Causal Effects in Twin Studies: the Role of Interference

• Activity Number: 81 - Contributed Poster Presentations: Section on Statistics in Epidemiology
• Type: Contributed
• Date/Time: Monday, August 3, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Bonnie Smith* and Elizabeth Ogburn and Saonli Basu and Matthew McGue and Daniel O Scharfstein
• Companies: Johns Hopkins Bloomberg School of Public Health and Johns Hopkins Bloomberg School of Public Health and Division of Biostatistics, University of Minnesota and University of Minnesota and Johns Hopkins University
• Keywords: Spillover effect; Semi-parametric efficient; Co-twin control method

Abstract

The use of twin designs to address causal questions is becoming increasingly popular. A standard assumption is that no twin's exposure has a causal impact on their co-twin's outcome--that is, that there is no interference between twins. However, in some settings this assumption may not hold, and this would impact the causal interpretation of the parameter estimated by commonly used existing twins methods. The presence of interference would also change which effects are of greatest interest, and impact the conditions under which we may estimate these effects. We explore these issues, and we derive semi-parametric efficient estimators for causal effects in the presence of interference between twins. Using data from the Minnesota Twin Family Study, we apply our estimators to assess whether twins' consumption of alcohol in early adolescence may have a causal impact on their co-twins' alcohol use later in life.