This Article Abstract Submit a response to this article Services E-mail this article to a friend Alert me to new issues of the journal

Read a related paper by Melinda Beeuwkes Buntin et al.

D A T A W A T C H : 
L O W - I N C O M E  C O V E R A G E
W E B  E X C L U S I V E   

24 September 2003  
  

Who Enrolls In A Program For Parents Of 
Publicly Insured Children?

Evidence from Alameda County, California,
highlights the promise of a program
to insure low-income working parents.


by Erin Fries Taylor, Jeffrey T. Kullgren, and Catherine G. McLaughlin

ABSTRACT:

Although interest in expanding SCHIP coverage to parents has grown over the past five years, few such expansions have actually been implemented. State governments and health plan administrators remain concerned that these expansions will attract only high-risk enrollees, resulting in costly premiums that require large subsidies. We examine characteristics of enrollees in an SCHIP-like expansion program in Alameda County, California. According to our survey data, the program did not experience unfavorable selection. Rather, it attracted a broad range of eligible adults. Enrollees were comparable to the overall low-income population in Alameda County in terms of age, health status, and various utilization measures.

Proposals to cover parents of publicly insured children have started to receive considerable attention in recent years, yet states and private entities have been slow to enact such expansions.¹ Uncertainties surrounding the characteristics of and costs associated with new enrollees appear to be a major obstacle. State governments and administrators often make assumptions about likely enrollees. One such assumption is that only high-risk enrollees who use large amounts of costly health care would enroll. A second is that the eligible population would not be able to afford or be willing to pay even subsidized premiums. Determining whether these assumptions are reasonable is important in understanding whether such expansions represent a viable policy option.

This paper describes the characteristics of adults who recently enrolled in Alliance Family Care, a subsidized health care plan in Alameda County, California, aimed at children and their parents who are in households with incomes under 300 percent of the federal poverty level and who do not qualify for public programs. This program is not a state-sponsored Medicaid or State Children’s Health Insurance Program (SCHIP) demonstration, but rather a local initiative by a public agency.² Examination of its adult enrollees provides an example of who participates when offered this type of coverage and is of value to both public and private organizations debating the expansion of coverage to uninsured parents.

After describing the Alliance Family Care program, we present data on enrollees’ sociodemographic characteristics, baseline health status and use of services, and previous insurance status and access to care. In general, we find that adult enrollees have self-reported health status measures and rates of use that are comparable to those of low-income nonelderly adult populations in Alameda County. Although we were not able to obtain cost data associated with utilization, these results suggest that fears of serious adverse selection were not realized. Future research will provide important information about changes in use and access after enrollees have been part of the program for six to twelve months.

Background

The Family Care program is administered by the Alameda Alliance for Health, a public, not-for-profit health maintenance organization (HMO). To be eligible for this program, a person must be an uninsured resident of Alameda County, have household income under 300 percent of poverty, and be ineligible for public coverage. Additionally, adults must have one or more children enrolled in the Alliance to be eligible for Alliance Family Care.³ The program does not require that a person be a legal permanent resident or U.S. citizen, nor does it require that a person be uninsured for any minimum period of time before enrollment. Importantly, the Alliance does engage in medical underwriting of applicants. This underwriting takes a very limited form, relying on the presence of various self-reported medical conditions at the time of application. As a result, approximately 0.7 percent of applicants were denied coverage in 2002.

Adult enrollees pay a monthly premium for individual coverage that ranges from $20 to $120 per month, depending on age. Enrollees must make these monthly payments to remain in the program.⁴ The benefit package includes physician services (including preventive care), laboratory services, prescription drugs, and inpatient and outpatient hospital care. Some of these services require small copayments ($5–$15). Alliance Family Care began in July 2000 and had almost 7,300 members (about 70 percent of whom are adults) as of March 2003. Much of the recruitment of adults into the plan occurs as a result of medical services referral, through encounters either for themselves or for their children.⁵

Data And Methods

Data for this analysis come from a telephone survey of newly enrolled Alliance Family Care adult participants, conducted within six to eight weeks of enrollment. This baseline survey used a pretested, structured interview format. Although all households enrolled between August and December 2001 were included in the sample, only adult enrollees were eligible for the survey, and only one person per household was included, to maintain independence.⁶ Furthermore, respondents were asked only about their own health and health care, not the health and health care of their spouse or family. The survey was conducted in English and Spanish by the University of Michigan’s Institute for Social Research and in Cantonese by Field Research Corporation of San Francisco. The response rate was more than 84 percent both for Cantonese- and Spanish-speaking enrollees and about 70 percent for English-speaking enrollees.⁷ Given the high response rate, we expect respondents generally to reflect the population of English-, Spanish-, and Cantonese-speaking enrollees in fall 2001.⁸

The baseline survey instrument consisted of questions about health insurance coverage, health care use, access, and unmet need during the previous year; satisfaction with and opinions of health care; and current health status, employment, and sociodemographic information. Many of the survey questions were taken from national surveys such as the Community Tracking Study, the National Survey of America’s Families, and the Medical Expenditure Panel Survey.

Key Findings

Descriptive information about Alliance Family Care participants can help to either confirm or contradict concerns about expanding coverage to uninsured parents. As noted above, eligibility for this program requires people to be low-income parents of either publicly insured children or children who do not qualify for public programs. Contrary to the notion that such an expansion would invite only parents who are heavy users of care, we find that enrollees’ health care use is comparable to and self-reported health status is better than that of low-income populations in Alameda County. In addition, despite fairly low average household incomes, these enrollees are willing to pay a subsidized monthly premium for coverage.

Sociodemographic characteristics of enrollees. Alliance Family Care’s enrollment comprised a diverse group of people in 2001 (Exhibit 1), approximately half of whom are undocumented.⁹ Although there is substantial heterogeneity, the “typical” Alliance Family Care respondent is a married female, age thirty-four, born outside the United States, living in a five-person household with an annual household income of $23,000. Of those respondents working at the time of the interview, about 82 percent worked in the private sector and 16 percent were self-employed.¹⁰

Exhibit 1 provides comparison data from the 2001 California Health Interview Survey (CHIS) for nonelderly adults in Alameda County. Specifically, we present data on parents (with children under age eighteen) and on uninsured adults from households with incomes under 300 percent of poverty. Given that Alliance Family Care adult enrollees are uninsured parents earning less than 300 percent of poverty, the two comparison groups represent a combination of our sample. Unfortunately, we could not include CHIS comparison data on uninsured parents from households with incomes under 300 percent of poverty because the sample size for this group was too small.

Relative to other low-income groups from Alameda County, Alliance Family Care enrollees in 2001 had lower levels of income (controlling for household size) and education, were less likely to be employed, and were more likely to be female and married. They were also more likely to be Hispanic or Asian, which reflects several factors. First, there is a strong relationship between the Alliance and a few clinics that serve primarily Asians and Hispanics. Much of the enrollment into Alliance Family Care has occurred at these clinics. In addition, Alliance Family Care offers subsidized coverage for low-income families regardless of documentation status, and Asians and Hispanics represent a large portion of undocumented Californians.

Baseline health status, use, and access. Baseline measures of health status and use of services might be useful in gauging the potential costs of Alliance Family Care adult enrollees (Exhibit 2). The majority of adult enrollees used some health care services (usually a physician visit) in the year before they joined Alliance Family Care. Of the 56 percent with a physician visit, the average number was 3.8 visits during the year. Few enrollees had an emergency department visit (17 percent), with an average of 1.6 visits among those who did. Median out-of-pocket spending for health care in the previous year was $100–$299 among those with any utilization.

Although 84 percent of those who used services in the previous year reported having a usual source of care during that time, only 67 percent of Alliance Family Care adult enrollees overall reported having either a usual source of care or a regular doctor in the previous year.¹¹ Thirty-eight percent of respondents with utilization and a usual source reported that source as “other clinic or health center,” and 35 percent reported the doctor’s office as their source. Almost no one with utilization in the previous year reported the emergency department as their usual source of care.¹²

Seven percent of respondents reported unmet health care need, and 11 percent reported unmet prescription drug need in the past year. The most frequently reported reasons for unmet need were cost or lack of affordability and not having health insurance. Although unmet need was not statistically significantly correlated with income or insurance status, it was positively correlated with having any utilization in the last year, having a chronic or limiting condition, and having a usual source of care.

Alliance Family Care enrollees had better self-reported health than and health care utilization rates between those of the two CHIS comparison groups. If the few applicants who were denied enrollment because of medical underwriting (0.7 percent) instead would have been enrolled, average self-reported health status might have been lower. Nonetheless, even if all of these people reported poor health, the overall distribution of health status across categories would have changed only slightly, and
Family Care enrollees would still appear slightly healthier than the comparison groups. Utilization rates of Alliance Family Care enrollees were higher than those of uninsured adults under 300 percent of poverty in Alameda County in 2001. However, utilization rates among enrollees who were uninsured in the previous year were similar to those of uninsured low-income adults. The percentages of Alliance Family Care enrollees reporting unmet need and a usual source of care were similar to those of Alameda County low-income groups, with rates among enrollees in Alliance Family Care falling between the rates of these comparison groups.

Previous health insurance coverage. Nearly half of enrollees surveyed had insurance coverage for at least a portion of the year before they joined Family Care (Exhibit 3).¹³ Almost one-third had never had insurance prior to joining. Those with insurance in the previous year were covered for an average of 8.7 months. Overall, Asians and others were more likely than Hispanics to be insured for six months or more. People who had never had insurance prior to enrollment were more likely to have a household income under 100 percent of poverty (Exhibit 4).

Being unemployed at the time of the survey was positively correlated with having insurance in the previous year, which highlights the fact that being unemployed is likely to be related to other important factors. Of those who were employed, we found that employer size was related to whether the employer offered health insurance generally but was not related to whether the respondent was eligible for coverage.¹⁴ People who worked for the same employer for a year or more were more likely to have had insurance in the past twelve months. Offer and eligibility rates did not differ significantly by employment sector (that is, private versus public) but did differ by race and ethnicity.

Relative to those who were uninsured for the entire year prior to enrollment, those with some form of insurance coverage in the past year were much more likely to have had a hospital stay, emergency department visit, or physician visit in that time. Insured people were also more likely to have received one or more of the four preventive services we inquired about.

Compared to those in other categories of insurance status, people with insurance for six months or more in the previous year were significantly more likely to report having a usual source of care, whereas those who had never had insurance were significantly less likely to report having a usual source. Furthermore, insured people were more likely to report the doctor’s office as their usual source of care, whereas the uninsured were more likely to report “other clinic or health center” as their usual source. The few who reported the emergency department as their usual source of care were never insured prior to joining Alliance Family Care.

Although there were notable differences in some measures of access and utilization by previous insurance status, we also found that people who had never had health insurance, or had gone without health insurance for a year or more, were not significantly different, on several measures, from people who had recently had health insurance. Enrollees who had never had health insurance or were uninsured for at least a year were no more likely than those with insurance in the previous year to report fair or poor general or mental health status, a chronic or limiting condition, or unmet need.

The most frequently reported source of previous coverage was a person’s own employer or union, followed by Medi-Cal (California Medicaid) and privately purchased plans (Exhibit 5). Hispanics were the most likely and Asians the least likely to have had Medi-Cal as their source of coverage.

Discussion And Policy Implications

In fall 2001 the Alliance Family Care program attracted a group of parents who reported better health and comparable health care utilization and unmet need, relative to other low-income adults in Alameda County. Because of the outreach approach, most enrollees had some contact with the health care system, through themselves or their children. However, the majority used little medical care prior to enrollment. Given the baseline characteristics, we expect that the average enrollee will not be a high user of health care; we will investigate this hypothesis with future waves of survey data.

Despite having low household incomes, Alliance Family Care participants were willing to pay a premium—albeit a highly subsidized one—for coverage when they enrolled. Alliance Family Care has experienced a low disenrollment rate (about 2 percent per month), which indicates that most people are willing and able to pay these monthly premiums over time. Furthermore, the Alliance has a waiting list of more than 1,000 adults who would like to join Family Care.

About one-third of adult enrollees have never had health insurance before, and it is unclear how these enrollees will respond to coverage through Alliance Family Care. Having coverage in the year prior to enrollment had important implications for enrollees. These enrollees were more likely to report having used some health care and received various forms of preventive care in the previous year.

Although we know that nearly half of Alliance Family Care adult enrollees had insurance for part of the previous year, we do not have adequate data with which to ascertain whether the program has crowded out employer-sponsored insurance—that is, whether it led low-income adults to either drop employer coverage or change employment and lose that coverage.¹⁵ However, we can provide descriptive information that provides some sense of the scope of this issue.¹⁶ Among the 249 enrollees who were employed at the time of the survey, only 18 percent reported that they were eligible for health insurance through an employer. Among people who had coverage through their own employer in the previous year (n = 93), almost half were unemployed at the time of the survey. For these people, Alliance Family Care may be helping to fill a gap when employer insurance is no longer available or is not affordable under provisions of the Consolidated Omnibus Budget Reconciliation Act (COBRA). Furthermore, among those with employer coverage in the previous year, another 25 percent (n = 23) had been at their main job for nine months or less at the time of the survey, which indicates that these people had experienced a job transition in the recent past. Only twenty-one people (23 percent of those who had employer coverage the previous year) were employed and had been at their job for a year or more at the time of the survey. This group is the most likely source of crowding out; seventeen of the twenty-one people reported that they were eligible for employer coverage.

The results presented here are subject to a number of limitations. First, as with any survey of program participants, this information reflects self-reported data that are subject to recall bias and misreporting. There also may be selection issues with our data, in that they reflect only those who were contacted (English, Spanish, and Cantonese speakers) and willing to take the survey. However, concerns about selection are mitigated to some degree—at least among English, Spanish, and Cantonese speakers—by the high response rate attained.

The data also are based on a small sample, which limits to some degree our ability to detect statistically significant differences, especially in subgroup analyses. The data reflect the characteristics of participants in a single, local program that does not represent a state policy reform effort. Furthermore, although adult Alliance Family Care enrollees appear to be young and healthy, this program exists in a state with a relatively generous Medicaid program. Moreover, it includes a substantial number of undocumented people, a situation that may be unique to a handful of states and thereby limits the generalizability of our results. For these reasons, similar programs elsewhere could attract a different set of enrollees. Finally, given that we do not have spending data for the enrollees included in our sample, we use self-reported utilization data as a proxy for expenditures. However, utilization data are not a perfect proxy for expenditures, so we are reluctant to make inferences about program costs. Furthermore, the expenditures of those few applicants denied coverage through medical underwriting could be large, potentially accounting for a substantial percentage of overall program costs had they been enrolled in the program.¹⁷

We are interviewing the same Alliance Family Care participants again at six and twelve months after initial enrollment. Information from the second and third waves of the telephone survey will provide important information about changes (if any) in access to care, care-seeking behavior, and health care use over time. Furthermore, a comparison between continuous enrollees and disenrollees on measures like utilization and previous insurance coverage could reveal important information about the characteristics of those who stay versus those who leave this type of expansion program.

This research is funded by the California HealthCare Foundation and the Robert Wood Johnson Foundation. The authors thank Nina Maruyama and Irene Ibarra of the Alameda Alliance for Health and Deborah Zahn of the Robert Wood Johnson Foundation–funded Communities in Charge and the W.K. Kellogg Foundation–funded Community Voices projects for their cooperation in this study. They also thank two anonymous reviewers for their useful comments and Cathy Huang for helpful research assistance. Any errors in this research are attributable solely to the authors.

NOTES

1. Efforts to extend coverage to low-income uninsured adults have not kept pace with expansions for children, and more than one-third of children covered through Medicaid or SCHIP have a primary parent who is uninsured. See E. Howell et al., “Early Experience in Covering Uninsured Parents under SCHIP,” New Federalism Series A, no. A-51 (Washington: Urban Institute, May 2002); and A. Davidoff et al., “Patterns of Child-Pattern Insurance Coverage: Implications for Coverage Expansions,” New Federalism Series B, no. B-39 (Washington: Urban Institute, November 2001). Despite this gap, and the positive externalities and efficiencies associated with covering entire families instead of just children, only a handful of states have expanded coverage to parents through either Medicaid or SCHIP demonstrations.
2. The Alameda Alliance for Health provides health care coverage for low-income uninsured parents through Alliance Family Care.
3. These children may be enrolled in the Alliance through Medicaid or SCHIP (known as Medi-Cal and Healthy Families, respectively, in California) or simply through Alliance Family Care if they are ineligible for public coverage but have a household income under 300 percent of poverty. According to estimates from the Alameda Alliance for Health, almost 80 percent of adult Alliance Family Care enrollees had children enrolled in the Alliance through Medi-Cal or Healthy Families in July 2002.
4. We estimate that more than 70 percent of adults who responded to our survey pay a monthly premium of $30 or less for their own coverage.
5. Outreach for Alliance Family Care has not yet included a media campaign.
6. When more than one adult in a household enrolled in a given month, we selected the adult with the earliest birth month (regardless of the year of birth) for inclusion in the sample frame. When another member of the household enrolled in a later month, we included only the first enrolled adult in the sample.
7. The response rate was calculated as the number of completed interviews divided by the total eligible sample, which excludes enrollees with incorrect telephone numbers who could not be traced or telephone numbers not in service (n = 25), and enrollees who could not complete the survey in one of the three languages (n = 7).
8. According to the Alameda Health Consortium, English, Spanish, and Cantonese speakers represented approximately 88 percent of Alliance Family Care enrollees in 2000. Other languages of enrollees include Vietnamese and Mandarin.
9. The category “Asian” includes only English- and Cantonese-speaking Asians. Other Asians, such as those who spoke only Mandarin or Vietnamese, were not included in our sample. Non-Hispanic whites and non-Hispanic blacks, as well as those who self-identified as “other race,” were grouped together as “other” simply because of the small number of people in each group, not because of expected similarities across these groups.
10. Unfortunately, we do not have information on spouses’ employment status.
11. Because of an error in survey administration, about 15 percent of respondents, all Cantonese, were not asked whether they had a usual source of care or whether they had a regular doctor. We did not include these respondents in our calculations.
12. Some research classifies people who report the emergency department for their usual source as not having a usual source; see, for example, J.L. Hargraves, P.J. Cunningham, and R.G. Hughes, “Racial and Ethnic Differences in Access to Medical Care in Managed Care Plans,” Health Services Research 36, no. 5 (2001): 853–868. However, we include these people here, because our results highlight the fact that very few people reported the emergency department as their usual source.
13. Insurance coverage is defined as full or partial coverage for any type of medical care, including Medicaid and HMO plans.
14. Self-employed people were not asked about an employer’s offer of and eligibility for insurance. If respondents believed that having no offer of employer coverage was a condition for Alliance Family Care eligibility, this question could be subject to reporting bias.
15. Estimates of crowding out of employer coverage from previous research range from 17 percent to 50 percent of new Medicaid enrollment. See L. Dubay and G. Kenney, “Did Medicaid Expansions for Pregnant Women Crowd Out Private Coverage?” Health Affairs (Jan/Feb 1997): 185–193; and D. Cutler and J. Gruber, “Does Public Insurance Crowd Out Private Insurance?” Quarterly Journal of Economics 111, no. 2 (1996): 391–430.
16. Note that the figures presented reflect people with coverage through their own employer. Unfortunately, we have very little information on employer coverage through a spouse or other family member.
17. Recent estimates indicate that the top 1 percent of the population (in terms of expenditures) is responsible for 27 percent of health care costs. See M. Berk and A. Monheit, “The Concentration of Health Expenditures, Revisited,” Health Affairs (Mar/Apr 2001): 9–18. However, this estimate reflects the overall population, including newborns in the neonatal intensive care unit, elderly people, and the chronically ill. Although those parents denied coverage through Alliance Family Care might be chronically ill, they cannot be in the first group and are highly unlikely to be in the second. Furthermore, underwriting based on self-reported medical conditions cannot perfectly target the top 1 percent of health care users.


Erin Taylor (etaylor{at}mathematica-mpr.com) is a health researcher at Mathematica Policy Research in Washington, D.C., but completed this work while at the University of Michigan, in Ann Arbor. Jeffrey Kullgren is a medical student at the Michigan State University College of Human Medicine in East Lansing. Catherine McLaughlin is a professor in the University of Michigan School of Public Health.

Read a related paper by Melinda Beeuwkes Buntin et al.

10.1377/hlthaff.W3.460
©2003 Project HOPE–The People-to-People Health Foundation, Inc.