Incidence density matching with a simple SAS computer program

Our plain-English summary of Incidence density matching with a simple SAS computer program, by N. Pearce (International Journal of Epidemiology, 1989).

The study

• Author: N. Pearce
• Journal: International Journal of Epidemiology
• Published: 1989;18(4):981-4
• Cited by: 17 articles, as counted by NIH iCite
• Read it: PubMed · Publisher
• In J-M's library: Johns-Manville's company medical library held a copy (J-M no. 002541).

Abstract

Incidence density matching is the method of choice for selecting controls in nested case-control studies, but its use has been limited by its computational complexity. We describe incidence density matching in nested case-control studies with a simple and flexible SAS computer program. The program contains no restrictions on the number of matching factors, although it is usually only necessary to match on age (in incidence density fashion) and on gender. The program is illustrated with data from an historical cohort study of the mortality experience of workers at an asbestos textile manufacturing plant.

In plain English

This was a methods paper, not a new study of asbestos health effects. It described a simple SAS computer program for choosing controls by incidence density matching in nested case-control studies, where each case is compared with people still at risk at the same age. To demonstrate it, the author used data from an earlier historical cohort study by Dement et al. of white male workers at an asbestos textile plant. The cases were 35 lung cancer deaths from 1940 to 1975, each matched on age at risk to four controls, giving 140 controls. Exposure histories existed for the whole cohort, so the case-control results could be checked against a full cohort analysis across five categories of cumulative exposure, measured in thousand fibre/cc days. For 40-99 thousand fibre/cc days, the cohort rate ratio was 6.82 (95% confidence interval 2.34-19.85) and the case-control odds ratio was 6.86 (1.74-27.04). The two methods generally gave similar findings, except in the highest category, where numbers were small.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Methods paper presenting a SAS computer program for incidence density matching, illustrated with existing data from a historical cohort study. The nested case-control analysis was done only for illustration.
• Who was studied: White male workers at an asbestos textile manufacturing plant, from the earlier Dement et al. cohort study (labelled South Carolina in Table 2). The cases were 35 lung cancer deaths during 1940 to 1975, and there were 140 matched worker controls.
• Exposure: Cumulative asbestos exposure, divided into five categories of thousand fibre/cc days. The lowest category was the reference.
• Main results:
  • A 4:1 matching ratio was used for the 35 lung cancer deaths, giving 140 worker controls matched on age at risk.
  • Cumulative exposure of 40-99 thousand fibre/cc days: the full cohort analysis gave a rate ratio of 6.82 (95% confidence interval 2.34-19.85), and the case-control analysis gave an odds ratio of 6.86 (95% confidence interval 1.74-27.04).
  • Cumulative exposure of 100+ thousand fibre/cc days: the cohort rate ratio was 8.10 (95% confidence interval 1.52-43.21), and the case-control odds ratio was 5.55. The upper confidence limit for that odds ratio is unreadable in the scanned text.
  • The nested case-control analysis generally gave similar findings to the full cohort analysis, except in the highest exposure category, where numbers were small.
  • The confidence limits for the case-control odds ratios were only slightly wider than those for the cohort rate ratios.
• The authors' conclusion: The author concluded that the nested case-control analysis using the program generally gave findings similar to the full cohort analysis, and that 4:1 matching gave relatively good precision.
• Limitations the authors state:
  • The program was kept simple for illustration and matched only on age at risk, not on other potential confounders.
  • Numbers were small in the highest exposure category, where the two analyses differed.
  • The case-control design was used only to illustrate the method, because exposure histories were available for the entire cohort.

Topics

Medical terms in this study, explained in our Lexicon: Cohort Study.

Related studies

• Role of postoperative radiotherapy in the management of malignant pleural mesothelioma : A propensity score matching of the SEER database — O. Abdel-Rahman (2017)
• Computer modeling of population exposure to a carcinogen: the example of asbestos and mesothelioma mortality in France — A. Banaei, B. Auvert, et al. (2000)
• A simple approach for fitting linear relative rate models in SAS — D.B. Richardson (2008)
• Simple benchmark for mesothelioma projection for Great Britain — M.D. Martínez-Miranda, B. Nielsen, et al. (2016)
• Occupational exposures contribute to educational inequalities in lung cancer incidence among men: Evidence from the EPIC prospective cohort study — G. Menvielle, H. Boshuizen, et al. (2010)
• Chronic obstructive lung diseases and risk of non-small cell lung cancer in women — A.G. Schwartz, M.L. Cote, et al. (2009)

Browse more research in International Journal of Epidemiology, or return to medical research.