Hospitalization for ambulatory care sensitive conditions and the role of primary care in Italian regions

Background: Hospitalization may often be prevented by timely and effective outpatient care either by preventing the onset of an illness, controlling an acute illness or managing a chronic disease with an appropriate follow-up. The objective of the study is to examine the variability of hospital admissions within Italian regions for Ambulatory Care Sensitive Conditions (ACSCs), and their relationship with primary care supply. Methods: Hospital discharge data aggregated at a regional level collected in 2005 were analysed by type of ACS conditions. Main outcome measures were regional hospital admission rates for ACSCs. Negative binomial models were used to analyse the association with individual risk factors (age and gender) and regional risk factors (propensity to hospitalisation and prevalence of specific conditions). Non-parametric correlation indexes between standardised hospital admission rates and quantitative measures of primary care services were calculated. Results: ACSC admissions accounted for 6.6% of total admissions, 35.7% were classified as acute conditions and 64.3% as chronic conditions. Admission rates for ACSCs varied widely across Italian regions with different patterns for chronic and acute conditions. Southern regions showed significantly higher rates for chronic conditions and North-eastern regions for acute conditions. We found a significant negative association between the provision of ambulatory specialist services and standardised hospitalization rates (SHR) for ACS chronic conditions (r=-0.50; p= 0.02) and an inverse correlation among SHR for ACS acute conditions and the rate of GPs per 1,000 residents, although the latter was not statistically significant. Conclusions: In Italy, about 480,000 inpatient hospital admissions in 2005 were attributable to ACSCs. Even adjusting for potential confounders, differences in hospital admissions for ACSCs among Italian regions were found. Such differences can be appropriately used to assess the effectiveness and/ or appropriateness of the primary care provided within different regions.