Assessing the impact of building footprint dataset choice for health programme planning: a case study of indoor residual spraying (IRS) in Zambia

Abstract

Background: The increasing availability globally of building footprint datasets has brought new opportunities to support a geographic approach to health programme planning. This is particularly acute in settings with high disease burdens but limited geospatial data available to support targeted planning. The comparability of building footprint datasets has recently started to be explored, but the impact of utilising a particular dataset in analyses to support decision making for health programme planning has not been studied. In this study, we quantify the impact of utilising four different building footprint datasets in analyses to support health programme planning, with an example of malaria vector control initiatives in Zambia.

Methods: Using the example of planning indoor residual spraying (IRS) campaigns in Zambia, we identify priority locations for deployment of this intervention based on criteria related to the area, proximity and counts of building footprints per settlement. We apply the same criteria to four different building footprint datasets and quantify the count and geographic variability in the priority settlements that are identified.

Results: We show that nationally the count of potential priority settlements for IRS varies by over 230% with different building footprint datasets, considering a minimum threshold of 25 sprayable buildings per settlement. Differences are most pronounced for rural settlements, indicating that the choice of dataset may bias the selection to include or exclude settlements, and consequently population groups, in some areas.

Conclusions: The results of this study show that the choice of building footprint dataset can have a considerable impact on the potential settlements identified for IRS, in terms of (i) their location and count, and (ii) the count of building footprints within priority settlements. The choice of dataset potentially has substantial implications for campaign planning, implementation and coverage assessment. Given the magnitude of the differences observed, further work should more broadly assess the sensitivity of health programme planning metrics to different building footprint datasets, and across a range of geographic contexts and health campaign types.

Keywords: Building footprints; Geospatial; Malaria; Microplanning; Satellite imagery.

Fig. 1

The count of building footprints per province, for each building footprint data product. Total counts are displayed as solid bars. The subset of “potentially residential” building footprints (≥ 9 m² and ≤ 330 m²) are indicated by the striped fill

Fig. 2

Mean counts of potentially residential building footprints per GRID3 v2.0 settlement extent, shown for each dataset and stratified by province and L1 degree of urbanisation

Fig. 3

The count of settlements in classes A–D, based on the count of potentially residential building footprints per settlement (upper row: GRID3 v2.0 settlement extents, lower row: derived settlement clusters), for each building footprint dataset. Derived settlement clusters are formed based on the presence of building footprints and so no Class A settlements are produced with this method. Note that the y-axis values differ between classes A–C and class D

Fig. 4

Mapped examples of class D settlement shown for two example locations (A and B). For these locations, the class D settlement extents (upper) and class D derived settlement clusters (lower) are shown for each building footprint dataset (left to right: Ecopia, Google, Microsoft and OSM). Three example locations with clear differences between datasets are marked as X, Y and Z

Fig. 5

Cumulative count of Class D settlement extents stratified by the count of building footprint datasets in agreement (n = 1–4). Within each stratum (X-axis), the total count of class D settlements is labelled by the combination of building footprint datasets (E = Ecopia year 2, G = Google v3, M = Microsoft (Oct. 2023 download) and O = OSM (Oct. 2023 download))

Fig. 6

For all settlement extents classified as class D with one or more building footprint datasets (n = 35,861), a the percentage that are classified as class D with each dataset, and b the percentage classified as class D with 1–4 datasets, is mapped at district level

Fig. 7

The count of class D settlement extents selected with each building footprint dataset, stratified by province, and the L1 degree of urbanisation class. Note that the y-axis values vary between the L1 degree of urbanisation strata