Evaluating tests for cluster-randomized trials with few clusters under generalized linear mixed models with covariate adjustment: A simulation study.

Generalized linear mixed models (GLMM) are commonly used to analyze clustered data, but when the number of clusters is small to moderate, standard statistical tests may produce elevated type I error rates. Small-sample corrections have been proposed for continuous or binary outcomes without covariate adjustment. However, appropriate tests to use for count outcomes or under covariate-adjusted models remains unknown. An important setting in which this issue arises is in cluster-randomized trials (CRTs). Because many CRTs have just a few clusters (eg, clinics or health systems), covariate adjustment is particularly critical to address potential chance imbalance and/or low power (eg, adjustment following stratified randomization or for the baseline value of the outcome). We conducted simulations to evaluate GLMM-based tests of the treatment effect that account for the small (10) or moderate (20) number of clusters under a parallel-group CRT setting across scenarios of covariate adjustment (including adjustment for one or more person-level or cluster-level covariates) for both binary and count outcomes. We find that when the intraclass correlation is non-negligible ( ≥ $$ \ge $$ 0.01) and the number of covariates is small ( ≤ $$ \le $$ 2), likelihood ratio tests with a between-within denominator degree of freedom have type I error rates close to the nominal level. When the number of covariates is moderate ( ≥ $$ \ge $$ 5), across our simulation scenarios, the relative performance of the tests varied considerably and no method performed uniformly well. Therefore, we recommend adjusting for no more than a few covariates and using likelihood ratio tests with a between-within denominator degree of freedom.

Differences in Weight Gain Following Residential Relocation in the Moving to Health (M2H) Study.

BACKGROUND: Neighborhoods may play an important role in shaping long-term weight trajectory and obesity risk. Studying the impact of moving to another neighborhood may be the most efficient way to determine the impact of the built environment on health. We explored whether residential moves were associated with changes in body weight. METHODS: Kaiser Permanente Washington electronic health records were used to identify 21,502 members aged 18-64 who moved within King County, WA between 2005 and 2017. We linked body weight measures to environment measures, including population, residential, and street intersection densities (800 m and 1,600 m Euclidian buffers) and access to supermarkets and fast foods (1,600 m and 5,000 m network distances). We used linear mixed models to estimate associations between postmove changes in environment and changes in body weight. RESULTS: In general, moving from high-density to moderate- or low-density neighborhoods was associated with greater weight gain postmove. For example, those moving from high to low residential density neighborhoods (within 1,600 m) gained an average of 4.5 (95% confidence interval [CI] = 3.0, 5.9) lbs 3 years after moving, whereas those moving from low to high-density neighborhoods gained an average of 1.3 (95% CI = -0.2, 2.9) lbs. Also, those moving from neighborhoods without fast-food access (within 1600m) to other neighborhoods without fast-food access gained less weight (average 1.6 lbs [95% CI = 0.9, 2.4]) than those moving from and to neighborhoods with fast-food access (average 2.8 lbs [95% CI = 2.5, 3.2]). CONCLUSIONS: Moving to higher-density neighborhoods may be associated with reductions in adult weight gain.

Practical Assessment of Alcohol Use Disorder in Routine Primary Care: Performance of an Alcohol Symptom Checklist.

BACKGROUND: Alcohol use disorder (AUD) is highly prevalent but underrecognized and undertreated in primary care settings. Alcohol Symptom Checklists can engage patients and providers in discussions of AUD-related care. However, the performance of Alcohol Symptom Checklists when they are used in routine care and documented in electronic health records (EHRs) remains unevaluated. OBJECTIVE: To evaluate the psychometric performance of an Alcohol Symptom Checklist in routine primary care. DESIGN: Cross-sectional study using item response theory (IRT) and differential item functioning analyses of measurement consistency across age, sex, race, and ethnicity. PATIENTS: Patients seen in primary care in the Kaiser Permanente Washington Healthcare System who reported high-risk drinking on the Alcohol Use Disorder Identification Test Consumption screening measure (AUDIT-C ≥ 7) and subsequently completed an Alcohol Symptom Checklist between October 2015 and February 2020. MAIN MEASURE: Alcohol Symptom Checklists with 11 items assessing AUD criteria defined in the Diagnostic and Statistical Manual for Mental Disorders, 5th edition (DSM-5), completed by patients during routine medical care and documented in EHRs. KEY RESULTS: Among 11,464 patients who screened positive for high-risk drinking and completed an Alcohol Symptom Checklist (mean age 43.6 years, 30.5% female), 54.1% reported ≥ 2 DSM-5 AUD criteria (threshold for AUD diagnosis). IRT analyses demonstrated that checklist items measured a unidimensional continuum of AUD severity. Differential item functioning was observed for some demographic subgroups but had minimal impact on accurate measurement of AUD severity, with differences between demographic subgroups attributable to differential item functioning never exceeding 0.42 points of the total symptom count (of a possible range of 0-11). CONCLUSIONS: Alcohol Symptom Checklists used in routine care discriminated AUD severity consistently with current definitions of AUD and performed equitably across age, sex, race, and ethnicity. Integrating symptom checklists into routine care may help inform clinical decision-making around diagnosing and managing AUD.

Bayesian kernel machine regression-causal mediation analysis.

Greater understanding of the pathways through which an environmental mixture operates is important to design effective interventions. We present new methodology to estimate natural direct and indirect effects and controlled direct effects of a complex mixture exposure on an outcome through a mediator variable. We implement Bayesian Kernel Machine Regression (BKMR) to allow for all possible interactions and nonlinear effects of (1) the co-exposures on the mediator, (2) the co-exposures and mediator on the outcome, and (3) selected covariates on the mediator and/or outcome. From the posterior predictive distributions of the mediator and outcome, we simulate counterfactuals to obtain posterior samples, estimates, and credible intervals of the mediation effects. Our simulation study demonstrates that when the exposure-mediator and exposure-mediator-outcome relationships are complex, BKMR-Causal Mediation Analysis performs better than current mediation methods. We applied our methodology to quantify the contribution of birth length as a mediator between in utero co-exposure to arsenic, manganese, and lead, and children's neurodevelopmental scores, in a prospective birth cohort in Bangladesh. Among younger children, we found a negative (adverse) association between the metal mixture and neurodevelopment. We also found evidence that birth length mediates the effect of exposure to the metal mixture on neurodevelopment for younger children. If birth length were fixed to its 75th percentile value, the harmful effect of the metal mixture on neurodevelopment is attenuated, suggesting nutritional interventions to help increase fetal growth, and thus birth length, could potentially block the harmful effect of the metal mixture on neurodevelopment.

Clarifying selection bias in cluster randomized trials.

BACKGROUND: In cluster randomized trials, patients are typically recruited after clusters are randomized, and the recruiters and patients may not be blinded to the assignment. This often leads to differential recruitment and consequently systematic differences in baseline characteristics of the recruited patients between intervention and control arms, inducing post-randomization selection bias. We aim to rigorously define causal estimands in the presence of selection bias. We elucidate the conditions under which standard covariate adjustment methods can validly estimate these estimands. We further discuss the additional data and assumptions necessary for estimating causal effects when such conditions are not met. METHODS: Adopting the principal stratification framework in causal inference, we clarify there are two average treatment effect (ATE) estimands in cluster randomized trials: one for the overall population and one for the recruited population. We derive analytical formula of the two estimands in terms of principal-stratum-specific causal effects. Furthermore, using simulation studies, we assess the empirical performance of the multivariable regression adjustment method under different data generating processes leading to selection bias. RESULTS: When treatment effects are heterogeneous across principal strata, the average treatment effect on the overall population generally differs from the average treatment effect on the recruited population. A naïve intention-to-treat analysis of the recruited sample leads to biased estimates of both average treatment effects. In the presence of post-randomization selection and without additional data on the non-recruited subjects, the average treatment effect on the recruited population is estimable only when the treatment effects are homogeneous between principal strata, and the average treatment effect on the overall population is generally not estimable. The extent to which covariate adjustment can remove selection bias depends on the degree of effect heterogeneity across principal strata. CONCLUSION: There is a need and opportunity to improve the analysis of cluster randomized trials that are subject to post-randomization selection bias. For studies prone to selection bias, it is important to explicitly specify the target population that the causal estimands are defined on and adopt design and estimation strategies accordingly. To draw valid inferences about treatment effects, investigators should (1) assess the possibility of heterogeneous treatment effects, and (2) consider collecting data on covariates that are predictive of the recruitment process, and on the non-recruited population from external sources such as electronic health records.

Baseline representativeness of patients in clinics enrolled in the PRimary care Opioid Use Disorders treatment (PROUD) trial: comparison of trial and non-trial clinics in the same health systems.

BACKGROUND: Pragmatic primary care trials aim to test interventions in "real world" health care settings, but clinics willing and able to participate in trials may not be representative of typical clinics. This analysis compared patients in participating and non-participating clinics from the same health systems at baseline in the PRimary care Opioid Use Disorders treatment (PROUD) trial. METHODS: This observational analysis relied on secondary electronic health record and administrative claims data in 5 of 6 health systems in the PROUD trial. The sample included patients 16-90 years at an eligible primary care visit in the 3 years before randomization. Each system contributed 2 randomized PROUD trial clinics and 4 similarly sized non-trial clinics. We summarized patient characteristics in trial and non-trial clinics in the 2 years before randomization ("baseline"). Using mixed-effect regression models, we compared trial and non-trial clinics on a baseline measure of the primary trial outcome (clinic-level patient-years of opioid use disorder (OUD) treatment, scaled per 10,000 primary care patients seen) and a baseline measure of the secondary trial outcome (patient-level days of acute care utilization among patients with OUD). RESULTS: Patients were generally similar between the 10 trial clinics (n = 248,436) and 20 non-trial clinics (n = 341,130), although trial clinics' patients were slightly younger, more likely to be Hispanic/Latinx, less likely to be white, more likely to have Medicaid/subsidized insurance, and lived in less wealthy neighborhoods. Baseline outcomes did not differ between trial and non-trial clinics: trial clinics had 1.0 more patient-year of OUD treatment per 10,000 patients (95% CI: - 2.9, 5.0) and a 4% higher rate of days of acute care utilization than non-trial clinics (rate ratio: 1.04; 95% CI: 0.76, 1.42). CONCLUSIONS: trial clinics and non-trial clinics were similar regarding most measured patient characteristics, and no differences were observed in baseline measures of trial primary and secondary outcomes. These findings suggest trial clinics were representative of comparably sized clinics within the same health systems. Although results do not reflect generalizability more broadly, this study illustrates an approach to assess representativeness of clinics in future pragmatic primary care trials.

Clinical documentation of patient-reported medical cannabis use in primary care: Toward scalable extraction using natural language processing methods.

Background: Most states have legalized medical cannabis, yet little is known about how medical cannabis use is documented in patients' electronic health records (EHRs). We used natural language processing (NLP) to calculate the prevalence of clinician-documented medical cannabis use among adults in an integrated health system in Washington State where medical and recreational use are legal. Methods: We analyzed EHRs of patients ≥18 years old screened for past-year cannabis use (November 1, 2017-October 31, 2018), to identify clinician-documented medical cannabis use. We defined medical use as any documentation of cannabis that was recommended by a clinician or described by the clinician or patient as intended to manage health conditions or symptoms. We developed and applied an NLP system that included NLP-assisted manual review to identify such documentation in encounter notes. Results: Medical cannabis use was documented for 16,684 (5.6%) of 299,597 outpatient encounters with routine screening for cannabis use among 203,489 patients seeing 1,274 clinicians. The validated NLP system identified 54% of documentation and NLP-assisted manual review the remainder. Language documenting reasons for cannabis use included 125 terms indicating medical use, 28 terms indicating non-medical use and 41 ambiguous terms. Implicit documentation of medical use (e.g., "edible THC nightly for lumbar pain") was more common than explicit (e.g., "continues medical cannabis use"). Conclusions: Clinicians use diverse and often ambiguous language to document patients' reasons for cannabis use. Automating extraction of documentation about patients' cannabis use could facilitate clinical decision support and epidemiological investigation but will require large amounts of gold standard training data.

Joint Associations of Multiple Dietary Components With Cardiovascular Disease Risk: A Machine-Learning Approach.

The human diet consists of a complex mixture of components. To realistically assess dietary impacts on health, new statistical tools that can better address nonlinear, collinear, and interactive relationships are necessary. Using data from 1,928 healthy participants in the Coronary Artery Risk Development in Young Adults (CARDIA) cohort (1985-2006), we explored the association between 12 dietary factors and 10-year predicted risk of atherosclerotic cardiovascular disease (ASCVD) using an innovative approach, Bayesian kernel machine regression (BKMR). Employing BKMR, we found that among women, unprocessed red meat was most strongly related to the outcome: An interquartile range increase in unprocessed red meat consumption was associated with a 0.07-unit (95% credible interval: 0.01, 0.13) increase in ASCVD risk when intakes of other dietary components were fixed at their median values (similar results were obtained when other components were fixed at their 25th and 75th percentile values). Among men, fruits had the strongest association: An interquartile range increase in fruit consumption was associated with -0.09-unit (95% credible interval (CrI): -0.16, -0.02), -0.10-unit (95% CrI: -0.16, -0.03), and -0.11-unit (95% CrI: -0.18, -0.04) lower ASCVD risk when other dietary components were fixed at their 25th, 50th (median), and 75th percentile values, respectively. Using BKMR to explore the complex structure of the total diet, we found distinct sex-specific diet-ASCVD relationships and synergistic interaction between whole grain and fruit consumption.

Differential associations of the built environment on weight gain by sex and race/ethnicity but not age.

OBJECTIVE: To explore the built environment (BE) and weight change relationship by age, sex, and racial/ethnic subgroups in adults. METHODS: Weight trajectories were estimated using electronic health records for 115,260 insured Kaiser Permanente Washington members age 18-64 years. Member home addresses were geocoded using ArcGIS. Population, residential, and road intersection densities and counts of area supermarkets and fast food restaurants were measured with SmartMaps (800 and 5000-meter buffers) and categorized into tertiles. Linear mixed-effect models tested whether associations between BE features and weight gain at 1, 3, and 5 years differed by age, sex, and race/ethnicity, adjusting for demographics, baseline weight, and residential property values. RESULTS: Denser urban form and greater availability of supermarkets and fast food restaurants were associated with differential weight change across sex and race/ethnicity. At 5 years, the mean difference in weight change comparing the 3rd versus 1st tertile of residential density was significantly different between males (-0.49 kg, 95% CI: -0.68, -0.30) and females (-0.17 kg, 95% CI: -0.33, -0.01) (P-value for interaction = 0.011). Across race/ethnicity, the mean difference in weight change at 5 years for residential density was significantly different among non-Hispanic (NH) Whites (-0.47 kg, 95% CI: -0.61, -0.32), NH Blacks (-0.86 kg, 95% CI: -1.37, -0.36), Hispanics (0.10 kg, 95% CI: -0.46, 0.65), and NH Asians (0.44 kg, 95% CI: 0.10, 0.78) (P-value for interaction <0.001). These findings were consistent for other BE measures. CONCLUSION: The relationship between the built environment and weight change differs across demographic groups. Careful consideration of demographic differences in associations of BE and weight trajectories is warranted for investigating etiological mechanisms and guiding intervention development.

Does the built environment have independent obesogenic power? Urban form and trajectories of weight gain.

OBJECTIVE: To determine whether selected features of the built environment can predict weight gain in a large longitudinal cohort of adults. METHODS: Weight trajectories over a 5-year period were obtained from electronic health records for 115,260 insured patients aged 18-64 years in the Kaiser Permanente Washington health care system. Home addresses were geocoded using ArcGIS. Built environment variables were population, residential unit, and road intersection densities captured using Euclidean-based SmartMaps at 800-m buffers. Counts of area supermarkets and fast food restaurants were obtained using network-based SmartMaps at 1600, and 5000-m buffers. Property values were a measure of socioeconomic status. Linear mixed effects models tested whether built environment variables at baseline were associated with long-term weight gain, adjusting for sex, age, race/ethnicity, Medicaid insurance, body weight, and residential property values. RESULTS: Built environment variables at baseline were associated with differences in baseline obesity prevalence and body mass index but had limited impact on weight trajectories. Mean weight gain for the full cohort was 0.06 kg at 1 year (95% CI: 0.03, 0.10); 0.64 kg at 3 years (95% CI: 0.59, 0.68), and 0.95 kg at 5 years (95% CI: 0.90, 1.00). In adjusted regression models, the top tertile of density metrics and frequency counts were associated with lower weight gain at 5-years follow-up compared to the bottom tertiles, though the mean differences in weight change for each follow-up year (1, 3, and 5) did not exceed 0.5 kg. CONCLUSIONS: Built environment variables that were associated with higher obesity prevalence at baseline had limited independent obesogenic power with respect to weight gain over time. Residential unit density had the strongest negative association with weight gain. Future work on the influence of built environment variables on health should also examine social context, including residential segregation and residential mobility.

Prevalence and Medication Treatment of Opioid Use Disorder Among Primary Care Patients with Hepatitis C and HIV.

BACKGROUND: Hepatitis C and HIV are associated with opioid use disorders (OUD) and injection drug use. Medications for OUD can prevent the spread of HCV and HIV. OBJECTIVE: To describe the prevalence of documented OUD, as well as receipt of office-based medication treatment, among primary care patients with HCV or HIV. DESIGN: Retrospective observational cohort study using electronic health record and insurance data. PARTICIPANTS: Adults ≥ 18 years with ≥ 2 visits to primary care during the study (2014-2016) at 6 healthcare systems across five states (CO, CA, OR, WA, and MN). MAIN MEASURES: The primary outcome was the diagnosis of OUD; the secondary outcome was OUD treatment with buprenorphine or oral/injectable naltrexone. Prevalence of OUD and OUD treatment was calculated across four groups: HCV only; HIV only; HCV and HIV; and neither HCV nor HIV. In addition, adjusted odds ratios (AOR) of OUD treatment associated with HCV and HIV (separately) were estimated, adjusting for age, gender, race/ethnicity, and site. KEY RESULTS: The sample included 1,368,604 persons, of whom 10,042 had HCV, 5821 HIV, and 422 both. The prevalence of diagnosed OUD varied across groups: 11.9% (95% CI: 11.3%, 12.5%) for those with HCV; 1.6% (1.3%, 2.0%) for those with HIV; 8.8% (6.2%, 11.9%) for those with both; and 0.92% (0.91%, 0.94%) among those with neither. Among those with diagnosed OUD, the prevalence of OUD medication treatment was 20.9%, 16.0%, 10.8%, and 22.3%, for those with HCV, HIV, both, and neither, respectively. HCV was not associated with OUD treatment (AOR = 1.03; 0.88, 1.21), whereas patients with HIV had a lower probability of OUD treatment (AOR = 0.43; 0.26, 0.72). CONCLUSIONS: Among patients receiving primary care, those diagnosed with HCV and HIV were more likely to have documented OUD than those without. Patients with HIV were less likely to have documented medication treatment for OUD.

Alcohol Use and Antiretroviral Adherence Among Patients Living with HIV: Is Change in Alcohol Use Associated with Change in Adherence?

Alcohol use increases non-adherence to antiretroviral therapy (ART) among persons living with HIV (PLWH). Dynamic longitudinal associations are understudied. Veterans Aging Cohort Study (VACS) data 2/1/2008-7/31/16 were used to fit linear regression models estimating changes in adherence (% days with ART medication fill) associated with changes in alcohol use based on annual clinically-ascertained AUDIT-C screening scores (range - 12 to + 12, 0 = no change) adjusting for demographics and initial adherence. Among 21,275 PLWH (67,330 observations), most reported no (48%) or low-level (39%) alcohol use initially, with no (55%) or small (39% ≤ 3 points) annual change. Mean initial adherence was 86% (SD 21%), mean annual change was - 3.1% (SD 21%). An inverted V-shaped association was observed: both increases and decreases in AUDIT-C were associated with greater adherence decreases relative to stable scores [p < 0.001, F (4, 21,274)]. PLWH with dynamic alcohol use (potentially indicative of alcohol use disorder) should be considered for adherence interventions.

RESUMEN: El consumo de alcohol aumenta el no-cumplimiento a la terapia antirretroviral (TARV) entre las personas que viven con VIH. No se han estudiado lo suficiente las dinámicas asociaciones longitudinales. Los datos del Estudio de la Envejecimiento de Cohorte de Veteranos (EECV) (1/2/2008­31/7/2016) fueron usados para encajar modelos de regresión lineal estimando los cambios en cumplimiento (% de días con medicaciones TARV surtidas) asociados con los cambios en el consumo de alcohol basado en los resultados anuales de las evaluaciones AUDIT-C, determinadas clínicamente, (una gama de -12 a + 12, 0 = cero cambio) adaptándose a las estadísticas demográficas y cumplimiento inicial. Entre 21,275 personas que viven con VIH (67,330 observaciones), la mayoría reportó ningún (48%) o bajos niveles del (39%) consumo de alcohol inicialmente, con ningún (55%) o muy pequeño (39% ≤ 3 puntos) cambio anual. la media inicial de cumplimiento fue 86% (DE 21%). La media de cambio anual fue -3.1% (DE 21%). Se observó una asociación de forma V invertida: tanto los aumentos como las disminuciones en AUDIT-C fueron asociados con mayor disminuciones de cumplimiento en comparación con resultados estables (p < 0.001, F (4, 21,274)). Personas que viven con VIH con el consumo dinámico de alcohol (potencialmente indicativo de un trastorno por consumo de alcohol) deben ser considerados por intervenciones de cumplimiento.

Prevalence of prescription opioid use during pregnancy in eight US health plans during 2001-2014.

PURPOSE: To estimate prevalence of prescription opioid use during pregnancy in eight US health plans during 2001-2014. METHODS: We conducted a cohort study of singleton live birth deliveries. Maternal characteristics were ascertained from health plan and/or birth certificate data and opioids dispensed during pregnancy from health plan pharmacy records. Prevalence of prescription opioid use during pregnancy was calculated for any use, cumulative days of use, and number of dispensings. RESULTS: We examined prevalence of prescription opioid use during pregnancy in each health plan. Tennessee Medicaid had appreciably greater prevalence of use compared to the seven other health plans. Thus, results for the two groups were reported separately. In the seven health plans (n = 587 093 deliveries), prevalence of use during pregnancy was relatively stable at 9%-11% throughout 2001-2014. In Tennessee Medicaid (n = 256 724 deliveries), prevalence increased from 29% in 2001 to a peak of 36%-37% in 2004-2010, and then declined to 28% in 2014. Use for ≥30 days during pregnancy was stable at 1% in the seven health plans and increased from 2% to 7% in Tennessee Medicaid during 2001-2014. Receipt of ≥5 opioid dispensings during pregnancy increased in the seven health plans (0.3%-0.6%) and Tennessee Medicaid (3%-5%) during 2001-2014. CONCLUSION: During 2001-2014, prescription opioid use during pregnancy was more common in Tennessee Medicaid (peak prevalence in late 2000s) compared to the seven health plans (relatively stable prevalence). Although a small percentage of women had opioid use during pregnancy for ≥30 days or ≥ 5 dispensings, they represent thousands of women during 2001-2014.

Routine Assessment of Symptoms of Substance Use Disorders in Primary Care: Prevalence and Severity of Reported Symptoms.

BACKGROUND: Most patients with substance use disorders (SUDs) never receive treatment and SUDs are under-recognized in primary care (PC) where patients can be treated or linked to treatment. Asking PC patients to directly report SUD symptoms on questionnaires might help identify SUDs but to our knowledge, this approach is previously untested. OBJECTIVE: To describe the prevalence and severity of DSM-5 SUD symptoms reported by PC patients as part of routine care. DESIGN: Cross-sectional study using secondary data. PARTICIPANTS: A total of 241,265 adult patients who visited one of 25 PC sites in an integrated health system in Washington state and had alcohol, cannabis, or other drug use screening documented in their EHRs (March 2015-July 2018) were included in main analyses if they had a positive screen for high-risk substance use defined as AUDIT-C score 7-12 points, or report of past-year daily cannabis use or any other drug use. MAIN MEASURES: The main outcome was number of SUD symptoms based on Diagnostic and Statistical Manual, 5th edition (DSM-5), reported on Symptom Checklists (0-11) for alcohol or other drugs: 2-3 mild; 4-5 moderate; 6-11 severe. RESULTS: Of screened patients, 16,776 (5.7%) reported high-risk use of alcohol (2.4%), cannabis (3.9%), and/or other drugs (1.7%), and 65.0-69.9% of those completed Symptom Checklists. Of those with high-risk alcohol use, 52.5% (95% CI 50.9-54.0%) reported ≥ 2 symptoms consistent with mild-severe alcohol use disorders. Of those reporting daily cannabis use, 29.8% (28.6-30.9%) reported ≥ 2 symptoms consistent with mild-severe SUDs. Of those reporting any other drug use, 37.5% (35.7-39.3%) reported ≥ 2 symptoms consistent with mild-severe SUDs. CONCLUSIONS AND RELEVANCE: Many PC patients who screened positive for high-risk substance use reported symptoms consistent with DSM-5 SUDs on self-report Symptom Checklists. Use of SUD Symptom Checklists could support PC providers in making SUD diagnoses and initiating discussions of substance use.

Validity of diagnosis and procedure codes for identifying neural tube defects in infants.

PURPOSE: The use of validated criteria to identify birth defects in electronic healthcare databases can avoid the cost and time-intensive efforts required to conduct chart reviews to confirm outcomes. This study evaluated the validity of various case-finding methodologies to identify neural tube defects (NTDs) in infants using an electronic healthcare database. METHODS: This analysis used data generated from a study whose primary aim was to evaluate the association between first-trimester maternal prescription opioid use and NTDs. The study was conducted within the Medication Exposure in Pregnancy Risk Evaluation Program. A broad approach was used to identify potential NTDs including diagnosis and procedure codes from inpatient and outpatient settings, death certificates and birth defect flags in birth certificates. Potential NTD cases were chart abstracted and confirmed by clinical experts. Positive predictive values (PPVs) and 95% confidence intervals (95% CI) are reported. RESULTS: The cohort included 113 168 singleton live-born infants: 55 960 infants with opioid exposure in pregnancy and 57 208 infants unexposed in pregnancy. Seventy-three potential NTD cases were available for the validation analysis. The overall PPV was 41% using all diagnosis and procedure codes plus birth certificates. Restricting approaches to codes recorded in the infants' medical record or to birth certificate flags increased the PPVs (72% and 80%, respectively) but missed a substantial proportion of confirmed NTDs. CONCLUSIONS: Codes in electronic healthcare data did not accurately identify confirmed NTDs. These results indicate that chart review with adjudication of outcomes is important when conducting observational studies of NTDs using electronic healthcare data.

Challenges and Opportunities for Using Big Health Care Data to Advance Medical Science and Public Health.

Methodological advancements in epidemiology, biostatistics, and data science have strengthened the research world's ability to use data captured from electronic health records (EHRs) to address pressing medical questions, but gaps remain. We describe methods investments that are needed to curate EHR data toward research quality and to integrate complementary data sources when EHR data alone are insufficient for research goals. We highlight new methods and directions for improving the integrity of medical evidence generated from pragmatic trials, observational studies, and predictive modeling. We also discuss needed methods contributions to further ease data sharing across multisite EHR data networks. Throughout, we identify opportunities for training and for bolstering collaboration among subject matter experts, methodologists, practicing clinicians, and health system leaders to help ensure that methods problems are identified and resulting advances are translated into mainstream research practice more quickly.

Lagged kernel machine regression for identifying time windows of susceptibility to exposures of complex mixtures.

The impact of neurotoxic chemical mixtures on children's health is a critical public health concern. It is well known that during early life, toxic exposures may impact cognitive function during critical time intervals of increased vulnerability, known as windows of susceptibility. Knowledge on time windows of susceptibility can help inform treatment and prevention strategies, as chemical mixtures may affect a developmental process that is operating at a specific life phase. There are several statistical challenges in estimating the health effects of time-varying exposures to multi-pollutant mixtures, such as: multi-collinearity among the exposures both within time points and across time points, and complex exposure-response relationships. To address these concerns, we develop a flexible statistical method, called lagged kernel machine regression (LKMR). LKMR identifies critical exposure windows of chemical mixtures, and accounts for complex non-linear and non-additive effects of the mixture at any given exposure window. Specifically, LKMR estimates how the effects of a mixture of exposures change with the exposure time window using a Bayesian formulation of a grouped, fused lasso penalty within a kernel machine regression (KMR) framework. A simulation study demonstrates the performance of LKMR under realistic exposure-response scenarios, and demonstrates large gains over approaches that consider each time window separately, particularly when serial correlation among the time-varying exposures is high. Furthermore, LKMR demonstrates gains over another approach that inputs all time-specific chemical concentrations together into a single KMR. We apply LKMR to estimate associations between neurodevelopment and metal mixtures in Early Life Exposures in Mexico and Neurotoxicology, a prospective cohort study of child health in Mexico City.

Effect of a Care Management Intervention on 12-Month Drinking Outcomes Among Patients With and Without DSM-IV Alcohol Dependence at Baseline.

BACKGROUND: The CHOICE care management intervention did not improve drinking relative to usual care (UC) for patients with frequent heavy drinking at high risk of alcohol use disorders. Patients with alcohol dependence were hypothesized to benefit most. We conducted preplanned secondary analyses to test whether the CHOICE intervention improved drinking relative to UC among patients with and without baseline DSM-IV alcohol dependence. METHODS: A total of 304 patients reporting frequent heavy drinking from 3 VA primary care clinics were randomized (stratified by DSM-IV alcohol dependence, sex, and site) to UC or the patient-centered, nurse-delivered, 12-month CHOICE care management intervention. Primary outcomes included percent heavy drinking days (%HDD) using 28-day timeline follow-back and a "good drinking outcome" (GDO)-abstaining or drinking below recommended limits and no alcohol-related symptoms on the Short Inventory of Problems at 12 months. Generalized estimating equation binomial regression models (clustered on provider) with interaction terms between dependence and intervention group were fit. RESULTS: At baseline, 59% of intervention and UC patients had DSM-IV alcohol dependence. Mean drinking outcomes improved for all subgroups. For participants with dependence, 12-month outcomes did not differ for intervention versus UC patients (%HDD 37% versus 38%, p = 0.76 and GDO 16% versus 16%, p = 0.77). For participants without dependence, %HDD did not differ between intervention (41%) and UC (31%) patients (p = 0.12), but the proportion with GDO was significantly higher among UC participants (26% versus 13%, p = 0.046). Neither outcome was significantly modified by dependence (interaction p values 0.19 for %HDD and 0.10 for GDO). CONCLUSIONS: Among participants with frequent heavy drinking, care management had no benefit relative to UC for patients with dependence, but UC may have had benefits for those without dependence. TRIAL REGISTRATION: ClinicalTrials.gov NCT01400581.

Bayesian varying coefficient kernel machine regression to assess neurodevelopmental trajectories associated with exposure to complex mixtures.

Exposure to environmental mixtures can exert wide-ranging effects on child neurodevelopment. However, there is a lack of statistical methods that can accommodate the complex exposure-response relationship between mixtures and neurodevelopment while simultaneously estimating neurodevelopmental trajectories. We introduce Bayesian varying coefficient kernel machine regression (BVCKMR), a hierarchical model that estimates how mixture exposures at a given time point are associated with health outcome trajectories. The BVCKMR flexibly captures the exposure-response relationship, incorporates prior knowledge, and accounts for potentially nonlinear and nonadditive effects of individual exposures. This model assesses the directionality and relative importance of a mixture component on health outcome trajectories and predicts health effects for unobserved exposure profiles. Using contour plots and cross-sectional plots, BVCKMR also provides information about interactions between complex mixture components. The BVCKMR is applied to a subset of data from PROGRESS, a prospective birth cohort study in Mexico city on exposure to metal mixtures and temporal changes in neurodevelopment. The mixture include metals such as manganese, arsenic, cobalt, chromium, cesium, copper, lead, cadmium, and antimony. Results from a subset of Programming Research in Obesity, Growth, Environment and Social Stressors data provide evidence of significant positive associations between second trimester exposure to copper and Bayley Scales of Infant and Toddler Development cognition score at 24 months, and cognitive trajectories across 6-24 months. We also detect an interaction effect between second trimester copper and lead exposures for cognition at 24 months. In summary, BVCKMR provides a framework for estimating neurodevelopmental trajectories associated with exposure to complex mixtures.

Medical diagnoses of heat wave-related hospital admissions in older adults.

Heat waves have been associated with adverse human health effects, including higher rates of all-cause and cardiovascular mortality, and these health effects may be exacerbated under continued climate change. However, specific causes of hospitalizations associated with heat waves have not been characterized on a national scale. We systematically estimated the risks of cause-specific hospitalizations during heat waves in a national cohort of 23.7 million Medicare enrollees residing in 1943 U.S. counties during 1999-2010. Heat waves were defined as ≥2 consecutive days exceeding the county's 99th percentile of daily temperatures, and were matched to non-heat wave periods by county and week. We considered 50 outcomes from broad disease groups previously associated with heat wave-related hospitalizations, and estimated cause-specific relative risks (RRs) of hospital admissions on heat wave days. We identified 11 diagnoses with a higher admission risk on heat wave days, with heat stroke and sunstroke having the highest risk (RR = 22.5, [95% CI 14.9-34.2]). Other diseases with elevated risks included fluid and electrolyte disorders [(Hyperosmolality RR = 1.4, [95% CI 1.1-1.3]; Hypoosmolaltiy RR = 1.2, [95% CI 1.1-1.3])] and acute kidney failure (RR = 1.1, [95% CI 1.1-1.2]). These risks tended to be higher under more severe heat wave events. In addition, risks were higher among adults in the oldest (≥85) category (reference: 65-74) for volume depletion and heat exhaustion. Several causes of hospitalization identified are preventable, and public health interventions, including early warning systems and plans targeting risk factors for these illnesses, could reduce adverse effects of heat in the present and under climate change.