Logistic regressionBMJ 2013; 347 doi: https://doi.org/10.1136/bmj.f4488 (Published 12 July 2013) Cite this as: BMJ 2013;347:f4488
- Philip Sedgwick, reader in medical statistics and medical education
- 1Centre for Medical and Healthcare Education, St George’s, University of London, Tooting, London, UK
Researchers investigated whether reoperation after breast conserving surgery was associated with patients’ characteristics. A cohort study design was used. Participants were 55 297 women who had primary breast conserving surgery in 156 English NHS trusts between 1 April 2005 and 31 March 2008. The primary outcome was at least one breast reoperation within three months of breast conserving surgery. During the three year study period, 11 032 (20.0%, 95% confidence interval 19.6% to 20.3%) women had at least one reoperation.1
Logistic regression analysis examined the association between breast reoperation and patients’ characteristics (age at admission, tumour group, comorbidity, and socioeconomic deprivation) (table⇓). Tumour type was grouped by whether a carcinoma in situ component was recorded at the time of primary breast conserving surgery. The researchers commented that reoperation was nearly twice as likely when the tumour had a carcinoma in situ component recorded.
Which of the following statements, if any, are true?
a) The outcome variable for logistic regression is continuous
b) The odds ratio of breast reoperation for categorised age 50-59 years was 1.0
c) It can be concluded that the type of tumour was independently associated with breast reoperation
d) Conditional logistic regression was used …