Interhospital Comparisons Research Articles

Comparison of antibiotic use and antibiotic resistance between a community hospital and tertiary care hospital for evaluation of the antimicrobial stewardship program in Japan.

Assessment of risk-adjusted antibiotic use (AU) is recommended to evaluate antimicrobial stewardship programs (ASPs). We aimed to compare the amount and diversity of AU and antimicrobial susceptibility of nosocomial isolates between a 266-bed community hospital (CH) and a 963-bed tertiary care hospital (TCH) in Japan. The days of therapy/100 bed days (DOT) was measured for four classes of broad-spectrum antibiotics predominantly used for hospital-onset infections. The diversity of AU was evaluated using the modified antibiotic heterogeneity index (AHI). With 10% relative DOT for fluoroquinolones and 30% for each of the remaining three classes, the modified AHI equals 1. Multidrug resistance (MDR) was defined as resistance to ≥ 3 anti-Pseudomonas antibiotic classes. The DOT was significantly higher in the TCH than in the CH (10.85 ± 1.32 vs. 3.89 ± 0.93, p < 0.001). For risk-adjusted AU, the DOT was 6.90 ± 1.50 for acute-phase medical wards in the CH, and 8.35 ± 1.05 in the TCH excluding the hematology department. In contrast, the DOT of antibiotics for community-acquired infections was higher in the CH than that in the TCH. As quality assessment of AU, higher modified AHI was observed in the TCH than in the CH (0.832 ± 0.044 vs. 0.721 ± 0.106, p = 0.003), indicating more diverse use in the TCH. The MDR rate in gram-negative rods was 5.1% in the TCH and 3.4% in the CH (p = 0.453). No significant difference was demonstrated in the MDR rate for Pseudomonas aeruginosa and Enterobacteriaceae species between hospitals. Broad-spectrum antibiotics were used differently in the TCH and CH. However, an increased antibiotic burden in the TCH did not cause poor susceptibility, possibly because of diversified AU. Considering the different patient populations, benchmarking AU according to the facility type is promising for inter-hospital comparisons of ASPs.

Contrasting definitions and incidence of healthcare-associated respiratory viral infections in a pediatric hospital.

To determine the difference in the incidence of healthcare-associated respiratory viral infection (HARVI) in a pediatric hospital depending on the definition used. Descriptive historical cohort study. Patients aged 0-21 years old who were admitted between July 2013 and June 2018 to a 490-bed primary to quaternary-care pediatric hospital serving northern Texas. HARVI was defined using microbiologic confirmation, development of new symptoms while hospitalized, and exposure time greater than the minimum incubation period for each specific virus. Events that occurred following the maximum incubation period for that virus were classified as definite, otherwise they were classified as possible. This definition was compared to definitions using alternate timing of onset and symptomatology requirements. Data pertaining to demographics, diagnoses, and illness severity were collected. In total, 498 HARVIs (320 definite and 178 possible) were identified, with an incidence rate of 0.98 per 1,000 patient days (0.63 and 0.35, respectively). Rhinovirus or enterovirus and respiratory syncytial virus were the most identified viruses (58% and 10%, respectively). The median time from admission until HARVI was 10.5 days (interquartile range [IQR], 5-30 days). When alternate definitions were employed, the incidence of HARVI ranged from 0.96 to 2.00 per 1,000 admitted patient days. HARVI remain a common nosocomial infection in pediatric hospitals and the measured incidence is dependent on the definition used. Because of the endemic and pandemic potential of respiratory viruses, standardized definitions are needed to facilitate intra- and interhospital comparisons.

A novel risk score for in-hospital perioperative mortality of five major surgeries.

Risk-scoring tools for perioperative mortality adjustment are essential for inter-hospital quality comparisons, which are still lacking in China. Existing scores had significant disadvantages when applied in managerial practice. This study aimed to develop a simple risk score using highly accessible information that could appropriately adjust the perioperative mortality from major surgeries across tertiary Chinese public hospitals and provide a reference for other underdeveloped countries with the same need. A study cohort from 19 hospitals was randomly split into a development set and an internal validation set in the ratio of 7:3. Another cohort from six hospitals was used as an external validation set. All data were obtained from the military-hospital public services database of the National Engineering Laboratory of Application Technology in Medical Big Data. Patients aged above 18 years undergoing one of the five categories of major surgical procedures between 1 January 2010 and 31 December 2020 were identified. The multivariate logistic regression analysis was used to predict the risk of mortality and derive the risk score. The area under the receiver operating characteristic curve (AUC) and the Hosmer-Lemeshow goodness-of-fit test were used to assess the discrimination and calibration of the model, respectively. The study set included 45 558 cases, divided into a development set containing 31 891 cases and an internal validation set with 13 667 cases. Another cohort with 14 956 cases was used as an external validation set. The final included predictor variables were age, Elixhauser Comorbidity Index, condition at admission, admission route and the procedure. The predicted risk score ranged from -21.5 to 37.0 points. The model discriminated well in the development set, internal validation set, and external validation set. The AUC for them were 0.753 (Standard Error(SE) 0.016, 95% Confidence Interval(CI): 0.721,0.784), 0.790 (SE 0.025, 95% CI: 0.742,0.839), and 0.766(SE 0.019, 95% CI: 0.728, 0.804), respectively. P values in the Hosmer-Lemeshow goodness-of-fit test were all above 0.05, indicating a good calibration. This risk-scoring model was proved to have satisfactory performance, allowing the rapid and effective risk adjustment of perioperative mortality when comparing the surgical quality in tertiary hospitals in China and other underdeveloped regions.

Predictors of a Top-Box Patient Experience: A Retrospective Observational Study of HCAHPS Data at a Safety Net Institution.

Hospitals commonly seek to improve patient experience as measured by the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey, yet there are limited data to guide improvement efforts. The HCAHPS survey was developed for interhospital comparisons, whereas its use in intrahospital comparisons has not been validated. We sought to better understand the validity of utilizing intrahospital score comparisons and to identify the factors that may predict top-box HCAHPS scores. We performed a retrospective observational cohort study at an academic urban safety-net hospital examining 4898 HCAHPS surveys completed by hospitalized patients. We found that while most Patient-Mix Adjustment factors for which HCAHPS scores are adjusted were associated with top-box scores on intrahospital comparisons, few additional variables were associated with top-box scores. Further, HCAHPS questions pertaining to nurse and doctor communication were highly correlated with overall hospital rating, suggesting that communication-related factors may influence a patient’s hospital experience more strongly than do administrative factors.

793. Expert Panel Consensus Ranking of Comorbid Conditions Causally Related to Clostridioides difficile Infection

BackgroundNumerous studies have identified comorbidities that are associated with Clostridioides difficile infection (CDI), but current CDC and CMS models for risk adjusting hospital CDI rates do not include patient comorbid conditions. Incorporating patient-level data could improve CDI risk adjustment, but comorbidities would need to be easily electronically available for widescale implementation. Ideally, they would also be causally related to CDI — i.e., true risk factors, not confounders — to facilitate more unbiased inter-hospital comparisons. The current study aimed to determine which comorbid conditions are causally related to CDI based upon expert consensus.MethodsWe used Delphi methodology to administer an iterative, two-round survey with an intervening teleconference, to eight infectious disease experts. Experts evaluated 40 comorbid conditions included in Charlson and Elixhauser comorbidity indices (and thus validated for electronic capture through administrative data), as well as other comorbidities commonly associated with CDI. Experts rated comorbid conditions from 1 (not at all related) to 5 (strongly related), based upon perceived relatedness with CDI. To assign causal relatedness, the following criteria had to be met at the end of round two: 1) majority (> 50%) of experts rating the condition at 3 (somewhat related) or higher; 2) inter-quartile range (IQR) < = 1; and 3) standard deviation (SD) < = 1.Results8/40 (20%) comorbid conditions were ranked as causally related to CDI, including patient age, three malignancy comorbidities, two transplant-related comorbidities, HIV/AIDS, and inflammatory bowel disease. A further 18/40 (45%) qualified as indeterminately related, and 14/40 (35%) were ranked as not causally related to CDI (Table). Three of the eight causally related factors were not components of Elixhauser or Charlson indices.Table ConclusionWe identified comorbid conditions that may be appropriate candidates to consider for inclusion in patient-level risk adjustment models. Some causal factors did not originate from established comorbidity indices. Thus, future work to validate electronic capture of these conditions could further reduce barriers to risk-adjustment implementation.Disclosures All Authors: No reported disclosures

Risk Factors Associated With Hospital-Onset MRSA Proportion—National Healthcare Safety Network, 2017–2018

Background:Staphylococcus aureus is frequently implicated in healthcare-associated infections in the United States, and a substantial proportion of these infections are attributed to methicillin-resistant Staphylococcus aureus (MRSA). Although MRSA infections have decreased in health care settings, accurate estimates of the rate of decline call for risk-adjusted methods for calculating the resistant proportion (%R), that is, the proportion of S. aureus resistant to cefoxitin or oxacillin. Risk-adjusted %R also enables more accurate interhospital comparisons and can serve as a quantitative guide and evaluation metric for prevention efforts. Methods: To develop a risk-adjusted %R for S. aureus, we analyzed the antimicrobial susceptibility test (AST) results for S. aureus isolates reported to the CDC NHSN Antimicrobial Resistance Option during 2017–2018. Isolates were reported for cerebrospinal fluid (CSF), blood, lower respiratory tract (LRT), and urine. Isolates without cefoxitin and oxacillin test results, or from the facilities that had &gt;10% missing test results were excluded. Test results were differentiated between those associated with community-onset and hospital-onset (HO) infections by defining the latter group as test results for isolates obtained 3 days or more after hospital admission. Logistic regression was used to evaluate the factors associated with oxacillin/cefoxitin resistance. Hospital, patient and isolate-level variables from NHSN annual survey and AR option were assessed as covariates. Variable entry into the models is based on significance level P &lt; .05. Results: Among 9,992 hospital-onset SA isolates from 9,019 patients in 315 facilities, 5,488 (54.9%) were MRSA. Logistic regression showed that a higher proportion of HO-MRSA was significantly associated with older age, female, particular sources of specimen (urine and LRT), and selected hospital characteristics: hospitals not serving as major teaching hospitals, hospitals with a higher proportion of MRSA among community-onset SA isolates, hospitals with lower percentage of beds in intensive care units, and hospitals outsourcing AST service (Table 1). Conclusions: HO-MRSA is independently associated with community burden of MRSA, older and female patient populations, and hospital teaching status and AST practices, which highlights the importance of public health engagement and regional collaborations to prevent MRSA. To provide a standardized MRSA proportion for public health surveillance, taking some of these factors into account in MRSA proportion standardization should be considered.Funding: NoneDisclosures: None

Administrative Coding Methods Impact Surgical Site Infection Rates

Background: Surveillance for surgical site infections (SSI) is recommended by the CDC. Currently, colon and abdominal hysterectomy SSI rates are publicly available and impact hospital reimbursement. However, the CDC NHSN allows surgical procedures to be abstracted based on International Classification of Diseases, Tenth Revision (ICD-10) or current procedural terminology (CPT) codes. We assessed the impact of using ICD and/or CPT codes on the number of cases abstracted and SSI rates. Methods: We retrieved administrative codes (ICD and/or CPT) for procedures performed at the University of Iowa Hospitals &amp; Clinics over 1 year: October 2018–September 2019. We included 10 procedure types: colon, hysterectomy, cesarean section, breast, cardiac, craniotomy, spinal fusion, laminectomy, hip prosthesis, and knee prosthesis surgeries. We then calculated the number of procedures that would be abstracted if we used different permutations in administration codes: (1) ICD codes only, (2) CPT codes only, (3) both ICD and CPT codes, and (4) at least 1 code from either ICD or CPT. We then calculated the impact on SSI rates based on any of the 4 coding permutations. Results: In total, 9,583 surgical procedures and 180 SSIs were detected during the study period using the fourth method (ICD or CPT codes). Denominators varied according to procedure type and coding method used. The number of procedures abstracted for breast surgery had a &gt;10-fold difference if reported based on ICD only versus ICD or CPT codes (104 vs 1,109). Hip prosthesis had the lowest variation (638 vs 767). For SSI rates, cesarean section showed almost a 3-fold increment (2.6% when using ICD only to 7.32% with both ICD &amp; CPT), whereas abdominal hysterectomy showed nearly a 2-fold increase (1.14% when using CPT only to 2.22% with both ICD &amp; CPT codes). However, SSI rates remained fairly similar for craniotomy (0.14% absolute difference), hip prosthesis (0.24% absolute difference), and colon (0.09% absolute difference) despite differences in the number of abstracted procedures and coding methods. Conclusions: Denominators and SSI rates vary depending on the coding method used. Variations in the number of procedures abstracted and their subsequent impact on SSI rates were not predictable. Variations in coding methods used by hospitals could impact interhospital comparisons and benchmarking, potentially leading to disparities in public reporting and hospital penalties.Funding: NoneDisclosures: None

Administrative coding methods impact surgical site infection rates.

We performed a retrospective analysis of the impact of using the International Classification of Diseases, Tenth Revision procedure coding system (ICD-10) or current procedural terminology (CPT) codes to calculate surgical site infection (SSI) rates. Denominators and SSI rates vary depending on the coding method used. The coding method used may influence interhospital performance comparisons.

Association of hospital-area deprivation with hospital performance on health care associated infection rates in 2018

Healthcare-associated infections (HAIs) are common and often preventable complications of care, with reduction emphasized in national policy. The Centers for Medicare and Medicaid Services introduced an HAI-focused Hospital Acquired Condition Reduction Program in 2015 to penalize poor-performing hospitals. Standardized infection ratios (SIRs) are used for comparisons between healthcare organizations, though they are not adjusted for socioeconomic risks known to impact infection. The objectives of this study were to assess the relationship between hospital-area deprivation with reported SIRs and reimbursement penalties. This was a cohort study using 2018 Hospital Compare, as well as area deprivation data and other hospital characteristics. Multivariable regression models were used to evaluate associations between hospital-area deprivation and SIR reporting as well as payment reduction, adjusting for case mix index and hospital ownership. Of the 2102 unique hospitals in our study, 12.8% reported at least one worse than national benchmark SIR and 23.7% had a payment reduction. After adjustment, there was a 17% increased risk of reporting worse than benchmarked SIRs with quartile increases in deprivation (95% confidence interval: 5%-30%, P = .004). Despite this, there were no significant relationships between reimbursement penalties and ADI (risk ratio: 1.00, 95% confidence interval: 0.997-1.005, P = .567). This study documented a significant relationship between hospital-area deprivation and the risk of reporting worse than national benchmark SIRs. Though this did not appear to translate to Hospital Acquired Condition Reduction Program penalties in this dataset, it reinforces problems with the current use of SIRs for interhospital comparisons.

Inter-hospital comparison of working time allocation among internal medicine residents using time-motion observations: an innovative benchmarking tool

ObjectivesThe vast majority of residents’ working time is spent away from patients. In hospital practice, many factors may influence the resident’s working day structure.Using an innovative method, we aimed to...

Agreement between units of measure for paediatric antibiotic utilisation surveillance using hospital pharmacy supply data.

Background:Drug utilisation studies from paediatric hospitals that do not have access to patient level data on medication use are limited by a lack of standardised units of measures that reflect the varying daily dosage requirements among patients. The World Health Organization’s defined daily dose is frequently used in adult hospitals for benchmarking and longitudinal analysis but is not endorsed for use in paediatric populations.Objective:Explore agreement between standard adult-based defined daily doses (DDD) and paediatric estimates of daily injectable antibiotic use in a Paediatric Intensive Care Unit that does not have access to individual patient-level data.Methods:Hospital pharmacy antibiotic use reports and age-specific occupied bed-day data from 1 January 2010 to 31 May 2016 were extracted. Paediatric reference dosages and frequencies for antibiotics were defined and applied to three paediatric units of measure. Measures were applied to extracted data, agreement between antibiotic use measured in the adult DDD and each of the paediatric measures was assessed visually via Bland-Altman plots and linear regression for each antibiotic.Results:Thirty one different antibiotics were used throughout the study period. Despite varying daily dosages in grams, the daily use of vials was unchanged from birth to 18 years for thirteen antibiotics. Agreement between DDD and vial-based measures was closer than the total recommended daily dose that did not account for wastage during preparation and administration. Vial-based measures were unaffected by vial size changes due to drug shortage.Conclusions:Agreement between the DDD and vial-based measures of use supports the use of DDD for select antibiotics that may be targeted by antimicrobial stewardship programs. Vial based measures should be further explored in hospitals with single vial policies; detailed understanding of hospital practice is needed before inter-hospital comparisons are made.

Drug utilization pattern in an intensive care unit of a teaching hospital in South India

Background: Polypharmacy and inappropriate usage of antibiotics are common in an Intensive Care Unit (ICU) which may increase morbidity, mortality, antimicrobial resistance and treatment cost. Hence, drug utilization research is crucial for measuring drug consumption using DDD/100 bed-days formula proposed by the WHO that would possibly be useful while formulating a comprehensive antibiotic policy for the institution and guide for future inter-hospital or institutional comparisons. Therefore, in this study, we proposed to evaluate the drug utilization patterns in the ICU.Methods: A prospective observational study was conducted for 3 months from June 2018 to August 2018, and the data were obtained from the ICU of a tertiary care hospital. The demographic data, disease data, the utilization of different classes of drugs (WHO-ATC classification) as well as individual drugs were recorded.Results:One hundred and twelve patient’s data were evaluated. About 90% patients were prescribed with antiulcer medications during their ICU stay followed by antibiotics in more than 89% patients. Azithromycin, levofloxacin, metronidazole, ofloxacin, ceftriaxone and amoxicillin/clavulanic acid were maximally utilized antibiotics with 24, 16.43, 14.27, 13.89, 12.22 and 10.97 DDD/100 bed-days respectively. Conclusions: Antiulcer medications were most commonly prescribed followed by antibiotics during ICU stay. Average numbers of drugs prescribed per patients were high. Regular prescription audit and modification of antibiotic policy is required to curtail the polypharmacy and inappropriate use of antibiotics.Background: Polypharmacy and inappropriate usage of antibiotics are common in an Intensive Care Unit (ICU) which may increase morbidity, mortality, antimicrobial resistance and treatment cost. Hence, drug utilization research is crucial for measuring drug consumption using DDD/100 bed-days formula proposed by the WHO that would possibly be useful while formulating a comprehensive antibiotic policy for the institution and guide for future inter-hospital or institutional comparisons. Therefore, in this study, we proposed to evaluate the drug utilization patterns in the ICU.Methods: A prospective observational study was conducted for 3 months from June 2018 to August 2018, and the data were obtained from the ICU of a tertiary care hospital. The demographic data, disease data, the utilization of different classes of drugs (WHO-ATC classification) as well as individual drugs were recorded.Results:One hundred and twelve patient’s data were evaluated. About 90% patients were prescribed with antiulcer medications during their ICU stay followed by antibiotics in more than 89% patients. Azithromycin, levofloxacin, metronidazole, ofloxacin, ceftriaxone and amoxicillin/clavulanic acid were maximally utilized antibiotics with 24, 16.43, 14.27, 13.89, 12.22 and 10.97 DDD/100 bed-days respectively.Conclusions: Antiulcer medications were most commonly prescribed followed by antibiotics during ICU stay. Average numbers of drugs prescribed per patients were high. Regular prescription audit and modification of antibiotic policy is required to curtail the polypharmacy and inappropriate use of antibiotics. Keywords: Antibiotics, Daily defined doses, Drug utilization, Intensive care unit

Unexpected discrepancies in hospital administrative databases can impact the accuracy of monitoring thyroid surgery outcomes in France.

ObjectiveTo determine the validity of hospital administrative databases compared to prospective collection of medical data assessing thyroid surgery complications.BackgroundAdministrative data are increasingly used to track surgical outcomes.MethodsAll patients undergoing thyroid surgery at three French university hospitals between April 2008 and April 2009 were prospectively included. Using diagnosis and procedural codes from hospital administrative database, we designed three indicators for measuring complications of thyroid surgery: recurrent laryngeal nerve palsy, postoperative hypoparathyroidism, and postoperative hemorrhage. Gold standard was obtained from a prospective collection of medical data after systematically screening each patient for the above-mentioned complications. Their ability to monitor surgical outcomes over time within individual hospitals was estimated using control charts. Spatial comparison between hospitals was performed by funnel plots.ResultsA total of 1909 patients were included. Complication rates extracted from administrative data were significantly lower compared to medical data (nerve palsy 2.4% vs. 6.7%, hypoparathyroidism 10.6% vs. 22.3%, p<0.0001). Indicator sensitivity was 30.4% for nerve palsy, 45.4% for hypoparathyroidism and 71.4% for postoperative hemorrhage. Corresponding positive predictive values were 84.4%, 95.1% and 68.2%. In two of the three hospitals, administrative data were not able to track temporal variations in complications rates. Regarding inter-hospital comparisons, 2 out of 3 hospitals were considered outliers according to administrative data despite having an average performance based on medical data.ConclusionsThe ability of indicators extracted from administrative databases to measure thyroid surgery outcomes depends on the quality of underlying data coding. Validation in every center should be a prerequisite before implementing such metrics for tracking performance

1820. Indirect Standardization: A Convenient Benchmarking Approach to Antibiotic Utilization Based on Patient Mix

BackgroundAntimicrobial stewardship programs are recommended to risk-adjust antimicrobial use in order to support intra- and inter-hospital comparisons. The purpose of our investigation was to evaluate a benchmarking strategy for its ability to accurately identify changes in risk-adjusted utilization as a result of known stewardship interventions.MethodsAntimicrobial drug use was measured as days of therapy (DOT) from billing records. Based on diagnosis-related group (DRG) assignment, we calculated expected (E) use determined by indirect standardization and compared with observed (O) use for three targeted groups of antimicrobials: anti-pseudomonals β-lactams; ceftriaxone; and fluoroquinolones. As a stewardship strategy, a clinical pharmacist-driven, individualized pseudomonal risk assessment based antibiotic prior authorization process was implemented in the third quarter of 2016, focusing on commonly encountered community-onset infections.ResultsThe 10-month time period prior to implementing the intervention was used to establish the benchmark with over 10,000 billing records. Utilization assigned to DRGs from this time period was used to predict expected utilization. As a result of the intervention, a decrease in anti-pseudomonal agent utilization at the cost of an increase in ceftriaxone utilization was observed (Figure 1), with the lack of a significant impact toward change in the utilization of the fluoroquinolones. Variability in use is explained by the treated patients within each DRG.ConclusionAntibiotic utilization was benchmarked to expected use adjusted for patient mix based on DRGs, and trends in changing antibiotic consumption were correctly identified. Differences between expected and observed use reflect usage patterns that take into consideration type of patients treated and provides the basis of evaluation of outcome measures for our stewardship interventions.Figure 1.O/E Ratios (95% CI) Calculated for the 12 Month Time Period Following Implementation of the Intervention.Disclosures All authors: No reported disclosures.

1873. Next Steps in Predicting Anti-MRSA Antibiotic Prescribing

BackgroundAntibiotic use metrics are utilized by antimicrobial stewardship programs to benchmark performance against peer institutions and inform stewardship efforts. Benchmarking requires risk adjustment for patient- and facility-level factors so that remaining differences are attributable only to prescribing practices. Antibiotics for the treatment of methicillin-resistant Staphylococcus aureus (MRSA) are one of the most frequently used drug classes. Our objective was to identify predictors of anti-MRSA antibiotic use in a nationwide network of hospitals.MethodsWe used data from inpatient encounters at facilities participating in the Vizient data repository between January 1, 2016 and December 31, 2016. The outcome, anti-MRSA antibiotic use, was calculated as days of therapy per patient-days present for each encounter. We constructed a multivariable negative binomial regression model and assessed the following predictors for inclusion: age, sex, race, ethnicity, diagnosis related groups (DRGs), ICU days, admit month, facility bed size, facility teaching status, and region. A clinical framework was used to categorize DRGs based on risk of anti-MRSA antibiotic use. A backwards stepwise approach was used to identify the final model. We evaluated predictor effect size and significance, and assessed model fit using a deviance-based pseudo-R2.ResultsOne hundred forty-five facilities representing 3,608,711 encounters met inclusion criteria. All predictors considered in our model were significant. Predictors with the greatest magnitude of association included DRG categories and patient age. The DRG categories with the strongest associations were DRGs for infections likely due to Staphylococcus aureus (RR = 1.66, P < 0.0001) or for diagnoses likely to receive long-term MRSA coverage (RR = 1.49, P < 0.0001). The age group with the strongest association was age 2–10 years (RR = 1.64; P < 0.001). The deviance-based pseudo-R2 of the final model was 0.19, indicating good model fit.ConclusionDRGs and patient-level characteristics can be utilized to account for variability in anti-MRSA antibiotic use beyond what is explained through facility-level characteristics. Incorporation of the significant predictors identified in this study may aid in more meaningful interhospital comparisons of anti-MRSA antibiotic use in both adults and pediatrics.Disclosures J. C. McGregor, Merck: Grant Investigator, Research grant.

Case mix adjustment of health outcomes, resource use and process indicators in childbirth care: a register-based study.

BackgroundUnwarranted variation in care practice and outcomes has gained attention and inter-hospital comparisons are increasingly being used to highlight and understand differences between hospitals. Adjustment for case mix is a prerequisite for meaningful comparisons between hospitals with different patient populations. The objective of this study was to identify and quantify maternal characteristics that impact a set of important indicators of health outcomes, resource use and care process and which could be used for case mix adjustment of comparisons between hospitals.MethodsIn this register-based study, 139 756 deliveries in 2011 and 2012 were identified in regional administrative systems from seven Swedish regions, which together cover 67 % of all deliveries in Sweden. Data were linked to the Medical birth register and Statistics Sweden’s population data. A number of important indicators in childbirth care were studied: Caesarean section (CS), induction of labour, length of stay, perineal tears, haemorrhage > 1000 ml and post-partum infections. Sociodemographic and clinical characteristics deemed relevant for case mix adjustment of outcomes and resource use were identified based on previous literature and based on clinical expertise. Adjustment using logistic and ordinary least squares regression analysis was performed to quantify the impact of these characteristics on the studied indicators.ResultsAlmost all case mix factors analysed had an impact on CS rate, induction rate and length of stay and the effect was highly statistically significant for most factors. Maternal age, parity, fetal presentation and multiple birth were strong predictors of all these indicators but a number of additional factors such as born outside the EU, body mass index (BMI) and several complications during pregnancy were also important risk factors. A number of maternal characteristics had a noticeable impact on risk of perineal tears, while the impact of case mix factors was less pronounced for risk of haemorrhage > 1000 ml and post-partum infections.ConclusionsMaternal characteristics have a large impact on care process, resource use and outcomes in childbirth care. For meaningful comparisons between hospitals and benchmarking, a broad spectrum of sociodemographic and clinical maternal characteristics should be accounted for.Electronic supplementary materialThe online version of this article (doi:10.1186/s12884-016-0921-0) contains supplementary material, which is available to authorized users.

Failure of Colorectal Surgical Site Infection Predictive Models Applied to an Independent Dataset: Do They Add Value or Just Confusion?

Colorectal surgical site infections (C-SSIs) are a major source of postoperative morbidity. Institutional C-SSI rates are modeled and scrutinized, and there is increasing movement in the direction of public reporting. External validation of C-SSI risk prediction models is lacking. Factors governing C-SSI occurrence are complicated and multifactorial. We hypothesized that existing C-SSI prediction models have limited ability to accurately predict C-SSI in independent data. Colorectal resections identified from our institutional ACS-NSQIP dataset (2006 to 2014) were reviewed. The primary outcome was any C-SSI according to the ACS-NSQIP definition. Emergency cases were excluded. Published C-SSI risk scores: the National Nosocomial Infection Surveillance (NNIS), Contamination, Obesity, Laparotomy, and American Society of Anesthesiologists (ASA) class (COLA), Preventie Ziekenhuisinfecties door Surveillance (PREZIES), and NSQIP-based models were compared with receiver operating characteristic (ROC) analysis to evaluate discriminatory quality. There were 2,376 cases included, with an overall C-SSI rate of 9% (213 cases). None of themodels produced reliable and high quality C-SSI predictions. For any C-SSI, the NNIS c-index was 0.57 vs 0.61 for COLA, 0.58 for PREZIES, and 0.62 for NSQIP: all well below the minimum "reasonably" predictive c-index of 0.7. Predictions for superficial, deep, and organ space SSI were similarly poor. Published C-SSI risk prediction models do not accurately predict C-SSI in our independent institutional dataset. Application of externally developed prediction models to any individual practice must be validated or modified to account for institution and case-mix specific factors. This questions the validity of using externally or nationally developed models for "expected" outcomes and interhospital comparisons.

Community-acquired pneumonia (CAP) hospitalizations and deaths: is there a role for quality improvement through inter-hospital comparisons?

To assess between-hospital variations in standardized in-hospital mortality ratios of community-acquired pneumonia (CAP), and identify possible leads for quality improvement. We used an administrative database to estimate standardized in-hospital mortality ratios for 111 Belgian hospitals, by carrying out a set of hierarchical logistic regression models, intended to disentangle therapeutic attitudes and biases. To facilitate the detection of false-negative/positive results, we added an inconclusive zone to the funnel plots, derived from the results of the study. Data quality was validated by comparison with (i) alternative data from the largest Belgian Sickness Fund, (ii) published German hospital data and (iii) the results of an on-site audit. All Belgian hospital discharge records from 2004 to 2007. A total of 111 776 adult patients were admitted for CAP. Risk-adjusted standardized in-hospital mortality ratios. Out of the 111 hospitals, we identified five and six outlying hospitals, with standardized mortality ratios of CAP consistently on the extremes of the distribution, as providing possibly better or worse care, respectively, and 18 other hospitals as having possible quality weaknesses/strengths. At the individuals' level of the analysis, adjusted odds ratios showed the paramount importance of old age, comorbidity and mechanical ventilation. The data compared well with the different validation sources. Despite the limitations inherent to administrative data, it seemed possible to establish inter-hospital differences in standardized in-hospital mortality ratios of CAP and to identify leads for quality improvement. Monitoring is needed to assess progress in quality.

Are hospitals in Japan with larger patient volume treating younger and earlier-stage cancer patients? An analysis of hospital-based cancer registry data in Japan.

Differences in hospital case-mix have not been adequately accounted for in hospital volume and patient outcome studies in Japan. We aimed to examine whether differences may exist by investigating the distribution of patients' stage and age across designated cancer treatment hospitals of varying patient volume across Japan. We analyzed data of gastric, breast, colorectal, lung and liver cancer patients who were included in the national database of hospital-based cancer registries between 2008 and 2011. We investigated the association between hospital volume, cancer stage and patient age. Hospitals were classified into five groups according to patient volume. In total, 676 713 patients met the inclusion criteria. The proportion of patients with early-stage (tumor-node-metastasis Stage 0 or I) cancer was higher among high-volume hospitals for all cancer types except small cell lung cancer. The proportion of older patients (age >75 years) was smaller among high-volume hospitals for all cancer types. The difference in the proportion of patients with early-stage cancers between very low-volume and very high-volume hospitals was greatest for non-small cell lung cancer (26.5% for very low and 43.5% for very high). This difference for the proportion of older patients was also greatest for non-small cell lung cancer (48.9% for very low and 30.3% for very high). We showed that the proportions of early-stage cancer patients and younger patients are greater in higher-volume hospitals compared with lower-volume hospitals in Japan. Researchers conducting volume-outcome studies and policymakers analyzing hospital performance should be cautious when making interhospital comparisons.

Complication rates are reduced for revision adult spine deformity surgery among high-volume hospitals and surgeons

Background contextRevision adult spinal deformity surgery (RASDS) is a particularly high-risk intervention. PurposeThe aim was to assess complication rates in RASDS by surgeon and hospital operative volume. Study design/settingThis was a retrospective analysis of prospectively collected data. Patient sampleBased on a Nationwide Inpatient Sample (NIS) database (2001–2010), patients aged older than 21 years (International Classification of Diseases, Ninth Revision, Clinical Modification) with spine arthrodesis for scoliosis were included. For longitudinal analysis, the 2008-2011 New York State Inpatient Database (NY SID) was queried. Outcome measuresThe outcome measures included complication rate after RASDS. MethodsCases were identified as primary or revision surgery with or without osteotomy performed. Annual surgeon and hospital volumes were stratified into quartiles via identifier codes. Case complexity was determined using a novel operative complexity index, based on available NIS operative parameters: levels fused, approach, osteotomy, and revision status. The primary end point was morbidity during the hospital stay. New York State Inpatient Database analysis allowed for identification of rate of reoperation for infection or pseudarthrosis/implant failure. One-way analysis of variance was used to assess continuous measures, chi-square for categorical measures. ResultsOf 139,150 adult spinal deformity surgery (ASDS) cases, 4,888 revision with hospital identifiers and 1,978 with surgeon identifiers were identified. Higher-volume surgeons performed more revision cases and cases requiring osteotomy. With increasing hospital volume, complication rate for RASDS decreased (9.7% vs. 12.9% at highest- vs. lowest-volume centers, p< .001). The highest-volume surgeons showed significant decreases in the rate of major complications for RASDS (8.8% vs. 10.7% for lowest-volume surgeons, p< .001). A similar trend was observed for ASDS cases requiring osteotomy. Multiple logistic regression analysis showed that the highest-volume hospitals and surgeons showed a reduced odds ratio for all complications compared with lowest-volume hospitals. For the NY SID, 528 RASDS cases indicated reoperation rates for infection and pseudarthrosis/implant failure after RASDS were increased for the lowest-volume hospitals and surgeons. ConclusionsPerioperative complication rate associated with RASDS is lower when patients are treated by high-volume surgeons at high-volume centers. As complex cases requiring osteotomy and combined approaches are more frequent at high-volume centers, an operative complexity index helps predict the likelihood of volume-dependent complication rates. Future interhospital and intersurgeon comparisons should account for these case characteristics so that similar case complexity is compared in these analyses.