MATERNAL RECALL AND MEDICAL RECORDS - AN EXAMINATION OF EVENTS DURINGPREGNANCY, CHILDBIRTH, AND EARLY INFANCY

It is not currently known whether sufficiently precise data on a previ ous pregnancy, labor and delivery, and early infancy can be obtained r etrospectively. We conducted a telephone survey in 1991 of women who d elivered babies between 1984 and 1986 at two teaching hospitals in Nas hville, Tennessee, to assess how well mothers recall information on fa ctors predictive of an adverse birth outcome. The survey yielded 102 u sable responses that were compared with hospital records for mothers a nd infants. Overall, 89 percent agreement was found between women's re sponses and their charts. Respondents were not reluctant to answer pot entially sensitive questions, and their technical knowledge was typica lly better for their own health than about some prenatal diagnostic pr ocedures and their infants' health. We found no difference in recall a ccuracy according to whether mothers experienced some adversity with t he index pregnancy. Accurate perinatal information can generally be ob tained with a recall period as high as four to six years.