Details

Autor(en) / Beteiligte

• Macharia, Peter M
• Ray, Nicolas
• Giorgi, Emanuele
• Okiro, Emelda A
• Snow, Robert W

Titel

Defining service catchment areas in low-resource settings

Ist Teil von

• BMJ global health, 2021-07, Vol.6 (7), p.e006381

Ort / Verlag

England: BMJ Publishing Group LTD

Erscheinungsjahr

2021

Quelle

MEDLINE

Beschreibungen/Notizen

• Correspondence to Dr Peter M Macharia; pmacharia@kemri-wellcome.org Summary box Defining an accurate, representative service catchment area is important for computing population denominators for disease mapping and efficient public planning, including health, education and social care. Disaggregating low-resolution census data to presumed high-resolution settlement patterns guided by satellite imagery and building footprints has increased in sophistication in recent years.8 The locations of schools, health facilities and other amenities have improved with crowd-sourced Global Positioning Systems initiatives, participatory mapping, improved national gazetteers, high-resolution satellite imagery and demand for such data sets.3 However, linking population to service delivery remains crude and ignores the complexities of service demand and/or supply mainly due to a lack of geocoded data on where the users of services reside in SSA. [...]women bypassed the nearest facility in Tanzania,9 Ghana,10 Kenya,11 Mozambique, India and Pakistan.12 As a result, it is common practice to have childhood immunisation coverage from routine data greater than 100% in administrative units that attract many people from neighbouring administrative units.13 14 This is likely due to a challenge of assigning population denominators to an administrative area instead of a well-defined service catchment area based on the actual users of a particular service. Gravity models assume that the flow of people from residential areas to service providers is proportional to the demand for services and inversely proportional to physical access.16 The demand for services is defined by morbidity, age, and social structure and the capacity of a facility.16 To be more realistic, this physical access can be estimated by modelling the travel time to the nearest provider, by accounting for travel factors and barriers mainly through cost distance algorithms and network analysis.17 The travel time is then binned to define a service catchment area1 based on either an arbitrary time threshold (figure 1) or known cut-offs, where available (eg, 2 hours for obstetrical complications).