Evaluating Public Health Interventions: 6. Modeling Ratios or Differences? Let the Data Tell Us

Learning about treatment effects in a new target population under transportability assumptions for relative effect measures

Investigators often believe that relative effect measures conditional on covariates, such as risk ratios and mean ratios, are "transportable" across populations. Here, we examine the identification of causal effects in a target population using an assumption that conditional relative effect measures are transportable from a trial to the target population. We show that transportability for relative effect measures is largely incompatible with transportability for difference effect measures, unless the treatment has no effect on average or one is willing to make even stronger transportability assumptions that imply the transportability of both relative and difference effect measures. We then describe how marginal (population-averaged) causal estimands in a target population can be identified under the assumption of transportability of relative effect measures, when we are interested in the effectiveness of a new experimental treatment in a target population where the only treatment in use is the control treatment evaluated in the trial. We extend these results to consider cases where the control treatment evaluated in the trial is only one of the treatments in use in the target population, under an additional partial exchangeability assumption in the target population (i.e., an assumption of no unmeasured confounding in the target population with respect to potential outcomes under the control treatment in the trial). We also develop identification results that allow for the covariates needed for transportability of relative effect measures to be only a small subset of the covariates needed to control confounding in the target population. Last, we propose estimators that can be easily implemented in standard statistical software and illustrate their use using data from a comprehensive cohort study of stable ischemic heart disease.

Using Parametric g-Computation for Time-to-Event Data and Distributed Lag Models to Identify Critical Exposure Windows for Preterm Birth: An Illustrative Example Using PM2.5 in a Retrospective Birth Cohort Based in Eastern Massachusetts (2011-2016)

BACKGROUND

Parametric g-computation is an attractive analytic framework to study the health effects of air pollution. Yet, the ability to explore biologically relevant exposure windows within this framework is underdeveloped.

OBJECTIVES

We outline a novel framework for how to incorporate complex lag-responses using distributed lag models (DLMs) into parametric g-computation analyses for survival data. We call this approach "g-survival-DLM" and illustrate its use examining the association between PM 2.5 during pregnancy and the risk of preterm birth (PTB).

METHODS

We applied the g-survival-DLM approach to estimate the hypothetical static intervention of reducing average PM 2.5 in each gestational week by 20% on the risk of PTB among 9,403 deliveries from Beth Israel Deaconess Medical Center, Boston, Massachusetts, 2011-2016. Daily PM 2.5 was taken from a 1 -km grid model and assigned to address at birth. Models were adjusted for sociodemographics, time trends, nitrogen dioxide, and temperature. To facilitate implementation, we provide a detailed description of the procedure and accompanying R syntax.

RESULTS

There were 762 (8.1%) PTBs in this cohort. The gestational week-specific median PM 2.5 concentration was relatively stable across pregnancy at ∼ 7 μ g / m 3 . We found that our hypothetical intervention strategy changed the cumulative risk of PTB at week 36 (i.e., the end of the preterm period) by - 0.009 (95% confidence interval: - 0.034 , 0.007) in comparison with the scenario had we not intervened, which translates to about 86 fewer PTBs in this cohort. We also observed that the critical exposure window appeared to be weeks 5-20.

DISCUSSION

We demonstrate that our g-survival-DLM approach produces easier-to-interpret, policy-relevant estimates (due to the g-computation); prevents immortal time bias (due to treating PTB as a time-to-event outcome); and allows for the exploration of critical exposure windows (due to the DLMs). In our illustrative example, we found that reducing fine particulate matter [particulate matter (PM) with aerodynamic diameter ≤ 2.5 μ m (PM 2.5 )] during gestational weeks 5-20 could potentially lower the risk of PTB. https://doi.org/10.1289/EHP13891.

Association between gestational exposure to solar activity and pregnancy loss using live births from a Massachusetts-based medical center

BACKGROUND

Solar activity has been linked to biological mechanisms important to pregnancy, including folate and melatonin levels and inflammatory markers. Thus, we aimed to investigate the association between gestational solar activity and pregnancy loss.

METHODS

Our study included 71,963 singleton births conceived in 2002-2016 and delivered at an academic medical center in Eastern Massachusetts. We studied several solar activity metrics, including sunspot number, Kp index, and ultraviolet radiation, with data from the NASA Goddard Space Flight Center and European Centre for Medium-Range Weather Forecasts. We used a novel time series analytic approach to investigate associations between each metric from conception through 24 weeks of gestation and the number of live birth-identified conceptions (LBICs) -the total number of conceptions in each week that result in a live birth. This approach fits distributed lag models to data on LBICs, adjusted for time trends, and allows us to infer associations between pregnancy exposure and pregnancy loss.

RESULTS

Overall, the association between solar activity during pregnancy and pregnancy loss varied by exposure metric. For sunspot number, we found that an interquartile range increase in sunspot number (78·7 sunspots) in all of the first 24 weeks of pregnancy was associated with 14·0 (95% CI: 6·5, 21·3) more pregnancy losses out of the average 92 LBICs in a week, and exposure in weeks ten through thirteen was identified as a critical window. Although not statistically significant, higher exposure to Kp index and to UV radiation across all 24 weeks of pregnancy was associated with more and less pregnancy losses, respectively.

CONCLUSION

While exposure to certain metrics of solar activity (i.e., sunspot number) throughout the first 24 weeks of pregnancy may be associated with pregnancy losses, exposure to other metrics were not. Solar activity is a complex phenomenon, and more studies are needed to clarify underlying pathways.

Disparities in Fatal Occupational Injury Rates in North Carolina, 1978-2017: Comparing Nonmanagerial Employees to Managers

BACKGROUND

We examined fatal occupational injuries among private-sector workers in North Carolina during the 40-year period 1978-2017, comparing the occurrence of fatal injuries among nonmanagerial employees to that experienced by managers.

METHODS

We estimated a standardized fatal occupational injury ratio by inverse probability of exposure weighting, taking nonmanagerial workers as the target population. When this ratio measure takes a value greater than unity it signals settings in which nonmanagerial employees are not provided as safe a work environment as that provided for managers.

RESULTS

Across all industries, nonmanagerial workers in North Carolina experienced fatal occupational injury rates 8.2 (95% CI = 7.0, 10.0) times the rate experienced by managers. Disparities in fatal injury rates between managers and the employees they supervise were greatest in forestry, rubber and metal manufacturing, wholesale trade, fishing and extractive industries, and construction.

CONCLUSIONS

The results may help focus discussion about workplace safety between labor and management upon equity, with a goal of providing a work environment for nonmanagerial employees as safe as the one provided for managers.

Understanding 'it depends' in ecology: a guide to hypothesising, visualising and interpreting statistical interactions

Ecologists routinely use statistical models to detect and explain interactions among ecological drivers, with a goal to evaluate whether an effect of interest changes in sign or magnitude in different contexts. Two fundamental properties of interactions are often overlooked during the process of hypothesising, visualising and interpreting interactions between drivers: the measurement scale - whether a response is analysed on an additive or multiplicative scale, such as a ratio or logarithmic scale; and the symmetry - whether dependencies are considered in both directions. Overlooking these properties can lead to one or more of three inferential errors: misinterpretation of (i) the detection and magnitude (Type-D error), and (ii) the sign of effect modification (Type-S error); and (iii) misidentification of the underlying processes (Type-A error). We illustrate each of these errors with a broad range of ecological questions applied to empirical and simulated data sets. We demonstrate how meta-analysis, a widely used approach that seeks explicitly to characterise context dependence, is especially prone to all three errors. Based on these insights, we propose guidelines to improve hypothesis generation, testing, visualisation and interpretation of interactions in ecology.

Using Randomized Controlled Trials to Estimate the Effect of Community Interventions for Childhood Asthma

INTRODUCTION

The Centers for Disease Control and Prevention's Controlling Childhood Asthma and Reducing Emergencies initiative aims to prevent 500,000 emergency department (ED) visits and hospitalizations within 5 years among children with asthma through implementation of evidence-based interventions and policies. Methods are needed for calculating the anticipated effects of planned asthma programs and the estimated effects of existing asthma programs. We describe and illustrate a method of using results from randomized control trials (RCTs) to estimate changes in rates of adverse asthma events (AAEs) that result from expanding access to asthma interventions.

METHODS

We use counterfactual arguments to justify a formula for the expected number of AAEs prevented by a given intervention. This formula employs a current rate of AAEs, a measure of the increase in access to the intervention, and the rate ratio estimated in an RCT.

RESULTS

We justified a formula for estimating the effect of expanding access to asthma interventions. For example, if 20% of patients with asthma in a community with 20,540 annual asthma-related ED visits were offered asthma self-management education, ED visits would decrease by an estimated 1,643; and annual hospitalizations would decrease from 2,639 to 617.

CONCLUSION

Our method draws on the best available evidence from RCTs to estimate effects on rates of AAEs in the community of interest that result from expanding access to asthma interventions.

Extent of Low-density Lipoprotein Cholesterol Reduction and All-cause and Cardiovascular Mortality Benefit: A Systematic Review and Meta-analysis

ABSTRACT

Lipid-modifying agents steadily lower low-density lipoprotein cholesterol (LDL-C) levels with the aim of reducing mortality. A systematic review and meta-analysis were conducted to determine whether all-cause or cardiovascular (CV) mortality effect size for lipid-lowering therapy varied according to the magnitude of LDL-C reduction. Electronic databases were searched, including PubMed and ClinicalTrials.gov , from inception to December 31, 2019. Eligible studies included randomized controlled trials that compared lipid-modifying agents (statins, ezetimibe, and PCSK-9 inhibitors) versus placebo, standard or usual care or intensive versus less-intensive LDL-C-lowering therapy in adults, with or without known history of CV disease with a follow-up of at least 52 weeks. All-cause and CV mortality as primary end points, myocardial infarction, stroke, and non-CV death as secondary end points. Absolute risk differences [ARD (ARDs) expressed as incident events per 1000 person-years], number needed to treat (NNT), and rate ratios (RR) were assessed. Sixty randomized controlled trials totaling 323,950 participants were included. Compared with placebo, usual care or less-intensive therapy, active or more potent lipid-lowering therapy reduced the risk of all-cause death [ARD -1.33 (-1.89 to -0.76); NNT 754 (529-1309); RR 0.92 (0.89-0.96)]. Intensive LDL-C percent lowering was not associated with further reductions in all-cause mortality [ARD -0.27 (-1.24 to 0.71); RR 1.00 (0.94-1.06)]. Intensive LDL-C percent lowering did not further reduce CV mortality [ARD -0.28 (-0.83 to 0.38); RR 1.02 (0.94-1.09)]. Our findings indicate that risk reduction varies across subgroups and that overall NNTs are high. Identifying patient subgroups who benefit the most from LDL-C levels reduction is clinically relevant and necessary.

A non-parametric approach for jointly combining evidence on progression free and overall survival time in network meta-analysis

Randomised controlled trials of cancer treatments typically report progression free survival (PFS) and overall survival (OS) outcomes. Existing methods to synthesise evidence on PFS and OS either rely on the proportional hazards assumption or make parametric assumptions which may not capture the diverse survival curve shapes across studies and treatments. Furthermore, PFS and OS are not independent: OS is the sum of PFS and post-progression survival (PPS). Our aim was to develop a non-parametric approach for jointly synthesising evidence from published Kaplan-Meier survival curves of PFS and OS without assuming proportional hazards. Restricted mean survival times (RMST) are estimated by the area under the survival curves (AUCs) up to a restricted follow-up time. The correlation between AUCs due to the constraint that OS>PFS is estimated using bootstrap re-sampling. Network meta-analysis models are given for RMST for PFS and PPS and ensure that OS=PFS + PPS. Both additive and multiplicative network meta-analysis models are presented to obtain relative treatment effects as either differences or ratios of RMST. The methods are illustrated with a network meta-analysis of treatments for Stage IIIA-N2 Non-Small Cell Lung Cancer. The approach has implications for health economic models of cancer treatments which require estimates of the mean time spent in the PFS and PPS health-states. The methods can be applied to a single time-to-event outcome, and so have wide applicability in any field where time-to-event outcomes are reported, the proportional hazards assumption is in doubt, and survival curve shapes differ across studies and interventions. This article is protected by copyright. All rights reserved.

Estimating the natural indirect effect and the mediation proportion via the product method

Background

The natural indirect effect (NIE) and mediation proportion (MP) are two measures of primary interest in mediation analysis. The standard approach for mediation analysis is through the product method, which involves a model for the outcome conditional on the mediator and exposure and another model describing the exposure–mediator relationship. The purpose of this article is to comprehensively develop and investigate the finite-sample performance of NIE and MP estimators via the product method.

Methods

With four common data types with a continuous/binary outcome and a continuous/binary mediator, we propose closed-form interval estimators for NIE and MP via the theory of multivariate delta method, and evaluate its empirical performance relative to the bootstrap approach. In addition, we have observed that the rare outcome assumption is frequently invoked to approximate the NIE and MP with a binary outcome, although this approximation may lead to non-negligible bias when the outcome is common. We therefore introduce the exact expressions for NIE and MP with a binary outcome without the rare outcome assumption and compare its performance with the approximate estimators.

Results

Simulation studies suggest that the proposed interval estimator provides satisfactory coverage when the sample size ≥500 for the scenarios with a continuous outcome and sample size ≥20,000 and number of cases ≥500 for the scenarios with a binary outcome. In the binary outcome scenarios, the approximate estimators based on the rare outcome assumption worked well when outcome prevalence less than 5% but could lead to substantial bias when the outcome is common; in contrast, the exact estimators always perform well under all outcome prevalences considered.

Conclusions

Under samples sizes commonly encountered in epidemiology and public health research, the proposed interval estimator is valid for constructing confidence interval. For a binary outcome, the exact estimator without the rare outcome assumption is more robust and stable to estimate NIE and MP. An R package mediateP is developed to implement the methods for point and variance estimation discussed in this paper.

Supplementary Information

The online version contains supplementary material available at (10.1186/s12874-021-01425-4).

Association Between Neighborhood Disadvantage and Functional Well-being in Community-Living Older Persons

IMPORTANCE

Neighborhood disadvantage is a novel social determinant of health that could adversely affect the functional well-being of older persons. Deficiencies in resource-poor environments can potentially be addressed through social and public health interventions.

OBJECTIVE

To evaluate whether estimates of active and disabled life expectancy differ on the basis of neighborhood disadvantage after accounting for individual-level socioeconomic characteristics and other prognostic factors.

DESIGN, SETTING, AND PARTICIPANTS

This prospective longitudinal cohort study included 754 nondisabled community-living persons, aged 70 years or older, who were members of the Precipitating Events Project in south central Connecticut from March 1998 to June 2020.

MAIN OUTCOMES AND MEASURES

Disability in 4 essential activities of daily living (bathing, dressing, walking, and transferring) was assessed each month. Scores on the Area Deprivation Index, a census-based socioeconomic measure with 17 education, employment, housing quality, and poverty indicators, were obtained through linkages with the 2000 Neighborhood Atlas. Area Deprivation Index scores were dichotomized at the 80th state percentile to distinguish neighborhoods that were disadvantaged (81-100) from those that were not (1-80).

RESULTS

Among the 754 participants, the mean (SD) age was 78.4 (5.3) years, and 487 (64.6%) were female. Within 5-year age increments from 70 to 90, active life expectancy was consistently lower in participants from neighborhoods that were disadvantaged vs not disadvantaged, and these differences persisted and remained statistically significant after adjustment for individual-level race and ethnicity, education, income, and other prognostic factors. At age 70 years, adjusted estimates (95% CI) for active life expectancy (in years) were 12.3 (11.5-13.1) in the disadvantaged group and 14.2 (13.5-14.7) in the nondisadvantaged group. At each age, participants from disadvantaged neighborhoods spent a greater percentage of their projected remaining life disabled, relative to those from nondisadvantaged neighborhoods, with adjusted values (SE) ranging from 17.7 (0.8) vs 15.3 (0.5) at age 70 years to 55.0 (1.7) vs 48.1 (1.3) at age 90 years.

CONCLUSIONS AND RELEVANCE

In this prospective longitudinal cohort study, living in a disadvantaged neighborhood was associated with lower active life expectancy and a greater percentage of projected remaining life with disability. By addressing deficiencies in resource-poor environments, new or expanded social and public health initiatives have the potential to improve the functional well-being of community-living older persons and, in turn, reduce health disparities in the US.

Treatment effect measures for culture conversion endpoints in phase IIb tuberculosis treatment trials

Phase IIb trials of tuberculosis therapy rely on early biomarkers of treatment effect. Despite limited predictive ability for clinical outcomes, culture conversion, the event in which an individual previously culture positive for Mycobacterium tuberculosis yields a negative culture after initiating treatment, is a commonly used endpoint. Lack of consensus on how to define the outcome and corresponding measure of treatment effect complicates interpretation and limits between-trial comparisons. We review common analytic approaches to measuring treatment effect and introduce difference in restricted mean survival times as an alternative to identify faster times to culture conversion and express magnitude of effect on the time scale. Findings from the PanACEA MAMSTB trial are reanalyzed as an illustrative example. In a systematic review we demonstrate variability in analytic approaches, sampling strategies, and outcome definitions in phase IIb tuberculosis trials. Harmonization would allow for larger meta-analyses, and may help expedite advancement of new TB therapeutics.

Meta-regression methods to characterize evidence strength using meaningful-effect percentages conditional on study characteristics

Meta-regression analyses usually focus on estimating and testing differences in average effect sizes between individual levels of each meta-regression covariate in turn. These metrics are useful but have limitations: they consider each covariate individually, rather than in combination, and they characterize only the mean of a potentially heterogeneous distribution of effects. We propose additional metrics that address both limitations. Given a chosen threshold representing a meaningfully strong effect size, these metrics address the questions: "For a given joint level of the covariates, what percentage of the population effects are meaningfully strong?" and "For any two joint levels of the covariates, what is the difference between these percentages of meaningfully strong effects?" We provide semiparametric methods for estimation and inference and assess their performance in a simulation study. We apply the proposed methods to meta-regression analyses on memory consolidation and on dietary behavior interventions, illustrating how the methods can provide more information than standard reporting alone. To facilitate implementing the methods in practice, we provide reporting guidelines and simple R code.

Modifiable Lifestyle Recommendations and Mortality in Denmark: A Cohort Study

INTRODUCTION

Modifiable lifestyle behaviors represent a central target for public health interventions. This study investigates the association between adherence to 4 modifiable lifestyle recommendations and all-cause, cancer, or cardiovascular disease mortality.

METHODS

Investigators used data from the Danish Diet, Cancer and Health cohort (1993-2013; N=54,276). Lifestyle recommendations included smoking (never smoking), diet (adherence to 6 national food-based dietary guidelines), alcohol consumption (≤7 units per week for women and ≤14 units per week for men), and physical activity (≥30 minutes per day of moderate-to-vigorous leisure-time physical activity). Pseudo-values were used to estimate the adjusted risk differences and 95% CIs for all-cause, cancer, or cardiovascular disease mortality. Data were analyzed in 2019-2020.

RESULTS

A total of 8,860 participants died during a median follow-up of 17.0 years. Adherence to all modifiable lifestyle recommendations was associated with an 18.46% (95% CI= -20.52%, -16.41%) lower absolute risk of all-cause mortality than no adherence. Never smokers had a 13.19% (95% CI= -13.95%, -12.44%) lower risk, those adhering to dietary guidelines (diet score ≥5) had a 7.52% (95% CI= -8.89%, -6.14%) lower risk, and those adhering to recommended levels of alcohol (2.11%, 95% CI= -2.75%, -1.48%) and physical activity (1.58%, 95% CI= -2.20%, -1.00%) had a lower risk than those who did not adhere. Stronger associations were observed in men than in women and in older than in middle-aged participants.

CONCLUSIONS

Findings suggest that adherence to modifiable lifestyle recommendations is associated with a lower risk of mortality from all causes, cancer, and cardiovascular disease, underlining the importance of supporting adherence to national guidelines for lifestyle recommendations.

Replacing Red Meat with Other Nonmeat Food Sources of Protein is Associated with a Reduced Risk of Type 2 Diabetes in a Danish Cohort of Middle-Aged Adults

BACKGROUND

Few cohort studies have modelled replacements of red meat with other sources of protein on subsequent risk of type 2 diabetes using dietary changes.

OBJECTIVES

To determine whether replacing red meat with other food sources of protein is associated with a lower risk of type 2 diabetes.

METHODS

We used data from the Danish Diet, Cancer, and Health cohort (n = 39,437) of middle-aged (55-72 years old) men and women who underwent 2 dietary assessments roughly 5 years apart to investigate dietary changes. The pseudo-observation method was used to model the average exposure effect of decreasing the intake of red meat while increasing the intake of either poultry, fish, eggs, milk, yogurt, cheese, whole grains, or refined grains on the subsequent 10-year risk of developing type 2 diabetes, compared with no changes in the intakes of these foods.

RESULTS

Replacing 1 serving/day (100 g/day) of red meat with 1 serving/day of eggs [risk difference (RD), -2.7%; 95% CI: -4.0 to -1.1%; serving size: 50 g/day], milk (RD, -1.2%; 95% CI: -2.1 to -0.4%; 200 g/day), yogurt (RD, -1.5%; 95% CI: -2.4 to -0.7%; 70 g/day), whole grains (RD, -1.7%; 95% CI: -2.5 to -0.9%; 30 g/day), or refined grains (RD, -1.2%; 95% CI: -2.0 to -0.3%; 30 g/day) was associated with a reduced risk of type 2 diabetes. Analyses of replacements with poultry or cheese, but not fish, also suggested a lower risk, but with wide CIs. After further adjustment for potential mediators (BMI, waist circumference, and history of hypertension or hypercholesterolemia), only the replacement with eggs was associated with a reduced risk (RD, -1.7%; 95% CI: -3.0 to -0.5%; 50 g/day).

CONCLUSIONS

Replacing red meat with eggs in middle-aged adults may reduce the risk of type 2 diabetes. In models not adjusted for potential mediators, replacing red meat with milk, yogurt, whole grains, or refined grains was also associated with a reduced risk of type 2 diabetes.

The Interaction Continuum

A common reason given for assessing interaction is to evaluate “whether the effect is larger in one group versus another”. It has long been known that the answer to this question is scale dependent: the “effect” may be larger for one subgroup on the difference scale, but smaller on the ratio scale. In this article, we show that if the relative magnitude of effects across subgroups is of interest then there exists an “interaction continuum” that characterizes the nature of these relations. When both main effects are positive then the placement on the continuum depends on the relative magnitude of the probability of the outcome in the doubly exposed group. For high probabilities of the outcome in the doubly exposed group, the interaction may be positive-multiplicative positive-additive, the strongest form of positive interaction on the “interaction continuum”. As the probability of the outcome in the doubly exposed group goes down, the form of interaction descends through ranks, of what we will refer to as the following: positive-multiplicative positive-additive, no-multiplicative positive-additive, negative-multiplicative positive-additive, negative-multiplicative zero-additive, negative-multiplicative negative-additive, single pure interaction, single qualitative interaction, single-qualitative single-pure interaction, double qualitative interaction, perfect antagonism, inverted interaction. One can thus place a particular set of outcome probabilities into one of these eleven states on the interaction continuum. Analogous results are also given when both exposures are protective, or when one is protective and one causative. The “interaction continuum” can allow for inquiries as to relative effects sizes, while also acknowledging the scale dependence of the notion of interaction itself.

Trends in excess mortality associated with atrial fibrillation over 45 years (Framingham Heart Study): community based cohort study

OBJECTIVE

To assess temporal trends in the association between newly diagnosed atrial fibrillation and death.

DESIGN

Community based cohort study.

SETTING

Framingham Heart Study cohort, in 1972-85, 1986-2000, and 2001-15 (periods 1-3, respectively), in Framingham, MA, USA.

PARTICIPANTS

Participants with no atrial fibrillation, aged 45-95 in each time period, and identified with newly diagnosed atrial fibrillation (or atrial flutter) during each time period.

MAIN OUTCOME MEASURES

The main outcome was all cause mortality. Hazard ratios for the association between time varying atrial fibrillation and all cause mortality were calculated with adjustment for time varying confounding factors. The difference in restricted mean survival times, adjusted for confounders, between participants with atrial fibrillation and matched referents at 10 years after a diagnosis of atrial fibrillation was estimated. Meta-regression was used to test for linear trends in hazard ratios and restricted mean survival times over the different time periods.

RESULTS

5671 participants were selected in time period 1, 6177 in period 2, and 6174 in period 3. Adjusted hazard ratios for all cause mortality between participants with and without atrial fibrillation were 1.9 (95% confidence interval 1.7 to 2.2) in time period 1, 1.4 (1.3 to 1.6) in period 2, and 1.7 (1.5 to 2.0) in period 3 (P_trend=0.70). Ten years after diagnosis of atrial fibrillation, the adjusted difference in restricted mean survival times between participants with atrial fibrillation and matched referents decreased by 31%, from -2.9 years (95% confidence interval -3.2 to -2.5) in period 1, to -2.1 years (-2.4 to -1.8) in period 2, to -2.0 years (-2.3 to -1.7) in period 3 (P_trend=0.03).

CONCLUSIONS

No evidence of a temporal trend in hazard ratios for the association between atrial fibrillation and all cause mortality was found. The mean number of life years lost to atrial fibrillation at 10 years had improved significantly, but a two year gap compared with individuals without atrial fibrillation still remained.

From Epidemiologic Knowledge to Improved Health: A Vision for Translational Epidemiology

Epidemiology should aim to improve population health; however, no consensus exists regarding the activities and skills that should be prioritized to achieve this goal. We performed a scoping review of articles addressing the translation of epidemiologic knowledge into improved population health outcomes. We identified 5 themes in the translational epidemiology literature: foundations of epidemiologic thinking, evidence-based public health or medicine, epidemiologic education, implementation science, and community-engaged research (including literature on community-based participatory research). We then identified 5 priority areas for advancing translational epidemiology: 1) scientific engagement with public health; 2) public health communication; 3) epidemiologic education; 4) epidemiology and implementation; and 5) community involvement. Using these priority areas as a starting point, we developed a conceptual framework of translational epidemiology that emphasizes interconnectedness and feedback among epidemiology, foundational science, and public health stakeholders. We also identified 2-5 representative principles in each priority area that could serve as the basis for advancing a vision of translational epidemiology. We believe an emphasis on translational epidemiology can help the broader field to increase the efficiency of translating epidemiologic knowledge into improved health outcomes and to achieve its goal of improving population health.

Effect heterogeneity and variable selection for standardizing causal effects to a target population

The participants in randomized trials and other studies used for causal inference are often not representative of the populations seen by clinical decision-makers. To account for differences between populations, researchers may consider standardizing results to a target population. We discuss several different types of homogeneity conditions that are relevant for standardization: Homogeneity of effect measures, homogeneity of counterfactual outcome state transition parameters, and homogeneity of counterfactual distributions. Each of these conditions can be used to show that a particular standardization procedure will result in an unbiased estimate of the effect in the target population, given assumptions about the relevant scientific context. We compare and contrast the homogeneity conditions, in particular their implications for selection of covariates for standardization and their implications for how to compute the standardized causal effect in the target population. While some of the recently developed counterfactual approaches to generalizability rely upon homogeneity conditions that avoid many of the problems associated with traditional approaches, they often require adjustment for a large (and possibly unfeasible) set of covariates.

Adjusted restricted mean survival times in observational studies

In observational studies with censored data, exposure-outcome associations are commonly measured with adjusted hazard ratios from multivariable Cox proportional hazards models. The difference in restricted mean survival times (RMSTs) up to a pre-specified time point is an alternative measure that offers a clinically meaningful interpretation. Several regression-based methods exist to estimate an adjusted difference in RMSTs, but they digress from the model-free method of taking the area under the survival function. We derive the adjusted RMST by integrating an adjusted Kaplan-Meier estimator with inverse probability weighting (IPW). The adjusted difference in RMSTs is the area between the two IPW-adjusted survival functions. In a Monte Carlo-type simulation study, we demonstrate that the proposed estimator performs as well as two regression-based approaches: the ANCOVA-type method of Tian et al and the pseudo-observation method of Andersen et al. We illustrate the methods by reexamining the association between total cholesterol and the 10-year risk of coronary heart disease in the Framingham Heart Study.

Risk related therapy in meta-analyses of critical care interventions: Bayesian meta-regression analysis

PURPOSE

The relationship between treatment efficacy and patient risk is explored in a series of meta-analyses from the critical care domain, focusing on mortality outcome.

METHODS

Systematic reviews of randomized controlled trials were identified by electronic search over the period 2002 to July 2018. A Bayesian meta-regression model was employed, using the risk difference metric to estimate the relationship between mortality difference and control arm risk, and estimate the mortality difference with and without adjusting for control arm risk.

RESULTS

Of 780 initially identified published systematic reviews, 113 had appropriate mortality data comprising 123 analysable groups. The 123 meta-analyses were pharmaceutical therapeutic (59.3%), non-pharmaceutical therapeutic (24.4%) and nutritional (16.3%), with a 25% overall average control arm mortality. In 25/123 (20%) analyses, meta-regression indicated significant baseline risk (Bayesian 95% credible intervals excluding zero). In all analyses, the relationship between risk-difference and control arm risk was negative indicating a positive treatment effect with increasing control arm risk. Adjusted estimates identified six studies with significant positive treatment effects, not evident until after adjustment for control arm risk.

CONCLUSION

Underlying risk-related therapy is apparent in meta-analyses of the critically-ill and identification is of importance to both the conduct and interpretation of these meta-analyses.

Evaluating Public Health Interventions: 8. Causal Inference for Time-Invariant Interventions

We provide an overview of classical and newer methods for the control of confounding of time-invariant interventions to permit causal inference in public health evaluations. We estimated the causal effect of gender on all-cause mortality in a large HIV care and treatment program supported by the President's Emergency Program for AIDS Relief in Dar es Salaam, Tanzania, between 2004 and 2012. We compared results from multivariable modeling, three propensity score methods, inverse-probability weighting, doubly robust methods, and targeted maximum likelihood estimation. Considerable confounding was evident, and, as expected by theory, all methods considered gave the same result, a statistically significant approximately 20% increased mortality rate in men. In general, there is no clear advantage of any of these methods for causal inference over classical multivariable modeling, from the point of view of either bias reduction or efficiency. Rather, given sufficient data to adequately fit the multivariable model to the data, multivariable modeling will yield causal estimates with the greatest statistical efficiency. All methods can adjust only for well-measured confounders-if there are unmeasured or poorly measured confounders, none of these methods will yield causal estimates.

Design of non-inferiority randomized trials using the difference in restricted mean survival times

Background/aims Non-inferiority trials with time-to-event outcomes are becoming increasingly common. Designing non-inferiority trials is challenging, in particular, they require very large sample sizes. We hypothesized that the difference in restricted mean survival time, an alternative to the hazard ratio, could lead to smaller required sample sizes. Methods We show how to convert a margin for the hazard ratio into a margin for the difference in restricted mean survival time and how to calculate the required sample size under a Weibull survival distribution. We systematically selected non-inferiority trials published between 2013 and 2016 in seven major journals. Based on the protocol and article of each trial, we determined the clinically relevant time horizon of interest. We reconstructed individual patient data for the primary outcome and fit a Weibull distribution to the comparator arm. We converted the margin for the hazard ratio into the margin for the difference in restricted mean survival time. We tested for non-inferiority using the difference in restricted mean survival time and hazard ratio. We determined the required sample size based on both measures, using the type I error risk and power from the original trial design. Results We included 35 trials. We found evidence of non-proportional hazards in five (14%) trials. The hazard ratio and the difference in restricted mean survival time were consistent regarding non-inferiority testing, except in one trial where the difference in restricted mean survival time led to evidence of non-inferiority while the hazard ratio did not. The median hazard ratio margin was 1.43 (Q1-Q3, 1.29-1.75). The median of the corresponding margins for the difference in restricted mean survival time was -21 days (Q1-Q3, -36 to -8) for a median time horizon of 2.0 years (Q1-Q3, 1-3 years). The required sample size according to the difference in restricted mean survival time was smaller in 71% of trials, with a median relative decrease of 8.5% (Q1-Q3, 0.4%-38.0%). Across all 35 trials, about 25,000 participants would have been spared from enrollment using the difference in restricted mean survival time compared to hazard ratio for trial design. Conclusion The margins for the hazard ratio may seem large but translate to relatively small differences in restricted mean survival time. The difference in restricted mean survival time offers meaningful interpretation and can result in considerable reductions in sample size. Restricted mean survival time-based measures should be considered more widely in the design and analysis of non-inferiority trials with time-to-event outcomes.

Evaluating Public Health Interventions: 7. Let the Subject Matter Choose the Effect Measure: Ratio, Difference, or Something Else Entirely

We define measures of effect used in public health evaluations, which include the risk difference and the risk ratio, the population-attributable risk, years of life lost or gained, disability-adjusted life years, quality-adjusted life years, and the incremental cost-effectiveness ratio. Except for the risk ratio, all of these are absolute effect measures. For constructing externally generalizable absolute measures of effect when there is superior fit of the multiplicative model, we suggest using the multiplicative model to estimate relative risks, which will often be obtained in simple linear form with no interactions, and then converting these to the desired absolute measure. The externally generalizable absolute measure of effect can be obtained by suitably standardizing to the risk factor distribution of the population to which the results are to be generalized. External generalizability will often be compromised when absolute measures are computed from study populations with risk factor distributions different from those of the population to whom the results are to be generalized, even when these risk factors are not confounders of the intervention effect.