A selective review of the first 20 years of instrumental variables models in health-services research and medicine

Abstract

Background:

The method of instrumental variables (IV) is useful for estimating causal effects. Intuitively, it exploits exogenous variation in the treatment, sometimes called natural experiments or instruments. This study reviews the literature in health-services research and medical research that applies the method of instrumental variables, documents trends in its use, and offers examples of various types of instruments.

Methods:

A literature search of the PubMed and EconLit research databases for English-language journal articles published after 1990 yielded a total of 522 original research articles. Citations counts for each article were derived from the Web of Science. A selective review was conducted, with articles prioritized based on number of citations, validity and power of the instrument, and type of instrument.

Results:

The average annual number of papers in health services research and medical research that apply the method of instrumental variables rose from 1.2 in 1991–1995 to 41.8 in 2006–2010. Commonly-used instruments (natural experiments) in health and medicine are relative distance to a medical care provider offering the treatment and the medical care provider’s historic tendency to administer the treatment. Less common but still noteworthy instruments include randomization of treatment for reasons other than research, randomized encouragement to undertake the treatment, day of week of admission as an instrument for waiting time for surgery, and genes as an instrument for whether the respondent has a heritable condition.

Conclusion:

The use of the method of IV has increased dramatically in the past 20 years, and a wide range of instruments have been used. Applications of the method of IV have in several cases upended conventional wisdom that was based on correlations and led to important insights about health and healthcare. Future research should pursue new applications of existing instruments and search for new instruments that are powerful and valid.

Keywords::

• Instrumental variables
• Econometrics
• Methods
• Health

Transparency

Declaration of funding

Funding for this paper was provided by Novo Nordisk, Inc., and the Cornell University Institute on Health Economics, Health Behaviors, and Disparities.

Declaration of financial/other relationships

JC has disclosed that he has no significant relationships with or financial interests in any commercial companies related to this study or article. JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

JC thanks three anonymous referees, Mette Hammer and Neil Wintfeld for their helpful comments. JC also thanks Anna Choi, Mariann Lucas, and Jeffrey Swigert for research assistance, and Eric Maroney for editorial assistance.

Notes

*This paper focuses on the model of instrumental variables, but there are other methods for estimating causal effects, for example: difference-in-differences models (see, e.g., Angrist and Pischke4, section 5.2), propensity score matching (see, e.g., Morgan and Winship8, chapter 48), and regression discontinuity designs (see, e.g., Angrist and Pischke4, chapter 6).

*This is not to say that use of IV has increased more than use of every other method of estimating causal effects. A search of PubMed indicates that articles satisfying the search term ‘propensity score matching’ rose from zero in 1991–1995 to over 350 in 2006–2010.

†The method of instrumental variables is probably most widely used in economics, but is also used in statistics13, epidemiology7, sociology8, and political science14.

*At the time of their writing (1994), the authors could accurately say that ‘The method, instrumental variables (IV) estimation, is well known in econometrics but generally has not been applied to estimate relationships between medical treatments and health outcomes’ (McClellan et al.9, pp. 859–60). Table 2 and Figure 1 in this paper confirm that statement.

*Few such articles appear to exist; the Soc Abstracts database was searched for journal articles satisfying the terms ‘instrumental variables’ and (‘health’ or ‘medicine’), and there were only 43 results for the entire period 1990–2013. Many of those were also included in the EconLit or PubMed databases and, thus, were included in this review.