In This Edition
- Pay-for-performance associated with limited benefits on quality in acute myocardial infarction patients.
- Meta-analysis finds increased risk of acute myocardial infarction with use of rosiglitazone.
- Rosiglitazone increases risk of heart failure, but not acute myocardial infarction or death, interim analysis shows.
- Surgery versus prolonged conservative treatment for severe sciatica.
- Predicting poor outcomes in upper gastrointestinal bleeds.
- Discharging patients with unresolved medical issues.
Background: In 2003, the Centers for Medicare and Medicaid Services (CMS) instituted a pay-for-performance (P4P) pilot program in which participating hospitals would be reimbursed more if they met specific quality standards of care for patients with certain conditions, including acute myocardial infarction (AMI). It is unknown if this type of financial incentive produces improvements in the processes or outcomes of care.
Study design: Observational cohort.
Setting: 500 hospitals across the U.S.
Synopsis: This study compared compliance with CMS quality indicators in the treatment of more than 100,000 patients with acute non-ST-elevation myocardial infarction at 54 participating and 446 non-participating hospitals in the P4P pilot. They found no significant difference in mortality or in a composite measure of the six quality indicators but a slight improvement in two of the six quality indicators (aspirin at discharge and smoking cessation counseling). They did not find that P4P adversely affected indicators not subject to financial incentives.
Bottom line: P4P is associated with limited improvements in compliance with CMS quality indicators in patients with AMI.
Citation: Glickman SW, Ou F-S, DeLong ER, et al. Pay for performance, quality of care, and outcomes in acute myocardial infarction. JAMA. 2007 Jun;297(21):2373-2380.
Background: Rosiglitazone (Avandia) is one of two approved oral thiazolidinedione drugs used for diabetic control. Muraglitazar, another thiazolidinedione drug, was not approved for market due to adverse cardiovascular outcomes. The cardiovascular effects of rosiglitazone had not previously been evaluated.
Study design: Meta-analysis.
Setting: All clinical trials (published and unpublished) involving rosiglitazone.
Synopsis: The authors reviewed data from all randomized trials of rosiglitazone versus placebo or other drugs for at least 24 weeks. From the 42 included trials (including more than 28,000 patients) researchers found a statistically significant increased risk of the odds of AMI (odds ratio 1.43, confidence interval 1.03-1.98) in the rosiglitazone group, and a non-significant risk of death from any cardiovascular cause (odds ratio 1.64, confidence interval 0.98-2.74) and all-cause mortality (odds ratio 1.18, confidence interval 0.89-1.55). The meta-analysis was criticized due to the small number of events (fewer than 100 acute AMIs in each group) and lack of patient-level data, but one expert wrote that “in view of the potential cardiovascular risks and in the absence of evidence of other health advantages ... the rationale for prescribing rosiglitazone at this time is unclear.”
The study raised larger concerns regarding Food and Drug Administration drug approvals, because the drug was approved due to its effect on lowering blood sugar levels (a surrogate outcome) without enough scrutiny of other patient outcomes.
Bottom line: Rosiglitazone is associated with increased risk of AMI. Alternative oral agents should be considered first for blood sugar control in diabetics.
Citation: Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med. 2007 Jun;356(24):2457-2471.
Editorial: Psaty BM, Furberg CD. Rosiglitazone and cardiovascular risk. N Engl J Med. 2007 Jun;356(24):2522-2524.
Background: In response to a meta-analysis, an interim analysis of an ongoing open-label manufacturer-sponsored trial was undertaken to determine the cardiovascular risks of rosiglitazone.
Study design: Unplanned interim analysis of a randomized, multicenter, open-label, non-inferiority trial.
Subjects: Outpatient, inadequately controlled type 2 diabetics.
Synopsis: This was an unplanned interim analysis of an open-label manufacturer-sponsored trial. There were 4,447 inadequately controlled type 2 diabetics on either metformin or sulfonylurea. The patients were randomized to receive both drugs (controls) or add-on rosiglitazone. After a mean follow up of 3.75 years, there was no statistically significant difference between the groups in the primary end point (hospitalization or death from cardiovascular causes), or other end points (MI and death from cardiovascular causes or any cause). However, rosiglitazone was associated with an increased risk of heart failure (hazard ratio 2.15, confidence interval 1.30-3.57). Because this was an unplanned interim analysis for a trial expected to continue for six years, experts caution that the results are inconclusive due to low statistical significance and small event rates.
Bottom line: Rosiglitazone is associated with an increased risk of heart failure, but the risks of hospitalization, death, and acute MI remain unclear.
Citation: Home PD, Phil D, Pockock SJ, et al. Rosiglitazone evaluated for cardiovascular outcomes—an interim analysis. N Engl J Med. 2007 Jul;357(1):28-38.
Background: Patients are often discharged from the hospital with incomplete workups, but it is unknown how often and what factors affect the completion of the intended workup.
Study design: Retrospective cohort.
Setting: Single institution teaching hospital.
Synopsis: The authors evaluated the inpatient and outpatient medical records of all patients discharged from the medicine or geriatric service over 18 months. Of almost 700 discharges, 28% of the patients had outpatient workups recommended (48% diagnostic procedures, 35% referrals, and 17% lab tests) by the discharging physician. Completion of the workup did not occur 36% of the time, and the likelihood of non-completion increased with time to the first follow-up appointment and lack of availability of the discharge summary.
Bottom line: Outpatient workups are intended in almost a third of discharged patient, the completion of which can likely be enhanced by timely follow-up and discharge summary availability.
Citation: Moore C, McGinn T, Halm E. Tying up loose ends: discharging patients with unresolved medical issues. Arch Intern Med. 2007;167(12):1305-1311.
Background: Although multiple risk-prediction scales exist for patients with upper gastrointestinal (UGI) bleeds, few have been prospectively validated or widely used in clinical practice.
Study design: Prospective cohort.
Setting: Veterans Affairs (VA) hospitals.
Synopsis: VA researchers created and validated a risk predictor in 391 patients with acute upper gastrointestinal bleeding. Data from the derivation set (two-thirds of the patients) was used to create the model tested on the validation set (one-third of the patients). Outcome one (re-bleeding, need for intervention to stop bleeding, or all-cause hospital mortality) was predicted by an APACHE score >11, stigmata of recent bleeding, or varices. Outcome two (outcome one plus new/worsening co-morbidity) was predicted by the above three factors plus an unstable co-morbidity at admission. In the validation group, outcome one occurred in 1%, 5%, and 25% of patients with zero, one, and two or more factors. Outcome two occurred in 6%, 18%, and 49%, respectively. A score of zero accurately identified 93% and 91% of patients for outcomes one and two. The authors speculated that these patients could be safely treated as outpatients. The study excluded patients on anticoagulation, and this VA cohort (99% male) may not be generalizable to other populations.
Bottom line: This validated prediction model can accurately predict more than 90% of patients at low-risk of poor outcomes with UGI bleeding, which could be used to stratify patients in need of hospital admission.
Citation: Imperiale TF, Dominitz JA, Provenzale DT, et al. Predicting poor outcome from acute upper gastrointestinal hemorrhage. Arch Intern Med. 2007 Jun;167(12):1291-1296.
Background: The optimal timing and benefit of lumbar-disk surgery in patients with symptomatic lumbar disk herniation is unknown.
Study design: Multicenter randomized trial.
Synopsis: 283 patients with severe sciatica were randomly chosen to receive early surgery or conservative treatment (with surgery as needed) for six to 12 weeks. The methods for determining the three primary outcomes were: score on the Roland Disability Questionnaire, leg pain score, and self-report of perceived recovery. At one year, 89% of the surgery group and 39% of the control group underwent surgery after a mean of 2.2 and 18.7 weeks, respectively. There was no difference between the groups in the disability score, but time to relief of leg pain and recovery was faster in the surgery group. At one year, 95% in each group reported perceived recovery.
Bottom line: Rates of pain relief and perceived recovery are faster with early surgery than conservative treatment in patients with severe sciatica, but one-year recovery rates are the same. TH
Citation: Peul WC, Van Houwelingen HC, van den Hout WB, et al. Surgery versus prolonged conservative treatment for sciatica. NEJM. 2007 May;356(22):2245-2256.