Skip to main content
SpringerLink
Log in

Instrumental variable estimation of early treatment effect in randomized screening trials

  • Published:
Lifetime Data Analysis Aims and scope Submit manuscript

Abstract

The primary analysis of randomized screening trials for cancer typically adheres to the intention-to-screen principle, measuring cancer-specific mortality reductions between screening and control arms. These mortality reductions result from a combination of the screening regimen, screening technology and the effect of the early, screening-induced, treatment. This motivates addressing these different aspects separately. Here we are interested in the causal effect of early versus delayed treatments on cancer mortality among the screening-detectable subgroup, which under certain assumptions is estimable from conventional randomized screening trial using instrumental variable type methods. To define the causal effect of interest, we formulate a simplified structural multi-state model for screening trials, based on a hypothetical intervention trial where screening detected individuals would be randomized into early versus delayed treatments. The cancer-specific mortality reductions after screening detection are quantified by a cause-specific hazard ratio. For this, we propose two estimators, based on an estimating equation and a likelihood expression. The methods extend existing instrumental variable methods for time-to-event and competing risks outcomes to time-dependent intermediate variables. Using the multi-state model as the basis of a data generating mechanism, we investigate the performance of the new estimators through simulation studies. In addition, we illustrate the proposed method in the context of CT screening for lung cancer using the US National Lung Screening Trial data.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

Data availability

The data that support the findings of this study are available from the NCI upon request. Restrictions apply the the availability of these data which were used under agreement for this study.

References

  • Aalen OO, Cook RJ, Røysland K (2015) Does Cox analysis of a randomized survival study yield a causal treatment effect? Lifetime Data Anal 21:579–593

    Article  MathSciNet  MATH  Google Scholar 

  • Altstein LL, Li G (2013) Latent subgroup analysis of a randomized clinical trial through a semiparametric accelerated failure time mixture model. Biometrics 69:52–61

    Article  MathSciNet  MATH  Google Scholar 

  • Altstein LL, Li G, Elashoff RM (2011) A method to estimate treatment efficacy among latent subgroups of a randomized clinical trial. Stat Med 30:709–717

    Article  MathSciNet  Google Scholar 

  • Angrist JD, Imbens GW, Rubin DB (1996) Identification of causal effects using instrumental variables. J Am Stat Assoc 91:444–455

    Article  MATH  Google Scholar 

  • Baiocchi M, Cheng J, Small DS (2014) Instrumental variable methods for causal inference. Stat Med 33:2297–2340

    Article  MathSciNet  Google Scholar 

  • Baker SG (1998) Analysis of survival data from a randomized trial with all-or-none compliance: estimating the cost-effectiveness of a cancer screening program. J Am Stat Assoc 93:929–934

    Article  Google Scholar 

  • Baker SG, Lindeman KS (1994) The paired availability design: a proposal for evaluating epidural analgesia during labor. Stat Med 13(21):2269–2278

    Article  Google Scholar 

  • Baker SG, Kramer BS, Lindeman KS (2016) Latent class instrumental variables: a clinical and biostatistical perspective. Stat Med 35(1):147–160

    Article  MathSciNet  Google Scholar 

  • Beyersmann J, Allignol A, Schumacher M (2011) Competing risks and multistate models with R. Springer Science & Business Media

  • Borgan Ø (2014) Aalen-Johansen estimator. Statistics Reference Online, Wiley StatsRef, pp 1–13

    Google Scholar 

  • Burgess S, Small DS, Thompson SG (2017) A review of instrumental variable estimators for mendelian randomization. Stat Methods Med Res 26:2333–2355

    Article  MathSciNet  Google Scholar 

  • Cube von M, Schumacher M, Wolkewitz M (2019) Causal inference with multistate models–estimands and estimators of the population attributable fraction. J R Stat Soc: Series A (Statistics in Society)

  • de Wreede LC, Fiocco M, Putter H (2011) mstate: An R package for the analysis of competing risks and multi-state models. J Stat Softw 38:1–30, http://www.jstatsoft.org/v38/i07/

  • Gray B (2019) cmprsk: Subdistribution Analysis of Competing Risks. https://CRAN.R-project.org/package=cmprsk, r package version 2.2-9

  • Habbema D (2018) Statistical analysis and decision making in cancer screening. Eur J Epidemiol 33:433–435

    Article  Google Scholar 

  • Hanley JA, Njor SH (2018) Disaggregating the mortality reductions due to cancer screening: model-based estimates from population-based data. Eur J Epidemiol 33:465–472

    Article  Google Scholar 

  • Hess K, Gentleman R (2019) muhaz: Hazard Function Estimation in Survival Analysis. https://CRAN.R-project.org/package=muhaz, r package version 1.2.6.1

  • Howe CJ, Cole SR, Lau B, Napravnik S, Eron JJ Jr (2016) Selection bias due to loss to follow up in cohort studies. Epidemiology (Cambridge, Mass) 27:91

    Article  Google Scholar 

  • Hu P, Zelen M (1997) Planning clinical trials to evaluate early detection programmes. Biometrika 84:817–829

    Article  MathSciNet  MATH  Google Scholar 

  • Imbens GW, Angrist JD (1994) Identification and estimation of local average treatment effects. Econometrica 62:467–475

    Article  MATH  Google Scholar 

  • Lee S, Zelen M (2006) Chapter 11: a stochastic model for predicting the mortality of breast cancer. JNCI Monographs 2006:79–86

  • Liu Z, Hanley JA, Strumpf EC (2013) Projecting the yearly mortality reductions due to a cancer screening programme. J Med Screen 20:157–164

    Article  Google Scholar 

  • Liu Z, Hanley JA, Saarela O, Dendukuri N (2015) A conditional approach to measure mortality reductions due to cancer screening. Int Stat Rev 83:493–510

    Article  MathSciNet  Google Scholar 

  • Loeys T, Goetghebeur E (2003) A causal proportional hazards estimator for the effect of treatment actually received in a randomized trial with all-or-nothing compliance. Biometrics 59:100–105

    Article  MathSciNet  MATH  Google Scholar 

  • Loeys T, Goetghebeur E, Vandebosch A (2005) Causal proportional hazards models and time-constant exposure in randomized clinical trials. Lifetime Data Anal 11:435–449

    Article  MathSciNet  MATH  Google Scholar 

  • Mark SD, Robins JM (1993a) Estimating the causal effect of smoking cessation in the presence of confounding factors using a rank preserving structural failure time model. Stat Med 12:1605–1628

    Article  Google Scholar 

  • Mark SD, Robins JM (1993b) A method for the analysis of randomized trials with compliance information: an application to the multiple risk factor intervention trial. Control Clin Trials 14:79–97

    Article  Google Scholar 

  • Martinussen T, Vansteelandt S (2020) Instrumental variables estimation with competing risk data. Biostatistics 21:158–171

    Article  MathSciNet  Google Scholar 

  • Martinussen T, Nørbo Sørensen D, Vansteelandt S (2019) Instrumental variables estimation under a structural Cox model. Biostatistics 20:65–79

    Article  MathSciNet  Google Scholar 

  • McIntosh MW (1999) Instrumental variables when evaluating screening trials: estimating the benefit of detecting cancer by screening. Stat Med 18:2775–2794

    Article  Google Scholar 

  • Miettinen OS (2014) Toward Scientific Medicine. Springer, Berlin

    Book  Google Scholar 

  • Miettinen OS (2015) ‘screening’ for breast cancer: Misguided research misinforming public policies. Epidemiologic Methods 4:3–10

    Google Scholar 

  • Nie H, Cheng J, Small DS (2011) Inference for the effect of treatment on survival probability in randomized trials with noncompliance and administrative censoring. Biometrics 67:1397–1405

    Article  MathSciNet  MATH  Google Scholar 

  • NLST Research Team (2011) Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med 365:395–409

    Article  Google Scholar 

  • Oken MM, Hocking WG, Kvale PA, Andriole GL, Buys SS, Church TR, Crawford ED, Fouad MN, Isaacs C, Reding DJ, Weissfeld JL, Yokochi LA, O’Brien B, Ragard LR, Rathmell JM, Riley TL, Wright P, Caparaso N, Hu P, Izmirlian G, Pinsky PF, Prorok PC, Kramer BS, Miller AB, Gohagan JK, Berg CD (2011) Screening by chest radiograph and lung cancer mortality: the Prostate, Lung, Colorectal, and Ovarian (PLCO) randomized trial. JAMA 306:1865–1873

    Article  Google Scholar 

  • Putter H, Geskus RB, Fiocco M (2007) Tutorial in biostatistics: Competing risks and multi-state models. Stat Med 26:2389–2430

    Article  MathSciNet  Google Scholar 

  • R Core Team (2017) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, https://www.R-project.org/

  • Richardson A, Hudgens MG, Fine JP, Brookhart MA (2017) Nonparametric binary instrumental variable analysis of competing risks data. Biostatistics 18:48–61

    Article  MathSciNet  Google Scholar 

  • Robins JM (1994) Correcting for non-compliance in randomized trials using structural nested mean models. Commun Stat A-Theor 23:2379–2412

    Article  MathSciNet  MATH  Google Scholar 

  • Roemeling S, Roobol MJ, Otto SJ, Habbema DF, Gosselaar C, Lous JJ, Cuzick J, Schröder FH (2007) Feasibility study of adjustment for contamination and non-compliance in a prostate cancer screening trial. The Prostate 67:1053–1060

    Article  Google Scholar 

  • Saha S, Liu ZA, Saarela O (2018) Estimating case-fatality reduction from randomized screening trials. Epidemiologic Methods 7

  • Shen Y, Zelen M (1999) Parametric estimation procedures for screening programmes: stable and nonstable disease models for multimodality case finding. Biometrika 86:503–515

    Article  MathSciNet  MATH  Google Scholar 

  • Sung NY, Jun JK, Kim YN, Jung I, Park S, Kim GR, Nam CM (2019) Estimating age group-dependent sensitivity and mean sojourn time in colorectal cancer screening. J Med Screen 26:19–25

    Article  Google Scholar 

  • Swanson SA, Robins JM, Miller M, Hernán MA (2015) Selecting on treatment: a pervasive form of bias in instrumental variable analyses. Am J Epidemiol 181:191–197

    Article  Google Scholar 

  • White IR (2005) Uses and limitations of randomization-based efficacy estimators. Stat Methods Med Res 14:327–347

    Article  MathSciNet  MATH  Google Scholar 

  • Young JG, Stensrud MJ, Tchetgen Tchetgen EJ, Hernán MA (2020) A causal framework for classical statistical estimands in failure-time settings with competing events. Stat Med 39:1199–1236

    Article  Google Scholar 

  • Zelen M (1993) Optimal scheduling of examinations for the early detection of disease. Biometrika 80:279–293

    Article  MathSciNet  MATH  Google Scholar 

  • Zheng C, Dai R, Hari PN, Zhang MJ (2017) Instrumental variable with competing risk model. Stat Med 36:1240–1255

    Article  MathSciNet  Google Scholar 

Download references

Acknowledgements

The authors thank the National Cancer Institute (NCI) for access to NCI’s data collected by the National Lung Screening Trial. The statements contained herein are solely those of the authors and do not represent or imply concurrence or endorsement by the NCI. This work was supported by the Ontario Institute for Cancer Research through funding provided by the Government of Ontario (to SS) and the Natural Sciences and Engineering Research Council of Canada (to OS).

Author information

Authors and Affiliations

  1. Dalla Lana School of Public Health, University of Toronto, 155 College Street, Toronto, Ontario, M5T 3M7, Canada

    Sudipta Saha & Olli Saarela

  2. Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario, Canada

    Zhihui Liu

Authors
  1. Sudipta Saha
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Zhihui Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Olli Saarela
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Olli Saarela.

Ethics declarations

Conflict of interest

The authors declare that they have no competing interests.

Code availability

The R code to reproduce the simulation study are available as online supplementary material.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary material 1 (zip 4 KB)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Saha, S., Liu, Z. & Saarela, O. Instrumental variable estimation of early treatment effect in randomized screening trials. Lifetime Data Anal 27, 537–560 (2021). https://doi.org/10.1007/s10985-021-09527-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10985-021-09527-3

Keywords

Search

Navigation