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.

Comorbidity risk adjustment methods have been used widely with administrative data, and the Charlson/Deyo method is perhaps the most commonly used in the literature. However, a new method defined by Elixhauser et al. has been introduced recently and could be superior, although it has not been validated widely. We compared the Charlson/Deyo and Elixhauser methods using Canadian administrative data on patients with myocardial infarction (MI). We conducted a historical cohort study. We used administrative hospital discharge data from a large Canadian city for all cases with acute MI coded as most responsible diagnosis between January 1, 1995, and March 31, 2001. We used each of the 2 methods to define comorbidity variables based on the International Classification of Diseases, 9th Revision, Clinical Modification codes present in each case record. We then compared 2 models predicting in-hospital mortality based on presence or absence of the variables defined by each of the methods. Frequency tables were produced and c-statistics and changes in −2 log likelihood (−2LogL) were calculated. We also visually assessed model performance by plotting observed and expected percentages of death for increasing risk categories defined by the 2 models. The Elixhauser model outperformed the Charlson/Deyo model in predicting mortality, with higher c-statistic values (0.793 vs. 0.704). Superior performance of the Elixhauser method is confirmed when plotting the expected and observed risks of death across groupings of increasing risk, in which the Elixhauser method yields a wider range of predicted and observed probabilities of death across groupings (2.5%–33%) than does the Charlson/Deyo method (5%–25%). The Elixhauser comorbidity measurement method performs better than the widely used Charlson/Deyo method in the Canadian acute MI cases studied.

Background— The optimal approach to revascularization in patients with kidney disease has not been determined. We studied survival by treatment group (CABG, percutaneous coronary intervention [PCI], or no revascularization) for patients with 3 categories of kidney function: dialysis-dependent kidney disease, non–dialysis-dependent kidney disease, and a reference group (serum creatinine <2.3 mg/dL). Methods and Results— Data were derived from the Alberta Provincial Project for Outcomes Assessment in Coronary Heart Disease (APPROACH), which captures information on all patients undergoing cardiac catheterization in Alberta, Canada. Characteristics and patient survival in 662 dialysis patients (1.6%) and 750 non–dialysis-dependent kidney disease patients (1.8%) were compared with the remainder of the 40 374 patients (96.6%). For the reference group, the adjusted 8-year survival rates for CABG, PCI, and no revascularization (NR) were 85.5%, 80.4%, and 72.3%, respectively ( P <0.001 for CABG versus NR; P <0.001 for PCI versus NR). Adjusted survival rates were 45.9% for CABG, 32.7% for PCI, and 29.7% for NR in the nondialysis kidney disease group ( P <0.001 for CABG versus NR; P =0.48 for PCI versus NR) and 44.8% for CABG, 41.2% for PCI, and 30.4% for NR in the dialysis group ( P =0.003 for CABG versus NR; P =0.03 for PCI versus NR). Conclusions— Compared with no revascularization, CABG was associated with better survival in all categories of kidney function. PCI was also associated with a lower risk of death than no revascularization in reference patients and dialysis-dependent kidney disease patients but not in patients with non–dialysis-dependent kidney disease. The presence of kidney disease or dependence on dialysis should not be a deterrent to revascularization, particularly with CABG.

Background: Low income appears to be associated with a higher prevalence of diabetes and diabetes related complications, however, little is known about how income influences access to diabetes care. The objective of the present study was to determine whether income is associated with referral to a diabetes centre within a universal health care system. Methods: Data on referral for diabetes care, diabetes prevalence and median household income were obtained from a regional Diabetes Education Centre (DEC) database, the Canadian National Diabetes Surveillance System (NDSS) and the 2001 Canadian Census respectively. Diabetes rate per capita, referral rate per capita and proportion with diabetes referred was determined for census dissemination areas. We used Chi square analyses to determine if diabetes prevalence or population rates of referral differed across income quintiles, and Poisson regression to model diabetes rate and referral rate in relation to income while controlling for education and age. Results: There was a significant gradient in both diabetes prevalence (χ² = 743.72, p < 0.0005) and population rates of referral (χ² = 168.435, p < 0.0005) across income quintiles, with the lowest income quintiles having the highest rates of diabetes and referral to the DEC. Referral rate among those with diabetes, however, was uniform across income quintiles. Controlling for age and education, Poisson regression models confirmed a significant socio-economic gradient in diabetes prevalence and population rates of referral. Conclusion: Low income is associated with a higher prevalence of diabetes and a higher population rate of referral to this regional DEC. After accounting for diabetes prevalence, however, the equal proportions referred to the DEC across income groups suggest that there is no access bias based on income.

Objectives: To determine the accuracy of CHADS₂ and CHA₂DS₂-VASc tools for predicting ischaemic stroke or transient ischaemic attack (TIA) and death in patients without a history of atrial fibrillation or flutter (AF). Methods: The study included 20 970 patients without known AF enrolled in the Alberta Provincial Project for Outcomes Assessment in Coronary Heart disease (APPROACH) prospective registry who were discharged after an acute coronary syndrome (ACS) between 2005 and 2011. The outcome measures were incident ischaemic stroke, TIA or death from any cause. Results: Over a median follow-up of 4.1 years, 453 patients (2.2%) had a stroke (n=297) or TIA (n=156) and 1903 (9.0%) died. The incidence of stroke or TIA increased with increases in each risk score (p<0.001), with an absolute annual incidence ≥1% with CHADS₂ ≥3 or CHA₂DS₂-VASc ≥4. Both CHADS₂ and CHA₂DS₂-VASc scores had acceptable discrimination performance (C-statistic=0.68 and 0.71, respectively). The mortality rate was also greater in patients with higher CHADS₂ and CHA₂DS₂-VASc scores (p<0.0001). Conclusions: In patients with ACS but no AF, the CHADS₂ and CHA₂DS₂-VASc scores predict ischaemic stroke/TIA events with similar accuracy to that observed in historical populations with non-valvular AF, but with lower absolute event rates. Further study of the utility of the CHADS₂ and CHA₂DS₂-VASc scores for the assessment of thromboembolic risk and selection of antithrombotic therapy in patients without AF is warranted.