Revisions Disenrollment - Response to Comments to OMB (4)_062813

Responses to Comments Received

Federal Register Notice on

CMS-10316

CMS received three comment letters on June 18, 2013, in response to the notice under the Paperwork Reduction Act concerning the “Implementation of the Medicare Prescription Drug Plan (PDP) and Medicare Advantage (MA) Plan Disenrollment Reasons Survey” (Form CMS-10316). As most of the comments were similar in nature, CMS will provide a summary response.

Comments on Survey Period and Sampling Window

There were three requests for clarification on the survey period and sampling window. There were some concerns about the recall bias and differing effects of year to year benefit changes across multiple contract years.

CMS Response

Although the survey was set for a 15-month period between December 2012 and February 2014, it is a rolling survey: a random sample of voluntary disenrollees at each contract will be surveyed as close as possible to the actual disenrollment. Each month, the sample is drawn from previous month’s disenrollment list. The sampled participants will receive 2 mailed packages within a 1-2 month window to reduce recall bias. Results will be analyzed by year (12 months) so there is no concern about the differing effects of year to year benefit changes.

Definition of Voluntary Disenrollees

There were two requests for clarification about who would be regarded as voluntary disenrollees and thus would be in the sampling pool.

CMS Response

CMS maintains a set of codes that indicate whether a beneficiary is voluntarily disenrolling, and all beneficiaries who disenroll for non-voluntary reasons will be excluded from the population from which the sample will be drawn. Voluntary disenrollment refers to a beneficiary either dropping prescription coverage or Medicare Advantage coverage entirely or willingly switching to another contract for coverage. The survey excludes beneficiaries who involuntarily disenrolled from contracts for eligibility reasons, moved out of service area, or are deceased. If a beneficiary moves from one plan to another within the same contract, he/she is not included in the survey. Low income Subsidy (LIS) reassignments are also excluded. There is interest at CMS in periodic surveying of this population for program improvement purposes. In order to minimize recall issues, CMS intends to survey voluntary disenrollees as close as possible to the time of their disenrollment.

Length of the Survey

Three comments were submitted expressing concerns that the survey was too long.

CMS Response

Based on standard metrics for survey administration time, CMS estimates it will take Medicare beneficiaries about 18 minutes to complete this survey. This length is shorter than the current Medicare CAHPS surveys, which achieved an overall response rate in 2009 of 61.8%. Separately, the response rates were: Medicare Advantage = 64.1%; the fee-for-service (FFS)-prescription drug = 57.7%; fee-for-service-only = 58.3%. Previous tests of different length versions of CAHPS surveys found that response rates were not sensitive to survey length. Should the survey be used beyond the currently funded data collection effort, content may be refined to meet the agency's information needs.

Percent of Disenrollees

One question was submitted requesting information on the percentage of disenrollees being surveyed per contract.

CMS Response

Sample size is not planned in terms of percentage of disenrollees by each contract. Per the previous CMS field test, it was concluded that in order to obtain reliable contract-level estimates, a minimum annual sample of 150 for each MA contract and 300 for each stand-alone PDP contract is needed to generate reliable estimates at the contract level. The average annual voluntary disenrollment rate is 11percent.

Contract Level Reports

One questioner supported the contract level reports, but asked for clarification on the timing.

CMS Response

The contract level reports will be available to the contracts by the end of August 2014.

Example on the First Page

One comment noted that the example on the page preceding the survey could be confusing, and therefore suggested the use of a different font or bolding for clarification.

CMS Response

CMS will take this suggestion into consideration.

Use of the Term “Monthly Fee”

Two comments suggested that the word “premium” was more familiar to respondents, and the survey’s use of the term “monthly fee” in Question 22 (MA-PD) could lead to confusion.

CMS Response

In our cognitive testing, some respondents did not understand the word of “premium,” so we switched to “monthly fee.” However, we understand this concern and will refine the survey question as follows to include both terms:

Q22. Some Medicare beneficiaries have to pay their health plan a monthly fee (also known as premium) out of their own pocket for coverage for health and prescription medicines.

Did you leave the plan because the monthly fee that the health plan charges to provide coverage for health care and prescription medicines went up?

 Yes

 No

Use of the Word “Dropping”

Two comments suggested that the word “dropping” in Question #2 was unclear.

CMS Response

CMS will keep the word “dropping” as it was evidenced in the cognitive testing that respondents understood the word with no ambiguity.

Question 19 “Rating Health Plan”

One comment noted that respondents might be confused by Question 19, as it followed a series of questions regarding prescriptions.

CMS Response

CMS will keep the question and order of the questions as is, since it tested well with respondents during the cognitive testing.

Primary Reason for Disenrollment

One comment noted that the survey does not provide a means to determine the primary reason for disenrollment.

CMS Response

Question 52 (MA-PD), shown below, asks respondents to check the single most important reason for disenrollment.

52. What was the one most important reason you left [PLAN_NAME]? (Check one.)

 Financial or cost reasons

 Problems getting prescription drugs through the plan

 Problems getting the care, tests, or treatment you needed through the plan

 Problems with plan not covering doctors or hospitals you wanted to see

 Problems getting information from the plan about prescription drugs

 Switched to another plan that offers better benefits or coverage

 Another reason. Please specify: _____________________

Primary Reason for Disenrollment (Question 52)

One comment suggested moving question 52 to the beginning of the survey.

CMS Response

It is important for respondents to have a chance to thoroughly review and answer all the specific reason questions before they rate the most important reason. The order of the questions tested well in cognitive testing. We will keep the current order.

Questions about Insurance Agents, Brokers, or Plan Representatives

One commenter questioned the rationale and purpose of questions 53 and 54, which asked about beneficiaries’ experience with insurance agents and brokers.

CMS Response

CMS acknowledges that these two questions did not directly ask about reasons for disenrollment. However, it is important for CMS to understand beneficiaries’ experiences with marketing activities.

Consistency between MA & PDP CAHPS and Disenrollment Survey

One commenter noted that questions 60 and 61 used 12 months as the recall period, while the same questions in MA & PDP CAHPS used 6 months.

CMS Response

Question 61, “In the past 12 months, have you seen a doctor or other health provider 3 or more times for the same condition or problem” is exactly the same question included in MA & PDP CAHPS.

Question 60, “In the last 12 months, how many different prescription medicines did you fill? (Don’t count the same prescriptions twice.)” is slightly different from the question included in MA & PDP CAHPS, which is “In the last 6 months, how many different prescription medicines did you fill or have refilled?” We identified problems with the question from MA & PDP CAHPS in cognitive testing—respondents were counting the number of refills separately instead of the number of medicines (that is, they were double-counting medicines). Therefore we removed the phrase "or have refilled" and added "(Don’t count the same prescriptions twice.)”   The rephrased question tested well in the cognitive testing.

Problems Getting Drugs Reimbursed

One comment suggested that plans should not be penalized because members did not check to see if their drugs were covered by the plans.

CMS response

Drug coverage is a key aspect of MA-PD and PDP contracts, and should be used as a benchmark to compare contracts. If a beneficiary leaves a contract due to dissatisfaction about drug coverage, this disenrollment reason is legitimate, and should be captured by the disenrollment reasons survey.