Misuse of prescribed opioids in the United States

Determinants of prescription opioid misuse among Black Americans: Evidence from the 2019 National Survey on Drug Use and Health

BACKGROUND

Opioid misuse, including prescription opioid misuse, remains a significant public health concern impacting various ethnoracial groups in the United States, including non-Hispanic Black Americans. This study provides more recent evidence on prescription opioid misuse among Black Americans.

METHODS

We used data from the 2019 National Survey on Drug Use and Health to examine the prevalence and determinants of prescription opioid misuse among Black American adults aged 18 and older. We compared these findings to non-Hispanic White American adults.

RESULTS

The prevalence rate of past-year prescription opioid misuse was very similar among Black (3.4%) and White respondents (3.8%). Adjusted multivariate logistic regression analyses found no significant racial differences in prescription opioid misuse. Religious importance and rurality were negatively associated with misuse only among Black respondents. Depressive episodes, other drug use, age, and risk-taking behaviors were associated with prescription opioid misuse among both Black and White respondents.

CONCLUSION

Black and White Americans remain at risk for prescription opioid-related problems. Religiosity and rurality require further investigation to understand how they may impact misuse among Black Americans.

Development of a machine learning algorithm for early detection of opioid use disorder

BACKGROUND

Opioid use disorder (OUD) affects an estimated 16 million people worldwide. The diagnosis of OUD is commonly delayed or missed altogether. We aimed to test the utility of machine learning in creating a prediction model and algorithm for early diagnosis of OUD.

SUBJECTS AND METHODS

We analyzed data gathered in a commercial claim database from January 1, 2006, to December 31, 2018 of 10 million medical insurance claims from 550 000 patient records. We compiled 436 predictor candidates, divided to six feature groups - demographics, chronic conditions, diagnosis and procedures features, medication features, medical costs, and episode counts. We employed the Word2Vec algorithm and the Gradient Boosting trees algorithm for the analysis.

RESULTS

The c-statistic for the model was 0.959, with a sensitivity of 0.85 and specificity of 0.882. Positive Predictive Value (PPV) was 0.362 and Negative Predictive Value (NPV) was 0.998. Significant differences between positive OUD- and negative OUD- controls were in the mean annual amount of opioid use days, number of overlaps in opioid prescriptions per year, mean annual opioid prescriptions, and annual benzodiazepine and muscle relaxant prescriptions. Notable differences were the count of intervertebral disc disorder-related complaints per year, post laminectomy syndrome diagnosed per year, and pain disorders diagnosis per year. Significant differences were also found in the episodes and costs categories.

CONCLUSIONS

The new algorithm offers a mean 14.4 months reduction in time to diagnosis of OUD, at potential saving in further morbidity, medical cost, addictions and mortality.

A comprehensive multivariate model of biopsychosocial factors associated with opioid misuse and use disorder in a 2017-2018 United States national survey

BACKGROUND

Few studies have comprehensively and contextually examined the relationship of variables associated with opioid use. Our purpose was to fill a critical gap in comprehensive risk models of opioid misuse and use disorder in the United States by identifying the most salient predictors.

METHODS

A multivariate logistic regression was used on the 2017 and 2018 National Survey on Drug Use and Health, which included all 50 states and the District of Columbia of the United States. The sample included all noninstitutionalized civilian adults aged 18 and older (N = 85,580; weighted N = 248,008,986). The outcome of opioid misuse and/or use disorder was based on reported prescription pain reliever and/or heroin use dependence, abuse, or misuse. Biopsychosocial predictors of opioid misuse and use disorder in addition to sociodemographic characteristics and other substance dependence or abuse were examined in our comprehensive model. Biopsychosocial characteristics included socioecological and health indicators. Criminality was the socioecological indicator. Health indicators included self-reported health, private health insurance, psychological distress, and suicidality. Sociodemographic variables included age, sex/gender, race/ethnicity, sexual identity, education, residence, income, and employment status. Substance dependence or abuse included both licit and illicit substances (i.e., nicotine, alcohol, marijuana, cocaine, inhalants, methamphetamine, tranquilizers, stimulants, sedatives).

RESULTS

The comprehensive model found that criminality (adjusted odds ratio [AOR] = 2.58, 95% confidence interval [CI] = 1.98-3.37, p < 0.001), self-reported health (i.e., excellent compared to fair/poor [AOR = 3.71, 95% CI = 2.19-6.29, p < 0.001], good [AOR = 3.43, 95% CI = 2.20-5.34, p < 0.001], and very good [AOR = 2.75, 95% CI = 1.90-3.98, p < 0.001]), no private health insurance (AOR = 2.12, 95% CI = 1.55-2.89, p < 0.001), serious psychological distress (AOR = 2.12, 95% CI = 1.55-2.89, p < 0.001), suicidality (AOR = 1.58, 95% CI = 1.17-2.14, p = 0.004), and other substance dependence or abuse were significant predictors of opioid misuse and/or use disorder. Substances associated were nicotine (AOR = 3.01, 95% CI = 2.30-3.93, p < 0.001), alcohol (AOR = 1.40, 95% CI = 1.02-1.92, p = 0.038), marijuana (AOR = 2.24, 95% CI = 1.40-3.58, p = 0.001), cocaine (AOR = 3.92, 95% CI = 2.14-7.17, p < 0.001), methamphetamine (AOR = 3.32, 95% CI = 1.96-5.64, p < 0.001), tranquilizers (AOR = 16.72, 95% CI = 9.75-28.65, p < 0.001), and stimulants (AOR = 2.45, 95% CI = 1.03-5.87, p = 0.044).

CONCLUSIONS

Biopsychosocial characteristics such as socioecological and health indicators, as well as other substance dependence or abuse were stronger predictors of opioid misuse and use disorder than sociodemographic characteristics.

National Patterns in Prescription Opioid Use and Misuse Among Cancer Survivors in the United States

IMPORTANCE

Prescription opioids are frequently prescribed to treat cancer-related pain. However, limited information exists regarding rates of prescription opioid use and misuse in populations with cancer.

OBJECTIVES

To estimate the prevalence and likelihood of prescription opioid use and misuse in adult cancer survivors compared with respondents without cancer and to identify characteristics associated with prescription opioid use and misuse in adult cancer survivors.

DESIGN, SETTING, AND PARTICIPANTS

This cross-sectional study is a retrospective, population-based study using data from 169 162 respondents to the National Survey on Drug Use and Health from January 2015 to December 2018. Survey data sets were queried for all respondents aged 18 years or older. Those with a reported history of cancer were termed cancer survivors and further divided into more recent (had cancer within 12 months of survey) and less recent (had cancer more than 12 months prior to survey) cohorts. Respondents with nonmelanoma skin cancer were excluded.

MAIN OUTCOMES AND MEASURES

Prescription opioid use and misuse within the past 12 months.

RESULTS

Among 169 162 respondents, 5139 (5.2%) were cancer survivors, with 1243 (1.2%) and 3896 (4.0%) reporting having more recent and less recent cancer histories, respectively. Higher rates of prescription opioid use were observed among more recent cancer survivors (54.3%; 95% CI, 50.2%-58.4%; odds ratio [OR], 1.86; 95% CI, 1.57-2.20; P < .001) and less recent cancer survivors (39.2%; 95% CI, 37.3%-41.2%; OR, 1.18; 95% CI, 1.08-1.28; P < .001) compared with respondents without cancer (30.5%, reference group). Rates of prescription opioid misuse were similar among more recent (3.5%; 95% CI, 2.4%-5.2%; OR, 1.27; 95% CI, 0.82-1.96; P = .36) and less recent (3.0%; 95% CI, 2.4%-3.6%; OR, 1.03; 95% CI, 0.83-1.28; P = .76) survivors compared with respondents without cancer (4.3%, reference group). Younger age (aged 18-34 years vs ≥65 years: OR, 7.06; 95% CI, 3.03-16.41; P < .001), alcohol use disorder (OR, 3.22; 95% CI, 1.45-7.14; P = .005), and nonopioid drug use disorder (OR, 14.76; 95% CI, 7.40-29.44; P < .001) were associated with prescription opioid misuse among cancer survivors.

CONCLUSIONS AND RELEVANCE

In this study, prescription opioid use was higher among more and less recent cancer survivors compared with the population without a history of cancer. Rates of prescription opioid misuse were low and similar among all 3 cohorts. These findings suggest that higher prescription opioid use among cancer survivors may not correspond to increased short-term or long-term misuse.

Associations of Opioid Prescriptions with Death and Hospitalization across the Spectrum of Estimated GFR

BACKGROUND AND OBJECTIVES

Most opioids undergo kidney excretion. The goal of this study was to evaluate opioid-associated risks of death and hospitalization across the range of eGFR.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS

The study population included adult primary care patients in Geisinger Health (Danville, PA) between 2008 and 2017. People receiving their first opioid prescription were propensity matched to people receiving NSAIDS (and, in sensitivity analysis, gabapentinoids) and the risk of death and hospitalization were compared, classifying opioid medication exposure as time-varying daily oral morphine milligram equivalents (MMEs) across time-varying eGFR.

RESULTS

The propensity-matched cohort included 46,246 patients prescribed either opioids or NSAIDs between 2008 and 2017 (mean [SD] age, 54 [16] years; 56% female; 3% of black race). Prescriptions for 1-59 and ≥60 MMEs were associated with higher risk of death (HR, 1.70; 95% CI, 1.41 to 2.05 for 1-59 MMEs; HR, 2.25; 95% CI, 1.82 to 2.79 for ≥60 MMEs) and hospitalization (HR, 1.38; 95% CI, 1.30 to 1.46 for 1-59 MMEs; HR, 1.68; 95% CI, 1.56 to 1.81 for ≥60 MMEs) compared with NSAID prescriptions, when evaluated at eGFR 80 ml/min per 1.73 m². The relative risk of death associated with ≥60 MMEs was higher at lower GFR (e.g., eGFR, 40 ml/min per 1.73 m²; HR, 3.94; 95% CI, 2.70 to 5.75; P for interaction, 0.01). When gabapentinoids were used as the comparison medication, only ≥60 MMEs were significantly associated with higher risk of death (HR, 2.72; 95% CI, 1.71 to 4.34), although both 1-59 and ≥60 MMEs were associated with risk of hospitalization (HR, 1.22; 95% CI, 1.04 to 1.43 for 1-59 MMEs; HR, 1.54; 95% CI, 1.28 to 1.86 for ≥60 MMEs).

CONCLUSIONS

The receipt of prescription opioids was associated with a higher risk of death and hospitalization compared with other pain medications, particularly with higher doses and at lower eGFR.

Information on doctor and pharmacy shopping for opioids adds little to the identification of presumptive opioid abuse disorders in health insurance claims data

Background

Doctor and pharmacy shopping (“Shopping”) for opioids is related to opioid abuse and is associated with opioid overdose and death. Lacking identifiers for prescribers and pharmacies, many data resources (notably the US FDA’s Sentinel System) cannot evaluate Shopping. We used data in which presumptive Shopping could be identified. We investigated whether US health insurance claims data could perform as well as Shopping to identify people with evidence for opioid abuse.

Methods

In this cross-sectional study, we examined health insurance claims from 164,923 persons with at least two dispensing of opioids in 18 months, the first occurring in 2012. Evidence for the presence of a possible opioid abuse disorder was drawn from predictive patterns of drug fills, diagnoses and care-seeking identified in a companion research project, and Shopping was determined using a published index. The prevalence of presumptive opioid abuse was examined across levels of Shopping. The comparison between Shopping and insurance-claims-derived covariates in the detection of apparent opioid abuse was examined in multiple regression analyses.

Results

Despite a strong correlation between presumptive opioid abuse and Shopping, most persons with extensive Shopping did not manifest presumptive opioid abuse, and half of the population with presumptive opioid abuse did not exhibit Shopping. As Shopping ranged from “None” to “Extensive,” the prevalence of presumptive opioid abuse increased from 0.28 to 5.0 per 100. The discriminating power of Shopping for identifying opioid abuse could be replaced using insurance claims data.

Conclusion

The results suggest that patient characteristics that can be inferred from insurance claims data provide as complete discrimination of persons with presumptive opioid abuse as does a full assessment of doctor and pharmacy shopping. The inference rests on patterns of health services and drug dispensing that are indicative of doctor–pharmacy shopping and of opioid abuse. There was no direct evaluation of patients. The extent to which the conclusions are generalizable beyond the study population – Americans with health insurance coverage in the early part of this decade – is uncertain in a quantitative sense. The qualitative conclusion is that diagnostic data in health insurance databases can be predictive of behaviors consistent with opioid abuse and that more elaborate indices such as doctor and pharmacy shopping may add little.

Registration: ClinicalTrials.gov study number: NCT02668549.