Referring and Specialist Physician Gender and Specialist Billing.

Importance: While a gender pay gap in medicine has been well documented, relatively little research has addressed mechanisms that mediate gender differences in referral income for specialists. Objective: To examine gender-based disparities in medical and surgical specialist referrals in Ontario, Canada. Design, Setting, and Participants: This cross-sectional study included referrals for specialist care ascertained from Ontario Health Insurance Plan physician billings for fiscal year 2018 to 2019. Participants were specialist physicians who received new patient consultations from April 1, 2018, to March 31, 2019, and the associated referring physicians. Data were analyzed from April 2018 to March 2020, including a 12-month follow-up period. Exposures: Specialist and referring physician gender (female or male). Main Outcomes and Measures: Revenue per referral was defined based on an episode-of-care approach as total billings for a 12-month period from the initial consultation. Mean total billings for female and male specialists were compared and the differential divided into the portion owing to referral volume vs referral revenue. Difference-in-differences multivariable regression analysis was used to estimate gender-based differences in revenue per referral. For each referring physician, gender-based differences in referral patterns were examined using case-control analysis, in which specialists who received a referral were compared with matched control specialists who did not receive a referral. This analysis considered the gender of the specialist and concordance between the gender of the referring physician and specialist, among other characteristics. Results: Of 7 621 365 new referrals, 32 824 referring physicians, of whom 13 512 (41.2%) were female (mean [SD] age, 46.3 [11.6] years) and 19 312 (58.8%) were male (mean [SD] age, 52.9 [13.5] years), made referrals to 13 582 specialists, of whom 4890 (36.0%) were female (mean [SD] age, 45.6 [11.0] years) and 8692 (64.0%) were male (mean [SD] age, 51.8 [13.0] years). Male specialists received more mean (SD) referrals than did female specialists (633 [666] vs 433 [515]), and the mean (SD) revenue per referral was higher for males ($350 [$474]) compared with females ($316 [$393]). Adjusted analysis demonstrated a -4.7% (95% CI, -4.9% to -4.5%) difference in the revenue per referral between male and female specialists. Multivariable regression analysis found that physicians referred more often to specialists of the same gender (odds ratio, 1.04; 95% CI, 1.03-1.04) but had higher odds of referring to male specialists (odds ratio, 1.10; 95% CI, 1.09-1.11). Conclusions and Relevance: In this cross-sectional study of the gender pay gap in specialist referral income, the number and revenue from referrals received differed by gender, as did the odds of receiving a referral from a physician of the same gender. Future research should examine the effectiveness of different policies to address this gap, such as a centralized, gender-blinded referral system.

Association between virtual primary care and emergency department use during the first year of the COVID-19 pandemic in Ontario, Canada.

BACKGROUND: Uptake of virtual care increased substantially during the first year of the COVID-19 pandemic. The aim of this study was to evaluate whether a shift from in-person to virtual visits by primary care physicians was associated with increased use of emergency departments among their enrolled patients. METHODS: We conducted an observational study of monthly virtual visits and emergency department visits from Apr. 1, 2020, to Mar. 31, 2021, using administrative data from Ontario, Canada. We used multivariable regression analysis to estimate the association between the proportion of a physician's visits that were delivered virtually and the number of emergency department visits among their enrolled patients. RESULTS: The proportion of virtual visits was higher among female, younger and urban physicians, and the number of emergency department visits was lower among patients of female and urban physicians. In an unadjusted analysis, a 1% increase in a physician's proportion of virtual visits was found to be associated with 11.0 (95% confidence interval [CI] 10.1-11.8) fewer emergency department visits per 1000 rostered patients. After controlling for covariates, we observed no statistically significant change in emergency department visits per 1% increase in the proportion of virtual visits (0.2, 95% CI -0.5 to 0.9). INTERPRETATION: We did not find evidence that patients substituted emergency department visits in the context of decreased availability of in-person care with their family physician during the first year of the COVID-19 pandemic. Future research should focus on the long-term impact of virtual care on access and quality of patient care.

Trends in prevalence of chronic disease and multimorbidity in Ontario, Canada.

BACKGROUND: New case-mix tools from the Canadian Institute for Health Information offer a novel way of exploring the prevalence of chronic disease and multimorbidity using diagnostic data. We took a comprehensive approach to determine whether the prevalence of chronic disease and multimorbidity has been rising in Ontario, Canada. METHODS: In this observational study, we applied case-mix methodology to a population-based cohort. We used 10 years of patient-level data (fiscal years 2008/09 to 2017/18) from multiple care settings to compute the rolling 5-year prevalence of 85 chronic diseases and multimorbidity (i.e., the co-occurrence of 2 or more diagnoses). Diseases were further classified based on type and severity. We report both crude and age- and sex-standardized trends. RESULTS: The number of patients with chronic disease increased by 11.0% over the 10-year study period to 9.8 million in 2017/18, and the number with multimorbidity increased 12.2% to 6.5 million. Overall increases from 2008/09 to 2017/18 in the crude prevalence of chronic conditions and multimorbidity were driven by population aging. After adjustments for age and sex, the prevalence of patients with ≥ 1 chronic conditions decreased from 70.2% to 69.1%, and the prevalence of multimorbidity decreased from 47.1% to 45.6%. This downward trend was concentrated in minor and moderate diseases, whereas the prevalence of many major chronic diseases rose, along with instances of extreme multimorbidity (≥ 8 conditions). Age- and sex-standardized resource intensity weights, which reflect relative expected costs associated with patient diagnostic profiles, increased 4.6%. INTERPRETATION: Evidence of an upward trend in the prevalence of chronic disease was mixed. However, the change in case mix toward more serious conditions, along with increasing patient resource intensity weights overall, may portend a future need for population health management and increased health system spending above that predicted by population aging.

Use of the Population Grouping Methodology of the Canadian Institute for Health Information to predict high-cost health system users in Ontario.

BACKGROUND: Prior research has consistently shown that the heaviest users account for a disproportionate share of health care costs. As such, predicting high-cost users may be a precondition for cost containment. We evaluated the ability of a new health risk predictive modelling tool, which was developed by the Canadian Institute for Health Information (CIHI), to identify future high-cost cases. METHODS: We ran the CIHI model using administrative health care data for Ontario (fiscal years 2014/15 and 2015/16) to predict the risk, for each individual in the study population, of being a high-cost user 1 year in the future. We also estimated actual costs for the prediction period. We evaluated model performance for selected percentiles of cost based on the discrimination and calibration of the model. RESULTS: A total of 11 684 427 individuals were included in the analysis. Overall, 10% of this population had annual costs exceeding $3050 per person in fiscal year 2016/17, accounting for 71.6% of total expenditures; 5% had costs above $6374 (58.2% of total expenditures); and 1% exceeded $22 995 (30.5% of total expenditures). Model performance increased with higher cost thresholds. The c-statistic was 0.78 (reasonable), 0.81 (strong) and 0.86 (very strong) at the 10%, 5% and 1% cost thresholds, respectively. INTERPRETATION: The CIHI Population Grouping Methodology was designed to predict the average user of health care services, yet performed adequately for predicting high-cost users. Although we recommend the development of a purpose-designed tool to improve model performance, the existing CIHI Population Grouping Methodology may be used - as is or in concert with additional information - for many applications requiring prediction of future high-cost users.

Using Diagnoses to Estimate Health Care Cost Risk in Canada.

OBJECTIVE: Until recently, the options for summarizing Canadian patient complexity were limited to health risk predictive modeling tools developed outside of Canada. This study aims to validate a new model created by the Canadian Institute for Health Information (CIHI) for Canada's health care environment. RESEARCH DESIGN: This was a cohort study. SUBJECTS: The rolling population eligible for coverage under Ontario's Universal Provincial Health Insurance Program in the fiscal years (FYs) 2006/2007-2016/2017 (12-13 million annually) comprised the subjects. MEASURES: To evaluate model performance, we compared predicted cost risk at the individual level, on the basis of diagnosis history, with estimates of actual patient-level cost using "out-of-the-box" cost weights created by running the CIHI software "as is." We next considered whether performance could be improved by recalibrating the model weights, censoring outliers, or adding prior cost. RESULTS: We were able to closely match model performance reported by CIHI for their 2010-2012 development sample (concurrent R=48.0%; prospective R=8.9%) and show that performance improved over time (concurrent R=51.9%; prospective R=9.7% in 2014-2016). Recalibrating the model did not substantively affect prospective period performance, even with the addition of prior cost and censoring of cost outliers. However, censoring substantively improved concurrent period explanatory power (from R=53.6% to 66.7%). CONCLUSIONS: We validated the CIHI model for 2 periods, FYs 2010/2011-2012/2013 and FYs 2014/2015-2016/2017. Out-of-the-box model performance for Ontario was as good as that reported by CIHI for the development sample based on 3-province data (British Columbia, Alberta, and Ontario). We found that performance was robust to variations in model specification, data sources, and time.

Cinacalcet versus standard treatment for chronic kidney disease: a systematic review and meta-analysis.

BACKGROUND: Chronic kidney disease-mineral and bone disorders (CKD-MBD) have been associated with poor health outcomes, including diminished quality and length of life. Standard management for CKD-MBD includes phosphate restricted diet, vitamin D and phosphate binders. Persistently elevated parathyroid hormone levels may require the addition of cinacalcet hydrochloride (cinacalcet), which sensitizes calcium receptors in the parathyroid gland. PURPOSE: The objective of this systematic review is to compare, in patients with CKD-MBD the effect of cinacalcet versus standard treatment on patient-important outcomes, including parathyroidectomy, fractures, hospitalizations due to cardiovascular events, cardiovascular mortality, all-cause mortality, and intermediate outcomes, in particular Kidney Disease Outcome Quality Initiative targets. METHODS: Data sources included MEDLINE, EMBASE, the Cochrane Register of Controlled Trials and Web of Science from 1996 to June 2015. Teams of two reviewers, independently and in duplicate, screened titles and abstracts and potentially eligible full text reports to determine eligibility, and subsequently abstracted data and assessed risk of bias in eligible trials. We calculated the effect estimates (risk ratios or mean differences) and 95% confidence intervals, as well as statistical measures of variability in results across studies using random effect models. We used the GRADE (Grading of Recommendations, Assessment, Development and Evaluation) approach to rate quality of evidence about estimates of effect on an outcome-by-outcome basis for all outcomes. We presented our results with a GRADE summary table. RESULTS: Twenty-four trials including 8311 CKD patients proved eligible. The results left considerable uncertainty regarding the impact of cinacalcet on reducing fractures (relative risk [RR] 0.59, 95% confidence interval [CI] 0.13-2.60; heterogeneity: p = 0.03, I(2)= 78%; very low quality evidence), and indicated that cinacalcet did not reduce hospitalizations due to cardiovascular events (RR 0.93, 95% CI 0.85-1.02, moderate quality of evidence), cardiovascular mortality (RR 0.95, 95% CI 0.84-1.07; heterogeneity p= 0.61, high quality evidence) or all-cause mortality (RR 0.96, 95% CI 0.89-1.04; heterogeneity: p= 0.98, I(2)= 0%; moderate quality evidence). Cinacalcet reduced the need for parathyroidectomy (RR 0.30, 95% CI 0.22-0.42; heterogeneity: p= 0.70, I(2)= 0%; absolute effect 55 fewer per 1000 [95% CI 61 fewer to 45 fewer], high quality of evidence). The most common adverse event associated with cinacalcet therapy was gastrointestinal side effects. Cinacalcet increased nausea (RR 2.16, 95% CI 1.46-3.21, absolute effect 158 more per 1000 [95% CI 82 more to 302 more]) and vomiting (RR 2.15, 95% CI 1.66-2.80, absolute effect 63 more per 1000 [95% CI 109 more to 171 more]). Cinacalcet treatment increased the rate of hypocalcemia (RR 6.0, 95% CI 3.65-9.87; heterogeneity: p= 0.71, I(2)= 0%, absolute effect 20 more per 1000 [95% CI 11 more to 36 more], high quality of evidence). CONCLUSIONS: In the hands of clinicians participating in these studies, cinacalcet decreased the rate of parathyroidectomy but had no influence on mortality. Patients and clinicians can trade of the benefit of fewer parathyroidectomies against the adverse effects.