Analysis of costs borne by families of patients hospitalized for stroke.

BACKGROUND

Studies focusing on the economic impact of cancer on families have emphasized that costs of chronic disease are substantial for patients and their families. However, little effort has been devoted to measuring the costs of care for families of patients hospitalized with stroke.

METHODS

A total of 215 stroke patients and their families from four teaching hospitals in the Taipei metropolitan area were monitored from the date of the patient's admission to hospital until the date of discharge. The value of labor contributed by families was estimated by assigning the current monetary market rate of providing health aide to the time families spent caring for patients in hospital. Lost earnings of patients and families, expenditure for medical care, and expenses for food, clothes, adult diapers, transportation and other miscellaneous items were determined and summed to arrive at the total family cost of providing care.

RESULTS

The average cost of care for one family per inpatient day was NT$4,358.20. A total of 98.6% of the families incurred labor costs, which accounted for about half of family costs for providing care. Hospital bills accounted for almost 19% of total family costs. The income loss for families and patients accounted for about 25% of total family costs. Expenses for food, clothes, transportation, diapers and other illness-related miscellaneous items accounted for about 12% of total family costs. Multiple regression analyses demonstrated that the number of family members involved in giving care and the length of stay are important predictors for the total cost of care. Average total family costs per day increased by 24.3% when an additional family member was involved in providing care. Total family costs increased 2.5% for each hospital day.

CONCLUSIONS

If direct and indirect nonmedical costs are not included in the total cost calculation for providing hospital care to stroke patients, the economic impact of care on families is likely to be underestimated.