Search

Studies have highlighted the potential importance of modeling interactions for suicide attempt prediction. This case-cohort study identified risk factors for suicide attempts among persons with depression in Denmark using statistical approaches that do (random forests) or do not (least absolute shrinkage and selection operator regression [LASSO]) model interactions. Cases made a nonfatal suicide attempt (n  = 6032) between 1995 and 2015. The comparison subcohort was a 5% random sample of all persons in Denmark on January 1, 1995 (n  = 11 963). We used random forests and LASSO for sex-stratified prediction of suicide attempts from demographic variables, psychiatric and somatic diagnoses, and treatments. Poisonings, psychiatric disorders, and medications were important predictors for both sexes. Area under the receiver-operating characteristic curve (AUC) values were higher in LASSO models (in men, 0.85, 95% CI, 0.84-0.86; in women, 0.89, 95% C, 0.88-0.90) than random forests (in men, 0.76, 95% CI, 0.74-0.78; in women, 0.79, 95% CI = 0.78-0.81). Automatic detection of interactions via random forests did not result in better model performance than LASSO models that did not model interactions. Due to the complex nature of psychiatric comorbidity and suicide, modeling interactions may not always be the optimal statistical approach to enhancing suicide attempt prediction in high-risk samples. This article is part of a Special Collection on Mental Health. [ABSTRACT FROM AUTHOR]

Suicide rates in the United States have increased over the past 15 years, with substantial geographic variation in these increases; yet there have been few attempts to cluster counties by the magnitude of suicide rate changes according to intercept and slope or to identify the economic precursors of increases. We used vital statistics data and growth mixture models to identify clusters of counties by their magnitude of suicide growth from 2008 to 2020 and examined associations with county economic and labor indices. Our models identified 5 clusters, each differentiated by intercept and slope magnitude, with the highest-rate cluster (4% of counties) being observed mainly in sparsely populated areas in the West and Alaska, starting the time series at 25.4 suicides per 100,000 population, and exhibiting the steepest increase in slope (0.69/100,000/year). There was no cluster for which the suicide rate was stable or declining. Counties in the highest-rate cluster were more likely to have agricultural and service economies and less likely to have urban professional economies. Given the increased burden of suicide, with no clusters of counties improving over time, additional policy and prevention efforts are needed, particularly targeted at rural areas in the West. [ABSTRACT FROM AUTHOR]

In their seminal 2002 paper, "Causal Knowledge as a Prerequisite for Confounding Evaluation: An Application to Birth Defects Epidemiology," Hernán et al. (Am J Epidemiol. 2002;155(2):176–184) emphasized the importance of using theory rather than data to guide confounding control, focusing on colliders as variables that share characteristics with confounders but whose control may actually introduce bias into analyses. In this commentary, we propose that the importance of this paper stems from the connection the authors made between nonexchangeability as the ultimate source of bias and structural representations of bias using directed acyclic graphs. This provided both a unified approach to conceptualizing bias and a means of distinguishing between different sources of bias, particularly confounding and selection bias. Drawing on examples from the paper, we also highlight unresolved questions about the relationship between collider bias, selection bias, and generalizability and argue that causal knowledge is a prerequisite not only for identifying confounders but also for developing any hypothesis about potential sources of bias. [ABSTRACT FROM AUTHOR]

Epidemiologic studies in the United States routinely report a lower or equal prevalence of major depressive disorder (MDD) for Black people relative to White people. Within racial groups, individuals with greater life stressor exposure experience greater prevalence of MDD; however, between racial groups this pattern does not hold. Informed by theoretical and empirical literature seeking to explain this "Black-White depression paradox," we outline 2 proposed models for the relationships between racial group membership, life stressor exposure, and MDD: an effect modification model and an inconsistent mediator model. Either model could explain the paradoxical within- and between-racial group patterns of life stressor exposure and MDD. We empirically estimated associations under each of the proposed models using data from 26,960 self-identified Black and White participants in the National Epidemiologic Survey on Alcohol and Related Conditions III (United States, 2012–2013). Under the effect modification model, we estimated relative risk effect modification using parametric regression with a cross-product term, and under the inconsistent mediation model, we estimated interventional direct and indirect effects using targeted minimum loss-based estimation. We found evidence of inconsistent mediation (i.e. direct and indirect effects operating in opposite directions), suggesting a need for greater consideration of explanations for racial patterns in MDD that operate independent of life stressor exposure. This article is part of a Special Collection on Mental Health. [ABSTRACT FROM AUTHOR]

Within a school grade, children who are young for grade are at increased risk of psychiatric diagnoses, but the long-term implications remain understudied, and associations with students who delay or accelerate entry underexplored. We used Norwegian birth cohort records (birth years: 1967–1976, n  = 626,928) linked to records in midlife. On-time school entry was socially patterned; among those born in December, 23.0% of children in the lowest socioeconomic position (SEP) delayed school entry, compared with 12.2% among the highest SEP. Among those who started school on time, there was no evidence for long-term associations between birth month and psychiatric/behavioral disorders or mortality. Controlling for SEP and other confounders, delayed school entry was associated with increased risk of psychiatric disorders and mortality. Children with delayed school entry were 1.31 times more likely to die by suicide (95% confidence interval: 1.07, 1.61) by midlife, and 1.96 times more likely to die from drug-related death (95% confidence interval: 1.59, 2.40) by midlife than those born late in the year who started school on time. Associations with delayed school entry are likely due to selection, and results thus underscore that long-term health risks can be tracked early in life, including through school entry timing, and are highly socially patterned. [ABSTRACT FROM AUTHOR]

Objectives This study aims to examine age differences in the intensity of chronic pain among middle-aged and older adults, where intensity is measured on a scale differentiating between chronic pain that is often troubling and likely requires intervention versus more endurable sensations. We aim to explore whether individual health and national gross domestic product (GDP) explain these differences as well. Methods Cross-nationally harmonized data from 20 countries on self-reported intensity of chronic pain (0 = no, 1 = mild, 2 = moderate, 3 = severe) in 104,826 individuals aged 50+ observed in 2012–2013. Two-level hierarchical ordinal linear models with individuals nested within countries were used to isolate estimations from heterogeneity explained by methodological differences across single-country studies. Results Overall, mean participant age was 66.9 (SD = 9.9), 56.1% were women, and 41.9% of respondents reported any chronic pain. Chronic pain intensity rose sharply with age in some countries (e.g. Korea and Slovenia), but this association waned or reversed in other countries (e.g. the United States and Denmark). Cross-country variation and age differences in chronic pain were partly explained (85.5% and 35.8%, respectively) by individual-level health (especially arthritis), country-level wealth (as indicated by GDP per capita), and demographics. Discussion Chronic pain intensity is not an inevitable consequence of chronological age, but the consequence of potential selection effects and lower activity levels combined with individual-level health and country-level wealth. Our findings suggest further investigation of health conditions and country affluence settings as potential targets of medical and policy interventions aiming to prevent, reduce, or manage chronic pain among older patients and aging populations. [ABSTRACT FROM AUTHOR]

Background: Whether changes in stroke mortality are affected by age distribution and birth cohorts, and if the decline in stroke mortality exhibits heterogeneity by stroke type, remains uncertain.Methods: We undertook a sequential time series analysis to examine stroke mortality trends in the USA among people aged 18-84 years between 1975 and 2019 (n = 4 332 220). Trends were examined for overall stroke and by ischaemic and haemorrhagic subtypes. Mortality data were extracted from the US death files, and age-sex population data were extracted from US census. Age-standardized stroke mortality rates and incidence rate ratio (IRR) with 95% confidence interval [CI] were derived from Poisson regression models.Results: Age-standardized stroke mortality declined for females from 87.5 in 1975 to 30.9 per 100 000 in 2019 (IRR 0.27, 95% CI 0.26, 0.27; average annual decline -2.78%, 95% CI -2.79, -2.78). Among males, age-standardized mortality rate declined from 112.1 in 1975 to 38.7 per 100 000 in 2019 (RR 0.26, 95% CI 0.26, 0.27; average annual decline -2.80%, 95% CI -2.81, -2.79). Stroke mortality increased sharply with advancing age. Decline in stroke mortality was steeper for ischaemic than haemorrhagic strokes.Conclusions: Stroke mortality rates have substantially declined, more so for ischaemic than haemorrhagic strokes. [ABSTRACT FROM AUTHOR]

Other forms of selection bias will also be over-represented in the siblings discordant for income, depending on the reason for income discordance (i.e. depending on how much non-random experiences contribute to the income difference between siblings). Evaluating the causal impact of income and socioeconomic status on mental health and wellbeing has been critically important yet methodologically challenging in psychiatric epidemiology since the very origins of the field. In fact, most family income in Sariaslan I et al i .[1] was stable and shared within families, and those families (and therefore effects of their income on study outcomes) were not included in the analysis. Yet the authors interpret the change in income of $15 000 per year as the total effect of all family income on psychiatric disorders and violent crime arrests. [Extracted from the article]

Adolescent internalizing symptoms have increased since 2010, whereas adequate sleep has declined for several decades. It remains unclear how self-reported sleep attainment has affected internalizing-symptoms trends. Using 1991–2019 data from the Monitoring the Future Study (n  ~ 390,000), we estimated age-period-cohort effects in adolescent internalizing symptoms (e.g. loneliness, self-esteem, self-derogation, depressive affect) and the association with yearly prevalence of a survey-assessed, self-reported measure of attaining ≥7 hours of sleep most nights. We focused our main analysis on loneliness and used median odds ratios to measure variance in loneliness associated with period differences. We observed limited signals for cohort effects and modeled only period effects. The feeling of loneliness increased by 0.83% per year; adolescents in 2019 had 0.68 (95% CI: 0.49, 0.87) increased log odds of loneliness compared with the mean, which was consistent by race/ethnicity and parental education. Girls experienced steeper increases in loneliness than boys (P  < 0.0001). The period-effect median odds ratio for loneliness was 1.16 (variance = 0.09; 95% CI: 0.06, 0.17) before adjustment for self-reported frequency of getting ≥7 hours sleep versus 1.07 (variance = 0.02; 95% CI: 0.01, 0.03) after adjustment. Adolescents across cohorts are experiencing worsening internalizing symptoms. Self-reported frequency of <7 hours sleep partially explains increases in loneliness, indicating the need for feasibility trials to study the effect of increasing sleep attainment on internalizing symptoms. [ABSTRACT FROM AUTHOR]

Agent-based modeling and g-computation can both be used to estimate impacts of intervening on complex systems. We explored each modeling approach within an applied example: interventions to reduce posttraumatic stress disorder (PTSD). We used data from a cohort of 2,282 adults representative of the adult population of the New York City metropolitan area from 2002–2006, of whom 16.3% developed PTSD over their lifetimes. We built 4 models: g-computation, an agent-based model (ABM) with no between-agent interactions, an ABM with violent-interaction dynamics, and an ABM with neighborhood dynamics. Three interventions were tested: 1) reducing violent victimization by 37.2% (real-world reduction); 2) reducing violent victimization by100%; and 3) supplementing the income of 20% of lower-income participants. The g-computation model estimated population-level PTSD risk reductions of 0.12% (95% confidence interval (CI): −0.16, 0.29), 0.28% (95% CI: −0.30, 0.70), and 1.55% (95% CI: 0.40, 2.12), respectively. The ABM with no interactions replicated the findings from g-computation. Introduction of interaction dynamics modestly decreased estimated intervention effects (income-supplement risk reduction dropped to 1.47%), whereas introduction of neighborhood dynamics modestly increased effectiveness (income-supplement risk reduction increased to 1.58%). Compared with g-computation, agent-based modeling permitted deeper exploration of complex systems dynamics at the cost of further assumptions. [ABSTRACT FROM AUTHOR]

Background: Loneliness and social isolation are emerging public health challenges for aging populations.Methods: We followed N = 11 302 U.S. Health and Retirement Study participants aged 50-95 from 2006 to 2014 to measure persistence of experiences of loneliness and exposure to social isolation. We tested associations of longitudinal loneliness and social isolation phenotypes with disability, morbidity, mortality, and biological aging through 2018.Results: During follow-up, 18% of older adults met criteria for loneliness, with 6% meeting criteria at 2 or more follow-up assessments. For social isolation, these fractions were 21% and 8%. Health and Retirement Study participants who experienced loneliness and were exposed to social isolation were at increased risk for disease, disability, and mortality. Those experiencing persistent loneliness were at a 57% increased hazard of mortality compared to those who never experienced loneliness. For social isolation, the increase was 28%. Effect sizes were somewhat larger for counts of prevalent activity limitations and somewhat smaller for counts of prevalent chronic diseases. Covariate adjustment for socioeconomic and psychological risks attenuated but did not fully explain associations. Older adults who experienced loneliness and were exposed to social isolation also exhibited physiological indications of advanced biological aging (Cohen's d for persistent loneliness and social isolation = 0.26 and 0.21, respectively). For loneliness, but not social isolation, persistence was associated with increased risk.Conclusions: Deficits in social connectedness prevalent in a national sample of U.S. older adults were associated with morbidity, disability, and mortality and with more advanced biological aging. Bolstering social connectedness to interrupt experiences of loneliness may promote healthy aging. [ABSTRACT FROM AUTHOR]

The depression gap refers to higher rates of depression among women than men. Change in the depression gap over time might elucidate social causes of this disparity—such as unequal college attendance or employment status. We conducted a meta-regression analysis to estimate variation in the depression gap over time by age, accounting for potential sources of variation between studies. Electronic databases and bibliographies were searched for English-language studies from January 1980 through October 2019; 144 independent estimates from US-representative samples met selection criteria (n  = 813,189). The depression gap was summarized as prevalence ratios among studies using diagnostic instruments and as standardized mean differences among symptom-based studies. Primary study measures were baseline study year (range, 1982–2017) and age (age groups ranging, in years, from 10–59 and 60 or older). Compared with respondents aged ≥60 years, depression prevalence was greater among respondents aged 10–19 (prevalence ratio = 1.26, 95% confidence interval: 1.02, 1.56). Over time, the depression gap did not change among adults, but it increased among adolescents (age-by-time interaction prevalence ratio = 1.05, 95% confidence interval: 1.01, 1.08). Results were similar for symptom-based studies. The present study finds no evidence of a change in the depression gender gap for US adults; however, the gap increased among adolescents. Greater attention to factors driving this widening disparity in adolescent depression is needed. [ABSTRACT FROM AUTHOR]

The opioid overdose crisis is driven by an intersecting set of social, structural, and economic forces. Simulation models are a tool to help us understand and address thiscomplex, dynamic, and nonlinear social phenomenon. We conducted a systematic review of the literature on simulation models of opioid use and overdose up to September 2019. We extracted modeling types, target populations, interventions, and findings; created a database of model parameters used for model calibration; and evaluated study transparency and reproducibility. Of the 1,398 articles screened, we identified 88 eligible articles. The most frequent types of models were compartmental (36%), Markov (20%), system dynamics (16%), and agent-based models (16%). Intervention cost-effectiveness was evaluated in 40% of the studies, and 39% focused on services for people with opioid use disorder (OUD). In 61% of the eligible articles, authors discussed calibrating their models to empirical data, and in 31%, validation approaches used in the modeling process were discussed. From the 63 studies that provided model parameters, we extracted the data sources on opioid use, OUD, OUD treatment, cessation or relapse, emergency medical services, and death parameters. From this database, potential model inputs can be identified and models can be compared with prior work. Simulation models should be used to tackle key methodological challenges, including the potential for bias in the choice of parameter inputs, investment in model calibration and validation, and transparency in the assumptions and mechanics of simulation models to facilitate reproducibility. [ABSTRACT FROM AUTHOR]

Prescription drug monitoring programs (PDMPs) are a crucial component of federal and state governments' response to the opioid epidemic. Evidence about the effectiveness of PDMPs in reducing prescription opioid-related adverse outcomes is mixed. We conducted a systematic review to examine whether PDMP implementation within the United States is associated with changes in 4 prescription opioid-related outcome domains: opioid prescribing behaviors, opioid diversion and supply, opioid-related morbidity and substance-use disorders, and opioid-related deaths. We searched for eligible publications in Embase, Google Scholar, MEDLINE, and Web of Science. A total of 29 studies, published between 2009 and 2019, met the inclusion criteria. Of the 16 studies examining PDMPs and prescribing behaviors, 11 found that implementing PDMPs reduced prescribing behaviors. All 3 studies on opioid diversion and supply reported reductions in the examined outcomes. In the opioid-related morbidity and substance-use disorders domain, 7 of 8 studies found associations with prescription opioid-related outcomes. Four of 8 studies in the opioid-related deaths domain reported reduced mortality rates. Despite the mixed findings, emerging evidence supports that the implementation of state PDMPs reduces opioid prescriptions, opioid diversion and supply, and opioid-related morbidity and substance-use disorder outcomes. When PDMP characteristics were examined, mandatory access provisions were associated with reductions in prescribing behaviors, diversion outcomes, hospital admissions, substance-use disorders, and mortality rates. Inconsistencies in the evidence base across outcome domains are due to analytical approaches across studies and, to some extent, heterogeneities in PDMP policies implemented across states and over time. [ABSTRACT FROM AUTHOR]

Systems science models use computer-based algorithms to model dynamic interactions between study units within and across levels and are characterized by nonlinear and feedback processes. They are particularly valuable approaches that complement the traditional epidemiologic toolbox in cases in which real data are not available and in cases in which traditional epidemiologic methods are limited by issues such as interference, spatial dependence, and dynamic feedback processes. In this commentary, we propose 2 key contributions that systems models can make to epidemiology: 1) the ability to test assumptions about underlying mechanisms that give rise to population distributions of disease; and 2) help in identifying the types of interventions that have the greatest potential to reduce population rates of disease in the future or in new sites where they have not yet been implemented. We discuss central challenges in the application of systems science approaches in epidemiology, propose potential solutions, and predict future developments in the role that systems science can play in epidemiology. [ABSTRACT FROM AUTHOR]

Increasing alcohol outlet density is well-documented to be associated with increased alcohol use and problems, leading to the policy recommendation that limiting outlet density will decrease alcohol problems. Yet few studies of decreasing problematic outlets and outlet density have been conducted. We estimated the association between closing alcohol outlets and alcohol use and alcohol-related violence, using an agent-based model of the adult population in New York City. The model was calibrated according to the empirical distribution of the parameters across the city's population, including the density of on- and off-premise alcohol outlets. Interventions capped the alcohol outlet distribution at the 90th to the 50th percentiles of the New York City density, and closed 5% to 25% of outlets with the highest levels of violence. Capping density led to a lower population of light drinkers (42.2% at baseline vs. 38.1% at the 50th percentile), while heavy drinking increased slightly (12.0% at baseline vs. 12.5% at the 50th percentile). Alcohol-related homicides and nonfatal violence remained unchanged. Closing the most violent outlets was not associated with changes in alcohol use or related problems. Results suggest that focusing solely on closing alcohol outlets might not be an effective strategy to reduce alcohol-related problems. [ABSTRACT FROM AUTHOR]

Many studies have shown inverse associations between childhood adversity and intelligence, although most are based on small clinical samples and fail to account for the effects of multiple co-occurring adversities. Using data from the 2001-2004 National Comorbidity Survey Adolescent Supplement, a cross-sectional US population study of adolescents aged 13-18 years (n = 10,073), we examined the associations between 11 childhood adversities and intelligence, using targeted maximum likelihood estimation. Targeted maximum likelihood estimation incorporates machine learning to identify the relationships between exposures and outcomes without overfitting, including interactions and nonlinearity. The nonverbal score from the Kaufman Brief Intelligence Test was used as a standardized measure of fluid reasoning. Childhood adversities were grouped into deprivation and threat types based on recent conceptual models. Adjusted marginal mean differences compared the mean intelligence score if all adolescents experienced each adversity to the mean in the absence of the adversity. The largest associations were observed for deprivation-type experiences, including poverty and low parental education, which were related to reduced intelligence. Although lower in magnitude, threat events related to intelligence included physical abuse and witnessing domestic violence. Violence prevention and poverty-reduction measures would likely improve childhood cognitive outcomes. [ABSTRACT FROM AUTHOR]

Although arrest rates among juveniles have substantially decreased since the 1990s, US national trends in conduct problems are unknown. Population variation in conduct problems would imply changes in the social environment, which would include emergent or receding risk factors. In the present study, we separated age, period, and cohort effects on conduct problems using nationally representative surveys of 375,879 US students conducted annually (1991-2015). The summed score of 7 items measuring the frequency of conduct problems was the outcome. Conduct problems have decreased during the past 25 years among boys; the total amount of the decrease was approximately 0.4 standard deviations (P < 0.01), and by item prevalence, the total amount of the decrease was 8%-11%. Declines are best explained by period effects beginning approximately in 2008, and a declining cohort effect beginning among those born after 1992, which suggests not only declines in population levels, but more rapid declines among younger cohorts of boys. Trends were also consistent with age-period-cohort effects on evenings spent out, which suggest a possible mechanism. Conduct problems among girls were lower than boys and did not demonstrate trends across time. These changes may reflect the changing nature of adolescence toward less unsupervised interaction. [ABSTRACT FROM AUTHOR]

Studies have shown that combat-area deployment is associated with increases in alcohol use; however, studying the influence of deployment on alcohol use faces 2 complications. First, the military considers a confluence of factors before determining whether to deploy a service member, creating a nonignorable exposure and unbalanced comparison groups that inevitably complicate inference about the role of deployment itself. Second, regression analysis assumes that a single effect estimate can approximate the population's change in postdeployment alcohol use, which ignores previous studies that have documented that respondents tend to exhibit heterogeneous postdeployment drinking behaviors. Therefore, we used propensity score matching to balance baseline covariates for the 2 comparison groups (deployed and nondeployed), followed by a variable-oriented difference-in-differences approach to account for the confounding and a person-oriented approach using a latent growth mixture model to account for the heterogeneous response to deployment in this prospective cohort study of the US Army National Guard (2009-2014). We observed a nonsignificant increase in estimated monthly drinks in the first year after deployment that regressed to predeployment drinking levels 2 years after deployment. We found a 4-class model that fit these data best, suggesting that common regression analyses likely conceal substantial interindividual heterogeneity in postdeployment alcohol-use behaviors. [ABSTRACT FROM AUTHOR]

The potentially deleterious effects on offspring health of excess maternal stress in pregnancy are important to understand--both whether observed associations are causal and through what mechanisms their effects may exert an influence. In this issue of the Journal, Räikkönen et al. (Am J Epidemiol. 2017;185(5):317-328) provide an ingenious test of a potential pathway through which maternal stress may influence offspring development. Licorice consumption is known to disrupt the ability of the placental enzyme 11β-hydroxysteroid dehydrogenase type 2 to inactivate cortisol before it reaches the fetus, leading to higher levels of cortisol exposure. Higher levels of cortisol exposure are also hypothesized to underlie the mechanism through which maternal stress may disrupt fetal development. Thus licorice consumption may serve, in some ways, to mimic maternal stress. The authors report associations between heavy licorice consumption during pregnancy and a wide range of offspring outcomes, including changes in pubertal timing, intelligence quotient, and mental health. In our view, these results should be considered preliminary; more work needs to be completed to determine the relationship of prenatal licorice consumption to these outcomes. Nonetheless, these intriguing and suggestive results demonstrate that this line of work should be given high priority, and they set the stage for additional research moving forward. [ABSTRACT FROM AUTHOR]

Introduction: To better understand the impact of the state-level tobacco environment (eg, tobacco control policies, attitudes towards smoking) on cigarette smoking, we examined whether the relationship of state tobacco environments to smoking is modified by individual-level vulnerability factors.Methods: In a nationally representative sample of US adults (N = 34 638), past-year smoking and heavy smoking were examined. State-level tobacco environment was defined by tobacco-related control policies and attitudes, ranging from permissive to restrictive; individual vulnerability was defined by childhood maltreatment and/or parental substance problems. Additive interaction tested differences in state-level tobacco environment effects on smoking and heavy smoking by individual-level vulnerability.Results: Significant interactions (P values < .01) indicated that the state tobacco environment had the strongest relationship to smoking outcomes among individuals with greatest individual vulnerability. For example, among respondents with childhood maltreatment and parental substance problems, those in states with permissive tobacco environments had 13.3% greater prevalence of smoking than those in restrictive states. Among respondents with neither individual-level risk factor, those in permissive states had 2.8% greater prevalence than those in restrictive states (interaction P value = .0002).Conclusions: Further restricting states' smoking environments could help reduce the prevalence of smoking and heavy smoking, particularly among those at increased individual risk in the general population.Implications: This study shows that the protective effect of restrictive state-level tobacco environments on smoking or heavy smoking was stronger among those especially vulnerable due to individual-level risk factors (parental substance problems, childhood maltreatment). Thus, public health campaigns to influence attitudes towards smoking or legislation to strengthen tobacco control could have a broad effect, particularly impacting those with vulnerability to smoking, which may help decrease smoking prevalence and reduce the massive public health burden of tobacco-related morbidity and mortality. [ABSTRACT FROM AUTHOR]

Epidemiology has long been concerned with understanding the causes of health and disease states so that we can improve the health of populations. Despite broad agreement on this definition of the field, we continue to debate certain core goals of epidemiology: whether epidemiology is a pragmatic science or not, which methods constitute epidemiologic methods, and what our gold-standard thinking should be to understand causation. We suggest that recognizing epidemiology as the quantitative heart of population health science can push these tensions aside and allow us to focus our science on the health of populations and on the processes that shape that health. Seeing epidemiology as the core quantitative health science has implications for the questions we ask, how we organize ourselves as a field, and how we train the next generation of epidemiologists. [ABSTRACT FROM AUTHOR]

Although depression symptoms have been associated with type 2 diabetes mellitus (T2DM) among adults, little is known about the association of adolescent-onset depression and development of T2DM in young adulthood and whether the association differs by sex. We examined the association between high levels of depressive symptoms in adolescence and T2DM in adulthood in the National Longitudinal Study of Adolescent to Adult Health (n = 12,657). Adolescents completed the 20-item version of Center for Epidemiologic Studies Depression Scale during wave 1 (mean age, 16 years) and the 10-item version during follow-up (mean age, 29 years). A high level of depressive symptoms was defined as a score of 16 or higher on the 20-item version or 11 or higher on the 10-item version. T2DM was identified 13 years after baseline on the basis of either a glycated hemoglobin concentration of at least 6.5% or use of hypoglycemic medication (with or without insulin). Participants who reported taking insulin alone were classified as having type 1 diabetes mellitus and excluded. In models adjusted for demographic characteristics, women were at a higher risk of developing T2DM if they experienced high levels of depressive symptoms during both adolescence and adulthood (odds ratio = 1.96, 95% confidence interval: 1.23, 3.11) than were those who did not experience a high level of symptoms at either time point. No statistically significant associations were noted among men (odds ratio = 0.46, 95% confidence interval: 0.20, 1.05). [ABSTRACT FROM AUTHOR]

The number of students and disciplines requiring basic instruction in epidemiologic methods is growing. As a field, we now have a lexicon of epidemiologic terminology and particular methods that have developed and become canonical through the historical development of the field. Yet, many of our basic concepts remain elusive to some students, particularly those not pursuing a career in epidemiology. Further, disagreement and redundancy across basic terms limit their utility in teaching epidemiology. Many approaches to teaching epidemiology generally start with labeling key concepts and then move on to explain them. We submit that an approach grounded not in labels but in foundational concepts may offer a useful adjunct to introductory epidemiology education. We propose 7 foundational steps in conducting an epidemiologic study and provide examples of how these steps can be operationalized, using simple graphics that articulate how populations are defined, samples are selected, and individuals are followed to count cases. A reorganization of introductory epidemiology around core first principles may be an effective way forward for educating the next generation of public health scientists. [ABSTRACT FROM PUBLISHER]

Agent-based models (ABMs) have grown in popularity in epidemiologic applications, but the assumptions necessary for valid inference have only partially been articulated. In this issue, Murray et al. (Am J Epidemiol. 2017;186(2):131-142) provided a much-needed analysis of the consequence of some of these assumptions, comparing analysis using an ABM to a similar analysis using the parametric g-formula. In particular, their work focused on the biases that can arise in ABMs that use parameters drawn from distinct populations whose causal structures and baseline outcome risks differ. This demonstration of the quantitative issues that arise in transporting effects between populations has implications not only for ABMs but for all epidemiologic applications, because making use of epidemiologic results requires application beyond a study sample. Broadly, because health arises within complex, dynamic, and hierarchical systems, many research questions cannot be answered statistically without strong assumptions. It will require every tool in our store of methods to properly understand population dynamics if we wish to build an evidence base that is adequate for action. Murray et al.'s results provide insight into these assumptions that epidemiologists can use when selecting a modeling approach. [ABSTRACT FROM AUTHOR]

An estimated 80% of sexually active young women in the United States use hormonal contraceptives during their reproductive years. Associations between hormonal contraceptive use and mood disturbances remain understudied, despite the hypothesis that estrogen and progesterone play a role in mood problems. In this study, we used data from 6,654 sexually active nonpregnant women across 4 waves of the National Longitudinal Study of Adolescent Health (1994–2008), focusing on women aged 25–34 years. Women were asked about hormonal contraceptive use in the context of a current sexual partnership; thus, contraceptive users were compared with other sexually active women who were using either nonhormonal contraception or no contraception. Depressive symptoms were assessed with the Center for Epidemiologic Studies Depression Scale. At ages 25–34 years, hormonal contraceptive users had lower mean levels of concurrent depressive symptoms (β = −1.04, 95% confidence interval: −1.73, −0.35) and were less likely to report a past-year suicide attempt (odds ratio = 0.37, 95% confidence interval: 0.14, 0.95) than women using low-efficacy contraception or no contraception, in models adjusted for propensity scores for hormonal contraceptive use. Longitudinal analyses indicated that associations between hormonal contraception and depressive symptoms were stable. Hormonal contraception may reduce levels of depressive symptoms among young women. Systematic investigation of exogenous hormones as a potential preventive factor in psychiatric epidemiology is warranted. [ABSTRACT FROM PUBLISHER]

Background The incidence and prevalence of autism have dramatically increased over the last 20 years. Decomposition of autism incidence rates into age, period and cohort effects disentangle underlying domains of causal factors linked to time trends. We estimate an age-period-cohort effect model for autism diagnostic incidence overall and by level of functioning.Methods Data are drawn from sequential cohorts of all 6 501 262 individuals born in California from 1992 to 2003. Autism diagnoses from 1994 to 2005 were ascertained from the California Department of Development Services Client Development and Evaluation Report.Results Compared with those born in 1992, each successively younger cohort has significantly higher odds of an autism diagnosis than the previous cohort, controlling for age and period effects. For example, individuals born in 2003 have 16.6 times the odds of an autism diagnosis compared with those born in 1992 [95% confidence interval (CI) 7.8–35.3]. The cohort effect observed in these data is stronger for high than for low-functioning children with an autism diagnosis.Discussion Autism incidence in California exhibits a robust and linear positive cohort effect that is stronger among high-functioning children with an autism diagnosis. This finding indicates that the primary drivers of the increases in autism diagnoses must be factors that: (i) have increased linearly year-to-year; (ii) aggregate in birth cohorts; and (iii) are stronger among children with higher levels of functioning. [ABSTRACT FROM AUTHOR]

A substantial proportion of injuries worldwide are attributable to alcohol consumption, and US estimates indicate that the drinking patterns of racial/ethnic groups vary considerably. The authors reviewed evidence from 19 publications regarding racial/ethnic differences in overall alcohol-attributable injury as well as percent blood alcohol content positivity for injury deaths in the United States. They found that Native Americans evidence higher rates of alcohol-attributable motor vehicle crash fatality, suicide, and falls compared with other racial/ethnic groups; conversely, Asians evidence lower rates of alcohol-attributable injury than other racial/ethnic groups. The rate of alcohol positivity and intoxication among Hispanics is disproportionately high relative to estimates of alcohol use. Black subgroups also evidence higher rates of alcohol positivity than would be expected given estimates of alcohol use, including for alcohol positivity among drivers of fatally injured black children and homicide. These findings highlight the continued need for public health focus on Native American populations with respect to alcohol consumption and injury. Further, the disparity in alcohol-attributable injury mortality among black and Hispanic groups relative to their reported rates of alcohol consumption is an overlooked area of research. The authors review potential social determinants of racial/ethnic disparities in alcohol-attributable injuries and identify directions for further research on these patterns. [ABSTRACT FROM PUBLISHER]

Background: Lesbian, gay and bisexual (LGB) populations evidence higher rates of psychiatric disorders than heterosexuals, but most LGB individuals do not have mental-health problems. The present study examined risk modifiers at the social/contextual level that may protect LGB individuals from the development of psychiatric disorders.Methods: Data are drawn from Wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions (N = 34,653), a nationally representative study of non-institutionalized US adults. Risk variables included social isolation and economic adversity. High state-level concentration of same-sex couples, obtained from the US Census, was examined as a protective factor.Results: The past-year prevalence of major depression and generalized anxiety disorder was lower among LGB respondents living in states with higher concentrations of same-sex couples, compared with LGB respondents in states with lower concentrations. Additionally, the increased risk for mood and anxiety disorders among LGB individuals exposed to economic adversity and social isolation was evident only in states with low concentrations of same-sex couples. These interactions between the risk and protective factors were not found among heterosexuals, suggesting specificity of the effects to LGB individuals. Results were not attenuated after controlling for socio-demographic factors, state-level income inequality, state-level policies targeting LGBs and state-level attitudes towards LGB-relevant issues.Conclusions: These results provide evidence for the protective effect of social/contextual influences on the prevalence of psychiatric disorders in LGB individuals. Measures of the social environment should be incorporated into future research on the mental health of LGB populations. [ABSTRACT FROM AUTHOR]

A response from the authors of the article "Association of Hormonal Contraceptive Use With Reduced Levels of Depressive Symptoms: A National Study of Sexually Active Women in the United States," in the 2013 issue is presented.