Publicity and Compliance in the Colombia Tutela

How to cite

Staton, Jeffrey; Gauri, Varun. 2015. "Publicity and Compliance in the Colombia Tutela." AEA RCT Registry. AEARCTR-0000571. https://doi.org/10.1257/rct.571-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Jeffrey Staton* , Varun Gauri

* Primary investigator

Registry ID
AEARCTR-0000571
DOI
https://doi.org/10.1257/rct.571-1.0
First registered
Last updated
Status
Completed
Trial dates
2013-10-01 – 2014-07-31
Countries
Colombia ()
Registry record
http://www.socialscienceregistry.org/trials/571

Abstract

The Colombian tutela provides a swift mechanism for resolving state violations of individual rights. The Colombian Constitutional Court has provided innovative solutions through tutela to a variety of systemic bureaucratic failures, yet, the practical impact of these innovations depends on the ability of claimants to obtain through a court order what they could not get on their own, and that depends on compliance. Compliance is commonly monitored in legal systems via indirect monitoring, where litigants are themselves responsible for alerting judges to problems. Recently courts have begun directly monitoring compliance by tracking their own orders. We conduct a cluster randomized trial designed to evaluate the effect of direct compliance monitoring. Specifically, we consider the effect of increasing health provider expectations that the Colombian public will become aware of their activities on compliance with tutela orders governing health rights. Our hypothesis is that providers that expect public scrutiny of their behavior will experience better compliance outcomes than providers who do not.

Intervention

The aim of the treatment was to inform Empresas Promotoras de Salud (EPS), health maintenance organizations responsible for the delivery of health services in Colombia, that their compliance record with respect to orders in tutela cases was being tracked and that it would be publicized upon review. We informed treated EPS's that we were working in partnership with the Constitutional Court. Treatment involved a letter addressed to the President and Managing Directors of a selected EPS's, as well as follow up discussions with any EPS that reached out to our research team. The letter explained the purpose of the study, reviewed the results of our pilot study reporting the overall average compliance rates in the health sector, and a promise to publish the results of the study in July of 2014.

Experimental design

We conduct a cluster-randomized field experiment. There are 32 clusters, each cluster reflecting a single EPS. EPS's are blocked by the health care regime that they manage. EPS's that manage insured in the contributing regime (where individuals contribute to their premiums) constitute one block. The other block includes EPS's that manage insured in the subsidized regime (where premiums are paid for by the state). Fifteen EPS's managed insured exclusively in the subsidized regime. Eleven managed insured exclusively in the contributing regime. Five EPS's managed the care of individuals in both regimes and the regime for a single EPS was unclear. We randomly assign these six EPS's to the subsidized or contributing regime blocks. We then randomize treatment within each block.

Design details

Randomization method
Randomization done in office by a computer.
Randomization unit
The randomization unit is an EPS.
Clusters
32
Observations
288
Arms
The sample size for each arm was 144.
Primary outcomes
The key outcome is a binary indicator of compliance problems, measured at the level or each tutela order. It takes on the value of 1 in the event that our research team determined (via a careful reading of each tutela case and an interview with the beneficiary of relief) that there was an instance of non-compliance, either partial or complete. The indicator also takes on the value of 1 in the event that there had been compliance but there was a delay that violated the terms of the order. Otherwise, the indicator is coded 0. We also have categorical measures of compliance and of timing, both judged subjectively by the interviewed litigant and by our team. In addition, we conduct semi-interviews with EPS's in the study and have qualitative measures of each EPS's protocol for dealing with tutela orders.

Cited by

No citation data available for this registration yet.

Keywords