Does area of residence influence the likelihood of hospitalisation for ambulatory care sensitive conditions in Ireland?

Problem

Conditions can be considered “ambulatory-care sensitive” if effective outpatient treatment could reasonably prevent hospitalisation. Geographical variation in hospitalisation for these conditions has previously been studied to evaluate equity of access to primary care. The aim of this study was to examine the extent of geographical variation of potentially avoidable hospitalisations for ambulatory care sensitive conditions in the Republic of Ireland. Our secondary aim was to explore potential causes for any variation identified.

Approach

Six ambulatory care sensitive conditions were defined based on work by the European Collaboration for Healthcare Optimization (ECHO) Consortium as follows: Asthma or chronic obstructive pulmonary disease (COPD) in patients aged ≥18 years, congestive heart failure (CHF) in patients aged ≥40, dehydration in patients aged ≥65, short-term complications of diabetes in patients aged ≥40, and angina without a cardiac procedure in patients aged ≥40. An anonymous dataset of emergency hospitalisations for these conditions over 5 years (2012-2016) was extracted from a national administrative database in the Republic of Ireland. Age and sex standardised discharge rates (SDRs) were calculated for 21 geographical areas nationally. Extremal quotients, coefficients of variation and systematic components of variance (SCV) were calculated. Regression analyses were conducted exploring the relationship between SDRs and year, unemployment rate, and the proportion of the population living in urban areas.

Findings

Between 2012 and 2016, across n=36 public hospitals, n=85,484 hospitalisations were included. This is an average annual rate of 49 hospitalisations per 10,000 adult population nationally. Hospitalisations for COPD accounted for the majority (55%) and overall 73% of hospitalisations were for patients aged ≥65. Based on SCV values, CHF showed low geographic variation, while asthma and COPD showed high variation in particular years. Geographic variation was high for hospitalisations for dehydration, high for diabetes complications and very high for angina. In multivariable analysis, higher unemployment at an area level was associated with higher standarised discharge rates for asthma and COPD. Higher rates of admission for diabetes complications were observed in urban areas, while higher rates of angina admissions were observed in rural areas.

Consequences

The average annual rate of hospitalisations overall for conditions under study is in line with previous European research, although COPD accounts for a higher proportion of these hospitalisations in the current study. Our findings suggest that there is significant geographical variation associated with management of chronic conditions, with angina showing particularly high variation. While there is currently an increased focus on the provision of national integrated primary care services for people with chronic diseases in Ireland, our results highlight the need for further research into local factors that may influence hospitalisation.