Instrumental variable specifications and assumptions for longitudinal analysis of mental health cost offsets
 A. James O’Malley
 … show all 1 hide
Abstract
Instrumental variables (IVs) enable causal estimates in observational studies to be obtained in the presence of unmeasured confounders. In practice, a diverse range of models and IV specifications can be brought to bear on a problem, particularly with longitudinal data where treatment effects can be estimated for various functions of current and past treatment. However, in practice the empirical consequences of different assumptions are seldom examined, despite the fact that IV analyses make strong assumptions that cannot be conclusively tested by the data. In this paper, we consider several longitudinal models and specifications of IVs. Methods are applied to data from a 7year study of mental health costs of atypical and conventional antipsychotics whose purpose was to evaluate whether the newer and more expensive atypical antipsychotic medications lead to a reduction in overall mental health costs.
 Angrist, JD, Imbens, GW, Rubin, DB (1996) Identification of causal effects using instrumental variables (Disc: P456472). J. Am. Stat. Assoc. 91: pp. 444455 CrossRef
 Angrist, JD, Pischke, JS (2009) Mostly Harmless Econometrics: An Empiricist’s Companion. Princeton University Press, Princeton
 Anselin, L (1990) Some robust approaches to testing and estimation in spatial econometrics. Reg. Sci. Urban Econom. 20: pp. 141163 CrossRef
 Arellano, M, Bond, S (1991) Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Rev. Econom. Stud. 58: pp. 277297 CrossRef
 Arellano, M, Honore, B (2001) Panel Data: Some Recent Developments. Elsevier, Amsterdam
 Arendt, J.N., Holm, A: Simple estimators for Probit models with dummy endogenous variables. Under Review, http://www.sam.sdu.dk/~jna/publicationsfiler/Arendtholm.pdf (2008).
 Brito, C., Pearl, J. (2002) A new identification condition for recursive models with correlated errors. Struct. Equ. Model. 9: pp. 459474 CrossRef
 Dowd, B (2011) Separated at birth: statisticians, social scientists, and causality in health services research. Health Serv. Res. 46: pp. 397420 CrossRef
 Greenland, S (2003) Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Epidemiology 14: pp. 300306
 Haining, RP (1978) The moving average model for spatial interaction. Transac. Inst. Br. Geogr. 3: pp. 202225 CrossRef
 Hogan, JW, Lancaster, T (2004) Instrumental variables and inverse probability weighting for causal inference from longitudinal observational studies. Stat. Methods Med. Res. 13: pp. 1748 CrossRef
 Hsiao, C (2003) Analysis of Panel Data. Cambridge University Press, Cambridge CrossRef
 Huber, P.J: The behavior of maximum likelihood estimation under nonstandard conditions. In: LeCam, L.M, Neyman, J. (eds.) Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, vol. 1, pp. 221–233. University of California Press, Berkeley (1967)
 Imbens, G, Angrist, J (1994) Identification and estimation of local average treatment effects. Econometrica 62: pp. 467476 CrossRef
 Joffe, M.M., Small, D., Brunelli, S., Ten Have, T., Feldman, H. I: Extended instrumental variables estimation of overall effects. Int. J. Biostat. 4, Epub 2008 Apr 7 (2008)
 Kelejian, HH, Robinson, DP (1993) A suggested method of estimation for spatial interdependent models with autocorrelated errors, and an application to a county expenditure model. Pap Reg. Sci. 72: pp. 297312 CrossRef
 Land, KC, Deane, G On the largesample estimation of regression models with spatial or network effect terms: a twostage leastsquares approach. In: Marsden, PV eds. (1992) Sociological Methodology. Basil Blackwell, Ltd, Oxford, pp. 221248
 Liang, KY, Zeger, SL (1986) Longitudinal data analysis using generalized linear models. Biometrika 73: pp. 1322 CrossRef
 Lichtenberg, F (2001) Are the benefits of new drugs worth their costs. Health Aff. 20: pp. 4151
 McClellan, M, Newhouse, JP (1997) The marginal costeffectiveness of medical technology: a panel instrumentalvariables approach. J. Econom. 77: pp. 3964 CrossRef
 Morgan, SL, Winship, C (2007) Counterfactuals and Causal Inference. Cambridge University Press, Cambridge CrossRef
 O’Malley, AJ, Frank, RG, Normand, SLT (2011) Estimating cost Offsets of new medications: use of new antipsychotics and mental health costs for schizophrenia. Stat. Med. 30: pp. 19711988 CrossRef
 Pepe, MS, Anderson, GL (1994) A cautionary note on inference for marginal regression models with longitudinal data and general correlated response data. Commun. Stat. Simul. Comput. 23: pp. 939951 CrossRef
 Rosenbaum, PR, Rubin, DB (1983) The central role of the propensity score in observational studies for causal effects. Biometrika 70: pp. 4155 CrossRef
 Rosenheck, RA, Leslie, DL, Sindelar, J, Miller, EA, Lin, H, Stroup, TS, McEvoy, J, Davis, SM, Keefe, RSE, Swartz, M, Perkins, DO, Hsiao, JK, Lieberman, J (2006) Cost effectiveness of second generation antipsychotics and perphenazine in a randomized trial of treatment for chronic schizophrenia. Am. J. Psychiatry 163: pp. 20802089 CrossRef
 Sargan, JD (1958) The estimation of econometric relationships using instrumental variables. Econometrica 26: pp. 393415 CrossRef
 Sharkey, P, Elwert, F (2010) The legacy of disadvantage: multigenerational effects on cognitive ability. Am. J. Sociol. 116: pp. 19341981 CrossRef
 Small, DS (2007) Sensitivity analysis for instrumental variables regression with overidentifying restrictions. J. Am. Stat. Assoc. 102: pp. 10491058 CrossRef
 Stock JH, Yogo M (2002) Testing for Weak Instruments in Linear IV Regression. NBER Technical Working Papers 0284, National Bureau of Economic Research, Inc, URL http://ideas.repec.org/p/nbr/nberte/0284.html
 VanderWeele, TJ (2011) Sensitivity analysis for contagion effects in social networks. Sociol. Methods Res. 40: pp. 240255 CrossRef
 Vansteelandt, S (2007) On confounding, prediction and efficiency in the analysis of longitudinal and crosssectional clustered data. Scand. J. Stat. 34: pp. 478498 CrossRef
 White, H (1982) Maximum likelihood estimation of misspecified models. Econometrica 50: pp. 125 CrossRef
 Wooldridge, JM (2002) Econometric Analysis of Cross Section and Panel Data. MIT Press, Cambridge
 Wright, PG (1928) The tariff on animal and vegetable oils: Appendix B. MacMillan, New York
 Zeger, SL, Liang, KY (1986) Longitudinal data analysis for discrete and continuous outcomes. Biometrics 42: pp. 121130 CrossRef
 Title
 Instrumental variable specifications and assumptions for longitudinal analysis of mental health cost offsets
 Open Access
 Available under Open Access This content is freely available online to anyone, anywhere at any time.
 Journal

Health Services and Outcomes Research Methodology
Volume 12, Issue 4 , pp 254272
 Cover Date
 20121201
 DOI
 10.1007/s1074201200977
 Print ISSN
 13873741
 Online ISSN
 15729400
 Publisher
 Springer US
 Additional Links
 Topics
 Keywords

 Causal inference
 Exclusion restriction
 Fixed differences
 Instrumental variable
 Longitudinal
 Mental health costs
 Industry Sectors
 Authors

 A. James O’Malley ^{(1)}
 Author Affiliations

 1. Department of Health Care Policy, Harvard Medical School, 180 Longwood Avenue, Boston, MA, 021155899, USA