|
|
Statistical Analysis Plan
The aim of this study is to determine if the treatment effect of the Resonator™ device for the active treatment group is greater than that for the placebo group.
The study will be considered a success if, using the Intent-To-Treat (ITT) Last Observation Carried Forward (LOCF) analysis, the primary endpoint is statistically significant at the 0.05 level.
PRIMARY EFFICACY OUTCOME MEASURE: The primary efficacy outcome measure in this study is the mean change from baseline to end of the eight-week treatment period (study endpoint) in the Summary Index score on the PDQ-39 (PDQ-39SI).
The PDQ-39SI is calculated for each individual subject as the sum of each of the 8 dimension scores divided by 8 (the total number of dimension scores). Each of the 8 dimension scores is calculated as a scale from 0-100 as follows: (sum of the scores for all questions in the dimension)/(4 X number of questions in the dimension) X 100.
Mean values will be calculated as the average of the individual subject single index scores within a treatment group at each time point. The mean change will be calculated as the difference in mean group score from baseline to study endpoint.
Study success will be established through the detection of a minimum mean difference of 4 points between test and placebo groups in the change in PDQ-39SI from baseline to study endpoint (end of treatment week 8) adjusted for the baseline PDQ-39SI score. The 4-point mean difference between treatment groups is considered clinically relevant with a SD of 5.55, a two-sided test with 80% power and a 5% level of significance.
Null Hypothesis: There is no difference in the mean change in summary index score on the PDQ-39 adjusted for the baseline PDQ-39SI score between active treatment and placebo groups, to the effect of less than 4.0.
Alternative Hypothesis: There is a difference in the mean change in summary index score on the PDQ-39 adjusted for the baseline PDQ-39SI score between active treatment and placebo groups, to the effect of 4.0 or greater.
The application of the detection of a mean difference of 4 points between test and placebo groups in the change in the PDQ-39SI being considered clinically relevant was determined according to the reference: Distribution-based criteria for change in health-related quality of life in Parkinson's disease, Ray Fitzpatrick, Josephine M. Norquist, Crispin Jenkinson, Journal of Clinical Epidemiology 57 (2004) 40-44. The complete published article is contained in Appendix D of this clinical study protocol document.
The purpose of the study was to calculate the minimum change scores required to be both a minimally meaningful and statistically meaningful change for the PDQ-39 Summary Index and eight dimension scores. Data was collected from both community and clinic samples and a distribution-based analysis approach was applied. The findings were also compared with results from other studies exploring minimally meaningful changes for the PDQ-39 in comparable populations employing an anchor-based approach.
Consistency was found in the calculated change required for minimally meaningful change between the distribution-based (both community and clinic samples) and the anchor-based approaches. It was found in the study that according to a one-SEM (standard error of measurement) criterion, a change of 3.80 points (clinic sample) to 3.98 points (community sample) is a minimally important difference. For the current study, a 4.0-change criterion was employed to ensure that minimal required change for meaningfulness and significance would be met.
A criterion of one-SEM means that the observed change is unlikely to be due to measurement error ,and may be considered on distributional grounds, a minimum true change in score. The criterion’s ability to adequately represent minimally important change in the PDQ-39 0-100 transformed scores for Parkinson's disease is consistent with and supported by the conclusion of several studies examining the same issue across a range of Health-Related Quality of Life (HRQoL) instruments in patients with cardiac problems, chronic obstructive pulmonary disease and asthma. In each of these studies, one-SEM provided the optimal statistical criterion consistent with anchor-based evidence. These references are the following:
- Wyrwich KW, Nienaber NA, Tierney WM, Wolinsky FD. Linking clinical relevance and statistical significance in evaluating intra-individual changes in health-related quality of life. Med Care 1999;37(5): 469–78.
Objective: To compare the standard error of measurement (SEM) with established standards for clinically relevant intra-individual change in an evaluation of health-related quality of life.
Design: Secondary analysis of data from a randomized controlled trial.
Subjects: Six hundred and five outpatients with a history of cardiac problems attending the general medicine clinics of a major academic medical center.
Measures: Baseline and follow-up interviews included a modified version of the Chronic Heart Failure Questionnaire (CHQ) and the SF-36. The SEM values corresponding to established standards for minimal clinically important differences (MCIDs) on the CHQ were determined. Individual change on the SF-36 was explored using the same SEM criterion.
Results: One-SEM changes in this population corresponded well to the patient-driven MCID standards on all CHQ dimensions (weighted kappas (0.87; P < 0.001). The distributions of outpatients who improved, remained stable, or declined (defined by the one-SEM criterion) were generally consistent between CHQ dimensions and SF-36 subscales.
Conclusions: The use of the SEM to evaluate individual patient change should be explored among other health-related quality of life instruments with established standards for clinically relevant differences. Only then can it be determined whether the one-SEM criterion can be consistently applied as a proxy for clinically meaningful change.
- Wyrwich KW, Tierney WM, Wolinsky FD. Further evidence supporting an SEM-based criterion for identifying meaningful intra-individual changes in health-related quality of life. J Clin Epidemiol 1999; 52(9):861–73.
This study used the standard error of measurement (SEM) to evaluate intra-individual change on both the Chronic Respiratory Disease Questionnaire (CRQ) and the SF-36. After analyzing the reliability and validity of both instruments at baseline among 471 COPD outpatients, the SEM was compared to established minimal clinically important difference (MCID) standards for three CRQ dimensions. A value of one SEM closely approximated the MCID standards for all CRQ dimensions. This SEM-based criterion was then validated by cross-classifying the change status (improved, stable, or declined) of 393 follow-up outpatients using the one-SEM criterion and the MCID standard. Excellent agreement was achieved for all three CRQ dimensions. Although MCID standards have not been established for the SF-36, the one-SEM criterion was explored in these change scores. Among SF-36 scales demonstrating acceptable reliability and reasonable variance, the percent of individuals within each change category was consistent with those seen in the CRQ dimensions. These results replicate previous findings where a value of one SEM also closely approximated MCIDs for all dimensions of the Chronic Heart Disease Questionnaire among cardiovascular outpatients. The one-SEM criterion should be explored in other health-related quality of life instruments with established MCIDs.
- Sloan DA, Donnelly MB, Schwartz RW, Felts JL, Blue AV, Strodel WE. The use of the objective structured clinical examination (OCSE) for evaluation and instruction in graduate medical education. J Surg Res 1996;63:225.
This study had two purposes: determining the reliability and validity of the Objective Structured Clinical Examination (OSCE) in assessing performance by trainees at all levels, including medical students and chief residents; and estimating the impact of providing OSCE participants with immediate feedback about their performance. A comprehensive 210-min OSCE was administered to 53 surgical residents and 6 junior medical students. Faculty experts proctored all patient stations and provided immediate feedback to participants after the patient interaction segments (Part A). The participants then answered questions about the patients seen (Part B). The reliability of the OSCE was high (.91), identical to that of a previous resident OSCE with no feedback. The standard error of measurement for both parts was approximately 4%. At the 95% confidence interval, each participant's actual level of clinical performance (Part A) and clinical knowledge (Part B) could be estimated with an error of ±8%. Participants showed significant differences in clinical performance (Part A,P< 0.01) and knowledge (Part B,P< 0.01) by level of training. Most participants (74%) rated the OSCE as an above average or outstanding educational method. The OSCE is a valid and reliable test of residents’ clinical skills. Feedback to participants during the OSCE was positively received and did not perturb test reliability.
- Wyrwich KW, Tierney RM, Wolinsky FD. Using the standard error of measurement to identify important changes on the Asthma Quality of Life Questionnaire. Qual Life Res 2002;11(2):1–7.
Objectives: To establish a link between the minimal important difference (MID) and the standard error of measurement (SEM) for all responsive dimensions of the Asthma Quality of Life Questionnaire (AQLQ). Methods: Secondary data analysis of baseline and follow-up interview data from 198 outpatients with asthma enrolled in a randomized controlled trial and receiving care at a major urban academic medical center''s general medicine clinics. Domain statistics for baseline and follow-up interviews were examined for the AQLQ. The baseline SEM values were compared with established AQLQ MID standards using weighted k values. Results: One SEM identified the MID in responsive AQLQ dimensions. Weighted k values (0.88–0.93) validated excellent agreement between these two criteria. Conclusion: This is the third study to support using one SEM to identify important individual change in health-related quality of life (HRQoL) measures. However, refinement of the process for determining a measure's clinically meaningful differences is still needed to secure a link between the SEM and the identification of relevant HRQoL change over time.
The complete published articles above are contained in Appendix D of this clinical study protocol document.
|
