Letters to the Editor: Addressing Duplicate Publication Concerns

Regarding Takahashi PY, Chandra A, Cha SS, Crane SJ. A predictive model for venous ulceration in older adults: results of a retrospective cohort study. Ostomy Wound Manage. 2010;56(4):60–66

Dear Editor,

  This brief letter of explanation and discussion describes the major findings in the two articles under comparison: A cross-sectional evaluation of the association between lower extremity venous ulceration and predictive risk factors¹ and A predictive model for venous ulceration in older adults: results of a retrospective cohort study.² While the editors of both journals believe there was overlap between the two articles, each article provides fundamentally different learning points and each possesses individual strengths and weaknesses.

  Three important points of difference involving the study designs, the study populations, and the main outcomes should be considered. These issues make each study unique and improve the reader’s understanding of venous ulcer risk factors. Possible areas of confusion include the initial population, which was the same for both studies. Many studies use a common population, including the Framingham Heart study with three generations of cohorts,³ the Women’s Health Initiative,⁴ and other large observational cohort studies. These studies define one population that is analyzed to provide different epidemiological designs and different outcomes.

  The first important point of difference between the articles was the different study designs. The initial study in November 2009 was cross-sectional and involved risk factors and outcomes at a single point of time (January 1, 2005). The cross-sectional design is often then used in the initial epidemiologic evaluation of a question. Cross-sectional designs cannot determine incidence (new ulcer formation) and cannot detect causality (which came first).⁵ Understanding these limitations, a different study is typically designed to answer these questions — we undertook a retrospective cohort study that looked for new ulcer development over the following two years (from January 1, 2005 to December 31, 2007). This provided new outcome data for all subjects in the population. This cohort study enables stronger evidence of causality because subjects had the risk factors before developing venous ulcers.⁵ Further, it allowed us to use venous ulcers as a risk factor, which was impossible in the cross-sectional study because it was the outcome.

  The second point involved different cohorts for each of the studies. Although the population was the same, we derived the cohorts differently and had different numbers of cases (those with outcomes) and controls (those without). This would be equivalent to using a deck of cards and drawing 12 cards and comparing them to the remaining 40 cards for average value, percentage of clubs, and the like. In a separate evaluation, 24 cards would be chosen and would be compared to the remaining 28 cards. In these studies, 215 patients developed a new ulcer in the cohort study² compared to 581 in the cross-sectional study.¹ Both studies had different cases and different controls that make them unique.

  The important learning point for the reader involves the fundamentally different outcomes for each study, the third concern. Understanding of risk factors evolved between studies. The cross-sectional study only found that venous insufficiency was a major risk for venous ulceration with an odds ratio >2.¹

  This provides an incomplete answer, given the vast number of patients with venous insufficiency. In the retrospective cohort study, the major risk factor was previous venous ulceration with an OR 18.66; CI 12.96-24.96.² Other interesting factors included previous decubitus ulcers as a risk factor.²

  Unfortunately, the risk model did not provide useful information for clinicians and the take-home point for prevention of venous ulcers is focusing on patients who have had previous venous ulcers. Those are the patients who should be targeted for aggressive edema compression.

This article was not subject to the Ostomy Wound Management peer-review process.

1. Takahashi PY, Chandra A, Cha SS. A cross-sectional evaluation of the association between lower extremity venous ulceration and predictive risk factors. Wounds. 2009;21(11):290–296.

2. Takahashi PY, Chandra A, Cha SS, Crane SJ. A predictive model for venous ulceration in older adults: results of a retrospective cohort study. Ostomy Wound Manage. 2010;56(4):60–66.

3. Splansky GL, Corey D, Yang Q, et al. The third generation cohort of the National Heart, Lung, and Blood Institute's Framingham Heart Study: design, recruitment, and initial examination. Am J Epidemiol. 2007;165(11):1328–1335.

4. Rossouw JE, Finnegan LP, Harlan WR, Pinn VW, Clifford C, McGowan JA. The evolution of the Women's Health Initiative: perspectives from the NIH. J Am Med Womens Assoc. 1995;50(2):50–55.

5. Lang TA, Secic M. How to Report Statistics in Medicine. Philadelphia, PA: American College of Physicians; 2006.