A Multicentered Randomized Controlled Trial Comparing the Effectiveness of Pain Treatment Communication Tools in Emergency Department Patients With Back or Kidney Stone Pain.

Objectives. To compare the effectiveness of 3 approaches for communicating opioid risk during an emergency department visit for a common painful condition. Methods. This parallel, multicenter randomized controlled trial was conducted at 6 geographically disparate emergency department sites in the United States. Participants included adult patients between 18 and 70 years of age presenting with kidney stone or musculoskeletal back pain. Participants were randomly assigned to 1 of 3 risk communication strategies: (1) a personalized probabilistic risk visual aid, (2) a visual aid and a video narrative, or 3) general risk information. The primary outcomes were accuracy of risk recall, reported opioid use, and treatment preference at time of discharge. Results. A total of 1301 participants were enrolled between June 2017 and August 2019. There was no difference in risk recall at 14 days between the narrative and probabilistic groups (43.7% vs 38.8%; absolute risk reduction = 4.9%; 95% confidence interval [CI] = -2.98, 12.75). The narrative group had lower rates of preference for opioids at discharge than the general risk information group (25.9% vs 33.0%; difference = 7.1%; 95% CI = 0.64, 0.97). There were no differences in reported opioid use at 14 days between the narrative, probabilistic, and general risk groups (10.5%, 10.3%, and 13.3%, respectively; P = .44). Conclusions. An emergency medicine communication tool incorporating probabilistic risk and patient narratives was more effective than general information in mitigating preferences for opioids in the treatment of pain but was not more effective with respect to opioid use or risk recall. Trial Registration. Clinical Trials.gov identifier: NCT03134092. (Am J Public Health. 2022;112(S1):S45-S55. https://doi.org/10.2105/AJPH.2021.306511).

Effect of Remediating Blighted Vacant Land on Shootings: A Citywide Cluster Randomized Trial.

Objectives. To determine if remediating blighted vacant urban land reduced firearm shooting incidents resulting in injury or death.Methods. We conducted a cluster randomized controlled trial in which we assigned 541 randomly selected vacant lots in Philadelphia, Pennsylvania, to 110 geographically contiguous clusters and randomly assigned these clusters to a greening intervention, a less-intensive mowing and trash cleanup intervention, or a no-intervention control condition. The random assignment to the trial occurred in April and June 2013 and lasted until March 2015. In a difference-in-differences analysis, we assessed whether the 2 treatment conditions relative to the control condition reduced firearm shootings around vacant lots.Results. During the trial, both the greening intervention, -6.8% (95% confidence interval [CI] = -10.6%, -2.7%), and the mowing and trash cleanup intervention, -9.2% (95% CI = -13.2%, -4.8%), significantly reduced shootings. There was no evidence that the interventions displaced shootings into adjacent areas.Conclusions. Remediating vacant land with inexpensive, scalable methods, including greening or minimal mowing and trash cleanup, significantly reduced shootings that result in serious injury or death.Public Health Implications. Cities should experiment with place-based interventions to develop effective firearm violence-reduction strategies.Trial Registration. This trial was registered with the International Standard Randomized Controlled Trial Number (study ID ISRCTN92582209; http://www.isrctn.com/ISRCTN92582209).

Development and Validation of an Algorithm for Identifying Patients with Hemophilia A in an Administrative Claims Database.

BACKGROUND: The accuracy with which hemophilia A can be identified in claims databases is unknown. OBJECTIVE: Develop and validate an algorithm using predictive modeling supported by machine learning to identify patients with hemophilia A in an administrative claims database. METHODS: We first created a screening algorithm using medical and pharmacy claims to identify potential hemophilia A patients in the US HealthCore Integrated Research Database between January 1, 2006 and April 30, 2015. Medical records for a random sample of patients were reviewed to confirm case status. In this validation sample, we used lasso logistic regression with cross-validation to select covariates in claims data and develop a predictive model to estimate the probability of being a confirmed hemophilia A case. RESULTS: The screening algorithm identified 2,252 patients and we reviewed medical records for 400 of these patients. The screening algorithm had a positive predictive value (PPV) of 65%. The predictive model identified 18 predictors of being a hemophilia A case or noncase. The strongest predictors of case status included male sex, factor VIII therapy, office visits for hemophilia A, and hospitalizations for hemophilia A. The strongest predictors of noncase status included hospitalizations for reasons other than hemophilia A and factor VIIa therapy. A probability threshold of ≥0.6 resulted in a PPV of 94.7% (95% CI: 92.0-97.5) and sensitivity of 94.4% (95% CI: 91.5-97.2). CONCLUSIONS: We developed and validated an algorithm to identify hemophilia A cases in an administrative claims database with high sensitivity and high PPV.

Health system and law enforcement synergies for injury surveillance, control and prevention: a scoping review.

BACKGROUND: Healthcare providers and law enforcement (LE) officers are among the most common first responders to injuring events. Despite frequent interface between the health system (HS) and LE sectors, the published evidence that supports their collaboration in injury surveillance, control and prevention has not been comprehensively reviewed. METHODS: We conducted a scoping review of literature published from 1990 to 2016 that focused on local and regional HS and LE collaborations in injury surveillance, control and prevention. Our aim was to describe what is known and what remains unexplored about these cross-sector efforts. RESULTS: 128 articles were included in the final review. These were categorised by their focus on either surveillance activities or partnerships in injury control and prevention programmes. The majority of surveillance articles focused on road traffic injuries. Conversely, articles describing partnerships and programme evaluations primarily targeted the prevention of interpersonal violence. DISCUSSION: This review yielded two major findings: overall, the combination of HS and LE injury data added value to surveillance systems, especially as HS data augmented LE data; and HS and LE partnerships have been developed to improve injury control and prevention. However, there are few studies that have evaluated the impact and sustainability of these partnerships. CONCLUSIONS: The current evidence to support HS and LE collaboration in injury surveillance and control and prevention programmes is heterogeneous. Notable gaps suggest ample opportunity for further research and programme evaluation across all types of injury.

Effect of Abandoned Housing Interventions on Gun Violence, Perceptions of Safety, and Substance Use in Black Neighborhoods: A Citywide Cluster Randomized Trial.

Importance: Structural racism has resulted in long-standing disinvestment and dilapidated environmental conditions in Black neighborhoods. Abandoned houses signal neglect and foster stress and fear for residents, weakening social ties and potentially contributing to poor health and safety. Objective: To determine whether abandoned house remediation reduces gun violence and substance-related outcomes and increases perceptions of safety and use of outdoor space. Design, Setting, and Participants: This cluster randomized trial was conducted from January 2017 to August 2020, with interventions occurring between August 2018 and March 2019. The study included abandoned houses across Philadelphia, Pennsylvania, and surveys completed by participants living nearby preintervention and postintervention. Data analysis was performed from March 2021 to September 2022. Interventions: The study consisted of 3 arms: (1) full remediation (installing working windows and doors, cleaning trash, weeding); (2) trash cleanup and weeding only; and (3) a no-intervention control. Main Outcomes and Measures: Difference-in-differences mixed-effects regression models were used to estimate the effect of the interventions on multiple primary outcomes: gun violence (weapons violations, gun assaults, and shootings), illegal substance trafficking and use, public drunkenness, and perceptions of safety and time outside for nearby residents. Results: A master list of 3265 abandoned houses was randomly sorted. From the top of this randomly sorted list, a total of 63 clusters containing 258 abandoned houses were formed and then randomly allocated to 3 study arms. Of the 301 participants interviewed during the preintervention period, 172 (57.1%) were interviewed during the postintervention period and were included in this analysis; participants were predominantly Black, and most were employed. Study neighborhoods were predominantly Black with high percentages of low-income households. Gun violence outcomes increased in all study arms, but increased the least in the full remediation arm. The full housing remediation arm, compared with the control condition, showed reduced weapons violations by -8.43% (95% CI, -14.68% to -1.19%), reduced gun assaults by -13.12% (95% CI, -21.32% to -3.01%), and reduced shootings by a nonsignificant -6.96% (95% CI, -15.32% to 3.03%). The trash cleanup arm was not associated with a significant differential change in any gun violence outcome. Instances of illegal substance trafficking and use and public drunkenness outcomes were not significantly affected by the housing remediation or trash cleanup treatment. Perceptions of neighborhood safety and time spent outside were unaffected by the intervention. The study arms did differ in a baseline characteristic and some preintervention trends, which raises questions regarding other potential nonmeasured differences between study arms that could have influenced estimates. No evidence of displacement of gun violence outcomes was found. Conclusions and Relevance: In this cluster randomized controlled trial among low-income, predominantly Black neighborhoods, inexpensive, straightforward abandoned housing remediation was directly linked to significant relative reductions in weapons violations and gun assaults, and suggestive reductions in shootings. Trial Registration: isrctn.org Identifier: ISRCTN14973997.

Interpreting the empirical evidence on illegal gun market dynamics.

Thousands of Americans are killed by gunfire each year, and hundreds of thousands more are injured or threatened with guns in robberies and assaults. The burden of gun violence in urban areas is particularly high. Critics suggest that the results of firearm trace data and gun trafficking investigation studies cannot be used to understand the illegal supply of guns to criminals and, therefore, that regulatory and enforcement efforts designed to disrupt illegal firearms markets are futile in addressing criminal access to firearms. In this paper, we present new data to address three key arguments used by skeptics to undermine research on illegal gun market dynamics. We find that criminals rely upon a diverse set of illegal diversion pathways to acquire guns, gun traffickers usually divert small numbers of guns, newer guns are diverted through close-to-retail diversions from legal firearms commerce, and that a diverse set of gun trafficking indicators are needed to identify and shut down gun trafficking pathways.

Benchmarking Danish hospitals on mortality and readmission rates after cardiovascular admission.

OBJECTIVE: The aim of this study was to examine hospital performance measures that account more comprehensively for unique mixes of patients' characteristics. DESIGN: Nationwide cohort registry-based study within a population-based health care system. PARTICIPANTS: In this study, 331,513 patients discharged with a primary cardiovascular diagnosis from 1 of 26 Danish hospitals during 2011-2015 were included. Data covering all Danish hospitals were drawn from the Danish National Patient Registry and the Danish National Health Service Prescription Database. MAIN OUTCOME MEASURES: Thirty-day post-admission mortality rates, 30-day post-discharge readmission rates, and the associated numbers needed to harm were measured. METHODS: For each index hospital, we used a non-parametric logistic regression model to compute propensity scores. Propensity score weighted patients treated at other hospitals collectively resembled patients treated at the index hospital in terms of age, sex, primary discharge diagnosis, diagnosis history, medications, previous cardiac procedures, and comorbidities. Outcomes for the weighted patients treated at other hospitals formed benchmarks for the index hospital. Doubly robust regression formally tested whether the outcomes of patients at the index hospital differed from the outcomes of the patients used to form the benchmarks. For each index hospital, we computed the false discovery rate, ie, the probability of being incorrect if we claimed the hospital differed from its benchmark. RESULTS: Five hospitals exceeded their benchmark for 30-day mortality rates, with the number needed to harm ranging between 55 and 137. Seven hospitals exceeded their benchmark for readmission, with the number needed to harm ranging from 22 to 71. Our benchmarking approach flagged fewer hospitals as outliers compared with conventional regression methods. CONCLUSION: Conventional methods flag more hospitals as outliers than our benchmarking approach. Our benchmarking approach accounts more thoroughly for differences in hospitals' patient case mix, reducing the risk of false-positive selection of suspected outliers. A more comprehensive system of hospital performance measurement could be based on this approach.

Estimating the effects of light rail transit on health care costs.

In recent years, there has been a proliferation of research on the effects of the built environment, including mass transit systems, on health-related outcomes. While there is general agreement that the built environment affects travel choices and physical activity, it remains unclear how much of a public health benefit (in dollars) can be derived from land use policies that support walking, biking, and transit. In the present study, we develop a model to assess the potential cost savings in public health that will be realized from the investment in a new light rail transit system in Charlotte, NC. Relying on estimates of future riders, area obesity rates, and the effects of public transit on physical activity (daily walking to and from the transit stations), we simulated the potential yearly public health cost savings associated with this infrastructure investment. Our results indicate that investing in light rail is associated with a 9-year cumulative public health cost savings of dollars 12.6 million. While these results suggest that there is a sizable public health benefit associated with the adoption of light rail, they also indicate that the effects are relatively small compared to the costs associated with constructing and operating such systems. These findings suggest that planning efforts that focus solely on the health impact of modifications in the built environment are likely to overstate the economic benefits. Public health benefits should be considered along with broader environmental health benefits.

Availability of Health-Related Programs in Private and Public Prisons.

Little is known about the resources available to protect inmates' health in private prisons compared to their public counterparts. This is the first national-level study that exclusively examined the availability of health-related programs in private and public prisons in the United States. We applied propensity score weighting and doubly robust estimation to compare private prisons to comparable public prisons. Data were self-reported by prison administrators as part of the 2005 Census of State and Federal Adult Correctional Facilities. We found that private prisons offered fewer substance dependency, psychological/psychiatric, and HIV/AIDS-related programs. But the differences were progressively reduced when the comparison was limited to public prisons most similar on a variety of facility-level characteristics. The extent to which the two types of prisons differ is closely tied to the characteristics of the facilities that are compared.

Examination of the association between syringe exchange program (SEP) dispensation policy and SEP client-level syringe coverage among injection drug users.

AIM: To determine whether syringe exchange programs' (SEPs) dispensation policy is associated with syringe coverage among SEP clients. DESIGN: Cross-sectional samples of SEPs and their clients. SETTING: SEPs in California, USA. PARTICIPANTS: Twenty-four SEPs and their injection drug using (IDU) clients (n = 1576). MEASUREMENTS: Clients were classified as having adequate syringe coverage if they received at least as many syringes from the SEP as their self-reported injections in the last 30 days. SEPs were classified based on their syringe dispensation policy. Dispensation schemes ranging from least restrictive to most are: unlimited needs-based distribution; unlimited one-for-one exchange plus a few additional syringes; per visit limited one-for-one plus a few additional syringes; unlimited one-for-one exchange; and per visit limited one-for-one exchange. FINDINGS: Adequate syringe coverage among SEP clients by dispensation policy is as follows: unlimited needs-based distribution = 61%; unlimited one-for-one plus = 50%; limited one-for-one plus = 41%; unlimited one-for-one = 42%; and limited one-for-one = 26%. In multivariate analysis, adequate syringe coverage was significantly higher for all dispensation policies compared to per visit limited one-for-one exchange. Using propensity scoring methods, we compared syringe coverage by dispensation policies while controlling for client-level differences. Providing additional syringes above one-for-one exchange (50% versus 38%, P = 0.009) and unlimited exchange (42% versus 27%, P = 0.05) generally resulted in more clients having adequate syringe coverage compared to one-for-one exchange and per visit limits. CONCLUSION: Providing less restrictive syringe dispensation is associated with increased prevalence of adequate syringe coverage among clients. SEPs should adopt syringe dispensation policies that provide IDUs sufficient syringes to attain adequate syringe coverage.

Interpreting treatment effects when cases are institutionalized after treatment.

Drug treatment clients are at high risk for institutionalization, i.e., spending a day or more in a controlled environment where their freedom to use drugs, commit crimes, or engage in risky behavior may be circumscribed. For example, in recent large studies of drug treatment outcomes, more than 40% of participants were institutionalized for a portion of the follow-up period. When longitudinal studies ignore institutionalization at follow-up, outcome measures and treatment effect estimates conflate treatment effects on institutionalization with effects on many of the outcomes of interest. In this paper, we develop a causal modeling framework for evaluating the four standard approaches for addressing this institutionalization confound, and illustrate the effects of each approach using a case study comparing drug use outcomes of youths who enter either residential or outpatient treatment modalities. Common methods provide biased estimates of the treatment effect except under improbable assumptions. In the case study, the effect of residential care ranged from beneficial and significant to detrimental and significant depending on the approach used to account for institutionalization. We discuss the implications of our analysis for longitudinal studies of all populations at high risk for institutionalization.

Flexible distributions for triple-goal estimates in two-stage hierarchical models.

Performance evaluations often aim to achieve goals such as obtaining estimates of unit-specific means, ranks, and the distribution of unit-specific parameters. The Bayesian approach provides a powerful way to structure models for achieving these goals. While no single estimate can be optimal for achieving all three inferential goals, the communication and credibility of results will be enhanced by reporting a single estimate that performs well for all three. Triple goal estimates [Shen and Louis, 1998. Triple-goal estimates in two-stage hierarchical models. J. Roy. Statist. Soc. Ser. B 60, 455-471] have this performance and are appealing for performance evaluations. Because triple-goal estimates rely more heavily on the entire distribution than do posterior means, they are more sensitive to misspecification of the population distribution and we present various strategies to robustify triple-goal estimates by using nonparametric distributions. We evaluate performance based on the correctness and efficiency of the robustified estimates under several scenarios and compare empirical Bayes and fully Bayesian approaches to model the population distribution. We find that when data are quite informative, conclusions are robust to model misspecification. However, with less information in the data, conclusions can be quite sensitive to the choice of population distribution. Generally, use of a nonparametric distribution pays very little in efficiency when a parametric population distribution is valid, but successfully protects against model misspecification.

Bayesian inference for the distribution of grams of marijuana in a joint.

BACKGROUND: The average amount of marijuana in a joint is unknown, yet this figure is a critical quantity for creating credible measures of marijuana consumption. It is essential for projecting tax revenues post-legalization, estimating the size of illicit marijuana markets, and learning about how much marijuana users are consuming in order to understand health and behavioral consequences. METHODS: Arrestee Drug Abuse Monitoring data collected between 2000 and 2010 contain relevant information on 10,628 marijuana transactions, joints and loose marijuana purchases, including the city in which the purchase occurred and the price paid for the marijuana. Using the Brown-Silverman drug pricing model to link marijuana price and weight, we are able to infer the distribution of grams of marijuana in a joint and provide a Bayesian posterior distribution for the mean weight of marijuana in a joint. RESULTS: We estimate that the mean weight of marijuana in a joint is 0.32g (95% Bayesian posterior interval: 0.30-0.35). CONCLUSIONS: Our estimate of the mean weight of marijuana in a joint is lower than figures commonly used to make estimates of marijuana consumption. These estimates can be incorporated into drug policy discussions to produce better understanding about illicit marijuana markets, the size of potential legalized marijuana markets, and health and behavior outcomes.

Propensity Score Analysis with Survey Weighted Data.

Propensity score analysis (PSA) is a common method for estimating treatment effects, but researchers dealing with data from survey designs are generally not properly accounting for the sampling weights in their analyses. Moreover, recommendations given in the few existing methodological articles on this subject are susceptible to bias. We show in this article through derivation, simulation, and a real data example that using sampling weights in the propensity score estimation stage and the outcome model stage results in an estimator that is robust to a variety of conditions that lead to bias for estimators currently recommended in the statistical literature. We highly recommend researchers use the more robust approach described here. This article provides much needed rigorous statistical guidance for researchers working with survey designs involving sampling weights and using PSAs.

Propensity score estimation with boosted regression for evaluating causal effects in observational studies.

Causal effect modeling with naturalistic rather than experimental data is challenging. In observational studies participants in different treatment conditions may also differ on pretreatment characteristics that influence outcomes. Propensity score methods can theoretically eliminate these confounds for all observed covariates, but accurate estimation of propensity scores is impeded by large numbers of covariates, uncertain functional forms for their associations with treatment selection, and other problems. This article demonstrates that boosting, a modern statistical technique, can overcome many of these obstacles. The authors illustrate this approach with a study of adolescent probationers in substance abuse treatment programs. Propensity score weights estimated using boosting eliminate most pretreatment group differences and substantially alter the apparent relative effects of adolescent substance abuse treatment.

Effectiveness of community-based treatment for substance-abusing adolescents: 12-month outcomes of youths entering phoenix academy or alternative probation dispositions.

Whereas strong efficacy research has been conducted on novel treatment approaches for adolescent substance abusers, little is known about the effectiveness of the substance abuse treatment approaches most commonly available to youths, their families, and referring agencies. This report compares the 12-month outcomes of adolescent probationers (N = 449) who received either Phoenix Academy, a therapeutic community for adolescents that uses a treatment model that is widely implemented across the U.S., or an alternative probation disposition. Across many pretreatment risk factors for relapse and recidivism, groups were well matched after case-mix adjustment. Repeated measures analyses of substance use, psychological functioning, and crime outcomes collected 3, 6, and 12 months after the baseline interview demonstrated that Phoenix Academy treatment is associated with superior substance use and psychological functioning outcomes over the period of observation. As one of the most rigorous evaluations of the effectiveness of a traditional community-based adolescent drug treatment program, this study provides evidence that one such program is effective. Implications of this finding for the dissemination of efficacious novel treatment approaches are discussed.

Bias and variance trade-offs when combining propensity score weighting and regression: with an application to HIV status and homeless men.

The quality of propensity scores is traditionally measured by assessing how well they make the distributions of covariates in the treatment and control groups match, which we refer to as "good balance". Good balance guarantees less biased estimates of the treatment effect. However, the cost of achieving good balance is that the variance of the estimates increases due to a reduction in effective sample size, either through the introduction of propensity score weights or dropping cases when propensity score matching. In this paper, we investigate whether it is best to optimize the balance or to settle for a less than optimal balance and use double robust estimation to adjust for remaining differences. We compare treatment effect estimates from regression, propensity score weighting, and double robust estimation with varying levels of effort expended to achieve balance using data from a study about the differences in outcomes by HIV status in heterosexually active homeless men residing in Los Angeles. Because of how costly data collection efforts are for this population, it is important to find an alternative estimation method that does not reduce effective sample size as much as methods that aggressively aim to optimize balance. Results from a simulation study suggest that there are instances in which we can obtain more precise treatment effect estimates without increasing bias too much by using a combination of regression and propensity score weights that achieve a less than optimal balance. There is a bias-variance tradeoff at work in propensity score estimation; every step toward better balance usually means an increase in variance and at some point a marginal decrease in bias may not be worth the associated increase in variance.

Behavioral impact of graduated driver licensing on teenage driving risk and exposure.

Graduated driver licensing (GDL) is a critical policy tool for potentially improving teenage driving while reducing teen accident exposure. While previous studies demonstrated that GDL reduces teenage involvement in fatal crashes, much remains unanswered. We explore the mechanisms through which GDL influences accident rates as well as its long term effectiveness on teen driving. In particular, we investigate: (1) whether GDL policies improve teenage driving behavior, or simply reduce teenage prevalence on the roads; (2) whether GDL exposed teens become better drivers in later years. We employ a unique data source, the State Data System, which contains all police reported accidents (fatal and non-fatal) during 1990-2005 for 12 states. We estimate a structural model that separately identifies GDL's effect on relative teenage prevalence and relative teenage riskiness. Identification of the model is driven by the relative numbers of crashes between two teenagers, two adults, or a teenager and an adult. We find that the GDL policies reduce the number of 15-17-year-old accidents by limiting the amount of teenage driving rather than by improving teenage driving. This prevalence reduction primarily occurs at night and stricter GDL policies, especially those with night-time driving restrictions, are the most effective. Finally, we find that teen driving quality does not improve ex post GDL exposure.

The effect of light rail transit on body mass index and physical activity.

BACKGROUND: The built environment can constrain or facilitate physical activity. Most studies of the health consequences of the built environment face problems of selection bias associated with confounding effects of residential choice and transportation decisions. PURPOSE: To examine the cross-sectional associations between objective and perceived measures of the built environment; BMI; obesity (BMI>30 kg/m(2)); and meeting weekly recommended physical activity (RPA) levels through walking and vigorous exercise. To assess the effect of using light rail transit (LRT) system on BMI, obesity, and weekly RPA levels. METHODS: Data were collected on individuals before (July 2006-February 2007) and after (March 2008-July 2008) completion of an LRT system in Charlotte NC. BMI, obesity, and physical activity levels were calculated for a comparison of these factors pre- and post-LRT construction. A propensity score weighting approach adjusted for differences in baseline characteristics among LRT and non-LRT users. Data were analyzed in 2009. RESULTS: More-positive perceptions of one's neighborhood at baseline were associated with a -0.36 (p<0.05) lower BMI; 15% lower odds (95% CI=0.77, 0.94) of obesity; 9% higher odds (95% CI=0.99, 1.20) of meeting weekly RPA through walking; and 11% higher odds (95% CI=1.01, 1.22) of meeting RPA levels of vigorous exercise. The use of LRT to commute to work was associated with an average -1.18 reduction in BMI (p<0.05) and an 81% reduced odds (95% CI=0.04, 0.92) of becoming obese over time. CONCLUSIONS: The results of this study suggest that improving neighborhood environments and increasing the public's use of LRT systems could provide improvements in health outcomes for millions of individuals.

Ranking USRDS provider specific SMRs from 1998-2001.

Provider profiling (ranking/percentiling) is prevalent in health services research. Bayesian models coupled with optimizing a loss function provide an effective framework for computing non-standard inferences such as ranks. Inferences depend on the posterior distribution and should be guided by inferential goals. However, even optimal methods might not lead to definitive results and ranks should be accompanied by valid uncertainty assessments. We outline the Bayesian approach and use estimated Standardized Mortality Ratios (SMRs) in 1998-2001 from the United States Renal Data System (USRDS) as a platform to identify issues and demonstrate approaches. Our analyses extend Liu et al. (2004) by computing estimates developed by Lin et al. (2006) that minimize errors in classifying providers above or below a percentile cut-point, by combining evidence over multiple years via a first-order, autoregressive model on log(SMR), and by use of a nonparametric prior. Results show that ranks/percentiles based on maximum likelihood estimates of the SMRs and those based on testing whether an SMR = 1 substantially under-perform the optimal estimates. Combining evidence over the four years using the autoregressive model reduces uncertainty, improving performance over percentiles based on only one year. Furthermore, percentiles based on posterior probabilities of exceeding a properly chosen SMR threshold are essentially identical to those produced by minimizing classification loss. Uncertainty measures effectively calibrate performance, showing that considerable uncertainty remains even when using optimal methods. Findings highlight the importance of using loss function guided percentiles and the necessity of accompanying estimates with uncertainty assessments.