Validity of Mother's Report of Father's Occupation in a Study of Paternal Occupation and Congenital Malformations

• Published in: American journal of epidemiology 1995-05, Vol.141 (9), p.872-877
• Main Authors: Schnitzer, Patricia G., Olshan, Andrew F., Savitz, David A., Erickson, J. David
• Format: Article
• Language: English
• Subjects: abnormalities; Biological and medical sciences; Case-Control Studies; Congenital Abnormalities - epidemiology; Congenital Abnormalities - etiology; Epidemiologic Methods; Epidemiology; Female; General aspects; Georgia; Humans; Infant, Newborn; Interviews; Interviews as Topic; Male; Medical sciences; Methodology; Occupations; Paternal Exposure - statistics & numerical data; Public health. Hygiene; Public health. Hygiene-occupational medicine; Registries; Sensitivity and Specificity; Socioeconomic Factors
• ISSN: 0002-9262; 1476-6256
• DOI: 10.1093/oxfordjournals.aje.a117523

Agreement between the mother's and father's report of the father's occupation was assessed in a case-control study of paternal occupation and birth defects. Cases were identified from births registered with the Metropolitan Atlanta Congenital Defects Program between 1968 and 1980; controls were selected from livebom infants without defects. Both parents were sought for interview, and each parent was asked about the father's job history for 2 years prior to the infant's birth. This concordance analysis is based on 3,739 case infants and 2,279 control infants for whom both parents were interviewed. The authors considered the father's report of his occupation as correct, and they assessed the ability of the mother to report the same occupation(s) during a 7-month period around conception. The exact agreement between mother's and father's report of the father's occupation was 59%. Agreement improved slightly with increasing family income and when fathers were college graduates. Female partners were not accurate proxy respondents in this study of paternal occupation and birth defects, which suggests that investigators should interview both parents in studies of paternal exposures and reproductive outcomes, i.e., mothers for pregnancy history and maternal confounders and fathers for occupational history and paternal confounders. Am J Epidemiol 1995; 141:872–7.