ABSTRACT Background Including romantic partners as supports in behavioral weight management programs targeting index participants could yield spillover effects, but few studies have explored the effects on partners. Methods Dyads included an index participant with BMI ≥ 27 kg/m 2 and a cohabitating partner with BMI ≥ 18 kg/m 2 randomized to a participant‐only ( N = 116) or partner‐assisted ( N = 115) intervention. All index participants received an evidence‐based group weight management program. Partners in the partner‐assisted group attended half of the sessions but were not required to make behavioral changes/lose weight. Linear mixed models compared changes in partner weight and self‐reported caloric intake between groups at 24 months. Results Most partners identified as male (63%) and White (88%). Mean baseline weight and BMI were 98 kg (SD 28.1) and 32 kg/m 2 (SD 7.8), respectively. At 24 months, partners in the partner‐assisted versus participant‐only group lost more weight (−2.5 kg vs. −0.4 kg; estimated mean difference: −2.0 kg [95% CI −3.5, −0.6]). There were no differences in caloric change (partner‐assisted: −367 kcal, participant‐only: −196 kcal; estimated mean difference: −171 kcal [95% CI −358, 17]). Conclusions Involving romantic partners is an efficient weight management approach that generates meaningful long‐term weight loss for cohabitating partners despite them not being the direct targets for weight loss. Trial Registration Clinicaltrials.gov NCT03801174

Does more schooling cause higher lifetime earnings? Social scientists have long sought to determine the economic returns to schooling given its importance to individual life chances and public policy. Prior estimates are limited by unobserved confounding in observational studies and the focus of popular quasi-experimental studies on increases in schooling at only one particular age. Genotyped data offer an opportunity to address unobserved confounding and to estimate the returns to an additional year of schooling at any age by using quasi-randomly assigned genetic variants related to education as instrumental variables [Mendelian randomization (MR)]. We analyze comprehensive Norwegian population registries with career-long earnings data and genotyped data from the Norwegian Mother, Father and Child Cohort Study (MoBa). We employ three identification strategies for triangulation: ordinary least squares (OLS) with covariate-adjustment, sibling and twin fixed-effects models, and MR. Estimated returns to schooling are 8.0% in MR (N = 109,800) and 6.3% in sibling-MR (N = 18,666). Extensive sensitivity analyses suggest that MR results are robust even to large potential violations of MR assumptions, including pleiotropy. MR estimates somewhat higher returns than OLS for the full population (5.9%, N = 1,255,604) and fixed-effects models for siblings (5.3%, N = 966,976) or monozygotic twins (3.2%, N = 2,630). The estimated internal rate of return to schooling exceeds opportunity costs of education as proxied by the market interest rate. The lifetime returns to schooling are positive and substantial across all models.

BACKGROUND:Children and adolescents with attention-deficit hyperactivity disorder (ADHD) have a higher likelihood of contact with child welfare services (CWS). Evidence on whether pharmacological treatment of ADHD reduces such contact is limited. AIMS:To estimate the causal effect of pharmacological treatment of ADHD on CWS contact. METHOD:In this quasi-experimental study, we used nationwide registry data covering all individuals aged 5-14 years and diagnosed with ADHD during 2009-2011 in Norway. We used linear probability models and instrument variable analyses to estimate the associations and causal effects of pharmacological treatment on CWS contact up to 4 years after diagnosis. As instrument variable analysis uses natural variation in treatment decisions between clinics as pseudo-randomisation, estimates inform effects for children and adolescents at the margin of treatment, i.e. patients whose treatment is more influenced by variation in treatment practice, e.g. due to less severe or atypical symptom presentation. RESULTS:A total of 5930 children and adolescents aged 5-14 years were diagnosed with ADHD between 2009 and 2011 (mean (s.d.) age 10.1 (2.4) years; 4380 males (73.9%)). Instrument variable analyses showed a reducing effect of pharmacological treatment on the use of supportive interventions by 11.9 percentage points (95% CI: -20.12, -3.80) and out-of-home-placement by 3.30 percentage points (95% CI: -6.44, -0.15) at 2-year follow-up. This corresponds to the numbers needed to treat estimates of 8 and 30, respectively. CONCLUSIONS:Pharmacological treatment of ADHD reduces CWS contact among children and adolescents at the margin of treatment, lowering the probability of receiving supportive interventions and out-of-home placements. Findings suggest that medication reduces behavioural symptoms, which may improve the family coping mechanism and reduces the need for CWS involvement. TRIAL REGISTRATION:ISRCTN11891971.

Social scientists often ask about the effect of increasing one's duration of exposure to a social context on one's outcomes, i.e. the overlap effect. Past studies adopted a unidimensional treatment effect framework to estimate the effect of overlap, imposing important restrictions. In this paper, we propose a new causal framework of multidimensional treatments where the overlap effects include both the duration and the content of overlap, under which, for instance, the grandparent overlap effect is defined as the union of all causal effects of a grandparent's observed and unobserved characteristics (i.e., the content) on the grandchild across their shared life course (i.e., the duration). The multidimensional framework allows for a more flexible and context rich approach to effect heterogeneity, where unobserved contextual characteristics play two roles as unobserved confounders and as integral components of overlap effects – overlap effects in this framework are not easily estimated with conventional fixed effects estimation. Hence, we develop a new cumulative fixed effects (CFE) approach that can estimate a range of interesting heterogeneous causal overlap effects from three-wave individual panel data. We show that the CFE approach is unbiased even in highly non-linear simulations, and we discuss assumptions and extensions.

We introduce a new nonparametric causal decomposition approach that identifies the mechanisms by which a treatment variable contributes to a group-based outcome disparity. Our approach distinguishes three mechanisms: group differences in: (1) treatment prevalence, (2) average treatment effects, and (3) selection into treatment based on individual-level treatment effects. Our approach reformulates classic Kitagawa-Blinder-Oaxaca decompositions in causal and nonparametric terms, complements causal mediation analysis by explaining group disparities instead of group effects, and isolates conceptually distinct mechanisms conflated in recent random equalization decompositions. In contrast to all prior approaches, our framework uniquely identifies differential selection into treatment as a novel disparity-generating mechanism. Our approach can be used for both the retrospective causal explanation of disparities and the prospective planning of interventions to change disparities. We present both an unconditional and a conditional decomposition, where the latter quantifies the contributions of the treatment within levels of certain covariates. We develop nonparametric estimators that are root n-consistent, asymptotically normal, semiparametrically efficient, and multiply robust. We apply our approach to analyze the mechanisms by which college graduation causally contributes to intergenerational income persistence (the disparity in adult income between the children of high-vs. lowincome parents). Empirically, we demonstrate a previously undiscovered role played by the new selection component in intergenerational income persistence.

Although behavioral weight-loss programs are efficacious for achieving clinically significant weight loss, adherence to such programs is variable. Positive reinforcement through small financial incentives has shown promise for increasing adherence to behaviors and clinical outcomes in some domains. Yet, for weight management, it is unknown whether to incent the behavior (calorie logging), the outcome (weight loss), or both. We report the rationale and design for a two-site clinical trial employing a three-phase, 2 × 2 factorial design to evaluate the impact of financial incentives for calorie logging and/or weekly weight loss. Phase I involved a 26-week incentivized weight-loss period with virtual group sessions every two weeks; Phase II involved a 26-week incentivized weight-loss maintenance period with three monthly group sessions and five individual counseling telephone calls; and Phase III involved a 26-week weight-loss maintenance period without incentives with three bimonthly calls. Participants were asked to record their caloric intake on a smartphone application and to weigh themselves regularly on a cellular scale. These data were processed each week to determine if participants qualified for a financial incentive ranging from $0 to $10. The primary outcome was the proportion of participants achieving clinically significant weight loss of at least 5 % at 26 weeks. Secondary outcomes were the same proportions measured at 52 and 78 weeks, and blood pressure and cardiovascular medication use at 78 weeks. Motivation for weight loss will be examined as a potential mediator of the treatment effect. Intervention cost and cost-effectiveness will be calculated to inform implementation.Clinicaltrials.gov registration: NCT04770909

Background Psychiatric comorbidity is frequent among persons with attention-deficit/hyperactivity disorder (ADHD). Whether pharmacological treatment of ADHD influences the incidence of psychiatric comorbidity is uncertain. Objective To investigate associations and causal relations between pharmacological treatment of ADHD and incidence of subsequent comorbid psychiatric diagnoses. Methods We employed registry data covering all individuals aged 5-18 years in Norway who were diagnosed with ADHD during 2009-2011 (n=8051), followed until 2020. We used linear probability models (LPM) and instrumental variable (IV) analyses to examine associations and causal effects, respectively, between pharmacological treatment and subsequent comorbidity. Findings From time of ADHD diagnosis to 9 years of follow-up, 63% of patients were registered with comorbid psychiatric disorders. For males, LPM showed associations between ADHD medication and several incident comorbidities, but strength and direction of associations and consistency over time varied. For females, no associations were statistically significant. IV analyses for selected categories isolating effects among patients 'on the margin of treatment' showed a protective effect for a category of stress-related disorders in females and for tic disorders in males for the first 2 and 3 years of pharmacological treatment, respectively. Conclusions Overall, LPM and IV analyses did not provide consistent or credible support for long-term effects of pharmacological treatment on later psychiatric comorbidity. However, IV results suggest that for patients on the margin of treatment, pharmacological treatment may initially reduce the incidence of certain categories of comorbid disorders. Clinical implications Clinicians working with persons with ADHD should monitor the effects of ADHD medication on later psychiatric comorbidity. Trial registration number ISRCTN11891971.

In causal mediation analysis, the natural direct and indirect effects (natural effects) are nonparametrically unidentifiable in the presence of treatment-induced confounding, which motivated the development of randomized interventional analogues (RIAs) of the natural effects. Being easier to identify, the RIAs are becoming widely used in practice. However, applied researchers often interpret RIA estimates as if they were the natural effects, even though the RIAs can be poor proxies for the natural effects. This calls for practical and theoretical guidance on when the RIAs differ from or coincide with the natural effects. We develop the first empirical test to detect the divergence between the natural effects and their RIAs under the weak assumptions sufficient for identifying the RIAs and illustrate the test using the Moving to Opportunity Study. We also provide new theoretical insights on the relationship between the natural effects and the RIAs both using a covariance formulation and from a structural equation perspective. This analysis also reveals previously undocumented connections between the natural effects, the RIAs, and estimands in instrumental variable analysis and Wilcoxon-Mann-Whitney tests.

Adverse health events within families can harm children's development, including their early literacy. Using data from a longitudinal Wisconsin birth cohort, we estimated the spillover effect of younger siblings' gestational ages on older siblings' kindergarten-level literacy. We sampled 20,014 sibling pairs born during 2007-2010 who took Phonological Awareness Literacy Screening-Kindergarten tests during 2012-2016. Exposures were gestational age (completed weeks), preterm birth (gestational age <37 weeks), and very preterm birth (gestational age <32 weeks). We used gain-score regression-a fixed effects strategy-to estimate spillover effect. A one-week increase in younger siblings' gestational age improved the older siblings' test score by 0.011 SD (95% confidence interval: 0.001, 0.021 SD). The estimated spillover effect was larger among siblings whose mothers reported having a high school diploma/equivalent only (0.024 SD; 95% CI: 0.004, 0.044 SD). The finding underscores the networked effects of one individual's early-life health shocks on their family members.

Little is known about the labor market consequences of becoming a grandparent. We estimate grandparenthood effects on labor supply and earnings using detailed multigenerational data from Danish population registers. Results show that the consequences of grandparenthood are unequally distributed and starkly patterned. Becoming a grandparent reduces hours worked and income, especially for grandmothers, more so when the grandchild is born to a daughter, and most when the grandmother's daughter gives birth as a teenager. Grandfathers also experience a reduction in hours worked (but not income) from their daughter's teen birth, but the reduction is much smaller than among grandmothers. The effects of a daughter's teen birth are further amplified for low-income grandmothers. Our results imply that childbearing has multigenerational consequences that are structured by gendered caregiving, the caregiving needs of the parent generation, and the delegating capacity of the grandparent generation.

Objective To discuss how methods to estimate heterogenous causal effects can be applied in Family Science and to supply empirical examples using the case of fatherhood and earnings. Background Many questions important to family scientists do not focus on one-size-fits-all average effects but rather on whether and how effects differ across groups. Recent methodological advances can assist this latter focus, offering new insights for theory and policy. Method Using Danish administrative data on all men who entered fatherhood 2005-2016 and on men of comparable age who did not, we focus on two types of heterogeneity in effects. First, effect heterogeneity across observed and unobserved covariates; second, treatment effect heterogeneity across the distribution of outcome variables. Results The fatherhood premium on annual labor income is, in fact, a fatherhood penalty on average and across most margins of heterogeneity. Substantial heterogeneity exists across observed and unobserved characteristics and across the distribution of labor market earnings, with results indicating larger penalties for lower earners and those least likely to become fathers. Conclusions Effect heterogeneity in Family Science holds great potential to inform policy and theory. However, causal interpretations always require assumptions, and researchers must be vigilant that the assumptions they make are warranted for each specific application.

Abstract Background Partner support is associated with better weight loss outcomes in observational studies, but randomized trials show mixed results for including partners. Unclear is whether teaching communication skills to couples will improve weight loss in a person attempting weight loss (index participant). Purpose To compare the efficacy of a partner-assisted intervention versus participant-only weight management program on 24-month weight loss. Methods This community-based study took place in Madison, WI. Index participants were eligible if they met obesity guideline criteria to receive weight loss counseling, were aged 18–74 years, lived with a partner, and had no medical contraindications to weight loss; partners were aged 18–74 years and not underweight. Couples were randomized 1:1 to a partner-assisted or participant-only intervention. Index participants in both arms received an evidence-based weight management program. In the partner-assisted arm, partners attended half of the intervention sessions, and couples were trained in communication skills. The primary outcome was index participant weight at 24 months, assessed by masked personnel; secondary outcomes were 24-month self-reported caloric intake and average daily steps assessed by an activity tracker. General linear mixed models were used to compare group differences in these outcomes following intent-to-treat principles. Results Among couples assigned to partner-assisted (n = 115) or participant-only intervention (n = 116), most index participants identified as female (67%) and non-Hispanic White (87%). Average baseline age was 47.27 years (SD 11.51 years) and weight was 106.55 kg (SD 19.41 kg). The estimated mean 24-month weight loss was similar in the partner-assisted (2.66 kg) and participant-only arms (2.89 kg) (estimated mean difference, 0.23 kg [95% CI, -1.58, 2.04 kg], p=0.80). There were no differences in 24-month average daily caloric intake (estimated mean difference 50 cal [95% CI: -233, 132 cal], p=0.59) or steps (estimated mean difference 806 steps [95% CI: -1675, 64 steps], p=0.07). The percentage of participants reporting an adverse event with at least possible attribution to the intervention did not differ by arm (partner-assisted: 9%, participant-only, 3%, p = 0.11). Conclusions Partner-assisted and individual weight management interventions led to similar outcomes in index participants. Trial registration Clinicaltrials.gov NCT03801174, January 11, 2019.

AbstractObjectiveExisting behavioral weight management interventions produce clinically meaningful weight loss. The onset of the COVID‐19 pandemic led to the quick transition of such interventions from in‐person to virtual platforms. This provided a unique opportunity to compare engagement and outcomes for an in‐person versus virtually delivered weight management intervention.MethodsA non‐randomized comparison of engagement and weight outcomes was performed between two cohorts who participated in a weight management intervention in person (N = 97) versus three who participated virtually via videoconference (N = 134). Various metrics of engagement were examined, including group class and individual phone call attendance and duration, and retention for weight assessments. Behavioral targets of daily caloric intake and step‐counts and the clinical weight outcome were explored.ResultsCohorts (mean [standard deviation] age 47.3 (11.5), 67.1% women: 86.8% White) that participated virtually attended more group sessions (p < 0.001) and had maintenance telephone calls that were of a longer duration (p < 0.001). No other engagement or weight outcomes significantly differed by delivery modality.ConclusionsVirtual weight management programs are promising and may generate similar outcomes to those delivered in‐person. Future research should seek to understand how best to promote and sustain engagement in virtual interventions.

Objective: Criminality rates are higher among persons with attention-de fi cit/hyperactivity disorder (ADHD), and evidence that medication reduces crime is limited. Medication rates between clinics vary widely even within universal health care systems, partly because of providers ' treatment preferences. We used this variation to estimate causal effects of pharmacological treatment of ADHD on 4 -year criminal outcomes. Method: We used Norwegian population-level registry data to identify all unique patients aged 10 to 18 years diagnosed with ADHD between 2009 and 2011 (n = 5,624), their use of ADHD medication, and subsequent criminal charges. An instrumental variable design, exploiting variation in provider preference for ADHD medication between clinics, was used to identify causal effects of ADHD medication on crime among patients on the margin of treatment, that is, patients who receive treatment because of their provider ' s preference. Results: Criminality was higher in patients with ADHD relative to the general population. Medication preference varied between clinics and strongly affected patients ' treatment. Instrumental variable analyses supported a protective effect of pharmacological treatment on violence-related and publicorder - related charges with numbers needed to treat of 14 and 8, respectively. There was no evidence for effects on drug-, traff i c-, sexual-, or propertyrelated charges. Conclusion: This is the fi rst study to demonstrate causal effects of pharmacological treatment of ADHD on some types of crimes in a populationbased natural experiment. Pharmacological treatment of ADHD reduced crime related to impulsive-reactive behavior in patients with ADHD on the margin of treatment. No effects were found on crimes requiring criminal intent, conspiracy, and planning. Study preregistration information: The ADHD controversy project: Long-term effects of ADHD medication; https://www.isrctn.com/; 11891971.

Importance Childhood exposure to mild traumatic brain injury (mTBI) is common. Individuals with a childhood history of mTBI experience more frequent criminal justice involvement in mid to late adolescence and adulthood. No study had been conducted to examine whether the link is causal or spurious. Objective To determine whether mTBI in childhood causes criminal justice involvement in mid to late adolescence. Design, Setting, and Participants This cohort study used population-based data for all children born between 1995 and 2000 in Denmark, with data linked to emergency department (ED) visits and hospitalizations before age 10 years and all criminal charges and convictions from ages 15 to 20 years. The exposure group contained all individuals diagnosed with mTBI before age 10 years without other intracranial or extracranial injuries; the comparison group was individuals not diagnosed with mTBI or intracranial or extracranial injuries. Sibling and twin fixed-effects models were used to evaluate the association after controlling for family-level confounding. Data were analyzed from May 2021 to July 2024. Exposures Mild TBI before age 10 years without other intracranial or extracranial injuries before or at the time of diagnosis. Main Outcomes and Measures Associations between mTBI before age 10 years and criminal charges and convictions from ages 15 to 20 for the entire study population and separately by sex at birth, controlling for additional covariates. Results The final analytic sample consisted of 343 027 individuals, 13 514 in the exposure group and 329 513 in the comparison group. Of the total sample, 166 455 (49%) were female and 176 572 were male (51%). A total of 326 191 participants (95%) had at least 1 parent with Danish citizenship, and 79 386 mothers (23%) held a college degree. There was a positive association between mTBI and criminal charges (odds ratio [OR], 1.26; 95% CI, 1.19-1.34) and convictions (OR, 1.24; 95% CI, 1.16-1.33). When controlling for family-level confounding, the associations became statistically insignificant and, in most models, greatly reduced. Results were robust across multiple model specifications. Conclusions and Relevance This study found that although mTBI in childhood was predictive of adolescent criminal justice involvement, there was no evidence that mTBI caused criminal charges or convictions.

Schools routinely employ seating charts to influence educational outcomes. Dependable evidence for the causal effects of seating charts on students' achievement levels and inequality, however, is scarce. We executed a large pre-registered field experiment to estimate causal peer effects on students' test scores and grades by randomizing the seating charts of 195 classrooms (N=3,365 students). We found that neither sitting next to a deskmate with higher prior achievement nor sitting next to a female deskmate affected learning outcomes on average. However, we also found that sitting next to the highest-achieving deskmates improved the educational outcomes of the lowest-achieving students; and sitting next to the lowest-achieving deskmates lowered the educational outcomes of the highest-achieving students. Therefore, compared to random seating charts, achievement-discordant seating charts would decrease inequality; whereas achievement concordant seating charts would increase inequality. We discuss policy implications.

OBJECTIVE:Criminality rates are higher among persons with attention-deficit/hyperactivity disorder (ADHD), and evidence that medication reduces crime is limited. Medication rates between clinics vary widely even within universal health care systems, partly because of providers' treatment preferences. We used this variation to estimate causal effects of pharmacological treatment of ADHD on 4-year criminal outcomes. METHOD:We used Norwegian population-level registry data to identify all unique patients aged 10 to 18 years diagnosed with ADHD between 2009 and 2011 (n = 5,624), their use of ADHD medication, and subsequent criminal charges. An instrumental variable design, exploiting variation in provider preference for ADHD medication between clinics, was used to identify causal effects of ADHD medication on crime among patients on the margin of treatment, that is, patients who receive treatment because of their provider's preference. RESULTS:Criminality was higher in patients with ADHD relative to the general population. Medication preference varied between clinics and strongly affected patients' treatment. Instrumental variable analyses supported a protective effect of pharmacological treatment on violence-related and public-order-related charges with numbers needed to treat of 14 and 8, respectively. There was no evidence for effects on drug-, traffic-, sexual-, or property-related charges. CONCLUSION:This is the first study to demonstrate causal effects of pharmacological treatment of ADHD on some types of crimes in a population-based natural experiment. Pharmacological treatment of ADHD reduced crime related to impulsive-reactive behavior in patients with ADHD on the margin of treatment. No effects were found on crimes requiring criminal intent, conspiracy, and planning. STUDY PREREGISTRATION INFORMATION:The ADHD controversy project: Long-term effects of ADHD medication; https://www.isrctn.com/; 11891971.