Inclusion of gender in emergency medicine research

Abstract

Background: Emergency physicians as front-line clinical specialists can directly advance patient care by understanding how gender-specific approaches may affect evaluation and management of diseases in the acute setting. Yet, it is unclear whether the role of gender is systematically examined in research focusing on emergency care.

Objectives: The objective was to determine if the effect of gender on health outcomes is examined in published studies targeting emergency medicine (EM).

Methods: Using MEDLINE, the term "emergency" was used to identify all English-language, EM-related studies of adult human subjects published between January 2006 and April 2009 in which the first, second, or last author belonged to an EM section, division, center, or institution functioning as an emergency department (ED). The alternative chance-corrected statistic was used for intercoder reliability, and chi-square was used to calculate odds ratios (OR) with 95% confidence intervals (CIs). Articles were coded for gender composition, as well as use of gender as a control variable, independent variable, or part of the primary hypothesis.

Results: The search revealed 2,487 articles using the selected "emergency" terms, and 750 original studies coded as EM-related publications were reviewed. The five topics contributing the most articles (44%) were administration/crowding, cardiovascular disease, emergency medical services, trauma, and sepsis. Seventy-nine percent of articles reported the gender composition of the sample, with 11% including gender as a control variable, 18% including gender as an independent variable, and 2% including gender in the primary hypothesis. The alternative chance-corrected statistic for evaluating gender composition was 0.90 (95% CI = 0.75 to 1.00). Use of gender in the analysis did not differ between federally funded studies versus non-federally funded studies (OR = 0.86; 95% CI = 0.5 to 1.4). The number of articles analyzing the effect of gender on a health outcome increased by 5% over the study period (27%-32%).

Conclusions: The majority of research articles targeted EM report gender as a demographic variable; however, few studies examined the effect of gender on health outcome. As the specialty advances into the next decade, the authors recommend that EM researchers 1) include both men and women in their study designs for appropriate gender comparisons; 2) report gender composition of study subjects and gender-specific comparisons study findings; and 3) report prognoses, outcomes, and interventions using gender as an independent variable in the study model.