Abstract

Warning View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

For this article...

Background
Keywords
Findings
Authors
What is already known on this subject?
What does this study add?

Background

Many studies of medicine use rely on self-reports. Based on pharmacy claims data, this analysis tests whether such self-reports constitute a valid and reliable data source.

Data and Methods

Linked data from the Canadian Community Health Survey and the Ontario Drug Benefit Program were used to estimate the agreement, based on kappa statistics, between seniors' self-reported medication use and the claims data. Health, demographic and socio-economic factors associated with the likelihood of agreement were modeled with logistic regression.

Results

The prevalence of antihypertensive medication use among Ontario residents aged 65 or older was about 40% in 2001, based on both self-report and pharmacy claims, and in 2005, it was 52% for self-report and 49% based on claims data. The prevalence of oral diabetes medication use was comparable between the two data sources. Overall agreement between self-reported and claims data was "good" to "very good" for oral diabetes medications (kappa = 0.79 in 2001; 0.87 in 2005), but "moderate" for antihypertensive medications (kappa = 0.46 in 2001; 0.55 in 2005). Agreement improved somewhat from 2001 to 2005, with implementation of a more targeted survey question.

Interpretation

Self-reports appear to be an accurate data source for measuring medication use; however, for antihypertensive medications, self-reports by the oldest and sickest subpopulations should be used cautiously.

Keywords

Aged, antihypertensive agents, diabetes, drug prescriptions, drug utilization, health surveys

Findings

Information on prescription drug use for populationlevel research is available from community surveys and from administrative data. While each data source offers advantages and disadvantages for the investigation of medication use, the two sources are rarely compared. The aim of this article is to examine the agreement between two sources of drug utilization data available for Ontario. [Full Text]

Author

Sara Allin (1-416-549-5413; sara.allin@utoronto.ca), Audrey Laporte and Ahmed M. Bayoumi are with the University of Toronto. Michael R. Law is with the University of British Columbia. Sara Allin is also with the Canadian Institute for Health Information. Sara Allin and Ahmed M. Bayoumi are also with the Centre for Research on Inner City Health, the Keenan Research Centre in the Li Ka Shing Knowledge Institute of St. Michael's Hospital.

What is already known on this subject?

  • Information on prescription drug use is available from community surveys and from administrative data.
  • Many studies use self-reported information on medication use, but no study in Canada has compared self-reported data on medication use with pharmacy claims data.
  • Results of earlier research suggest that the accuracy of self-reported medication use is affected by the design of the surveys, the drug classes investigated, and the characteristics of respondents.

What does this study add?

  • For the household population aged 65 or older, this study found "good" to "very good" agreement between Ontario Drug Benefit claims data and Canadian Community Health Survey (CCHS) data for oral diabetes medications, and "moderate" agreement for antihypertensive medications.
  • Agreement for both drug classes was higher based on the more targeted 2005 CCHS question than on the 2001 question.
  • The odds of agreement between data sources for antihypertensive medications were lower among older seniors and those in poorer health.
  • Self-reports generally appear to be an accurate data source for measuring medication use; however, for antihypertensive medications, self-reported data (particularly reports from the oldest and sickest subpopulations) should be used cautiously.