Bridging the Gap for Subnational Tailoring: Towards a More Accurate Assessment of Malaria Burden

doi:10.21203/rs.3.rs-4808762/v1

Download PDF

Research Article

Bridging the Gap for Subnational Tailoring: Towards a More Accurate Assessment of Malaria Burden

https://doi.org/10.21203/rs.3.rs-4808762/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: Accurate estimates of malaria burden are crucial for allocating resources and designing effective control strategies. However, global reports often underestimate the burden in low- and middle-income countries due to limitations in surveillance systems. This study addresses this gap by providing a detailed analysis of malaria burden and spatio-temporal distribution in Sindh province, Pakistan.

Methods: We utilized data on monthly suspected malaria cases reported from 1,211 primary healthcare facilities across twenty-three districts (excluding Karachi city) for an 11-year period (2012-2022). Facility weighted and population-weighted estimates calculated to account for variations in population size across districts.

Findings: An incidence of 76 per 1000 population per annum suspected malaria cases were reported at primary public healthcare facilities. Pooled estimates of 1.16 million cases occurred annually. Marked heterogeneity observed in malaria burden across districts and high burden districts were identified. The study revealed a distinct seasonal pattern with peaks coinciding with the wet season and post-monsoon period. Malaria persisted heavily in few districts over the last decade.

Conclusions: These findings highlight the substantial malaria burden in Sindh, particularly in specific high-burden districts. Since 60-70% of population seek healthcare from private providers, the reported figures are gross underestimate. Also, the analysis suggest that sub-national analysis and tailoring of strategies are needed. The underestimation by global reports emphasizes the need for country-level and subnational analyses for informed decision-making. By addressing these gaps and refining burden estimates, Pakistan can develop more targeted strategies towards malaria control.

Malaria

high-burden

spatiotemporal distribution

time-series

incidence

Globally, in 2022, there were 249 million reported malaria cases across eighty-five endemic countries, with an incidence rate of fifty-eight per 1,000 people at risk. Africa withstands most of the disease, accounting for 94% (233 million) of cases. Within Africa, Nigeria, the Democratic Republic of Congo, Uganda, and Mozambique contribute half of all cases [1]. However, Pakistan stands out with the largest case number increase from 2021 to 2022 (2.1 million - a five-fold rise), primarily due to devastating floods. This surge has placed Pakistan second in the Eastern Mediterranean region, a position it has never held before [1]. These estimates were weighted by modelled estimates based on parasite index from high burden countries, primarily from Africa [1]. However, this method has limitations due to weak surveillance systems in high-burden, low-and middle-income countries [2]. Improved methods of malaria estimates would help efficiency of malaria control programs.

Pakistan's focus on provincial health governance since the 18th Amendment (2010) [3] necessitates provincial-level information for effective malaria control programs. The national malaria figures though informative, offer limited practical value for program implementation at the provincial or district level. To address this gap, the World Health Organization (WHO) provides sub-national tailoring (SNT) support, aiding in the development of context-specific malaria control strategies. However, reliable provincial estimates are crucial for the success of such SNT efforts [4].

Accurate epidemiological data on malaria burden is essential for all aspects of malaria control, including resource allocation, targeted interventions, monitoring progress, advocacy, and funding [5]. Knowing the most affected areas allows for targeted interventions like insecticide-treated bed nets, indoor residual spraying, and access to rapid diagnostic tests and effective antimalarial treatments. Understanding the geographical distribution of malaria burden helps tailor intervention plans. Regularly monitoring malaria burden allows for tracking progress toward national and international malaria elimination goals. This data helps identify sub-national areas where interventions are successful and areas needing further attention. Finally, reliable data on the burden of malaria strengthens advocacy efforts. It allows policymakers and healthcare officials to secure funding for prevention, control, and potential elimination programs [4–5].

Pakistan lacks a comprehensive understanding of its malaria burden due to the absence of systematic efforts to estimate prevalence. A recent (2023) systematic review and meta-analysis revealed a national pooled prevalence of 23.3%, with variation ranging from 1.68–99.79% [6]. All included studies were cross-sectional, originated from different geographic locations, points in time, seasons, and covered varying age groups. Another national level study reported an average of 265000 cases per annum in Pakistan, clearly a gross underestimate [7]. Another health facility-based study suggested an overall 8.2% prevalence however, reported unexplained variation of prevalence from 4% among pregnant women to 72.3% among non-pregnant women [8]. The above studies highlight the gap in estimates of malaria burden in Pakistan and make the above assessments questionable. We believe that longitudinal data which is distributed according to location, malaria cases at multiple points in time including seasons at same location, may provide a more accurate estimate of the burden of malaria and may also identify context-specific malaria control strategies.

To address this gap, we aimed to determine the burden of malaria according to districts and estimate the population-weighted malaria burden in the province of Sindh, Pakistan using spatio-temporal data for over a decade. We analysed the time trend of malaria burden according to district and determined seasonal pattern and identified malaria density regions using cumulative data over a decade in the province.

Study design

To assess the provincial burden of malaria and estimate population-adjusted prevalence, we employed a longitudinal data distributed across a broad geographic area. The study utilized data from 1211 primary healthcare facilities across twenty-three districts in Sindh province of Pakistan. We analysed monthly reports of suspected malaria (SM) cases reported between 2012 and 2022. To account for population changes, we employed population estimates from the 2017 census, which best approximates the mid-point population for our study period (2012–2022).

Study site

The study site, Sindh province in southern Pakistan, borders the Arabian Sea and features diverse landscapes. Fertile plains along the Indus River, with stagnant water bodies, provide breeding grounds for malaria-transmitting mosquitoes. The riverine belt, with its high density of healthcare facilities, experiences a favourable climate for malaria transmission during the wet season (July-August). Conversely, arid regions face significant temperature fluctuations, impacting mosquito breeding patterns. The temperatures frequently rise above 46°C (115°F) between May and August, and the minimum average temperature of 2°C (36°F) occurs during December and January. The annual rainfall averages about seven inches, falling during July and August, which can influence mosquito breeding patterns and potentially affect malaria seasonality [9]

People's Primary Healthcare Initiative (PPHI) has been managing public sector primary healthcare facilities in Sindh province since 2007. This public-private partnership (PPP) with the Government of Sindh aims to improve healthcare access and quality across the province [10]. PPHI currently operates facilities in twenty-six districts, including three within Karachi. To focus on facilities with established malaria data, we excluded forty-five secondary healthcare facilities offering specialized services and 102 facilities established after 2012 (25 in Karachi and seventy-seven in other districts).

Our analysis focused on 1211 primary healthcare facilities. These facilities, established before 2012, provided consistent monthly malaria data. They comprised diverse types of facilities, including 690 basic health units (BHUs), 384 dispensaries, fifty-five maternal and child health centres (MCHs), and 82 others. These facilities offer essential maternal, newborn, and child health (MNCH) services such as antenatal care, childbirth support, postnatal care, family planning, and childhood immunizations. Of 1211, 76.2% (923) work between 9am to 2:00pm and provide outpatient services for six days, while 23.8% (288) also provide inpatient services and work 24/7 [10].

Participants and suspected malaria definition

This study included suspected malaria (SM) cases reported at healthcare facilities in Sindh province, Pakistan. Data was gathered for all ages and genders.

Healthcare providers used the World Health Organization's (WHO) definition of suspected malaria. This definition classifies an illness as suspected malaria if a healthcare worker believes it could be malaria, typically based on the presence of fever, with or without other symptoms, as established by the national malaria control program. In these cases, patients receive antimalarial treatment without confirmatory testing. According to WHO, suspected cases are "presumed cases" [10].

Statistical analysis

We analysed monthly suspected malaria (SM) cases reported at each healthcare facility. The malaria case was analysed according to health facility and district. To calculate a malaria incidence rate for each facility, we divided the total number of reported cases from 2012–2022 by the corresponding facility's catchment population for the entire 11-year period. Then, district-level average incidence rates of malaria were obtained, by averaging the facility-level rates within each district. Weighted averages for each facility were used. Finally, the district-level average incidence rate was multiplied by the total population of the district (from 2017 census) [11], to estimate the overall malaria burden for that district. While the 2017 census, data provides the most relevant mid-point population estimate for our study period (2012–2022), it is important to acknowledge that population changes might have occurred over this timeframe.

Mapping and visual analysis:

To determine the spatial distribution of malaria, burden each healthcare facility was assigned a geographic location using GPS coordinates. We used ArcMap 10.8 (Esri, Redlands, CA, USA) for spatial data [12].

First, we developed yearly district-level malaria burden maps based on pooled averages from each health facility without population weights. Secondly, we created yearly district level malaria maps for population weighted estimates. We created a point density map to depict the spatial distribution of calculated malaria burden at the district level. Each point on the map represents an estimated burden of 1,000 suspected malaria cases per total population within a district over the study period of 2012–2022. In simpler terms, denser clusters of points indicate areas with a higher burden of malaria relative to the population size.

District-wise distribution of health facilities in Sindh

Figure 2 visually depicts the distribution of healthcare facilities across Sindh province, Pakistan. As the figure shows, a clear pattern emerges in facility placement. The highest concentration of facilities was in the riverine zone, areas bordering the Indus River and its tributaries. This strategic placement aligns well with the fact that population density is also highest in these riverine areas. This strategic distribution ensures greater accessibility to healthcare services for the most populated regions.

Figure 2 further highlights the uniform distribution of Basic Health Units (BHUs) across the province. BHUs are the most basic level of healthcare facility, offering essential services like vaccinations, primary care consultations, and basic diagnostics. Their uniform distribution suggests an attempt to provide a foundational level of healthcare accessibility throughout Sindh.

However, a contrasting pattern is evident in the desert areas, particularly in districts like Tharparkar and Dadu. Figure 2 reveals a sparse distribution of healthcare facilities in these regions. This disparity could be attributed to several factors, such as lower population density, challenging infrastructure development in desert environments, or potential resource limitations.

Burden of suspected malaria cases in the province

Table 1 summarizes the burden of suspected malaria (SM) cases reported across Sindh province from 2012 to 2022. Public sector healthcare facilities in twenty-three districts reported a total of 12.81 million suspected malaria cases during this period. This translates to an average of 1.16 million cases reported annually. On average, each healthcare facility observed approximately 978 suspected cases per year.

The number of healthcare facilities varied across districts, ranging from twenty-two in Thatta to 144 in Khairpur, with an average of 52.6 per district. There was a moderate positive correlation (correlation coefficient: 0.68) between the number of facilities and the total population in each district.

The total number of suspected malaria cases also varied significantly by district, ranging from 189,489 to 1,959,017. The average annual number of cases per district showed a strong positive correlation (correlation coefficient: 0.86) with the number of healthcare facilities. Hyderabad reported the lowest average annual cases (17,226), while Khairpur reported the highest (178,092). The percentage of the total population covered by healthcare facilities in each district ranged from 19% (Hyderabad) to 83% (Khairpur), with an average coverage of 49%. This translates to an average population served by each facility of 12,951.

The overall annual incidence of suspected malaria was seventy-six cases per 1,000 people. This ranged from thirty-four cases per 1,000 in Jamshoro to 161 cases per 1,000 in Thatta. When considering population-weighted estimates, Naushero Feroze district had the highest annual burden (214,655 cases over 11 years), while Jamshoro had the lowest (33,848 cases over 11 years).

Table 1

Burden of suspected malaria cases in the province of Sindh (2012–2022)
Districts	# of health facilities in the district	Total cases (2012–2022)	Average cases per annum	Average cases /year /facility	Population covered by health facilities (% of total population)	Average population covered per facility	Incidence /1000/year	District population (Census 2017)	Population weighted district-wise total incidence per year
Badin	102	1002932	91176	894	1182795 (66)	11596	77	1804958	139135
Dadu	49	463326	42121	860	579960 (37)	11836	73	1550390	112600
Ghotki	47	329570	29961	637	818868 (50)	17423	37	1648708	60323
Hyderabad	40	189489	17226	431	425930 (19)	10648	40	2199928	88974
Jacobabad	36	389221	35384	983	559234 (56)	15534	63	1007009	63715
Jamshoro	47	202395	18400	391	540278 (54)	11495	34	993908	33848
Kamber	51	478897	43536	854	719354 (54)	14105	61	1338035	80979
Kashmore	40	366418	33311	833	527032 (48)	13176	63	1090336	68914
Khairpur	144	1959017	178092	1237	1998086 (83)	13876	89	2405190	214378
Larkana	48	713196	64836	1351	642438 (42)	13384	101	1521786	153581
Matiari	34	227665	20697	609	300207 (39)	8830	69	770040	53088
Mirpurkhas	95	827289	75208	792	1101848 (73)	11598	68	1504440	102688
Naushero Feroze	46	913990	83090	1806	624000 (39)	13565	133	1612047	214655
Sanghar	89	939070	85370	959	1394676 (68)	15671	61	2049873	125475
Shaheed Benazirabad	56	387101	35191	628	449536 (28)	8027	78	1613506	126310
Shikarpur	27	321482	29226	1082	411576 (33)	15244	71	1233760	87608
Sujawal	29	346277	31480	1081	262995 (34)	9069	120	779062	93251
Sukkur	36	291658	26514	737	651679 (44)	18102	41	1488372	60556
Tando Allahyar	46	397334	36121	785	601377 (72)	13073	60	838527	50366
Tando Muhammad Khan	34	326303	29664	872	395083 (58)	11620	75	677098	50838
Tharparkar	49	399376	36307	741	478901 (29)	9773	76	1647036	124867
Thatta	22	553997	50363	2289	312410 (32)	14200	161	982138	158330
Umerkot	44	788571	71688	1629	705292 (66)	16029	102	1073469	109111
Total/Average*	1211	12814574	1164961	978*	*15683555 (49)**	12951	76*	31829616	2373592

Time series and seasonality of malaria

Time series and seasonality of malaria further inform the spatial distribution of the burden observed in Table 1 and Fig. 2. Figure 3 illustrates the yearly pattern of suspected malaria cases across all twenty-three districts. Interestingly, the figure reveals a synchronous pattern, with all districts experiencing peaks and troughs at similar times. This suggests a potential influence of province-wide factors, such as climatic conditions, on malaria transmission.

The year 2022 stands out with a significant surge in cases across all districts due to the super floods that struck Sindh province. Three districts namely Larkana, Thatta and Kamber had the highest the number of cases of malaria during the super floods of 2022. After the 2012 flood two districts namely Khairpur and Naushero Feroze reported the highest cases of malaria. Notably, increases in cases also reported during 2012, 2015, and 2019, all years coinciding with major flooding events. These observations highlight the established link between floods and increased malaria transmission, due to the creation of stagnant water bodies that serve as mosquito breeding grounds.

The time series data also reveals a low point in suspected cases around 2017 and 2021. The information from Table 1 and Fig. 2 is useful to categorize districts into low-burden and high-burden groups.

Figure 4 depicts the distinct seasonal pattern of suspected malaria cases throughout the year. This figure complements the time series data in Fig. 3, which highlighted the impact of major flooding events. Here, we see a clear trend of increased cases during the wet season, with a significant decline observed in drier months. This seasonal pattern aligns with the known biology of the mosquitos that transmit malaria, as they thrive in stagnant water bodies created during periods of heavy rainfall.

The graph in Fig. 4 reveals two distinct peaks in malaria cases. One peak occurs in March, potentially coinciding with the tail end of the winter season and the start of pre-monsoon rains. The second peak occurs during the post-monsoon season, due to the presence of stagnant water bodies remaining after the heavy rainfall. Notably, the months of July to December, corresponding to the wet season and post-monsoon period, experience two times higher malaria incidence compared to the rest of the year.

Spatial distribution of malaria by district

Building upon the insights gained from time series data (Fig. 3), Fig. 5 offers a detailed perspective on the spatial distribution of the annual suspected malaria burden across districts in Sindh province. This figure depicts the malaria burden for each district throughout the entire study period (2012–2022). By aggregating suspected cases reported at all healthcare facilities within each district, the figure allows us to visualize the total annual cases across the twenty-three districts.

A critical observation from Fig. 5 is the persistence of malaria cases in districts like Khairpur, Larkana, Sanghar, Thatta, Mirpur Khas, and Badin. This finding aligns with the identification of high-burden districts based on overall case numbers (Table 1).

Figure 6 delves deeper into the spatial distribution of the malaria burden by incorporating population-weighted estimates. While the previous figures focused on raw case numbers reported by healthcare facilities, population-weighted estimates account for the varying population sizes across districts. This adjustment provides a more accurate picture of the burden relative to the population at risk.

The figure reveals a good correlation between the actual estimates based on health facility and catchment population data, and the population-weighted estimates. This suggests that the spatial distribution patterns observed in previous figures (e.g., higher burden in districts like Khairpur) are valid representations of the true burden when population size is considered. Persistently high burden districts should be prioritized for targeted interventions due to the significant public health impact of malaria in these regions.

The density map shown in Fig. 7, depicts the cumulative spatial distribution of suspected malaria burden in Sindh, Pakistan, over a period of 11 years. Our spatial analysis revealed a concentration of suspected malaria cases along the Indus River corridor, extending from the Jacobabad/Kashmore border (where the river enters Sindh from Punjab) to its discharge point at the Arabian Sea near Thatta. The districts bordering the river, including Larkana, Khairpur, Naushero Feroze, Thatta, and Hyderabad, exhibited the highest burden of malaria persistently in the last decade. This suggests a potential need for targeted interventions in these specific regions.

This study provided the most recent population-based estimates of malaria burden in the second most populated province of Pakistan. The estimates are based on monthly reported suspected cases over 11 years from 1211 primary healthcare facilities. The large malaria datasets encompassing twenty-three districts and 31.8 million population over 11 years. On average 1.16 million cases per year were reported in twenty-three districts alone with a cumulative incidence of 12.81 million between 2012–2022. The average annual cases per facility was approximately one thousand (978) (see Table 1). Also, there is high variability of malaria cases across districts (189,489–1,959,017). This study provided insight in subnational analysis of malaria for decision-making, in particular district level estimates for strategic malaria control in the province. To best of our knowledge burden of malaria based on estimates of more than a decade was previously not available in the region.

Comparatively, the global report of malaria (2021) estimated 4.8 to 6.4 million annual cases, both suspected (presumed) and confirmed in Pakistan in last decade (2010–2021) [2]. The global malaria report of 2022 estimated that around 217 million people in Pakistan are at moderate risk of malaria, and sixty-three million are at high risk where approximately 0.47 million malaria cases and eight hundred deaths were reported [2]. Our study estimated the incidence of suspected malaria cases (1.16 malaria cases annually) in approximately 15% of the population of Pakistan, which lives in twenty-three districts of the province of Sindh excluding the population of Karachi. Thus, approximately eight million malaria cases (both presumed and confirmed) occurred annually in Pakistan during the last decade and these are reported only to public sector healthcare facilities.

The current study estimated the incidence of seventy-six suspected malaria cases per one thousand population per year and the range observed across districts between 34–161 cases per 1,000 people in public sector healthcare facilities only (Table 1). It is important to note that in Pakistan about 60–70% of the population use private health sector [13] in Pakistan therefore the above tally is an underestimate. Considering that at least three times the cases are seeking healthcare from private health sector, the incidence rate might go above two hundred per one thousand per annum and the overall incidence may cross twenty million cases of malaria annually. Furthermore, the study did not include secondary and tertiary care hospitals in these district which may further underestimate the overall burden. The global malaria report estimated the burden of malaria on varied quality of data. In areas with limited surveillance data, sub-Saharan African and low-income countries, malaria case estimates are generated using modelled parasite prevalence and geographic information. In contrast, countries with robust surveillance systems can directly use reported case data, adjusted for factors like healthcare-seeking behaviour and population coverage. We contend that population-weighted estimates, derived from actual data, are more dependable than model-based estimates reliant on weaker surveillance systems. Our study employed data from over 1200 healthcare facilities collected over eleven years, a previously unprecedented dataset for Pakistan.

Moreover, studies from Pakistan estimated that 15–25% of suspected malaria patients have confirmed malaria [14–15]. The global malaria reported that the percentage of confirmed cases among suspected ranged between 18.1–49.7%. Therefore, from the above estimates of seventy-six suspected cases, we assume that the incidence of confirmed malaria cases in Sindh province ranges between 11–18 cases per one thousand per year, which are reported to public healthcare sector facilities only. If we include an overall estimate including those who might be reporting to private sector health facilities, then approximately 30–50 confirmed cases occur per one thousand population per year. It seems that 1 in 5 presumed cases are diagnosed as confirmed cases of malaria.

The study also revealed that the average coverage of the population by public sector healthcare facilities was 49% in the province of Sindh, ranging between 19% − 83% across twenty-three districts (Table 1). There was moderate correlation between the number of healthcare facilities and total population per district (correlation coefficient: 0.68) suggesting that the distribution of health facilities according to population size can be enhanced to improve coverage. However, given the meagre resources for health sector, it provides a good population representation of health facilities for primary healthcare facilities across districts in the province [16].

There is a large disparity in the malaria burden in the province. The study highlights the high burden districts where targeted intervention can be organized to control malaria. This also highlights the disparity between districts based on population-weighted estimates (Naushero Feroze highest, Jamshoro lowest). Spatial distribution clearly identifies river Indus as a risk factor for malaria. The districts closer to the river have one of the highest burdens of malaria in the province targeted for intervention [17–18].

The study provided a trend of suspected malaria cases according to seasons (monthly) and spatial distribution across districts identifying high-burden districts/hotspots. Malaria exhibits a pronounced seasonal pattern, and two peaks were consistently visible: one in March and the other is a higher peak post-monsoon. To our knowledge, this information is novel at least for the region. The wet season witnessed a surge in cases, while drier months experienced a significant decline. This trend can be attributed to the ideal breeding conditions for mosquitoes, the primary vectors transmitting malaria are during periods of high rainfall and humidity. These peaks highlight the periods when factors like rainfall and humidity create an environment conducive to amplified malaria transmission [19–21]. It is important to note that our study used suspected cases for the seasonality trend of malaria. However, results are consistent with other studies and robust in indicating malaria seasonality. Two distinct peaks inform decision makers for organizing control strategies before such surge which may include targeted bed net and chemoprophylaxis. This strong seasonal pattern underscores the importance of incorporating seasonal malaria chemoprophylaxis campaigns as part of malaria control strategies in Sindh province. These campaigns involve the targeted administration of antimalarial medication during peak transmission periods, which can significantly reduce malaria burden [22]. Additionally, the presence of water around the riverbanks during periods of overflow and its stagnancy in bordering areas may contribute to the high burden of malaria in these areas. Addressing such factors could be crucial in implementing effective malaria control measures and reducing the disease burden in these regions. Although a significant contributing factor to the high case load in these locations is the stagnant water, the province's green belts also serve as breeding grounds for mosquitoes [23–24]

While considering the above estimates following limitations may be considered. The main methods of malaria surveillance in Pakistan are passive case detection at hospitals and rapid diagnosis and treatment at laboratories. However, passive case detection of malaria for burden estimation is a uniform method globally [1]. This study used suspected cases instead of confirmed cases of malaria, which is an inherent limitation of the health system, as laboratory services in primary healthcare facilities are not available. Only a handful of primary healthcare facilities had referral laboratories and that data was also not organized. Secondly, we only utilized public sector primary healthcare malaria data for estimation as private sector healthcare facilities do not have systematic data collection methods in Pakistan. Furthermore, secondary and tertiary care hospital data was not included which may further underestimate the overall burden of malaria in studied districts.

We believe that implementing standardized methods for collecting and reporting malaria cases across the country ensures consistency and reliability of data and help in informed decision-making. In future expanding access to rapid diagnostic tests allows for faster and more accurate diagnosis, leading to quicker treatment and better disease management. Strengthening surveillance systems, especially in high-risk (burden) areas, helps identify outbreaks. By working towards better alignment and estimation of the malaria burden, a country can develop more effective strategies for controlling and eliminating this preventable disease.

Furthermore, Larkana, Thatta and Kamber districts have the highest case load. These districts had undergone free malaria testing services through an NGO, which might have encouraged healthcare providers to identify and evaluate more suspects in these districts.

The population at risk of malaria in Pakistan is not well defined. The funding agencies (Global Fund to Fight AIDS, Tuberculosis and Malaria started to support the National Malaria Control Program (NMCP) in 2007) would find this information useful for allocation efficiency.

Furthermore, one of the implications of this study that it has identified high malaria burden districts with low health facility coverage for targeted interventions. Identified districts with targeted interventions may improve the efficiency in prevention, diagnosis, and treatment of malaria.

This study provides a comprehensive analysis of the malaria burden and distribution in Sindh province, Pakistan, utilizing a large dataset spanning 11 years (2012–2022) and encompassing twenty-three districts. By leveraging monthly suspected malaria cases reported from 1,211 healthcare facilities, we were able to estimate the district-wise burden using population-weighted methods. This approach, potentially the first of its kind in the region, offers a more accurate picture of the malaria situation in Sindh.

Our findings reveal a substantial burden of suspected malaria cases across Sindh, with an average of 1.16 million cases reported annually. Importantly, the burden varies significantly between districts, highlighting the need for targeted interventions. Population-weighted estimates identified Naushero Feroze and other districts bordering the Indus River corridor as particularly high-burden areas.

The analysis also revealed a distinct seasonal pattern with two peaks: one in March and another during the post-monsoon period. This finding aligns with the known biology of malaria vectors and underscores the importance of incorporating seasonal malaria chemoprophylaxis (SMC) campaigns as part of the control strategy in Sindh.

While this study relied on suspected malaria cases due to limitations in diagnostic accessibility, it provides valuable insights for informing public health decisions. Our findings highlight the critical need for:

Strengthened surveillance systems: Implementing standardized data collection and reporting methods across healthcare facilities would enhance data quality and facilitate informed decision-making.
Expanded access to rapid diagnostic tests: This would enable faster and more accurate diagnosis, leading to improved treatment outcomes and better disease management.
Targeted interventions in high-burden areas: Focusing resources and control efforts on districts like Naushero Feroze and the Indus River corridor can significantly improve the efficiency of malaria prevention, diagnosis, and treatment in Sindh.

By addressing these crucial aspects, policymakers and public health officials can develop a more effective and targeted approach to controlling malaria in Sindh province, Pakistan.

BHU: Basic Health Units

GIS: Geographic Information System

MCH: Mother and Child Health

MNCH: Maternal, Newborn, and Child Health

NMCP: National Malaria Control Program

PPHI: People’s Primary Healthcare Initiative

SMC: Seasonal Malaria Chemoprophylaxis

SM: Suspected Malaria

SNT: Sub-National Tailoring

WHO: World Health Organization

Ethics approval and consent to participate: Not Applicable. No human interaction occurred for collection of data.

Consent for publication: Not Applicable.

Availability of data and materials: Materials available on request.

Competing interests: The authors declare no competing interests.

Funding: This was a non funded research.

Authors' contributions: ZF, ZA, BAU conceptualized the study. NB, WZJ and ZA compiled the initial data. NB, ZA, RHR, JAJ provided insight into operational details and interpretation of information. SR, BAU and MA conducted analysis and developed maps for spatial distribution. ZF, BAU, WZJ, ZA wrote the draft manuscript. All authors critically reviewed and approved the final manuscript.

World Health Organization. Global malaria report 2023. Geneva: World Health Organization; 2023.
World Health Organization. Global malaria report 2022. Geneva: World Health Organization; 2022.
Pakistan. 18th Amendment to the Constitution of Pakistan. Islamabad: Government of Pakistan; 2010.
World Health Organization. Global technical strategy for malaria 2016-2030 (2021 update). Geneva: World Health Organization; 2021.
Scott N, Hussain SA, Martin-Hughes R. et al. Maximizing the impact of malaria funding through allocative efficiency: using the right interventions in the right locations. Malar J16, 368 (2017). https://doi.org/10.1186/s12936-017-2019-1.
Khan, M I, Qureshi, H., Bae, S. J., Khattak, A. A., Anwar, M. S., Ahmad, S., Hassan, F., & Ahmad, S. (2023). Malaria prevalence in Pakistan: A systematic review and meta-analysis (2006-2021). Heliyon, 9(4), e15373.
Umer MF, Zofeen S, Majeed A, Hu W, Qi X, Zhuang G. Spatiotemporal Clustering Analysis of Malaria Infection in Pakistan. Int J Environ Res Public Health. 2018 Jun 7;15(6):1202. doi: 10.3390/ijerph15061202. PMID: 29880778; PMCID: PMC6025434.
Qureshi H, Khan MI, Ahmad A, Awan UA, Khattak AA, Khan AA and Sun Y (2021) Passive Surveillance of Malaria in Pregnant Women, Non-pregnant Women and Children Under 5 Years of Age in Bannu District, Khyber Pakhtunkhwa Pakistan. Front. Med. 8:751456. doi: 10.3389/fmed.2021.751456.
Sindh Weather Portal (SWP). Climate of Sindh | Sindh Weather Portal (SWP) [Internet]. Karachi, Pakistan: WordPress.com; [cited 2024 Jul 24]. Available from: https://sindhweatherportal.wordpress.com/climate-of-sindh/
People's Primary Healthcare Initiatives (PPHI). PPHI Sindh [Internet]. Karachi: PPHI; [cited 2024 Jul 24]. Available from: http://pphisindh.org/
Pakistan Bureau of Statistics. Pakistan Population and Housing Census 2017. Islamabad: Author; 2017.
Esri, Redlands, CA, USA. Retrieved from https://www.esri.com/
National Institute of Population Studies (NIPS) [Pakistan] and ICF. 2019. Pakistan Demographic and Health Survey 2017-18. Islamabad, Pakistan, and Rockville, Maryland, USA: NIPS and ICF
Qureshi H, Khan MI, Ambachew H, Pan HF, Ye DQ. Baseline survey for malaria prevalence in Khyber Pakhtunkhwa Province, Pakistan. East Mediterr Health J. 2020 Apr 16;26(4):453-460. doi: 10.26719/emhj.19.015. PMID: 32338364.
Khan W, Rahman AU, Khan W, Shafiq S, Ihsan H, Khan K. Malaria prevalence in Malakand district, the north western region of Pakistan. J Pak Med Assoc. 2019 Jul;69(7):946-949. PMID: 31308559.
Health Department Sindh. Health Department Sindh [Internet]. Karachi, Pakistan: Health Department Sindh; [cited 2024 Jul 24]. Available from: [Insert website URL].
Dieng, S., Adebayo-Ojo, T.C., Kruger, T. et al. Geo-epidemiology of malaria incidence in the Vhembe District to guide targeted elimination strategies, South-Africa, 2015–2018: a local resurgence. Sci Rep 13, 11049 (2023). https://doi.org/10.1038/s41598-023-38147-0.
Oesterholt MJ, Bousema JT, Mwerinde OK, Harris C, Lushino P, Masokoto A, Mwerinde H, Mosha FW, Drakeley CJ. Spatial and temporal variation in malaria transmission in a low endemicity area in northern Tanzania. Malaria Journal. 2006 Dec;5:1-7.
Stadler, E., Cromer, D., Ogunlade, S. et al. Evidence for exposure dependent carriage of malaria parasites across the dry season: modelling analysis of longitudinal data. Malar J 22, 42 (2023). https://doi.org/10.1186/s12936-023-04461-1.
Liu Z, Wang S, Zhang Y, Xiang J, Tong MX, Gao Q, Zhang Y, Sun S, Liu Q, Jiang B, Bi P. Effect of temperature and its interactions with relative humidity and rainfall on malaria in a temperate city Suzhou, China. Environmental Science and Pollution Research. 2021 Apr;28:16830-42.
Matthew OJ. Investigating climate suitability conditions for malaria transmission and impacts of climate variability on mosquito survival in the humid tropical region: a case study of Obafemi Awolowo University Campus, Ile-Ife, south-western Nigeria. International journal of biometeorology. 2020 Mar;64(3):355-65.
ACCESS-SMC Partnership. Effectiveness of seasonal malaria chemoprevention at scale in west and central Africa: an observational study. Lancet. 2020 Dec 5;396(10265):1829-1840. doi: 10.1016/S0140-6736(20)32227-3. PMID: 33278936; PMCID: PMC7718580.
Ali SA, Ahmad A. Mapping of mosquito-borne diseases in Kolkata Municipal Corporation using GIS and AHP based decision making approach. Spatial Information Research. 2019 Jun 1;27(3):351-72.
Sowilem MM, El-Zeiny AM, Mohamed ES. Mosquito larval species and geographical information system (GIS) mapping of environmental vulnerable areas, Dakhla Oasis, Egypt. International Journal of Environment and Climate Change. 2019 Apr 1;9(1):17-28.

No competing interests reported.

Download PDF

Reviewers invited by journal
01 Aug, 2024
Editor assigned by journal
27 Jul, 2024
Submission checks completed at journal
27 Jul, 2024
First submitted to journal
26 Jul, 2024

You are reading this latest preprint version

Bridging the Gap for Subnational Tailoring: Towards a More Accurate Assessment of Malaria Burden

Status:

Version 1

Abstract

Figures

Background

Methods

Mapping and visual analysis:

Results

District-wise distribution of health facilities in Sindh

Burden of suspected malaria cases in the province

Time series and seasonality of malaria

Spatial distribution of malaria by district

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1