Written informed consent and selection bias in observational studies using medical records: systematic review

Abstract

Objectives To determine whether informed consent introduces selection bias in prospective observational studies using data from medical records, and consent rates for such studies. Design Systematic review. Data sources Embase, Medline, and the Cochrane Library up to March 2008, reference lists from pertinent articles, and searches of electronic citations. Study selection Prospective observational studies reporting characteristics of participants and nonparticipants approached for informed consent to use their medical records. Studies were selected independently in duplicate; a third reviewer resolved disagreements. Data extraction Age, sex, race, education, income, or health status of participants and non-participants, the participation rate in each study, and susceptibility of these calculations to threats of selection and reporting bias. Results Of 1650 citations 17 unique studies met inclusion criteria and had analysable data. Across all outcomes there were differences between participants and nonparticipants; however, there was a lack of consistency in the direction and the magnitude of effect. Of 161 604 eligible patients, 66.9% consented to use of data from their medical records. Conclusions Significant differences between participants and non-participants may threaten the validity of results from observational studies that require consent for use of data from medical records. To ensure that legislation on privacy does not unduly bias observational studies using medical records, thoughtful decision making by research ethics boards on the need for mandatory consent is necessary.

Large panel-survey data demonstrated country-level and ethnic minority variation in consent for health record linkage

“…The overall consent rate observed in the current study is consistent with findings from a recent systematic review which found that 8 out of 11 studies reviewed had proportions of consent greater than 72% [38], the other three having proportions of 53% or lower. We found the highest consent rate in the 16-24 age range, with little difference across the rest of the age spectrum, and although the association between age and consent in other studies has been inconsistent [38], this may reflect a greater willingness to share information secondary to greater exposure to and use of social media among this age group.…”

“…We found the highest consent rate in the 16-24 age range, with little difference across the rest of the age spectrum, and although the association between age and consent in other studies has been inconsistent [38], this may reflect a greater willingness to share information secondary to greater exposure to and use of social media among this age group. The study also confirms the positive association between poorer health and consent to record linkage [14,30], which may arise because individuals with poor health, who are utilising the health service, can readily appreciate the benefits of linkage or want to give something back to the service.…”

“…The potential for selection bias invalidating the study findings has also prompted some researchers to question the universal need for mandatory consent for linkage to medical and other records [38,56,57].…”

Aspectos eticos em estudos longitudinais: o caso do ELSA-Brasil

“…In this context, the requirement of consent to use identifi ed secondary data has been a hotly debated topic. 15,18,22 Subjects' consent to access functional and health records has been considered a fundamental condition to participate in the study, as this information is essential to investigate the outcomes 2 and, consequently, to fulfi ll the objectives of a longitudinal study. Therefore, the following sentence was included in the document's fi nal statement, after the general agreement with the conditions to participate in the study: "I state that I authorize the researchers of the Brazilian Longitudinal Study for Adult Health -ELSA-Brasil, to obtain information on my health history from health institutions, medical records at outpatient clinics, emergency services and/or hospitals, according to the specifi c situation.…”

The PIERS Experience: Research or Quality Improvement?

“…A systematic review of 17 non-pregnancy studies identified highly variable published consent rates of 37% to 93% for data abstraction from existing medical records. 11 These often low and variable rates of consent were also seen in the Canadian Acute Stroke Registry when it was forced to transition from anonymous record review to informed consent; the consent rate was 51% even after a year's experience in recruitment methods, and the inter-site variation in consent was 17%. 12,13 This registry also identified the need for multiple visits to introduce the study, explain the procedures, and give ample opportunity to ask questions and make a decision about participation.…”