A comparison of sampling error and standard errorBMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h3577 (Published 03 July 2015) Cite this as: BMJ 2015;351:h3577
- Philip Sedgwick, reader in medical statistics and medical education
- 1Institute for Medical and Biomedical Education, St George’s, University of London, London, UK
- Correspondence to: P Sedgwick
The impact of a diet and physical activity programme on body weight in overweight or obese people initiated through a national colorectal cancer screening programme was investigated. A multicentre randomised controlled trial was performed. The intervention consisted of a personalised, behaviourally focused weight loss programme, delivered over 12 months. The control treatment consisted of a weight loss booklet only. Participants were overweight or obese adults (aged 50-74 years) who had undergone colonoscopy after a positive faecal occult blood test result and had a diagnosis of adenoma confirmed by histopathological examination. In total, 329 participants were recruited and randomised to the intervention (n=163) or control (n=166).1
The primary outcome was weight change over 12 months. At follow-up, data were available on the primary outcome for 148 (91%) participants in the intervention group and 157 (95%) in the control group. At 12 months, mean weight loss was 3.50 kg (95% confidence interval 2.70 to 4.30; standard error 0.40) in the intervention group compared with 0.78 kg (0.19 to 1.38; 0.30) in the control group. The intervention group lost significantly more weight (mean difference 2.69 kg, 1.70 to 3.67; P<0.001). It was concluded that significant weight loss can be achieved by a diet and physical activity intervention initiated within a national colorectal cancer screening programme in older adults, offering considerable potential for reducing the risk of disease.
Which of the following statements, if any, are true?
a) The sample estimates were …