Environmental conditions during childhood impact gene transcription, resulting in health issues during adulthood

The opioid epidemic toll likely higher than originally reported

Patient age may bias cardiac health care

Environmental conditions during childhood impact gene transcription, resulting in health issues during adulthood

Researchers supported by grants from the NICHD, NIA, and the University of Zürich recently published findings that may explain how sociodemographic factors early in life affect future health. Health later in life varies significantly by individual demographic characteristics such as age, sex, and race/ethnicity, as well as by social factors including socioeconomic status and geographic region. However, the biological mechanisms of these disparities remain poorly understood. Neuroendocrine regulation of gene expression has been hypothesized as one pathway by which social environmental conditions could contribute to health disparities. For example, experiencing chronic stress, like poverty, can lead to changes in genes expression resulting in a shift called a "conserved transcriptional response to adversity". Previous research has indicated that this shift is characterized by more activity in inflammatory genes, a down-regulation in genes that are protective against viral infections and are associated with cancer and cardiovascular disease. The aim of this study was to determine if sociodemographic variations in the gene regulation of immune and inflammatory molecular pathways could contribute to the social gradients in chronic disease risk later in life.

The researchers conducted a transcriptome profiling study of inflammatory and antiviral gene activity using data from 1,069 young adults (mean age = 37) from the National Longitudinal Study of Adolescent to Adult Health (Add Health), a large, nationally representative and ethnically diverse sample with peripheral blood transcriptome profiles. The analyses focused on two molecular pathways involved in the pathogenesis of several chronic diseases. These included the genes involved in inflammation and genes involved in type I interferon (IFN) responses. The variation in the expression of genes involved in inflammation and type I interferon (IFN) response was analyzed as a function of individual demographic factors, sociodemographic conditions, and biobehavioral factors (smoking, drinking, and body mass index).

The results indicated that sociodemographic variations in the gene activity of these molecular pathways emerge by young adulthood, well before the presentation of chronic illness. Differential gene expression was most pronounced by sex, race/ethnicity, and body mass index (BMI), but transcriptome correlates were identified for every demographic dimension analyzed, including individual demographic characteristics (age, sex, race/ethnicity) and social context (family poverty, region of residence). Inflammation-related gene expression exhibited the most pronounced variation as a function of biobehavioral factors (BMI and smoking), whereas type I IFN-related immune transcripts varied as a function of individual demographic characteristics (sex and race/ethnicity). Bioinformatic analyses of transcription factor and immune-cell activation based on transcriptome-wide empirical differences identified additional effects of family poverty and geographic region. These results identify pervasive sociodemographic differences in immune-cell gene regulation that emerge by young adulthood and may help explain social disparities in the development of chronic illness and premature mortality at older ages. This study provides detailed evidence that this shift in gene transcription due to societal context exists, and it happens when people are very young. Additionally, these findings may provide a framework for reducing health disparities by mitigating molecular risk gradients by starting social, behavioral, and/or policy interventions early in life before the development of overt disease.

Citation:
Cole SW, Shanahan MJ, Gaydosh L, Harris KM. 2020. Population-based RNA profiling in Add Health finds social disparities in inflammatory and antiviral gene regulation to emerge by young adulthood, PNAS https://doi.org/10.1073/pnas.1821367117.

The opioid epidemic toll likely higher than originally reported

Researchers supported by grants from the NIH Common Fund shed light on the extent of the opioid epidemic. According to the Centers for Disease Control and Prevention (CDC), over 700,000 people died in the United States from drug overdoses between 1999 and 2017. In 2017, there were 70,237 deaths, of which 67.8% involved an opioid. However, fatal opioid overdoses maybe underreported since the drug involved in an overdose is not always specified on death certificates. “Other and unspecified drugs” were implicated in 21–25% of drug overdoses in 1999–2013 and 15–19% in 2014–2016. Additionally, many reports of fatal drug overdoses are missing information on specific drug involvement, which could lead to underreporting of opioid‐related death rates and a misrepresentation of the extent of the opioid epidemic. In order to address this problem, this study compared methodological approaches used to predict opioid involvement in unclassified drug overdoses in United States death records and then used the best-performing method to estimate the number of fatal opioid overdoses from 1999 to 2016.

The researchers performed a secondary data analysis of all drug overdoses in the United States from 1999–2016 obtained from the National Center for Health Statistics’ detailed Multiple Cause of Death records (n = 632,331 drug overdose deaths). Drug overdoses with known drug classification comprised 78.2% of the cases (n = 494,316), and unclassified drug overdoses comprised 21.8% (n = 138,015). Known opioid involvement was defined using ICD‐10 codes recorded in the set of contributing causes of death. Opioid involvement in unclassified drug overdoses was predicted using multiple methodological approaches: logistic regression and machine learning techniques, inclusion/exclusion of contributing causes of death, and inclusion/exclusion of county‐level characteristics. Logistic regression and random forest models performed similarly, thus the logistic regression was chosen for its ease of use. Including contributing causes substantially improved predictive accuracy.

Using the best-performing prediction model, the researchers found that 71.8% of unclassified drug overdoses in 1999–2016 involved opioids, which then translates into an additional 99,160 opioid‐related deaths, or approximately 28% more deaths than previously reported. Interestingly, there was a striking geographic variation in undercounting of opioid overdoses. When modeling opioid involvement in unclassified drug overdoses, the highest predictive accuracy is achieved using a statistical model (logistic regression or random forest ensemble) that includes decedent characteristics and contributing causes of death as predictors. These findings may facilitate more accurate assessments of the national and local opioid epidemics and improve the response to the crisis.

Citation:
Boslett AJ, Denham A, Hill EL. 2020. Using contributing causes of death improves prediction of opioid involvement in unclassified drug overdoses in US death records, Addiction, https://doi.org/10.1111/add.14943.

Patient age may bias cardiac health care

Behavioral heuristics (mental shortcuts that simplify decision making) are common in medicine and can lead to cognitive biases that affect clinical decisions. Researchers supported by grants from the NIH Common Fund recently published research that investigated if cognitive biases play a role in cardiac care for older patients. Cognitive biases refer to a range of systematic errors in human decision-making stemming from the tendency to use mental shortcuts. One example of cognitive bias is left-digit bias, which is the tendency to categorize continuous variables based on the left-most numeric digit. This type of bias explains why items are often priced at $9.99 as opposed to $10.00. This bias may affect treatment decisions. For example, a patient who is has recently turned 80 years of age may be perceived by doctors as being at greater risk for complications than a patient who is 79 years and 50 weeks of age. The doctor’s cognitive bias leads them to discretely categorize patients as being “in their 80s” rather than “in their 70s” which could lead to a different, less aggressive treatment. In this study the researchers wanted to determine if this type of bias was influencing treatment decisions for patients who were admitted with acute myocardial infarction (AMI) in the 2 weeks before their 80th birthday as compared with those who were admitted in the 2 weeks after their 80th birthday. They focused on the coronary-artery bypass grafting (CABG) procedure, since it is a complex procedure that may improve long-term survival; however, it has a risk of short-term complications and a long recovery period, a tradeoff that may lead doctors to treat patients they perceive to be at a higher risk for complications more conservatively.

The researchers used data on Medicare beneficiaries from 2006 through 2012 to evaluate how frequently inpatient CABG was performed in patients who were admitted with in the 2 weeks before their 80th birthday (n = 4,426) as compared with those who were admitted in the 2 weeks after their 80th birthday (n = 5,036). The baseline analysis was a comparison of primary (CABG performed or not) and secondary (mortality following hospital admission) outcomes between individuals admitted with AMI in the 2 weeks before compared to 2 weeks after their 80th birthday. This statistical comparison relied on the assumption that patients admitted with AMI in the 2-week window surrounding their birthdays were essentially identical, except for any influence on physician behavior of patient age being heuristically categorized according to the leftmost integer. To test this assumption, they first compared the characteristics of patients admitted in the two weeks before versus after their 80th birthday, then performed unadjusted comparisons of outcomes, and then estimated a hospitalization-level multivariable linear regression model.

Patients with AMI who were admitted in the 2 weeks after their 80th birthday were similar to those admitted before their 80th birthday with regard to various baseline characteristics (race/ethnicity, sex, chronic conditions, Medicare status, disability). However, those admitted after their 80th birthday were significantly less likely to undergo CABG than those admitted before their 80th birthday (5.3% versus 7.0%, respectively). There was no corresponding difference for those that were admitted with 2 weeks before their 77th through 79th or 81st through 83rd birthdays. The adjusted 30-day mortality rate after hospitalization for AMI was 17.7% among patients admitted before their 80th birthday and 19.8% among those admitted after their 80th birthday. These results are consistent with the occurrence of left-digit bias in clinical decision-making for older patients receiving cardiac care following an AMI. Awareness of these cognitive biases can help to reduce them in clinical decision-making as well as the development risk assessment tools that rely on more objective factors.

Citation:
Olenski AR, Zimerman A, Coussens S, Jena AB. 2020. Behavioral Heuristics in Coronary-Artery Bypass Graft Surgery. N Engl J Med, 382:778-779 DOI: 10.1056/NEJMc1911289.