Early Performance in Medicaid Accountable Care Organizations: A Comparison of Oregon and Colorado.

Importance: Several state Medicaid reforms are under way, but the relative performance of different approaches is unclear. Objective: To compare the performance of Oregon's and Colorado's Medicaid Accountable Care Organization (ACO) models. Design, Setting, and Participants: Oregon initiated its Medicaid transformation in 2012, supported by a $1.9 billion investment from the federal government, moving most Medicaid enrollees into 16 Coordinated Care Organizations, which managed care within a global budget. Colorado initiated its Medicaid Accountable Care Collaborative in 2011, creating 7 Regional Care Collaborative Organizations that received funding to coordinate care with providers and connect Medicaid enrollees with community services. Data spanning July 1, 2010, through December 31, 2014 (18 months before intervention and 24 months after intervention, treating 2012 as a transition year) were analyzed for 452 371 Oregon and 330 511 Colorado Medicaid enrollees, assessing changes in outcomes using difference-in-differences analyses of regional focus, primary care homes, and care coordination. Oregon's Coordinated Care Organization model was more comprehensive in its reform goals and in the imposition of downside financial risk. Exposures: Regional focus, primary care homes, and care coordination in Medicaid ACOs. Main Outcomes and Measures: Performance on claims-based measures of standardized expenditures and utilization for selected services, access, preventable hospitalizations, and appropriateness of care. Results: In a total of 782 882 Medicaid enrollees, 45.0% were male, with mean (SD) age 16.74 (14.41) years. Standardized expenditures for selected services declined in both states during the 2010-2014 period, but these decreases were not significantly different between the 2 states. Oregon's model was associated with reductions in emergency department visits (-6.28 per 1000 beneficiary-months; 95% CI, -10.51 to -2.05) and primary care visits (-15.09 visits per 1000 beneficiary-months; 95% CI, -26.57 to -3.61), improvements in acute preventable hospital admissions (-1.01 admissions per 1000 beneficiary-months; 95% CI, -1.61 to -0.42), 3 of 4 measures of access (well-child visits, ages 3-6 years, 2.69%; 95% CI, 1.20% to 4.19%; adolescent well-care visits, 6.77%; 95% CI, 5.22% to 8.32%; and adult access to preventive ambulatory care, 1.26%; 95% CI, 0.28% to 2.25%), and 1 of 4 measures of appropriateness of care (avoidance of head imaging for uncomplicated headache, 2.59%; 95% CI, 1.35% to 3.83%). Conclusions and Relevance: Two years into implementation, Oregon's and Colorado's Medicaid ACO models exhibited similar performance on standardized expenditures for selected services. Oregon's model, marked by a large federal investment and movement to global budgets, was associated with improvements in some measures of utilization, access, and quality, but Colorado's model paralleled Oregon's on several other metrics.

Surveillance after treatment for cervical intraepithelial neoplasia: outcomes, costs, and cost-effectiveness.

OBJECTIVE: To estimate outcomes and costs of surveillance strategies after treatment for high-grade cervical intraepithelial neoplasia (CIN). METHODS: A hypothetical cohort of women was evaluated after treatment for CIN 2 or 3 using a Markov model incorporating data from a large study of women treated for CIN, systematic reviews of test accuracy, and individual preferences. Surveillance strategies included initial conventional or liquid-based cytology, human papillomavirus testing, or colposcopy 6 months after treatment, followed by annual or triennial cytology. Estimated outcomes included CIN, cervical cancer, cervical cancer deaths, life expectancy, costs, cost per life-year, and cost per quality-adjusted life-year. RESULTS: Conventional cytology at 6 and 12 months, followed by triennial cytology, was least costly. Compared with triennial cytology, annual cytology follow-up reduced expected cervical cancer deaths by 73% to 77% and had an average incremental cost per life-year gained of $69,000 to $81,000. For colposcopy followed by annual cytology, the incremental cost per life-year gained ranged from $70,000 to more than $1 million, depending on risk. Between-strategy differences in mean additional life expectancy per woman were less than 4 days; differences in mean incremental costs per woman were as high as $822. In the cost-utility analysis, colposcopy at 6 months followed by annual cytology had an incremental cost per quality-adjusted life-year of less than $5,500. Human papillomavirus testing or liquid-based cytology added little to no improvement to life-expectancy with higher costs. CONCLUSION: Annual conventional cytology surveillance reduced cervical cancers and cancer deaths compared with triennial cytology. For high risk of recurrence, a strategy of colposcopy at 6 months increased life expectancy and quality-adjusted life expectancy. Human papillomavirus testing and liquid-based cytology increased costs, but not effectiveness, compared with traditional approaches.

Periodontal health of older men: the MrOS dental study.

OBJECTIVE: The purpose of this study was to evaluate the prevalence and severity of periodontitis in men of 65+ years and identify demographic and lifestyle factors associated with its presence. METHODS: Participants were recruited from the Osteoporotic Fractures in Men Study, a longitudinal study of risk factors for fractures in older men. Dental measures included clinical attachment loss (CAL), pocket depth (PD), calculus, plaque and bleeding on a random half-mouth, plus a questionnaire regarding access to care, symptoms and previous diagnosis. RESULTS: 1210 dentate men completed the dental visit. Average age was 75 years, 39% reported some graduate school education, 32% smoked 20 + pack years and 88% reported their overall health as excellent/good. In terms of periodontal health, 38% had sub-gingival calculus, 53% gingival bleeding, 82% CAL > or =5 mm and 34% PD > or =6 mm. The prevalence of severe periodontitis was 38%. Significant demographic and lifestyle factors associated with severe periodontitis in multivariate analyses included age > or =75 (OR 1.4, 95% CI 1.1-1.7) non-white race (OR 1.9, 95% CI 1.3-2.8), less than an annual dental visit (OR 1.5, 95% CI 1.1-2.0), and 20 + pack years (OR 2.1, 95% CI 1.6-2.7). CONCLUSION: A high proportion of healthy older men have evidence of periodontal destruction which could, given the growing ageing population, have a significant impact on the dental profession's ability to provide preventive and therapeutic care. The population at highest risk of periodontitis in MrOS is older minority men who smoke and do not have annual dental visits.

Posttreatment human papillomavirus testing for recurrent cervical intraepithelial neoplasia: a systematic review.

OBJECTIVE: We conducted a systematic review to evaluate the characteristics of human papillomavirus testing, particularly Hybrid Capture 2, in follow-up evaluations after treatment for cervical intraepithelial neoplasia for the detection of residual or recurrent cervical intraepithelial neoplasia grade >/= 2. STUDY DESIGN: Medline was searched for relevant studies that were published between 1992 and September 2007. Of the 1107 citations that were identified, 20 articles met the inclusion criteria. RESULTS: Studies that used polymerase chain reaction testing were too heterogeneous to combine. We identified 5 studies that performed both Hybrid Capture 2 and colposcopy. Pooled sensitivity for Hybrid Capture 2 was 90.7% (95% CI, 75.4-96.9%), and pooled specificity was 74.6% (95% CI, 60.4-85.0%). Pooled sensitivity for cervical cytologic testing was 76.6% (95% CI, 62.0-86.8%), and pooled specificity was 89.7% (95% CI, 22.7-99.6%). CONCLUSION: Hybrid Capture 2 testing can identify approximately 91% of women with residual or recurrent cervical intraepithelial neoplasia grade >/= 2; however, approximately 30% of women would undergo colposcopy in follow-up evaluation.

Gabapentin versus tricyclic antidepressants for diabetic neuropathy and post-herpetic neuralgia: discrepancies between direct and indirect meta-analyses of randomized controlled trials.

BACKGROUND: Previous systematic reviews concluded that tricyclics antidepressants are superior to gabapentin for neuropathic pain, but were based on indirect comparisons from placebo-controlled trials. PURPOSE: To evaluate gabapentin versus tricyclic antidepressants for diabetic neuropathy and post-herpetic neuralgia, using direct and indirect comparisons. DATA SOURCES: MEDLINE (1966 to March Week 4 2008), the Cochrane central register of controlled trials (1st quarter 2008), and reference lists. STUDY SELECTION: We selected randomized trials directly comparing gabapentin versus tricyclic antidepressants or comparing either of these medications versus placebo. DATA EXTRACTION: Studies were reviewed, abstracted, and quality-rated by two independent investigators using predefined criteria. DATA SYNTHESIS: We performed a meta-analysis of head-to-head trials using a random effects model and compared the results to an adjusted indirect analysis of placebo-controlled trials. RESULTS: In three head-to-head trials, there was no difference between gabapentin and tricyclic antidepressants for achieving pain relief (RR 0.99, 95% CI 0.76 to 1.29). In adjusted indirect analyses, gabapentin was worse than tricyclic antidepressants for achieving pain relief (RR = 0.41, 95% CI 0.23 to 0.74). The discrepancy between direct and indirect analyses was statistically significant (p = 0.008). Placebo-controlled tricyclic trials were conducted earlier than the gabapentin trials, reported lower placebo response rates, had more methodological shortcomings, and were associated with funnel plot asymmetry. CONCLUSIONS: Though direct evidence is limited, we found no difference in likelihood of achieving pain relief between gabapentin and tricyclic antidepressants for diabetic neuropathy and post-herpetic neuralgia. Indirect analyses that combine data from sets of trials conducted in different eras can be unreliable.

Association of low bone mineral density with selective serotonin reuptake inhibitor use by older men.

BACKGROUND: Selective serotonin reuptake inhibitors (SSRIs) are a widely used class of antidepressants that block the serotonin transporter. Osteoblasts and osteocytes express functional serotonin transporters; serotonin transporter gene disruption in mice results in osteopenia; and SSRI use has been associated with increased risk of hip fracture. METHODS: To determine whether SSRI use is associated with lower bone mineral density (BMD) in older men and to compare the results for SSRIs with those of other antidepressants, we performed a cross-sectional analysis of data from 5995 men 65 years and older participating in the prospective cohort Osteoporotic Fractures in Men Study. Main outcome measures included medication use; BMD at the femoral neck, greater trochanter, and lumbar spine measured by dual-energy x-ray absorptiometry; and potential covariates. RESULTS: In adjusted analyses, mean BMD among SSRI users (n=160) was 3.9% lower at the total hip and 5.9% lower at the lumbar spine compared with BMD in men reporting no antidepressant use (n=5708 [P=.002 for total hip; P<.001 for lumbar spine]). There was no significant difference among users of trazodone hydrochloride (n=52) or tricyclic antidepressants (n=99) compared with nonusers. Adjusting for variables that could be associated with BMD and/or SSRI use did not significantly alter these results. The observed difference in BMD for SSRIs is similar to that seen with glucocorticoids. CONCLUSIONS: In this population of men, BMD was lower among those reporting current SSRI use, but not among users of other antidepressants. Further research is needed to confirm this finding in light of widespread SSRI use and potentially important clinical implications.

Cost-effectiveness of echocardiography to identify intracardiac thrombus among patients with first stroke or transient ischemic attack.

BACKGROUND AND PURPOSE: Echocardiography to select stroke patients for targeted treatments, such as anticoagulation (AC), to reduce recurrent stroke risk is controversial. The authors' objective was to evaluate the cost-effectiveness of imaging strategies that use transthoracic (TTE) and transesophageal (TEE) echocardiography for identifying intracardiac thrombus in new stroke patients. METHODS: Model-based cost-effectiveness analysis of 7 echocardiographic imaging strategies and 2 nontesting strategies with model parameters based on systematic evidence review related to effectiveness of echocardiography in newly diagnosed ischemic stroke patients (white males aged 65 years in base case). Primary outcome was cost per quality-adjusted life year (QALY). RESULTS: All strategies containing TTE were dominated by others and were eliminated from the analysis. Assuming that AC reduces recurrent stroke risk from intracardiac thrombus by 43% over 1 year, TEE generated a cost per QALY of $137,000 (relative to standard treatment) among patients with 5% thrombus prevalence. Cost per QALY dropped to $50,000 in patients with at least 15% intracardiac thrombus prevalence, or, if an 86% relative risk reduction with AC is assumed, in patients with thrombus prevalence of at least 6%. Probabilistic analyses indicate considerable uncertainty around the cost-effectiveness of echocardiography across a wide range of intracardiac thrombus prevalence (pretest probability). CONCLUSIONS: Current evidence on cost-effectiveness is insufficient to justify widespread use of echocardiography in stroke patients. Additional research on recurrent stroke risk in patients with intracardiac thrombus and on the efficacy of AC in reducing that risk may contribute to a better understanding of the circumstances under which echocardiography will be cost-effective.

Incident fall risk and physical activity and physical performance among older men: the Osteoporotic Fractures in Men Study.

Physical activity and physical performance have been linked to fall risk in the elderly. The authors examined the relation between physical activity and physical performance with incident falls in the Osteoporotic Fractures in Men Study, a large prospective cohort study of 5,995 community-dwelling men in the United States at least 65 years of age. The authors also examined what types of activities are associated with falling. Incident falls between 2000 and 2005 were captured from up to 17 triannual follow-up questionnaires per participant and analyzed with generalized estimating equations. Follow-up averaged 4.5 years. The average risk of falling in the first 4 months of follow-up was 6.6%. The most active quartile had a significantly greater fall risk than did the least active quartile (relative risk = 1.18, 95% confidence interval (CI): 1.07, 1.29). Men with greater leg power and grip strength had significantly reduced fall risk (for highest leg power quartile vs. lowest: relative risk = 0.82, 95% CI: 0.73, 0.92; for highest grip strength quartile vs. lowest: relative risk = 0.76, 95% CI: 0.69, 0.85). Partitioning components of activity showed no association between fall risk and leisure activities but a positive association with household activities (for highest quartile vs.lowest: relative risk = 1.17, 95% CI: 1.07, 1.28).

Relationship between resident workload and self-perceived learning on inpatient medicine wards: a longitudinal study.

BACKGROUND: Despite recent residency workload and hour limitations, little research on the relationship between workload and learning has been done. We sought to define residents' perceptions of the optimal patient workload for learning, and to determine how certain variables contribute to those perceptions. Our hypothesis was that the relationship between perceived workload and learning has a maximum point (forming a parabolic curve): that either too many or too few patients results in sub-optimal learning. METHODS: Residents on inpatient services at two academic teaching hospitals reported their team and individual patient censuses, and rated their perception of their learning; the patient acuity; case variety; and how challenged they felt. To estimate maximum learning scores, linear regression models with quadratic terms were fit on learning score. RESULTS: Resident self-perceived learning correlated with higher acuity and greater heterogeneity of case variety. The equation of census versus learning score, adjusted for perception of acuity and case mix scores, showed a parabolic curve in some cases but not in others. CONCLUSION: These data suggest that perceived resident workload is complex, and impacted by additional variables including patient acuity and heterogeneity of case variety. Parabolic curves exist for interns with regard to overall census and for senior residents with regard to new admissions on long call days.

Physicians' responses to patient-requested cesarean delivery.

BACKGROUND: The issue about whether a woman's autonomy in childbirth should include the choice of cesarean delivery in the absence of medical indications has become a major source of debate. Our objective was to examine factors that determined physicians' responses to patient-requested cesarean delivery. METHODS: Surveys were distributed to all obstetrician-gynecologists in the greater Portland, Oregon, metropolitan area in Spring, 2000. Physicians were asked to respond to scenarios involving a term patient with a singleton pregnancy requesting primary cesarean delivery. RESULTS: One hundred and seventy of 255 physicians (67%) responded, of whom 68 to 98 percent agreed to cesarean delivery in cases with clear medical indications. Without a clear medical indication, most practitioners would not perform a cesarean delivery. In cases where medical indications were unclear, responses were divided. Physician male gender and patient high socioeconomic status were associated with increased likelihood of physician agreement to patient-requested cesarean delivery. Age, years in practice, and practice type were not associated with agreement. CONCLUSIONS: Physicians are reluctant to agree to patient request for primary cesarean delivery without a clear medical indication. Male physicians were more likely to agree to a patient's request for cesarean delivery than female physicians.

Systematic review of the incidence and consequences of uterine rupture in women with previous caesarean section.

OBJECTIVE: To evaluate the incidence and consequences of uterine rupture in women who have had a delivery by caesarean section. DESIGN: Systematic review. DATA SOURCES: Medline, HealthSTAR, Cochrane Database of Systematic Reviews, Cochrane Controlled Trials Register, National Centre for Reviews and Dissemination, reference lists, and national experts. Studies in all languages were eligible if published in full. REVIEW METHODS: Methodological quality was evaluated for each study by using criteria from the United States Preventive Services Task Force and the National Health Service Centre for Reviews and Dissemination. Uterine rupture was categorised as asymptomatic or symptomatic. RESULTS: We reviewed 568 full text articles to identify 71 potentially eligible studies, 21 of which were rated at least fair in quality. Compared with elective repeat caesarean delivery, trial of labour increased the risk of uterine rupture by 2.7 (95% confidence interval 0.73 to 4.73) per 1000 cases. No maternal deaths were related to rupture. For women attempting vaginal delivery, the additional risk of perinatal death from rupture of a uterine scar was 1.4 (0 to 9.8) per 10,000 and the additional risk of hysterectomy was 3.4 (0 to 12.6) per 10 000. The rates of asymptomatic uterine rupture in trial of labour and elective repeat caesarean did not differ significantly. CONCLUSIONS: Although the literature on uterine rupture is imprecise and inconsistent, existing studies indicate that 370 (213 to 1370) elective caesarean deliveries would need to be performed to prevent one symptomatic uterine rupture.

Medical student evaluation of the quality of hospitalist and nonhospitalist teaching faculty on inpatient medicine rotations.

PURPOSE: To evaluate the impact of academic hospitalists on third-year medical students during inpatient medicine rotations. METHOD: The authors conducted a retrospective quantitative assessment of medical student evaluations of hospitalist and nonhospitalist Department of Medicine faculty at Oregon Health & Science University, for the 1998-00 academic years. Using a nine-point Likert-type scale, students evaluated the faculty on the following characteristics: communication of rotation goals, establishing a favorable learning climate, use of educational time, teaching style, evaluation and feedback, contributions to the student's growth and development, and overall effectiveness as a clinical teacher. RESULTS: A total of 138 students rotated on the university wards during the study period; 100 with hospitalists, and 38 with nonhospitalists. Of these students, 99 (71.7%) returned evaluations. The hospitalists received higher numeric evaluations for all individual attending characteristics. Significance was achieved comparing communication of goals (p =.011), effectiveness as a clinical teacher (p =.016), and for the combined analysis of all parameters (p <.001). Despite lack of achieving statistical significance, there was a trend toward hospitalists being more likely to contribute to the medical student's perception of growth and development during the period evaluated (p =.065). CONCLUSIONS: In addition to performing the responsibilities required of full-time hospital-based physicians, hospitalists were able to provide at least as positive an educational experience as did highly rated nonhospitalist teaching faculty and in some areas performed better. A hospitalist model can be an effective method of delivering inpatient education to medical students.

Potentially avoidable rehospitalizations following acute myocardial infarction by insurance status.

Few studies have explored the impact of health insurance on patients with severe, chronic diseases. This retrospective study examined the association between health insurance and the risk of potentially avoidable rehospitalization in the 3 years following validated acute myocardial infarction (AMI) for a community-based probability sample of 683 patients admitted to 30 California hospitals in 1990-1991. In a multivariate analysis adjusted for measures of comorbidity burden, severity of illness, and AMI-related inpatient care, the risk of readmission was not significantly different among patients with no insurance, Medicare insurance, and non-Medicaid, non-Medicare ("private or other") insurance. However, compared to the latter group, patients with Medicaid were 2.6 times more likely to be readmitted for an AMI-related process (risk ratio. 2.61; 95% confidence interval, 1.33 to 5.11). Additional studies are needed to define the role of health insurance on clinical outcomes and health care access across a broader range of conditions and communities.

The effectiveness of primary care-based interventions to promote breastfeeding: systematic evidence review and meta-analysis for the US Preventive Services Task Force.

PURPOSE: We wanted to systematically review whether primary care-based interventions improve initiation and duration of breastfeeding. METHODS: Studies were found by searching MEDLINE (1966-2001), Health-STAR, the Cochrane Database of Systematic Reviews, the National Health Service Centre for Reviews and Dissemination Databases, and bibliographies of identified trials and review articles. Studies were included if they originated in the primary care setting and were conducted in a developed country, written in English, and contained a concurrent control group. RESULTS: Thirty randomized and nonrandomized controlled trials and 5 systematic reviews of breastfeeding counseling were included. Educational programs had the greatest effect of any single intervention on both initiation (difference 0.23; 95% confidence interval [CI], 0.12-0.34) and short-term duration (difference 0.39; 95% CI, 0.27-0.50). Support programs conducted by telephone, in person, or both increased short-term (difference 0.11; 95% CI, 0.03-0.19) and long-term duration (difference 0.08; 95% CI, 0.02-0.16). In contrast, written materials such as pamphlets did not significantly increase breastfeeding. Data were insufficient to determine whether the combination of education with support was more effective than education alone. CONCLUSIONS: Educational programs were the most effective single intervention. One woman would breast-feed for up to 3 months for every 3 to 5 women attending breastfeeding educational programs. Future research and policy should focus on translating these findings into more widespread practice in diverse primary care settings.

Management of the low-grade abnormal Pap smear: What are women's preferences?

OBJECTIVE: To evaluate preferences among ethnically diverse women for the management of a low-grade abnormal Pap smear result: early colposcopy or observation with repeat Pap smears. STUDY DESIGN: Structured interviews were conducted with 170 women of diverse ethnic backgrounds to assess their preferences. Trained personnel conducted standardized interviews. A standard description of all tests and procedures was read to participants. The participants were presented with scenarios of contrasting management approaches for a hypothetical low-grade abnormal Pap smear result-observation with repeat Pap smear vs. immediate colposcopy. POPULATION: Study participants were recruited from the waiting rooms of 5 family planning clinics in Northern Californiaamprsquos Central Valley. OUTCOMES MEASURED: The primary outcome measures for each scenario were utilities (quantified preferences for specific health states) measured by the Standard Gamble. RESULTS: The range in utilities was large for all scenarios. Mean utilities (SD) for observation: 0.96 ( 0.13) followed by resolution; 0.93 ( 0.17) followed by cryotherapy; 0.91 ( 0.21) followed by cone biopsy. Mean utilities for early colposcopy: 0.93 ( 0.20) followed by resolution; 0.95 ( 0.14) followed by cryotherapy; and 0.92 ( 0.16) followed by cone biopsy. Subject characteristics explained less than 20% of the variance in utilities. Decision analysis gave a slightly higher overall utility for early colposcopy (0.940 vs 0.932 for observation), but was sensitive to small changes in branch utilities. CONCLUSIONS: Womenamprsquos preferences for management of a low-grade abnormal Pap result vary widely. Clinicians should adopt a flexible approach to the management of low-grade abnormal Pap smears to incorporate individual preferences.

Breast cancer screening: a summary of the evidence for the U.S. Preventive Services Task Force.

PURPOSE: To synthesize new data on breast cancer screening for the U.S. Preventive Services Task Force. DATA SOURCES: MEDLINE; the Cochrane Controlled Trials Registry; and reference lists of reviews, editorials, and original studies. STUDY SELECTION: Eight randomized, controlled trials of mammography and 2 trials evaluating breast self-examination were included. One hundred fifty-four publications of the results of these trials, as well as selected articles about the test characteristics and harms associated with screening, were examined. DATA EXTRACTION: Predefined criteria were used to assess the quality of each study. Meta-analyses using a Bayesian random-effects model were conducted to provide summary relative risk estimates and credible intervals (CrIs) for the effectiveness of screening with mammography in reducing death from breast cancer. DATA SYNTHESIS: For studies of fair quality or better, the summary relative risk was 0.84 (95% CrI, 0.77 to 0.91) and the number needed to screen to prevent one death from breast cancer after approximately 14 years of observation was 1224 (CrI, 665 to 2564). Among women younger than 50 years of age, the summary relative risk associated with mammography was 0.85 (CrI, 0.73 to 0.99) and the number needed to screen to prevent one death from breast cancer after 14 years of observation was 1792 (CrI, 764 to 10 540). For clinical breast examination and breast self-examination, evidence from randomized trials is inconclusive. CONCLUSIONS: In the randomized, controlled trials, mammography reduced breast cancer mortality rates among women 40 to 74 years of age. Greater absolute risk reduction was seen among older women. Because these results incorporate several rounds of screening, the actual number of mammograms needed to prevent one death from breast cancer is higher. In addition, each screening has associated risks and costs.

Postmenopausal hormone replacement therapy and the primary prevention of cardiovascular disease.

PURPOSE: To evaluate the value of hormone replacement therapy (HRT) in the primary prevention of cardiovascular disease (CVD) and coronary artery disease (CAD). DATA SOURCES: MEDLINE and Cochrane databases were searched for all primary prevention studies reporting CVD or CAD incidence, mortality, or both in association with HRT; reference lists, letters, editorials, and reviews were also reviewed. DATA EXTRACTION: All studies were reviewed, abstracted, and rated for quality. STUDY SELECTION: Only studies of good or fair quality, according to U.S. Preventive Services Task Force (USPSTF) criteria, were included in the detailed review and meta-analysis. DATA SYNTHESIS: The summary relative risk with any HRT use was 0.75 (95% credible interval [CrI], 0.42 to 1.23) for CVD mortality and 0.74 (CrI, 0.36 to 1.45) for CAD mortality. The summary relative risk with any use was 1.28 (CrI, 0.86 to 2.00) for CVD incidence and 0.87 (CrI, 0.62 to 1.21) for CAD incidence. Further analysis of studies adjusting for socioeconomic status, as well as other major CAD risk factors, showed a summary relative risk of 1.07 (CrI, 0.79 to 1.48) for CAD incidence associated with any HRT use. Similar results were found when the analysis was stratified by studies adjusting for alcohol consumption, exercise, or both, in addition to other major risk factors, suggesting confounding by these factors. CONCLUSIONS: This meta-analysis differs from previous meta-analyses by evaluating potential explanatory variables of the relationship between HRT, CVD, and CAD. The adjusted meta-analysis is consistent with recent randomized trials that have shown no benefit in the secondary or primary prevention of CVD events. A valid answer to the role of HRT in the primary prevention of CVD will best come from randomized, controlled trials.

Postmenopausal estrogen replacement and risk for venous thromboembolism: a systematic review and meta-analysis for the U.S. Preventive Services Task Force.

BACKGROUND: Postmenopausal estrogen replacement is widely used in the United States but poses important health risks. PURPOSE: To assess the risk for venous thromboembolism with postmenopausal estrogen replacement by using literature review and meta-analysis. DATA SOURCES: All relevant English-language studies identified in searches of the MEDLINE (1966 to December 2000), HealthSTAR (1975 to December 2000), and Cochrane Library databases, and references lists of key articles. STUDY SELECTION: All published studies of postmenopausal estrogen replacement reporting venous thromboembolism as an outcome or adverse event. DATA EXTRACTION: 12 studies of estrogen were identified (3 randomized, controlled trials; 8 case-control studies; and 1 cohort study). Data were extracted on participants, interventions, event rates, and confounders. Two reviewers independently rated study quality on the basis of established criteria. DATA SYNTHESIS: A Bayesian meta-analysis was conducted. When data from all studies were pooled, current estrogen use was associated with an increased risk for venous thromboembolism (relative risk, 2.14 [95% credible interval, 1.64 to 2.81]). Estimates did not significantly change when studies were pooled according to study design, quality score, or whether participants had preexisting coronary artery disease. The absolute rate increase was 1.5 venous thromboembolic events per 10 000 women in 1 year. Six case-control studies that reported risk according to duration of use found that risk was highest in the first year of use (relative risk, 3.49 [credible interval, 2.33 to 5.59]). CONCLUSION: Postmenopausal estrogen replacement is associated with an increased risk for venous thromboembolism, and this risk may be highest in the first year of use.