Goldenberg R L, Hickey C A, Cliver S P, Gotlieb S, Woolley T W, Hoffman H J
Department of Obstetrics and Gynecology, School of Medicine, University of Alabama at Birmingham, USA.
Acta Obstet Gynecol Scand Suppl. 1997;165:19-29.
Data from five existing psychosocial scales were used to develop an abbreviated scale for the assessment of psychosocial status during pregnancy.
Scales were self-administered by 842 black and 381 white low-income multiparous women at risk for poor pregnancy outcome. Trait anxiety (Speilberger), self-esteem (Rosenberg), mastery (Pearlin), and depression (CES-D) were assessed at 24-26 weeks' gestation; subjective stress (Schar) was assessed at 30-32 weeks' gestation. The 59 pooled items were examined for redundancy and the discernment of primary factors using principal factor analysis. Regression analysis was used to determine if the resulting abbreviated scale (28 items) would provide information similar to that obtained with the 59 item pool (full scale) in predicting gestational age (GA), birth weight (BW), fetal growth restriction (FGR), and preterm delivery (PTD).
The abbreviated scale was highly correlated (r = 0.97) with the 59-item pool and the six factors isolated were generally compatible with the major characteristics assessed by the five original scales. The distribution of FGR and PTD by scale quartile was similar for the abbreviated and the combined scales. Logistic regression analysis of scores for all women revealed that poor (high) scores on both the full (p = 0.0151) and the abbreviated scales (p = 0.0131) were positively associated with FGR, but not with PTD. In linear regression analysis poor (high) scores on both the full (p = 0.0024) and the abbreviated scale (p = 0.0019) were negatively related to BW, but not to GA. When data for black and white women were examined separately, the two scales provided comparable information.
The abbreviated psychosocial scale provided information similar to that obtained with 59 pooled items in predicting GA, BW, FGR, and PTD.
