Development of a disease-based hospital-level diagnostic intensity index

OBJECTIVES

Low-value care is associated with increased healthcare costs and direct harm to patients. We sought to develop and validate a simple diagnostic intensity index (DII) to quantify hospital-level diagnostic intensity, defined by the prevalence of advanced imaging among patients with selected clinical diagnoses that may not require imaging, and to describe hospital characteristics associated with high diagnostic intensity.

METHODS

We utilized State Inpatient Database data for inpatient hospitalizations with one or more pre-defined discharge diagnoses at acute care hospitals. We measured receipt of advanced imaging for an associated diagnosis. Candidate metrics were defined by the proportion of inpatients at a hospital with a given diagnosis who underwent associated imaging. Candidate metrics exhibiting temporal stability and internal consistency were included in the final DII. Hospitals were stratified according to the DII, and the relationship between hospital characteristics and DII score was described. Multilevel regression was used to externally validate the index using pre-specified Medicare county-level cost measures, a Dartmouth Atlas measure, and a previously developed hospital-level utilization index.

RESULTS

This novel DII, comprised of eight metrics, correlated in a dose-dependent fashion with four of these five measures. The strongest relationship was with imaging costs (odds ratio of 3.41 of being in a higher DII tertile when comparing tertiles three and one of imaging costs (95 % CI 2.02-5.75)).

CONCLUSIONS

A small set of medical conditions and related imaging can be used to draw meaningful inferences more broadly on hospital diagnostic intensity. This could be used to better understand hospital characteristics associated with low-value care.

National Improvement of Waiting Times: First Results From the Dutch Head and Neck Audit

OBJECTIVE

Timely treatment initiation in head and neck cancer (HNC) care is of great importance regarding survival, oncological, functional, and psychological outcomes. Therefore, waiting times are assessed in the Dutch Head and Neck Audit (DHNA). This audit aims to assess and improve the quality of care through feedback and benchmarking. For this study, we examined how waiting times evolved since the start of the DHNA.

STUDY DESIGN

Prospective cohort study.

SETTING

National multicentre study.

METHODS

The DHNA was established in 2014 and reached national coverage of all patients treated for primary HNC in 2019. DHNA data on curative patients from 2015 to 2021 was extracted on national (benchmark) and hospital level. We determined 3 measures for waiting time: (1) the care pathway interval (CPI, first visit to start treatment), (2) the time to treatment interval (TTI, biopsy to start treatment), and (3) CPI-/TTI-indicators (percentage of patients starting treatment ≤30 days). The Dutch national quality norm for the CPI-indicator is 80%.

RESULTS

The benchmark median CPI and TTI improved between 2015 and 2021 from 37 to 26 days and 37 to 33 days, respectively. Correspondingly, the CPI- and TTI-indicators, respectively, increased from 39% to 64% and 35% to 40% in 2015 to 2021. Outcomes for all hospitals improved and dispersion between hospitals declined. Four hospitals exceeded the 80% quality norm in 2021.

CONCLUSION

Waiting times improved gradually over time, with 4 hospitals exceeding the quality standard in 2021. On the hospital-level, process improvement plans have been initiated. Systematic registration, auditing, and feedback of data support the improvement of quality of care.

Hospital Variation in Skilled Nursing Facility Use After Coronary Artery Bypass Graft Surgery

BACKGROUND

Over 20% of patients are discharged to a skilled nursing facility (SNF) after coronary artery bypass graft surgery, but little is known about specific drivers for postdischarge SNF use. The purpose of this study was to evaluate hospital variation in SNF use and its association with postoperative outcomes after coronary artery bypass graft.

METHODS AND RESULTS

A retrospective study design utilizing Medicare Provider Analysis and Review files was used to evaluate SNF use among 70 509 beneficiaries undergoing coronary artery bypass graft, with or without valve procedures, between 2016 and 2018. A total of 17 328 (24.6%) were discharged to a SNF, ranging from 0% to 88% across 871 hospitals. Multilevel logistic regression models identified significant patient-level predictors of discharge to SNF including increasing age, comorbidities, female sex, Black race, dual eligibility, and postoperative complications. After adjusting for patient and hospital factors, 15.6% of the variation in hospital SNF use was attributed to the discharging hospital. Compared with the lower quartile of hospital SNF use, hospitals in the top quartile of SNF use had lower risk-adjusted 1-year mortality (12.5% versus 8.6%, P<0.001) and readmission (59.9% versus 49.8%, P<0.001) rates for patients discharged to a SNF.

CONCLUSIONS

There is high variability in SNF use among hospitals that is only partially explained by patient characteristics. Hospitals with higher SNF utilization had lower risk-adjusted 1-year mortality and readmission rates for patients discharged to a SNF. More work is needed to better understand underlying provider and hospital-level factors contributing to SNF use variability.

Hospital-level variation in racial disparities in low-risk nulliparous cesarean delivery rates

BACKGROUND

Nationally, rates of cesarean delivery are highest among Black patients compared with other racial/ethnic groups. These observed inequities are a relatively new phenomenon (in the 1980s, cesarean delivery rates among Black patients were lower than average), indicating an opportunity to narrow the gap. Cesarean delivery rates vary greatly among hospitals, masking racial disparities that are unseen when rates are reported in aggregate.

OBJECTIVE

This study aimed to explore reasons for the current large Black-White disparity in first-birth cesarean delivery rates by first examining the hospital-level variation in first-birth cesarean delivery rates among different racial/ethnic groups. We then identified hospitals that had low first-birth cesarean delivery rates among Black patients and compared them with hospitals with high rates. We sought to identify differences in facility or patient characteristics that could explain the racial disparity.

STUDY DESIGN

A population cross-sectional study was performed on 1,267,493 California live births from 2018 through 2020 using birth certificate data linked with maternal patient discharge records. Annual nulliparous term singleton vertex cesarean delivery (first-birth) rates were calculated for the most common racial/ethnic groups statewide and for each hospital. Self-identified race/ethnicity categories as selected on the birth certificate were used. Relative risk and 95% confidence intervals for first-birth cesarean delivery comparing 2019 with 2015 were estimated using a log-binomial model for each racial/ethnic group. Patient and hospital characteristics were compared between hospitals with first-birth cesarean delivery rates <23.9% for Black patients and hospitals with rates ≥23.9% for Black patients.

RESULTS

Hospitals with at least 30 nulliparous term singleton vertex Asian, Black, Hispanic, and White patients each were identified. Black patients had a very different distribution, with a significantly higher rate (28.4%) and wider standard deviation (7.1) and interquartile range (6.5) than other racial groups (P<.01). A total of 29 hospitals with a low first-birth cesarean delivery rate among Black patients were identified using the Healthy People 2020 target of 23.9% and compared with 106 hospitals with higher rates. The low-rate group had a cesarean delivery rate of 19.9%, as opposed to 30.7% in the higher-rate group. There were no significant differences between the groups in hospital characteristics (ownership, delivery volume, neonatal level of care, proportion of midwife deliveries) or patient characteristics (age, education, insurance, onset of prenatal care, body mass index, hypertension, diabetes mellitus). Among the 106 hospitals that did not meet the target for Black patients, 63 met it for White patients with a mean rate of 21.4%. In the same hospitals, the mean rate for Black patients was 29.5%. Among Black patients in the group that did not meet the 23.9% target, there were significantly higher rates of all cesarean delivery indications: labor dystocia, fetal concern (spontaneous labor), and no labor (eg, macrosomia), which are all indications with a high degree of subjectivity.

CONCLUSION

The statewide cesarean delivery rate of Black patients is significantly higher and has substantially greater hospital variation compared with other racial or ethnic groups. The lack of difference in facility or patient characteristics between hospitals with low cesarean delivery rates among Black patients and those with high rates suggests that unconscious bias and structural racism potentially play important roles in creating these racial differences.

Among-Hospital Variation in Intensive Care Unit Admission Practices and Associated Outcomes for Patients with Acute Respiratory Failure

Rationale: We have previously shown that hospital strain is associated with intensive care unit (ICU) admission and that ICU admission, compared with ward admission, may benefit certain patients with acute respiratory failure (ARF). Objectives: To understand how strain-process-outcomes relationships in patients with ARF may vary among hospitals and what hospital practice differences may account for such variation. Methods: We examined high-acuity patients with ARF who did not require mechanical ventilation or vasopressors in the emergency department (ED) and were admitted to 27 U.S. hospitals from 2013 to 2018. Stratifying by hospital, we compared hospital strain-ICU admission relationships and hospital length of stay (LOS) and mortality among patients initially admitted to the ICU versus the ward using hospital strain as a previously validated instrumental variable. We also surveyed hospital practices and, in exploratory analyses, evaluated their associations with the above processes and outcomes. Results: There was significant among-hospital variation in ICU admission rates, in hospital strain-ICU admission relationships, and in the association of ICU admission with hospital LOS and hospital mortality. Overall, ED patients with ARF (n = 45,339) experienced a 0.82-day shorter median hospital LOS if admitted initially to the ICU compared with the ward, but among the 27 hospitals (n = 224-3,324), this effect varied from 5.85 days shorter (95% confidence interval [CI], -8.84 to -2.86; P < 0.001) to 4.38 days longer (95% CI, 1.86-6.90; P = 0.001). Corresponding ranges for in-hospital mortality with ICU compared with ward admission revealed odds ratios from 0.08 (95% CI, 0.01-0.56; P < 0.007) to 8.89 (95% CI, 1.60-79.85; P = 0.016) among patients with ARF (pooled odds ratio, 0.75). In exploratory analyses, only a small number of measured hospital practices-the presence of a sepsis ED disposition guideline and maximum ED patient capacity-were potentially associated with hospital strain-ICU admission relationships. Conclusions: Hospitals vary considerably in ICU admission rates, the sensitivity of those rates to hospital capacity strain, and the benefits of ICU admission for patients with ARF not requiring life support therapies in the ED. Future work is needed to more fully identify hospital-level factors contributing to these relationships.

Variation in hospital utilization of palliative interventions for patients with advanced gastrointestinal cancer near end of life

BACKGROUND

Patients with advanced gastrointestinal (GI) cancer often undergo noncurative interventions with palliative intent to relieve high symptom burden near end of life. Hospital-level variation in intervention utilization remains unclear.

METHODS

National cohort study of 142 304 patients with stage III or IV GI cancer within the National Cancer Database (2004-2014) who died within 1-year of diagnosis. Hospitals were stratified by palliative intervention utilization (surgery, chemotherapy, radiation, pain management). Multivariable, multinomial regression evaluated the association between patient/hospital factors and palliative intervention utilization.

RESULTS

Across 1322 hospitals, median hospital palliative intervention utilization was 12.0% [interquartile range: 0.0%-26.1%]. Utilization increased over time in all but lowest utilizing hospitals. Relative to lowest utilizing hospitals, factors associated with a lower likelihood of care at highest utilizing hospitals included: race (White [ref]; Black-Relative Risk Ratio [RRR] 0.81, 95% confidence interval [0.77-0.85]) and lower income (RRR 0.81 [0.78-0.84]). Factors associated with a higher likelihood included: lower education level (RRR 1.62 [1.55-1.69]) and hospital type (community program [ref]; comprehensive community-RRR 1.33 [1.26-1.41]; academic-RRR 1.88 [1.77-1.99]; integrated network-RRR 1.79 [1.66-1.93]).

CONCLUSION

Hospital variation in palliative intervention use is substantial and potentially associated with sociodemographic and hospital characteristics. Future work can examine how differences in hospital care processes translate to quantity/quality of life for cancer patients.

Hospital Strain and Variation in Sepsis ICU Admission Practices and Associated Outcomes

To understand how strain-process-outcome relationships in patients with sepsis may vary among hospitals.

DESIGN

Retrospective cohort study using a validated hospital capacity strain index as a within-hospital instrumental variable governing ICU versus ward admission, stratified by hospital.

SETTING

Twenty-seven U.S. hospitals from 2013 to 2018.

PATIENTS

High-acuity emergency department patients with sepsis who do not require life support therapies.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The mean predicted probability of ICU admission across strain deciles ranged from 4.9% (lowest ICU-utilizing hospital for sepsis without life support) to 61.2% (highest ICU-utilizing hospital for sepsis without life support). The difference in the predicted probabilities of ICU admission between the lowest and highest strain deciles ranged from 9.0% (least strain-sensitive hospital) to 45.2% (most strain-sensitive hospital). In pooled analyses, emergency department patients with sepsis (n = 90,150) experienced a 1.3-day longer median hospital length of stay (LOS) if admitted initially to the ICU compared with the ward, but across the 27 study hospitals (n = 517-6,564), this effect varied from 9.0 days shorter (95% CI, -10.8 to -7.2; p < 0.001) to 19.0 days longer (95% CI, 16.7-21.3; p < 0.001). Corresponding ranges for inhospital mortality with ICU compared with ward admission revealed odds ratios (ORs) from 0.16 (95% CI, 0.03-0.99; p = 0.04) to 4.62 (95% CI, 1.16-18.22; p = 0.02) among patients with sepsis (pooled OR = 1.48).

CONCLUSIONS

There is significant among-hospital variation in ICU admission rates for patients with sepsis not requiring life support therapies, how sensitive those ICU admission decisions are to hospital capacity strain, and the association of ICU admission with hospital LOS and hospital mortality. Hospital-level heterogeneity should be considered alongside patient-level heterogeneity in critical and acute care study design and interpretation.

Hospital Variation in Mortality and Ventilator Management among Mechanically Ventilated Patients with ARDS

RATIONALE

Acute Respiratory Distress Syndrome (ARDS) is associated with significant mortality. Despite the mortality benefits of lung protective ventilation, adherence rates to evidence-based ventilator practice have remained low and ARDS mortality has remained high.

OBJECTIVE

Determine variation in ARDS mortality and adherence to low tidal volume ventilation (LTV) across US hospitals.

MATERIALS AND METHODS

We identified mechanically ventilated patients with ARDS using data from Philips eICU (2014-2015). We then used multi-variable hierarchical logistic regression models with hospital site as the random effect and patient and hospital level factors as fixed effects to assess the hospital risk adjusted mortality rate and median odds ratio for the association between mortality and hospital site. We then assessed associations between adherence to LTV (defined as 4-8 mL/kg PBW) and hospital risk adjusted mortality rates using Spearman correlation.

RESULTS

Among 4441 patients admitted at 110 hospitals with ARDS, the hospital risk-adjusted mortality rate ranged from 19% to 39%, and the MOR for hospital of admission was 1.33 (95% CI 1.25-1.41). Among 3070 patients at 72 hospitals with available ventilator data, 73% of patients had a median set Vt between 4 to 8 mL/kg PBW; hospital adherence rates to LTV ranged from 13% to 95%. There was no association between hospital adherence to LTV and risk-adjusted mortality rate (spearman correlation coefficient -0.01, p = .93). Similarly, among 956 patients who started with a Vt > 8 mL/kg PBW, there was no association between the percent of patients at each hospital whose Vt was decreased to ≤ 8 mL/kg PBW and risk adjusted mortality rate (spearman correlation coefficient .05, p = .73).

CONCLUSION

Risk adjusted mortality and use of LTV for patients with ARDS varied widely across hospitals. However, hospital adherence to LTV was not associated with ARDS mortality rates. Further evaluation of hospital practices associated with lower ARDS mortality are warranted.

Hospital-Level Variation in Ticagrelor Use in Patients With Acute Coronary Syndrome

Background

Despite improved outcomes associated with ticagrelor compared with clopidogrel in acute coronary syndrome (ACS), many studies have demonstrated slow adoption of ticagrelor in the United States because of its increased cost. Less is known about how ticagrelor is adopted when there is no added cost consideration. Our objectives were to determine patterns of use of ticagrelor, hospital‐level adoption of ticagrelor use, and factors associated with its use after ACS in a publicly funded health care system.

Methods and Results

We conducted a population‐based cohort study including patients (≥65 years) hospitalized with their first ACS from April 2014 to March 2018 in Ontario, Canada. We determined temporal trends in ticagrelor use and hospital‐level adoption of its use post‐ACS discharge. Using hierarchical regression models, we identified significant predictors of ticagrelor use. There were 23 962 patients with ACS (mean age 76.3 years, 59.7% men) hospitalized in 156 hospitals. Overall ticagrelor use increased from 32.6% in 2014/2015 to 51.8% in 2017/2018. There was substantial variation in ticagrelor use post‐ACS across hospitals, with hospital‐specific prescribing rates ranging from 0% to 83.6%. Lower odds of ticagrelor use was associated with advanced age and the presence of comorbidities. Besides patient factors, being admitted to a rurally located hospital more than halved the odds of being prescribed ticagrelor (odds ratio [OR], 0.49; 95% CI, 0.32–0.77). Being managed by a cardiologist during the index ACS hospitalization was associated with higher odds of having a ticagrelor prescription after ACS (OR, 2.80; 95% CI, 2.36–3.33).

Conclusions

Ticagrelor use rates varied substantially across hospitals and were strongly associated with physician and hospital factors independent of patient characteristics.

Neonatal Intensive Care Utilization and Postdischarge Newborn Outcomes: A Population-based Study of Texas Medicaid Insured Infants

OBJECTIVE

To test the hypothesis that newborn infants cared for in hospitals with greater utilization of neonatal intensive care experienced fewer postdischarge adverse events.

STUDY DESIGN

We developed 3 retrospective population-based cohorts of Texas Medicaid insured singletons born in 2010-2014 (very low birth weight [VLBW n = 11 139], late preterm [n = 57 509], and non-preterm [n = 664 447]) who received care in higher volume hospitals with level III/IV neonatal intensive care units (NICUs). Measures of NICU care were hospital-level risk adjusted NICU admission rates, special care days (days of nonroutine care) per infant, and the percent of intensive (highest billable care code) special care days. The units of analysis were hospitals (n = 80) and the primary outcome was an adverse event, (defined as admission, emergency department visit, or death) within 30 days postdischarge.

RESULTS

Higher use of NICU care at a hospital level was not associated with lower postdischarge 30-day adverse event. Infants cared for in hospitals with above vs below median special care day rates experienced slightly higher postdischarge adverse event per 100 infants (VLBW: 14.01 [95% CI 12.74-15.27] vs 11.84 [10.52-13.16], P < .05; late preterm: 7.33 [6.68-7.97] vs 6.28 [5.87-6.69], P < .01; non-preterm: 4.47 [4.17-4.76] vs 3.97 [3.75-4.18], P < .01). Weak positive associations (Pearson correlations of 0.31-0.37, P < .01) were observed for adverse event with special care days; in no instance was a negative association observed between NICU utilization and adverse event.

CONCLUSION

Higher utilization of NICU care was not associated with lower rates of short-term events suggesting that there may be opportunities to safely decrease admission rates and length of NICU stays.

Hospital variation in responses to safety warnings about power morcellation in hysterectomy

BACKGROUND

Safety warnings about power morcellation in 2014 considerably changed hysterectomy practice, especially for laparoscopic supracervical hysterectomy that typically requires morcellation to remove the corpus uteri while preserving the cervix. Hospitals might vary in how they respond to safety warnings and altered hysterectomy procedures to avoid use of power morcellation. However, there has been little data on how hospitals differ in their practice changes.

OBJECTIVE

This study aimed to examine whether hospitals varied in their use of laparoscopic supracervical hysterectomy after safety warnings about power morcellation and compare the risk of surgical complications at hospitals that had different response trajectories in use of laparoscopic supracervical hysterectomy.

STUDY DESIGN

This was a retrospective analysis of data from the New York Statewide Planning and Research Cooperative System and the State Inpatient Databases and State Ambulatory Surgery and Services Databases from 14 other states. We identified women aged ≥18 years undergoing hysterectomy for benign indications in the hospital inpatient and outpatient settings from October 1, 2013 to September 30, 2015. We calculated a risk-adjusted utilization rate of laparoscopic supracervical hysterectomy for each hospital in each calendar quarter after accounting for patient clinical risk factors. Applying a growth mixture modeling approach, we identified distinct groups of hospitals that exhibited different trajectories of using laparoscopic supracervical hysterectomy over time. Within each trajectory group, we compared patients' risk of surgical complications in the prewarning (2013Q4-2014Q1), transition (2014Q2-2014Q4), and postwarning (2015Q1-2015Q3) period using multivariable regressions.

RESULTS

Among 212,146 women undergoing benign hysterectomy at 511 hospitals, the use of laparoscopic supracervical hysterectomy decreased from 15.1% in 2013Q4 to 6.2% in 2015Q3. The use of laparoscopic supracervical hysterectomy at these 511 hospitals exhibited 4 distinct trajectory patterns: persistent low use (mean risk-adjusted utilization rate of laparoscopic supracervical hysterectomy changed from 2.8% in 2013Q4 to 0.6% in 2015Q3), decreased medium use (17.0% to 6.9%), decreased high use (51.4% to 24.2%), and rapid abandonment (30.5% to 0.8%). In the meantime, use of open abdominal hysterectomy increased by 2.1, 4.1, 7.8, and 11.8 percentage points between the prewarning and postwarning periods in these 4 trajectory groups, respectively. Compared with the prewarning period, the risk of major complications in the postwarning period decreased among patients at "persistent low use" hospitals (adjusted odds ratio, 0.88; 95% confidence interval, 0.81-0.94). In contrast, the risk of major complications increased among patients at "rapid abandonment" hospitals (adjusted odds ratio, 1.48; 95% confidence interval, 1.11-1.98), and the risk of minor complications increased among patients at "decreased high use" hospitals (adjusted odds ratio, 1.31; 95% confidence interval, 1.01-1.72).

CONCLUSION

Hospitals varied in their use of laparoscopic supracervical hysterectomy after safety warnings about power morcellation. Complication risk increased at hospitals that shifted considerably toward open abdominal hysterectomy.

Hospital Variation in 30-Day Readmissions Following Transcatheter Aortic Valve Replacement

Background Data on hospital variation in 30-day readmission rates after transcatheter aortic valve replacement (TAVR) are limited. Further, whether such variation is explained by differences in hospital characteristics and hospital practice patterns remains unknown. Methods and Results We used the 2017 Nationwide Readmissions Database to identify hospitals that performed at least 5 TAVRs. Hierarchical logistic regression models were used to examine between-hospital variation in 30-day all-cause risk-standardized readmission rate (RSRR) after TAVR and to explore reasons underlying hospital variation in 30-day RSRR. The study included 27 091 index TAVRs performed across 325 hospitals. The median (interquartile range) hospital-level 30-day RSRR was 11.9% (11.1%-12.8%) ranging from 8.8% to 16.5%. After adjusting for differences in patient characteristics, there was significant between-hospital variation in 30-day RSRR (hospital odds ratio, 1.59; 95% CI, 1.39-1.77). Differences in length of stay and discharge disposition accounted for 15% of the between-hospital variance in RSRRs. There was no significant association between hospital characteristics and 30-day readmission rates after TAVR. There was statistically significant but weak correlation between 30-day RSRR after TAVR and that after surgical aortic valve replacement, percutaneous coronary intervention, acute myocardial infarction, heart failure, and pneumonia (r=0.132-0.298; P<0.001 for all). Causes of 30-day readmission varied across hospitals, with noncardiac readmissions being more common at the bottom 5% hospitals (ie, those with the highest RSRRs). Conclusions There is significant variation in 30-day RSRR after TAVR across hospitals that is not entirely explained by differences in patient or hospital characteristics as well as hospital-wide practice patterns. Noncardiac readmissions are more common in hospitals with the highest RSRRs.

Impact of surgeon and hospital factors on surgical decision-making for grade 1 degenerative lumbar spondylolisthesis: a Quality Outcomes Database analysis

OBJECTIVE

Surgical treatment for degenerative spondylolisthesis has been proven to be clinically challenging and cost-effective. However, there is a range of thresholds that surgeons utilize for incorporating fusion in addition to decompressive laminectomy in these cases. This study investigates these surgeon- and site-specific factors by using the Quality Outcomes Database (QOD).

METHODS

The QOD was queried for all cases that had undergone surgery for grade 1 spondylolisthesis from database inception to February 2019. In addition to patient-specific covariates, surgeon-specific covariates included age, sex, race, years in practice (0-10, 11-20, 21-30, > 30 years), and fellowship training. Site-specific variables included hospital location (rural, suburban, urban), teaching versus nonteaching status, and hospital type (government, nonfederal; private, nonprofit; private, investor owned). Multivariable regression and predictor importance analyses were performed to identify predictors of the treatment performed (decompression alone vs decompression and fusion). The model was clustered by site to account for site-specific heterogeneity in treatment selection.

RESULTS

A total of 12,322 cases were included with 1988 (16.1%) that had undergone decompression alone. On multivariable regression analysis clustered by site, adjusting for patient-level clinical covariates, no surgeon-specific factors were found to be significantly associated with the odds of selecting decompression alone as the surgery performed. However, sites located in suburban areas (OR 2.32, 95% CI 1.09-4.84, p = 0.03) were more likely to perform decompression alone (reference = urban). Sites located in rural areas had higher odds of performing decompression alone than hospitals located in urban areas, although the results were not statistically significant (OR 1.33, 95% CI 0.59-2.61, p = 0.49). Nonteaching status was independently associated with lower odds of performing decompression alone (OR 0.40, 95% CI 0.19-0.97, p = 0.04). Predictor importance analysis revealed that the most important determinants of treatment selection were dominant symptom (Wald χ2 = 34.7, accounting for 13.6% of total χ2) and concurrent diagnosis of disc herniation (Wald χ2 = 31.7, accounting for 12.4% of total χ2). Hospital teaching status was also found to be relatively important (Wald χ2 = 4.2, accounting for 1.6% of total χ2) but less important than other patient-level predictors.

CONCLUSIONS

Nonteaching centers were more likely to perform decompressive laminectomy with supplemental fusion for spondylolisthesis. Suburban hospitals were more likely to perform decompression only. Surgeon characteristics were not found to influence treatment selection after adjustment for clinical covariates. Further large database registry experience from surgeons at high-volume academic centers at which surgically and medically complex patients are treated may provide additional insight into factors associated with treatment preference for degenerative spondylolisthesis.

From Multiple Quality Indicators of Breast Cancer Care Toward Hospital Variation of a Summary Measure

OBJECTIVES

To improve quality in breast cancer care, large numbers of quality indicators are collected per hospital, but benchmarking remains complex. We aimed to assess the validity of indicators, develop a textbook outcome summary measure, and compare case-mix adjusted hospital performance.

METHODS

From a nationwide population-based registry, all 79 690 nonmetastatic breast cancer patients surgically treated between 2011 and 2016 in 91 hospitals in The Netherlands were included. Twenty-one indicators were calculated and their construct validity tested by Spearman's rho. Between-hospital variation was expressed by interquartile range (IQR), and all valid indicators were included in the summary measure. Standardized scores (observed/expected based on case mix) were calculated as above (>100) or below (<100) expected. The textbook outcome was presented as a continuous and all-or-none score.

RESULTS

The size of between-hospital variation varied between indicators. Sixteen (76%) of 21 quality indicators showed construct validity, and 13 were included in the summary measure after excluding redundant indicators that showed collinearity with others owing to strong construct validity. The median all-or-none textbook outcome score was 49% (IQR 42%-54%) before and 49% (IQR 48%-51%) after case-mix adjustment. From the total of 91 hospitals, 3 hospitals were positive (3%) and 9 (10%) were negative outliers.

CONCLUSIONS

The textbook outcome summary measure showed discriminative ability when hospital performance was presented as an all-or-none score. Although indicator scores and outlier hospitals should always be interpreted cautiously, the summary measure presented here has the potential to improve Dutch breast cancer quality indicator efforts and could be implemented to further test its validity, feasibility, and usefulness.

Evidence-practice gaps in P2Y12 inhibitor use after hospitalisation for acute myocardial infarction: findings from a new population-level data linkage in Australia

Background

P2Y₁₂ inhibitor therapy is recommended for 12 months in patients hospitalised for acute myocardial infarction (AMI) unless the bleeding risk is high.

Aims

To describe real‐world use of P2Y₁₂ inhibitor therapy following AMI hospitalisation.

Methods

We used population‐level linked hospital data to identify all patients discharged from a public hospital with a primary diagnosis of AMI between July 2011 and June 2013 in New South Wales and Victoria, Australia. We used dispensing claims to examine dispensing of a P2Y₁₂ inhibitor (clopidogrel, prasugrel or ticagrelor) within 30 days of discharge and multilevel models to identify predictors of post‐discharge dispensing and persistence of therapy to 1 year.

Results

We identified 31 848 patients hospitalised for AMI, of whom 56.8% were dispensed a P2Y₁₂ inhibitor within 30 days of discharge. The proportion of patients with post‐discharge dispensing varied between hospitals (interquartile range: 25.0–56.5%), and significant between‐hospital variation remained after adjusting for patient characteristics. Patient factors associated with the lowest likelihood of post‐discharge dispensing were: having undergone coronary artery bypass grafting (odds ratio (OR): 0.17; 95% confidence intervals (CI): 0.15–0.20); having oral anticoagulants dispensed 180 days before or 30 days after discharge (OR: 0.39, 95% CI: 0.35–0.44); major bleeding (OR: 0.68, 95% CI: 0.61–0.76); or being aged ≥85 years (OR: 0.68, 95% CI: 0.62–0.75). A total of 26.8% of patients who were dispensed a P2Y₁₂ inhibitor post‐discharge discontinued therapy within 1 year.

Conclusion

Post‐hospitalisation use of P2Y₁₂ inhibitor therapy in AMI patients is low and varies substantially by hospital of discharge. Our findings suggest strategies addressing both health system (hospital and physician) and patient factors are needed to close this evidence‐practice gap.

Hospital Variation in Geriatric Surgical Safety for Emergency Operation

BACKGROUND

The American College of Surgeons maintains that surgical care in the US has not reached optimal safety and quality. This can be driven partially by higher-risk, emergency operations in geriatric patients. We therefore sought to answer 2 questions: First, to what degree does standardized postoperative mortality vary in hospitals performing nonelective operations in geriatric patients? Second, can the differences in hospital-based mortality be explained by patient-, operative-, and hospital-level characteristics among outlier institutions?

STUDY DESIGN

Patients 65 years and older who underwent 1 of 8 common emergency general surgery operations were identified using the California State Inpatient Database (2010 to 2011). Expected mortality was obtained from hierarchical, Bayesian mixed-effects logistic regression models. A risk-adjusted hospital-level standardized mortality ratio (SMR) was calculated from observed-to-expected in-hospital deaths. "Outlier" hospitals had an SMR 80% CI that did not cross the mean SMR of 1.0. High-mortality (SMR >1.0) and low-mortality (SMR <1.0) outliers were compared.

RESULTS

We included 24,207 patients from 107 hospitals. SMRs varied widely, from 2.3 (highest) to 0.3 (lowest). Eleven hospitals (10.3%) were poor-performing high-SMR outliers, and 10 hospitals (9.3%) were exceptional-performing low-SMR outliers. SMR was 3 times worse in the high-SMR compared with the low-SMR group (1.7 vs 0.6; p < 0.001). Patient-, operation-, and hospital-level characteristics were equivalent among outlier-hospitals.

CONCLUSIONS

Significant hospital variation exists in standardized mortality after common general surgery operations done emergently in older patients. More than 10% of institutions have substantial excess mortality. These findings confirm that the safety of emergency operation in geriatric patients can be significantly improved by decreasing the wide variability in mortality outcomes.

Traditional Chinese Medicine Use in the Treatment of Acute Heart Failure in Western Medicine Hospitals in China: Analysis From the China PEACE Retrospective Heart Failure Study

Background

Traditional Chinese medicine (TCM) is used in the treatment of many conditions, including heart failure (HF), although it is not well characterized.

Methods and Results

We conducted a retrospective analysis of TCM use in a random sample of hospitalizations for HF within a random sample of Western medicine hospitals in China in 2015 using data from the China PEACE 5r‐HF (China Patient‐Centered Evaluative Assessment of Cardiac Events 5 Retrospective Heart Failure Study). We describe the frequency of TCM use and its association with patient characteristics, in‐hospital use of evidence‐based therapies, and hospital characteristics using hierarchical logistic regression models. Finally, we assessed risk‐adjusted in‐hospital bleeding and mortality. Among 10 004 patients hospitalized with HF (median age, 73 years; 48.9% women) from 189 hospitals, 74.7% received TCM (83.3% administered intravenously). The most commonly used agent was Salvia miltiorrhiza (51.2%). Patients with coronary artery disease (odds ratio [OR], 1.73; 95% CI, 1.53–1.95) or stroke (OR, 1.32; 95% CI, 1.15–1.51) were more likely to receive TCM; there was no correlation with evidence‐based therapy use. Nearly all hospitals (99.4%) used TCM, with substantial variation across hospitals (median OR, 3.29; 95% CI, 2.82–3.76). In‐patient bleeding (OR, 1.39; 95% CI, 1.03–1.88) and mortality (OR, 1.36; 95% CI, 1.04–1.79) were higher with Salvia miltiorrhiza, although not with other TCMs.

Conclusions

In a nationally representative sample of patients hospitalized with acute HF in China, three fourths received TCM. Nearly all hospitals used TCM, although use varied substantially by hospital. Although TCM was not used in lieu of evidence‐based therapies for HF, we found a signal for harm with the most commonly used TCM.

Clinical Trial Registration

URL: https://www.clinicaltrials.gov. Unique identifier: NCT02877914.

Optimal maternal and neonatal outcomes and associated hospital characteristics

BACKGROUND

This study aims to examine hospital variation in both maternal and neonatal morbidities and identify institutional characteristics associated with hospital performance in a combined measure of maternal and neonatal outcomes.

METHODS

Using the California Linked Birth File containing data from birth certificate and hospital discharge records, we identified 1 322 713 term births delivered at 248 hospitals during 2010-2012. For each hospital, a risk-standardized rate of severe maternal morbidities and a risk-standardized rate of severe newborn morbidities were calculated after adjusting for patient clinical risk factors. Hospitals were ranked based on combined information on their maternal and newborn morbidity rates.

RESULTS

Risk-standardized severe maternal and severe newborn morbidity rates varied substantially across hospitals (10th to 90th percentile range = 67.5-148.2 and 141.8-508.0 per 10 000 term births, respectively), although there was no significant association between the two (P = 0.15). Government hospitals (non-Federal) were more likely than other hospitals to be in worse rank quartiles (P value for trend = 0.004), whereas larger volume was associated with better rank among hospitals in the first three quartiles (P = 0.004). The most prevalent morbidities that differed progressively across hospital rank quartiles were severe hemorrhage, disseminated intravascular coagulation, and heart failure during procedure/surgery for mothers, and severe infection, respiratory complication, and shock/resuscitation for neonates.

CONCLUSIONS

Hospitals with low maternal morbidity rates may not have low neonatal morbidity rates and vice versa, highlighting the importance of assessing joint maternal-newborn outcomes in order to fully characterize a hospital's obstetrical performance. Hospitals with smaller volume and government ownership tend to have less desirable outcomes and warrant additional attention in future quality improvement efforts.

Unexpected term NICU admissions: a marker of obstetrical care quality?

BACKGROUND

Unexpected admissions of term neonates to the neonatal intensive care unit and unexpected postnatal complications have been proposed as neonatal-focused quality metrics for intrapartum care. Previous studies have noted significant variation in overall hospital neonatal intensive care unit admission rates; however, little is known about the influence of obstetric practices on these rates or whether variation among unanticipated admissions in low-risk, term neonates can be attributed to systemic hospital practices.

OBJECTIVE

The objective of the study was to examine the relative effects of patient characteristics and intrapartum events on unexpected neonatal intensive care unit admissions and to quantify the between-hospital variation in neonatal intensive care unit admission rates among this group of neonates.

STUDY DESIGN

We performed a retrospective cross-sectional study using data collected as part of the Consortium for Safe Labor study. Women who delivered term (≥37 weeks), singleton, nonanomalous, liveborn infants without an a priori risk for neonatal intensive care unit admission were included. The primary outcome was neonatal intensive care unit admission among this population. Multilevel mixed-effect models were used to calculate adjusted odds ratios for demographics (age, race, insurer), pregnancy characteristics (parity, gestational age, tobacco use, birthweight), maternal comorbidities (chronic and pregnancy-induced hypertension), hospital characteristics (delivery volume, hospital and neonatal intensive care unit level, academic affiliation), and intrapartum events (prolonged second stage, induction of labor, trial of labor after cesarean delivery, chorioamnionitis, meconium-stained amniotic fluid, and abruption). Intraclass correlation coefficients were used to estimate the between-hospital variance in a series of hierarchical models.

RESULTS

Of the 143,951 infants meeting all patient and hospital inclusion criteria, 7995 (5.6%) were admitted to the neonatal intensive care unit after birth. In the fully adjusted model, the factors associated with the highest odds for neonatal intensive care unit admission included: nulliparity (adjusted odds ratio, 1.62 [95% confidence interval, 1.53-1.71]), large for gestational age (adjusted odds ratio, 1.59 [95% confidence interval, 1.47-1.71]), and small for gestational age (adjusted odds ratio, 1.60 [95% confidence interval, 1.47-1.73]). Induction of labor (adjusted odds ratio, 0.95 [95% confidence interval, 0.89-1.01]) was not associated with increased odds of neonatal intensive care unit admission compared with women who labored spontaneously. The events associated with higher odds of neonatal intensive care unit admission included: prolonged second stage (adjusted odds ratio, 1.66 [95% confidence interval, 1.51-1.83]); chorioamnionitis (adjusted odds ratio, 3.89 [95% confidence interval, 3.42-4.44]), meconium-stained amniotic fluid (adjusted odds ratio, 1.96 [95% confidence interval, 1.82-2.10]), and abruption (adjusted odds ratio, 2.64 [95% confidence interval, 2.16-.21]). Compared with women who did not labor, the odds of neonatal intensive care unit admission were lower for women who labored: adjusted odds ratio, 0.48 (95% confidence interval, 0.45-0.52) for women with no uterine scar and adjusted odds ratio, 0.83 (95% confidence interval, 0.73-0.94) for women with a uterine scar. There was significant variation in neonatal intensive care unit admission rates by hospital, ranging from 2.9% to 11.2%. After accounting for case mix and hospital characteristics, the between-hospital variance was 1.9%, suggesting that little of the variation was explained by the effect of the hospital.

CONCLUSION

This study contributes to the currently limited understanding of term, neonatal intensive care unit admission rates as a marker of obstetrical care quality. We demonstrated that significant variation exists in hospital unexpected neonatal intensive care unit admission rates and that certain intrapartum events are associated with an increased risk for neonatal intensive care unit admission after delivery. However, the between-hospital variation was low. Unmeasured confounders and extrinsic factors, such as neonatal intensive care unit bed availability, may limit the ability of unexpected term neonatal intensive care unit admissions to meaningfully reflect obstetrical care quality.

Hospital variation in cost of childbirth and contributing factors: a cross-sectional study

OBJECTIVE

To examine hospital variation in cost of childbirth hospitalisations and identify factors that contribute to the variation.

DESIGN

Cross-sectional analysis of linked birth certificate and hospital discharge data.

SETTING

Two hundred and twenty hospitals in California delivering ≥ 100 births per year.

POPULATION

A total of 405 908 nulliparous term singleton vertex births during 2010-2012.

METHODS

Cost of childbirth hospitalisations was compared across hospitals after accounting for differences in patient clinical risk factors. Relative contributions of patient sociodemographic, obstetric intervention, birth attendant and institutional characteristics to variation in cost were assessed by further adjusting for these factors in hierarchical generalised linear models.

MAIN OUTCOME MEASURES

Cost of childbirth hospitalisation.

RESULTS

Median risk-standardised cost of childbirth was $7149 among the hospitals (10^th -90th percentile range: $4760-$10,644). Maternal sociodemographic characteristics and type of birth attendant did not explain hospital variation in cost. Adjustment for obstetric interventions overall reduced within-hospital variance by 15.8% (P < 0.001), while adjusting for caesarean delivery alone reduced within-hospital variance by 14.4% (P < 0.001). However, obstetric interventions did not explain between-hospital variation in cost. In contrast, adjustment for institutional characteristics reduced between-hospital variance by 30.3% (P = 0.002). Hospital type of ownership, teaching/urban-rural status, neonatal care capacity and geographic region were most impactful. Risk-standardised cost was positively correlated with risk-standardised rate of severe newborn morbidities (correlation coefficient 0.22, P = 0.001), but not associated with risk-standardised rate of severe maternal morbidities.

CONCLUSIONS

Cost of childbirth hospitalisations varied widely among hospitals in California. Institutional characteristics significantly contributed to this variation. Higher-cost hospitals did not have better outcomes, suggesting potential opportunities to enhance value in care.

TWEETABLE ABSTRACT

Hospitals vary in cost of childbirth. Institutional characteristics significantly contribute to the variation.

Patterns of Care in the Administration of Neo-adjuvant Chemotherapy for Breast Cancer. A Population-Based Study

Neo-adjuvant chemotherapy (NAC) is used to facilitate radical surgery for initially irresectable or locally advanced breast cancer. The indication for NAC has been extended to clinically node negative (cN0) patients in whom adjuvant systemic therapy is foreseen. A population-based study was conducted to evaluate the increasing use of NAC, breast conserving surgery (BCS) after NAC and timing of the sentinel node biopsy (SNB). All female breast cancer patients, treated in 10 hospitals in the Eindhoven Cancer Registry area in the Netherlands between January 2003 and June 2012 were included (N = 18,427). In total, 1,402 patients (7.6%) received NAC. The administration increased from 2.5% in 2003 to 13.0% in 2011 (p < 0.001). Use of NAC increased from 0.5% to 2.3% for cT1 tumors, from 2.8% to 27.0% for cT2, from 30.6% to 70.9% for cT3, and from 40.5% to 58.1% for cT4 tumors (p < 0.001). In cN0 patients, use of NAC increased from 1.0% to 4.4% and in clinically node positive patients from 12.0% to 57.5% (p < 0.001). Downsizing of the tumor and BCS are achieved increasingly. In 2011, in three hospitals NAC was administered in <10% of patients, in five hospitals in 10-15% and in two hospitals the proportion of patients receiving NAC was >20% (p < 0.001). Of the 1,402 patients with NAC, 495 patients underwent SNB, 91.5% of whom prior to NAC. In the Netherlands up to one in eight patients receive NAC. The administration of NAC and the percentage of BCS increased over the past decade, especially in cT2 tumors. Considerable hospital variation in the administration of NAC exists.

Comparing variation in hospital rates of cesarean delivery among low-risk women using 3 different measures

This report describes the development of a measure of low-risk cesarean delivery by the Society for Maternal-Fetal Medicine (SMFM). Safely lowering the cesarean delivery rate is a priority for maternity care clinicians and health care delivery systems. Therefore, hospital quality assurance programs are increasingly tracking cesarean delivery rates among low-risk pregnancies. Two commonly used definitions of "low risk" are available, the Joint Commission (JC) and the Agency for Healthcare Research and Quality (AHRQ) measures, but these measures are not clinically comprehensive. We sought to refine the definition of the low-risk cesarean delivery rate to enhance the validity of the metric for quality measurement. We created this refined definition-called the SMFM definition-and compared it to the JC and AHRQ measures using claims-based data from the 2011 Nationwide Inpatient Sample of >863,000 births in 612 hospitals. Using these definitions, we calculated means and interquartile ranges (25th-75th percentile range) for hospital low-risk cesarean delivery rates, stratified by hospital size, teaching status, urban/rural location, and payer mix. Across all hospitals, the mean low-risk cesarean delivery rate was lowest for the SMFM definition (12.65%), but not substantially different from the JC and AHRQ measures (13.12% and 13.29%, respectively). We empirically examined the SMFM definition to ensure its validity and utility. This refined definition performs similarly to existing measures and has the added advantage of clinical perspective, enhanced face validity, and ease of use.

Hospital variation in cesarean delivery rates: contribution of individual and hospital factors in Florida

BACKGROUND

Primary cesarean deliveries are a major contributor to the large increase in cesarean delivery rates in the United States over the past 2 decades and are an essential focus for the reduction of related morbidity and costs. Studies have shown that primary cesarean delivery rates among low-risk women in the United States vary 3-fold across hospitals and are not explained by differences in patient case-mix. However, the extent to which maternal vs hospital characteristics contribute to this variation remains poorly understood because previous studies were limited in scope and did not assess the influence of factors such as maternal ethnicity subgroups or prepregnancy obesity.

OBJECTIVE

We assessed the contribution of individual- and hospital-level risk factors to the hospital variation in primary cesarean delivery rates among low-risk women in Florida.

STUDY DESIGN

Our population-based retrospective cohort study used Florida's linked birth certificate and hospital discharge records for the period of 2004-2011. The study population was comprised of 412,192 nulliparous, singleton, vertex, live births with labor at 37-40 weeks gestation in 122 nonmilitary delivery hospitals. Data were analyzed with logistic mixed-effects regression with cesarean delivery as the outcome. This approach provided adjusted risk estimates at an individual and hospital level and the estimated percent of hospital variation statewide that was explained by these factors.

RESULTS

The primary cesarean delivery rate in the study population was 23.9%, with hospital-specific estimates that ranged from 12.8-47.3%. Leading risk factors for cesarean delivery were maternal age ≥35 years (adjusted relative risk, 2.22), prepregnancy obesity (body mass index, ≥30 kg/m(2); adjusted relative risk, 1.73), medical risk conditions (adjusted relative risk, 1.72), labor induction (adjusted relative risk, 1.52), and delivery in hospitals located in Miami-Dade County (adjusted relative risk, 1.73). Hospital geographic location was a significant effect modifier for prepregnancy obesity, medical conditions, and labor induction (P < .05), with a tendency towards lower adjusted relative risks for these factors in Miami-Dade County relative to other Florida regions. Conversely, Miami-Dade County had an increased prevalence of higher-risk ethnic subgroups, such as Cuban or Puerto Rican mothers, and also substantially higher adjusted relative risks that were associated with practice-related factors, such as delivery during weekday hours. Whereas hospital geographic location contributed to 39.6% of the observed variation statewide, the estimated contribution of maternal ethnicity ranged from 1.6-15.7% among Florida regions.

CONCLUSIONS

Hospital geographic location contributes to hospital variation in primary cesarean delivery rates among low-risk women in Florida. In contrast to previous studies, our findings suggest that individual level risk factors such as maternal ethnicity also contribute to some of this variation, with differing extent by region. These individual factors likely interact with practice factors and add to the variation. This study was limited by not including maternal Bishop score before induction or obstetrics provider in the analysis. These were not available on the dataset but likely contribute to the variation. Our findings suggest potential issues to consider in quality improvement efforts, such as the need for future qualitative research that focuses on mothers in higher-risk ethnic subgroups and providers in high-rate hospitals, particularly those in Miami-Dade County. These studies may help to identify potential cultural differences in maternal beliefs and expectations for delivery and maternal reasons for differences in obstetrics practices.

Hospital Variation in Cesarean Delivery: A Multilevel Analysis

OBJECTIVES

To assess the issue of hospital variations in Colombia and to contribute to the methodology on health care variations by using a model that clusters the variance between hospitals while accounting for individual-level reimbursement rates and objective health-status variables.

METHODS

We used data on all births (N = 11,954) taking place in a contributory-regimen insurer network in Colombia during 2007. A multilevel logistic regression model was used to account for the share of unexplained variance between hospitals. In addition, an alternative variance decomposition specification was further carried out to measure the proportion of such unexplained variance due to the region effect.

RESULTS

Hospitals account for 20% of the variation in performing cesarean sections, whereas region explains only one-third of such variance. Variables accounting for preferences on the demand side as well as reimbursement rates are found to predict the probability of performing cesarean sections.

CONCLUSIONS

Hospital variations explain large variances within a single-payer's network. Because this insurer company is highly regarded in terms of performance and finance, these results might provide a lower bound for the scale of hospital variation in the Colombian health care market. Such lower bound provides guidance on the relevance of this issue for Colombia. Some factors such as demand-side preferences and physician reimbursement rates increase variations in health care even within a single-payer network. This is a source of inefficiencies, threatening the quality of health care and financial sustainability. The proposed methodology should be considered in further research on health care variations.

Do laborists improve delivery outcomes for laboring women in California community hospitals?

OBJECTIVE

We sought to determine the impact of the laborist staffing model on cesarean rates and maternal morbidity in California community hospitals.

STUDY DESIGN

This is a cross-sectional study comparing cesarean rates, vaginal birth after cesarean rates, composite maternal morbidity, and severe maternal morbidity for laboring women in California community hospitals with and without laborists. We conducted interviews with nurse managers to obtain data regarding hospital policies, practices, and the presence of laborists, and linked this information with patient-level hospital discharge data for all deliveries in 2012.

RESULTS

Of 248 childbirth hospitals, 239 (96.4%) participated; 182 community hospitals were studied, and these hospitals provided 221,247 deliveries for analysis. Hospitals with laborists (n = 43, 23.6%) were busier, had more clinical resources, and cared for higher-risk patients. There was no difference in the unadjusted primary cesarean rate for laborist vs nonlaborist hospitals (11.3% vs 11.7%; P = .382) but there was a higher maternal composite morbidity rate (14.4% vs 12.0%; P = .0006). After adjusting for patient and hospital characteristics, there were no differences in laborist vs nonlaborist hospitals for any of the specified outcomes. Hospitals with laborists had higher attempted trial of labor after cesarean rates, and lower repeat cesarean rates (90.9% vs 95.9%; P < .0001). However, among women attempting trial of labor after cesarean, there was no difference in the vaginal birth after cesarean success rate.

CONCLUSION

We were unable to demonstrate differences in cesarean and maternal childbirth complication rates in community hospitals with and without laborists. Further efforts are needed to understand how the laborist staffing model contributes to neonatal outcomes, cost and efficiency of care, and patient and physician satisfaction.

Variation in critical care unit admission rates and outcomes for patients with acute coronary syndromes or heart failure among high- and low-volume cardiac hospitals

Background

Little is known about cross‐hospital differences in critical care units admission rates and related resource utilization and outcomes among patients hospitalized with acute coronary syndromes (ACS) or heart failure (HF).

Methods and Results

Using a population‐based sample of 16 078 patients admitted to a critical care unit with a primary diagnosis of ACS (n=14 610) or HF (n=1467) between April 1, 2003 and March 31, 2013 in Alberta, Canada, we stratified hospitals into high (>250), medium (200 to 250), or low (<200) volume based on their annual volume of all ACS and HF hospitalization. The percentage of hospitalized patients admitted to critical care units varied across low, medium, and high‐volume hospitals for both ACS and HF as follows: 77.9%, 81.3%, and 76.3% (P<0.001), and 18.0%, 16.3%, and 13.0% (P<0.001), respectively. Compared to low‐volume units, critical care patients with ACS and HF admitted to high‐volume hospitals had shorter mean critical care stays (56.6 versus 95.6 hours, P<0.001), more critical care procedures (1.9 versus 1.2 per patient, <0.001), and higher resource‐intensive weighting (2.8 versus 1.5, P<0.001). No differences in in‐hospital mortality (5.5% versus 6.2%, adjusted odds ratio 0.93; 95% CI, 0.61 to 1.41) were observed between high‐ and low‐volume hospitals; however, 30‐day cardiovascular readmissions (4.6% versus 6.8%, odds ratio 0.77; 95% CI, 0.60 to 0.99) and cardiovascular emergency‐room visits (6.6% versus 9.5%, odds ratio 0.80; 95% CI, 0.69 to 0.94) were lower in high‐volume compared to low‐volume hospitals. Outcomes stratified by ACS or HF admission diagnosis were similar.

Conclusions

Cardiac patients hospitalized in low‐volume hospitals were more frequently admitted to critical care units and had longer hospitals stays despite lower resource‐intensive weighting. These findings may provide opportunities to standardize critical care utilization for ACS and HF patients across high‐ and low‐volume hospitals.

Variation in hospital treatment patterns for metastatic colorectal cancer

BACKGROUND

There are many treatment options for metastatic colorectal cancer (CRC). However, to the authors' knowledge, national treatment patterns for metastatic CRC, and the stability of hospital treatment patterns over time, have not been well described.

METHODS

Data from the 2006 through 2011 National Cancer Data Base were used to study adults with newly diagnosed metastatic CRC (84,161 patients from 1051 hospitals). Using hierarchical models, the authors characterized hospital volume in the use of different treatment modalities (primary site resection, metastatic site resection, chemotherapy, and palliative care). The authors then assessed variation in the receipt of treatment according to the hospitals' relative volume of services used. Finally, the extent to which hospital treatment patterns changed over the past decade was examined.

RESULTS

Overall use of volume of services varied widely (5.0% in the hospitals with low volumes of service to 22.3% in the hospitals with high volumes of service). As hospitals' volumes of services increased, adjusted rates of metastatic site surgery (6.6% to 30.8%; P<.001) and multiagent chemotherapy (37.8% to 57.4%; P<.001) use increased, but primary site resection demonstrated little variation (56.8% vs 59.5%; P = .024). It is interesting to note that use of palliative care also increased (8.1% to 11.3%; P = .002). Hospital treatment patterns did not change over time, with hospitals with high volumes of service consistently using more metastatic site resection and multiagent chemotherapy than hospitals with low volumes of service.

CONCLUSIONS

There is wide variation in hospital treatment patterns for patients with metastatic CRC, and these patterns have been stable over time. It appears that much of the approach for metastatic CRC treatment depends on the hospital in which the patient presents.

No improvement in median survival for patients with metastatic gastric cancer despite increased use of chemotherapy

BACKGROUND

Gastric cancer often presents in a metastasized stage. We conducted a population-based study to evaluate trends in systemic treatment and survival of metastatic noncardia gastric cancer.

PATIENTS AND METHODS

All patients with noncardia adenocarcinoma of the stomach, diagnosed between 1990 and 2011 in the Eindhoven Cancer Registry area in the Netherlands were included (N = 4797). We conducted multivariable logistic regression analysis to evaluate trends in administration of palliative chemotherapy and multivariable proportional hazards regression analyses to evaluate trends in crude overall survival.

RESULTS

The proportion of patients presenting with metastatic gastric cancer increased from 24% in 1990 to 44% in 2011 (P < 0.0001). The use of palliative chemotherapy increased, from 5% in 1990 to 36% in 2011, with a strong increase in particular after 2006 (P < 0.0001). Younger patients [<50 years: adjusted odds ratio (ORadj) 3.9, P < 0.001; 50-59 years: ORadj 1.7, P = 0.01] and patients with a high socioeconomic status (ORadj 1.7, P = 0.01) more often received chemotherapy. In contrast, older patients (70-79 years: ORadj 0.3, P < 0.001; 80+ years: ORadj 0.02, P < 0.001), patients with comorbidity (ORadj 0.6, P = 0.03), linitis plastica (ORadj 0.5, P = 0.03) and multiple distant metastases (ORadj 0.5, P = 0.01) were less often treated with chemotherapy. A large hospital variation was observed in the administration of palliative chemotherapy (9%-27%). Median overall survival remained constant between 15 [95% confidence interval (CI) 11.9-17.7] and 17 (95% CI 15.0-20.0) weeks (P = 0.10).

CONCLUSIONS

The increased administration of chemotherapy in patients with metastatic gastric cancer did not lead to an increase in population-based overall survival. Identification of the subgroup of patients which benefits from palliative chemotherapy is of utmost importance to avoid unnecessary treatment.