Intended for healthcare professionals

Endgames Statistical Question

Bias in observational study designs: case-control studies

BMJ 2015; 350 doi: (Published 30 January 2015) Cite this as: BMJ 2015;350:h560
  1. Philip Sedgwick, reader in medical statistics and medical education1
  1. 1Institute for Medical and Biomedical Education, St George’s, University of London, London, UK
  1. p.sedgwick{at}

Researchers investigated the association between sun exposure and risk of multiple sclerosis. A population based case-control study was performed. The participants were recruited from residents of Tasmania, Australia, who were aged under 60 years and had at least one grandparent born in Tasmania. Cases were people with multiple sclerosis who volunteered after information evenings at local multiple sclerosis societies, or after having been invited by a healthcare professional. In total, 136 people with a diagnosis of multiple sclerosis, as defined by clinical and magnetic resonance imaging criteria, were included as cases. For each case, two controls matched for sex and year of birth were randomly selected from the community. In total, 359 eligible controls were approached and the response rate was 76%.1

A validated questionnaire was used to record the amount of time participants would normally have spent in the sun during weekends and holidays in winter and summer before the age of 16 years. Actin damage was used as a marker of cumulative lifetime sun exposure. Silicone casts of the skin surface of the hand were obtained and graded by the researchers from 1 (undamaged skin) to 6 (severe deterioration).

The researchers reported that greater sun exposure when aged 6-15 years (average two to three hours or more a day in summer during weekends and holidays) was associated with a reduced risk of multiple sclerosis (adjusted odds ratio 0.31, 95% confidence interval 0.16 to 0.59). Greater actinic damage (disease grades 4-6) was also independently associated with a reduced risk of multiple sclerosis (0.32, 0.11 to 0.88). It was concluded that higher sun exposure during childhood and early adolescence was associated with a reduced risk of multiple sclerosis. Insufficient ultraviolet radiation may therefore influence the development of multiple sclerosis.

Which of the following, if any, might the above case-control …

View Full Text

Log in

Log in through your institution


* For online subscription