling and health care services); (iv) percentage of people who identify the mosquito as a cause/vector of malaria; (v) percentage of people who recognize the main signs and symptoms of malaria; (vi) percentage of people who know about treatment for malaria; (vii) percentage of people who know malaria prevention measures; (viii) percentage of households with at least one ITN; (ix) percentage of households with one ITN for every 2 people; (x) percentage of people with access to mosquito nets; xi) percentage of people who slept under an ITN the night before the survey; and, xii) use/access ratio of mosquito nets: behaviour indicator.

Data and statistical analysis

After conducting the study, the previously coded questionnaires were reviewed to verify the responses and their validation; later, the data were entered in SPSS for Windows, version 23.0 (IBM; Armonk, NY, USA). Data analysis was based on descriptive and inferential statistical analysis,

Sociodemographic characteristics of participants

A total of 773 household heads were interviewed of whom 59% were females and 41% males (Table 1 ). The mean age was 34.6 years (range 18–90 years; standard deviation: 7.7). About 27.4% were illiterate, 63% had completed primary school, and 8.4% had basic education. About half of the respondents (49.9%) lived in households with 4 to 6 people, a quarter (26.3%) in households composed of 1 to 3 members, and the remainder in households with 7 members or more (23.8%). Detailed socio-demographic characteristics are presented in Table 1 .

Knowledge regarding malaria prevention and treatment

Table 2 shows that within 773 household heads, about 96% reported having heard and 3.3% never having heard about malaria. However, the majority of respondents (96.4%), recalled hearing about malaria in the previous 6 months.

Most respondents (83.4%) reported that malaria is transmitted through mosquito bites. Regarding recognition of malaria symptoms, headache was pointed to as one of the main symptoms of malaria by both male (48.9%) and female (48.7%) respondents. Body pain was the second most mentioned symptom by 39.1% of males and 34.2% of females. About 3.9% of respondents were unable to identify any symptoms of malaria. In Table 3 , it can be noted that the most frequently reported malaria preventive measure was the use of ITNs (72.2%). More men (76%) reported the use of an ITN than women (70%), burning garbage and creating smoke to chase away mosquitoes (35%) and improving the cleanliness and hygiene of house and yard (24%) were other preventive measures mentioned by more than a fifth of respondents. IRS was one of the least mentioned forms of prevention (4.3%). The use of insecticide products and repellents were rarely mentioned, 3.1% and 1.8%, respectively.

Practices of malaria prevention, ownership and use of bed nets in target communities

About 80% of respondents reported having at least one bed net hanging at home, 21% reported having only one, 37.2% having two, 26.7% having three, and 13.3% having four or more. Most of the respondents (97.7%) reported that they received bed nets through the mass universal coverage ITN distribution campaign, and 82.7% reported sleeping under the bed net the night before the survey (Tables 4 and 5 ).

Beneficiaries’ attitudes towards malaria prevention, diagnosis and treatment

From the 721 respondents that reported exposure to information about malaria, 86.3% felt confident about their knowledge of how to prevent malaria and 96.4% knew where to get treatment. Regarding the use of bed nets at night, 96.9% of respondents considered it an important prevention from malaria. Most of the respondents (69.2%) reported that family members, friends and neighbours influence their decision-making regarding their health and 23.6% disagreed with this assertion (Table 6 ).

Table 7 shows attitudes regarding measures adopted for malaria diagnosis and febrile symptoms. From the 721 heads of household that had heard of malaria, 539 (74.8%) reported a household member with a fever in the previous 6 months. In 409 households a family member was reported to have had a fever in the two weeks prior to the survey, among whom 395 reported seeking counselling and treatment from health facilities, 10 from the market, and 4 from other places. Participants reported that they sought care at the health facility services because it had better quality/was more efficient (66.1%), and was less expensive (22.0%). The person who decided where to seek counselling and treatment was generally the head of the household (72.9%), followed by the spouse of head of household (16%), then the person with fever (9.0%). From 395 households, 359 reported a rapid diagnostic test (RDT) was taken by a health worker and the remaining 39 were not tested. Of 359 respondents who were tested, 89.7% obtained a positive result, 0.8% negative, and 9.5% did not know their result.

Attitudes of beneficiaries regarding malaria prevention

As shown in Table 8 , around 70.5% of respondents felt that community volunteers were ready to disseminate key malaria prevention messages. From 721 respondents, 76.1% identified worship places (churches and mosques) where they heard key messages on malaria prevention. Some 79.1% strongly agreed that community dialogue sessions helped them better understand how to prevent malaria. For the respondents, volunteers/activists/teachers played an important role in the dissemination of key malaria prevention messages.

Most malaria prevention strategies are centred on human behaviour and SBC interventions are a key part of the NMCP malaria strategy. This study assessed malaria prevention and treatment KAP in two rural Mozambican districts, Namacurra and Nicoadala.

The results show that almost all respondents had heard about malaria in the previous 6 months and those in these rural Zambezia districts have at least some knowledge of malaria causes, symptoms, treatment, and preventive measures. These results are similar to those obtained in other studies [ 7 , 8 , 9 , 15 ], implying that the SBC campaigns of previous years have been successful at reaching people in rural Mozambique and somewhat successful at disseminating education messages. However, it is important to note that although most respondents knew that malaria is transmitted by the mosquito bite, they did not associate it with other people (i.e., with “bites of mosquito which bit a malarial patient”). This lack of knowledge has been reported before [ 8 , 15 ], and is an indication that messaging on this aspect of transmission needs to be improved. Headaches were identified as the main symptom of malaria, similar to a study conducted by Khumbulani et al. [ 16 ]. However, despite relatively good knowledge of malaria symptoms and signs, respondents failed to name anaemia and convulsions. This lack of information could lead to a delay in seeking appropriate care from health facilities or community health workers. It is important to improve the training of local health community actors and subsequently improve dissemination and explanation to beneficiaries.

Most respondents felt confident and knew about malaria prevention methods and where to seek treatment, and considered the use of bed nets important to prevent and protect from malaria, which is similar to a study conducted in Ethiopia where the majority of respondents considered the mosquito net a protective measure against mosquito bites [ 17 ]. ITNs are a key part of malaria prevention strategies. ITNs are distributed through key channels, with most distributed through mass distribution campaigns and through antenatal care consultations [ 1 ]. In Mozambique 68% of the population sleeps under a bed net (40% under LLINs). These results show that previous SBC campaigns on bed nets have likely been successful in rural Zambezia as this were among the most recognized prevention form among the respondents, most of them having at least one bed net hanging at home which they use every night. Additionally, most reported sleeping under a bed net the night before the survey, which is similar to other studies conducted in the country [ 7 , 15 , 18 ].

Despite the wide availability of bed nets in the region and the indication from the results that the community uses them, the prevalence of malaria in Zambezia Province in children aged 6 to 59 months (using the malaria diagnostic test) increased from 38.3% in 2011 [ 19 ], 40.2% in 2015 [ 20 ] and 39% in 2018 [ 15 ]. Whilst efforts to support improved and regular use of nets by all is required, which may need more nuanced messaging for different audiences, other methods of vector control may also need to be investigated for a further reduction in prevalence.

In this study, health facilities were most commonly used for malaria treatment. This observation is similar to other studies [ 7 , 21 ], and it was pointed out as being the more effective and less expensive place to go to. The decision about where to go to receive treatment was the responsibility of the head of the household. Data show that family members, friends and neighbours were the greatest influencers on decision-makers regarding health of members of the household, similar to a study conducted in Nigeria, where it was indicated that family members play a role on health decisions [ 22 ]. This reinforces the importance of having strategies and approaches in SBC to target greatest influencers.

The majority of respondents consider community volunteers/activists/teachers are very well trained and they play an important role in disseminating key malaria preventive messages. These findings are collaborated by a study conducted in Kenya which showed that community actors are very well accepted during community implementation of SBC interventions [ 23 ]. Worship places (churches and mosques) are where respondents heard malaria preventive messages, and community dialogue sessions helped them better understand how to prevent malaria. Similarly, in a study conducted in Nampula Province, Mozambique, respondents affirmed that community dialogues helped communication and ultimately encouraged malaria-related behaviour [ 24 ]. These results indicate that the people implementing SBC interventions in Mozambique aid in dissemination of information and are accepted interlocutors by the beneficiaries.

This study has some limitations: it targeted the head of household as a proxy to KAP held by all members of a household. Ideally, a broader sampling method across the range of adults within Nicoadala and Namacurra communities should have been used. However, this was not possible due to funding constraints. The results may not accurately represent the community’s perspectives as a whole. Another limitation of the study is that people in these communities could have obtained information from sources other than the formal SBC interventions, e.g., social media or informal conversations. One cannot attribute all of the success to previous formal SBC interventions. All reported behaviours were self-reported and may have been affected by social desirability bias. Additionally, the questionnaire was not designed to document net quality. Low net quality might have influenced the effectiveness of the intervention, as noted in other studies from Mozambique [ 25 ].

This study confirms that SBC interventions carried out in rural Mozambique have had many successes. Namely, most people have knowledge of malaria prevention and treatment. The study shows that beneficiaries have bed nets and widely report the use of them. Beneficiaries recognized the role of community actors (teachers, community health workers, religious leaders) in dissemination of malaria key preventive messages. It was identified that there is room for improvement on SBC messaging regarding malaria symptoms. Specifically, that people recognize anaemia and convulsions as malaria symptoms so that they quickly seek health care services. Moreover, the study raises the need for further research into the main drivers of malaria prevalence in the country and the need to conduct operational research to refine SBC approaches.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Abbreviations

Insecticide-treated nets

Social and behavioural change

Universal coverage campaign

Antenatal care

National Malaria Control Programme

Insecticide-treated net

Indoor residual spraying

Knowledge, attitude and practice

Rapid diagnostic test

WHO. World Malaria Report. Geneva, World Health Organization; 2019. Available from: https://www.who.int/publications-detail/world-malaria-report-2019 . Accessed 29 January 2020.

WHO. Global technical strategy for malaria 2016–30. Geneva, World Health Organization, 2015. Available in: https://www.who.int/malaria/areas/global_technical_strategy/en/ . Accessed 20 September 2019.

Ministério da Saúde: National Malaria Controlo Programme. Malaria Strategic Plan 2017–2022. Maputo, Mozambique, 2017.

Zalisk K, Herrera S, Inyang U, Mohammed AB, Uhomoibhi P, Yé Y. Caregiver exposure to malaria social and behaviour change messages can improve bed net use among children in an endemic country: secondary analysis of the 2015 Nigeria Malaria Indicator Survey. Malar J. 2019;8:121.

Article   Google Scholar  

President Malaria Initiative (PMI). Malaria Initiative Technical Guidance, 2018

Roll Back Malaria Partnership to End Malaria. The Strategic Framework for Malaria Social and Behavior Change Communication 2018–2030. Available in: https://endmalaria.org/sites/default/files/RBM%20SBCC%20Framework%202018-2030%20English.pdf . Accessed 29 January 2020.

Mazingo HD, Obasy E, Mauka W, Manyiri P, Zinga M, Kweka EJ, et al. Knowledge, attitudes, and practices about malaria and its control in rural northwest Tanzania. Malar Res Treat. 2020;2010:794261.

Nejati J, Moosa-Kazemi SH, Saghafipour A, Soofi K. Knowledge, attitude and practice (KAP) on malaria, from high malaria burden rural communities, southeastern Iran. J Parasit Dis. 2018;42:62–7.

Musoke D, Karani G, Sempebwa JC, Etajak S, Guwatudde D, Musoke MB. Knowledge and practices on malaria prevention in two rural communities in Wakiso District. Uganda Afr Health Sci. 2015;15:401–12.

Ministerio da Administracao Estatal. Perfil do distrito de Nicoadala, 2014. Available from: http://www.maefp.gov.mz/wp-content/uploads/2017/04/Nicoadala.pdft . [Accessed in 8 February 2020]

Ministério da Saúde. Programa Nacional de Control da Malaria. Direção Provincial de Saúde Zambézia. Relatório anual. Boletim Epidemiológico de Saúde (BES), SISMA. Maputo, Mozambique, 2017.

Ahmed SM, Haque R, Haque U, Hossain A. Knowledge on the transmission, prevention and treatment of malaria among two endemic populations of Bangladesh and their health-seeking behaviour. Malar J. 2009;8:173.

Ministerio da Saude: Direccao nacional de Saude Publica. Estrategia Nacional de Promocao de Saude (2015 – 2019). Maputo, Mozambique, 2015.

Ministerio da Saúde. Direcção Provincial da Saúde da Zambezia. Relatório anual 2017. Maputo, Mozambique, 2017.

Instituto Nacional de Saúde (INS). Inquérito Nacional Sobre Indicadores de Malária 2018. Maputo, Moçambique, 2018.

Hlongwana KW, Mabaso ML, Kunene S, Govender D, Maharaj R. Community knowledge, attitudes and practices (KAP) on malaria in Swaziland: a country earmarked for malaria elimination. Malar J. 2009;8:29.

Astatkie A. Knowledge and practice of malaria prevention methods among residents of Arba Minch Town and Arba Minch Zuria district, southern Ethiopia. Ethiop J Health Sci. 2010;20:185–93.

PubMed   PubMed Central   Google Scholar  

Arroz JAH, Candrinho B, Mendis C, Varela P, Pinto J, Martins MDRO. Effectiveness of a new long-lasting insecticidal nets delivery model in two rural districts of Mozambique: a before-after study. Malar J. 2018;17:66.

República de Moçambique, Ministério da Saúde, Instituto Nacional de Estatística; MEASURE DHS/ICF International. Inquérito demográfico e de saúde 2011. Maputo, Mozambique, 2011. Available from: http://www.dhsprogram.com/pubs/pdf/FR266/FR266.pdf [Accessed 19th September 2017].

Ministério da Saúde (MISAU), Instituto Nacional de Estatística (INE), ICF Internacional, Inquérito de Indicadores de Imunização, Malária e HIV/SIDA em Moçambique 2015. Maputo, Moçambique. Rockville, Maryland, EUA: INS, INE and ICF International.

Sumba PO, Wong SL, Kanzaria HK, Johnson KA, John CC. Malaria treatment-seeking behaviour and recovery from malaria in a highland area of Kenya. Malar J. 2008;7:245.

Campbell TH, McDaniel SH, Cole-Kelly K. Family issues in Health care. In: Taylor RB, Ed. Family Medicine Principles and Practices. 6th Edn. New York: Springer-Verlag; 2002.

Owek CJ, Oluoch E, Wachira J, Estambale B, Afrane YA. Community perceptions and attitudes on malaria case management and the role of community health workers. Malar J. 2017;16:272.

Hildon ZJ, Escorcio-Ymayo M, Zulliger R, de Aramburu RA, Lewicky N, Harig H, et al. “We have this, with my husband, we live in harmony”: exploring the gendered decision-making matrix for malaria prevention and treatment in Nampula Province. Mozambique Malar J. 2020;19:133.

Article   CAS   Google Scholar  

Abílio AP, Obi E, Kilian A, Koenker H, Babalola S, Saifodine A, et al. Monitoring the durability of the long-lasting insecticidal nets MAGNet and Royal Sentry in three ecological zones of Mozambique. Malar J. 2020;19:209.

Download references

Acknowledgements

The authors would like to acknowledge the contributions of the anonymous community members who participated in the interviewing process. They would like to especially thank the household heads for participating in this study. The authors also would like to acknowledge the Zambezia health authorities for the administrative authorization for the study.

This study was funded by the Principal Investigator as part of her Ph.D. thesis. The findings and conclusions in this report are those of the authors.

Author information

Authors and affiliations.

Mozambique Medical Coucil, Maputo, Mozambique

Liliana de Sousa Pinto & Jorge A. H. Arroz

Global Health and Tropical Medicine, GHTM, Instituto de Higiene E Medicina Tropical, IHMT, Universidade Nova de Lisboa, UNL, Rua da Junqueira 100, 1349-008, Lisboa, Portugal

Maria do Rosário O. Martins & Zulmira Hartz

Cactus Communications, A/603, Satellite Gazebo, Guru Hargovindji Marg, Andheri East, Mumbai, 400093, Maharashtra, India

Nuria Negrao

Plataforma Inter-Religiosa de Comunicação para Saúde–PIRCOM, Maputo, Mozambique

Victor Muchanga & Amadeu Cossa

Independent Consultant, Maputo, Mozambique

Rose Zulliger

You can also search for this author in PubMed   Google Scholar

Contributions

LP conceived and designed the study protocol, analysed the data, drafted the manuscript, and made final revisions. JA and MROM supported the protocol design, performed sample calculations, analysed data, and reviewed the manuscript. RZ supported the design of data collection tool (questionnaire) and reviewed the final manuscript. ZH critically reviewed the manuscript. VM and AM analysed the data and reviewed the manuscript. NN contributed to the writing, editing, and revision of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Liliana de Sousa Pinto .

Ethics declarations

Ethics approval and consent to participate.

The study was administratively authorized by the Provincial Health Directorate of Zambezia and received authorization from the National Committee on Bioethics in Health (Ref 308/CNBS/2018).

The participants were informed about the objectives of the study. They signed an informed consent document to ensure the willingness of participation and they were free to withdraw from the study at any time. Identification numbers were used instead of participant names to maintain the confidentiality throughout the study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

de Sousa Pinto, L., Arroz, J.A.H., Martins, M.d.R.O. et al. Malaria prevention knowledge, attitudes, and practices in Zambezia Province, Mozambique. Malar J 20 , 293 (2021). https://doi.org/10.1186/s12936-021-03825-9

Download citation

Received : 07 September 2020

Accepted : 17 June 2021

Published : 30 June 2021

DOI : https://doi.org/10.1186/s12936-021-03825-9

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Social and behaviour change

Malaria Journal

ISSN: 1475-2875

• Submission enquiries: [email protected]

Europe PMC requires Javascript to function effectively.

Either your web browser doesn't support Javascript or it is currently turned off. In the latter case, please turn on Javascript support in your web browser and reload this page.

• Open access
• Published: 21 July 2022

Tracking malaria health disbursements by source in Zambia, 2009–2018: an economic modelling study

• Michael Mtalimanja 1 ,
• Kassim Said Abasse 2 ,
• Muhammad Abbas 3 ,
• James Lamon Mtalimanja 4 ,
• Xu Zhengyuan 1 ,
• DuWenwen 1 ,
• Andre Cote 2 &
• Wei Xu 1 , 3  

Cost Effectiveness and Resource Allocation volume  20 , Article number:  34 ( 2022 ) Cite this article

2390 Accesses

1 Altmetric

Metrics details

Zambia has made profound strides in reducing both the incidence and prevalence of malaria followed by reducing malaria related deaths between 2009 and 2018. The number of partners providing malaria funding has significantly increased in the same period. The increasing number of partners and the subsequent reduction of the number of reported malaria cases in the Ministry of Health main data repository Health Management Information System (HMIS) stimulated this research. The study aimed at (1) identifying major sources of malaria funding in Zambia; (2) describe malaria funding per targeted interventions and (3) relating malaria funding with malaria disease burden.

Data was collected using extensive literature review of institutional strategic document between the year 2009 to 2018, assuming one-year time lag between investment and the health outcome across all interventions. The National’s Health Management Information System (HMIS) provided information on annual malaria admission cases and outpatient clinic record. The statistical package for social sciences (SPSS) alongside Microsoft excel was used to analyze data in the year 2019.

The investigation observed that about 30% of the funding came from PMI/USAID, 26% from the global funds, the government of Zambia contributed 17% and other partners sharing the remaining 27%. Multivariate regression analysis suggests a positive correlation between reducing reported malaria disease burden in HMIS 2009–2018 and concurrent increasing program/intervention funding towards ITNs, IRS, MDA, and Case Management with r 2  = 77% (r 2  > 0.77; 95% CI: 0.72–0.81). Furthermore, IRS showed a p-value 0.018 while ITNs, Case Management and MDA having 0.029, 0.030 and 0.040 respectively.

Our findings highlight annual funding towards specific malaria intervention reduced the number of malaria admission cases.

Malaria infections are caused by a plasmodium parasite which spreads to people through the bites of infected female Anopheles mosquitoes. In 2017, an estimated 229 million cases of malaria were recorded worldwide and 409,000 people died, mostly children in the African region [ 1 ]. Malaria occurs in more than 100 countries and territories. About half of the world's population is at risk. Large areas of Africa and South Asia and parts of Central and South America, the Caribbean, Southeast Asia, the Middle East, and Oceania are considered areas where malaria transmission occurs. In the global community, about 111 countries globally have eliminated malaria and another 35 countries, are making progress toward elimination of the disease [ 2 ]. The African region accounts for most global cases of malaria (88%), followed by the Southeast Asia region (10%) and the Eastern Mediterranean region (2%).

In Zambia plasmodium falciparum malaria is endemic throughout the country, with the main transmission season being between November and March every-year. The Malaria Indicator Survey (2008) indicates that the country’s average parasite rate was found to be 10%, with some parts of the country reporting less than 1%, while others still have high parasite prevalence rates of up to 20%-30% [ 3 , 4 ].

Malaria has for a long time remained the leading cause of morbidity and mortality in Zambia with recent statistics suggesting malaria still being the leading cause of morbidity and the second leading cause of mortality, surpassed only by HIV and AIDS. Also, malaria accounts for up to 40% of all infant mortality and 20% of all maternal mortality in Zambia, and represents a major socio-economic burden on the country, particularly on the communities living in malaria endemic areas [ 5 , 6 ].

Zambia has shown strong growth in the last decade, reaching lower-middle-income status. Nevertheless, the health sector continues to be dependent on external resources, which has accounted for over 60% percent of health expenditure in recent years. Out-of-pocket health expenditure contributes 12.8% to total health expenditure. Private medical schemes and insurance account for about 5% [ 7 ].

Malaria remains a major public health problem in Zambia, despite significant progress made in fighting the disease in the last decade. Malaria prevalence varies across all provinces and districts with 18 million people at risk, including the most vulnerable groups, such as pregnant women and children. The country’s last two iterations of the National Malaria Strategic Plan aimed to reduce transmission and in the current NMSP (2017–2021) the government of Zambia through the ministry of health and the national malaria elimination program (NMEP) adopted an ambitious agenda to eliminate malaria by use of scientific proven interventions for prevention, control, curative, and inclusion of new tools of innovations in strengthening of routine surveillance at all levels. The efforts towards nationwide malaria elimination with regard to malaria case management, emphasizes the need to have diagnostic and curative services as close to home as possible, utilizing community health workers as extensions for the health facility within the community [ 8 , 9 , 10 ]. Despite a better understanding of pathophysiology and management of malaria, childhood mortality remains unacceptably high [ 11 ].

Over the past 10 years, Zambia has significantly intensified efforts against malaria by initiating and scaling up the implementation of internationally accepted strategies and best practices for prevention, treatment, and care for malaria [ 12 , 13 ]. These include: vector control, through indoor residual spraying (IRS) and the promotion of ownership and correct use of insecticide-treated bed nets (ITNs); intermittent preventive treatment in pregnancy (IPTp); prompt and effective malaria case management; Coartem (artemether/lumefantriine) use; and the introduction and scaling up of Rapid Diagnostic Tests (RDTs) in health facilities that do not have microscopy services [ 14 ]. On the other hand, if no new control measures are developed, the malaria death toll is expected to raise as actual figures could be potentially higher as a result of under-reporting and challenges in diagnosis [ 15 ].

It was observed that the increasing malaria disease burden can only be contained through harnessing, harmonization and coordination of all the available resources, to maximize the benefits from synergies. In this respect, the national malaria control program (NMCP) has successfully established strong partnerships with the communities, other government line ministries and departments, the faith-based health sector under the coordination of the private sector, civil society, and the global community. Strong, effective, and coordinated partnerships have been established with the global community, through the RBM Partnerships, leading to significant technical, financial, and logistical support. Also, reference documents such as the world malaria report present a scope on source of funding partners and reflects their mandate and area of support. However, provided funding is sustained there is need to harness the fragmented information into one document thereby providing an opportunity to gauge and categories related interventions. In addition, the main data repository for ministry of health (DHIS) only present data in its raw form without linking the disease burden to available funds and interventions. The researcher believes that malaria disease burden, (trends and prevalence) cannot be understood fully unless there is an attempt to discuss the disease burden in light of the country’s funding profile and specific malaria interventions. Therefore, study aims at (1) identifying major sources of malaria funding in Zambia; (2) describing malaria funding per targeted interventions and (3) relating malaria funding with malaria disease burden between 2009 and 2018.

Study design

A retrospective cross-sectional study focused on institutional documents of the ministry of health main data repository. We performed a time series regression analysis with delta of one year factoring in data collected at time point 2009 to 2018 to capture any change that could have been repeated at each successive equally spaced time points to follow a sequence and an average of all time point taken to enable investigation of the pattern of change over time. Statistical package for social sciences (SPSS) alongside STATA version 16, were used to describe data and explain the relationship between the dependent variable malaria disease incidence by adopting and including seven (7) independent variables; insecticide treated nets (ITNs), case management (CM), indoor residual spray (IRS), mass drug administration (MDA), monitoring and evaluation (M&E), entomological studies (ES), information education and communication (IEC).Also stationarity test and variance inflation factor were employed on variables to investigate predictability and collinearity among variables respectively.). Thus, we adopted and included all the seven predictors related to the dependent variable into the model taking a form of standard multiple-linear equitation of the form Y = a + Bx 1  + BX 2  + BX 3 .

Statistical analysis used descriptive statistics that were quantitatively summarized from a collection of information from documents which included annual financial reports, and partner country operational plans which provide data on actual annual disbursement while other documents like medium term expenditure frameworks, strategic plans, evaluation of country programs, to mention but a few provide data on the budgets estimates and their assumption. Interestingly, all contributions received by the ministry of health from varies partners are expressed as annual disbursement. Also, administrative data such as malaria incidence and admissions were extracted from the District Health Management Information System accessed on ( http://www.dhis2.org.zm/hmis ).

Hypothesis testing was employed using the ANOVA single factor analysis of variance on malaria.

In addition, prior to data collection, ethical clearance was sought and granted in China from the institutional Review Board of China pharmaceutical University and approval from relevant authorities including Ministry of Health Public Health and Research Unit Zambia.

Additional file 1 Shows funding partners and reflects their mandate and area of support in the period under review. Contributors to malaria funding for malaria prevention, treatment and control in Zambia included Presidential Malaria Initiative (PMI-USAID), The Global Fund (GF), World Bank (WB), World Health Organization (WHO), United Nations International Children's Emergency Fund (UNICEF), Program for Appropriate Technology in Health (PATH), Malaria Control and Elimination Partnership in Africa (MACEPA) and Department for International Development (DfID) United Kingdom among others.

Disbursements directed to specific interventions in the study period showed a huge and continuous fluctuation in the annual funding disbursement from government and other stakeholders. According to the proportion of funds towards interventions, about 30% (95% CI: 24.43–33.62) of the funding came from PMI/USIAD, 26% (95% CI: 24.72–28.53) from the global funds. The government contributed 17% (95% CI: 16.03–18.46) with other partners sharing the remaining percentage.

Our findings indicated insecticide treated nets (ITNs) receiving a substantial amount of funding in the past 10 years representing about 34.7% of the total funding followed by IRS with 26.9%, clinical case management 19.0%, monitoring and evaluation 11.2% and information communication education 7.8%. While Support towards provision of mass drug administration and entomological surveillance were at 0.4% and 0.08% respectively Additional file 2 . Additional file 3 Shows distribution of malaria intervention summarized as simple counts of proven support provided to ministry of health between the years 2009 and 2018. The results clearly indicates 26 institutions representing 24.1% of all institutions provided support towards Information Education and Communication (IEC), 23 institutions representing 21.3% support towards Indoor Residue Spray (IRS), 22 institutions representing (20.4%) support towards Case Management, 21 institutions representing 19.4% support towards provision of Insecticide Treated Nets (ITNs), 8 institutions representing 7.4% support towards M&E, 5 institutions representing 4.6% support towards Mass Drug Administration (MDA) and only 3 institutions representing 2.8% support towards entomological studies (ES).

Table 1 The District Health Information Management System (DHIS) indicated a systematic decline in number for patients admitted with malaria. Trend analysis of malaria admission showed over 60% (95% CI: 56.42–62.32) reduction from 176,664 to 68,898 admission cases denoting on average a reduction of 140,533 cases between the year 2009 and 2018. Data also suggest variation in annual malaria admission.

Table 2 Shows a notable difference in terms of amount of funds received per intervention, with ITNs, IRS, Case management, Monitoring & evaluation, and IEC receiving comparably larger allocation of funds. While mass drug administration and entomological surveillance showed missing data in the first 6 consecutive years of the review period. The average annual disbursement for all interventions from the year 2009 to 2018 was K32, 739,863 equivalent to USD $ 2,751,248. Furthermore, the highest annual average disbursement was towards ITNs at $5,923,341 while the least annual average was entomological surveillance at $807.

Table 3 The single factor analysis of variance tests using the following null hypothesis: the number of admissions is equal and the alternative hypothesis: the number admissions is not equal, was performed to test equity of variances of reported malaria admissions by province. Results showed p -value < 0.00001 and significant at p  < 0.05. Thus, we reject the null hypothesis indicating that there is sufficient evidence to suggest that reported malaria cases by provinces are statistically different.

Table 4 Predictor variables explaining the relationship between funding and malaria incidence using the 95% confidence interval in the overall regression predictive model found a strong association between increasing funding towards selected key interventions and reducing occurrence of new cases of malaria.

Ministry of health has benefited from various malaria program interventions funds from more than 47 known partners. Local institutions and companies are also recorded to provide support mostly in interventions related to system straightening, program management and public health awareness. Thus, the study highlights the ministry of health having a long-standing, well established partnerships with a range of multilateral, governmental, nongovernmental, and private organizations providing support towards prevention, treatment and control of malaria in Zambia. It is worth noting that, through the district health management information system of the ministry of health which is mandated to collect routine data on service coverage and disease burden, analysis of malaria data indicated that there is a general decline in the number of reported malaria cases in the country [ 16 ]. Trend of malaria admissions falling over the years depicting over 60% (95% CI: 58.6–63.4) reduction between 2009 and 2018. However, the declining malaria disease burden is associated with geographical location [ 17 ]. Our research outcomes on single factor analysis of variance established a significant geographical variation in the number of reported malaria cases countywide with Eastern province recording highest number of patients admitted, followed by Luapula, North Western, Muchinga and Northern. While Southern and Lusaka reported lowest number of cases and a similar decline in annual malaria reported cases. On the other hand, Luapula province reported highest number of malaria deaths per annum. Copperbelt, Northern and Eastern provinces also reported high numbers of malaria related deaths. Although the number of reported malaria hospital admissions are seemingly high, the trend analysis showed declining malaria admission across all ten (10) provinces.

Furthermore, research results on insecticide treated nets showed a significant predictor value of 2.02612E−7 third after case management and indoor residual spraying in reducing the malaria burden provided one unit of addition investment towards funding. Interestingly, the use of ITNs in the fight against malaria in our study agrees with what was documented by Lengeler (2004) that in areas of stable malaria transmission, provisions of ITNs have potential to reduced parasite prevalence by 13%, uncomplicated malaria episodes by 50%, and severe malaria by 45% compared to equivalent populations with no nets [ 18 , 19 , 20 ]. Also, a study conducted in Zambia to determine ITNs integrity and insecticide content highlighted more availability of ITNs in households compared to previous years due to increases in funding for malaria control and continuous distribution of ITNs achieved through channels such as Antenatal Care, Expanded Program for Immunization and selected primary schools [ 21 ].

Analysis on indoor residual spraying was found to be significant in reducing the malaria disease burden with a 3.52625E-6 predictor value denoting greater change in health outcome per additional investment compared to other interventions in the study. Also, showed a consistent association between implementation of IRS and confirmed malaria case incidence, and a stronger association in reducing malaria disease burden. In Zambia IRS activities are conducted annually and these activities routinely include district-level planning and budgeting for targeted areas, assessment of spray structures, training of spray teams, supervision and monitoring of spray activities [ 22 ]

According to MOH (2011), Zambia IRS operations expanded from 5 districts in 2003, to 54 districts by 2014, with support from various partners including US President’s Malaria Initiative (PMI). The 2011–2015 National Malaria Strategic Plan recommended IRS in high-risk areas (a minimum of 85% of all targeted structures) with focalized IRS mounted in response to malaria surveillance data [ 23 , 24 ].

In rural areas access to health care still remains a significant obstacle due to long distances to health facilities and challenges with transport among others. Consequently, many rural patients with malaria do not present to a health facility in time for treatment [ 25 ]. Owing to this, community health workers accustomed to high malaria incidence presumptively treat patients for malaria based on their clinical symptoms and this could potentially reduce the number of confirmed malaria cases recorded at health facilities as these interventions are not captured on time. In addition, clinicians had for a long time practiced a common non-evidence based “fever equals malaria” and treated patients as such without laboratory test /RDT results [ 26 ]. Owing to the interventions mentioned, our research findings showed case management to be significant in reducing malaria disease burden from the perspective of routine malaria cases recorded in the HMIS-based clinical records.

Mass drug administration is also a well know malaria prevention and control intervention worldwide. A community randomized step-wedged control trail was conducted in Southern Zambia to access effectiveness of population-wide malaria testing and treatment with rapid diagnostic tests and Artemether-Lumefantrine showed a strong inverse relationship. Moreover, A clear relationship between provision of MDA and reducing HMIS reported malaria burden was noticed in the lower transmission strata, malaria parasite prevalence declined from 7.7% at baseline to 0.5% after the first two MDA rounds, an 87% larger decline than seen in the control group [ 27 ]. Similarly, our results showed MDA being significant in reducing disease burden though with the least predictor value of 1.37886E-8 when compared with other interventions analyzed and to the previous studies indicating mass drug administration having a strong predictor power to prevent the spread of the disease. The lower predictor power can be attributed to limited data in the years reported in the study.

However, data for information education communication and M&E were not significant in reducing disease burden and showed a week positive relationship between the intervention and annual malaria reported cases. Thus, this would imply that the more the community is informed or educated about malaria, the more likely they will visit the hospital to seek medical services and as such the number of reported cases recorded is expected to increase. In addition, M&E would suggest the number of reported malaria cases recorded increase as more activities to monitor collection of such data are employed.

Malaria transmission is driven by a complex interaction of the vector, parasite, human host, and the environment, and governed by different ecological and social determinants. Human population increase, developmental activities and associated ecological transformations have a significant impact on malaria epidemiology and have invariably exacerbated the situation. Malaria transmission depends markedly on local environmental conditions and other compounding factors, that is, presence of drug-resistant parasites and insecticide resistant vectors, environmental changes, and economy, poverty levels, climatic changes, natural disasters and political instability, adaptability of malaria vectors to changing environments and limited investment in research, and optimization of malaria vector control programs [ 28 ]. Thus, entomological surveillance is also important in the estimation of the expected impact of the various control measures on reducing disease burden though our research outcomes were not significant owing to missing data and also having the least percentage of 2.8% representing the number of institutions supporting the intervention. Data is suggestive of need to sustained investment in entomological surveillance to have adequate samples to measure mean mosquito densities.

In the period reported approximately $ 2.8 million from the selected key interventions was invested into different malaria prevention and curative and diagnostic programs each year. Though the finances ploughed into the key interventions are believed to be potentially higher than the estimation captured in this study, as donations and actual budget allocations from the government of Zambia through the ministry of health would increase the financial investment towards reducing malaria burden if incorporated.

Our findings are suggestive of the pronounced decline in malaria disease burden over the study periods as a consequence of increase in annual funding towards preventive, control and curative interventions. Also, much of this success in the decline could be credited to a combination of sustained economic growth, behavioral change, change in the use of agricultural pesticides or insecticide-like compounds not directly applied for targeting malaria vectors, changes in housing from traditional to modern houses, better surveillance, and improved access to health services in rural areas as a result of newly built health posts reducing the distance from homes to health facilities and an increased number of community health workers [ 29 , 30 , 31 ]. The overall regression predictive model in this study found a strong association between increasing funding towards selected key interventions and reducing occurrence of new cases of of malaria owing to research outcomes indicating about 77% of variations in HMIS-recorded clinical malaria burden could statistically be concurrent to increase in funding towards programs/interventions between the years 2009 and 2018. It is worth noting that, there may be an intriguing link to recent changes in social and ecological sittings that requires further investigation and other potential explanations for the observed decline in disease burden. Hence, need to include an assessment of the potential role of changes in global warming, proper water drainage during rain seasons, and agricultural related use of chemicals, land use, deforestation, and entomological surveillance to reducing the disease burden. Relatedly, the Zambia national malaria program performance review report of 2010, confirms that combined funding for malaria prevention methods (IRS, ITNS, and other vector bone methods) are more compared to treatment and diagnostics methods due to their huge impact on the disease burden [ 32 ].

Methodological problems were encountered in this research with primary limitations of this research method being the high cost of research owing to the number of study areas needed in time-series design, the challenges in developing generalizable theoretical principles about community change processes through retrospective data collected, the obscuring of relationships that are unique to a subset of communities, and the problem of diffusion of intervention activities from intervention to control the area of study.

Moreover, it is a cross-sectional study in which the analysis was descriptive and recall bias due to the retrospective nature of data collection related to variables as well as desirability bias may further have an affect the results outcomes of the analysis. In addition, bias-variance may affect the results. Ideally, averaging multiple observations using information from larger regions would reduce the bias—variance from high variance resulting from algorithm modeling the random noise in the data in a model that is over fitted. Also, we conducted our research from the perspective of malaria disease burden routinely recorded in the HMIS-based clinical records while there is also the wider population-level burden seen through population-based household surveys e.g. malaria parasite infection prevalence in DHS, MICS or Malaria indicator survey which typically does not fully access records. In addition, investment will have a lag before the service reaches its beneficiaries. However, an assumption of a one-year time lag was suggested across all interventions while specific interventions such as ITNs distributions leads to protection over 2–3 years, IRS impact limited between 6 and 12 months following spraying and case management benefits in terms of avoiding hospitalization are relatively immediate. Hence, using fixed effects potentially produce similar results and differing on predictor power.

Malaria funding support is largely targeted towards mass drug administration, indoor residue spray, insecticide treated bed nets, clinical case management (provision of anti-malarial drugs, laboratory diagnostic equipment), entomological intervention, monitoring and evaluation, and information Education/ communication. Further studies are needed to evaluate the relationship between all know malaria interventions against disease burden. Also, finding more efficient and cost-effective ways of developing and evaluating interventions which could be of great benefit to individual health and well-being of a nation or region as addition lives will be saved if we learn how to develop area study interventions more efficiently.

Availability of data and materials

The dataset used/analysed are available from the corresponding author on request.

Romero M, Leiba E, Carrión-Nessi FS, et al. Malaria in pregnancy complications in Southern Venezuela. Malar J. 2021;20:186. https://doi.org/10.1186/s12936-021-03728-9 .

Article   PubMed   PubMed Central   Google Scholar  

WHO (World Health Organization). World malaria report 2015. Geneva: WHO; 2015.

Book   Google Scholar  

https://reliefweb.int/report/zambia/zambia-national-malaria-indicator-survey-2008

Nawa M, Hangoma P, Morse AP, et al. Investigating the upsurge of malaria prevalence in Zambia between 2010 and 2015: a decomposition of determinants. Malar J. 2019;18:61. https://doi.org/10.1186/s12936-019-2698-x .

https://www.who.int/teams/global-malaria-programme/reports/world-malaria-report-2018

Korenromp E, Hamilton M, Sanders R, et al. Impact of malaria interventions on child mortality in endemic African settings: comparison and alignment between LiST and Spectrum-Malaria model. BMC Public Health. 2017;17:781. https://doi.org/10.1186/s12889-017-4739-0 .

https://www.scidev.net/sub-saharan-africa/news/malaria-research-finds-new-home-in-zambia-ssa/

Kaonga O, Banda C, Masiye F. Hardship financing of out-of-pocket payments in the context of free healthcare in Zambia. PLoS ONE. 2019;14(4):e0214750. https://doi.org/10.1371/journal.pone.02147 ( Published 2019 Apr 10 ).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Chipukuma HM, Zulu JM, Jacobs C, et al. Towards a framework for analyzing determinants of performance of community health workers in malaria prevention and control: a systematic review. Hum Resour Health. 2018;16:22. https://doi.org/10.1186/s12960-018-0284-x .

Kok MC, Dieleman M, Taegtmeyer M, Broerse JE, Kane SS, Ormel H, Tijm MM, de Koning KA. Which intervention design factors influence performance of community health workers in low- and middle-income countries? A systematic review. Health Policy Plan. 2015;30(9):1207–27. https://doi.org/10.1093/heapol/czu126 .

Article   PubMed   Google Scholar  

https://www.moh.gov.zm/docs/Community%20Health%20Strategy.pdf

Dzeing-Ella A, Nze Obiang PC, Tchoua R, et al. Severe falciparum malaria in Gabonese children: clinical and laboratory features. Malar J. 2005;4:1. https://doi.org/10.1186/1475-2875-4-1 .

Chizema-Kawesha E, Miller JM, Steketee RW, Mukonka VM, Mukuka C, Mohamed AD, Miti SK, Campbell CC. Scaling up malaria control in Zambia: progress and impact 2005–2008. Am J Trop Med Hyg. 2010;83(3):480–8. https://doi.org/10.4269/ajtmh.2010.10-0035 .

Okumu FO, Moore SJ. Combining indoor residual spraying and insecticide-treated nets for malaria control in Africa: a review of possible outcomes and an outline of suggestions for the future. Malar J. 2011;10:208. https://doi.org/10.1186/1475-2875-10-208 .

Manyando C, Njunju EM, Chileshe J, et al. Rapid diagnostic tests for malaria and health workers’ adherence to test results at health facilities in Zambia. Malar J. 2014;13:166. https://doi.org/10.1186/1475-2875-13-166 .

Mutemwa RI. HMIS and decision-making in Zambia: re-thinking information solutions for district health management in decentralized health systems. Health Policy Plan. 2005;21(1):40–52. https://doi.org/10.1093/heapol/czj003 .

Lubinda J, Bi Y, Hamainza B, Haque U, Moore AJ. Modelling of malaria risk, rates, and trends: a spatiotemporal approach for identifying and targeting sub-national areas of high and low burden. PLoS Comput Biol. 2021;17(3): e1008669. https://doi.org/10.1371/journal.pcbi.1008669 .

Christian Lengeler (2004). Insecticide‐treated bed nets and curtains for preventing malaria cochrane systematic review—intervention version published: 19 April 2004 see what's https://doi.org/10.1002/14651858.CD000363.pub2 . Retrieved March 9, 2019.

https://www.who.int/malaria/publications/atoz/cochrane_reviewitns2004.pdf

https://www.who.int/teams/global-malaria-programme/reports/world-malaria-report-2020

Craig AS, Muleba M, Smith SC, et al. Long-lasting insecticidal nets in Zambia: a cross-sectional analysis of net integrity and insecticide content. Malar J. 2015;14:239. https://doi.org/10.1186/s12936-015-0754-8 .

Keating J, Yukich JO, Miller JM, et al. Retrospective evaluation of the effectiveness of indoor residual spray with pirimiphos-methyl (Actellic) on malaria transmission in Zambia. Malar J. 2021;20:173. https://doi.org/10.1186/s12936-021-03710-5) .

https://www.moh.gov.zm/docs/nhsp.pdf

Pinchoff J, Larsen DA, Renn S, Pollard D, Fornadel C, Maire M, Sikaala C, Sinyangwe C, Winters B, Bridges DJ, Winters AM. Targeting indoor residual spraying for malaria using epidemiological data: a case study of the Zambia experience. Malar J. 2016;15:11. https://doi.org/10.1186/s12936-015-1073-9 .

Littrell M, Miller JM, Ndhlovu M, Hamainza B, Hawela M, Kamuliwo M, Hamer DH, Steketee RW. Documenting malaria case management coverage in Zambia: a systems effectiveness approach. Malar J. 2013;12:371. https://doi.org/10.1186/1475-2875-12-371 .

Masaninga F, Sekeseke-Chinyama M, Malambo T, et al. Finding parasites and finding challenges: improved diagnostic access and trends in reported malaria and anti-malarial drug use in Livingstone district, Zambia. Malar J. 2012;11:341. https://doi.org/10.1186/1475-2875-11-341 .

Silumbe K, Yukich JO, Hamainza B, Bennett A, Earle D, Kamuliwo M, Steketee RW, Eisele TP, Miller JM. Costs and cost-effectiveness of a large-scale mass testing and treatment intervention for malaria in Southern Province, Zambia. Malar J. 2015;14:211. https://doi.org/10.1186/s12936-015-0722-3 .

Chanda E, Baboo KS, Shinondo CJ. Transmission attributes of periurban malaria in Lusaka, Zambia, precedent to the integrated vector management strategy: an entomological input. J Trop Med. 2012;2012:1–8. https://doi.org/10.1155/2012/873852 ( Article ID 873852 ).

Article   Google Scholar  

Chanda E, Hemingway J, Kleinschmidt I, Rehman AM, Ramdeen V, Phiri FN, et al. Insecticide resistance and the future of malaria control in Zambia. PLoS ONE. 2011;6(9): e24336. https://doi.org/10.1371/journal.pone.0024336 .

Ippolito MM, Searle KM, Hamapumbu H, Shields TM, Stevenson JC, Thuma PE, Moss WJ. House structure is associated with plasmodium falciparum infection in a low-transmission setting in Southern Zambia. Am J Trop Med Hyg. 2017;97(5):1561–7. https://doi.org/10.4269/ajtmh.17-0299 ( Epub 2017 Aug 18. PMID: 28820722; PMCID: PMC5817773 ).

Chipukuma HM, Halwiindi H, Zulu JM, et al. Evaluating fidelity of community health worker roles in malaria prevention and control programs in Livingstone District, Zambia-a bottleneck analysis. BMC Health Serv Res. 2020;20:612. https://doi.org/10.1186/s12913-020-05458-1 .

https://endmalaria.org/sites/default/files/Zambia-The-malaria-program-performance-review-2010-Report.pdf

Download references

Acknowledgements

We would like to express our special thanks to the Zambian Ministry of Health, Monitoring and Evaluation unit for giving us access to the district health information system (DHIS2) to extract primary data for this study.

There is no funding for this study.

Author information

Authors and affiliations.

School of International Pharmaceutical Business, China Pharmaceutical University, Nanjing, 211198, Jiangsu, China

Michael Mtalimanja, Xu Zhengyuan,  DuWenwen & Wei Xu

Centre de Recherche en Gestion Des Services de Sante, Faculté Des Sciences de L’administration (FSA), Université Laval (UL), Centre Hospitalière Universitaire (CHU) de Québec UL-IUCPQ-UL, Québec, QC, Canada

Kassim Said Abasse & Andre Cote

Riphah Institute of Pharmaceutical Sciences, Riphah International University, Islamabad, Pakistan

Muhammad Abbas & Wei Xu

Department of Monitoring and Evaluation, Ministry of Health, P.O Box, 30205, Lusaka, Zambia

James Lamon Mtalimanja

You can also search for this author in PubMed   Google Scholar

Contributions

MM, WX, and AC contributed to the conceptualization of the study. The design by MM, KSA, MA, JLM and DWW. Data collection by MM, JLM, MM, XZY, DWW, KSA and JLM analysed and interpreted the data.MM and XZY wrote the first draft of the manuscript. MA, and WX revised the article for important intellectual content. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Wei Xu .

Ethics declarations

Ethics approval and consent to participate.

This study received ethical approval in China from the institutional review Board of China Pharmaceutical University (Nanjing, China). While in Zambia from the National Health Research Authority (NHRA), Ministry of Health Public Health and Research Unit and relevant authorities in the Ministry of Health.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interest.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1.

. Sources of Malaria Disbursement in Zambia.

Additional file 2

. Proportion of disbursement towards interventions.

Additional file 3

. Support towards malaria intervention.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Mtalimanja, M., Abasse, K.S., Abbas, M. et al. Tracking malaria health disbursements by source in Zambia, 2009–2018: an economic modelling study. Cost Eff Resour Alloc 20 , 34 (2022). https://doi.org/10.1186/s12962-022-00371-2

Download citation

Received : 06 February 2022

Accepted : 05 July 2022

Published : 21 July 2022

DOI : https://doi.org/10.1186/s12962-022-00371-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Malaria interventions
• Disease Burden
• Malaria funding

Cost Effectiveness and Resource Allocation

ISSN: 1478-7547

• General enquiries: [email protected]

• Research article
• Open access
• Published: 02 July 2020

Evaluating fidelity of community health worker roles in malaria prevention and control programs in Livingstone District, Zambia-A bottleneck analysis

• Helen Mwiinga Chipukuma 1 ,
• Hikabasa Halwiindi 2 ,
• Joseph Mumba Zulu 3 ,
• Steven Chifundo Azizi 4 &
• Choolwe Jacobs 4  

BMC Health Services Research volume  20 , Article number:  612 ( 2020 ) Cite this article

3677 Accesses

7 Citations

Metrics details

Community Health Workers (CHWs) are an important human resource in improving community malaria intervention coverages and success in reducing malaria incidence has been attributed to them. However, despite this attribution, malaria resurgence cases have been reported in various countries including Zambia. This study aims to evaluate the implementation fidelity of CHW roles in malaria prevention and control programs in Livingstone through performance and service quality assessment.

A mixed method concurrent cross-sectional study based on quantitative and qualitative approaches was used to evaluate performance and service quality of the CHW roles for selected catchments areas in Livingstone district. For the quantitative approach, (34) CHWs were interviewed and a community survey was also done with 464 community participants. For qualitative approach, two focused group discussions with CHWs and three key informant interviews from the CHW supervisors were done.

Overall implementation fidelity to the CHW roles was low with only 5(14.7%) of the CHWs having good performance and least good quality service while 29 (85.3%) performed poorly with substandard service. About 30% of house-holds reported having experienced malaria cases but CHWs had low coverage in testing with RDT (27%) for malaria index case service response with treatment at 14% coverage and provision of health education at 23%. For other households without malaria cases, only 27% had received malaria health education and 15% were screened for malaria. However, ITN distribution, sensitization for IRS were among other CHW services received by the community but were not documented in CHW registers for evaluation. Factors that shaped fidelity were being married, record for reports, supervision, and work experience as significant factors associated with performance. Lack of supplies, insufficient remuneration and lack of ownership by the supervising district were reported to hinder ideal implementation of the CHW strategy.

Fidelity to the malaria CHW roles was low as performance and quality of service was poor. A systems approach for malaria CHW facilitation considering supervision, stock supply and recruiting more CHWs on a more standardized level of recognition and remuneration would render an effective quality implementation of the CHW roles in malaria.

Peer Review reports

Global reports indicate a decline in Malaria incidence by 37% and Malaria mortality rate by 60% between 2000 and 2015 [ 1 , 2 ]. This has been in part been attributed to Community interventions through Community Health Workers (CHWs) who are a link between the community and the health facility [ 2 , 3 , 4 ]. However, the threat of resurgent Malaria is present in various countries including Zambia, Madagascar and Cameroon because of weakening of the Malaria control programs that has been linked to lack of funding, complacency with the malaria situation coupled with poor malaria program execution, purposeful cessation of control activities and lack of cooperation from the community. Great awareness of the threat of resurgent malaria and the development of systems to minimize the decline in malaria incidence and mortality are key to further progress in malaria control [ 5 ].

The World Health Organization (WHO) has proposed three main pillars that, if implemented would to move efforts towards elimination of Malaria. These pillar are; to ensure universal access to malaria prevention, diagnosis and treatment; to accelerate efforts towards elimination of Malaria and attainment of Malaria-free status, and; to transform malaria surveillance into a core intervention [ 2 , 3 ]. These pillars can be best achieved through the primary health care (PHC) strategy that most countries adopted as policy after the 1978 Declaration of Alma-Ata. Community Health Workers are key actors in PHC as links to the community members, and their involvement has proved to be a good initiative in Malaria control activities [ 4 , 6 , 7 ].

Community Health Workers are men and women chosen by the community, and trained to deal with individual and community health problems, and working closely with the formal health care system [ 2 , 8 , 9 ]. Community Health Worker is a general term used for all health volunteers, assisting in community health regardless of a type of disease. However, different names are used in different countries but for the purpose of literature search in the document, the term Community health worker was dominant in the search engines and as such the term was used predominantly. However, national wide CHWs who deal with malaria issues are called Community Malaria Agents (CMA). Over the past decades, studies have shown that CHWs can help reduce morbidity and mortality in settings that have traditionally lacked access to health care [ 7 , 10 , 11 ]. Crucial to sustainable success of CHW programs is strengthening the health system capacity to support with commodity supply, supervision, and appropriate treatment of referred [ 2 , 4 , 12 ]. The CHW strategy, as adopted by most countries in relation to helping improve intervention coverages has proved to be a good strategy in Malaria control activities [ 6 ]. However, despite the attribution of reduction in incidence of morbidity and mortality, malaria resurgence cases have been reported in various countries including Zambia thus raising several questions on the implementation of CHW programs [ 5 ].

In Zambia, malaria continues to be a public health problem, a major cause of morbidity and mortality in the country, resulting in approximately 6 million cases and 2000 deaths despite substantial progress made over the past decade (HMIS, 2016). Efforts to control malaria are currently being scaled up through coordinated efforts in community interventions such as Indoor residual spraying in the communities (IRS), distribution of long lasting Insecticide treated nets (LLIN), Larval control, Intermittent presumptive treatment of malaria in pregnancy, Diagnostic testing and Malaria case management [ 13 ]. These activities were implemented through a phased roll-out in selected districts in the Southern Province of Zambia starting with Choma and Livingstone districts, with the goal of improving surveillance and interrupting transmission [ 14 ]. This is because Southern province had made a substantial decline in malaria and is in the low transmission zone (Zone II) [ 15 ]. Livingstone district of Zambia however, has had a steady rise in the cases of malaria from 3.5/100,000 population in 2009 to 10.6/100,000 population in 2014 (LDHIS, 2014). Factors contributing to these observations are not well known. It is not known whether this situation is due to the weakening of malaria control programs at the community level or could be partly due to a decrease in community acceptance or participation in malaria programs [ 5 ].

In 2013, the Zambia government adopted a Community health worker strategy for early diagnosis and treatment of uncomplicated malaria to achieve malaria elimination in the low transmission zones (zone II) [ 16 ]. According to the Zambia Malaria Operational Plan FY 2017, the CHWs in the ICCM approach are provided with diagnostic tools and medicines for the management of common childhood illnesses including the treatment of uncomplicated malaria. Their roles in the management of uncomplicated malaria include; Carrying out diagnoses according to their training and recognizing danger signs, Using RDTs in all cases of fever to confirm malaria before treatment, administering the first-line medicine, referring to the next level and administer pre-referral treatment when danger signs are recognized, instituting measures to reduce body temperature, following up with patients, particularly children under 5 years of age, providing education to the community on the need for compliance to treatment, recognition of danger signs, and prevention of malaria in terms of environmental management, ITN distribution and inspection of ITN use and dissemination of preventive messages such as IRS for instance, advising when to return if the condition persists, and reporting to the facility in their catchment area.

Understanding fidelity and factors shaping CHWs performance in relation to adhering to the CHW components or roles is important for a positive implementation outcome [ 5 ]. Periodic performance assessments through surveys and prompt feedback of results on the implementation of interventions to stakeholders in the locality may help to improve malaria control in malaria-endemic countries [ 17 , 18 ]. Livingstone district is implementing the CHW strategy using voluntary CHWs in Libuyu and Nakatindi communities. The aim of the study was to evaluate the implementation fidelity of CHW roles in malaria prevention and control programs in Livingstone, Zambia, despite the attribution to decline of malaria incidence of morbidity and mortality to CHWs.

The evaluation of the CHW fidelity to roles was based on a framework by Carroll and co-workers. Components of the framework are the intervention, in this case the CHW worker intervention and potential moderators such as facilitation strategies to implementation of the CHW malaria strategy, quality of service delivery and participant responsiveness. Adherence to roles as a measure of fidelity was evaluated through CHW role performance coverage [ 16 ] which are Health Education of malaria prevention and control, Community testing with RDT, treatment with ACT for uncomplicated malaria and reporting [ 19 ], and also and quality of execution of malaria activities in terms of whether they carry out their roles as expected according to training.

Study design

The study employed a concurrent mixed methods design. The quantitative design used a cross-sectional approach while the qualitative component used a case study. The qualitative study informed quantitative findings hence in the results and discussion, the two methods are mixed in presentation.

Research setting

This study was conducted in Livingstone districts of Zambia, a tourist capital involved mostly in hospitality industry. About a third of the Livingstone population lives in poverty. Despite others being in formal employment, majority of people are involved in informal employment such as small – scale businesses, cross border trading and fish mongering. The district was chosen based on the fact that despite having moved towards the elimination of malaria, the district currently experiencing resurgent malaria threats (LDHIS 2015) hence the need of assessing extent of fidelity to CHW roles. The study was conducted in the catchment areas for two health facilities - Libuyu urban health facility with a catchment population of 18,057 and Nakatindi health facility with a population of 4925 according to central statistics Office in 2015. Libuyu records the highest malaria cases in the district and both facilities are the only clinic catchments (Nakatindi and Libuyu) where there were active CHWs for malaria prevention and control programs.

Study population

The study participants for the quantitative component consisted of all CHWs in the two clinics participating in malaria community health activities. Community members were also included as end users through a household survey and because they are key to the measurement of CHW performance in validating or qualifying the performance measure of CHWs through the services rendered by CHWs in the community.

The qualitative study participants included key informants - the program implementers such as the district malaria program officer and facility malaria community focal persons who were interviewed on strategies that affected implementation of the CHW strategy. In addition 10 CHWs from each of the two study sites participated in two FGDs to discuss perspectives on responsiveness.

Sample size and sampling

For the community survey, the catchment areas were purposively sampled as they are areas were the community malaria agents are found. All the zones in each catchment area were selected (complete enumeration) as they have representation of a community malaria agent in each zone.

Community survey sample size was calculated using the single proportion formula n \( =\frac{{\boldsymbol{z}}^{\mathbf{2}}\boldsymbol{p}\left(\mathbf{1}-\boldsymbol{p}\right)}{{\boldsymbol{e}}^{\mathbf{2}}} \) = \( \frac{{\mathbf{1.96}}^{\mathbf{2}}\times \mathbf{0.5}\times \mathbf{0.5}}{{\mathbf{0.05}}^{\mathbf{2}}} \) = 384 + 20% (77) non-response rate: n  = 463.

The proportion 0.5 was used since the prevalence of a CHW strategy service was unknown. The study sample size of 463 was apportioned to the two catchment areas using probability proportion to size as follows: (4100/11,387)*463 which gave Libuyu 167 Households and (7287/11,387)*463 which gave Nakatindi 296 Households. An equal number of households were selected from each zone. The household head from the selected households, above 18 years of age, male or female were interviewed.

The key informants (KI) for the qualitative study (District Malaria focal point person and the two malaria focal point persons at the two clinics) were purposively sampled. For the two FGDs, 10 CHWs from each study site were purposively sampled from the 40 CHWs. The selection of FGD participants was guided by the CHW facility supervisors.

Data collection

A checklist was used for collection of quantitative data. Questions for the checklist were based on the conceptual framework for CHW performance [ 20 ]. Semi-structured interviews were used in the household survey.

Questionnaires administered to the individual CHWs were adopted from studies that used validated questionnaires using Cronbach’s reliability alpha [ 20 , 21 ]. The independent variables included age, sex, level of education, marital status, monthly income or incentives, number of household members (those with more than five members tend to have more support at home), number of children under 5 (those with no child under 5 tend to be more active), knowledge in the CHWs understanding of malaria, reason for being CHW, duration of being CHW, financing, health systems factors like training, supply of commodities, and technical support supervision. For the survey, data was collected from sampled households using a semi-structured interview guide to assess coverage of CHW strategy in the community. Household heads above 18 years were interviewed. The survey questionnaire was administered by the trained research assistants. Pretesting of the tool was done manually with a questionnaire in both study sites which allowed for modification of the data collection tool.

For qualitative study, interview guides that were developed based on the implementation fidelity framework by Carroll and co-workers [ 19 ] were administered to the CHWs and the KIs (Additional file 1 ).

Performance assessment

This is the accomplishment of CHW tasks in Malaria prevention and control against given roles and targets in a given period considering how many CHWs were adherent to the roles. The dependent variable was the level of each CHW’s performance which was developed using five indicators: monthly reporting rate within the previous 6 months, malaria knowledge, health education or sensitizations done, percentage of people tested from the expected target of 40 within 140 m radius of the index case [ 14 ], percentage treatment of positive cases by the CHW in their zones. Each indicator was categorized into quintiles (0–4) to standardize the scores making a total of 20 points as highest integrated score. Knowledge of malaria epidemiology and vector ecology, was measured by the respondents’ correct answers on items related to service quality and actions on malaria. There was no prior assessment done to test reliability of the tool but the indicators picked to assess performance was based on a previous study that was published [ 20 ] but all the indicators had equal weighting. Reference was made to the reports submitted to the Health center by the CHWs (Table  1 ). Malaria transmission is throughout the year but the denominator for calculating percentage coverage for performance sorely depended on what the records indicated during the previous 6 months period under review, of which part of peak periods October to December (hot-rainy season was assessed [ 20 ].

Quality assessment

Assessing quality of CHW services was done by quantifying CHWs who always carried out active detection, diagnosis and treatment, prescription of anti-malarial, follow-up of patients, and dissemination of preventive measures; as a fidelity measure to community malaria activities with scores given that was used in the previous study [ 21 ]. Each item was detailed in how each activity is done in terms of regularity or frequency categorized as “always” = 2, “sometimes” = 1 or “never” = 0. The score for each of the five items was calculated as [total points divided by maximum points] so that each item is given a maximum of one point. A CHW was said to be offering good quality service if they scored at least 4 and above representing at least 80% of the CHW activities always done by CHWs (Table  2 ). For the variable of follow up of index cases, the assessment was done with an index case present in the register. All the CHW assessed had malaria positive cases that needed follow up.

Data analysis and management

Quantitative data entry was done using Epi info, and analysis using STATA version 13. In order to identify factors associated with performance for responsiveness and quality of CHWs’ service delivery, Chi square fishers exact test was done to determine whether there was significance of association. The quality of data was assessed by comparing data in the community registers at the health facility level and numbers provided in the reports at the district health office. Data was also checked for completeness and consistency. Further, only data that was complete was included for analysis.

For performance measurement, a chi square fishers exact test was run as performance levels was at two levels (poor< 12 and good 12+). For quality assessment, a chi square fishers exact was used to measure association between the predictor and dependent variables (< 4-less than 80% substandard and ≥ 4–80% + good). Each indicator of the five CHW roles were analyzed based on CHW who always carried out the expected activities for that particular indicator and categorized as substandard if scored less than 0.8 for each indicator and at least good quality if scored more than 0.8 for an indicator. Good coverage is at least 80% achievement according to WHO guidelines.

Qualitative data was managed using n-vivo 10 software after verbatim transcription of all the recordings. Data was analyzed using thematic analysis. This approach allows for categorizing of data into themes so as to identify patterns and trends. Preliminary reading of transcripts allowed for development of a code-list that was imported into the software for coding. Code reports from the coding activity allowed for analysis and interpretation of results.

Ethical approval to conduct the study was obtained from the University of Zambia Biomedical Research Ethics Committee IRB00001131 Of IORG0000777. Authority to conduct the study was also obtained from the Ministry of Health-Livingstone District Medical office. During the data collection process, written informed consent was obtained from the study participants.

CHW demographic characteristics

A total of 34 CHWs participated in the study. Nakatindi had 19 (56%) CHWs while Libuyu had 15 (44%). Most CHWs were females (71%) while 29% were males. The age range was between 22 and 62 years with a mean age of 23 years. CHWs age was categorized based on quartiles and the bench march for two lower quartiles was below 40 years and two upper quartiles above 40 years of which 18% were under 40 and 62% were above 40 years old. About 71% had a secondary school education while 3% had no formal education. Only half (50%) of the participants were married. About 47% of them only did some piece work for survival and earned less than K500 (US 53) only on monthly basis while 53% were involved in diverse businesses. A total of 12 (35%) had more than six household members, and 17 (50%) had under five children in their households. Also 17 (50%) had more than 1 year of working as a community malaria agent. Table  3 provides a summary of the background characteristics of the participants.

CHW roles and performance

Assessment of CHW roles for the previous 6 months according to records showed that 73% of CHW had good performance in report submission as they had submitted at least four (4) reports from the expected six (6) reports. Only 44% of the CHW managed to give more than 13 sessions of health education in malaria in the community from the expected 19 health education schedules. For index testing, according to the 2013 Malaria training protocols used at the time, Community health workers for malaria were expected to test at least 40 people in households that are within the radius of 140 m [ 14 ] and only 29% of CHWs were able to test at least 80% (32) of the expected target of 40. For treatment, only 15% of CHW were able to treat positive malaria cases. However, 97% of CHWs were knowledgeable about malaria by scoring at least 9 out of 16 questions and yet did not perform as expected. The knowledge performance range was from 8 to 13 and the mean knowledge score was 10 with a standard error of 0.22 and a confidence interval of 10.2–11.1. Overall, only 15.7% (5 out of 34) CHWs had good performance in all the CHW malaria roles. The performance scores ranged from 3 to 16 out of the maximum score of 20. The mean performance score was 11 with a standard error of 0.67 and a confidence interval of 9.8–12.5 (Table  4 ).

Implementation fidelity to the CHW malaria roles

Fidelity to the CHW roles, measured in terms of quality adherence to program components such as active detection, patient follow up, diagnosis and treatment, dissemination of preventive measures, was generally low as only 7(21%) CHW carried out the CHW program as it was intended while 27 (79%) had substandard quality as they carried out less than 80% of the community malaria activities. Details on the fidelity levels are provided in Table 3 .

For the active detection of cases, 23 (68%) reported going in the community and find malaria cases and were able to find out if the patient had recovered. For follow up role, 16 (47%) always followed up patients in the community to monitor the progress of treatment. And only 11 (32%) always disseminated preventive messages. Concerning the use of RDT for diagnosis and treatment, only 9 (26%) of the CHWs always used RDT for diagnosis and treatment as others (74%) reported stock outs for RDT. This therefore meant that those that had stock out of RDT could not give malaria treatment but referred to the facility for assessment as treatment is based on a positive slide. Prescription information was only given sufficiently by only 3(8.8%) of CHWs risking non-adherence to drug dosage guidelines (Table  5 ).

Community coverage of CHW malaria services

According to guidelines, all malaria index cases are supposed to be followed up by CHWs for screening, including other household members and other households within 140 m radius. CHWs were also to give health education and treating positive cases as a CHW role in community malaria surveillance. Results indicate that out of the 140 households that reported having had positive malaria, only 39 (27%) were tested by CHW and others 101 (72%) from the health facility which indicates high utilization of health facility services by the community. Out of the 140 positive cases, only 33 (24%) reported having been given health education, 20 (14%) were treated by CHW and the others 120 (86%) were treated at the facility. There was low community coverage in terms of health education, testing and treatment in relation to community follow up of index case response by CHWs. From households that did not report malaria, only 49 (15%) of them were screened and 87 (27%) received health education messages from CHWs (Table  6 ).

Qualitative findings indicated that the number of households to be covered by CHWs (11,367) was high for the small number of CHWs available (36) which made it unrealistic to meet targets. CHWs were overwhelmed with work as they are a few. Sometimes the same CHWs were taken to other zones that did not have the CHW strategy for malaria activities. Multitasking also increased the workload as they were the same people participating in many other projects, especially those that were offering incentives.

“ We need more CHWs to be trained as the area is too big for one CHW to cover a Zone” P1 , FGD2

Factors affecting fidelity to CHWs roles

Proportions of CHWs who adhered to the CHW roles measured by performance outcomes was low and the service delivery was substandard despite having high knowledge on Malaria. The factors that were associated with CHWs’ fidelity to roles were marital status, work experience, supervision and reporting records. From the qualitative findings, issues of coverage, remuneration, supplies, ownership and capacity building shaped adherence to the CHW roles.

A chi square fishers exact tests showed that being married was significantly associated to performance of CHW roles with a p value of 0.015 of which all good performers 5 (29.4%) were among the married while no one from the unmarried had good performance (Table  7 ). We also found no correlation between age and work experience, marital status and age or marital status and commodity stock. However, CHWs who were married were more likely to have three or more years of experience than those who were single P  = 0.016 (Table  8 ). Supervision by professionals at facility (80%) or NGO was significantly associated with performance ( P  = 0.002) as no one supervised by the NHC chairperson had good performance. Supervision was at facility level but was carried out quarterly, by the Anglican Church supporting the malaria program. Supervision is done by checking books, observing tests and having meetings where discussions of challenges are done. Sometimes Community Malaria Agents (CMAs) were supervised when the health facility workers went for the outreach activities or the NHC chairperson upon submission of reports. However the environmental health officer in charge of the CMAs did not manage to constantly supervise the CHWs due to transport logistics and that other professional staff do not supervise them leaving the CMAs demotivated for work.

Experience and self confidence

Among those with 6–12 months’ work experience no one performed well compared to 5 (29%) that had at least 12 months’ work experience with p  = 0.04 (Table 7 ). Most CHWs 29(85%) seemed not to have confidence in the work they did especially in treating. Treatment protocols were at the clinic and this maybe one of the contributing factors to lacking confidence to administer treatment as they had nothing to refer to in terms of treatment according to age categories. Only those that have more than 12 months experience could administer coartem.

“They administer drugs and they have drugs though there is a challenge. Some are experienced and others are not experienced. They have fear to administer drugs hence they refer to the clinic even if client is positive”. KI 2

Inconsistant supplies

Insufficient stock for use was found to be a challenge. This included RDTs, anti-malarial drugs (ACT) and storage equipment like bags for carriage or keeping RDTs and drugs which risks compromising potency if put on direct sunlight. Sometimes when clients with fever were attended to but with a negative result, there was no antipyretic or pain killer but were just referred to the health facility. For ACT, only CHWs from peri-urban areas were given but those near the health facility just collected for clients in order to treat, with instructions on how it should be taken. Lack of thermometers, scales and referral forms was also reported. Insecticide Treated Nets were given but demand was higher than supply. Pregnant women and under five children were prioritized to be given an ITN as each bed space may not have an ITN.

The reporting record was found to be a factor associated with performance where 92% of non-performers and 8% for good performers used books and not standard registers with p value =0.05. However, a variety of reporting tools were reported to have been used like phone, paper and book which have no standardized way of recording (Table 7 ).

“We who work are free but we have challenges with reagents in some cases. We need RDT consistency for us to continue working, as people are now saying the CHW has stopped testing. People will neglect us, so give us RDT so that we continue working. A long time ago people came because we had access to RDTs but now people come but can’t be tested. Now people come and bounce, we feel bad as we can’t do anything for them as we don’t want to help meanwhile it is just things we lack to use.p4 fgd2

Financial resources

Key informants indicated that it was very difficult to achieve the malaria targets without CHW incentives. This affects their performance since the community malaria agents would rather work where there is an offer of an incentive as they even feel the malaria program is lagging behind in terms of incentives compared to other health programs. However, Community health workers indicated that what motivated them to continue working even without incentives was the fact that they were told that they were volunteers from the beginning and so they were self-driven to serve the community and wanted to be part of malaria reduction. Difficulties in mobility affected their performance as they had to travel distant places hence difficulty to cover vast areas.

“ From the beginning, we were told we were volunteers and we understood. We do sacrifice so at some point they should remember us that we have families. We don’t work and at the end of the day we need to see to it that the family has food on the table. The world we live in now is different from the way it was before. Times are different from times back. Let them consider us even just a bit for us to be able to continue”. P1 FDG

Ownership by the district

The CHWs felt discouraged to work as they felt that the district did not pay much attention to them. They however received support from the Anglican Church which offered support in terms of training the CHWs and provision of mosquito nets for distribution, ITN use inspection and health education. The district only come in if asked especially if there were leakages they help talk to the city council who also supports by ensuring water leakages are controlled as children play in the stagnated water.

“Let the district also plan for us besides the Anglican. They should claim ownership and have their own program. They should also provide refresher courses. After all the Anglican found us at a clinic for the district. Like the days of old they should give us refresher course, they should not relax, not waiting for Anglican who get from other donors and us the church.” P3 FGD1.

Recognition and community connectedness

The CHWs are known in the community and therefore needed some formal recognition with certificates of training in the work they did including the similar identification with identity cards and aprons. This enthusiasm for recognition made them want to work at the center instead of working from the community and have to constantly remind them to go work in the community as the health center was just for them to bring reports. This may be the reason for poor community coverage with regards to their community malaria roles.

“We need certificates of training and also identity cards, and refresher courses to have latest updates so that we are not left too much behind. We just have apron and T- Shirts of which others still do not have, especially in rainy season, we need rain gear you can be protected and the books do not get wet.” P2 P8 FGD2.

This study evaluated fidelity to the CHW roles for malaria considering the moderating processes as it was intended in terms of content and coverage. Overall, Fidelity was found to be low considering that a lot of factors were affecting the implementation of roles and these were individual CHW factors and a great deal of health system factors. In the qualitative findings, community factors were found to affect fidelity of CHW to their roles in community malaria programs, a finding similar with a systematic review study by Rowe on the effectiveness of strategies to improve health-care provider practices in low-income and middle-income countries which reviewed that training alone for CHWs was not effective but with added community support [ 22 ].

Facilitation strategies shaping fidelity to CHW roles

Individual chw factors.

Only 15% of CHW were able to work according to the way the program was intended with diagnosis, treatment, follow up, health education and reporting despite them being knowledgeable unlike in a Cambodian study were VHW were not knowledgeable as a reason for poor quality service [ 21 ].

Being married was a significant factor of performance in this study as all good performers were married. This is similar to study were married CHWs gave a higher performance than others [ 20 ]. This could be because they have more family members to help with household duties or that they may have an extra source of income from the other partner. Having fewer household duties encourages CHWs to work more actively and reduces the dropout rate as one of the barriers preventing a good CHW performance was a heavy amount of household duties [ 23 ] though this was not significant in this study. Recruitment of married CHW for malaria interventions with priority placing may help in sustaining good performance in improving coverages. This however is a hypothesis that should be tested to see if it is valid in this setting.

Work experience was another significant factor that related to CHW performance. This was indicated by the fact that CHWs who were good performers all had worked for more than 1 year. Longer work experience was also a similar finding in this study as confidence is belt more with the longer the period one works [ 20 , 24 , 25 , 26 ]. The program identified experienced CHW to be receiving zonal reports from the CHWs that had little experience. Longer work experience entails having more opportunity to receive effective training, supervision and any incentives and to build a confidential relationship with community members [ 25 , 27 ]. Familiarity with the work motivates CHWs to apply for position like that of a Community Health Assistant (CHA), another strategy being rolled out in Zambia [ 28 ].

Most CHW were able to conduct diagnosis with RDTs skills that they had acquired through the training programs [ 29 , 30 ] but prescription messages were poor during follow ups which risks resistance due to non-adherence to treatment. This could also have been due to the issues of confidence to treat, a finding different from a study in Ghana where CHW adequately treated even children with malaria [ 31 ]. Occasional onsite quality supervision to actually see what they do is vital in order to ensure to quality service especially that some of them are supervised by the fellow malaria CHW and that the facility supervisor is facing challenges in mobility to enable quality supervision.

CHWs drop-out rate in this setting was reported to be at 5% (2/36) reasons being lack of monitory incentives. Demotivation due to unfunded program for CHWs to make ends meet [ 32 , 33 ] affected fidelity hence CHWs just stayed in the community or sought other programs that have incentives. Improving non-monetary incentives such as providing them with materials that identify them as community-based health workers e.g. badges, t-shirts, and so on; frequent refresher courses and the exchange visits. To avoid demotivating CHW and health workers alike, sufficient remuneration, supplies of RDT, drugs and ITNs and job aids need to be consistent, including relevant infrastructure and supportive supervision may improve adherence to CHW roles [ 28 , 32 , 33 ]. This may also improve their social prestigious need for recognition in the community and improve community connectedness as indicated in a study by Strachan DL that CHWs value feedback and feeling connected to the health system and their community, are motivated by status and community standing, and want to be provided with the necessary tools to perform [ 34 ].

Health system factors

CHWs are aware that they are volunteers always willing to work with 97% of them being knowledgeable but this knowledge did not yield good performance or good quality service because the organization system has not put in place adequate necessities for this strategy to work as intended. The main qualitative factors that surrounded performance and fidelity of CHWs to the malaria program included insufficient remuneration, reduced mobility, inadequate quality supervision, inadequate funding, and work overload for CHWs, inconsistent supplies of stocks, poor coordination with partners and lack of initiative for capacity building. CHW performance is hard to achieve and to maintain without sufficient consideration for funding and other motivating factors like transport and remuneration [ 20 , 25 , 35 , 36 ].

Supervision was a significant factor associated with CHW performance specifically the supervising organization. The CHWs were being supervised by the existing NGO, the clinic and chairman of the program. Those who had good performance were supervised by the clinic though the facility supervisors faced challenges with mobility for following up CHWs in the field. Supervision not only needs frequency but quality input as high quality supervision is one of the key factors in improving a CHW’s performance [ 2 , 20 ] as evidence from a systematic review on impact and implementation of supervision suggests that improving supervision quality has a greater impact than increasing frequency of supervision alone with supportive supervision packages, community monitoring and quality improvement/problem-solving approaches, though evaluation of all strategies is weak [ 37 ]. For instance in Kalabo study, frequent supervision did not have a positive impact on CHW performance as quality was reported to be poor and almost half of the community health workers do not experience any benefit from the supervision [ 26 ]. Supervisors should therefore have adequate health knowledge and conduct routine supervisions to sustain a high performance and responsiveness from the CHWs and they should have standardized method or checklist for the supervision of community health workers. This is because supervisors do not receive specialized training as mentors, but assume that role based on their academic training and orientation to program performance assessment through workshops.

Reporting system

Reporting record used was found to be a significant determinant of performance. There is poor CHW program coordination and collaboration with regards to the supporting organization and ownership by the supervising district. A standard register and a reporting tool that is common to both supervising organizations is necessary for a common goal. The stake holders have a direct influence on the health system factors and are to produce guidelines, registers, reporting tools, checklists and evaluation tools [ 33 ] though an innovation to improve the information system through use of phones for reporting has been effected [ 38 ] without data being captured by the local district. CHW had no standard reporting tools and were expected to submit a report to their zonal chosen CHW supervisor who takes to the chairman of the Malaria CHW program selected among the CHWs. Other CHW activities were not recorded and CHWs concentrated only on diagnosis and treatment, neglecting other CHW roles for malaria prevention and control. This is similar to a study by Yasuoka et al. where Village Health workers (VHW) concentrated only on diagnosis and treatment [ 21 ]. Apart from the health education, testing, and treatment from the survey, a lot more services are rendered by CHWs but not reported and these included ITN distribution, ITN utilization inspection, households sprayed after CHW sensitization, referral, burying of trenches. The qualitative results however indicated doubt in the authenticity of CHW reports as they are rarely supervised and their register records are taken as gospel truth. There is therefore need for a standard reporting tool that will cover all the CHW roles capturing all details required as other details were missing in the improvised registers. This may also aid in supervision by the supervising officers visits [ 26 ].

Service delivery

Fidelity to the malaria CHW roles was poor due to lack of supplies a cause of inactivity of some CHWs in terms of active detection and treatment [ 26 ] similar with the implementation challenge of the test and treat intervention in another part of Zambia where it was concluded that with limited resources, coverage and diagnostic tools, reactive screen-and-treat would not likely be sufficient to achieve malaria elimination but with reactive focal drug administration as an alternative strategy [ 14 ].

Human resource

It is almost impossible for 34 CHWs to cover 11,387 households without inefficiency as CHWs tend perform poorly due to large population coverage and multiple tasks CHWs and get overwhelmed with so many programs [ 39 ] . Scaling up in terms of capacity building for more man power with priority placing may foster good performance in the malaria CHW program. Scaling up of these malaria CHW interventions, promoting continued use of CHWs in national programs as an important human resource that contributes to long term impact of interventions [ 6 , 40 ]. The government endorsed a CHA assistant program to help meet the human resource demands who are to be liaison between community and the health facilities but discussions with CHAs showed that because of the limited number of trained staff at health posts, it was resolved that CHAs should spend more time at the health posts than in the community [ 28 ] hence local malaria CHWs who are always present in their communities will continue being an effective strategy to in the elimination of malaria for community surveillance improvement as one of the WHO pillars in malaria elimination.

The strength of the study is that it a mixed methods study which explores both quantitative and qualitative and this gives an in-depth explanation of barriers to implementation fidelity of the malaria CHW program. It also assesses the services received through the community survey and hence is a comprehensive performance and fidelity assessment of CHW intervention in malaria. However, the limitations were lack of a standard measurement tool for an integrated approach for CHW community malaria interventions hence performance measurement and quality assessment indicators were adopted from previous studies. To evaluate CHW’ service quality, only self-reported data were used, and the actual community experiences were not taken into account in terms of utilization of malaria CHW services. The validation of self-reported indices regarding service quality needs improvement. However, possible attempts were made: for instance, self-reported data were double-checked with CHW’ records in their monthly reports; data used to assess performance, submitted to the CHW malaria supervisor regularly. The analysis of association did not take into consideration the confounding variables as multivariate analysis was not done as the sample size for CHW was small. However, multivariate analysis will be done with survey data when analyzing utilization of CHW for malaria interventions.

The study findings indicate that fidelity to the malaria CHW roles was low in that adherence to the program as it was intended through performance outcome was poor and quality of service was substandard. Findings suggest that CHWs can still adequately contribute to the elimination of malaria in this setting with attention to health system support. A systems approach for malaria CHW facilitation considering supervision, stock supply and recruiting more CHWs on a more standardized level of recognition and remuneration would render an effective quality high implementation fidelity of the CHW malaria roles for this setting.

Availability of data and materials

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Abbreviations

Artemisinin Based Combination Therapy

Community Health Workers

Community Malaria Agents

Central Statistics Office

Focused Group Discussion

Indoor Residual Spraying

Intermittent Presumptive Treatment

Long Lasting Insecticide Treated Net

Primary Health Care

Rapid Diagnostic Test

Livingstone District Health Information System

House Holds

Health Education

WHO. Regional profiles 2015. Geneva: World Health Organization; 2015.

Google Scholar  

Chipukuma HM, et al. Towards a framework for analyzing determinants of performance of community health workers in malaria prevention and control: a systematic review. Hum Resour Health. 2018;16(1):22.

PubMed   PubMed Central   Google Scholar  

WHO. World Malaria Report 2015. Geneva: World Health Organization; 2015.

Mushi AK, et al. Acceptability of malaria rapid diagnostic tests administered by village health workers in Pangani District, north eastern Tanzania. Malar J. 2016;15(1):439.

Cohen JM, et al. Malaria resurgence: a systematic review and assessment of its causes. Malar J. 2012;11:122.

Christopher JB, et al. Thirty years after Alma-Ata: a systematic review of the impact of community health workers delivering curative interventions against malaria, pneumonia and diarrhoea on child mortality and morbidity in sub-Saharan Africa. Hum Resour Health. 2011;9:27.

Pallas SW, et al. Community health workers in low-and middle-income countries: what do we know about scaling up and sustainability? Am J Public Health. 2013;103(7):e74–82.

Lehman U, Sanders D. Community health workers : what do we know about them? 2007.

Zulu JM, et al. Developing the national community health assistant strategy in Zambia: a policy analysis. Health Res Policy Syst. 2013;11(24):10.1186.

WHO, Community health workers: What do we know about them? 2007.

Linn AM, et al. Reduction in symptomatic malaria prevalence through proactive community treatment in rural Senegal. Trop Med Int Health. 2015;20(11):1438–46.

PubMed   Google Scholar  

Kelly JM, et al. Community health worker performance in the management of multiple childhood illnesses: Siaya District, Kenya, 1997-2001. Am J Public Health. 2001;91(10):1617–24.

CAS   PubMed   PubMed Central   Google Scholar  

MOH. Focus on Zambia-Roll back malaria. 2011; Progress and impact series -country Report: Available from: www.rollbackmalaria.org/microsites/progress/impactseries . Cited 2015 November 24.

Searle KM, et al. Evaluation of the operational challenges in implementing reactive screen-and-treat and implications of reactive case detection strategies for malaria elimination in a region of low transmission in southern Zambia. Malar J. 2016;15(1):412.

MOH. Zambia National Malaria Indictor Survey. 2010; Available from: http://nmcc.org.zm/files/2010_Zambia_Malaria_Indicator_Survey.pdf . Cited 2015 November 24.

Kouznetsov, R., P.F. Beales, and W.H. Organization, Malaria: a manual for community health workers . 1996.

Larson DA. Malaria surveillance in low-transmission areas of Zambia using reactive case detection. BMC Malaria J. 2015;14:465.

Owusu-Agyei S, et al. Assessing malaria control in the Kassena-Nankana district of northern Ghana through repeated surveys using the RBM tools. Malar J. 2007;6:103.

Carroll C, et al. A conceptual framework for implementation fidelity. Implement Sci. 2007;2(1):40.

Kawakatsu Y, et al. Individual and contextual factors associated with community health workers’ performance in Nyanza Province, Kenya: a multilevel analysis. BMC Health Serv Res. 2015;15(1):1.

Yasuoka J, et al. Assessing the quality of service of village malaria workers to strengthen community-based malaria control in Cambodia. Malar J. 2010;9:109.

Rowe AK, et al. Effectiveness of strategies to improve health-care provider practices in low-income and middle-income countries: a systematic review. Lancet Glob Health. 2018;6(11):e1163–75.

Kok MC, et al. Which intervention design factors influence performance of community health workers in low-and middle-income countries? A systematic review. Health Policy Plan. 2015;30(9):1207–27.

Bagonza J, Kibira S, Rutebemberwa E. Performance of community health workers managing malaria, pneumonia and diarrhoea under the community case management programme in Central Uganda: a cross sectional study. Malar J. 2014;13(1):367.

Kalyango JN, et al. Performance of community health workers under integrated community case management of childhood illnesses in eastern Uganda. Malar J. 2012;11:282.

Stekelenburg J, Kyanamina SS, Wolffers I. Poor performance of community health workers in Kalabo District, Zambia. Health Policy. 2003;65(2):109–18.

Alam K, Tasneem S, Oliveras E. Performance of female volunteer community health workers in Dhaka urban slums. Soc Sci Med. 2012;75(3):511–5.

Zulu JM, et al. Hope and despair: community health assistants’ experiences of working in a rural district in Zambia. Hum Resour Health. 2014;12(1):30.

Chanda P, et al. Community case management of malaria using ACT and RDT in two districts in Zambia: achieving high adherence to test results using community health workers. Malar J. 2011;10(1):1.

Counihan H, et al. Community health workers use malaria rapid diagnostic tests (RDTs) safely and accurately: results of a longitudinal study in Zambia. Am J Trop Med Hyg. 2012;87(1):57–63.

Chinbuah MA, et al. Assessment of the adherence of community health workers to dosing and referral guidelines for the management of fever in children under 5 years: a study in Dangme West District , Ghana. Int Health. 2013;5(2):148–56.

Thu NTH, Wilson A, McDonald F. Motivation or demotivation of health workers providing maternal health services in rural areas in Vietnam: findings from a mixed-methods study. Hum Resour Health. 2015;13(1):91.

Nsona H, et al. Scaling up integrated community case management of childhood illness: update from Malawi. Am J Trop Med Hyg. 2012;87(5 Suppl):54–60.

Strachan DL, et al. Using theory and formative research to design interventions to improve community health worker motivation, retention and performance in Mozambique and Uganda. Hum Resour Health. 2015;13(1):25.

Perez F, et al. The role of community health workers in improving child health programmes in Mali. BMC Int Health Hum Rights. 2009;9(1):28.

Druetz T, et al. Utilization of community health workers for malaria treatment: results from a three-year panel study in the districts of Kaya and Zorgho , Burkina Faso. Malaria J. 2015;14:71.

Hill Z, et al. Supervising community health workers in low-income countries - a review of impact and implementation issues. Glob Health Action. 2014;7(1):24085.

Hamainza B, et al. Comparison of a mobile phone-based malaria reporting system with source participant register data for capturing spatial and temporal trends in epidemiological indicators of malaria transmission collected by community health workers in rural Zambia. Malar J. 2014;13:489.

Wanduru P, et al. The performance of community health workers in the management of multiple childhood infectious diseases in lira, northern Uganda - a mixed methods cross-sectional study. Glob Health Action. 2016;9(1):33194.

Kamal-Yanni MM, Potet J, Saunders PM. Scaling-up malaria treatment: a review of the performance of different providers. Malar J. 2012;11:414.

Download references

Acknowledgements

We are grateful to UNICEF/UNDP/World Bank/ WHO Special program for Research and Training in Tropical Diseases (TDR). Also to my colleague Adam Silumbwe for the contribution made towards this study. I also appreciate my husband Kelvin Mwiinga and my son, Lushomo Mwiinga, for the support and for understanding my absence.

Helen Mwiinga Chipukuma /H.M.C is a recipient of a TDR scholarship under the Postgraduate Training Scheme in Implementation Research at the University of Zambia. We are grateful to the financial support for the training scheme as provided by the UNICEF/UNDP/World Bank/ WHO Special program for Research and Training in Tropical Diseases (TDR). The funding Organization provided financial support for data collection process.

Author information

Authors and affiliations.

Department of Health Policy and Management, School of Public Health, University of Zambia, P. O. Box 50110, Lusaka, Zambia

Helen Mwiinga Chipukuma

Department of Environmental Health, School of Public Health, University of Zambia, P.O. Box 50110, Lusaka, Zambia

Hikabasa Halwiindi

Department of Health promotion and Education, School of Public Health, University of Zambia, P.O. Box 50110, Lusaka, Zambia

Joseph Mumba Zulu

Department of Epidemiology and Biostatistics, School of Public Health, University of Zambia, P. O. Box 50110, Lusaka, Zambia

Steven Chifundo Azizi & Choolwe Jacobs

You can also search for this author in PubMed   Google Scholar

Contributions

HMC conceived the article, did the literature search, data collection analysis and reporting. The data analysis was done by SCA and HMC. J M Z, HH and CJ refined the title, aided in drafting manuscript and JMZ structured the article. JMZ and CJ reviewed and edited the manuscript for intellectual content. The opinions expressed are those of authors alone. All authors read and approved final manuscript.

Authors’ information

The main author; Helen Mwiinga Chipukuma ( [email protected] ) is a 2nd year Masters student at the University of Zambia, School of Public Health in the department of health policy and management. She is currently pursuing a career in Health Policy and Management with implementation science.

Corresponding author

Correspondence to Helen Mwiinga Chipukuma .

Ethics declarations

Ethics approval and consent to participate.

Ethical approval to conduct the study was obtained from the University Of Zambia Biomedical Research Ethics Committee IRB00001131 Of IORG0000777. Written informed consent was obtained from each participants before administering the questionnaire.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Additional file 1., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Chipukuma, H.M., Halwiindi, H., Zulu, J.M. et al. Evaluating fidelity of community health worker roles in malaria prevention and control programs in Livingstone District, Zambia-A bottleneck analysis. BMC Health Serv Res 20 , 612 (2020). https://doi.org/10.1186/s12913-020-05458-1

Download citation

Received : 08 February 2019

Accepted : 23 June 2020

Published : 02 July 2020

DOI : https://doi.org/10.1186/s12913-020-05458-1

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Community health worker
• Community malaria agent
• Performance
• Implementation

BMC Health Services Research

ISSN: 1472-6963

• General enquiries: [email protected]

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

Tracking malaria health disbursements by source in Zambia, 2009-2018: an economic modelling study

Affiliations.

• 1 School of International Pharmaceutical Business, China Pharmaceutical University, Nanjing, 211198, Jiangsu, China.
• 2 Centre de Recherche en Gestion Des Services de Sante, Faculté Des Sciences de L'administration (FSA), Université Laval (UL), Centre Hospitalière Universitaire (CHU) de Québec UL-IUCPQ-UL, Québec, QC, Canada.
• 3 Riphah Institute of Pharmaceutical Sciences, Riphah International University, Islamabad, Pakistan.
• 4 Department of Monitoring and Evaluation, Ministry of Health, P.O Box, 30205, Lusaka, Zambia.
• 5 School of International Pharmaceutical Business, China Pharmaceutical University, Nanjing, 211198, Jiangsu, China. [email protected].
• 6 Riphah Institute of Pharmaceutical Sciences, Riphah International University, Islamabad, Pakistan. [email protected].
• PMID: 35864530
• PMCID: PMC9306103
• DOI: 10.1186/s12962-022-00371-2

Background: Zambia has made profound strides in reducing both the incidence and prevalence of malaria followed by reducing malaria related deaths between 2009 and 2018. The number of partners providing malaria funding has significantly increased in the same period. The increasing number of partners and the subsequent reduction of the number of reported malaria cases in the Ministry of Health main data repository Health Management Information System (HMIS) stimulated this research. The study aimed at (1) identifying major sources of malaria funding in Zambia; (2) describe malaria funding per targeted interventions and (3) relating malaria funding with malaria disease burden.

Methods: Data was collected using extensive literature review of institutional strategic document between the year 2009 to 2018, assuming one-year time lag between investment and the health outcome across all interventions. The National's Health Management Information System (HMIS) provided information on annual malaria admission cases and outpatient clinic record. The statistical package for social sciences (SPSS) alongside Microsoft excel was used to analyze data in the year 2019.

Results: The investigation observed that about 30% of the funding came from PMI/USAID, 26% from the global funds, the government of Zambia contributed 17% and other partners sharing the remaining 27%. Multivariate regression analysis suggests a positive correlation between reducing reported malaria disease burden in HMIS 2009-2018 and concurrent increasing program/intervention funding towards ITNs, IRS, MDA, and Case Management with r 2 = 77% (r 2 > 0.77; 95% CI: 0.72-0.81). Furthermore, IRS showed a p-value 0.018 while ITNs, Case Management and MDA having 0.029, 0.030 and 0.040 respectively.

Conclusion: Our findings highlight annual funding towards specific malaria intervention reduced the number of malaria admission cases.

Keywords: Disease Burden; Malaria funding; Malaria interventions; Zambia.

© 2022. The Author(s).

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no competing interest.

Similar articles

• Factors associated with private health facilities reporting malaria in the national health management information system in Zambia: a cross sectional study. Gama A, Banda I, Kapaya F, Ingwe MM, Chiwaula J, Yeta A, Kawesha EC, Hamainza B, Chongwe G. Gama A, et al. Pan Afr Med J. 2020 Oct 29;37:203. doi: 10.11604/pamj.2020.37.203.18829. eCollection 2020. Pan Afr Med J. 2020. PMID: 33505572 Free PMC article.
• Country size bias in global health: cross-country comparison of malaria policy and foreign aid. Jezek T. Jezek T. Glob Health Res Policy. 2021 Feb 3;6(1):4. doi: 10.1186/s41256-020-00176-x. Glob Health Res Policy. 2021. PMID: 33531079 Free PMC article.
• The US President's Malaria Initiative, Plasmodium falciparum transmission and mortality: A modelling study. Winskill P, Slater HC, Griffin JT, Ghani AC, Walker PGT. Winskill P, et al. PLoS Med. 2017 Nov 21;14(11):e1002448. doi: 10.1371/journal.pmed.1002448. eCollection 2017 Nov. PLoS Med. 2017. PMID: 29161259 Free PMC article.
• Impact of the large-scale deployment of artemether/lumefantrine on the malaria disease burden in Africa: case studies of South Africa, Zambia and Ethiopia. Barnes KI, Chanda P, Ab Barnabas G. Barnes KI, et al. Malar J. 2009 Oct 12;8 Suppl 1(Suppl 1):S8. doi: 10.1186/1475-2875-8-S1-S8. Malar J. 2009. PMID: 19818175 Free PMC article. Review.
• Review of the malaria epidemiology and trends in Zambia. Masaninga F, Chanda E, Chanda-Kapata P, Hamainza B, Masendu HT, Kamuliwo M, Kapelwa W, Chimumbwa J, Govere J, Otten M, Fall IS, Babaniyi O. Masaninga F, et al. Asian Pac J Trop Biomed. 2013 Feb;3(2):89-94. doi: 10.1016/S2221-1691(13)60030-1. Asian Pac J Trop Biomed. 2013. PMID: 23593585 Free PMC article. Review.
• Romero M, Leiba E, Carrión-Nessi FS, et al. Malaria in pregnancy complications in Southern Venezuela. Malar J. 2021;20:186. doi: 10.1186/s12936-021-03728-9. - DOI - PMC - PubMed
• WHO (World Health Organization) World malaria report 2015. Geneva: WHO; 2015.
• https://reliefweb.int/report/zambia/zambia-national-malaria-indicator-su...
• Nawa M, Hangoma P, Morse AP, et al. Investigating the upsurge of malaria prevalence in Zambia between 2010 and 2015: a decomposition of determinants. Malar J. 2019;18:61. doi: 10.1186/s12936-019-2698-x. - DOI - PMC - PubMed
• https://www.who.int/teams/global-malaria-programme/reports/world-malaria...

Related information

Linkout - more resources, full text sources.

• BioMed Central
• Europe PubMed Central
• PubMed Central

• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

Loading metrics

Open Access

Peer-reviewed

Research Article

Data-driven nexus between malaria incidence and World Bank indicators in the Mekong River during 2000–2022

Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

* E-mail: [email protected] , [email protected]

Affiliation Department of Plant Biotechnology and Biotransformation, Faculty of Biology and Biotechnology, University of Science, Vietnam National University of Ho Chi Minh City (VNUHCM-US), Ho Chi Minh City, Vietnam

• Phuong Hoang Ngoc Nguyen

• Published: September 23, 2024
• https://doi.org/10.1371/journal.pgph.0003764
• Reader Comments

The increase in hydro dams in the Mekong River amidst the prevalence of multidrug-resistant malaria in Cambodia has raised concerns about global public health. Political conflicts during Covid-19 pandemic led cross-border movements of malaria cases from Myanmar and caused health care burden in Thailand. While previous publications used climatic indicators for predicting mosquito-borne diseases, this research used globally recognizable World Bank indicators to find the most impactful indicators related with malaria and shed light on the predictability of mosquito-borne diseases. The World Bank datasets of the World Development Indicators and Climate Change Knowledge Portal contain 1494 time series indicators. They were stepwise screened by Pearson and Distance correlation. The sets of five and four contain respectively 19 and 149 indicators highly correlated with malaria incidence which were found similarly among five and four GMS countries. Living areas, ages, career, income, technology accessibility, infrastructural facilities, unclean fuel use, tobacco smoking, and health care deficiency have affected malaria incidence. Tonle Sap Lake, the largest freshwater lake in Southeast Asia, could contribute to the larval habitat. Seven groups of indicator topics containing 92 indicators with not-null datapoints were analyzed by regression models, including Multiple Linear, Ridge, Lasso, and Elastic Net models to choose 7 crucial features for malaria prediction via Long Short Time Memory network. The indicator of people using at least basic sanitation services and people practicing open defecation were health factors had most impacts on regression models. Malaria incidence could be predicted by one indicator to reach the optimal mean absolute error which was lower than 10 malaria cases (per 1,000 population at risk) in the Long Short Time Memory model. However, public health crises caused by political problems should be analyzed by political indexes for more precise predictions.

Citation: Nguyen PHN (2024) Data-driven nexus between malaria incidence and World Bank indicators in the Mekong River during 2000–2022. PLOS Glob Public Health 4(9): e0003764. https://doi.org/10.1371/journal.pgph.0003764

Editor: Meghnath Dhimal, Nepal Health Research Council, NEPAL

Received: January 30, 2024; Accepted: August 29, 2024; Published: September 23, 2024

Copyright: © 2024 Phuong Hoang Ngoc Nguyen. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data were downloaded free of charge on Worldbank.org . All relevant data are within the manuscript and its Supporting Information files.

Funding: The author received no specific funding for this work.

Competing interests: The author has declared that no competing interests exist.

Introduction

According to a report from the WHO, vector-borne diseases killed more than 700,000 people yearly and accounted for 17% of the deaths caused by all infectious diseases in 2020 [ 1 ]. Among mosquito-borne diseases, malaria, a parasitic infection, caused estimated 249 million cases and 608,000 deaths globally in 2022 [ 2 ]. Southeast Asia is among the regions impacted most by malaria after Sub-Saharan Africa. Recently, reported cases of multidrug-resistant malaria in Cambodia could spread worldwide rapidly, increasing the risk to public health worldwide [ 3 ].

Malaria is caused by the blood parasites Plasmodium ( P . falciparum , P . knowlesi , P . malariae , P . ovale , and P . vivax ) through the infected female Anopheles mosquitoes, which can cause symptoms such as fever and shaking. Without treatment, patients can have severe health problems, such as seizures, brain damage, trouble breathing, organ failure, and death [ 4 ].

Affected by climate change, approximately one billion people may be exposed to mosquito-borne diseases for the first time by 2080 during extreme global warming [ 5 ]. Globally, the number of dengue cases has increased 30-fold in the past 50 years [ 6 ]. Some Southeast Asian countries, such as Vietnam, the Philippines, and Malaysia, have been seriously affected by severe dengue epidemics [ 7 ]. Malaria situations in GMS were divergent in the past two decades. Myanmar contributed 92.5% of total malaria cases while China was certified as malaria free in 2021 [ 2 ].

Furthermore, more than 200 large dams planned, completed, or under construction on the Mekong mainstream and its tributaries have raised concerns about increased mosquito habitats and water-related vector-borne diseases [ 8 , 9 ]. In addition, international travel and globalization could increase the spread of infectious diseases worldwide [ 10 , 11 ]. Several severe viruses, such as the Zika virus [ 12 ], MERS-CoV, and SARS-CoV-2, have spread far beyond their origin by air travel, causing the COVID-19 pandemic. In Lower Mekong (LM) countries, where the growth rate of tourism was the fastest among Asia-Pacific regions, nearly 60 million tourists visited these regions in 2017 [ 13 ]. Multidrug-resistant infectious diseases in that region could be dispersed globally through international tourism and transport [ 3 ].

Controlling the spread of infectious diseases requires effective coordination between the health and informational sectors of transnational governments in the Greater Mekong Subregion (GMS). However, several barriers in LM countries, such as information systems based on paper reports, a low level of computerization, and a shortage of health workers with data science training in the field of preventive health care, have been common challenges in addition to their ecological distribution conflicts [ 14 ]. These factors have slowed down data sharing for disease surveillance and effective policymaking.

Published studies in the region have shown correlations between mosquito-borne diseases with many sociodemographic and environmental factors. A multivariate analysis of dengue-like diseases in suburban communities in Laos and Thailand revealed that age, education, and occupation were associated with infection rates in suburban Laos and rural Thailand [ 15 ]. Many studies in Mekong countries indicate a strong relationship between dengue incidence with temperature and humidity [ 16 , 17 ]. A study in rubber forests confirmed that industrial rubber plantations provided shelter for mosquitoes and increased the incidence of mosquito-borne diseases [ 18 ]. In the Mekong Delta, a waterborne disease index was used to map dengue for a remote sensing study in Vietnam, which revealed clear seasonal variation in dengue fever according to changes in climatic factors [ 19 ]. In general, previous studies about factors affecting mosquito-borne diseases in LM were scattered and inadequate because inconsistent data and information have not been shared between countries. In addition, those studies in the region have not employed global open data portals or widely recognizable indicators, e.g., the World Bank indicators, the Air Quality Index, or the Human Development Index. These global indicators are being used in reports by many international organizations and have become references for information exchange worldwide.

Several climatic factors, such as temperature, rainfall, and humidity, have been widely used for time series forecasting via generalized linear models (GLMs), autoregressive integrated moving averages (ARIs), seasonal autoregressive integrated moving averages (SARIMAs), and Holt-Winters models [ 16 ]. The GLM assumes a linear relationship between targets and features, while the ARIMA model assumes a linear relationship between past values and current values. This finding is inconsistent with the nonlinear seasonal climatic features in the real world. SARIMA and Holt-Winters can deal with seasonal data better than the former. However, they assume that the data are stationery and deal with stable short-term data better than long-term data. ARIMA, SARIMA, and Holt-Winters are used for univariate data; hence, they lack explanatory power, which provides insights into the underlying factors affecting the models.

Long Short Time Memory (LSTM) is an advanced Recurrent neural network (RNN) that can be used for multivariate sequential data and addressing the vanishing or exploding gradient problem in traditional RNNs [ 20 ]. They have been applied for the classification, processing, and prediction of sequential data such as time series, handwriting, and voice data. In GMS, LSTM has been used for predicting malaria in China [ 21 ] or dengue in Vietnam [ 22 ] by climatic factors as the features.

Using too many features in prediction can cause overfitting and computational inefficiency. Ridge is a regularization technique used for preventing overfitting in linear regression models [ 23 , 24 ]. It uses L2 regularization, which is controlled by the tuning parameter alpha, to shrink the coefficients toward zero and prevent overfitting, hence resulting in a more robust and accurate model. Similarly, Lasso [ 25 ] is another regularization technique that uses L1 regularization to select features, but this approach could cause information loss due to the limited recognizability of collinearity between important indicators and their collinear counterparts. Elastic Net combines both Lasso and Ridge by learning from their shortcomings to improve the regularization.

This research aims to find critically impactful World Development Indicators (WDIs) correlated with malaria incidence in the GMS and used multiple linear regression combined with Ridge, Lasso, and Elastic Net to select features for time series forecasting by LSTM.

The methodology used in this work is described in Fig 1 , which includes the following steps: data preprocessing, correlational analysis, regression, and time series forecasting.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pgph.0003764.g001

The World Bank datasets included 1494 time series indicators from 2000–2022 from six countries in the Greater Mekong Subregion. Among the set of indicators strongly correlated with malaria incidence in most countries, seven were selected as the features for Multiple Linear Regression models combined with Ridge, Lasso, and Elastic Net techniques. One indicator critically affecting the linear models was chosen for time series forecasting by Long Short Time Memory Network. The vectors of map were downloaded from public domain https://www.naturalearthdata.com/ . The clipart icons were downloaded from public domain https://www.flaticon.com/ including:

https://www.flaticon.com/free-icon/data-collection_2103533

https://www.flaticon.com/free-icon/analysis_1239623

https://www.flaticon.com/free-icon/deep-learning_12031359

https://www.flaticon.com/free-icon/engineer_9321656

https://www.flaticon.com/free-icon/harvest_10132205

https://www.flaticon.com/free-icon/cityscape_2451728

https://www.flaticon.com/free-icon/field_4614470

https://www.flaticon.com/free-icon/malaria_6037989 .

The WDI dataset of the updated version (May 30, 2024) for GMS, which included Cambodia, the People’s Republic of China, the Lao People’s Democratic Republic (Lao PDR), Myanmar, Thailand, and Vietnam, was downloaded from the World Bank Group [ 26 ]. The dataset contained 1492 time series indicators, including the indicator of malaria incidence per 1000 population at risk (WDI code: SH.MLR.INCD.P3), from 2000 to 2022. In addition, another dataset of climatic indicators containing annual mean average temperature and precipitation in the GMS from 2000 to 2022 was downloaded from the Climate Change Knowledge Portal (CCKP) [ 27 ]. The WDI and CCKP datasets were combined into the final dataset containing 1494 indicators from 2000 to 2022 for six countries of the Mekong River. The datasets are provided in S1 Table .

Correlation

Distance correlation was calculated by the equation in Eq ( 2 ), in which D(x i , x j ) is the Euclidean distance between two sets of indicators, with x for malaria and y for the other indicators [ 29 ]. The Distance correlation ranges from 0 to 1, where 0 implies independence between two indicators and 1 implies a linear relationship.

The indicators with high correlation coefficients, > = 0.8 or < = -0.8 for Pearson and > = 0.8 for Distance correlation, which were found similarly between GMS countries, were put into the sets from 1 to 6.

The features in each topic or topic group for the regression models were selected from the indicators containing > = 20 datapoints that were strongly correlated with malaria and found similarly in most countries. The MLR models which were established in Eq ( 3 ) with malaria (y), the number of other independent variables (k), the k th feature (x k ), the regression coefficient (weight) of the k th feature (β k ), and the intercept (β 0 ). The datapoint values in each topic or topic group were normalized between 0 and 1. Ridge, Lasso, and Elastic Net algorithms were applied with an alpha range from 10 −5 to 10 2 for feature selection in time series forecasting.

Time series forecasting

The many-to-one architecture of LSTM consists of the following four layers. The 1 st LSTM layer takes mini batches in the sliding window from the time series input and returns the whole sequence. The number of neurons in the 1 st LSTM layer equals the number of features multiplied by the size of the sliding window. The 2 nd LSTM layer with the same number of neurons receives the sequence from the 1 st LSTM layer but only returns the same number of features. The next dense layer with the number of neurons is the same as the size of the features. The last dense layer outputs the predicted value. The LSTM setting parameters for the sequential model include normalization between 0 and 1, a training ratio from 0.6 to 0.8, 10 to 30 epochs, a batch size from 10 to 30, and a sliding window size (window size) from 1 to 9.

Performance metrics for evaluation

The performance metrics for model evaluation were calculated by Eqs ( 4 )–( 7 ). The absolute error (MAE), mean absolute percentage error (MAPE), mean square error (MSE), and coefficient of determination (R2) were used for evaluating the MLR and LSTM models.

The calculation and visualization were implemented in Python 3.9.16 with the following main packages: Cartopy 0.22.0, Geopandas 0.9.0, Geoplot 0.5.1, Keras 2.12.0, Matplotlib 3.7.1, Numpy 1.23.5, Pandas 1.5.3, Scikit-Learn 1.2.1, and TensorFlow 2.12.0.

Trends in malaria incidence

In the GMS, Myanmar, Cambodia, and Laos were among the top countries with the most cases of malaria during 2000–2022 ( Fig 2 ). The spread of a multidrug-resistant co-lineage of P . falciparum malaria, named KEL1/PLA1, across Cambodia could lead to a peak time of malaria during 2008–2013 [ 30 ]. Similar peaks repeated later in Myanmar, Laos, and Thailand. Myanmar contributed 92.4% indigenous malaria cases and 95.0% indigenous P . falciparum cases in 2021–2022 [ 2 ]. The increase of malaria in Myanmar from 78 000 cases in 2019 to 584 000 cases in 2022 amid political instability, which was shown at the end of line plot of Fig 2A , led cross-border movements of individuals to seek health care in Thailand. Nevertheless, the other GMS countries are aiming for certification of malaria elimination like China, which was successfully certified malaria free in 2021.

(A) The trends in malaria incidence in the Greater Mekong Subregion during 2000–2022. (B) The Greater Mekong Subregion in 2022. The vectors of map were downloaded from public domain https://www.naturalearthdata.com/ .

https://doi.org/10.1371/journal.pgph.0003764.g002

Among a total of 1494 combined indicators of WDI and CCKP, 994 and 394 unique indicators, respectively, were strongly correlated with malaria incidence in the GMS countries according to Pearson and Distance correlation ( Fig 3 ).

(A) The number of indicators strongly correlated with malaria incidence according to Pearson correlation analysis; the results were similar for one to six countries, and the values are shown in parentheses. (B) The number of indicators strongly correlated with malaria incidence according to the Distance correlation coefficient, which was similar for one to six countries, with the corresponding indicators in parentheses. (C) The set of 4, containing 149 indicators highly correlated with malaria incidence, was found similarly for four countries with 109 indicators were identified by Pearson correlation, and 88 indicators were identified by Distance correlation, with 48 intersecting indicators. (D) The set of 5, containing 19 indicators highly correlated with malaria incidence, was found similarly for five countries with 19 indicators were identified by Pearson correlation, and 0 indicators were identified by Distance correlation.

https://doi.org/10.1371/journal.pgph.0003764.g003

Table 1 summarizes the number of indicators in the set of 4 and 5 which were counted by topic groups, number of datapoints, urbanization, and gender. Fig 4 and Table 2 present the Pearson correlation coefficients in the set of 5. The coefficients in the set of 4 are provided in S2 Table .

The linkage for hierarchical clustering uses the complete method and Euclidean metric. Zero (0.00) values contain null values.

https://doi.org/10.1371/journal.pgph.0003764.g004

https://doi.org/10.1371/journal.pgph.0003764.t001

https://doi.org/10.1371/journal.pgph.0003764.t002

In the set of 5 containing 19 indicators which were found similarly for five GMS countries, most of them (10/19) belong the topic group of Environment ( Table 1 ). The other topic groups are Health (7/19) and Poverty (2/19). Among 149 indicators in the set of 4, half of them (71/149) belong to the topic group of Health. The second and third leading topic groups are Economic Policy & Debt (23/149) and Social Protection & Labor (17/149). The set of 4 contains 92 indicators with > = 20 datapoints during 2000–2022. However, most of indicators in the set of 5 contain less than 20 datapoints, from two to seven datapoints per country, except the only indicator of people practicing open defecation in urban (SH.STA.ODFC.UR.ZS). The number of datapoints in each indicator is provided in S3 Table .

Environment & demographics

In the topic group of Environment, the rural land area (AG.LND.TOTL.RU.K2, Fig 4 ) was perfectly positively correlated with malaria incidence, as opposed to the urban land area (AG.LND.TOTL.UR.K2, Fig 4 ) in all countries. Similarly, the rural land area where the elevation is less than 5 meters (AG.LND.EL5M.RU._, Fig 4 ) was perfectly positively correlated with malaria, in contrast to the urban land area where the elevation is less than 5 meters (AG.LND.EL5M.UR._, Fig 4 ) in all four countries except Cambodia.

Moreover, the population living in areas where the elevation is less than 5 meters (EN.POP.EL5M._, Fig 4 ) was completely positively correlated with malaria cases in Cambodia and Vietnam as opposed to Myanmar and Thailand. Meanwhile, the population density (EN.POP.DNST, S2 Table ), population in largest city (EN.URB.LCTY, S2 Table ), and the population in urban agglomerations of more than one million (EN.URB.MCTY.TL.ZS, S2 Table ) were strongly negatively correlated with malaria incidence in most countries.

On the other hand, the population ages under 4 (SP.POP.0004., S2 Table ), population ages 0–14 (SP.POP.0014., S2 Table ), and population ages 10–14 (SP.POP.1004., S2 Table ) were correlated strongly positively with malaria incidence, in contrast to the population ages 55–59 (SP.POP.5559.MA.5Y, S2 Table ) and population ages above 80 (SP.POP.80UP.MA.5Y, S2 Table ).

Income per capita (SI.SPR._, Fig 4 ), GDP per capita (NY.GDP.PCAP._, S2 Table ), wage and salaried workers (SL.EMP.WORK._, S2 Table ), and coverage of social protection & labor programs (per_allsp.cov_pop_tot, S2 Table ) were strongly negatively correlated with malaria incidence. In contrast, employment in agriculture (SL.AGR.EMPL._, S2 Table ) was strongly positively correlated with malaria incidence, which was like self-employed (SL.EMP.SELF._, S2 Table ), and vulnerable employment (SL.EMP.VULN._, S2 Table ).

The indicators of environment and demographics suggest living areas, ages, careers, income, social securities have affected malaria incidence.

Manufacturing (NV.IND.MANF.CN, S2 Table ), industry (NV.IND.TOTL._, S2 Table ), service (NV.SRV.TOTL._, S2 Table ), merchandise imports from low- and middle-income economies (TM.VAL.MRCH._, S2 Table ), researchers in R&D (SP.POP.SCIE.RD.P6, S2 Table ) and statistical performance indicators (IQ.SPI.PIL1, S2 Table ) were strongly negatively correlated with malaria incidence.

Similarly, individuals using the Internet (IT.NET.USER.ZS, S2 Table ), account ownership at a financial institution or with a mobile-money-service provider (FX.OWN.TOTL._, S2 Table ), accessibility to clean fuels and technologies for cooking (EG.CFT.ACCS._, S2 Table ), and average time to clear exports through customs (IC.CUS.DURS.EX, S2 Table ) were strongly negatively correlated with malaria cases, in contrast to power outages in firms in a typical month (IC.ELC.OUTG, S2 Table ).

The indicators of economics suggest economic improvement, technology accessibility, infrastructural facilities of information, transportation, and energy help to decrease malaria incidence.

In the topic groups of health, the cause of death by communicable diseases and maternal, prenatal and nutrition conditions (SH.DTH.COMM.ZS, Fig 4 ) was strongly positively with malaria incidence, which was similar to infant deaths (SH.DTH.IMRT, S2 Table ), neonatal deaths (SH.DYN.NMRT, S2 Table ), under-five deaths (.SH.DYN.MORT._., S2 Table ), and incidence of HIV (SH.HIV.INCD._, S2 Table ). Besides, prevalence of current tobacco use (SH.PRV.SMOK._, Fig 4 ) was also strongly positively correlated with malaria incidence, which was like people practicing open defecation (SH.STA.ODFC._, Fig 4 ).

In contrast, universal health coverage (UHC) index (SH.UHC.SRVS.CV.XD, Fig 4 ) was strongly negatively correlated with malaria incidence, which was similar to people using at least basic drinking water services (SH.H2O.BASW._, S2 Table ), people using at least basic sanitation services (SH.STA.BASS._, S2 Table ), people with basic handwashing facilities including soap and water (SH.STA.HYGN.ZS, S2 Table ).

The indicators of health suggest vulnerable dependent population, health safety and habits have affected malaria incidence.

In the suspected indicators which were previously reported about the correlation with malaria ( Table 3 ), forest area (AG.LND.FRST._) was associated with malaria incidence in some countries. Climatic indicators of precipitation (PR.) and temperature (TAS.) were uncorrelated with malaria incidence in all countries, which was like air transport (IS.AIR._) in both freight and passenger. Meanwhile, the renewable internal freshwater resources per capita (ER.H2O.INTR.PC) were moderately positively correlated with malaria incidence, especially strong in Cambodia and Vietnam.

https://doi.org/10.1371/journal.pgph.0003764.t003

Multiple linear regression model

Although most of the 19 indicators in the set of 5 show a perfect correlation with malaria incidence, they contain very few observations for building dependable models except the indicator of people practicing open defecation in urban (SH.STA.ODFC.UR.ZS). Therefore, 92 indicators containing > = 20 datapoints were considered as the features for regression models from the set of 4. There was high collinearity between these indicators via Pearson correlational analysis as shown in S4 Table . S5 Table provided MLR coefficients when using all indicators in each topic group for building MLR models to predict malaria incidence.

When using 50 indicators from the group topic of Health for building MLR models ( Fig 5 and S5 Table ), each country, each locality (urban and rural regions), and each gender reveals different patterns. The MLR coefficient of people practicing open defecation in urban (SH.STA.ODFC.UR.ZS) in Lao PDR is negative while they were positive in the other countries, especially high in China. However, the MLR coefficient of people practicing open defecation in rural is negative in China while they were positive in the others. In Myanmar, two out-standing indicators with negative coefficients including people using safely managed sanitation services in urban (SH.STA.SMSS.UR.ZS) and population ages 65 and above in female (SP.POP.65UP.FE.IN) while population ages 65–69 in male (SP.POP.6569.MA.5Y) had an out-standing positive coefficient. Nevertheless, the coefficients of population ages 80 and above (SP.POP.80UP._) are noticeably positively high in Vietnam.

https://doi.org/10.1371/journal.pgph.0003764.g005

Using 50 features such as the indicators from topic group of Health at the same time for MLR regression could cause overfitting models but using too few features such as one or two indicators from the topic group of Infrastructure or Private Sector & Trade could cause under fitting models. Therefore, 92 indicators were divided into 7 groups based on their topics or topic groups for feature selection by MLR, Ridge, Lasso, and Elastic Net regressions. The MLR, Ridge, Lasso, and Elastic Net coefficients for each group are provided in S6 Table . The more the alpha increases in the paths of the regularized coefficients, the coefficients of the more affecting indicators in the models tended more slowly to zero.

Table 4 presents the most impacting indicator per each group in each country. The indicator SH.STA.ODFC.RU.ZS was the most impacting feature of the group 1 of Health in most countries, except Myanmar. In Myanmar, SH.STA.SMSS.UR.ZS was the most impacting feature in the group 1 of Health ( Fig 6 ). Moreover, it was also the most impacting feature among seven representative indicators selected from 7 groups in Myanmar ( Fig 7 ).

(A) Ridge. (B) Lasso. (C) Elastic Net.

https://doi.org/10.1371/journal.pgph.0003764.g006

https://doi.org/10.1371/journal.pgph.0003764.g007

https://doi.org/10.1371/journal.pgph.0003764.t004

The number of features, training ratio, and window size played significant roles in tuning the optimal models ( Fig 8 ). The MAE values range from 2–32 (mean = 11, median = 10). The lower MAE values indicate the better models. The training ratio 0.6 releases MAE values larger than the models with training ratio 0.7 and 0.8. The window sizes which were lower than 3 or higher than 4 at big epochs and batches also result the higher MAE values. There was no significant difference between the one-feature model ( Fig 9A & 9B ) and three-feature model ( Fig 9C & 9D ). Both feature sets were better than the 7 feature sets ( Fig 9E & 9F ). In general, the predictability of malaria incidence could be predicted by World Bank indicators via machine learning and artificial intelligence at national or subregion levels yearly.

The parameters of the LSTM model included the number of features, window size, training ratio, epochs, and batch size. The evaluation data of the LSTM network are provided in S7 Table . One-feature models used each of the indicators ’SH.STA.SMSS.UR.ZS’, ’SP.DYN.IMRT.IN’, and ’SP.POP.TOTL.MA.ZS’. Three-feature model used all of three indicators ’SH.STA.SMSS.UR.ZS’, ’SP.DYN.IMRT.IN’, and ’SP.POP.TOTL.MA.ZS’. Seven-feature model used all of seven indicators ’SH.STA.SMSS.UR.ZS’, ’SP.DYN.IMRT.IN’, ’SP.POP.TOTL.MA.ZS’, ’NY.GDP.PCAP.KD’, ’NY.GNP.PCAP.PP.CD’, ’SL.AGR.EMPL.FE.ZS’, and ’SP.RUR.TOTL.ZS’.

https://doi.org/10.1371/journal.pgph.0003764.g008

(A & B) One-feature model used the indicators ’SH.STA.SMSS.UR.ZS’. (C & D) Three-feature model used all of three indicators ’SH.STA.SMSS.UR.ZS’, ’SP.DYN.IMRT.IN’, and ’SP.POP.TOTL.MA.ZS’. (E & F) Seven-feature model used all of seven indicators ’SH.STA.SMSS.UR.ZS’, ’SP.DYN.IMRT.IN’, ’SP.POP.TOTL.MA.ZS’, ’NY.GDP.PCAP.KD’, ’NY.GNP.PCAP.PP.CD’, ’SL.AGR.EMPL.FE.ZS’, and ’SP.RUR.TOTL.ZS’.

https://doi.org/10.1371/journal.pgph.0003764.g009

Many previous publications have shown that climatic factors such as temperature, moisture, and precipitation are strongly correlated with mosquito-borne diseases [ 31 – 33 ]. However, this study showed that climatic factors were uncorrelated with malaria incidence in the GMS during 2000–2022. The reason could be that the input data in previous studies were collected locally where latitude, longitude, elevation, and weather changed in limited regions compared to a very wide range of climatic patterns across the different regions of the GMS in this research. This provides hope for local communities in hot and humid regions to combat against not only malaria but also mosquito-borne diseases by considering and affecting other key factors.

Other reports have indicated that many sociodemographic factors, such as urbanization, education, and occupation are strongly correlated with mosquito-borne diseases [ 15 , 34 ]. This research also reinforces these findings and provides other noticeable indicators. Here, people using at least basic sanitation services and people practicing open defecation are among the crucial indicators affecting regression models. Many previous studies have shown that poor sanitation, open defecation, and improper wastewater management are ideal breeding conditions for mosquitoes [ 35 – 38 ]. Although rapid urbanization in developing countries could cause urban health problems such as air pollution, garbage, heat, the urban island effect, and water containers for larval habitats [ 39 ], the results here show that rural land area and rural population living in area where elevation is below 5 meters were correlated positively strongly with malaria incidence, as opposed to urban ones in most GMS countries. However, the limited number of datapoints in land area and population should be considered to conclude urbanization or ruralization facilitates malaria. This pattern was reversed in Cambodia. Tonle Sap Lake, the largest freshwater lake in Southeast Asia [ 40 ], could contribute greatly to the larval habitat in Cambodia. This research also revealed that the renewable internal freshwater resources per capita were correlated positively strongly with malaria incidence in most of the GMS, especially in Cambodia and Vietnam. The increase in hydro dams in the Mekong River to satisfy the unstoppable demands of economic development will be one of the main roots for spreading mosquito borne diseases in future.

Furthermore, behaviors at the household level, for example, improving housing quality and removing larval habitats, provided evidences for preventing mosquito-borne diseases [ 41 – 43 ]. A prior research on socioeconomic and household risk factors with malaria showed using wood and dung cakes as cooking fuel were significantly more at risk to have malaria cases [ 44 ]. Others reported carbon dioxide as a mosquito attractant on Aedes [ 45 , 46 ], Culex [ 47 ] and Anopheles [ 48 , 49 ]. Here, the indicators of accessibility of clean fuels and technologies for cooking is correlated negatively strongly with malaria incidence in most GMS countries. Smartphone geospatial apps and other mobile technology-tools have been used for disease surveillance in community [ 50 ]. This research also shows that individuals using the Internet is correlated negatively strongly with malaria incidence.

In addition, nutritional conditions, and healthy habits are also among the important key factors for reducing mosquito-borne diseases according to this study. Mosquitoes are attracted by several specific chemicals, such as carbon dioxide, lactic acid, and oct-3-enol, that are emitted by tobacco smokers [ 45 , 46 , 51 – 55 ]. Here, the prevalence of current tobacco use was also strongly correlated with the incidence of malaria in all GMS countries. In addition to tobacco users, pregnant women who exhale more carbon dioxide also attract more mosquitoes [ 56 ]. According to this research, female individuals aged 15–19 years are among most affecting features in regression models. Some odorous compounds produced by skin bacteria and emitted strongly by young adults that might attract mosquitoes [ 57 , 58 ]. Beside young adults, dependent population such as elderly individuals and infants who usually have weak immune systems and limited mobility were critically affected by malaria.

While most previous reports about predicting malaria and other mosquito-borne diseases by regression and time series forecasting models used climatic factors as features [ 13 , 14 , 18 , 19 , 41 ], the results here provide another approach using globally referenceable World Bank indicators. Some indicators, such as land area, have changed little for a long time, so it is difficult and expensive to collect data yearly. Although the yearly datasets in specified countries cause a challenge in building high resolution models for certain geographic locality, the recommended features used for building predicting models are among the indicators that are strongly correlated with malaria and are found similarly in most Mekong countries. Those results will deliver stakeholders and policymakers important references to make national and transnational decisions. However, political upheaval and humanitarian crisis which were main reasons of malaria increasing in Myanmar since 2021 have not been estimated by WDI.

Despite many steady efforts towards malaria elimination in GM, antimalarial drug resistance is still a concern in GMS. Malaria is among the most common fatal vector-borne diseases, especially in low- and middle-income countries. LM has been thriving in tourism, a green industry, owing to its diversified original culture and nature, which are less impacted by humans even though this also increases the risk of the spread of communicable diseases. To pursue sustainable development goals, transnational governments in GMS need to communicate effectively by sharing electronic data and information about communicable diseases, including malaria.

Malaria is still one of the most severe public health problem midst the multidrug-resistant malaria in Cambodia and the increase in hydro dams in the Mekong River. From 1494 indicators from WDI and CCKP, this research provided the sets of 4 and 5 containing respectively 19 and 149 indicators highly correlated with malaria incidence which was found respectively similarly for five and four GMS countries by Pearson and Distance correlation. They indicate malaria incidence are correlated with the living areas, ages, careers, health habits, economic status, technology accessibility, infrastructural facilities of information, transportation, and energy. From the set of 4, 92 indicators containing > = 20 datapoints were analyzed by MLR, Ridge, Lasso, and Elastic Net regressions. Seven most impacting features from seven topic groups were chosen for LSTM model. WDI can be used for predicting malaria incidence by LSTM model with one feature. While WDI could be referred for transnational or national level decisions, certain geographic areas still need high resolution time series indicators such as climatic indicators for disease surveillance. However, public health crises in GMS caused by political instability should be analyzed by political indexes for more precise prediction.

Supporting information

S1 table. dataset of world development indicator..

https://doi.org/10.1371/journal.pgph.0003764.s001

S2 Table. Peason and Distance correlation coefficients.

https://doi.org/10.1371/journal.pgph.0003764.s002

S3 Table. Number of datapoints in the set of 4 and 5.

https://doi.org/10.1371/journal.pgph.0003764.s003

S4 Table. Collinearity via Pearson correlation between the indicators of the set of 4.

https://doi.org/10.1371/journal.pgph.0003764.s004

S5 Table. MLR coefficients of total of not-null indicators in the set of 4.

https://doi.org/10.1371/journal.pgph.0003764.s005

S6 Table. MLR, Ridge, Lasso, and Elastic Net coefficients of the indicators in each group of topics.

https://doi.org/10.1371/journal.pgph.0003764.s006

S7 Table. The evaluation and coefficients of long short time memory network for Myanmar.

https://doi.org/10.1371/journal.pgph.0003764.s007

Acknowledgments

The author would like to express the gratitude to her family and communities, who support her unconditionally. The author would like to acknowledge the Mekong-U.S. Partnership Young Scientist Program in the second year known as “Lower Mekong Initiative 2019 (LMI)—Public Health & Bioinformatics: Using Information Technologies to Address Public Health Challenges”. The program was sponsored by the U.S. State Department and implemented by Arizona State University (ASU) and the National University of Laos (NUOL).

• View Article
• Google Scholar
• PubMed/NCBI
• 28. Kirch W, editor. Pearson’s Correlation Coefficient. Encyclopedia of Public Health. Dordrecht: Springer Netherlands; 2008. pp. 1090–1091. https://doi.org/10.1007/978-1-4020-5614-7_2569
• 38. Tyagi V, Saikia S, Dey D, Naorem AS, Tyagi V, Chattopadhyay P, et al. Sewage and Wastewater Management to Combat Different Mosquito Vector Species. In: Kashyap BK, Solanki MK, editors. Current Research Trends and Applications in Waste Management. Singapore: Springer Nature Singapore; 2023. pp. 101–117. https://doi.org/10.1007/978-981-99-3106-4_4

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

The PMC website is updating on October 15, 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• ISRN Prev Med
• v.2013; 2013

An Overview of the Malaria Control Programme in Zambia

Emmanuel chanda.

1 Directorate of Public Health and Research, National Malaria Control Centre, Ministry of Health, P.O. Box 32509, Lusaka, Zambia

Mulakwa Kamuliwo

Richard w. steketee.

2 Malaria Control and Evaluation Partnership in Africa, PATH, P.O. Box 900922, Seattle, WA 98109, USA

Michael B. Macdonald

3 Global Malaria Programme, WHO Headquarters, World Health Organization, Avenue Appia 20, 1211 Geneva, Switzerland

Olusegun Babaniyi

4 WHO Country Office, World Health Organization, P.O. Box 51449, Ridgeway, Lusaka, Zambia

Victor M. Mukonka

5 Department of Public Health, School of Medicine, Copperbelt University, P.O. Box 71191, Ndola, Zambia

The Zambian national malaria control programme has made great progress in the fight against Malaria. The country has solid, consistent, and coordinated policies, strategies, and guidelines for malaria control, with government prioritizing malaria in both the National Health Strategic Plan and the National Development Plan. This has translated into high coverage of proven and effective key preventive, curative, and supportive interventions with concomitant marked reduction in both malaria cases and deaths. The achievements attained can be attributed to increased advocacy, communication and behaviour changes, efficient partnership coordination including strong community engagement, increased financial resources, and evidence-based deployment of key technical interventions in accordance with the national malaria control programme policy and strategic direction. The three-ones strategy has been key for increased and successful public-private sector partner coordination, strengthening, and mobilization. However, maintaining the momentum and the gains is critical as the programme strives to achieve universal coverage of evidence-based and proven interventions. The malaria control programme's focus is to maintain the accomplishments, by mobilizing more resources and partners, increasing the government funding towards malaria control, scaling up and directing interventions based on epidemiological evidence, and strengthen active malaria surveillance and response to reduce transmission and to begin considering elimination.

1. Introduction

Malaria continues to be a disease of major public health significance in Zambia despite recent successes in scaling up interventions and documented reductions in malaria burden among children [ 1 – 4 ]. The report article entitled “Achievements in Malaria Control: The Zambian Story 2000–2010” was published in 2010 by the Directorate of Public Health and Research of the Ministry of Health (MoH) in Zambia [ 2 ]. The publication indicates that in the 10–20 years leading up to the year 2000, relatively limited malaria prevention existed in the country and much of the activities were focused on treatment of malaria. This led to steady increase in the disease burden, with hospital admissions increasing from 8.8% in 1976 to over 20% in the 1990s. Accordingly, case fatality rates in hospitalized patients increased from 10.6 deaths per 1000 malaria admissions in 1976 to 51 deaths per 1000 malaria admissions in 1994 [ 5 ]. In 1999, approximately 3.46 million malaria cases were recorded for a population of 10.8 million inhabitants. The malaria case rate was 4- to 5-fold higher in children under 5 years of age compared to those above 5 years of age. The situation prompted the Zambian Government to place malaria as a priority area and clearly outlined it in both the National Health Strategic Plan and the National Development Plan [ 6 – 8 ]. In an effort to reduce the impact of malaria and contribute to the attainment of the Roll Back Malaria (RBM) targets and health related Millennium Development Goals (MDGs), malaria control measures using an integrated approach with evidence-based proven prevention, control and management interventions were reintroduced in Zambia [ 6 – 9 ].

Major malaria vectors in the country are Anopheles gambiae s.s. An. arabiensis and An. funestus s.s [ 10 ]. The predominant malaria parasite species is Plasmodium falciparum , with Plasmodium malariae and Plasmodium ovale accounting for less than 5 percent [ 11 ]. Zambia's initial National Malaria Control Strategic Plan covered the period from 2000 to 2005; the plan was updated for 2006 to 2010, setting ambitious goals to scale up a package of malaria interventions [ 7 , 9 ]. The key malaria prevention, control and management strategies that Zambia took to mitigate the disease are: (1) vector control using indoor residual spraying (IRS) and promotion of ownership and use of insecticide-treated nets (ITNs); (2) malaria case management using effective diagnostics and lifesaving drugs-artemisinin-based combination therapy (ACTs); (3) control of malaria in pregnancy through intermittent presumptive treatment (IPTp) strategy; (4) information, education, and communication (IEC)/behavioural change communication (BCC) strategies.

The country has made great progress in the fight against malaria (Tables ​ (Tables1 1 and ​ and2). 2 ). The operational scale deployment of effective control tools has transformed the epidemiological profile from country-wide high endemicity to three distinct epidemiological strata: very low transmission and parasite prevalence of <1%, low transmission (1–10%), and persistent high transmission (>10%) [ 2 – 4 ]. Intermittent presumptive treatment in pregnancy (IPTp) uptake has reached the RBM target at 86% including uptake of two to three doses of IPTp representing 70% which is one of the highest in Africa [ 2 – 4 ] (Tables ​ (Tables1 1 and ​ and3). 3 ). The incidence of malaria has declined by 39% between 2006 and 2008, and a more than 60% decline in inpatient malaria cases between 2001 and 2008, in both under 5 and 5 to 15 year age groups [ 6 – 8 , 12 – 14 ]. Parasite prevalence among children under five in Zambia declined from 22% to 16% in 2010 [ 2 – 4 ] ( Table 1 ).

Benchmarking change in Zambia.

Source of data: DHS, MIS, and reports (2001 to 2010).

Changes in child mortality rates in 2001/02 and 2007.

Source: Zambia Demographic and Health Survey, 2001/2 and 2007.

Summary of progress in MIP interventions.

Source: Zambia Demographic and Health Survey, 2001/2, MIS; 2006, 2008, and 2010.

The report aims at sharing with the rest of the malaria community the achievements made by the malaria control programme in Zambia and highlighting the need to maintain the thrust and the gains as the programme strives towards achieving universal coverage of evidence-based and proven interventions. Particularly, the need to scale up and direct interventions based on epidemiological and entomological evidence (including insecticide susceptibility and management of resistance) strengthens active malaria surveillance and response to reduce transmission, to address the epidemiological differences across the country, and utilize the evidence for ongoing refinement of policy and strategy and strengthens malaria control operations at provincial, district, and community levels in accordance with national policies based on decentralization programs to consolidate partnership and performance management in order to address human and financial resource needs, commodity requirements, and program action, as well as addressing the low utilisation and acceptance of interventions through increased advocacy, education, and communication for behaviour change.

The main findings or arguments of the report are that (1) sustaining high levels of transmission-reducing interventions is critical to the long-term success of malaria control and its future elimination; (2) a solid and predictable resource base is absolutely required for effective planning and efficient programme implementation; (3) mobilization and efficient coordination of partners have markedly contributed to the success of the malaria control efforts in Zambia; (4) advocacy, communication, and behavioural change are key for strengthened political will, national leadership, community ownership and involvement, and concerted efforts from all stakeholders; (5) all these aspects together could facilitate for the ultimate attainment of a malaria-free Zambia.

Thus, the success that Zambia has achieved in malaria control can be attributed to the strong partnerships, increased resources, and evidence-based deployment of interventions in accordance with the national malaria control programme (NMCP) policy and strategic direction [ 15 ]. In light of enhanced advocacy and strengthened partnerships, there is unequivocally strong need for thorough evaluation of the performance of different aspects of the control programme. Herein we provide an in-depth evaluation of the strengths, weaknesses, and key issues of the report on the achievements of malaria control in Zambia.

The report is well written and attractively produced but there are some notable gaps. For example, the report does not give from the outset a clear background of the country's demographic and epidemiologic description. However, Zambia is situated in the Southern African region between 8° and 18° degrees south latitude and between 20° and 35° degrees east longitude with a population of approximately 13 million [ 16 ] in 10 provinces ( Figure 1 ). There are three distinct seasons: a cool and dry season from April to August, a hot and dry season from August to November, and a warm and rainy season from November to April. Malaria is endemic with regular and moderate to high transmission across the entire country with a seasonal pattern of high transmission peaks between December and May coinciding with the rainy season [ 5 ].

Map of Zambia showing the location of the neighbouring countries in Southern Africa.

There is a clear indication that the MoH implements a sector wide approach (SWAp) which harnesses the pooling of financial resources into the district basket funding leading to regular, predictable, and sustained flow of resources [ 17 ]. However, the report does not bring out strongly the challenges of the dwindling financial resources that have followed in the wake of diminishing donor support and the limited government funding for malaria control. Equally, most challenges are not addressed adequately but rather confined to specific interventions even when they relate to all aspects of malaria control. To illustrate, the lack of adequate competent human resource pool in the health sector necessary for driving forward the malaria control agenda is minimally addressed. The report only alludes to this challenge in relation to operational research and malaria case management and diagnosis.

In the same vein, coordination and partnerships at district level that remains a major stumbling block to effective deployment of interventions received little attention, as indicated by the need to reinforce partnership engagement for IRS, particularly with the local authorities, at this level. Generally flaws in the supply chain management of commodities and equipment have resulted in delayed implementation of key preventive interventions and timely management of the disease. The report only mentions the need to collaborate with Medical Stores Limited and the reproductive health department to assure supplies of sulphadoxine-pyrimethamine (SP) for IPTp and streamline the distribution of SP to all ante natal clinic facilities. Most statements in the report are either not or are inadequately referenced. The success story could have been greatly enhanced if the foregoing shortfalls were addressed.

2.1. Policy and Strategic Direction

Since 2000, when the RBM initiative was launched in Zambia, the number of malaria programme partners has increased, translating into increased financial, technical, material, and human resources for malaria control. Zambia is fully committed to reducing the impact of malaria and contributing to the attainment of the Abuja Declaration, Millennium Development Goals (MDGs), and the RBM targets. The country has developed the National Health Strategic Plan 2011–2015, which is very critical to achieving the MDGs, and the sixth National Development Plan that focuses on malaria elimination as one of the key health priorities. A comprehensive National Malaria Control Strategic Plan from 2011 to 2015 including several intervention-specific guidelines has also been developed.

2.2. Technical Interventions

Scaling up of malaria prevention and control programme interventions has been intensified by the MoH with substantial and important scores made towards achieving the health-related Millennium Development Goals (MDGs) and other key national achievements in relation to RBM targets. With assistance from valuable partners, strong leadership, and political will, the MoH has expanded the availability and access to ITNs with over seven million having been distributed since 2004, with increased ownership of ITNs from 38 percent (2006) to 64 percent in 2010 (Figures ​ (Figures3 3 and ​ and4). 4 ). Coverage of IRS has been scaled up from five initial districts in 2003 to 15 in 2006, 36 in 2008, 54 in 2010, and 72 in 2012 ( Figure 5 ), and uptake of full dosing of IPTp has increased significantly from 59 percent (2006) to 70 percent 2010. Equally, maternal mortality rate per 100,000 population decreased markedly from 729 in 2001 to 591 in 2007. The government has ensured availability of malaria commodities such as diagnostic tools and efficacious drugs at all of public health facilities including at community level using community health workers for home management of malaria.

ITN used by children under age of five years in rural and urban areas (source: Zambia Malaria Indicator Surveys 2006–2010 ) [ 3 , 8 ].

District level ITN coverage expressed as percentage of 3 ITNs distributed per district household estimate, by three-year intervals. (Source: Zambia Malaria Indicator Surveys 2006–2010) [ 3 , 8 ].

Progressive scale-up of indoor residual spraying from 2003 to 2010.

Nevertheless, the publication covers the achievements of malaria control in Zambia broadly and yet with sufficient information to be useful to other control programmes that are intending to scale up interventions. It highlights the integrated approach that the national malaria control programme has implemented [ 6 – 9 ]. The strategy emphasizes a core set of evidence-based proven preventive and treatment interventions for malaria control [ 18 , 19 ] including: ITNs with a vision of attaining universal coverage with all sleeping spaces in all households [ 13 , 14 , 20 ] (Figures ​ (Figures3 3 and ​ and4); 4 ); IRS to ensure that at least 80% of the targeted structures in IRS eligible districts are protected [ 21 , 22 ]; case management and parasite detection to ensure that at least 80% of malaria patients to receive prompt and effective diagnosis and treatment within 24 hours of onset of symptoms [ 23 , 24 ]; IPTp to ensure that at least 80% of pregnant women have access to the package of interventions (SP and ITN) to reduce the burden of malaria in pregnancy [ 6 – 8 , 19 ].

2.3. Monitoring and Evaluation

The MoH has further conducted surveys and reviews to assess the impact of malaria prevention and control interventions. These include the Demographic and Health Survey [ 25 , 26 ], Malaria Indicator Surveys [ 3 , 8 , 14 ], and the Malaria Programme Review [ 4 ]. The impact of Zambia's interventions is visible through the reduction of the annual number of malaria deaths by over 60 percent between 2000 and 2008 [ 1 ]; under five malaria deaths by 41 percent between 2006 and 2008; reduced severe anemia rates in children by 56 percent (2006–2010) ( Table 1 ). According to the WHO assessment conducted in 2008, Zambia recorded a decline in malaria cases by 66 percent [ 1 ]. With this achievement, the country has surpassed targets set by (i) the Abuja Declaration by Heads of States in 2000 of reducing malaria illness and deaths by fifty percent by 2010, (ii) the RBM goal of reducing the global malaria burden by fifty percent by 2010.

One notable innovation worth grasping is the efficient utilization of supportive strategies to streamline uptake and purposeful deployment of key preventive and treatment tools. In Zambia, implementation of key malaria control interventions is augmented with cross-cutting supportive approaches. The report highlights an interactive advocacy, communication, and behaviour change to enhance utilization of interventions through promotion of appropriate care seeking behaviour [ 8 ]; viable operations research (OR) feeding into and providing timely and sound evidence to guide implementation of malaria control and inform policy decision making. Here a unique Zambia feature is coordination of the OR network, with strong collaborations of various local and international research institutions, whose information is shared with all stakeholders such as implementers, policy makers, funding agencies, and academic institutions. There is strong evidence-based monitoring and evaluation to facilitate for the documentation of progress made towards the achievement of goals and targets of the United Nations MDGs by 2015. Zambia also has solid, consistent, and coordinated policies and strategies for malaria control in place. This includes a comprehensive national malaria strategic plan for 2011–2015, policy guidelines for key interventions, and support services as well as budgeted annual work plans.

With the country-wide scaling up of vector control interventions, entomological monitoring and management of insecticide resistance is the major challenge [ 10 ]. In response to this, Zambia is again unique in developing a robust network of local Malaria Institute at Macha (MIAM), Zambia Integrated System Strengthening Programme (ZISSP), University of Zambia (UNZA), Tropical Disease Research Centre (TDRC), Zambia Environmental Management Agency (ZEMA), international-World Health Organization (WHO), Centres for Disease Control and Prevention (CDC), United States Agency for International Development (USAID), Malaria Transmission Consortium (MTC), Innovative Vector Control Consortium (IVCC), John Hopkins Malaria Research Institute (JHMRI), and the Liverpool School of Tropical Medicine (LSTM) entomology partners with clear terms of reference to consolidate and coordinate resistance monitoring and data collation to make recommendations for pesticide procurement [ 27 ].

2.4. Partnership and Coordination

The Ministry of Health leads malaria control efforts in Zambia through its National Malaria Control Centre (NMCC), provincial and district health offices and health facilities. Many multilateral agencies, nongovernmental organizations, research institutions, and community-based organizations are engaged in malaria control efforts throughout the country in implementing interventions, training health workers, and strengthening IEC/BCC. Increased community and private sector engagement coupled with strong partnership coordination is striking. Notably the national IRS program was built upon collaboration with Konkola Copper Mines (KCM), Mopani Copper Mines, and Zambia Sugar programs. The success of the malaria control can be ascribed to exceptional efforts towards establishment of strong partnership coordination; engaging community leaders and health workers as front line in the fight, an emphasis that echoes the mission of the ministry of health to provide quality health care as close to the family as possible; involvement of private sector to complement the public sector efforts and strengthening of malaria operations research to facilitate for evidence-based programming. The strengthening and coordination of partners under the stewardship and leadership of government has contributed to the increased and sustained number of multilateral, bilateral, national, faith-based, private sector, and community organizations. More specifically, the three-ones approach: one coordinating mechanism; one implementation plan, and one monitoring plan is largely responsible for the success in partner coordination, strengthening, and mobilization.

Zambia has some unique stories to tell. One of the unique feature of the Zambian NMCP is the partnership with community-based organizations such asestablishment of the Zambia Malaria Foundation to operationalize the concept of an NGO umbrella group. This provided a forum to engage and coordinate with a very broad range of NGOs, from the Zambia Scouts Association (who used to help in the net retreatment campaigns) to the small youth and church groups and to business groups such as Rotary, as well as the Zambia Association of Chambers of Commerce and Industry (ZACCI). In addition, there has been an exceptional partnership with the HIV/AIDS programs. Both in information, education, and communication (IEC) and behaviour change communication (BCC) in support for ITNs targeting people living with HIV/AIDS (PLWHA) through home-based care groups such as the “Reaching HIV/AIDS Affected People with Integrated Development and Support” (RAPIDS) project. The Zambian programme was one of the first to really embrace integrated management of child illnesses (IMCI), then the first for “Fresh Air,” NGO coordination, the first for nation-wide roll out of ACT (Coartem), the second (after the small district distribution in Ghana by IFRC) for a mass-free distribution of LLINs, and the first for the integrated vector management strategy (IVM) policy [ 18 ].

The NMCC is pivotal in providing technical guidance, leadership, and coordination of malaria control and preventive activities. It ensures full participation and involvement of partners in the development of key documents: strategic plans, annual action plans, and policy guidelines through intervention specific multisectoral technical working groups for vector control; information, education, and communication; monitoring, evaluation, and operations research.

2.5. Financing and Human Resources

The NMCP receives financial and technical support from a variety of organizations to enable a coordinated approach to scaling-up interventions and tracking progress. Partners providing the largest financial contributions to malaria control efforts in Zambia apart from government includes the World Bank, Global Fund to fight HIV, TB, and malaria (GFATM), and the United States' President's Malaria Initiative (PMI) ( Figure 2 ). The challenge of limited human resources for malaria control is circumvented by increased capacity building at provincial and district levels and by increasing collaboration with other implementing partners.

Zambia: external funding (in millions, US$) for the Zambia Malaria Control Programme, 2003–2010. Source: Roll Back Malaria Progress and Impact Series—focus on Zambia.

As funding for malaria control gets tighter it is important for countries to demonstrate “the business case” that investments in malaria control reap economic and social benefits. Zambia has solid evidence from the private sector that is, the programme has a unique collaboration with the private sector; such as the mining industry; Konkola Copper Mines, Mopani Copper Mines, and the agricultural sector; Zambia Sugar company programs that have shown a “positive return for investment” for their workplace malaria programs. There has also been a lot of engagement with Zambia Association of Chambers of Commerce and Industry (ZACCI) to try to expand “the business case” to other sectors.

3. Conclusions

The malaria control programme in Zambia has made great achievements in its control efforts through provision of high coverage of malaria prevention and curative services. The success can be attributed to the strong partnerships including community engagement, increased resources, and evidence-based deployment of key technical and supportive interventions in accordance with the national malaria control programme policy and strategic direction. The country offers some unique models and experiences that could really benefit other programmes in the region. Community level integrated entomological and case surveillance, prompt effective treatment, and sustained high levels of contemporary malaria prevention tools, are pivotal to the long-term success of malaria control and future malaria elimination. However, there is great need for increased resource mobilization by broadening the partnership base and increasing the government commitment to malaria control.

Authors' Contribution

E. Chanda systematically evaluated the Zambian story text and drafted the review paper. M. Kamuliwo, R. W. Steketee, M. B. Macdonald and O. Babaniyi collaborated and critically reviewed the paper. V. M. Mukonka conceived the idea, developed the technical frame work, and participated in the drafting process of the paper.

Acknowledgments

The authors thank Dr. Peter Mwaba, Permanent Secretary, and Dr. Elizabeth Chizema-Kawesha, Director Public Health and Research, Ministry of Health, Zambia. The authors particularly acknowledge Mrs Pauline K. Wamulume for conceiving the idea of publishing the achievements of the malaria control programme in Zambia.