Experts Doubt Survey Findings on Health Plan Owners’ Satisfaction

Published February 1, 2006

A December 2005 survey found people with consumer-driven health care (CDHC) plans were less satisfied than those in comprehensive health plans. The Internet-based survey by the Employee Benefit Research Institute (EBRI) and Commonwealth Fund paints a far bleaker picture than CDHC proponents believe is true.

Nearly two-thirds (63 percent) of people with comprehensive health plans were either extremely satisfied or very satisfied with their plan, according to the survey. Nearly all of these (95 percent) reported they would like to stay with their existing plan. By contrast, the EBRI/Commonwealth data showed only 42 percent of people in consumer-driven health plans and one-third of people in high-deductible health plans were as satisfied with their respective plan.

The findings are at odds with the experiences of insurers offering CDHC plans, who have surveyed members and found satisfaction to be high. Lumenos, a company that has offered health reimbursement arrangements (HRA) for several years, found 96 percent of its enrollees were satisfied and nearly all said they would enroll again.

Aetna, which also has offered HRAs for several years, found about 90 percent of its enrollees were satisfied and would enroll again.

Sample Was Small

Proponents of consumerism in health care say there are several reasons why respondents to the EBRI/Commonwealth survey reported lower satisfaction with their consumer-driven health plans.

One explanation is potential problems with the survey’s methodology. As with any survey, one weakness is the characteristics of those who volunteer to participate. The people who volunteer to take a survey might not be a representative sample. Those most apt to volunteer are those with strong opinions: They either love or hate their health plan.

Another possible weakness is sample size. Strong opinions could be magnified by a small sample.

Only 17 of the 1,204 survey respondents had health plans qualifying as consumer-driven. Statisticians understand surveys with small response rates (or small sample sizes) are less reliable than those with a large number of responses. A mere 17 people responding to an Internet-based survey might not be representative of all consumers with CDHC plans.

Over-Sampling Another Problem

The authors of the EBRI/Commonwealth survey used a statistical technique known as “over-sampling” to improve accuracy and find out more about people with consumer-driven and high-deductible health plans (HDHPs). They intentionally sought out 505 additional people with high-deductible health plans (168 of which qualified as consumer-driven) and asked them about their preferences. The methodology section of the survey does not say where this additional sample came from, making it difficult for social scientists to evaluate whether the sample was representative of all people with similar health plans.

According to economist Linda Gorman, director of the Health Care Policy Center at the Denver-based Independence Institute, “Since they over-sampled after the fact, this could mean that they simply went out and found 168 more people with a CDHP.” If these respondents all came from the same source–a sampling technique known as a “snowball” sample–they may be biased.

For instance, if EBRI over-sampled individuals from an organization where less generous consumer-driven plans replaced low-deductible comprehensive plans, the sample may be biased by a high proportion of workers irritated at losing their previous health plan.

The EBRI/Commonwealth survey is consistent with a June 2005 McKinsey & Company survey that found “[o]verall, only 44 percent of the CDHP consumers stated that they were as or more satisfied with these plans than they had been with their previous, typically more generous, health benefits.”

Time Periods Inconsistent

The idea that some of the people in the EBRI/Commonwealth sample had recently been forced by employers into less-generous health plans is plausible because the survey found those in CDHC plans had been in their current plan for shorter periods of time than those in traditional health plans.

For instance, among those with CDHC plans, nearly half (46 percent) had owned the policy less than one year, while only about one-quarter (24 percent) had been with the plan three years. For those with traditional health plans, the ratio was reversed. More than half (54 percent) had been with their current traditional health plan for three years or more, compared to 21 percent who had been with the plan for less than one year.

Sample Respondents Less Healthy

Gorman also points out the out-of-pocket medical expenditure for people in the survey with either high-deductible health plans or consumer-driven health plans appears far higher than the average found in the Census Bureau’s Medical Expenditure Panel Survey (MEPS). “This doesn’t square with the fact that 50 percent of the population spends less than $664 a year total on health care according to MEPS,” according to Gorman.

She suggests those in the sample with high-deductible health plans appear to be sicker than average. Health costs tend to increase with age, and the study showed those with high deductible health plans (including CDHC) were older than respondents in traditional health plans.

Dr. Robert Hamilton, a physician who follows consumer-driven health care, adds, “I would expect a higher percentage of people who have high expenses to be dissatisfied in comparing a CDHP or a HDHP to a standard insurance plan. These would also be people who might be more likely to respond to an online questionnaire.”

Dr. Robert Hopper, a health benefits broker based in Santa Barbara, California and author of a recent book on health savings accounts, thinks satisfaction with health plans may be a matter of perception. “Workers often do not realize the cost of their health insurance impacts their take-home pay,” said Hopper. “When employees discover they are paying 100 percent of their health benefits, they will no longer want to pay such large amounts for health insurance. … They will no longer have the attitude of trying to get as much out of the insurance as they possibly can.”

HDHP Consumers Cost-Conscious

One point both CDHC proponents and the authors of the EBRI/Commonwealth survey agree on is that individuals enrolled in high-deductible health plans are more cost-conscious than those in traditional health plans. Individuals enrolled in CDHC (or HDHP) were nearly twice as likely to “delay or avoid getting health care due to cost” as those with comprehensive coverage, according to the EBRI/Commonwealth findings. Letting consumers choose between health care and other uses for their money is something supporters of CDHC have championed for a long time.

In traditional, comprehensive health plans, those using less care cross-subsidize those using more services. Thus individuals with higher than average out-of-pocket medical expenses and those who believe health plans are a free benefit may not be as satisfied with a higher deductible. Proponents and opponents alike largely agree that when people have an incentive to control costs, they tend to consume less health care.

Devon M. Herrick ([email protected]) is a health economist and senior fellow at the National Center for Policy Analysis.

For more information …

“Early Experience with High-Deductible and Consumer-Driven Health Plans: Findings from the EBRI/ Commonwealth Fund Consumerism in Health Care Survey,” by Paul Fronstin, EBRI, and Sara R. Collins, is available online at