Using natural language processing of clinical text to enhance identification of opioid-related overdoses in electronic health records data.

PURPOSE: To enhance automated methods for accurately identifying opioid-related overdoses and classifying types of overdose using electronic health record (EHR) databases. METHODS: We developed a natural language processing (NLP) software application to code clinical text documentation of overdose, including identification of intention for self-harm, substances involved, substance abuse, and error in medication usage. Using datasets balanced with cases of suspected overdose and records of individuals at elevated risk for overdose, we developed and validated the application using Kaiser Permanente Northwest data, then tested portability of the application using Kaiser Permanente Washington data. Datasets were chart-reviewed to provide a gold standard for comparison and evaluation of the automated method. RESULTS: The method performed well in identifying overdose (sensitivity = 0.80, specificity = 0.93), intentional overdose (sensitivity = 0.81, specificity = 0.98), and involvement of opioids (excluding heroin, sensitivity = 0.72, specificity = 0.96) and heroin (sensitivity = 0.84, specificity = 1.0). The method performed poorly at identifying adverse drug reactions and overdose due to patient error and fairly at identifying substance abuse in opioid-related unintentional overdose (sensitivity = 0.67, specificity = 0.96). Evaluation using validation datasets yielded significant reductions, in specificity and negative predictive values only, for many classifications mentioned above. However, these measures remained above 0.80, thus, performance observed during development was largely maintained during validation. Similar results were obtained when evaluating portability, although there was a significant reduction in sensitivity for unintentional overdose that was attributed to missing text clinical notes in the database. CONCLUSIONS: Methods that process text clinical notes show promise for improving accuracy and fidelity at identifying and classifying overdoses according to type using EHR data.

Development of an algorithm to identify inpatient opioid-related overdoses and oversedation using electronic data.

PURPOSE: To facilitate surveillance and evaluate interventions addressing opioid-related overdoses, algorithms are needed for use in large health care databases to identify and differentiate community-occurring opioid-related overdoses from inpatient-occurring opioid-related overdose/oversedation. METHODS: Data were from Kaiser Permanente Northwest (KPNW), a large integrated health plan. We iteratively developed and evaluated an algorithm for electronically identifying inpatient overdose/oversedation in KPNW hospitals from 1 January 2008 to 31 December 2014. Chart audits assessed accuracy; data sources included administrative and clinical records. RESULTS: The best-performing algorithm used these rules: (1) Include events with opioids administered in an inpatient setting (including emergency department/urgent care) followed by naloxone administration within 275 hours of continuous inpatient stay; (2) exclude events with electroconvulsive therapy procedure codes; and (3) exclude events in which an opioid was administered prior to hospital discharge and followed by readmission with subsequent naloxone administration. Using this algorithm, we identified 870 suspect inpatient overdose/oversedation events and chart audited a random sample of 235. Of the random sample, 185 (78.7%) were deemed overdoses/oversedation, 37 (15.5%) were not, and 13 (5.5%) were possible cases. The number of hours between time of opioid and naloxone administration did not affect algorithm accuracy. When "possible" overdoses/oversedations were included with confirmed events, overall positive predictive value (PPV) was very good (PPV = 84.0%). Additionally, PPV was reasonable when evaluated specifically for hospital stays with emergency/urgent care admissions (PPV = 77.0%) and excellent for elective surgery admissions (PPV = 97.0%). CONCLUSIONS: Algorithm performance was reasonable for identifying inpatient overdose/oversedation with best performance among elective surgery patients.

Identifying and classifying opioid-related overdoses: A validation study.

PURPOSE: The study aims to develop and validate algorithms to identify and classify opioid overdoses using claims and other coded data, and clinical text extracted from electronic health records using natural language processing (NLP). METHODS: Primary data were derived from Kaiser Permanente Northwest (2008-2014), an integrated health care system (~n > 475 000 unique individuals per year). Data included International Classification of Diseases, Ninth Revision (ICD-9) codes for nonfatal diagnoses, International Classification of Diseases, Tenth Revision (ICD-10) codes for fatal events, clinical notes, and prescription medication records. We assessed sensitivity, specificity, positive predictive value, and negative predictive value for algorithms relative to medical chart review and conducted assessments of algorithm portability in Kaiser Permanente Washington, Tennessee State Medicaid, and Optum. RESULTS: Code-based algorithm performance was excellent for opioid-related overdoses (sensitivity = 97.2%, specificity = 84.6%) and classification of heroin-involved overdoses (sensitivity = 91.8%, specificity = 99.0%). Performance was acceptable for code-based suicide/suicide attempt classifications (sensitivity = 70.7%, specificity = 90.5%); sensitivity improved with NLP (sensitivity = 78.7%, specificity = 91.0%). Performance was acceptable for the code-based substance abuse-involved classification (sensitivity = 75.3%, specificity = 79.5%); sensitivity improved with the NLP-enhanced algorithm (sensitivity = 80.5%, specificity = 76.3%). The opioid-related overdose algorithm performed well across portability assessment sites, with sensitivity greater than 96% and specificity greater than 84%. Cross-site sensitivity for heroin-involved overdose was greater than 87%, specificity greater than or equal to 99%. CONCLUSIONS: Code-based algorithms developed to detect opioid-related overdoses and classify them according to heroin involvement perform well. Algorithms for classifying suicides/attempts and abuse-related opioid overdoses perform adequately for use for research, particularly given the complexity of classifying such overdoses. The NLP-enhanced algorithms for suicides/suicide attempts and abuse-related overdoses perform significantly better than code-based algorithms and are appropriate for use in settings that have data and capacity to use NLP.

Effects of primary care clinician beliefs and perceived organizational facilitators on the delivery of preventive care to individuals with mental illnesses.

BACKGROUND: Although many studies have documented patient-, clinician-, and organizational barriers/facilitators of primary care among people with mental illnesses, few have examined whether these factors predict actual rates of preventive service use. We assessed whether clinician behaviors, beliefs, characteristics, and clinician-reported organizational characteristics, predicted delivery of preventive services in this population. METHODS: Primary care clinicians (n = 247) at Kaiser Permanente Northwest (KPNW) or community health centers and safety-net clinics (CHCs), in six states, completed clinician surveys in 2014. Using electronic health record data, we calculated preventive care-gap rates for patients with mental illnesses empaneled to survey respondents (n = 37,251). Using separate multi-level regression models for each setting, we tested whether survey responses predicted preventive service care-gap rates. RESULTS: After controlling for patient-level characteristics, patients of clinicians who reported a greater likelihood of providing preventive care to psychiatrically asymptomatic patients experienced lower care-gap rates (KPNW γ= - .05, p = .041; CHCs γ= - .05, p = .033). In KPNW, patients of female clinicians had fewer care gaps than patients of male clinicians (γ= - .07, p = .011). In CHCs, patients of clinicians who had practiced longer had fewer care gaps (γ= - .004, p = .010), as did patients whose clinicians believed that organizational quality goals facilitate preventive service provision (γ= - .06, p = .006). Case manager availability in CHCs was associated with higher care-gap rates (γ=.06, p = .028). CONCLUSIONS: Clinicians who report they are likely to address preventive concerns when their mentally ill patients present without apparent psychiatric symptoms had patients with fewer care gaps. In CHCs, care quality goals may facilitate preventive care whereas case managers may not.

Assessing the accuracy of opioid overdose and poisoning codes in diagnostic information from electronic health records, claims data, and death records.

PURPOSE: The purpose of this study is to assess positive predictive value (PPV), relative to medical chart review, of International Classification of Diseases (ICD)-9/10 diagnostic codes for identifying opioid overdoses and poisonings. METHODS: Data were obtained from Kaiser Permanente Northwest and Northern California. Diagnostic data from electronic health records, submitted claims, and state death records from Oregon, Washington, and California were linked. Individual opioid-related poisoning codes (e.g., 965.xx and X42), and adverse effects of opioids codes (e.g., E935.xx) combined with diagnoses possibly indicative of overdoses (e.g., respiratory depression), were evaluated by comparison with chart audits. RESULTS: Opioid adverse effects codes had low PPV to detect overdoses (13.4%) as assessed in 127 charts and were not pursued. Instead, opioid poisoning codes were assessed in 2100 individuals who had those codes present in electronic health records in the period between the years 2008 and 2012. Of these, 10/2100 had no available information and 241/2100 were excluded potentially as anesthesia-related. Among the 1849 remaining individuals with opioid poisoning codes, 1495 events were accurately identified as opioid overdoses; 69 were miscodes or misidentified, and 285 were opioid adverse effects, not overdoses. Thus, PPV was 81%. Opioid adverse effects or overdoses were accurately identified in 1780 of 1849 events (96.3%). CONCLUSIONS: Opioid poisoning codes have a predictive value of 81% to identify opioid overdoses, suggesting ICD opioid poisoning codes can be used to monitor overdose rates and evaluate interventions to reduce overdose. Further research to assess sensitivity, specificity, and negative predictive value are ongoing. Copyright © 2017 John Wiley & Sons, Ltd.

A qualitative analysis of family involvement in prescribed opioid medication monitoring among individuals who have experienced opioid overdoses.

BACKGROUND: Little is known about the role, extent, or effects of family member involvement in monitoring and managing opioid analgesics. Knowing when or how family members monitor prescribed opioid medication taking, whether it is acceptable to patients, or how family relationships may be affected by monitoring, are not well documented. METHODS: The study was conducted at Kaiser Permanente Northwest, an integrated health plan in Oregon and Washington. Semistructured in-depth interviews (N = 87) assessed circumstances surrounding overdose events among individuals who either experienced an opioid-related overdose or were family members of patients who died as a result of such an overdose. A subset of participants (n = 20) described family members' roles in monitoring opioid medications before or after overdoses. Interviews were transcribed verbatim and coded using Atlas.ti. We used a modified grounded theory approach to categorize emergent data and to identify common themes. RESULTS: When family members played roles in monitoring and managing opioid medications, clinicians were often unaware of their involvement. Patients and family members reported better outcomes when the patient, caregiver, and clinician developed a shared treatment plan. Negative outcomes included relationship stress, particularly when patients and caregivers had differing perspectives about what constituted effective pain management versus misuse and abuse. CONCLUSIONS: When families are concerned about opioid medications, coordination between clinicians, patients, and family carers appears to clarify roles and foster better outcomes. Increased stress and worse outcomes were reported when clinicians were not actively involved and when they did not attend to carers' concerns.

An Economic Evaluation of a Weight Loss Intervention Program for People with Serious Mental Illnesses Taking Antipsychotic Medications.

Individuals with serious mental illnesses suffer from obesity and cardiometabolic diseases at high rates, and antipsychotic medications exacerbate these conditions. While studies have shown weight loss and lifestyle interventions can be effective in this population, few have assessed intervention cost-effectiveness. We present results from a 12-month randomized controlled trial that reduced weight, fasting glucose, and medical hospitalizations in intervention participants. Costs per participant ranged from $4365 to $5687. Costs to reduce weight by one kilogram ranged from $1623 to $2114; costs to reduce fasting glucose by 1 mg/dL ranged from $467 to $608. Medical hospitalization costs were reduced by $137,500.

Dual recovery among people with serious mental illnesses and substance problems: a qualitative analysis.

OBJECTIVE: Individuals with serious mental illnesses are more likely to have substance-related problems than those without mental health problems. They also face more difficult recovery trajectories as they cope with dual disorders. Nevertheless, little is known about individuals' perspectives regarding their dual recovery experiences. METHODS: This qualitative analysis was conducted as part of an exploratory mixed-methods study of mental health recovery. Members of Kaiser Permanente Northwest (a group-model, not-for-profit, integrated health plan) who had serious mental illness diagnoses were interviewed four times over two years about factors affecting their mental health recovery. Interviews were recorded, transcribed, and coded with inductively derived codes. Themes were identified by reviewing text coded "alcohol or other drugs." RESULTS: Participants (N = 177) had diagnosed schizophrenia/schizoaffective disorder (n = 75, 42%), bipolar I/II disorder (n = 84, 48%), or affective psychosis (n = 18, 10%). At baseline, 63% (n = 112) spontaneously described addressing substance use as part of their mental health recovery. When asked at follow-up, 97% (n = 171) provided codeable answers about substances and mental health. We identified differing pathways to recovery, including through formal treatment, self-help groups or peer support, "natural" recovery (without the help of others), and continued but controlled use of alcohol. We found three overarching themes in participants' experiences of recovering from serious mental illnesses and substance-related problems: Learning about the effects of alcohol and drugs provided motivation and a foundation for sobriety; achieving sobriety helped people to initiate their mental health recovery processes; and achieving and maintaining sobriety built self-efficacy, self-confidence, improved functioning and a sense of personal growth. Non-judgmental support from clinicians adopting chronic disease approaches also facilitated recovery. CONCLUSIONS: Irrespective of how people achieved sobriety, quitting or severely limiting use of substances was important to initiating and continuing mental health recovery processes. Substance abuse treatment approaches that are flexible, reduce barriers to engagement, support learning about effects of substances on mental health and quality of life, and adopt a chronic disease model of addiction may increase engagement and success. Peer-based support like Alcoholics or Narcotics Anonymous can be helpful for people with serious mental illnesses, particularly when programs accept use of mental health medications.

Purposeful Sampling for Qualitative Data Collection and Analysis in Mixed Method Implementation Research.

Purposeful sampling is widely used in qualitative research for the identification and selection of information-rich cases related to the phenomenon of interest. Although there are several different purposeful sampling strategies, criterion sampling appears to be used most commonly in implementation research. However, combining sampling strategies may be more appropriate to the aims of implementation research and more consistent with recent developments in quantitative methods. This paper reviews the principles and practice of purposeful sampling in implementation research, summarizes types and categories of purposeful sampling strategies and provides a set of recommendations for use of single strategy or multistage strategy designs, particularly for state implementation research.

Approaches to Mixed Methods Dissemination and Implementation Research: Methods, Strengths, Caveats, and Opportunities.

Limited translation of research into practice has prompted study of diffusion and implementation, and development of effective methods of encouraging adoption, dissemination and implementation. Mixed methods techniques offer approaches for assessing and addressing processes affecting implementation of evidence-based interventions. We describe common mixed methods approaches used in dissemination and implementation research, discuss strengths and limitations of mixed methods approaches to data collection, and suggest promising methods not yet widely used in implementation research. We review qualitative, quantitative, and hybrid approaches to mixed methods dissemination and implementation studies, and describe methods for integrating multiple methods to increase depth of understanding while improving reliability and validity of findings.

A 12-week weight reduction intervention for overweight individuals taking antipsychotic medications.

People taking antipsychotic medications are at increased risk for obesity, diabetes, and early mortality. Few weight loss interventions have targeted this population. Thirty-six individuals were randomized to an evidence-based 12-week weight loss intervention (PREMIER with DASH diet, n = 18) or to usual care (n = 18) in this feasibility trial. Average attendance was 8.6 of 12 sessions. Intent-to-treat analyses of covariance, adjusted for baseline weight, showed significant changes in weight: Mean weight in intervention participants declined from 213.3 to 206.6 pounds, while control participants' weight was unchanged. It is possible to recruit, assess, intervene with, and retain participants taking antipsychotic medications in a dietary and exercise lifestyle change trial. Participants reported high levels of satisfaction with the intervention.

STRIDE: a randomized trial of a lifestyle intervention to promote weight loss among individuals taking antipsychotic medications.

BACKGROUND: Individuals diagnosed with serious mental illnesses are at increased risk of obesity- and cardiovascular-related morbidity and early mortality. Lifestyle interventions aimed at weight loss, even those adapted to suit the needs of this particular subgroup, have rarely produced clinically meaningful reductions in weight. METHODS/DESIGN: The STRIDE study is a multi-site, parallel, two-arm randomized controlled translational trial. Participants were recruited from community mental health clinics and an integrated not-for-profit health system. Participants were randomized either to usual care or to a 12-month intervention that consisted of: 1) weekly group participation for six months covering topics on nutrition, physical activity and lifestyle changes; 2) monthly group participation for an additional six month maintenance period; and 3) individual monthly contacts from intervention group facilitators during the second six month phase. All participants are assessed at baseline, 6, 12, and 24 months post-enrollment. Process and implementation evaluations are included and the study design includes a cost-utility analysis. Participants include 200 individuals with serious mental illness with an average age of 47.1 years, a mean body-mass index of 38.3 kg/m(2) and taking an average of 3.2 psychiatric medications at baseline. Baseline physiological measures included mean blood pressure (SBP/DBP) measurements of 119.2 (SD = 14.7)/79.4 (SD = 10.1); 35% reported a hypertension diagnosis and 11% took antihypertensive medications. Average lipid levels (mg/dL) were: a) triglycerides 188.0 (SD = 138.6), ranged from 43 to 1145; b) LDL 101.4 (SD = 32.9) and ranged from 17 to 185; c) HDL 45.8 (SD = 12.7) and ranged from 22 to 89; and d) total cholesterol 181.6 (SD = 39.7) and ranged from 50 to 324. Average fasting glucose levels were 108.9 (SD = 32.5) and ranged from 24 to 289. Average fasting insulin levels were 13.0 (SD=11.9) and ranged from 2 to 99. DISCUSSION: The STRIDE study is based on a modified version of the PREMIER comprehensive lifestyle intervention, DASH diet arm. STRIDE has successfully enrolled 200 individuals with serious mental illness in community-based settings. Baseline characteristics present a population at high risk for obesity-related negative health outcomes and demonstrate the need for evidence-based interventions to reduce these risks. TRIAL REGISTRATION: Clinical Trials.gov NCT00790517.

Engaging youths with serious mental illnesses in treatment: STARS study consumer recommendations.

OBJECTIVE: The purpose of this study was to identify better methods of engaging youths in mental health services by asking experienced mental health consumers for suggestions for clinicians. METHODS: 177 members of an integrated health plan, ranging in age from 16-84 years and diagnosed with schizophrenia, schizoaffective disorder, bipolar disorder, or affective psychosis, completed four in-depth semistructured interviews over 24 months as part of a study of recovery from serious mental illness. We transcribed and coded interviews, extracted a set of common themes addressing consumer recommendations to clinicians, and compared these themes across age groups. RESULTS: Five primary themes emerged in participants' recommendations: (1) use an age-appropriate approach that reflects youth culture and lifestyles; (2) foster development of autonomy; (3) take a personal, rather than diagnostic, approach; (4) be empathetic and authentic; and (5) create a safe and supportive environment. Consumers age 30 and older described three additional areas in which clinicians could contribute to youths' well being: (1) help find the right diagnosis and the right medication, (2) counsel youths to avoid using alcohol and drugs, and (3) take steps to help prevent social isolation. CONCLUSIONS: Study findings suggest that many strategies recommended for working with adults may benefit young people, but that developmentally appropriate modifications to these approaches are needed to foster treatment engagement among youths.

An exploration of site effects in a multisite trial of OROS-methylphenidate for smokers with attention deficit/hyperactivity disorder.

BACKGROUND: Multisite trials, the gold standard for conducting studies in community-based settings, can mask variability across sites resulting in misrepresentation of effects in specific sites. In a placebo-controlled trial of osmotic-release oral system methylphenidate (OROS-MPH) as augmentation treatment for smokers with attention deficit hyperactivity/impulsivity disorder (ADHD), three types of sites were selected according to their clinical research specialty (ADHD, smoking cessation, and general mental health). OBJECTIVE: Analysis was conducted to determine if clinical outcomes, that is, reduction in ADHD symptoms and smoking cessation rates, and the effect of treatment on these outcomes would differ by type of site. METHOD: A total of 255 adult smokers diagnosed with ADHD were enrolled in three clinic types: 72 in ADHD, 79 in tobacco dependence, and 104 in the mental health clinics. RESULTS: The three site-types were similar in demographic characteristics, smoking history, baseline level of ADHD symptoms, and history of psychiatric illness. Site-type but not a site-type by treatment interaction predicted prolonged smoking abstinence. A significant three-way interaction of site-type, treatment, and time-predicted improvement in ADHD symptoms. Moderate to strong effects of OROS-MPH relative to placebo were observed in the mental health and the ADHD clinics; a weak effect was observed in the tobacco dependence clinics. CONCLUSION: OROS-MPH benefit varied by site for reducing ADHD symptoms but not for improving smoking abstinence. SCIENTIFIC SIGNIFICANCE: Assessment of site-type effects can indicate the generalizability of findings from multisite trials and should be routinely incorporated in the design of multisite trials.

Gender research in the National Institute on Drug Abuse National Treatment Clinical Trials Network: a summary of findings.

BACKGROUND: The National Institute of Drug Abuse's National Drug Abuse Treatment Clinical Trials Network (CTN) was established to foster translation of research into practice in substance abuse treatment settings. The CTN provides a unique opportunity to examine in multi-site, translational clinical trials, the outcomes of treatment interventions targeting vulnerable subgroups of women; the comparative effectiveness of gender-specific protocols to reduce risk behaviors; and gender differences in clinical outcomes. OBJECTIVES: To review gender-related findings from published CTN clinical trials and related studies from January 2000 to March 2010. METHODS: CTN studies were selected for review if they focused on treatment outcomes or services for special populations of women with substance use disorders (SUDs) including those with trauma histories, pregnancy, co-occurring eating and other psychiatric disorders, and HIV risk behaviors; or implemented gender-specific protocols. The CTN has randomized 11,500 participants (41% women) across 200 clinics in 24 randomized controlled trials in community settings, of which 4 have been gender-specific. RESULTS: This article summarizes gender-related findings from CTN clinical trials and related studies, focusing on trauma histories, pregnancy, co-occurring eating and other psychiatric disorders, and HIV risk behaviors. CONCLUSIONS: These published studies have expanded the evidence base regarding interventions for vulnerable groups of women with SUDs as well as gender-specific interventions to reduce HIV risk behaviors in substance-using men and women. The results also underscore the complexity of accounting for gender in the design of clinical trials and analysis of results. SCIENTIFIC SIGNIFICANCE: To fully understand the relevance of gender-specific moderators and mediators of outcome, it is essential that future translational studies adopt more sophisticated approaches to understanding and measuring gender-relevant factors and plan sample sizes that are adequate to support more nuanced analytic methods.

Development of the Patient Activation Measure for mental health.

Our objective was to adapt the physical health Patient Activation Measure (PAM) for use among people with mental health conditions (PAM-MH). Data came from three studies among people with chronic mental health conditions and were combined in Rasch analyses. The PAM-MH's psychometric properties equal those of the original 13-item PAM. Test-retest reliability and concurrent validity were good, and the PAM-MH showed sensitivity to change. The PAM-MH appears to be a reliable and valid measure of patient activation among individuals with mental health problems. It appears to have potential for use in assessing change in activation.

Drinking Patterns, Gender and Health III: Avoiding vs. Seeking Healthcare.

BACKGROUND: Inability to predict most health services use and costs using demographics and health status suggests that other factors affect use, including attitudes and practices that influence health and willingness to seek care. Alcohol consumption has generated interest because heavy, chronic consumption causes adverse health consequences, acute consumption increases injury, and moderate drinking is linked to better health while hazardous drinking and alcohol-related problems are stigmatized and may affect willingness to seek care. METHODS: A stratified random sample of health-plan members completed a mail survey, yielding 7884 respondents (2995 male/4889 female). We linked survey data to 24 months of health-plan records to examine relationships between alcohol use, gender, health-related attitudes, practices, health, and service use. In-depth interviews with a stratified 150-respondent subsample explored individuals' reasons for seeking or avoiding care. RESULTS: Quantitative results suggest health-related practices and attitudes predict subsequent service use. Consistent predictors of care were having quit drinking, current at-risk consumption, cigarette smoking, higher BMI, disliking visiting doctors, and strong religious/spiritual beliefs. Qualitative analyses suggest embarrassment and shame are strong motivators for avoiding care. CONCLUSIONS: Although models included numerous health, functional status, attitudinal and behavioral predictors, variance explained was similar to previous reports, suggesting more complex relationships than expected. Qualitative analyses suggest several potential predictive factors not typically measured in service-use studies: embarrassment and shame, fear, faith that the body will heal, expectations about likelihood of becoming seriously ill, disliking the care process, the need to understand health problems, and the effects of self-assessments of health-related functional limitations.

Drinking Patterns, Gender and Health II: Predictors of Preventive Service Use.

BACKGROUND: Chronic diseases and injuries are elevated among people with substance use problems/dependence, yet heavier drinkers use fewer routine and preventive health services than non-drinkers and moderate drinkers, while former drinkers and abstainers use more than moderate drinkers. Researchers hypothesize that drinking clusters with attitudes and practices that produce better health among moderate drinkers and that heavy drinkers avoid doctors until becoming ill, subsequently quitting and using more services. Gender differences in alcohol consumption, health-related attitudes, practices, and prevention-services use may affect these relationships. METHODS: A stratified random sample of health-plan members (7884; 2995 males, 4889 females) completed a mail survey that was linked to 24 months of health-plan records. Data were used to examine relationships between alcohol use, gender, health-related attitudes/practices, health, and prevention-service use. RESULTS: Controlling for attitudes, practices, and health, female lifelong abstainers and former drinkers were less likely to have mammograms; individuals with alcohol use disorders and positive AUDIT scores were less likely to obtain influenza vaccinations. AUDIT-positive women were less likely to undergo colorectal screening than AUDIT-positive men. Consistent predictors of prevention-services use were: self-report of having a primary care provider (positive); disliking visiting the doctor (negative); smoking cigarettes (negative), and higher BMI (negative). CONCLUSIONS: When factors associated with drinking are controlled, patterns of alcohol consumption have limited effects on preventive service use. Individuals with stigmatized behaviors (e.g., hazardous/harmful drinking, smoking, or high BMIs) are less likely to receive care. Making care experiences positive and carefully addressing stigmatized health practices could increase preventive service use.

Drinking Patterns, Gender and Health I: Attitudes and Health Practices.

BACKGROUND: Despite considerable research, relationships among gender, alcohol consumption, and health remain controversial, due to potential confounding by health-related attitudes and practices associated with drinking, measurement challenges, and marked gender differences in drinking. We examined gender/alcohol consumption differences in health-related attitudes and practices, and evaluated how these factors affected relationships among gender, alcohol consumption, and health status. METHODS: A stratified random sample of adult health-plan members completed a mail survey, yielding 7884 respondents (2995 male/4889 female). Using MANCOVAs and adjusting for health-related attitudes, values, and practices, we examined gender differences in relationships between alcohol consumption and health. RESULTS: More frequent heavy drinking was associated with worse health-related attitudes and values, worse feelings about visiting the doctor, and worse health-related practices. Relationships between health-related practices and alcohol use differed by gender, and daily or almost daily heavy drinking was associated with significantly lower physical and mental health for women compared to men. Drinking status (lifelong abstainers, former drinkers, and level of regular alcohol consumption) was related to health status and vitality, even after adjusting for health-related attitudes, values, and practices. Relationships did not differ by gender. Former drinkers reported lower physical and mental health status than either lifelong abstainers or current drinkers. CONCLUSIONS: Drinking status is independently related to physical health, mental health, and vitality, even after controlling for the health-related attitudes, values, and practices expected to confound these relationships. Among current drinkers, women who engage in very frequent heavy drinking have worse physical and mental health than their male counterparts.