Standards for Privacy of Individually Identifiable Health Information. Final Privacy Rule Preamble.. Cost/Benefit Analysis


Comment: Many commenters made general statements to the effect that the cost estimates for implementing the provisions of the proposed regulation were incomplete or greatly understated.

Response: The proposal, including the cost analysis, is, in effect, a first draft. The purpose of the proposal was to solicit public comment and to use those comments to refine the final regulation. As a result of the public comment, the Department has significantly refined our initial cost estimates for implementing this regulation. The cost analysis below reflects a much more complete analysis of the major components of the regulation than was presented in the proposal.

Comment: Numerous commenters noted that significant areas of potential cost had not been estimated and that if they were estimated, they would greatly increase the total cost of the regulation. Potential cost areas identified by various respondents as omitted from the analyses include the minimum disclosure requirements; the requisite monitoring by covered entities of business partners with whom they share private health information; creation of de-identified information; internal complaint processes; sanctions and enforcement; the designation of a privacy official and creation of a privacy board; new requirements for research/optional disclosures; and future litigation costs.

Response: We noted in the proposed rule that we did not have data from which to estimate the costs of many provisions, and solicited comments providing such data. The final analysis below reflects the best estimate possible for these areas, based on the information available. The data and the underlying assumptions are explained in the cost analysis section below.

Comment: A number of comments suggested that the final regulation be delayed until more thorough analyses could be undertaken and completed. One commenter stated that the Department should refrain from implementing the regulation until a more realistic assessment of costs could be made and include local governments in the process. Similarly, a commenter requested that the Department assemble an outside panel of health industry experts, including systems analysts, legal counsel, and management consultants to develop stronger estimates.

Response: The Department has engaged in extensive research, data collection and fact-finding to improve the quality of its economic analysis. This has included comments from and discussions with the kinds of experts one commenter suggested. The estimates represent a reasonable assessment of the policies proposed.

Comment: Several commenters indicated that the proposed regulation would impose significant new costs on providers' practices. Furthermore, they believe that it runs counter to the explicit statutory intent of HIPAA's Administrative Simplification provisions which require that "any standard adopted . . . shall be consistent with the objective of reducing the administrative costs of providing and paying for health care."

Response: As the Department explained in the Transactions Rule, this provision applies to the administrative simplification regulations of HIPAA in the aggregate. The Transactions Rule is estimated to save the health care system $29.9 billion in nominal dollars over ten years. Other regulations published pursuant to the administrative simplification authority in HIPAA, including the privacy regulation, will result in costs, but these costs are within the statutory directive so long as they do not exceed the $29.9 billion in estimated savings. Furthermore, as explained in the Transactions Rule, and the preamble to this rule, assuring privacy is essential to sustaining many of the advances that computers will provide. If people do not have confidence that their medical privacy will be protected, they will be much less likely to allow their records to be used for any purpose or might even avoid obtaining necessary medical care.

Comment: Several commenters criticized the omission of aggregate, quantifiable benefit estimates in the proposed rule. Some respondents argued that the analysis in the proposed rule used "de minimus" cost estimates to argue only that benefits would certainly exceed such a low barrier. These commenters further characterized the benefits analysis in the Notice of Proposed Rulemaking as "hand waving" used to divert attention from the fact that no real cost-benefit comparison is presented. Another commenter stated that the benefit estimates rely heavily on anecdotal and unsubstantiated inferences. This respondent believes that the benefit estimates are based on postulated, but largely unsubstantiated causal linkages between increased privacy and earlier diagnosis and medical treatment.

Response: The benefits of privacy are diffused and intangible but real. Medical privacy is not a good people buy or sell in a market; therefore, it is very difficult to quantify. The benefits discussion in the proposal reflects this difficulty. The examples presented in the proposal were meant to be illustrative of the benefits based on a few areas of medicine where some relevant data was available. Unfortunately, no commenters provided either a better methodological approach or better data for assessing the overall benefits of privacy. Therefore, we believe the analysis in the proposal represents a valid illustration of the benefits of privacy, and we do not believe it is feasible to provide an overall dollar estimate of the benefits of privacy in the aggregate.

Comment: One commenter criticized the benefit analysis as being incomplete because it did not consider the potential cost of new treatments that might be engendered by increased confidence in medical privacy resulting from the regulation.

Response: There is no data or model to reliably assess such long-term behavioral and scientific changes, nor to determine what portion of the increasingly rapid evolution of new improved treatments might stem from improved privacy protections. Moreover, to be complete, such analysis would have to include the savings that might be realized from earlier detection and treatment. It is not possible at this time to project the magnitude or even the direction of the net effects of the response to privacy that the commenter suggests.