Results of report cards for patients with congestive heart failure depend on the method used to adjust for severity

Association of Clinical Characteristics With Variation in Emergency Physician Preferences for Patients

Importance

Much of the wide variation in health care has been associated with practice variation among physicians. Physicians choosing to see patients with more (or fewer) care needs could also produce variations in care observed across physicians.

Objective

To quantify emergency physician preferences by measuring nonrandom variations in patients they choose to see.

Design, Setting, and Participants

This cross-sectional study used a large, detailed clinical data set from an electronic health record system of a single academic hospital. The data set included all emergency department (ED) encounters of adult patients from January 1, 2010, to May 31, 2015, as well as ED visits information. Data were analyzed from September 1, 2018, to March 31, 2019.

Exposure

Patient assignment to a particular emergency physician.

Main Outcomes and Measures

Variation in patient characteristics (age, sex, acuity [Emergency Severity Index score], and comorbidities) seen by emergency physicians before patient selection, adjusted for temporal factors (seasonal, weekly, and hourly variation in patient mix).

Results

This study analyzed 294 915 visits to the ED seen by 62 attending physicians. Of the 294 915 patients seen, the mean (SD) age was 48.6 (19.8) years and 176 690 patients (59.9%) were women. Many patient characteristics, such as age (F = 2.2; P < .001), comorbidities (F = 1.7; P < .001), and acuity (F = 4.7; P < .001), varied statistically significantly. Compared with the lowest-quintile physicians for each respective characteristic, the highest-quintile physicians saw patients who were older (mean age, 47.9 [95% CI, 47.8-48.1] vs 49.7 [95% CI, 49.5-49.9] years, respectively; difference, +1.8 years; 95% CI, 1.5-2.0 years) and sicker (mean comorbidity score: 0.4 [95% CI, 0.3-0.5] vs 1.8 [95% CI, 1.7-1.8], respectively; difference, +1.3; 95% CI, 1.2-1.4). These differences were absent or highly attenuated during overnight shifts, when only 1 physician was on duty and there was limited room for patient selection. Compared with earlier in the shift, the same physician later in the shift saw patients who were younger (mean age, 49.7 [95% CI, 49.4-49.7] vs 44.6 [95 % CI, 44.3-44.9] years, respectively; difference, -5.1 years; 95% CI, 4.8-5.5) and less sick (mean comorbidity score: 0.7 [95% CI, 0.7-0.8] vs 1.1 [95% CI, 1.1-1.1], respectively; difference, -0.4; 95% CI, 0.4-0.4). Accounting for preference variation resulted in substantial reordering of physician ranking by care intensity, as measured by ED charges, with 48 of 62 physicians (77%) being reclassified into a different quintile and 9 of 12 physicians (75%) in the highest care intensity quintile moving into a lower quintile. A regression model demonstrated that 22% of reported ED charges were associated with physician preference.

Conclusions and Relevance

This study found preference variation across physicians and within physicians during the course of a shift. These findings suggest that current efforts to reduce practice variation may not affect the variation associated with physician preferences, which reflect underlying differences in patient needs and not physician practice.

Heart failure in primary care: co-morbidity and utilization of health care resources

BACKGROUND

In order to ensure proper management of primary care (PC) services, the efficiency of the health professionals tasked with such services must be known. Patients with heart failure (HF) are characterized by advanced age, high co-morbidity and high resource utilization.

OBJECTIVE

To ascertain PC resource utilization by HF patients and variability in the management of such patients by GPs.

METHODS

Descriptive, cross-sectional study targeting a population attended by 129 GPs over the course of 1 year. All patients with diagnosis of HF in their clinical histories were included, classified using the Adjusted Clinical Group system and then grouped into six resource utilization bands (RUBs). Resource utilization and Efficiency Index were both calculated.

RESULTS

One hundred per cent of patients with HF were ranked in RUBs 3, 4 and 5. The highest GP visit rate was 20 and the lowest in excess of 10 visits per year. Prescription drug costs for these patients ranged from €885 to €1422 per patient per year. Health professional efficiency varied notably, even after adjustment for co-morbidity (Efficiency Index Variation Ratio of 28.27 for visits and 404.29 for prescription drug cost).

CONCLUSIONS

Patients with HF register a high utilization of resources, and there is great variability in the management of such patients by health professionals, which cannot be accounted for by the degree of case complexity.

Predicting risk of hospitalization or death among patients with heart failure in the veterans health administration

Patients with heart failure (HF) are at high risk of hospitalization or death. The objective of this study was to develop prediction models to identify patients with HF at highest risk for hospitalization or death. Using clinical and administrative databases, we identified 198,460 patients who received care from the Veterans Health Administration and had ≥1 primary or secondary diagnosis of HF that occurred within 1 year before June 1, 2009. We then tracked their outcomes of hospitalization and death during the subsequent 30 days and 1 year. Predictor variables chosen from 6 clinically relevant categories of sociodemographics, medical conditions, vital signs, use of health services, laboratory tests, and medications were used in multinomial regression models to predict outcomes of hospitalization and death. In patients who were in the ≥95th predicted risk percentile, observed event rates of hospitalization or death within 30 days and 1 year were 27% and 80% respectively, compared to population averages of 5% and 31%, respectively. The c-statistics for the 30-day outcomes were 0.82, 0.80, and 0.80 for hospitalization, death, and hospitalization or death, respectively, and 0.82, 0.76, and 0.77, respectively, for 1-year outcomes. In conclusion, prediction models using electronic health records can accurately identify patients who are at highest risk for hospitalization or death. This information can be used to assist care managers in selecting patients for interventions to decrease their risk of hospitalization or death.

Risk Adjustment: Towards Achieving Meaningful Comparison of Health Outcomes in the Real World

Health outcomes evaluation seeks to compare a new treatment or novel programme with the current standard of care, or to identify variation of outcomes across different healthcare providers. In the real world, it is not always possible to conduct randomised controlled trials to address the issue of comparator groups being different with respect to baseline risk factors for the outcomes. Therefore, risk adjustment is required to address patient factors that may lead to biases in estimates of treatment effects. It is essential when conducting outcomes evaluation of more than trivial significance. Risk adjustment begins by asking 4 questions: what outcome, what time frame, what population, and what purpose. Next, design issues are considered. This involves choosing the data source, planning data collection, defining the sample required, and selecting the variables carefully. Finally, analytical issues are considered. Regression modelling is central to every analytic strategy. Other methods that may augment regression include restriction, stratification, propensity scores, instrumental variables, and difference-in-differences. The construction of risk adjustment models is an iterative process requiring both art and science. Derived models should be validated. Limitations of risk adjustment include reliance on data availability and quality, imperfect method, ineffectiveness when comparators are very different, and sensitivity to different methods used. Thoughtful application of risk adjustment can improve the validity of comparisons between different treatments, programmes and providers. The extent of risk adjustment should be guided by its purpose. Finally, its methodology should be made explicit, so that informed readers can judge the robustness of results obtained. Key words: Health Services Research, Outcome assessement, Regression analysis, Risk adjustment

Coding and documentation: Medicare severity diagnosis-related groups and present-on-admission documentation

Effective October 1, 2007, the Centers for Medicare and Medicaid Services has changed its methodology for determining the diagnosis-related group for hospitalized patients. In an effort to more accurately reflect severity of illness, the 538 diagnosis-related groups have been converted to 745 new Medicare severity diagnosis-related groups. In addition, selected hospital-acquired complications not identified as present on admission will no longer be reimbursed. The changes will have profound effects on reimbursement for hospitalizations. To minimize financial losses under the new rules, hospitals and physicians will have to devote significant resources and attention to improved documentation. This article will discuss the new payment system, the physician's role in ensuring that all clinically important diagnoses are captured by coding specialists, and strategies that can be employed to respond proactively to the challenge.

Predicting in-hospital mortality in patients with cirrhosis: results differ across risk adjustment methods

Risk-adjusted health outcomes are often used to measure the quality of hospital care, yet the optimal approach in patients with liver disease is unclear. We sought to determine whether assessments of illness severity, defined as risk for in-hospital mortality, vary across methods in patients with cirrhosis. We identified 258,731 patients with cirrhosis hospitalized in the Nationwide Inpatient Sample between 2002 and 2005. The performance of four common risk adjustment methods (the Charlson/Deyo and Elixhauser comorbidity algorithms, Disease Staging, and All Patient Refined Diagnosis Related Groups [APR-DRGs]) for predicting in-hospital mortality was determined using the c-statistic. Subgroup analyses were conducted according to a primary versus secondary diagnosis of cirrhosis and in homogeneous patient subgroups (hepatic encephalopathy, hepatocellular carcinoma, congestive heart failure, pneumonia, hip fracture, and cholelithiasis). Patients were also ranked according to the probability of death as predicted by each method, and rankings were compared across methods. Predicted mortality according to the risk adjustment methods agreed for only 55%-67% of patients. Similarly, performance of the methods for predicting in-hospital mortality varied significantly. Overall, the c-statistics (95% confidence interval) for the Charlson/Deyo and Elixhauser algorithms, Disease Staging, and APR-DRGs were 0.683 (0.680-0.687), 0.749 (0.746-0.752), 0.832 (0.829-0.834), and 0.875 (0.873-0.878), respectively. Results were robust across diagnostic subgroups, but performance was lower in patients with a primary versus secondary diagnosis of cirrhosis.

CONCLUSION

Mortality analyses in patients with cirrhosis require sensitivity to the method of risk adjustment. Because different methods often produce divergent severity rankings, analyses of provider-specific outcomes may be biased depending on the method used.

A single health status question had important prognostic value among outpatients with chronic heart failure

OBJECTIVE

Health status is an important marker of the impact of disease on function among patients with chronic heart failure (CHF). However, the prognostic value of CHF-specific health status on long-term mortality has not been adequately evaluated. Our objective was to assess CHF-specific health status and 5-year mortality among outpatients with CHF.

STUDY DESIGN AND SETTING

We analyzed data from 494 Veterans Affairs outpatients with diagnoses of CHF and objective evidence of left ventricular dysfunction who enrolled in a quality improvement intervention. We extracted information about comorbid diagnoses, severity of illness (Charlson index), health care utilization, drug therapy, laboratory, and vital sign data along with generic and CHF-specific health status. We then identified multivariate correlates of subsequent mortality at 5 years.

RESULTS

Five-year mortality was 44%. Age (chi2=26.1, hazard ratio [HR]=1.63, confidence interval [CI]: 1.35, 1.97; P<0.0001) and Charlson index (chi2=12.9, HR=1.39, CI: 1.16, 1.67; P=0.0003) were significantly associated with 5-year mortality. Controlling for clinical, lab, medication, and administrative data, a single-item assessing change in CHF-specific health status was independently associated with 5-year mortality (chi2=11.4, HR=0.87, CI: 0.80, 0.94, P=0.0007).

CONCLUSIONS

Given the strength of the association with mortality, health care providers should routinely assess this single-item change in health status among outpatients with CHF to identify higher risk patients and guide therapy.

Validation of a Risk Score for Dying Within 1 Year of an Admission for Heart Failure

BACKGROUND

Development of heart failure greatly reduces life expectancy. Accurate estimates of the risk of dying are needed in clinical practice and for risk adjustment in observational studies. A relatively simple risk score has been developed to determine the risk of dying within 1-year of an admission for heart failure. We wanted to evaluate the risk score's predictive validity.

METHODS AND RESULTS

Data were abstracted from the electronic medical records of 769 patients admitted to the Minneapolis Veterans Administration medical center with a primary diagnosis of heart failure. Mortality within 1 year of admission was 25%. The c-index for the risk score was 0.71 (95% confidence interval 0.67-0.76). Similar to the original derivation cohort, mortality in risk score groups was 7% for a score lower than 60 (n = 44), 14% for 61 to 90 (n = 246), 26% for 91 to 120 (n = 222), 51% for 121 to 150 (n = 106), and 50% for scores greater than 150 (n = 8).

CONCLUSION

A previously developed risk score for 1-year mortality after an admission for heart failure provided a moderate degree of discrimination in a validation cohort from a different setting. Mortality in risk score groups was consistent with the original patient cohort. These results support the validity of the risk score and its application to a different patient population.

Variation by race in factors contributing to heart failure hospitalizations

BACKGROUND

Previous studies have shown a paradox of lower mortality in black compared with white patients after hospitalization for heart failure, in contrast to the overall higher mortality reported for nonhospitalized black patients with heart failure. We examined racial differences in factors contributing to hospitalization and in severity of illness in a cohort of black and white patients with heart failure who were hospitalized within a financially "equal access" health care system.

METHODS AND RESULTS

We performed a retrospective cohort study on 100 black or white male veterans admitted with heart failure to a VA Medical Center (black, n=52; white, n=48). Severity of illness as measured by the APACHE II score, a generic severity score, was similar between black and white patients (P=.72). However, using a recently developed heart failure-specific risk score, we found that white patients had higher severity of illness (P=.03). White patients had a higher number of coexisting comorbidities than black patients (P=.01), while black patients more frequently had uncontrolled hypertension at the time of admission (P=.004). Nonclinical factors contributing to hospitalization-such as nonadherence with medications or diet, inadequate outpatient follow-up, poor social support, and substance abuse-were documented more frequently for black patients compared with white patients.

CONCLUSIONS

At the time of hospitalization for heart failure, black patients may have an overall lower burden of disease and may more frequently have heart failure exacerbation precipitated by nonclinical factors. These findings may partly account for better long-term survival after hospitalization in black patients compared with white patients.

Common comorbidity scales were similar in their ability to predict health care costs and mortality

OBJECTIVE

To compare the ability of commonly used measures of medical comorbidity (ambulatory care groups [ACGs], Charlson comorbidity index, chronic disease score, number of prescribed medications, and number of chronic diseases) to predict mortality and health care costs over 1 year.

STUDY DESIGN AND SETTING

A prospective cohort study of community-dwelling older adults (n=3,496) attending a large primary care practice.

RESULTS

For predicting health care charges, the number of medications had the highest predictive validity (R(2)=13.6%) after adjusting for demographics. ACGs (R(2)=16.4%) and the number of medications (15.0%) had the highest predictive validity for predicting ambulatory visits. ACGs and the Charlson comorbidity index (area under the receiver operator characteristic [ROC] curve=0.695-0.767) performed better than medication-based measures (area under the ROC curve=0.662-0.679) for predicting mortality. There is relatively little difference, however, in the predictive validity across these scales.

CONCLUSION

In an outpatient setting, a simple count of medications may be the most efficient comorbidity measure for predicting utilization and health-care charges over the ensuing year. In contrast, diagnosis-based measures have greater predictive validity for 1-year mortality. Current comorbidity measures, however, have only poor to moderate predictive validity for costs or mortality over 1 year.

Diagnosis of heart failure in older adults: predictive value of dyspnea at rest

The objective of this study was to describe the predictive value of dyspnea at rest in the diagnosis of heart failure in older adults. We have conducted a retrospective chart review of older adults hospitalized with heart failure in 11 hospitals in Alabama, US. We defined dyspnea at rest as dyspnea not accompanied by activities, sleep, or lying down. We confirmed diagnosis of heart failure using modified Framingham criteria (based on available variables). We estimated values for sensitivity, specificity, predictive value positive and likelihood ratio positive of dyspnea at rest along with their 95% confidence intervals (95% CI). Patients had a mean (+/-S.D.) age 79 (+/-7.5) years, were 60% female and 18% African-American. The prevalence (pre-test probability) of heart failure was 77%. Dyspnea at rest was reported by 975 (89%) patients. Dyspnea on exertion, orthopnea and paroxysmal nocturnal dyspnea were, respectively, reported by 279 (26%), 356 (33%) and 230 (21%) patients. The sensitivity, specificity and positive predictive value (post-test probability) of dyspnea at rest were, respectively, 92% (95% CI = 90 - 94%), 19% (95% CI = 14 - 24%) and 79% (95% CI = 77 - 82%). Patients with dyspnea at rest were 13% (likelihood = 1.13; 95% CI = 1.06 - 1.20) more likely to have heart failure than those without. Presence of any three of the symptoms of dyspnea at rest, dyspnea on exertion, orthopnea, paroxysmal nocturnal dyspnea, fatigue and lower extremity edema increased the positive predictive value to 90% (95% CI = 87 - 93%). Dyspnea at rest was the commonest symptom among elderly hospitalized HF patients and had high sensitivity, but low specificity and positive predictive value, which increased in the presence of other symptoms.

Aligning quality and payment for heart failure care: Defining the challenges

Hospitals may not support programs that improve the quality of care delivered to heart failure patients because these programs lower readmission rates and empty beds, and therefore further diminish already-declining revenues. A conflict between the highest quality of care and financial solvency does not serve the interests of patients, physicians, hospitals, or payers. In principle, resolution of this conflict is simple: reimbursement systems should reward higher quality care. In practice, resolving the conflict is not simple. A recent roundtable discussion sponsored by the Heart Failure Society of America identified 4 major challenges to the design and implementation of reimbursement schemes that promote higher quality care for heart failure: defining quality, accounting for differences in disease severity, crafting novel payment mechanisms, and overcoming professional parochialism. This article describes each of these challenges in turn.

Do hospitals with lower mortality have higher patient satisfaction? A regional analysis of patients with medical diagnoses

Although patient satisfaction is a widely used indicator of quality, relationships between satisfaction and other indicators are poorly studied. The current study examined hospital-level correlations between patient satisfaction and severity-adjusted mortality for 29 hospitals in northeastern Ohio during 1993-1997. Satisfaction with 6 dimensions of care was measured using a validated survey of 42,255 randomly selected patients with medical diagnoses. Severity-adjusted mortality rates were determined for 200,562 consecutive patients with 6 high-volume medical diagnoses. Analyses found that satisfaction scores were inversely correlated with mortality rates. For the cumulative 5-year period, correlations were significant or of borderline significance for 5 of the 6 dimensions (coordination [R = -0.40; P = .03], discharge instructions [R = -0.39; P = .04], overall quality [R = -0.38; P = .04], information provided [R = -0.37; P = .05], and nursing [R = -0.35; P = .06]). The correlation was weakest for physician care (R = -0.07; P = .72). These findings indicate that hospitals with higher patient satisfaction also tended to have lower severity-adjusted mortality. Associations were strongest for dimensions of satisfaction measuring patient communication, coordination of care, and nursing care and weakest for physician care.

Can we be certain of PRO accuracy and accountability? Inaccurate "report cards" generated by medicare claims data as a marker for laboratory testing in diabetic patients

Quality of care attributed to individual physicians will be made public more frequently in the future. Inaccurate reports could damage physicians both professionally and financially. Using claims data for services rendered to Medicare patients from 1997 through 2000, The New Jersey Peer Review Organization (PRO) reported the percentage of each physician's patients who received testing for HgB A1c and for lipid profiles. We reviewed the patient records kept by these physicians to calculate the actual percentage of patients who received these tests. A statistically significant difference existed between the PRO's data and our results. Use of claims data as a marker for services provided can lead to inaccurate inferences about the quality of care provided by individual physicians, which can have potentially negative consequences to these providers.

Measuring health care quality: comparing and contrasting the medical and the marketing approaches

Health care quality, a key concept for medical practice and research, is also a widely used construct in health care administration and marketing research. We explored discipline-specific differences in the definition of quality, with the intent of finding a more unified approach. We summarized definitions and basic conceptual approaches to quality in both disciplines and then compared them on several attributes: basic goals, sources of measurement, role of patient perceptions, role of health care personnel, and need for risk adjustment. We developed a conceptual model combining the 2 approaches. Both disciplines could benefit from broadening their outcome measures. Patient satisfaction deserves more attention from medical researchers, whereas marketing approaches should go beyond using patient satisfaction as the only outcome of interest. It is conceptually feasible to integrate medical and marketing approaches to quality, with important insights resulting from this integration.

The impact of statistical adjustment on economic profiles of interventional cardiologists

OBJECTIVES

The objective of this study was to identify preprocedure patient factors associated with percutaneous intervention costs and to examine the impact of these patient factors on economic profiles of interventional cardiologists.

BACKGROUND

There is increasing demand for information about comparative resource use patterns of interventional cardiologists. Economic provider profiles, however, often fail to account for patient characteristics.

METHODS

Data were obtained from Duke Medical Center cost and clinical information systems for 1,949 procedures performed by 13 providers between July 1, 1997, and December 31, 1998. Patient factors that influenced cost were identified using multiple regression analysis. After assessing interprovider variation in unadjusted cost, mixed linear models were used to examine how much cost variability was associated with the provider when patient characteristics were taken into account.

RESULTS

Total hospital costs averaged $15,643 (median, $13,809), $6,515 of which represented catheterization laboratory costs. Disease severity, acuity, comorbid illness and lesion type influenced total costs (R(2) = 38%), whereas catheterization costs were affected by lesion type and acuity (R(2) = 32%). Patient characteristics varied significantly among providers. Unadjusted total costs were weakly associated with provider, and this association disappeared after accounting for patient factors. The provider influence on catheterization costs persisted after adjusting for patient characteristics. Furthermore, the pattern of variation changed: the adjusted analysis identified three new outliers, and two providers lost their outlier status. Only one provider was consistently identified as an outlier in the unadjusted and adjusted analyses.

CONCLUSIONS

Economic profiles of interventional cardiologists may be misleading if they do not adequately adjust for patient characteristics before procedure.

Length of stay as a source of bias in comparing performance in VA and private sector facilities: lessons learned from a regional evaluation of intensive care outcomes

OBJECTIVES

Compare intensive care unit (ICU) mortality and length of stay (LOS) in a VA hospital and private sector hospitals and examine the impact of hospital utilization on mortality comparisons.

RESEARCH DESIGN

Retrospective cohort study.

SUBJECTS

Consecutive ICU admissions to a VA hospital (n = 1,142) and 27 private sector hospitals (n = 51,249) serving the same health care market in 1994 to 1995.

MEASURES

Mortality and ICU LOS were adjusted for severity of illness using a validated method that considers physiologic data from the first 24 hours of ICU admission. Mortality comparisons were made using two different multivariable techniques.

RESULTS

Unadjusted in-hospital mortality was higher in VA patients (14.5% vs. 12.0%; P = 0.01), as was hospital (28.3 vs. 11.3 days; P <0.001) and ICU (4.3 vs. 3.9 days; P <0.001) LOS. Using logistic regression to adjust for severity, the odds of death was similar in VA patients, relative to private sector patients (OR 1.16, 95% CI 0.93-1.44; P = 0.18). However, a higher proportion of VA deaths occurred after 21 hospital days (33% vs. 13%; P <0.001). Using proportional hazards regression and censoring patients at hospital discharge, the risk for death was lower in VA patients (hazard ratio 0.70; 95% CI 0.59-0.82; P <0.001). After adjusting for severity, differences in ICU LOS were no longer significant (P = 0.19).

CONCLUSIONS

Severity-adjusted mortality in ICU patients was lower in a VA hospital than in private sector hospitals in the same health care market, based on proportional hazards regression. This finding differed from logistic regression analysis, in which mortality was similar, suggesting that comparisons of hospital mortality between systems with different hospital utilization patterns may be biased if LOS is not considered. If generalizable to other markets, our findings further suggest that ICU outcomes are at least similar in VA hospitals.

Risk-adjusted surgical outcomes

Measures of risk-adjusted outcome are particularly suited for the assessment of the quality of surgical care. The reliability of measures of quality that use surgical outcomes is enhanced by prospective data acquisition and should be adjusted for the preoperative severity of illness. Such measures should be based only on reliable and validated data, and they should apply state-of-the-art analytical methods. The risk-adjusted postoperative mortality rate is useful as a quality measure only in specialties and operations expected to have a high rate of postoperative deaths. Risk-adjusted complications are more common but are limited as a comparative measure of quality by a lack of uniform definitions and data collection mechanisms. In specialties in which the expected postoperative mortality is low, risk-adjusted functional outcomes are promising measures for the assessment of the quality of surgical care. Measures of cost and patient satisfaction should also be incorporated in systems designed to measure the quality and cost-effectiveness of surgical care.