Evaluating the Mediating Role of Recall of Intervention Knowledge in the Relationship Between a Peer-Driven Intervention and HIV Risk Behaviors Among People Who Inject Drugs.

Peer-driven interventions can be effective in reducing HIV injection risk behaviors among people who inject drugs (PWID). We employed a causal mediation framework to examine the mediating role of recall of intervention knowledge in the relationship between a peer-driven intervention and subsequent self-reported HIV injection-related risk behavior among PWID in the HIV Prevention Trials Network (HPTN) 037 study. For each intervention network, the index participant received training at baseline to become a peer educator, while non-index participants and all participants in the control networks received only HIV testing and counseling; recall of intervention knowledge was measured at the 6-month visit for each participant, and each participant was followed to ascertain HIV injection-related risk behaviors at the 12-month visit. We used inverse probability weighting to fit marginal structural models to estimate the total effect (TE) and controlled direct effect (CDE) of the intervention on the outcome. The proportion eliminated (PE) by intervening to remove mediation by the recall of intervention knowledge was computed. There were 385 participants (47% in intervention networks) included in the analysis. The TE and CDE risk ratios for the intervention were 0.47 [95% confidence interval (CI): 0.28, 0.78] and 0.73 (95% CI: 0.26, 2.06) and the PE was 49%. Compared to participants in the control networks, the peer-driven intervention reduced the risk of HIV injection-related risk behavior by 53%. The mediating role of recall of intervention knowledge accounted for less than 50% of the total effect of the intervention, suggesting that other potential causal pathways between the intervention and the outcome, such as motivation and skill, self-efficacy, social norms and behavior modeling, should be considered in future studies.

Network-based Analysis of Prescription Opioids Dispensing Using Exponential Random Graph Models (ERGMs).

The United States has been experiencing an unprecedented level of opioid overdose-related mortality due in part to excessive use of prescription opioids. Peer-driven network interventions may be beneficial. A key assumption of social network interventions is that of some members of the network act as key players and can influence the behavior of others in the network. We used opioid prescription records to create a social network of patients who use prescription opioid in the state of Rhode Island. The study population was restricted to patients on stable opioid regimens who used one source of payment and received the same opioid medication from ≥ 3 prescribers and pharmacies. An exponential random graph model (ERGM) was employed to examine the relationship between patient attributes and the likelihood of tie formation and modularity was used to assess for homophily (the tendency of individuals to associate with similar people). We used multivariable logistic regression to assess predictors of high betweenness centrality, a measure of influence within the network. 372 patients were included in the analysis; average age was 51 years; 53% were female; 57% were prescribed oxycodone, 34% were prescribed hydrocodone and 9% were prescribed buprenorphine/naloxone. After controlling for the main effects in the ERGM model, homophily was associated with age group, method of payment, number and type of opioid prescriptions filled, mean daily dose, and number of providers seen. Type of opioid and number of prescribers were identified as significant predictors of high betweenness centrality. We conclude that patients who use multiple prescribers or have a diagnosis of opioid use disorder may help promote positive health behaviors or disrupt harmful behaviors in an opioid prescription network.

Toward evaluation of disseminated effects of medications for opioid use disorder within provider-based clusters using routinely-collected health data.

Routinely-collected health data can be employed to emulate a target trial when randomized trial data are not available. Patients within provider-based clusters likely exert and share influence on each other's treatment preferences and subsequent health outcomes and this is known as dissemination or spillover. Extending a framework to replicate an idealized two-stage randomized trial using routinely-collected health data, an evaluation of disseminated effects within provider-based clusters is possible. In this article, we propose a novel application of causal inference methods for dissemination to retrospective cohort studies in administrative claims data and evaluate the impact of the normality of the random effects distribution for the cluster-level propensity score on estimation of the causal parameters. An extensive simulation study was conducted to study the robustness of the methods under different distributions of the random effects. We applied these methods to evaluate baseline prescription for medications for opioid use disorder among a cohort of patients diagnosed with opioid use disorder and adjust for baseline confounders using information obtained from an administrative claims database. We discuss future research directions in this setting to better address unmeasured confounding in the presence of disseminated effects.

Early buprenorphine-naloxone initiation for opioid use disorder reduces opioid overdose, emergency room visits, and healthcare cost compared to late initiation.

Background: Although the effectiveness of buprenorphine-naloxone (BUP-NX) has been established, real-world evidence on the benefits of early treatment initiation is limited.Objective: To evaluate the association between early BUP-NX initiation and health-related outcomes among insured adults with opioid use disorder (OUD).Methods: We conducted a cross-sectional analysis using the Optum's de-identified Clinformatics® Data Mart Database from 2010 to 2018. Patients who initiated BUP-NX within 30 days of OUD diagnosis were classified as early initiators. Patients who initiated BUP-NX later, but within the one-year follow-up, were defined as late initiators. Outcomes included opioid overdose, opioid overdose-related emergency department (ED) visits, and all-cause healthcare cost during the year after OUD diagnosis. We employed generalized linear models to compare outcomes between early and late initiators, adjusting for baseline covariates and accounting for missing information for covariates using multiple imputation.Results: A total of 8,388 patients with OUD were identified; mean age was 39.9 years; 36% were female; and 67.6% were early initiators. Early initiators had an estimated 42% lower rate of opioid overdose (adjusted rate ratio (aRR) = 0.58; 95% confidence interval (CI): 0.52, 0.64); 51% lower rate of opioid overdose-related ED visits (aRR = 0.49; 95% CI: 0.44, 0.55); and 31% lower total healthcare cost (adjusted cost ratio = 0.69; 95% CI: 0.66, 0.72), compared to late initiators.Conclusion: Compared to late BUP-NX initiation, early initiation was associated with a lower risk of opioid overdose and opioid overdose-related ED visits, and reduced total healthcare cost among insured adult patients with OUD.

Initial Patterns of Prescription Opioid Supply and Risk of Mortality Among Insured Adults in the United States.

OBJECTIVE: To examine the association between initial patterns of prescription opioid supply (POS) and risk of all-cause mortality among an insured opioid-naïve patient population in the United States (US). METHODS: This retrospective observational cohort study used de-identified, administrative health care claims data from a large national insurer (Optum Clinformatics Data Mart) from 2010 to 2015. Participants included insured, cancer-free adults prescribed opioid analgesics. Prescription opioids received during the first 6 months of therapy were used to categorize initial patterns of POS as daily or nondaily. Cox regression was used to estimate the association of initial patterns of POS with all-cause mortality within one year of follow-up, adjusting for baseline covariates to control for confounding. RESULTS: A total of 4,054,417 patients were included, of which 2.75% had incident daily POS; 54.8% were female; median age was 50 years; mean Charlson comorbidity index (CCI) was 0.21 (standard deviation = 0.77); and mean daily morphine milligram equivalent was 34.61 (95% confidence intervals: 34.59, 34.63). There were 2068 more deaths per 100,000 person-years among patients who were prescribed opioids daily than nondaily. After adjusting for baseline covariates, the hazard of all-cause mortality among patients with incident daily POS was nearly twice that among those prescribed nondaily (hazard ratio [HR] = 1.94; 95% confidence intervals: 1.84, 2.04). CONCLUSIONS: Among insured adult patients with noncancer pain, incident chronic POS was associated with a significantly increased risk of all-cause mortality over at most 1 year of follow-up. Because these results may be susceptible to bias, more research is needed to establish causality.

Guideline-Concordant Treatment Among Elderly Women With HER2-Positive Metastatic Breast Cancer in the United States.

BACKGROUND: It is crucial to identify whether women with HER2-positive (HER2+) metastatic breast cancer (MBC) are treated according to treatment guidelines and whether treatment disparities exist. This study examined guideline-concordant treatment among women with HER2+ MBC and determined the magnitude of differences in treatment between those with positive and negative hormone receptor (HR) status using a nonlinear decomposition technique. METHODS: A retrospective observational cohort study was conducted using the SEER-Medicare linked database. The study cohort consisted of women aged ≥66 years diagnosed with HER2+ MBC in 2010 through 2013 (n=241). Guideline-concordant initial treatment after cancer diagnosis was defined based on the NCCN Clinical Practice Guidelines in Oncology for Breast Cancer. A multivariable logistic regression was performed to identify significant predictors of guideline-concordant treatment. A postregression decomposition was conducted to identify the magnitude of disparities in treatment by HR status. RESULTS: Of 241 women included in the study, a total of 76.8% received guideline-concordant treatment. These women were significantly more likely to have positive HR status (P=.0298), have good performance status (P=.0009), and more oncology visits (P<.0001). With 1-year increments in age at cancer diagnosis, the likelihood of receiving guideline-concordant treatment reduced by 5% (P=.0356). The decomposition analysis revealed that 19.0% of the disparity in guideline-concordant treatment between women with positive and negative HR status was explained by differences in their characteristics. Enabling characteristics (marital status, income, and education) explained the highest (22.8%) proportion of the disparity. CONCLUSIONS: Nearly one-quarter of the study cohort did not receive guideline-concordant treatment. Our findings suggest opportunities to improve cancer care for elderly women with negative HR status who are unpartnered or have lower socioeconomic status. The high unexplained portion of the disparity by HR status can be due to patient treatment preferences, propensity to seek care, and organizational and physician-level characteristics that were not included in the study.

Chronic opioid use in women following hysterectomy: Patterns and predictors.

BACKGROUND: Most women are prescribed an opioid after hysterectomy. The goal of this study was to determine the association between initial opioid prescribing characteristics and chronic opioid use after hysterectomy. METHODS: This study included women enrolled in a commercial health plan who had a hysterectomy between 1 July 2010 and 31 March 2015. We used trajectory models to define chronic opioid use as patients with the highest probability of having an opioid prescription filled during the 6 months post-surgery. A multivariable logistic regression was applied to examine the association between initial opioid dispensing (amount prescribed and duration of treatment) and chronic opioid use after adjusting for potential confounders. RESULTS: A total of 693 of 50 127 (1.38%) opioid-naïve women met the criteria for chronic opioid use following hysterectomy. The baseline variables and initial opioid prescription characteristics predicted the pattern of long-term opioid use with moderate discrimination (c statistic = 0.70). Significant predictors of chronic opioid use included initial opioid daily dose (≥60 MME vs <40 MME, aOR: 1.43, 95% CI: 1.14-1.79) and days' supply (4-7 days vs 1-3 days, aOR: 1.28, 95% CI: 1.06-1.54; ≥8 days vs 1-3 days, aOR: 1.41, 95% CI: 1.05-1.89). Other significant baseline predictors included older age, abdominal or laparoscopic/robotic hysterectomy, tobacco use, psychiatric medication use, back pain, and headache. CONCLUSION: Initial opioid prescribing characteristics are associated with the risk of chronic opioid use after hysterectomy. Prescribing lower daily doses and shorter days' supply of opioids to women after hysterectomy may result in lower risk of chronic opioid use.

Real-World Direct Health Care Costs Associated with Psychotropic Polypharmacy Among Adults with Common Cancer Types in the United States.

BACKGROUND: Psychotropic polypharmacy is not uncommon among cancer patients and may contribute to the increased direct health care cost burden in this population. OBJECTIVE: To estimate average direct health care costs in the year following cancer diagnosis among cancer patients receiving psychotropic polypharmacy compared with those without psychotropic polypharmacy, using a multivariable analysis framework. METHODS: A retrospective cross-sectional study was conducted among patients aged 18 years and older diagnosed with the most commonly occurring cancers (breast, prostate, lung, and colorectal) in the United States during 2011-2012 using the deidentified Optum Clinformatics Data Mart commercial claims database. Psychotropic polypharmacy was defined as concurrent use of 2 or more psychotropic medications for at least 90 days. Direct health care costs in the year following cancer diagnosis were estimated as total medical payments made by the health plans and were derived from claims files. A generalized linear regression model with log-link function and gamma distribution was used to model average direct health care costs, controlling for baseline patient demographic and clinical covariates. RESULTS: Average annual direct health care costs for cancer patients with psychotropic polypharmacy ($53,497; SD $72,590) were higher than those without psychotropic polypharmacy ($38,255; SD $59,844), with an unadjusted average cost difference of $15,242 (P < 0.0001). In the adjusted regression model, the average difference in costs shrunk to $5,888 but remained notable. When examined by type of cancer, average direct health care costs for all cancer patients with psychotropic polypharmacy were significantly higher than those for patients without psychotropic polypharmacy, except for colorectal cancer patients. CONCLUSIONS: Overall health care costs were higher among cancer patients with psychotropic polypharmacy compared with those without psychotropic polypharmacy. Our findings support the need for future research to better understand the benefits and risks of psychotropic polypharmacy, given its potential to cause adverse health outcomes and avoidable health care utilization and costs for this vulnerable patient population. DISCLOSURES: This study was funded by the American Association of Colleges of Pharmacy (AACP) New Investigator Award mechanism, which was received by Vyas. Aroke was partially supported by the AACP grant for conducting data analysis of the study. Kogut is partially supported by Institutional Development Award Number U54GM115677 from the National Institute of General Medical Sciences of the National Institutes of Health, which funds Advance Clinical and Translational Research (Advance-CTR). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health and the AACP. The authors report no conflicts of interest. An abstract of this study was presented as a poster at the American Association of Colleges of Pharmacy Annual Meeting on July 22, 2018, in Boston, MA.

Prevalence of Psychotropic Polypharmacy and Associated Healthcare Resource Utilization during Initial Phase of Care among Adults with Cancer in USA.

BACKGROUND: The use of psychotropic medications is not uncommon among patients with newly diagnosed cancer. However, the impact of psychotropic polypharmacy on healthcare utilization during the initial phase of cancer care is largely unknown. METHODS: We used a claims database to identify adults with incident breast, prostate, lung, and colorectal cancers diagnosed during 2011-12. Psychotropic polypharmacy was defined as concurrent use of two or more psychotropic medication classes for at least 90 days. A multivariable logistic regression was performed to identify significant predictors of psychotropic polypharmacy. Multivariable Poisson and negative binomial regressions were used to assess the associations between psychotropic polypharmacy and healthcare utilization. RESULTS: Among 5604 patients included in the study, 52.6% had breast cancer, 30.6% had prostate cancer, 11.4% had colorectal cancer, and 5.5% had lung cancer. During the year following incident cancer diagnosis, psychotropic polypharmacy was reported in 7.4% of patients, with the highest prevalence among patients with lung cancer (14.4%). Compared with patients without psychotropic polypharmacy during the initial phase of care, patients with newly diagnosed cancer with psychotropic polypharmacy had a 30% higher rate of physician office visits, an 18% higher rate of hospitalization, and a 30% higher rate of outpatient visits. The rate of emergency room visits was similar between the two groups. CONCLUSION: Psychotropic polypharmacy during the initial phase of cancer care was associated with significantly increased healthcare resource utilization, and the proportion of patients receiving psychotropic polypharmacy differed by type of cancer. IMPACT: Findings emphasize the importance of evidence-based psychotropic prescribing and close surveillance of events causing increased healthcare utilization among patients with cancer receiving psychotropic polypharmacy.

Estimating the Direct Costs of Outpatient Opioid Prescriptions: A Retrospective Analysis of Data from the Rhode Island Prescription Drug Monitoring Program.

BACKGROUND: Overuse and misuse of prescription opioids is associated with increased morbidity and mortality and places a significant cost burden on health systems. OBJECTIVE: To estimate annual statewide spending for prescription opioids in Rhode Island. METHODS: A cross-sectional study of opioids dispensed from retail pharmacies using data from the Rhode Island Prescription Drug Monitoring Program (PDMP) was performed. The study sample consisted of 651,227 opioid prescriptions dispensed to 197,062 patients between January 1, 2015, and December 31, 2015. The mean, median, and total cost of opioid use was estimated using prescription dispensings and patients as units of analysis. A generalized linear model with gamma distribution with an identity link function, and separately with a log link function, was used to estimate the absolute and relative differences in per-patient annual adjusted average opioid prescription cost, respectively, by potential predictors. RESULTS: The estimated 2015 annual expenditure for opioid prescriptions in Rhode Island was $44,271,827. The average and median costs of an opioid prescription were $67.98 (SD $210.91) and $21.08 (quartile 1 to quartile 3 = $7.65-$47.51), respectively. Prescriptions for branded opioid products accounted for $17,380,279.05, which was approximately 39.3% of overall spending, although only 6% of all opioids dispensed were for branded drugs. On average, patients aged 45-54 years and 55-64 years had overall adjusted spending for opioids that were 1.53 (95% CI = 1.49-1.57) and 1.75 (95% CI = 1.71-1.80) times higher than patients aged 65 years and older, respectively. Per patient Medicaid and Medicare average annual spending for opioid prescriptions were 1.19 (95% CI = 1.16-1.22) and 2.01 (95% CI = 1.96-2.06) times higher than commercial insurance spending, respectively. Annual opioid prescription spending was 2.01 (95% CI = 1.98-2.04) and 1.50 (95% CI = 1.45-1.55) times higher among patients who also had at least 1 dispensing of a benzodiazepine or sympathomimetic stimulant, respectively. Average total spending for prescription opioids per patient increased with the average daily dosage: from 3-fold for patients using 50-90 morphine milligrams equivalent (MME) daily to 22-fold for those receiving 90 or more MME daily compared with those receiving less than 50 MME daily. CONCLUSIONS: This study provides the first estimate of the statewide direct cost burden of prescription opioid use using PDMP data and standardized pricing benchmarks. Total annual cost increased with age up to 65 years, mean daily dose, and concurrent use of benzodiazepines or stimulants. Commercial insurance bore the majority of the cost of prescription opioid use, but cost per patient was highest among Medicare beneficiaries. In addition to reducing harms associated with opioid overuse and misuse, substantial cost savings could be realized by reducing unnecessary opioid use, especially among middle-aged adults. DISCLOSURES: This study was funded by the Rhode Island Department of Health. Aroke and Kogut report grants from the Rhode Island Department of Health during this study. Kogut is partially supported by Institutional Development Award Number U54GM115677 from the National Institute of General Medical Sciences of the National Institutes of Health, which funds Advance Clinical and Translational Research (Advance-CTR). Koziol reports grants from the Centers for Disease Control and Prevention during this study. The other authors have nothing to disclose. The content of this study is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Study concept and design were contributed by Koziol, Ragosta, and Kogut, along with Aroke. Koziol, Ragosta, Aroke, and Kogut collected the data, and data interpretation was performed by Aroke, Buchanan, Wen, and Kogut. The manuscript was primarily written by Aroke, along with Buchanan and Kogut, and revised by Aroke, Buchanan, Wen, and Kogut.