Beneficiaries' Perceptions of New Medicare Health Plan Choice Print Materials

Abstract

This article presents findings from a study involving seven focus groups with aged and disabled Medicare beneficiaries in the Kansas City area regarding their impressions of a pilot version of the Medicare & You 1999 handbook and the Medicare Consumer Assessment of Health Plans Study (CAHPS^®) survey report. Beneficiaries generally had positive reactions to both booklets and viewed the handbook as an important reference tool. Based on the findings, we present policy recommendations for the development and dissemination of Medicare health plan information to beneficiaries.

Introduction

Today, more than 1 in 8 Americans, or 34.5 million adults, is age 65 or over. By 2030, older adults are projected to outnumber those under age 18 (Fowles, Duncker, and Greenberg, 2000). Medical advancements have increased our life span, but they cannot ensure quality of life (extending our health span). In fact, increasing numbers of older adults have chronic health conditions. In this context, making sure that older Americans understand their health care options and can make informed choices about their health insurance coverage is an increasingly important societal issue.

An intersection of forces—including the public policy environment, the dynamic health care financing and delivery market, and the challenges Medicare beneficiaries face in processing unfamiliar, complex information—all point to the need to develop effective ways to inform, assist, and educate beneficiaries about their Medicare plan choices. The Balanced Budget Act (BBA) of 1997 legislated new health insurance options for Medicare beneficiaries through the Medicare+Choice program; and required CMS for the first time to give beneficiaries local comparative health plan information on all options available to them. Medicare beneficiaries are now being offered a greater variety of health insurance options, to add to an already complex mix. For many beneficiaries, the choices may now include up to 10 standardized commercial supplemental plans, various forms of Medicare managed care plans, basic Medicare, employer-sponsored retirement supplemental options, and a Medicare private fee-for-service option.

This increasing complexity comes at a time when research finds that Medicare beneficiaries do not understand: (1) today's current health care system; (2) the differences between managed care (e.g., network) and fee-for-service insurance options; (3) that managed care plans are both insurer and care deliverer; or, (4) that plans play a role in access to and quality of care (Eppig and Poisal, 1996; Hibbard, Jewett, Engleman, and Tusler, 1998; McGee, Sofaer, and Kreling, 1996). This is not surprising considering that age 65 is the first time many persons have to navigate these choices on their own, without help from an employer (Frederick Sneiders Research, 1995; Gibbs, 1995). The Medicare+Choice plan lock-in (scheduled to become effective beginning in 2002) also would remove the safety valve feature that currently enables beneficiaries to change plans monthly if they make a plan choice that does not work well for them.

One result of the increased plan choices available is the large amount of information that beneficiaries need to process to make meaningful plan comparisons. Considering the different types of information needed to make an informed choice—cost, coverage, providers, quality, and other relevant comparative information—it is easy to foresee that beneficiaries can easily get lost in information. Some policy-minded researchers question whether the informed choice policy approach is even appropriate for Medicare beneficiaries, in light of empirical findings that many beneficiaries have difficulty using comparative plan information (Hibbard et al., 2001) and lack basic knowledge of their health plan choices (Hibbard et al., 1998). One thing is certain—more information, in and of itself, will not necessarily improve the quality of beneficiaries' plan decisions. The way that comparative plan information is presented to Medicare beneficiaries is key to their understanding and use of it and its influence on their decisionmaking (Harris-Kojetin et al., 2001).

While CMS launched the National Medicare Education Program (NMEP) in 1996, prior to the 1997 BBA mandate, the NMEP supports the educational objectives of the BBA. The Medicare & You handbook is an integral component of the NMEP. In preparation for implementing the NMEP, CMS pilot tested Medicare & You in five States and the Kansas City metropolitan statistical area (MSA) in fall 1998. CMS commissioned RTI to evaluate the pilot version of the handbook in the Kansas City MSA. As part of the study, RTI also evaluated the 1998 CAHPS^® Kansas City survey report. The Medicare CAHPS^® survey report provided Medicare beneficiaries in the Kansas City MSA with information comparing the quality of care provided by the five local Medicare health maintenance organizations (HMOs) (Carman et al., 1999). The Kansas City pilot study included both an outcomes survey and focus groups with Medicare beneficiaries who were largely drawn from among the outcome survey respondents. This article presents findings from the focus group portion of the study and limited survey results (McCormack et al., 2001c).

We examined what aged, disabled, and dually eligible Medicare beneficiaries thought of the handbook and CAHPS^® survey report, how they used these booklets, and how the booklets could be improved to increase their usability and utility among beneficiaries in plan choice decisionmaking. The main research questions addressed through the focus groups were:

• What are beneficiaries' overall impressions of the Medicare & You handbook and the Medicare CAHPS^® survey report?

• Do beneficiaries understand the purpose and intent of each booklet?

• How useful do beneficiaries find each booklet and how would they use each booklet?

• How much do beneficiaries trust the information in the booklets?

• Are there any aspects of the booklets that are problematic for beneficiaries or sub-groups of beneficiaries?

The tactical goal of the focus groups was to identify ways to improve the booklets to be more useful and likely to be used by beneficiaries when disseminated more broadly. The more strategic, policy goal was to determine whether Medicare beneficiaries could use these materials to help launch and support a long-term, informed choice initiative.

Methods

Booklets Examined

CMS developed the 52-page Medicare & You 1999 pilot version of the handbook. The handbook was significantly different in many ways in its content and appearance from its predecessor, known as the Medicare Handbook. In addition to describing the Medicare program, the 1999 handbook included information about Medicare costs and benefits, new managed care options, patient rights, and multiple information sources. It also provided comparative cost and benefit information for local Medicare HMOs and worksheets to facilitate plan comparisons. CMS used the handbook to convey key messages to Medicare beneficiaries at a time of significant change in the Medicare program (ushered in largely by the 1997 BBA) and public concerns about pressures to enroll more beneficiaries in managed care. Finally, CMS incorporated beneficiary perspectives in the development process,¹ not solely based on what researchers or other experts thought beneficiaries should have or want (Schriver, 1994).

RTI developed the 1998 Medicare CAHPS^® survey report, by adapting the report template created by RAND, Harvard Medical School, and RTI through the CAHPS^® study funded by the Agency for Healthcare Research and Quality. The CAHPS^® Medicare survey report is a 22-page booklet that provides comparative information on 15 quality of care measures for 5 Kansas City Medicare HMOs. These quality of care measures are based on beneficiaries' perceptions of their ability to get care and their experiences with the providers, the medical office staff, and the health plan. The data in the CAHPS^® survey report were collected via a survey from a representative sample of Medicare beneficiaries enrolled in Medicare HMOs in the Kansas City area in 1998. The CAHPS^® survey report includes an introductory section on the Medicare program and related plan options, the absolute health plan ratings displayed using bar graphs, the relative plan ratings using star icons, and a section suggesting topics that consumers should consider when choosing a plan. The survey report was designed to address some of the cognitive challenges that some older adults face, by using a large font size, sufficient amounts of white space, and a logical, organized and simple format, and ordering of contents.

Focus Group Methodology

Focus groups are structured as informal but guided discussions addressing a predetermined set of issues (Greenbaum, 1988). During the focus group, participants are encouraged to share openly their opinions, experiences, and ideas. Participants are guided and probed by a trained moderator who encourages participants to express their thoughts and experiences. The group moderator uses a topic guide that provides the requisite structure for the meeting. An observer is also present to take notes.

Strengths

Focus groups are an ideal methodology for gaining insight on perceptions, preferences, and general understanding of written materials (Morgan, 2000). Focus groups can provide a rich data source through which to obtain indepth findings using an interactive format and the capacity to pursue topics or questions that require further investigation.

We used the focus groups to supplement the survey component of the outcome evaluation of the Kansas City pilot study to enable us to examine not just what beneficiaries thought of the booklets, but why they thought the way they did. The focus groups allowed us to examine in more depth answers to some of the responses from the survey of new and experienced Medicare beneficiaries in the Kansas City MSA. Focus groups were also an appropriate methodology for our purposes because we wanted to examine the experiences and beliefs about the booklets that beneficiaries revealed as they interacted in the focus group discussions. In the focus groups, we learned about knowledge gaps or misperceptions (e.g., that Medicare HMOs cover all costs). Through the groups we also gleaned the extent of consensus and diversity among participants regarding the booklets. Compared with quantitative methods, focus groups allow more flexibility to followup on topics and provided a more natural environment than an experiment. The group format is one which research has shown Medicare beneficiaries tend to favor. With a small group, there is less pressure on any one participant to “be on” during the entire session. Thus, people may be more likely to relax in a group than in a one-on-one interview.

Weaknesses

Compared with other qualitative methods, focus groups are less flexible than one-on-one interviews because of group dynamics and the group findings are not generalizable with statistical precision to a larger population. Focus groups lack the representativeness of probability-based survey samples because they use small, purposeful samples. Focus groups also lack the conclusivenss of experimental designs, because they use open-ended questions and uncontrolled conditions (Morgan, 2000). In addition, dynamics within a group can produce a bias, since some people may speak less because of others or a group may be swayed by the opinions of one or two more vocal participants. Moderators in our groups tried wherever possible to encourage all participants to share in the discussion. For example, if one participant tended to talk a lot more than others, the moderator asked others what they thought of what that participant said. Moderators also tried at least once to ask silent participants to share their thoughts. Furthermore, we purposely planned to conduct more than one group for each of the subpopulations of interest (new aged, experienced aged, dually eligible, and disabled beneficiaries). Having more than one of each type of group safeguarded us against drawing conclusions about a subpopulation based on only one group whose experience may have been a fluke.

Individuals recruited for groups likely overrepresent the more physically- and cognitively-able Medicare beneficiaries. We tried to address the former obstacle by providing and paying for transportation in cases where the participant (usually a disabled beneficiary) would otherwise not have been able to attend the group. Participants in these focus groups were likely also more interested than the average Medicare beneficiary in discussing the booklets, since they had to take time out of their lives to participate. While we cannot generalize the focus group results, our confidence in group findings increases to the extent that patterns emerge regardless of group dynamics, location, and moderator. It is these patterns that we present in this article.

Study Population and Sites

RTI conducted 7 focus groups with a total of 56 participants. Two of the groups were conducted with primarily new beneficiaries (age 65) who had recently aged into Medicare and two groups were conducted with experienced beneficiaries (age 66 to 85) who had aged into Medicare more than 1 year prior to the focus group. Three groups were also conducted with individuals who were dually eligible for Medicaid and Medicare or who had become eligible for Medicare due to a disability. We conducted five of the groups in the Kansas City, Kansas area and two of the groups in the Kansas City, Missouri area. The focus groups varied in size from 7 to 10 participants.

We used multiple avenues for recruiting participants. We recruited experienced and new aged beneficiaries from a Kansas City MSA-area sample of individuals who participated in the outcome survey component of the project and had received the Medicare & You handbook (with or without the Medicare CAHPS^® survey report). To recruit dually eligible beneficiaries and beneficiaries with a disability, we identified potential participants using contacts at CMS' Kansas City Regional Office. Approximately 10 parti-cipants were recruited through an advertisement in a local newspaper. All participants were screened to ensure their eligibility for the study.

Table 1 provides the distribution of the Kansas City focus group Medicare beneficiaries on selected sociodemographic and health status characteristics. Focus group participants were relatively diverse in age, sex, education, race/ethnicity, health status, and income. Also shown is the distribution of all Medicare beneficiaries in the United States on these same sociodemographic and health status characteristics. This table allows us to compare the Kansas City Medicare beneficiaries who participated in the focus groups with the U.S. Medicare population (aged and non-aged disabled), to show the extent to which these focus group participants reflect the composition of the U.S. Medicare population. Our focus groups overrepresented both non-aged disabled Medicare beneficiaries and aged beneficiaries under age 74, relative to the national population. Just over one-half of the focus group participants were female, reflecting the national distribution. Group participants were more highly educated than Medicare beneficiaries at the national level. This is understandable since we overrepresented younger aged and underrepresented older aged beneficiaries in the groups. Older aged beneficiaries tend to be less educated than younger aged beneficiaries (U.S. Bureau of the Census, 1996).

Table 1. Selected Characteristics of Kansas City Medicare Focus Group Participants Compared with Medicare Beneficiaries in the United States: 1998.

Characteristic	Kansas City Medicare Focus Group Participants	Medicare Beneficiaries in the United States
	Percent
Age
Under 65 Years	29	13
65-74 Years	57	45
75 Years or Over	14	42
Sex
Male	45	44
Female	55	56
Education
Grades 0-8	2	20
Grades 9-11	8	16
Grade 12 (High School Graduate or General Equivalency Degree)	27	34
Some College or Technical School	37	16
College Graduate or Post-Graduate School	25	14
Not Ascertained	—	—
Race/Ethnicity
White Non-Hispanic	67	82
Black Non-Hispanic	30	9
Hispanic	0	7
Other	4	3
Not Ascertained	—	—
Health Status
Excellent	11	14
Very Good	32	25
Good	21	31
Fair	21	20
Poor	15	10
Not Ascertained	—	—
Annual Household Income
$10,000 or Less	40	28
$10,001 to $15,000	14	17
$15,001 to $20,000	14	12
$20,001 or More	32	43
Not Ascertained	—	—

NOTE: This table compares the Kansas City Medicare beneficiaries who participated in the focus groups to the general U.S. Medicare population (aged and non-aged disabled), to show the extent to which the focus group participants reflect the composition of the U.S. Medicare population on selected sociodemographic and health status characteristics.

SOURCES: Focus group post-questionnaire conducted by Research Triangle Institute in December 1998. The Characteristics and Perceptions of the Medicare Population: Data from the 1998 Medicare Current Beneficiary Survey, 1999.

We overrepresented black non-Hispanic beneficiaries relative to the national composition (30 percent to 9 percent) but underrepresented Hispanic beneficiaries (0 percent to 7 percent). Focus group participants rated their health slightly less well than Medicare beneficiaries at the national level (64 percent to 70 percent, respectively, rated their health as excellent, very good, or good). Group participants also tended to have somewhat lower incomes than Medicare beneficiaries at the national level (40 percent versus 28 percent had an annual income of $10,000 or less). Though not shown in a table, there were some differences between the aged, the disabled, and dually eligible beneficiaries. The aged beneficiaries tended to have higher incomes, better self-assessed health status, and were more likely to be white than the disabled and dually eligible beneficiaries.

Data Collection

We conducted the groups in conveniently located public libraries and municipal centers. Those participants who were determined during the recruitment screening not to have a copy of the Medicare & You 1999 handbook and the Medicare CAHPS^® survey report were sent a copy of each before the session. All participants were asked to review and be familiar with both materials prior to the groups, and to bring the materials with them to the groups.

A team of four researchers (two moderators and two note takers) working in teams of two conducted the seven focus groups. Each two-person team consisted of a senior researcher who moderated the group accompanied by a junior researcher to observe and record notes on verbal and non-verbal exchanges in each group. We purposely used more than one moderator and note taker to avoid potential bias from having the same person do all groups. Having these teams allowed us to compare findings and in an informal way conduct a reliability check on the findings. The moderators were trained in focus group moderation either through formal instruction and/or on-the-job through observing and conducting groups. Each moderator also had 5 years of focus group moderating experience. Each note taker had previous experience observing and taking notes on focus groups.

Each moderator used a semi-structured topic guide (or protocol). RTI's Institutional Review Board approved the project's focus group protocol. The topic guide addressed participants' impressions of and trust in the Medicare & You handbook and the Medicare CAHPS^® survey report, respectively. The protocol also tapped participants' comprehension of the messages in each booklet, perceptions of the usefulness and utility of each booklet, and how participants used each booklet. Also contained in the protocol is introductory language describing the study, information about confidentiality, a warm up exercise, a wrap-up where additional participant comments were requested, and administration of the post group questionnaire.² Moderators were instructed to read the introduction and the main probes in the protocol as written, to avoid potential bias from asking different questions of different groups.

Before the groups were conducted, the moderators and note takers reviewed and familiarized themselves with the topic guide. The full focus group team then met to discuss the main research questions to be addressed during the groups and to prioritize discussion topics in the event that a group risked running out of time. To decrease potential ordering bias, one-half of the groups were asked first about their impression of the handbook while the others were asked first about the Medicare CAHPS^® survey report.

Groups lasted approximately 2 hours each, refreshments were served, and every participant received $40 for their participation. Transportation was arranged for individuals with special needs. Groups were tape recorded with participant permission. At the end of the group, participants were asked to complete a short questionnaire to obtain basic demographic information and to provide all participants a last opportunity to share opinions about the booklets that they may not have expressed in the group.

Analysis

Immediately following each focus group, the moderator and note taker conducted a detailed debriefing on the group and completed a standardized debriefing form to convey main themes and illustrative participant quotes. These debriefing forms were analyzed for key common themes and differences among the groups by the lead analyst, who had also developed the topic guide and conducted two of the seven groups.

In addition, each group discussion was transcribed and analyzed using qualitative software. The transcriptions were coded according to a combination of predetermined and newly-appearing themes using a qualitative data analysis software program called Non-Numerical Unstructured Data Indexing Searching & Theorizing, or Nud*ist4. Nud*ist4 was developed by Qualitative Solutions and Research, an Australian company that develops qualitative research products. Nud*ist4 is one of the most widely used qualitative data analysis software products in the world (Quality Solutions and Research, 2001).

We used Nud*ist4 software to code and manage the transcribed group discussions in preparation for analysis. One team member who participated in the focus groups developed a set of theme and subtheme codes (also called nodes) using the focus group protocol discussion areas as a guide. The team member also reviewed a subset of the transcripts and derived additional nodes based on that review. There were 10 main nodes, corresponding to the 8 main sections of the focus group topic guide plus 2 additional nodes for areas not specifically probed in the groups but which were discussed extemporaneously by at least some of the groups. Within most of these 10 nodes, there were several sub-nodes, most often corresponding to the subprobes in the protocol.

The transcripts were divided among a team of four analysts (all of whom participated in conducting the groups), with each analyst responsible for reviewing each of the seven transcripts and coding these transcripts for 2 to 3 assigned nodes and their corresponding subnodes. The coded transcript data were then grouped by node and subnode, resulting in text files organized by theme. These coded raw data were then analyzed for content and group variation. Using a qualitative software program allowed the team of four analysts to verify the findings based on the debriefing forms as well as to glean findings from the focus groups that were “off the beaten path,” in that they were not directly derived from the focus group topic guide probes.

Findings

This section is based on three main data sources collected from the focus groups: (1) notes and quotations recorded by the note taker and moderator from each respective group debriefing session; (2) responses to the post-focus group questionnaire that RTI administered at the end of each group³; and (3) results from coding the transcribed tapes from the groups using the Nud*ist4 qualitative software program. Since the results from the different data sources corroborate each other, there are similar themes found for each booklet, and results across the seven groups are often similar, we present results thematically rather than by data source or by booklet. Differences among groups and by booklet are noted where they occur. To triangulate the group findings, we include limited results from the survey of new and experienced Medicare beneficiaries in the Kansas City MSA.

Perceived Utility

Beneficiaries generally had a positive response to both booklets, seeing the handbook as a reference tool and the Medicare CAHPS^® survey report as a short, easy-to-read booklet. Beneficiaries felt that neither booklet could stand alone, however. Beneficiaries would want to have additional information before making a plan choice, though these booklets were seen as a useful starting point.

People viewed Medicare & You as comprehensive, understandable, and a good reference to save and consult over time as the need arose. The majority of beneficiaries read at least some of the handbook. About one-half the beneficiaries said they read the entire handbook while others skimmed or read specific sections.

This finding is in line with research on adult learning. For adults, learning is a means to an end, not an end in itself. Adults come to informational materials with expectations that the information will help them answer their particular questions or complete a task they want to accomplish. Conversely, if adults do not see themselves as having a particular, immediate use for informational material, they will be less likely to read it (Keenan, 2001). In this case, many participants saw the handbook as a resource to be called upon when they had a specific question.

Participants generally felt that both booklets would be most useful to beneficiaries with particular information needs. Specifically, some participants saw the Medicare CAHPS^® survey report as primarily useful for people considering or choosing an HMO, and Medicare & You as most useful for people changing coverage or making an initial health plan decision. This was particularly true in the groups composed predominantly of newly enrolled beneficiaries. A few new beneficiaries went so far as to say that they might have selected a different plan had they received the comparative plan information in the booklets before having to make their current Medicare plan choice.

Table 2 shows selected results from the focus group post-questionnaire about participants' perceptions of the handbook. Most group participants thought the handbook was (somewhat or very) easy to understand (not shown in table). However, the higher educated were significantly more likely than the lower educated to think that the handbook was easy to understand (97 percent compared with 67 percent). Participants found particularly useful the comparative information about the different Medicare health plans available locally and an explanation of what Original Medicare entails. Higher educated focus group participants were significantly more likely than lower educated participants to find the information about different Medicare health plans useful. This may be because the higher educated were better able to manage and wade through the information about the various new plan types. When asked in the groups what other sections of the handbook they found most useful, the most frequent responses were the telephone numbers and the information about patients' rights. Telephone numbers were seen by group participants as a way to enable people to get additional information, or to get information for their specific needs that could not be included in the handbook (because of length limitations).

Table 2. Selected Responses to Kansas City Medicare Focus Group Post-Questionnaire About the Medicare & You Handbook, by Education: 1998.

Question and Response Option	High School Graduate or Less		At Least Some College
	Number	Percent	Number	Percent
In general how hard or easy is the Medicare & You handbook to understand?1
Very Hard	1	6	—	—
Somewhat Hard	5	28	1	3
Somewhat Easy	7	39	17	59
Very Easy	5	28	11	38
Which sections of the Medicare & You handbook, if any, did you find most useful? (all that apply)2,3
Learning About Medicare Health Plans1	6	35	19	61
What is the Original Medicare Plan?	7	41	16	52
Telephone Numbers for Assistance	9	53	14	45
Medicare Patients' Rights	9	53	12	39

¹Differences between educational groups are significant at the 0.05 level.

²Percentages total to more than 100 because participants could choose more than one response to the question.

³The four topics shown are those chosen by almost one-half of the focus group participants as being the most useful sections of the handbook.

SOURCE: Focus group post-questionnaire conducted by Research Triangle Institute in December 1998.

In the Medicare CAHPS^® survey report, there is a two-page section on “Things to Think About” that guides the reader through the process of comparing plans using the CAHPS^® data, obtaining other comparative information, and asking questions to consider when making a plan choice. The handbook includes a four-page work sheet that provides a more detailed comparison process than in the Medicare CAHPS^® survey report. Focus group participants found both of these tools particularly useful, with some participants noting that some of the questions that are in the booklets they might not have thought of themselves.

Some beneficiaries also mentioned using Medicare & You and the report to verify information from other sources to increase their confidence in their current health plan choices. Participants also found the Medicare CAHPS^® survey report useful because it helps people compare their Medicare HMO choices and it shows differences in quality among the plans. Some participants also thought it valuable to be able to see the opinions that other beneficiaries have of the Medicare HMOs.

As with the handbook, most group participants (85 percent, not shown in table) thought the survey report was (somewhat or very) easy to understand. Table 3 shows selected results from the focus group post-questionnaire about participants' perceptions of the Medicare CAHPS^® survey report broken out by education. Higher educated group participants were significantly more likely than lower educated persons to find the report easy to understand (96 percent to 64 percent), as shown in Table 3.

Table 3. Selected Responses to the Kansas City Medicare Focus Group Post-Questionnaire About the Medicare CAHPS^® Survey Report, by Education: 1998.

Questions and Response Options	High School Graduate or Less		At Least Some College
	Number	Percent1	Number	Percent2
In general how hard or easy is the Medicare CAHPS survey report to understand?2
Very Hard	1	6	—	—
Somewhat Hard	5	29	1	3
Somewhat Easy	6	35	16	53
Very Easy	5	29	13	43

¹Percentages may not total to exactly 100 due to rounding.

²Differences between educational groups are significant at the 0.05 level.

NOTE: CAHPS^® is Consumer Assessment of Health Plans Study.

SOURCE: Focus group post-questionnaire conducted by Research Triangle Institute in December 1998.

Some participants acknowledged that since Medicare is a complex system, they appreciated that Medicare & You and the Medicare CAHPS^® survey report were an attempt to help beneficiaries understand it. In general, participants wanted any helpful information they could get about this often challenging topic. One participant went so far as to say that, while she appreciates the handbook information, the Medicare program is so complex that no written materials would ever successfully explain it in a way that beneficiaries could understand it.

“Medicare is getting more and more complicated and us senior citizens are getting less and less capable of making good decisions. It's almost as if they are muddying the waters so we can't pick out the best thing for us.”

Ways to Increase Utility of Booklets

Some participants noted that because the Medicare CAHPS^® survey report was short and easy to read, they were more likely to read it fully than they were to read Medicare & You in its entirety. This suggests that more information is not necessarily better. Beyond some threshold level of information beneficiaries may be less likely to read a document or some of the information in it. Some beneficiaries noted multiple times the absence from the survey report of information about beneficiary costs and would like to have seen it included.

Trust in Booklets

While more group participants trusted the handbook and the Medicare CAHPS^® survey report than did not, some tempered their trust with skepticism. The majority of participants felt that knowing that Medicare & You came from the Federal Government made them more likely to trust the handbook, seeing it as less biased and more impartial compared with other sources such as health plans. However, several participants noted that the government is not as trustworthy now as it was when they were young.

The group participants expressed somewhat less trust in the Medicare CAHPS^® survey report than in the handbook. Just over one-half of the focus group participants found the report trustworthy, either because they mistakenly thought it came from the government or because they attributed some validity to the booklet being copyrighted. This is in line with the results of the survey of new and experienced Medicare beneficiaries in the Kansas City MSA, in which almost 10 percent of respondents did not trust the survey report at all (not shown in table). In addition, according to the Kansas City MSA survey results, trust levels varied significantly with beneficiary education, with lower educated beneficiaries being more skeptical about the survey report than higher educated beneficiaries. Specifically, 50 percent of college educated beneficiaries trusted the survey report “a lot” and none trusted it “not at all,” compared with 28 percent and 18 percent, respectively, of those with less than a high school education (not shown in table).

One reason some group participants tended to trust Medicare & You more than they trusted the Medicare CAHPS^® survey report was because they thought that the report was “pushing HMOs.” Some Medicare beneficiaries, like some of the larger population, have preconceived negative ideas about HMOs that are informed by what they hear in the media and learn about from family and friends. Even if informational material about HMOs is unbiased and balanced, readers may see information as partial to HMOs because of these preconceived notions or because, as in the Medicare CAHPS^® survey report, only Medicare HMOs are shown. Some participants wondered why the survey report only contained HMOs unless it was encouraging people to enroll in HMOs. Some thought that the survey report should state more clearly and strongly that beneficiaries can keep the health insurance option they have and do not need to enroll in an HMO. Other participants also showed distrust in the Medicare CAHPS^® survey report because of a general skepticism of, unfamiliarity with, or discomfort with surveys and related statistical issues (e.g., sampling, statistical adjustments) presented in the report. This unfamiliarity could have played a role in the degree to which educational level was associated with trust (as previously discussed). Nevertheless, most participants (regardless of education level) said they trusted the Medicare CAHPS^® survey report more than they trusted information from individual health plans.

Special Subgroup Needs

Some disabled and dually eligible focus group participants were confused about their eligibility for the Medicare plans shown in the booklets. These participants were frustrated both over the lack of clarity about this in the booklets as well the possibility that they may, in fact, not be eligible for some plans due to being under age 65. Among the dually eligible participants, there was also a sentiment that the cost information in Medicare & You was less relevant to them because Medicaid paid for gaps in Medicare. The dually eligible beneficiaries would like to have seen more information about how the Medicare and Medicaid programs coordinate services and coverage. However, dually eligible beneficiaries also acknowledged that it would not be feasible to incorporate all of the information they needed for their particular circumstances into a booklet intended for all Medicare beneficiaries.

If they wrote this [handbook] for specific problems it would be four feet thick and no one would never read it [sic]. This is more of an introductory-type thing.

Summary

Medicare beneficiaries were generally positive about both Medicare & You and the Medicare CAHPS^® survey report, seeing them as good starting points for comparing plans. Beneficiaries found the inclusion of comparative cost information (in Medicare & You only) and telephone numbers for further information to be a particularly useful part of the materials. The majority of focus group participants trusted the materials, with their attitude chiefly influenced by their trust in the Medicare program. The booklets are not entirely responsive to the needs of beneficiaries with special situations, such as disability or dual eligibility. Most beneficiaries were interested in learning about others' experiences in Medicare plans (presented in the CAHPS^® survey results). These focus group results may overestimate the level of beneficiary interest in similar materials. These participants may have a higher than average interest, they tended to be more highly educated than Medicare beneficiaries as a whole, and they were asked to review the materials before attending the focus group discussions.

Discussion

We draw the following policy implications based on the focus group findings. Consistent with previous research (Goldstein, 1999), it appears that most beneficiaries who have been in Medicare for awhile will use comparative information only when they need to make a plan change. For this reason, a broad campaign that stresses the availability of comparative cost, benefit, and quality information (“for when you may need it down the road”) may help raise awareness of this information among beneficiaries (and those soon to age into Medicare). Combining this broad-based awareness campaign with a targeted set of information tools for only those beneficiaries who are looking to change plans or choose a new plan may be more effective and less costly than mailing comparative plan choice materials to all beneficiaries every year.⁴

Participants in the groups liked the worksheets in both booklets intended to help them compare plans and make a Medicare plan choice. They also thought that the sections of the materials containing questions to consider when comparing plans were quite valuable. However, few focus group participants actually filled out the worksheets, either because they were happy with their choice or had already made a choice before receiving the handbook.⁵ Sending out informational materials to new beneficiaries up to one year before they become eligible for Medicare could increase the likelihood that new beneficiaries actually use the work sheets to help them make a Medicare plan choice.

Some of the participants also expressed concern that the government is trying to push beneficiaries into HMOs to save money. More attention and additional effort should be paid to persuading beneficiaries that the CAHPS^® survey information can be trusted. This is particularly important because in fall 2001 (for 2002 plan choices) plan-level quality of care data will be available only on the Medicare Web site (not in the handbook). The relationship observed between trust and education in the Kansas City MSA survey results may have the potential to turn into a story of the “haves” and the “have nots” regarding quality data, as higher educated beneficiaries are more likely than lower educated beneficiaries to use the Medicare Web site. Plan choice materials should simply and directly address beneficiary concerns about feeling pressured to enroll in HMOs. Including comparative CAHPS^® survey data for all of the local Medicare health coverage options, including fee-for-service (as CMS plans do), should also help to alleviate some of the concern that HMOs are being promoted.

Those developing plan information materials need to recognize more fully that many beneficiaries may be experiencing information saturation. Some key messages may need to be significantly highlighted, stated more simply or directly, and repeated across multiple venues (as is the goal under the NMEP). Although many beneficiaries in the groups understood from Medicare & You that they did not need to change their Medicare plan if they were satisfied with what they had, some beneficiaries did not get this message. This might be accomplished on the inside cover letter of the handbook by simply rewording “…you don't have to do anything” to something more like “you don't have to change the type of Medicare coverage you now have” or “you can stay with the Medicare coverage you now have.”

Special groups of beneficiaries (such as Medicaid recipients and military retirees) have information needs that may not be fully met by Medicare documents intended for the majority of Medicare beneficiaries. Supplemental materials may be necessary to target the specific situations of these beneficiaries. Similarly, some topics that are currently in Medicare & You (e.g., specified low-income Medicare beneficiary, qualified Medicare beneficiary, Medicaid) are relevant only to a subset of beneficiaries. Beneficiaries who may benefit from these specific programs may not even be aware of them. These specialized sections could be made more easily accessible to such readers, for example by including them in the table of contents and possibly highlighting them in a specific section of the handbook with references to additional information that can be obtained outside of Medicare & You. Additional research with targeted beneficiary sub-populations is also needed, to examine whether alternative ways to present this targeted information will help them more easily access the information.

No matter how simple the materials created, many beneficiaries will need assistance from formal (e.g., employee benefits staff, State Health Insurance Program counselors, Medicare helpline)⁶ or informal (e.g., family, friends) intermediaries (Kiefer, 2001; McCormack et al., 2001a) to use them. For this reason, understanding the role of these different intermediaries and what information and assistance they need is becoming increasingly important in the field of Medicare education.

With Medicare beneficiaries having greater responsibility for choosing their health care coverage, the challenge to develop and disseminate useful written information is a critical part of the effort to help beneficiaries make more informed choices. The focus group results suggest that the Medicare information materials tested have much to recommend them as well as areas for improvement as the long-term, informed choice initiative evolves.