Dataset Information

Two-Stage Residual Inclusion Estimation in Health Services Research and Health Economics.

ABSTRACT:

Objectives

Empirical analyses in health services research and health economics often require implementation of nonlinear models whose regressors include one or more endogenous variables-regressors that are correlated with the unobserved random component of the model. In such cases, implementation of conventional regression methods that ignore endogeneity will likely produce results that are biased and not causally interpretable. Terza et al. (2008) discuss a relatively simple estimation method that avoids endogeneity bias and is applicable in a wide variety of nonlinear regression contexts. They call this method two-stage residual inclusion (2SRI). In the present paper, I offer a 2SRI how-to guide for practitioners and a step-by-step protocol that can be implemented with any of the popular statistical or econometric software packages.

Study design

We introduce the protocol and its Stata implementation in the context of a real data example. Implementation of 2SRI for a very broad class of nonlinear models is then discussed. Additional examples are given.

Empirical application

We analyze cigarette smoking as a determinant of infant birthweight using data from Mullahy (1997).

Conclusion

It is hoped that the discussion will serve as a practical guide to implementation of the 2SRI protocol for applied researchers.

SUBMITTER: Terza JV

PROVIDER: S-EPMC5980262 | biostudies-literature | 2018 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Two-Stage Residual Inclusion Estimation in Health Services Research and Health Economics.

Terza Joseph V JV

Health services research 20170531 3

<h4>Objectives</h4>Empirical analyses in health services research and health economics often require implementation of nonlinear models whose regressors include one or more endogenous variables-regressors that are correlated with the unobserved random component of the model. In such cases, implementation of conventional regression methods that ignore endogeneity will likely produce results that are biased and not causally interpretable. Terza et al. (2008) discuss a relatively simple estimation ...[more]

PMID: 28568477

Similar Datasets

Project description:BackgroundParticipant recruitment is an ongoing challenge in health research. Recruitment may be especially difficult for studies of access to health care because, even among those who are in care, people using services least often also may be hardest to contact and recruit. Opt-out recruitment methods (in which potential participants are given the opportunity to decline further contact about the study (opt out) following an initial mailing, and are then contacted directly if they have not opted out within a specified period) can be used for such studies. However, there is a dearth of literature on the effort needed for effective opt-out recruitment.MethodsIn this paper we describe opt-out recruitment procedures for two studies on access to health care within the U.S. Department of Veterans Affairs. We report resource requirements for recruitment efforts (number of opt-out packets mailed and number of phone calls made). We also compare the characteristics of study participants to potential participants via t-tests, Fisher's exact tests, and chi-squared tests.ResultsRecruitment rates for our two studies were 12 and 21%, respectively. Across multiple study sites, we had to send between 4.3 and 9.2 opt-out packets to recruit one participant. The number of phone calls required to arrive at a final status for each potentially eligible Veteran (i.e. study participation or the termination of recruitment efforts) were 2.9 and 6.1 in the two studies, respectively. Study participants differed as expected from the population of potentially eligible Veterans based on planned oversampling of certain subpopulations. The final samples of participants did not differ statistically from those who were mailed opt-out packets, with one exception: in one of our two studies, participants had higher rates of mental health service use in the past year than did those mailed opt-out packets (64 vs. 47%).ConclusionsOur results emphasize the practicality of using opt-out methods for studies of access to health care. Despite the benefits of these methods, opt-out alone may be insufficient to eliminate non-response bias on key variables. Researchers will need to balance considerations of sample representativeness and feasibility when designing studies investigating access to care.

Project description:BackgroundDiagnostic testing and treatment recommendations can vary when medical care is sought by individuals for low back pain (LBP), leading to variation in quality and costs of care. We examine how the first provider seen by an individual at initial diagnosis of LBP influences downstream utilization and costs.MethodsUsing national private health insurance claims data, individuals age 18 or older were retrospectively assigned to cohorts based on the first provider seen at the index date of LBP diagnosis. Exclusion criteria included individuals with a diagnosis of LBP or any serious medical conditions or an opioid prescription recorded in the 6 months prior to the index date. Outcome measures included use of imaging, back surgery rates, hospitalization rates, emergency department visits, early- and long-term opioid use, and costs (out-of-pocket and total costs of care) twelve months post-index date. We used a two-stage residual inclusion (2SRI) estimation approach comparing copay for the initial provider visit and differential distance as the instrumental variable to reduce selection bias in the choice of first provider, controlling for demographics.ResultsAmong 3,799,593 individuals, cost and utilization varied considerably based on the first provider seen by the patient. Copay and differential distance provided similar results, with copay preserving a greater sample size. The frequency of early opioid prescription was significantly lower when care began with an acupuncturist or chiropractor, and highest for those who began with an emergency medicine physician or advanced practice registered nurse (APRN). Long-term opioid prescriptions were low across most providers except physical medicine and rehabilitation physicians and APRNs. The frequency and time to serious illness varied little across providers. Total cost of care was lowest when starting with a chiropractor ($5093) or primary care physician ($5660), and highest when starting with an orthopedist ($9434) or acupuncturist ($9205).ConclusionThe first provider seen by individuals with LBP was associated with large differences in health care utilization, opioid prescriptions, and cost while there were no differences in delays in diagnosis of serious illness.

Dataset Information

Two-Stage Residual Inclusion Estimation in Health Services Research and Health Economics.

Objectives

Study design

Empirical application

Conclusion

Publications

Two-Stage Residual Inclusion Estimation in Health Services Research and Health Economics.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets