Adverse drug reactions in an elderly hospitalised population: inappropriate prescription is a leading cause

Drugs Aging. 2005;22(9):767-77. doi: 10.2165/00002512-200522090-00005.


Objective: Adverse drug reactions (ADRs) represent a major public health problem in the aged. In order to better evaluate this problem in Brazil, this study was designed to assess the prevalence of ADRs in an elderly hospitalised population, identify the most common ADRs and the principal medications involved, evaluate the appropriateness of use of these drugs in elderly people and determine the risk factors implicated in the appearance of such ADRs.

Methods: The study population was comprised of 186 elderly patients (> or =60 years of age) admitted to the internal medicine service of a teaching hospital. The patients were assessed by a single observer using an intensive drug surveillance method to identify and report ADRs. The degree of probability for each adverse reaction was determined using the Naranjo algorithm.

Results: The mean (+/- SD) age of the patients studied was 73.6 +/- 9.1 years. Up to 115 patients (61.8%) of the study population presented at least one ADR. A total of 199 ADRs were found, at an average of 1.7 per patient. The ADRs appeared during hospitalisation in 46.2% of the study population, were present at the time of the admission but did not cause hospitalisation in 17.2% of patients, and were the cause of admission in 11.3% of patients. The most frequent ADR that caused hospitalisation was digitalis toxicity (22.7% of such ADRs). Hypokalaemia as a result of diuretics was the most prevalent ADR both during hospitalisation (11.8%) and at the time of admission without being the cause (12.1%). Captopril was the most frequently prescribed drug (138 prescriptions), and led to an ADR in 18.1% of patients who received the drug. Almost a quarter of the patients presenting an ADR were prescribed drugs considered inappropriate for the elderly. By means of a multiple logistic regression model, the following were considered to be significant risk factors for the appearance of ADRs: number of diagnoses (odds ratio [OR] 1.40; 95% CI 1.06, 1.86), number of drugs (OR 1.07; 95% CI 1.01, 1.13) and use of drug that is inappropriate for the elderly (OR 2.32; 95% CI 1.17, 4.59).

Discussion: The main contribution of the present study was identification of use of drugs that are considered inappropriate for elderly populations as a major risk factor for presenting an ADR. This finding is useful for continuous education programmes, therapeutic committees and policy makers, because adverse effects complicate the course of diseases in aged patients, cause hospitalisation and/or require the prescription of additional drugs. In addition to contributing to a reduction in healthcare costs, continuous efforts to promote rational drug use could also benefit elderly patients by preventing some avoidable drug toxicity.

Conclusion: A significant prevalence of ADRs was found among hospitalised elderly people. The risk factors associated with ADRs in this population included use of drugs considered to be inappropriate for that population, number of previous diagnoses and number of administered drugs. More appropriate drug prescription could avoid part of this burden of disease by minimising preventable ADRs.

MeSH terms

  • Adverse Drug Reaction Reporting Systems* / statistics & numerical data
  • Aged
  • Aged, 80 and over
  • Anti-Inflammatory Agents, Non-Steroidal / adverse effects
  • Brazil
  • Captopril / adverse effects
  • Digitalis Glycosides / adverse effects
  • Drug Interactions
  • Drug Prescriptions*
  • Female
  • Hospitalization* / statistics & numerical data
  • Hospitals, Teaching*
  • Humans
  • Hypoglycemic Agents / adverse effects
  • Male
  • Medication Errors / statistics & numerical data
  • Middle Aged
  • Risk Factors


  • Anti-Inflammatory Agents, Non-Steroidal
  • Digitalis Glycosides
  • Hypoglycemic Agents
  • Captopril