Association between physician specialty and volumes of treated patients and mortality among patients hospitalized for newly diagnosed heart failure

Physician-level variation in clinical outcomes and resource use in inpatient general internal medicine: an observational study

Background

Variations in inpatient medical care are typically attributed to system, hospital or patient factors. Little is known about variations at the physician level within hospitals. We described the physician-level variation in clinical outcomes and resource use in general internal medicine (GIM).

Methods

This was an observational study of all emergency admissions to GIM at seven hospitals in Ontario, Canada, over a 5-year period between 2010 and 2015. Physician-level variations in inpatient mortality, hospital length of stay, 30-day readmission and use of ‘advanced imaging’ (CT, MRI or ultrasound scans) were measured. Physicians were categorised into quartiles within each hospital for each outcome and then quartiles were pooled across all hospitals (eg, physicians in the highest quartile at each hospital were grouped together). We report absolute differences between physicians in the highest and lowest quartiles after matching admissions based on propensity scores to account for patient-level variation.

Results

The sample included 103 085 admissions to 135 attending physicians. After propensity score matching, the difference between physicians in the highest and lowest quartiles for in-hospital mortality was 2.4% (95% CI 0.6% to 4.3%, p<0.01); for readmission was 3.3% (95% CI 0.7% to 5.9%, p<0.01); for advanced imaging was 0.32 tests per admission (95% CI 0.12 to 0.52, p<0.01); and for hospital length of stay was 1.2 additional days per admission (95% CI 0.5 to 1.9, p<0.01). Physician-level differences in length of stay and imaging use were consistent across numerous sensitivity analyses and stable over time. Differences in mortality and readmission were consistent across most sensitivity analyses but were not stable over time and estimates were limited by sample size.

Conclusions

Patient outcomes and resource use in inpatient medical care varied substantially across physicians in this study. Physician-level variations in length of stay and imaging use were unlikely to be explained by patient factors whereas differences in mortality and readmission should be interpreted with caution and could be explained by unmeasured confounders. Physician-level variations may represent practice differences that highlight quality improvement opportunities.

Appropriateness of peripherally inserted central catheter use among general medical inpatients: an observational study using routinely collected data

Background

Peripherally inserted central catheters (PICC) are among the most commonly used medical devices in hospital. This study sought to determine the appropriateness of inpatient PICC use in general medicine at five academic hospitals in Toronto, Ontario, Canada, based on the Michigan Appropriateness Guide for Intravenous Catheters (MAGIC).

Methods

This was a retrospective, cross-sectional study of general internal medicine patients discharged between 1 April 2010 and 31 March 2015 who received a PICC during hospitalisation. The primary outcomes were the proportions of appropriate and inappropriate inpatient PICC use based on MAGIC recommendations. Hospital administrative data and electronic clinical data were used to determine appropriateness of each PICC placement. Multivariable regression models were fit to explore patient predictors of inappropriate use.

Results

Among 3479 PICC placements, 1848 (53%, 95% CI 51% to 55%) were appropriate, 573 (16%, 95% CI 15% to 18%) were inappropriate and 1058 (30%, 95% CI 29% to 32%) were of uncertain appropriateness. The proportion of appropriate and inappropriate PICCs ranged from 44% to 61% (p<0.001) and 13% to 21% (p<0.001) across hospitals, respectively. The most common reasons for inappropriate PICC use were placement in patients with advanced chronic kidney disease (n=500, 14%) and use for fewer than 15 days in patients who are critically ill (n=53), which represented 14% of all PICC placements in the intensive care unit. Patients who were older, female, had a Charlson Comorbidity Index score greater than 0 and more severe illness based on the Laboratory-based Acute Physiology Score were more likely to receive an inappropriate PICC.

Conclusions

Clinical practice recommendations can be operationalised into measurable domains to estimate the appropriateness of PICC insertions using routinely collected hospital data. Inappropriate PICC use was common and varied substantially across hospitals in this study, suggesting that there are important opportunities to improve care.

Do outcomes for patients with heart failure vary by emergency department volume?

BACKGROUND

Heart failure is a common Emergency Department (ED) presentation but whether ED volume influences patient outcomes is unknown.

METHODS AND RESULTS

Retrospective cohort of all adults presenting to 93 EDs between 1999 and 2009 with a most responsible diagnosis of heart failure (n=44 925 ED visits; mean age, 76.4 years). Cases seen in low-volume EDs had less comorbidities and were less likely to be hospitalized (54.5%) than those seen in medium (61.8%; adjusted odds ratio [aOR] 1.16, [95% confidence interval {CI} 1.10-1.23]) or high-volume EDs (73.6%; aOR, 1.95 [95% CI, 1.83-2.07]). Of patients treated and released, low-volume ED cases exhibited higher risk of death/hospitalization/ED visit in the subsequent 7 (22.0%) and 30 days (44.9%) than medium (16.3%; aOR, 0.81 [95% CI, 0.73-0.90], and 35.3%; aOR, 0.79 [95% CI, 0.73-0.86]) or high-volume ED cases (13.0%; aOR, 0.69 [95% CI, 0.61-0.78], and 30.2%; aOR, 0.67 [95% CI, 0.61-0.74]). Of patients hospitalized at the time of their index ED visit, low-volume ED cases exhibited a higher risk of 30-day death/all-cause readmission (24.3%) than those seen in medium (21.9%; aOR, 0.83 [95% CI, 0.76-0.91]) or high-volume EDs (18.1%; aOR, 0.77 [95% CI, 0.70-0.85]).

CONCLUSIONS

Low-volume EDs were more likely to discharge patients with heart failure home, but low-volume ED cases exhibited worse outcomes (driven largely by readmissions or repeat ED visits). Interventions to improve management of acute heart failure are required at low-volume sites.

Improved outcomes with early collaborative care of ambulatory heart failure patients discharged from the emergency department

BACKGROUND

The type of outpatient physician care after an emergency department visit for heart failure may affect patients' outcomes.

METHODS AND RESULTS

Using the National Ambulatory Care Reporting System, we examined the care and outcomes of heart failure patients who visited and were discharged from the emergency department in Ontario, Canada (April 2004 to March 2007). Early collaborative care by a cardiologist and primary care (PC) physician within 30 days after discharge was compared with PC alone. Care for 10 599 patients (age, 74.9±11.9 years; 50.2% male) was provided by PC alone (n=6596), cardiologist alone (n=535), or concurrently by both cardiologist and PC (n=1478); 1990 did not visit a physician. Collaborative care patients were more likely to undergo assessment of left ventricular function (57.4% versus 28.7%), noninvasive stress testing (20.1% versus 7.8%), and cardiac catheterization (11.6% versus 2.7%) compared with PC. Drug prescriptions (patients ≥65 years of age) demonstrated higher use of angiotensin-converting enzyme inhibitors (58.8% versus 54.6%), angiotensin receptor blockers (22.7% versus 18.1%), β-adrenoceptor antagonists (63.4% versus 48.0%), loop diuretics (84.2% versus 79.6%), metolazone (4.8% versus 3.4%), and spironolactone (19.8% versus 12.7%) within 100 days after emergency department discharge for collaborative care compared with PC. In a propensity-matched model, mortality was lower with PC compared with no physician visit (hazard ratio, 0.75; 95% confidence interval, 0.64 to 0.87; P<0.001). Collaborative care reduced mortality compared with PC (hazard ratio, 0.79; 95% confidence interval, 0.63 to 1.00; P=0.045). Sole cardiology care conferred a trend to increased mortality (hazard ratio, 1.41 versus collaborative care; 95% confidence interval, 0.98 to 2.03; P=0.067).

CONCLUSIONS

Early collaborative heart failure care was associated with increased use of drug therapies and cardiovascular diagnostic tests and better outcomes compared with PC alone.

Validation of coding algorithms for the identification of patients with primary biliary cirrhosis using administrative data

BACKGROUND

Large-scale epidemiological studies of primary biliary cirrhosis (PBC) have been hindered by difficulties in case ascertainment.

OBJECTIVE

To develop coding algorithms for identifying PBC patients using administrative data--a widely available data source.

METHODS

Population-based administrative databases were used to identify patients with a diagnosis code for PBC from 1994 to 2002. Coding algorithms for confirmed PBC (two or more of antimitochondrial antibody positivity, cholestatic liver biochemistry and/or compatible liver histology) were derived using chart abstraction data as the reference. Patients with a recorded PBC diagnosis but insufficient confirmatory data were classified as 'suspected PBC'.

RESULTS

Of 189 potential PBC cases, 119 (60%) had confirmed PBC and 28 (14%) had suspected PBC. The optimal algorithm including two or more uses of a PBC code had a sensitivity of 94% (95% CI 71% to 100%) and positive predictive values of 73% (95% CI 61% to 75%) for confirmed PBC, and 89% (95% CI 82% to 94%) for confirmed or suspected PBC. Sensitivity analyses revealed greater accuracy among women, and with the use of multiple data sources and one or more years of data. Inclusion of diagnosis codes for conditions frequently misclassified as PBC did not improve algorithm performance.

CONCLUSIONS

Administrative databases can reliably identify patients with PBC and may facilitate epidemiological investigations of this condition.

Validation of ICD-9-CM/ICD-10 coding algorithms for the identification of patients with acetaminophen overdose and hepatotoxicity using administrative data

Background

Acetaminophen overdose is the most common cause of acute liver failure (ALF). Our objective was to develop coding algorithms using administrative data for identifying patients with acetaminophen overdose and hepatic complications.

Methods

Patients hospitalized for acetaminophen overdose were identified using population-based administrative data (1995–2004). Coding algorithms for acetaminophen overdose, hepatotoxicity (alanine aminotransferase >1,000 U/L) and ALF (encephalopathy and international normalized ratio >1.5) were derived using chart abstraction data as the reference and logistic regression analyses.

Results

Of 1,776 potential acetaminophen overdose cases, the charts of 181 patients were reviewed; 139 (77%) had confirmed acetaminophen overdose. An algorithm including codes 965.4 (ICD-9-CM) and T39.1 (ICD-10) was highly accurate (sensitivity 90% [95% confidence interval 84–94%], specificity 83% [69–93%], positive predictive value 95% [89–98%], negative predictive value 71% [57–83%], c-statistic 0.87 [0.80–0.93]). Algorithms for hepatotoxicity (including codes for hepatic necrosis, toxic hepatitis and encephalopathy) and ALF (hepatic necrosis and encephalopathy) were also highly predictive (c-statistics = 0.88). The accuracy of the algorithms was not affected by age, gender, or ICD coding system, but the acetaminophen overdose algorithm varied between hospitals (c-statistics 0.84–0.98; P = 0.003).

Conclusion

Administrative databases can be used to identify patients with acetaminophen overdose and hepatic complications. If externally validated, these algorithms will facilitate investigations of the epidemiology and outcomes of acetaminophen overdose.