[Potentially avoidable hospitalisation in Bologna, 1997-2000: temporal trend and differences by income level]

To describe the temporal trend of hospitalisations for Ambulatory Care Sensitive Conditions (ACSCs) from 1997 to 2000 in Bologna (Italy) and to analyze the association with the income level.

We have selected two panels of ACSCs: eight conditions for the paediatric/young population (<20 years of age) and fourteen for the adult population (> or =20 years of age). All discharges for ACSCs of residents in Bologna from Emilia-Romagna hospitals have been selected in the years 1997-2000. An indicator of social position was computed: the median per capita equivalent income by census block, obtained through record linkage between the Italian Tax Register (income earned in 1998) and the Population Register of Bologna.

The direct age-standardized rates and the rate ratio by income level have been calculated. The Poisson regression model has been used to calculate the relative risk (RR) of hospitalizations for ACSCs.

2359 (17.6% of the total) hospitalisations have been selected among the paediatric/young population and 27822 (11.1% of the total) among the adult population. The annual age-adjusted rate of ACSC is 122.68 per 10000 persons among children and 176.60 among adults. The hospitalisation forACSCs among children is associated with a middle-low level of income (RR 1.55; CI 95% 1.35-1.78 for the lowest level vs. highest level), male gender and age <5 years old. In the adult population the risk of hospitalisation for ACSCs is higher among those with lower levels of income (RR 1.80; CI 95% 1.66-1.95 for the lowest level vs. highest level), moreover the RR is higher for men at every age. The admissions forACSCs among adults show a decreasing temporal trend.

The disadvantaged groups of the population experience the highest risk of hospitalisation for ACSCs, with differences by gender and age groups. Although it is difficult to specifically identify the mechanisms potentially involved in the relationship between socioeconomic status and excess of hospitalisation, the admissions for ACSCs may represent an interesting indicator of quality ofprimary care.