Assessing causality in epidemiology: revisiting Bradford Hill to incorporate developments in causal thinking

Guide to Decomposition of Causal Effects Into Mediation, Interaction, and Direct Effects: Case Study on Aerobic Exercise and Parkinson Disease

Mediation analysis can be applied in medical research with the aim of understanding the pathways that operate between an exposure and its effects on an outcome. This method can help to improve our understanding of pathophysiologic mechanisms and may guide the choice of potential treatment strategies. Traditional mediation analysis decomposes the total effect of an intervention on the outcome into 2 effects: (1) an indirect effect, from exposure using a mediator to the outcome, and (2) a direct effect, directly from exposure to outcome. A limitation of this method is that it assumes no interaction between the exposure and the mediator, which can either lead to an over- or underestimation of clinically relevant effects. The "4-way decomposition" method has the advantage of overcoming this limitation. Specifically, the total effect of an exposure on the outcome is decomposed into 4 elements: (1) reference interaction (interaction only), (2) mediated interaction (mediation and interaction), (3) the pure indirect effect (mediation but not interaction), and (4) the direct effect (no mediation and no interaction). We provide a guide to select the most appropriate method to investigate and decompose any causal effect given the research question at hand. We explain the application of the 4-way decomposition and illustrate this with a real-world example of how aerobic exercise may influence motor function in persons with Parkinson disease.

Comparative Risk of Injury with Concurrent Use of Opioids and Skeletal Muscle Relaxants

Concurrent use of skeletal muscle relaxants (SMRs) and opioids has been linked to an increased risk of injury. However, it remains unclear whether the injury risks differ by specific SMR when combined with opioids. We conducted nine retrospective cohort studies within a US Medicaid population. Each cohort consisted exclusively of person-time exposed to both an SMR and one of the three most dispensed opioids-hydrocodone, oxycodone, and tramadol. Opioid users were further divided into three cohorts based on the initiation order of SMRs and opioids-synchronically triggered, opioid-triggered, and SMR-triggered. Within each cohort, we used Cox proportional hazard models to compare the injury rates for different SMRs compared to methocarbamol, adjusting for covariates. We identified 349,543, 139,458, and 218,967 concurrent users of SMRs with hydrocodone, oxycodone, and tramadol, respectively. In the oxycodone-SMR-triggered cohort, the adjusted hazard ratios (HRs) were 1.86 (95% CI, 1.23-2.82) for carisoprodol and 1.73 (1.09-2.73) for tizanidine. In the tramadol-synchronically triggered cohort, the adjusted HRs were 0.69 (0.49-0.97) for metaxalone and 0.62 (0.42-0.90) for tizanidine. In the tramadol-SMR-triggered cohort, the adjusted HRs were 1.51 (1.01-2.26) for baclofen and 1.48 (1.03-2.11) for cyclobenzaprine. All other HRs were statistically nonsignificant. In conclusion, the relative injury rate associated with different SMRs used concurrently with the three most dispensed opioids appears to vary depending on the specific opioid and the order of combination initiation. If confirmed by future studies, clinicians should consider the varying injury rates when prescribing SMRs to individuals using hydrocodone, oxycodone, and tramadol.

A practical guide to estimating treatment effects in patients with rheumatic diseases using real-world data

OBJECTIVE

Randomized controlled trials are considered the gold standard in study methodology. However, due to their study design and inclusion criteria, these studies may not capture the heterogeneity of real-world patient populations. In contrast, the lack of randomization and the presence of both measured and unmeasured confounding factors could bias the estimated treatment effect when using observational data. While causal inference methods allow for the estimation of treatment effects, their mathematical complexity may hinder their application in clinical research.

METHODS

We present a practical, nontechnical guide using a common statistical package (Stata) and a motivational simulated dataset that mirrors real-world observational data from patients with rheumatic diseases. We demonstrate regression analysis, regression adjustment, inverse-probability weighting, propensity score (PS) matching and two robust estimation methods.

RESULTS

Although the methods applied to control for confounding factors produced similar results, the commonly used one-to-one PS matching method could yield biased results if not thoroughly assessed.

CONCLUSION

The guide we propose aims to facilitate the use of readily available methods in a common statistical package. It may contribute to robust and transparent epidemiological and statistical methods, thereby enhancing effectiveness research using observational data in rheumatology.

Access to civil justice as a social determinant of health: a legal epidemiological cross-sectional study

BACKGROUND

Although it is widely acknowledged that access to civil justice (ATJ) is a key social determinant of health (SDOH), the existing literature lacks empirical evidence supporting ATJ as a SDOH for specific dimensions of health.

METHODS

A legal epidemiological, cross-sectional, postal survey was conducted on n = 908 randomly sampled participants in Hong Kong in March 2023. Data collected were perceptions of the civil justice system, health, and sociodemographics. Perceived ATJ was assessed using a modified version of the Inaccessibility of Justice scale (IOJ) and Perceived Inequality of Justice scale (PIJ), i.e. the "modified IOJ-PIJ", consisting of 12 of the original 13 items from both scales divided into two subdomains: "procedural fairness", and "outcome neutrality". For health data, quality of life was assessed using the Hong Kong version of the Abbreviated World Health Organization Quality of Life questionnaire (WHOQOL-BREF(HK)), psychological distress (including symptoms of anxiety and depression) was assessed using the four-Item Patient Health Questionnaire (PHQ-4), and having comorbidities was assessed using Sangha's Self-Administered Comorbidity Questionnaire (SCQ). Structural equation modelling (SEM) was used to investigate the relationships between perceived ATJ and the measured health outcomes.

RESULTS

SEM demonstrated that both subdomains for ATJ had significantly negative associations (B < 0; p < 0.05) with all quality-of-life subdomains, except for between outcome neutrality with social relationships; both subdomains for ATJ had significantly positive association (B > 0; p < 0.05) with both anxiety and depression; and, after adjusting for age, only "procedural fairness" had significantly positive association (B > 0; p < 0.05) with having comorbidities.

CONCLUSION

This study provided empirical evidence that ATJ is a SDOH for specific dimensions of health. The results of this study encourage laws, policies, and initiatives aimed at improving ATJ, as well as collaborative efforts from the legal and health sectors through health-justice partnerships, and from the broader community, to safeguard and promote public health by strengthening ATJ.

A blueprint for a new commercial driving epidemiology: An emerging paradigm grounded in integrative exposome and network epistemologies

Excess health and safety risks of commercial drivers are largely determined by, embedded in, or operate as complex, dynamic, and randomly determined systems with interacting parts. Yet, prevailing epidemiology is entrenched in narrow, deterministic, and static exposure-response frameworks along with ensuing inadequate data and limiting methods, thereby perpetuating an incomplete understanding of commercial drivers' health and safety risks. This paper is grounded in our ongoing research that conceptualizes health and safety challenges of working people as multilayered "wholes" of interacting work and nonwork factors, exemplified by complex-systems epistemologies. Building upon and expanding these assumptions, herein we: (a) discuss how insights from integrative exposome and network-science-based frameworks can enhance our understanding of commercial drivers' chronic disease and injury burden; (b) introduce the "working life exposome of commercial driving" (WLE-CD)-an array of multifactorial and interdependent work and nonwork exposures and associated biological responses that concurrently or sequentially impact commercial drivers' health and safety during and beyond their work tenure; (c) conceptualize commercial drivers' health and safety risks as multilayered networks centered on the WLE-CD and network relational patterns and topological properties-that is, arrangement, connections, and relationships among network components-that largely govern risk dynamics; and (d) elucidate how integrative exposome and network-science-based innovations can contribute to a more comprehensive understanding of commercial drivers' chronic disease and injury risk dynamics. Development, validation, and proliferation of this emerging discourse can move commercial driving epidemiology to the frontier of science with implications for policy, action, other working populations, and population health at large.

PM2.5, component cause of severe metabolically abnormal obesity: An in silico, observational and analytical study

Obesity is currently one of the most alarming pathological conditions due to the progressive increase in its prevalence. In the last decade, it has been associated with fine particulate matter suspended in the air (PM2.5). The purpose of this study was to explore the mechanistic interaction of PM2.5 with a high-fat diet (HFD) through the differential regulation of transcriptional signatures, aiming to identify the association of these particles with metabolically abnormal obesity. The research design was observational, using bioinformatic methods and an explanatory approach based on Rothman's causal model. We propose three new transcriptional signatures in murine adipose tissue. The sum of transcriptional differences between the group exposed to an HFD and PM2.5, compared to the control group, were 0.851, 0.265, and -0.047 (p > 0.05). The HFD group increased body mass by 20% with two positive biomarkers of metabolic impact. The group exposed to PM2.5 maintained a similar weight to the control group but exhibited three positive biomarkers. Enriched biological pathways (p < 0.05) included PPAR signaling, small molecule transport, adipogenesis genes, cytokine-cytokine receptor interaction, and HIF-1 signaling. Transcriptional regulation predictions revealed CpG islands and common transcription factors. We propose three new transcriptional signatures: FAT-PM2.5-CEJUS, FAT-PM2.5-UP, and FAT-PM2.5-DN, whose transcriptional regulation profile in adipocytes was statistically similar by dietary intake and HFD and exposure to PM2.5 in mice; suggesting a mechanistic interaction between both factors. However, HFD-exposed murines developed moderate metabolically abnormal obesity, and PM2.5-exposed murines developed severe abnormal metabolism without obesity. Therefore, in Rothman's terms, it is concluded that HFD is a sufficient cause of the development of obesity, and PM2.5 is a component cause of severe abnormal metabolism of obesity. These signatures would be integrated into a systemic biological process that would induce transcriptional regulation in trans, activating obesogenic biological pathways, restricting lipid mobilization pathways, decreasing adaptive thermogenesis and angiogenesis, and altering vascular tone thus inducing a severe metabolically abnormal obesity.

Building Learning Healthcare Systems for Critical Care Medicine

Learning healthcare systems are an evolving way of integrating informatics, analytics, and continuous improvement into daily practice in healthcare. This article discusses strategies to build learning healthcare systems for critical care medicine.

The association between prenatal cannabis use and congenital birth defects in offspring: A cumulative meta-analysis

OBJECTIVE

To examine the association between prenatal cannabis use and structural birth defects in exposed offspring.

METHODS

In line with the preregistered protocol (PROSPERO: CRD42022368623), we systematically searched PubMed/Medline, CINHAL, EMBASE, Web of Science, ProQuest, Psych-Info, and Google Scholar for published articles until 25 January 2024. The methodological quality of the included studies was appraised by the Newcastle-Ottawa Quality Assessment Scale (NOS). A meta-analysis was carried out to report the pooled effect estimates from the included studies. We further performed subgroup, leave-one-out sensitivity, and meta-regression analyses, which increased the robustness of our findings.

RESULTS

In this cumulative meta-analysis, thirty-six observational studies, consisting of 18 case-control and 18 cohort studies, with 230, 816 cases of birth defects and 18,049,013 controls (healthy babies) were included in the final analysis. We found that offspring exposed to maternal prenatal cannabis are at greater risks of a wide range of structural birth defects: cardiovascular/heart [OR = 2.35: 95 % CI 1.63 - 3.39], gastrointestinal [OR = 2.42: 95 % CI 1.61 - 3.64], central nervous system [OR = 2.87: 95 % CI 1.51 - 5.46], genitourinary [OR = 2.39: 95 % CI 1.11 - 5.17], and any (unclassified) birth defects [OR = 1.25: 95 % CI 1.12 - 1.41].

CONCLUSION

The findings from the current study suggest that maternal prenatal cannabis exposure is associated with a higher risk of different forms of structural birth defects in offspring. The findings underscore the significance of implementing preventive strategies, including enhanced preconception counselling, to address cannabis use during pregnancy and mitigate the risk of birth defects in offspring.

Assessing cardiovascular disease risk in women with a history of hypertensive disorders of pregnancy: A guidance paper for studies using administrative data

BACKGROUND

Hypertensive disorders of pregnancy (HDP) are a major cause of maternal morbidity and mortality, and their association with increased cardiovascular disease (CVD) risk represents a major public health concern. However, assessing CVD risk in women with a history of these conditions presents unique challenges, especially when studies are carried out using routinely collected data.

OBJECTIVES

To summarise and describe key challenges related to the design and conduct of administrative studies assessing CVD risk in women with a history of HDP and provide concrete recommendations for addressing them in future research.

METHODS

This is a methodological guidance paper.

RESULTS

Several conceptual and methodological factors related to the data-generating mechanism and study conceptualisation, design/data management and analysis, as well as the interpretation and reporting of study findings should be considered and addressed when designing and carrying out administrative studies on this topic. Researchers should develop an a priori conceptual framework within which the research question is articulated, important study variables are identified and their interrelationships are carefully considered.

CONCLUSIONS

To advance our understanding of CVD risk in women with a history of HDP, future studies should carefully consider and address the conceptual and methodological considerations outlined in this guidance paper. In highlighting these challenges, and providing specific recommendations for how to address them, our goal is to improve the quality of research carried out on this topic.

Advances in methodologies of negative controls: a scoping review

OBJECTIVES

Negative controls are considered an important tool to mitigate biases in observational studies. The aim of this scoping review was to summarize current methodologies of negative controls (both negative control exposure [NCE] and negative control outcome [NCO]).

STUDY DESIGN AND SETTING

We searched PubMed, Web of Science, Embase, and Cochrane Library (up to March 9, 2023) for articles on methodologies of negative controls. Two reviewers selected eligible studies and collected relevant data independently and in duplicate. We reported total numbers and percentages, and summarized methodologies narratively.

RESULTS

A total of 37 relevant methodological articles were included in our review. These publications covered NCE (n = 11, 29.8%), NCO (n = 13, 35.1%), or both (n = 13, 35.1%), with most focused on bias detection (n = 14, 37.8%), bias correction (n = 16, 43.3%), and P value or confidence interval (CI) calibration (n = 5, 13.5%). For the two remaining articles (5.4%), one discussed bias detection and P value or CI calibration and the other covered all the three functions. For bias detection, the existence of an association between the NCE (NCO) and outcome (exposure) variables of interest simply indicates that results may suffer from confounding bias, selection bias and/or information bias. For bias correction, however, the algorithms of negative control methods need more stringent assumptions such as rank preservation, monotonicity, and linearity.

CONCLUSION

Negative controls can be leveraged for bias detection, P value or CI calibration, and bias correction, among which bias correction has been the most studied methodologically. The current available methods need some stringent assumptions to detect or remove bias. More methodological research is needed to optimize the use of negative controls.

GRADE GUIDANCE 38: Updated guidance for rating up certainty of evidence due to a dose-response gradient

INTRODUCTION

This updated guidance from the Grading of Recommendations Assessment, Development, and Evaluation addresses rating up certainty of evidence due to a dose-response gradient (DRG) observed in synthesis of intervention and exposure studies.

STUDY DESIGN AND SETTING

This guidance was developed using iterative discussions and consensus in multiple meetings and was presented to attendees of the Grading of Recommendations Assessment, Development, and Evaluation Working Group meeting for feedback in November 2022 and for final approval in May 2023.

RESULTS

The guidance consists of two steps. The first is to determine whether the DRG is credible. We describe five items for assessing credibility: a) is DRG identified using a proper analytical approach; b) is confounding the cause of the DRG; c) is there serious concern about ecological bias; d) is the DRG consistent across studies; and e) is there indirect evidence supporting the DRG. The first two of these items are the most critical. If the DRG was judged to be credible, then the second step is to apply the DRG domain and consider rating up, but only by one level due to the concern about residual confounding.

CONCLUSION

Systematic review authors should only rate up certainty in evidence when a DRG is deemed credible.

Toxic neuropathies

PURPOSE OF REVIEW

Immunotherapy has had a significant impact on the treatment of an increasing number of cancers as well as in inflammatory, rheumatological and gastroenterological conditions.Recreational nitrous oxide use is now a global epidemic. Linezolid is now recommended for the treatment of drug-resistant tuberculosis (TB); neuropathy is a significant cause of morbidity.Global warming will result in increasing toxin exposure, such as ciguatera, in previously unaffected areas.

RECENT FINDINGS

With increasing experience, the pathophysiology underlying the neuropathic complications of these drugs has become clear with guidelines now available, for the complications of immune check-point inhibitors and nitrous oxide toxicity. The optimum dose and duration of treatment for resistant TB with regimens, including linezolid, has been ascertained.

SUMMARY

Although neuropathic complications with immunotherapy are relatively rare, it is essential that they are recognized and treated early. Nitrous oxide toxicity should be in the differential diagnosis for all patients, particularly those of younger age, presenting with a neuropathy or myleo-neuropathy. Ciguatera toxicity is under recognized and its geographical spread will increase due to global warming. Further research is necessary on the mechanisms and treatment of both acute and chronic effects, which at present, are only symptomatic.

Association between Vaccination Status for COVID-19 and the Risk of Severe Symptoms during the Endemic Phase of the Disease

The global health emergency caused by COVID-19 concluded in May 2023, marking the beginning of an endemic phase. This study aimed to evaluate the association between vaccination status and other patient characteristics and the risk of severe disease during this new endemic period. A nationwide cohort study was conducted in Mexico, where we analyzed data from 646 adults who had received positive confirmation of COVID-19 through PCR testing from May to August 2023. The overall risk of severe symptoms in the study sample was 5.3%. The average time elapsed from the last vaccine shot to symptom onset was over six months in all the immunized groups (1, 2 or 3 vaccine doses). Compared to unvaccinated patients, those with three vaccine doses showed an elevated risk of severe symptoms. Advancing age and various chronic comorbidities (specifically cardiovascular, kidney, and obstructive pulmonary conditions) were associated with a heightened risk of severe COVID-19 manifestations. These findings underscore the ongoing seriousness of COVID-19, even in an endemic phase, underscoring the urgent need for tailored interventions aimed at high-risk patients.

Understanding Training Load as Exposure and Dose

Various terms used in sport and exercise science, and medicine, are derived from other fields such as epidemiology, pharmacology and causal inference. Conceptual and nomological frameworks have described training load as a multidimensional construct manifested by two causally related subdimensions: external and internal training load. In this article, we explain how the concepts of training load and its subdimensions can be aligned to classifications used in occupational medicine and epidemiology, where exposure can also be differentiated into external and internal dose. The meanings of terms used in epidemiology such as exposure, external dose, internal dose and dose-response are therefore explored from a causal perspective and their underlying concepts are contextualised to the physical training process. We also explain how these concepts can assist in the validation process of training load measures. Specifically, to optimise training (i.e. within a causal context), a measure of exposure should be reflective of the mediating mechanisms of the primary outcome. Additionally, understanding the difference between intermediate and surrogate outcomes allows for the correct investigation of the effects of exposure measures and their interpretation in research and applied settings. Finally, whilst the dose-response relationship can provide evidence of the validity of a measure, conceptual and computational differentiation between causal (explanatory) and non-causal (descriptive and predictive) dose-response relationships is needed. Regardless of how sophisticated or "advanced" a training load measure (and metric) appears, in a causal context, if it cannot be connected to a plausible mediator of a relevant response (outcome), it is likely of little use in practice to support and optimise the training process.

Association between SARS-CoV-2 infection and newly diagnosed hypertension during pregnancy: prospective, population based cohort study

Objective

To study the association between SARS-CoV-2 infection and newly diagnosed hypertension during pregnancy.

Design

Prospective, population based cohort study.

Setting

All singleton pregnancies after 22 completed gestational weeks registered in the Swedish Pregnancy Register and the Medical Birth Registry of Norway, from 1 March 2020 to 24 May 2022.

Participants

312 456 individuals available for analysis (201 770 in Sweden and 110 686 in Norway), with pregnancies that reached 42 completed gestational weeks by the end of follow-up in the pregnancy registries, excluding individuals with SARS-CoV-2 infection before pregnancy and those with a diagnosis of pre-existing hypertension or onset of hypertension before 20 gestational weeks.

Main outcome measures

Newly diagnosed hypertension during pregnancy was defined as a composite outcome of a diagnosis of gestational hypertension, pre-eclampsia, HELLP (haemolysis, elevated liver enzymes, low platelets) syndrome, or eclampsia, from gestational week 20 to one week after delivery. The association between SARS-CoV-2 infection and hypertension during pregnancy was investigated with a stratified Cox proportional hazard model, adjusting for maternal age, body mass index, parity, smoking, region of birth, education, income, coexisting medical conditions, previous hypertension during pregnancy, number of healthcare visits during the past year, and vaccination against SARS-CoV-2. Pre-eclampsia was also analysed as a separate outcome.

Results

Of 312 456 individuals available for analysis, 8% (n=24 566) had SARS-CoV-2 infection any time during pregnancy, 6% (n=18 051) had a diagnosis of hypertension during pregnancy, and 3% (9899) had pre-eclampsia. SARS-CoV-2 infection during pregnancy was not associated with an increased risk of hypertension during pregnancy (adjusted hazard ratio 0.99, 95% confidence interval 0.93 to 1.04) or pre-eclampsia (0.98, 0.87 to 1.10). The results were similar for SARS-CoV-2 infection in all gestational trimesters and in different time periods that corresponded to dominance of different variants of the SARS-CoV-2 virus.

Conclusions

This population based study did not find any evidence of an association between SARS-CoV-2 infection during pregnancy and an increased risk of hypertension during pregnancy or pre-eclampsia.

Subjective and objective sleep and circadian parameters as predictors of depression-related outcomes: A machine learning approach in UK Biobank

BACKGROUND

Sleep and circadian disruption are associated with depression onset and severity, but it is unclear which features (e.g., sleep duration, chronotype) are important and whether they can identify individuals showing poorer outcomes.

METHODS

Within a subset of the UK Biobank with actigraphy and mental health data (n = 64,353), penalised regression identified the most useful of 51 sleep/rest-activity predictors of depression-related outcomes; including case-control (Major Depression (MD) vs. controls; postnatal depression vs. controls) and within-case comparisons (severe vs. moderate MD; early vs. later onset, atypical vs. typical symptoms; comorbid anxiety; suicidality). Best models (of lasso, ridge, and elastic net) were selected based on Area Under the Curve (AUC).

RESULTS

For MD vs. controls (n_(MD) = 24,229; n_(control) = 40,124), lasso AUC was 0.68, 95 % confidence interval (CI) 0.67-0.69. Discrimination was reasonable for atypical vs. typical symptoms (n_(atypical) = 958; n_(typical) = 18,722; ridge: AUC 0.74, 95 % CI 0.71-0.77) but poor for remaining models (AUCs 0.59-0.67). Key predictors across most models included: difficulty getting up, insomnia symptoms, snoring, actigraphy-measured daytime inactivity and lower morning activity (~8 am). In a distinct subset (n = 310,718), the number of these factors shown was associated with all depression outcomes.

LIMITATIONS

Analyses were cross-sectional and in middle-/older aged adults: comparison with longitudinal investigations and younger cohorts is necessary.

DISCUSSION

Sleep and circadian measures alone provided poor to moderate discrimination of depression outcomes, but several characteristics were identified that may be clinically useful. Future work should assess these features alongside broader sociodemographic, lifestyle and genetic features.

Beyond the biomedical, towards the agentic: A paradigm shift for population health science

Life expectancy in the United States is decreasing. Health disparities are widening. Growing evidence for and integration of social and structural determinants into theory and practice has not yet improved outcomes. The COVID-19 pandemic reinforced the fact. In this paper, we argue that the biomedical model and its underlying scientific paradigm of causal determinism, which currently dominate population health, cannot meet population health needs. While criticism of the biomedical model is not new, this paper advances the field by going beyond criticism to recognize the need for a paradigm shift. In the first half of the paper, we present a critical analysis of the biomedical model and the paradigm of causal determinism. In the second half, we outline the agentic paradigm and present a structural model of health based on generalizable, group-level processes. We use the experience of the COVID-19 pandemic to illustrate the practical applications of our model. It will be important for future work to investigate the empirical and pragmatic applications of our structural model of population health.

Improving interventional causal predictions in regulatory risk assessment

In 2022, the US EPA published an important risk assessment concluding that "Compared to the current annual standard, meeting a revised annual standard with a lower level is estimated to reduce PM2.5-associated health risks in the 30 annually-controlled study areas by about 7-9% for a level of 11.0 µg/m3… and 30-37% for a level of 8.0 µg/m3." These are interventional causal predictions: they predict percentage reductions in mortality risks caused by different counterfactual reductions in fine particulate (PM2.5) levels. Valid causal predictions are possible if: (1) Study designs are used that can support valid causal inferences about the effects of interventions (e.g., quasi-experiments with appropriate control groups); (2) Appropriate causal models and methods are used to analyze the data; (3) Model assumptions are satisfied (at least approximately); and (4) Non-causal sources of exposure-response associations such as confounding, measurement error, and model misspecification are appropriately modeled and adjusted for. This paper examines two long-term mortality studies selected by the EPA to predict reductions in PM2.5-associated risk. Both papers use Cox proportional hazards (PH) models. For these models, none of these four conditions is satisfied, making it difficult to interpret or validate their causal predictions. Scientists, reviewers, regulators, and members of the public can benefit from more trustworthy and credible risk assessments and causal predictions by insisting that risk assessments supporting interventional causal conclusions be based on study designs, methods, and models that are appropriate for predicting effects caused by interventions.

Re-assessing human mortality risks attributed to PM2.5-mediated effects of agricultural ammonia

How can and should epidemiologists and risk assessors assemble and present evidence for causation of mortality or morbidities by identified agents such as fine particulate matter or other air pollutants? As a motivating example, some scientists have warned recently that ammonia from the production of meat significantly increases human mortality rates in exposed populations by increasing the ambient concentration of fine particulate matter (PM2.5) in air. We reexamine the support for such conclusions, including quantitative calculations that attribute deaths to PM2.5 air pollution by applying associational results such as relative risks, odds ratios, or slope coefficients from regression models to predict the effects on mortality or morbidity of reducing PM2.5 exposures. Taking an outside perspective from the field of causal artificial intelligence (CAI), we conclude that these attribution calculations are methodologically unsound. They produce unreliable conclusions because they ignore an essential distinction between differences in outcomes observed at different levels of exposure and changes in outcomes caused by changing exposure. We find that multiple studies that have examined associations between changes over time in particulate exposure and mortality risk instead of differences in exposures and corresponding mortality risks have found no clear evidence that observed changes in exposure help to predict or explain subsequent changes in mortality risks. We conclude that there is no sound theoretical or empirical reason to believe that reducing ammonia emissions from farms has reduced or would reduce human mortality risks. More generally, applying CAI principles and methods can potentially improve current widespread practices of unsound causal inferences and policy-relevant causal claims that are made without the benefit of formal causal analysis in air pollution health effects research and in other areas of applied epidemiology and public health risk assessment.

Understanding clinical outcome measures reported in HIV pregnancy studies involving antiretroviral-naive and antiretroviral-experienced women

HIV infection is a clinically significant public health disease and contributes to increased risk of maternal and fetal morbidity and mortality. HIV pregnancy studies use outcome measures as metrics to show how people with HIV feel, function, or survive. These endpoints are crucial for tracking the evolution of HIV illness over time, assessing the effectiveness of antiretroviral therapy (ART), and comparing outcomes across studies. Although the need for ideal outcome measures is widely acknowledged, selecting acceptable outcome measures for these HIV pregnancy studies can be challenging. We discuss the many outcome measures that have been implemented over time to assess HIV in pregnancy studies, their benefits, and drawbacks. Finally, we offer suggestions for improving the reporting of outcome measures in HIV in pregnancy studies. Medical professionals can best care for pregnant women living with HIV receiving ART by having a thorough understanding of these outcome metrics.

Low level exposure to hydrogen sulfide: a review of emissions, community exposure, health effects, and exposure guidelines

Hydrogen sulfide (H2S) is a toxic gas that is well-known for its acute health risks in occupational settings, but less is known about effects of chronic and low-level exposures. This critical review investigates toxicological and experimental studies, exposure sources, standards, and epidemiological studies pertaining to chronic exposure to H2S from both natural and anthropogenic sources. H2S releases, while poorly documented, appear to have increased in recent years from oil and gas and possibly other facilities. Chronic exposures below 10 ppm have long been associated with odor aversion, ocular, nasal, respiratory and neurological effects. However, exposure to much lower levels, below 0.03 ppm (30 ppb), has been associated with increased prevalence of neurological effects, and increments below 0.001 ppm (1 ppb) in H2S concentrations have been associated with ocular, nasal, and respiratory effects. Many of the studies in the epidemiological literature are limited by exposure measurement error, co-pollutant exposures and potential confounding, small sample size, and concerns of representativeness, and studies have yet to consider vulnerable populations. Long-term community-based studies are needed to confirm the low concentration findings and to refine exposure guidelines. Revised guidelines that incorporate both short- and long-term limits are needed to protect communities, especially sensitive populations living near H2S sources.

How to Distinguish Correlation From Causation in Orthopaedic Research

Correlations in observational studies are commonly misinterpreted as causation. Although correlation is necessary to establish a causal relationship between two variables, correlations may also arise due to chance, reverse causality, or confounding. There are several methods available to orthopaedic researchers to determine whether the observed correlations are causal. These methods depend on the key components of the study including, but not limited to, study design and data availability on confounders. In this article, we illustrate the main concepts surrounding correlation and causation using intuitive real-world examples from the orthopaedic literature. Please visit the following https://www.youtube.com/watch?v=WW7pFudZbHA&t=52s for a video that explains the highlights of the paper in practical terms.

Relationship between parental exposure to radiofrequency electromagnetic fields and primarily hematopoietic neoplasms (lymphoma, leukemia) and tumors in the central nervous system in children: a systematic review

Low-frequency electromagnetic fields have grown exponentially in recent years due to technological development and modernization. The World Health Organization (WHO)/International Agency for Research on Cancer (IARC) has classified radiofrequency electromagnetic fields (RF-EMFs) as possibly carcinogenic to humans (Group 2B), and recent studies have investigated the association between exposure to electromagnetic fields in parents and possible health effects in children, especially the development of tumours of the central nervous system (CNS). The objective of this systematic review was to collate all evidence on the relationship between parental occupational exposure to electromagnetic fields and the development of CNS cancer in children and to evaluate this association. This review was prepared according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. PubMed, Embase, and Web of Science were searched from January 1990 to April 2021. The search was conducted using the following search string: "occupational" AND "child" AND "electromagnetic" AND "cancer". Seventeen articles met our inclusion criteria: 13 case-control studies, two cohort studies, and 2 meta-analyses. Most of the studies showed several methodological weaknesses that limited their results. Due to a lack of consistency regarding the outcome as well as the heterogeneity in the reviewed studies, the body of evidence for the effects of parental exposure to electromagnetic fields is not clear. Methodological heterogeneity in the way that studies were conducted could be responsible for the lack of consistency in the findings. Overall, the body of evidence allows no conclusion on the relationship between parental exposure to electromagnetic fields and the occurrence of CNS tumours in children.

COVID-19 Vaccination and New Onset Glomerular Disease: Results from the IRocGN2 International Registry

Key Points

IgAN and MCD are the most common de novo glomerular diseases reported after COVID-19 vaccination, particularly after mRNA vaccination. Membranous nephropathy, pauci-immune GN, and collapsing GN have also been attributed to COVID-19 vaccination, some with dual histologies. Recovery of kidney function and proteinuria remission is more likely in IgAN and MCD by 4–6 months compared with the other glomerular diseases.

Background

Patients with de novo glomerular disease (GD) with various renal histologies have been reported after vaccination against SARS-CoV-2. Causality has not been established, and the long-term outcomes are not known. To better characterize the GDs and clinical courses/outcomes, we created the International Registry of COVID-19 vaccination and Glomerulonephritis to study in aggregate patients with de novo GN suspected after COVID-19 vaccine exposure.

Methods

A REDCap survey was used for anonymized data collection. Detailed information on vaccination type and timing and GD histology were recorded in the registry. We collected serial information on laboratory values (before and after vaccination and during follow-up), treatments, and kidney-related outcomes.

Results

Ninety-eight patients with GD were entered into the registry over 11 months from 44 centers throughout the world. Median follow-up was 89 days after diagnosis. IgA nephropathy (IgAN) and minimal change disease (MCD) were the most common kidney diseases reported. Recovery of kidney function and remission of proteinuria were more likely in IgAN and MCD at 4–6 months than with pauci-immune GN/vasculitis and membranous nephropathy.

Conclusions

The development of GD after vaccination against SARS-CoV-2 may be a very rare adverse event. Temporal association is present for IgAN and MCD, but causality is not firmly established. Kidney outcomes for IgAN and MCD are favorable. No changes in vaccination risk-benefit assessment are recommended based on these findings.

Assigning cause for emerging diseases of aquatic organisms

Resolving the cause of disease (= aetiology) in aquatic organisms is a challenging but essential goal, heightened by increasing disease prevalence in a changing climate and an interconnected world of anthropogenic pathogen spread. Emerging diseases play important roles in evolutionary ecology, wildlife conservation, the seafood industry, recreation, cultural practices, and human health. As we emerge from a global pandemic of zoonotic origin, we must focus on timely diagnosis to confirm aetiology and enable response to diseases in aquatic ecosystems. Those systems' resilience, and our own sustainable use of seafood, depend on it. Synchronising traditional and recent advances in microbiology that span ecological, veterinary, and medical fields will enable definitive assignment of risk factors and causal agents for better biosecurity management and healthier aquatic ecosystems.

High Prevalence of Syphilis and Syphilis/HIV Coinfection among Men Who Have Sex with Men Who Attend Meeting Places in Mexico

Men who have sex with men (MSM) are disproportionately affected by syphilis, HIV, and syphilis/HIV coinfection. Antiretroviral therapy (ART) prevents HIV transmission but does not impede the spread or acquisition of syphilis. Information about syphilis/HIV coinfection among MSM is scarce. We aimed to determine the prevalence of syphilis/HIV coinfection in a national sample of MSM who attend meeting places (such as movies, clubs, gay bars, saunas, shopping malls, and others referred to by the same MSM participants of the study) in Mexico to evaluate factors associated with syphilis, and to compare the prevalence rates of syphilis between the current survey and DGE data. We performed a laboratory diagnosis to determine the rates of syphilis and HIV among the included MSM. The national and regional prevalence of syphilis was calculated. HIV and coinfection prevalence were determined only for the survey. All prevalence rates included 95%CIs. Descriptive, bivariate, and multivariate analyses were performed. The national prevalence rates of syphilis, HIV, and coinfection were 15.2%, 10.2%, and 5.7%, respectively. The region with the highest prevalence rate was Mexico City (39.4%). The center region, minimal “goods” (i.e., a minimal number of material possessions such as a car or dryer, etc., which served as a proxy for low economic income level), use of “inhalant drugs”, “HIV infection”, “sexual intercourse” only with men, “rewarded sex”, and “youngest age at first sexual encounter or debut” were risk factors for syphilis. In general, regional prevalence of syphilis was higher in the survey (2013) and DGE data from 2019 than in the DGE data from 2013. Similar to other countries, Mexico needs to assess elements around not only syphilis and HIV infections but also syphilis/HIV coinfection, and preventive measures focusing on MSM are needed.

Application of AOPs to assist regulatory assessment of chemical risks - Case studies, needs and recommendations

While human regulatory risk assessment (RA) still largely relies on animal studies, new approach methodologies (NAMs) based on in vitro, in silico or non-mammalian alternative models are increasingly used to evaluate chemical hazards. Moreover, human epidemiological studies with biomarkers of effect (BoE) also play an invaluable role in identifying health effects associated with chemical exposures. To move towards the next generation risk assessment (NGRA), it is therefore crucial to establish bridges between NAMs and standard approaches, and to establish processes for increasing mechanistically-based biological plausibility in human studies. The Adverse Outcome Pathway (AOP) framework constitutes an important tool to address these needs but, despite a significant increase in knowledge and awareness, the use of AOPs in chemical RA remains limited. The objective of this paper is to address issues related to using AOPs in a regulatory context from various perspectives as it was discussed in a workshop organized within the European Union partnerships HBM4EU and PARC in spring 2022. The paper presents examples where the AOP framework has been proven useful for the human RA process, particularly in hazard prioritization and characterization, in integrated approaches to testing and assessment (IATA), and in the identification and validation of BoE in epidemiological studies. Nevertheless, several limitations were identified that hinder the optimal usability and acceptance of AOPs by the regulatory community including the lack of quantitative information on response-response relationships and of efficient ways to map chemical data (exposure and toxicity) onto AOPs. The paper summarizes suggestions, ongoing initiatives and third-party tools that may help to overcome these obstacles and thus assure better implementation of AOPs in the NGRA.

An application of the stress-diathesis model: A review about the association between smoking tobacco, smoking cessation, and mental health

Background

Worldwide, approximately 24% of all adults smoke, but smoking is up to twice as prevalent in people with mental ill-health. There is growing evidence that smoking may be a causal risk factor in the development of mental illness, and that smoking cessation leads to improved mental health.

Methods

In this scholarly review we have: (1) used a modern adaptation of the Bradford-Hill criteria to bolster the argument that smoking could cause mental ill-health and that smoking cessation could reverse these effects, and (2) by considering psychological, biological, and environmental factors, we have structured the evidence to-date into a stress-diathesis model.

Results

Our model suggests that smoking is a psychobiological stressor, but that the magnitude of this effect is mediated and modulated by the individual's diathesis to develop mental ill-health and other vulnerability and protective factors. We explore biological mechanisms that underpin the model, such as tobacco induced damage to neurological systems and oxidative stress pathways. Furthermore, we discuss evidence indicating that it is likely that these systems repair after smoking cessation, leading to better mental health.

Conclusion

Based on a large body of literature including experimental, observational, and novel causal inference studies, there is consistent evidence showing that smoking can negatively affect the brain and mental health, and that smoking cessation could reverse the mental ill-health caused by smoking. Our model suggests that smoking prevention and treatment strategies have a role in preventing and treating mental illness as well as physical illness.

Causal attribution fractions, and the attribution of smoking and BMI to the landscape of disease incidence in UK Biobank

Unlike conventional epidemiological studies that use observational data to estimate "associations" between risk factors and disease, the science of causal inference has identified situations where causal estimates can be made from observational data, using results such as the "backdoor criteria". Here these results are combined with established epidemiological methods, to calculate simple population attribution fractions that estimate the causal influence of risk factors on disease incidence, and can be estimated using conventional proportional hazards methods. A counterfactual argument gives an attribution fraction for individuals. Causally meaningful attribution fractions cannot be constructed for all risk factors or confounders, but they can for the important established risk factors of smoking and body mass index (BMI). Using the new results, the causal attribution of smoking and BMI to the incidence of 226 diseases in the UK Biobank are estimated, and summarised in terms of disease chapters from the International Classification of Diseases (ICD-10). The diseases most strongly attributed to smoking and BMI are identified, finding 11 with attribution fractions greater than 0.5, and a small number with protective associations. The results provide new tools to quantify the causal influence of risk factors such as smoking and BMI on disease, and survey the causal influence of smoking and BMI on the landscape of disease incidence in the UK Biobank population.

Spironolactone as a potential new pharmacotherapy for alcohol use disorder: convergent evidence from rodent and human studies

Evidence suggests that spironolactone, a nonselective mineralocorticoid receptor (MR) antagonist, modulates alcohol seeking and consumption. Therefore, spironolactone may represent a novel pharmacotherapy for alcohol use disorder (AUD). In this study, we tested the effects of spironolactone in a mouse model of alcohol drinking (drinking-in-the-dark) and in a rat model of alcohol dependence (vapor exposure). We also investigated the association between spironolactone receipt for at least 60 continuous days and change in self-reported alcohol consumption, using the Alcohol Use Disorders Identification Test-Consumption (AUDIT-C), in a pharmacoepidemiologic cohort study in the largest integrated healthcare system in the US. Spironolactone dose-dependently reduced the intake of sweetened or unsweetened alcohol solutions in male and female mice. No effects of spironolactone were observed on drinking of a sweet solution without alcohol, food or water intake, motor coordination, alcohol-induced ataxia, or blood alcohol levels. Spironolactone dose-dependently reduced operant alcohol self-administration in dependent and nondependent male and female rats. In humans, a greater reduction in alcohol consumption was observed among those who received spironolactone, compared to propensity score-matched individuals who did not receive spironolactone. The largest effects were among those who reported hazardous/heavy episodic alcohol consumption at baseline (AUDIT-C ≥ 8) and those exposed to ≥ 50 mg/day of spironolactone. These convergent findings across rodent and human studies demonstrate that spironolactone reduces alcohol use and support the hypothesis that this medication may be further studied as a novel pharmacotherapy for AUD.

Free Sugar Intake and Periodontal Diseases: A Systematic Review

High free sugar intake is associated with an increased risk of various non-communicable diseases. We aimed to systematically review articles investigating the association between free sugar intake and periodontal diseases. This systematic review was conducted according to PRISMA guidelines and was registered in the PROSPERO database (CRD42022337828). We obtained articles from PubMed, Web of Science, and Scopus in April 2022. The study selection was performed according to predefined eligibility criteria based on the following PECOS: (P) general population, (E/C) free-sugar-containing food/beverage intake, (O) clinically measured periodontal diseases, and (S) observational study and clinical trial. Of the 839 screened records, 13 studies were included in the review. Most studies (n = 12) had a cross-sectional design. The age groups in the included studies were children/adolescents (n = 5) and adults (n = 8). Among the included studies, 11 reported a significant association between the frequent intake of free-sugar-containing food or beverages and a higher prevalence or incidence of periodontal diseases. The quality of most of the included studies was scored "fair" based on the Newcastle-Ottawa Quality Assessment Scale. Although the majority of the included studies reported a significant positive association between high free sugar intake and periodontal diseases, the evidence is considered to be limited due to the study designs.

White Blood Cells, COVID-19, and Mendelian Randomization

Many observational studies have shown an association between the severity of COVID-19 and the different white blood cell counts, most frequently neutrophils, lymphocytes, and eosinophils. The studies aimed to predict the prognosis, and therefore, a causal relationship was unnecessary. However, if we begin to look at these biomarkers as potential therapeutic targets, then causality is essential. Observational studies cannot prove a causal relationship, and randomized trials are not always feasible. In this case, Mendelian randomization studies, considered more valid than observational studies, could add to the arguments for causality. Two Mendelian randomization studies tested for a causal relationship between the number of different white cell populations and COVID-19 severity, but their results are different; therefore, the problem of causality is not settled in this case.

Causal assessment in evidence synthesis: A methodological review of reviews

In fields (such as population health) where randomised trials are often lacking, systematic reviews (SRs) can harness diversity in study design, settings and populations to assess the evidence for a putative causal relationship. SRs may incorporate causal assessment approaches (CAAs), sometimes called 'causal reviews', but there is currently no consensus on how these should be conducted. We conducted a methodological review of self-identifying 'causal reviews' within the field of population health to establish: (1) which CAAs are used; (2) differences in how CAAs are implemented; (3) how methods were modified to incorporate causal assessment in SRs. Three databases were searched and two independent reviewers selected reviews for inclusion. Data were extracted using a standardised form and summarised using tabulation and narratively. Fifty-three reviews incorporated CAAs: 46/53 applied Bradford Hill (BH) viewpoints/criteria, with the remainder taking alternative approaches: Medical Research Council guidance on natural experiments (2/53, 3.8%); realist reviews (2/53, 3.8%); horizontal SRs (1/53, 1.9%); 'sign test' of causal mechanisms (1/53, 1.9%); and a causal cascade model (1/53, 1.9%). Though most SRs incorporated BH, there was variation in application and transparency. There was considerable overlap across the CAAs, with a trade-off between breadth (BH viewpoints considered a greater range of causal characteristics) and depth (many alternative CAAs focused on one viewpoint). Improved transparency in the implementation of CAA in SRs in needed to ensure their validity and allow robust assessments of causality within evidence synthesis.

"Sugar-Sweetened Beverages" Is an Independent Risk From Pancreatic Cancer: Based on Half a Million Asian Cohort Followed for 25 Years

Although the link between sugar-sweetened beverages (SSB) and pancreatic cancer has been suggested for its insulin-stimulating connection, most epidemiological studies showed inconclusive relationship. Whether the result was limited by sample size is explored. This prospective study followed 491,929 adults, consisting of 235,427 men and 256,502 women (mean age: 39.9, standard deviation: 13.2), from a health surveillance program and there were 523 pancreatic cancer deaths between 1994 and 2017. The individual identification numbers of the cohort were matched with the National Death file for mortality, and Cox models were used to assess the risk. The amount of SSB intake was recorded based on the average consumption in the month before interview by a structured questionnaire. We classified the amount of SSB intake into 4 categories: 0–<0.5 serving/day, ≥0.5–<1 serving per day, ≥1–<2 servings per day, and ≥2 servings per day. One serving was defined as equivalent to 12 oz and contained 35 g added sugar. We used the age and the variables at cohort enrolment as the reported risks of pancreatic cancers. The cohort was divided into 3 age groups, 20–39, 40–59, and ≥60. We found young people (age <40) had higher prevalence and frequency of sugar-sweetened beverages than the elderly. Those consuming 2 servings/day had a 50% increase in pancreatic cancer mortality (HR = 1.55, 95% CI: 1.08–2.24) for the total cohort, but a 3-fold increase (HR: 3.09, 95% CI: 1.44–6.62) for the young. The risk started at 1 serving every other day, with a dose–response relationship. The association of SSB intake of ≥2 servings/day with pancreatic cancer mortality among the total cohort remained significant after excluding those who smoke or have diabetes (HR: 2.12, 97% CI: 1.26–3.57), are obese (HR: 1.57, 95% CI: 1.08–2.30), have hypertension (HR: 1.90, 95% CI: 1.20–3.00), or excluding who died within 3 years after enrollment (HR: 1.67, 95% CI: 1.15–2.45). Risks remained in the sensitivity analyses, implying its independent nature. We concluded that frequent drinking of SSB increased pancreatic cancer in adults, with highest risk among young people.

Assessing the causal relationship between income inequality and mortality and self-rated health: protocol for systematic review and meta-analysis

BACKGROUND

Income inequality has been linked to health and mortality. While there has been extensive research exploring the relationship, the evidence for whether the relationship is causal remains disputed. We describe the methods for a systematic review that will transparently assess whether a causal relationship exists between income inequality and mortality and self-rated health.

METHODS

We will identify relevant studies using search terms relating to income inequality, mortality, and self-rated health (SRH). Four databases will be searched: MEDLINE, ISI Web of Science, EMBASE, and the National Bureau of Economic Research. The inclusion criteria have been developed to identify the study designs best suited to assess causality: multilevel studies that have conditioned upon individual income (or a comparable measure, such as socioeconomic position) and natural experiment studies. Risk of bias assessment of included studies will be conducted using ROBINS-I. Where possible, we will convert all measures of income inequality into Gini coefficients and standardize the effect estimate of income inequality on mortality/SRH. We will conduct random-effects meta-analysis to estimate pooled effect estimates when possible. We will assess causality using modified Bradford Hill viewpoints and assess certainty using GRADE.

DISCUSSION

This systematic review protocol lays out the complexity of the relationship between income inequality and individual health, as well as our approach for assessing causality. Understanding whether income inequality impacts the health of individuals within a population has major policy implications. By setting out our methods and approach as transparently as we can, we hope this systematic review can provide clarity to an important topic for public policy and public health, as well as acting as an exemplar for other "causal reviews".

Impact of Noise Exposure on Risk of Developing Stress-Related Health Effects Related to the Cardiovascular System: A Systematic Review and Meta-Analysis

Background

: Exposure to acute noise can cause an increase in biological stress reactions, which provides biological plausibility for a potential association between sustained noise exposure and stress-related health effects. However, the certainty in the evidence for an association between exposures to noise on short- and long-term biomarkers of stress has not been widely explored. The objective of this review was to evaluate the strength of evidence between noise exposure and changes in the biological parameters known to contribute to the development of stress-related adverse cardiovascular responses.

Materials and Methods

This systematic review comprises English language comparative studies available in PubMed, Cochrane Central, EMBASE, and CINAHL databases from January 1, 1980 to December 29, 2021. Where possible, random-effects meta-analyses were used to examine the effect of noise exposure from various sources on stress-related cardiovascular biomarkers. The risk of bias of individual studies was assessed using the risk of bias of nonrandomized studies of exposures instrument. The certainty of the body of evidence for each outcome was assessed using the Grading of Recommendations Assessment, Development, and Evaluation approach.

Results

: The search identified 133 primary studies reporting on blood pressure, hypertension, heart rate, cardiac arrhythmia, vascular resistance, and cardiac output. Meta-analyses of blood pressure, hypertension, and heart rate suggested there may be signals of increased risk in response to a higher noise threshold or incrementally higher levels of noise. Across all outcomes, the certainty of the evidence was very low due to concerns with the risk of bias, inconsistency across exposure sources, populations, and studies and imprecision in the estimates of effects.

Conclusions

: This review identifies that exposure to higher levels of noise may increase the risk of some short- and long-term cardiovascular events; however, the certainty of the evidence was very low. This likely represents the inability to compare across the totality of the evidence for each outcome, underscoring the value of continued research in this area. Findings from this review may be used to inform policies of noise reduction or mitigation interventions.

SARS-CoV-2 infection during pregnancy and risk of preeclampsia: a systematic review and meta-analysis

OBJECTIVE

To examine the relationship between SARS-CoV-2 infection during pregnancy and the risk for preeclampsia.

DATA SOURCES

MEDLINE, Embase, POPLINE, CINAHL, LILACS, and the World Health Organization COVID-19, Chinese, and preprint databases (all from December 1, 2019, to May 31, 2021). Google Scholar, bibliographies, and conference proceedings were also searched.

STUDY ELIGIBILITY CRITERIA

Observational studies that assessed the association between SARS-CoV-2 infection during pregnancy and preeclampsia and that reported unadjusted and/or adjusted risk estimates and 95% confidence intervals or data to calculate them.

STUDY APPRAISAL AND SYNTHESIS METHODS

The primary outcome was preeclampsia. Secondary outcomes included preeclampsia with severe features, preeclampsia without severe features, eclampsia, and hemolysis, elevated liver enzymes, and low platelet count (HELLP) syndrome. Two reviewers independently reviewed studies for inclusion, assessed their risk of bias, and extracted data. Pooled unadjusted and adjusted odds ratios with 95% confidence intervals, and 95% prediction interval were calculated. Heterogeneity was quantified using the І2 statistic, for which І2≥30% indicated substantial heterogeneity. Subgroup and sensitivity analyses were performed to test the robustness of the overall findings.

RESULTS

A total of 28 studies comprising 790,954 pregnant women, among which 15,524 were diagnosed with SARS-CoV-2 infection, met the inclusion criteria. The meta-analysis of unadjusted odds ratios showed that the odds of developing preeclampsia were significantly higher among pregnant women with SARS-CoV-2 infection than among those without SARS-CoV-2 infection (7.0% vs 4.8%; pooled odds ratio, 1.62; 95% confidence interval, 1.45-1.82; P<.00001; І2=17%; 26 studies; 95% prediction interval of the odds ratio, 1.28-2.05). The meta-analysis of adjusted odds ratios also showed that SARS-CoV-2 infection during pregnancy was associated with a significant increase in the odds of preeclampsia (pooled odds ratio, 1.58; 95% confidence interval, 1.39-1.80; P<.0001; І2=0%; 11 studies). There was a statistically significant increase in the odds of preeclampsia with severe features (odds ratio, 1.76; 95% confidence interval, 1.18-2.63; І2=58%; 7 studies), eclampsia (odds ratio, 1.97; 95% confidence interval, 1.01-3.84; І2=0%, 3 studies), and HELLP syndrome (odds ratio, 2.10; 95% confidence interval, 1.48-2.97; 1 study) among pregnant women with SARS-CoV-2 infection when compared to those without the infection. Overall, the direction and magnitude of the effect of SARS-CoV-2 infection during pregnancy on preeclampsia was consistent across most prespecified subgroup and sensitivity analyses. Both asymptomatic and symptomatic SARS-CoV-2 infections significantly increased the odds of developing preeclampsial; however, it was higher among patients with symptomatic illness (odds ratio, 2.11; 95% confidence interval, 1.59-2.81) than among those with asymptomatic illness (odds ratio, 1.59; 95% confidence interval, 1.21-2.10).

CONCLUSION

SARS-CoV-2 during pregnancy is associated with higher odds of preeclampsia.

An approach to quantifying the potential importance of residual confounding in systematic reviews of observational studies: A GRADE concept paper

Small relative effect sizes are common in observational studies of exposure in environmental and public health. However, such effects can still have considerable policy importance when the baseline rate of the health outcome is high, and many persons are exposed. Assessing the certainty of the evidence based on these effect sizes is challenging because they can be prone to residual confounding due to the non-randomized nature of the evidence. When applying GRADE, a precise relative risk >2.0 increases the certainty in an existing effect because residual confounding is unlikely to explain the association. GRADE also suggests rating up when opposing plausible residual confounding exists for other effect sizes. In this concept paper, we propose using the E-value, defined as the smallest effect size of a confounder that still can reduce an observed RR to the null value, and a reference confounder to assess the likelihood of residual confounding. We propose a 4-step approach. 1. Assess the association of interest for relevant exposure levels. 2. Calculate the E-value for this observed association. 3. Choose a reference confounder with sufficient strength and information and assess its effect on the observed association using the E-value. 4. Assess how likely it is that residual confounding will still bias the observed RR. We present three case studies and discuss the feasibility of the approach.

Strength in causality: discerning causal mechanisms in the sufficient cause model

The assessment of causality is fundamental to epidemiology and biomedical sciences. One well-known approach to distinguishing causal from noncausal explanations is the nine Bradford Hill viewpoints. A recent article in this journal revisited the viewpoints to incorporate developments in causal thinking, suggesting that the sufficient cause model is useful in elucidating the theoretical underpinning of the first of the nine viewpoints-strength of association. In this article, we discuss how to discern the causal mechanisms of interest in the sufficient cause model, which pays closer attention to the relationship between the sufficient cause model and the Bradford Hill viewpoints. To this end, we explicate the link between the sufficient cause model and the potential-outcome model, both of which have become the cornerstone of causal thinking in epidemiology and biomedicine. A clearer understanding of the link between the two models provides significant implications for interpretation of the observed risks in the subpopulations defined by exposure and confounder. We also show that the concept of potential completion times of sufficient causes is useful to fully discerning completed sufficient causes, which leads us to pay closer attention to the fourth of the nine Bradford Hill viewpoints-temporality. Decades after its introduction, the sufficient cause model may be vaguely understood and thus implicitly used under unreasonably strict assumptions. To strengthen our assessment in the face of multifactorial causality, it is significant to carefully scrutinize the observed associations in a complementary manner, using the sufficient cause model as well as its relevant causal models.

Applying Concepts of Causal Inference to Infectious Bovine Keratoconjunctivitis

Establishing causation, otherwise known as causal assessment, is a difficult task, made more difficult by the variety of causal assessment frameworks available to consider. In this article, Bradford Hill viewpoints are used to discuss the evidence base for Moraxella bovis and Moraxella bovoculi being component causes of infectious bovine keratoconjunctivitis. Each of the nine Bradford Hill viewpoints are introduced and explained: strength, consistency, specificity, temporality, biologic gradient, plausibility, coherence, experiment, and analogy. Examples of how the viewpoints have been applied for other causal relations are provided, and then the evidence base for M bovis and M bovoculi is discussed.

Impact of educational level and employment status on short-term and long-term pain relief from supervised exercise therapy and education: an observational study of 22 588 patients with knee and hip osteoarthritis

OBJECTIVES

To investigate the impact of educational level and employment status on change in pain intensity after treatment among patients with knee and hip osteoarthritis (OA).

DESIGN

A prospective cohort study.

SETTING AND PARTICIPANTS

We analysed 22 588 patients participating in the Good Life with osteoArthritis in Denmark (GLA:D). GLA:D consists of two patient education sessions and 12 supervised exercise sessions.

PRIMARY OUTCOME

Baseline educational level and employment status were used as exposures. We investigated the impact of both exposures separately on mean change in pain intensity (visual analogue scale 0-100 mm) from baseline to immediately after treatment (approximately 3 months) and at 12 months, using linear mixed models.

RESULTS

On average, all patients improved in pain intensity. The average improvement in pain did not differ by educational level, except for one group. Patients with long-term education had less improvement after treatment (2.0 mm, 95% CI 0.8 to 3.1) and at 12 months (2.0 mm, 95% CI 0.6 to 3.4) compared with primary school only (reference). According to employment status, patients on sick leave had the greatest improvement in pain after treatment (-3.4, 95% CI -4.9 to -1.9) and at 12 months (-4.5, 95% CI -6.4 to -2.6) compared with retired patients (reference).

CONCLUSIONS

On average, all patients reported improvement in pain at short-term and long-term follow-up. Change in pain intensity did not substantially differ by educational level or employment status, as the absolute differences were small and most likely not clinically important.