Examining the Psychometric Properties of Three Standardized Screening Tools in a Pregnant and Parenting Population

Significance Although standardized tools such as the Speilberger State Anxiety scale (1970) and Cohen's Perceived Stress Index (1983) have often been used in pregnant and postpartum populations to evaluate psychosocial risk, there exists very little literature on the reliability of these scales in this population. Given the potentially grave consequences of undiagnosed pre and postpartum mental distress, it is of paramount importance that the tools that are commonly used are evaluated for their psychometric properties within the pregnant and parenting population. Methods The psychometric properties of three mental health scales were evaluated using the All Our Babies data. The All Our Babies cohort is a prospective longitudinal study based in Calgary, Alberta. A total of 3300 women were recruited from prenatal care sites from all four quadrants of the city. Participants were mailed questionnaires at 24-26 weeks gestation, 34-36 weeks and at 4 months postpartum. Reliability for the Speilberger State Anxiety Scale, the Perceived Stress Index and the Life Optimism Test-revised were evaluated by calculating Cronbach's alphas. Validity with related constructs were tested for the State Anxiety Scale and the Perceived Stress Index by calculating Pearson Correlation Coefficients with closely related constructs. Sample size varied from 2670 to 3376, according to the response rate and time point. Depression as a related construct was evaluated using the Edinburgh Postpartum Depression Scale. Results The Cronbach's Alphas for the Speilberger State Anxiety scale were 0.92, 0.92 and 0.93 for 24-26 weeks, 34-36 weeks and 4 months postpartum respectively. The Alpha's for the Perceived Stress index were 0.88, 0.88 and 0.89 respectively. The Life Optimism Test Revised was only measured in the third trimester and the Cronbach's alpha was 0.83. The Pearson Correlation Coefficient for Anxiety with depression were r = 0.73, r = 0.72 and r = 0.77 respectively. The coefficients for stress and depression were r = 0.75, r = 0.75 and r = 0.77 respectively. Discussion/Conclusion The psychometric properties for all three scales were strong, with alphas that were comparable or higher to literature values. These data provide evidence that the use of these scales, previously validated in other populations are appropriate for use in among pregnant and parenting women at risk for mental distress.