Search

BackgroundIn the context of the Affordable Care Act, there is extensive emphasis on making provider quality transparent and publicly available. Online public reports of quality exist, but little is known about how visitors find reports or about their purpose in visiting. ObjectiveTo address this gap, we gathered website analytics data from a national group of online public reports of hospital or physician quality and surveyed real-time visitors to those websites. MethodsWebsites were recruited from a national group of online public reports of hospital or physician quality. Analytics data were gathered from each website: number of unique visitors, method of arrival for each unique visitor, and search terms resulting in visits. Depending on the website, a survey invitation was launched for unique visitors on landing pages or on pages with quality information. Survey topics included type of respondent (eg, consumer, health care professional), purpose of visit, areas of interest, website experience, and demographics. ResultsThere were 116,657 unique visitors to the 18 participating websites (1440 unique visitors/month per website), with most unique visitors arriving through search (63.95%, 74,606/116,657). Websites with a higher percent of traffic from search engines garnered more unique visitors (P=.001). The most common search terms were for individual hospitals (23.25%, 27,122/74,606) and website names (19.43%, 22,672/74,606); medical condition terms were uncommon (0.81%, 605/74,606). Survey view rate was 42.48% (49,560/116,657 invited) resulting in 1755 respondents (participation rate=3.6%). There were substantial proportions of consumer (48.43%, 850/1755) and health care professional respondents (31.39%, 551/1755). Across websites, proportions of consumer (21%-71%) and health care professional respondents (16%-48%) varied. Consumers were frequently interested in using the information to choose providers or assess the quality of their provider (52.7%, 225/427); the majority of those choosing a provider reported that they had used the information to do so (78%, 40/51). Health care professional (26.6%, 115/443) and consumer (20.8%, 92/442) respondents wanted cost information and consumers wanted patient narrative comments (31.5%, 139/442) on the public reports. Health care professional respondents rated the experience on the reports higher than consumers did (mean 7.2, SD 2.2 vs mean 6.2, SD 2.7; scale 0-10; P

BackgroundOpioids and multimodal analgesia are widely administered to manage postoperative pain. However, little is known on how improvements in inpatient pain control are correlated with high-risk (> 90 daily OME) discharge opioid prescriptions for opioid naïve surgical patients.MethodsWe conducted a retrospective observational study of adult opioid-naïve patients undergoing surgery from June 2012 through December 2018 at a large academic medical center. We used multivariate logistic regression to assess whether multimodal analgesic drugs consumed in the 24 h prior to discharge was associated with a reduction in high-risk opioid discharge prescriptions. We identified other risk factors for receiving a high-risk discharge opioid prescription.ResultsAmong the 32,511 patients, 83% of patients were discharged with an opioid prescription. In 2013, 34.1% of patients with a discharge opioid prescription received a high-risk prescription and this declined to 17.7% by 2018. Use of multimodal analgesic agents during the final 24 h of hospitalization increased each year, with over 80% receiving at least one multimodal analgesic agent by 2018. The median OME consumed in the 24 h prior to discharge peaked in 2013 at 31 and steadily decreased to 19.8 by 2018. There was a significant association between the use of acetaminophen in the 24 h prior to discharge and a high-risk prescription at discharge (p < 0.01). OMEs consumed in the 24 h prior to discharge was a significant predictor of receiving a high-risk discharge prescription, even at low doses. Other factors associated with receipt of a high-risk discharge opioid prescription included male gender, race, history of anxiety disorder, and discharge service.DiscussionUse of multimodal analgesia regimens in hospitalized surgical patients in the 24 h prior to hospital discharge increased between 2012 and 2018. Simultaneously, opioid use prior to hospital discharge decreased. Despite these gains, approximately one in five discharge prescriptions was high-risk (> 90 daily OME). In addition, we found that prescribing of discharge opioids above inpatient opioid requirements remains common in opioid naive surgical patients.ConclusionProviders should account for pre-discharge opioid consumption and use of multimodal analgesia when considering the total and daily OME's that may be appropriate for an individual surgical patient on the discharge opioid prescription.

To evaluate whether different approaches in note text preparation (known as preprocessing) can impact machine learning model performance in the case of mortality prediction ICU.DesignClinical note text was used to build machine learning models for adults admitted to the ICU. Preprocessing strategies studied were none (raw text), cleaning text, stemming, term frequency-inverse document frequency vectorization, and creation of n-grams. Model performance was assessed by the area under the receiver operating characteristic curve. Models were trained and internally validated on University of California San Francisco data using 10-fold cross validation. These models were then externally validated on Beth Israel Deaconess Medical Center data.SettingICUs at University of California San Francisco and Beth Israel Deaconess Medical Center.SubjectsTen thousand patients in the University of California San Francisco training and internal testing dataset and 27,058 patients in the external validation dataset, Beth Israel Deaconess Medical Center.InterventionsNone.Measurements and main resultsMortality rate at Beth Israel Deaconess Medical Center and University of California San Francisco was 10.9% and 7.4%, respectively. Data are presented as area under the receiver operating characteristic curve (95% CI) for models validated at University of California San Francisco and area under the receiver operating characteristic curve for models validated at Beth Israel Deaconess Medical Center. Models built and trained on University of California San Francisco data for the prediction of inhospital mortality improved from the raw note text model (AUROC, 0.84; CI, 0.80-0.89) to the term frequency-inverse document frequency model (AUROC, 0.89; CI, 0.85-0.94). When applying the models developed at University of California San Francisco to Beth Israel Deaconess Medical Center data, there was a similar increase in model performance from raw note text (area under the receiver operating characteristic curve at Beth Israel Deaconess Medical Center: 0.72) to the term frequency-inverse document frequency model (area under the receiver operating characteristic curve at Beth Israel Deaconess Medical Center: 0.83).ConclusionsDifferences in preprocessing strategies for note text impacted model discrimination. Completing a preprocessing pathway including cleaning, stemming, and term frequency-inverse document frequency vectorization resulted in the preprocessing strategy with the greatest improvement in model performance. Further study is needed, with particular emphasis on how to manage author implicit bias present in note text, before natural language processing algorithms are implemented in the clinical setting.

ObjectiveDescribe NICU admission rate variation among hospitals in infants with birthweight ≥2500 g and low illness acuity, and describe factors that predict NICU admission.Study designRetrospective study from the Vizient Clinical Data Base/Resource Manager®. Support vector machine methodology was used to develop statistical models using (1) patient characteristics (2) only the indicator for the inborn hospital and (3) patient characteristics plus indicator for the inborn hospital.ResultsNICU admission rates of 427,449 infants from 154 hospitals ranged from 0 to 28.6%. C-statistics for the patient characteristics model: 0.64 (Confidence Interval (CI) 0.62-0.65), hospital only model: 0.81 (CI, 0.81-0.82), and patient characteristic plus hospital variable model: 0.84 (CI, 0.83-0.84).Conclusion/relevanceThere is wide variation in NICU admission rates in infants with low acuity diagnoses. In all cohorts, birth hospital better predicted NICU admission than patient characteristics alone.

PurposeDelaying cataract surgery is associated with an increased risk of falls, but whether routine preoperative testing delays cataract surgery long enough to cause clinical harm is unknown. We sought to determine whether the use of routine preoperative testing leads to harm in the form of delayed surgery and falls in Medicare beneficiaries awaiting cataract surgery.DesignRetrospective, observational cohort study using 2006-2014 Medicare claims.ParticipantsMedicare beneficiaries 66+ years of age with a Current Procedural Terminology claim for ocular biometry.MethodsWe measured the mean and median number of days between biometry and cataract surgery, calculated the proportion of patients waiting ≥ 30 days or ≥ 90 days for surgery, and determined the odds of sustaining a fall within 90 days of biometry among patients of high-testing physicians (testing performed in ≥ 75% of their patients) compared with patients of low-testing physicians. We also estimated the number of days of delay attributable to high-testing physicians.Main outcome measuresIncidence of falls occurring between biometry and surgery, odds of falling within 90 days of biometry, and estimated delay associated with physician testing behavior.ResultsOf 248 345 beneficiaries, 16.4% were patients of high-testing physicians. More patients of high-testing physicians waited ≥ 30 days and ≥ 90 days to undergo surgery (31.4% and 8.2% vs. 25.0% and 5.5%, respectively; P < 0.0001 for both). Falls before surgery in patients of high-testing physicians increased by 43% within the 90 days after ocular biometry (1.0% vs. 0.7%; P < 0.0001). The adjusted odds ratio of falling within 90 days of biometry in patients of high-testing physicians versus low-testing physicians was 1.10 (95% confidence interval [CI], 1.03-1.19; P = 0.008). After adjusting for surgical wait time, the odds ratio decreased to 1.07 (95% CI, 1.00-1.15; P = 0.06). The delay associated with having a high-testing physician was approximately 8 days (estimate, 7.97 days; 95% CI, 6.40-9.55 days; P < 0.0001). Other factors associated with delayed surgery included patient race (non-White), Northeast region, ophthalmologist ≤ 40 years of age, and low surgical volume.ConclusionsOveruse of routine preoperative medical testing by high-testing physicians is associated with delayed surgery and increased falls in cataract patients awaiting surgery.

To develop a method for calculating rates of testing for breast cancer recurrence in patients who have already undergone initial treatment for breast cancer, we calculated rates in a cohort of Medicare breast cancer patients and an age-matched noncancer cohort. We first used only tests with claims including diagnosis codes indicating invasive breast cancer and then used all tests regardless of diagnosis code. For each method, we calculated testing rates in the breast cancer cohort above the background rate in the noncancer population. The two methods provided similar estimates of testing prevalence and frequency, with exception of prevalence of CT.

Background As patients with congenital heart disease (CHD) are living longer, understanding the comorbidities they develop as they age is increasingly important. However, there are no published population-based estimates of the comorbidity burden among the US adult patients with CHD. Methods and Results Using the IBM MarketScan commercial claims database from 2010 to 2016, we identified adults aged ≥18 years with CHD and 2 full years of continuous enrollment. These were frequency matched with adults without CHD within categories jointly defined by age, sex, and dates of enrollment in the database. A total of 40 127 patients with CHD met the inclusion criteria (mean [SD] age, 36.8 [14.6] years; and 48.2% were women). Adults with CHD were nearly twice as likely to have any comorbidity than those without CHD (P

IntroductionDementia is among the costliest of medical conditions, but it is not known how these costs vary by dementia subtype.MethodsThe effect of dementia diagnosis subtype on direct health care costs and utilization was estimated using 2015 California Medicare fee-for-service data. Potential drivers of increased costs in Lewy body dementia (LBD), in comparison to Alzheimer's disease, were tested.Results3,001,987 Medicare beneficiaries were identified, of which 8.2% had a dementia diagnosis. Unspecified dementia was the most common diagnostic category (59.6%), followed by Alzheimer's disease (23.2%). LBD was the costliest subtype to Medicare, on average, followed by vascular dementia. The higher costs in LBD were explained in part by falls, urinary incontinence or infection, depression, anxiety, dehydration, and delirium.DiscussionDementia subtype is an important predictor of health care costs. Earlier identification and targeted treatment might mitigate the costs associated with co-occurring conditions in LBD.

Introduction/objectivesPneumocystis jirovecii pneumonia (PJP) is a rare but potentially fatal opportunistic infection; however, consensus varies around which conditions or medications confer a level of risk sufficient to justify antibiotic prophylaxis for PJP. We used electronic health record (EHR) data to assess the current patterns of PJP prophylaxis, PJP outcomes, and prophylaxis-related adverse events among patients with rheumatic diseases who were receiving high-risk immunosuppressant drugs.MethodsData derive from the EHR of a large health system. We included new immunosuppressant users with diagnoses of vasculitis, myositis, or systemic lupus erythematosus. We calculated the proportion of patients who received PJP prophylaxis for each diagnosis and drug combination. We also calculated the number of PJP infections and the number of antibiotic adverse drug events (ADEs) per patient-year of exposure.ResultsWe followed 316 patients for 23.2 + /- 14.2 months. Overall, 124 (39%) of patients received prophylactic antibiotics for PJP. At least 25% of patients with the highest risk conditions (e.g. vasculitis) or highest risk immunosuppressants (e.g. cyclophosphamide) did not receive PJP prophylaxis. We found no cases of PJP infection over 640 patient-years of follow up, including among those not receiving prophylaxis, and an overall incidence rate of ADEs of 2.2% per patient-year.ConclusionsPJP prophylaxis for patients with rheumatic conditions is inconsistent, with one quarter of patients who have high risk conditions or high risk immunosuppressants not receiving prophylaxis. However, given extremely low rates of PJP infection, but detectable ADEs to prophylactic antibiotics, our findings suggest that evidence to guide more personalized risk assessments are needed to inform PJP prophylaxis.

Hepatitis B virus (HBV) reactivation in the setting of rituximab use is a potentially fatal but preventable safety event. The rate of HBV screening and proportion of patients at risk who receive antiviral prophylaxis in patients initiating rituximab is unknown.We analyzed electronic health record (EHR) data from 2 health systems, a university center and a safety net health system, including diagnosis grouper codes, problem lists, medications, laboratory results, procedures codes, clinical encounter notes, and scanned documents. We identified all patients who received rituximab between 6/1/2012 and 1/1/2016. We calculated the proportion of rituximab users with inadequate screening for HBV according to the Centers for Disease Control guidelines for detecting latent HBV infection before their first rituximab infusion during the study period. We also assessed the proportion of patients with positive hepatitis B screening tests who were prescribed antiviral prophylaxis. Finally, we characterized safety failures and adverse events.We included 926 patients from the university and 132 patients from the safety net health system. Sixty-one percent of patients from the university had adequate screening for HBV compared with 90% from the safety net. Among patients at risk for reactivation based on results of HBV testing, 66% and 92% received antiviral prophylaxis at the university and safety net, respectively.We found wide variations in hepatitis B screening practices among patients receiving rituximab, resulting in unnecessary risks to patients. Interventions should be developed to improve patient safety procedures in this high-risk patient population.

ImportanceDespite the significant contribution of surgical spending to health care costs, most surgeons are unaware of their operating room costs.ObjectiveTo examine the association between providing surgeons with individualized cost feedback and surgical supply costs in the operating room.Design, setting, and participantsThe OR Surgical Cost Reduction (OR SCORE) project was a single-health system, multihospital, multidepartmental prospective controlled study in an urban academic setting. Intervention participants were attending surgeons in orthopedic surgery, otolaryngology-head and neck surgery, and neurological surgery (n = 63). Control participants were attending surgeons in cardiothoracic surgery, general surgery, vascular surgery, pediatric surgery, obstetrics/gynecology, ophthalmology, and urology (n = 186).InterventionsFrom January 1 to December 31, 2015, each surgeon in the intervention group received standardized monthly scorecards showing the median surgical supply direct cost for each procedure type performed in the prior month compared with the surgeon's baseline (July 1, 2012, to November 30, 2014) and compared with all surgeons at the institution performing the same procedure at baseline. All surgical departments were eligible for a financial incentive if they met a 5% cost reduction goal.Main outcomes and measuresThe primary outcome was each group's median surgical supply cost per case. Secondary outcome measures included total departmental surgical supply costs, case mix index-adjusted median surgical supply costs, patient outcomes (30-day readmission, 30-day mortality, and discharge status), and surgeon responses to a postintervention study-specific health care value survey.ResultsThe median surgical supply direct costs per case decreased 6.54% in the intervention group, from $1398 (interquartile range [IQR], $316-$5181) (10 637 cases) in 2014 to $1307 (IQR, $319-$5037) (11 820 cases) in 2015. In contrast, the median surgical supply direct cost increased 7.42% in the control group, from $712 (IQR, $202-$1602) (16 441 cases) in 2014 to $765 (IQR, $233-$1719) (17 227 cases) in 2015. This decrease represents a total savings of $836 147 in the intervention group during the 1-year study. After controlling for surgeon, department, patient demographics, and clinical indicators in a mixed-effects model, there was a 9.95% (95% CI, 3.55%-15.93%; P = .003) surgical supply cost decrease in the intervention group over 1 year. Patient outcomes were equivalent or improved after the intervention, and surgeons who received scorecards reported higher levels of cost awareness on the health care value survey compared with controls.Conclusions and relevanceCost feedback to surgeons, combined with a small departmental financial incentive, was associated with significantly reduced surgical supply costs, without negatively affecting patient outcomes.

BackgroundProviders wish to help patients with prescription costs but often lack drug cost information. We examined whether giving providers formulary and drug cost information was associated with changes in their diabetes patients' drug costs and use. We conducted a longitudinal non-randomized evaluation of the web-based Prescribing Guide ( www.PrescribingGuide.com ), a free resource available to Hawaii's providers since 2006, which summarizes the formularies and copayments of six health plans for drugs to treat 16 common health conditions. All adult primary care physicians in Hawaii were offered the Prescribing Guide, and providers who enrolled received a link to the website and regular hardcopy updates.MethodsWe analyzed prescription claims from a large health plan in Hawaii for 5,883 members with diabetes from 2007 (baseline) to 2009 (follow-up). Patients were linked to 299 "main prescribing" providers, who on average, accounted for >88 % of patients' prescriptions and drug costs. We compared changes in drug costs and use for "study" patients whose main provider enrolled to receive the Prescribing Guide, versus "control" patients whose main provider did not enroll to receive the Prescribing Guide.ResultsIn multivariate analyses controlling for provider specialty and clustering of patients by providers, both patient groups experienced similar increases in number of prescriptions (+3.2 vs. +2.7 increase, p = 0.24), and days supply of medications (+141 vs. +129 increase, p = 0.40) averaged across all drugs. Total and out-of-pocket drug costs also increased for both control and study patients. However, control patients showed higher increases in yearly total drug costs of $208 per patient (+$792 vs. +$584 increase, p = 0.02) and in 30-day supply costs (+$9.40 vs. +$6.08 increase, p = 0.03). Both groups experienced similar changes in yearly out-of-pocket costs (+$41 vs + $31 increase, p = 0.36) and per 30-day supply (-$0.23 vs. -$0.19 decrease, p = 0.996).ConclusionGiving formulary and drug cost information to providers was associated with lower increases in total drug costs but not with lower out-of-pocket costs or greater medication use. Insurers and health information technology businesses should continue to increase providers' access to formulary and drug cost information at the point of care.

ObjectivePatient-centred care has become a priority in many countries. It is unknown whether current tools capture aspects of care patients and their surrogates consider important. We investigated whether online narrative reviews from patients and surrogates reflect domains in the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) and we described additional potential domains.DesignWe used thematic analysis to assess online narrative reviews for reference to HCAHPS domains and salient non-HCAHPS domains and compared results by reviewer type (patient vs surrogate).SettingWe identified hospitals for review from the American Hospital Association database using a stratified random sampling approach. This approach ensured inclusion of reviews of a diverse set of hospitals. We searched online in February 2013 for narrative reviews from any source for each hospital.ParticipantsWe included up to two narrative reviews for each hospital.ExclusionsOutpatient or emergency department reviews, reviews from self-identified hospital employees, or reviews of

ObjectivesTo characterize patients who received tracheostomies for airway compromise or were initiated on long-term ventilation for chronic respiratory failure in PICUs and to examine variation in the incidence of initiation, patient characteristics, and modalities across sites.DesignRetrospective cross-sectional analysis.SettingsSeventy-three North American PICUs that participated in the Virtual Pediatric Systems, LLC.PatientsPICU patients admitted between 2009 and 2011.InterventionsNone.Measurements and main resultsAmong 115,437 PICU patients, 1.8% received a tracheostomy or were initiated on long-term ventilation; 1,034 received a tracheostomy only, 717 were initiated on invasive ventilation, and 381 were initiated on noninvasive ventilation. Ninety percent had substantial chronic conditions and comorbidities, including more than 50% with moderate or worse cerebral disability upon discharge. Seven percent were initiated after a catastrophic injury/event. Across sites, there was variation in incidence of tracheotomy and initiation of long-term ventilation, ranging from 0% to 4.6%. There also was variation in patient characteristics, time to tracheotomy, number of extubations prior to tracheostomy, and the use of invasive ventilation versus noninvasive ventilation.ConclusionsAlthough the PICU incidence of initiation of tracheostomies and long-term ventilation was relatively uncommon, it suggests that thousands of children and young adults receive these interventions each year in North American PICUs. The majority of them have conditions and comorbidities that impose on-going care needs, beyond those required by artificial airways and long-term ventilation themselves.

ImportanceCommercial virtual visits are an increasingly popular model of health care for the management of common acute illnesses. In commercial virtual visits, patients access a website to be connected synchronously-via videoconference, telephone, or webchat-to a physician with whom they have no prior relationship. To date, whether the care delivered through those websites is similar or quality varies among the sites has not been assessed.ObjectiveTo assess the variation in the quality of urgent health care among virtual visit companies.Design, setting, and participantsThis audit study used 67 trained standardized patients who presented to commercial virtual visit companies with the following 6 common acute illnesses: ankle pain, streptococcal pharyngitis, viral pharyngitis, acute rhinosinusitis, low back pain, and recurrent female urinary tract infection. The 8 commercial virtual visit websites with the highest web traffic were selected for audit, for a total of 599 visits. Data were collected from May 1, 2013, to July 30, 2014, and analyzed from July 1, 2014, to September 1, 2015.Main outcomes and measuresCompleteness of histories and physical examinations, the correct diagnosis (vs an incorrect or no diagnosis), and adherence to guidelines of key management decisions.ResultsSixty-seven standardized patients completed 599 commercial virtual visits during the study period. Histories and physical examinations were complete in 417 visits (69.6%; 95% CI, 67.7%-71.6%); diagnoses were correctly named in 458 visits (76.5%; 95% CI, 72.9%-79.9%), and key management decisions were adherent to guidelines in 325 visits (54.3%; 95% CI, 50.2%-58.3%). Rates of guideline-adherent care ranged from 206 visits (34.4%) to 396 visits (66.1%) across the 8 websites. Variation across websites was significantly greater for viral pharyngitis and acute rhinosinusitis (adjusted rates, 12.8% to 82.1%) than for streptococcal pharyngitis and low back pain (adjusted rates, 74.6% to 96.5%) or ankle pain and recurrent urinary tract infection (adjusted rates, 3.4% to 40.4%). No statistically significant variation in guideline adherence by mode of communication (videoconference vs telephone vs webchat) was found.Conclusions and relevanceSignificant variation in quality was found among companies providing virtual visits for management of common acute illnesses. More variation was found in performance for some conditions than for others, but no variation by mode of communication.

ObjectivesNarrative review of the impact of pay-for-performance (P4P) and public reporting (PR) on health care outcomes, including spillover effects and impact on disparities.Principal findingsThe impact of P4P and PR is dependent on the underlying payment system (fee-for-service, salary, capitation) into which these schemes are introduced. Both have the potential to improve care, but they can also have substantial unintended consequences. Evidence from the behavioral economics literature suggests that individual physicians will vary in how they respond to incentives. We also discuss issues to be considered when including patient-reported outcome measures (PROMs) or patient-reported experience measures into P4P and PR schemes.ConclusionWe provide guidance to payers and policy makers on the design of P4P and PR programs so as to maximize their benefits and minimize their unintended consequences. These include involving clinicians in the design of the program, taking into account the payment system into which new incentives are introduced, designing the structure of reward programs to maximize the likelihood of intended outcomes and minimize the likelihood of unintended consequences, designing schemes that minimize the risk of increasing disparities, providing stability of incentives over some years, and including outcomes that are relevant to patients' priorities. In addition, because of the limitations of PR and P4P as effective interventions in their own right, it is important that they are combined with other policies and interventions intended to improve quality to maximize their likely impact.

BackgroundHospital report cards and financial incentives linked to performance require clinical data that are reliable, appropriate, timely, and cost-effective to process. Pay-for-performance plans are transitioning to automated electronic health record (EHR) data as an efficient method to generate data needed for these programs.ObjectiveTo determine how well data from automated processing of structured fields in the electronic health record (AP-EHR) reflect data from manual chart review and the impact of these data on performance rewards.Research designCross-sectional analysis of performance measures used in a cluster randomized trial assessing the impact of financial incentives on guideline-recommended care for hypertension.SubjectsA total of 2840 patients with hypertension assigned to participating physicians at 12 Veterans Affairs hospital-based outpatient clinics. Fifty-two physicians and 33 primary care personnel received incentive payments.MeasuresOverall, positive and negative agreement indices and Cohen's kappa were calculated for assessments of guideline-recommended antihypertensive medication use, blood pressure (BP) control, and appropriate response to uncontrolled BP. Pearson's correlation coefficient was used to assess how similar participants' calculated earnings were between the data sources.ResultsBy manual chart review data, 72.3% of patients were considered to have received guideline-recommended antihypertensive medications compared with 65.0% by AP-EHR review (κ=0.51). Manual review indicated 69.5% of patients had controlled BP compared with 66.8% by AP-EHR review (κ=0.87). Compared with 52.2% of patients per the manual review, 39.8% received an appropriate response by AP-EHR review (κ=0.28). Participants' incentive payments calculated using the 2 methods were highly correlated (r≥0.98). Using the AP-EHR data to calculate earnings, participants' payment changes ranged from a decrease of $91.00 (-30.3%) to an increase of $18.20 (+7.4%) for medication use (interquartile range, -14.4% to 0%) and a decrease of $100.10 (-31.4%) to an increase of $36.40 (+15.4%) for BP control or appropriate response to uncontrolled BP (interquartile range, -11.9% to -6.1%).ConclusionsPay-for-performance plans that use only EHR data should carefully consider the measures and the structure of the EHR before data collection and financial incentive disbursement. For this study, we feel that a 10% difference in the total amount of incentive earnings disbursed based on AP-EHR data compared with manual review is acceptable given the time and resources required to abstract data from medical records.

Background and objectivesAnemia guidelines for CKD recommend withholding intravenous iron in the setting of active infection, although no data specifically support this recommendation. This study aimed to examine the association between intravenous iron and clinical outcomes among hemodialysis patients hospitalized for infection.Design, setting, participants, & measurementsThis was a retrospective observational cohort study using data from the US Renal Data System of 22,820 adult Medicare beneficiaries on in-center hemodialysis who had received intravenous iron in the 14 days preceding their first hospitalization for bacterial infection in 2010. In multivariable analyses, the association between receipt of intravenous iron at any point from the day of hospital admission to discharge and all-cause 30-day mortality, mortality in 2010, length of hospital stay, and readmission for infection or death within 30 days of discharge was evaluated.ResultsThere were 2463 patients (10.8%) who received intravenous iron at any point from the day of admission to discharge. Receipt of intravenous iron was not associated with age, dialysis vintage, or comorbidities. There were 2618 deaths within 30 days of admission and 6921 deaths in 2010 (median follow-up 173 days; 25th and 75th percentiles, 78-271 days). The median length of stay was 7 days (25th and 75th percentiles, 5-12 days). Receipt of intravenous iron was not associated with higher 30-day mortality (odds ratio, 0.86; 95% confidence interval [95% CI], 0.74 to 1.00), higher mortality in 2010 (hazard ratio, 0.92; 95% CI, 0.85 to 1.00), longer mean length of stay (10.1 days [95% CI, 9.7 to 10.5] versus 10.5 days [95% CI, 10.3 to 10.7]; P=0.05), or readmission for infection or death within 30 days of discharge (odds ratio, 1.08; 95% CI, 0.96 to 1.22) compared with no receipt of intravenous iron.ConclusionsThis analysis does not support withholding intravenous iron upon admission for bacterial infection in hemodialysis patients, although clinical trials are required to make definitive recommendations.

ObjectiveTo recommend methods for assessing quality of care via patient-reported outcome-based performance measures (PRO-PMs) of symptoms, functional status, and quality of life.MethodsA Technical Expert Panel was assembled by the American Medical Association-convened Physician Consortium for Performance Improvement. An environmental scan and structured literature review were conducted to identify quality programs that integrate PRO-PMs. Key methodological considerations in the design, implementation, and analysis of these PRO-PM data were systematically identified. Recommended methods for addressing each identified consideration were developed on the basis of published patient-reported outcome (PRO) standards and refined through public comment. Literature review focused on programs using PROs to assess performance and on PRO guidance documents.ResultsThirteen PRO programs and 10 guidance documents were identified. Nine best practices were developed, including the following: provide a rationale for measuring the outcome and for using a PRO-PM; describe the context of use; select a measure that is meaningful to patients with adequate psychometric properties; provide evidence of the measure's sensitivity to differences in care; address missing data and risk adjustment; and provide a framework for implementation, interpretation, dissemination, and continuous refinement.ConclusionMethods for integrating PROs into performance measurement are available.

ObjectiveMore than 1 in 4 Medicare beneficiaries with rheumatoid arthritis use high-cost biologic disease-modifying antirheumatic drugs (DMARDs), and spending for these drugs has risen sharply for Medicare Part D. Our aim was to conduct the first systematic, national investigation of how Part D plans cover biologic DMARDs and to determine patients' financial burden under current cost-sharing structures.MethodsWe performed a cross-sectional analysis of Part D plan formularies (n = 2,737) in 50 states and Washington, DC using the January 2013 Centers for Medicare and Medicaid Services Prescription Drug Plan Formulary and Pharmacy Network Files. We calculated the percentage of plans covering each DMARD as well as the percentage requiring prior authorization and/or coinsurance. We also compared biologic drug coverage in Medicare Advantage prescription drug plans to that in Medicare Part D stand-alone plans.ResultsAll plans covered at least 1 biologic DMARD, but the vast majority required prior authorizations (95%). Nearly all plans (81-100%) required patients to pay a coinsurance percentage (averaging 29.6% of drug costs) rather than a fixed dollar copayment. This translated into mean out-of-pocket costs of $2,712-$2,774 before reaching the catastrophic phase of coverage, during which beneficiaries pay 5% of drug costs. Medicare Advantage prescription drug plans covered more individual biologic DMARDs (55-100%) than did Medicare Part D stand-alone plans (22-100%), but Medicare Advantage prescription drug plans required higher average coinsurance (31.1% versus 29.0%). In contrast, 6 of 9 nonbiologic DMARDs were covered by nearly all plans without prior authorizations at fixed copayments averaging $5-$10/month.ConclusionNationally, nearly all Part D plans cover at least 1 biologic DMARD, but the vast majority require cost sharing sufficiently high to risk significant financial burden to patients.

Sparsity is often a desirable property of statistical models, and various feature selection methods exist so as to yield sparser and interpretable models. However, their application to biomedical text classification, particularly to mortality risk stratification among intensive care unit (ICU) patients, has not been thoroughly studied.To develop and characterize sparse classifiers based on the free text of nursing notes in order to predict ICU mortality risk and to discover text features most strongly associated with mortality.We selected nursing notes from the first 24h of ICU admission for 25,826 adult ICU patients from the MIMIC-II database. We then developed a pair of stochastic gradient descent-based classifiers with elastic-net regularization. We also studied the performance-sparsity tradeoffs of both classifiers as their regularization parameters were varied.The best-performing classifier achieved a 10-fold cross-validated AUC of 0.897 under the log loss function and full L2 regularization, while full L1 regularization used just 0.00025% of candidate input features and resulted in an AUC of 0.889. Using the log loss (range of AUCs 0.889-0.897) yielded better performance compared to the hinge loss (0.850-0.876), but the latter yielded even sparser models.Most features selected by both classifiers appear clinically relevant and correspond to predictors already present in existing ICU mortality models. The sparser classifiers were also able to discover a number of informative - albeit nonclinical - features.The elastic-net-regularized classifiers perform reasonably well and are capable of reducing the number of features required by over a thousandfold, with only a modest impact on performance.

BackgroundRoutine preoperative testing is not recommended for patients undergoing cataract surgery, because testing neither decreases adverse events nor improves outcomes. We sought to assess adherence to this guideline, estimate expenditures from potentially unnecessary testing, and identify patient and health care system characteristics associated with potentially unnecessary testing.MethodsUsing an observational cohort of Medicare beneficiaries undergoing cataract surgery in 2011, we determined the prevalence and cost of preoperative testing in the month before surgery. We compared the prevalence of preoperative testing and office visits with the mean percentage of beneficiaries who underwent tests and had office visits during the preceding 11 months. Using multivariate hierarchical analyses, we examined the relationship between preoperative testing and characteristics of patients, health system characteristics, surgical setting, care team, and occurrence of a preoperative office visit.ResultsOf 440,857 patients, 53% had at least one preoperative test in the month before surgery. Expenditures on testing during that month were $4.8 million higher and expenditures on office visits $12.4 million higher (42% and 78% higher, respectively) than the mean monthly expenditures during the preceding 11 months. Testing varied widely among ophthalmologists; 36% of ophthalmologists ordered preoperative tests for more than 75% of their patients. A patient's probability of undergoing testing was associated mainly with the ophthalmologist who managed the preoperative evaluation.ConclusionsPreoperative testing before cataract surgery occurred frequently and was more strongly associated with provider practice patterns than with patient characteristics. (Funded by the Foundation for Anesthesia Education and Research and the Grove Foundation.).

PurposeThe purpose of the study is to compare demographics, intensive care unit (ICU) admission characteristics, and ICU outcomes among adults with childhood-onset chronic conditions (COCCs) admitted to US pediatric and adult ICUs.Materials and methodsRetrospective cross-sectional analyses of 6088 adults aged 19 to 40 years admitted in 2008 to 70 pediatric ICUs that participated in the Virtual Pediatric Intensive Care Unit Performance Systems and 50 adult ICUs that participated in Project IMPACT.ResultsChildhood-onset chronic conditions were present in 53% of young adults admitted to pediatric units, compared with 9% of those in adult units. The most common COCC in both groups were congenital cardiac abnormalities, cerebral palsy, and chromosomal abnormalities. Adults with COCC admitted to pediatric units were significantly more likely to be younger, have lower functional status, and be nontrauma patients than those in adult units. The median ICU length of stay was 2 days, and the intensive care unit mortality rate was 5% for all COCC patients with no statistical difference between pediatric or adult units.ConclusionsThere are marked differences in characteristics between young adults with COCC admitted to pediatric ICUs and adult ICUs. Barriers to accommodating these young adults may be reasons why many such adults have not transitioned from pediatric to adult critical care.

ObjectiveSystemic lupus erythematosus (SLE) has one of the highest hospital readmission rates among chronic conditions. This study was undertaken to identify patient-level, hospital-level, and geographic predictors of 30-day hospital readmissions associated with SLE.MethodsUsing hospital discharge databases from 5 geographically dispersed states, we studied all-cause readmission of SLE patients between 2008 and 2009. We evaluated each hospitalization as a possible index event leading up to a readmission, our primary outcome. We accounted for clustering of hospitalizations within patients and within hospitals and adjusted for hospital case mix. Using multilevel mixed-effects logistic regression, we examined factors associated with 30-day readmission and calculated risk-standardized hospital-level and state-level readmission rates.ResultsWe examined 55,936 hospitalizations among 31,903 patients with SLE. Of these hospitalizations, 9,244 (16.5%) resulted in readmission within 30 days. In adjusted analyses, age was inversely related to risk of readmission. African American and Hispanic patients were more likely to be readmitted than white patients, as were those with Medicare or Medicaid insurance (versus private insurance). Several clinical characteristics of lupus, including nephritis, serositis, and thrombocytopenia, were associated with readmission. Readmission rates varied significantly between hospitals after accounting for patient-level clustering and hospital case mix. We also found geographic variation, with risk-adjusted readmission rates lower in New York and higher in Florida as compared to California.ConclusionWe found that ~1 in 6 hospitalized patients with SLE were readmitted within 30 days of discharge, with higher rates among historically underserved populations. Significant geographic and hospital-level variation in risk-adjusted readmission rates suggests potential for quality improvement.

BackgroundExisting risk adjustment models for intensive care unit (ICU) outcomes rely on manual abstraction of patient-level predictors from medical charts. Developing an automated method for abstracting these data from free text might reduce cost and data collection times.ObjectiveTo develop a support vector machine (SVM) classifier capable of identifying a range of procedures and diagnoses in ICU clinical notes for use in risk adjustment.Materials and methodsWe selected notes from 2001-2008 for 4191 neonatal ICU (NICU) and 2198 adult ICU patients from the MIMIC-II database from the Beth Israel Deaconess Medical Center. Using these notes, we developed an implementation of the SVM classifier to identify procedures (mechanical ventilation and phototherapy in NICU notes) and diagnoses (jaundice in NICU and intracranial hemorrhage (ICH) in adult ICU). On the jaundice classification task, we also compared classifier performance using n-gram features to unigrams with application of a negation algorithm (NegEx).ResultsOur classifier accurately identified mechanical ventilation (accuracy=0.982, F1=0.954) and phototherapy use (accuracy=0.940, F1=0.912), as well as jaundice (accuracy=0.898, F1=0.884) and ICH diagnoses (accuracy=0.938, F1=0.943). Including bigram features improved performance on the jaundice (accuracy=0.898 vs 0.865) and ICH (0.938 vs 0.927) tasks, and outperformed NegEx-derived unigram features (accuracy=0.898 vs 0.863) on the jaundice task.DiscussionOverall, a classifier using n-gram support vectors displayed excellent performance characteristics. The classifier generalizes to diverse patient populations, diagnoses, and procedures.ConclusionsSVM-based classifiers can accurately identify procedure status and diagnoses among ICU patients, and including n-gram features improves performance, compared to existing methods.

BackgroundThe choice of antiplatelet therapy after acute coronary syndrome (ACS) is complicated: Ticagrelor and prasugrel are novel alternatives to clopidogrel, patients with some genotypes may not respond to clopidogrel, and low-cost generic formulations of clopidogrel are available.ObjectiveTo determine the most cost-effective strategy for dual antiplatelet therapy after percutaneous coronary intervention for ACS.DesignDecision-analytic model.Data sourcesPublished literature, Medicare claims, and life tables.Target populationPatients having percutaneous coronary intervention for ACS.Time horizonLifetime.PerspectiveSocietal.InterventionFive strategies were examined: generic clopidogrel, prasugrel, ticagrelor, and genotyping for polymorphisms of CYP2C19 with carriers of loss-of-function alleles receiving either ticagrelor (genotyping with ticagrelor) or prasugrel (genotyping with prasugrel) and noncarriers receiving clopidogrel.Outcome measuresDirect medical costs, quality-adjusted life years(QALYs), and incremental cost-effectiveness ratios (ICERs).Results of base-case analysisThe clopidogrel strategy produced$179 301 in costs and 9.428 QALYs. Genotyping with prasugrel was superior to prasugrel alone, with an ICER of $35 800 per QALY relative to clopidogrel. Genotyping with ticagrelor was more effective than genotyping with prasugrel ($30 200 per QALY relative to clopidogrel). Ticagrelor was the most effective strategy($52 600 per QALY relative to genotyping with ticagrelor).Results of sensitivity analysisStronger associations between genotype and thrombotic outcomes rendered ticagrelor substantially less cost-effective ($104 800 per QALY). Genotyping with prasugrel was the preferred therapy among patients who could not tolerate ticagrelor.LimitationNo randomized trials have directly compared genotyping strategies or prasugrel with ticagrelor.ConclusionGenotype-guided personalization may improve the cost-effectiveness of prasugrel and ticagrelor after percutaneous coronary intervention for ACS, but ticagrelor for all patients may bean economically reasonable alternative in some settings.

BackgroundExamination of at least 12 lymph nodes (LNs) in the staging of colon cancer (CC) was recommended by the National Comprehensive Cancer Network in 2000; however, rates of an adequate examination remain low. This study compares the impact of the hospital contextual variance against that of the operating surgeon on delivery of an adequate LN examination.Study designRetrospective analysis of California Cancer Registry data for all CC operations (2001-2006). Hierarchical models predicted the adequacy of LN examination as a function of patient, surgeon, and hospital characteristics. Models were created using penalized quasi-likelihood approximation with second order Taylor linearization as implemented in MLwiN 2.15.ResultsA total of 25,606 resections involving 3376 surgeons operating in 346 hospitals were analyzed. Half of cases had an adequate examination. Hierarchical models showed the median odds of an adequate examination associated with the hospital context [(MORhosp 2.05; 95% confidence interval, 1.9-2.2) was much higher than that associated with the surgeon (MORsurg 1.34; 95% confidence interval, 1.2-1.4)]. Hospital characteristics teaching and high volume predicted higher odds of an adequate examination. There was no association with hospital revenue.ConclusionsApproximately half of patients undergoing surgery for CC received an adequate LN examination. Hospital contextual factors had a stronger association with receipt of an adequate examination than surgeon factors. Our results suggest that quality improvement initiatives and incentives should be targeted at the hospital level to achieve the highest impact. Furthermore, we have identified nonteaching and low volume settings as rational targets for these efforts.

ObjectivesTo determine the rate of unplanned PICU readmissions, examine the characteristics of index admissions associated with readmission, and compare outcomes of readmissions versus index admissions.DesignRetrospective cohort analysis.SettingNinety North American PICUs that participated in the Virtual Pediatric Intensive Care Unit Systems.PatientsOne hundred five thousand four hundred thirty-seven admissions between July 2009 and March 2011.InterventionsNone.Measurements and main resultsUnplanned PICU readmission within 48 hours of index discharge was the primary outcome. Summary statistics, bivariate analyses, and mixed-effects logistic regression model with random effects for each hospital were performed.There were 1,161 readmissions (1.2%). The readmission rate varied among PICUs (0-3.3%), and acute respiratory (56%), infectious (35%), neurological (28%), and cardiovascular (20%) diagnoses were often present on readmission. Readmission risk increased in patients with two or more complex chronic conditions (adjusted odds ratio, 1.72; p < 0.001), unscheduled index admission (adjusted odds ratio, 1.37; p < 0.001), and transfer to an intermediate unit (adjusted odds ratio, 1.29; p = 0.004, compared with ward). Trauma patients had a decreased risk of readmission (adjusted odds ratio, 0.67; p = 0.003). Gender, race, insurance, age more than 6 months, perioperative status, and nighttime transfer were not associated with readmission. Compared with index admissions, readmissions had longer median PICU length of stay (3.1 vs 1.7 d, p < 0.001) and higher mortality (4% vs 2.5%, p = 0.002).ConclusionsUnplanned PICU readmissions were relatively uncommon, but were associated with worse outcomes. Several patient and admission characteristics were associated with readmission. These data help identify high-risk patient groups and inform risk-adjustment for standardized readmission rates.

ObjectivesTo assess clinician attitudes and experiences in Health eHearts, a quality recognition and financial incentive program using health information technology.Study designSurvey of physicians.MethodsA survey was administered to 140 lead clinicians at each participating practice. Survey domains included clinicians' experiences and attitudes toward the selected clinical quality measures focused on cardiovascular care, use of electronic health records (EHRs), technical assistance visits, quality measurement reports, and incentive payments. Responses were compared across groups of practices receiving financial incentives with those in the control (no financial rewards).ResultsSurvey response rate was 74%. The majority of respondents reported receiving and reviewing the quality reports (89%), agreed with the prioritization of measures (89%), and understood the information given in the quality reports (95%). Over half of the respondents had a quality improvement visit (56%), with incentive clinicians more likely to have had a visit compared with the control group (68% vs 43%, P = .01). The incentive group respondents (92%) were more likely to report using clinical decision support system alerts than control group respondents (82%, P = .11).ConclusionsClinicians in both incentive and control groups reported positive experiences with the program. No differences were detected between groups regarding agreement with selected clinical measures or their relevance to the patient population. However, clinicians in the incentive group were more likely to review quarterly performance reports and access quality improvement visits. Incentives may be used to further engage clinicians operating in small independently owned practices to participate in quality improvement activities.

ImportancePay for performance is intended to align incentives to promote high-quality care, but results have been contradictory.ObjectiveTo test the effect of explicit financial incentives to reward guideline-recommended hypertension care.Design, setting, and participantsCluster randomized trial of 12 Veterans Affairs outpatient clinics with 5 performance periods and a 12-month washout that enrolled 83 primary care physicians and 42 nonphysician personnel (eg, nurses, pharmacists).InterventionsPhysician-level (individual) incentives, practice-level incentives, both, or none. Intervention participants received up to 5 payments every 4 months; all participants could access feedback reports.Main outcomes and measuresAmong a random sample, number of patients achieving guideline-recommended blood pressure thresholds or receiving an appropriate response to uncontrolled blood pressure, number of patients prescribed guideline-recommended medications, and number who developed hypotension.ResultsMean (SD) total payments over the study were $4270 ($459), $2672 ($153), and $1648 ($248) for the combined, individual, and practice-level interventions, respectively. The unadjusted baseline and final percentages and the adjusted absolute change over the study in patients meeting the combined blood pressure/appropriate response measure were 75% to 84% and 8.84% (95% CI, 4.20% to 11.80%) for the individual group, 80% to 85% and 3.70% (95% CI, 0.24% to 7.68%) for the practice-level group, 79% to 88% and 5.54% (95% CI, 1.92% to 9.52%) for the combined group, and 86% to 86% and 0.47% (95% CI, -3.12% to 4.04%) for the control group. The adjusted absolute estimated difference in the change between the proportion of patients with blood pressure control/appropriate response for individual incentive and control groups was 8.36% (95% CI, 2.40% to 13.00%; P=.005). The other incentive groups did not show a significant change compared with controls for this outcome. For medications, the unadjusted baseline and final percentages and the adjusted absolute change were 61% to 73% and 9.07% (95% CI, 4.52% to 13.44%), 56% to 65% and 4.98% (95% CI, 0.64% to 10.08%), 65% to 80% and 7.26% (95% CI, 2.92% to 12.48%), and 63% to 72% and 4.35% (95% CI, -0.28% to 9.28%), respectively. These changes in the use of guideline-recommended medications were not significant in any of the incentive groups compared with controls, nor was the incidence of hypotension. The effect of the incentive was not sustained after a washout.Conclusions and relevanceIndividual financial incentives, but not practice-level or combined incentives, resulted in greater blood pressure control or appropriate response to uncontrolled blood pressure; none of the incentives resulted in greater use of guideline-recommended medications or increased incidence of hypotension compared with controls. Further research is needed on the factors that contributed to these findings.Trial registrationclinicaltrials.gov Identifier: NCT00302718.

ImportanceMost evaluations of pay-for-performance (P4P) incentives have focused on large-group practices. Thus, the effect of P4P in small practices, where many US residents receive care, is largely unknown. Furthermore, whether electronic health records (EHRs) with chronic disease management capabilities support small-practice response to P4P has not been studied.ObjectiveTo assess the effect of P4P incentives on quality in EHR-enabled small practices in the context of an established quality improvement initiative.Design, setting, and participantsA cluster-randomized trial of small (

ObjectiveTo assess variation among hospitals on pediatric readmission and revisit rates and to determine the number of high- and low-performing hospitals.MethodsIn a retrospective analysis using the State Inpatient and Emergency Department Databases from the Healthcare Cost and Utilization Project with revisit linkages available, we identified pediatric (ages 1-20 years) visits with 1 of 7 common inpatient pediatric conditions (asthma, dehydration, pneumonia, appendicitis, skin infections, mood disorders, and epilepsy). For each condition, we calculated rates of all-cause readmissions and rates of revisits (readmission or presentation to the emergency department) within 30 and 60 days of discharge. We used mixed logistic models to estimate hospital-level risk-standardized 30-day revisit rates and to identify hospitals that had performance statistically different from the group mean.ResultsThirty-day readmission rates were low (1% of hospitals labeled as different from the mean on 30-day risk-standardized revisit rates was mood disorders (4.2% of hospitals [n = 15], range of hospital performance 6.3%-15.9%).ConclusionsWe found that when comparing hospitals' performances to the average, few hospitals that care for children are identified as high- or low-performers for revisits, even for common pediatric diagnoses, likely due to low hospital volumes. This limits the usefulness of condition-specific readmission or revisit measures in pediatric quality measurement.

Background and objectivesLittle is known about geographic differences in the incidence of AKI. The objective of this study was to determine if regional variation exists in the population incidence of dialysis-requiring AKI in the United States.Design, setting, participants, & methodsData from the Nationwide Inpatient Sample, a US nationally representative sample of hospitalizations, were used to determine the incidence rates of dialysis-requiring AKI between 2007 and 2009 among the four US Census-designated regions. Cases were identified using validated discharge codes. Poisson regression models were used to estimate overall regional rates, accounting for the data's sampling scheme.ResultsIn 2007-2009, the population incidence rates of dialysis-requiring AKI differed across the four Census-designated regions (P=0.04). Incidence was highest in the Midwest (523 cases/million person-yr, 95% confidence interval=483 to 568) and lowest in the Northeast (457 cases/million person-yr, 95% confidence interval=426 to 492). The pattern of regional variation in the incidence of dialysis-requiring AKI was not the same as the pattern of regional variation in the incidence of renal replacement therapy-requiring ESRD (obtained from the US Renal Data System). In-hospital mortality associated with dialysis-requiring AKI differed across the four regions, with the highest case fatality in the Northeast (25.9%) and the lowest case fatality in the Midwest (19.4%).ConclusionsSignificant regional variation exists in the population incidence of dialysis-requiring AKI in the United States, and additional investigation is warranted to uncover potential causes behind these geographic differences.

BackgroundAdjustment for differing risks among patients is usually incorporated into newer payment approaches, and current risk models rely on age, sex, and diagnosis codes. It is unknown the extent to which controlling additionally for disease severity improves cost prediction. Failure to adjust for within-disease variation may create incentives to avoid sicker patients. We address this issue among patients with chronic obstructive pulmonary disease (COPD).MethodsCost and clinical data were collected prospectively from 1202 COPD patients at Kaiser Permanente. Baseline analysis included age, sex, and diagnosis codes (using the Diagnostic Cost Group Relative Risk Score) in a general linear model predicting total medical costs in the following year. We determined whether adding COPD severity measures-forced expiratory volume in 1 second, 6-Minute Walk Test, dyspnea score, body mass index, and BODE Index (composite of the other 4 measures)-improved predictions. Separately, we examined household income as a cost predictor.ResultsMean costs were $12,334/y. Controlling for Relative Risk Score, each ½ SD worsening in COPD severity factor was associated with $629 to $1135 in increased annual costs (all P

ImportanceGrowing numbers of persons with childhood-onset chronic illnesses are surviving to adulthood. Many use pediatric hospitals for their inpatient needs. To our knowledge, the prevalence and characteristics of adult pediatric intensive care unit patients have not been reported.ObjectivesTo estimate the proportion of adults admitted to pediatric intensive care units (PICUs), characterize them, and compare them with older adolescents.DesignOne-year cross-sectional analysis.SettingPediatric intensive care units in the United States that participated in the Virtual Pediatric Intensive Care Unit Systems.ParticipantsPediatric intensive care unit patients 15 years or older admitted in 2008.Main outcome measuresWe compared adults with adolescents across clinical characteristics and outcomes. Mixed-effects logistic regression was used to estimate the independent association of age with PICU mortality.ResultsSeventy PICUs had 67 629 admissions; 1954 admissions (2.7%) were patients 19 years or older; and 9105 admissions (13.5%) were patients aged 15 to 18 years. The proportion of adults (≥19 years) varied considerably by PICU (range, 0%-9.2%). As age increased, the proportion of patients who had a complex chronic condition and planned or perioperative admissions increased; the proportion of trauma-related admissions decreased. Patients aged 21 to 29 years had a 2 times (95% CI, 1.3-3.2; P = .004) greater odds of PICU mortality compared with adolescent patients, after adjusting for Paediatric Index of Mortality score, sex, trauma, and having a complex chronic condition. Being 30 years or older was associated with a 3.5 (95% CI, 1.3-9.7; P = .01) greater odds of mortality.Conclusions and relevanceIn this multi-institutional study, adults constituted a small but high-risk proportion of patients in some PICUs, suggesting that these PICUs should have plans and protocols specifically focused on this group.

ObjectiveThis study investigated clinicians' perspectives during a quality improvement project to promote breastmilk feedings in premature infants.Study designFrom 2009 to 2010, 11 hospitals in the California Perinatal Quality Care Collaborative participated in a project to promote breastmilk feedings in premature infants. Audio recordings of monthly meetings held to encourage sharing of ideas were analyzed using qualitative methods to identify common themes related to barriers and solutions to breastmilk feeding promotion.ResultsTwo broad categories were noted: communication and team composition. Communication subthemes included (1) communication among hospital staff, including consistent documentation, (2) communication with family, and (3) communication between transfer hospitals. Team composition subthemes included (4) importance of physician buy-in and (5) integrated teams designed to empower leaders.ConclusionsOptimizing communication among health professionals and parents and improving team composition may be key components of facilitating breastmilk feeding promotion in premature infants.

BackgroundOur goal was to compare hospital scores from the most widely used commercial website in the USA to hospital scores from more systematic measures of patient experience and outcomes, and to assess what drives variation in the commercial website scores.MethodsFor a national sample of US hospitals, we compared scores on Yelp.com, which aggregates website visitor ratings (1-5 stars), with traditional measures of hospital quality. We calculated correlations between hospital Yelp scores and the following: hospital percent high ratings (9 or 10, scale 0-10) on the 'Overall' item on the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey; hospital individual HCAHPS domain scores (eg, nurse communication, pain control); hospital 30-day mortality; and hospital 30-day readmission rates.ResultsOf hospitals reporting HCAHPS (n=3796), 962 (25%) had scores on Yelp. Among hospitals with >5 Yelp ratings, the correlation of percent high ratings between Yelp and HCAHPS was 0.49 (p

ObjectiveTo identify an approach to summarizing publicly reported hospital performance data for acute myocardial infarction (AMI) or heart failure (HF) that best predicts current year hospital mortality rates.SettingA total of 1,868 U.S. hospitals reporting process and outcome measures for AMI and HF to the Centers for Medicare and Medicaid Services (CMS) from July 2005 to June 2006 (Year 0) and July 2006 to June 2007 (Year 1).DesignObservational cohort study measuring the percentage variation in Year 1 hospital 30-day risk-adjusted mortality rate explained by denominator-based weighted composite scores summarizing hospital Year 0 performance.Data collectionData were prospectively collected from hospitalcompare.gov.ResultsPercentage variation in Year 1 mortality was best explained by mortality rate alone in Year 0 over other composites including process performance. If only Year 0 mortality rates were reported, and consumers using hospitals in the highest decile of mortality instead chose hospitals in the lowest decile of mortality rate, the number of deaths at 30 days that potentially could have been avoided was 1.31 per 100 patients for AMI and 2.12 for HF (p < .001).ConclusionPublic reports focused on 30-day risk-adjusted mortality rate may more directly address policymakers' goals of facilitating consumer identification of hospitals with better outcomes.

The population epidemiology of AKI is not well described. Here, we analyzed data from the Nationwide Inpatient Sample, a nationally representative dataset, to identify cases of dialysis-requiring AKI using validated International Classification of Diseases, Ninth Revision (ICD-9) codes. From 2000 to 2009, the incidence of dialysis-requiring AKI increased from 222 to 533 cases per million person-years, averaging a 10% increase per year (incidence rate ratio=1.10, 95% CI=1.10-1.11 per year). Older age, male sex, and black race associated with higher incidence of dialysis-requiring AKI. The rapid increase in incidence was evident in all age, sex, and race subgroups examined. Temporal changes in the population distribution of age, race, and sex as well as trends of sepsis, acute heart failure, and receipt of cardiac catheterization and mechanical ventilation accounted for about one third of the observed increase in dialysis-requiring AKI among hospitalized patients. The total number of deaths associated with dialysis-requiring AKI rose from 18,000 in 2000 to nearly 39,000 in 2009. In conclusion, the incidence of dialysis-requiring AKI increased rapidly in all patient subgroups in the past decade in the United States, and the number of deaths associated with dialysis-requiring AKI more than doubled.

45 Background: There is growing interest to use performance measures that integrate patient-reported outcomes (PROs), but there are no existing methodological standards for developing, implementing, or analyzing such measures in this context.To develop standards for evaluating PRO-based performance measures submitted to or developed by the American Medical Association-convened Physician Consortium for Performance Improvement, a multi-disciplinary expert technical panel was assembled. A systematic literature review and landscape overview were conducted to identify use cases and existing PRO standards in related contexts. Lessons learned from use cases and existing standards guided development of discrete methodological standards.The systematic review identified only one use case, whereas the landscape overview identified eight. Five purposes were identified for which PROs might be measured to assess performance, including one categorized as process and four as outcome. Eight discrete methodological standards were developed which will be detailed at the meeting.The patient perspective is an essential but missing component of performance evaluation. When appropriately elicited, data from patients can be informative. Lack of adherence to good practices and inadequate patient input when developing measurement strategies are potential pitfalls to be avoided.