A Survey of Employers Offering Group Long-Term Care Insurance to Their Employees

U.S. Department of Health and Human Services

A Survey of Employers Offering Group Long-Term Care Insurance to Their Employees

Final Report

Steven Lutzky, John Corea, and Lisa Alecxih

The Lewin Group

May 31, 2000

---

This report was prepared under contract #HHS-100-97-0011 between the U.S. Department of Health and Human Services (HHS), Office of Disability, Aging and Long-Term Care Policy (DALTCP) and The Lewin Group. For additional information about the study, you may visit the DALTCP home page at http://aspe.hhs.gov/daltcp/home.htm or contact the ASPE Project Officer, Hunter McKay, at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, SW, Washington, DC 20201. His e-mail address is: Hunter.McKay@osaspe.dhhs.gov.

---

TABLE OF CONTENTS

EXECUTIVE SUMMARY

I. INTRODUCTION

II. DATA AND METHODS

A. Survey Development

B. Defining the Respondent Universe

C. Data Collection Process

D. Survey Methods and Maximizing Response Rates

E. Sample Characteristics

III. FINDINGS

A. Current Practices and Products

1. Policy Provisions
2. Plan Complexity
3. Eligibility and Price
4. Motivation for Offering LTC Insurance
5. Implementation of the Offering
6. Employers' Satisfaction with the Offering

B. Factors Related to Participation Rates

IV. DISCUSSION

A. Potential for Growth in the Private LTC Insurance Market

B. Potential Limitations of Employer-Based Private Coverage

C. Importance of Education

V. FUTURE RESEARCH

REFERENCES

NOTES

 

EXECUTIVE SUMMARY

Background

Private long-term care insurance provides one of the few available mechanisms for individuals to protect themselves against the catastrophic costs of long-term care. Long-term care (LTC) insurance sold through employers has advantages over policies available through the individual market that may increase sales, including:

• Lower premiums due to:
  • More effective marketing to individuals at younger ages, when premiums are lower;
  • Administrative and agent commission savings; and

  • The potential for employers to use bargaining power to reduce insurers' profit percentages;

• Increased access to coverage, due to less stringent screening criteria or absence of screening (i.e., guarantee issue); and

• Increased ease and comfort of purchase, due to the broker role of the employer and fewer coverage decisions required.

Because a majority of adults already receive some insurance benefits through the work place, the employer group LTC insurance market holds impressive potential for protecting millions more Americans from catastrophic long-term care costs. Making group LTC insurance available to federal employees could be a major stimulus for increasing the size of this market.

Purpose

The Office of the Assistant Secretary for Planning and Evaluation (ASPE) within the Department of Health and Human Services sponsored this study to improve current understanding of employer group LTC insurance. The primary purpose of this research is to inform employers' (including the Federal Government's) decision about how to structure and market a LTC insurance offering to employees.

This final report provides information about current and best practices in the employer group LTC insurance market that can inform federal policy makers and employers in deciding how to construct a group LTC insurance offering. Current practices were collected from a random and select sample of employers and best practices were compiled from the select sample (similar to government agencies, innovative, or successful in enrollment) of employers. The random sample survey generated a response rate of 58 percent, which is in line with response rates achieved with other employer surveys. The total sample included 93 employers.

Principal Findings

Current Practices

The survey data indicate that, among employers offering the product, the employer market provides greater access to coverage than in the individual market. A majority of employers offered less restrictive underwriting or even guarantee issue policies (i.e., did not require health information) during initial offerings to active full-time employees. In addition, a majority offered coverage to at least one group in addition to full-time active employees (i.e., parents/in-laws, spouses, and retired employees), potentially extending the benefit well beyond the size of the employee population.

Most of the employers surveyed usually limited the number of benefit choices. For example, a majority offered two to four benefit amount options and a set package rather than allowing the employee to select every option separately. Nearly all employers used a single LTC insurer. These practices simplify the multiplicity of choices generally related to purchasing this product.

Despite the limited choices of benefits, data from a random sample of employers suggest that the benefit features of employer group plans generally resemble the most common individually purchased policies:

• Employers offered a full range of coverage for most recognized LTC services, with most including a 60 or 90 day elimination period (the deductible period between qualification for benefits and the first day benefits can be received).

• All surveyed employers offered some type of inflation protection, with about half offering it immediately at a higher initial premium and half offering only the option to upgrade benefit levels in the future (future purchase option). A recommended area for further investigation is whether the higher premiums associated with inflation protection deter employees from upgrading their policies.

• Just under half of the surveyed employers offered some type of "non-forfeiture" benefit that would provide the purchaser some level of benefits if he or she lapses (i.e., if coverage is stopped because the individual stops paying premiums). The most common non-forfeiture benefit provided was the "reduced paid up benefit," which provides a reduced benefit amount over the same benefit period as defined in the policy. Reduced paid up benefits are much less likely to offer real protection against the costs associated with needing long-term care than other more expensive non-forfeiture provisions, most notably "shortened benefit period."

• Almost as common as the non-forfeiture benefit was "return of premium at death," which was offered by 44 percent of employers. This feature allows a portion of premiums paid to be returned to the insured's estate upon his or her death.

• The vast majority of plans (all but two) required the employee to pay the entire premium.

• Employee purchase rates varied considerably by employer. While over 40 percent of employers experienced participation rates below 2 percent, 20 percent achieved participation of 10 percent or higher.

Best Practices

• More simple policy offerings may be related to higher participation. Companies that offered fewer choices were more likely to have higher enrollment rates.

• The survey also revealed that companies considered educating employees about the benefit very important, but also challenging. When asked what they would have improved about their LTC insurance offering, employers (particularly those with low participation rates) most cited education and communication during the initial offering. Insurers corroborated this, and added that employers appeared to have better enrollment rates when senior management was actively involved.

Limitations of Employer-Based Coverage

There are potential limitations of employer-based private coverage, including: (1) insufficient numbers of individuals having any LTC protection when they need it because of low enrollment rates and possibly high lapses (2) insufficient protection for those who do retain their coverage until they need benefits because of a failure to purchase inflation protection (3) a lack of value for individuals who pay substantial premiums and then lapse because of a lack of interest in non-forfeiture benefits, and (4) limitations in policies ability to adapt to changes in the LTC delivery system.

Future Research Recommendations

This study raises questions that could be addressed in future research on employer-based LTC insurance. Studies could examine the types of insurance employees are buying and the rate of lapses, the role employers are playing in managing policies, the potential role of employer coalitions, employee demand for LTC insurance, and the type of employee typically purchasing LTC insurance.

I. INTRODUCTION

Private long-term care insurance (LTC insurance) provides one of the few available mechanisms for individuals to protect themselves against the catastrophic costs of long-term care. Although private LTC insurance finances only 1 percent of total long-term care spending¹, the market has grown at an average annual rate of 21 percent each year between 1987 and 1997².

Between January 1997 and June 1998, the employer market accounted for 15 percent of overall private LTC insurance sales. By mid-1998, 800,000 policies had been sold in a market consisting of approximately 2,100 employers offering policies (HIAA, 2000).³ In 2000, the Life Insurance and Marketing Research Association (LIMRA), a member-owned organization dedicated to meeting the needs of companies offering benefits, found that employer sponsored group LTC insurance sales (as measured by new participants) were up 126 percent in 1999, and in-force sales were up 24 percent.⁴

An overview of the long-term care insurance market completed for the Health Care Financing Administration (HCFA) in 1996 found that regulators, consumer groups, and industry representatives generally agreed that expanding the employer-based market for private LTC insurance would put the product within the reach of substantially more people.⁵ The employer group LTC insurance market possesses a number of important advantages over the individual market for LTC insurance, including:

• Better vehicle for growth -- Because most working-age persons are employed and a large majority already receive some employee insurance benefits, including heath and life insurance,⁶ the workplace represents a possible vehicle for substantial LTC insurance sales growth. However, for the potential of the group LTC market to be realized, there must be widespread employer sponsorship of group LTC plans and significant participation levels among eligible employees in these plans.⁷ If just 10 percent of the working-age population over age 45 decided to purchase a policy, the number of long-term care insurance policies sold through employers would increase to nearly seven times current sales.⁸

• Lower premiums -- In focus group discussions, insurers estimated that group policies are generally 10 to 20 percent less expensive than comparable individual policies.⁹ Employer-group policies have three advantages that make premiums less expensive than in the individual market:

  • Employer-based plans allow LTC insurance to be marketed more effectively to individuals at younger ages. Younger purchasers face substantially lower annual premiums. At age 50, a long-term care policy costs less than half what it would if purchased at age 65.¹⁰ Although younger individuals can purchase LTC insurance policies in the individual market, they may be more likely to do so if it is offered through an employer.

  • Group policies can be sold and processed en masse, which results in administrative and agent commission savings.

  • Employers can use their bargaining power to obtain concessions from the insurer, such as lower profit percentages that will lower premiums, while still building adequate reserves.

• Increased access to coverage -- In the individual market, insurance companies screen all potential insureds applying for coverage and deny coverage to those considered high risk. In the employer group market, many plans use less stringent screening criteria, or do not require any screening (i.e., guarantee issue).

• Increased ease and comfort in purchasing LTC insurance -- In the employer group market, the employer acts as an intermediary between the insurer and the potential insured. The employer can simplify the coverage decisions their employees face by simplifying the benefit structure. Employees may also be more likely to purchase a policy if they have affinity for their employer and the employer demonstrates support for the offering.

We found relatively limited research studies specifically addressing the employer LTC insurance market. The recent emergence of employer-based LTC insurance and its small portion of LTC insurance sales may account for the lack of research. The Health Insurance Association of America (HIAA) periodically surveys the major LTC insurers and publishes information about the number of employers offering LTC insurance and size of those employers on a periodic basis.¹¹ Unfortunately, these reports do not provide any further description of what employers are offering.

A report published by William M. Mercer, Inc. in 1997 included information about offerings by 66 employers. However, while this study provides an introduction to the employer LTC insurance market, the information contained in the four-page report does not contain enough detail to guide decisions about whether and how to implement an offering. The International Foundation of Employee Benefit Plans (IFEBP) surveyed its members about characteristics of employer-sponsored LTC insurance. This report compares our results to that of the Mercer and IFEBP studies where appropriate. In 1997, The Segal Company prepared a report on LTC insurance for state employees¹². This report only addressed whether a state offered or was considering offering LTC insurance to its employees and, if so, through which insurer, and whether the state paid for a portion of the premiums. Like the Mercer report, this report provided insufficient information to aid design decisions.

LIMRA released a report in 2000 of sales and in-force data for 26 U.S. organizations that insure or administer employer-sponsored group LTC plans and group association plans that include master contracts between insurers and (1) employers or trusts set up by employers, and; (2) associations or trusts set up by associations. The report did not address the components of offerings with high participation rates. The Employee Benefit Research Institute (EBRI) released an issue brief in April 2000 which details the best practices for increasing employer sponsorship of LTC insurance. Reviewing results from a number of prior surveys, they cited lack of awareness and low priority as primary barriers in employer's decision to offer a LTC insurance benefit. A forthcoming issue brief from EBRI will address the question of employee participation. Some insurers, most notably John Hancock, maintain detailed databases and have conducted more detailed analyses, however, these data are highly proprietary.

The Office of the Assistant Secretary for Planning and Evaluation (ASPE) within the Department of Health and Human Services sponsored this study to improve the current understanding of employer group LTC insurance and to identify best practices. The primary purpose of this research is to inform employers' (including the Federal Government's) decision about how to structure and market a LTC insurance offering to employees.

II. DATA AND METHODS

This report contains data collected primarily through a written survey mailed to large and medium sized employers identified as offering LTC insurance to their employees. This section describes the development and field operations of the survey and presents some general descriptive information about the study samples.

A. Survey Development

This study had the dual purpose of gathering information about: (1) current employer LTC insurance offerings; and (2) best practices, especially those that might be of interest to the federal government as it designed a LTC insurance offering. We therefore included survey questions on a broad range of topics related to the LTC insurance offering, including: (1) plan features; (2) the decision to offer LTC insurance; (3) the relationship between LTC insurance and other employee benefits; (4) financing and arrangements with the insurer; (5) education of and marketing to employees; (6) enrollment; (7) evaluation of the LTC insurance offering; and (8) employer demographics.

The survey had two components: a written survey and a brief, optional follow-up telephone call. The written survey took an average of 50 minutes to complete. The length of time required for the follow-up call varied considerably depending on the level of innovation of the employer's LTC insurance plan. Relatively standard plans took about fifteen minutes while highly innovative plans lasted up to one hour. Follow-up calls were not conducted for employers with standard offerings and relatively complete surveys.

We conducted a pilot test of the employer survey instrument on nine employers prior to fielding the survey for the random and select samples. Based on the results of this pilot testing, we modified the survey instrument to substantially reduce the level of response burden.

B. Defining the Respondent Universe

Because the study had the dual purpose of gathering information about current offerings and best practices, we drew two samples. To obtain a representative snapshot of current offerings, we drew a random sample of medium and large employers with LTC insurance offerings.

We then augmented the random sample in order to capture a sufficient number of employers exhibiting innovative offerings or other "best practices." ASPE was particularly interested in employers that exhibited innovative plan design or marketing, or had successful enrollment efforts. In addition, public employers and very large employers were expected to have experiences that would be particularly instructive to the federal government as an employer. Therefore, we surveyed an additional sample of these "relevant" employers not selected for the random sample. For the analysis, we would combine these additional records with records of employers from the random sample that met the same criteria.

The common sampling universe from which we drew both samples included all medium and large employers that make a group LTC insurance benefit available to their employees. We defined medium employers as those with between 501 and 5,000 employees, and large employers as those with more than 5,000 employees. We restricted the universe to medium and large employers for two reasons. First, ASPE was chiefly interested in identifying best practices (e.g., leading edge plan features, innovative marketing strategies) in the employer group LTC insurance market. Small businesses are generally less likely to demonstrate innovative LTC insurance offerings because they do not have the volume to demand concessions and innovations from insurers. Second, the experience of larger employers is more likely to provide information relevant to other large public employers considering offering LTC insurance, such as the Federal Government.

We estimated the universe for the random sample of medium and large employers to consist of approximately 825 employers in 1998. We based this number on the estimated number of all firms offering LTC insurance in 1996 (1,500), projected to 1998 using the annualized growth rate in all employer offerings from 1994 to 1996 (20.8 percent annually), multiplied by the proportion of employers offering LTC insurance that were medium or large in the latest year observed (i.e., 38 percent in 1995).¹³ This estimate assumed equal growth among small, medium, and large employers, and relatively steady overall growth from 1994 to 1998.

The universe for the additional sample of relevant employers consists of all medium or large organizations not selected for the random sample but which we identified as being able to provide best practices or other information particularly relevant to a large public employer. These employers fell into at least one of the following groups:

• very large employers (at least 25,000 employees);

• state or local government sponsors;

• employers exhibiting other distinguishing characteristics in either the way it approaches the presentation to employees and other prospective purchasers (e.g., spouses and other relatives) or in how it designed the product; or

• employers having unusual LTC insurance enrollment success, defined as employee participation of at least 10 percent.

We constructed a "select" sample of employers by combining these additional records with employers from the random sample that met the criteria for the select sample listed above. The report presents data separately for the select and random samples where appropriate.

C. Data Collection Process

The process of collecting data on the universe of employers described above included the following steps:

• Gathering information on employers offering LTC insurance from insurers;
• Building the sampling frame;
• Selecting the random sample;
• Selecting additional employers for the "select" sample; and
• Collecting data using a mailed survey instrument and follow-up telephone interviews.

We used major LTC insurers to assist us in identifying employers to survey. These insurers had incentive to cooperate with this survey effort, because they saw the study as an activity that could promote the development of the employer group LTC insurance market. We gathered information for the sampling frame from these insurer contacts, combined with information about additional employers suggested by the HIAA. These insurers, listed below, accounted for over 90 percent of the employer group LTC insurance market¹⁴:

• Aetna
• John Hancock
• Mutual of Omaha
• TIAA-CREF
• CNA
• Prudential Health Group
• Metropolitan Life Long-Term Care Group
• Fortis Long-Term Care
• UNUM

From these insurers, we requested a complete list of all their group LTC insurance clients that were medium or large employers/organizations (i.e., 500 or more employees). We randomly selected 125 employers for the first sample. We obtained the name of a contact person at each of these employers from the employer's insurer.

We then asked each insurer to identify as many employers as possible that met the criteria for the additional sample. In addition, we attempted to collect data on as many plans as possible offered by state employers, using information on state LTC insurance offerings from the 1997 Segal Company state survey. While we intended to survey up to 35 employers for the additional sample, we only identified and contacted thirteen such employers.

To create the "select sample," we combined these thirteen additional employers with eleven employers from the random sample and seven employers from the pilot sample that also met the select sample criteria.

D. Survey Methods and Maximizing Response Rates

We employed a time-intensive strategy to maximize the response rate. To alleviate employer concern about participating, the insurer made the first contact with the employer, either through a letter or a telephone call, to endorse the study.

We followed up by calling the employer contact to secure their participation. Provided they agreed to participate, we sent them the survey instrument, introduced by a cover letter from ASPE's Deputy Assistant Secretary for Disability, Aging and Long-Term Care Policy. Approximately one week after sending the survey, we called the respondent to ensure that he/she received it and to answer any questions. We called multiple times to increase participation.

In addition to these explicit efforts to increase response rates, we made efforts to reduce response burden. We minimized response burden by using survey questions only when the information could not be obtained elsewhere, including:

• The survey asked the employer to provide us with existing materials that describe key aspects of the employer's LTC insurance program, including plan features and premium schedules rather than having the employer describe their plan on the survey. This information is typically available in the enrollment kit assembled for the employer by the LTC insurer.

• We sought most information about specific financial arrangements from the employers' insurers rather than employers themselves. We found during our pilot interviews that insurers are able to provide this information easily from administrative records and that they were willing to cooperate.

• We used follow-up telephone interviews to fill in gaps in our understanding of the offering and to obtain more detail about innovative practices.

• For employers that either failed to respond after several months of follow-up or that refused to complete the full survey, we developed a one-page fax survey that contained several key survey questions. The one page "fast fax" included the following survey items: year of initial offering, use of multiple LTC insurers, LTC insurance premium subsidization, premium contributions for any other voluntary benefits, eligibility and underwriting practices, LTC insurance participation rates, employer size, and employer industry.

From the random sample of 125 employers, we collected full survey responses from 43 employers, for a 34 percent response rate. We collected at least partial responses (i.e., including shortened surveys and enrollment kits) from 71 employers, for a response rate of 58 percent. These response rates are generally in line with rates achieved by other employer surveys. One important example is the Employee Benefits Survey, a major survey of employers conducted regularly by the Bureau of Labor Statistics, which achieved a response rate of 53 percent in 1997.¹⁵ The International Foundation of Employee Benefit Plans (IFEBP) provides a recent example of an employer survey devoted entirely to LTC insurance issues. The survey of IFEBP members achieved a 32 percent response rate.

Exhibit 2.1 presents our final sample counts presented for the random sample, the select sample and pilot test group. While all but two of the select sample employers returned the original survey, 28 of the 71 random sample employers provided only one-page fast-fax responses or their LTC insurance enrollment kits. Exhibit 2.2 displays response rates by insurer.

Exhibit 2.3 shows that the individual completing the survey was typically someone from the benefits or human resources department of the organization. Some employers had benefits consultants respond to the survey, while an equal number had members of senior management fill out the survey. In some cases, the principal contact recruited other individuals in the organization to fill out sections of the survey they could not complete themselves (for example, if the principal contact did not have knowledge of an initial offering that occurred several years prior).

EXHIBIT 2.1. Type of Survey Response, by Sample
Sample	Full Survey	"Fast Fax," No Full Survey	Enrollment Kit Only	Total Response
Random Sample	43	17	11	71
Select Sample	29	1	1	31
TOTAL SAMPLE	63	18	12	93
Sample types do not sum to total sample because some employers belong to more than one sample. Specifically, 11 employers are members of both the random and select samples. In addition, the total sample includes two employers fro the pilot sample that were in neither the random nor the select group.

EXHIBIT 2.2. Response Rates in the Random Sample by Insurer
By Insurer	Sample	Respondents1	Any Response Rate	Full Survey Response Rate
UNUM	47	32	68%	40%
Hancock	13	12	92%	31%
CNA	33	9	27%	24%
Metropolitan Life	14	10	71%	36%
Aetna	13	6	46%	15%
Mutual of Omaha	2	2	100%	100%
TIAA2	--	--	--	--
Prudential2	--	--	--	--
Fortis2	--	--	--	--
TOTAL	122	71	58%	32%
Indicates any response from employer. These insurers were included in the sampling frame based on information from HIAA about the number of employer group policies insurers had in orce in 1997. However, when we contacted these insurers, they indicated that their policies were actually individual policies, not true group policies.

EXHIBIT 2.3. Characteristics of Individuals Completing the Survey

E. Sample Characteristics

Exhibit 2.4, Exhibit 2.5, and Exhibit 2.6 display some basic characteristics of the sample on which we report. Exhibit 2.4 shows the distribution of sample respondents by insurer for both the random and select samples. The distribution by insurer in the random sample differs somewhat from the distribution in the sampling frame (and also from the true distribution in the employer LTC insurance market), because some insurers were more successful than others in persuading their employer clients to participate in the survey. Thus, UNUM is slightly over-represented in the random sample, while CNA is under-represented.¹⁶

EXHIBIT 2.4. Survey Respondents by Insurer and Sample Type1
By Insurer	All		Random		Select
UNUM	35	38%	30	42%	8	26%
Hancock	15	16%	12	17%	4	13%
CNA	13	14%	10	14%	5	16%
Metropolitan Life	11	12%	10	14%	2	6%
Aetna	10	11%	6	8%	6	19%
Mutual of Omaha	4	4%	1	1%	3	10%
TIAA	2	2%	1	1%	0	0%
MedAmerica2	1	1%	0	0%	1	3%
New York Life3	1	1%	1	1%	1	3%
U.S. Care4	1	1%	0	0%	1	3%
TOTAL	93	100%	71	100%	31	100%
Survey respondents include employers providing at least one of the following: a completed full survey, an enrollment kit, or a completed "fast fax" shortened survey. Insurer for one state employer identified through the Segal Company survey, and not through our initial list of cooperating insurers. Employer initially identified through a cooperating insurer, but had switched to New York Life by the time the survey was fielded. The only employer in the sample that self-funded for LTC insurance used U.S. Care as a third party administrator. This employer was identified through discussions with industry representatives, and provided only written enrollment materials.

Exhibit 2.5 compares the size distribution (i.e., by number of employees) of employers in the random sample with the size distribution of employers in the Mercer survey. Our random sample had a notably higher concentration of employers with 1,000 to 5,000 employees, while the Mercer sample had a majority with 10,000 or more employees.

EXHIBIT 2.5. Percent of Random Sampled Employers and Mercer Study Employers by Number of Employees
Employees	Random Sample	Mercer Sample
Fewer than 1,000	16%	6%
1,000 to 4,999	46%	20%
5,000 to 9,999	19%	20%
10,000 or more employees	19%	54%

Exhibit 2.6 displays the industries of the random employer sample. Sixty-one (61) percent were services, 28 percent manufacturing, 9 percent government, and 2 percent unknown. In the Mercer study, 48 percent of the survey respondents were manufacturing, 49 percent were services, and 3 percent were government.

EXHIBIT 2.6. Percent of Random Employers by Industry

III. FINDINGS

This section focuses on the characteristics of employer group LTC insurance offerings. The first part provides a snapshot of the market, describing the policy provisions offered, as well as how plans are marketed and administered. It also provides detailed information about the factors employers considered when selecting an insurer, constructing an offering, and rolling out a benefit, as well as employers' views on what they would change if they had it to do again.

The second part carries the discussion of the proper design of a LTC insurance offering further, with a discussion of factors that appear to be related to increased enrollment.

The following discussion draws upon survey data from both the random and select samples of employers. The current practices and products results are based on the random sample unless the select sample is specifically enumerated. The number of available responses varies by survey question, often significantly, because many employers provided answers to only certain sections of the original survey instrument.

A. Current Practices and Products

Employers offered a variety of policy provisions. However, employers tended to limit the number of options presented to employees.

1. Policy Provisions

a. Service versus disability model policies

Employers offered two models of LTC insurance policies: a service model and a disability model. The service model policy is the traditional plan that provides payments for covered services. The disability model provides a per-diem payment of a set dollar amount to be used at a beneficiary's discretion.

Exhibit 3.1 shows the proportion of employers using disability model versus service model LTC insurance plans. Among employers in the total sample providing information on model type, 43 percent used disability model LTC insurance plans. Interviews with representatives from the major insurers at the outset of the survey effort indicated that the vast majority of employers offered service model plans. The high response rate from employers offering UNUM plans in the sample (see methodology section) most likely explains the higher-than-expected proportion of disability model plans in the sample.

EXHIBIT 3.1. Insurance Model Used, by Sample
LTC Insurance Model	All		Random		Select
Disability Model	27	43%	19	44%	14	50%
Service Model	36	57%	24	56%	14	50%
TOTAL	63	100%	43	100%	28	100%

b. Services Covered

i. Direct Services

Skilled nursing facility coverage was standard in all employer LTC insurance plans (Exhibit 3.2). Home health care coverage and adult day care were offered by almost all employers surveyed, and were included in most plans. However, some employers offered nursing home only policies. The majority of employers offered plans that covered respite care, homemaker/chore services, assisted living, and reimbursement of a family member.

EXHIBIT 3.2. Services Covered by Surveyed Employers' LTC Insurance Plans

ii. Care Management

Most care management included in LTC insurance plans offered by employers consisted of mechanisms to assist beneficiaries in finding needed services. Eleven employers' plans offered telephone consultations with a care manager who would assist in accessing and coordinating services (Exhibit 3.3). Nearly all also offered face-to-face consultations with a care manager. Typically, these services are provided by care managers who work for (or contract with) the LTC insurance carrier as opposed to allowing the individual to identify and hire his or her own care manager. Reimbursement for an individual care manager was much less common.

Care management appeared to serve a gate keeping function in a subset of plans. About 40 percent of plans required the insurer or an insurer approved care management organization to pre-approve services and/or providers.

EXHIBIT 3.3. Percent of Surveyed Employers Offering Care Management by Type, Random Sample

c. Coverage levels

The amount of coverage varied across employers sampled, and most employers offered more than one level of benefits. Exhibit 3.4 summarizes the benefit levels offered by the surveyed employers.

Most employers defined benefits in terms of a daily benefit amount (DBA), or daily maximum that capped the amount covered per day. Among those with DBAs, $100 was the most common DBA for nursing home care. Some employers offered DBAs for nursing facility care as low as $40 or as high as $300.

Most employers set the home care DBA at between 60 percent and 100 percent of the nursing home DBA. Home care DBAs of $75 to $120 were common, but some plans offered home care benefits as low as $20 per day or as high as $200 per day. Little difference was seen between random and select groups for DBAs.

Almost one-third of employers defined nursing home and home health benefits in terms of a monthly benefit amount (MBA) rather than a DBA, and most offered a wide range of monthly benefits (i.e., from $1,000 to $6,000 per month for nursing home care). Most employers with MBAs set home care benefits at half the nursing home MBA.

All plans also set lifetime benefit limits. While usually set at a dollar amount (dollar-defined), about one-third used a time-defined benefit or benefit period, expressed in years. Dollar-defined benefit limits varied considerably by employer, ranging from as low as $24,000 to as high as $1 million. Plans that set lifetime limits in terms of years of care provided used limits that ranged from two to six years.

All surveyed plans used a pool-of-money approach for their lifetime limits, rather than having separate lifetime limits for nursing home care and home care. This pool-of-money approach covers claimants for whichever mix of covered home care or nursing facility services they wish, up to the lifetime maximum.

EXHIBIT 3.4. Amount of LTC Insurance Coverage Offered by Surveyed Employers
Daily/Monthly Benefit Amounts	Random Sample	Select Sample
Total Number of Observations	58	30
Nursing Home
Number of Employers Offering Daily Amounts	34	24
Most Common Daily Benefit Amount	$100	$100
Range of Daily Benefit Amounts (lowest and highest)	$40-$300	$40-$250
Number of Employers Offering Monthly Amounts	18	5
Most Common Monthly Benefit Amount	$1,000	$1,000
Range of Monthly Benefit Amounts (lowest and highest)	$1,000- $6,000	$1,000- $6,000
Home Health Care
Number of Employers Offering Daily Amounts	34	22
Most Common Daily Benefit Amount	$75/$120	$60
Range of Daily Benefit Amounts (lowest and highest)	$20-$200	$20-$200
Number of Employers Offering Monthly Amounts	18	7
Most Common Monthly Benefit Amount	$500	$500
Range of Monthly Benefit Amounts (lowest and highest)	$500- $3,000	$500- $3,000
Lifetime Maximums
Number of Employers Offering Dollar-Defined Lifetime Benefit Amounts	39	25
Most Common Dollar Maximum	$200,000	$36,000/ $109,500
Range of Dollar Maximums (lowest and highest)	$24,000- $1,000,000	$36,000- $1,000,000
Number of Employers Offering Time-Defined Lifetime Benefit Amounts	18	8
Most Common Time-Defined Limit	3 years	5 years
Range of Time-Defined Limit (lowest and highest)	2-6 years	2-6 years
Note: The number of employers can sum to more than the total because some employers offer more than one benefit package.

Exhibit 3.5 presents the nursing home DBAs displays the range of each employer's DBAs. This chart includes plans with MBAs, but displays them converted to DBAs for comparison purposes. It is important to note that a monthly benefit provides the beneficiary with more flexibility in daily spending than a daily limit allows. Spending on any given day may be above the average daily amount, as long as the monthly amount is not exceeded by the month's end. Similarly, Exhibit 3.6 displays visually each offering's range of lifetime benefit limits, for offerings that use dollar amount lifetime benefits. We do not include offerings in which lifetime benefits are limited to time periods.

EXHIBIT 3.5. Range of Nursing Home1 Daily Benefit Amounts Offered by Each Surveyed Employer

Home care daily benefit amounts follow a similar pattern, but generally are set at about 60 percent of the nursing home daily benefit.

EXHIBIT 3.6. Range of Lifetime Benefit Maximums Offered by Each Surveyed Employer

Exhibit 3.7 shows that the majority of offerings included two to four choices of DBAs. However, several plans offered numerous benefit choices.

EXHIBIT 3.7. Employers by Number of Benefit Levels Choices They Offer

Note: Benefit level choices are for Nursing Home Daily/Monthly. Lifetime Maximum benefits had a similar number of choices.

d. Consumer protection features

i. Inflation Protection

Inflation protection options provide an annual or periodic adjustment of the benefit amount to adjust for increases in the cost of LTC over time. Some plans offer the purchase of inflation protection at the time of the LTC insurance purchase; benefit levels increase annually from the date of purchase while premiums remain level. This automatic built-in inflation protection typically increases DBAs and MBAs by 5 percent that may or may not be compounded annually (compound versus simple).

Future purchase options (FPO), which allows the insured to periodically purchase increased indemnity payments without medical underwriting, offer an alternative to automatic built-in inflation protection. However, if the insured selects not to upgrade their coverage a certain number of times, typically she or he become ineligible for future upgrade opportunities. The additional coverage is priced at the insured's current age (i.e., as opposed to his/her age at the time of initial LTC insurance purchase) and is guaranteed (i.e., it is not underwritten). A similar amount of coverage will cost more than if the coverage had been bought when the insured first purchased the policy. However, increases in wages might compensate for increases in premiums.

All surveyed employers offered some type of inflation protection in their LTC insurance plans (Exhibit 3.8). Just over half offered only a future purchase option, which does not add to the initial cost of the policy. About one-quarter (26 percent) of employers offered only 5 percent compound inflation protection, while an additional 9 percent offered only 5 percent simple inflation protection. The remainder offered multiple options.

EXHIBIT 3.8. Percent of Surveyed Employers by Type of Inflation Protection Offered, Random Sample

Note: No employer had both 5% simple and 5% compound inflation protection available.

ii. Non-Forfeiture Benefits

Insured persons generally do not require benefits until years after purchasing a LTC insurance policy. Therefore, they need to consistently pay premiums over a long period of time to keep the policy in force. An individual may lapse (i.e., stop paying premiums) if, for example, the individual cannot afford to continue paying premiums. To provide some benefit if the policyholder does lapse, LTC insurance policies may include non-forfeiture benefits. These benefits are typically offered as an option at an additional cost. Non-forfeiture protection can take a number of forms, including:

• Reduced paid-up -- The policy pays a reduced daily benefit amount but over the same benefit period defined in the policy (e.g., $20 instead of $100 per day in a nursing home for four years). This benefit offers only limited protection and individuals are likely to incur substantial out-of-pocket costs if they need long-term care.

• Shortened benefit period -- The policy pays the same daily benefit defined in the policy but over a shorter benefit period (e.g., $100 a day in a nursing home for only one year rather than the four years specified in the policy). This feature may offer substantial protection against shorter long-term care episodes, such as breaking a hip and entering a nursing facility for several months.

• Return of premium at lapse -- The policy returns a portion of the premiums paid in if the policyholder lapses. The amount returned to the policyholder is typically tied to the number of years premiums have been paid in (e.g., less than 5 years: no premiums returned; 5-9 years: 25 percent of premiums returned; 10-14 years: 50 percent returned; 15-20 years: 75 percent returned; 20 or more years: 100 percent returned). Typically, only premiums and not interest accrued on those premium dollars, are paid. Over time, the interest accounts for substantially more of the value the policy has accrued than the actual premium dollars. In addition, this non-forfeiture option offers no long-term care coverage in the event the individual becomes impaired.

• Benefit bank -- Payments are made to an account as premiums accrue. This account can then be used to pay claims at the maximum DBA in force at the date of lapse, until the account is exhausted.

• Extended term -- The individual is eligible for full benefits if he or she should become impaired for a period of time after premium payments cease. After this extended term expires, the individual is not eligible for any benefits. Because individuals are less likely to need benefits shortly after they let their policies lapse, this option may offer little value to consumers.

Just under one-half of the random sample offered the employees the option of some form of non-forfeiture protection. Exhibit 3.9 displays the types of non-forfeiture benefits reported by employers. The most common non-forfeiture benefit for random employers was reduced paid-up at over 40 percent.

iii. Return of Premium at Death

Almost as common as the non-forfeiture benefit was the return of premium at death feature. This feature returns a portion of the premiums, less long-term care claims paid, to the insured's estate when the insured dies. This feature only considers premiums paid and not the interest earned on those premium dollars. This form of non-forfeiture is usually built into the policy. Exhibit 3.10 indicates that about 44 percent of the random sample included return of premium at death in the policies offered.

EXHIBIT 3.9. Percent of Surveyed Employers Offering Non-Forfeiture Benefits by Type

Note: These were voluntary responses of surveyed employers. The denominators used for percentages reflect the number of missing responses.

EXHIBIT 3.10. Percent of Surveyed Employers with Return of Premium at Death

2. Plan Complexity

Our interviews with insurers suggested that they strongly encouraged employers to limit the complexity of their offerings. Insurers indicated that if the purchase decision was too complicated, employees would be less likely to reach any decision and would end up not purchasing a policy.

The most obvious way that employer-based LTC insurance limits complexity is by limiting the number of insurance companies included in the offering. The survey indicated that only three employers (of 46 responding) had selected to offer policies from more than one insurer. One of these employers used one insurer for a basic benefit and another insurer for an add-on. This employer remarked that, in retrospect, using two carriers was not a good idea because it added too much complexity to the offering.

Measuring the complexity of the offering was somewhat complicated because some companies marketed policies as packaged plans, others allowed employees to decide on individual plan features, and still others offered packages to which certain features, most notably inflation protection, could be added.

We constructed a measure of the complexity of each of the offerings by summarizing across several of the variables included in the survey. We assumed that the employee had to make a decision for each of the following optional benefits if they were offered: non-forfeiture benefits, inflation protection, and nursing home only versus coverage of both nursing facility and home care. We also tried to quantify the number of decisions individuals had to make about benefit levels. If the individual had a choice of two or three DBAs, we classified this as one decision. Four to Five DBAs were classified as two decisions. More than five DBAs were classified as three decisions. If the lifetime benefit amount differed in ways that were not directly related to the DBA chosen, this was classified as one decision. For example, if the plan offered two DBAs of $50 and $100 a day and two lifetime benefit amounts (LBA) that corresponded to four years of coverage at $50 or $100 a day, we classified this as one decision on the DBA. If, on the other hand, the plan offered either of two or four years of coverage at $50 or $100 a day, we classified this as one decision for the DBA and one decision for the LBA.

We present the summarized simplicity scores, which ranged from zero (least complex) to five (most complex), in Exhibit 3.11. All employers required at least one decision (aside from the decision to purchase) on the part of the enrollee, however, half required only one or two decisions. About one-third of employers presented employees with relatively complex offerings, requiring four or five decisions.

EXHIBIT 3.11. Random Sample Employers by Number of Decisions Required of Enrollees1

The score presented was designed to denote the level of complexity posed by the LTC insurance offering, and may not represent exactly the number of decisions faced by the enrollee (see discussion above).

3. Eligibility and Price

a. Eligibility

Employer-based policies may broaden the number of LTC insurance policyholders beyond active employees by making policies available to retired employees or relatives of active employees. Exhibit 3.12 shows that the majority of employers offer almost all LTC insurance to at least one group in addition to full-time active employees. Among the sample of employers that was randomly selected to reflect the universe of employers offering LTC insurance, over 50 percent include all four of the listed groups in the LTC insurance offering (full-time active employees, their spouses, parents/in-laws, and retired employees and their spouses). Almost 65 percent of employers from the select sample offered LTC insurance to all four groups. The data on eligibility is consistent with that of the Mercer (1997)¹⁷ study, which reported that over 90 percent of employers also covered employees, spouses, parents/in-laws and 73 percent reported covering retirees and their spouses. A majority of both the random employers also extended the benefit to part-time permanent employees (Exhibit 3.13).

EXHIBIT 3.12. Percent of Random Sample Employers Offering LTC Insurance to Eligible Groups

EXHIBIT 3.13. Percent of Random Sample Employers Offering LTC Insurance to Employees by Employee Category

b. Underwriting

An important advantage of the group LTC insurance market over the individual market is the wider access to coverage it allows. In the individual market, insurers typically determine if someone applying for a policy should not be sold a policy for health reasons through a process called underwriting. Insurance companies underwrite policies in an attempt to avoid enrolling individuals likely to use services soon after purchasing a policy. More restrictive underwriting can contribute to lower premiums, but also to reduced access to coverage, by avoiding this adverse selection.

While underwriting is standard in the individual LTC insurance market, many employer group plans are made more accessible through one of the following:

• Guarantee issue, where applicants are approved for coverage without underwriting. Employers may make a guarantee issue offering to certain groups (e.g., active employees, new hires) or at certain times (e.g., during the initial offering).

• Short-form underwriting, which allows an individual to apply for coverage using a shortened form that does not include detailed medical history questions, and does not require the applicant to have a physical exam. The form is typically one page and consists of as few as six questions, all of which are directly related to current need or recent use of long-term care services.

Exhibit 3.14 illustrates the increased access that employer sponsored LTC insurance can provide. All employers in the random sample reduced the barriers to obtaining LTC insurance to some degree, in the form of either reduced levels of underwriting (i.e., short-form) or a waiving of the underwriting requirement at certain times (e.g., during the initial offering, or when an employee is hired). Nearly all employers (96 percent) also offered LTC insurance with reduced underwriting to retirees, and majorities of those offering to spouses and parents required reduced underwriting for these groups (67 percent of those offering to spouses, and 56 percent of those offering to parents). While not presented in Exhibit 3.14, it appears that the majority of employers required at least short-form underwriting of their retirees, whereas active employees were more likely to have underwriting waived at least for a certain period of time.

EXHIBIT 3.14. Percent of Sampled Employers Using Minimal or No Underwriting1, by Eligibility Group

Offerings with "minimal or no underwriting" are defined as offerings that either allow guarantee issue at some point for the listed group or use short-form underwriting for the listed group. For each eligibility group, the sample includes only those employers offering LTC insurance to that group.

c. Premiums

Premiums varied widely across plans, reflecting the range of benefit levels available and other plan features that differ across plans, such as inflation protection and non-forfeiture. Premiums also increase sharply with age. The variation in LTC insurance plan features across employers makes it difficult to compare premiums across the entire sample of employers for the same benefit. We found that for an individual at age 50 with a plan containing both nursing home and home care coverage, premiums ranged from $7.30 to $92.10 per month; however, much of this variation is attributable to differences in plan features.

Optional plan features can increase the monthly premium well above the base plan premium. Exhibit 3.15 shows the range of monthly premiums available through the LTC insurance plan of one sampled employer, as well as the percent increase in premium over the base plan that results from the addition of the specified plan feature. For a person purchasing LTC insurance at age 50 through this plan, premiums range from $31 to $132, with a single optional plan feature increasing the base premium anywhere from 3 percent to 84 percent.

EXHIBIT 3.15. Monthly Premiums for a Sampled Employer's LTC Insurance Plan by Plan Features
Plan Features	Premium	Premium Increase Over Base Plan
Base Plan Only	$31	N/A
Base Plan plus Return of Premium at Death	$32	3%
Base Plan plus Reduced Paid Up Non-forfeiture	$44	43%
Base Plan plus 5% Compound Inflation Protection	$118	281%
Base Plan plus all optional plan features	$132	330%

For reasons outlined earlier (i.e., administrative and commission savings, employer negotiation of lower insurer profit margins), employer group plans tend to have considerably lower premiums than individually purchased policies for the same age. However, because the survey only included employer-based plans, we cannot specifically compare premiums in the employer group LTC insurance market to those in the individually purchased market. Insurer representatives we interviewed suggested that employer-based plans are on average 10 to 20 percent less expensive than comparable individually sold plans.

In addition to lower premiums for the same age group, premiums for employer sponsored products are on average lower than individually purchased products because of the younger purchase age. Exhibit 3.16 shows how monthly premiums increase with issue age for one sampled employer's LTC insurance plan both with and without optional plan features. Monthly premiums for this insurer are more than four times as high if LTC insurance is purchased at age 70 than if purchased at age 50.

EXHIBIT 3.16. Monthly Premiums by Issue Age for a Sampled Employer's LTC Insurance Plan With and Without Options

d. Employer Contributions

The vast majority of surveyed employers (94 percent, or all but two) did not contribute to the LTC insurance premium (Exhibit 3.17). One employer paid for a basic product that covered nursing home care, but not home care. Another employer paid 20 percent of the LTC insurance premium for a basic LTC insurance plan covering both nursing home and home care. Another employer contributed indirectly through a flex plan. Employees then paid the full premium for additional buy-up options, such as inflation protection.

Recent and proposed legislative changes may increase employers' willingness to contribute to LTC insurance premiums. The Health Insurance Portability Accountability Act (HIPAA) of 1996 clarified the tax status of LTC insurance and established product criteria for tax qualification. However, HIPAA reinforced the prohibition LTC insurance in cafeteria plans that are tax deductible. Employers often use a cafeteria plan as a vehicle for contributing to employees' insurance premiums for health insurance or life insurance. The HIPAA prohibition may discourage employers from contributing to LTC insurance premiums, because employers may be more likely to contribute to LTC insurance premiums only if LTC insurance were one of several benefits among which their employees could choose. Several pieces of recent legislation have proposed allowing tax deductions for LTC insurance premiums. If legislation passes that allows employers to deduct the cost of premiums, they may be more likely to contribute.

EXHIBIT 3.17. Percent of Surveyed Employers Contributing to LTC Insurance Premiums

4. Motivation for Offering LTC Insurance

The survey provides information about how and why employers developed their LTC insurance offering. The most common motivation for offering LTC insurance was a desire to offer leading edge benefits (see Exhibit 3.18). Employers also cited the low cost to the organization and the fit with their particular work force as strong motivating factors. This data is consistent with data from the Mercer study and the IFEBP¹⁸ surveys.

5. Implementation of the Offering

a. History of Offering

In recent years, LTC insurance has more frequently become an option for employers. The survey data reflect this recent growth (Exhibit 3.19).

b. Identifying Potential Insurers

The survey provides an overview of how employers solicited information from insurers. While approximately one-fifth of employers considered using more than one insurer for their LTC insurance offering, only three of 46 surveyed employers ultimately used more than one insurer for their LTC insurance benefit.

EXHIBIT 3.18. Factors Cited as Reasons for Employers to Offer LTC Insurance
Importance in decision to offer LTC insurance(1=not important5=extremely important):	Random Sample(n=37)		Select Sample(n=18)		Total(n=55)
	Mean	Rank	Mean	Rank	Mean	Rank
Chance to offer leading-edge benefits	3.95	1	3.61	1	3.84	1
Offered at low cost to organization	3.69	2	3.61	1	3.67	2
Good fit for workforce	3.22	3	3.22	2	3.24	3
Employees wanted it	2.81	4	3.00	3	3.00	4
Senior management wanted it	2.51	7	2.85	4	2.65	5
Helped achieve human resource benefit	2.81	4	3.00	3	2.63	6
Positive word of mouth	2.65	6	2.44	6	2.51	7
Uncertainty about the future of Medicare and Medicaid	2.41	8	2.52	5	2.42	8
Retirees wanted it	2.25	9	2.44	6	2.33	9
Offered by competitors	2.14	10	2.07	7	2.18	10
Advised by employee benefit consultant	2.08	11	1.93	8	1.93	11
Compensate for reductions in or greater contributions to other benefits	1.46	12	1.33	10	1.45	12
New govenment tax incentives	1.46	12	1.55	9	1.45	12
Collective bargaining agreement	1.06	13	1.35	10	1.19	13

EXHIBIT 3.19. Year Surveyed Employers First Offered LTC Insurance

i. Process for Obtaining Proposals from Insurers

Most employers sampled choose to send out a request for proposal (RFP) to solicit bids from insurers (Exhibit 3.20). Sending out RFPs indicates that the employers actively solicited information from multiple insurers, rather than passively receiving information marketed to them. Employers in our select sample which tended to be larger and invested more resources into developing a LTC insurance offering, were much more likely than randomly selected employers to issue an RFP. Most of both samples of employers had a benefit consultant assist in the preparation of the RFP. A few relied on the assistance of an insurance broker in the preparation of the RFP or the selection of an insurer.

Employers issuing RFPs sent them to an average of between 5 and 11 insurers. Employers in the random sample were more likely to rely on the advice of a benefit consultant in determining which insurers to send an RFP. Employers in the select sample may have relied less on a benefit consultant because they tended to be larger and have more in-house knowledge and, therefore, had less need for an outside benefit consultant. Another explanation may be that because the employers started offering LTC insurance earlier, there may not have been a substantial pool of LTC insurance benefit consultants available. Select sample employers appeared to be much more likely to send the RFP out to all top insurers.

EXHIBIT 3.20. Elements of the Process of Sending out Request for Proposals (RFP)
	Random(n=38)	Select(n=28)	Total(n=57)
Considering using more than one insurer	21%	21%	21%
Chose to use more than one insurer	6%	7%	7%
Percentage of employers that issued an RFP	58%	82%	68%
Average number of insurers to which RFPs were sent	5	11	8
Percentage that used benefits consultant in preparation of RFP	76%	67%	69%
Percentage for which broker assisted in preparation of RFP/selection of insurer	35%	10%	26%
Factors influencing which insurer was sent RFP
Advice of consultant	80%	63%	66%
Sent to top insurers	58%	72%	66%
Sent to all insurers offering LTC insurance	6%	31%	21%
Sent to insurers with relationship with company	22%	39%	37%
Other	8%	17%	19%

ii.Information Requested in RFP

Most RFPs requested a broad array of information from the insurer (see Exhibit 3.21). The majority of RFPs requested information about premiums, underwriting, burden on the employer and the background of the insurer. RFPs, especially those put forth by the random sample, were less likely to ask more detailed underwriting and pricing information, such as the level of denials that could be expected given the proposed underwriting criteria or information about the assumptions used to develop premiums. Employers requested financial information regarding the assumptions used for developing pricing and the insurer's history of increasing rates most frequently.

EXHIBIT 3.21. Information Requested in RFP
	Random(n=37)	Select(n=28)	Total(n=56)
Design information sought
Affordability of premiums	89%	93%	93%
Whether plans were underwritten	81%	93%	86%
Which groups underwritten	81%	93%	86%
Underwriting criteria	73%	79%	77%
Anticipated percentage of denials	35%	43%	39%
Length of time the insurer has marketed employer LTC insurance plans	76%	82%	79%
Estimates of cost to company in dollars	68%	82%	77%
Financial ratings of insurer	65%	89%	77%
Marketing strategy of insurer	65%	57%	63%
Estimates of cost to company (employee time)	43%	50%	45%
	Random(n=35)	Select(n=25)	Total(n=50)
Financial information requested from insurer
Pricing assumptions	61%	72%	67%
Premium rate history	47%	68%	55%
Formulas for calculating premium rates	47%	68%	55%
Loss ratios	37%	60%	50%
Expense factors	49%	48%	49%
	Random(n=35)	Select(n=26)	Total(n=53)
In soliciting information from insurers or in RFP, the employer:
Specified their desired benefit structure at the outset	17%	25%	20%
Began with a benefit structure but allowed the insurer to make suggestions	26%	29%	30%
Did not have benefit structure fleshed out at outset, and followed the lead of the insurer	51%	46%	47%

Employers in the select sample were somewhat more likely than randomly selected employers to request specific information in their RFPs. These employers were also somewhat more likely to specify the type of benefit structure at the outset and not allow the insurer to make suggestions to that structure than the random sample. A higher percentage of the random sampled employers followed the lead of the insurer in defining the benefit structure.

c. Selecting an Insurer

The survey indicated that plan features were the most important factor in an employer's selection of an insurer (see Exhibit 3.22). For employers in the random sample, the insurer's reputation or name recognition was the second most important factor, followed by customer service and price. Among the select sample, employers ranked plan features offered as the leading deciding factor followed by initial price of plans and customer service.

EXHIBIT 3.22. Factors Employer Deemed Important in Selecting an Insurer
(scale 1=not important; 5=extremely important)	Random(n=37)		Select(n=27)		Total(n=55)
	Mean	Rank	Mean	Rank	Mean	Rank
Plan features offered	4.4	1	4.5	1	4.5	1
Customer service	4.2	3	4.2	3	4.2	2
Reputation/brand name	4.3	2	4.0	4	4.2	3
Initial price of plans	4.0	4	4.2	2	4.2	4
Potential for price increases	3.7	6	3.6	4	3.8	5
Marketing and communication plan	3.8	5	3.5	5	3.8	6
Current customers/references	3.6	7	3.3	6	3.6	7

d. Selecting the Plan Design

The survey indicated that the employers' personnel staff had the most influence over selecting plan design and plan features (see Exhibit 3.23). This may have been prejudiced by the fact that personnel staff members were most likely to have filled out responses to the survey. Respondents considered senior management and the insurer as the second most important influence. Specially appointed committees or task forces and benefit consultants were also rated as being influential.

e. Marketing the Offering

Employers and insurers approached educating employees about long-term care and the potential protection offered by LTC insurance in a variety of ways. The most common method employed was brochures and flyers (see Exhibit 3.24). Enrollment meetings were the second most common technique for providing information to employees. Seventy percent of randomly sampled employers held enrollment meetings, and all of those who did held them on the company's premise and employees were given time off to attend meetings. A majority of employers said that these meetings were well attended. However, senior management generally did not have a high profile during these meetings. A majority of offerings also provided information about LTC insurance during new employee orientation, had a cover letter from the employer endorsing the purchase of LTC insurance, mentioned LTC insurance in the company newsletter and had posters or tables with information about LTC insurance.

EXHIBIT 3.23. Relative Influence of Different Parties in Selecting Plans or Plan Features
Scale: 1=no influence; 5=extremely influential)	Random(n=37)		Select(n=27)		Total(n=54)
	Mean	Rank	Mean	Rank	Mean	Rank
Human resources/personnel staff	3.3	1	3.2	1	3.4	1
Senior management	2.9	3	3.3	2	2.9	2
The insurer	3.0	2	3.0	3	2.9	2
A specially appointed committee/task force	2.7	4	2.8	4	2.7	4
A benefit consultant	2.6	5	2.8	5	2.6	5
Survey of employees	2.1	6	1.9	8	2.0	6
State regulations	1.8	8	2.2	6	2.0	7
A broker	2.1	6	1.6	12	1.9	8
HIPAA requirements	1.7	9	1.7	7	1.7	9
Focus group of employees	1.7	10	1.5	10	1.6	10
Experiences with a previous LTC insurance plan	1.3	11	1.6	9	1.3	11
Collective bargaining	1.1	12	1.3	10	1.2	12
Third party administrator (TPA)	1.1	13	1.2	13	1.1	13

EXHIBIT 3.24. Techniques Used to Inform Employees about LTC Insurance
	Random(n=40)	Select(n=29)	Total(n=59)
Highlights brochures/flyers	65%	71%	68%
Placed in payroll envelopes	18%	15%	17%
Enrollment meetings1	70%	76%	75%
Percent held in company premises	70%	76%	75%
Percent held on company time	70%	72%	75%
Percent where senior management introduced meetings	25%	25%	23%
Percent that were well attended	50%	62%	57%
LTC insurance education part of new employee orientation	60%	66%	65%
Cover letter from employer sent to employees	65%	62%	63%
Mentioned in company newsletter	58%	71%	63%
Posters/table tents	53%	55%	55%
Included in employee handbook	45%	64%	40%
Videotape	28%	45%	33%
Employer contributed to communication/education budget	18%	25%	20%
E-mail	20%	18%	18%
Emails from senior management	3%	0%	2%
Voice mail	8%	11%	7%
Voice mail from senior management	3%	4%	3%
Company intranet	23%	14%	18%
Percentages of enrollment meetings are calculated from those employers who reported having enrollment meetings.

f. Enrollment Process

The majority of employers set a time limit in which employees were able to enroll (see Exhibit 3.25). However, employers were much more likely to limit the enrollment period for initial offerings and for new hires than for the most recent offering. The average time period ranged between 1.0 and 1.5 months.

EXHIBIT 3.25. Summary of Enrollment Period Features
	Random(n=43)	Select(n=27)	Total(n=58)
Percentage setting time limit for initial offering	88%	85%	86%
Average number of days	41	42	42
Percentage setting time limit for most recent offering	54%	63%	67%
Average number of days	34	39	37
Percentage setting time limit for most new hires	72%	76%	76%
Average number of days	34	36	36

Beyond allowing direct deduction of LTC insurance premiums from pay or pension checks, employers generally did not take advantage of mechanisms that could make enrollment in LTC insurance easier for employees, such as making on-site insurance staff available, as well as telephone and intranet enrollment (Exhibit 3.26). Randomly selected employers in our survey were generally more reliant on outside personnel from the insurer coming in and facilitating the enrollment process than the select sample. We speculate, given the larger size of the employers, the select sample may have had better mechanisms for enrolling employees themselves. Employers in the select sample were also more likely to allow employees to enroll over the telephone.

EXHIBIT 3.26. Percentage of Employers Using Mechanisms that Could Facilitate Enrollment
	Random(n=40)	Select(n=29)	Total(n=59)
On-site enrollment by insurer personnel	33%	17%	25%
Telephone enrollment	20%	28%	24%
Intranet enrollment	3%	3%	3%
Direct deduction of premiums from pay or pension check	82%	79%	83%

6. Employers' Satisfaction with the Offering

Exhibit 3.27 outlines employer satisfaction with the LTC insurance offering. Over 40 percent of random and select sampled employers reported that administering the offering was extremely easy. This finding is remarkably consistent with the Mercer study. The majority of the sampled employers were glad they had the offering. A higher percentage of the select employers reported they were extremely glad they had the offering. Again, these findings mirror the Mercer study.

EXHIBIT 3.27. Employer Satisfaction with Ease of Administration

Overall Employer Satisfaction with LTC Insurance Program

Major Changes Employer Would Make

Almost 50 percent of the randomly sampled employers indicated that if they had to do it all over again, they would focus on improving communication to individuals during initial enrollment and subsequent open enrollments. The need for better communication during initial and subsequent enrollment periods also was the most consistently cited change in the Mercer study. Over 20 percent of the random sample, less than 5 percent of the select sample, indicated that they would tie long-term care more closely to employees' retirement planning. Almost 20 percent of the random sample and less than 5 percent of the select sample indicated that they would find out more about employee wants and needs beforehand if they were to implement the offering all over again. Thirty-five (35) percent of select employers said they would make no change whereas only 3 percent of the random sample said they would make no change. The findings regarding employer satisfaction for the random sample were similar to what Mercer found. The findings for the select sample can be interpreted as indicating that they were, in fact, "select" employers who demonstrated best practices.

B. Factors Related to Participation Rates

In this section, we discuss key features related to successful enrollment efforts. As shown in Exhibit 3.28, LTC insurance purchase rates in the random sample were generally low. The median participation rate among active employees was 3.1 percent. Over 40 percent achieved employee enrollment rates no higher than 2 percent, and two-thirds had rates below 6 percent.

Participation rates among our random sample were lower than those reported by Mercer from their 1997 survey.¹⁹ For example, while Mercer reported that 51 percent of surveyed employers had participation rates of at least 5 percent, only 38 percent of our random sample reported participation that high.

Our sample size generally prevents us from identifying whether certain factors have a statistically significant relation to participation rates. However, we identify practices that appear to be related to successful enrollment in this section.

a. Simplicity of Purchase Decision

More simple policy offerings may be related to higher participation. The complexity measure we described in section III.A.2 was correlated with participation in the predicted direction (r=-0.15); greater complexity was associated with lower participation. However, this correlation was not significant. The lack of significance is not surprising given the small sample size and inability to control for confounding variables. The insurers that we interviewed consistently stated that individuals are less likely to enroll if they must contend with many decisions and choices. One large insurer indicated that, using their more extensive proprietary database of their employer offerings, complexity was significantly negatively correlated with participation.

EXHIBIT 3.28. LTC Insurance Participation Among Employees

LTC Insurance Participation Among Retirees

b. Backing from Senior Management

Visible support for the LTC insurance offering from senior management also appeared to encourage enrollment. Among employers that "announced and visibly supported the program," 39 percent (or 9 of 23 responding) reported participation rates of 10 percent or greater. In contrast, only 10 percent (3 of 29 responding) of employers who did not show such support reported participation rates of 10 or more percent.

Viewed another way, three-quarters of employers who experienced high participation (9 of 12) had demonstrated support for the offering from senior management. These survey results corroborate comments made by all the major insurers we interviewed. These insurers stressed that the most successful offerings were often those where senior management demonstrated strong commitment and support for the offering, sometimes relating to employees their own difficulties when trying to obtain long-term care services for a frail relative.

c. Education and Marketing

Related to the level of commitment to the offering senior management demonstrates is the extent and quality of the education and marketing efforts the employer put in place. We created an education score equal to the number of vehicles the employer used to educate their potential enrollees about the offering (these techniques are listed in Exhibit 3.24 above). Employers with high education scores were much more likely to have high participation rates compared to employers with lower education scores. For example, 6 of the 9 employers (67 percent) who had education scores of 13 or greater (of a possible 19) achieved participation rates of 6 percent or higher. In contrast, only 12 of the 43 employers (28 percent) who had education scores of 12 or lower achieved participation rates of six percent or greater. Further, the two employers with the highest education scores also reported the highest participation rates.

This is consistent with the responses employers gave when asked what they would change if they could do the initial enrollment again. Communication of the offering was the most commonly cited area employers wished to improve, and employers with low participation rates were much more likely to report needing improvement in this area than employers with higher participation rates. Among employers with participation rates of less than 2 percent, 44 percent stated that they would have liked to improve the way they communicated the offering to potential enrollees. In contrast, only 17 percent of employers with participation of 10 percent or more stated that they would have liked to improve communication, suggesting that they believed they had done an adequate job of communicating the offering to their employees.

d. Underwriting

Employers that permitted guarantee issue or minimal underwriting appeared to have higher participation rates than those who used standard underwriting, although the correlation is not statistically significant. Only two employers (17 percent) who did not permit minimal or no underwriting had enrollment rates of 6 percent or higher, while 16 (41 percent) of those who allowed minimal or no underwriting achieved participation rates of at least six percent. This suggests four implications: (1) these companies may have been highly committed to the offering and also employed other strategies that increased enrollment; (2) employees may be more receptive to an offering that does not require them to reveal personal health information; (3) the additional enrollment that these companies experienced could be individuals that might have otherwise been screened out, which would increase adverse selection; and (4) employers (or their selected insurer) who anticipated high participation rates may have been more comfortable using a minimal or no underwriting.

IV. DISCUSSION

The data presented in this report highlight a number of policy issues, and also raise some concerns. These are discussed below.

A. Potential for Growth in the Private LTC Insurance Market

The rapid growth in the employer LTC insurance market may offer the best opportunity for LTC insurance to protect a large number of consumers. In addition to advantages that make employer-based LTC insurance more accessible than LTC insurance sold to individuals discussed in the introduction of this paper, there are a number of reasons to expect the current growth in employer-based LTC insurance to continue.

First, products are significantly more attractive to consumers wants and are evolving in ways that may make them even more marketable in the future. Unlike many LTC insurance policies offered in the 1980s, current policies may allow beneficiaries to be cared for in their home and maintain a high degree of choice over their care. The flexibility of benefits offered by policies is likely to continue as insurers gain more experience in providing benefits and have blocks of claimants large enough to justify developing more sophisticated care management mechanisms and negotiating more discounts from providers.

Second, insurers and employers are becoming more sophisticated about rolling out employee offerings. Leading insurers are analyzing their experience with employers to determine which marketing and education techniques are the most effective. In addition, insurers are learning to tailor marketing to an individual employer. For example, an insurer marketing to an employer with a relatively young workforce may emphasize that LTC insurance will offer protection to younger people who become seriously impaired.

Third, proposed legislation authorizing an offering of LTC insurance to federal government employees may result in other employers viewing LTC insurance as a more standard benefit. Employers that benchmark their benefits to the federal government would be much more likely to offer a policy. The extent to which the federal government presents a model that other employers will emulate will, to a large extent, depend on the perception or the degree of success for that offering. If the offering is viewed as a way of providing a needed benefit to substantial numbers of employees at little cost to the employer, many more employers are likely to follow suit. However, if the federal offering is viewed as a substantial administrative hassle that only a small minority of employees purchase, other employers may be even more reluctant to offer LTC insurance in the future.

B. Potential Limitations of Employer-Based Private Coverage

Even if employer-based LTC insurance continues to grow, several trends threaten its ability to pay for a substantial portion of long-term care costs. This survey effort revealed several issues that will need to be addressed before employer-based LTC insurance can become a major source of long-term care financing. These include:

1. insufficient numbers of individuals having any long-term care protection when they need it because of low enrollment rates and probably high lapses (persistency);

2. insufficient protection for those who do retain their coverage until they need benefits because many employees may fail to purchase inflation protection;

3. a lack of value for individuals who pay substantial premiums and then lapse because of a lack of interest in non-forfeiture benefits; and

4. limited provision included in policies to adapt to changes in the long-term care delivery system

Low enrollment rates and possibly high lapses -- If current trends prevail, only a small minority of employees will have LTC insurance coverage when they need benefits. As the findings of this study indicated, even for the most successful offerings, participation rates represent only a fraction of employees. In addition, if the employer market is similar to the individual market, where persistency has been a problem, then a related question is how many will retain coverage.

Possible failure to elect inflation protection -- Many individuals who purchase policies may find that the value of their benefits have eroded substantially by the time they need care. This study found that most employer-based LTC insurance offerings include the option of buying more coverage in the future as the only mechanism for protecting benefits from erosion caused by inflation. Unfortunately, we did not have any data regarding the percentage of employees without built-in inflation protection who upgraded their coverage to keep pace with inflation.

While the NAIC model regulations include a mandatory offering of 5 percent compound inflation protection built into the initial premium, insurers are required only to offer this as an option to employers. It is up to employers to then offer this option as part of their employee offering. This study indicates that only small proportions of employers are choosing to do so. In addition, our interviews with employers and insurer representatives suggests that when employers offer this type of inflation protection, only a minority of employees choose it, apparently because it adds considerably to the cost of premiums. Thus, the challenge is to find a way to deliver inflation protection to younger ages or do a better job of locking in the option as employees age.

Lack of interest in non-forfeiture benefits -- Only about one-half of employers offer non-forfeiture benefits and the forms offered provide the least amount of protection for consumers in need of long-term care. Because purchasers of employer-based LTC insurance are likely to be younger, the time between purchase and receipt of benefits tends to be substantially longer than for individually purchased policies. This may increase the likelihood that purchasers will let policies lapse, even after having paid substantial premiums. These individuals will receive no benefits if they lapse without electing non-forfeiture benefits.

This study found that non-forfeiture benefits that are included in employer LTC insurance offerings typically fall into the categories of "return of premium on death" or "reduced paid up." A return of premium on death provision pays a portion of premiums paid if the individual dies without going into claims. A major concern about this type of non-forfeiture is that it offers no protection against long-term care. In addition, the amount paid does not consider interest earned on premiums paid, which will account for most of the equity an individual builds in a LTC insurance policy over time. However, it is likely, employees view this benefit as of more value because they will not have "wasted" their money if they do not use the policy.

Reduced paid up benefits (RPU) pay long-term care benefits to individuals requiring care for the same benefit period included in their policy. However, the daily and lifetime maximum benefit amounts are reduced. This benefit may offer much less protection than a "shortened benefit period" (SBP) form of non-forfeiture, which provides the full daily benefit amount for a shorter period. This form (SBP) of non-forfeiture will still protect individuals who have relatively short periods in which they need long-term care (e.g., the individual breaks a hip and needs to go into a nursing facility for several months). In contrast, a RPU benefit of $10 a day is not likely to prevent someone from spending down their assets in a nursing facility or provide them with enough home care to prevent a nursing facility stay.

Limited provision included in policies to adapt to changes in the long-term care delivery system -- Changes in the long-term care delivery system will likely necessitate changes in current policies. Because the current long-term care delivery system is rapidly changing and the lag time between the purchase of employer-based LTC insurance policy and receipt of benefits is typically long, the range of services covered in current policies may not correspond well with service options available when the individual goes into claims. Insurers and employers will likely have to update benefits periodically to respond to these changes. Our interviews with insurers indicated that insurers have been updating coverage for employers by unilaterally expanding covered services.

Employer involvement in their LTC insurance offering beyond helping to enroll employees could substantially improve the likelihood that employees will benefit from purchasing this coverage. Because the long-term care delivery system and needs are likely to change over time, employers will need to renegotiate benefits and premiums on a periodic basis. Two mechanisms may assist employers in this endeavor:

• As opposed to current policies that proscribe covered services and benefit levels, premiums paid into employer-based LTC insurance offerings could be used to create a "trust" fund which would be directed with input from both the employer and insurer to meet the needs of current and former employees. This trust fund could be used in much more flexible ways to provide assistance as the long-term care needs and delivery systems evolve.

• Employers may want to consider forming coalitions to reduce administrative costs and ensure that they have sufficient clout to influence decisions regarding how benefits should be restructured as long-term care needs and services evolve.

C. Importance of Education

Both employers and employees require substantial education about the value and pitfalls of LTC insurance. Education of employers and employees is vital to alleviating the concerns about employer-based LTC insurance raised above. The survey results indicated that education was valued in the select sample of employers. These companies used a variety of means to educate employees about LTC insurance. Three-quarters of the companies used flyers, new employee orientation, and the company newsletter as opportunities for education. Over half of the companies used the employee handbook, a poster or table tent, and a cover letter to inform employees.

The focus groups and interviews we conducted with insurers and employers highlighted that insurers serve as the major source of information about long-term care needs and LTC insurance for both employers and employees. While many employers rely on benefit consultants to develop an offering (the survey found that two-thirds of employers used a benefit consultant), insurers interviewed noted that many benefit consultants lack experience with LTC insurance. Insurers noted that they often have to spend substantial amounts of time educating the benefit consultants. This lack of experience would likely hamper the benefit consultants' ability to negotiate on the behalf of employers with the insurers. Employers in the random sample were much more likely to use a benefit consultant in the process of selecting an insurer indicating that they may have a disadvantage in negotiating with insurers if the benefit consultant is not knowledgeable about long-term care issues.

Presumably, one of the benefits of employer-offered LTC insurance is that the employers would educate themselves and negotiate with insurers as advocates for their employees rather than simply serving as a market entry point for insurers. To accomplish this both employers and employees need to be educated about the pitfalls, as well as the benefits, of LTC insurance.

V. FUTURE RESEARCH

While this survey adds to our understanding of the LTC insurance coverage offered by employers, several issues need to be investigated to more fully assess the value of employer-based LTC insurance. Future research could focus on addressing the following questions:

• What types of LTC insurance are employees buying and are they retaining their coverage? While this research described employer offerings, identifying actual policies that employees purchased was beyond the scope of the study. Assessing lapse rates and the extent to which employees who purchase a policy protect their benefit from erosion caused by inflation either through built-in protection or choosing to purchase more coverage in the future (i.e., exercising the future purchase option) are key issues. Research that has been done in this area has, for the most part, been proprietary and focused on improving the marketing of employer LTC insurance.²⁰

• What role do and should employers play in managing LTC insurance reserves and renegotiating LTC insurance contracts? The survey and the interviews suggests that employers currently only play a limited role in designing policies and that changes to covered benefits and management of reserves is almost always controlled by the insurer. However, the survey only addressed this item in a limited fashion. The creation of guidelines for employers about roles they can play in managing reserves and updating benefits may be of the greatest use to employers.

• Are employer coalitions to purchase and manage LTC insurance feasible and if so, what are the best strategies for building and maintaining these coalitions? Small and medium size employers may not be able to independently devote resources to developing and monitoring a LTC insurance offering. Employers may wish to pool necessary resources in a purchasing coalition. Research could investigate whether employers would be interested in developing these coalitions and whether current mechanisms for pooling administrative responsibilities (e.g., outsourcing payroll or other employee benefits) could be expanded to include LTC insurance.

• To what extent do employees want LTC insurance and what features would they most like to have included in a policy? What difference does the underwriting mechanism make in enrollment rates? Independent market research regarding the appeal of LTC insurance to employees is lacking. Insurers probably have conducted their own proprietary market research. Independent research could assist employers in making design decisions, especially if those decisions are contrary to the interests of long-term care insurers (e.g., offering a self-funded policy).

• What do the demographics of the population purchasing LTC insurance look like? What types of employees are purchasing the plans and how many employees have enough assets to protect with LTC insurance? Are the same employees buying LTC insurance as are contributing to company 401(k) plans or company stock purchase plans? Would more people purchase LTC insurance if the company contributed to the premiums?

REFERENCES

Bureau of Labor Statistics, United States Department of Labor. (1999, September). Employee Benefits in Medium and Large Private Establishments, 1997. Bulletin 2517. Washington, DC.

Cheung, M. (1997, November). John Hancock presentation at the Private Long-Term Care Insurance Conference, San Diego, CA.

Employee Benefit Research Institute. (1995, February). Sources of Health Insurance and Characteristics of the Uninsured (Issue Brief, No. 158). Washington, DC:

Employee Benefit Research Institute. (2000, April). Employer-Sponsored Long Term Care Insurance: Best Practices for Increasing Sponsorship (Issue Brief, No. 220). Washington, DC: Jeremy Pincus.

Health Care Financing Administration. (1996). Key issues for long-term care insurance: Ensuring quality products, increasing access to coverage, and enabling consumer choice. Report prepared by Lewin-VHI and The Brookings Institution for HCFA. Contract No. 500-89-0047.

Health Insurance Association of America. (1997, May). Long Term Care Insurance in 1995: Research Findings. Washington, DC: Cornel and Fulton.

Health Insurance Association of America. (2000, March). Research Findings: Long-Term Care Insurance in 1997-1998. Washington, DC: Coronel, S.A.

International Foundation of Employee Benefit Plans (1999, June). Employer-Sponsored Long-Term Care Insurance: Did HIPPA matter? Brookfield, WI.

William M. Mercer, Incorporated. (1997). State-of-the-Art in Long-Term Care Insurance: Results of a Mercer Study of Employers. San Francisco, CA.

Life Insurance Marketing Research International. (1999, April). U.S. Group Long-Term Care Sales Show Mixed Success in 1998. Windsor, CT.

Life Insurance Marketing Research International (2000). U.S. Group Long-Term Care Insurance: Sales and In force. Windsor, CT: Patricia A. Ash.

The Segal Company. (1997, October). Long Term Care for State Employees. A Survey by The Segal Company. Unpublished memorandum. Yaggy, Clark J.

U.S. Bureau of the Census. (1998). Statistical Abstract of the United States: 1998 (118th edition). Washington, DC.

NOTES

1. Based on estimates from the Lewin Group Long-Term Care Financing Model (LTCFM): April 2000.

2. HIAA (March 2000). Research Findings : Long-Term Care Insurance in 1997-1998. Coronel, S.A., Washington, DC.

3. Ibid

4. Life Insurance Marketing Research (2000). U.S. Group Long-Term Care Insurance: Sales and In force. Hartford, CT: Patricia A. Ash. Compared with 1998 results, employer groups increased 36 percent, participants increased 24 percent, and premiums or their equivalents collected increased 28 percent.

5. Health Care Financing Administration. (1996). Key issues for long-term care insurance: Ensuring quality products, increasing access to coverage, and enabling consumer choice. Report prepared by Lewin-VHI and The Brookings Institution for HCFA. Contract No. 500-89-0047.

6. Employee Benefit Research Institute. (1995, February). Sources of Health Insurance and Characteristics of the Uninsured (Issue Brief, No.158). Washington, DC:

7. Employee Benefit Research Institute. (2000, April). Employer-Sponsored Long Term Care Insurance: Best Practices for Increasing Sponsorship (Issue Brief, No. 220). Washington, DC: Jeremy Pincus.

8. Based on civilian employment data from the Statistical Abstract of the United States, the number of workers over age 45-64 in 1998 is projected to have been just over 42 million. The Life Insurance and Marketing Research Association (LIMRA) reports approximately 645,000 policies were sold through the employer market by 1998. Thus, assuming 4.2 million persons bought policies, the total employer sales would be 6.5 times the 1998 estimates. U.S. Bureau of the Census, Statistical Abstract of the United States: 1998 (118th edition). Washington, DC, 1998; LIMRA, U.S. Group Long-Term Care Sales Show Mixed Success in 1998, (April 1999).

9. Focus groups of industry representatives conducted at the Private Long Term Care Insurance Conference in San Diego, CA in November 1997.

10. Health Insurance Association of America. (1997, May). Long Term Care Insurance in 1995: Research Findings. Washington, DC: Cornel and Fulton.

11. Health Insurance Association of America. (2000, March). Research Findings: Long-Term Care Insurance in 1997-1998. Washington, DC: Coronel, S.A.

12. The Segal Company. (1997, October). Long Term Care for State Employees. A Survey by The Segal Company. Unpublished memorandum. Yaggy, Clark J.

13. Health Insurance Association of America. (1997, May). Long Term Care Insurance in 1995: Research Findings. Washington, DC.

14. HIAA, 1997.

15. The Bureau of Labor Statistics conducts the Employee Benefits Survey annually, but divides the data collection into two biannual series collected in alternating years based on establishment size. The most recently available survey of medium and large private establishments, from 1997, reports a response rate of 53 percent. (Bureau of Labor Statistics 1997, p. 160").

16. CNA is better represented among employers submitting full survey responses. Employers using Aetna, however, were less well represented among those with full responses, because a majority of Aetna employers submitted partial responses.

17. William M. Mercer, Incorporated. (1997). State-of-the-Art in Long-Term Care Insurance: Results of a Mercer Study of Employers. San Francisco, CA.

18. International Foundation of Employee Benefit Plans (1999, June). Employer-Sponsored Long-Term Care Insurance: Did HIPPA matter? Brookfield, WI.

19. William M. Mercer, Incorporated. (1997). State-of-the-Art in Long-Term Care Insurance: Results of a Mercer Study of Employers. San Francisco, CA.

20. Examples include a presentation by Malcolm Cheung of John Hancock at the Private Long-Term Care Insurance Conference in San Diego, CA, November 1997.