Natural Language Processing for the Ascertainment and Phenotyping of Left Ventricular Hypertrophy and Hypertrophic Cardiomyopathy on Echocardiogram Reports

Extracting and accurately phenotyping electronic health documentation is critical for medical research and clinical care. We sought to develop a highly accurate and open-source natural language processing (NLP) module to ascertain and phenotype left ventricular hypertrophy (LVH) and hypertrophic cardiomyopathy (HCM) diagnoses from echocardiogram reports within a diverse hospital network. After the initial development on 17,250 echocardiogram reports, 700 unique reports from 6 hospitals were randomly selected from data repositories within the Mass General Brigham healthcare system and manually adjudicated by physicians for 10 subtypes of LVH and diagnoses of HCM. Using an open-source NLP system, the module was formally tested on 300 training set reports and validated on 400 reports. The sensitivity, specificity, positive predictive value, and negative predictive value were calculated to assess the discriminative accuracy of the NLP module. The NLP demonstrated robust performance across the 10 LVH subtypes, with the overall sensitivity and specificity exceeding 96%. In addition, the NLP module demonstrated excellent performance in detecting HCM diagnoses, with sensitivity and specificity exceeding 93%. In conclusion, we designed a highly accurate NLP module to determine the presence of LVH and HCM on echocardiogram reports. Our work demonstrates the feasibility and accuracy of NLP to detect diagnoses on imaging reports, even when described in free text. This module has been placed in the public domain to advance research, trial recruitment, and population health management for patients with LVH-associated conditions.

A pragmatic clinical trial assessing the effect of a targeted notification and clinical support pathway on the diagnostic evaluation and treatment of individuals with left ventricular hypertrophy (NOTIFY-LVH)

BACKGROUND

Electronic health records contain vast amounts of cardiovascular data, including potential clues suggesting unrecognized conditions. One important example is the identification of left ventricular hypertrophy (LVH) on echocardiography. If the underlying causes are untreated, individuals are at increased risk of developing clinically significant pathology. As the most common cause of LVH, hypertension accounts for more cardiovascular deaths than any other modifiable risk factor. Contemporary healthcare systems have suboptimal mechanisms for detecting and effectively implementing hypertension treatment before downstream consequences develop. Thus, there is an urgent need to validate alternative intervention strategies for individuals with preexisting-but potentially unrecognized-LVH.

METHODS

Through a randomized pragmatic trial within a large integrated healthcare system, we will study the impact of a centralized clinical support pathway on the diagnosis and treatment of hypertension and other LVH-associated diseases in individuals with echocardiographic evidence of concentric LVH. Approximately 600 individuals who are not treated for hypertension and who do not have a known cardiomyopathy will be randomized. The intervention will be directed by population health coordinators who will notify longitudinal clinicians and offer to assist with the diagnostic evaluation of LVH. Our hypothesis is that an intervention that alerts clinicians to the presence of LVH will increase the detection and treatment of hypertension and the diagnosis of alternative causes of thickened myocardium. The primary outcome is the initiation of an antihypertensive medication. Secondary outcomes include new hypertension diagnoses and new cardiomyopathy diagnoses. The trial began in March 2023 and outcomes will be assessed 12 months from the start of follow-up.

CONCLUSION

The NOTIFY-LVH trial will assess the efficacy of a centralized intervention to improve the detection and treatment of hypertension and LVH-associated diseases. Additionally, it will serve as a proof-of-concept for how to effectively utilize previously collected electronic health data to improve the recognition and management of a broad range of chronic cardiovascular conditions.

TRIAL REGISTRATION

NCT05713916.

Patterns in Physician Burnout in a Stable-Linked Cohort

Importance

Physician burnout is widely reported to be an increasing problem in the US. Although prior analyses suggest physician burnout is rising nationally, these analyses have substantial limitations, including different physicians joining and leaving clinical practice.

Objective

To examine the prevalence of burnout among physicians in a large multispecialty group over a 5-year period.

Design, Setting, and Participants

This survey study was conducted in 2017, 2019, and 2021 and involved physician faculty members of the Massachusetts General Physicians Organization. Participants represented different clinical specialties and a full range of career stages. The online survey instrument had 4 domains: physician career and compensation satisfaction, physician well-being, administrative workload on physicians, and leadership and diversity.

Exposure

Time.

Main Outcomes and Measures

Physician burnout, which was assessed with the Maslach Burnout Inventory. A binary burnout measure was used, which defined burnout as a high score in 2 of the 3 burnout subscales: Exhaustion, Cynicism, and Reduced Personal Efficacy.

Results

A total of 1373 physicians (72.9% of the original 2017 cohort) participated in all 3 surveys. The cohort included 690 (50.3%) male, 921 (67.1%) White, and 1189 (86.6%) non-Hispanic individuals. The response rates were 93.0% in 2017, 93.0% in 2019, and 92.0% in 2021. Concerning years of experience, the cohort was relatively well distributed, with the highest number and proportion of physicians (478 [34.8%]) reporting between 11 and 20 years of experience. Within this group, burnout declined from 44.4% (610 physicians) in 2017 to 41.9% (575) in 2019 (P = .18) before increasing to 50.4% (692) in 2021 (P < .001).

Conclusions and Relevance

Findings of this survey study suggest that the physician burnout rate in the US is increasing. This pattern represents a potential threat to the ability of the US health care system to care for patients and needs urgent solutions.

Association between implanted cardioverter-defibrillators and mortality for patients with left ventricular ejection fraction between 30% and 35

BACKGROUND

Consensus guidelines support the use of implanted cardioverter-defibrillators (ICD) for primary prevention of sudden cardiac death in patients with either non-ischaemic or ischaemic cardiomyopathy with left ventricular ejection fraction (LVEF) ≤35%. However, evidence from trials for efficacy specifically for patients with LVEF near 35% is weak. Past trials are underpowered for this population and future trials are unlikely to be performed.

METHODS

Patients with lowest LVEF between 30% and 35% without an ICD prior to the lowest-LVEF echo (defined as 'time zero') were identified by querying echocardiography data from 28 November 2001 to 9 July 2020 at the Massachusetts General Hospital linked to ICD treatment status. To assess the association between ICD and mortality, propensity score matching followed by Cox proportional hazards models considering treatment status as a time-dependent covariate was used. A secondary analysis was performed for LVEF 36%-40%.

RESULTS

Initially, 526 440 echocardiograms representing 266 601 unique patients were identified. After inclusion and exclusion criteria were applied, 6109 patients remained for the analytical cohort. In bivariate unadjusted comparisons, patients who received ICDs were substantially more often male (79.8% vs 65.4%, p<0.0001), more often white (87.5% vs 83.7%, p<0.046) and more often had a history of ventricular tachycardia (74.5% vs 19.1%, p<0.0001) and myocardial infarction (56.1% vs 38.2%, p<0.0001). In the propensity matched sample, after accounting for time-dependence, there was no association between ICD and mortality (HR 0.93, 95% CI 0.75 to 1.15, p=0.482).

CONCLUSIONS

ICD therapy was not associated with reduced mortality near the conventional LVEF threshold of 35%. Although this treatment design cannot definitively demonstrate lack of efficacy, our results are concordant with available prior trial data. A definitive, well-powered trial is needed to answer the important clinical question of primary prevention ICD efficacy between LVEF 30% and 35%.

The Effect of Standardized Hospitalist Information Cards on the Patient Experience: a Quasi-Experimental Prospective Cohort Study

BACKGROUND

Communication with clinicians is an important component of a hospitalized patient's experience.

OBJECTIVE

To test the impact of standardized hospitalist information cards on the patient experience.

DESIGN

Quasi-experimental study in a U.S. tertiary-care center.

PARTICIPANTS

All-comer medicine inpatients.

INTERVENTIONS

Standardized hospitalist information cards containing name and information on a hospitalist's role and availability vs. usual care.

MAIN MEASURES

Patients' rating of the overall communication as excellent ("top-box" score); qualitative feedback summarized via inductive coding.

KEY RESULTS

Five hundred sixty-six surveys from 418 patients were collected for analysis. In a multivariate regression model, standardized hospitalist information cards significantly improved the odds of a "top-box" score on overall communication (odds ratio: 2.32; 95% confidence intervals: 1.07-5.06). Other statistically significant covariates were patient age (0.98, 0.97-0.99), hospitalist role (physician vs. advanced practice provider, 0.56; 0.38-0.81), and hospitalist-patient gender combination (female-female vs. male-male, 2.14; 1.35-3.40). Eighty-seven percent of patients found the standardized hospitalist information cards useful, the perceived most useful information being how to contact the hospitalist and knowing their schedule.

CONCLUSIONS

Hospitalized patients' experience of their communication with hospitalists may be improved by using standardized hospitalist information cards. Younger patients cared for by a team with an advanced practice provider, as well as female patients paired with female providers, were more likely to be satisfied with the overall communication. Assessing the impact of information cards should be studied in other settings to confirm generalizability.

Association of Self-reported Primary Care Physician Tolerance for Uncertainty With Variations in Resource Use and Patient Experience

IMPORTANCE

Inappropriate variations in clinical practice are a known cause of poor quality and safety, with variations often associated with nonclinical factors, such as individual differences in cognitive processing. The differential response of physicians to uncertainty may explain some of the variations in resource use and patient experience.

OBJECTIVE

To examine the association of physician tolerance for uncertainty with variations in resource use and patient experience.

DESIGN, SETTING, AND PARTICIPANTS

This survey study linked physician survey data (May to June 2019), patient experience survey data (January 2016 to December 2019), and billing data (January 2019 to December 2019) among primary care physicians (PCPs) at Massachusetts General Hospital with at least 10 visits in 2019. The statistical analysis was performed in 2021.

MAIN OUTCOMES AND MEASURES

The analysis examined associations of PCP tolerance for uncertainty with the tendency to order diagnostic tests, the frequency of outpatient visits, hospital admissions, emergency department visits, and patient experience data (focused on physician communication and overall rating). A 2-stage hierarchical framework was used to account for clustering of patients under PCPs. Binary outcomes were modeled using a hierarchical logistic model, and count outcomes were modeled using hierarchical Poisson or negative binomial models. The analysis was adjusted for patient demographic variables (age, sex, and race and ethnicity), socioeconomic factors (payer and neighborhood income), and clinical comorbidities.

RESULTS

Of 217 included physicians, 137 (63.1%) were women, and 174 (80.2%) were adult PCPs. A total of 62 physicians (28.6%) reported low tolerance, 59 (27.2%) reported medium tolerance, and 96 (44.2%) reported high tolerance for uncertainty. Physicians with a low tolerance for uncertainty were less likely to order complete blood cell counts (odds ratio [OR], 0.66; 95% CI, 0.50-0.88), thyroid tests (OR, 0.67; 95% CI, 0.52-0.88), a basic metabolic profile (OR, 0.78; 95% CI, 0.60-1.00), and liver function tests (OR, 0.72; 95% CI, 0.53-0.99) than physicians with a high tolerance for uncertainty. Physicians who reported higher tolerance for uncertainty were more likely to receive higher patient experience scores for listening to patients carefully (OR, 0.65; 95% CI, 0.50-0.83) and higher overall ratings (OR, 0.80; 95% CI, 0.66-0.98) than physicians with medium tolerance. Conversely, no association was found between physician tolerance for uncertainty and patient outpatient visits, hospital admissions, or emergency department visits.

CONCLUSIONS AND RELEVANCE

In clinical practice, identifying and effectively managing inappropriate variations and improving patient experience have proven to be difficult, despite increased attention to these issues. This study supports the hypothesis that physicians' tolerance for uncertainty is associated with differences in resource use and patient experience. Whether enhancing physicians' tolerance for uncertainty could help reduce unwarranted practice variations, improve quality and patient safety, and improve patient's experience remains to be established.

Impact of Gabapentin on PACU Length of Stay and Perioperative Intravenous Opioid Use for ERAS Hysterectomy Patients

We investigated the impact of preoperative gabapentin on perioperative intravenous opioid requirements and post anesthesia care unit length of stay (PACU LOS) for patients undergoing laparoscopic and vaginal hysterectomies within an Enhanced Recovery After Surgery (ERAS) pathway. A multidisciplinary team retrospectively examined 2,015 patients who underwent laparoscopic or vaginal hysterectomies between October 2016 and January 2020 at a single academic institution. The average PACU LOS was 168 min among patients who did not receive gabapentin vs. 180 min both among patients who received ≤ 300 mg of gabapentin and patients who received > 300 mg of gabapentin. After adjusting for demographics and medical comorbidities, PACU LOS for patients given ≤ 300 mg gabapentin was 6% longer (rate ratio (RR) = 1.06, 95% CI = 1.01-1.11) than for patients who were not given gabapentin, and for patients who received > 300 mg of gabapentin was 7% longer (RR = 1.07, 95%CI = 1.01-1.13) than for those who did not receive gabapentin. Patients who received ≤ 300 mg gabapentin received 9% less perioperative intravenous hydromorphone than patients who did not receive gabapentin (RR = 0.91, 95% CI = 0.86 - 0.97); patients who received > 300 mg of gabapentin received 12% less perioperative intravenous hydromorphone than patients who did not receive gabapentin (RR = 0.88, 95% CI = 0.82 - 0.95). These findings represent an absolute difference of 0.09 mg intravenous hydromorphone. There were no statistically significant differences in total intravenous fentanyl received. Preoperative gabapentin given as part of an ERAS pathway is associated with statistically but not clinically significant increases in PACU LOS and decreases in total perioperative intravenous opioid use.

Implementing ERAS: how we achieved success within an anesthesia department

Background

The Massachusetts General Hospital is a large, quaternary care institution with 58 operating rooms, 164 anesthesiologists, 76 certified nurse anesthetists (CRNAs), an anesthesiology residency program that admits 25 residents annually, and 35 surgeons who perform laparoscopic, vaginal, and open hysterectomies. In March of 2018, our institution launched an Enhanced Recovery After Surgery (ERAS) pathway for patients undergoing hysterectomy. To implement the anesthesia bundle of this pathway, an intensive 14-month educational endeavor was created and put into effect. There were no subsequent additional educational interventions.

Methods

We retrospectively reviewed records of 2570 patients who underwent hysterectomy between October 2016 and March 2020 to determine adherence to the anesthesia bundle of the ERAS Hysterectomy pathway. RESULTS: Increased adherence to the four elements of the anesthesia bundle (p < 0.001) was achieved during the intervention period. Compliance with the pathway was sustained in the post-intervention period despite no additional actions.

Conclusions

Implementing the anesthesia bundle of an ERAS pathway in a large anesthesia group with diverse providers successfully occurred using implementation science-based approach of intense interventions, and these results were maintained after the intervention ceased.

Association of an Acute Myocardial Infarction Readmission-Reduction Program With Mortality and Readmission

BACKGROUND

Reducing hospital readmission after acute myocardial infarction (AMI) has the potential to both improve quality and reduce costs. As such, readmission after AMI has been a target of financial penalties through Medicare. However, substantial concern exists about potential adverse effects and efficacious readmission-reduction strategies are not well validated.

METHODS AND RESULTS

We started an AMI readmissions reduction program in November 2017. Between July 2016 and February 2019, hospital billing data were queried to detect all inpatient hospitalizations at the Massachusetts General Hospital for AMI. Thirty-day readmission was identified through hospital billing data, and mortality was extracted from our electronic health record. The data set was merged with claims data for patients in accountable care organizations to detect readmission at other hospitals. We performed segmented linear regression, adjusting for secular trend and case mix, to assess the independent association of our program on both outcome variables. After inclusion and exclusion criteria were applied, the study population included 2020 patients. The overall 30-day readmission rate was higher before the intervention than after the intervention (15.5% versus 10.7%, P=0.002). The overall 30-day mortality rate was similar in both time periods (1.8% versus 1.4%, P=0.457). The program was associated with initial reduction in 30-day readmission (-9.8%, P=0.0002) and 30-day mortality (-2.6%, P=0.041). The program did not change trend in 30-day readmission (+0.19% readmissions/mo, P=0.554) and trend in 30-day mortality (-0.21% deaths/mo, P=0.119).

CONCLUSIONS

An AMI readmissions reduction program that increases outpatient and emergency department (ED) access to cardiology care is associated with reduced 30-day readmission and 30-day mortality. Similar statistical techniques can be used to conduct a rigorous, mechanistic program evaluation of other quality improvement initiatives.

Physician Burnout, Engagement and Career Satisfaction in a Large Academic Medical Practice

OBJECTIVE

To determine (1) if engagement among physicians impacted plans to stay in current role and job satisfaction, (2) what factors impact engagement and burnout, and (3) the relationship between engagement and burnout. Burnout has been described as a syndrome characterized by depersonalization, emotional exhaustion, and a sense of low personal accomplishment resulting in decreased effectiveness at work. Engagement may be regarded as the antonym to burnout and has been described as a connection to one's work characterized by dedication, vigor, and absorption.

DESIGN

We extracted data from an academic practice-wide survey conducted at two time-points and evaluated physician burnout and engagement. We used the Maslach Burnout Inventory and the Utrecht Work Engagement Scale to evaluate the association between burnout and engagement and the impact of engagement on mitigating the effect of burnout in a large physician academic faculty practice.

SETTING

Large academic practice PARTICIPANTS: Academic physicians METHODS: The authors conducted a hospital-wide physician practice survey in 2014 and 2017 assessing physician burnout and engagement.

RESULTS

Of eligible physicians (n=1882), 92.0% completed a survey. High levels of engagement and burnout were shown in 59.5% and 45.6%, respectively. Compared to physicians with high levels of engagement and low levels of burnout, physicians with low engagement and low burnout were less satisfied with their career (OR=0.20, 95% CI=0.11-0.35) and less likely to stay in their current role (OR=0.52, 95% CI= 0.37-0.73). Among physicians with high levels of burnout, highly engaged physicians were more satisfied (OR=0.21; 95% CI=0.12-0.36 vs OR=0.08; 95% CI=0.05-0.12) and more likely to stay in their career (OR=0.34; 95% CI=0.25-0.45 vs OR=0.27; 95% CI=0.21-0.34) than non-engaged physicians.

CONCLUSION

Engaged physicians have higher career satisfaction. There are many actionable ways to improve engagement.

Trends and Factors Associated With Physician Burnout at a Multispecialty Academic Faculty Practice Organization

IMPORTANCE

Physician burnout is common, and prevalence may differ throughout a clinician's career. Burnout has negative consequences for physician wellness, patient care, and the health care system. Identifying factors associated with burnout is critical in designing and implementing initiatives to reduce burnout.

OBJECTIVE

To measure trends and identify factors associated with physician burnout.

DESIGN, SETTING, AND PARTICIPANTS

Survey study conducted from May 16 to June 15, 2014, and again from May 16 to June 15, 2017, measuring rates of physician burnout in a large academic medical practice. Factors associated with burnout out were evaluated. In 2014, 1774 of 1850 eligible physicians (95.9%) completed the survey. In 2017, 1882 of 2031 (92.7%) completed the survey.

EXPOSURES

Medical specialty, demographic characteristics, years in practice, and reported rates of burnout.

MAIN OUTCOMES AND MEASURES

Burnout rates measured at 2 points and risk factors associated with burnout.

RESULTS

Respondents included 1027 men (57.9%) and 747 women (42.1%) in 2014 and 962 men (51.1%) and 759 women (40.3%) in 2017. The mean (SD) number of years since training completion was 15.3 (11.3) in the 2014 survey data and 15.1 (11.3) in the 2017 data. Burnout increased from 40.6% to 45.6% between the 2 points. The increased rate was associated with an increase in exhaustion (from 52.9% in 2014 to 57.7% in 2017; difference, 4.8%; 95% CI, 1.6%-8.0%; P = .004) and cynicism (from 44.8% in 2014 to 51.1% in 2017; difference, 6.3%; 95% CI, 3.1%-9.6%; P < .001). Compared with midcareer physicians (11-20 years since training), early-career physicians (≤10 years since training) were more susceptible to burnout (odds ratio, 1.36; 95% CI, 1.05-1.77), while physicians in their late career (>30 years since training) were less vulnerable (odds ratio, 0.59; 95% CI, 0.40-0.88).

CONCLUSIONS AND RELEVANCE

Efforts to alleviate physician burnout and administrative burden require a combination of a shared commitment from physicians and organizations and central and locally implemented programs. Continued research is necessary to establish the most effective initiatives to decrease physician burnout at the individual and organizational level.

Measuring individual physician clinical productivity in an era of consolidated group practices

BACKGROUND

As physician groups consolidate and value-based payment replaces traditional fee-for-service systems, physician practices have greater need to accurately measure individual physician clinical productivity within team-based systems. We compared methodologies to measure individual physician outpatient clinical productivity after adjustment for shared practice resources.

METHODS

For cardiologists at our hospital between January 2015 and June 2016, we assessed productivity by examining completed patient visits per clinical session per week. Using mixed-effects models, we sequentially accounted for shared practice resources and underlying baseline characteristics. We compared mixed-effects and Generalized Estimating Equations (GEE) models using K-fold cross validation, and compared mixed-effect, GEE, and Data Envelopment Analysis (DEA) models based on ranking of physicians by productivity.

RESULTS

A mixed-effects model adjusting for shared practice resources reduced variation in productivity among providers by 63% compared to an unadjusted model. Mixed-effects productivity rankings correlated strongly with GEE rankings (Spearman 0.99), but outperformed GEE on K-fold cross validation (root mean squared error 2.66 vs 3.02; mean absolute error 1.89 vs 2.20, respectively). Mixed-effects model rankings had moderate correlation with DEA model rankings (Spearman 0.692), though this improved upon exclusion of outliers (Spearman 0.755).

CONCLUSIONS

Mixed-effects modeling accounts for significant variation in productivity secondary to shared practice resources, outperforms GEE in predictive power, and is less vulnerable to outliers than DEA.

IMPLICATIONS

With mixed-effects regression analysis using otherwise easily accessible administrative data, practices can evaluate physician clinical productivity more fairly and make more informed management decisions on physician compensation and resource allocation.

Variation in the Echocardiographic Surveillance of Primary Mitral Regurgitation

BACKGROUND

Clinical outcomes after surgical treatment of mitral regurgitation are worse if intervention occurs after deterioration of left ventricular size and function. Transthoracic echocardiographic (TTE) surveillance of patients with mitral regurgitation is indicated to avoid adverse ventricular remodeling. Overly frequent TTEs can impair patient access and reduce value in care delivery. This balance between timely surveillance and overutilization of TTE in valvular disease provides a model to study variation in the delivery of healthcare services. We investigated patient and provider factors contributing to variation in TTE utilization and hypothesized that variation was attributable to provider practice even after adjustment for patient characteristics.

METHODS AND RESULTS

We obtained records of all TTEs from 2001 to 2016 completed at a large echocardiography laboratory. The outcome variable was time interval between TTEs. We constructed a mixed-effects linear regression model with the individual physician as the random effect in the model and used intraclass correlation coefficient to assess the proportion of outcome variation because of provider practice. Our study cohort was 55 773 TTEs corresponding to 37 843 intervals ordered by 635 providers. The mean interval between TTEs was 12.4 months, 17.0 months, 18.3 months, and 17.4 months for severe, moderate, mild, and trace mitral regurgitation, respectively, with 20% of providers deemed overutilizers of TTEs and 25% underutilizers.

CONCLUSIONS

We conclude that there is substantial variation in follow-up intervals for TTE assessment of mitral regurgitation, despite risk-adjustment for patient variables, likely because of provider factors.

The Impact of Administrative Burden on Academic Physicians: Results of a Hospital-Wide Physician Survey

PURPOSE

To determine the characteristics of clinically active academic physicians most affected by administrative burden; the correlation between administrative burden, burnout, and career satisfaction among academic physicians; and the relative value and burden of specific administrative tasks.

METHOD

The authors analyzed data from the 2014 Massachusetts General Physicians Organization Survey. Respondents reported the percentage of time they spent on patient-related administrative duties and rated the value and burden associated with specific administrative tasks. A five-point Likert scale and multivariate regression identified predictors of administrative burden and assessed the impact of administrative burden on perceived quality of care, career satisfaction, and burnout.

RESULTS

Of the eligible workforce, 1,774 physicians (96%) responded to the survey. On average, 24% of working hours were spent on administrative duties. Primary care physicians and women reported spending more time on administrative duties compared with other physicians. Two-thirds of respondents reported that administrative duties negatively affect their ability to deliver high-quality care. Physicians who reported higher percentages of time spent on administrative duties had lower levels of career satisfaction, higher levels of burnout, and were more likely to be considering seeing fewer patients in the future. Prior authorizations, clinical documentation, and medication reconciliation were rated the most burdensome tasks.

CONCLUSIONS

Administrative duties required substantial physician time and affected physicians' perceptions of being able to deliver high-quality care, career satisfaction, burnout, and likelihood to continue clinical practice. There is variation in administrative burden across specialties, and multiple areas of work contribute to overall administrative workload.

Differences Among Cardiologists in Rates of Positive Coronary Angiograms

Background

Understanding the sources of variation for high‐cost services has the potential to improve both patient outcomes and value in health care delivery. Nationally, the overall diagnostic yield of coronary angiography is relatively low, suggesting overutilization. Understanding how individual cardiologists request catheterization may suggest opportunities for improving quality and value. We aimed to assess and explain variation in positive angiograms among referring cardiologists.

Methods and Results

We identified all cases of diagnostic coronary angiography at Massachusetts General Hospital from January 1, 2012, to June 30, 2013. We excluded angiograms for acute coronary syndrome. For each angiogram, we identified clinical features of the patients and characteristics of the requesting cardiologists. We also identified angiogram positivity, defined as at least 1 epicardial coronary stenosis ≥50% luminal narrowing. We then constructed a series of mixed‐effects logistic regression models to analyze predictors of positive coronary angiograms. We assessed variation by physician in the models with median odds ratios. Over this time period, 5015 angiograms were identified. We excluded angiograms ordered by cardiologists requesting <10 angiograms. Among the remaining 2925 angiograms, 1450 (49.6%) were positive. Significant predictors of positive angiograms included age, male patients, and peripheral arterial disease. After adjustment for clinical variables only, the median odds ratio was 1.23 (95% CI 1.0–1.36), consistent with only borderline clinical variation after adjustment. In the full clinical and nonclinical model, the median odds ratio was 1.07 (95% CI 1.07–1.20), also consistent with clinically insignificant variation.

Conclusions

Substantial variation exists among requesting cardiologists with respect to positive and negative coronary angiograms. After adjustment for clinical variables, there was only borderline clinically significant variation. These results emphasize the importance of risk adjustment in reporting related to quality and value.