A method for combining matched and unmatched binary data

Application to randomized, controlled trials of photocoagulation in the treatment of diabetic retinopathy

Stephen W. Duffy, Thomas E. Rohan, Douglas G. Altman

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

A method for combining matched and unmatched data is described and was applied to the results of randomized, controlled trials of photocoagulation in the treatment of diabetic retinopathy. A pooled estimate from the matched and unmatched studies war obtained by adaptation of the Mantel-Haenszel method, where the strata were unmatched studies and matched pairs within studies. A test of significance was based on the Mantel-Haenszel chi-square statistic, the latter also being used to calculate test-based confidence intervals. A test of homogeneity was performed by combining Mantel-Haenszel chi-square statistics from the matched and unmatched studies. By these methods, the combined estimate of the risk of deterioration of visual acuity for those receiving photocoagulation (relative to a risk of unity for those not receiving photocoagulation) was 0.37 (95 per cent confidence interval 0.29-0.46). The chi-square statistic (1 df) for an effect of treatment was 87.75 (p < 0.0001). The chi-square statistic for homogeneity of relative risk among studies was 11.20 (4 df, p < 0.05). However, this result was influenced disproportionately by one small matched study.

Original languageEnglish (US)
Pages (from-to)371-378
Number of pages8
JournalAmerican Journal of Epidemiology
Volume130
Issue number2
StatePublished - Aug 1989
Externally publishedYes

Fingerprint

Light Coagulation
Diabetic Retinopathy
Randomized Controlled Trials
Confidence Intervals
Therapeutics
Visual Acuity

Keywords

  • Epidemiologic methods;
  • Statistics

ASJC Scopus subject areas

  • Geriatrics and Gerontology
  • Epidemiology

Cite this

@article{14c0c7fc3412425ca8ec46b3cb5ba570,
title = "A method for combining matched and unmatched binary data: Application to randomized, controlled trials of photocoagulation in the treatment of diabetic retinopathy",
abstract = "A method for combining matched and unmatched data is described and was applied to the results of randomized, controlled trials of photocoagulation in the treatment of diabetic retinopathy. A pooled estimate from the matched and unmatched studies war obtained by adaptation of the Mantel-Haenszel method, where the strata were unmatched studies and matched pairs within studies. A test of significance was based on the Mantel-Haenszel chi-square statistic, the latter also being used to calculate test-based confidence intervals. A test of homogeneity was performed by combining Mantel-Haenszel chi-square statistics from the matched and unmatched studies. By these methods, the combined estimate of the risk of deterioration of visual acuity for those receiving photocoagulation (relative to a risk of unity for those not receiving photocoagulation) was 0.37 (95 per cent confidence interval 0.29-0.46). The chi-square statistic (1 df) for an effect of treatment was 87.75 (p < 0.0001). The chi-square statistic for homogeneity of relative risk among studies was 11.20 (4 df, p < 0.05). However, this result was influenced disproportionately by one small matched study.",
keywords = "Epidemiologic methods;, Statistics",
author = "Duffy, {Stephen W.} and Rohan, {Thomas E.} and Altman, {Douglas G.}",
year = "1989",
month = "8",
language = "English (US)",
volume = "130",
pages = "371--378",
journal = "American Journal of Epidemiology",
issn = "0002-9262",
publisher = "Oxford University Press",
number = "2",

}

TY - JOUR

T1 - A method for combining matched and unmatched binary data

T2 - Application to randomized, controlled trials of photocoagulation in the treatment of diabetic retinopathy

AU - Duffy, Stephen W.

AU - Rohan, Thomas E.

AU - Altman, Douglas G.

PY - 1989/8

Y1 - 1989/8

N2 - A method for combining matched and unmatched data is described and was applied to the results of randomized, controlled trials of photocoagulation in the treatment of diabetic retinopathy. A pooled estimate from the matched and unmatched studies war obtained by adaptation of the Mantel-Haenszel method, where the strata were unmatched studies and matched pairs within studies. A test of significance was based on the Mantel-Haenszel chi-square statistic, the latter also being used to calculate test-based confidence intervals. A test of homogeneity was performed by combining Mantel-Haenszel chi-square statistics from the matched and unmatched studies. By these methods, the combined estimate of the risk of deterioration of visual acuity for those receiving photocoagulation (relative to a risk of unity for those not receiving photocoagulation) was 0.37 (95 per cent confidence interval 0.29-0.46). The chi-square statistic (1 df) for an effect of treatment was 87.75 (p < 0.0001). The chi-square statistic for homogeneity of relative risk among studies was 11.20 (4 df, p < 0.05). However, this result was influenced disproportionately by one small matched study.

AB - A method for combining matched and unmatched data is described and was applied to the results of randomized, controlled trials of photocoagulation in the treatment of diabetic retinopathy. A pooled estimate from the matched and unmatched studies war obtained by adaptation of the Mantel-Haenszel method, where the strata were unmatched studies and matched pairs within studies. A test of significance was based on the Mantel-Haenszel chi-square statistic, the latter also being used to calculate test-based confidence intervals. A test of homogeneity was performed by combining Mantel-Haenszel chi-square statistics from the matched and unmatched studies. By these methods, the combined estimate of the risk of deterioration of visual acuity for those receiving photocoagulation (relative to a risk of unity for those not receiving photocoagulation) was 0.37 (95 per cent confidence interval 0.29-0.46). The chi-square statistic (1 df) for an effect of treatment was 87.75 (p < 0.0001). The chi-square statistic for homogeneity of relative risk among studies was 11.20 (4 df, p < 0.05). However, this result was influenced disproportionately by one small matched study.

KW - Epidemiologic methods;

KW - Statistics

UR - http://www.scopus.com/inward/record.url?scp=0024387008&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024387008&partnerID=8YFLogxK

M3 - Article

VL - 130

SP - 371

EP - 378

JO - American Journal of Epidemiology

JF - American Journal of Epidemiology

SN - 0002-9262

IS - 2

ER -