In this issue, Wang et al explore patterns of outpatient hypertension treatment in 5 western Europe countries and the United States. Using a representative sample of 21 053 office visits in 2004, they find that the initial pretreatment blood pressure levels were lowest and the use of combination drug therapy was highest in the United States. Findings from multivariate analyses indicated that compared with US patients, European patients had higher last-measured blood pressure, a smaller likelihood of hypertension control, and a smaller likelihood of medication increase for inadequately controlled hypertension. Better hypertension control in the United States likely results from lower thresholds for initiation and more intensive treatment.
Some studies have found that psychosocial factors are risk factors for coronary artery disease. Inflammation could be one of the biological mechanisms linking psychosocial stress to the development of coronary artery disease. Ranjit et al investigated cross-sectional associations between psychosocial factors and inflammatory markers IL-6, C-reactive protein, and fibrinogen, in a US study of 6814 men and women aged 45 to 84 years and free of cardiovascular disease. Cynical distrust and chronic stress were associated with elevated levels of IL-6 and C-reactive protein. Depression was positively associated with IL-6. No associations remained after controlling for behaviors, body mass index, and diabetes. The findings indicate that psychosocial factors may play a role in inflammatory processes, most likely through their association with behaviors, body mass index, and diabetes mellitus.
Kapoor et al performed a meta-analysis of diagnostic imaging studies of suspected foot osteomyelitis using summary receiver operating characteristic curves. They evaluated the performance of magnetic resonance imaging (MRI) and compared this performance with that of plain radiography, technetium bone scanning, and white blood cell (WBC)-labeled studies. Sixteen studies met inclusion criteria. The diagnostic odds ratio for MRI was 42.1 (95% confidence interval, 14.8-119.9), which corresponds to a specificity of 82.5% at a preselected cutpoint of 90% sensitivity. In studies in which bone scanning, WBC studies, or plain radiograph were compared with MRI as diagnostic tests, the performance for MRI was significantly higher. These results indicate that MRI has strong diagnostic test performance for foot osteomyelitis.
To determine whether repeat bone mineral density (BMD) measurement adds benefit beyond the initial BMD measurement in predicting fractures, Hillier et al prospectively measured total hip BMD in 4124 older women (mean ± SD age, 72 ± 4 years) in 1989-1990 and 8 years later. Initial and repeat BMD were similarly associated with fracture risk (per unit standard deviation lower in BMD) for nonspine (hazard ratio, 1.6), spine (odds ratio, 1.8-1.9), and hip fracture (hazard ratio, 2.0-2.2) (P<.001 for all models). Areas under the receiver operating characteristic curves (AUCs) revealed no significant differences to discriminate nonspine (AUC, 0.65), spine (AUC, 0.67-0.68), or hip fractures (AUC, 0.73-0.74) between models with initial BMD, repeat BMD, or initial BMD + change in BMD. Our results indicate that in healthy older postmenopausal women, repeating a measurement of BMD up to 8 years later provides little additional value besides the initial BMD measurement for predicting incident fractures.
Verification bias occurs when test findings influence the decision to perform a gold standard test. It adversely influences diagnostic test accuracy by inflating sensitivity and deflating specificity. Lauer et al studied the impact of verification bias on the estimated accuracy of positron emission tomography (PET) scanning in 534 patients with suspected lung cancer, without excluding patients who did not undergo subsequent performance of a gold standard tissue biopsy. Clinicians were much less likely to refer patients with stage 0 PET scans for tissue diagnosis; once this was accounted for, the apparent sensitivity of PET for diagnosis of any cancer fell from 95% to 85%. On the other hand, clinicians largely ignored PET scan results when deciding whether to biopsy mediastinal tissue, and diagnostic test accuracy for mediastinal metastases was not affected by verification bias. Although patients with stage 0 PET disease were much less likely to undergo biopsy, their prognosis was poor, with a 14% per year death rate and with one third of deaths due to eventually diagnosed lung cancer.
In This Issue of Archives of Internal Medicine. Arch Intern Med. 2007;167(2):111. doi:10.1001/archinte.167.2.111