Objective. To explore the relationship between functional status and physic
ian cost (general practitioner/specialist) in an elderly population.
Design, setting and participants. A longitudinal study involving 328 patien
ts aged 65 years or over admitted to medical and surgical wards of a Sydney
metropolitan hospital over a 10-month period.
Main outcome measures. Two predictive cost models were developed using mult
iple linear regression analyses. Nine predictors were modelled including fu
nctional status (Short Form 36; SF-36) and major diagnostic categories. The
se models were then applied to the Australian SF-36 norms to produce a prof
ile of cost by level of functioning.
Results. After adjusting for potential confounders, five variables were fou
nd to be predictive of general practitioner cost at a 5% significance level
. Females and age were positively associated, whereas case note mention of
post-discharge services and high SF-SG vitality and role emotional scores w
ere negatively predictive. For specialist cost, five variables were statist
ically significant. The SF-36 domains of physical functioning and mental he
alth were positively associated. Higher vitality, role emotional scores and
case note mention of post-discharge services were negatively associated.
Conclusions. Cost models can be used to highlight the differences between g
eneral practitioner and specialist attendances, guide future physician care
of the aged, and facilitate informed decision making.