Research on predictors of birth outcomes tends to focus on maternal characteristics. Less is known about the role of paternal factors. Missing paternal data on administrative records may be a marker for risk of adverse birth outcomes.

Data and Methods

Analyses were performed on a cohort of births that occurred from May 16, 2004 through May 15, 2006, which was created by linking birth and death registration data with the 2006 Canadian census. Log-binomial and binomial regression were used to estimate relative risks and risk differences for preterm birth, small-for-gestational-age birth, stillbirth and infant mortality associated with the absence of paternal information. Analyses controlled for maternal age, education, household income, parity, marital status, ethnicity and birthplace.


The analyses pertained to 135,285 singleton births. Paternal data were missing from the birth registration for 7,461 births (4.6%) and from the census data for 17,713 births (11.4%). The adjusted relative risks associated with missing paternal data on the birth registration were 1.12 (95% CI: 0.99, 1.26) for preterm birth; 1.15 (1.05, 1.26) for small-for-gestational-age birth; 1.86 (1.27, 2.73) for stillbirth; and 1.53 (1.00, 2.34) for infant mortality. Estimates were robust to varying definitions of missing paternal information, based on the birth registration, census data, or both.


This study suggests that missing paternal data is a marker for increased risk of adverse birth outcomes, over and above maternal characteristics.


Birth certificates, fetal death, infant mortality, medical record linkage, premature infant, small-for-gestational-age infant, stillbirth


In recent decades, the focus of perinatal research has extended beyond biomedical risk factors to include psychosocial variables, families and communities, and the role of paternal characteristics. However, as the range of factors investigated expands, so does the amount of missing data, a situation that could adversely affect analyses. [Full Text]


Gabriel D. Shapiro (gabriel.shapiro@mail.mcgill.ca), Michael S. Kramer, Jay S. Kaufman and Seungmi Yang are with the Department of Epidemiology, Biostatistics and Occupational Health at McGill University. Tracey Bushnik is with the Health Analysis Division at Statistics Canada. Amanda J. Sheppard is with The Hospital for Sick Children in Toronto. Michael S. Kramer is also with the Department of Pediatrics at McGill University.

Start of text box


What is already known on this subject?

  • Increasingly, perinatal studies are incorporating paternal risk factors.
  • Missing paternal data may negatively affect analyses.
  • The nature and implications of missing paternal data have not been systematically examined.

What does this study add?

  • Missing paternal information from birth registrations and from census results is a marker for increased risk for adverse birth outcomes, over and above maternal characteristics.
  • Future studies should aim to determine the extent to which missing paternal information is associated with the absence or limited involvement of the father.

End of text box

Date modified: