Hospital use by an ageing cohort: an investigation into the association between biological, behavioural and social risk markers and subsequent hospital utilization.

BACKGROUND: The aims of the study were to describe the pattern of hospital utilization (acute and mental health sectors) of the Paisley-Renfrew MIDSPAN cohort and assess the influence of biological, behavioural and social 'risk factors' (established at the time of screening) on subsequent hospital admissions.

METHOD: A cohort analysis was carried out in Paisley and Renfrew, two post-industrial towns in West Central Scotland. This used a linked data set covering a 23 year follow-up period to combine original 'risk'-related data with subsequent routine hospital admissions data. The subjects were 8349 women and 7057 men, aged 45-64 in the early to mid-1970s, and representing approximately 80 per cent of the eligible population. The main outcome measures were patterns of hospital utilization (acute and mental health sectors), 'any acute hospital admission', 'a serious acute hospital admission' and 'death' (relative risks of each outcome were calculated for all risk factors).

RESULTS: The following patterns of hospital utilization were found. Only 5 per cent experienced a mental health admission but mean stay was long (265 bed days per cohort member admitted). In contrast, 79 per cent experienced at least one acute hospital stay. The age-specific proportions of cohort members requiring admission increased over time but the growth in acute episodes was even higher (suggesting increasing rates of multiple admission). For non-survivors, 42 per cent of all acute episodes (55 per cent of bed days) took place during the 12 months before death. Analysis of risk factors (using Cox's proportional hazards model) of 'any admission' and 'a serious admission' showed forced expiratory volume (FEV1), age, sex, smoking status, blood pressure, blood sugar, body mass index, cholesterol and deprivation category to be important predictors.

CONCLUSIONS: Despite the desirability of alternative settings of care for the chronically ill and dying, a high proportion of hospital bed days were required near the time of death. The absolute size of the demand for hospital services within the cohort was strikingly large and increasing over time. Strategies to address the tide of rising admissions will have to confront the increasing proportion of individuals requiring admission as well as the growth in multiple admissions. Those who were at higher risk of admission were the older members of the cohort (especially men), those with low FEV1, smokers, those who were underweight or obese, the small number with abnormal levels of blood sugar, those with high blood pressure and those who lived in the most deprived areas. Thus, programmes which affect these determinants of ill health may be useful in reducing age-specific admission rates.