Primary care doctor and nurse consultations among people who live in slums: a retrospective, cross-sectional survey in four countries

Objectives 

To survey on the availability and use of primary care services in slum populations. 

Design 

Retrospective, cross-sectional, household, individual and healthcare provider surveys. 

Setting 

Seven slum sites in four countries (Nigeria, Kenya, Pakistan and Bangladesh). 

Participants 

Residents of slums and informal settlements. 

Primary and secondary outcome measures 

Primary care consultation rates by type of provider and facility. 

Results 

We completed 7692 household, 7451 individual adult and 2633 individual child surveys across seven sites. The majority of consultations were to doctors/nurses (in clinics or hospitals) and pharmacies rather than single-handed providers or traditional healers. Consultation rates with a doctor or nurse varied from 0.2 to 1.5 visits per person-year, which was higher than visit rates to any other type of provider in all sites except Bangladesh, where pharmacies predominated. Approximately half the doctor/nurse visits were in hospital outpatient departments and most of the remainder were to clinics. Over 90% of visits across all sites were for acute symptoms rather than chronic disease. Median travel times were between 15 and 45 min and the median cost per visit was between 2% and 10% of a household's monthly total expenditure. Medicines comprised most of the cost. More respondents reported proximity (54%-78%) and service quality (31%-95%) being a reason for choosing a provider than fees (23%-43%). Demand was relatively inelastic with respect to both price of consultation and travel time. 

Conclusions 

People in slums tend to live sufficiently close to formal doctor/nurse facilities for their health-seeking behaviour to be influenced by preference for provider type over distance and cost. However, costs, especially for medicines are high in relation to income and use rates remain significantly below those of high-income countries.