Table of Contents

About one half of patients who take efavirenz develop disturbing neuropsychiatric side effects. This randomized trial involving 114 patients compared full-dose efavirenz with stepped-dose efavirenz during the first 2 weeks of treatment. Patients who received stepped-dose therapy had fewer side effects during the first week of treatment, and after 24 weeks, HIV RNA levels and CD4+ cell counts were similar in both groups.

Rapid control of HIV infection is especially important in resource-limited countries that have high rates of opportunistic infections. The investigators created a computer simulation of HIV disease using data from randomized trials and observational cohorts in South Africa and then performed a cost-effectiveness analysis of 3 options: no treatment and treatment initiation at CD4 count thresholds of 0.250 and 0.350 × 10⁹ cells/L. Compared with a threshold of 0.250 × 10⁹ cells/L, starting therapy at 0.350 × 10⁹ cells/L was highly cost-effective. Therapy at a higher CD4 count may prove superior in South Africa.

Darwish Murad and colleagues describe 163 patients with newly diagnosed hepatic venous outflow obstruction (the Budd–Chiari syndrome). Most of the patients had identifiable risk factors for clot, most commonly myeloproliferative disorders. About half were managed noninvasively with diuretics and anticoagulation, while half were managed with transjugular shunting, recanalization, and liver transplantation. Survival rates were good and were better than previously described.

Clostridium difficile infection (CDI) seems to be increasing in virulence and incidence, becoming more resistant to metronidazole, and producing worse outcomes. Three common misconceptions lead to misdiagnoses: CDI is possible when the patient has fewer than 3 loose stools per day; the glutamate dehydrogenase test is a good initial test for CDI; and repeating an insensitive laboratory test for CDI is useful. Diagnoses of CDI will be more accurate if clinicians reduce the frequency of testing for a single episode of diarrhea, use tests with a higher sensitivity, and give more attention to the patient's history.

The body of evidence regarding associations between long-distance travel and risk for venous thromboembolism (VTE) is heterogeneous and inconclusive. Chandra and colleagues sought to clarify the risk for VTE in travelers and identify reasons for the contradictory results of previous studies. Among 14 eligible studies, the pooled relative risk for VTE was 2.0 (95% CI, 1.5 to 2.7). After 6 case–control studies were eliminated, the relative risk was 2.8 (CI, 2.2 to 3.7). The excluded studies had control participants with a different risk for VTE than the source population for the case patients.

Terasawa and colleagues performed a systematic review to determine whether radiofrequency catheter ablation is better than medical therapy for patients with atrial fibrillation. They found that radiofrequency ablation after a failed drug course maintained sinus rhythm more often than continuation of drug therapy alone. Some studies indicated that ablation improved quality of life but did not necessarily reduce stroke rates compared with medical therapy. Fewer than 5% of patients undergoing ablation reportedly experienced major adverse events, such as pulmonary-vein stenosis or cardiac tamponade.

The U.S. Congress asked the Institute of Medicine to elicit input from stakeholders on which research topics should have the highest priority for comparative effectiveness research (CER) funding and to develop a list of the highest-priority topics for the Secretary of Health and Human Services to consider. This article describes the Institute of Medicine committee's process in defining CER, identifying high-priority research topics, and establishing a sustainable national program of CER.

This companion article to Sox and Greenfield's report describes better research methods for CER. It addresses the fundamental limitations of traditional randomized, controlled trials for meeting CER objectives and offers 3 potentially transformational approaches to enhance operational efficiency, analytic efficiency, and generalizability of trials for CER.

In this issue, Walensky and colleagues report the results of a mathematical model of the effects of antiretroviral therapy for HIV in South Africa. The editorialists discuss the merits and limitations of Walensky and colleagues' methods and the implications of modeling studies, which are likely to continue to be an essential tool for designing clinical research and for extrapolating its results.

In this issue, Chandra and colleagues present a meta-analysis that is based on case–control studies. This editorial discusses the selection of controls for case–control studies, using Chandra and colleagues' study as an example.

When we arrived at the oncology clinic, Mr. Liu's smile changed to a frown. The department consisted of row upon row of beds, crowded with patients and their relatives.