Table 1.
Examples of non-pharmaceutical and pharmaceutical health-protective behaviors.
Note that this table represents examples of each type of behavior rather than a comprehensive list of all behaviors included in this analysis.
Fig 1.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of search strategy.
A flow diagram providing the organization of the article search and selection process along with values for article retention numbers at each state.
Table 2.
Reasons for exclusion and their associated frequency in full text screening.
Fig 2.
Map of global study distribution.
A map visualizing the number of studies addressing populations from each country.
Fig 3.
Density graph showing the sets of log odds ratios for pharmaceutical and non-pharmaceutical behaviors addressed for the 88 included study populations.
Males are used as the reference; positive log odds ratios correspond to females being more likely to adopt/practice a given behavior, and negative log odds ratios correspond to males being more likely to adopt/practice a given behavior. The set of non-pharmaceutical behaviors shown is trimmed such that the log odds ratio falling outside of three standard deviations from the mean is excluded.
Table 3.
Fixed- and random-effects model results.
Includes the non-pharmaceutical and pharmaceutical study sets and the three corresponding sensitivity analysis sets for each.
Fig 4.
Forest plot of the associations between gender and non-pharmaceutical behaviors.
The effect size and confidence interval of each study are indicated by a square and a horizontal line, respectively. The weight of each study in the model is indicated by the size of its square. A log odds ratio of 0, indicated by the dashed reference line, corresponds to no gender difference in behavioral response. Positive log odds ratios correspond to greater behavioral response by females, while negative log odds ratios correspond to greater behavioral response by males. The population mean effect size of the random-effects model incorporating these studies is given by the placement of the diamond, while the horizontal corners of the diamond illustrate the 95% CI of this mean effect size.
Fig 5.
Forest plot of the associations between gender and pharmaceutical behaviors.
The effect size and confidence interval of each study are indicated by a square and a horizontal line, respectively. The weight of each study in the model is indicated by the size of its square. A log odds ratio of 0, indicated by the dashed reference line, corresponds to no gender difference in behavioral response. Positive log odds ratios correspond to greater behavioral response by females, while negative log odds ratios correspond to greater behavioral response by males. The population mean effect size of the random-effects model incorporating these studies is given by the placement of the diamond, while the horizontal corners of the diamond illustrate the 95% CI of this mean effect size. Publications with the same author(s) and year of publication are differentiated by the first word of their title. Publications including multiple studies are denoted by labeling the studies A, B, etc.