Predictors for high costs of hospital care in elderly hypertensive patients

Abstract

Objective. To analyse predictors of high cost of care in elderly hypertensive patients, in particular costs related to short‐term (<10 days) and long‐term (⩽10 days) institutional care. Design. Health Economy (HE) sub‐study in the Swedish Trial in Old Patients with Hypertension‐2 (STOP Hypertension‐2). Setting. Outpatient clinics, hospitals and nursing homes in Sweden. Subjects. Elderly (70–84 years) patients (n = 6614) from the STOP Hypertension‐2 cohort with a systolic or diastolic hypertension, or a combination thereof, were included. Methods. Costs of institutional care were analyzed and categorized as short‐term (<10 days) or long‐term care (⩾10 days). Costs were related to individual patients and calculations were made during follow‐up in STOP Hypertension‐2 from inclusion to end of study. Data was available from 99% of all scheduled visits during the median 5.3 years of follow‐up in the 6614 elderly hypertensive patients. Results. A multivariate analysis of potential predictors for inpatient short‐term or long‐term care demonstrated that several clinical factors within the groups of target organ damage (TOD), associated clinical conditions (ACC), as well as additional risk factors (RF) predicted for an increased probability of inpatient care in elderly hypertensives. Specifically, predictors for short‐term care (<10 days) were: previous myocardial infarct (OR 1.50, p = 0.008), stroke (OR 1.41, p = 0.013), congestive heart failure (OR 1.73, p = 0.005), diabetes (OR 1.36, p<0.0005) and older age (OR 1.05, p<0.0001). Predictors at entry for long‐term care (⩾10 days) were; presence of ischaemic heart disease (OR 1.65, p<0.0001), diabetes mellitus (OR 1.32, p = 0.012), female gender (OR 0.80, p = 0.0003) as well as older age (OR 1.02, p = 0.046). High total costs for this cohort of elderly hypertensive patients were recorded in the group subjected to long‐term care for cardiovascular as well as non‐cardiovascular reasons. Male gender (p = 0.004) and stroke (p = 0.06) remained predictors for high costs for hospital care while stroke (p<0.0001) and old age (p<0.0001) predicted for high costs for nursing home care. Conclusion. In elderly hypertensives in STOP Hypertension‐2, presence of cardiac disease, stroke, diabetes and older age at entry increased the probability as well as costs for both short‐ and long‐term care. Level of systolic or diastolic blood pressure did not predict for hospitalization or cost outcome. Our results provide an economic argument for a strict risk reduction focus in the management of elderly high‐risk hypertensive patients.