Determinants of locally acquired malaria infections in Zanzibar: a cross-sectional study

doi:10.21203/rs.3.rs-4887684/v1

Download PDF

Research Article

Determinants of locally acquired malaria infections in Zanzibar: a cross-sectional study

https://doi.org/10.21203/rs.3.rs-4887684/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Zanzibar has made significant progress in malaria control reaching a population prevalence of around 1% with program-led interventions. Nevertheless, malaria infections persist in people with no recent travel history implying locally acquired infections. Understanding risk factors of local transmission is crucial to refine elimination strategies.

Methods

From May 2017 to October 2019, a rolling cross-sectional survey was conducted in five Zanzibar districts, linked to routine malaria surveillance procedures. The study involved testing all individuals living in households of clinical malaria cases (index cases) routinely detected at a health facility, and a sample of neighboring households using malaria rapid diagnostic tests and qPCR. Information on socio-demographic and household characteristics, recent travel and bed net use were collected during interviews.

Results

Among 17,891 surveyed individuals, 15,151 (85%) had not travelled outside Zanzibar in the last sixty days. Of these, 7286 were tested for malaria by qPCR and 2.6% [95% confidence interval (CI): 2–3%] tested positive. Members of index households were more likely to be infected (adjusted odds ratio [aOR ] = 14.9, 95% CI 9.7–23.0) than neighbours, as were individuals aged 5–15 and 16–25 years compared to older age groups (aOR = 4.0, 95% CI 2.5–6.5 and aOR = 2.0, 95% CI 1.3–3.1, respectively). Infections were more likely in Unguja residents (aOR = 1.3 95% CI 0.9–2.0), in people living in houses with no windows (aOR = 2.1, 95% CI 1.4–3.1), in those with no bed net (aOR = 2.3, 95% CI 1.4–3.1), those going to sleep early and in people living in areas with a higher local index of wetness.

Conclusion

Combatting locally acquired malaria in Zanzibar requires interventions and strategies that promote uptake of existing interventions such as bed nets, housing improvements, and the identification and targeting of individuals at higher risk of infection.

Malaria

residual transmission

Zanzibar

Tanzania

elimination

In Zanzibar, an archipelago located off the eastern coast of Tanzania, major progress in malaria control has been made over the past decades, and the malaria parasite prevalence in the community reached historically low levels of around 1% as of 2011 [1]. This achievement appears to be due to a combination of program-led interventions. They included vector control with Long-Lasting Insecticidal Nets (LLINs) and Indoor Residual Spraying (IRS), wide availability of malaria Rapid Diagnostic Tests (RDT) and efficacious Artemisinin-based Combination Therapy [2, 3]. A near real-time electronic surveillance system notifies individual malaria cases diagnosed at health facilities to the Zanzibar Malaria Elimination Programme (ZAMEP), which triggers household investigations in search for additional cases [4]. The system classifies cases into imported or locally acquired infections based on self-reported recent travel outside Zanzibar [4].

In low-transmission settings, asymptomatic carriers of malaria parasites play a crucial role in sustaining malaria transmission [5]. Transmission is dependent on the receptivity of a geographical area (as determined by the abundance of anopheline vectors and favourable ecological and climatic conditions), and on factors that put people at risk of exposure to the mosquitoes [6]. Heterogeneity and changes in temperature, rainfall, and humidity influence the variability of malaria transmission in time and space [7].

Zanzibar had previously set out on an ambitious comprehensive strategy aimed at achieving malaria elimination by the year 2023 [8]. Yet recently, a resurgence of cases was observed [9, 10] and the realization of elimination seems to have become considerably less likely in the near future. The importation of parasites from mainland Tanzania to Zanzibar has been documented as a challenge and source of many infections detected during reactive case detection [6, 11–13]. However, local malaria transmission also still persists in Zanzibar. Routine surveillance data and modelling studies suggest that a majority of infections in Zanzibar are indigenous in spite of frequent reported importation [13]. Evidence of local transmission is provided, for example, by infections in people without recent travel outside of Zanzibar [12] and the identification of sporozoite-positive mosquitoes [14].

On the northern island of Pemba, where fewer imported malaria cases are found than on Unguja, indigenous cases appear more isolated and possibly more amenable to local elimination efforts. The parasite population on Unguja appears to be more admixed (most likely due to more frequent importation from mainland Tanzania) and thus the chance of reintroduction of infections is higher [11]. Hence overall, both imported infections and local malaria transmission hinder the achievement of the local elimination goal [13, 15].

In order to implement suitable interventions to address local transmission in Zanzibar, it is imperative to understand drivers of local transmission. Previous investigations found that infections are clustered in time and space [11]. Determinants of transmission in local foci in Zanzibar may be related to human behaviour – particularly outdoor activities that increase the risk of exposure to vector mosquitoes [16], socioeconomic factors [17], and the environmental context [18]. Yet to date, investigations into specific risk factors of locally acquired infections in Zanzibar are limited in spite of their importance for targeting interventions.

This study investigated risk factors associated with malaria infections in individuals who had no recent travel history outside of Zanzibar to enhance our understanding of local malaria transmission. This information aims to guide tailored interventions to curtail local transmission and progress towards zero indigenous cases.

Study setting

This study was conducted in Zanzibar (United Republic of Tanzania), in two districts of Unguja island, namely Magharibi (West) and Kusini (South); and in three districts of Pemba, namely Micheweni, Chake Chake, and Mkoani. The current population of the Zanzibar archipelago is approximately 1.8 million [19]. Unguja, being more developed and densely populated, serves as the primary hub for travel between Zanzibar, mainland Tanzania and across the globe. Zanzibar, being politically semi-autonomous, has its own Ministry of Health, which oversees the operations of ZAMEP.

Since 2012, ZAMEP has been implementing a reactive case detection (RACD) system. This system involves the prompt notification of clinical cases diagnosed in health facilities through an electronic surveillance system and the active search for additional infections in their households by dedicated district malaria surveillance officers (DMSO) [4].

The malaria case incidence in Zanzibar distinctly follows a seasonal pattern that aligns with the bimodal rainfall on the islands: the two rainy seasons occur from March to June and from October to November [4, 20]. The islands have rich extensive vegetation covers which include mangroves, wild date palms (Phamix reclinata), scrub bush, and tropical forest [21].

Survey procedures

A rolling cross-sectional survey was carried out in connection with routine RACD activities executed by ZAMEP. The study included households of clinical malaria cases (index cases) diagnosed at health facilities, as well as four additional neighbouring households and five households located along a 200-meter transect leading away from the index household in a random direction. All individuals aged three months and older residing in the selected households were eligible to participate in the study, as described in detail elsewhere [20].

The data collection took place from May 2017 to October 2018 by local field research assistants who were trained for this purpose. The research assistants accompanied DMSOs during their routine visits to households, usually during their first visit of the day. They carried out the collection of study-specific data and collected blood samples once the DMSOs had finished their routine duties.

The information collected by the field team included household information on sociodemographic background, water and sanitation, dwelling structure, and geographical coordinates of the household. Additionally, information pertaining to household members’ behaviours relevant with regard to exposure to mosquito bites, such as timing of indoor and outdoor activities, as well as ownership and utilization of mosquito nets were recorded.

A blood sample was taken from a finger prick from every person in the household who was present and consenting. An RDT (SD BIOLINE Malaria Ag P.f/pan, Standard Diagnostics Inc., Republic of South Korea) was used to diagnose malaria according to the manufacturer's instructions. Individuals who tested positive for malaria by RDT were given free antimalarial treatment by the DMSOs as per Zanzibar’s standard treatment guidelines. A dried blood spot was collected on filter paper for the detection of Plasmodium falciparum DNA using quantitative polymerase chain reaction (qPCR), as described in detail elsewhere [20]. Due to resource constraints, not all dried blood spot samples collected during the survey could be analysed by qPCR. Samples from household clusters with a large number of individual samples from a complete set of index and neighbouring/transect households were prioritised.

Climatic datasets

A shape file of 1023 point features was generated using unique latitude and longitude coordinates of surveyed houses. The shape file was used to extract climatological variables from TerraClimate in Google Earth Engine using a Python script executed in a product from Google Research - Colab. Terraclimate is a gridded array of global data that has a high spatial resolution of 1/24° (~ 4 km in the equator) and provides monthly records spanning from 1958 to current. This dataset is generated by combining climatological normals from a set of global climate layers - WorldClim with time-varying data from other resources, mainly Climatic Research Unit Timeseries (CRU TS) Ts4.0 and The Japanese 55-year Reanalysis (JRA-55) [22]. Hence, monthly gridded datasets of maximum and minimum temperature (°C), precipitation (mm), vapour pressure (kPa), and soil moisture (mm) were retrieved two months prior for the survey period 2017 and 2018. The monthly mean temperature was calculated from the maximum and minimum temperature. In addition, relative humidity was computed, using Tetens empirical formula [23] for saturation vapour pressure, applied in the following equation:

RH = Ea/ (0.6108 * exp((17.27 * T) / (T + 237.3)))

Where: RH = Relative Humidity (%), T = mean temperature (degree C), Ea = actual vapour pressure (kPa). Then, averages of the climatic variables for the survey period were calculated and integrated back with the household members file.

Temperature and precipitation data were used to estimate the Palmer Drought Severity Index (PDSI) for the 1023 points. The PDSI is a standardized index that generally ranges from − 10 (extremely dry) to + 10 (extremely wet), representing relative dryness.

Participants, data management and analysis

The primary analyses included individuals who reported that they had not travelled outside of Zanzibar in the 60 days prior to the survey. The variables pertaining to socio-demographics and malaria infection were summarized as frequencies and proportions. Logistic regression was used to assess the association between qPCR positivity and specific predictors with household identifier as random effect to account for clustering at household level. For bivariate analyses, predictors were selected based on their plausible relevance to malaria infection. For the multivariable logistic regression model. We selected covariates based on plausibility, although in some cases, a higher cut-off p-value of < 0.2 was also considered. In the final model, covariates with a p-value p < 0.05 were considered as statistically significantly associated with the malaria infection. Season was included as possible confounding variable to the multivariable logistic regression model.

A socioeconomic status (SES) index was created by using principal component analysis (PCA) of household characteristics, namely the source of drinking water, the energy sources used for cooking and lighting, the type of toilet and flooring, and ownership of a fridge and cell phone, as described in detail elsewhere [24, 25]. Based on the wealth index, households were divided into wealth quintiles.

Net use was captured using Likert scales, with categories: 1; "not often" (merging "seldom," "sometimes," and "often"), 2; "always," and 3; "never"

STATA version 16 (StataCorp, College Station TX, USA) was used for data management and analysis.

Study population

The survey included 3,473 households with 17,891 individuals, 15,151 (85%) of whom reported no recent travel outside Zanzibar. A qPCR result was available for 7,286 (48%) of those who had not travelled (Additional file: Figure S1). About half of these individuals were from Unguja (51%), mostly from Magharibi district (34%), the majority (88%) from a neighbour or transect household. More than half of the participants (54%) were older than 26 years (median age 16 years [interquartile range (IQR): 7–30)] (Table 1, Additional file: Table S2 and Table S3). Only 3% of individuals had a wage job. (Table 1).

Most respondents (88%) lived in households that owned at least one bed net and 81% reported having used a net the night before the survey and 70% reported “always” using a net. More than half of the participants (59%) reported going to sleep between 20:00h and 22:00h (Additional file: Table S2).

Table 1

Characteristics of study participants with a qPCR result available
Variables	n (%)
Sex
Male	3,020 (41)
Female	4266 (59)
Age group (years)
Less than 5	547 (8)
5–15	2150 (30)
16–25	646 (9)
26 +	3943 (54)
Occupation
Enterpreneur	1706 (23)
Not employed	2912 (40)
Student	2421 (33)
Wage Job	239 (3)
Missing	8 (0)
Household type
Index households members	889 (12)
Other households members	6397 (88)
Wealth quintile
Lowest	1472 (20)
2nd	1538 (21)
Middle	1395 (19)
4th	1718 (24)
Highest	1163 (16)
District of residence
Magharibi (Unguja)	2473 (34)
Kusini (Unguja)	1227 (17)
Micheweni (Pemba)	1338 (18)
Chakechake (Pemba)	1280 (18)
Mkoani (Pemba)	968 (13)
Island of residence
Pemba	3586 (49)
Unguja	3700 (51)
Total	7286 (100)

Malaria infection

Of the 7,286 survey participants who had not recently travelled and had an available qPCR result, 2.6% [95% confidence interval (CI): 2.0–3.0%] tested positive for malaria by qPCR; of those with an RDT test result available (N = 10,689), 0.6% (95% CI 0.4–0.8) had a positive RDT. Of 123 qPCR-detected infections, 71.5% (95% CI 62.9–78.9) were not detected by RDT.

Individual characteristics

Infection prevalence by qPCR did not significantly differ between male and female participants, or reported occupations (Table 2) but between age groups (P < 0.001), with the highest prevalence in participants aged 5–15 years (4.4%, 95% CI 3.5–5.6%) and 16–25 years (3.4%, 95% CI 2.6–4.5%). Mosquito net use the previous night was reported by 81% of study participants and regular use (“always”) by 70%. Regular net use was lowest in male survey participants aged 16–25 years (56.5%, 95% CI 52.0–61.0; N = 467) (Figure S3). Malaria prevalence was higher in participants never using a net (4.0%; 95% CI 3.2–4.9) compared to those always using a net (2.0; 95% CI 1.6–2.4). Prevalence was also higher in individuals who reported going to sleep between 18:00 and 20:00 (27% of participants) compared to those who reported to sleep later (Table 2 and Additional file: Table S2).

Table 2

Univariable regression model with odds ratios for individual-level factors associated with malaria infection among individuals with no recent travel outside Zanzibar, detected by qPCR
Variables	N = 7286		Univariable model¹
	Negative	Positive	OR	95% CI	p-value
Socio-demographic
Gender
Male	2938 (97)	82 (3)	ref	-
Female	4160 (98)	106 (2)	1.0	0.73–1.43	0.884
Age group
Less than 5 years	538 (98)	9 (2)	0.9	0.46–2.10	< 0.001
5–15	1383 (96)	64 (4)	2.9	1.99–4.46
16–25	1303 (97)	46 (3)	2.0	1.31–3.13
26 +	3874 (98)	69 (2)	ref	-
Occupation
Entrepreneur	1650 (97)	56 (3)	ref	-	0.439
Not employed	2844 (98)	68 (2)	0.7	0.47–1.08
Student	2364(98)	57 (2)	0.7	0.27–2.05
Wage Job	233 (97)	6 (3)	0.8	0.49–1.18
Human behavior
Net use (n = 7271)
Yes	5741 (98)	139 (2)	ref	-	0.038
No	1343 (97)	48 (3)	1.6	1.02–2.48
Net use (how often)
Always	5007 (98)	101 (2)	ref	-	< 0.001
Not often	207 (96)	9 (4)	2.0	0.86–4.85
Never	1870 (96)	77 (4)	2.2	1.44–3.21
Missing	14 (93)	1 (7)	1.5	0.12–19.51
Indoor time (n = 7264)
Early evening (18-20hrs)	3537 (97)	103 (3)	0.9	0.56–1.44
Early night (20-22hrs)	2384 (98)	48 (2)	0.6	0.35–1.02
More than 22hrs	1158 (97)	34 (3)	ref	-	0.088
Sleeping time (n = 7228)
Early evening (18-20hrs)	1862 (96)	72 (4)	1.5	0.82–2.57	0.013
Early night (20-22hrs)	4244 (98)	90 (2)	0.8	0.48–1.42
More than 22hrs	937 (98)	23 (2)	ref	-
¹Models include random effect to account for clustering at household level

Household characteristics

Malaria prevalence was highest among index household members (12.4%, 95% CI 10.4–14.7) and higher in households in which index cases had travelled (21.1%, 95% CI 7.6–46.5; N = 19), though the difference did not reach statistical significance (OR = 2.1, 95% CI 0.6–8.0, P = 0.267). Prevalence did not differ significantly between participants in Unguja and Pemba (Table 3).

Several housing characteristics that reflect either a more rudimentary house design, or a possibly lower socio-economic status, were associated with a higher prevalence among the household members (Table 3). This includes living in houses covered with natural roofing materials (e.g. thatched roofs), in houses with windows that cannot be closed, in households with wells or boreholes as drinking water source (28%) and in households without toilet facilities (20%) (all P < 0.05). Overall, prevalence was higher in individuals living in a “low quality house” (4% of study participants; prevalence 6.7%; 95% CI 4.3–10.1%) (Table 3 and Additional file: Table S1).

The majority of participants (88%) lives in households that owned at least one bed net. At the same time, only 38% of the study participants reported that their house had been sprayed with insecticide (IRS) within twelve months prior to the survey; 143 individuals provided no information regarding their houses being sprayed. Participants in households that did not own a net were significantly more likely to be infected with malaria (OR = 2.5, 95% CI 1.45–4.16) but reported IRS spraying was not associated with malaria infection (Table 3 and Additional file: Table S2).

Table 3

Univariable regression model with odds ratios for household-level factors associated with malaria infection among individuals with no recent travel outside Zanzibar, detected by qPCR
Variables	N = 7286		Univariable model¹
	Negative	Positive	OR	95% CI	p-value
Wealth quintile
Lowest	1416 (96)	56 (4)	1.9	0.98–3.84	0.113
2nd	1500 (98)	38 (2)	1.2	0.58–2.44
Middle	1363 (98)	32 (2)	1.1	1.52–2.27
4th	1681 (98)	37 (2)	0.9	0.44–1.86
Highest	1138 (98)	25 (2)	ref	-
Water and sanitation
Main source of drinking water (n = 7282)
Piped water	5138 (98)	113 (2)	ref	-	0.016
Well / borehole	1957 (96)	47 (4)	1.9	1.22–2.94
Households toilets facility (n = 7282)
With toilet facility	5720 (98)	131 (2)	ref		0.003
No facility/bush/field	1375 (96)	56 (4)	2.1	1.27–3.30
Houses structures
Roofing (n = 7282)
Finished roofing	6013 (98)	141 (3)	ref	-	0.003
Natural/unfinished roofing	1082 (96)	46 (4)	2.2	1.29–3.58
Window can be closed (n = 7282)
Yes	3772 (98)	68 (2)	ref	-	< 0.001
No	3323 (97)	119 (3)	2.2	1.42–3.31
Eaves open
Open	5322 (97)	148 (3)	ref	-	0.590
Closed	1773 (98)	41 (2)	0.9	0.58–1.58
Low quality house
No	6818 (98)	168 (2)	ref	-
Yes	280 (93)	20 (7)	3.9	1.82–8.49	0.0005
Number of rooms (n = 7282)
One room	701 (96)	28 (4)	1.6	0.87–2.85
2–3 rooms	4737 (98)	120 (2)	ref	-	0.279
4 + rooms	1657 (98)	39 (2)	0.9	0.55–1.61
Vector control measures
IRS any time in the past 12 months (n = 7143)
Yes	2712 (97)	75 (3)	1	0.66–1.58
No	4250 (98)	106 (2)	ref	-	0.184
Missing	136 (95)	7 (5)	2.9	0.93–8.79
Household with any net (n = 7282)
Yes	6257 (98)	142 (2)	ref	-	0.001
No	838 (95)	45 (5)	2.5	1.45–4.16
Seasons
Long rains	1765 (99)	19 (1)	ref	-	< 0.001
Dry season	3358 (97)	98 (3)	3.0	1.57–5.80
Short rains	1975 (97)	71 (3)	3.7	1.85–7.43
Households type
Non-index households member	6319 (99)	78 (1)	ref	-
Index households member	779 (88)	110 (12)	18.4	11.40-29.68	< 0.001
Island of residence
Pemba	3489 (97)	97 (3)	ref	-	0.527
Unguja	3609 (98)	91 (3)	0.9	0.58–1.33
¹Models include random effect to account for clustering at household level

Environmental characteristics

Several climatic indicators for the place of residence and time period prior to the survey appear to be inversely associated with malaria prevalence, including minimum temperature, soil moisture, precipitation and relative humidity. The strongest positive association was found with the composite Palmer Drought Severity Index, suggesting the highest prevalence in the wettest areas (Fig. 1 and Additional file: Table S6).

Combined risk factor analysis

A multivariable logistic regression model with random effect (Table 4) confirmed that age was associated with malaria (P < 0001) with individuals 5–15 years and 16–25 years having significantly higher odds of malaria infection than adults over 25 years (aOR = 4.0, 95% CI 2.5–6.5 and aOR = 2.0, 95% CI 1.3–3.1, respectively). Individuals engaged in entrepreneurial activities (which may include any type of self-employed business) demonstrated higher odds of malaria infections (aOR = 1.8, 95% CI 1.1–2.7), compared to others who were unemployed. Rather unexpectedly, individuals who reported going to sleep later than 20:00 exhibited a decrease likelihood of being infected with malaria (aOR = 0.4, 95% CI 0.3–0.6, and aOR = 0.5, 95% CI 0.3–0.9), compared to those who reported to sleep between 18:00 and 20:00 hours.

The individuals who resided in poorer house structures with window openings which cannot be closed demonstrated two times higher odds of being infected with malaria (aOR = 2.1, 95% CI 1.4–3.1) than those who lived in better house structures with either screened windows or windows which can be closed. Independently, the odds of malaria infection were also significantly higher among members of households that did not own any bed nets (aOR = 2.3, 95% CI 1.4–3.6). As determined in earlier analyses [12], people living in index households of symptomatic malaria cases were much more likely to be infected with malaria (aOR = 14.9, 95% CI 9.7–23.0, P = < 0.001) compared to those living in neighbouring and nearby households.

Malaria infection was substantially higher in individuals living in areas with a moderate or high Palmer Drought Severity Index (aOR = 4.0, 95% CI 2.5–6.4, and aOR = 2.9, 95% CI 1.7–5.0), than those who lived in areas with a low Index. The difference between residents of Unguja and Pemba was not found to be significant (Table 4).

Table 4

Adjusted odds ratios for factors associated with malaria infection among individuals with no recent travel outside Zanzibar, detected by qPCR
Variables (N = 7282)	Negative n (%)	Positive n (%)	aOR*	95% CI	p-value
Age groups
< 5-Years	538 (98)	9 (2)	0.2	0.0–3.0	< 0.001
5–15	1383 (96)	64 (4)	4.0	2.5–6.5
16–25	1303 (97)	46 (3)	2.0	1.3–3.1
Over 25	3874 (98)	69 (2)	ref
Members occupation
Not employed	2511 (98)	64 (2)	ref		0.008
Entrepreneur	1650 (97)	56 (3)	1.8	1.1–2.7
Wage jobs	232 (97)	6 (3)	1.2	0.4–3.4
Students	2160 (98)	52 (2)	0.7	0.4–1.1
Missing	545 (98)	10 (2)	4.5	0.4–55.2
Members sleep time
18:00–20:00	1862 (96)	72 (4)	ref		< 0.001
20:00-22.00hrs	4244 (98)	90 (2)	0.4	0.3–0.6
Later than 22hrs	937 (98)	23 (2)	0.5	0.3–0.9
missing	55 (95)	3 (5)	0.7	0.2–2.9
Houses structures
Windows that can be closed/screened windows	3772 (98)	68 (2)	ref		< 0.001
Window can't be closed	3323 (97)	119 (3)	2.1	1.4 − 3.1
N/A	3 (75)	1 (25)	1
Household with any mosquito net
Yes	6257 (98)	142 (2)	ref		< 0.001
No	838 (95)	45 (5)	2.3	1.4–3.6
N/A	3 (75)	1 (25)	1.0
Households types
Neighbour/transect	6319 (99)	78 (1)	ref		< 0.001
Index households	779 (88)	110 (12)	14.9	9.7–23.0	< 0.001
Palmer Drought Severity Index
Low	3400 (99)	40 (1)	ref		< 0.001
Moderate	2127 (96)	93 (4)	4.0	2.5 - 6.4
High	1357 (96)	50 (4)	2.9	1.7–5.0
Missing	214 (98)	5 (2)	1.4	0.4 - 4.4
Island of residence
Pemba	3489 (97)	97 (3)	ref		0.067
Unguja	3609 (98)	91 (3)	1.3	0.9–2.0	0.067

Adjusted model with random effect to account for clustering at household level

This study investigated risk factors of locally acquired malaria infections, as detected in the population that had not recently travelled outside Zanzibar. Since with decreasing transmission, an increasing proportion of infected individuals [20, 26, 27] and of gametocyte carriers [28, 29] have a low-density parasitaemia, our study included not only infections detected by routine RDT but all qPCR-detected infections. Results from this study complement findings from the same setting documenting that asymptomatic and low-density infections are common [14, 26]. Studies conducted in other near-elimination settings have shown that low-density malaria infections are an obstacle to the attainment of the malaria elimination goal [30, 31].

As widely documented [2, 27], prevalence peaks shift to older age groups as transmission decreases. In this study, school-age children, adolescents and younger individuals were more likely to be infected than children < 5 years of adults above 25 years. An analysis of routine data from Zanzibar found lower test-positivity by RDT in adults ≥ 45 years compared to younger age groups, though this analysis would have missed the majority of low-density infections due to the limited sensitivity of RDTs [17, 32]. A study in north-eastern Tanzania, where transmission is also low, found a similar age group of school age children to be at increased risk of malaria infection [33] and research in Ethiopia found submicroscopic asymptomatic infections to account for 70% of the overall prevalence of 5% in school children aged 6–15 years [34]. Explanations for the increased risk among school-age children and adolescents may include delayed exposure due to low transmission, as well as behavioural factors such as different use patterns of bed nets or increased outdoor activities [16]. In Zanzibar, extra-curricular activities or extra classes happening indoor after sunset are also common, as well as early evening outdoor football TV-watching sessions - behaviour that might have exposed this younger age group to mosquito bites. Tailoring interventions to protect specific age groups would benefit from additional empirical evidence on the malaria transmission risk linked to specific activities that is currently not available for Zanzibar. A possible approach to specifically target school-aged children could include intermittent preventive therapy in this age group, as implemented in a study in north-eastern mainland Tanzania that led to significant declines in malaria cases [33, 35]. While routinely reported clinical cases in Zanzibar tend to include more male than female patients (especially during outbreaks [ZAMEP unpublished data]), no differences in odds of infections between genders was found in this study. Differences between these two different study populations could be a result of distinct treatment seeking behaviour. An important conclusion is nevertheless that efforts to prevent transmission and detect cases should certainly include both male and female community members.

Our study found that residents of dwellings that may be considered of poorer quality face a higher risk of malaria infection. In general terms, this finding showcases the relationship between socio-economic vulnerability, in this case reflected in poor house structures, and increased susceptibility to malaria. The association between lower living standards and malaria risk observed in Zanzibar aligns with existing evidence from elsewhere [36]. Architectural features that allow mosquitoes to enter and exit houses freely, or provide save and suitable spaces for indoor-resting mosquitoes, exacerbate ongoing local transmission. Our findings resonate with findings from mainland Tanzania, where poor house structures with open eaves and unscreened windows were associated with malaria infections in their residents [37–39]. As a mitigating measure, house improvement projects have been documented to effectively reduce malaria transmission in sub-Saharan Africa [40], an approach that should also be considered in Zanzibar at macro-scale or clusters of proximity.

Our study found the high-risk occupational group for malaria infection being entrepreneurs. In this study, the individuals categorized as entrepreneurs are engaged in various sectors including agriculture (specifically food crop farming), construction (focusing on building and dwelling), retail (operating small shops), food vending, and fishing. In some parts in Africa, occupational activities include agriculture [41, 42], and itinerant vendors [42]. In other places but not Zanzibar, occupational risk which exposes individuals to malaria transmission for several days includes forest goers [43, 44].

Rather unexpectedly, this study found people who go to sleep earlier in the evening (between 18:00 to 20:00 hours) to have a higher odds of malaria infection. Explanations could include an increased exposure to infected endophilic and endophagic mosquitoes, possibly linked with lower bed net use by early sleepers. A substantial number of individuals (27%) reported never using bed nets in this study. Similar as in other studies conducted in the same setting where substantial number of individuals reported not using bed net [45, 46]. The changing trend of mosquitoes to early evening biting in Zanzibar further might be a contributing factor for those who go to sleep early, as active vectors might be seeking human blood meals during this period posing an increased threat of infections [47, 48]. Evidence from mainland Tanzania and across Africa suggests changing patterns of mosquito biting times to the early evening [49–51] underlining the need to adapt vector control strategies to evolving mosquito behaviours. Targeted efforts to ensure universal access of mosquito nets and consistent usage, especially among those who go to sleep early, could play a pivotal role in reducing malaria transmission.

Our study found an association of malaria infection with people living in areas with a higher local index of wetness. As the relationship between meteorological, environmental, and geographical parameters, human behaviour, and malaria infection is complex, a better understanding of environmental and behavioural drivers of transmission is particularly important in the context of climate change and changes in patterns of malaria transmission [7, 52, 53]. The Zanzibar islands are not immune to the global trends of climate change, such as shifts in seasons, unpredictable weather patterns, and their potential effect on infectious disease dynamics [7, 54]. Unpredictable rainfall patterns and harsh weather conditions have been implicated in a substantial increase in malaria cases in countries like Pakistan [7, 55], jeopardizing previous gains in malaria control and elimination gains. Not to forget that high relative humidity and optimal temperature have been documented as often forgotten key parameters of sustained malaria transmission [7, 56–58]. Further analyses of environmental and climatic variables and routine surveillance data may help to unravel the complex effect on local malaria transmission.

Because of their isolated geographical settings, islands may still present a unique opportunity for malaria elimination due to comparably limited influx of people and parasites [59]. This study found that residing in Unguja carries a relatively higher risk of malaria infection, a finding supported by genomic research in the same setting [59]. These findings emphasize the importance of tailoring malaria control efforts to the specific transmission dynamics, recognizing the unique risk factors associated with the different island settings.

Previous epidemiological and modelling studies have provided evidence that, in order to accelerate progress towards malaria elimination in Zanzibar, increased efforts are required to address both the importation and the local transmission of malaria parasites [13, 46, 60]. Evidence as to why some individuals and geographical areas within the islands exhibit a higher risk of infection than others is warranted to identify targeted measures to curtail local infections.

One of the limitations is that the data collection in this study did not cover the general population but it was linked to reported clinical ‘index’ cases and the follow-up of their household clusters. The sample is hence biased towards households with clinical cases; however, neighbouring households of clinical cases were found to be comparable in terms of malaria infection prevalence to a general population sample [20]. Anecdotal evidence suggests that behaviour related to a person’s occupation (e.g. night-time jobs of security jobs, protecting crops, hunting, and fishing) or leisure activities may be risk factors for malaria transmission, as explored by Monroe et al. in a qualitative study [16]. Assessing details on study participants’ occupation, outdoor activities or indoor activities before going to bed was beyond the scope of data collection for this study. Further investigations into specific human behavioural patterns that expose people to infectious mosquito bites are therefore crucial to properly target interventions.

Another limitation of this analysis is that the study did not collect data throughout the entire calendar years of 2017 and 2018 (data was not collected during February, March, and April of both years and data collection ended in October of 2018). This implies that seasonality is only partially captured in this analysis. In addition, TerraClimate's raster datasets was incomplete, leading to missing climatic records for 219 households in the analysis (this could be due to missing of station data, or processing of the data into grid cells). However, as the missing of the data points is unlikely to be related to the study outcome, there was sufficiently unbiased information available for the analyses.

Additionally, this study is the recent malaria treatment received by participants, which could significantly affect the observed prevalence rates. This treatment may hinder the true impact of the risk factors on malaria infection, as the administration of antimalarial drugs reduces detectable prevalence, potentially distorting the relationship with risk factors. Therefore, reanalyzing the data without including recently treated individuals (if the sample size is statistically sufficient) might be essential to more accurately determine the influence of risk factors on malaria prevalence.

Findings from this study would certainly not have identified all factors of ongoing local malaria transmission in Zanzibar – but they point to a number of issues that can be addressed by ZAMEP and partners. A well designed approach of improved reactive focal interventions with mass drug administration and vector control combined with broader measures to reduce importation of infections (e.g. by reducing transmission in places of origin) and local transmission can be expected to accelerate progression towards malaria elimination in Zanzibar, as demonstrated in field studies elsewhere and in modelling analyses [13, 61].

As Zanzibar strives towards achieving malaria elimination, specific drivers of locally acquired malaria infection need to be addressed alongside the reduction of malaria importation. This study emphasizes the importance of targeting interventions to specific risk groups including school-age children, adolescent and young adults irrespective of their sex. Our findings further point to the likely contribution of poor living conditions and under-use of bed nets to ongoing local transmission. An integrated and multisectoral approach to vector control measures is required alongside ensuring universal access and use of bed nets in order to interrupt transmission and accelerate towards malaria elimination in Zanzibar.

ACT

Artemisinin-based Combination Therapy

confidence interval

DMSO

District Malaria Surveillance Officer

odds ratio

qPCR

quantitative Polymerase Chain Reaction

RACD

Reactive Case Detection

RDT

Rapid Diagnostic Test

ZAMEP

Zanzibar Malaria Elimination Programme.

Author information and contributions

BSF: contributed to planning, coordinating, data analysis, implementing of field study, developed the first draft of the manuscript, and writing – review & editing. OS: Environmental & climatological variables data extraction, spatial data – review & editing. AR: statistical support, analysis, writing – review & editing. RA: data management, and writing – review & editing. AM: planning, and implementing of field studies – review & editing. SS: writing – review & editing. SA: writing – review & editing. AA: writing – review & editing. GF: supervision, writing – review & editing. JY: conceiving, design, supervision, planning the study, implementing of field study, interpretation and editing of the final manuscript. MWH: conceiving, design, supervision, planning the study, implementing of field study, interpretation and editing of the final manuscript. All authors read and approved the final version of the manuscript.

Ethics approval and consent to participate

The Ethics Commission of Northwest and Central Switzerland (EKNZ Reg-2017-00162), Tulane University (Study Number: 993573), Ifakara Health Institute IRB (IHI/IRB/No: 003–2017), and the Zanzibar Medical Research and Ethics Committee all granted ethical approval for the initial data collection (ZAMREC/0001/February/17). In all study districts, administrative authorization was acquired prior to conducting the study including sensitization workshops with community leaders (Shehas). Written informed consent was acquired in advance from all participants or adult and guardians of children before conducting interviews and collecting blood samples. Assent was sought from minors before any blood samples were collected. In order to ensure the confidentiality of the participants' information, they were assigned software-generated IDs and all personal identifiers were removed from the final databases.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Funding Statement

Bakar S. Fakih was supported by the Department of Education of the Canton of Basel-Stadt (PhD scholarship), the Freiwillige Akademische Gesellschaft, Basel, Switzerland, and the Swiss Tropical and Public Health Institute. This is a secondary analysis using data generated in the Reactive Case Detection in Zanzibar: Effectiveness and Cost (RADZEC) project [20].

Author Contribution

Acknowledgements

We would like to extend our sincere acknowledgement to everyone who participated in this study, our study team members, the community leaders (Shehas) in all the study districts, all DMSOs in the study districts, and the management and staff at ZAMEP for their full support. We would like to extend our heartfelt appreciation to PD Dr. Amanda Ross for her invaluable advice on statistical methods and data analysis. Furthermore, we would like to acknowledge Gloria Shirima for her tireless efforts to clarify and guide data visualisation aspects.

Data Availability

Datasets underlying the calculations in this study are available upon reasonable request to the co-authors.

Le Menach A, Tatem AJ, Cohen JM, Hay SI, Randell H, Patil AP, Smith DL. Travel risk, malaria importation and malaria transmission in Zanzibar. Sci Rep. 2011;1:93.
Bhattarai A, Ali AS, Kachur SP, Mårtensson A, Abbas AK, Khatib R, Al-Mafazy A-w, Ramsan M, Rotllant G, Gerstenmaier JF. Impact of artemisinin-based combination therapy and insecticide-treated nets on malaria burden in Zanzibar. PLoS Med. 2007;4:e309.
Ashton RA, Bennett A, Al-Mafazy A-W, Abass AK, Msellem MI, McElroy P, Kachur SP, Ali AS, Yukich J, Eisele TP. Use of routine health information system data to evaluate impact of malaria control interventions in Zanzibar, Tanzania from 2000 to 2015. EClinicalMedicine 2019, 12:11–19.
van der Horst T, Al-Mafazy AW, Fakih BS, Stuck L, Ali A, Yukich J, Hetzel MW. Operational Coverage and Timeliness of Reactive Case Detection for Malaria Elimination in Zanzibar, Tanzania. Am J Trop Med Hyg. 2020;102:298–306.
Rulisa S, Kateera F, Bizimana JP, Agaba S, Dukuzumuremyi J, Baas L, de Dieu Harelimana J, Mens PF, Boer KR, de Vries PJ. Malaria prevalence, spatial clustering and risk factors in a low endemic area of Eastern Rwanda: a cross sectional study. PLoS ONE. 2013;8:e69443.
Le Menach A, Tatem AJ, Cohen JM, Hay SI, Randell H, Patil AP, Smith DL. Travel risk, malaria importation and malaria transmission in Zanzibar. Sci Rep. 2011;1:1–7.
WHO. World malaria report 2023. World Health Organisation; 2023.
Revolutionary Government of Zanzibar Ministry Of Health. Zanzibar Malaria Elimination Programme: Zanzibar Malaria Elimination Strategic Plan IV 2018/19-2022/23. 2027.
Ali MH, Kitau J, Ali AS, Al-Mafazy A-w, Tegegne SG, Ussi O, Musanhu C, Shija SJ, Khatib BO, Mkali H. Malaria elimination in Zanzibar: where next? Pan Afr Med J 2023, 45.
Mkali HR, Lalji SM, Al-Mafazy A-W, Joseph JJ, Mwaipape OS, Ali AS, Abbas FB, Ali MH, Hassan WS, Reaves EJ. How Real-Time Case-Based Malaria Surveillance Helps Zanzibar Get a Step Closer to Malaria Elimination: Description of Operational Platform and Resources. Global Health: Sci Pract 2023, 11.
Holzschuh A, Lerch A, Gerlovina I, Fakih BS, Al-Mafazy A-WH, Reaves EJ, Ali A, Abbas F, Ali MH, Ali MA. Multiplexed ddPCR-amplicon sequencing reveals isolated Plasmodium falciparum populations amenable to local elimination in Zanzibar, Tanzania. Nat Commun. 2023;14:3699.
Stuck L, Fakih BS, Abdul-wahid H, Hofmann NE, Holzschuh A, Grossenbacher B, Bennett A, Cotter C, Reaves E, Ali A. Malaria infection prevalence and sensitivity of reactive case detection in Zanzibar. Int J Infect Dis. 2020;97:337–46.
Das AM, Hetzel MW, Yukich JO, Stuck L, Fakih BS, Al-Mafazy A-WH, Ali A, Chitnis N. Modelling the impact of interventions on imported, introduced and indigenous malaria infections in Zanzibar, Tanzania. Nat Commun. 2023;14:2750.
Björkman A, Shakely D, Ali A, Morris U, Mkali H, Abbas A, Al-Mafazy A, Haji K, Mcha J, Omar R. From high to low malaria transmission in Zanzibar—challenges and opportunities to achieve elimination. BMC Med. 2019;17:1–15.
Shakely D. Aiming at malaria elimination in Zanzibar. Karolinska Institutet (Sweden); 2015.
Monroe A, Mihayo K, Okumu F, Finda M, Moore S, Koenker H, Lynch M, Haji K, Abbas F, Ali A. Human behaviour and residual malaria transmission in Zanzibar: findings from in-depth interviews and direct observation of community events. Malar J. 2019;18:1–13.
Mkali HR, Reaves EJ, Lalji SM, Al-Mafazy A-W, Joseph JJ, Ali AS, Abbas FB, Ali MH, Hassan WS, Kitojo C. Risk factors associated with malaria infection identified through reactive case detection in Zanzibar, 2012–2019. Malar J. 2021;20:1–10.
Hardy A, Mageni Z, Dongus S, Killeen G, Macklin MG, Majambare S, Ali A, Msellem M, Al-Mafazy A-W, Smith M. Mapping hotspots of malaria transmission from pre-existing hydrology, geology and geomorphology data in the pre-elimination context of Zanzibar, United Republic of Tanzania. Parasites vectors. 2015;8:1–15.
National Bureau of Statistics. population and housing census 2022. Population Distribution by Administrative Areas; 2023.
Stuck L, Fakih BS, Al-Mafazy AH, Hofmann NE, Holzschuh A, Grossenbacher B, Bennett A, Cotter C, Reaves E, Ali A, et al. Malaria infection prevalence and sensitivity of reactive case detection in Zanzibar. Int J Infect Dis. 2020;97:337–46.
Ingrams WH. Zanzibar: its history and its people. Routledge; 2012.
Abatzoglou JT, Dobrowski SZ, Parks SA, Hegewisch KC. TerraClimate, a high-resolution global dataset of monthly climate and climatic water balance from 1958–2015. Sci data. 2018;5:1–12.
Tetens O. Uber einige meteorologische Begriffe. Z geophys. 1930;6:297–309.
Mkopi A, Mtenga S, Festo C, Mhalu G, Shabani J, Tillya R, Masemo A, Kheir K, Nassor M, Mwengee W. Factors affecting non-coverage of measles-rubella vaccination among children aged 9–59 months in Tanzania. Vaccine. 2021;39:6041–9.
Vyas S, Kumaranayake L. Constructing socio-economic status indices: how to use principal components analysis. Health Policy Plann. 2006;21:459–68.
Björkman A, Morris U. Why asymptomatic Plasmodium falciparum infections are common in low-transmission settings. Trends Parasitol. 2020;36:898–905.
Koepfli C, Ome-Kaius M, Jally S, Malau E, Maripal S, Ginny J, Timinao L, Kattenberg JH, Obadia T, White M. Sustained malaria control over an 8-year period in Papua New Guinea: the challenge of low-density asymptomatic Plasmodium infections. J Infect Dis. 2017;216:1434–43.
Essuman E, Grabias B, Verma N, Chorazeczewski JK, Tripathi AK, Mlambo G, Addison EA, Amoah AG, Quakyi I, Oakley MS. A novel gametocyte biomarker for superior molecular detection of the Plasmodium falciparum infectious reservoirs. J Infect Dis. 2017;216:1264–72.
Slater HC, Ross A, Felger I, Hofmann NE, Robinson L, Cook J, Gonçalves BP, Björkman A, Ouedraogo AL, Morris U. The temporal dynamics and infectiousness of subpatent Plasmodium falciparum infections in relation to parasite density. Nat Commun. 2019;10:1433.
Harris I, Sharrock WW, Bain LM, Gray K-A, Bobogare A, Boaz L, Lilley K, Krause D, Vallely A, Johnson M-L. A large proportion of asymptomatic Plasmodium infections with low and sub-microscopic parasite densities in the low transmission setting of Temotu Province, Solomon Islands: challenges for malaria diagnostics in an elimination setting. Malar J. 2010;9:1–8.
Drakeley C, Gonçalves B, Okell L, Slater H. Understanding the importance of asymptomatic and low-density infections for malaria elimination. Towards Malar Elimination-A Leap Forw 2018:1–20.
Grossenbacher B, Holzschuh A, Hofmann NE, Omar KA, Stuck L, Fakih BS, Ali A, Yukich J, Hetzel MW, Felger I. Molecular methods for tracking residual Plasmodium falciparum transmission in a close-to-elimination setting in Zanzibar. Malar J. 2020;19:1–12.
Hayuma PM, Wang CW, Liheluka E, Baraka V, Madebe RA, Minja DT, Misinzo G, Alifrangis M, Lusingu JP. Prevalence of asymptomatic malaria, submicroscopic parasitaemia and anaemia in Korogwe District, north-eastern Tanzania. Malar J. 2021;20:1–9.
Biruksew A, Demeke A, Birhanu Z, Golassa L, Getnet M, Yewhalaw D. Schoolchildren with asymptomatic malaria are potential hotspot for malaria reservoir in Ethiopia: implications for malaria control and elimination efforts. Malar J. 2023;22:311.
Makenga G, Baraka V, Francis F, Nakato S, Gesase S, Mtove G, Madebe R, Kyaruzi E, Minja DT, Lusingu JP. Effectiveness and safety of intermittent preventive treatment with dihydroartemisinin–piperaquine or artesunate–amodiaquine for reducing malaria and related morbidities in schoolchildren in Tanzania: a randomised controlled trial. Lancet Global Health. 2023;11:e1277–89.
Tusting LS, Bottomley C, Gibson H, Kleinschmidt I, Tatem AJ, Lindsay SW, Gething PW. Housing improvements and malaria risk in sub-Saharan Africa: a multi-country analysis of survey data. PLoS Med. 2017;14:e1002234.
Liu JX, Bousema T, Zelman B, Gesase S, Hashim R, Maxwell C, Chandramohan D, Gosling R. Is housing quality associated with malaria incidence among young children and mosquito vector numbers? Evidence from Korogwe, Tanzania. PLoS ONE. 2014;9:e87358.
Lwetoijera DW, Kiware SS, Mageni ZD, Dongus S, Harris C, Devine GJ, Majambere S. A need for better housing to further reduce indoor malaria transmission in areas with high bed net coverage. Parasites vectors. 2013;6:1–9.
Okumu F, Finda M. Key characteristics of residual malaria transmission in two districts in south-eastern Tanzania—implications for improved control. J Infect Dis. 2021;223:S143–54.
Tusting LS, Bisanzio D, Alabaster G, Cameron E, Cibulskis R, Davies M, Flaxman S, Gibson HS, Knudsen J, Mbogo C. Mapping changes in housing in sub-Saharan Africa from 2000 to 2015. Nature. 2019;568:391–4.
Smith JL, Mumbengegwi D, Haindongo E, Cueto C, Roberts KW, Gosling R, Uusiku P, Kleinschmidt I, Bennett A, Sturrock HJ. Malaria risk factors in northern Namibia: the importance of occupation, age and mobility in characterizing high-risk populations. PLoS ONE. 2021;16:e0252690.
Gebreegziabher E, Raoliarison A, Ramananjato A, Fanomezana A, Rafaliarisoa M, Ralisata S, Razafindrakoto J, Smith JL, Ahmed J, Smith Gueye C. Identifying and characterizing high-risk populations in pilot malaria elimination districts in Madagascar: a mixed-methods study. Malar J. 2024;23:121.
Sandfort M, Vantaux A, Kim S, Obadia T, Pepey A, Gardais S, Khim N, Lek D, White M, Robinson LJ. Forest malaria in Cambodia: the occupational and spatial clustering of Plasmodium vivax and Plasmodium falciparum infection risk in a cross-sectional survey in Mondulkiri province, Cambodia. Malar J. 2020;19:1–12.
Dapari R, Mohd Yusop MZF, Chinnasamy D, Zakaria NI, Mohd Shoaib SM, Edros ME. A systematic review of the factors associated with malaria infection among forest rangers. PLoS ONE. 2024;19:e0303406.
Bauch JA, Gu JJ, Msellem M, Mårtensson A, Ali AS, Gosling R, Baltzell KA. Perception of malaria risk in a setting of reduced malaria transmission: a qualitative study in Zanzibar. Malar J. 2013;12:1–10.
Fakih BS, Holzschuh A, Ross A, Stuck L, Abdul R, Al-Mafazy A-WH, Irema I, Mbena A, Thawer SG, Shija SJ. Risk of imported malaria infections in Zanzibar: a cross-sectional study. Infect Dis Poverty. 2023;12:80.
Musiba RM, Tarimo BB, Monroe A, Msaky D, Ngowo H, Mihayo K, Limwagu A, Chilla GT, Shubis GK, Ibrahim A. Outdoor biting and pyrethroid resistance as potential drivers of persistent malaria transmission in Zanzibar. Malar J. 2022;21:1–14.
Haji KA, Khatib BO, Smith S, Ali AS, Devine GJ, Coetzee M, Majambere S. Challenges for malaria elimination in Zanzibar: pyrethroid resistance in malaria vectors and poor performance of long-lasting insecticide nets. Parasites vectors. 2013;6:1–9.
Degefa T, Githeko AK, Lee M-C, Yan G, Yewhalaw D. Patterns of human exposure to early evening and outdoor biting mosquitoes and residual malaria transmission in Ethiopia. Acta Trop. 2021;216:105837.
Sangbakembi-Ngounou C, Costantini C, Longo-Pendy NM, Ngoagouni C, Akone-Ella O, Rahola N, Cornelie S, Kengne P, Nakouné ER, Komas NP. Diurnal biting of malaria mosquitoes in the Central African Republic indicates residual transmission may be out of control. Proceedings of the National Academy of Sciences 2022, 119:e2104282119.
Mathania MM, Kimera SI, Silayo RS. Knowledge and awareness of malaria and mosquito biting behaviour in selected sites within Morogoro and Dodoma regions Tanzania. Malar J. 2016;15:1–9.
Tanser FC, Sharp B, Le Sueur D. Potential effect of climate change on malaria transmission in Africa. lancet. 2003;362:1792–8.
Paaijmans KP, Blanford S, Bell AS, Blanford JI, Read AF, Thomas MB. Influence of climate on malaria transmission depends on daily temperature variation. Proceedings of the National Academy of Sciences 2010, 107:15135–15139.
Abdalla AH, Kai KH, Khamis SA, Kondowe AL, Osima SE, King’uza PH, Hamad AO. The Influence of Climate Change and Variability on Spatio-Temporal Rainfall and Temperature Distribution in Zanzibar. Atmospheric Clim Sci. 2023;13:282–313.
Iqbal MP. Effect of Climate Change on Health in Pakistan: Climate Change and Health in Pakistan. Proceedings of the Pakistan Academy of Sciences: B Life and Environmental Sciences 2020, 57:1–12.
Santos-Vega M, Martinez P, Vaishnav K, Kohli V, Desai V, Bouma M, Pascual M. The neglected role of relative humidity in the interannual variability of urban malaria in Indian cities. Nat Commun. 2022;13:533.
Brown JJ, Pascual M, Wimberly MC, Johnson LR, Murdock CC. Humidity–The overlooked variable in the thermal biology of mosquito-borne disease. Ecol Lett 2023.
Lyon B, Dinku T, Raman A, Thomson MC. Temperature suitability for malaria climbing the Ethiopian Highlands. Environ Res Lett. 2017;12:064015.
Holzschuh A, Lerch A, Gerlovina I, Fakih BS, Al-mafazy A-wH, Reaves EJ, Ali A, Abbas F, Ali MH, Ali MA, et al. Multiplexed ddPCR-amplicon sequencing reveals isolated Plasmodium falciparum populations amenable to local elimination in Zanzibar, Tanzania. Nat Commun. 2023;14:3699.
Le Menach A, Tatem AJ, Cohen JM, Hay SI, Randell H, Patil AP, Smith DL. Travel risk, malaria importation and malaria transmission in Zanzibar. Sci Rep. 2011;1:93.
Hsiang MS, Ntuku H, Roberts KW, Dufour M-SK, Whittemore B, Tambo M, McCreesh P, Medzihradsky OF, Prach LM, Siloka G. Effectiveness of reactive focal mass drug administration and reactive focal vector control to reduce malaria transmission in the low malaria-endemic setting of Namibia: a cluster-randomised controlled, open-label, two-by-two factorial design trial. Lancet. 2020;395:1361–73.

No competing interests reported.

SuplimentaryTablesandFiguresFakihetal19.07.2024.docx

Download PDF

Reviewers agreed at journal
28 Oct, 2024
Reviewers invited by journal
31 Aug, 2024
Editor assigned by journal
10 Aug, 2024
Submission checks completed at journal
10 Aug, 2024
First submitted to journal
09 Aug, 2024

You are reading this latest preprint version

Determinants of locally acquired malaria infections in Zanzibar: a cross-sectional study

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Background

Methods

Study setting

Survey procedures

Climatic datasets

Participants, data management and analysis

Results

Study population

Malaria infection

Individual characteristics

Household characteristics

Environmental characteristics

Combined risk factor analysis

Discussion

Conclusion

Abbreviations

Declarations

Author information and contributions

Funding Statement

Author Contribution

Acknowledgements

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1