http://orcid.org/0000-0001-9024-3126
Figures
Abstract
Introduction
Malaria is a public health emergency in India and Odisha. The national malaria elimination programme aims to expedite early identification, treatment and follow-up of malaria cases in hot-spots through a robust health system, besides focusing on efficient vector control. This study, a result of mass screening conducted in a hot-spot in Odisha, aimed to assess prevalence, identify and estimate the risks and develop a management tool for malaria elimination.
Methods
Through a cross-sectional study and using WHO recommended Rapid Diagnostic Test (RDT), 13221 individuals were screened. Information about age, gender, education and health practices were collected along with blood sample (5 μl) for malaria testing. Altitude, forestation, availability of a village health worker and distance from secondary health center were captured using panel technique. A multi-level poisson regression model was used to analyze association between risk factors and prevalence of malaria, and to estimate risk scores.
Results
The prevalence of malaria was 5.8% and afebrile malaria accounted for 79 percent of all confirmed cases. Higher proportion of Pv infections were afebrile (81%). We found the prevalence to be 1.38 (1.1664–1.6457) times higher in villages where the Accredited Social Health Activist (ASHA) didn’t stay; the risk increased by 1.38 (1.0428–1.8272) and 1.92 (1.4428–2.5764) times in mid- and high-altitude tertiles. With regard to forest coverage, villages falling under mid- and highest-tertiles were 2.01 times (1.6194–2.5129) and 2.03 times (1.5477–2.6809), respectively, more likely affected by malaria. Similarly, villages of mid tertile and lowest tertile of education had 1.73 times (1.3392–2.2586) and 2.50 times (2.009–3.1244) higher prevalence of malaria.
Conclusion
Presence of ASHA worker in villages, altitude, forestation, and education emerged as principal predictors of malaria infection in the study area. An easy-to-use risk-scoring system for ranking villages based on these risk factors could facilitate resource prioritization for malaria elimination.
Citation: Panda B, Mohapatra MK, Paital S, Kumbhakar S, Dutta A, Kadam S, et al. (2019) Prevalence of afebrile malaria and development of risk-scores for gradation of villages: A study from a hot-spot in Odisha. PLoS ONE 14(9): e0221223. https://doi.org/10.1371/journal.pone.0221223
Editor: Luzia Helena Carvalho, Instituto Rene Rachou, BRAZIL
Received: June 3, 2019; Accepted: August 1, 2019; Published: September 6, 2019
Copyright: © 2019 Panda et al. 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: The data is available with DRYAD as given: Data from: Prevalence of afebrile malaria and development of risk-scores for gradation of villages: A study from a hot-spot in Odisha; Journal: PLOS ONE; DOI: doi:10.5061/dryad.443s9p4; Data files: individual_screening; DOI: doi:10.5061/dryad.67bf868; Data file(s):village_level.
Funding: This manuscript emanated from a research study funded by the CCRH, Ministry of Ayurveda, Yoga and Naturopathy, Unani, Siddha and Homoeopathy. Co-authors AK, DN, and RKM belong to the funding agency and have contributed to editing and finalization of this manuscript. The specific contributions of these authors are articulated in the 'author contributions' section.
Competing interests: Co-authors AK, DN, and RKM belong to the funding agency which funded the research underlying this paper, CCRH, Ministry of Ayurveda, Yoga and Naturopathy, Unani, Siddha and Homoeopathy. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Malaria, one of the highly prevalent infectious diseases, accounted for about 216 million new cases and 0.45 million deaths in 2016, globally [1]. The most common manifestation of malaria is the typical characteristic presentation a cyclical syndrome: fever, chill, sweat, headache, and vomiting—mostly in non-immune individuals. Yet another lesser known type of ‘asymptomatic’ malaria continue to exist amongst individuals who have had partial or complete immunity to the disease—mostly in population residing in malaria endemic areas [2,3]. Asymptomatic cases pose greater challenges to the program managers as they act as hidden reservoirs of active infection that perpetuates sustained transmission. In settings that adopt passive surveillance such as in India, these reservoirs pose formidable challenges for malaria elimination [4]. Therefore, strategies to eliminate asymptomatic infections would have greater public health consequences not only from the point of view of caseload but also achievement the elimination objectives.
India is a signatory to the National Framework for Malaria Elimination (NFME). Government of India in close alignment with the Global Technical Strategy has detailed a roadmap for implementation of the national strategy for malaria elimination as to achieve the national goals by 2030 [5]. Under this national framework, about 40.0% reduction of incidences and mortalities by 2020 is envisaged as compared to that of 2015. Interventions such as mass distribution of Long-Lasting Insecticidal Nets (LLINs), intensification of Indoor Residual Spray (IRS) and administration of Artemisinin-based Combination Therapy (ACT), have resulted in temporary reduction of malaria cases, mostly febrile, though such cases represent only the tip of the iceberg. Further, the national malaria elimination program adopts passive surveillance for case identification, treatment and reporting[6]. Thus, the chances of ‘asymptomatic cases’ going unnoticed continue to remain high. Identification of asymptomatic “hotspots” and the strategy of targeted interventions ought to have strong strategic, operational and policy implications.
Taiwan, China, and Brazil have resorted to active case detection and treatment (ACDT), also referred to as mass test and treatment (MTAT), in which mass screening of at-risk population is undertaken through rapid diagnostic test (RDT), followed by effective treatment of detected cases as to reduce transmission of Malaria [4]. However, owing to the nature of transmission, mass interventions for malaria control were historically short-lived [7] as they required incessant mass screening and treatment options to eliminate possibility of any outbreak. Therefore, malaria elimination needs a systematic and targeted ACDT approach, driven by context-specific evidences about ‘hotspots’. The Recent success in reduction of malaria in Odisha [8] need to be considered with a word of caution, as our earlier experiences in dealing with Malaria in the State has had mixed outcomes.
Therefore, this paper, the first of its kind, aims to present evidence on the prevalence of asymptomatic reservoir in one of the highly endemic and inaccessible pockets of Odisha [9], and to offer a ‘risk-score’ as an easy-to-use tool for program managers to rank villages as per existence of risk factors, which has the potential to facilitate malaria elimination in the State.
Materials and methods
Study design
A cross-sectional baseline mass screening for Malaria was conducted in 47 villages spread across four sub centres (SC) of Pallahara Block, Angul district, Odisha. The main objective was to estimate symptomatic and asymptomatic malaria cases in the community, using WHO recommended Rapid Diagnostic Test (RDT). We also aimed to identify the predictors that could explain the variation in the prevalence of malaria in these sample villages.
Study setting and sampling
This study was conducted in four neighboring malaria endemic sub-centers of Pallahara block in Angul district of Odisha (21°29'N and 85°14'E). The average annual rainfall of Pallahara is reported 1551 mm. We used multistage stratified random sampling to sample the subcenters. First, 30 districts of the state were divided into four categories on the basis of reported API of 2016–17. Angul district was randomly selected from the top priority districts, reporting high API. In the second stage, three high API blocks were considered for Block selection. Pallahara Block was selected at random from the list of three. In the third stage, all the sub-centres in the selected block were divided into two groups based on their API and four sub-centres of Pallahara were chosen randomly from the higher API group. All villages of three sub-centres and seven villages, selected randomly, of the fourth sub-centre were included in the study. The villages had similar thick forest coverage and were dominated by tribal population. Data from the health department over past three years indicated P. Falciparum (Pf) as the predominant malarial species in that area. Mass screening was conducted in all 47 villages. A total of 13221 individuals out of 17552 enumerated population were screened.
Study procedures
Eight field investigators (FI), 2 research assistants (RA) and one senior research assistant were hired and trained under the project. The field team was divided into four sub-teams of two FIs and one RA. The senior research assistant (SRA) cross checked the validity of data and confirmation of fever cases in community. Screening was done in camps and through door-to-door visit for the left-out population. Rapid diagnostic tests were performed using WHO-approved EzDx antigen Pf/Pv kit manufactured by ADVY chemicals, India at the field sites that contained a monoclonal anti-P. Falciparum histidine-rich protein II (HRP-II) specific antibody and an anti-P. Vivax (Pv) p-LDH-specific antibody to detect PF and PV malarial parasite infections. All positive cases were treated with an artemisinin-based combination therapy (ACT), a combination of artesunate (AS) plus sulfadoxine-pyrimethamine (SP), as per the National Vector Borne Disease Control Program (NVBDCP) guidelines.
Definitions
A test using RDT is considered malaria positive if either or all of the two test bands (PF test line and PV test line) along with the control band are noticeable. A case of febrile malaria was defined as an individual with a history of fever within the past 48 hrs and with an axillary temperature of >37.5°C at the time of survey, while an afebrile case of malaria was defined as absence of fever within the past 48 hrs and an axillary temperature ~37.5°C at the time of survey.
Data collection and management
A client-centric register was used for data collection. Questions related to basic demographic characteristics, such as, age, gender, education and health practices were asked. Blood samples (5 μl) were collected from the individuals to test PF and PV infection. Further, data related to village characteristics (eg., altitude, forestation, availability of a village health worker, distance from sub centre (SC) and primary health centre (PHC), etc) were collected in a separate register through expert-group consultations. The expert groups for each village consisted of representatives from the forest department, health department and senior citizens. Similarly, Altimeter Android application was used to measure the altitude of each village.
Statistical analysis
Distribution of fever cases, education, mosquito net use and all the other relevant co-variates were examined in descriptive tables. Statistical significance of the distribution differentials was tested using chi-square test. Further, the individual data was aggregated to calculate the village wise malaria prevalence (which is our principle outcome variable) and average education.
To estimate the association, we modelled the principle outcome variable using a multilevel model to account for the clustered nature of the data, i.e. villages were clustered within a SC. We used the poisson regression framework in the screening population as offsetting variable for modelling the unadjusted, the adjusted and the final analysis. The final model was constructed based on statistical significance of variables in unadjusted and fully adjusted models. Through fully-adjusted estimations, the impact of individual explanatory variables on the estimates of regression were measured using marginal effect of each variable at a given level, keeping other variables at average—also called Average Marginal Effect (AME)—this is an intuitive method to interpret complex outcomes of a general linear model with log links.
The marginal effect measures the change in the expected value of y as one independent variable increases by unity while all other variables are kept constant. Therefore, the average marginal effect computes the average of all the individual marginal effects[10]. We computed the final poisson model and then obtained the average marginal effects for all covariates by using the command “margins” available in package “margins” for R-software. The formula to compute AMEs for the ith explanatory variable is
Where βxk denotes the linear combination of parameters and variables for the k th observation.
F(.) denotes the cumulative distribution function.
We created algorithms based on the AMEs in order to rank and score the villages according to their probability of reporting higher prevalence. The scores against each factor were then summed up to create a village score which was used to rank the villages—the highest ranked village being the most risk-prone.
Ethical issues
The study obtained ethics approval from the Institutional Review Board of Indian Institute of Public Health, Bhubaneswar and subsequently from the State Research and Ethics Committee of Government of Odisha. Written informed consent and assent (translated into the local language) was obtained from all participants. No material benefits were offered to any of the study participants. Confirmed malaria cases were treated on the spot by suitable anti-malarial drugs and referred to the government institutions. Identities of all participants were anonymized using identity numbers and the decode keys were maintained only by the principal investigator.
Results
Our principal explanatory variables were village characteristics that posed a greater risk for malaria transmission and infection: availability of a village health worker (Accredited Social Health Activist), altitude of the village, percentage of land covered by forestation, distance of village from the nearest SC and the PHC, use of long-lasting Insecticidal nets (LLIN) and average years of education (Table 1).
We found hardly any difference in gender distribution across groups, while age distribution was significantly different. P. falciparum prevalence was more among 5–14 years age group, while Pv was prevalent among under-five children. The mean years of education was significantly higher among non-infected (3.7 years) than Pf infected (2.6 years), Pv infected (2 years) and mixed infected (2.4 years) population. Further, infection was significantly lower among professionals, skilled labors and homemakers and was comparatively higher among students and unemployed. Afebrile fever cases were more among Pv infected (81%) individuals than with Pf and mixed-infection individuals (Table 2).
Afebrile malaria accounted for 79 percent of all confirmed cases. And the prevalence of asymptomatic malaria among adults is higher than that of 0–5 years group. Four out of six independent variables were found to be significantly associated with prevalence of malaria in the final model. We found the prevalence 1.38 times higher in villages where ASHA didn’t stay as compared to villages where ASHA resided. Further, the risk increased by 1.38 and 1.92 times in mid- and high-altitude tertiles. With regard to forest coverage, villages falling under mid- and highest-tertiles were 2.01 times and 2.03 times, respectively, more likely to have been affected by malaria than villages under the lowest tertile. Similarly, villages of mid tertile and lowest tertile of education had 1.73 times and 2.50 times higher prevalence of malaria than those in the highest tertile (Table 3). These findings are also reflected in Fig 1.
The marginal effect from fully adjusted model showed that availability of ASHA in villages played a critical role in the prevalence of malaria. Non-residence of ASHA at villages contributed to 2.2 percent increase in malaria prevalence across the study sites. Similarly, mid- and highest-altitude tertiles had 1.017 and 1.042 AME as compared to the lowest tertile. The AME of medium forested villages and highly forested villages on Malaria prevalence was approximately 1.041. Finally, the AME for mid- and lowest-education tertiles were 1.028 and 1.056, respectively (Table 4).
Risk score estimation
We used poisson regression with population as offset to model the association of independent variables of interest. Though this model offers flexibility to specify potentially complex multivariate relationships, it also brings the risk of misinterpretation and miscalculation [11,12]. Moreover, the coefficient estimates derived from such generalized linear models (GLM) don’t provide unconditional marginal effects and therefore lose their direct interpretational value. Hence, in order to make the interpretations more intuitive, we used ‘average marginal effects’ estimate as it provides information about the rate at which the dependent variable changes at a given point in the covariate space with respect to one covariate dimension while holding all other covariate values constant [13,14]. To rank villages according to their probability of reporting high prevalence of malaria, we multiplied the AMEs with 100 to get the effect on a percentage scale which were rounded off to make the ranking tool easier for use (Table 5).
The risk estimation table (Table 5) gives specific scores to a village according to its place in the spectrum of risk factors. For example, a village in third tertile of altitude, second tertile of forestation and second tertile of average education with no ASHA residing in it will get scores of: 4, 4, 2.5 and 2, respectively, and a cumulative score of 12.5. All villages in a given geographic area could be scored and ranked from the highest to the lowest rank and prioritized from the point of anti-malaria programme implementation.
Discussion
This paper examined the prevalence of malaria, especially afebrile cases in one of the hotspots (Pallahara block) of Odisha, the state contributing maximum caseload and mortalities in India. It also analyzed the village level risk factors that influenced malaria prevalence. The results highlight a disproportionately high prevalence of afebrile cases. Further, availability of ASHA at village level had a direct and significant association with prevalence of malaria–this held true even after adjusting for known covariates such as altitude, forestation and education each of which had significant associations as well. The prevalence of afebrile malaria in all four SC areas was alarming, though the scenario may not be similar across all hotspots in Odisha.
A study conducted in the Kondagaon district of Chhattisgarh in India, a very similar terrain, found 65% asymptomatic cases among all positive cases [15]. Similarly another study from west Bengal found a very high prevalence of asymptomatic malaria (8.4%) in healthy tribal population in a malaria endemic area [16]. The burden of such asymptomatic malaria in endemic areas can be attributed to high herd immunity and low density of the infection [17]. Further, such findings are “textbook” examples of the “Iceberg” phenomena [18] and suggest that the clinically ill cases are only a handful of what lies beneath. No malaria elimination campaign would achieve the objectives without addressing these sub-clinical infections. This further emphasizes the importance of ACDT to detect and stop transmission of malaria infection.
Higher prevalence of infection was observed among children (0–14 yrs) as compared to adults that implies development of protective immunity with age. Several past studies re-affirm that acquired immunity increases with age, leading to higher prevalence of asymptomatic cases among adults [2,15,16,19,20]. Education as a principal factor for determining health outcomes was also found to be strongly associated with malaria prevalence [21–24].
Residence of ASHA, altitude, forestation and average education of village were found to be critical factors contributing to prevalence of malaria. Past studies have demonstrated the impact of community level health workers on prevalence of Tuberculosis (TB) and childhood mortality [25–28]. Ours is the first study showing strong association of ASHA’s residence with prevalence of malaria. It is also logical to relate the role of ASHA (screening fever cases, treating and monitoring) with anti-malaria programme implementation in a village. If an ASHA worker doesn’t stay in a village, these functions are likely to be delayed or disturbed.
Studies on malaria and altitude dynamics reveal that a fall in temperature is not conducive for vectors that transmit malaria [29,30], but in our study there is a direct association between altitude and prevalence of malaria–this may be mainly due to the fact that the range of altitude variation in this study settings is quite narrow, rather acts more as a proxy to forest dwelling for food habits, lesser acceptance of preventive measures leading to uninterrupted malaria transmission [31–34].
It is well established that forest ecosystem promotes vector sustainability and malaria transmission. A global assessment by Guerra et.al estimated that half of the risks for malaria is attributable to people living in forested areas [35]. Factors such as rainfall [36,37], humidity [38–40], tree canopy [41], and high organic content in breeding pools [31] in these forest regions influence malaria transmission as compared to other regions.
Finally, number of years of education emerged as one of the strongest associated factors, among all, with a clear gradient across tertiles. Education is a crucial social determinant of health across the globe [42] and it’s no different for Odisha. The strength of association between malaria and education need to be seen from the perspective of delivering better education to residents of hotspots which could potentially influence the probability of a sustained anti-malaria campaign and achievement of malaria elimination goals of the State. With education the health seeking behavior among the general public in such areas would also improve significantly [43], where educational attainment is poor in general.
Identification of risk factors and strengths of associations offered a unique opportunity to device an easy-to-use scoring tool for managers and administrators to rank villages which would have a huge implication in utilizing the limited resources efficiently in similar settings. Past studies have highlighted about the importance of predicting and placing early warning systems to combat malaria [36] as well as to delineate hotspots [44]. Our tool has the potential to add a chapter to further prioritization of high burden villages in hotspots for evolving a sustained and effective malaria elimination strategy.
Limitations
This is the first ever study in one of the hotspots of Odisha in terms of mass screening of the entire population living under the sample clusters. The tool developed for risk-scoring of villages is also a unique product. However, it has some limitations as well: the study was conducted in a typical tropical region with moderately elevated lands: generalization of findings to a wider population, especially hilly terrains should be done with caution. Secondly, we have used RDTs that have limited detectability of low parasitemia, and hence, the prevalence of asymptomatic malaria as found in this study might be lower than if tested with Polymerase Chain Reaction (PCR)-based kit. Finally, temperature variations and existence of water bodies have not been included in this analysis, which have effects on vector breeding and malaria transmission.
Conclusion
Presence of ASHA worker in villages, altitude, forestation, and education are strongly associated with malaria prevalence in a village. An easy-to-use risk-scoring system for ranking villages and prioritizing resource allocation for malaria elimination has the potential to change the paradigm of government’s approach in addressing malaria in the State of Odisha. Further in-depth investigation into the socio-cultural and behavioral patterns of people living in hotspots could through more light into the phenomenon of vector breeding and malaria transmission in the State.
References
- 1.
World Health Organisation. World Malaria Report 2017. Geneva: World Health Organization. World Health Organization. 2017. https://doi.org/10.1071/EC12504
- 2. Laishram DD, Sutton PL, Nanda N, Sharma VL, Sobti RC, Carlton JM, et al. The complexities of malaria disease manifestations with a focus on asymptomatic malaria. Malar J. 2012;11: 29. pmid:22289302
- 3. Chen I, Clarke SE, Gosling R, Hamainza B, Killeen G, Magill A, et al. “Asymptomatic” Malaria: A Chronic and Debilitating Infection That Should Be Treated. PLoS Med. 2016;13: 1–11. pmid:26783752
- 4. Sturrock HJW, Hsiang MS, Cohen JM, Smith DL, Greenhouse B, Bousema T, et al. Targeting Asymptomatic Malaria Infections: Active Surveillance in Control and Elimination. PLoS Med. 2013;10: e1001467. pmid:23853551
- 5.
DIRECTORATE OF NATIONAL VECTOR BORNE DISEASE CONTROL PROGRAMME (NVBDCP). National Framework for Malaria Elimination in India. 2016; 63. Available: http://nvbdcp.gov.in/Doc/National-framework-for-malaria-elimination-in-India-2016–2030.pdf
- 6.
DIRECTORATE OF NATIONAL VECTOR BORNE DISEASE CONTROL PROGRAMME (NVBDCP). National Framework for Malaria Elimination in India. 2016; 63.
- 7. Larsen DA, Bennett A, Silumbe K, Hamainza B, Yukich JO, Keating J, et al. Population-wide malaria testing and treatment with rapid diagnostic tests and artemether-lumefantrine in Southern Zambia: A community Randomized Step-Wedge Control Trial Design. Am J Trop Med Hyg. 2015;92: 913–921. pmid:25802434
- 8.
WHO | In its quest to eliminate malaria, India focuses on Odisha and the tribal states. WHO. 2018; Available: https://www.who.int/malaria/news/2018/india-elimination-odisha/en/
- 9.
National Vector Borne Disease Control Programme (NVBDCP) [Internet]. [cited 3 Jan 2018]. Available: http://nvbdcp.gov.in/aboutus.html
- 10. Bartus T. Estimation of marginal effects using margeff. Stata J. 2005;5: 309–329.
- 11.
Brambor T, Clark WR, Golder M. Understanding Interaction Models: Improving Empirical Analyses [Internet]. Political Analysis. Cambridge University PressSociety for Political Methodology; 2006. pp. 63–82. https://doi.org/10.2307/25791835
- 12. Norton EC. Computing interaction effects and standard errors in logit and probit models [Internet]. The Stata Journal. 2004. Available: www.jstor.org
- 13. King G, Tomz M, Wittenberg J. Making the Most of Statistical Analyses: Improving Interpretation and Presentation. Am J Pol Sci. 2007;44: 347.
- 14. Leeper TJ. Interpreting regression results using average marginal effects with R’s margins. Reference manual. 2017; 32.
- 15. Chourasia MK, Raghavendra K, Bhatt RM, Swain DK, Valecha N, Kleinschmidt I. Burden of asymptomatic malaria among a tribal population in a forested village of central India: a hidden challenge for malaria control in India. Public Health. 2017;147: 92–97. pmid:28404504
- 16. S. G, P. S, S.K. G, A. B, S. D, P.K. K, et al. High prevalence of asymptomatic malaria in a tribal population in eastern India. J Clin Microbiol. 2013;51: 1439–1444. pmid:23426929
- 17. Björkman AB. Asymptomatic low-density malaria infections: a parasite survival strategy? Lancet Infect Dis. 2018;18: 485–486. pmid:29398387
- 18.
Evans AS. Viral Infections of Humans: Epidemiology and Control. Springer US; 2013.
- 19. Chourasia MK, Raghavendra K, Bhatt RM, Swain DK, Meshram HM, Meshram JK, et al. Additional burden of asymptomatic and sub-patent malaria infections during low transmission season in forested tribal villages in Chhattisgarh, India. Malar J. 2017;16: 1–9.
- 20. Zhao Y, Zeng J, Zhao Y, Liu Q, He Y, Zhang J, et al. Risk factors for asymptomatic malaria infections from seasonal cross-sectional surveys along the China-Myanmar border. Malar J. 2018;17: 1–13.
- 21. Thuilliez J, Sissoko MS, Toure OB, Kamate P, Berthelemy J-C, Paris U, et al. "Malaria and Primary education in Mali: a longitudinal study in the village of Donéguébougou." Soc Sci Med. 2010;71: 324–334 pmid:20413198
- 22. Deressa W. Individual and household factors associated with ownership of long-lasting insecticidal nets and malaria infection in south-central Ethiopia: a case-control study. Malar J. 2048;16: 402. pmid:28985734
- 23. Ma C, Claude M, Tsongo Kibendelwa Z, Brooks H, Zheng X, Hawkes M. Is maternal education a social vaccine for childhood malaria infection? A cross-sectional study from war-torn Democratic Republic of Congo. Pathog Glob Health. 2017;111: 98–106. pmid:28220714
- 24. Njau JD, Stephenson R, Menon MP, Kachur SP, Mcfarland DA. Investigating the Important Correlates of Maternal Education and Childhood Malaria Infections. Am J Trop Med Hyg. 2014;91: 509–519. pmid:25002302
- 25.
Bhutta ZA, Lassi ZS, Pariyo G, Huicho L. Global Evidence of Community Health Workers Global Experience of Community Health Workers for Delivery of Health Related Millennium Development Goals: A Systematic Review, Country Case Studies, and Recommendations for Integration into National Health Syst.
- 26. Perry H, Associate S, Zulliger R, Candidate D. HOW EFFECTIVE ARE COMMUNITY HEALTH WORKERS? 2012.
- 27. Saprii L, Richards E, Kokho P, Theobald S. Community health workers in rural India: analysing the opportunities and challenges Accredited Social Health Activists (ASHAs) face in realising their multiple roles. Hum Resour Health. 2015;13: 95. pmid:26646109
- 28. Bhargava Madhavi. Accredited Social Health Activists Epitomize Rural Social Change and Women’s Empowerment in India. 2017; pmid:28406116
- 29. Dhiman RC. YKYSSPS. Altitude, temperature, and malaria vectors in Nainital and Udham Singh Nagar districts of Uttarakhand, India: an evidence-based study Nagar districts of Uttarakhand, India: An evidence-based study. 2013. pmid:24220082
- 30. Tchuinkam T, Nyih-Kong B, Fopa F, Simard F, Antonio-Nkondjio C, Awono-Ambene H-P, et al. Distribution of Plasmodium falciparum gametocytes and malaria-attributable fraction of fever episodes along an altitudinal transect in Western Cameroon. 2011; pmid:25889511
- 31. Okech BA, Gouagna LC, Yan G, Githure JI, Beier JC. Larval habitats of Anopheles gambiae s.s. (Diptera: Culicidae) influences vector competence to Plasmodium falciparum parasites. Malar J. 2007;6: 1–7.
- 32. Gryseels C, Durnez L, Gerrets R, Uk S, Suon S, Set S, et al. Re-imagining malaria: Heterogeneity of human and mosquito behaviour in relation to residual malaria transmission in Cambodia. Malar J. 2015; pmid:25908498
- 33. Erhart A, Thang ND, Van Ky P, Tinh TT, Van Overmeir C, Speybroeck N, et al. Epidemiology of forest malaria in central Vietnam: A large scale cross-sectional survey. Malaria Journal. 2005. pmid:16336671
- 34. Ribera JM, Hausmann-Muela S. The straw that breaks the camel’s back. Redirecting health-seeking behavior studies on malaria and vulnerability. Med Anthropol Q. 2011;25: 103–21. pmid:21495497
- 35. Guerra CA, Snow RW, Hay SI. A global assessment of closed forests, deforestation and malaria risk Europe PMC Funders Group. Ann Trop Med Parasitol. 2006;100: 189–204. pmid:16630376
- 36. Thomson MC, Mason SJ, Phindela T, Connor SJ. Use of rainfall and sea surface temperature monitoring for malaria early warning in Botswana. Am J Trop Med Hyg. 2005;
- 37. Briët OJT, Vounatsou P, Gunawardena DM, Galappaththy GNL, Amerasinghe PH. Temporal correlation between malaria and rainfall in Sri Lanka. Malar J. 2008; pmid:18460205
- 38. Haque U, Hashizume M, Glass GE, Dewan AM, Overgaard HJ, Yamamoto T. The role of climate variability in the spread of malaria in bangladeshi highlands. PLoS One. 2010; pmid:21179555
- 39. Yamana TK, Eltahir EAB. Incorporating the effects of humidity in a mechanistic model of Anopheles gambiae mosquito population dynamics in the Sahel region of Africa. Parasites and Vectors. 2013; pmid:23938022
- 40. Tonnang HEZ, Kangalawe RYM, Yanda PZ. Review Predicting and mapping malaria under climate change scenarios: The potential redistribution of malaria vectors in Africa. Malaria Journal. 2010. pmid:20416059
- 41. Yan G, Githeko AK, Zhou G, Minakawa N, Munga S. Spatial Relationship between Adult Malaria Vector Abundance and Environmental Factors in Western Kenya Highlands. Am J Trop Med Hyg. 2018;77: 29–35.
- 42. Ross CE, Wu C. The Links Between Education and Health. Am Sociol Rev. 2006;
- 43. Degarege A, Fennie K, Degarege D, Id SC, Madhivanan P. Improving socioeconomic status may reduce the burden of malaria in sub Saharan Africa: A systematic review and meta-analysis. 2019; pmid:30677102
- 44. Akinbobola A, Ikiroma IA. Determining Malaria Hotspot Using Climatic Variables and Geospatial Technique in Central Urban Area of Ibadan, Southwest, Nigeria. J Climatol Weather Forecast. 2018;