Cancer Causes & Control

, Volume 16, Issue 4, pp 383–388 | Cite as

Choice of exposure scores for categorical regression in meta-analysis: a case study of a common problem

  • Dora Il’yasovaEmail author
  • Irva Hertz-Picciotto
  • Ulrike Peters
  • Jesse A. Berlin
  • Charles Poole


Objective: Reporting categorical relative risk estimates for a series of exposure levels versus a common reference category is a widespread practice. In meta-analysis, categorical regression estimates a dose–response trend from such results. This method requires the assignment of a single score to each exposure category. We examined how closely meta-analytical categorical regression approximates the results of analysis based on the individual-level continuous exposure.

Methods: The analysis included five studies on tea intake and outcomes related to colorectal cancer. In addition, we derived categorical mean and median values from published distributions of tea consumption in similar populations to assign scores to the categories of tea intake when possible. We examined whether these derived mean and median values well approximate the individual-level results.

Results: In meta-analytical categorical regression, using the midrange scores approximated the individual-level continuous analyses reasonably well, if the value assigned to the uppermost, open-ended category was at least as high as the lower bound plus the width of the second-highest category. Categorical mean values derived from the published distributions of regular tea (in the US) and green tea (in Japan) well approximated the slope obtained from individual-level analysis.

Conclusion: Publication of both the categorical and the continuous estimates of effect in primary studies, with their standard errors, can enhance the quality of meta-analysis, as well as providing intrinsically valuable information on dose–response.


colorectal cancer dose–response relationships epidemiology epidemiologic method meta-analyses tea. 



confidence limit ratios


relative risk


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Greenland, S 1998Basic methods for sensitivity analysis and external adjustmentRothman, KJGreenland, S eds. Modern EpidemiologyLippincott-Raven PublishersPhiladelphia343358Google Scholar
  2. 2.
    Greenland, S, Longnecker, MP 1992Methods for trend estimation from summarized dose–response data, with applications to meta-analysisAm J Epidemiol13513011309PubMedGoogle Scholar
  3. 3.
    Berlin, JA, Longnecker, MP, Greenland, S 1993Meta-analysis of epidemiologic dose–response dataEpidemiology4218228PubMedGoogle Scholar
  4. 4.
    Brown, KG 1999Lung cancer and environmental tobacco smoke: occupational risk to nonsmokersEnviron Health Perspect107885890PubMedGoogle Scholar
  5. 5.
    Villanueva, CM, Fernandez, F, Malats, N, Grimalt, JO, Kogevinas, M 2003Meta-analysis of studies on individual consumption of chlorinated drinking water and bladder cancerJ Epidemiol Community Health57166173PubMedGoogle Scholar
  6. 6.
    Ford, ES, Smith, SJ, Stroup, DF, Steinberg, KK, Mueller, PW, Thacker, SB 2002Homocyst(e)ine and cardiovascular disease: a systematic review of the evidence with special emphasis on case–control studies and nested case-control studiesInt J Epidemiol315970PubMedGoogle Scholar
  7. 7.
    Reynolds, K, Lewis, B, Nolen, JD, Kinney, GL, Sathya, B, He, J 2003Alcohol consumption and risk of stroke: a meta-analysisJAMA289579588PubMedGoogle Scholar
  8. 8.
    Norat, T, Lukanova, A, Ferrari, P, Riboli, E 2002Meat consumption and colorectal cancer risk: dose–response meta-analysis of epidemiological studiesInt J Cancer98241256PubMedGoogle Scholar
  9. 9.
    Spitzer, WO, Faith, JM, MacRae, KD 2002Myocardial infarction and third generation oral ontraceptives: aggregation of recent studiesHum Reprod1723072314PubMedGoogle Scholar
  10. 10.
    Bergstrom, A, Hsieh, CC, Lindblad, P, Lu, CM, Cook, NR, Wolk, A 2001Obesity and renal cell cancer – a quantitative reviewBr J Cancer85984990PubMedGoogle Scholar
  11. 11.
    Shi, JQ, Copas, JB 2004Meta-analysis for trend estimationStat Med23319PubMedGoogle Scholar
  12. 12.
    Mantel, N 1963Chi-Square Tests with One Degree of Freedom: Extensions of the Mantel-Haenszel ProcedureJ Am Stat Assoc53690700Google Scholar
  13. 13.
    Rothman, KJ 1986Modern epidemiologyLittle, BrownBostonGoogle Scholar
  14. 14.
    Peters, U, Poole, C, Arab, L 2001Does tea affect cardiovascular disease? A meta-analysisAm J Epidemiol154495503PubMedGoogle Scholar
  15. 15.
    Baron, JA, Gerhardsson, D, Ekbom, A 1994Coffee, tea, tobacco, and cancer of the large bowelCancer Epidemiol Biomarkers Prev3565570PubMedGoogle Scholar
  16. 16.
    Baron, JA, Greenberg, ER, Haile, R, Mandel, J, Sandler, RS, Mott, L 1997Coffee and tea and the risk of recurrent colorectal adenomasCancer Epidemiol Biomarkers Prev6710PubMedGoogle Scholar
  17. 17.
    Il’yasova, D, Arab, L, Martinchik, A, Weisgerber, UM, Sdvizhkov, A, Urbanovich, L 2003Black tea consumption and risk of rectal cancer in Moscow populationAnn Epidemiol13405411CrossRefPubMedGoogle Scholar
  18. 18.
    Kono, S, Shinchi, K, Ikeda, N, Yanai, F, Imanishi, K 1991Physical activity, dietary habits and adenomatous polyps of the sigmoid colon: a study of self-defense officials in JapanJ Clin Epidemiol4412551261PubMedGoogle Scholar
  19. 19.
    Su, LJ, Arab, L 2002Tea consumption and the reduced risk of colon cancer – results from a national prospective cohort studyPublic Health Nutr5419425PubMedGoogle Scholar
  20. 20.
    Poole, C 2001Low P-values or narrow confidence intervals: which are more durable?Epidemiology12291294PubMedGoogle Scholar
  21. 21.
    US Department of Health and Human Services (DHHS). Second National Health and Nutrition Examination Survey I Epidemiologic Followup Study, 1982–1984. NCHS. 2001. NCHS.Google Scholar
  22. 22.
    Inoue, M, Tajima, K, Hirose, K,  et al. 1997Epidemiological features of first-visit outpatients in Japan: comparison with general population and variation by sex, age, and seasonJ Clin Epidemiol506977PubMedGoogle Scholar
  23. 23.
    U.S. Environmental Protection Agency1986Guidelines for Carcinogen Risk AssessmentFed. Reg.513399234003Google Scholar
  24. 24.
    Hertz-Picciotto, I 1995Epidemiology and quantitative risk assessment: a bridge from science to policyAm J Public Health85484491PubMedGoogle Scholar
  25. 25.
    Shore, RE 1995Epidemiologic data in risk assessment – imperfect but valuableAm J Public Health85474476PubMedGoogle Scholar
  26. 26.
    Rohan, TE, McMichael, AJ 1988Alcohol consumption and risk of breast cancerInt J Cancer41695699PubMedGoogle Scholar
  27. 27.
    Cook, RJ, Brumback, BB, Wigg, MB, Ryan, LM 2001Synthesis of evidence from epidemiological studies with interval-censored exposure due to groupingBiometrics57671680PubMedMathSciNetGoogle Scholar
  28. 28.
    Bagnardi, V, Zambon, A, Quatto, P, Corrao, G 2004Flexible meta-regression functions for modeling aggregate dose–response data, with an application to alcohol and mortalityAm J Epidemiol15910771086PubMedGoogle Scholar
  29. 29.
    Hertz-Picciotto, I, Smith, AH 1993Observations on the dose–response curve for arsenic exposure and lung cancerScand J Work Environ Health19217226PubMedGoogle Scholar
  30. 30.
    Poole, C, Greenland, S 1999Random-effects meta-analyses are not always conservativeAm J Epidemiol150469475PubMedGoogle Scholar

Copyright information

© Springer 2005

Authors and Affiliations

  • Dora Il’yasova
    • 1
    • 4
    Email author
  • Irva Hertz-Picciotto
    • 1
    • 5
  • Ulrike Peters
    • 2
  • Jesse A. Berlin
    • 3
  • Charles Poole
    • 1
  1. 1.Department Epidemiology, School of Public HealthUniversity of North CarolinaChapel HillUSA
  2. 2.Division of Cancer Epidemiology and GeneticsNational Cancer Institute, NIH, DHHSBethesdaUSA
  3. 3.Center for Clinical Epidemiology and Biostatistics and Department of Biostatistics and EpidemiologyUniversity of Pennsylvania School of MedicineUSA
  4. 4.Medical Center BoulevardWake Forest University School of MedicineWinston-Salem
  5. 5.Department of Epidemiology and Preventive MedicineUniversity of CaliforniaDavisUSA

Personalised recommendations