Impact of Changing the Statistical Methodology on Hospital and Surgeon Ranking

Glance, Laurent G.; Dick, Andrew W.; Osler, Turner M.; Li, Yue; Mukamel, Dana B.

doi:10.1097/01.mlr.0000204106.64619.2a

Cited by 77 publications

(68 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, it was found that hospital rankings may change considerably on the COPD AND SMOKING-RELATED DISORDERS N. AGABITI ET AL. basis of the risk-adjustment models used and that the classification of ''outlier status'' depends on the chosen threshold for statistical significance [22,23]. In our study, a less conservative approach, using p,0.1, could have implied more high and low outliers.…”

Section: Discussionmentioning

confidence: 88%

Profiling hospital performance to monitor the quality of care: the case of COPD

Agabiti¹,

Belleudi²,

Davoli³

et al. 2009

European Respiratory Journal

View full text Add to dashboard Cite

Comparative outcomes data are widely used to monitor quality of care in the cardiovascular area; little is available in the respiratory field. We applied validated methods to compare hospital outcomes for chronic obstructive pulmonary disease (COPD) exacerbation.From the hospital information system, we selected all hospital admissions for COPD exacerbation in Rome (for 2001Rome (for -2005. Vital status within 30 days was obtained from the municipality mortality register. Each hospital was compared to a pool of hospitals with the lowest adjusted mortality rate (the benchmark). Age, sex and several potential clinical predictors were covariates in logistic regression analysis.12,756 exacerbated COPD patients were analysed (mean age 74 yrs, 71% males). Diabetes, hypertension, ischaemic heart disease, heart failure and arrhythmia were the most common coexisting conditions. The average crude mortality in the benchmark group was 3.8%; in the remaining population it was 7.5% (range 5.2-17.2%). In comparison with the benchmark, the relative risk of 30-day mortality varied widely across the hospitals (range 1.5-5.9%).A large variability in 30-day mortality after COPD exacerbation exists even considering patients' characteristics. Although these results do not detect mechanisms related to worse outcomes, they may be useful to stimulate providers to revision and improvement of COPD care management.

show abstract

Section: Discussionmentioning

confidence: 88%

Profiling hospital performance to monitor the quality of care: the case of COPD

Agabiti¹,

Belleudi²,

Davoli³

et al. 2009

European Respiratory Journal

View full text Add to dashboard Cite

show abstract

“…15 A fixed-effects models results in an unbiased estimate of the hospital effect when correlated with observable patient risk factors. The model was evaluated for discrimination using the c-statistic.…”

Section: Main Outcome Measure and Statistical Analysismentioning

confidence: 99%

“…Multiplying the O:E ratio by the outcome's mean yields the risk-adjusted rates. 15 All statistical analyses were conducted using STATA special edition (version 13, StataCorp, College Station, TX).…”

Section: Main Outcome Measure and Statistical Analysismentioning

confidence: 99%

Multi-institutional assessment of sphincter preservation for rectal cancer.

Abdelsattar¹,

Wong²,

Birkmeyer³

et al. 2014

JCO

View full text Add to dashboard Cite

BACKGROUND-Sphincter preserving surgery (SPS) has been proposed as a quality measure for rectal cancer (RC) surgery. However, previous studies on SPS-rates lack critical clinical characteristics, rendering it unclear if variation in SPS-rates is due to unmeasured case-mix differences or surgeons' selection criteria. In this context, we investigate the variation in SPS-rates at various practice settings.

show abstract

“…From this we derived relative risks (RRs) for each hospital using predicted probabilities from only the fixed effects part of the model. 27,28 We used the noblup ilink options within PROC GLIMMIX to achieve this. These RRs are akin to standardised mortality ratios (SMRs), which represent the ratio of the hospital's rate to the national average rate.…”

Section: Modelling Frameworkmentioning

confidence: 99%

Can valid and practical risk-prediction or casemix adjustment models, including adjustment for comorbidity, be generated from English hospital administrative data (Hospital Episode Statistics)? A national observational study

Bottle

Gaudoin

Goudie

et al. 2014

Health Services and Delivery Research

View full text Add to dashboard Cite

Background NHS hospitals collect a wealth of administrative data covering accident and emergency (A&E) department attendances, inpatient and day case activity, and outpatient appointments. Such data are increasingly being used to compare units and services, but adjusting for risk is difficult. Objectives To derive robust risk-adjustment models for various patient groups, including those admitted for heart failure (HF), acute myocardial infarction, colorectal and orthopaedic surgery, and outcomes adjusting for available patient factors such as comorbidity, using England’s Hospital Episode Statistics (HES) data. To assess if more sophisticated statistical methods based on machine learning such as artificial neural networks (ANNs) outperform traditional logistic regression (LR) for risk prediction. To update and assess for the NHS the Charlson index for comorbidity. To assess the usefulness of outpatient data for these models. Main outcome measures Mortality, readmission, return to theatre, outpatient non-attendance. For HF patients we considered various readmission measures such as diagnosis-specific and total within a year. Methods We systematically reviewed studies comparing two or more comorbidity indices. Logistic regression, ANNs, support vector machines and random forests were compared for mortality and readmission. Models were assessed using discrimination and calibration statistics. Competing risks proportional hazards regression and various count models were used for future admissions and bed-days. Results Our systematic review and empirical analysis suggested that for general purposes comorbidity is currently best described by the set of 30 Elixhauser comorbidities plus dementia. Model discrimination was often high for mortality and poor, or at best moderate, for other outcomes, for example c = 0.62 for readmission and c = 0.73 for death following stroke. Calibration was often good for procedure groups but poorer for diagnosis groups, with overprediction of low risk a common cause. The machine learning methods we investigated offered little beyond LR for their greater complexity and implementation difficulties. For HF, some patient-level predictors differed by primary diagnosis of readmission but not by length of follow-up. Prior non-attendance at outpatient appointments was a useful, strong predictor of readmission. Hospital-level readmission rates for HF did not correlate with readmission rates for non-HF; hospital performance on national audit process measures largely correlated only with HF readmission rates. Conclusions Many practical risk-prediction or casemix adjustment models can be generated from HES data using LR, though an extra step is often required for accurate calibration. Including outpatient data in readmission models is useful. The three machine learning methods we assessed added little with these data. Readmission rates for HF patients should be divided by diagnosis on readmission when used for quality improvement. Future work As HES data continue to develop and improve in scop...

show abstract

Impact of Changing the Statistical Methodology on Hospital and Surgeon Ranking

Cited by 77 publications

References 36 publications

Profiling hospital performance to monitor the quality of care: the case of COPD

Profiling hospital performance to monitor the quality of care: the case of COPD

Multi-institutional assessment of sphincter preservation for rectal cancer.

Can valid and practical risk-prediction or casemix adjustment models, including adjustment for comorbidity, be generated from English hospital administrative data (Hospital Episode Statistics)? A national observational study

Contact Info

Product

Resources

About