gatekeeping procedure

Login or register (free and only takes a few minutes) to participate in this question.

You will also have access to many other tools and opportunities designed for those who have language-related jobs (or are passionate about them). Participation is free and the site has a strict confidentiality policy.

20:15 Jul 6, 2018
English to Polish translations [PRO]
Science - Mathematics & Statistics / badania kliniczne
English term or phrase: gatekeeping procedure
To control the overall Type I error rate for the primary efficacy endpoints, a gatekeeping procedure will be used.
kgas
Poland
Local time: 03:01

Summary of reference entries provided
gatekeeping procedures
Frank Szmulowicz, Ph. D.
Gatekeeping Testing Strategies
geopiet

Discussion entries: 1





  

Reference comments


23 mins
Reference: gatekeeping procedures

Reference information:
The objective of this paper is to give an overview of a relatively new area of multiplicity research that deals with the analysis of hierarchically ordered multiple objectives. Testing procedures for this problem are known as gatekeeping procedures and have found a variety of applications in clinical trials. This paper reviews main classes of these procedures, including serial and parallel gatekeeping procedures, and tree gatekeeping procedures that account for logical restrictions among multiple objectives.

--------------------------------------------------
Note added at 27 mins (2018-07-06 20:43:09 GMT)
--------------------------------------------------

This paper discusses multiple testing problems in which families of null hypotheses are tested in a sequential manner and each family serves as a gatekeeper for the subsequent families. Gatekeeping testing strategies of this type arise frequently in clinical trials with multiple objectives, e.g., multiple endpoints and/or multiple dose-control comparisons. It is demonstrated in this paper that the parallel gatekeeping procedure of Dmitrienko,
https://www.ncbi.nlm.nih.gov/pubmed/17240656


    https://onlinelibrary.wiley.com/doi/pdf/10.1002/pst.291
Frank Szmulowicz, Ph. D.
United States
Native speaker of: Native in EnglishEnglish, Native in PolishPolish
PRO pts in category: 113
Login to enter a peer comment (or grade)

1 hr
Reference: Gatekeeping Testing Strategies

Reference information:
1318 Clinical trials commonly assess efficacy of a treatment on multiple endpoints, usually grouped
1319 into a primary endpoint or endpoint family, and a secondary endpoint or endpoint family (see
1320 sections II.A and III.A). The usual strategy is to test all endpoints in the primary family
1321 according to one of the previously discussed methods (e.g., Bonferroni, fallback) and proceed to
1322 the secondary family of endpoints only if there has been statistical success in the primary family.
1323 This allows all of the available alpha level to be distributed within the primary family
1324 (containing the most important study endpoints) and thus maximizes the study power for those
1325 endpoints. In contrast, if the available alpha were distributed among all of the endpoints in the
1326 primary and secondary families, power would be reduced for the primary endpoints. Although it
1327 is not generally recommended, if there were an additional family of endpoints for which it was
1328 also important to control the Type I error rate, that family could be designated as third in the
1329 sequence.
1330
1331 This approach of testing the primary family first, and then the secondary family contingent upon
1332 the results within the primary family is called the gatekeeping testing strategy to highlight the
1333 fact that the endpoint families are analyzed in a sequence, with each family serving as a
1334 gatekeeper for the next one. The tests for the secondary family (and subsequent families if any)
1335 are carried out with appropriate multiplicity adjustments within that family, but only if the tests
1336 in the primary family have been successful.

https://www.fda.gov/downloads/drugs/guidancecomplianceregula... page 35

--------------------------------------------------
Note added at 1 hr (2018-07-06 22:10:34 GMT)
--------------------------------------------------

Two types of gatekeeping testing strategies are common in clinical trials, serial and parallel,
1339 determined by how the endpoints are tested within the primary family. The term serial strategy
1340 is applied when the endpoints of the primary family are tested as co-primary endpoints (section
1341 III.C). If all endpoints in the primary family are statistically significant at the same alpha level
1342 (e.g., α = 0.05), the endpoints in the second family are examined. The endpoints in the second
1343 family are tested by any one of several possible methods (e.g., Holm procedure, the fixed1344
sequence method, or others described in section IV.C). If, however, at least one of the null
1345 hypotheses of the primary family fails to be rejected, the primary family criterion has not been
1346 met and the secondary endpoint family is not tested.
1347
1348 The term parallel gatekeeping strategy is applied when the endpoints in the primary family are
1349 not all co-primary endpoints, and a testing method that allows the passing along of alpha from an
1350 individual test to a subsequent test (e.g., Bonferroni method or Truncated Holm method
1351 described next) is specified. In this strategy, the second endpoint family is examined when at
1352 least one of the endpoints in the first family has shown statistical significance.
1353
1354 The Bonferroni method is sometimes used for the parallel gatekeeping strategy, as it is the
1355 simplest approach. The secondary endpoint family may use a different method (e.g., the fixed1356
sequence method or Holm method). In this approach, if an endpoint comparison within the
1357 primary family is statistically significant at its allocated (or accumulated) endpoint-specific alpha
1358 level, then this alpha level can be validly passed on to the next family. On the other hand, if an
1359 endpoint comparison in a family is not significant at its endpoint-specific alpha level, that alpha
1360 is not passed on to the next family. The overall alpha available for testing the secondary family
1361 is the accumulated (unused) endpoint-specific alpha levels of those comparisons in the primary
1362 family that were found significant.

geopiet
Native speaker of: Native in PolishPolish
PRO pts in category: 12
Login to enter a peer comment (or grade)



KudoZ™ translation help

The KudoZ network provides a framework for translators and others to assist each other with translations or explanations of terms and short phrases.


See also:

Your current localization setting

English

Select a language

Term search
  • All of ProZ.com
  • Term search
  • Jobs
  • Forums
  • Multiple search