Combining Dependent Tests with Incomplete Repeated Measurements

Abstract

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Biometrika Trust is collaborating with JSTOR to digitize, preserve and extend access to Biometrika. SUMMARY In comparing the effectiveness of two treatments, repeated measurements of the same characteristic are often taken under two or more distinct conditions for each experimental subject. A locally best linear combination of individual test statistics obtained under different conditions is proposed to test the equality of the two treatments. The test procedure allows different patterns of missing observations for the two groups to be compared. Special cases such as combining Wilcoxon tests, t tests and 2 x 2 tables are discussed in detail with real examples.Some key words: Pitman local alternative; Repeated measurement; Two by two table; U-statistic; Wilcoxon test.

Analysis and design issues for studies using censored biomarker measurements with an example of viral load measurements in HIV clinical trials

“…Furthermore, some of the non-parametric analyses can be extended to repeated measurements. For example, the Wei-Johnson test [19] can be applied to repeated censored measurements, and it is possible to construct (censored) summary statistics, for example, the area under the curve, for censored repeated measurements. Alternatively, it is possible to construct (censored) summary statistics such as the area under the curve of repeated measurements, and to conduct hypothesis tests using rank or permutation tests [20].…”

Performance of Selected Nonparametric Tests for Discrete Longitudinal Data Under Different Patterns of Missing Data

“…Another test statistic used was the one proposed by Wei and Johnson (1985) which is valid when the patterns of incompleteness are different among the groups. Under this procedure, a test statistic of the U form (Mann-Whitney statistic) is computed separately for each course, and an overall test statistic is obtained by combining these per course test statistics.…”

“…However, most of these have been applied to continuous response variables and mainly through parametric tests. There is very little similar work on discrete responses and nonparametric tests (Koziol and Maxwell, 1981;Wei and Lachin, 1984;Wei and Johnson, 1985;Wu et al, 1994). Furthermore, there are plenty of examples of published work comparing the performance of parametric tests (Lindsey, 2000;Sutradhar and Bartlett, 1993) for longitudinal data, but very little is known on nonparametric tests (Davis, 1994;Davis and Wei, 1988) whose applications are low.…”

Quality of Life Analyses in a Clinical Trial of DPPE (tesmilifene) Plus Doxorubicin Versus Doxorubicin in Patients with Advanced or Metastatic Breast Cancer: NCIC CTG Trial MA.19

“…The Wilcoxon rank-sum test was then performed on each of these four summary measures for the comparison of QOL between the two treatment arms. (2) Analyses based on the Wei-Lachin [8] There were no differences in objective response rates [1]. Drug related adverse effects assessed by NCIC CTG Expanded Toxicity Criteria are summarized in Table 2; the most common events were gastrointestinal (nausea and vomiting), neurologic (hallucinations, vertigo, dizziness) and fatigue.…”