Annals Reports on Postpartum Health, Comorbidity in Research, Pay-for-Performance

Employed mothers at five weeks postpartum show a need for rest and recovery beyond the traditional leave period of about six weeks; the exclusion of patients with comorbidities from randomized controlled trials limits the applicability of research results; and payment for performance in primary care in accord with guidelines should be undertaken only when the scientific evidence for the guidelines is relevant to primary care. These are a sampling of conclusions in articles in the March/April Annals of Family Medicine.

"The Postpartum Health of Employed Mothers 5 Weeks After Childbirth" by Pat McGovern, Ph.D., of the University of Minnesota, Minneapolis, and others reports on interviews with 716 women who intended to return to work after childbirth. "Most women contend with several minor to moderate discomforts for weeks … and some may face serious problems, such as depression, that may limit daily activities for months," say the researchers.

Health concerns were greater among women who were breastfeeding and those whose babies were delivered by cesarean section, the researchers found. "Amidst the growing trend toward cesarean sections, parents and clinicians need to acknowledge the risks of operative deliveries," say the authors. "Some mothers may adapt better to their return to work by using an intermittent rather than straight-time family and medical leave." The Family and Medical Leave Act allows return to work on a gradual, part-time basis for a longer time than is typical, but physicians must certify the intermittent leave as necessary for the mother's serious health condition related to childbirth and distinct from time for infant bonding, the authors note.

"Randomized Controlled Trials: Do They Have External Validity for Patients With Multiple Comorbidities?" by Martin Fortin, M.D., M.Sc., of Sherbrooke University, Sherbrooke, Quebec, and others calls into question the practice of excluding patients with comorbidities from randomized controlled trials, or RCTs, to help ensure internal validity. Even when patients with comorbid conditions are not excluded from RCTs, patients' comorbidities may not be reported in the final results, say the authors. In either case, the external validity of these trials may be compromised, limiting their widespread applicability.

Out of 424 patients in the data set for Fortin's study who would have matched criteria for participation in any of five RCTs studying hypertension, between 89 percent and 100 percent had comorbidities. "The increasing health problem of comorbidity should not be underestimated," say the authors. They warn that in this instance, the treatment of the majority of patients with hypertension "could be inadequate because of the evidence it is based on."

In the editorial "Threads and Yarns: Weaving the Tapestry of Comorbidity," Barbara Starfield, M.D., M.P.H., of Johns Hopkins University, Baltimore, comments on the article by Fortin and others. Rules for clinical trials should mandate characterization of participants according to their total morbidity burden, says Starfield.

Moreover, she adds, "Payment for performance (based on adherence to guidelines) in primary care should be undertaken only when scientific evidence is known to be relevant to primary care practice."

Developers of performance indicators for general practitioners in the United Kingdom "recognized that the application of disease-specific guidelines is inappropriate for everyone with the disease. … In the United States, where performance measurement is a thriving enterprise, there has been little attention to developing a basis for judging the appropriateness of performance measures in patients with comorbidity," Starfield observes.

Other research reports and reflections in this issue of Annals include