Through an intervention program-8 weeks, I am working on behavior change. Due to time constraints and, ethically, not letting anyone unexposed to the intervention as the targeted behavior may cause danger to health, I have chosen non-randomly selected one group for pre and post test. Post test will be repeated at 10th, 12th, and 14th week.
How can I provide sound rationale for using this design and how can i address the well-known biases associated with this design e.g. maturation, RTM, history, and pre-test etc. I will be thankful for your suggestions, please.
You can minimize regression-to-the-mean effect by NOT selecting people with extreme test results (either low or high). You may randomly select the people to study from the larger group, select people with average pre-test results, or calculate the average of 2-3 pre-test results and use it as a pre-test value.
You cannot control for maturation or natural history effect, using a single arm design. You may use historical controls (outside group or equivalent previous observation period on the same group), however, this will be of course a weak control.
If your pre-test and post-test results are on continuous scale, given the stated constraints you have, you may consider a regression-discontinuity design. You would give your intervention to individuals who need it most (using some cut point on pre-test results) and then plot the post-test results against the pre-test results for both treated and untreated. The effect of your intervention, if any, would be seen as a discontinuity in the regression line.
We had a similar problem in a s study addressing communication of probabilistic risk on preventive behaviour in a single occupational cohort.
see Leigh J Harrison J . J Occ Health Safety ANZ 1991 ; 7:467-472
paper on my rg site.
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts