A cross-sectional analysis of ITN and IRS coverage in Namibia in 2013

Malaria is a global public health concern, causing approximately 438,000 deaths, worldwide, in 2015 [1]. The World Health Organization (WHO) Africa Region experiences a disproportionately high burden of malaria, with 88% of global cases in 2015 occurring in the region [1]. Namibia is one of eight sub-Saharan African countries aiming to eliminate malaria, and intends to eliminate by 2020.

Interventions for malaria control and elimination include indoor residual spraying (IRS), insecticide-treated nets (ITNs) and long-lasting insecticide-treated nets (LLINs). These are effective tools for reducing the adult mosquito population density and longevity, and are therefore fundamental for interrupting transmission [2]. ITNs and LLINs have successfully reduced the risk of infection in a number of settings [3‐5], with up to 90% reductions in malaria transmission recorded following ITN implementation in some high-transmission settings [6]. High coverage of ITNs and IRS can both result in community-level protection [7, 8], highlighting the importance of high coverage and uptake of these interventions. There is also evidence to suggest that using IRS and ITNs in combination is more effective at reducing the vector population and interrupting transmission than ITNs alone [9, 10].

Malaria transmission in Namibia is heterogeneous. In 2013, it was estimated that 67% of Namibia’s population were living in the highest transmission areas [11]. Prevalence of malaria is highest in the northern regions that border Angola [12]. Namibia has experienced fluctuations in malaria incidence with reported cases rising from 4911 in 2013 [11] to 15,915 in 2014 [1], with two outbreaks occurring in 2016 and 2017 [13‐16]. Importantly, between 2000 and 2015, Namibia’s overall malaria incidence and mortality rates increased by over 20% [17], highlighting the need for an effective elimination programme.

Namibia’s 2010–2016 Malaria Strategic Plan (MSP) aimed to achieve at least 95% coverage with a combination of vector control interventions in all malaria endemic areas and identified transmission foci by 2013 [12]. However, the 2013 Namibia Demographic and Health Survey (DHS), a nationally-representative survey that collected data on IRS and ITN coverage, reported that only 24% of households had at least one ITN, and just 16% of households had received IRS during the previous 12 months [18]. By contrast, a governmental report indicated that IRS was successfully completed in the eight malaria regions, with 93% coverage of targeted households achieved by the end of January 2013 [19]. To understand these discordant findings, a detailed analysis of ITN and IRS coverage was conducted as a function of DHS data, malaria transmission patterns and government intervention zones across Namibia in 2013.

These detailed analyses indicate that the prevalence of IRS and ITN interventions for malaria in 2013 did not reflect governmental malaria intervention targets in Namibia. In this DHS sample of 9846 households representing 41,314 individuals, and malaria transmission intensity data, the prevalence of at least one intervention (ITN or IRS) was around 34% across Namibia, and 49% in the highest transmission regions in 2013. These analyses highlight the need to include quantitative monitoring of such interventions, to provide a framework to reliably evaluate intervention strategies for malaria.

Operational constraints for IRS delivery have been reported in previous years in Namibia [33]. These constraints included a lack of community acceptability, shortage of human resources, late payments of spray men and challenges in access due to rains and flooding [33]. High levels of community acceptability of ITNs have also historically been difficult to achieve [33]. The need for the government to address operational constraints, particularly human resource capacity to implement these interventions, has been identified [33] and may partly explain the low coverage observed in the DHS.

Low coverage of malaria control interventions is not unique to Namibia; this has also been observed in DHS reports for other southern African countries. The Zambia 2013–2014 DHS found ITN coverage to be 68% and IRS coverage to be 28% [34]. In Zimbabwe, 48% of households in 2015 had an ITN and 21% had IRS [35]. Angola had a lower coverage of interventions than reported by the Namibia DHS, with 7% of households with IRS and 35% that owned an ITN in 2011 [36]. ITN coverage was higher than IRS coverage in all of these countries, consistent with the findings in Namibia.

By contrast to the 2013 Namibia DHS Report and these findings, the Namibia Ministry of Health and Social Services Annual Report 2012/2013 stated that at least 93% of households received these interventions [19]. It was reported that across eight malaria regions, 669,578 households out of a targeted 719,412 structures received IRS by the end of January 2013 [19]. The report also stated that 87,900 ITNs were procured but only 7000 ITNs were distributed at the time of the Annual Report [19]. The DHS identified around 4300 ITNs owned by the survey households collectively, 49% of which were obtained in the previous 12 months but 37% of which were obtained more than 3 years prior to the survey (Additional file 1: Table S10).

The DHS was carried out at least 4 months following the end of the spray season and at the end of the malaria transmission season. As such, the survey was well timed to provide a nationally-representative estimate of the coverage of these interventions by the end of the transmission season.

One of the limitations of this analysis is that it uses data collected as part of a survey investigating a range of health and disease indices, not only malaria. Data were not collected on larviciding practices so it was not possible to include larviciding in vector control estimates. Further, differences in intervention coverage reported by the DHS and the Annual Report could be due to a lack of overlap between households surveyed in the DHS and the households reported to be sprayed in the Annual Report or targeted by the IRS programme. However, the DHS surveyed approximately 2% of all households recorded in the 2011 Population and Housing Census and was designed to be nationally representative. In this context, the DHS appropriately represents the Namibian population based on national census data.

The nature of data collection on vector control methods as part of the DHS may be a further limitation. Information on whether a household received IRS was obtained by asking a household member whether the household had been sprayed against mosquitoes in the last 12 months. It is possible that the household member who answered the question may not have remembered this event or may not have been present at the time of spraying, for example, which would result in under-reporting of IRS. It was also not possible to ascertain which specific households were targeted for spraying by the IRS programme; thus, these estimates may not reflect programme efforts in targeted areas. However, it was observed that, as well as regional transmission intensity, MSP target zones additionally explained the variation in ITN and IRS in the DHS data—suggesting that the DHS data reflect the coverage of government malaria interventions.

Information on ITN ownership and use was gathered from survey questions, in addition to an inventory conducted by the interviewer. The number of ITNs could have been underestimated if not all the ITNs originally distributed were shown to the interviewer, for example, or if the nets were discarded, sold, or used for other purposes, as has been reported elsewhere [37‐40]. The number of ITNs may also have been overestimated, as not all ITNs that were reported as part of the DHS were actually observed by the interviewer. The source of the ITNs is also unknown. Whilst these factors together may result in some uncertainty around the estimates of IRS and ITN coverage, they are unlikely to fully account for the difference between estimates provided by the DHS data and governmental reports.

Study inferences may also be limited by the classification of regions and EAs based on MAP data. MAP PfPR_2–10 values are predicted values and do not necessarily reflect actual levels of transmission intensity in 2013. Using MAP data, Zambezi is classified as being in the 1 to < 5% regional PfPR_2–10 category; however, other studies and reports have identified this as one of the higher risk regions in Namibia [12, 33, 41]. Re-categorizing Zambezi resulted in only a minor increase in ITN and IRS coverage of no more than 4% for any intervention in the highest transmission regions (Additional file 1: Table S1). PfPR_2–10 data were also analyzed at both EA and regional levels, with regional level analyses found to better explain variation in ITN and IRS distribution in the DHS data—suggesting that finer-scale geographical data on transmission intensity data does not explain these patterns of ITN and IRS distribution.

These findings indicate that the prevalence of IRS and ITN interventions for malaria in 2013 did not reflect governmental malaria intervention targets in Namibia. The WHO recommends that “Malaria control and elimination programmes should prioritize delivering either LLINs or IRS at high coverage and to a high standard rather than introducing the second intervention as a means of compensating for deficiencies in the implementation of the first” [42]. Given the relatively low malaria transmission in Namibia and the operational challenges of delivering vector control interventions, it will be relevant to identify the barriers to implement interventions or prioritize the implementation of a single intervention. As countries such as Namibia work towards malaria elimination, high coverage of vector control interventions will be critical, not only to reduce the incidence of malaria but also to prevent resurgence. Such efforts will require quantitative monitoring to assess implementation and provide a framework to reliably evaluate the effectiveness of these interventions and inform future strategies for malaria elimination.