Faculty Bibliography

Indiscriminate use of predictive models incorporating race can reinforce biases present in source data and lead to an exacerbation of health disparities. In some countries, such as the United States, there is therefore a push to remove race from prediction models; however, there are still many prediction models that use race as an input. Biomedical informaticists who are given the responsibility of using these predictive models in healthcare environments are likely to be faced with questions like how to deal with race covariates in these models. Thus, there is a need for a pragmatic framework to help model users think through how to include race in their chosen model so as to avoid inadvertently exacerbating disparities. In this paper, we use the case study of lung cancer screening to propose a simple framework to guide how model users can approach the use (or non-use) of race inputs in the predictive models they are tasked with leveraging in electronic health records and clinical workflows.

OBJECTIVES:Few studies have considered how trends in opioid poisonings have changed among older adults. The objective of this study was to examine trends in fatal and nonfatal opioid-related poisonings ("exposures") among older adults. METHODS:National poison center data were used to examine trends in characteristics of reported exposures to commonly prescribed opioids between 2015 and 2021 among adults 60 years or older. We estimated the proportion of opioid exposures by demographic characteristics, the specific opioid(s) involved, exposure type, route of administration, other substances co-used, and medical outcomes for each calendar year. We estimated whether there were linear changes in prevalence by year using logistic regression. RESULTS:Although there was a decrease in the number of opioid exposures within the study population from 7706 in 2015 to 7337 in 2021 (a 4.8% decrease, P = 0.04), exposures increased for adults aged 70 to 79 years (a 14.0% increase, P < 0.001). The proportion classified as "abuse" increased by 63.3% ( P < 0.001). There were significant decreases in the proportion involving hydromorphone (a 23.3% decrease, P < 0.001) and morphine (a 22.0% decrease, P < 0.001), with an increase involving buprenorphine (a 216.0% increase, P < 0.001). The proportion increased for co-use of cocaine (a 488.9% increase, P < 0.001) and methamphetamine (a 220.0% increase, P = 0.02), with a decrease in co-use of benzodiazepines (a 25.5% decrease, P < 0.001). The proportion of major medical outcomes increased by 93.9% ( P < 0.001). CONCLUSIONS:National patterns of opioid-related poisonings are shifting among older adults, including the types of opioids involved and co-use of other drugs. These results can inform prevention and harm reduction efforts aimed at older adults.

The Score for Trauma Triage in the Geriatric and Middle-Aged (STTGMA) is a risk stratification tool. We evaluated the STTGMA's accuracy in predicting 30-day mortality and the odds of unfavorable clinical trajectories among crash-related trauma patients. This retrospective cohort study (n = 912) pooled adults aged 55 years and older from a single institutional trauma database. The data were split into training and test data sets (70:30 ratio) for the receiver operating curve analysis and internal validation, respectively. The outcome variables were 30-day mortality and measures of clinical trajectory. The predictor variable was the high-energy STTGMA score (STTGMAHE). We adjusted for the American Society of Anesthesiologists Physical Status. Using the training and test data sets, STTGMAHE exhibited 82% (95% CI: 65.5-98.3) and 96% (90.7-100.0) accuracies in predicting 30-day mortality, respectively. The STTGMA risk categories significantly stratified the proportions of orthopedic trauma patients who required intensive care unit (ICU) admissions, major and minor complications, and the length of stay (LOS). The odds of ICU admissions, major and minor complications, and the median difference in the LOS increased across the risk categories in a dose-response pattern. STTGMAHE exhibited an excellent level of accuracy in identifying middle-aged and geriatric trauma patients at risk of 30-day mortality and unfavorable clinical trajectories.

BACKGROUND:Management of orthostatic hypotension (OH) prioritizes prevention of standing hypotension, sometimes at the expense of supine hypertension. It is unclear whether supine hypertension is associated with adverse outcomes relative to standing hypotension. OBJECTIVES:To compare the long-term clinical consequences of supine hypertension and standing hypotension among middle-aged adults with and without OH. METHODS:The ARIC study (Atherosclerosis Risk in Communities) measured supine and standing blood pressure (BP) in adults aged 45 to 64 years, without neurogenic OH, between 1987 and 1989. We defined OH as a positional drop in systolic BP ≥20 mm Hg or diastolic BP ≥10 mm Hg, supine hypertension as supine BP≥140/≥90 mm Hg, and standing hypotension as standing BP≤105/≤65 mm Hg. Participants were followed for >30 years. We used Cox regression models to examine associations with cardiovascular disease events, all-cause mortality, falls, and syncope. RESULTS:-interactions >0.25). Supine hypertension was associated with heart failure (hazard ratio, 1.83 [95% CI, 1.68-1.99]), falls (hazard ratio, 1.12 [95% CI, 1.02-1.22]), and all-cause mortality (hazard ratio, 1.45 [95% CI, 1.37-1.54]), while standing hypotension was only significantly associated with mortality (hazard ratio, 1.06 [95% CI, 1.00-1.14]). CONCLUSIONS:Supine hypertension was associated with higher risk of adverse events than standing hypotension, regardless of OH status. This challenges conventional OH management, which prioritizes standing hypotension over supine hypertension.

BACKGROUND:Reporting race and ethnicity in clinical trial publications is critical for determining the generalizability and effectiveness of new treatments. This is particularly important for breast cancer, in which Black women have been shown to have between 40 and 100% higher mortality rate yet are underrepresented in trials. Our objective was to describe changes over time in the reporting of race/ethnicity in breast trial publications. PATIENTS AND METHODS/METHODS:We searched ClinicalTrials.gov to identify the primary publication linked to trials with results posted from May 2010-2022. Statistical analysis included summed frequencies and a linear regression model of the proportion of articles reporting race/ethnicity and the proportion of non-White enrollees over time. RESULTS:A proportion of 72 of the 98 (73.4%) studies that met inclusion criteria reported race/ethnicity. In a linear regression model of the proportion of studies reporting race/ethnicity as a function of time, there was no statistically significant change, although we detected a signal toward a decreasing trend (coefficient for quarter = -2.2, p = 0.2). Among all studies reporting race and ethnicity over the study period, the overall percentage of non-White enrollees during the study period was 21.9%, [standard error (s.e.) 1.8, 95% confidence interval (CI) 18.4, 25.5] with a signal towards a decreasing trend in Non-White enrollment [coefficient for year-quarter = -0.8 (p = 0.2)]. CONCLUSION/CONCLUSIONS:Our data demonstrate that both race reporting and overall representation of minority groups in breast cancer clinical trials did not improve over the last 12 years and may have, in fact, decreased. Increased reporting of race and ethnicity data forces the medical community to confront disparities in access to clinical trials. This may improve efforts to recruit and retain members of minority groups in clinical trials, and over time, reduce racial disparities in oncologic outcomes.

The growing number of people living with dementia (PLWD) requires a coordinated clinical response to deliver pragmatic, evidence-based interventions in frontline care settings. However, infrastructure to support such a response is lacking. Moreover, there are too few researchers conducting rigorous embedded pragmatic clinical trials (ePCTs) to make the vision of high quality, widely accessible dementia care a reality. National Institute on Aging (NIA) Imbedded Pragmatic Alzheimer's disease and Related Dementias Clinical Trials (IMPACT) Collaboratory seeks to improve the pipeline of early career researchers qualified to lead ePCTs by funding career development awards. Even with support from the Collaboratory, awardees face practical and methodological challenges to success, recently exacerbated by the COVID-19 pandemic. We first describe the training opportunities and support network for the IMPACT CDA recipients. This report then describes the unique career development challenges faced by early-career researchers involved in ePCTs for dementia care. Topics addressed include challenges in establishing a laboratory, academic promotion, mentoring and professional development, and work-life balance. Concrete suggestions to address these challenges are offered for early-career investigators, their mentors, and their supporting institutions. While some of these challenges are faced by researchers in other fields, this report seeks to provide a roadmap for expanding the work of the IMPACT Collaboratory and initiating future efforts to recruit, train, and retain talented early-career researchers involved in ePCTs for dementia care.

RATIONALE & OBJECTIVE/UNASSIGNED:Biomarkers of kidney disease progression have been identified in individuals with diabetes and underlying chronic kidney disease (CKD). Whether or not these markers are associated with the development of CKD in a general population without diabetes or CKD is not well established. STUDY DESIGN/UNASSIGNED:Prospective observational cohort. SETTING & PARTICIPANTS/UNASSIGNED:In the Atherosclerosis Risk in Communities) study, 948 participants were studied. EXPOSURES/UNASSIGNED:The baseline plasma biomarkers of kidney injury molecule-1 (KIM-1), monocyte chemoattractant protein-1 (MCP-1), soluble urokinase plasminogen activator receptor (suPAR), tumor necrosis factor receptor 1 (TNFR-1), tumor necrosis factor receptor 2 (TNFR-2), and human cartilage glycoprotein-39 (YKL-40) measured in 1996-1998. OUTCOME/UNASSIGNED:or dialysis dependence through United States Renal Data System linkage. ANALYTICAL APPROACH/UNASSIGNED:Logistic regression and C statistics. RESULTS/UNASSIGNED: < 0.01) and using the observed risk of 12% for incident CKD, the predicted risk gradient changed from 5%-40% (for the 1st-5th quintile) to 4%-44%. LIMITATIONS/UNASSIGNED:Biomarkers and creatinine were measured at one time point. CONCLUSIONS/UNASSIGNED:Higher levels of KIM-1, suPAR, TNFR-1, and TNFR-2 were associated with higher odds of incident CKD among individuals without diabetes. PLAIN-LANGUAGE SUMMARY/UNASSIGNED:For people with diabetes or kidney disease, several biomarkers have been shown to be associated with worsening kidney disease. Whether these biomarkers have prognostic significance in people without diabetes or kidney disease is less studied. Using the Atherosclerosis Risk in Communities study, we followed individuals without diabetes or kidney disease for an average of 15 years after biomarker measurement to see if these biomarkers were associated with the development of kidney disease. We found that elevated levels of KIM-1, suPAR, TNFR-1, and TNFR-2 were associated with the development of kidney disease. These biomarkers may help identify individuals who would benefit from interventions to prevent the development of kidney disease.

BACKGROUND/UNASSIGNED:Traditional Chinese medicine (TCM) body constitution (BC), primarily determined by physiological and clinical characteristics, is an important process for clinical diagnosis and treatment and play a critical role in precision medicine in TCM. The purpose of the study was to explore whether the distributions of BC types differed by obesity status. METHODS/UNASSIGNED:We conducted a study to evaluate BC type in US population during 2012-2016. A total of 191 White participants from Personalized Prevention of Colorectal Cancer Trial (PPCCT) completed a self-administered Traditional Chinese Medicine Questionnaire (TCMQ, English version). In this study, we further compared the distribution of major types of TCM BC in the PPCCT to those Chinese populations stratified by obesity status. RESULTS/UNASSIGNED:We found the Blood-stasis frequency was higher in US White adults, 22.6% for individuals with BMI <30 and 11.2% for obese individuals, compared to 1.4% and 1.8%, respectively, in Chinese populations. We also found the percentages Inherited-special and Qi-stagnation were higher in US White adults than those in Chinese populations regardless of obesity status. However, the proportions of Yang-deficiency were higher in Chinese populations than those in our study conducted in US White adults regardless of obesity status. CONCLUSIONS/UNASSIGNED:These new findings indicate the difference in distribution of BC types we observed between US and Chinese populations cannot be explained by the differences in prevalence of obesity. Further studies are needed to confirm our findings and understand the potential mechanism including genetic background and/or environmental factors.

The 2030 Agenda for Sustainable Development sets out, through 17 Sustainable Development Goals (SDGs), a path for the prosperity of people and the planet. SDG 3 in particular aims to ensure healthy lives and promote well-being for all at all ages and includes several targets to enhance health. This review presents a "headache-tailored" perspective on how to achieve SDG 3 by focusing on six specific actions: targeting chronic headaches; reducing the overuse of acute pain-relieving medications; promoting the education of healthcare professionals; granting access to medication in low- and middle-income countries (LMIC); implementing training and educational opportunities for healthcare professionals in low and middle income countries; building a global alliance against headache disorders. Addressing the burden of headache disorders directly impacts on populations' health, as well as on the possibility to improve the productivity of people aged below 50, women in particular. Our analysis pointed out several elements, and included: moving forward from frequency-based parameters to define headache severity; recognizing and managing comorbid diseases and risk factors; implementing a disease management multi-modal management model that incorporates pharmacological and non-pharmacological treatments; early recognizing and managing the overuse of acute pain-relieving medications; promoting undergraduate, postgraduate, and continuing medical education of healthcare professionals with specific training on headache; and promoting a culture that favors the recognition of headaches as diseases with a neurobiological basis, where this is not yet recognized. Making headache care more sustainable is an achievable objective, which will require multi-stakeholder collaborations across all sectors of society, both health-related and not health-related. Robust investments will be needed; however, considering the high prevalence of headache disorders and the associated disability, these investments will surely improve multiple health outcomes and lift development and well-being globally.