To Örebro University

BACKGROUND: Adverse drug reactions (ADRs) are common in elderly patients. There are various reasons for this, including age- and disease-related alterations in pharmacokinetics and pharmacodynamics as well as the common practice of polypharmacy. The decline in renal function in elderly patients may also predispose them to pharmacological ADRs (type A, augmented). Patients receiving home healthcare may be at even higher risk.

OBJECTIVES: To study ADRs as a cause of acute hospital admissions in a defined cohort of elderly patients (aged >or=65 years) registered to receive home healthcare services, with special reference to impaired renal function as a possible risk factor.

METHODS: This was a retrospective study of 154 elderly patients aged >or=65 years admitted to the emergency department of a university hospital in Stockholm, Sweden, in October-November 2002. Estimated creatinine clearance (eCL(CR)) was calculated from the Cockcroft-Gault formula, and estimated glomerular filtration rate (eGFR) by the Modification of Diet in Renal Disease (MDRD) equation. ADRs were defined according to WHO criteria. All medications administered to patients at admission and at discharge were collated. These and other data were collected from computerized hospital records.

RESULTS: ADRs were judged to contribute to or be the primary cause of hospitalization in 22 patients, i.e. 14% of 154 patients registered to receive home healthcare. Eleven of the 22 patients were women. All but one ADR were type A. Excessive doses or drugs unsuitable in renal insufficiency were present in seven patients in the ADR group compared with only four patients in the group without ADRs (p = 0.0001). Patients with ADRs did not differ significantly from those without ADRs in relation to age, plasma creatinine, eCL(CR), weight or number of drugs prescribed at admission. However, women with ADRs were significantly older than women without ADRs (mean +/- SD age 88.8 +/- 5.7 years vs 82.5 +/- 8.0 years, respectively; p = 0.014) and had significantly lower mean +/- SD eCL(CR) values (25.5 +/- 10.8 and 37.1 +/- 17.1 mL/min, respectively; p = 0.035). Median MDRD eGFR was significantly higher than median eCL(CR) (59 [range 6-172] mL/min/1.73 m2 vs 38 [range 5-117] mL/min, respectively; p = 0.0001).

CONCLUSIONS: In elderly patients registered to receive home healthcare, 14% of hospital admissions were primarily caused by ADRs. One-third of these ADRs were related to impaired renal function, generally in very old women. These ADRs may be avoided by close monitoring of renal function and adjustments to pharmacotherapy (drug selection and dose), particularly in very elderly women.