We often seek to estimate the impact of an exposure naturally occurring or randomly assigned at the cluster-level. For example, the literature on neighborhood determinants of health continues to grow. Likewise, community randomized trials are applied to learn about real-world implementation, sustainability, and population effects of interventions with proven individual-level efficacy. In these settings, individual-level outcomes are correlated due to shared cluster-level factors, including the exposure, as well as social or biological interactions between individuals. To flexibly and efficiently estimate the effect of a cluster-level exposure, we present two targeted maximum likelihood estimators (TMLEs). The first TMLE is developed under a non-parametric causal model, which allows for arbitrary interactions between individuals within a cluster. These interactions include direct transmission of the outcome (i.e. contagion) and influence of one individual’s covariates on another’s outcome (i.e. covariate interference). The second TMLE is developed under a causal sub-model assuming the cluster-level and individual-specific covariates are sufficient to control for confounding. Simulations compare the alternative estimators and illustrate the potential gains from pairing individual-level risk factors and outcomes during estimation, while avoiding unwarranted assumptions. Our results suggest that estimation under the sub-model can result in bias and misleading inference in an observational setting. Incorporating working assumptions during estimation is more robust than assuming they hold in the underlying causal model. We illustrate our approach with an application to HIV prevention and treatment.
KawachiIBerkmanL. Neighborhoods and health, Oxford, NY: Oxford University Press, 2003.
2.
OakesJ. The (mis)estimation of neighborhood effects: causal inference for a practicable social epidemiology (with discussion). Soc Sci Med2004; 58: 1929–1952.
3.
SobelM. What do randomized studies of housing mobility demonstrate?: Causal inference in the face of interference. J Am Stat Assoc Asoc2006; 101: 1398–1407.
University of California, San Francisco. Sustainable East Africa Research in Community Health (SEARCH). ClinicalTrials.gov, http://clinicaltrials.gov/show/NCT01864603 (2013).
GardinerJLuoZRomanL. Fixed effects, random effects and GEE: what are the differences?Stat Med2009; 28: 221–239.
12.
HubbardAAhernJFleischerN, et al.To GEE or not to GEE comparing population average and mixed models for estimating the associations between neighborhood risk factors and health. Epidemiology2010; 21: 467–474.
13.
StephensATchetgen TchetgenEDeGruttolaV. Augmented generalized estimating equations for improving efficiency and validity of estimation in cluster randomized trials by leveraging cluster-level and individual-level covariates. Stat Med2012; 31: 915–930.
14.
StephensATchetgen TchetgenEDeGruttolaV. Locally efficient estimation of marginal treatment effects when outcomes are correlated: is the prize worth the chase?Int J Biostat2014; 10: 59–75.
15.
PearlJ. Probabilistic reasoning in intelligent systems, San Mateo, CA: Morgan Kaufmann, 1988.
16.
PearlJ. Causality: models, reasoning and inference, 2nd ed. New York, NY: Cambridge University Press, 2009.
17.
van der LaanMRoseS. Targeted learning: causal inference for observational and experimental data, London: Springer, 2011.
18.
CoxD. Planning of experiments, New York, NY: Wiley, 1958.
19.
RubinD. Randomization analysis of experimental data: the fisher randomization test comment. J Am Stat Assoc1980; 75: 591–593.
20.
HalloranMStruchinerC. Study designs for dependent happenings. Epidemiology1991; 2: 331–338.
21.
HalloranMStruchinerC. Causal inference in infectious diseases. Epidemiology1995; 6: 142–151.
22.
HudgensMHalloranM. Toward causal inference with interference. J Am Stat Assoc2008; 103: 832–842.
23.
SeamanSPavlouMCopasA. Review of methods for handling confounding by cluster and informative cluster size in clustered data. Stat Med2014; 33: 5371–5387.
24.
RosenbaumPRubinD. The central role of the propensity score in observational studies for causal effects. Biometrika1983; 70: 41–55.
25.
AbadieAImbensG. Large sample properties of matching estimators for average treatment effects. Econometrica2006; 74: 235–267.
26.
SekhonJ. Multivariate and propensity score matching software with automated balance optimization: the matching package for R. J Stat Software2011; 42: 1–52.
27.
RobinsJ. A new approach to causal inference in mortality studies with sustained exposure periods–application to control of the healthy worker survivor effect. Math Model1986; 7: 1393–1512.
28.
TaubmanSRobinsJMittlemanM, et al.Intervening on risk factors for coronary heart disease: an application of the parametric G-formula. Int J Epidemiol2009; 38: 1599–1611.
29.
YoungJCainLRobinsJ, et al.Comparative effectiveness of dynamic treatment regimes: An application of the parametric g-formula. Stat Biosci2011; 3: 119–143.
30.
SnowdenJRoseSMortimerK. Implementation of g-computation on a simulated data set: demonstration of a causal inference technique. Am J Epidemiol2011; 173: 731–738.
31.
HorvitzDThompsonD. A generalization of sampling without replacement from a finite universe. J Am Stat Assoc1952; 47: 663–685.
32.
RobinsJHernánMBrumbackB. Marginal structural models and causal inference in epidemiology. Epidemiology2000; 11: 550–560.
33.
HernánMBrumbackBRobinsJ. Marginal structural models to estimate the causal effect of zidovudine on the survival of HIV-positive men. Epidemiology2000; 11: 561–570.
34.
BodnarLDavidianMSiega-RizA, et al.Marginal structural models for analyzing causal effects of time-dependent treatments: an application in perinatal epidemiology. Am J Epidemiol2004; 159: 926–934.
ShenCLiXLiL. Inverse probability weighting for covariate adjustment in randomized studies. Stat Med2014; 33: 555–568.
37.
Robins J and Rotnitzky A. Recovery of information and adjustment for dependent censoring using surrogate markers. In: Jewell NP, Dietz K and Farewell VT (eds) AIDS epidemiology. Basel: Birkhäuser, 1992.
38.
RobinsJRotnitzkyAZhaoL. Estimation of regression coefficients when some regressors are not always observed. J Am Stat Assoc1994; 89: 846–866.
39.
van der LaanMRobinsJ. Unified methods for censored longitudinal data and causality, Heidelberg: Springer-Verlag, 2003.
40.
BangHRobinsJ. Doubly robust estimation in missing data and causal inference models. Biometrics2005; 61: 962–972.
41.
van der LaanMRubinD. Targeted maximum likelihood learning. Int J Biostat2006; 2. Article 11.
42.
PragueMWangRStephensA, et al.Accounting for interactions and complex inter-subject dependency in estimating treatment effect in cluster-randomized trials with missing outcomes. Biometrics2016; 72: 1066–1077.
43.
CohenMChenYQMcCauleyM, et al.Prevention of HIV-1 infection with early antiretroviral therapy. New Engl J Med2011; 365: 493–505.
The TEMPRANO ANRS 12136 Study Group. A trial of early antiretrovirals and isoniazid preventive therapy in Africa. New Engl J Med2015; 373: 808–822.
47.
The INSIGHT START Study Group. Initiation of antiretroviral therapy in early asymptomatic HIV infection. New Engl J Med2015; 373: 795–807.
48.
CohenMChenYMccauleyM, et al.Final results of the HPTN 052 randomized controlled trial: antiretroviral therapy prevents HIV transmission. J Acquired Immune Deficiency Syndrome2015; 18: 20479–20479.
49.
MooreKvan der LaanM. Covariate adjustment in randomized trials with binary outcomes: targeted maximum likelihood estimation. Stat Med2009; 28: 39–64.
50.
GruberSvan der LaanM. tmle: An R package for targeted maximum likelihood estimation. J Stat Software2012; 51: 1–35.
51.
LendleSSchwabJPetersenM, et al.ltmle: An R package implementing targeted minimum loss-based estimation for longitudinal data. J Stat Software2017; 81: 1–21.
52.
GruberSvan der LaanM. A targeted maximum likelihood estimator of a causal effect on a bounded continuous outcome. Int J Biostat2010; 6. 1–14. DOI: 10.2202/ 1557-4679.1260.
53.
RoseSvan der LaanMUnderstanding TMLE. In: van der LaanMRoseS (eds). Targeted learning: causal inference for observational and experimental data, New York, NY: Springer, 2011.
54.
van der LaanMStarmansR. Entering the era of data science: targeted learning and the integration of statistics and computational data analysis. Adv Stat2014; 14: 502678–502678.
55.
GruberSvan der LaanM. Targeted minimum loss based estimator that outperforms a given estimator. Int J Biostat2012; 8: 1–20.
56.
RosenblumMvan der LaanM. Simple, efficient estimators of treatment effects in randomized trials using generalized linear models to leverage baseline variables. Int J Biostat2010; 6. 1–41. DOI:10.2202/1557-4679.1138.
57.
BickelPKlaassenCRitovY, et al.Efficient and adaptive estimation for semiparametric models, New York, NY: Springer, 1997.
58.
van der LaanMRobinsJ. Unified methods for censored longitudinal data and causality, New York, NY: Springer-Verlag, 2003.
59.
BalzerLvan der LaanMPetersenM, et al.Adaptive pre-specification in randomized trials with and without pair-matching. Stat Med2016; 35: 4528–4545.
60.
van der LaanMPolleyEHubbardA. Super learner. Stat Appl Genetics Mol Biol2007; 6: 25–25.
61.
PolleyERoseSvan der LaanMSuper learner. In: van der LaanMRoseS (eds). Targeted learning: causal inference for observational and experimental data, New York, NY: Springer, 2011.
TibshiraniR. Regression shrinkage and selection via the lasso. J R Stat Soc Ser B1996; 58: 267–288.
67.
SchnitzerMvan der LaanMMoodieE, et al.Effect of breastfeeding on gastrointestinal infection in infants: a targeted maximum likelihood approach for clustered longitudinal data. Ann Appl Stat2014; 8: 703–725.
68.
R Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing2017. http://www.R-project.org
69.
KarrerBNewmanM. Stochastic block models and community structure in networks. Phys Rev E2011; 83: 016107–016107.
70.
AndersonRMayR. Infectious diseases of humans: dynamics and control, New York, NY: Oxford University Press, 1992.
71.
GaolatheTWirthKHolmeM, et al.Botswana’s progress toward achieving the 2020 UNAIDS 90-90-90 antiretroviral therapy and virological suppression goals: a population-based survey. Lancet HIV2016; 3: e221–e230. DOI: 10.1016/S2352-3018(16)00037-0.
72.
Fidler S, Floyd S, Yang B, et al. Towards the second UNAIDS target: Population-level ART coverage in HPTN 071 (PopART). In: Conference on retroviruses and opportunistic infections (CROI), Boston, MA, 22–25 February 2016.
73.
Petersen M, Balzer L, Kwarsiima D, et al. SEARCH test and treat study in Uganda and Kenya exceeds the UNAIDs 90-90-90 cascade target by achieving over 80% population-level viral suppression after 2 years. In: 21st International AIDS conference, Durban, South Africa, 18–22 July 2016.
GruberSvan der LaanM. An application of collaborative targeted maximum likelihood estimation in causal inference and genomics. Int J Biostat2010; 6. 1–29.
76.
FisherR. Statistical methods for research workers, 4th ed. Edinburgh: Oliver and Boyd Ltd., 1932.
77.
CochranW. Analysis of covariance: its nature and uses. Biometrics1957; 13: 261–281.
78.
CoxDMcCullaghP. Some aspects of analysis of covariance. Biometrics1982; 38: 541–561.
79.
TsiatisADavidianMZhangM, et al.Covariate adjustment for two-sample treatment comparisons in randomized clinical trials: a principled yet flexible approach. Stat Med2008; 27: 4658–4677.
80.
ChamieGClarkTKabamiJ, et al.A hybrid mobile HIV testing approach for population-wide HIV testing in rural East Africa. Lancet HIV2016; 3: e111–e119.
Friedman J, Hastie T and Tibshirani R. Regularization paths for generalized linear models via coordinate descent. J Stat Software 2010; 33: 1–22, http://CRAN.R-project.org/package=glmnet. R package version 1.1-5.
84.
Balzer L, Petersen M, Schwab J, et al. Estimating the impact of community-level interventions: the SEARCH trial and HIV prevention in Sub-Saharan Africa. In: Western North American Region (WNAR) of the International Biometric Society (IBS) conference, June 2012. Fort Collins, CO.
85.
MorozovaOCohenTCrawfordF. Risk ratios for contagious outcomes. J R Soc Interface2018. 15(20170696): 1–12.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
