Sage Journals: Discover world-class research

Abstract

Many public health interventions are conducted in settings where individuals are connected and the intervention assigned to some individuals may spill over to other individuals. In these settings, we can assess: (a) the individual effect on the treated, (b) the spillover effect on untreated individuals through an indirect exposure to the intervention, and (c) the overall effect on the whole population. Here, we consider an egocentric network-based randomized design in which a set of index participants is recruited and randomly assigned to treatment, while data are also collected on their untreated network members. Such a design is common in peer education interventions conceived to leverage behavioral influence among peers. Using the potential outcomes framework, we first clarify the assumptions required to rely on an identification strategy that is commonly used in the well-studied two-stage randomized design. Under these assumptions, causal effects can be jointly estimated using a regression model with a block-diagonal structure. We then develop sample size formulas for detecting individual, spillover, and overall effects for single and joint hypothesis tests, and investigate the role of different parameters. Finally, we illustrate the use of our sample size formulas for an egocentric network-based randomized experiment to evaluate a peer education intervention for HIV prevention.

Keywords

Casual inference design of the experiments interference sample size calculations social networks

Get full access to this article

View all access options for this article.

References

Cox

. Planning of Experiments. New York: Wiley, 1958.

Rosenbaum

. Interference between units in randomized experiments. J Am Stat Assoc 2007; 102: 191–200.

Rubin

. Estimating causal effects of treatments in randomized and non randomized studies. J Educ Psychol 1974; 66: 688–701.

Ross

. An application of the theory of probabilities to the study of a priori pathometry.—part i. Proc R Soc London Ser A, Contain Papers Math Phys Character 1916; 92: 204–230.

Forastiere

Airoldi

Mealli

. Identification and estimation of treatment and interference effects in observational studies on networks. J Am Stat Assoc 2020; 116: 1–18.

Halloran

Hudgens

. Dependent happenings: A recent methodological review. Curr Epidemiol Rep 2016; 3: 297–305.

Hudgens

Halloran

. Towards causal inference with interference. J Am Stat Assoc 2008; 103: 832–842.

Liu

Hudgens

. Large sample randomization inference of causal effects in the presence of interference. J Am Stat Assoc 2014; 109: 288–301.

Liu

Hudgens

Becker-Dreps

. On inverse probability-weighted estimators in the presence of interference. Biometrika 2016; 103: 829–842.

10.

Perez-Heydrich

Hudgens

Halloran

, et al. Assessing effects of cholera vaccination in the presence of interference. Biometrics 2014; 70: 731–741.

11.

Sinclair

McConnell

Green

. Detecting spillover effects: design and analysis of multilevel experiments. Am J Pol Sci 2012; 56: 1055–1069.

12.

Tchetgen Tchetgen

VanderWeele

. On causal inference in the presence of interference. Stat Methods Med Res 2012; 21: 55–75.

13.

Aronow

Samii

. Estimating average causal effects under general interference, with application to a social network experiment. Ann Appl Stat 2017; 11: 1912–1947.

14.

Lee

Buchanan

Forastiere

KNVL

, et al. Estimating causal effects of non-randomized hiv prevention interventions with interference in network-based studies among people who inject drugs. Ann Appl Stat In Press 2023; 17: 2165–2191.

15.

Leung

. Treatment and spillover effects under network interference. Rev Econ Stat 2020; 102: 368–380.

16.

Loh

Hudgens

Clemens

, et al. Randomization inference with general interference and censoring. Biometrics 2018; 76: 235–245.

17.

Ogburn

Sofrygin

Diaz

, et al. Causal inference for social network data. arXiv preprint arXiv:1705.08527, 2017.

18.

Sofrygin

van der Laan

. Semi-parametric estimation and inference for the mean outcome of the single time-point intervention in a causally connected population. J Causal Inference 2016; 5: 20160003.

19.

Sävje

Aronow

Hudgens

. Average treatment effects in the presence of unknown interference. Ann Stat 2021; 49: 673–701.

20.

Tchetgen Tchetgen

Fulcher

Shpitser

. Auto-g-computation of causal effects on a network. J Am Stat Assoc 2020; 116: 833–844.

21.

Buchanan

Vermund

Friedman

, et al. Assessing individual and disseminated effects in network-randomized studies. Am J Epidemiol 2018; 187: 2449–2459.

22.

Davey-Rothwell

Tobin

Yang

, et al. Results of a randomized controlled trial of a peer mentor HIV/STI prevention intervention for women over an 18 month follow-up. AIDS Behav 2011; 15: 1654–1663.

23.

Latkin

Donnell

Sherman

DMS

, et al. The efficacy of a network intervention to reduce HIV risk behaviors among drug users and risk partners in Chiang Mai, Thailand and Philadelphia, USA. Soc Sci Med 2009; 68: 740–748.

24.

Tobin

Kuramoto

Davey-Rothwell

, et al. The step into action study: a peer-based, personal risk network-focused HIV prevention intervention with injection drug users in Baltimore, Maryland. Addict (Abingdon, Engl) 2010; 106: 366–375.

25.

Donner

Klar

. Design and analysis of cluster randomization trials in health research. London: Arnold, 2000.

26.

Hayes

Moulton

. Cluster randomised trials. New York: Chapman & Hall/CRC Press, 2009.

27.

Hemming

Eldridge

Forbes

, et al. How to design efficient cluster randomised trials. BMJ 2017; 358.

28.

Murray

. Design and Analysis of Group-Randomized Trials. New York, NY: Oxford University Press, 1998.

29.

Raudenbush

. Statistical analysis and optimal design for cluster randomized trials. Psychol Methods 1997; 2: 173–185.

30.

Walters

Jacques

dos Anjos Henriques-Cadby

, et al. Sample size estimation for randomised controlled trials with repeated assessment of patient-reported outcomes: what correlation between baseline and follow-up outcomes should we assume?. Trials 2019; 20: 1–16.

31.

Wittes

. Sample size calculations for randomized controlled trials. Epidemiol Rev 2002; 24: 39–53.

32.

Baird

Bohren

McIntosh

, et al. Optimal design of experiments in the presence of interference. Rev Econ Stat 2018; 100: 844–860.

33.

Jiang

Imai

Malani

. Statistical inference and power analysis for direct and spillover effects in two-stage randomized experiments. Biometrics 2023; 3: 2370–2381.

34.

Brookes

Whitely

Egger

, et al. Subgroup analyses in randomized trials: risks of subgroup-specific analyses: power and sample size for the interaction test. J Clin Epidemiol 2004; 57: 229–236.

35.

Shieh

. Detecting interaction effects in moderated multiple regression with continuous variables power and sample size considerations. Organ Res Methods 2009; 12: 510–528.

36.

Perry

Pescosolido

Borgatti

. Egocentric network analysis: foundations, methods, and models. Cambridge: Cambridge University Press, 2018.

37.

Hernán

Robins

. Causal Inference: What If. Boca Raton: Chapman & Hall/CRC Press, 2020.

38.

Pinheiro

Bates

. Mixed-effects models in S and S-PLUS. Headquarters: Springer Science & Business Media, 2000.

39.

Tian

Esserman

Tong

, et al. Sample size calculation in hierarchical 2

\times

2 factorial trials with unequal cluster sizes. Stat Med 2022; 41: 645–664.

40.

Manatunga

Hudgens

Chen

. Sample size estimation in cluster randomized studies with varying cluster size. Biomet J: J Mathemat Methods Biosci 2001; 43: 75–86.

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.

0.00 MB

0.23 MB

Design of egocentric network-based studies to estimate causal effects under interference

Abstract

Keywords

Get full access to this article

References

Supplementary Material