Mediation effects that emulate a target randomised trial: Simulation-based evaluation of ill-defined interventions on multiple mediators

Decomposition analysis to assess intervention strategies for reducing health-related exposure disparities in Michigan public drinking water systems

OBJECTIVE

Contamination in U.S. public drinking water systems (PWS) is estimated to cause millions of illnesses and billions of dollars in medical expenditures annually. Few prior studies have explored intervention strategies, including environmental enforcement, to reduce estimated health-related exposure disparities (exposure disparity) in PWS, which are driven partially by socioeconomic status (SES), racism, and PWS characteristics.

METHOD

This study used a longitudinal measurement method to estimate the annual health-related exposure level (health level) of each PWS in Michigan, based on data from the Enforcement and Compliance Online (ECHO) and U.S. Census Bureau databases. Using a decomposition model with four strategies, we analyzed how eliminating disparities in SES, proportion minority, environmental enforcement, and PWS characteristics across communities would affect adjusted exposure disparities.

RESULTS

This study found that adjusted race- and poverty-based exposure disparities have existed since the 1980s but might have decreased in the last one or two decades. PWS characteristics strongly impacted the crude and adjusted exposure disparity. Environmental enforcement, although less effective in minority-concentrated communities, reduced the adjusted race-based exposure disparity by 10%-20% in the 1980s, 8% in the 1990s, and 0.012% in the 2010s. Equalizing the poverty rate distribution reduced the adjusted race-based exposure disparity by 0.72% in the 1980s and 6.8% in the 2010s. However, equalizing racial and ethnic composition distribution increased the adjusted poverty-based exposure disparity in the 2000s.

CONCLUSION

These findings indicate that economically disadvantaged or minority-concentrated communities in Michigan disproportionately suffer from poorer PWS quality. Enhanced environmental enforcement, increased household income, PWS investment, and other actions are needed to address these exposure disparities effectively.

Do age at diagnosis, tumour thickness and tumour site explain sex differences in melanoma survival? A causal mediation analysis using cancer registry data

Women diagnosed with melanoma have better survival than men, but little is known about potential intervention targets to reduce this survival gap by sex. We conducted a population-based study using Victorian Cancer Registry data including 5833 women and 6780 men aged 15 to 70 years when diagnosed with first primary melanoma between 2007 and 2015. Deaths to the end of 2020 were identified through linkage to the Victorian and national death registries. We estimated the effect of age at diagnosis, tumour thickness and tumour site on reducing the melanoma-specific survival gap by sex (ie, interventional indirect effects [IIEs]) on risk difference (RD) scale. Compared to women, there were 211 (95% CI: 145-278) additional deaths per 10 000 in men within 5 years following diagnosis. We estimated that 44% of this gap would be reduced by a hypothetical intervention shifting the distribution of melanoma thickness in men to be the same as that observed for women (IIEthickness RD 93 [95% CI: 75-118] per 10 000) and 20% by an intervention on tumour site (head and neck/trunk vs upper limb/lower limb; IIEsite RD 42 [95% CI: 15-72] per 10 000), while an intervention on age at diagnosis would have a negligible effect. Tumour thickness, tumour site and age at diagnosis mediated 65% of the effect of sex on 5-year melanoma survival in Victoria. Of these factors, tumour thickness had the most considerable mediating effect, suggesting that effective promotion of earlier detection of melanoma in men could potentially nearly halve the gap in melanoma-specific survival by sex.

Mediation Analyses of the Mechanisms by Which Socioeconomic Status, Comorbidity, Stroke Severity, and Acute Care Influence Stroke Outcome

BACKGROUND AND OBJECTIVES

Low socioeconomic status (SES) is associated with increased risk of death and disability after stroke, but interventional targets to minimize disparities remain unclear. We aim to assess the extent to which SES-based disparities in the association between low SES and death and dependency at 3 months after stroke could be eliminated by offsetting differences in comorbidity, stroke severity, and acute care.

METHODS

This nationwide register-based cohort study included all 72 hospitals caring for patients with acute stroke in Sweden. All patients registered with an acute ischemic stroke in the Swedish Stroke Register in 2015-2016 who were independent in activities of daily living (ADL) during stroke were included. Data on survival and SES the year before stroke were retrieved by cross-linkage with other national registers. SES was defined by education and income and categorized into low, mid, and high. Causal mediation analysis was used to study the absolute risk of death and ADL dependency at 3 months depending on SES and to what extent hypothetical interventions on comorbidities, stroke severity, and acute care would equalize outcomes.

RESULTS

Of the 25,846 patients in the study, 6,798 (26.3%) were dead or ADL dependent 3 months after stroke. Adjusted for sex and age, low SES was associated with an increased absolute risk of 5.4% (95% CI 3.9%-6.9%; p < 0.001) compared with mid SES and 10.1% (95% CI 8.1%-12.2%; p < 0.001) compared with high SES. Intervening to shift the distribution of all mediators among patients with low SES to those of the more privileged groups would result in absolute reductions of these effects by 2.2% (95% CI 1.2%-3.2%; p < 0.001) and 4.0% (95% CI 2.6%-5.5%; p < 0.001), respectively, with the largest reduction accomplished by equalizing stroke severity.

DISCUSSION

Low SES patients have substantially increased risks of death and ADL dependency 3 months after stroke compared with more privileged patient groups. This study suggests that if we could intervene to equalize SES-related differences in the distributions of comorbidity, acute care, and stroke severity, up to 40 of every 1,000 patients with low SES could be prevented from dying or becoming ADL dependent.

Interventional Effects Analysis of Dating Violence and Sexual Assault Victimization in LGBTQ + Adolescents: Quantifying the Roles of Inequities in School and Family Factors

Lesbian, gay, bisexual, transgender/nonbinary, and queer (LGBTQ +) adolescents experience considerable disparities in dating violence and sexual assault victimization relative to heterosexual and cisgender peers. These disparities may be driven in part by the disruptive effects of heterosexism and cissexism on school-based and family relationships. To quantify the potential roles of these processes and identify priorities for prevention efforts, we estimated the extent to which dating violence and sexual assault victimization in LGBTQ + adolescents could be reduced by eliminating sexual orientation and gender modality inequities in school adult support, bullying victimization, and family adversity. We analyzed data from a cross-sectional, population-based survey of high school students in Dane County, Wisconsin (N = 15,467; 13% sexual minority; 4% transgender/nonbinary; 72% White) using interventional effects analysis, adjusting for grade, race/ethnicity, and family financial status. We found that eliminating inequities in bullying victimization and family adversity could significantly reduce dating violence and sexual assault victimization in LGBTQ + adolescents, particularly sexual minority cisgender girls and transgender/nonbinary adolescents. For instance, eliminating gender modality inequities in family adversity could reduce sexual assault victimization in transgender/nonbinary adolescents by 2.4 percentage points, representing 27% of the existing sexual assault victimization disparity between transgender/nonbinary and cisgender adolescents (P < 0.001). Results suggest that dating violence and sexual assault victimization in LGBTQ + adolescents could be meaningfully reduced by policies and practices addressing anti-LGBTQ + bullying as well as heterosexism- and cissexism-related stress in LGBTQ + adolescents' families.

Mediation Analysis to Untangle Opposing Associations of High-Dose Docosahexaenoic Acid With IQ and Bronchopulmonary Dysplasia in Children Born Preterm

Importance

High-dose omega-3 docosahexaenoic acid (DHA) supplementation of children born at less than 29 weeks' gestation has been shown to improve IQ despite increasing the risk of bronchopulmonary dysplasia (BPD). Given that BPD is associated with poorer cognitive outcomes, it is unclear whether the increased risk of BPD with DHA supplementation is associated with decreased benefit to IQ.

Objective

To investigate whether the increased risk of BPD with DHA supplementation was associated with diminished IQ benefit.

Design, Setting, and Participants

This cohort study used data collected from a multicenter, blinded, randomized controlled trial of DHA supplementation in children born at less than 29 weeks' gestation. Participants were recruited from 2012 to 2015 and followed up until 5 years' corrected age. Data were analyzed from November 2022 to February 2023.

Interventions

Enteral DHA emulsion (60 mg/kg/d, to match the estimated in-utero requirement) or a control emulsion from the first 3 days of enteral feeds until 36 weeks' postmenstrual age or discharge home.

Main Outcomes and Measures

Physiological BPD was assessed at 36 weeks' postmenstrual age. IQ was assessed at 5 years' corrected age using the Wechsler Preschool and Primary Scale of Intelligence, 4th Edition; children from the 5 highest-recruiting Australian hospitals were assessed. The total effect of DHA supplementation on IQ was divided into direct and indirect effects using mediation analysis, with BPD as the presumed mediating variable.

Results

Among 656 surviving children from hospitals involved in IQ follow-up (mean [SD] gestational age at birth, 26.8 [1.4] weeks; 346 males [52.7%]), there were 323 children with DHA supplementation and 333 children in the control group. Mean IQ was 3.45 points (95% CI, 0.38 to 6.53 points) higher in the DHA group than the control group, despite an increase in the risk of BPD (160 children [49.7%] vs 143 children [42.8%] with BPD). The indirect effect of DHA on IQ via BPD was not statistically significant (-0.17 points; 95% CI, -0.62 to 0.13 points), with most of the effect of DHA on IQ occurring independently of BPD (direct effect = 3.62 points; 95% CI, 0.55 to 6.81 points).

Conclusions and Relevance

This study found that associations of DHA with BPD and IQ were largely independent. This finding suggests that if clinicians supplement children born preterm with high-dose DHA, any resulting increase in BPD risk would not be associated with meaningful reductions in the IQ benefit.

Complete effect decomposition for an arbitrary number of multiple ordered mediators with time-varying confounders: A method for generalized causal multi-mediation analysis

Causal mediation analysis is advantageous for mechanism investigation. In settings with multiple causally ordered mediators, path-specific effects have been introduced to specify the effects of certain combinations of mediators. However, most path-specific effects are unidentifiable. An interventional analog of path-specific effects is adapted to address the non-identifiability problem. Moreover, previous studies only focused on cases with two or three mediators due to the complexity of the mediation formula in a large number of mediators. In this study, we provide a generalized definition of traditional path-specific effects and interventional path-specific effects with a recursive formula, along with the required assumptions for nonparametric identification. Subsequently, a general approach is developed with an arbitrary number of multiple ordered mediators and with time-varying confounders. All methods and software proposed in this study contribute to comprehensively decomposing a causal effect confirmed by data science and help disentangling causal mechanisms in the presence of complicated causal structures among multiple mediators.

Impact of early intervention on the population prevalence of common mental disorders: 20-year prospective study

BACKGROUND

The potential for early interventions to reduce the later prevalence of common mental disorders (CMD) first experienced in adolescence is unclear.

AIMS

To examine the course of CMD and evaluate the extent to which the prevalence of CMD could be reduced by preventing adolescent CMD, or by intervening to change four young adult processes, between the ages of 20 and 29 years, that could be mediating the link between adolescent and adult disorder.

METHOD

This was a prospective cohort study of 1923 Australian participants assessed repeatedly from adolescence (wave 1, mean age 14 years) to adulthood (wave 10, mean age 35 years). Causal mediation analysis was undertaken to evaluate the extent to which the prevalence of CMD at age 35 years in those with adolescent CMD could be reduced by either preventing adolescent CMD, or by intervening on four young adult mediating processes: the occurrence of young adult CMD, frequent cannabis use, parenting a child by age 24 years, and engagement in higher education and employment.

RESULTS

At age 35, 19.2% of participants reported CMD; a quarter of these participants experienced CMD during both adolescence and young adulthood. In total, 49% of those with CMD during both adolescence and young adulthood went on to report CMD at age 35 years. Preventing adolescent CMD reduced the population prevalence at age 35 years by 3.9%. Intervening on all four young adult processes among those with adolescent CMD, reduced this prevalence by 1.6%.

CONCLUSIONS

In this Australian cohort, a large proportion of adolescent CMD resolved by adulthood, and by age 35 years, the largest proportion of CMD emerged among individuals without prior CMD. Time-limited, early intervention in those with earlier adolescent disorder is unlikely to substantially reduce the prevalence of CMD in midlife.

Socioeconomic status and stroke severity: Understanding indirect effects via risk factors and stroke prevention using innovative statistical methods for mediation analysis

Background

Those with low socioeconomic status have an increased risk of stroke, more severe strokes, reduced access to treatment, and more adverse outcomes after stroke. The question is why these differences are present. In this study we investigate to which extent the association between low socioeconomic status and stroke severity can be explained by differences in risk factors and stroke prevention drugs.

Methods

The study included 86 316 patients registered with an ischemic stroke in the Swedish Stroke Register (Riksstroke) 2012–2016. Data on socioeconomic status was retrieved from the Longitudinal integrated database for health insurance and labour market studies (LISA) by individual linkage. We used education level as proxy for socioeconomic status, with primary school education classified as low education. Stroke severity was measured using the Reaction Level Scale, with values above 1 classified as severe strokes. To investigate the pathways via risk factors and stroke prevention drugs we performed a mediation analysis estimating indirect and direct effects.

Results

Low education was associated with an excess risk of a severe stroke compared to mid/high education (absolute risk difference 1.4%, 95% CI: 1.0%-1.8%), adjusting for confounders. Of this association 28.5% was an indirect effect via risk factors (absolute risk difference 0.4%, 95% CI: 0.3%-0.5%), while the indirect effect via stroke prevention drugs was negligible.

Conclusion

Almost one third of the association between low education and severe stroke was explained by risk factors, and clinical effort should be taken to reduce these risk factors to decrease stroke severity among those with low socioeconomic status.

Mechanisms for the sex-specific effect of H. pylori on risk of gastroesophageal reflux disease and Barrett's oesophagus

BACKGROUND

Mechanisms for how Helicobacter pylori infection affects risk of gastroesophageal reflux disease (GERD) and Barrett's oesophagus (BE) are incompletely understood and might differ by sex.

METHODS

In a case-control study nested in the Melbourne Collaborative Cohort Study with 425 GERD cases and 169 BE cases (identified at 2007-10 follow-up), we estimated sex-specific odds ratios for participants who were H. pylori seronegative versus seropositive at baseline (1990-94). To explore possible mechanisms, we 1) compared patterns of H. pylori-induced gastritis by sex using serum pepsinogen-I and gastrin-17 data and 2) quantified the effect of H. pylori seronegativity on BE mediated by GERD using causal mediation analysis.

RESULTS

For men, H. pylori seronegativity was associated with 1.69-fold (CI:1.03-2.75) and 2.28-fold (CI:1.27-4.12) higher odds of GERD and BE, respectively. No association was observed for women. H. pylori-induced atrophic antral gastritis was more common in men (68%) than in women (56%; p=0.015). For men, 5 of the 15 per 1,000 excess BE risk from being seronegative was mediated by GERD.

CONCLUSIONS

Men, but not women, who were H. pylori seronegative had increased risks of GERD and BE. A possible explanation might be sex-differences in patterns of H. pylori-induced atrophic antral gastritis, which could lead to less erosive reflux for men. Evidence of GERD mediating the effect of H. pylori on BE risk among men supports this proposed mechanism.

IMPACT

The findings highlight the importance of investigating sex differences in the effect of H. pylori on risk of GERD and BE in future studies.

Path-specific effects in the presence of a survival outcome and causally ordered multiple mediators with application to genomic data

Causal multimediation analysis (i.e. the causal mediation analysis with multiple mediators) is critical for understanding the effectiveness of interventions, especially in medical research. Deriving the path-specific effects of exposure on the outcome through a set of mediators can provide detail about the causal mechanism of interest However, existing models are usually restricted to partial decomposition, which can only be used to evaluate the cumulative effect of several paths. In genetics studies, partial decomposition fails to reflect the real causal effects mediated by genes, especially in complex gene regulatory networks. Moreover, because of the lack of a generalized identification procedure, the current multimediation analysis cannot be applied to the estimation of path-specific effects for any number of mediators. In this study, we derive the interventional analogs of path-specific effect for complete decomposition to address the difficulty of nonidentifiability. On the basis of two survival models of the outcome, we derive the generalized analytic forms for interventional analogs of path-specific effects by assuming the normal distributions of mediators. We apply the new methodology to investigate the causal mechanism of signature genes in lung cancer based on the cell cycle pathway, and the results clarify the gene pathway in cancer.

Adiposity and breast, endometrial, and colorectal cancer risk in postmenopausal women: Quantification of the mediating effects of leptin, C-reactive protein, fasting insulin, and estradiol

BACKGROUND

Mechanisms underlying the adiposity-cancer relationship are incompletely understood. We quantified the mediating roles of C-reactive protein (CRP), leptin, fasting insulin, and estradiol in the effect of adiposity on estrogen receptor (ER)-positive breast, endometrial, and colorectal cancer risk in postmenopausal women.

METHODS

We used a case-cohort study within the Women's Health Initiative Observational Study, analyzed as a cumulative sampling case-control study. The study included 188 breast cancer cases, 98 endometrial cancer cases, 193 colorectal cancer cases, and 285 controls. Interventional indirect and direct effects on the risk ratio (RR) scale were estimated using causal mediation analysis.

RESULTS

For breast cancer, the total effect RR for BMI ≥30 versus ≥18.5-<25 kg/m2 was 1.87 (95%CI,1.11-3.13). The indirect effect RRs were 1.38 (0.79-2.33) through leptin and CRP, 1.58 (1.17-2.43) through insulin, and 1.11 (0.98-1.30) through estradiol. The direct effect RR was 0.82 (0.39-1.68). For endometrial cancer, the total effect RR was 2.12 (1.12-4.00). The indirect effect RRs were 1.72 (0.85-3.98) through leptin and CRP, 1.42 (0.96-2.26) through insulin, and 1.24 (1.03-1.65) through estradiol. The direct effect RR was 0.70 (0.23-2.04). For colorectal cancer, the total effect RR was 1.70 (1.03-2.79). The indirect effect RRs were 1.04 (0.61-1.72) through leptin and CRP, 1.36 (1.00-1.88) through insulin, and 1.02 (0.88-1.17) through estradiol. The direct effect RR was 1.16 (0.58-2.43).

CONCLUSION

Leptin, CRP, fasting insulin, and estradiol appear to mediate the effect of high BMI on cancer risk to different extents, with likely varying degrees of importance between cancers. These insights might be important in developing interventions to modify obesity-associated cancer risk in postmenopausal women.

Preventing postnatal depression: a causal mediation analysis of a 20-year preconception cohort

Postnatal depression (PND) is common and predicts a range of adverse maternal and offspring outcomes. PND rates are highest among women with persistent mental health problems before pregnancy, and antenatal healthcare provides ideal opportunity to intervene. We examined antenatal perceived social support as a potential intervention target in preventing PND symptoms among women with prior mental health problems. A total of 398 Australian women (600 pregnancies) were assessed repeatedly for mental health problems before pregnancy (ages 14-29 years, 1992-2006), and again during pregnancy, two months postpartum and one year postpartum (2006-2014). Causal mediation analysis found that intervention on perceived antenatal social support has the potential to reduce rates of PND symptoms by up to 3% (from 15 to 12%) in women with persistent preconception symptoms. Supplementary analyses found that the role of low antenatal social support was independent of concurrent antenatal depressive symptoms. Combined, these two factors mediated up to more than half of the association between preconception mental health problems and PND symptoms. Trialling dual interventions on antenatal depressive symptoms and perceived social support represents one promising strategy to prevent PND in women with persistent preconception symptoms. Interventions promoting mental health before pregnancy may yield an even greater reduction in PND symptoms by disrupting a developmental cascade of risks via these and other pathways. This article is part of the theme issue 'Multidisciplinary perspectives on social support and maternal-child health'.