Response to Public Comments

OMB Revision Package (CMS-10316, OMB 0938-1113)

Summary of Public comments / responses

Two commenters submitted feedback on the PDP and MA Plan Disenrollment Reasons Survey. Both commenters recommended that CMS consider reducing the length of the disenrollment survey.

CMS has shortened the PDP and MA Plan Disenrollment Reasons Survey from the survey that was initially piloted in 2010 to the survey that is proposed for 2016. CMS has used information from the survey fielding to inform the shortening of the survey. The table below shows item counts for three versions of the survey (MA-PD, PDP, and MA-Only). Note that the MA-Only survey form was not fielded in the pilot phase.

	2010 Pilot	2013-2015 fielding	2016 (proposed)
MA-PD	78	74	63
PDP	68	65	54
MA-Only	NA	56	45

CMS has shortened the survey forms by approximately 20% since the 2010 pilot survey; however, additional cuts are more difficult given the need to report composite scores for reasons for disenrollment and standalone reasons for disenrollment items in the contract-level plan reports and as part of the data available to consumers on Medicare Plan Finder.

Annually, CMS works with the contractor that is implementing the disenrollment survey to examine whether there are problems with specific survey items based on responses to the survey. The goal is to improve survey screen-in rates and understand whether and how shortening the survey affects response rates, the mix of respondents, and reasons cited for disenrollment.

The two commenters also made specific remarks about item wording and response options as summarized in the table below.

Public Comments	Response
General comment #1: We urge CMS to consider shortening the MAPD disenrollment survey, as in current format the survey may come across as overwhelming and burdensome to the individual. Many of these questions could be combined and/or condensed, while still gathering the same amount of information needed.	Please see the response that precedes the table.
In addition, we noticed numerous inconsistencies relating to the numbering of the questions, phrasing, and the answer options.	Please see responses to specific examples below.
For instance, with regard to the numbering, there are a different number of questions proposed for English-speaking survey takers, versus Spanish-speaking ones. Also, the ordering of the numbers seems to be off in several instances.	The Spanish-language versions of the surveys include the following two items that do not appear on the English surveys. The additional items account for the difference in item numbers between English and Spanish-language survey forms: Did you ever need written information from the plan in Spanish? (Yes / No) [if no, skip next question] How often did the plan give you written information in Spanish? (never, sometimes, usually, always)
Respecting the inconsistencies with phrasing, for instance in some cases the survey uses "former health plan" and in others it uses "[plan marketing name]." We recommend that CMS consider substituting all "former health plan" references with "[plan marketing name]" in order to avoid confusion of the definition of "former health plan."	CMS believes that the use of the term “former health plan” may serve as a helpful reminder that the survey is focused on patient experiences with their former (and not current) plan. The recommendation also contradicts the instruction provided on page 3 of the survey:  “If you were not enrolled in [plan marketing name/contract number] recently, please answer the survey based on your experiences with the plan you had before you enrolled in your current plan.”  Using the “plan marketing name” throughout would be confusing for respondents who do not recognize the plan named as the former plan and these respondents would be more likely to not fill out the survey if it specifically refers to the plan by name throughout.
Finally, pertaining to the answer options, we noticed that some questions do not give the option of selecting "N/A," while others do.	Thank you for this comment. We agree and will revise the survey to include the “N/A” response option for consistency. This revision does not change any requirements or estimated burden.
Question 2: we recommend adding a generic pre-question stating "Did you voluntary choose to leave the plan or were you required to leave the plan by Medicare?" We believe adding this precursor will lessen beneficiary confusion with regard to what CMS is asking and will further CMS' efforts in streamlining the survey process.	Thank you for this comment. The proposed question is complex for beneficiaries to understand. We believe that it would be difficult for beneficiaries to be able to accurately say whether their reason to leave the plan was voluntary or not, and would be confused by language about being “required to leave the plan.”
Question 48: we suggest CMS add language asking the individual whether he or she left the plan because a provider or provider staff told them they should. In conducting our own disenrollment surveys, we have encountered instances where individuals have chosen to leave the plan due to provider influence, so adding this as an option to your survey would provide more clarity to your team with regard to how often providers and/or provider staff have a direct effect on beneficiaries choosing to leave a health plan.	Thank you for this comment. CMS periodically conducts interviews with Medicare beneficiaries to understand whether additional reasons for disenrollment are not currently captured in the survey; to date we have not heard this as a reason from beneficiaries. This is an area that CMS may consider exploring in future formative work with Medicare beneficiaries who have disenrolled.
Questions 73: it appears that questions 68-72, which have now been removed, may have led to this question to explain the definition of "person." If these were removed, we recommend this question be reworded with a preamble so that beneficiary confusion is eliminated.	The proposed item #62 (“How did that person help you?”) is immediately preceded by the question below: “Did someone help you complete this survey?” We believe the wording is clear and that the question does not need a preamble. CMS has not identified any problems with this.
General comment#2: CMS is proposing to remove a number of survey questions, which would reduce the total number of questions in the disenrollment reasons survey. The total number of questions in the English-version of the proposed survey is as follows: 63 (MA-PD survey), 54 (PDP survey), and 45 (MA-Only survey). We appreciate CMS’ proposal to shorten the length of the current survey. However, we believe that beneficiaries who have voluntarily disenrolled from a plan may still find the survey too long or time consuming and therefore choose not to respond. We recommend that CMS consider further reducing the number of questions in the survey to encourage beneficiary participation	Please see the response that precedes the table.
Getting Information or Help from Your Former Health Plan/ Getting Health Care and the Prescription Medicines You Needed from Your Former Health Plan. (MA-PD survey, pages 29-31). These sections of the proposed survey include sets of questions on a particular topic. If beneficiaries select the response “Yes,” they are directed to the next question in the set. If beneficiaries select the response “No,” they are instructed to skip the next question and move to the following set of questions. We have a concern with the fifth response option under Question #4 and how it interacts with Question #3. Under Question #3, beneficiaries are directed to Question #4 only if they respond that they did try to get information or help from their former plan’s customer service. However, the fifth response option under Question #4 indicates that the beneficiary did not try to get information or help from his/her former plan’s customer service, which seems inappropriate given the instructions under Question#3 to skip Question #4 in this case. We are concerned that this response option creates the potential for inconsistent answers. This problem also exists under Questions #6, 8, 14 and 16. We believe that CMS should either combine the questions within these sets or eliminate the fifth response option under the applicable questions. This comment also applies to comparable sections and questions under the PDP survey (pages 42-43) and the MA-Only survey (page 53).	CMS added a response option that mirrors the preceding screener question in these instances because respondents sometimes miss the skip instruction.
Your Experience with Insurance Agents, Brokers, or Plan Representatives. (Section removed from MA-PD, PDP and MA-Only surveys). CMS is proposing to remove the set of questions related to beneficiary experience with insurance agents, brokers, or plan representatives from the survey instrument. As indicated in our previous comments regarding this section of the survey, we believe such questions cannot be directly linked to a beneficiary’s decision to voluntarily disenroll and should therefore not be included in the survey. AHIP agrees with CMS’ proposal to remove these questions. Further, we believe that removing these questions is an important aspect of the agency’s efforts to shorten the length of the survey.	Thank you for these comments.

2

6/20/2016