Tips on overlapping confidence intervals and univariate linear models
Adefowope Odueyungbo , Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton ON, Canada
Lehana Thabane Associate professor, Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton ON, Canada, Centre for Evaluation of Medicines, St Joseph’s Healthcare Hamilton – a Division of St Joseph’s Health System, Hamilton ON, Canada
Maureen Markle-Reid Assistant professor, School of Nursing, McMaster University, Hamilton ON, Canada, Ontario Ministry of Health and Long-Term Care, ON, Canada
Adefowope Odueyungbo, Lehana Thabane and Maureen Markle-Reid discuss ways to improve estimates from various linear regression models and derive findings when confidence intervals overlap
In randomised controlled trials, an overlap of confidence intervals is often cited as evidence of ‘no statistically significant difference’ between intervention groups. This paper illustrates the limitations of this strategy and compares different univariate linear regression models with baseline and follow-up response measures. The researchers also demonstrate that using ‘change in response’ or exit score as a function of the baseline response in clinical studies leads to the same results. Further, using a model that includes baseline response as covariate leads to more precise estimates. The implications for future trials are discussed.
16, 4, 73-83.
Want to read more?
Already subscribed? Log in
Unlock full access to RCNi Plus today
Save over 50% on your first 3 months
Your subscription package includes:
- Unlimited online access to all 10 RCNi Journals and their archives
- Customisable dashboard featuring 200+ topics
- RCNi Learning featuring 180+ RCN accredited learning modules
- RCNi Portfolio to build evidence for revalidation
- Personalised newsletters tailored to your interests
Alternatively, you can purchase access to this article for the next seven days. Buy now