Multivariate Analysis of Patterns of Informal and Formal Caregiving among Privately Insured and Non-Privately Insured Disabled Elders Living in the Community. D. Consumer Satisfaction


The average age of individuals who purchase private long-term care insurance is 68 years. Yet, the average age of claimants is closer to age 80. This means that most LTC insurance purchasers can expect to hold their policy for 10 or more years before they might actually need to access benefits. During this premium payment period, the primary "benefit" of a policy is psychic in nature: individuals derive utility from knowing that the certain premium they pay reduces the uncertainty and ultimate financial risk they might face in the absence of a policy. Consumer satisfaction during this phase is a small matter. Once benefits commence, however, consumer satisfaction with the policy is a more relevant issue.

We asked claimants a series of questions related to general satisfaction with their insurance policy as well as with claims management practices. The focus of this section is on gaining a better understanding of the factors underlying individuals' general satisfaction with their policy. As shown in Figure 4, the vast majority of claimants is indeed satisfied with their policy. In fact, only about one in seven indicated that they were dissatisfied or very dissatisfied with their policy.

FIGURE 4: Degree of Satisfaction with LTC Insurance Policy
Bar Chart: Very Satisfied (60%); Satisfied (26%); Dissatisfied (10%); Very Dissatisfied (4%).
SOURCE: 1999 National Claimant Study (n= 697 claimants).

The general concept of satisfaction and an individual's rating of their degree of satisfaction is necessarily subjective. Still, one might expect that certain aspects of a policy or an insurer's interaction with a claimant would influence a claimant's general sense of satisfaction. For example, if a claimant reports having undermet needs even though they are receiving insurance benefits, one might expect that they would be somewhat less satisfied with their policy. Or perhaps individuals with policies that have low daily benefit amounts are unsatisfied. Table 8 shows the relationship between three classes of variables and policyholder satisfaction.

  TABLE 8. The Characteristics Associated with Satisfaction Levels  
Characteristic Satisfied
  with Policy  
  with Policy  
Demographic Characteristics
Average Age (years) 79* 80
Gender (percent female) 68% 64%
Income $37,000 $40,000
Number of ADL dependencies 3.2* 3.6
Presence of Informal Caregiver 77% 75%
Presence of Undermet ADL Need 23% 28%
Insurance Policy Characteristics
Policy type: Disability policy 10%* 5%
Daily Benefit Amount $79 $81
Duration of Coverage 4.6 years* 3.9 years
Claims Administration and Claimant Experience
Had no difficulty filing policy 77%*** 28%
Had difficulty understanding the policy 19%*** 62%
Had a disagreement with the company 12%*** 62%
Coverage is adequate to meet most needs   65%*** 30%
Chose enough coverage 75%*** 51%
ANOVA F-statistic significance level: *.10; **.05; *** .01.

As shown, nine of the 14 variables tested, were significantly associated with satisfaction.20 Older individuals and those with more ADL limitations are more likely to be dissatisfied with their policy than are younger, less disabled claimants. With respect to policy characteristics, while daily benefit amount is not related to overall satisfaction, both the policy type and the duration of coverage are. Having a disability policy design and longer coverage durations are positively associated with satisfaction.

A claimant's interaction with the insurance company is very important to their sense of satisfaction with the insurance policy itself. Individuals who have not had difficulty filing their policy, have understood what their policy covers, and have not had any disagreements with the insurer about covered services also tend to be more satisfied with their policy. If the policy enables claimants to meet their care needs, and if they believe that the coverage they chose at initial purchase was adequate, they are also more likely to be satisfied with their policy.

We conducted a factor analysis in order to understand better the underlying "dimensionality" of the concept of satisfaction. In essence we wished to identify a small set of factors that could be used to represent relationships among sets of many interrelated variables. The basic assumption of factor analysis is that observed correlations between variables result from their sharing certain factors that cannot actually be observed but can be used to explain complex phenomena like, for example, satisfaction.

Table 9 displays the component score coefficient matrix for the dependent variable satisfaction. Using Principal Components Analysis, five factors were extracted, each of which had an Eigenvalue greater than 1.0. The cumulative variance in the dependent variable accounted for by these factors is 62%.21

An analysis of the component score coefficient matrix highlights five distinct dimensions related to satisfaction. An analysis of the factor loadings for each suggests that these dimensions can be broadly thought of as:

  1. Disability and Service Use
  2. Interaction with Insurer
  3. Policy Characteristics
  4. Payment Type
  5. Impact on Living Arrangement

The variables with the largest factor loadings in the first factor include those related to disability status -- ADLs and IADLs -- and to the provision of care -- the total number of hours of care and the total number of caregivers. Taken together, this dimension of quality accounts for 20% of the variance in the dependent variable. It can be interpreted to mean that one's disability status, and the amount of care received to compensate for disabilities, are the most important dimensions of claimants' sense of satisfaction with their policy.

TABLE 9. Component Score Coefficient Matrix for Individuals Satisfied with Their Long-Term Care Insurance Policy
Variable   Factor 1     Factor 2     Factor 3     Factor 4     Factor 5  
Number of ADL Limitations .351 -.006 -.041 .065 .061
Number of IADL Limitations .315 .018 .022 .125 .127
Number of ADL Helpers .258 -.080 -.131 -.148 .082
Total Hours of Care .330 -.060 .039 -.055 -.097
Had Disagreement with Insurance Company -.006 .386 .060 .254 .106
Had Trouble Understanding Policy Coverage .023 .451 -.039 -.069 .148
Was Easy to File the Policy -.015 -.442 -.056 .091 .161
Home Care Daily Benefit Greater than $40 .062 .012 .537 -.302 -.204
Home Care Duration Greater than 2 Years .033 .003 .547 .259 -.353
Disability Policy Design .023 -.027 -.130 .754 -.274
Policy is Meeting Care Needs -.087 -.251 .236 .107 .256
Having the Policy Allows Individual to Stay in Community   .003 .029 .297 .213 .772
Percent of Variance 20% 13% 11% 9% 8%
NOTE: Cumulative percent of variance captured by factors is 62%; Extraction Method is Principal Components; Factors for Eigenvalues greater than 1; KMO measure of sampling adequacy is 69%.

The next most important dimension of satisfaction relates to the interaction between the claimant and the insurance company. Whether or not an individual has had a disagreement with the insurance company, understands their policy, or found the claims filing process to be easy are all related to overall sense of satisfaction. This factor accounts for an additional 13% of the variance. It is also interesting to note that this dimension is more important than the policy parameters themselves. That is, how a claimant feels about their policy has less to do with the policy features themselves and more to do with the way that they are treated at claim time. This is evidenced by the fact that the third and fourth factors account for slightly less variance than the "interaction with insurer" factor; the variables with the largest unique factor loadings include policy duration, daily benefit amount and having a policy that pays cash benefits, i.e. the disability policy design. Finally, the fifth dimension of satisfaction has to do with individuals' perceptions of what the policy is enabling them to do vis-à-vis living arrangement. Individuals who believe that their policy is enabling them to avoid having to seek institutional alternatives by meeting their care needs are also more likely to be satisfied with their policy.


Although policy design is important to claimant satisfaction, it is not the most important variable. The interaction between the insurance company and the claimant at claim time and how a claim is serviced are critical to the claimant's overall evaluation of the policy. If the administration of claims is handled well and the claimant understands their coverage -- something that may be the responsibility of the agents selling the policy on behalf of the insurance company -- then there is a much greater likelihood of he/she being satisfied with the policy in general. This is true even when account is taken of the disability status of the claimant and the presence of undermet need. The "servicing" of the claim can be as important as the financial flexibility provided by the policy itself as a factor in the claimants' feeling about the policy.

View full report


"multanal.pdf" (pdf, 597.33Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®