Harlow, S.D., Cohen, M. Ohmit, S.E. and others (2003, April). "Substance use and psychotherapeutic medications: A likely contributor to menstrual disorders in women who are seropositive for human immunodeficiency virus." (Cosponsored by AHRQ, NIH, and CDC). American Journal of Obstetrics & Gynecology 188, pp. 881-886.
Women with HIV disease commonly report irregular periods. These investigators previously reported that HIV infection per se, in the absence of advanced immunodeficiency, did not appear to have a substantial direct effect on the length of menstrual cycles in otherwise healthy women. The researchers prospectively collected the menstrual calendars for 1,075 women who were HIV seropositive or seronegative and enrolled in the Women's Interagency HIV Study (WIHS) or the HIV Epidemiology Research Study. Women who received methadone maintenance and who used injection drugs were substantially more likely to have a menstrual cycle of 90 days or longer. Women who used psychotherapeutic medications such as antidepressants nearly doubled the risk that they would have very short cycles (less than 18 days) or cycles of 90 or more days. Clinicians should consider the neuroendocrinologic effects of these medications as a potential cause of menstrual disruptions in women who have HIV infection.
Ioannidis, J.P., and Lau, J. (2003). "F-FDG PET for the diagnosis and grading of soft-tissue sarcoma: A meta-analysis." (AHRQ contract 290-97-0019) Journal of Nuclear Medicine 44(5), pp. 717-724.
F-FDG positron emission tomography (PET) is considered a potential advance in clinical practice. It may offer information not only about the diagnosis and grading of tumors, but also the behavior of tumors, thus helping to guide therapeutic choices. This meta-analysis of studies of the diagnostic and grading performance of F-FDG PET for soft-tissue sarcoma concluded that it has very good discriminating ability in the evaluation of both primary and recurrent soft-tissue lesions. F-FDG PET may be helpful in tumor grading, but it offers inadequate discrimination between low-grade tumors and benign lesions. The meta-analysis included 15 studies with 441 soft-tissue lesions (227 malignant and 214 benign).
Kaushal, R., Shojania, K.G., and Bates, D.W. (2003, June). "Effects of computerized physician order entry and clinical decision support systems on medication safety." (AHRQ contract no. 290-97-0013) Archives of Internal Medicine 163, pp. 1409-1416.
Computerized physician order entry (CPOE) and clinical decision support systems (CDSSs) can substantially reduce medication error rates. However, most studies have not been powered to detect differences in adverse drug events and have evaluated a small number of "homegrown" systems, concludes this study. After a thorough search of the research literature, the investigators identified five trials that met criteria for assessing CPOE and seven for assessing isolated CDSSs. Of the CPOE studies, two demonstrated a marked decrease in the serious medication error rate, one an improvement in corollary orders, one an improvement in five prescribing behaviors, and one an improvement in nephrotoxic drug dose and frequency. Of the seven studies evaluating isolated CDSSs, three demonstrated significant improvements in antibiotic-associated medication errors or adverse drug events and one an improvement in theophylline-associated medication errors. The remaining three studies had nonsignificant results.
McCabe, M., Morgan, F., Smith, M., and others (2003, June). "Challenges in interpreting diabetes concepts in the Navajo language." (AHRQ grant HS10637). Diabetes Care 26(6), pp. 1913-1914.
During development of a pamphlet on diabetes for Navajo Indians, these researchers found that they had to pay attention to cultural factors, regional language differences, and the possibility of a lack of word-for-word translation. Translators initially recorded an oral Navajo translation of the Michigan Diabetes Knowledge Test. A Navajo language expert translated the taped version back into written English and then prepared another Navajo version. The revised translation was verified with selected elderly community members, health care professionals, and others. By probing the linguistic, cultural, and regional language issues in translating an apparently simple questionnaire from English into Navajo, the translators were able to articulate approaches that can be used in explaining diabetes management in an appropriate cultural context.
Nemeth, L.S. (2003). "Implementing change for effective outcomes." (AHRQ grant HS11132). Outcomes Management 7(3), pp. 134-139.
This article analyzes the concept of change illustrated through a quality improvement intervention-based research project at six practice sites. The researcher disseminated clinical guidelines for cardiovascular disease and stroke prevention to various U.S. primary care practices. Ten intervention sites were offered education-related academic detailing and assistance with action planning to improve clinical practice adherence. The researcher assessed the change process underway at six practices and concluded that to implement change effectively, it is necessary to provide clear vision, leadership, and adequate time to develop followers. Coordination of activities and integration of changes in practice to promote positive outcomes are also needed for success.
Pathman, D.E., Konrad, T.R., and Agnew, C.R. (2003, Summer). "Predictive accuracy of rural physicians' stated retention plans." (AHRQ grants HS10654 and HS06544). Journal of Rural Health 19(3), pp. 236-244.
To test the predictive accuracy of rural physicians' stated plans to stay in a rural area, these investigators analyzed responses to a 1991 mail survey of rural physicians' retention plans and responses to a followup survey 5 to 6 years later to determine if and when respondents had moved. Predictions for individuals were moderately accurate: four of five physicians who predicted remaining at least 5 years did so; two of three who predicted remaining less than 5 years did leave before 5 years. Predictions of job changes in less than 2 years tended to be more accurate than predictions of 2 to 5 years. The researchers conclude that rural generalist physicians are moderately accurate when reporting how much longer they will remain in their jobs. This validates the use of anticipated retention in rural health workforce studies.
Terrin, N., Schmid, C.H., Lau, J., and Olkin, I. (2003). "Adjusting for publication bias in the presence of heterogeneity." (AHRQ grant HS10254). Statistics in Medicine 22, pp. 2113-2126.
The existence of publication bias can influence the conclusions of a meta-analysis. Some methods have been developed to deal with publication bias, but issues remain. One particular adjustment method called "trim and fill" is intuitively appealing and comprehensible by nonstatisticians. It is based on a simple and popular graphical tool called the funnel plot. These authors present a simulation study designed to evaluate the behavior of this method. They found that when the studies are heterogeneous (that is, when they estimate different effects), trim and fill may inappropriately adjust for publication bias where none exists. Thus, this approach is inappropriate for heterogeneous meta-analyses.
Internet Citation:
Research Activities newsletter. September 2003, No. 277. AHRQ Publication No. 03-0046. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/sep03/
540 Gaither Road Rockville, MD 20850