"We adjusted for race": now what? A systematic review of utilization and reporting of race in American Journal of Epidemiology and Epidemiology, 2020-2021.

Race is a social construct, commonly used in epidemiologic research to adjust for confounding. However, adjustment of race may mask racial disparities, thereby perpetuating structural racism. We conducted a systematic review of articles published in Epidemiology and American Journal of Epidemiology between 2020 and 2021 to (1) understand how race, ethnicity, and similar social constructs were operationalized, used, and reported; and (2) characterize good and poor practices of utilization and reporting of race data on the basis of the extent to which they reveal or mask systemic racism. Original research articles were considered for full review and data extraction if race data were used in the study analysis. We extracted how race was categorized, used-as a descriptor, confounder, or for effect measure modification (EMM)-and reported if the authors discussed racial disparities and systemic bias-related mechanisms responsible for perpetuating the disparities. Of the 561 articles, 299 had race data available and 192 (34.2%) used race data in analyses. Among the 160 US-based studies, 81 different racial categorizations were used. Race was most often used as a confounder (52%), followed by effect measure modifier (33%), and descriptive variable (12%). Fewer than 1 in 4 articles (22.9%) exhibited good practices (EMM along with discussing disparities and mechanisms), 63.5% of the articles exhibited poor practices (confounding only or not discussing mechanisms), and 13.5% were considered neither poor nor good practices. We discuss implications and provide 13 recommendations for operationalization, utilization, and reporting of race in epidemiologic and public health research.

Development and validation of an electronic health records-based opioid use disorder algorithm by expert clinical adjudication among patients with prescribed opioids.

BACKGROUND: In the US, over 200 lives are lost from opioid overdoses each day. Accurate and prompt diagnosis of opioid use disorders (OUD) may help prevent overdose deaths. However, international classification of disease (ICD) codes for OUD are known to underestimate prevalence, and their specificity and sensitivity are unknown. We developed and validated algorithms to identify OUD in electronic health records (EHR) and examined the validity of OUD ICD codes. METHODS: Through four iterations, we developed EHR-based OUD identification algorithms among patients who were prescribed opioids from 2014 to 2017. The algorithms and OUD ICD codes were validated against 169 independent "gold standard" EHR chart reviews conducted by an expert adjudication panel across four healthcare systems. After using 2014-2020 EHR for validating iteration 1, the experts were advised to use 2014-2017 EHR thereafter. RESULTS: Of the 169 EHR charts, 81 (48%) were reviewed by more than one expert and exhibited 85% expert agreement. The experts identified 54 OUD cases. The experts endorsed all 11 OUD criteria from the Diagnostic and Statistical Manual of Mental Disorders-5, including craving (72%), tolerance (65%), withdrawal (56%), and recurrent use in physically hazardous conditions (50%). The OUD ICD codes had 10% sensitivity and 99% specificity, underscoring large underestimation. In comparison our algorithm identified OUD with 23% sensitivity and 98% specificity. CONCLUSIONS AND RELEVANCE: This is the first study to estimate the validity of OUD ICD codes and develop validated EHR-based OUD identification algorithms. This work will inform future research on early intervention and prevention of OUD.

Trajectories and correlates of opioid prescription receipt among patients experiencing interpersonal violence.

Interpersonal violence increases vulnerability to the deleterious effects of opioid use. Increased opioid prescription receipt is a major contributor to the opioid crisis; however, our understanding of prescription patterns and risk factors among those with a history of interpersonal violence remains elusive. This study sought to identify 5-year longitudinal patterns of opioid prescription receipt among patients experiencing interpersonal violence within a large healthcare system and sociodemographic and clinical characteristics associated with prescription patterns. This secondary analysis examined electronic health record data from January 2004-August 2019 for a cohort of patients (N = 1,587) referred for interpersonal violence services. Latent class growth analysis was used to estimate trajectories of opioid prescription receipt over a 5-year period. Standardized differences were calculated to assess variation in sociodemographic and clinical characteristics between classes. Our cohort had a high prevalence of prescription opioid receipt (73.3%) and underlying co-morbidities, including chronic pain (54.6%), substance use disorders (39.0%), and mental health diagnoses (76.9%). Six prescription opioid receipt classes emerged, characterized by probability of any prescription opioid receipt at the start and end of the study period (high, medium, low, never) and change in probability over time (increasing, decreasing, stable). Classes with the highest probability of prescription opioids also had the highest proportions of males, chronic pain diagnoses, substance use disorders, and mental health diagnoses. Black, non-Hispanic and Hispanic patients were more likely to be in low or no prescription opioid receipt classes. These findings highlight the importance of monitoring for synergistic co-morbidities when providing pain management and offering treatment that is trauma-informed, destigmatizing, and integrated into routine care.

LGBT young adults' awareness of and receptivity to the This Free Life tobacco public education campaign.

PURPOSE: This study measures awareness of and receptivity to the Food and Drug Administration's This Free Life campaign seeking to change tobacco-related attitudes and beliefs among lesbian, gay, bisexual and/or transgender (LGBT) young adults. METHODS: Participants were young adults who self-identify as LGBT. The evaluation uses a treatment-control design. This study includes data from four survey rounds with participants from each round invited to participate in subsequent rounds and new participants invited to account for attrition. Bivariate analyses assess treatment-control differences in campaign awareness by round. We used multivariable logistic regression models with a time×treatment interaction and covariates to assess whether increases in awareness were greater in treatment than control from follow-ups 1 to 4. Descriptive statistics describe perceived effectiveness and models explore covariates of perceived effectiveness. RESULTS: At each round, an increasing number of participants in treatment were brand aware (25%-67%) and reported high (16%-34%) and medium (16%-25%) video awareness compared with control (all p<0.001). Regressions revealed interactions in brand and video awareness, wherein the effect of treatment on awareness increased more over time, with significant treatment-control differences in change from follow-up 1 to 4 (all p<0.05). Reactions to all but one ad were positive (one neutral) with mean perceived effectiveness scores from 3.21 to 3.92 ('neither disagree nor agree' to 'agree' on 5-point scale). Perceived effectiveness differed by LGBT identity (all p<0.05). CONCLUSIONS: At follow-up 4, This Free Life reached most of the campaign audience in treatment markets and has achieved higher awareness in treatment than control markets, at individual survey rounds and over time.

Awareness of and Receptivity to the Fresh Empire Tobacco Public Education Campaign Among Hip Hop Youth.

PURPOSE: The aim of the study was to assess awareness of and receptivity to the U.S. Food and Drug Administration's Fresh Empire tobacco public education campaign designed to reach Hip Hop-identified youth, who are at higher smoking risk. METHODS: The evaluation uses a randomized treatment-control design with 15 campaign-targeted treatment and 15 control markets. We conducted surveys among 12- to 17-year-olds before campaign launch and at approximately 6-month intervals. Analyses explore treatment-control differences in Fresh Empire brand and video advertisement awareness at individual survey rounds and over time and perceived effectiveness of advertisements. RESULTS: Awareness of the Fresh Empire brand was higher among youth in treatment than control markets following campaign launch (ps < .01). Awareness of the Fresh Empire brand increased more in treatment than control over time (adjusted odds ratio = 3.26, 95% confidence interval = 1.90-5.58). At follow-ups 1 and 3, youth in treatment (vs. control) were more likely to report high and less likely to report low awareness of video advertisements (ps < .05). There were no treatment-control differences in video awareness at follow-up 2 (not significant). Fresh Empire video advertisements had perceived effectiveness scores ranging from 3.66 to 4.11 (1-5 scale) across three survey rounds. CONCLUSIONS: Among the campaign audience of Hip Hop-identified youth, awareness of the Fresh Empire campaign was higher in treatment than control markets at individual survey rounds, and increases in campaign awareness were greater in treatment than control markets over time. Campaign advertisements also elicited positive audience reactions. Findings suggest that heavily digital campaigns may take longer to achieve Centers for Disease Control and Prevention-recommended 75% quarterly awareness.

Recruiting Hard-to-Reach Populations for Survey Research: Using Facebook and Instagram Advertisements and In-Person Intercept in LGBT Bars and Nightclubs to Recruit LGBT Young Adults.

BACKGROUND: Tobacco public education campaigns focus increasingly on hard-to-reach populations at higher risk for smoking, prompting campaign creators and evaluators to develop strategies to reach hard-to-reach populations in virtual and physical spaces where they spend time. OBJECTIVE: The aim of this study was to describe two novel recruitment strategies (in-person intercept interviews in lesbian, gay, bisexual, and transgender [LGBT] social venues and targeted social media ads) and compares characteristics of participants recruited via these strategies for the US Food and Drug Administration's This Free Life campaign evaluation targeting LGBT young adults who smoke cigarettes occasionally. METHODS: We recruited LGBT adults aged 18-24 years in the United States via Facebook and Instagram ads (N=1709, mean age 20.94, SD 1.94) or intercept in LGBT social venues (N=2348, mean age 21.98, SD 1.69) for the baseline evaluation survey. Covariates related to recruitment strategy were age; race or ethnicity; LGBT identity; education; pride event attendance; and alcohol, cigarette, and social media use. RESULTS: Lesbian or gay women (adjusted odds ratio, AOR 1.88, 95% CI 1.54-2.29, P<.001), bisexual men and women (AOR 1.46, 95% CI 1.17-1.82, P=.001), gender minorities (AOR 1.68, 95% CI 1.26-2.25, P<.001), and other sexual minorities (AOR 2.48, 95% CI 1.62-3.80, P<.001) were more likely than gay men to be recruited via social media (than intercept). Hispanic (AOR 0.73, 95% CI 0.61-0.89, P=.001) and other or multiracial, non-Hispanic participants (AOR 0.70, 95% CI 0.54-0.90, P=.006) were less likely than white, non-Hispanic participants to be recruited via social media. As age increased, odds of recruitment via social media decreased (AOR 0.76, 95% CI 0.72-0.80, P<.001). Participants with some college education (AOR 1.27, 95% CI 1.03-1.56, P=.03) were more likely than those with a college degree to be recruited via social media. Participants reporting past 30-day alcohol use were less likely to be recruited via social media (AOR 0.33, 95% CI 0.24-0.44, P<.001). Participants who reported past-year pride event attendance were more likely to be recruited via social media (AOR 1.31, 95% CI 1.06-1.64, P=.02), as well as those who used Facebook at least once daily (AOR 1.43, 95% CI 1.14-1.80, P=.002). Participants who reported using Instagram at least once daily were less likely to be recruited via social media (AOR 0.73, 95% CI 0.62-0.86, P<.001). Social media recruitment was faster (incidence rate ratio, IRR=3.31, 95% CI 3.11-3.52, P<.001) and less expensive (2.2% of combined social media and intercept recruitment cost) but had greater data quality issues-a larger percentage of social media respondents were lost because of duplicate and low-quality responses (374/4446, 8.41%) compared with intercept respondents lost to interviewer misrepresentation (15/4446, 0.34%; P<.001). CONCLUSIONS: Social media combined with intercept provided access to important LGBT subpopulations (eg, gender and other sexual minorities) and a more diverse sample. Social media methods have more data quality issues but are faster and less expensive than intercept. Recruiting hard-to-reach populations via audience-tailored strategies enabled recruitment of one of the largest LGBT young adult samples, suggesting these methods' promise for accessing hard-to-reach populations.