The impact of using different costing methods on the results of an economic evaluation of cardiac care: microcosting vs gross-costing approaches.

BACKGROUND: Published guidelines on the conduct of economic evaluations provide little guidance regarding the use and potential bias of the different costing methods. OBJECTIVES: Using microcosting and two gross-costing methods, we (1) compared the cost estimates within and across subjects, and (2) determined the impact on the results of an economic evaluation. METHODS: Microcosting estimates were obtained from the local health region and gross-costing estimates were obtained from two government bodies (one provincial and one national). Total inpatient costs were described for each method. Using an economic evaluation of sirolimus-eluting stents, we compared the incremental cost-utility ratios that resulted from applying each method. RESULTS: Microcosting, Case-Mix-Grouper (CMG) gross-costing, and Refined-Diagnosis-Related grouper (rDRG) gross-costing resulted in 4-year mean cost estimates of $16,684, $16,232, and $10,474, respectively. Using Monte Carlo simulation, the cost per QALY gained was $41,764 (95% CI: $41,182-$42 346), $42,538 (95% CI: $42 167-$42 907), and $36,566 (95% CI: $36,172-$36,960) for microcosting, rDRG-derived and CMG-derived estimates, respectively (P<0.001). CONCLUSIONS: Within subject, the three costing methods produced markedly different cost estimates. The difference in cost-utility values produced by each method is modest but of a magnitude that could influence a decision to fund a new intervention.

An administrative data merging solution for dealing with missing data in a clinical registry: adaptation from ICD-9 to ICD-10.

BACKGROUND: We have previously described a method for dealing with missing data in a prospective cardiac registry initiative. The method involves merging registry data to corresponding ICD-9-CM administrative data to fill in missing data 'holes'. Here, we describe the process of translating our data merging solution to ICD-10, and then validating its performance. METHODS: A multi-step translation process was undertaken to produce an ICD-10 algorithm, and merging was then implemented to produce complete datasets for 1995-2001 based on the ICD-9-CM coding algorithm, and for 2002-2005 based on the ICD-10 algorithm. We used cardiac registry data for patients undergoing cardiac catheterization in fiscal years 1995-2005. The corresponding administrative data records were coded in ICD-9-CM for 1995-2001 and in ICD-10 for 2002-2005. The resulting datasets were then evaluated for their ability to predict death at one year. RESULTS: The prevalence of the individual clinical risk factors increased gradually across years. There was, however, no evidence of either an abrupt drop or rise in prevalence of any of the risk factors. The performance of the new data merging model was comparable to that of our previously reported methodology: c-statistic = 0.788 (95% CI 0.775, 0.802) for the ICD-10 model versus c-statistic = 0.784 (95% CI 0.780, 0.790) for the ICD-9-CM model. The two models also exhibited similar goodness-of-fit. CONCLUSION: The ICD-10 implementation of our data merging method performs as well as the previously-validated ICD-9-CM method. Such methodological research is an essential prerequisite for research with administrative data now that most health systems are transitioning to ICD-10.

Public versus private institutional performance reporting: what is mandatory for quality improvement?

BACKGROUND: In the past 11 years, Ontario has generated institution-level performance report cards on outcomes of coronary artery bypass graft (CABG) surgery. The objective of this study was to evaluate the differences in patient characteristics and outcomes observed during the transition from no reporting to confidential, and ultimately public performance report cards for CABG surgery in a public health system. METHODS: We used clinical and administrative data to assess crude, expected, and risk-adjusted 30-day mortality rates after isolated CABG surgery in Ontario for 67693 patients from September 1, 1991, to March 31, 2002. Confidence intervals on relative mortality reductions were determined by bootstrapping. We compared 30-day mortality trends to a control outcome (risk-adjusted 30-day all-cause readmission). We analyzed inhospital mortality trends for Ontario compared with the rest of Canada for the period from 1992 to 1998. RESULTS: The risk-adjusted 30-day mortality rate decreased 29% (95% CI 21-39) from the era of no reporting (1991-1993) to confidential reporting (1994-1998). There was no further decrease with public reporting (1999-2001). The control outcome of 30-day readmission did not decrease across reporting eras. Inhospital mortality fell significantly faster in Ontario during the period of confidential reporting than in other parts of Canada. CONCLUSION: Ontario CABG mortality outcomes improved sharply after provider results were confidentially disclosed at an institutional level. No such changes were seen for nondisclosed outcomes or regions outside Ontario. Further public reporting of outcomes had no discernible impact on performance. These results are consistent with the hypothesis that confidential disclosure of outcomes was sufficient to accelerate quality improvement in a public system with little competition for patients between hospitals.

Revascularization in patients with heart failure.

BACKGROUND: Although practice guidelines recommend coronary revascularization for patients with heart failure, the evidence to support this recommendation is weak. The objective of our study was to determine the association of coronary revascularization with survival in patients who have had heart failure. METHODS: Data were obtained from the Alberta Provincial Project for Outcomes Assessment in Coronary Heart Disease (APPROACH), a clinical outcome-monitoring initiative that has captured data on all patients undergoing cardiac catheterization in the province of Alberta since 1995. Our study included data from patients with a history of heart failure and with documented coronary artery disease; patients with normal coronary arteries or prior coronary artery bypass grafting (CABG) were excluded. We constructed survival curves and adjusted them by the corrected group prognosis method (incorporating all clinical variables in APPROACH). Propensity scores were used to account for clinical characteristics that could influence the decision to revascularize. RESULTS: A total of 2538 patients (mean age 68 yr, standard deviation [SD] 11 yr, 31% female) underwent revascularization; 1690 patients (mean age 69 [SD 11] yr, 34% female) did not. Crude 1-year mortality was 11.8% among patients who underwent revascularization, compared with 21.6% among those who did not. Adjusted survival curves diverged early and continued up to 7 years of follow-up (hazard ratio 0.50, 95% confidence interval 0.44-0.57). Propensity scores showed improved survival with revascularization across all quintiles of likelihood of revascularization. INTERPRETATION: This new evidence lends support to practice guidelines, which recommend revascularization in patients with heart failure and coronary disease.

Dealing with missing data in a multi-question depression scale: a comparison of imputation methods.

BACKGROUND: Missing data present a challenge to many research projects. The problem is often pronounced in studies utilizing self-report scales, and literature addressing different strategies for dealing with missing data in such circumstances is scarce. The objective of this study was to compare six different imputation techniques for dealing with missing data in the Zung Self-reported Depression scale (SDS). METHODS: 1580 participants from a surgical outcomes study completed the SDS. The SDS is a 20 question scale that respondents complete by circling a value of 1 to 4 for each question. The sum of the responses is calculated and respondents are classified as exhibiting depressive symptoms when their total score is over 40. Missing values were simulated by randomly selecting questions whose values were then deleted (a missing completely at random simulation). Additionally, a missing at random and missing not at random simulation were completed. Six imputation methods were then considered; 1) multiple imputation, 2) single regression, 3) individual mean, 4) overall mean, 5) participant's preceding response, and 6) random selection of a value from 1 to 4. For each method, the imputed mean SDS score and standard deviation were compared to the population statistics. The Spearman correlation coefficient, percent misclassified and the Kappa statistic were also calculated. RESULTS: When 10% of values are missing, all the imputation methods except random selection produce Kappa statistics greater than 0.80 indicating 'near perfect' agreement. MI produces the most valid imputed values with a high Kappa statistic (0.89), although both single regression and individual mean imputation also produced favorable results. As the percent of missing information increased to 30%, or when unbalanced missing data were introduced, MI maintained a high Kappa statistic. The individual mean and single regression method produced Kappas in the 'substantial agreement' range (0.76 and 0.74 respectively). CONCLUSION: Multiple imputation is the most accurate method for dealing with missing data in most of the missind data scenarios we assessed for the SDS. Imputing the individual's mean is also an appropriate and simple method for dealing with missing data that may be more interpretable to the majority of medical readers. Researchers should consider conducting methodological assessments such as this one when confronted with missing data. The optimal method should balance validity, ease of interpretability for readers, and analysis expertise of the research team.

Association between completeness of percutaneous coronary revascularization and postprocedure outcomes.

BACKGROUND: Multivessel coronary artery revascularization may be accomplished by percutaneous coronary intervention (PCI) or coronary artery bypass graft surgery (CABG). The importance of complete revascularization is emphasized in the surgical literature, but little is known about its impact on PCI outcomes. This study evaluated multivessel PCI patients to determine the predictors of complete revascularization and the association of complete revascularization with survival, subsequent CABG, and repeat PCI. METHODS: The Alberta Provincial Project for Outcomes Assessment in Coronary Heart Disease (APPROACH) is a clinical data collection and outcome-monitoring initiative capturing all patients undergoing cardiac catheterization and revascularization in the province of Alberta, Canada. Characteristics and long-term outcomes of 1308 patients undergoing multivessel PCI with complete revascularization were compared with those of 648 patients with incomplete revascularization. RESULTS: The significant independent predictors of complete revascularization were pre-PCI Duke jeopardy score, the presence of a total occlusion, year of PCI, age > 65 years, renal failure, and left ventricular function. With a median follow-up time of 3.0 +/- 1.8 years, the adjusted hazard ratio (HR) (95% CI) for the association between complete revascularization and outcome was 0.75 (0.54-1.04) for death, 0.55 (0.37-0.84) for subsequent CABG, and 0.93 (0.65-1.34) for repeat PCI. CONCLUSIONS: Baseline angiographic characteristics and other clinical factors can predict complete revascularization in patients undergoing multivessel PCI. Complete multivessel PCI is associated with reduced need for future CABG, a trend toward better survival, and no difference in repeat PCI.

Economic evaluation of sirolimus-eluting stents.

BACKGROUND: Sirolimus-eluting stents have recently been shown to reduce the risk of restenosis among patients who undergo percutaneous coronary intervention (PCI). Given that sirolimus-eluting stents cost about 4 times as much as conventional stents, and considering the volume of PCI procedures, the decision to use sirolimus-eluting stents has large economic implications. METHODS: We performed an economic evaluation comparing treatment with sirolimus-eluting and conventional stents in patients undergoing PCI and in subgroups based on age and diabetes mellitus status. The probabilities of transition between clinical states and estimates of resource use and health-related quality of life were derived from the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease (APPROACH) database. Information on effectiveness was based on a meta-analysis of randomized controlled clinical trials (RCTs) comparing sirolimus-eluting and conventional stents. RESULTS: Cost per quality-adjusted life year (QALY) gained in the baseline analysis was Can58,721 dollars. Sirolimus-eluting stents were more cost-effective in patients with diabetes and in those over 75 years of age, the costs per QALY gained being 44,135 dollars and 40,129 dollars, respectively. The results were sensitive to plausible variations in the cost of stents, the estimate of the effectiveness of sirolimus-eluting stents and the assumption that sirolimus-eluting stents would prevent the need for cardiac catheterizations in the subsequent year when no revascularization procedure was performed to treat restenosis. INTERPRETATION: The use of sirolimus-eluting stents is associated with a cost per QALY that is similar to or higher than that of other accepted medical forms of therapy and is associated with a significant incremental cost. Sirolimus-eluting stents are more economically attractive for patients who are at higher risk of restenosis or at a high risk of death if a second revascularization procedure were to be required.

Moving beyond the cost per quality-adjusted life year: modelling the budgetary impact and clinical outcomes associated with the use of sirolimus-eluting stents.

Restenosis is a major limitation to the long-term success of percutaneous coronary intervention. Drug-eluting stents are the most recent technological advance in restenosis prevention. While they are effective, their use is associated with a significant incremental cost, and a recent economic evaluation performed by the authors suggested that their use is associated with a cost per quality-adjusted life year of $58,721. How should decision-makers react to this value, particularly given that the use of sirolimus-eluting stents appears more attractive in certain patient subgroups, such as those with complex coronary lesions? In the present paper, the authors explore an alternative method of presenting the results of their economic evaluation, rather than the usual cost per quality-adjusted life year rubric, in an attempt to assist decision-makers in deciding whether, and for whom, to fund sirolimus-eluting stents. Several issues that decision-makers and providers may wish to consider when making such funding decisions are discussed.

Outcomes after coronary artery bypass graft surgery in Canada: 1992/93 to 2000/01.

BACKGROUND: The authors have previously reported on Canada-wide outcomes of coronary artery bypass graft (CABG) surgery for 1992/93 through 1995/96. OBJECTIVE: To provide an updated Canada-wide CABG surgery outcome report with outcome data organized by province and by year for 1992/93 through 2000/01. METHODS: Hospital discharge abstract data were obtained from the Canadian Institute for Health Information and were used to identify all patients who underwent isolated CABG surgery in eight provinces from fiscal year 1992/93 through 2000/01. Crude data from Quebec hospitals were available for calendar years 1998 and 1999. Logistic regression modelling was used to calculate risk-adjusted in-hospital mortality rates by year and province. RESULTS: Patients undergoing CABG surgery in the later years studied were on average older and had more comorbidities than did patients undergoing this surgery in earlier years. Despite increasing case complexity, risk-adjusted mortality rates decreased significantly from 3.5% (95% CI 3.2% to 3.8%) to 2.0% (95% CI 1.8% to 2.3%). Risk-adjusted mortality rates varied between provinces. Provincial risk-adjusted mortality rates ranged from 2.0% to 3.3%. However, all provinces studied had either persistently low mortality rates (Nova Scotia) or declining mortality rates across years studied, such that all provinces achieved risk-adjusted mortality rates of 2.7% or lower in 2000/01. CONCLUSIONS: This evaluation of Canadian CABG surgery outcomes demonstrates a pattern of either steadily improving or persistently favourable provincial in-hospital mortality rates after isolated CABG surgery. These favourable provincial outcome trends have been achieved despite an accompanying increase in the average case complexity of patients undergoing CABG in Canada.

In-hospital outcomes after percutaneous coronary intervention in Canada: 1992/93 to 2000/01.

BACKGROUND: Despite existing research on outcomes of cardiac care in Canada, little is known about Canada-wide trends and interprovincial differences in outcomes after percutaneous coronary intervention (PCI). OBJECTIVES: To examine Canadian trends in rates of in-hospital mortality and same-admission coronary artery bypass grafting (CABG) after PCI and to compare provincial risk-adjusted in-hospital death and same-admission CABG rates. METHODS: Hospital discharge abstract data were obtained from the Canadian Institute for Health Information and were used to identify cohorts of patients who underwent PCI in eight provinces in fiscal years 1992/93 through 2000/01. Crude data from Quebec hospitals were available for calendar years 1998 and 1999. Logistic regression modelling was used to calculate risk-adjusted in-hospital death and same-admission CABG rates by year and province. RESULTS: A total of 127,103 PCI cases performed in 23 hospitals across eight provinces were examined, with an overall unadjusted death rate of 1.4% and an overall unadjusted CABG rate of 1.6%. A national trend of stable in-hospital mortality rates was observed with a risk-adjusted death rate of 1.4% in 1992/93 versus 1.4% in 2000/01. An overall decline was seen in rates of same-admission CABG with a risk-adjusted rate of 2.7% in 1992/93 versus 0.9% in 2000/01 (relative decrease 67%, P<0.01). New Brunswick, Manitoba and British Columbia achieved overall declines in risk-adjusted death rates over the study period, while the other provinces experienced a slight increase (Newfoundland, Nova Scotia, Ontario, Alberta and Saskatchewan). All provinces displayed a similar decline in risk-adjusted same-admission CABG rates post-PCI. INTERPRETATION: Risk-adjusted rates of in-hospital death after PCI in Canada have remained stable over nine years, while risk-adjusted rates of same-admission CABG have decreased. The presence of interprovincial differences in risk-adjusted outcomes raises the possibility of variable quality of care for patients undergoing PCI across the Canadian provinces.

Prognostic implications of C-reactive protein and troponin following percutaneous coronary intervention.

BACKGROUND: C-reactive protein (CRP), a marker of inflammation, plays a role in the pathophysiology of atherosclerotic events. The relationship between CRP levels and myocardial necrosis assessed by troponin T (TnT) in patients undergoing percutaneous coronary intervention (PCI) has not been established. In addition, the long-term significance of TnT rise following PCI is not clear. OBJECTIVES: To examine the relationship between CRP and the rise in TnT levels, and evaluate the long-term prognostic implications of TnT rise following PCI. METHODS: A total of 1208 patients underwent successful nonemergent PCI. Baseline demographic characteristics, CRP and TnT levels were prospectively collected before and 12 h to 18 h following PCI. Long-term follow-up data over two years were available. RESULTS: Among the patients studied (mean age 62 years), 64% presented with acute coronary syndrome. A PCI procedure was associated with a significant increase in TnT levels (higher than 0.1 microg/L) in 238 patients (20%). Multivariate logistic regression identified presentation with acute coronary syndrome or myocardial infarction, no statin use at the time of the procedure, increased CRP and increasing length of stent as independent predictors of TnT rise following PCI. Periprocedural TnT rise was not associated with adverse events in follow-up examinations (OR 1.09, 95% CI 0.73 to 1.65). CONCLUSIONS: Myocardial necrosis commonly occurred in otherwise successful PCI and was particularly prevalent in the proinflammatory milieu of a recent myocardial infarction. This response was blunted with statin therapy. However, there was no long-term adverse sequelae of these troponin rises following otherwise uncomplicated PCI.

Prognostic implication of hyperglycemia in myocardial infarction and primary angioplasty.

PURPOSE: The study assessed the relationship of admission blood glucose level to in-hospital mortality in patients presenting with an ST-segment elevation myocardial infarction and treated with primary angioplasty. METHODS: A total of 980 patients presenting with an ST-segment elevation myocardial infarction and treated exclusively with primary angioplasty were evaluated. Patients were divided into quartiles based on their admission blood glucose level: group 1 (< or =6.6 mmol/L [< or =119 mg/dL]), group 2 (6.7-7.8 mmol/L [120-140 mg/dL]), group 3 (7.9-10.0 mmol/L [141-180 mg/dL], and group 4 (> or =10.1 mmol/L [> or =181 mg/dL]. The primary end point was in-hospital mortality. RESULTS: The mean age of the patient cohort was 62 years, 260 (27%) of whom were female. The mean admission blood glucose level was 9.1+/-4.4 mmol/L (164+/-79 mg/dL). At admission, 16% of this group were known to have diabetes. The in-hospital mortality rate was 3.8% (n=37), 5.2% in the diabetic group (n=8) and 3.5% (n=29) in the nondiabetic group. In-hospital mortality rates were significantly increased in patients with an elevated admission blood glucose level (P<.001). The in-hospital deaths in each admission blood glucose level quartile were 0.4% (n=1) in group 1, 2% (n=6) in group 2, 2% (n=6) in group 3, and 10% (n=24) in group 4. CONCLUSIONS: In this cohort of patients who were admitted with an ST-segment elevation myocardial infarction and treated exclusively with primary angioplasty, elevated admission blood glucose level is significantly associated with an increase in in-hospital mortality.

Use of the U.S. and U.K. scoring algorithm for the EuroQol-5D in an economic evaluation of cardiac care.

BACKGROUND: Most studies that have used the EuroQol-5D instrument (EQ-5D) have used a scoring algorithm based on preferences solicited from the U.K. population. An algorithm recently was developed for the U.S. population, with studies showing meaningful differences in the results obtained using the 2 algorithms. We recently published an economic evaluation assessing the use of drug-eluting stents in patients undergoing percutaneous coronary intervention (PCI). OBJECTIVES: Using the aforementioned economic evaluation, we describe the EQ-5D utility scores resulting from use of U.S. and U.K. algorithms and explore the differences in the incremental cost-utility ratio (ICER) resulting from use of the different EQ-5D estimates. METHODS: EQ-5D data were obtained from the Alberta Provincial Project for Outcomes Assessment in Coronary Heart (APPROACH) disease registry. Individual responses were scored once with each algorithm. The within-individual difference was calculated (U.S. score-U.K. score). The mean, SD, and range were compared using paired t tests. The resulting ICERs were compared using probabilistic sensitivity analysis. RESULTS: The U.K. mean was statistically different from the U.S. mean (0.83, SD 0.20 vs. 0.87, SD 0.15, P<0.001). The mean within individual difference was 0.04 with a wide range (-0.02 to +0.41). The resulting ICER are CAN $58,635 (95% confidence interval $198,248-$34,406) per quality-adjusted life year and CAN $58,229 (95% confidence interval $116,818-$38,779) per quality-adjusted life year for the U.K. and U.S. algorithms, respectively (P value: 0.07). CONCLUSIONS: The algorithms produce quite notable differences within individuals. The effect on the mean score is less pronounced. In the context of our economic evaluation, however, the impact of using the U.S. algorithm on the ICER is negligible.

Defining the optimal treatment of locally advanced esophageal cancer: a systematic review and decision analysis.

BACKGROUND: The objective of this study was to combine systematic review and decision analytic techniques to determine the optimal treatment strategy for patients with locally advanced esophageal cancer. METHODS: We performed a systematic review of all randomized trials of patients with locally advanced esophageal cancer that included one of the following strategies compared with surgery alone: chemoradiotherapy followed by surgery, chemotherapy followed by surgery, or surgery with adjuvant chemoradiotherapy. Using the estimates of relative risk for mortality and overall quality of life we constructed a decision model. The outcome of interest was expected quality-adjusted life-years (QALY). RESULTS: The meta-analysis showed for the first year, the relative risk (95% confidence interval) of death for treatments compared with surgery were 0.87 (0.75 to 1.02) for chemoradiotherapy followed by surgery, 0.94 (0.82 to 1.08) for chemotherapy followed by surgery, and 1.33 (0.93 to 1.93) for surgery with adjuvant chemoradiotherapy. The QALYs gained for surgery alone, chemoradiotherapy followed by surgery, chemotherapy followed by surgery, and surgery with adjuvant chemoradiotherapy strategies were 2.07, 2.18, 2.14, and 1.99, respectively. If the reduction in utility for multimodality treatment was increased to 21%, the QALYs gained for surgery alone, chemoradiotherapy followed by surgery, chemotherapy followed by surgery, and surgery with adjuvant chemoradiotherapy were 2.07, 2.03, 1.99, and 1.85, respectively. CONCLUSIONS: Chemoradiotherapy followed by surgery appears to be associated with the best survival and the largest expected gain in QALYs. However, the improvement in quality-adjusted life expectancy is modest at 40 days, and surgery alone becomes the preferred strategy if the reduction in utility associated with multimodality treatment is increased to 21%.

The prognostic importance of nonsignificant left main coronary artery disease in patients undergoing percutaneous coronary intervention.

OBJECTIVES: The purpose of this research was to study the association between nonsignificant (<50%) left main coronary artery disease (LMCAD) and short- and long-term survival in patients undergoing percutaneous coronary intervention (PCI). BACKGROUND: The prognostic importance of nonsignificant LMCAD is unknown; however, the co-existence of nonsignificant LMCAD may influence revascularization decisions. METHODS: We analyzed mortality and repeat catheterization rates of 11,855 patients in a prospective cardiac registry database who underwent single-vessel or multivessel PCI from January 1996 through December 2001. Of this cohort, 11.7% (n = 1,385) had nonsignificant (<50%) LMCAD. Outcomes were compared with those without LMCAD. A secondary analysis was performed on a larger cohort of 34,586 patients undergoing cardiac catheterization, irrespective of mode of revascularization therapy. RESULTS: Patients with nonsignificant LMCAD had more co-morbidities, and a significantly higher crude mortality rate at 1 year compared with those without LMCAD (4.4% vs. 3.4%; p = 0.05). The 7-year crude mortality hazard ratio (HR) of PCI patients with <50% LMCAD versus those with no LMCAD was 1.18 (95% confidence interval [CI] 0.94 to 1.46). After risk adjustment for differences in baseline clinical profile, however, the HR decreased to 0.98 (95% CI 0.79 to 1.23). Repeat catheterization rates at 1 year did not differ between groups. The secondary analysis in all patients with nonsignificant LMCAD showed an adjusted HR of 1.03 (95% CI 0.94 to 1.14). CONCLUSIONS: Patients undergoing single-vessel or multivessel PCI who have <50% LMCAD have a nonsignificantly increased 18% relative risk for mortality compared with those without detectable LMCAD that appears to be related to these patients' higher incidence of co-morbidities rather than the left main stenosis itself.