Alhassan et al. Malaria Journal (2022) 21:120 https://doi.org/10.1186/s12936-022-04136-3 Malaria Journal RESEARCH Open Access Impact of insecticide-treated nets and indoor residual spraying on self-reported malaria prevalence among women of reproductive age in Ghana: implication for malaria control and elimination Yakubu Alhassan1, Duah Dwomoh2* , Susan Ama Amuasi3, Justice Nonvignon1, Harriet Bonful4, Mary Tetteh1, Kofi Agyabeng2, Martha Kotey4, Alfred E. Yawson5 and Samuel Bosomprah2 Abstract Background: The Global Fund alone contributed 56% of all international financing for malaria and has invested more than US$13.5 billion in malaria treatment, prevention, and control programmes by June 2021. These investments include interventions such as mosquito nets, indoor residual spraying, and preventive treatment for children and pregnant women. However, there is paucity of studies for assessment of such investments to a reduction in malaria prevalence. This study was aimed at quantifying the impact of household access to insecticide-treated nets (ITNs) and the indoor residual spraying (IRS) on self-reported malaria prevalence among women of reproductive age in Ghana. Methods: The study analysed the 2016 Ghana Malaria Indicator Survey (MIS) data. The MIS is a nationwide survey that included women aged 15–49 years. Poisson regression model with inverse probability to treatment weighting was used to determine average treatment effect estimate of the two malaria interventions on self-reported malaria prevalence among women of reproductive age in Ghana. Results: A total sample of 4861 women interviewed from the 2016 Ghana MIS was used for analysis. The preva- lence of self-reported malaria in 2016 was 34.4% (95% CI [32.4%, 36.4%]). Approximately 80.0% of women lived in households with access to ITNs [Percentage (Pr) = 79.9%, (95% CI [78.0%, 81.7%])], 12.4% (95% CI [7.5%, 19.8%]) of the households had access to IRS and 11.4% (95% CI [7.0%, 18.0%]) of the households had access to both ITNs and IRS. Household access to only ITN contributed to 7.1 percentage point (pt) reduction in the self-reported malaria among women (95% CI [− 12.0%, − 2.1%], p = 0.005) whilst IRS at the households contributed to 6.8pt reduction in malaria prevalence (95% CI [− 12.0%, − 2.1%], p = 0.005). Households with access to both ITNs and IRS contributed to a 27.1pt reduction in self-reported malaria prevalence among women (95% CI [− 12.0%, − 2.1%], p = 0.005). Conclusion: Access to both ITNs and application of IRS at the household level contributed to a significant reduction in self-reported malaria prevalence among women of reproductive age in Ghana. This finding confirms the need for integration of malaria control interventions to facilitate attainment of malaria elimination in Ghana. *Correspondence: duahdwomoh@ug.edu.gh 2 Department of Biostatistics, School of Public Health, University of Ghana, Accra, Ghana Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://c reati veco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http://c reat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Alhassan et al. Malaria Journal (2022) 21:120 Page 2 of 17 Background through school-based distribution, and USD 3.90–4.55 Malaria is a life-threatening disease caused by Plasmo- through health facilities [5]. The median cost of protect- dium parasites transmitted through the infected bite of ing an individual each year using ITNs was estimated as female Anopheles mosquitoes. There was an estimated USD 2.20 (range: USD 0.88–9.54) whilst IRS was USD 241 million malaria cases and 627,000 malaria deaths in 6.70 (range: USD 2.22–12.85) [6]. Between May 2010 and 2020 compared to 228 million cases and 411,000 deaths October 2012, a total of 12.5 million ITNs were distrib- in 2018 [1, 2]. The disease disproportionately affects uted across Ghana with an incurred cost of USD 6.51 per children under the age of five years, accounting for ITN [7]. However, there is paucity of studies quantifying approximately 274,000 (67%) of all malaria deaths glob- the impact of these investments in terms of the distribu- ally in 2019. Countries in the World Health Organization tion of ITNs and the application of IRS towards reduction (WHO) African Region have a disproportionately high of malaria prevalence in Ghana. Therefore, the aim of this share of the global malaria burden, accounting for about study is to estimate the impact of ITNs distribution and 94% of malaria cases and deaths. In 2019, the total fund- application of IRS on malaria prevalence among women ing for malaria control and elimination was estimated as of reproductive age (15–49  years) in Ghana using self- USD 3 billion globally, of which about USD 900 million reported malaria as a proxy for true malaria prevalence. (31%) were contributed from governments of endemic countries [1]. Methods One of the overarching objectives of the sustainable Study design and participants development goals (SDG) is to attain the highest stand- Data for this study were derived from the Ghana Malaria ard of health care for everyone within all communities by Indicator Survey (GMIS). The GMIS is a nationally rep- preventing the occurrence of diseases [3]. Vector control resentative survey conducted by the Ghana Statistical has been identified as an important preventive strategy Service from October 2016 to December 2016. For this for malaria. The WHO recommends insecticide-treated study, only women of reproductive age 15–49 years from nets (ITNs) and indoor residual spraying (IRS) as part of the survey were considered. Women who had data for all this strategy. These preventive strategies came at a huge the variables were included in the analysis. cost with an estimated USD 3.1 billion invested in 2017 The Ghana MIS used a multi-stage cluster sampling of which USD 2.2 billion were invested in the WHO Afri- procedure across all 10 regions of country at the time of can regions [2]. A total of 624 million mosquito nets were the survey in 2016. The country was divided into 20 strata delivered from 2015 to 2017, of which 459 million (83%) (10 regions and residential types—urban/rural). A clus- ITNs were delivered in sub-Saharan Africa [2]. In 2019, ter was defined as a census enumeration area (EA) com- it was estimated that about 46% of all people at risk of prising approximately between 300 and 500 households. malaria in Africa were protected by an ITN, compared In the first stage of sampling, for each stratum, clusters to 2% in 2000 [1]. However, ITN coverage has plateaued were selected using probability proportion to size. A total since 2016 [1]. In contrast, globally, IRS protection of 200 clusters were selected. In the second stage of sam- declined from a peak of 5% in 2010 to 2% in 2019, with pling, a fixed number of 30 households were randomly decreases recorded across all WHO regions. The declines selected from each selected cluster without replacement. in IRS coverage are occurring as countries switch from Women aged 15–49 were interviewed from each house- pyrethroid insecticides to more expensive alternatives to hold if available [8]. In the original survey, 5150 women mitigate mosquito resistance to pyrethroids [1]. were interviewed. However, due to missing responses for In Ghana, over 13 million ITNs had been distributed some of the variables, a total of 4861 women were used as of September 2017 with about 1.5 million of those for this study representing 94.4% of the sampled women. distributed in 2017 only [4]. Again, over 300,000 house- The data includes information on housing, household, holds were sprayed against mosquitoes protecting over women characteristics, malaria prevention, and knowl- 840,000 household residents through the indoor residual edge on malaria. Computer-assisted personal inter- spraying programme [4]. Funding from the US Presi- viewing (CAPI) system on tablet computers and paper dent Malaria Initiative (PMI) over the years from an ini- questionnaires were used to collected data. The Census tial annual funding of USD 5 million in 2008 increased and Survey Processing (CSPro) system was used for data to USD 28 million in 2017 cumulating to over USD 275 editing and management by the data curators [8]. million within the 10 years period. A budget of USD 26 million was made for the malaria operational plan for Variable definition the 2018 fiscal year through the PMI [4]. The median Primary outcome cost of distribution of each ITN was estimated as USD The primary outcome for this study was prevalence of 4.34–4.55 through mass distribution, USD 3.30 to 3.69 self-reported malaria among women of reproductive A lhassan et al. Malaria Journal (2022) 21:120 Page 3 of 17 ages 15–49  years, defined as women who reported to of Ghana covers malaria. Women who scored 0–2 were have experienced at least one episode of malaria within considered to have low knowledge, those who scored 3 12-months preceding the survey.  That is,  self-reported or 4 were considered to have moderate knowledge and malaria  prevalence among the women  aged 15–49  was those who scored five were considered to have compre- used as proxy for  actual malaria RDT or microscopy hensive knowledge on malaria. The selected variables are positivity among the women because these tests were not associated with access to ITNs, IRS or malaria prevalence performed among the women during the survey. in literature [10–18]. Statistical analysis Intervention Background characteristics of women were summa- The interventions were household access to ITNs, and rized using frequencies and percentages for categorical application of IRS in households within 12 months prior variables whereas continuous variables were summa- to the survey. Households which had received both rized using mean and standard deviation. The  char- interventions were considered as integrated interven- acteristics were summarized by intervention status to tion. Household access to ITNs was defined as women examine potential imbalance and population structure, who were living in households with access to at least one which is an indication of potential confounding  bias. insecticide-treated net while household application of Choropleth maps were used to describe prevalence of IRS was defined as women living in households that had self-reported malaria among women and coverage of been sprayed against mosquitoes within the 12  months the two interventions by geographical location. The preceding the survey. Rao’s Scott’s chi-square  test statistic that accounts for survey design characteristics (i.e., stratification, cluster- ing, and sampling weight) was used to assess the asso- Potential confounders ciation between self-reported malaria prevalence and The study considered two main categories of confound- access to the two interventions and background char- ing variables, namely, household, and individual charac- acteristics. Self-reported malaria prevalence was calcu- teristics. Household characteristics included; regions, lated as the number of women who experienced at least type of residence (rural–urban), sex of household head, one episode of malaria in the 12 months preceding the household size, household access to electricity, type of survey divided by the total eligible women interviewed cooking fuel (solid or non-solid), main floor material, in the survey. main wall material, roof material, source of drinking A modified weighted Poisson regression model was water (improved or unimproved), type of toilet facility used to estimate the impact of access to the malaria (improved or unimproved) and household wealth cat- interventions on self-reported malaria prevalence egory (Poor, middle and rich). Categories of the house- among women after adjusting for the inverse prob- hold characteristics were recoded according to the DHS ability of treatment weight (IPTW) and survey weight reporting standards in the 2016 GMIS and 2014 Ghana using the “svy linearized” model in Stata 16 IC (Stata Demographic Health Survey (GDHS) reports [8, 9]. Indi- Corp, College Station, TX, USA). The inverse probabil- vidual characteristics considered were current age of the ity of treatment weight (IPTW) for intervention “i” and woman, highest level of education, pregnancy status at woman “j” was estimated as: time of survey, health insurance status, religion, exposure i 1− i to malaria messages in the 6 months prior to the survey IPTWij = + and the knowledge level of the woman on malaria issues. pwij 1− pwij The knowledge level of the woman was assessed using where, IPTW is the inverse probability of treatment five knowledge questions including woman’s knowledge ijweight for intervention i for woman j, pw is the estimated on causes of malaria, symptoms of malaria, methods of ijprobabitlity of woman j having access to intervention i, preventing malaria, treatment of malaria and aware- ness that the national health insurance scheme (NHIS) { 0if individual j does not have access to intervention i i is the indicator variable 1 if individual j have access to the intervention i Alhassan et al. Malaria Journal (2022) 21:120 Page 4 of 17 . living in household of 4–6 members. Majority of the The final weighting variable to be used in the Poisson households had access to electricity (79.5%), improved regression model was then adjusted as follows: source of drinking water (87.2%), improved toilet facil- ity (71.4%) and uses solid cooking fuel (76.6%) (Table 1). fwi = IPTWi ∗ swi The mean (SD) age of the women was 29.8 (9.5) years. where, fw is the final weighting variable for individual In most (55.9%) cases, the women had up to secondary ij j and intervention i, sw is the sampling weight from the level of education while few of them had beyond sec-ij 2016 GMIS for individual j and intervention i. ondary education. Christianity was the most (77.4%) The command “margins, dydx (intervention_i)” post affiliated religion among the women. Over a quarter estimation command in Stata was then used to estimate (28.6%) of the women had never given birth, another the marginal difference (impact) of access to interven- 28.9% had given birth once or twice whilst a fifth tion “i” on self-reported malaria prevalence among (20.0%) had given birth for more than four times. About women after the modified weighted Poisson regression seven in every ten women sampled (68.2%) had a com- model was fitted controlling for all observed confound- prehensive knowledge of malaria. However, more than ing variables. half (54.2%) of the women had been exposed to malaria As a sensitivity analysis, three different regression messages in the past 6 months (Table 1). models, the binary logistic regression, the probit regres- sion, and the linear regression models were also used to Prevalence of self‑reported malaria and access to malaria estimate the impact of each of the malaria interventions interventions on self-reported malaria prevalence among women in The prevalence of self-reported malaria in the last Ghana. The 95% confidence interval was estimated for 12  months prior to the survey was 34.4% (95% CI all the point prevalence estimates, prevalence ratios as 32.4–36.4%). The percentage of women with access to well as impact estimates. All statistical analyses in this ITNs was 79.9% (95% CI 78.0–81.7%) whereas women study were considered significant at an alpha level of living in household sprayed against mosquitoes (IRS) 0.050. Stata IC version 16 (StataCorp, Texas, USA) was was 12.4% (95% CI 7.5–19.8%). Access to only IRS was used for statistical analysis. 1.0% (95% CI 17.1–21.2%), only ITNs was 68.5% (95% CI 62.9–73.6%) and both IRS and ITNs was 11.4% (95% CI 7.0–18.0%) (Fig. 1). Ethical statement Access to ITNs was significantly associated with The Demographic and Health Survey (DHS) program region (p < 0.001), area of residence (p < 0.001), house- approved and granted permission to use the data for hold size (p < 0.001), Sex of household head (p < 0.001), this paper. The data was accessed from the DHS pro- age of household head (p = 0.041), household wealth gram website (http:// dhspr ogram. com) on 8th Septem- index category (p < 0.001), source of drinking water ber 2020. The data was already de-identified and can (p = 0.004), type of toilet facility (p < 0.001), access to longer be linked to any individual participant in the electricity (p < 0.001), type of cooking fuel (p < 0.001) survey. and housing characteristics such as main wall mate- rial (p < 0.001) and main roof material (p < 0.001). In addition, women characteristics such as education Results (p = 0.009), number of births (p = 0.007) and knowl- Characteristics of households and women in the study edge of malaria (p < 0.001) were also associated with A total of 4861 women aged 15–49  years interviewed access to ITNs (Table 1). in the 2016 GMIS survey were involved in this study. Household characteristics associated with access A majority (53.1%) were from the urban areas of to IRS included region (p < 0.001), place of residence the country. The Ashanti (19.8%) and Greater Accra (p = 0.005), household size (p < 0.001), sex of household (18.1%) regions had the highest percentage of partici- head (p = 0.007), wealth index (p = 0.011), type of toilet pants whilst the Upper East (4.0%) and Upper West facility (p = 0.007), main wall material (p = 0.004) and (2.7%) regions had the least percentage of participants. main roof material (p < 0.001). The women character- Approximately 36.1% of the households were headed istics associated with access to IRS among the women by males. The mean (SD) age of the household head included education (p < 0.001), health insurance status was 43.8 (13.5) years. Most (45.8%) of the women were (p < 0.001) and religion (p = 0.005) (Table 1). A lhassan et al. Malaria Journal (2022) 21:120 Page 5 of 17 Table 1 Background characteristics of women by intervention status Variables Number of women (%) Access to ITNs Household Sprayed Combination of interventions (IRS) None IRS only ITNs only Both IRS & ITNs P‑value % P‑value % P‑value % % % % Total 4861 (100%) 79.91 12.43 19.07 1.02 68.5 11.41 Region of residence < 0.001 < 0.001 < 0.001 Western 398 (8.19) 5.99 1.15 2.06 0.14 4.98 1.01 Central 511 (10.51) 9.42 0.78 1.09 0.00 8.64 0.78 Greater Accra 882 (18.15) 12.86 0.29 5.29 0.00 12.57 0.29 Volta 401 (8.24) 6.79 0.00 1.46 0.00 6.79 0.00 Eastern 451 (9.28) 7.09 0.00 2.19 0.00 7.09 0.00 Ashanti 964 (19.84) 14.87 2.00 4.59 0.37 13.25 1.62 Brong Ahafo 420 (8.64) 7.27 0.23 1.35 0.01 7.05 0.22 Northern 509 (10.48) 9.21 4.47 0.93 0.34 5.08 4.13 Upper East 192 (3.95) 3.86 1.01 0.08 0.01 2.85 1.00 Upper West 133 (2.73) 2.56 2.50 0.03 0.15 0.20 2.35 Place of residence < 0.001 0.005 < 0.001 Urban 2579 (53.06) 39.16 3.27 13.62 0.29 36.17 2.99 Rural 2282 (46.94) 40.75 9.15 5.45 0.73 32.33 8.42 Household characteristics Household size < 0.001 < 0.001 < 0.001 < 4 members 1448 (29.79) 21.79 2.21 7.60 0.40 19.98 1.81 4–6 members 2225 (45.78) 37.29 5.28 8.22 0.27 32.28 5.01 7–9 members 904 (18.6) 15.73 3.61 2.52 0.35 12.46 3.27 10+ members 284 (5.83) 5.11 1.32 0.73 0.00 3.78 1.32 Sex of household head < 0.001 0.007 < 0.001 Male 3108 (63.93) 53.06 9.87 10.2 0.67 43.86 9.20 Female 1753 (36.07) 26.85 2.56 8.86 0.36 24.64 2.21 Age of household head (mean ± SD) 43.75 ± 13.46 0.041 0.183 0.054 < 30 668 (13.74) 10.14 1.20 3.55 0.05 8.99 1.16 30–49 2746 (56.49) 45.84 7.60 9.92 0.73 38.97 6.87 50–69 1206 (24.81) 19.75 3.07 4.84 0.22 16.90 2.85 > 69 241 (4.961) 4.17 0.55 0.76 0.02 3.64 0.53 Wealth index < 0.001 0.011 < 0.001 Poor 1705 (35.08) 30.82 6.48 4.09 0.17 24.51 6.31 Middle 1000 (20.56) 16.60 2.82 3.66 0.30 14.09 2.52 Rich 2156 (44.36) 32.48 3.13 11.32 0.56 29.90 2.58 Alhassan et al. Malaria Journal (2022) 21:120 Page 6 of 17 Table 1 (continued) Variables Number of women (%) Access to ITNs Household Sprayed Combination of interventions (IRS) None IRS only ITNs only Both IRS & ITNs P‑value % P‑value % P‑value % % % % Source of water 0.004 0.857 0.324 Improved water source 4239 (87.22) 68.87 10.95 17.37 0.97 58.89 9.98 Unimproved water source 621 (12.78) 11.03 1.48 1.70 0.05 9.61 1.42 Toilet facility < 0.001 0.007 0.002 Improved toilet facility 3468 (71.35) 55.38 6.73 15.34 0.63 49.28 6.10 Unimproved toilet facility 1393 (28.65) 24.52 5.70 3.73 0.39 19.22 5.31 Access to electricity < 0.001 0.524 0.045 No 996 (20.48) 18.29 3.04 2.06 0.13 15.39 2.91 Yes 3865 (79.52) 61.61 9.39 17.01 0.89 53.11 8.50 Main floor materials 0.058 0.076 0.087 Ceramic/tiles/carpet 1165 (23.97) 18.20 2.05 5.51 0.26 16.41 1.79 Cement 3051 (62.78) 50.66 8.16 11.55 0.57 43.07 7.59 Sand/earth/wooden planks 644 (13.25) 11.05 2.22 2.01 0.19 9.02 2.02 Main wall materials < 0.001 0.004 < 0.001 Cement/bricks 3143 (64.66) 48.77 4.97 15.27 0.61 44.41 4.36 Others (clay, woods etc.,) 1718 (35.34) 31.13 7.46 3.80 0.41 24.09 7.05 Main roof materials 0.002 < 0.001 < 0.001 Asbestos/shingles/concrete 871 (17.92) 12.90 0.26 4.96 0.06 12.70 0.21 Zinc/aluminium 3810 (78.37) 63.75 11.38 13.68 0.95 53.32 10.43 Thatch/palm leaves/wood 180 (3.709) 3.25 0.78 0.44 0.02 2.49 0.77 Cooking fuel < 0.001 0.064 0.003 Non-solid (LPG, electricity) 1137 (23.39) 16.48 1.55 6.64 0.27 15.21 1.27 Solid (charcoal, woods, etc.) 3724 (76.61) 63.43 10.88 12.43 0.75 53.29 10.13 Women characteristics Woman’s age 29.80 ± 9.51 0.91 0.439 0.7016 15–19 854 (17.56) 14.20 2.55 3.08 0.28 11.93 2.27 20–29 1610 (33.12) 26.36 3.88 6.47 0.29 22.77 3.59 30–39 1458 (29.99) 23.88 3.66 5.86 0.25 20.47 3.41 40–49 939 (19.33) 15.46 2.33 3.66 0.21 13.34 2.13 Woman’s education 0.009 < 0.001 < 0.001 No education 955 (19.65) 16.46 4.62 2.75 0.44 12.28 4.18 A lhassan et al. Malaria Journal (2022) 21:120 Page 7 of 17 Table 1 (continued) Variables Number of women (%) Access to ITNs Household Sprayed Combination of interventions (IRS) None IRS only ITNs only Both IRS & ITNs P‑value % P‑value % P‑value % % % % Primary 832 (17.12) 13.91 2.23 2.92 0.28 11.96 1.95 Secondary 2719 (55.94) 44.18 4.80 11.46 0.30 39.68 4.50 Higher/tertiary 355 (7.29) 5.35 0.77 1.94 0.00 4.58 0.77 Number of births 0.007 0.204 0.076 None 1391 (28.61) 21.81 3.12 6.44 0.36 19.05 2.76 1–2 births 1406 (28.92) 22.89 3.39 5.75 0.28 19.77 3.12 3–4 births 1095 (22.52) 18.56 2.73 3.85 0.10 15.93 2.63 > 4 births 970 (19.96) 16.65 3.18 3.02 0.29 13.75 2.89 Woman’s currently pregnant 0.079 0.134 0.164 No/unsure 4511 (92.8) 73.84 11.28 18.05 0.90 63.47 10.38 Yes 350 (7.2) 6.06 1.15 1.02 0.12 5.03 1.03 Covered by health insurance 0.135 < 0.001 0.003 No 2011 (41.38) 32.44 3.98 8.71 0.22 28.68 3.76 Yes 2850 (58.62) 47.47 8.45 10.36 0.80 39.82 7.65 Woman’s religion 0.277 0.005 0.004 Christians 3760 (77.35) 61.74 6.48 15.12 0.48 55.74 6.00 Islam 946 (19.46) 15.83 5.71 3.13 0.49 10.61 5.22 Tradition/no religion/others 155 (3.19) 2.33 0.24 0.81 0.05 2.15 0.18 Knowledge of malaria < 0.001 0.1254 0.016 Low knowledge 81 (1.66) 1.25 0.15 0.38 0.04 1.13 0.11 Moderate knowledge 1464 (30.11) 22.90 3.10 6.89 0.33 20.12 2.77 Comprehensive knowledge 3316 (68.22) 55.77 9.18 11.80 0.66 47.24 8.52 Exposure to malaria messages in the 0.942 0.167 0.194 past 6 months Not exposed 2633 (54.17) 43.31 7.67 10.07 0.79 36.43 6.88 Exposed 2228 (45.83) 36.59 4.76 9.00 0.24 32.07 4.52 ITN, insecticide treated net; IRS, indoor residual spraying; CI, confidence interval All percentages are column percentages P-values are from the Rao Scott’s chi-square tests Alhassan et al. Malaria Journal (2022) 21:120 Page 8 of 17 Fig. 1 Prevalence of self-reported malaria and access to malaria interventions among women aged 15–49 years in Ghana Regional distribution of self‑reported malaria prevalence Factors associated with self‑reported malaria prevalence and access to malaria interventions among women in the past 12 months The Upper East (42.8%) and the Central (45.3%) Prevalence of self-reported malaria was significantly recorded the highest self-reported malaria prevalence associated with the region of residence of the women whilst the Upper West (23.1%) and Ashanti (28.4%) (χ2 = 4.38, p < 0.001). Self-reported malaria prevalence recorded the least prevalence. Access to ITNs was was highest among women with access to improved highest in the Upper West (93.6%) and the Upper East water sources (35.4%, 95% CI 33.4–37.5%) compared to (97.7%) regions whilst Greater Accra (70.9%), Western the 27.3% (95% CI 23.5–31.5%) prevalence among women (73.1%) and Ashanti (75.0%) recorded the least percent- with access to unimproved water sources (χ2 = 12.57, age access. The percentage of women with access to IRS p < 0.001). Also, self-reported malaria prevalence was was highest in the Upper West region (91.7%) followed lowest among women in the age range 15–19  years by the Northern region with 42.7% and Upper East with (25.4%, 95% CI 22.3–28.7%) compared to women in the 25.6% whilst the rest of the southern regions recorded age groups 20–29  years (35.6%, 95% CI 31.7–38.5%), less than 15% each with the Volta and Eastern regions 30–39 years (37.1%, 95% CI 33.8–40.6%) and those aged recording 0%. Access to both ITNs and IRS was high- 40–49 years (36.3%, 95% CI 32.5–40.3%). The age group est in the three northern regions, Upper West (86.3%), of women was significantly associated with self-reported Northern (39.4%) and Upper East (25.4%) (Fig. 2). malaria prevalence (χ2 = 8.14, p < 0.001). Self-reported malaria was lowest among women with low knowledge Prevalence of self‑reported malaria among women on malaria (11.4%, 95% CI 6.3–19.7%) compared to 12 month before the survey by access to malaria women with moderate (33.8%, 95% CI 30.4–37.5%) or interventions comprehensive (35.2%, 95% CI 32.9–37.5%) knowledge Prevalence of self-reported malaria among women with (χ2 = 7.03, p = 0.002). Also, self-reported malaria preva- access to ITNs was 33.3% (95% CI 31.2–35.4%) which was lence was highest among women exposed to malaria significantly lower compared to the 38.7% (95% CI 33.9– messages (40.1%, 95% CI 37.3–43.0%) compared to 43.7%) among women with no access to ITNs (χ2 = 4.32, women not exposed to malaria messages (29.5%, 95% CI p = 0.039). Self-reported malaria did not significantly 27.1–32.0%) (χ2 = 34.07, p < 0.001) (Table 2). vary between women with access to IRS (32.3%, 95% CI 28.1–36.9%) compared to women with no access to IRS 2= = The impact of household access to ITNs and application (34.7%, 95% CI 32.6–36.8%) (χ 0.91, p 0.342). Also, of IRS on self‑reported malaria prevalence self-reported malaria among the women did not signifi- Women living in households with access to ITNs cantly differ across the combination of access to the two 2= = recorded a 7.05% significant absolute reduction in self-malaria interventions (χ 1.65, p 0.188) (Table 2). reported malaria prevalence [ATE −  7.05%, 95% CI A lhassan et al. Malaria Journal (2022) 21:120 Page 9 of 17 Fig. 2 Prevalence of self-reported malaria and access to malaria interventions among women by regions (− 11.96%, − 2.14%), p = 0.005]. Women living in house- those with access to IRS only, access to both ITNs and holds with access IRS had a 6.81% significant reduction in IRS did not show significant reduction in malaria preva- self-reported malaria prevalence [ATE: − 6.81%, 95% CI lence in the final model (Table 3). (− 13.06%, − 0.55%), p = 0.033] (Table 3). Compared to those with no access to both ITNs and Compared to those with access to only ITNs, access to IRS, access to both ITNs and IRS contributed a 27.09% both ITNs and IRS did not show significant reduction significant absolute reduction in self-reported malaria in self-reported malaria prevalence among the women prevalence among the women [ATE: −  27.09, 95% CI in any of the four regression models. Also, compared to (− 34.94%, − 19.25%), p < 0.001] (Table 3). Alhassan et al. Malaria Journal (2022) 21:120 Page 10 of 17 Table 2 Prevalence of self-reported malaria among women 12 month before the survey by access to malaria interventions Experienced malaria in the past 12 months Rao Scott’s Chi‑ P‑value square No Yes % [95% CI] % [95% CI] 65.62 [63.63, 67.57] 34.38 [32.43, 36.37] Interventions Insecticide treated nets (ITNs) 4.32 0.039 No access to ITNS 61.33 [56.31, 66.11] 38.67 [33.89, 43.69] Access to ITNs 66.71 [64.59, 68.76] 33.29 [31.24, 35.41] Indoor residual spraying (IRS) 0.91 0.342 Household not sprayed 65.34 [63.18, 67.43] 34.66 [32.57, 36.82] Household sprayed 67.63 [63.07, 71.87] 32.37 [28.13, 36.93] Both ITNs and IRS 1.65 0.188 No access to both ITNs & IRS 61.31 [56.44, 65.96] 38.69 [34.04, 43.56] Access to only IRS 61.65 [39.00, 80.17] 38.35 [19.83, 61.00] Access to only ITNs 66.46 [64.27, 68.59] 33.54 [31.41, 35.73] Access to both ITNs & IRS 68.16 [62.41, 73.42] 31.84 [26.58, 37.59] Region of residence 4.38 < 0.001 Western 63.19 [57.76,68.30] 36.81 [31.70,42.24] Central 54.72 [48.15,61.13] 45.28 [38.87,51.85] Greater Accra 66.59 [61.80,71.07] 33.41 [28.93,38.20] Volta 69.30 [64.06,74.08] 30.70 [25.92,35.94] Eastern 65.02 [59.91,69.80] 34.98 [30.20,40.09] Ashanti 71.64 [65.87,76.78] 28.36 [23.22,34.13] Brong Ahafo 60.56 [53.22,67.44] 39.44 [32.56,46.78] Northern 67.44 [62.39,72.11] 32.56 [27.89,37.61] Upper East 57.25 [50.71,63.54] 42.75 [36.46,49.29] Upper West 76.95 [70.86,82.09] 23.05 [17.91,29.14] Place of residence 0.00 0.970 Urban 65.59 [62.88,68.20] 34.41 [31.80,37.12] Rural 65.66 [62.68,68.53] 34.34 [31.47,37.32] Household characteristics Household size 0.89 0.440 < 4 members 64.02 [60.42,67.46] 35.98 [32.54,39.58] 4–6 members 65.52 [62.53,68.39] 34.48 [31.61,37.47] 7–9 members 67.45 [63.50,71.17] 32.55 [28.83,36.50] 10+ members 68.81 [62.58,74.43] 31.19 [25.57,37.42] Sex of household head 0.63 0.430 Male 66.14 [63.87,68.33] 33.86 [31.67,36.13] Female 64.71 [61.47,67.83] 35.29 [32.17,38.53] Age of household head (mean ± SD) 0.44 0.702 < 30 64.54 [59.04,69.69] 35.46 [30.31,40.96] 30–49 65.47 [62.57,68.27] 34.53 [31.73,37.43] 50–69 65.69 [62.45,68.79] 34.31 [31.21,37.55] > 69 69.99 [62.22,76.76] 30.01 [23.24,37.78] Wealth index 2.26 0.106 Poor 67.68 [64.29,70.88] 32.32 [29.12,35.71] Middle 61.91 [57.74,65.90] 38.09 [34.10,42.26] Rich 65.73 [62.61,68.71] 34.27 [31.29,37.39] Source of water 12.57 < 0.001 Improved water source 64.59 [62.48,66.64] 35.41 [33.36,37.52] Alhassan et al. Malaria Journal (2022) 21:120 Page 11 of 17 Table 2 (continued) Experienced malaria in the past 12 months Rao Scott’s Chi‑ P‑value square No Yes % [95% CI] % [95% CI] Unimproved water source 72.70 [68.54,76.49] 27.30 [23.51,31.46] Toilet facility 0.14 0.713 Improved toilet facility 65.39 [62.95,67.74] 34.61 [32.26,37.05] Unimproved toilet facility 66.22 [62.45,69.80] 33.78 [30.20,37.55] Access to electricity 1.98 0.161 No 68.52 [63.69,72.97] 31.48 [27.03,36.31] Yes 64.88 [62.75,66.95] 35.12 [33.05,37.25] Main floor materials 2.50 0.083 Ceramic/tiles/carpet 62.18 [58.53,65.70] 37.82 [34.30,41.47] Cement 66.35 [63.91,68.71] 33.65 [31.29,36.09] Sand/earth/wooden planks 68.41 [63.05,73.33] 31.59 [26.67,36.95] Main wall materials 0.49 0.485 Cement/bricks 65.20 [62.70,67.61] 34.80 [32.39,37.30] Others (clay, woods etc.,) 66.40 [63.63,69.06] 33.60 [30.94,36.37] Main roof materials 0.50 0.587 Asbestos/shingles/concrete 66.37 [61.43,70.97] 33.63 [29.03,38.57] Zinc/aluminium 65.27 [63.05,67.43] 34.73 [32.57,36.95] Thatch/palm leaves/wood 69.50 [62.36,75.81] 30.50 [24.19,37.64] Cooking fuel 0.61 0.435 Non-solid (LPG, electricity) 64.47 [60.94,67.84] 35.53 [32.16,39.06] Solid (charcoal, woods, etc.) 65.98 [63.73,68.15] 34.02 [31.85,36.27] Women characteristics Woman’s age 8.14 < 0.001 15–19 74.63 [71.32,77.68] 25.37 [22.32,28.68] 20–29 64.45 [60.87,67.87] 35.55 [32.13,39.13] 30–39 62.91 [59.45,66.24] 37.09 [33.76,40.55] 40–49 63.67 [59.66,67.50] 36.33 [32.50,40.34] Woman’s education 0.68 0.537 No education 68.35 [63.02,73.24] 31.65 [26.76,36.98] Primary 64.97 [61.44,68.35] 35.03 [31.65,38.56] Secondary 64.99 [62.40,67.50] 35.01 [32.50,37.60] Higher/tertiary 64.66 [57.68,71.07] 35.34 [28.93,42.32] Number of births 2.24 0.087 None 68.29 [65.33,71.12] 31.71 [28.88,34.67] 1–2 births 64.80 [61.32,68.13] 35.20 [31.87,38.68] 3–4 births 66.40 [62.68,69.92] 33.60 [30.08,37.32] > 4 births 62.12 [57.80,66.25] 37.88 [33.75,42.20] Woman’s currently pregnant 0.07 0.787 No/unsure 65.70 [63.61,67.73] 34.30 [32.27,36.39] Yes 64.67 [57.16,71.51] 35.33 [28.49,42.84] Covered by health insurance 0.23 0.630 No 66.18 [62.98,69.24] 33.82 [30.76,37.02] Yes 65.23 [62.77,67.62] 34.77 [32.38,37.23] Woman’s religion 1.97 0.143 Christians 65.61 [63.28,67.87] 34.39 [32.13,36.72] Islam 64.35 [60.67,67.87] 35.65 [32.13,39.33] Tradition/no religion/others 73.63 [65.73,80.25] 26.37 [19.75,34.27] Alhassan et al. Malaria Journal (2022) 21:120 Page 12 of 17 Table 2 (continued) Experienced malaria in the past 12 months Rao Scott’s Chi‑ P‑value square No Yes % [95% CI] % [95% CI] Knowledge of malaria 7.03 0.002 Low knowledge 88.59 [80.26,93.69] 11.41 [6.31,19.74] Moderate knowledge 66.17 [62.50,69.65] 33.83 [30.35,37.50] Comprehensive knowledge 64.82 [62.50,67.08] 35.18 [32.92,37.50] Exposure to malaria messages in the past 34.07 < 0.001 6 months Not exposed 70.47 [67.96,72.87] 29.53 [27.13,32.04] Exposed 59.89 [57.03,62.69] 40.11 [37.31,42.97] ITN, insecticide treated net; IRS, indoor residual spraying; CI, confidence interval All percentages are row percentages Subgroup analysis of the impact of household access Access to both ITNs and IRS contributed to significant to ITNs and application of IRS on self‑reported malaria reduction in self-reported malaria prevalence in the cen- prevalence tral (ATE: − 25.77%, 95% CI [− 49.52, − 2.01], p = 0.034), Access to ITNs contributed to a significant reduc- Greater Accra (ATE: − 10.84%, 95% CI [− 18.40, − 3.28], tion in self-reported malaria prevalence in the central p = 0.005), Volta (ATE: −  15.04%, 95% CI [−  22.18, (ATE: −  8.71%, 95% CI [−  16.49, −  0.92], p = 0.029), −  7.90], p < 0.001), the Eastern (ATE: −  23.54%, 95% Greater Accra (ATE: − 6.49%, 95% CI [− 11.14, − 1.79], CI [−  35.43, −  11.65], p < 0.001) and the Ashanti (ATE: p = 0.007), Volta (ATE: −  6.33%, 95% CI [−  10.51, − 29.34%, 95% CI [− 56.51, − 2.18], p = 0.034) regions. − 2.15], p = 0.003), and the Eastern (ATE: − 7.89%, 95% Also, access to both ITNs and IRS contributed to signifi- CI [− 13.66, − 2.07], p = 0.008) regions. Also, access to cant reduction in both the urban (ATE: − 24.22%, 95% CI ITNs contributed over 7% significant reduction in both [−  32.65, −  15.78], p < 0.001) and the rural areas (ATE: the urban (ATE: −  7.14%, 95% CI: [−  12.13, −  2.14], − 30.94%, 95% CI [− 39.66, − 22.22], p < 0.001). All the p = 0.005) and the rural areas (ATE: −  7.88%, 95% CI other subgroups of the household characteristics and [− 13.60, − 2.16], p = 0.007). All other subgroups of the women individual characteristics also showed varying household characteristics and women individual char- significant reduction in self-reported malaria prevalence acteristics also showed varying significant reduction in among women with access to both ITNs and IRS rang- self-reported malaria prevalence among women with ing from over 11% among women with low knowledge access to ITNs ranging from over 2% reduction among on malaria (ATE: −  11.69, 95% CI [−  21.42, −  1.96], women with low knowledge on malaria (ATE: − 2.67%, p = 0.019) to over 36% reduction among women living 95% CI: [− 5.53, − 0.02], p = 0.048) to over 8% reduc- in household with no access to electricity (ATE: − 36.96, tion among women with more than 4 births (ATE: 95% CI [−  52.52, −  21.40], p < 0.001)(Figs. 3 and 4, and −  8.92%, 95% CI [−  15.69, −  2.15], p = 0.010) (Figs.  3 Additional file 1: Table S1). and 4, and Additional file 1: Table S1). Access to IRS contributed to significant reduction in Discussion self-reported malaria prevalence in the Greater Accra The package of vector controlled preventive strategy (ATE: − 4.10%, 95% CI [− 7.37, − 0.83], p = 0.014), Volta for malaria contributed to significant reduction in self- (ATE: − 7.29%, 95% CI [− 12.78, − 1.81], p = 0.009), and reported malaria prevalence among women of repro- the Eastern (ATE: −  8.20%, 95% CI [−  14.89, −  1.52], ductive age in Ghana. Access to both ITNs and IRS p = 0.016) regions. Also, access to IRS contributed to among women recorded a 27% reduction in self-reported over 8% significant reduction in both the urban areas malaria prevalence. This finding is consistent with the (ATE: − 8.35%, 95% CI [− 14.96, − 1.75], p = 0.013) and results from a randomized controlled trial which showed the rural areas (ATE: − 8.30%, 95% CI [− 14.64, − 1.96], evidence of significant reduction in malaria RDT positiv- p = 0.011). Results of the impact of IRS on self-reported ity among IRS users compared to non-IRS users in a high malaria reduction among women by both household malaria endemic but high standard ITNs access area in characteristics and women individual characteristics are Mozambique [19]. In Northern Tanzania, the combina- shown in Figs. 3 and 4, and Additional file 1: Table S1. tion of ITNs and IRS recorded a significant reduction A lhassan et al. Malaria Journal (2022) 21:120 Page 13 of 17 Table 3 The impact of malaria control interventions on self-reported malaria prevalence among women in the past 12 months Intervention arm vs. Non‑ Impact estimates of the malaria interventions on self‑reported malaria prevalence among women aged 15–49 years intervention arm Poisson regression model Sensitivity analysis Binary logistic regression model Probit regression model Linear regression model ATE [95% CI] P‑value ATE [95% CI] P‑value ATE [95% CI] P‑value ATE [95% CI] P‑value Access to ITNs vs. No access − 7.05 [− 11.96, − 2.14] 0.005 − 7.88 [− 13.14, − 2.62] 0.004 − 7.16 [− 12.26, − 2.07] 0.006 − 7.39 [− 12.60, − 2.17] 0.006 to ITNs Household sprayed (IRS) vs. − 6.81 [− 13.06, − 0.55] 0.033 − 6.36 [− 13.03, 0.32] 0.062 − 7.34 [− 14.10, − 0.58] 0.033 − 5.99 [− 12.20, 0.23] 0.059 Household not sprayed Access to ITNs & IRS vs. − 6.88 [− 14.69, 0.93] 0.084 − 6.83 [− 14.61, 0.94] 0.085 − 7.68 [− 15.63, 0.28] 0.059 − 6.41 [− 13.86, 1.03] 0.091 Access to ITNs only Access to ITNs & IRS vs. − 4.70 [− 9.76, 0.37] 0.068 − 4.12 [− 8.15, − 0.09] 0.045 − 3.25 [− 7.09, 0.59] 0.095 − 19.08 [− 38.97, 0.80] 0.060 Access to IRS only Access to ITNs & IRS vs. No − 27.09 [− 34.94, − 19.25] < 0.001 − 27.99 [− 35.58, − 20.41] < 0.001 − 28.66 [− 36.33, − 21.00] < 0.001 − 27.12 [− 35.62, − 18.63] < 0.001 access to ITNs & IRS Household characteristics (region, residence, household size, sex of household head, age of household head, household wealth quintile, household water source, toilet facility, access to electricity, main floor materials, main wall materials, main roof material and type of cooking fuel) and women individual characteristics (age of woman, highest education, religion, number of children, currently pregnant, covered by health insurance, knowledge on malaria and exposure to malaria message in the past 6 months) were controlled for ATE, average treatment effect. Percentage difference in malaria prevalence (Intervention − No Intervention); CI, confidence interval; ITNs, insecticide treated nets; IRS, indoor residual spraying Alhassan et al. Malaria Journal (2022) 21:120 Page 14 of 17 Fig. 3 Impact of access to malaria intervention on self-reported malaria prevalence among women by household characteristics in the Anopheles density and entomological inoculation found that women living in households sprayed against rates [20]. The finding on an integrated vector-controlled mosquitoes or treated with indoor residual insecticide preventive strategy for malaria is further supported by a recorded a 6.8% absolute reduction in self-reported community-based survey conducted in Nyanza province malaria prevalence. This was also consistent with find- in Western Kenya which found that the combination of ings from a district-level observational study in the indoor residual spraying and insecticide-treated nets northern region of Ghana in which there was a 39%, provided added protection against malaria compared 26% and 58% relative reduction in confirmed malaria with insecticide-treated nets alone [21]. cases in 2015, 2016 and 2017 respectively among IRS There was a 7% reduction in reported malaria prev- campaigned districts compared to non-IRS campaigned alence among women with access to ITNs with a 95% districts [24]. In another study in the Bunkpurugu- confidence reduction of 2–12% from this study. Com- yunyoo district in northern region of Ghana, there was parable results were reported from a trend of malaria an estimated 5% marginal decline in asexual parasitae- cases in health sentinel sites in Papua New Guinea mia prevalence among children from 52% in November which also recorded a reduction in malaria cases 2010 to 48% in October 2012 during a high transmis- because of the repeated distribution of long-lasting sion season after application of alpha-cypermethrin IRS insecticidal nets [22]. In the Tombel Health District, between the two periods. There was a further decline in Southwest region of Cameroon, the distribution of malaria parasitaemia prevalence from 48% in October ITNs recorded a short-lived reduction of malaria cases 2012 to 20.6% in October 2013 after pirimiphos-methyl from three health facilities in 2012 (22.7%) following IRS application [25]. the distribution of ITNs compared to post-distribu- This study estimated that 34% of women had malaria tion cases in 2010 (26.7%) and 2011 (30.7%). However, episode 12  months before the survey with a 95% con- the cases recorded an increase to 29.5% in 2013 from fidence interval estimate of 32–36%. The prevalence of 22.7% in 2012 [23]. Comparable results were recorded self-reported malaria episode among women living in a for IRS alone. For example, compared to no IRS, we household with access to ITNs (33%) was significantly A lhassan et al. Malaria Journal (2022) 21:120 Page 15 of 17 Fig. 4 Impact of access to malaria intervention on self-reported malaria prevalence among women by women characteristics lower than women living in households with no access had the intervention should they not have the interven- to ITNs (39%). Similarly, the prevalence of self-reported tion as well as those whose households did not have the malaria among women living in a household that intervention should they have. had been sprayed against mosquitoes was 32% com- Secondly, access to ITNs does not necessarily imply pared to 35% in household that had not been sprayed. utilization of ITNs, therefore, care must be taken in Unimproved toilet facilities and poor sanitary condi- the interpretation of results and conclusions from this tions mostly includes open spaces where dirty water study. Thirdly, the outcome for this study self-reported is stagnated. These stagnated dirty water bodies are malaria was a proxy to malaria prevalence among the optimal environment for breeding the anopheles’ women in the past 12  months, hence could be biased mosquitoes which is the main vector for transmitting by the knowledge level of the women on malaria, espe- malaria in Ghana. Unimproved toilet facilities and sani- cially the unconfirmed positive cases. tary condition, therefore, directly lead to increase in Finally, the study did not account for multiple epi- community spread of malaria. Efforts towards the pro- sodes of malaria cases per participants within the vision of improved toilet facilities and sanitations in 1-year reference period as well as the exact timing of households and communities should be strengthen. having the malaria episode and the interventions. Given the limitations of the observational study, a more Study limitations robust randomized controlled trial would be an impor- This study had several important limitations. First, the tant consideration for future research study. study used data from a cross-sectional survey which makes it difficult to measure causality. To overcome this limitation, causal inference statistical methodolo- Conclusion gies was used to estimate average treatment effects of Households with access to both ITNs and IRS had a the interventions. The analysis adjusted for the treat- lower prevalence of self-reported malaria compared to ment assignment with important variables in estimat- households with none of the two interventions. This ing potential outcomes of women whose households finding confirms the call for integrating malaria control Alhassan et al. Malaria Journal (2022) 21:120 Page 16 of 17 interventions to facilitate attainment of malaria elimi- Received: 15 October 2021 Accepted: 22 March 2022 nation in Ghana. Abbreviations References DHS: Demographic health survey; GMIS: Ghana malaria indicator survey; IRS: 1. WHO. World Malaria Report 2019. Geneva, World Health Organization. Indoor residual spraying; ITNs: Insecticide-treated nets; LLINs: Long-lasting 2019. https:// www.w ho.i nt/ publi catio ns-d etail/ world- malari a-r eport- insecticidal nets; MIS: Malaria indicator survey; NMCP: National Malaria Control 2019. Accessed 2 Mar 2021. Programme; WHO: World Health Organization. 2. WHO. Malaria report 2018. Geneva: World Health Organization; 2018. 3. WHO. Mission vision immunization vaccines 2015–2030. Geneva: World Supplementary Information Health Organization; 2015. 4. USAID. President’s malaria initiative. President’s malaria initiative—Ghana The online version contains supplementary material available at https://d oi. country profile. Ghana malaria operational plan FY 2018. 2018. p. 1–2. org/ 10.1 186/ s12936- 022- 04136-3. https:// www. pmi. gov/ docs/ defau lt- source/ defau lt- docum ent- libra ry/ count ry- profi les/ ghana_ profi le. pdf? sfvrsn=2 2. Accessed 2 Mar 2021. Additional file 1: Table S1. Sub analysis of the impact of malaria inter- 5. Scates SS, Finn TP, Wisniewski J, Dadi D, Mandike R, Khamis M, et al. Costs vention on self-reported malaria prevalence among women by household of insecticide-treated bed net distribution systems in sub-Saharan Africa. and women characteristics. Malar J. 2020;19:105. 6. White MT, Conteh L, Cibulskis R, Ghani AC. Costs, and cost-effective- ness of malaria control interventions—a systematic review. Malar J. Acknowledgements 2011;10:337. We would like to thank the Ghana statistical services, the DHS program for 7. Smith Paintain L, Awini E, Addei S, Kukula V, Nikoi C, Sarpong D, et al. organizing and conducting the 2016 Ghana Malaria Indicator Survey. We also Evaluation of a universal long-lasting insecticidal net (LLIN) distribution thank the entire Biostatistics and Health Policy, Planning and Management campaign in Ghana: cost effectiveness of distribution and hang-up activi- departments of School of Public Health from the University of Ghana for build- ties. Malar J. 2014;13:71. ing the capacity of the first author. 8. Ghana Statistical Service (GSS), Ghana Health Service (GHS), ICF. Ghana malaria indicator survey 2016. Accra, Ghana and Rockville, Maryland, USA. Authors’ contributions 2017. DD, YA and KA developed and designed the concept of the study. YA, DD and 9. Ghana Statistical Service (GSS), Ghana Health Service (GHS), ICF Inter- KA performed statistical analysis. The discussions section of the manuscript national. Ghana demographic health survey, 2014. Accra, Ghana and was done by all authors. HB, SAA, SB, JN, AEY, YA, DD, MT, MK reviewed the Rockville, Maryland, USA. 2015. manuscript critically for intellectual content. All authors read and approved 10. Agegnehu F, Shimeka A, Berihun F, Tamir M. Determinants of malaria the final manuscript. infection in Dembia district, Northwest Ethiopia: a case-control study. BMC Public Health. 2018;18:480. Funding 11. Birhanu Z, Yihdego YY, Emana D, Feyissa D, Kenate S, Kebede E, et al. No funding was received for this study. Relationship between exposure to malaria and haemoglobin level of children 2–9 years old in low malaria transmission settings. Acta Trop. Availability of data and materials 2017;173:1–10. The GMIS data is available online at no cost at the DHS portal. It can be access 12. Lindblade KA, Mwandama D, Mzilahowa T, Steinhardt L, Gimnig J, Shah M, through the website https://d hspr ogram.c om/ data/ upon request [26]. et al. A cohort study of the effectiveness of insecticide-treated bed nets to prevent malaria in an area of moderate pyrethroid resistance, Malawi. Declarations Malar J. 2015;14:31. 13. Kanyangarara M, Hamapumbu H, Mamini E, Lupiya J, Stevenson JC, Mharakurwa S, et al. Malaria knowledge and bed net use in three trans- Ethical approval and consent to participate mission settings in southern Africa. Malar J. 2018;17:41. The Ghana MIS survey sought ethical approval from relevant institutions 14. Nkoka O, Chipeta MS, Chuang YC, Fergus D, Chuang KY. A comparative before data collection. Consent from all relevant individuals were also sought study of the prevalence of and factors associated with insecticide-treated from the field during data collection. This study required no consent from nets usage among children under 5 years of age in households that participants as it had no direct contact from the interviewed participants. already own nets in Malawi. Malar J. 2019;18:43. However, approval for the use of the GMIS data as secondary data was sought 15. Hetzel MW, Gideon G, Lote N, Makita L, Siba PM, Mueller I. Ownership, from the Demographic Health Survey (DHS) program through their online and usage of mosquito nets after four years of large-scale free distribu- portal using the students DHS account. tion in Papua New Guinea. Malar J. 2012;11:192. 16. Fokam EB, Kindzeka GF, Ngimuh L, Dzi KTJ, Wanji S. Determination of Consent for publication the predictive factors of long-lasting insecticide-treated net ownership Not applicable. and utilisation in the Bamenda Health District of Cameroon. BMC Public Health. 2017;17:263. Competing interests 17. Tugume A, Muneza F, Oporia F, Kiconco A, Kihembo C, Kisakye AN, et al. The authors declare that they have no competing interests. Effects and factors associated with indoor residual spraying with Actellic 300 CS on malaria morbidity in Lira District, Northern Uganda. Malar J. Author details 1 2019;18:44. Department of Health Policy, Planning and Management, School of Public 18. Sakeni M, Khorram A, Majdzadeh R, Raiesi A. Indoor residual spraying Health, University of Ghana, Accra, Ghana. 2 Department of Biostatistics, School 3 coverage and acceptability rates to control malaria and the householders’ of Public Health, University of Ghana, Accra, Ghana. Department of Physician reasons of acceptance or rejection of spraying, in South-East of Iran. Int J Assistantship and Public Health, School of Medicine and Health Sciences, Cen- Infect. 2015;2:192. tral University College, Accra, Ghana. 4 Department of Epidemiology, School 5 19. Chaccour C, Zulliger R, Wagman J, Casellas A, Nacima A, Elobolobo E, et al. of Public Health, University of Ghana, Accra, Ghana. Department of Com- Incremental impact on malaria incidence following indoor residual spray- munity Health, University of Ghana Medical School, University of Ghana, Accra, ing in a highly endemic area with high standard ITN access in Mozam- Ghana. bique: results from a cluster—randomized study. Malar J. 2021;20:84. A lhassan et al. Malaria Journal (2022) 21:120 Page 17 of 17 20. Protopopoff N, Wright A, West PA, Tigererwa R, Mosha FW, Kisinza W, et al. Combination of insecticide treated nets and indoor residual spraying in Northern Tanzania provides additional reduction in vector population density and malaria transmission rates compared to insecticide treated nets alone: a randomised control trial. PLoS ONE. 2015;10:e0142671. 21. Hamel MJ, Otieno P, Bayoh N, Kariuki S, Were V, Marwanga D, et al. The combination of indoor residual spraying and insecticide-treated nets provides added protection against malaria compared with insecticide- treated nets alone. Am J Trop Med Hyg. 2011;85:1080–6. 22. Rodriguez-Rodriguez D, Maraga S, Lorry L, Robinson LJ, Siba PM, Mueller I, et al. Repeated mosquito net distributions, improved treatment, and trends in malaria cases in sentinel health facilities in Papua New Guinea. Malar J. 2019;18:364. 23. Fokam EB, Dzi KTJ, Ngimuh L, Enyong P. The effect of long-lasting insec- ticide bed net use on malaria prevalence in the Tombel Health District, southwest region-Cameroon. Malar Res Treat. 2016;2016:e3216017. 24. Gogue C, Wagman J, Tynuv K, Saibu A, Yihdego Y, Malm K, et al. An obser- vational analysis of the impact of indoor residual spraying in Northern, Upper East, and Upper West Regions of Ghana: 2014 through 2017. Malar J. 2020;19:242. 25. Abuaku B, Ahorlu C, Psychas P, Ricks P, Oppong S, Mensah S, et al. Impact of indoor residual spraying on malaria parasitaemia in the Bunkpurugu- Yunyoo District in northern Ghana. Parasit Vectors. 2018;11:555. 26. The DHS Program—Data. https:// dhspro gram. com/d ata/. Accessed 25 Oct 2020. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced rese archers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions