Dataset Information

Choice Inconsistencies among the Elderly: Evidence from Plan Choice in the Medicare Part D Program: Reply.

ABSTRACT: We explore the in- and out- of sample robustness of tests for choice inconsistencies based on parameter restrictions in parametric models, focusing on tests proposed by Ketcham, Kuminoff and Powers (KKP). We argue that their non-parametric alternatives are inherently conservative with respect to detecting mistakes. We then show that our parametric model is robust to KKP's suggested specification checks, and that comprehensive goodness of fit measures perform better with our model than the expected utility model. Finally, we explore the robustness of our 2011 results to alternative normative assumptions highlighting the role of brand fixed effects and unobservable characteristics.

SUBMITTER: Abaluck J

PROVIDER: S-EPMC5695563 | biostudies-literature | 2017 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Choice Inconsistencies among the Elderly: Evidence from Plan Choice in the Medicare Part D Program: Reply.

Abaluck Jason J Gruber Jonathan J

The American economic review 20171201 12

We explore the in- and out- of sample robustness of tests for choice inconsistencies based on parameter restrictions in parametric models, focusing on tests proposed by Ketcham, Kuminoff and Powers (KKP). We argue that their non-parametric alternatives are inherently conservative with respect to detecting mistakes. We then show that our parametric model is robust to KKP's suggested specification checks, and that comprehensive goodness of fit measures perform better with our model than the expect ...[more]

PMID: 29170561

Similar Datasets

Project description:BackgroundAlthough the Medicare Part D coverage gap phase-out should reduce cost-related nonadherence (CRN) among seniors with diabetes, preferential generic prescribing may have already decreased CRN, while smaller numbers of patients using more costly branded oral anti-diabetic (OAD) medications remain vulnerable to CRN.ObjectiveTo estimate the effects of cost sharing in the Part D standard (non-LIS) benefit on adherence to different OAD classes, comparing two classes dominated by inexpensive generic medications and two by more costly branded medications.Design and patientsRetrospective cohort study using dispensed prescription data for elderly non-LIS (N=81,047) and LIS (low-income subsidy) (N=150,359) beneficiaries using same class OAD(s) in 2008 and 2009. Logistic regression modeled non-LIS likelihood; LIS and non-LIS patients matched using propensity outcome (N=38,054). Logistic regression, controlling for demographic and health status characteristics, modeled effects of non-LIS coverage on 2009 OAD class adherence.Main measuresMain outcome measures were within-class OAD coverage year adherence, with patients considered adherent when days supplied to calendar days ratio at least 0.8.Key resultsNon-LIS patients had 0.52 and 0.57 times the odds of branded-only DPP-4 Inhibitor (N=1,812; 95 % CI: 0.43, 0.63; P<0.001) and Thiazolidinedione (TZD) (N=6,290; 95 % CI: 0.52, 0.63; P<0.001) adherence. Most patients (N=32,510; 82 %) used OADs in primarily generic classes, where we found no significant (Biguanides; N=21,377) or small differences (Sulfonylureas/Glinides [N=19,240; OR: 0.91; 95 % CI: 0.86, 0.97; P=0.002]) in adherence odds. Crude adherence rates were sub-optimal when CRN was not a factor (Non-LIS/LIS: Biguanides: 65 %/65 %; Sulfonylureas/Glinides: 66 %/68 %; LIS: DPP-4 Inhibitors: 66 %; TZDs: 67 %).ConclusionsGap elimination would not affect generic, but should reduce branded OAD CRN. Branded copayments may continue to lead to CRN. Policy initiatives and benefit changes targeting both cost deterrents for patients with more complex disease and non-cost generic OAD underuse are recommended.

Dataset Information

Choice Inconsistencies among the Elderly: Evidence from Plan Choice in the Medicare Part D Program: Reply.

Publications

Choice Inconsistencies among the Elderly: Evidence from Plan Choice in the Medicare Part D Program: Reply.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets