Search

Modifications of risk-adjustment systems used to pay health plans in individual health insurance markets typically seek to reduce selection incentives at the individual and group levels by adding variables to the payment formula. Adding variables can be costly and lead to unintended incentives for upcoding or service utilization. While these drawbacks are recognized, they are hard to quantify and difficult to balance against the concrete, measurable improvements in fit that may be achieved by adding variables to the formula. This paper takes a different approach to improving the performance of health plan payment systems. Using the HHS-HHC V0519 model from the Marketplaces as a starting point, we constrain fit at the individual and group level to be as good or better than the current payment model while reducing the number of variables in the model. We introduce three elements in the design of plan payment: reinsurance, constrained regressions, and machine learning methods for variable selection. The fit performance of our alternative formulas with many fewer variables is as good or better than the current HHS-HHC V0519 formula.

Machine learning algorithms are increasingly used in the clinical literature, claiming advantages over logistic regression. However, they are generally designed to maximize the area under the receiver operating characteristic curve. While area under the receiver operating characteristic curve and other measures of accuracy are commonly reported for evaluating binary prediction problems, these metrics can be misleading. We aim to give clinical and machine learning researchers a realistic medical example of the dangers of relying on a single measure of discriminatory performance to evaluate binary prediction questions. Prediction of medical complications after surgery is a frequent but challenging task because many post-surgery outcomes are rare. We predicted post-surgery mortality among patients in a clinical registry who received at least one aortic valve replacement. Estimation incorporated multiple evaluation metrics and algorithms typically regarded as performing well with rare outcomes, as well as an ensemble and a new extension of the lasso for multiple unordered treatments. Results demonstrated high accuracy for all algorithms with moderate measures of cross-validated area under the receiver operating characteristic curve. False positive rates were [Formula: see text]1%, however, true positive rates were [Formula: see text]7%, even when paired with a 100% positive predictive value, and graphical representations of calibration were poor. Similar results were seen in simulations, with the addition of high area under the receiver operating characteristic curve ([Formula: see text]90%) accompanying low true positive rates. Clinical studies should not primarily report only area under the receiver operating characteristic curve or accuracy.

The use of machine learning (ML) in health care raises numerous ethical concerns, especially as models can amplify existing health inequities. Here, we outline ethical considerations for equitable ML in the advancement of health care. Specifically, we frame ethics of ML in health care through the lens of social justice. We describe ongoing efforts and outline challenges in a proposed pipeline of ethical ML in health, ranging from problem selection to post-deployment considerations. We close by summarizing recommendations to address these challenges., Comment: Annual Reviews in Biomedical Data Science 2021

Access to specialty mental health care remains challenging for people with serious mental illnesses, such as schizophrenia and bipolar disorder. Whether expansion of telemedicine is associated with improved access and quality of care for these patients is unclear.To assess whether greater telemedicine use in a nonmetropolitan county is associated with quality measures, including use of specialty mental health care and medication adherence.In this cohort study, the variable uptake of telemental health visits was examined across a national sample of fee-for-service claims from Medicare beneficiaries in 2916 nonmetropolitan counties between January 1, 2010, and December 31, 2018. Beneficiaries with schizophrenia and related psychotic disorders and/or bipolar I disorder during the study period were included. For each year of the study, each county was categorized based on per capita telemental health service use (none, low, moderate, and high). The association between telemental health service use in the county and quality measures was tested using a multivariate model controlling for both patient characteristics and county fixed effects. Analyses were conducted from January 1 to April 11, 2022. Before the COVID-19 pandemic, telemedicine reimbursement was limited to nonmetropolitan beneficiaries.Receipt of a minimum of 2 specialty mental health service visits (telemedicine or in-person) in the year, number of months per year with medication, hospitalization rate, and outpatient follow-up visits after a mental health hospitalization in a year.In 2018, there were 2916 counties with 118 170 patients (77 068 [65.2%] men; mean [SD] age, 58.3 [15.6] years) in the sample. The fraction of counties that had high telemental health service use increased from 2% in 2010 to 17% in 2018. In 2018 there were 1.08 telemental health service visits per patient in the high telemental health counties. Compared with no telemental health care in the county, patients in high-use counties were 1.2 percentage points (95% CI, 0.81-1.60 percentage points) (8.0% relative increase) more likely to have a minimum number of specialty mental health service visits, 13.7 percentage points (95% CI, 5.1-22.3 percentage points) (6.5% relative increase) more likely to have outpatient follow-up within 7 days of a mental health hospitalization, and 0.47 percentage points (95% CI, 0.25-0.69 percentage points) (7.6% relative increase) more likely to be hospitalized in a year. Telemental health service use was not associated with changes in medication adherence.The findings of this study suggest that greater use of telemental health visits in a county was associated with modest increases in contact with outpatient specialty mental health care professionals and greater likelihood of follow-up after hospitalization. No substantive changes in medication adherence were noted and an increase in mental health hospitalizations occurred.

BackgroundAs highlighted by the COVID-19 pandemic, researchers are eager to make use of a wide variety of data sources, both government-sponsored and alternative, to characterize the epidemiology of infectious diseases. To date, few studies have investigated the strengths and limitations of sources currently being used for such research. These are critical for policy makers to understand when interpreting study findings.MethodsTo fill this gap in the literature, we compared infectious disease reporting for three diseases (measles, mumps, and varicella) across four different data sources: Optum (health insurance billing claims data), HealthMap (online news surveillance data), Morbidity and Mortality Weekly Reports (official government reports), and National Notifiable Disease Surveillance System (government case surveillance data). We reported the yearly number of national- and state-level disease-specific case counts and disease clusters according to each of our sources during a five-year study period (2013–2017).FindingsOur study demonstrated drastic differences in reported infectious disease incidence across data sources. When compared against the other three sources of interest, Optum data showed substantially higher, implausible standardized case counts for all three diseases. Although there was some concordance in identified state-level case counts and disease clusters, all four sources identified variations in state-level reporting.InterpretationResearchers should consider data source limitations when attempting to characterize the epidemiology of infectious diseases. Some data sources, such as billing claims data, may be unsuitable for epidemiological research within the infectious disease context.

The potential for bias in non-representative, large-scale, low-cost survey data can limit their utility for population health measurement and public health decision-making. We developed a multi-step regression framework to bias-adjust vaccination coverage predictions from the large-scale US COVID-19 Trends and Impact Survey that included post-stratification to the American Community Survey and secondary normalization to an unbiased reference indicator. As a case study, we applied this framework to generate county-level predictions of long-run vaccination coverage among children ages 5 to 11 years. Our vaccination coverage predictions suggest a low ceiling on long-term national coverage (46%), detect substantial geographic heterogeneity (ranging from 11% to 91% across counties in the US), and highlight widespread disparities in the pace of scale-up in the three months following Emergency Use Authorization of COVID-19 vaccination for 5 to 11 year-olds. Generally, our analysis demonstrates an approach to leverage differing strengths of multiple sources of information to produce estimates on the time-scale and geographic-scale necessary for proactive decision-making. The utility of large-scale, low-cost survey data for improving population health measurement is amplified when these data are combined with other representative sources of data.

We know very little about how the pandemic impacted outpatient alcohol use disorder (AUD) care and the role of telemedicine.Using OptumLabs® Data Warehouse de-identified administrative claims, we identified AUD cohorts in 2018 (N = 23,204) and 2019 (N = 23,445) and examined outpatient visits the following year, focusing on week 12, corresponding to the March 2020 US COVID-19 emergency declaration, through week 52. Using multivariable logistic regression, we examined the association between patient demographic and clinical characteristics and receipt of any outpatient AUD visits in 2020 vs. 2019.In 2020, weekly AUD visit utilization decreased maximally at the pandemic start (week 12) by 22.5 % (2019: 3.8 %, 2020: 3.0 %, percentage point change [95 % CI] = -0.86[-1.19, -0.05]) but was similar to 2019 utilization by mid-April 2020 (week 16). Telemedicine accounted for 50.1 % of AUD visits by early July 2020 (week 27). Individual therapy returned to 2019 levels within 1 week (i.e., week 13) whereas group therapy did not consistently do so until mid-August 2020 (week 31). Further, individual therapy exceeded 2019 levels by as much as 50 % starting mid-October 2020. The study found no substantial differences in visits by patient demographic or clinical characteristics.Among patients with known AUD, initial outpatient care disruptions were relatively brief. However, substantial shifts occurred in care delivery-an embrace of telemedicine but also more pronounced, longer disruptions in group therapy vs. individual and an increase in individual therapy use. Further research needs to help us understand the implications of these findings for clinical outcomes.

Sherri Rose, Ph.D. is an associate professor at Stanford University in the Center for Health Policy and Center for Primary Care and Outcomes Research as well as Co-Director of the joint Harvard–Stanford Health Policy Data Science Lab. A renowned expert in machine learning methodology for causal inference and prediction, her applied work has focused on risk adjustment, algorithmic fairness, health program evaluation, and comparative effectiveness research. Dr. Rose’s leadership positions include current roles as Co-Editor of Biostatistics and Chair of the American Statistical Association’s Biometrics Section. She is also a Fellow of the American Statistical Association. Dr. Rose earned a BS in Statistics from The George Washington University and a PhD in Biostatistics from the University of California, Berkeley before completing an NSF Mathematical Sciences Postdoctoral Research Fellowship at Johns Hopkins University. Prior to joining the faculty at Stanford University, she was on the faculty at Harvard Medical School in the Department of Health Care Policy. Below, an interview of Dr. Rose, conducted by her colleague, Dr. Laura Hatfield, on the occasion of her 2020 Mid-Career Award from the Health Policy Statistics Section (HPSS) of the American Statistical Association. This award recognizes leaders in health care policy and health services research who have made outstanding contributions through methodological or applied work and who show a promise of continued excellence at the frontier of statistical practice that advances the aims of HPSS.

The field of health services research is broad and seeks to answer questions about the health care system. It is inherently interdisciplinary, and epidemiologists have made crucial contributions. Parametric regression techniques remain standard practice in health services research with machine learning techniques currently having low penetrance in comparison. However, studies in several prominent areas, including health care spending, outcomes and quality, have begun deploying machine learning tools for these applications. Nevertheless, major advances in epidemiological methods are also as yet underleveraged in health services research. This article summarizes the current state of machine learning in key areas of health services research, and discusses important future directions at the intersection of machine learning and epidemiological methods for health services research.

PURPOSE Cancer stage is a key determinant of outcomes; however, stage is not available in claims-based data sources used for real-world evaluations. We compare multiple methods for classifying lung cancer stage from claims data. METHODS Our study used the linked SEER-Medicare data. The patient samples included fee-for-service Medicare beneficiaries diagnosed with lung cancer from 2010 to 2011 (development cohort) and 2012 to 2013 (validation cohort) who received chemotherapy. Classification algorithms considered Medicare Part A and B claims for care in the 3 months before and after chemotherapy initiation. We developed a clinical algorithm to predict stage IV ( v I to III) cancer on the basis of treatment patterns (surgery, radiotherapy, chemotherapy). We also considered an ensemble of claims-based machine learning algorithms. Classification methods were trained in the development cohort, and performance was measured in both cohorts. The SEER data were the gold standard for cancer stage. RESULTS Development and validation cohorts included 14,760 and 14,620 patients with lung cancer, respectively. Validation analyses assessed clinical, random forest, and simple logistic regression algorithms. The best performing classifier within the development cohort was the random forests, but this performance was not replicated in validation analysis. Logistic regression had stable performance across cohorts. Compared with the clinical algorithm, the 14-variable logistic regression algorithm demonstrated higher accuracy in both the development (77% v 71%) and validation cohorts (77% v 73%), with improved specificity for stage IV disease. CONCLUSION Machine learning algorithms have potential to improve lung cancer stage classification but may be prone to overfitting. Use of ensembles, cross-validation, and external validation can aid generalizability. Degradation of accuracy between development and validation cohorts suggests the need for caution in implementing machine learning in research or care delivery.

While much of the causal inference literature has focused on addressing internal validity biases, both internal and external validity are necessary for unbiased estimates in a target population of interest. However, few generalizability approaches exist for estimating causal quantities in a target population when the target population is not well-represented by a randomized study but is reflected when additionally incorporating observational data. To generalize to a target population represented by a union of these data, we propose a class of novel conditional cross-design synthesis estimators that combine randomized and observational data, while addressing their respective biases. The estimators include outcome regression, propensity weighting, and double robust approaches. All use the covariate overlap between the randomized and observational data to remove potential unmeasured confounding bias. We apply these methods to estimate the causal effect of managed care plans on health care spending among Medicaid beneficiaries in New York City., 25 pages, 4 figures; supplement of 31 pages, 12 figures, and 4 tables

Difficulty in identifying cancer stage in health care claims data has limited oncology quality of care and health outcomes research. We fit prediction algorithms for classifying lung cancer stage into three classes (stages I/II, stage III, and stage IV) using claims data, and then demonstrate a method for incorporating the classification uncertainty in outcomes estimation. Leveraging set-valued classification and split conformal inference, we show how a fixed algorithm developed in one cohort of data may be deployed in another, while rigorously accounting for uncertainty from the initial classification step. We demonstrate this process using SEER cancer registry data linked with Medicare claims data., Code available at: https://github.com/sl-bergquist/cancer_classification

Background Use of telemental health has increased among rural Medicare beneficiaries, particularly among individuals with serious mental illness (SMI). Little is known about what leads to the initiation of telemental health. Objective To categorize the different patterns of mental health care use before initiation of telemental health services among individuals with SMI. Methods A cohort of rural beneficiaries with SMI (defined as schizophrenia/related psychotic disorders or bipolar disorder) with an index telemental health visit in 2010-2017 was built using claims for a 20% random sample of fee-for-service Medicare beneficiaries. The authors used latent class analysis to identify classes of mental health care use in the 6 months before the index telemental health visits. Across the classes, the authors also described characteristics of index and subsequent mental health visits. Results The cohort included 4930 rural Medicare beneficiaries with SMI. Three classes of mental health care use before initiation of telemental health were identified. The largest class (n=3066) had minimal use of primary care provider mental health care and the second largest class (n=1537) had minimal specialty mental health care. The smallest class (n=327) was characterized by recent hospitalization or emergency department care. In the overall cohort, index visits were frequently established visits and were often with specialty prescribers. Conclusions Our findings highlight 3 distinct patterns of care before telemental health initiation, providing insight into the role that telemedicine may play in mental health care for rural Medicare beneficiaries with SMI. Overall, telemental health was most often used to maintain care with existing providers.

Purpose: The purpose of this study was to characterize the health status of privately insured gender minority individuals. Methods: We created a diagnosis-based algorithm to identify gende...

Only a small proportion of individuals with a substance use disorder (SUD) receive treatment. The shortage of SUD providers, particularly in rural areas, is an important driver of this treatment gap. Telemedicine could be a means of expanding access to SUD treatment. However, several key regulatory and reimbursement barriers to greater tele-SUD use exist, and both the Congress and the states are considering or have recently passed legislation to address these barriers. To inform these efforts, we describe how tele-SUD is currently being used. Using 2010–2017 claims data from a large commercial insurer, we identify characteristics of tele-SUD users and examine how tele-SUD is being used in conjunction with in-person SUD care. Despite a rapid increase in tele-SUD over the period, we find low use rates overall, particularly relative to the growth in tele-mental health. Tele-SUD is primarily being used as a complement to in-person care and is disproportionately used by those with relatively severe SUD. Given the severity of the opioid epidemic, the low-rates of tele-SUD use that we observe represent a missed opportunity. As availability of tele-SUD is expanded, it will be important to monitor closely which tele-SUD delivery models are being deployed and their impact on access and outcomes.

Background & ObjectiveDuring infectious disease outbreaks, health agencies often share text-based information about cases and deaths. This information is usually text-based and rarely machine-readable, thus creating challenges for outbreak researchers. Here, we introduce a generalizable data assembly algorithm that automatically curates text-based, outbreak-related information and demonstrate its performance across three outbreaks.MethodsAfter developing an algorithm with regular expressions, we automatically curated data from health agencies via three information sources: formal reports, email newsletters, and Twitter. A validation data set was also curated manually for each outbreak.FindingsWhen compared against the validation data sets, the overall cumulative missingness and misidentification of the algorithmically curated data were ≤2% and ≤1%, respectively, for all three outbreaks.ConclusionsWithin the context of outbreak research, our work successfully addresses the need for generalizable tools that can transform text-based information into machine-readable data across varied information sources and infectious diseases.

In human challenge trials, volunteers are deliberately infected with a pathogen to accelerate vaccine development and answer key scientific questions. In the U.S., preparations for challenge trials with the novel coronavirus are complete, and in the U.K., challenge trials have recently begun. However, ethical concerns have been raised about the potential for invalid consent or exploitation. These concerns largely reflect worries that challenge trial volunteers may be unusually risk-seeking or too economically vulnerable to refuse the payments these trials provide, rather than being motivated primarily by altruistic goals. We conducted the first large-scale survey of intended human challenge trial volunteers and found that SARS-CoV-2 challenge trial volunteers exhibit high levels of altruistic motivations without any special indication of poor risk perception or economic vulnerability. Findings indicate that challenge trials with the novel coronavirus can attract volunteers with background conditions, attitudes, and motivations that should allay key ethical concerns.

Percutaneous coronary interventions (PCIs) are nonsurgical procedures to open blocked blood vessels to the heart, frequently using a catheter to place a stent. The catheter can be inserted into the blood vessels using an artery in the groin or an artery in the wrist. Because clinical trials have indicated that access via the wrist may result in fewer post procedure complications, shortening the length of stay, and ultimately cost less than groin access, adoption of access via the wrist has been encouraged. However, patients treated in usual care are likely to differ from those participating in clinical trials, and there is reason to believe that the effectiveness of wrist access may differ between males and females. Moreover, the choice of artery access strategy is likely to be influenced by patient or physician unmeasured factors. To study the effectiveness of the two artery access site strategies on hospitalization charges, we use data from a state-mandated clinical registry including 7,963 patients undergoing PCI. A hierarchical Bayesian likelihood-based instrumental variable analysis under a latent index modeling framework is introduced to jointly model outcomes and treatment status. Our approach accounts for unobserved heterogeneity via a latent factor structure, and permits nonparametric error distributions with Dirichlet process mixture models. Our results demonstrate that artery access in the wrist reduces hospitalization charges compared to access in the groin, with higher mean reduction for male patients., Comment: 11 tables, 5 figures

We consider an extension of Leo Breiman's thesis from "Statistical Modeling: The Two Cultures" to include a bifurcation of algorithmic modeling, focusing on parametric regressions, interpretable algorithms, and complex (possibly explainable) algorithms., Comment: 7 pages, forthcoming in Observational Studies

Primary care is the largest healthcare delivery platform in the US. Facing the Artificial Intelligence and Machine Learning technology (AI/ML) revolution, the primary care community would benefit from a roadmap revealing priority areas and opportunities for developing and integrating AI/ML-driven clinical tools. This article presents a framework that identifies five domains for AI/ML integration in primary care to support care delivery transformation and achieve the Quintuple Aims of the healthcare system. We concluded that primary care plays a critical role in developing, introducing, implementing, and monitoring AI/ML tools in healthcare and must not be overlooked as AI/ML transforms healthcare.

Risk-adjustment systems used to pay health plans in individual health insurance markets have evolved towards better “fit” of payments to plan spending, at the individual and group levels, generally achieved by adding variables used for risk adjustment. Adding variables demands further plan and provider-supplied data. Some data called for in the more complex systems may be easily manipulated by providers, leading to unintended “upcoding” or to unnecessary service utilization. While these drawbacks are recognized, they are hard to quantify and are difficult to balance against the concrete, measurable improvements in fit that may be attained by adding variables to the formula. This paper takes a different approach to improving the performance of health plan payment systems. Using the HHS-HHC V0519 model of plan payment in the Marketplaces as a starting point, we constrain fit at the individual and group level to be as good or better than the current payment model while reducing the number of variables called for in the model. Opportunities for simplification are created by the introduction of three elements in design of plan payment: reinsurance (based on high spending or plan losses), constrained regressions, and powerful machine learning methods for variable selection. We first drop all variables relying on drug claims. Further major reductions in the number of diagnostic-based risk adjustors are possible using machine learning integrated with our constrained regressions. The fit performance of our simpler alternatives is as good or better than the current HHS-HHC V0519 formula.

Key Points Question Was the COVID-19 pandemic associated with changes in mental health–related utilization for individuals with severe mental illness in the United States? Findings In this cohort study of more than 650 000 Medicare beneficiaries with schizophrenia or bipolar I disorders, there were statistically significant decreases in mental health–related outpatient, emergency department, and inpatient use as well as medication fills, particularly early in the pandemic, despite widespread use of telemedicine. Statistically significant differentially greater decreases in outpatient utilization were observed among individuals who were Black, had dual Medicare-Medicaid eligibility, or had disability. Meaning In this study, individuals with severe mental illness experienced substantial disruptions in care during the pandemic, and these disruptions were greater among disadvantaged populations., This cohort study examines changes in mental health care during the pandemic and the use of telemedicine in outpatient care among Medicare beneficiaries with severe mental illness., Importance Little is known about changes in care for individuals with severe mental illness during the COVID-19 pandemic. Objective To examine changes in mental health care during the pandemic and the use of telemedicine in outpatient care among Medicare beneficiaries with severe mental illness. Design, Setting, and Participants This population-based cohort study included Medicare beneficiaries (age ≥18 years) diagnosed with schizophrenia and schizophrenia-related disorders or bipolar I disorder. Care patterns during January to September 2020 for a cohort defined in 2019 were compared with those during January to September 2019 for a cohort defined in 2018. Exposures Start of COVID-19 pandemic in the United States, defined as week 12 of 2020. Main Outcomes and Measures Use of mental health–related outpatient visits, emergency department visits, inpatient care, and oral prescription fills for antipsychotics and mood stabilizers during 4-week intervals. Multivariable logistic regression analyses examined whether the pandemic was associated with differential changes in outpatient care across patient characteristics. Results The 2019 cohort of 686 214 individuals included 389 245 (53.8%) women, 114 073 (15.8%) Black and 526 301 (72.8%) White individuals, and 477 353 individuals (66.0%) younger than 65 years; the 2020 cohort of 723 045 individuals included 367 140 (53.5%) women, 106 699 (15.6%) Black and 497 885 (72.6%) White individuals, and 442 645 individuals (64.5%) younger than 65 years. Compared with 2019, there were large decreases during the pandemic’s first month (calendar weeks 12-15) in individuals with outpatient visits (265 169 [36.7%] vs 200 590 [29.2%]; 20.3% decrease), with antipsychotic and mood stabilizer medication prescription fills (216 468 [29.9%] vs 163 796 [23.9%]; 20.3% decrease), with emergency department visits (12 383 [1.7%] vs 8503 [1.2%]; 27.7% decrease), and with hospital admissions (11 564 [1.6%] vs 7912 [1.2%]; 27.9% decrease). By weeks 32 to 35 of 2020, utilization rebounded but remained lower than in 2019, ranging from a relative decrease of 2.5% (outpatient visits) to 12.9% (admissions). During the full pandemic period (weeks 12-39) in 2020, 1 556 403 of 2 743 553 outpatient visits (56.7%) were provided via telemedicine. In multivariable analyses, outpatient visit use during weeks 12 to 25 of 2020 was lower among those with disability (odds ratio, 0.95; 95% CI, 0.93-0.96), and during weeks 26 to 39 of 2020, it was lower among Black vs non-Hispanic White individuals (OR, 0.97; 95% CI, 0.95-0.99) and those with dual Medicaid eligibility (OR, 0.96; 95% CI, 0.95-0.98). Conclusions and Relevance In this cohort study, despite greater use of telemedicine, individuals with severe mental illness experienced large disruptions in care early in the pandemic. These narrowed but persisted through September 2020. Disruptions were greater for several disadvantaged populations.

Objective To assess the issue of nonrepresentative sampling in Medicare Advantage (MA) risk adjustment. Data sources Medicare enrollment and claims data from 2008 to 2011. Data extraction Risk adjustment predictor variables were created from 2008 to 2010 Part A and B claims and the Medicare Beneficiary Summary File. Spending is based on 2009-2011 Part A and B, Durable Medical Equipment, and Home Health Agency claims files. Study design A propensity-score matched sample of Traditional Medicare (TM) beneficiaries who resembled MA enrollees was created. Risk adjustment formulas were estimated using multiple techniques, and performance was evaluated based on R2 , predictive ratios, and formula coefficients in the matched sample and a random sample of TM beneficiaries. Principal findings Matching improved balance on observables, but performance metrics were similar when comparing risk adjustment formula results fit on and evaluated in the matched sample versus fit on the random sample and evaluated in the matched sample. Conclusions Fitting MA risk adjustment formulas on a random sample versus a matched sample yields little difference in MA plan payments. This does not rule out potential improvements via the matching method should reliable MA encounter data and additional variables become available for risk adjustment.

OBJECTIVE Prompt initiation and intensification of antidiabetic therapy can delay or prevent complications from diabetes. We sought to understand the rates of and factors associated with the initiation and intensification of antidiabetic therapy among commercially insured patients in the U.S. RESEARCH DESIGN AND METHODS Using 2008–2015 commercial claims linked with laboratory and pharmacy data, we created an initiation cohort with no prior antidiabetic drug use and an HbA1c ≥8% (64 mmol/mol) and an intensification cohort of patients with an HbA1c ≥8% (64 mmol/mol) who were on a stable dose of one noninsulin diabetes drug. Using multivariable logistic regression, we determined the rates of and factors associated with initiation and intensification. In addition, we determined the percent of variation in treatment patterns explained by measurable patient factors. RESULTS In the initiation cohort (n = 9,799), 63% of patients received an antidiabetic drug within 6 months of the elevated HbA1c test. In the intensification cohort (n = 10,941), 82% had their existing antidiabetic therapy intensified within 6 months of the elevated HbA1c test. Higher HbA1c levels, lower generic drug copayments, and more frequent office visits were associated with higher rates of both initiation and intensification. Better patient adherence prior to the elevated HbA1c level, existing therapy with a second-generation antidiabetic drug, and lower doses of existing therapy were also associated with intensification. Patient factors explained 7.96% of the variation in initiation and 7.35% of the variation in intensification. CONCLUSIONS Approximately two-thirds of patients were newly initiated on antidiabetic therapy, and four-fifths of those already receiving antidiabetic therapy had it intensified within 6 months of an elevated HbA1c in a commercially insured population. Patient factors explain 7–8% of the variation in diabetes treatment patterns.

This article examines depictions of blindness across painting, poetry and theatre. Bruegel’s sixteenth-century painting, The Parable of the Blind, strongly resonates with Baudelaire’s and Maeterlin...

Earthquakes are a common and deadly natural disaster, with roughly one-quarter of survivors subsequently developing posttraumatic stress disorder (PTSD). Despite progress identifying risk factors, limited research has examined how to combine variables into an optimized post-earthquake PTSD prediction tool that could be used to triage survivors to mental health services. The current study developed a post-earthquake PTSD risk score using machine learning methods designed to optimize prediction. The data were from a two-wave survey of Chileans exposed to the 8.8 magnitude earthquake that occurred in February 2010. Respondents (n = 23,907) were interviewed roughly three months prior to and again three months after the earthquake. Probable post-earthquake PTSD was assessed using the Davidson Trauma Scale. We applied super learning, an ensembling machine learning method, to develop the PTSD risk score from 67 risk factors that could be assessed within one week of earthquake occurrence. The super learner algorithm had better cross-validated performance than the 39 individual algorithms from which it was developed, including conventional logistic regression. The super learner also had a better area under the receiver operating characteristic curve (0.79) than existing post-disaster PTSD risk tools. Individuals in the top 5%, 10%, and 20% of the predicted risk distribution accounted for 17.5%, 32.2%, and 51.4% of all probable cases of PTSD, respectively. In addition to developing a risk score that could be implemented in the near future, these results more broadly support the utility of super learning to develop optimized prediction functions for mental health outcomes.

This cross-sectional study investigates which community factors may be associated with the increase in telemedicine use during the COVID-19 pandemic.

While family purchase of health insurance may benefit insurance markets by pooling individual risk into family groups, the correlation across illness types in families could exacerbate adverse selection. We analyze the impact of family pooling on risk for health insurers to inform policy about family-level insurance plans. Using data on 8,927,918 enrollees in fee-for-service commercial health plans in the 2013 Truven MarketScan database, we compare the distribution of annual individual health spending across four pooling scenarios: (1) “Individual” where there is no pooling into families; (2) “real families” where costs are pooled within families; (3) “random groups” where costs are pooled within randomly generated small groups that mimic families in group size; and (4) “the Sims” where costs are pooled within random small groups which match families in demographics and size. These four simulations allow us to identify the separate contributions of group size, group composition, and family affinity in family risk pooling. Variation in individual spending under family pooling is very similar to that within “simulated families” and to that within random groups, and substantially lower than when there is no family pooling and individuals choose independently (standard deviation $12,526 vs. $11,919, $12,521 and $17,890 respectively). Within-family correlations in health status and utilization do not “undo” the gains from family pooling of risks. Family pooling can mitigate selection and improve the functioning of health insurance markets.

Objective: Widespread application of clinical natural language processing (NLP) systems requires taking existing NLP systems and adapting them to diverse and heterogeneous settings. We describe the challenges faced and lessons learned in adapting an existing NLP system for measuring colonoscopy quality.Materials and Methods: Colonoscopy and pathology reports from 4 settings during 2013–2015, varying by geographic location, practice type, compensation structure, and electronic health record.Results: Though successful, adaptation required considerably more time and effort than anticipated. Typical NLP challenges in assembling corpora, diverse report structures, and idiosyncratic linguistic content were greatly magnified.Discussion: Strategies for addressing adaptation challenges include assessing site-specific diversity, setting realistic timelines, leveraging local electronic health record expertise, and undertaking extensive iterative development. More research is needed on how to make it easier to adapt NLP systems to new clinical settings.Conclusions: A key challenge in widespread application of NLP is adapting existing systems to new clinical settings.

SUMMARY Health insurers may attempt to design their health plans to attract profitable enrollees while deterring unprofitable ones. Such insurers would not be delivering socially efficient levels of care by providing health plans that maximize societal benefit, but rather intentionally distorting plan benefits to avoid high-cost enrollees, potentially to the detriment of health and efficiency. In this work, we focus on a specific component of health plan design at risk for health insurer distortion in the Health Insurance Marketplaces: the prescription drug formulary. We introduce an ensembled machine learning function to determine whether drug utilization variables are predictive of a new measure of enrollee unprofitability we derive, and thus vulnerable to distortions by insurers. Our implementation also contains a unique application-specific variable selection tool. This study demonstrates that super learning is effective in extracting the relevant signal for this prediction problem, and that a small number of drug variables can be used to identify unprofitable enrollees. The results are both encouraging and concerning. While risk adjustment appears to have been reasonably successful at weakening the relationship between therapeutic-class-specific drug utilization and unprofitability, some classes remain predictive of insurer losses. The vulnerable enrollees whose prescription drug regimens include drugs in these classes may need special protection from regulators in health insurance market design.