Representativeness of child controls recruited by random digit dialling

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Case-control studies rely on representative control groups.
• Recruiting representative controls is a significant challenge in epidemiological research.
• Socio-economic status (SES) can influence participation and introduce bias.

Purpose of the Study:

• To evaluate the effectiveness of random digit dialling (RDD) for recruiting control subjects.
• To assess the representativeness of an RDD-recruited sample concerning socio-economic status.
• To investigate potential biases in a population-based case-control study of childhood acute lymphoblastic leukaemia (ALL).

Main Methods:

• Utilized data from a national, population-based case-control study of childhood ALL in Australia (2003-2006).
• Linked control families' addresses to Australian Bureau of Statistics Census 2006 Collection Districts for Socio-Economic Indexes for Area (SEIFA) scores.
• Compared SEIFA scores of participating families with the general population distribution.

Main Results:

• An estimated 55% of eligible families participated in the RDD sample.
• Participation rates were directly associated with socio-economic status; higher SES families were more likely to enroll.
• Completeness of data collection also correlated with higher socio-economic status.

Conclusions:

• RDD recruitment for this childhood ALL case-control study exhibited selection bias related to socio-economic status.
• This SES-based selection may indicate potential bias regarding other crucial factors in ALL aetiology.
• Findings highlight the need for methods to mitigate selection bias in population-based case-control studies.