Marquette Elder's AdvisorVolume 3Issue 1 Summer Article 3

Consumer Preferences for a Consumer DirectedCash Option Versus Traditional ServicesSharon M. DesmondUniversity of Maryland

Kevin J. MahoneyBoston College

Lori Simon-RusinowitzUniversity of Maryland Center on Aging and Department of Public and Community Health

Dawn M. ShoopUniversity of Maryland Center on Aging

Follow this and additional works at: http://scholarship.law.marquette.edu/eldersPart of the Elder Law Commons

This Featured Article is brought to you for free and open access by the Journals at Marquette Law Scholarly Commons. It has been accepted forinclusion in Marquette Elder's Advisor by an authorized administrator of Marquette Law Scholarly Commons. For more information, please contactmegan.obrien@marquette.edu.

Repository CitationDesmond, Sharon M.; Mahoney, Kevin J.; Simon-Rusinowitz, Lori; and Shoop, Dawn M. (2001) "Consumer Preferences for aConsumer Directed Cash Option Versus Traditional Services," Marquette Elder's Advisor: Vol. 3: Iss. 1, Article 3.Available at: http://scholarship.law.marquette.edu/elders/vol3/iss1/3

Consumer Preferencesfor a Consumer-DirectedCash Option VersusTraditional ServicesTelephone Survey Findings of Florida Eldersand Adults with Physical Disabilities

This report evaluates the findings of a

telephone survey inquiring into the

preferences of Florida elders and adults

with physical disabilities with regard to

various options for receiving assistance

and services.

By Sharon M. Desmond,Kevin J. Mahoney, Lori Simon-Rusinowitz, Dawn M. Shoop,Marie R. Squillace, Rob A. Fay

oday, in most states, whether you arean elderly individual or a younger per-son with disabilities, if you need Med-icaid assistance to perform major ac-tivities of daily living (ADLs) like bath-

ing, dressing, toileting, transferring, or eating youwill not have much say over who helps you, whenthey come, or what they actually do. However, formany years, people from the disability communityhave been saying, "If I had more control over myservices, my quality of life would improve and I couldmeet my needs for the same amount of money orless." The Cash and Counseling Demonstration andEvaluation (CCDE)1 is, at its heart, a policy-drivenevaluation of this basic belief. The CCDE is a test ofone of the most unfettered forms of consumer-di-rected services--offering consumers a cash allowancein lieu of agency-delivered services.

Sharon M. Desmond, Ph.D., is an associate professor in the Department of Public and Community Health at theUniversity of Maryland.KevinJ. Mahoney, Ph.D., is the National Project Director of the Cash and Counseling Demonstration and Evalua-tion at Boston College.Lori Simon-Rusinowitz, Ph.D., is the Deputy Project Director of the Cash and Counseling Demonstration andEvaluation at the University of Maryland Center on Aging and Department of Public and Community Health.Dawn M. Shoop, Ph.D., is a faculty research associate at the University of Maryland Center on Aging.Marie R. Squillace, M.A., was a faculty research associate at the University of Maryland Center on Aging.Rob A. Fay, B.S., was the telephone survey lab coordinator at the University of Maryland Center on Aging.

2 1 Elder's Advisor

This article reports on background research con-ducted to inform the three-state (Arkansas, Florida,and New Jersey) CCDE design and further our un-derstanding about implementing consumer-directedservices. A consumer-directed approach to servicesencourages maximum consumer choice and controlto people needing assistance with ADLs and otherpersonal care tasks. Efforts to better understand theintricacies of implementing consumer-directed ser-vices, especially in the aging community, have onlyrecently begun.2 Early in the CCDE development,program planners realized that key information es-sential to program implementation was lacking-data indicating consumers' preferences for a con-sumer-directed cash option versus traditionalagency-delivered services. For that reason, the Uni-versity of Maryland Center on Aging conductedtelephone surveys in each demonstration state toassess consumers' preferences. This article reportsfindings from the survey conducted in Florida.

As this preference survey provides background in-formation for the CCDE, we begin with a briefdescription of the CCDE and existing personal assis-tance services to provide a context for the Floridafindings. Although the survey results guided severalaspects of program development, we focus on one ap-plication--guidance in developing communications andsocial marketing materials to assist Florida in its ef-forts to inform consumers about the project. We alsohighlight key policy issues addressed by the survey.

BackgroundThe idea of consumer-directed services originated

more than two decades ago among younger peoplewith disabilities in the disability rights and indepen-dent living movements.' The aging community beganto adopt consumer-direction principles more recentlywhen a coalition between the aging and younger dis-ability communities emerged in the mid-1980s.4Interest in consumer choice expanded among someaging leaders in the early 1990s, in part due to abelief that consumer-directed care might lead to muchneeded cost savings.5 The emphasis on consumerchoice and control in the language of the 1994 HealthSecurity Act 6 exemplifies this increased interest.

Typically personal assistance services are financedby public or private third-party payers in one of threeways:

1. Cash benefits (payments to qualified clientsor their representative payees);

2. Vendor payments (a case manager determinesthe types/amounts of covered services, andarranges for and pays authorized providersto deliver the services); and

3. Vouchers (clients use funds for authorizedpurchases).

In the United States, most existing public programsthat finance personal care services follow the ven-dor payment model where the program purchasesservices for consumers from authorized vendors (i.e.,service providers or equipment suppliers).7

Cash allowance programs are currently verysmall because they involve "state-only" funds. Statescannot use Medicaid to fund cash allowances thatpermit clients to purchase their own services becauseof federal restrictions on direct payments to clients.Until recently, the prohibition on cash payments toMedicaid clients has rarely been questioned. How-ever, many state program officials have come to sharethe concerns of disability rights advocates who wantprograms that promote consumer choice (such as aconsumer-directed cash benefit program) rather thanthose that may foster dependency in the name ofconsumer protection and/or public accountability.8

Additionally, state officials have a strong interest inachieving program economies. Most Medicaid per-sonal care programs mandate that case managers(registered nurses and/or social workers) assess cli-ents, develop and monitor care plans, and authorizeprovider payments. Case management can be ex-pensive, and researchers and administrators questionwhether it should be uniformly required.9

The cash and counseling model offers a cash al-lowance and information to clients so they canpurchase the services, assistive devices, or homemodifications that best meet their individual needs.In theory, consumers who shop for the most cost-effective providers would then (through such savings)have funds to purchase additional services. 10 To de-termine the effectiveness of this consumer-directedoption, the CCDE was designed. The evaluationcompares cost, quality, and satisfaction of consum-ers receiving traditional personal care services withthose receiving the cash option.11

Determining Consumers' Preferences for aCash OptionLittle research exists to indicate clients' preferencesfor consumer directed programs, specifically the cashoption. This information was essential to help the

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services 3

demonstration states design various cash option com-ponents (including counseling services) and socialmarketing approaches that would enable consumersand surrogates to make an informed choice betweenthe cash option and their current program.

Policy makers, program planners, and othershave speculated that age would be a strong indica-tor of interest in a consumer-directed option (i.e.,younger consumers would be more likely to be in-terested). While research on consumers' preferencesfor consumer-directed services is limited, there isevidence that consumers of all ages, including elders,would like to be more involved in directing theircare.12 However, there is much more to be learnedabout age-specific preferences for consumer direc-tion in general, and particularly for a cash option.Additionally, there is scant information regardingother demographic and background characteristicsthat may influence interest in consumer direction.For example, studies have found greater informalcare and less nursing home use among some minor-ity groups compared to their white counterparts,providing speculation that preferences for consumerdirection may differ among racial and ethnicgroups.13 More recently, Sciegaj and Kyriacou 14

found that consumers' preferences for types of per-sonal assistance services (consumer-directed,negotiated care managed, and traditional case-man-aged services) varied among racial/ethnic groups.

In the present study, Florida elders and adultswith physical disabilities who are receiving Medic-aid Home and Community-Based Waiver serviceswere randomly selected to participate in a telephonesurvey assessing their perceptions regarding the cashoption. Specifically, we wanted to:

1. Determine preferences for consumer-directedservices and the cash option;

2. Determine the percentage of consumers andsurrogates choosing the cash option versustraditional services and identify reasons fortheir choices;

3. Identify demographic and background char-acteristics of consumers and surrogates withspecific preferences;

4. Identify cash option features that are at-tractive or unattractive to consumers and sur-rogates;

5. Identify what information consumers andsurrogates need to decide whether to choosethe cash option;

6. Identify consumers' and surrogates' needs forcounseling and support services; and

7. Develop strategies to market the cash option.

Methods

SampleThere were 950 adults with physical disabilities 18to 64 years of age and 11,499 adults 65 and olderwho received services in 1997 from Florida's Agingand Disabled Waiver Program. Based on this popu-lation, and using the relevant standard error formulato insure adequate power, a sample size of 273 adults18 to 64 years of age and 330 adults 65 and olderwas needed. It was assumed phone numbers wouldbe at least 80 percent accurate and that the responserate would be 20 percent. Florida forwarded thenames and phone numbers of 368 randomly selectedadults 18 to 64 years of age with physical disabili-ties and 2,344 randomly selected older adults (65and older) to the University of Maryland's Interdis-ciplinary Health Research Lab (IHRL).

InstrumentationA survey measuring client perceptions of the cash

option was developed. It consisted of four sections:

1. Type of services received and satisfaction withthose services (34 items);

2. Perceptions regarding the cash option (32items);

3. Consumer demographic and backgroundvariables (35 items); and

4. Perceptions and demographics of surrogates(33 items).

To explain the cash option, interviewers read ascenario about a woman, Mrs. Green, who neededpersonal care services. The scenario described sev-eral different ways she could use her cash benefit.Then, subsequent survey items informed respondentsabout various cash option features and support ser-vices. The questionnaire contained two attitudinalsubscales-satisfaction with services (7 items) andwillingness to assume responsibility for personal as-sistance services (6 items). Background variablesincluded a measure of functional status based onADLs, an assessment of overall physical health, theindividual's living arrangement, number of informalcaregivers, and experience interviewing, training,hiring, or supervising workers, among others.

4 I Elder's Advisor

Content validity of the survey was establishedvia an expert panel (n = 7) knowledgeable in aging,disabilities, and survey design and evaluation. Inaddition, the survey was pilot tested with three dis-abled and elderly individuals to assess administrationtime as well as acceptability and understandabilityof the items. A pilot test with 120 clients was alsoconducted to determine internal consistency andreliability of the subscales (both were reliable, .73and .79). Finally, the instrument was translated intoSpanish, and then back-translated, to insure accu-racy and enable Spanish-speaking consumers'participation. We expected to have a large numberof Spanish-speaking respondents in Florida, given thelarge Hispanic population in the state. However,only 63 surveys were conducted in Spanish."5

ProceduresThe randomly selected potential participant phonefiles from Florida were entered into the MacIntoshComputer-Assisted Telephone Interview (MaCATI)system. Telephone interviews took place in 1997from September through November. On average,interviews lasted 40 minutes. Individuals were calledup to six times before they were considerednonrespondents. If respondents were unable to an-swer survey items themselves (for a variety of reasons,including language barriers, communication difficul-ties, or other health problems), they provided theinterviewer with the name and phone number of asurrogate responder (a friend or relative). A seriesof questions were then used to identify surrogatesand the type of decisions they helped the consumermake (i.e., financial, medical, living arrangements,or all). Surrogates were instructed to represent theconsumer and respond to survey items with the con-sumer present, so the consumer could clarifyresponses if necessary.

Of the 2,712 randomly selected names and phonenumbers sent to the IHRL, 899 were unusable num-bers (i.e., not in service, wrong locale, business phone,person no longer receiving services, etc.), 813 refusedto participate, 257 were not used because the appro-priate sample size had been reached, and 743 completedthe survey (48 percent response rate). One hundredand sixty-five of the 743 individuals who completedthe survey were "partial completes," meaning they hadanswered up to and including the item that asked abouttheir interest in the cash option. A total of 578 re-spondents actually completed the entire survey, and43 of those responded in Spanish.

Nonrespondents were asked why they did notwant to participate in the survey prior to ending theinterview, and the two most frequent reasons givenwere that the individual was too sick, disabled, orold (n = 235, 34 percent), or that she or he was notinterested in participating in any survey (n = 179, 26percent). Other reasons were that the consumer wasnot interested in changing services (n = 60, 9 per-cent) or not interested in discussing her or hispersonal care services (n = 60, 9 percent). We wereable to determine if nonrespondents differed fromparticipants based on age and average amount ofmoney the state spent on services over a nine-monthperiod, from July 1, 1996 through March 31, 1997.Age and cost data were provided to the researchersvia the phone files sent by Florida. There was a sig-nificant difference (p < .01) between participants(n = 743) and nonrespondents (n = 697) on bothvariables. The average amount of money spent onservices over the nine months for participants was$2,581 (SD = $2,312), compared to $2,217 (SD =$1,980) for nonrespondents. Participants were alsoyounger than nonrespondents; mean ages were68.7 years (SD = 16.2) and 77.4 years (SD = 12.0)respectively.

Results

Description of the SampleBetween 579 and 591 respondents provided con-sumer demographic and background information.The sample size varied slightly due to missing dataon a particular item. The majority of consumerswere female (80 percent), had a high school educa-tion or less (73 percent), and were 65 years of age orolder (66 percent) (Exhibit 1). Fifty-eight percentwere Caucasian, 24 percent were African American,and 12 percent identified themselves as Hispanic.Forty-five percent were widowed, while 20 percentwere divorced. Ninety-one percent reported havingbeen employed, although the majority of these indi-viduals were currently retired (71 percent). Amajority had an informal caregiver (61 percent), and33 percent had an informal live-in caregiver. Thirty-four percent had experience hiring, firing, orinterviewing workers, while 44 percent had experi-ence supervising or training workers (Exhibit 1).

Consumers were asked to rate their overall healthstatus, and 74 percent rated it "fair" or "poor," com-pared to 24 percent who rated it "good," "verygood," or "excellent." A measure of functional

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services 5

Exhibit 1. Demographic and Background Characteristics of Florida Elderly and Adult Consumers

Demographic Variable N (%)

Adult Consumers (18 to 64)Elderly Consumers (65 and older)

GenderMaleFemale

Race/EthnicityAfrican American/BlackNative American/Alaskan NativeAsianHispanicCaucasian/WhiteBiracialDon't know/Refused question

Education levelLess than high schoolHigh school graduateTrade or vocational schoolSome collegeBaccalaureate degreeSome graduate schoolGraduate degreeDon't know/Refused question

Marital statusMarriedWidowedDivorcedSeparatedSingleLive with partner

Living arrangementAloneWith spouse and/or childrenWith friend, partner, or relativeOther

Do you own your own home?YesNoDon't know/Refused question

Demographic Variable

Do you have an informal caregiver?YesNoDon't know/Refused question

199 (34.0)387 (66.0)

120 (20.3)470 (79.7)

14352

69343

821

267160

88132

51024

84262118

3480

1

292179113

1

(24.2)(00.8)(00.3)(11.7)(58.0)(01.4)(03.5)

(45.5)(27.3)(01.4)(13.8)(05.5)(00.9)(01.7)(04.1)

(14.4)(44.8)(20.2)(05.8)(13.7)(00.2)

(49.9)(30.6)(19.3)(00.2)

208 (35.6)371 (63.4)

6 (01.1)

N (%)

358223

4

(61.2)(38.1)(00.6) (continued)

I Elder's Advisor

Exhibit 1. Demographic and Background Characteristics of Florida Elderly and Adult Consumers (cont'd)

Does informal caregiver live with you?

No informal caregiverDon't know

Have you ever been employed?YesNoDon't know/Refused question

Consumer employment status (if ever employed)Employed part timeUnemployedRetiredHomemakerVolunteerDon't know/Refused question

193 (33.0)168 (28.7)222 (37.9)

2 (00.3)

529 (90.7)49 (08.4)

5 (00.8)

4 (00.7)131 (24.5)380 (71.0)

6 (01.1)2 (00.4)

12 (02.2)

Any experience hiring, firing, or interviewing any type of worker?Yes 196 (33.5)No 375 (64.1)Don't know/Refused question 14 (02.5)

Any experience supervising or training any type of worker?YesNoDon't know/Refused question

Current overall physical healthExcellentVery goodGoodFairPoorDon't know/Refused question

Disability levelMildModerateSevere

257 (44.1)310 (53.2)

16 (02.8)

1819

103192238

13

(03.1)(03.3)(17.7)(32.9)(40.8)(02.2)

288 (49.7)153 (26.4)138 (23.8)

Length in program0-6 months6-11 months1-2 years3 years or moreDon't know

3249

209424

28

(04.3)(06.6)(28.2)(57.1)(03.8)

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services 7

status was also obtained, using the five ADLs (bath-ing, dressing, transferring, toileting, and eating).Consumers were asked if they needed help with eachactivity and could respond "yes," "no," or "some-times." A "yes" response received a score of 1, a"no" a 0, and a "sometimes" a 0.5. The functional-status scale could thus range from 0 to 5. Individualswho scored between 0 and 1.5 were consideredmildly disabled (50 percent), those who scored be-tween 2 and 3.5 were labeled moderately disabled(26 percent), and those with a 4 or 5 were deter-mined to be severely disabled (24 percent). Finally,742 respondents answered the item regarding howlong they had been in the Florida Home Care pro-gram. The majority (57 percent, n = 424) had beenin the program more than three years, 28 percent(n = 209) had participated one to two years, and theremainder were enrolled for an even shorter periodof time (Exhibit 1).

Services Received and Satisfaction LevelsEither 742 or 743 respondents answered the itemsconcerning services received, as one person discon-tinued the survey during this sequence of questions.Respondents were asked if they currently receivedeach of 14 different services provided by Florida'sHome Care program. They were also asked howmuch of the service they received and if they weresatisfied with it. A majority of consumers received 3services-case manager (73 percent, n = 543), home-maker (72 percent, n = 531), and personal care (61percent, n = 454) (Exhibit 2). The percentage ofconsumers who obtained other services ranged from4 percent (n = 26), who received caregiver trainingand support, to 40 percent (n = 294), who receivedhome-delivered meals. The vast majority of consum-ers who received each service were satisfied with theservice. Rates of satisfaction ranged from a high of94 percent (n = 257), who were satisfied with con-sumable medical supplies, to a low of 75 percent(n = 21), who were satisfied with caregiver trainingand support (Exhibit 2). The number of servicesclients received ranged anywhere from 0 to 10, witha distribution mode of 3 services received (23 per-cent, n = 168). The average number of servicesreceived was 3.68 (SD = 1.85). Analysis revealed asignificant (p < .01) negative association betweennumber of services received and interest in the cashoption. Descriptive statistics and frequencies forservices received are presented in Exhibit 3.

Five other items related to consumer satisfactionwere included on the survey, as well as the 7-pointsatisfaction subscale. One indication of satisfactionis whether a consumer has had to dismiss a worker;23 percent (n = 171) of these respondents had thisexperience. Of those, 53 percent (n = 91) stated thereason for dismissal was that the worker wasn't do-ing a good job. An additional 23 percent (n = 39)indicated the worker "had a bad attitude," 10 per-cent (n = 17) said they didn't get along with theworker, and 8 percent (n = 14) stated their workerstole from them. Another sign of satisfaction has todo with consumers' perception of the adequacy ofhours of home care services they receive. A majorityof respondents (55 percent, n = 394) believed thenumber they received was "just about right," while28 percent (n = 201) stated they were "a little low"and 14 percent (n = 103) said "much too low." Fi-nally, 2 percent (n = 2) said the number of hoursthey received was too high, and an additional 2 per-cent (n = 2) did not have an opinion.

Fifty-seven percent of respondents (n = 411)stated they were satisfied with the availability ofhome care services in an emergency, while 15 per-cent (n = 105) were dissatisfied and 24 percent(n = 171) chose the "don't know" option for thisitem. Five percent (n = 33) stated the question wasn'tapplicable. We asked consumers if one of the thingsthey didn't like about their services was that the con-sumer and worker did not speak the same language.Nine percent of respondents (n = 66) agreed withthis statement, and the remainder either didn't know,said the statement wasn't applicable, or disagreed(82 percent, n = 610).16

Satisfaction was also measured via the seven-itemsubscale mentioned previously. Three items werepositively worded with the following stem: "One ofthe things I like most about my current home careservices is:" followed by (a) the worker, (b) the sched-ule, and (c) that the agency makes sure the worker isdoing her/his job. Four items were negatively wordedwith the stem: "One of the things I really don't likeabout my home care services is:" followed by (a) thetime of day my worker arrives, (b) my lack of con-trol over the services, (c) that the agency doesn'tinform me of changes being made, and (d) that myworker is not properly trained. Positive responsesto the first three items and negative responses to thelast four each received a score of 1, resulting in asatisfaction subscale score that ranged from 0 (most

8 I Elder's Advisor

Exhibit 2. Services Received and Satisfaction with Services Among Florida Elderly and Adult Consumers

Service

Case manager servicesYesNoDon't know

Case aide servicesYesNoDon't know

Homemaker servicesYesNoDon't know

Personal care servicesYesNoDon't know

Adult day health servicesYesNoDon't know

Home modification servicesYesNoDon't know

Chore servicesYesNoDon't know

Res pite servicesYesNoDon't know

Personal emergency response system servicesYesNoDon't know

Consumable medical suppliesYesNoDon't know

Counseling servicesYesNoDon't know

Nutritional servicesYesNoDon't know

Home delivered mealsYesNoDon't know

Care iver training and supportYesNoDon't know

Receives ServiceN (%)

543 (73.1)158 (21.3)42 (05.7)

150 (20.2578 77.8)

15 02.0'

531 (71.6)

209 (28.2)2 (00.3)

454 (61.1)

285 (38.4)4 (00.5)

60 (08.1)

675 91.0)7 00.9'

177 (23.9)

563 75.9)2 00.3'

106 (14.3)

632 (85.2)4 (00.5)

142 (19.2)571 (77.2)27 (3.6)

264 (35.6)

473 (63.7)5 (00.7)

273 (36.8)

461 (62.1)8 (01.1)

72 (09.7)

667 (89.9)3 (00.4)

39 (05.3)

697 (93.9)6 (00.8)

294 (39.6)

447 (60.2)1 (00.1)

26 (03.5)

687 (92.7)28 (03.8)

Satisfied with ServiceN (%)

490 90.7)

25 04.6)25 04.6)

132 (89.86 (04.19 06.1'

437 (82.8)69 (13.1)22 (04.2)

401 (88.5)

38 (08.414 03.1'

54 (90.0)

4 (06.7)2 (03.3)

160 (90.4)13 07.3)4 02.3)

85 (80.2)

13 (12.3)8 (07.5)

129 (90.8)

7 (04.9)6 (04.2)

236 (90.1)

6 (02.3)20 (07.6)

257 93.5)

15 05.5)3 01.1)

64 (88.93 04.2)5 06.9'

36 (92.3)

3 07.7)0 00.0'

232 (78.9)42 (14.3)20 (06.8)

21 (75.0)

4 (14.3)3 (10.7)

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services 9

Exhibit 3. Descriptive Statistics for Services Received

Service Received

Case manager services (contacts per month)

Case aid services (hours per week)

Homemaker services (hours per week)

Personal care services (hours per week)

Adult day health service (hours per week)

Home modifications (number ever)

Chore services (hours per month)

Respite service (days per month)

Personal emergency response service (times ever)

Consumable medical supplies (times per month)

Counseling services (hours per week)

Nutritional service (times ever)

Home delivered meals (hours per week)

Caregiver training and support (times per month)

dissatisfied) to 7 (most satisfied). Eighty-one per-cent of respondents who answered all seven items(573/706) scored a 4 or higher on this subscale.

Interest In the Cash OptionValid responses for the item concerning interest inthe cash option were received from 511 consumersand 43 surrogates answering on behalf of the con-sumer. Approximately 59 percent (n = 301) ofconsumers answering for themselves indicated inter-est in the cash option, with an additional 20 percent(n = 102) indicating they were not sure of their inter-est (Exhibit 4). When examining responses ofsurrogates who answered for consumers, 49 percent(n = 21) indicated that the consumer would be inter-ested while 26 percent (n = 11) were not sure of theconsumer's interest level. Finally, surrogates wereasked for their own opinion; 62 percent (n = 31)

Min Max Mean SD Median Mode N

1 33 2.00 3.66 1 1 394

1 50 7.30 8.23 5 3 128

1 49 3.93 3.64 3 2 515

1 50 5.87 5.12 5 3 433

1 56 3.93 2.74 10 2 53

1 13 1.94 1.57 1 1 166

1 25 4.51 5.92 2 1 84

1 28 5.63 5.17 4 4 122

1 60 3.27 6.47 2 1 120

1 30 1.72 2.93 1 1 236

1 23 2.35 4.11 1 1 49

1 12 3.29 2.45 3 1 32

1 14 4.25 2.20 5 5 287

1 36 7.15 9.91 2 1 13

stated they liked the idea of the cash option, while24 percent (n = 12) said they didn't know (Exhibit4). Overall, 58 percent (322/554) of the consumersand 62 percent (31/50) of the surrogate respondentswere interested in the cash option.

Bivariate analyses were conducted to assess fac-tors associated with interest in the cash option. Caseswere included in the bivariate analysis if respondentshad answered the interest in the cash option itemand the specific demographic or background vari-able being examined. Consumers differed signifi-cantly (p < .001) in their interest in the cash optionbased on their age. There were 189 consumers un-der age 65 who answered the item regarding interestlevel, and of these, 71 percent (n = 135) were inter-ested in the option, 12 percent (n = 23) were notcertain of their interest, and 16 percent (n = 31) werenot interested. Examination of the 355 consumers

I Elder's Advisor

Exhibit 4. Interest in the Cash Option by Respondent Status

Respondent Status

Consumer answering for self

Surrogate answering for consumer

Surrogate answering for self

age 65 and older who answered the item showed 51percent (n = 182) were interested, 25 percent (n =89) were not sure, and 24 percent (n = 84) were notinterested in the option.

Consumer interest also differed significantly (p< .01) by level of disability, as measured by the ac-tivities of daily living scale. There were 122consumers with severe, 146 with moderate, and 270with mild disability levels who also answered theinterest item. Consumers with more severe disabili-ties were more likely to be interested in the option.Seventy-one percent of consumers with the severedisability rating were interested compared to 57 per-cent of those with the moderate rating and 55 percentof those with the mild rating. Consumer interest didnot differ significantly by gender, race, level of edu-cation, marital status, living arrangement, homeownership, having ever been employed, current em-ployment status, length of time in the Florida HomeCare program or self-rating of overall health status.

Consumers who desired more involvement withtheir current personal care services (n = 215) weresignificantly more likely to be interested in the cashoption compared to those who desired the same orless involvement (n = 292) (73 percent versus 49percent, p < .001). Consumers who reported havingexperience hiring, firing, or interviewing workers(n = 187) and consumers who had experience super-vising or training workers (n = 245) were significantlymore likely to be interested when compared to thosewho did not have such experience (71 percent ver-sus 53 percent and 68 percent versus 51 percent,respectively) (p < .001).

Consumers who reported having an informalcaregiver (n = 331) were significantly more likely(p < .01) to be interested in the cash option when

InterestedN (%)

301 (58.9)

21 (48.8)

31 (62.0)

Not SureN (%)

102 (20.0)

11 (25.6)

7 (14.0)

Not InterestedN (%)

108 (21.1)

11 (25.6)

12 (24.0)

compared to those who did not have one (n = 209),(63 percent versus 52 percent). For those who havean informal caregiver, no significant difference in in-terest was found based on whether the informalcaregiver lived with the consumer or not. Consum-ers who felt the number of hours of home care servicewere much too low (n = 89) were more likely to beinterested in the option than those who felt the hourswere a little low (n = 164) or those who indicatedthe hours were just about right (n = 284) (75 percentversus 67 percent versus 49 percent, p < .001). Con-sumers dissatisfied with the availability of home careworkers in an emergency (n = 83) were more likelyto be interested in the cash option when comparedto those who were not sure of their satisfaction withthis service (n = 143) or those who were satisfied(n = 301) (70 percent versus 56 percent versus 55percent, p < .05).

Respondents were asked if they would be will-ing to sign up for the program even if, due torandomization procedures, there was a chance theymight not get into the program. Of those who indi-cated interest in the program, 80 percent (n = 258)were willing to sign up, and of those who indicatedthey were unsure of their interest in the program, 35percent (n = 40) were willing to sign up. Respon-dents were also asked if it would be more importantto know the exact amount of money they would re-ceive under the cash and counseling option, or toknow that the amount was close to what the statenow pays the agency. Of those interested in the op-tion, 59 percent thought it was more important toknow the exact amount, while for those unsure oftheir interest (n = 113) and those not interested(n = 119), the corresponding percentages were 54percent and 33 percent, respectively (p < .001).

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services 31

Examination of the surrogate subsample foundthat their interest in the option differed significantlyby age, with the 29 surrogates under 65 more likelyto be interested (79 percent) versus the 17 surrogates65 or older (35 percent) (p < .01). Surrogate interestin the option did not differ significantly by gender,race, level of education, marital status, relationshipto consumer, or current living arrangement.

Attractive Cash Option CharacteristicsConsumer ratings of the importance of various pro-gram characteristics differed significantly by level ofinterest in the cash and counseling option for each

of the four program characteristic items. Consum-ers who indicated interest (n varied from 319 to 321by question) were more likely than those not inter-ested (n varied from 115 to 117 by question) toconsider it important to be able to hire their currentworker (76 percent versus 44 percent), to pay theirworker more money (67 percent versus 29 percent),to know a group of other consumers who were par-ticipating (74 percent versus 29 percent), and to beable to back out of the cash and counseling optionif they so desired (83 percent versus 58 percent)(p < .001) (Exhibit 5). Significant differences werealso found by consumer age for three of these

Exhibit 5. Consumers' Perceptions of the Importance of Cash Option Characteristics by Interest Level

Cash Option Characteristic

Hire current workerImportantDon't knowNot importantNot applicableRefused question

Pay worker moreImportantDon't knowNot importantNot applicableRefused question

Know a group of other consumersImportantDon't knowNot importantNot applicableRefused question

Ability to back out of programImportantDon't knowNot importantNot applicableRefused question

InterestedN (%)

24626420602

21665320603

2382652051

26722270303

(76.4)(08.1)(13.0)(01.9)(00.6)

(67.1)(20.2)(09.9)(01.9)(00.9)

(73.9)(08.1)(16.1)(01.6)(00.3)

(82.9)(06.8)(08.4)(00.9)(00.9)

Not SureN (%)

(51.3)(29.2)(14.2)(05.3)

(39.8)(44.2)(11.5)(03.5)(00.9)

(55.8)(23.0)(18.6)(02.7)

(69.9)(21.2)(05.3)(03.5)

Not InterestedN (%)

(43.7)*(17.6)(28.6)(07.6)(02.5)

(29.4) *(22.7)(37.0)(07.6)(03.4)

(29.4)*(16.8)(42.9)(09.2)(01.7)

(58.0)*(14.3)(17.6)(07.6)(02.5)

* p < .001

12 1 Elder's Advisor

program characteristics, with consumers under 65(n varied from 197 to 198) more likely than those65 or older (n varied from 376 to 380) to consider itimportant to hire their current worker (74 percentversus 56 percent) (p < .001), to pay their workermore money (62 percent versus 46 percent) (p < .01),and know a group of other consumers in the pro-gram (69 percent versus 53 percent) (p <.01). Whenexamining surrogate responses concerning these im-portant program characteristics, no significantdifferences were found by interest in the cash optionor age of respondent.

Consumers were asked whether particular pro-gram characteristics (i.e., getting services when youwant them, hiring whomever you want including afriend or relative, and buying different services)would make them interested in the cash option. Foreach program characteristic, interested consumerswere significantly (p < .001) more likely to say the

characteristic made them more interested in the pro-gram when compared to those who weren't sureof their interest and those who were not interested(Exhibit 6).

Surrogates were also asked a series of questionsto ascertain reasons for their interest in the cash op-tion. Those interested in the option were significantly(p < .001) more likely to agree with the reason statedwhen compared to those not sure of their interest ornot interested (Exhibit 7). For example, interestedsurrogates were more likely to believe the consumerwould like to participate, that the cash option wouldoffer more choice, flexibility and independence tothe consumer and themselves, and that they woulduse the cash option to hire a friend or relative. Sur-rogates interested in the cash option were alsosignificantly more likely to be willing to hire a per-sonal care worker, show the worker what todo, arrange a worker's schedule, and pay a worker

Exhibit 6. Consumers' Reasons for Interest in the Cash Option by Actual Interest Level

Consumer Interest in Cash and Counseling OptionInterestedN (%)Reason for Interest

Get services when you wantInterestedDon't knowNot interestedNot applicableRefused question

Hire whomever you wantInterestedDon't knowNot interestedNot applicableRefused question

Buy different servicesInterestedDon't knowNot interestedNot applicableRefused question

2961115

29507160202

29015150101

Not SureN (%)

(91.9)(03.4)(04.7)

(91.6)(02.2)(05.0)(00.6)(00.6)

(90.1)(04.7)(04.7)(00.3)(00.3)

(55.8)(23.9)(17.7)(02.7)

(58.4)(20.4)(17.7)(03.5)

(61.9)(18.6)(15.9)(03.5)

Not InterestedN (%)

(28.6)*(10.9)(56.3)(03.4)(00.8)

(20.2)*(05.0)(70.6)(03.4)(00.8)

(26.1)*(03.4)(67.2)(02.5)(00.8)

* p < .001

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services 13

Exhibit 7. Surrogates' Reasons for Interest in the Cash Option by Interest Level

Surrogate Interest in Cash and Counseling Option

Reason for Interest

Thinks consumer would like to participateAgreeDisagreeNot applicableDon't know

Cash could offer more choice, flexibility,and independence for consumer

AgreeDisagreeDon't know

Cash could offer more choice, flexibility,and independence for the surrogate

AgreeDisagreeDon't know

Would like cash option to interview andhire workers

AgreeDisagreeDon't know

Would like cash option to hirea friend or relative

AgreeDisagreeDon't know

InterestedN (%)

(62.5)(03.1)(03.1)(31.3)

(68.8)(06.3)(25.0)

(78.1)

(21.9)

(90.3)

(06.5)

(77.4)(16.1)(06.5)

Not SureN (%)

Not InterestedN (%)

(14.3)(14.3)(14.3)(57.1)

(28.6)(28.6)(42.9)

(42.9)

(57.1)

(42.9)(03.2)(57.1)

(57.1)(14.3)(28.6)

(08.3)*(75.0)

(16.7)

(16.7)*(75.0)(08.3)

(16.7)*(75.0)(08.3)

(08.3)*(66.7)(25.0)

(16.7)*(75.0)(08.3)

* p < .001

(p < .001), as well as to fire and supervise a worker(p < .05). No significant differences were found forthe surrogate's willingness to perform duties associ-ated with the cash option by the age of the surrogate.Finally, surrogates were asked if they thought thecash option would make it easier or harder on them.Thirty-four percent (n = 16) believed the optionwould make their job easier, while 17 percent (n = 8)believed it would be harder.

Services Consumers Want to PurchaseRespondents were asked about their interest in pur-chasing various services, and again, those interestedin the cash option were significantly (p < .001) morelikely to want to purchase each of the services whencompared to those not interested or not sure of theirinterest level. For example, a majority of respon-dents interested in the option (n = 301) wanted topurchase more hours of service (80 percent, n = 240),

I Elder's Advisor

housekeeping services (80 percent, n = 240), trans-portation services (70 percent, n = 209), respite careservices (62 percent, n = 185), laundry services (59percent, n = 176), grab bars or shower equipment

(55 percent, n = 166), and wheelchairs or other equip-,ment (50 percent, n = 151) (Exhibit 8). A majority ofthose not sure of their interest in the option (n = 102).were interested in purchasing housekeeping services

Exhibit 8. Services Consumers Would Like to Purchase by Cash Option Interest Level

Consumer Interest in Cash and Counseling Option

Services Would Like to PurchaseInterestedN (%)

Not SureN (%)

Not InterestedN (%)

More hours of serviceYesNo

Grab bars or shower equipmentYesNo

Adult day health programYesNo

Home modificationsYesNo

Transportation servicesYesNo

Laundry servicesYesNo

Housekeeping servicesYesNo

Respite careYesNo

Wheelchairs and other equipmentYesNo

240 79.7 4861 20.3 54

166 55.1 31135 44.9 71

107 35.5 19194 64.5 83

138 45.8 22162 53.8 80

209 69.492 30.6

176 58.5125 41.5

240 79.761 20.3

185 61.5 33116 38.5 69

151 50.2 34150 49.8 68

* p < .001

47.152.9

30.469.6

18.681.4

21.678.4

47.152.9

39.260.8

55.944.1

32.467.6

33.366.7

21.3*78.7

13.9*86.1

13.0*87.0

12.0*88.0

25.9*74.1

23.1*76.9

34.3*65.7

22.2*77.8

25.9*74.1

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services 15

(56%, n = 57) (Exhibit 8).Consumers under 65 (n = 199) were significantly

more likely than those 65 or older (n = 387) to beinterested in purchasing home modification services(41% versus 26%) and laundry services (55% ver-sus 37%) (p < .001), as well as grab bars and showerequipment (47% versus 35%), transportation ser-vices (60% versus 47%), and housekeeping services(68% versus 56%) (p < .01). Consumers underage 65 were also more likely to be interested in pur-chasing more hours of service (64% versus 54%),adult day health services (32% versus 23%), andwheelchairs or other equipment (46% versus 36%)(p < .05).

Consumer and Surrogate Training andSupport NeedsSeven different tasks associated with the cash op-tion were included in the survey, and the majority ofconsumers wanted assistance or training in each ofthese areas. However, respondents interested in thecash option (n varied from 312 to 318 by item) weresignificantly (p < .001) more likely to want help ortraining on each task when compared to the othertwo groups (Exhibit 9). They were more likely towant help doing a background check on a worker(82 percent), deciding how much to pay a worker(79 percent), assistance with payroll taxes (77 per-cent), and knowing what to do when a workerdoesn't show (73 percent). No significant differenceswere found for the need for help or training on thesetasks by the age of the consumer.

Examination of surrogates' perceptions, whenresponding to the items about their training needs,found no significant differences in need for help ortraining by interest level in the cash option on six ofthe seven tasks. However, surrogates interested inthe cash option (n = 31) and those not certain oftheir interest (n = 7) were significantly (p < .05) morelikely to indicate a need for help in knowing what todo if a worker did not show up than were those notinterested in the option (n = 9), (84 percent versus86 percent versus 44 percent, p < .05) (Exhibit 10).

Additional Information NeededOverall, before deciding whether to be involved inthe cash option, the majority of consumers, regard-less of their interest level, wanted more information.However, there were statistically significant differ-ences (p < .001) among the three interest-level groups

on each of these items-respondents were even morelikely to want information if they stated they wereinterested in the option or not certain. Specifically,when asked if they needed to know more financialdetails, 92 percent (n = 293) of those interested, 87percent (n = 96) of those not sure, and 69 percent (n= 75) of those not interested answered "yes" (over-all 86 percent). When asked if they needed to knowwhether their current worker could be retained, 85percent (n = 268) of those interested, 82 percent (n =90) of those not sure, and 77 percent (n = 82) ofthose not interested responded "yes" (overall 83percent). When asked if they needed to know howother current benefits they received would be af-fected, the percentage breakdown was 95 percent(n = 300), 86 percent (n = 95), and 84 percent(n = 86) respectively (overall 91 percent). Finally,when asked if they needed to know more about theirrights and responsibilities under the cash option, 96percent (n = 305) of the interested consumers, 92percent (n = 101) of those not sure, and 84 percent(n = 85) of those not interested responded "yes"(overall 93 percent). There were no statistically sig-nificant differences in need for more information byage of respondents. Again, the majority of respon-dents (84 percent to 93 percent) wanted additionalinformation, regardless of their age.

Discussion and RecommendationsSurvey results provide Florida with informationabout the type of services their clients actually re-ceive and satisfaction with those services. Resultshave guided Florida in designing numerous aspectsof the cash option; however, this discussion will fo-cus on the implications of survey findings forFlorida's communications and social marketing ef-forts, as well as policy issues.

Approximately three-fourths of Florida consum-ers received case management, homemaker, andpersonal care services, and 40 percent received home-delivered meals. There was a negative associationbetween number of services received and interest inthe cash option; as the number of services decreased,respondents were more likely to be interested in theoption. It is possible these respondents believe theymay receive more services with the cash option, asthey would have more control, personally decidingwhich services to purchase. Alternatively, individu-als who receive many services may be loath to giveup case management, as the sheer number of ser-

I6 I Elder's Advisor

Exhibit 9. Consumers Who Want Help or Training with Task by Cash Option Interest Level

Consumer Interest in Cash and Counseling Option

Task Want Help With

Help finding a workerYesNoDon't know

Help interviewing a workerYesNoDon't know

Help doing a background checkYesNoDon't know

Help deciding how much to payYesNoDon't know

Help if worker didn't showYesNoDon't know

Help firing workerYesNoDon't know

Help with payroll taxesYesNoDon't know

InterestedN (%)

1939510

1701208

242475

2375211

228619

Not SureN (%)

64.831.93.4

57.040.32.7

82.316.01.7

79.017.33.7

73.125.21.7

50.248.81.0

76.520.53.0

67.315.317.3

52.031.616.3

76.513.310.2

73.511.215.3

68.419.412.2

45.939.814.3

68.418.413.3

Not InterestedN (%)

52.0*36.711.2

45.3*

47.47.4

52.6*37.99.5

51.0*37.511.5

56.4*33.010.6

44.7*48.96.4

49.5*41.19.5

* p < .001

vices they receive requires much coordination andthe complexity of operationalizing the cash plan maybe greater.

Satisfaction with current services was measuredin several ways, and it appears that the vast major-

ity of Florida consumers were very satisfied with theircurrent services (75 percent to 94 percent), and 81percent scored a 4 or higher on the 7-point satisfac-tion subscale, indicating they were satisfied.However, it appears dissatisfaction with specific char-

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services 17

Exhibit 10. Surrogates Who Want Help or Training with Task by Cash Option Interest Level

Surrogate Interest in Cash and Counseling OptionInterestedN (%)Task Want Help With

Finding workerYesNoDon't know

Interviewing workerYesNoDon't know

Doing background checkYesNoDon't know

Deciding how much to payYesNoDon't know

What to do if worker doesn't showYesNo

Firing worker if doesn't work outYesNoDon't know

Payroll taxesYesNoDon't know

p < .05

acteristics of current services creates more interestin the cash option. If consumers believed the ser-vices they currently received were much too low, theywere more likely to be interested in the cash option.Also, if they were dissatisfied with the availability of

64.532.3

54.838.7

6.5

77.419.43.2

77.416.16.5

83.916.1

38.758.13.2

71.025.83.2

Not SureN (%)

57.114.3

3.2

57.114.328.6

57.114.328.6

71.414.314.3

85.714.3

28.642.928.6

71.414.314.3

Not InterestedN (%)

70.020.028.6

37.562.5

55.644.4

44.455.6

44.4*55.6

44.455.6

66.722.211.1

a worker in an emergency situation, they were sig-nificantly more likely to be interested in the cashoption. Florida can be very pleased with the highlevel of consumer satisfaction. For social marketingand enrollment purposes however, it would not be

18 I Elder's Advisor

sufficient for Florida to focus on dissatisfied con-sumers as they are in the minority.

Identifying Interested and Uncertain Consumersand SurrogatesIt is likely Florida can achieve its evaluation enroll-ment requirement by focusing on individuals whoindicated an initial interest in the cash option-54percent of the sample was interested. In addition,23 percent of respondents stated they were unsureof their interest in the cash option. These individu-als may decide to enroll if their concerns aresatisfactorily addressed. When surrogates were askedif the consumer would be interested in the cash op-tion, 49 percent said "yes," but when asked if theywere personally interested, an even higher percent-age said "yes" (62 percent). It is unclear whysurrogates believed fewer consumers would be in-terested in the option than they were themselves.Perhaps this is simply an indication that surrogatestruly tried to answer for the consumer and did notallow their personal opinion to color the consumer'schoice.

One of the survey's major research questions in-quired about age as a factor influencing interest inthe cash option. Although there was a significantdifference in interest in the cash option between theyounger (< 65) and older (2! 65) age groups (71 per-cent versus 51 percent), a high percentage of olderconsumers were interested. (Younger surrogates werealso more likely to be interested in the option whencompared to older surrogates (79 percent versus 35percent)). These findings support the work of oth-ers, 17 who suggest that there may be sizable interestin consumer-directed services among older clients.In Florida this was certainly the case. Therefore,social marketing efforts should focus on consumersof all ages, not just younger consumers. Bivariateanalyses indicated that in Florida (unlike the otherthree demonstration states) there were no significantdifferences in interest in the cash option based onrace or gender. Surrogates also did not differ signifi-cantly on any of the demographic variables, with theexception of age as previously mentioned. This maymake social marketing efforts easier given the broaderpotential client base.

There were significant differences among thoseinterested, not interested, or not sure of their inter-est on a variety of variables. For example, those whowere severely disabled were more likely to be inter-

ested in the option when compared to those whowere mildly or moderately disabled. Prior to datacollection, there was speculation regarding this vari-able. Some believed the most disabled individualswould not be able to manage all the tasks associatedwith the option and that the majority of those par-ticipating would be only mildly disabled. However,the data did not support this speculation; perhapsseverely disabled consumers were especially excitedabout the flexibility and control offered via the cashoption. Age of onset of disability also significantlyincreased interest level; the younger the age of on-set, the more likely the consumer was to be interestedin the cash option. However, age of onset of disabil-ity was also related to the consumer's current age(p < .001). When current age was statistically con-trolled for using partial correlation, the relationshipbetween age of onset and interest in the cash optionwas no longer significant (p > .60).

Thirty-four percent of Florida consumers hadexperience hiring, firing, and interviewing workers,and 44 percent had experience supervising and train-ing workers. Those who had these experiences weresignificantly more interested in the cash option whencompared to those who had not. Perhaps those withpast experience supervising, training, or hiring oth-ers (in any capacity) are more comfortable takingon some of the tasks related to the cash option, asthey already know they can be successful.

Survey data indicated the highest level of inter-est in the cash option was among surrogate decisionmakers when expressing their own views (62 per-cent) followed by consumers answering forthemselves (60 percent). In addition, respondentswho had an informal caregiver were more interestedin the cash option when compared to those who didnot. These findings may be related, as surrogatedecision makers are likely to be informal caregivers.Florida needs to learn more about the reasons forsurrogates' high level of interest in the cash option,as well as their role in working with a consumer tochoose the cash option. The same is true for theinformal caregiver-how does the presence of thisindividual influence a consumer's decision to selector not select a cash option? One possible explana-tion is that the informal caregiver could serve as theemergency back-up person if the paid worker didn'tshow up, an important concern expressed by con-sumers. Without support from informal caregivers,many consumers may lack the confidence to take on

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services 19

the additional responsibilities required in the cashoption.

Communications and Social MarketingMessages to EmphasizeAn important factor to further explore in relation tosocial marketing efforts is educational level. Forty-six percent of Florida consumers had less than a highschool education, and 27 percent were high schoolgraduates but did not go on to college. These find-ings serve as potent reminders that outreach andtraining materials must be simple and straightfor-ward (and/or many consumers may need theassistance of surrogates and direct contact with coun-selors or trainers).

The survey data offer detailed guidance in as-sisting Florida in the development of communicationsand social marketing materials, and more impor-tantly, provide direction for designing the cashoption. For example, the majority of interested con-sumers thought it would be important to hire theirown worker, to pay the worker more money if theydesired, to know a group of others participating inthe program, and to be able to back out of the pro-gram if they wished."i In fact, being able to backout of the program was important to a majority ofall consumers (70 percent to 83 percent), regardlessof interest level. Obviously, having a safety net (i.e.,being able to back out without being penalized) iscritical. This was expected, as change is difficult forpeople, even a desired change. This may be espe-cially true for personal assistance services as they areessential to daily living and functioning. Finally,consumers appear to appreciate the importance ofsocial support when embarking on a change.

Seventy-six percent of interested consumerswanted to be able to hire their current worker. Thisconsumer preference has important implications forFlorida. It is understandable that once consumershave a worker they like they would want to con-tinue with that person, especially as focus-groupfindings indicated many consumers had unsatisfac-tory experiences before finding a worker they liked.However, this desire poses difficult organizationalissues. It is possible provider agencies may try tolimit this practice. In addition, a worker may needfull-time employment but only work part-timehours for a specific consumer. This issue is likelyto be less important for new consumers entering theMedicaid program, as they would be less attached

to the existing arrangement.Surrogates' reasons for being interested in the

cash option also offer messages to include when ad-dressing that group. Surrogates' interest in the cashoption was significantly related to their willingnessto assume responsibility for employer tasks, a con-cept to highlight in social marketing messages.Materials should be sure to highlight the surrogate'sability to interview and hire workers (even a friendor relative), the possibility that the cash option wouldmake things easier for surrogates, increase flexibil-ity for both consumers and surrogates, as well asprovide potential benefits to the consumer.

Effective social marketing materials also need toaddress the consumer's ability to "get services on thedays and at the times you want,". "hire whomeveryou want to provide personal care services, even afriend or a relative," and "buy different services,"as the vast majority of interested consumers foundthese program characteristics appealing. The ma-jority of consumers interested in the cash optionwanted to purchase more hours of service, house-keeping services, and transportation services, so eachof these should also be included in social marketingmaterials.

Consumers interested in the cash option weremore likely to express a need for help or training inemployer tasks. Social marketing materials shouldbe sure to inform consumers that they could havehelp or training with the most requested tasks. Asthe majority of consumers wanted more informa-tion before deciding whether to choose the cashoption, social marketing materials and in-personcommunication should be as specific as possible re-garding the following issues: consumers' rights andresponsibilities under the cash option, assurance thatother current benefits would not be affected, cashoption financial details, and an explanation of howthe current worker would be affected.

Policy IssuesThe CCDE is a policy-driven project addressing nu-merous policy concerns. While comprehensiverecommendations will not be available until theevaluation is complete, the Florida preference sur-vey offers insight into policy issues concerning theimportance of offering consumers a choice of per-sonal assistance services options as well as insightsregarding potential fraud and abuse, and servicequality. The CCDE is based on the premise that the

20 I Elder's Advisor

cash option is a choice available to those consumerswho want consumer-direction. It is not intended toreplace traditional services, as the cash option isunlikely to be appropriate for or desirable to all con-sumers.

Fraud and abuse concerns, related to the possi-bility that consumers and/or their families mightmisuse the cash benefit or be exploited by others, 19

must also be considered. While the demonstrationneeds to address these concerns, procedures to mini-mize fraud and abuse must maintain the consumerempowerment principles being tested in the CCDE.Overly restrictive measures could negate the effectof the consumer-directed intervention.

Misuse of the cash benefit includes the possibil-ity that consumers might not pay taxes or theirworkers. Florida survey data indicate these possi-bilities are limited as a majority of consumers andsurrogates interested in the cash option (77 percentand 71 percent, respectively) said they wanted helpor training with payroll and taxes. Most clients arelikely to elect to have the payroll and tax withhold-ing for their workers done by accountingprofessionals. This would greatly reduce the amountof cash consumers receive and manage. 20 Those con-sumers electing not to use accounting professionalswill need to participate in a training program andpass a skills test in payroll tasks. Any training pro-gram Florida develops must take into considerationthat 46 percent of respondents did not complete highschool and only 27 percent graduated from highschool.

One may also question whether the cash benefitwill be adequate to provide the level of service thatconsumers need as Florida offers a relatively small ben-efit level ($389 monthly is the weighted average fordifferent programs). This may be especially pertinentfor severely disabled consumers with high levels of need.While it is not possible to comment on the adequacyof the benefit until the evaluation is complete, it is im-portant to note that the amount of the cash benefitwill approximate the dollar amount spent on theconsumer's current service plan (and will be based uponthe same assessment process used in the traditionalprogram). The evaluation will compare cost, quality,and satisfaction with service in both the cash optionand traditional services and will be able to compareadequacy of service in both models.

To prevent consumer exploitation by others (andsubsequent suffering of ill effects), the cash option

allows and encourages the use of surrogate decisionmakers to represent consumers who are unable tomake all decisions independently. (Surrogates arenot paid for their assistance.) While there are manyquestions to consider regarding surrogate decisionmakers, we know from the Florida survey that 9percent of consumers utilized surrogates and 62 per-cent of surrogates responding for themselves (versusrepresenting a consumer) were interested in the cashoption. In the event of possible exploitation by asurrogate, it is important to note that, under the cashoption, counselors will have a role in monitoring allconsumers-even those with surrogates.

For those consumers functioning independently,without surrogates, the cash option training and sup-port services offer further protection againstconsumer exploitation. When asked whether theywould want help or training with various cash op-tion tasks, the vast majority of consumers whowere interested in the cash option wanted these sup-port services. In Florida, all consumers andsurrogates will be required to participate in a train-ing program.

Florida survey respondents found the ability to"hire whomever you want to provide personal careservices, even a friend or relative" an attractive fea-ture of the cash option. This finding indicates thatconsumers are likely to hire friends or relatives astheir workers. Policy makers often raise concernsabout the quality of care provided by friends or rela-tives as they may lack formal training. Yet, twostudies of California's In-Home Support Services pro-gram21 found that consumers rated family membersand friends as more reliable than workers who werestrangers. In addition, a study of elderly Medicaidpersonal care recipients in Michigan, Texas, andMaryland found that client satisfaction was relatedto several indicators of greater client control, andspecifically, to Michigan's policy of encouragingclients to hire family, friends, and neighbors asattendants. 22 The Florida survey indicates theCCDE will further our understanding about the qual-ity of services when friends and relatives become paidproviders.

SummaryThis article has presented results from a telephonesurvey conducted to assess the preferences of eldersand adults with physical disabilities for a cash op-tion versus traditional services in Florida, one

ARTICLE I Consumer Preferences for a Consumer-Directed Cash Option Versus Traditional Services

demonstration state in the CCDE. The telephonesurvey was conducted as background research. Sur-vey findings have guided Florida in designing the cashoption and developing much needed communica-tions, training, and social marketing materials. Theseefforts are essential to informing Florida consumersabout the cash option so they can make thoughtful

Endnotes

1. The CCDE is cosponsored by the Robert WoodJohnson Foundation and the U.S. Department ofHealth and Human Services, Office of the AssistantSecretary for Planning and Evaluation.

2. Lori Simon-Rusinowitz et al., The Status of aCommon Agenda Between the Aging and DisabilityCommunities: Where Has It Been? Where Is ItGoing? Views from Policy Experts, in ANNUAL

REVIEW OF ETHIcs, LAW, AND AGING, at 85-106(Marshall Kapp ed., 2000).

3. Gerben DeJong et al., The Independent LivingModel of Personal Assistance in National Long-Term Care Policy, 16 GENERATIONS 89-95 (1992).

4. Edward F. Ansello & Nancy N. Eustis, A CommonStake? Investigating the Emerging Intersection ofAging and Disabilities, 16 GENERATIONS 5-8 (1992);UNIV. OF CAL. & WORLD INST. ON DISABILITY, TOWARD

A UNIFIED AGENDA: PROCEEDINGS OF A NATIONAL

CONFERENCE ON DISABILrTY AND AGING (C.W.Mahoney et al. eds., 1986); Lori Simon-Rusinowitz& B.F. Hofland, Adopting a Disability Approach toHome Care Services for Older Adults, 33 GERON-TOLOGIST 159-67 (1993).

5. Simon-Rusinowitz et al., supra note 2.

6. Health Security Act, H.R. 3600, 103d Congress (2dSess. 1994), S1757, 103d Congress (2d Sess. 1994),cited in Marshall Kapp, Enhancing Autonomy andChoice in Selecting and Directing Long-Term CareServices, 4 ELDER L.J. 55-97 (1996).

7. Much of this background section comes frombackground materials written by Pamela Doty, theCCDE's project officer at DHHS, ASPE.

8. SIMI LrrvAK ET AL., WORLD INST. ON DISABILITY,ATTENDING TO AMERICA: PERSONAL ASSISTANCE FOR

INDEPENDENT LIVING, A SURVEY OF ATTENDANT SERVICE

PROGRAMS IN THE UNITED STATES FOR PEOPLE OF ALL

decisions about choosing a consumer-directed op-tion or staying in the traditional program. As theCCDE continues, we will not only learn how con-sumers fare in the cash option, but also how thisrigorous social experiment will offer valuable les-sons about how to implement consumer-directedprograms in a "real world" setting.

AGES WITH DISABILITIES (1987); SIMI LITvAK & J.KENNEDY, WORLD INST. ON DISABILITY, NEW MODELS

FOR THE PROVISION OF PERSONNEL ASSISTANCE SERVICES

app. B (1990); SImI LrrvAK & J. KENNEDY, WORLDINST. ON DISABILITY, POLICY ISSUES AND QUESTIONS

AFFECTING THE MEDICAID PERSONAL CARE SERVICES

OPTIONAL BENEFIT (1991).

9. S. GERON & D. CHASSLER, CONN. CMTY. CARE,

GUIDELINES FOR CASE MANAGEMENT PRACTICE ACROSS

THE LONG-TERM CARE CONTINUUM (1994); M.E.JACKSON, RATIONING CASE MANAGEMENT: SIX CASE

STUDIES (1994) (report prepared for the Office of theAssistant Secretary for Planning and Evaluation,Department of Health and Human Services).

10. Kapp, supra note 6.

11. For a detailed description of the CCDE design, seeKevin J. Mahoney et al., Early ImplementationLessons from the Cash and Counseling Demonstra-tion and Evaluation, 24 GENERATIONS 41-46 (2000).

12. C. BARNES & S. SUTHELAND, CAL. STATE UNIV.,CONTEXT OF CARE, PROVIDER CHARACTERISTICS, AND

QUALITY OF CARE IN THE IHSS PROGRAM: IMPLICATIONS

FOR PROVIDER STANDARDS (Apr. 17, 1995); A.E.BENJAMIN ET AL., U.C.L.A., COMPARING CLIENT-

DIRECTED AND AGENCY MODELS FOR PROVIDING

DISABILITY-RELATED SUPPORTIVE SERVICES AT HOME(Sept. 1998); Pamela Doty et al., Consumer-Directed Models of Personal Care: Lesson fromMedicaid, 74 MILBANK MEMORIAL FUND 377-409(1996); Nancy N. Eustis & L.R. Fischer, CommonNeeds, Different Solutions? Younger and OlderHomecare Clients, 16 GENERATIONS 17-22 (1992);L.L. GLICKMAN ET AL., UNIV. OF MASS., SELF-DIREC-

TION IN HOME CARE FOR OLDER PEOPLE (1994).

13. Sharon Tennestedt & Bei-Hung Chang, TheRelative Contribution of Ethnicity versus Socioeco-nomic Status in Explaining Differences in Disabilityand Receipt of Informal Care, 53B(2) J. GERONTOL-

OGY SOC. SC. S61-70 (1998); Steven P. Wallace etal., The Persistence of Race and Ethnicity, 53B(2)

I Elder's Advisor

J. GERONTOLOGY SOC. Sci. S104-112 (1998); S.Rimer, Blacks Carry Load of Care for TheirElderly, N.Y. TIMES, Mar. 15, 1998, at 1.

14. Mark Sciegaj & C.K. Kyriacou, Study ExaminesRacial and Ethnic Differences in Preferences forConsumer Direction, 4(4) CONSUMER CHOICE NEWS

5 (Spring 2000).

15. Recent welfare reform legislation enacted close tothe time the survey was conducted may haveincreased fear within the Hispanic population, thuscreating reluctance to respond.

16. We anticipated a language problem in Florida withmany English-speaking workers and Spanish-speaking clients, or the other way around, sohaving only 9 percent of respondents state they hadlanguage difficulties was unexpected.

17. BARNES & SUTHERLAND, supra note 12; Doty et al.,supra note 12; Eustis & Fischer, supra note 12;GLICKMAN ET AL., supra note 12.

18. Younger consumers were also significantly morelikely than older consumers to believe the first threecharacteristics mentioned above were important.

19. Pamela Doty, U.S. D.H.H.S., internal briefing paperaddressing possible fraud and abuse issues in thecash option (1997) (unpublished).

20. Id.

21. BARNES & SUTHERLAND, supra note 12; BENJAMIN ET

AL., supra note 12.

22. Doty et al., supra note 12.