Publications

Using 2003−2018 National Ambulatory Medical Care Survey data for office-based visits and 2003−2018 National Hospital Ambulatory Medical Care Survey data for emergency department (ED) visits, we conducted cross-sectional analyses to examine cough medication (CM) use trends in the United States (US) ambulatory care settings. We included adult (≥18 years) patient visits with respiratory-infection-related or non-infection-related cough as reason-for-visit or diagnosis without malignant cancer or benign respiratory tumor diagnoses. Using multivariable logistic regressions, we examined opioid antitussive, benzonatate, dextromethorphan-containing antitussive, and gabapentinoid use trends. From 2003−2005 to 2015−2018, opioid antitussive use decreased in office-based visits (8.8% to 6.4%, Ptrend = 0.03) but remained stable in ED visits (6.3% to 5.9%, Ptrend = 0.99). In both settings, hydrocodone-containing antitussive use declined over 50%. Benzonatate use more than tripled (office-based:1.6% to 4.8%; ED:1.5% to 8.0%; both Ptrend < 0.001). Dextromethorphan-containing antitussive use increased in ED visits (1.8% to 2.6%, Ptrend = 0.003) but stayed unchanged in office-based visits (3.8% to 2.7%; Ptrend = 0.60). Gabapentinoid use doubled in office-based visits (1.1% in 2006−2008 to 2.4% in 2015−2018, Ptrend < 0.001) but was negligible in ED visits. In US office-based and ED ambulatory care settings, hydrocodone-containing antitussive use substantially declined from 2003 to 2018, while benzonatate use more than tripled, and dextromethorphan-containing antitussive and gabapentinoid use remained low (<3%).

IMPORTANCE: State Medicaid programs have implemented initiatives to expand treatment coverage for opioid use disorder (OUD); however, some Medicaid programs still require prior authorizations (PAs) for filling buprenorphine prescriptions.

OBJECTIVE: To evaluate the changes in buprenorphine use for OUD among Medicaid enrollees in states that completely removed buprenorphine PA requirements.

DESIGN SETTING AND PARTICIPANTS: This retrospective cross-sectional study analyzed the immediate and trend changes on buprenorphine use during 2013 to 2020 associated with removal of PA requirements using a controlled interrupted time series analysis to account for autocorrelation. Data were collected from Medicaid State Drug Utilization Data for 2 states (California and Illinois) that completely removed a buprenorphine PA during the study period, and buprenorphine prescriptions for OUD treatment were identified among Medicaid enrollees.

MAIN OUTCOMES AND MEASURES: Quarterly total number of buprenorphine prescriptions for each state was calculated, and stratification analyses were conducted by dosage form (films and tablets).

RESULTS: Among the 2 state Medicaid programs (California and Illinois) that removed buprenorphine PAs, there was a total of 702 643 and 415 115 eligible buprenorphine prescription claims, respectively. After removing PA requirements for buprenorphine, there was an immediate increase that was not statistically significant (rate ratio [RR], 1.11; 95% CI, 0.76-1.61) in the number of all buprenorphine prescriptions in California and a statistically significant increase (RR, 6.99; 95% CI, 4.67-10.47) in the number of all buprenorphine prescriptions in Illinois relative to the change in the control states (Alabama, Florida, Idaho, Kansas, Mississippi, Nevada, South Dakota, and Wyoming). Additionally, there was a statistically significant decreasing trend in the number of all buprenorphine prescriptions in California (RR, 0.88; 95% CI, 0.82-0.94) and a statistically significant increasing trend in Illinois (RR, 1.11; 95% CI, 1.05-1.19) relative to the trend in control states.

CONCLUSIONS AND RELEVANCE: In this cross-sectional study, removal of buprenorphine PA requirements was associated with a statistically significant increase in the number of buprenorphine prescription fills among Medicaid populations in 1 of the 2 included states.

Suicide is a leading cause of death in the US. Patients with pain conditions have higher suicidal risks. In a systematic review searching observational studies from multiple sources (e.g., MEDLINE) from 1 January 2000-12 September 2020, we evaluated existing suicide prediction models' (SPMs) performance and identified risk factors and their derived data sources among patients with pain conditions. The suicide-related outcomes included suicidal ideation, suicide attempts, suicide deaths, and suicide behaviors. Among the 87 studies included (with 8 SPM studies), 107 suicide risk factors (grouped into 27 categories) were identified. The most frequently occurring risk factor category was depression and their severity (33%). Approximately 20% of the risk factor categories would require identification from data sources beyond structured data (e.g., clinical notes). For 8 SPM studies (only 2 performing validation), the reported prediction metrics/performance varied: C-statistics (n = 3 studies) ranged 0.67-0.84, overall accuracy(n = 5): 0.78-0.96, sensitivity(n = 2): 0.65-0.91, and positive predictive values(n = 3): 0.01-0.43. Using the modified Quality in Prognosis Studies tool to assess the risk of biases, four SPM studies had moderate-to-high risk of biases. This systematic review identified a comprehensive list of risk factors that may improve predicting suicidal risks for patients with pain conditions. Future studies need to examine reasons for performance variations and SPM's clinical utility.

PURPOSE: To conduct a systematic review to identify studies that assessed the association between CYP2C19 polymorphisms and clinical outcomes in peripheral artery disease (PAD) patients who took clopidogrel.

METHODS: We systematically searched Ovid EMBASE, PubMed, and Web of Science from November 1997 (inception) to September 2020. We included observational studies evaluating how CYP2C19 polymorphism is associated with clopidogrel's effectiveness and safety among patients with PAD. We extracted relevant information details from eligible studies (e.g., study type, patient population, study outcomes). We used the Risk of Bias in Non-randomized Studies-of Interventions (ROBINS-I) Tool to assess the risk of bias for included observational studies.

RESULTS: The outcomes of interest were the effectiveness and safety of clopidogrel. The effectiveness outcomes included clinical ineffectiveness (e.g., restenosis). The safety outcomes included bleeding and death related to the use of clopidogrel. We identified four observational studies with a sample size ranging from 50 to 278. Outcomes and comparison groups of the studies varied. Three studies (75%) had an overall low risk of bias. All included studies demonstrated that carrying CYP2C19 loss of function (LOF) alleles was significantly associated with reduced clinical effectiveness and safety of clopidogrel.

CONCLUSIONS: Our systematic review showed an association between CYP2C19 LOF alleles and reduced functions of clopidogrel. The use of CYP2C19 testing in PAD patients prescribed clopidogrel may help improve the clinical outcomes. However, based on the limited evidence, there is a need for randomized clinical trials in PAD patients to test both the effectiveness and safety outcomes of clopidogrel.

BACKGROUND AND AIMS: The time lag encountered when accessing health-care data is one major barrier to implementing opioid overdose prediction measures in practice. Little is known regarding how one's opioid overdose risk changes over time. We aimed to identify longitudinal patterns of individual predicted overdose risks among Medicaid beneficiaries after initiation of opioid prescriptions.

DESIGN, SETTING AND PARTICIPANTS: A retrospective cohort study in Pennsylvania, USA among Pennsylvania Medicaid beneficiaries aged 18-64 years who initiated opioid prescriptions between July 2017 and September 2018 (318 585 eligible beneficiaries (mean age = 39 ± 12 years, female = 65.7%, White = 62.2% and Black = 24.9%).

MEASUREMENTS: We first applied a previously developed and validated machine-learning algorithm to obtain risk scores for opioid overdose emergency room or hospital visits in 3-month intervals for each beneficiary who initiated opioid therapy, until disenrollment from Medicaid, death or the end of observation (December 2018). We performed group-based trajectory modeling to identify trajectories of these predicted overdose risk scores over time.

FINDINGS: Among eligible beneficiaries, 0.61% had one or more occurrences of opioid overdose in a median follow-up of 15 months. We identified five unique opioid overdose risk trajectories: three trajectories (accounting for 92% of the cohort) had consistent overdose risk over time, including consistent low-risk (63%), consistent medium-risk (25%) and consistent high-risk (4%) groups; another two trajectories (accounting for 8%) had overdose risks that substantially changed over time, including a group that transitioned from high- to medium-risk (3%) and another group that increased from medium- to high-risk over time (5%).

CONCLUSIONS: More than 90% of Medicaid beneficiaries in Pennsylvania USA with one or more opioid prescriptions had consistent, predicted opioid overdose risks over 15 months. Applying opioid prediction algorithms developed from historical data may not be a major barrier to implementation in practice for the large majority of individuals.

BACKGROUND AND AIMS: We aimed to develop and validate machine learning algorithms to predict direct-acting antiviral (DAA) treatment failure among patients with HCV infection.

APPROACH AND RESULTS: We used HCV-TARGET registry data to identify HCV-infected adults receiving all-oral DAA treatment and having virologic outcome. Potential pretreatment predictors (n = 179) included sociodemographic, clinical characteristics, and virologic data. We applied multivariable logistic regression as well as elastic net, random forest, gradient boosting machine (GBM), and feedforward neural network machine learning algorithms to predict DAA treatment failure. Training (n = 4894) and validation (n = 1631) patient samples had similar sociodemographic and clinical characteristics (mean age, 57 years; 60% male; 66% White; 36% with cirrhosis). Of 6525 HCV-infected adults, 95.3% achieved sustained virologic response, whereas 4.7% experienced DAA treatment failure. In the validation sample, machine learning approaches performed similarly in predicting DAA treatment failure (C statistic [95% CI]: GBM, 0.69 [0.64-0.74]; random forest, 0.68 [0.63-0.73]; feedforward neural network, 0.66 [0.60-0.71]; elastic net, 0.64 [0.59-0.70]), and all four outperformed multivariable logistic regression (0.51 [0.46-0.57]). Using the Youden index to identify the balanced risk score threshold, GBM had 66.2% sensitivity and 65.1% specificity, and 12 individuals were needed to evaluate to identify 1 DAA treatment failure. Over 55% of patients with treatment failure were classified by the GBM in the top three risk decile subgroups (positive predictive value: 6%-14%). The top 10 GBM-identified predictors included albumin, liver enzymes (aspartate aminotransferase, alkaline phosphatase), total bilirubin levels, sex, HCV viral loads, sodium level, HCC, platelet levels, and tobacco use.

CONCLUSIONS: Machine learning algorithms performed effectively for risk prediction and stratification of DAA treatment failure.

BACKGROUND: Gabapentinoids are recommended by guidelines as a component of multimodal analgesia to manage postoperative pain and reduce opioid use. It remains unknown whether perioperative use of gabapentinoids is associated with a reduced or increased risk of postoperative long-term opioid use (LTOU) after total knee or hip arthroplasty (TKA/THA).

METHODS: Using Medicare claims data from 2011 to 2018, we identified fee-for-service beneficiaries aged ≥ 65 years who were hospitalized for a primary TKA/THA and had no LTOU before the surgery. Perioperative use of gabapentinoids was measured from 7 days preadmission through 7 days postdischarge. Patients were required to receive opioids during the perioperative period and were followed from day 7 postdischarge for 180 days to assess postoperative LTOU (ie, ≥90 consecutive days). A modified Poisson regression was used to estimate the relative risk (RR) of postoperative LTOU in patients with versus without perioperative use of gabapentinoids, adjusting for confounders through propensity score weighting.

RESULTS: Of 52,788 eligible Medicare older beneficiaries (mean standard deviation [SD] age 72.7 [5.3]; 62.5% females; 89.7% White), 3,967 (7.5%) received gabapentinoids during the perioperative period. Postoperative LTOU was 3.8% in patients with and 4.0% in those without perioperative gabapentinoids. After adjusting for confounders, the risk of postoperative LTOU was similar comparing patients with versus without perioperative gabapentinoids (RR = 1.07; 95% confidence interval [CI] = 0.91-1.26, P = .408). Sensitivity and bias analyses yielded consistent results.

CONCLUSION: Among older Medicare beneficiaries undergoing a primary TKA/THA, perioperative use of gabapentinoids was not associated with a reduced or increased risk for postoperative LTOU.

Despite the availability of efficacious direct-acting antiviral (DAA) therapy, the number of people infected with hepatitis C virus (HCV) continues to rise, and HCV remains a leading cause of liver-related morbidity, liver transplantation, and mortality. We developed and validated machine learning (ML) algorithms to predict DAA treatment failure. Using the HCV-TARGET registry of adults who initiated all-oral DAA treatment, we developed elastic net (EN), random forest (RF), gradient boosting machine (GBM), and feedforward neural network (FNN) ML algorithms. Model performances were compared with multivariable logistic regression (MLR) by assessing C statistics and other prediction evaluation metrics. Among 6525 HCV-infected adults, 308 patients (4.7%) experienced DAA treatment failure. ML models performed similarly in predicting DAA treatment failure (C statistic [95% CI]: EN, 0.74 [0.69-0.79]; RF, 0.74 [0.69-0.80]; GBM, 0.72 [0.67-0.78]; FNN, 0.75 [0.70-0.80]), and all 4 outperformed MLR (C statistic [95% CI]: 0.51 [0.46-0.57]), and EN used the fewest predictors (n = 27). With Youden index, the EN had 58.4% sensitivity and 77.8% specificity, and nine patients were needed to evaluate to identify 1 DAA treatment failure. Over 60% treatment failure were classified in top three risk decile subgroups. EN-identified predictors included male sex, treatment < 8 weeks, treatment discontinuation due to adverse events, albumin level < 3.5 g/dL, total bilirubin level > 1.2 g/dL, advanced liver disease, and use of tobacco, alcohol, or vitamins. Addressing modifiable factors of DAA treatment failure may reduce the burden of retreatment. Machine learning algorithms have the potential to inform public health policies regarding curative treatment of HCV.

BACKGROUND: Although opioids are initiated on hospital discharge in millions of older adults each year, there are no studies examining patient- and prescribing-related risk factors for opioid-related adverse drug events (ADEs) after hospital discharge among medical patients.

METHODS: A retrospective cohort study of a national sample of Medicare beneficiaries aged 65 years and older, hospitalized for a medical reason, with at least one claim for an opioid within 2 days of hospital discharge. We excluded patients receiving hospice care and patients admitted from or discharged to a facility. We used administrative billing codes and medication claims to define potential opioid-related ADEs within 30 days of hospital discharge, and competing risks regression to identify risk factors for these events.

RESULTS: Among 22,879 medical hospitalizations (median age 74, 36.9% female) with an opioid claim within 2 days of hospital discharge, a potential opioid-related ADE occurred in 1604 (7.0%). Independent risk factors included age of 80 years and older (HR 1.18, 95% CI 1.05-1.33); clinical conditions, including kidney disease (HR 1.22, 95% CI 1.08-1.37), dementia/delirium (HR 1.38, 95% CI 1.22-1.56), anxiety disorder (HR 1.20, 95% CI 1.06-1.36), opioid use disorder (HR 1.20, 95% CI 1.03-1.39), intestinal disorders (HR 1.31, 95% CI 1.15-1.49), pancreaticobiliary disorders (HR 1.32, 95% CI 1.09-1.61), and musculoskeletal and nervous system injuries (HR 1.35, 95% CI 1.17-1.54); red flags for opioid misuse (HR 1.37, 95% CI 1.04-1.80); opioid use in the 30 days before hospitalization (HR 1.20, 95% CI 1.08-1.34); and prescription of long-acting opioids (HR 1.34, 95% CI 1.06-1.70).

CONCLUSIONS: Potential opioid-related ADEs occurred within 30 days of hospital discharge in 7.0% of older adults discharged from a medical hospitalization with an opioid prescription. Identified risk factors can be used to inform physician decision-making, conversations with older adults about risk, and development and targeting of harm reduction strategies.