Protocol for Designing a Model to Predict the Likelihood of Psychosis From Electronic Health Records Using Natural Language Processing and Machine Learning.

INTRODUCTION: Rapid identification of individuals developing a psychotic spectrum disorder (PSD) is crucial because untreated psychosis is associated with poor outcomes and decreased treatment response. Lack of recognition of early psychotic symptoms often delays diagnosis, further worsening these outcomes. METHODS: The proposed study is a cross-sectional, retrospective analysis of electronic health record data including clinician documentation and patient-clinician secure messages for patients aged 15-29 years with ≥ 1 primary care encounter between 2017 and 2019 within 2 Kaiser Permanente regions. Patients with new-onset PSD will be distinguished from those without a diagnosis if they have ≥ 1 PSD diagnosis within 12 months following the primary care encounter. The prediction model will be trained using a trisourced natural language processing feature extraction design and validated both within each region separately and in a modified combined sample. DISCUSSION: This proposed model leverages the strengths of the large volume of patient-specific data from an integrated electronic health record with natural language processing to identify patients at elevated chance of developing a PSD. This project carries the potential to reduce the duration of untreated psychosis and thereby improve long-term patient outcomes.

Racial and ethnic disparities in receipt of specialty treatment across risk profiles of adults with heavy alcohol use.

BACKGROUND: Variation in specialty treatment utilization for alcohol use disorder (AUD) by patient subgroups is poorly understood. This study examined whether and how patient risk profiles predict receipt of specialty treatment and whether there are disparities by race and ethnicity. METHODS: This cohort study included 206,956 adults with heavy alcohol use (that which exceeded National Institute on Alcohol Abuse and Alcoholism guidelines) between June 1, 2013 and December 31, 2014, using electronic health record data from Kaiser Permanente Northern California. Five risk profiles (characterized by daily or weekly heavy drinking and level of health risks) were identified in latent class analysis. Logistic regression models were fit to examine associations between risk profiles, race, ethnicity, and receipt of specialty treatment (including addiction medicine, psychiatry, or integrated behavioral health visits, and AUD pharmacotherapy), adjusting for other patient characteristics. Variation in the association between risk profiles and receipt of specialty treatment by race/ethnicity was also examined. RESULTS: Overall, 4.0% of patients received specialty treatment. Latino/Hispanic and Asian/Pacific Islander patients had lower odds of receiving specialty treatment than White patients (adjusted odds ratio [aOR] [95% CI] = 0.80 [0.75, 0.85], and 0.64 [0.59, 0.70], respectively). The substance use disorder and mental health disorder (SUD/MH) risk profile had the highest odds of receiving specialty treatment (10.46 [9.65, 11.34]). Associations between risk profiles and receipt of specialty treatment significantly differed by race/ethnicity. Black patients in the SUD/MH risk profile, and Hispanic/Latino patients in the risk profile with heavy daily drinking and more health risks, had lower odds of receiving specialty treatment than their White counterparts (adjusted ratio of odds ratios [aROR] [95% CI] = 0.69 [0.50, 0.94], and 0.79 [0.67, 0.92], respectively). CONCLUSIONS: This study provides new insights into racial/ethnic disparities in specialty treatment utilization for alcohol problems. Findings may help inform strategies for tailoring interventions to address heavy alcohol use.

Unhealthy Alcohol Use among Adults with Depression or Anxiety: Changes During COVID-19 and Associations with Mental Health Treatment.

OBJECTIVE: Individuals with unhealthy alcohol use and comorbid depression or anxiety may be vulnerable to alcohol use escalation in times of stress such as the COVID-19 pandemic. Among a cohort of individuals with pre-pandemic unhealthy drinking, we compared changes in alcohol use by whether people had a depression or anxiety diagnosis, and examined whether mental health treatment was related to these changes. METHODS: Using electronic health record data from Kaiser Permanente Northern California, we analyzed drinking changes during the pandemic (3/1/2020-6/30/2022) among adults identified in primary care with unhealthy alcohol use (exceeding daily/weekly recommended limits) pre-pandemic (1/1/2019-2/29/2020). Outcomes were mean changes in number of heavy drinking days (prior three months), drinks/week, drinks/day, and drinking days/week. Multivariable linear regression models were fit to: 1) compare outcomes of patients with depression or anxiety diagnoses to those without, and 2) among patients with depression or anxiety, estimate associations between mental health treatment and outcomes. RESULTS: The sample included 62,924 adults with unhealthy alcohol use, of whom 12,281 (19.5%) had depression or anxiety. On average, alcohol use significantly decreased across all measures during the pandemic, but patients with depression or anxiety had greater decreases in drinks/week (adjusted mean difference [aMD] [CI]=-0.34 [-0.55, -0.12]) and drinking days/week (-0.15 [-0.20, -0.10]). No associations were found between mental health treatment and changes in drinking. CONCLUSIONS: Contrary to expectations, patients with unhealthy alcohol use and depression or anxiety decreased alcohol use more than those without depression or anxiety during COVID-19, whether or not they accessed mental health services.

Adolescent utilization of eating disorder higher level of care: roles of family-based treatment adherence and demographic factors.

BACKGROUND: Outpatient family-based treatment (FBT) is effective in treating restrictive eating disorders among adolescents. However, little is known about whether FBT reduces higher level of care (HLOC) utilization or if utilization of HLOC is associated with patient characteristics. This study examined associations between utilization of eating disorder related care (HLOC and outpatient treatment) and reported adherence to FBT and patient characteristics in a large integrated health system. METHODS: This retrospective cohort study examined 4101 adolescents who received care for restrictive eating disorders at Kaiser Permanente Northern California. A survey was sent to each medical center to identify treatment teams as high FBT adherence (hFBT) and low FBT adherence (lFBT). Outpatient medical and psychiatry encounters and HLOC, including medical hospitalizations and higher-level psychiatric care as well as patient characteristics were extracted from the EHR and examined over 12 months post-index. RESULTS: 2111 and 1990 adolescents were treated in the hFBT and lFBT, respectively. After adjusting for age, sex, race/ethnicity, initial percent median BMI, and comorbid mental health diagnoses, there were no differences in HLOC or outpatient utilization between hFBT and lFBT. Females had higher odds of any utilization compared with males. Compared to White adolescents, Latinos/Hispanics had lower odds of HLOC utilization. Asian, Black, and Latino/Hispanic adolescents had lower odds of psychiatric outpatient care than Whites. CONCLUSIONS: Reported FBT adherence was not associated with HLOC utilization in this sample. However, significant disparities across patient characteristics were found in the utilization of psychiatric care for eating disorders. More efforts are needed to understand treatment pathways that are accessible and effective for all populations with eating disorders.

Adolescents with restrictive eating treated by Family-Based Treatment (FBT) teams had better early weight gain but no differences in the use of intensive outpatient, residential, partial hospital programs or inpatient psychiatry care when compared to those treated by teams with a low adherence to the FBT approach. Factors such as sex, race, ethnicity, mood disorders, and suicidality were associated with the use of psychiatric services. These findings are consistent with previously documented systematic disparities in accessing psychiatric services across patient demographics and should be used to inform the development of proposed care models that are more inclusive and accessible to all patients.

A patient activation intervention in primary care for patients with chronic pain on long term opioid therapy: results from a randomized control trial.

BACKGROUND: Given significant risks associated with long-term prescription opioid use, there is a need for non-pharmacological interventions for treating chronic pain. Activating patients to manage chronic pain has the potential to improve health outcomes. The ACTIVATE study was designed to evaluate the effectiveness of a 4-session patient activation intervention in primary care for patients on long-term opioid therapy. METHODS: The two-arm, pragmatic, randomized trial was conducted in two primary care clinics in an integrated health system from June 2015-August 2018. Consenting participants were randomized to the intervention (n = 189) or usual care (n = 187). Participants completed online and interviewer-administered surveys at baseline, 6- and 12- months follow-up. Prescription opioid use was extracted from the EHR. The primary outcome was patient activation assessed by the Patient Activation Measure (PAM). Secondary outcomes included mood, function, overall health, non-pharmacologic pain management strategies, and patient portal use. We conducted a repeated measure analysis and reported between-group differences at 12 months. RESULTS: At 12 months, the intervention and usual care arms had similar PAM scores. However, compared to usual care at 12 months, the intervention arm demonstrated: less moderate/severe depression (odds ratio [OR] = 0.40, 95%CI 0.18-0.87); higher overall health (OR = 3.14, 95%CI 1.64-6.01); greater use of the patient portal's health/wellness resources (OR = 2.50, 95%CI 1.42-4.40) and lab/immunization history (OR = 2.70, 95%CI 1.29-5.65); and greater use of meditation (OR = 2.72; 95%CI 1.61-4.58) and exercise/physical therapy (OR = 2.24, 95%CI 1.29-3.88). At 12 months, the intervention arm had a higher physical health measure (mean difference 1.63; 95%CI: 0.27-2.98). CONCLUSION: This trial evaluated the effectiveness of a primary care intervention in improving patient activation and patient-reported outcomes among adults with chronic pain on long-term opioid therapy. Despite a lack of improvement in patient activation, a brief intervention in primary care can improve outcomes such as depression, overall health, non-pharmacologic pain management, and engagement with the health system. TRIAL REGISTRATION: The study was registered on 10/27/14 on ClinicalTrials.gov (NCT02290223).

Risk profiles of adults with heavy alcohol use: Drinking patterns, behavioral and metabolic factors, health problems, and racial and ethnic disparities.

BACKGROUND: Heavy alcohol use is a growing risk factor for chronic disease, yet little is known about its co-occurrence with other risk factors and health problems. This study aimed to identify risk profiles of adults with heavy alcohol use and examined potential disparities by race and ethnicity. METHODS: This cross-sectional study included 211,333 adults with heavy alcohol use (in excess of daily or weekly limits recommended by National Institute on Alcohol Abuse and Alcoholism) between June 1, 2013 and December 31, 2014 in Kaiser Permanente Northern California. Latent class analysis was used to examine how heavy drinking patterns clustered with other behavioral and metabolic risk factors and health problems to form risk profiles. Multinomial logistic regression models were fit to examine associations between race, ethnicity, and risk profiles. RESULTS: A 5-class model was selected as best fitting the data and representing clinically meaningful risk profiles: (1) "heavy daily drinking and lower health risks" (DAILY, 44.3%); (2) "substance use disorder and mental health disorder" (SUD/MH, 2.3%); (3) "heavy weekly drinking and lower health risks" (WEEKLY, 19.6%); (4) "heavy daily drinking and more health risks" (DAILY-R, 18.5%); (5) "heavy weekly drinking and more health risks" (WEEKLY-R, 15.3%). American Indian or Alaska Native (AIAN) and Black patients had higher odds than White patients of being in the SUD/MH, DAILY-R, and WEEKLY-R profiles than the DAILY profile. AIAN, Black, and Latino/Hispanic patients had higher odds than White patients of being in the SUD/MH, DAILY-R, and WEEKLY-R profiles rather than the WEEKLY profile. CONCLUSIONS: AIAN, Black, and Latino/Hispanic patients with self-reported heavy drinking were more likely to be in risk profiles with greater alcohol consumption, more health risks, and higher morbidity. Targeted, culturally appropriate interventions for heavy alcohol use that may address other modifiable risk factors are needed to work towards health equity.

Associations between alcohol brief intervention in primary care and drinking and health outcomes in adults with hypertension and type 2 diabetes: a population-based observational study.

OBJECTIVES: To evaluate associations between alcohol brief intervention (BI) in primary care and 12-month drinking outcomes and 18-month health outcomes among adults with hypertension and type 2 diabetes (T2D). DESIGN: A population-based observational study using electronic health records data. SETTING: An integrated healthcare system that implemented system-wide alcohol screening, BI and referral to treatment in adult primary care. PARTICIPANTS: Adult primary care patients with hypertension (N=72 979) or T2D (N=19 642) who screened positive for unhealthy alcohol use between 2014 and 2017. MAIN OUTCOME MEASURES: We examined four drinking outcomes: changes in heavy drinking days/past 3 months, drinking days/week, drinks/drinking day and drinks/week from baseline to 12-month follow-up, based on results of alcohol screens conducted in routine care. Health outcome measures were changes in measured systolic and diastolic blood pressure (BP) and BP reduction ≥3 mm Hg at 18-month follow-up. For patients with T2D, we also examined change in glycohaemoglobin (HbA1c) level and 'controlled HbA1c' (HbA1c<8%) at 18-month follow-up. RESULTS: For patients with hypertension, those who received BI had a modest but significant additional -0.06 reduction in drinks/drinking day (95% CI -0.11 to -0.01) and additional -0.30 reduction in drinks/week (95% CI -0.59 to -0.01) at 12 months, compared with those who did not. Patients with hypertension who received BI also had higher odds for having clinically meaningful reduction of diastolic BP at 18 months (OR 1.05, 95% CI 1.00 to 1.09). Among patients with T2D, no significant associations were found between BI and drinking or health outcomes examined. CONCLUSIONS: Alcohol BI holds promise for reducing drinking and helping to improve health outcomes among patients with hypertension who screened positive for unhealthy drinking. However, similar associations were not observed among patients with T2D. More research is needed to understand the heterogeneity across diverse subpopulations and to study BI's long-term public health impact.

Young Adult Substance Use and Healthcare Use Associated With Screening, Brief Intervention and Referral to Treatment in Pediatric Primary Care.

PURPOSE: Screening, brief intervention, and referral to treatment (SBIRT) may impact future comorbidity and healthcare utilization among adolescents screening positive for substance use or mood problems. METHODS: In a randomized trial sample, we compared an SBIRT group to usual care for substance use, mental health, medical diagnoses, and healthcare utilization over 7 years postscreening. RESULTS: In logistic regression models adjusting for patient characteristics, the SBIRT group had lower odds of any substance (Odds Ratio[OR] = 0.80, 95% Confidence Interval [CI] = 0.66-.98), alcohol (OR = 0.69, 95% CI = 0.51-0.94), any drug (OR = 0.73, 95% CI = 0.54-0.98), marijuana (OR = 0.70, 95% CI = 0.50-0.98), and tobacco (OR = 0.83, 95% CI = 0.69-1.00) diagnoses, and lower odds of any inpatient hospitalizations (OR = 0.59, 95% CI = 0.41-0.85) compared with usual care. Negative binomial models examining number of visits among adolescents with at least one visit of that type found that those in the SBIRT group had fewer primary care (incidence rate ratio[iRR] = 0.90, p < .05) and psychiatry (iRR = 0.64, p < .01) and more addiction medicine (iRR = 1.52, p < .01) visits over 7 years compared with usual care. In posthoc analyses, we found that among Hispanic patients, those in the SBIRT group had lower odds of any substance, any drug and marijuana use disorder diagnoses compared with usual care, and among Black/African American patients, those in the SBIRT group had lower odds of alcohol use disorder diagnoses compared with usual care. DISCUSSION: Beneficial effects of adolescent SBIRT on substance use and healthcare utilization may persist into young adulthood.

Predicting severe alcohol use disorders in primary care using number of heavy drinking days.

BACKGROUND AND AIMS: Although screening for unhealthy alcohol use is becoming more common, severe alcohol use disorders (AUDs) associated with the most severe medical and socio-economic sequelae still often go unidentified in primary care. To improve identification of severe AUDs and aid clinical decision-making, we aimed to identify a threshold of heavy drinking days (HDDs) associated with severe AUDs. DESIGN, SETTING AND CASES: This cohort study analyzed electronic health record data of 138 765 adults who reported ≥ 1 HDD (4+ drinks/occasion for women and men aged ≥ 65 years, 5+ for men aged 18-64 years) during a 3-month period at a routine alcohol screening in primary care in a large Northern California, USA health-care system from 2014 to 2017. Our sample was 66.5% male, 59.7% white, 11.0% Asian/Pacific Islander, 5.0% black, 17.4% Latino/Hispanic and 7.0% other/unknown race/ethnicity; the mean age was 40.6 years (standard deviation = 15.2). MEASUREMENTS: We compared sensitivity and specificity of different thresholds of the reported number of HDDs during a 3-month period for predicting severe AUD diagnoses in the following year, in the full sample and by sex and age. FINDINGS: The prevalence of severe AUD diagnoses in the year after the screening was 0.6%. The optimal threshold predicting future severe AUD diagnoses in the full sample was ≥ 5 HDDs during a 3-month period [sensitivity = 68.9%, 95% confidence interval (CI) = 65.9, 72.0; specificity = 63.2%, 95% CI = 62.9, 63.4], but varied by sex and age. Women had a lower threshold than men (4 versus 6 HDDs), which decreased as women aged (from 5 HDDs among 18-24 years to 4 HDDs ≥ 25 years), but increased as men aged (from 5 HDDs among 18-24 years to 6 HDDs among 25-64 years, to 7 HDDs ≥ 65 years). CONCLUSIONS: Five or more heavy drinking days in a 3-month period may indicate heightened risk of future severe alcohol use disorder in an adult primary care population. The optimal thresholds are lower for women than for men, and thresholds decrease as women age but increase as men age.

Alcohol brief intervention, specialty treatment and drinking outcomes at 12 months: Results from a systematic alcohol screening and brief intervention initiative in adult primary care.

BACKGROUND: Alcohol screening, brief intervention and referral to treatment (SBIRT) in adult primary care is an evidence-based, public health strategy to address unhealthy alcohol use, but evidence of effectiveness of alcohol brief intervention (ABI) in real-world implementation is lacking. METHODS: We fit marginal structural models with inverse probability weighting to estimate the causal effects of ABI on 12-month drinking outcomes using longitudinal electronic health records data for 312,056 adults with a positive screening result for unhealthy drinking between 2014 and 2017 in a large healthcare system that implemented systematic primary care-based SBIRT. We examined effects of ABI with and without adjusting for receipt of specialty alcohol use disorder (AUD) treatment, and whether effects varied by patient demographic characteristics and alcohol use patterns. RESULTS: Receiving ABI resulted in significantly greater reductions in heavy drinking days (mean difference [95% CI] = -0.26 [-0.45, -0.08]), drinking days per week (-0.04 [-0.07, -0.01]), drinks per drinking day (-0.05 [-0.08, -0.02]) and drinks per week (-0.16 [-0.27, -0.04]). Effects of ABI on 12-month drinking outcomes varied by baseline consumption level, age group and whether patients already had an AUD, with better improvement in those who were drinking at levels exceeding only daily limits, younger, and without an AUD. CONCLUSIONS: Systematic ABI in adult primary care has the potential to reduce drinking among people with unhealthy drinking considerably on both an individual and population level. More research is needed to help optimize ABI, in particular tailoring it to diverse sub-populations, and studying its long-term public health impact.

Predictors of early and sustained cessation of heavy drinking over 5 years among adult primary care patients.

AIMS: To identify factors asociated with early and sustained cessation of heavy drinking. DESIGN: Retrospective cohort study over 5 years. SETTING: Kaiser Permanente Northern California, United States. PARTICIPANTS: Adults reporting heavy drinking during primary care-based alcohol screening between 1 June 2013 and 31 May 2014. The sample (n = 85 434) was 40.7% female and 33.8% non-white; mean age was 50.3 years (standard deviation = 18.1). MEASUREMENTS: Following US guidelines, early and sustained cessation of heavy drinking was defined as reporting lower-risk drinking or abstinence at 1 year and to 5 years after achieving early cessation, respectively. Associations between patient characteristics and service use and cessation outcomes were examined using logistic regression with inverse probability weights addressing attrition. FINDINGS: Nearly two-thirds of participants achieved early cessation of heavy drinking. Women [odds ratio (OR) = 1.39, 95% confidence interval (CI) = 1.35, 1.44], middle-age (35-64 years: ORs = 1.16-1.19), non-white race/ethnicity (ORs = 1.03-1.57), medical conditions (OR = 1.05, 95% CI = 1.04, 1.06), psychiatric (OR = 1.10, 95% CI = 1.06, 1.15) and drug use disorders (OR = 1.35, 95% CI = 1.17, 1.56) and addiction treatment (OR = 1.19, 95% CI = 1.09, 1.30) were associated with higher odds of early cessation, while older age (≥ 65 years: OR = 0.91, 95% CI = 0.86, 0.96), smoking (OR = 0.81, 95% CI = 0.77, 0.84), higher index drinking levels (exceeding both daily and weekly limits: OR = 0.30, 95% CI = 0.29, 0.32) and psychiatric treatment (OR = 0.91, 95% CI = 0.84, 0.99) were associated with lower odds. Among those who achieved early cessation (n = 19 200), 60.0% sustained cessation. Associations between patient factors and sustained cessation paralleled those observed in analyses of early cessation. Additionally, routine primary care (OR = 1.57, 95% CI = 1.44, 1.71) and addiction treatment post-1 year (OR = 1.41, 95% CI = 1.19, 1.66) were associated with higher odds of sustained cessation. Lower-risk drinking versus abstinence at 1 year was associated with lower odds of sustained cessation (OR = 0.62, 95% CI = 0.57, 0.66). CONCLUSIONS: Nearly two-thirds of a large, diverse sample of patients who reported heavy drinking in a Californian health-care system achieved early and sustained cessation of heavy drinking. Vulnerable subgroups (i.e. non-white patients and those with psychiatric disorders), patients who received routine primary care and those who received addiction treatment were more likely to sustain cessation of heavy drinking than other participants.

Patient and provider factors associated with receipt and delivery of brief interventions for unhealthy alcohol use in primary care.

BACKGROUND: Unhealthy alcohol use is a serious and costly public health problem. Alcohol screening and brief interventions are effective in reducing unhealthy alcohol consumption. However, rates of receipt and delivery of brief interventions vary significantly across healthcare settings, and relatively little is known about the associated patient and provider factors. METHODS: This study examines patient and provider factors associated with the receipt of brief interventions for unhealthy alcohol use in an integrated healthcare system, based on documented brief interventions in the electronic health record. Using multilevel logistic regression models, we retrospectively analyzed 287,551 adult primary care patients (and their 2952 providers) who screened positive for unhealthy drinking between 2014 and 2017. RESULTS: We found lower odds of receiving a brief intervention among patients exceeding daily or weekly drinking limits (vs. exceeding both limits), females, older age groups, those with higher medical complexity, and those already diagnosed with alcohol use disorders. Patients with other unhealthy lifestyle activities (e.g., smoking, no/insufficient exercise) were more likely to receive a brief intervention. We also found that female providers and those with longer tenure in the health system were more likely to deliver brief interventions. CONCLUSIONS: These findings point to characteristics that can be targeted to improve universal receipt of brief intervention.

Effectiveness of spironolactone dispensation in reducing weekly alcohol use: a retrospective high-dimensional propensity score-matched cohort study.

There is a need to increase the armamentarium of pharmacotherapies for alcohol use disorder (AUD). Recent research suggests that mineralocorticoid receptor (MR) antagonism via spironolactone may represent a novel pharmacological treatment for AUD. We conducted a pharmacoepidemiologic retrospective cohort study (June 1, 2014 to May 31, 2018) to examine whether spironolactone dispensation (≥90 continuous days), for any indication, is associated with changes in weekly alcohol use about 6 months later. We compared 523 spironolactone-treated adults and 2305 untreated adults, matched on high-dimensional propensity scores created from a set of predefined (sociodemographic and health characteristics, diagnoses, and service utilization) and empirical electronic health record-derived covariates. The sample was 57% female and 27% non-White with a mean age of 59.2 years (SD = 19.3). Treated patients reduced their weekly alcohol use by 3.50 drinks (95% CI = -4.22, -2.79), while untreated patients reduced by 2.74 drinks (95% CI = -3.22, -2.26), yielding a significant difference of 0.76 fewer drinks (95% CI = -1.43, -0.11). Among those who drank >7 drinks/week at baseline, treated patients, compared to untreated patients, reported a greater reduction in weekly alcohol use by 4.18 drinks (95% CI = -5.38, -2.97), while there was no significant difference among those who drank less. There was a significant dose-response relationship between spironolactone dosage and change in drinks/week. Pending additional evidence on its safety and efficacy in individuals with AUD, spironolactone (and MR blockade, at large) may hold promise as a pharmacotherapy for AUD.

Three-Year Outcomes After Brief Treatment of Substance Use and Mood Symptoms.

BACKGROUND: Screening, brief intervention, and referral to treatment (SBIRT) for adolescents exhibiting co-occurring substance use and mental health problems may improve outcomes and have long-lasting effects. This study examined the relationship between access to SBIRT and substance use, depression and medical diagnoses, and health services use at 1 and 3 years postscreening for such adolescents. METHODS: The study draws from a cluster-randomized trial comparing SBIRT to usual care (UC) for adolescents endorsing past-year substance use and recent mood symptoms during visits to a general pediatrics clinic between November 1, 2011, and October 31, 2013, in a large, integrated health system (N = 1851); this sample examined the subset of adolescents endorsing both problems (n = 289). Outcomes included depression, substance use and medical diagnoses, and emergency department and outpatient visits 1 and 3 years later. RESULTS: The SBIRT group had lower odds of depression diagnoses at 1 (odds ratio [OR] = 0.31; confidence interval [CI] = 0.11-0.87) and 3 years (OR = 0.51; CI = 0.28-0.94) compared with the UC group. At 3 years, the SBIRT group had lower odds of a substance use diagnosis (OR = 0.46; CI = 0.23-0.92), and fewer emergency department visits (rate ratio = 0.65; CI = 0.44-0.97) than UC group. CONCLUSIONS: The findings suggest that SBIRT may prevent health complications and avert costly services use among adolescents with both mental health and substance use problems. As SBIRT is implemented widely in pediatric primary care, training pediatricians to discuss substance use and mental health problems can translate to positive outcomes for these vulnerable adolescents.

Premature mortality associated with severe mental illness or substance use disorder in an integrated health care system.

OBJECTIVE: Research has reported shortened lifespans (by 15-30 years) for those with severe mental illness (SMI) or substance use disorder (SUD), particularly among public mental health treatment consumers. We assessed SMI- and SUD-associated mortality in the understudied setting of a large, nonprofit integrated health care system. METHOD: This retrospective cohort study examined 2010-2017 health system and death records for 564,592 adult patients. Half had SMI/SUD diagnosis; half were a demographically matched comparison group without SMI, other mental health, or SUD diagnoses. We estimated mortality risks, adjusting for demographic and physical health factors. RESULTS: Having SMI or SUD was associated with higher odds of death (adjusted odds ratio = 1.87) and an average 6.3 years of earlier death relative to comparison individuals. Co-occurring SMI and SUD conferred higher mortality risk from major natural and unnatural causes than did SMI with no SUD. CONCLUSIONS: Some indicators of premature mortality were lower than those reported for U.S. public mental health consumers, but risk level varied widely by diagnosis. While patients' having insurance and broad access to care may lower risk, access to care may be insufficient to overcome the many patient-, provider-, and system-level factors contributing to poor physical health in SMI and SUD.

Associations Between Psychiatric Disorders and Alcohol Consumption Levels in an Adult Primary Care Population.

BACKGROUND: Unhealthy alcohol use frequently co-occurs with psychiatric disorders; however, little is known about the relationship between psychiatric disorders and alcohol consumption levels. Understanding varying levels of unhealthy alcohol use among individuals with a variety of psychiatric disorders in primary care would provide valuable insight for tailoring interventions. METHODS: We conducted a cross-sectional study of 2,720,231 adult primary care patients screened for unhealthy alcohol use between 2014 and 2017 at Kaiser Permanente Northern California, using electronic health record data. Alcohol consumption level was classified as no reported use, low-risk use, and unhealthy use, per National Institute on Alcohol Abuse and Alcoholism guidelines. Unhealthy use was further differentiated into mutually exclusive groups: exceeding only daily limits, exceeding only weekly limits, or exceeding both daily and weekly limits. Multivariable multinomial logistic regression models were fit to examine associations between 8 past-year psychiatric disorders (depression, bipolar disorder, anxiety disorder, obsessive-compulsive disorder, schizophrenia, schizoaffective disorder, anorexia nervosa, and bulimia nervosa) and alcohol consumption levels, adjusting for sociodemographic and health characteristics. RESULTS: In the full sample [53% female, 48% White, mean (SD) age = 46 (18) years], patients with psychiatric disorders (except eating disorders), compared to those without, had lower odds of reporting low-risk and unhealthy alcohol use relative to no use. Among patients who reported any alcohol use (n = 861,427), patients with depression and anxiety disorder, compared to those without, had higher odds of exceeding only weekly limits and both limits; patients with bulimia nervosa were also more likely to exceed both limits. CONCLUSIONS: Findings suggest that patients with anxiety disorder, depression, and bulimia nervosa who drink alcohol are more likely to exceed recommended limits, increasing risk of developing more serious problems. Health systems and clinicians may wish to consider implementing more robust screening, assessment, and intervention approaches to support these vulnerable subgroups in limiting their drinking.

Remission From Unhealthy Drinking Among Patients With an Alcohol Use Disorder: A Longitudinal Study Using Systematic, Primary Care-Based Alcohol Screening Data.

OBJECTIVE: Using electronic health record (EHR) data from a systematic, primary care-based alcohol screening, brief intervention, and referral to treatment (SBIRT) initiative within a health system, we examined correlates of remission from unhealthy drinking among patients with an alcohol use disorder (AUD). METHOD: We conducted a longitudinal study of 4,078 adults with AUD who screened positive for unhealthy drinking between October 1, 2015, and September 30, 2016. We extracted EHR data up to 3 years after screening until October 1, 2018. We used survival analysis to examine associations between remission (i.e., reporting abstinence or low-risk drinking at a subsequent screening) and patient characteristics, comorbidities, and treatment utilization. RESULTS: The median time to remission from unhealthy drinking was 1.7 years. Factors significantly associated with greater odds of remitting from unhealthy drinking during follow-up were female gender; older age (50-64 years); Black or Latino/Hispanic race/ethnicity; having more medical comorbidities; not having a comorbid drug use disorder; lower alcohol consumption levels; and receiving addiction medicine treatment before the index screening. In the first follow-up year, individuals with mental health comorbidities were more likely to remit, but those in psychiatric treatment were less likely. Receiving addiction treatment during follow-up was not associated with remission. CONCLUSIONS: Ethnic minorities and individuals with mental illness were more likely to remit, which is encouraging given the health disparities observed among these clinically important subgroups and warrants further research. Our findings may inform research on AUD recovery and clinical practice, as remission from unhealthy drinking is a crucial component of the early stages of recovery.

The Kaiser Permanente Northern California Adult Alcohol Registry, an Electronic Health Records-Based Registry of Patients With Alcohol Problems: Development and Implementation.

BACKGROUND: Electronic health record (EHR)-based disease registries have aided health care professionals and researchers in increasing their understanding of chronic illnesses, including identifying patients with (or at risk of developing) conditions and tracking treatment progress and recovery. Despite excessive alcohol use being a major contributor to the global burden of disease and disability, no registries of alcohol problems exist. EHR-based data in Kaiser Permanente Northern California (KPNC), an integrated health system that conducts systematic alcohol screening, which provides specialty addiction medicine treatment internally and has a membership of over 4 million members that are highly representative of the US population with access to care, provide a unique opportunity to develop such a registry. OBJECTIVE: Our objectives were to describe the development and implementation of a protocol for assembling the KPNC Adult Alcohol Registry, which may be useful to other researchers and health systems, and to characterize the registry cohort descriptively, including underlying health conditions. METHODS: Inclusion criteria were adult members with unhealthy alcohol use (using National Institute on Alcohol Abuse and Alcoholism guidelines), an alcohol use disorder (AUD) diagnosis, or an alcohol-related health problem between June 1, 2013, and May 31, 2019. We extracted patients' longitudinal, multidimensional EHR data from 1 year before their date of eligibility through May 31, 2019, and conducted descriptive analyses. RESULTS: We identified 723,604 adult patients who met the registry inclusion criteria at any time during the study period: 631,780 with unhealthy alcohol use, 143,690 with an AUD diagnosis, and 18,985 with an alcohol-related health problem. We identified 65,064 patients who met two or more criteria. Of the 4,973,195 adult patients with at least one encounter with the health system during the study period, the prevalence of unhealthy alcohol use was 13% (631,780/4,973,195), the prevalence of AUD diagnoses was 3% (143,690/4,973,195), and the prevalence of alcohol-related health problems was 0.4% (18,985/4,973,195). The registry cohort was 60% male (n=432,847) and 41% non-White (n=295,998) and had a median age of 41 years (IQR=27). About 48% (n=346,408) had a chronic medical condition, 18% (n=130,031) had a mental health condition, and 4% (n=30,429) had a drug use disorder diagnosis. CONCLUSIONS: We demonstrated that EHR-based data collected during clinical care within an integrated health system could be leveraged to develop a registry of patients with alcohol problems that is flexible and can be easily updated. The registry's comprehensive patient-level data over multiyear periods provides a strong foundation for robust research addressing critical public health questions related to the full course and spectrum of alcohol problems, including recovery, which would complement other methods used in alcohol research (eg, population-based surveys, clinical trials).

Associations Between Medical Conditions and Alcohol Consumption Levels in an Adult Primary Care Population.

Importance: Excessive alcohol consumption is associated with increased incidence of several medical conditions, but few nonveteran, population-based studies have assessed levels of alcohol use across medical conditions. Objective: To examine associations between medical conditions and alcohol consumption levels in a population-based sample of primary care patients using electronic health record data. Design, Setting, and Participants: This cross-sectional study used separate multinomial logistic regression models to estimate adjusted associations between 26 medical conditions and alcohol consumption levels in a sample of 2 720 231 adult primary care patients screened for unhealthy drinking between January 1, 2014, and December 31, 2017, then only among those reporting alcohol use. The study was conducted at Kaiser Permanente Northern California, a large, integrated health care delivery system that incorporated alcohol screening into its adult primary care workflow. Data were analyzed from June 29, 2018, to February 7, 2020. Main Outcomes and Measures: The main outcome was level of alcohol use, classified as no reported use, low-risk use, exceeding daily limits only, exceeding weekly limits only, or exceeding daily and weekly limits, per National Institute on Alcohol Abuse and Alcoholism guidelines. Other measures included sociodemographic, body mass index, smoking, inpatient and emergency department use, and a dichotomous indicator for the presence of 26 medical conditions in the year prior to the alcohol screening identified using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10-CM diagnosis codes. Results: Among the 2 720 231 included patients, 1 439 361 (52.9%) were female, 1 308 659 (48.1%) were white, and 883 276 (32.5%) were aged 18 to 34 years. Patients with any of the conditions (except injury or poisoning) had lower odds of drinking at low-risk and unhealthy levels relative to no reported use compared with those without the condition. Among 861 427 patients reporting alcohol use, patients with diabetes (odds ratio [OR], 1.11; 95% CI, 1.08-1.15), hypertension (OR, 1.11; 95% CI, 1.09-1.13), chronic obstructive pulmonary disease (COPD; OR, 1.16; 95% CI, 1.10-1.22), or injury or poisoning (OR, 1.06; 95% CI, 1.04-1.07) had higher odds of exceeding daily limits only; those with atrial fibrillation (OR, 1.12; 95% CI, 1.06-1.18), cancer (OR, 1.06; 95% CI, 1.03-1.10), COPD (OR, 1.15; 95% CI, 1.09-1.20), or hypertension (OR, 1.37; 95% CI, 1.34-1.40) had higher odds of exceeding weekly limits only; and those with COPD (OR, 1.15; 95% CI, 1.07-1.23), chronic liver disease (OR, 1.42; 95% CI, 1.32-1.53), or hypertension (OR, 1.48; 95% CI, 1.44-1.52) had higher odds of exceeding both daily and weekly limits. Conclusions and Relevance: Findings suggest that patients with certain medical conditions are more likely to have elevated levels of alcohol use. Health systems and clinicians may want to consider approaches to help targeted patient subgroups limit unhealthy alcohol use and reduce health risks.