NSG702 Biostatistics for Advanced Practice Nurses
Case Study #3: Paired t-Test (16 points)
You are a nurse practitioner who works in a pain clinic. You have several patients with Piriform Syndrome, a pelvic condition that involves malfunction of the piriformis muscle (a deep buttock muscle), which often causes lumbar and buttock pain with sciatica (pain radiating down the leg). These patients have been asking about new options for pain management. You are curious if any other medications promote pain relief in this population. You identify a local researcher who is conducting a clinical trial of medications to treat Piriform Syndrome and join the research team. This team conducted a randomized double-blind clinical trial for patients with Piriform Syndrome; whereby, patients were injected with one of three substances.
- Group 1 received an injection of a combination of triamcinolone and lidocaine (the TL group).
- Group 2 received a placebo injection.
- Group 3 received an injection of Botox.
The randomization schedule was set up such that, approximately, for every six patients, three were assigned to group 1, one to group 2, and two to group 3. All injections were directly into the piriformis muscle. Patients were asked to come back at 2 weeks post-injection (0.5 month), 1 month post-injection, and monthly thereafter up to 17 months, although there were many missed visits. At each visit the patients rated their percentage of improvement of pain vs. baseline on a visual analog scale, with a maximum of 100% improvement (indicated by 100). Negative numbers indicate percentage of worsening.
You are on the team that asked if the TL group had a statically significant difference in pain level between the 0.5-week appointment and the 1-month appointment. The data are in BOTOX (no missing data).DAT, with a description in BOTOX.DOC.
- State the null and alternative hypothesis (2 points)
- Define the significance level and the critical value for the computed t-test statistic. (2 points)
- Evaluate the data to determine if the all necessary assumptions are met. State whether you will use a parametric (paired t-test) or a non-parametric test (Wilcoxon Matched-Pairs Signed Rank test). (3 points)
- Compute the appropriate test statistic. (3 points)
- Determine statistical significance. (3 points)
- State a conclusion. Consider how the results will apply to your patients with Piriform Syndrome who are asking for other medication options. (3 points)