Objective. To determine the prevalence of symptoms of depression and the factors aecting their presence in an elderly population at the start of rehabilitation. To assess changes in the severity of these symptoms during rehabilitation and the correlation between these changes and corresponding changes in the patient's functional state.
Design. A population-based prospective study.
Setting. A geriatric ward in a general university hospital in southern Israel.
Participants. Two hundred and seventy-six elderly patients hospitalized for physical rehabilitation, 150 following hip fracture (HF) and 126 after stroke.
Measurements. Symptoms of depression were measured by the Geriatric Depression Screening Scale (GDS). The functional state was assessed using the FIM scale. A broad spectrum of clinical, functional, social and demographic variables was measured using conventional tests. The association between the GDS and these variables was tested by stepwise multiple regression.
Results. One hundred and thirteen patients (41%) showed signs of depressions (GDS 4 10), with 12 (4%) patients having severe symptoms (GDS 4 20). No signi®cant dierence was found between HF and stroke patients in symptoms of depression. Only four of the 41 variables tested were found to be signi®cantly and independently associated with the GDS: pre-event functional state (b À0.311, p 5 0.001), the self-care component of the FIM scale on admission to the hospital (b À0.267, p 5 0.001), living alone (b 0.149, p 0.015) and impaired visual acuity (b 0.137, p 0.026). The total variance in GDS accounted for by these four variables (adjusted R-square) was 0.24. The severity of depression symptoms decreased signi®cantly during rehabilitation and the GDS at discharge was signi®cantly lower than on admission ( p 0.008). This change correlated signi®cantly with the corresponding change in functional state (R À0.15, p 0.03).
Conclusions. Symptoms of depression are common in elderly patients beginning rehabilitation. These symptoms are aected independently, and almost exclusively, by the functional state of the patient, both prior to the event and after its occurrence. The depressed condition improves towards the end of hospitalization and the degree of improvement is correlated with the corresponding change in the patient's functional state.