Predictors and Impact of Non-Response in a Population-Based Case Control Study: Findings from the Georgia Study to Explore Early Development

Background: Participation in epidemiologic studies has declined over time, raising concerns about selection bias. While estimates derived from epidemiologic studies have been shown to be robust under a wide range of scenarios, additional empiric study is needed. Empiric assessments of non-response are inherently challenging given the lack of information available on non-responders in most studies, including those of autism spectrum disorder (ASD). We explore factors associated with non-participation and its potential impact on associations between ASD and various risk factors in the Georgia Study to Explore Early Development (GA SEED).

Objectives: N/A

Methods: GA SEED is a population-based case-control study of risk factors for ASD that recruited children at age 2-5 years residing in the metropolitan-Atlanta area. Children with ASD were identified from multiple schools and healthcare providers serving children with disabilities; children from the general population (POP) were randomly sampled from birth records. Recruitment was via mailed invitation letter with follow-up phone calls. Eligibility criteria included birth and current residence in study area and an English-speaking caregiver who could provide legal consent. Many children identified for potential inclusion could not be contacted. We examined demographic and perinatal factors associated with study completion using birth certificate data available for both participants and non-participants. Using the birth-certificate data, we also compared ASD-risk factor associations for the final sample of children who completed the study with the initial sample of all likely ASD and POP children invited to potentially participate in the study. Finally, we derived post-stratification sampling weights for participants who completed the study and compared weighted and unweighted associations between ASD and two maternally-reported factors collected post-enrollment.

Results: Maternal age >35 years and maternal education >12 years were independently associated with participation in the POP group. Maternal education >12 years was independently associated with participation in the ASD group. However, numerous other demographic and perinatal factors were not associated with participation. Moreover, unadjusted and adjusted odds ratios for associations between ASD and several demographic and perinatal factors were similar between the final sample of study completers and the total invited sample. Odds ratios for associations between ASD and maternally-reported reproductive health factors -- infertility and reproductive stoppage -- were also similar in unweighted and weighted analyses of the study completion sample.

Conclusions: We demonstrated empirically that while select demographic factors were directly associated with participation in a population-based case-control study of ASD risk factors, other demographic and biologic factors were not. Moreover, associations of biologic factors – both perinatal factors on the birth certificate and reproductive health history factors captured via maternal interview – were not impacted by differential participation. Additionally, while differential participation limited our ability to examine associations between ASD and two demographic factors – maternal age and education – this study demonstrated that the effect estimates for associations with several other demographic factors were unbiased. SEED is an important source of information on ASD risk factors. The findings from this analysis of GA SEED data indicate that the estimates from most SEED risk factor analyses are robust.