Endgames Statistical Question

# Measuring the benefit of treatment: number needed to treat

BMJ 2015; 350 (Published 27 April 2015) Cite this as: BMJ 2015;350:h2206
1. Philip Sedgwick, reader in medical statistics and medical education
1. 1Institute for Medical and Biomedical Education, St George’s, University of London, London, UK
1. p.sedgwick{at}sgul.ac.uk

The effectiveness of offering financial incentives to help pregnant smokers quit was investigated using a randomised controlled trial. The intervention consisted of routine care plus the offer of up to £400 (€559; \$600) for women who attended specialist pregnancy stop smoking services and were abstinent from smoking at 34-38 weeks’ gestation. The control group received routine care, which included offers of face to face appointments to discuss smoking and cessation, free nicotine replacement therapy, and weekly support phone calls. The setting was a materially deprived inner city population in the west of Scotland. Participants were 612 self reported pregnant smokers aged at least 16 years and less than 24 weeks pregnant. Women were randomised to the intervention (n=306) or control (n=306).1

The primary outcome was abstinence from smoking at 34-38 weeks’ gestation. Secondary outcomes included birth weight. The proportion of smokers who were abstinent at 34-38 weeks’ gestation was significantly higher in the financial incentives group than in the control group (22.55% (n=63) v 8.58% (n=26); absolute risk difference 13.97%, 95% confidence interval 8.2% to 19.7%; P<0.001). The number needed to treat (NNT) was 7.2 (5.1 to 12.2). The relative risk of abstinence at 34-38 weeks’ gestation for the intervention group compared with the control group was 2.63 (1.73 to 4.01). No significant difference was seen between treatment groups in mean birth weight (3140 (standard deviation 600) v 3120 (590) g; P=0.67). The researchers concluded that financial incentives were effective in helping pregnant smokers quit.

Which one of the following statements is the best interpretation of the NNT?

• a) On average, for every 7.2 pregnant smokers offered financial incentives, one would be abstinent at 34-38 weeks’ gestation …

View Full Text