The unadjusted frequency of cost responses in the overall population weighted for sampling proportions. Limit lines indicate 95% confidence intervals. ED indicates emergency department.
Fung V, Graetz I, Galbraith A, Hamity C, Huang J, Vollmer WM, Hsu J, Wu AC. Financial Barriers to Care Among Low-Income Children With AsthmaHealth Care Reform Implications. JAMA Pediatr. 2014;168(7):649–656. doi:10.1001/jamapediatrics.2014.79
Copyright 2014 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
The Patient Protection and Affordable Care Act (ACA) includes subsidies that reduce patient cost sharing for low-income families. Limited information on the effects of cost sharing among children is available to guide these efforts.
To examine the associations between cost sharing, income, and care seeking and financial stress among children with asthma.
Design, Setting, and Participants
A telephone survey in 2012 about experiences during the prior year within an integrated health care delivery system. Respondents included 769 parents of children aged 4 to 11 years with asthma. Of these, 25.9% of children received public subsidies; 21.7% were commercially insured with household incomes at or below 250% of the federal poverty level (FPL) and 18.2% had higher cost-sharing levels for all services (eg, ≥$75 for emergency department visits). We classified children with asthma based on (1) current receipt of a subsidy (ie, Medicaid or Children’s Health Insurance Program) or potential eligibility for ACA low-income cost sharing or premium subsidies in 2014 (ie, income ≤250%, 251%-400%, or >400% of the FPL) and (2) cost-sharing levels for prescription drugs, office visits, and emergency department visits. We examined the frequency of changes in care seeking and financial stress due to asthma care costs across these groups using logistic regression, adjusted for patient/family characteristics.
Main Outcomes and Measures
Switching to cheaper asthma drugs, using less medication than prescribed, delaying/avoiding any office or emergency department visits, and financial stress (eg, cutting back on necessities) because of the costs of asthma care.
After adjustment, parents at or below 250% of the FPL with lower vs higher cost-sharing levels were less likely to delay or avoid taking their children to a physician’s office visit (3.8% vs 31.6%; odds ratio, 0.07 [95% CI, 0.01-0.39]) and the emergency department (1.2% vs 19.4%; 0.05 [0.01-0.25]) because of cost; higher-income parents and those whose children were receiving public subsidies (eg, Medicaid) were also less likely to forego their children’s care than parents at or below 250% of the FPL with higher cost-sharing levels. Overall, 15.6% of parents borrowed money or cut back on necessities to pay for their children’s asthma care.
Conclusions and Relevance
Cost-related barriers to care among children with asthma were concentrated among low-income families with higher cost-sharing levels. The ACA’s low-income subsidies could reduce these barriers for many families, but millions of dependents for whom employer-sponsored family coverage is unaffordable could remain at risk for cost-related problems because of ACA subsidy eligibility rules.
Paying for health care is challenging. Most private insurance plans require individuals to pay a portion of their total health care costs out-of-pocket (cost sharing) to reduce spending.1 Higher levels of cost sharing have been shown to reduce the use of unnecessary and necessary care; the latter may be especially concerning for patients with chronic conditions or lower incomes.2- 7 At present, Medicaid and Children’s Health Insurance Programs (CHIP) subsidize coverage for the lowest-income children and include limited or no cost sharing. Starting in 2014, the Patient Protection and Affordable Care Act (ACA)8 expands Medicaid eligibility and increases the availability of private coverage through new state health insurance marketplaces (exchanges). Premium subsidies will be available to families with incomes at or below 400% of the federal poverty level (FPL) and cost-sharing subsidies to families at or below 250% of the FPL who purchase insurance through the marketplaces.
These expansions in coverage and subsidies could be especially valuable for children with chronic conditions such as asthma who have greater health care costs.9- 12 Studies examining the effects of expanding subsidized coverage to poor children, such as through Medicaid or CHIP, have found improvements in access to care.13- 17 Less evidence is available on the effects of cost sharing on care seeking among children covered by commercial insurance.18 One recent study19 of commercially insured children with asthma found that higher cost sharing for drugs was associated with modest reductions in the use of medication and increases in asthma-related hospitalizations for older children. However, limited evidence exists on responses to cost sharing for care other than drugs and on variations across income levels to inform low-income subsidies mandated by the ACA.20
We examined the effects of cost sharing on adherence to medication regimens, visits to a physician’s office or an emergency department (ED), and financial stress among children with asthma. We compared children receiving public subsidies (Medicaid or CHIP) with those covered by commercial insurance, stratified by income levels that correspond to ACA subsidy eligibility at or below 250% (cost sharing and premium), 251% to 400% (premium only), and more than 400% (no subsidies) of the FPL. We further examined variations among children with higher vs lower cost-sharing levels and hypothesized that responses to cost sharing would be concentrated among lower-income children (≤250% of the FPL) with higher levels of cost sharing who were not receiving public subsidies.
This study was conducted in Kaiser Permanente Northern California, an integrated health care delivery system providing comprehensive medical care to more than 3 million members. Our target population included members aged 4 to 11 years on January 1, 2011, with at least 1 diagnosis of asthma (code 493.xx from the International Classification of Diseases, Ninth Revision, Clinical Modification) in 2004 through 2008 and at least 1 controller medication dispensed in 2011 (ie, an inhaled corticosteroid, a leukotriene receptor agonist, or an inhaled corticosteroid plus a long-acting β-agonist). We excluded children receiving omalizumab or long-term oral corticosteroid therapy (a supply for ≥180 continuous days) before the survey or with diagnoses of cystic fibrosis, bronchiectasis, pulmonary hypertension, pulmonary embolism, immunodeficiency, hereditary and degenerative diseases, psychoses, and mental retardation.
To identify children with more severe asthma, half of the sample was selected to have 1 or more asthma-related hospitalization or ED visit in 2011. These participants were matched to children without events by age (±1 year) and sex. All results are weighted by the sampling fraction, accounting for the matched design and nonresponse to represent the target population.
Starting in April 2012, we mailed 1400 parents of children in our sample an introductory letter and questionnaire. Recipients could decline participation via the reply postcard or telephone or could return the completed questionnaire by mail (implied consent). If no response was received after 3 weeks, trained interviewers called parents by telephone, obtained verbal consent, and conducted the interview. Participants received a $25 gift card. The Kaiser Permanente Northern California Institutional Review Board approved the study.
Parents were ineligible to participate if they could not complete the survey in English, were no longer Kaiser Permanente Northern California members, reported that their child did not have asthma, or had incorrect contact information (n = 96). Among eligible parents, 161 declined to participate and 374 were unreachable after at least 10 attempts by telephone. In sum, 769 parents completed the surveys (59.0% response rate). Parents of children with recent asthma-related events were more likely to respond than those without (57.7% vs 52.1%; P = .03); other differences between responders and nonresponders were not significant.
We asked parents whether, in the last 12 months, they switched their child’s asthma medication to a cheaper (eg, generic) drug or their child used less medication than prescribed (eg, by skipping doses, by not filling or refilling a prescription) because of the amount they had to pay for asthma medications. We also asked if they delayed or avoided taking their child to any physician’s office visit or to the ED because of the amount they had to pay. To assess financial stress, we asked whether parents borrowed money from a friend/family member, credit card, or bank or cut back on necessities like food, rent, or other basics in the last 12 months because of the cost of their child’s asthma care.
We obtained children’s levels of cost sharing for prescription drugs and office and ED visits from membership data as of December 2011. We focused on cost sharing for brand-name drugs because most asthma controller medications are available only as such. We defined a dichotomous variable of a higher cost-sharing level if copayments were at least $25 for brand drugs, at least $20 for office visits, and at least $75 for ED visits or if the participant had a deductible for the service (3.6% had a deductible for all services; 9.9%, for hospital and ED visits; and 3.0%, for prescription drugs). We selected these thresholds because they represented approximately the top quartile in cost sharing; sensitivity analyses examining different thresholds yielded similar findings.
We obtained Medicaid and CHIP enrollment from membership data as of December 2011; in California, the income limits for children and adolescents aged 6 to 18 years were 100% of the FPL for Medicaid and 250% of the FPL for CHIP. For commercially insured children, we classified income into categories that corresponded with the ACA limits for cost sharing (≤250% of the FPL) and premium subsidies (≤400% of the FPL) using self-reported household income. We asked parents to report annual income for the household where their child lives, including income from all sources in $10 000 increments to $100 000 or more. We used the midpoint of each category ($150 000 for the highest category) and the number of persons living in the household to calculate income as a percentage of the FPL using the 2012 Health & Human Services Poverty Guidelines (eg, $23 050 for a family of four).21 Income was missing for 16 participants who were coded as having an unknown income.
To examine the associations among responses to costs, cost-sharing levels, and income/subsidy status, we used multivariate logistic regression (svy: logistic in STATA/MP, version 12.0; StataCorp). For each outcome, we used the relevant cost-sharing level (eg, cost sharing for brand-name drugs for medication-related responses). To examine financial stress outcomes, we included an indicator for whether subjects had high levels of cost sharing for all services.
To examine whether cost-sharing levels influenced the associations between income and responses to costs, we stratified unsubsidized participants into the following higher and lower cost-sharing groups resulting in the following 8 comparison groups: Medicaid enrollees; CHIP enrollees; and commercially insured children at or below 250% (reference group), 251% to 400%, and more than 400% of the FPL with higher cost-sharing levels (reference group); and commercially insured children at or below 250%, 251% to 400%, and more than 400% of the FPL with lower cost-sharing levels.
We calculated the adjusted percentages of participants in each group reporting each outcome based on the model results. The standard population used in the direct adjustment procedure was the mix of covariate values in the target population.
We adjusted for age (7-9 or 10-11 vs 4-6 years), sex, race/ethnicity (white vs nonwhite), and parental educational level (high school degree or less vs some college or more). To adjust for disease severity, we included indicators for having an asthma-related hospitalization or an ED visit in 2011 based on health system records and for receiving any combination or multidrug asthma regimen from the survey.
About one-quarter of the population was enrolled in Medicaid (9.1%) or CHIP (16.8%); the remainder were commercially insured, with 21.7% of the study population with household income at or below 250% of the FPL; 14.0%, 251% to 400% of the FPL; and 38.4%, more than 400% of the FPL (Table 1). Medicaid enrollees had no cost sharing, whereas CHIP enrollees had modest levels of cost sharing (eg, $0-$15). Overall, 18.2% had higher levels of cost sharing for all services (Table 2). Among the commercially insured children, 85.2% were covered by medium or large employer-sponsored plans; the remaining were covered by plans purchased by the families or sponsored by small employers.
Overall, 2.6% of parents reported that they switched their child to a cheaper asthma medication, and 9.3% used less medication than prescribed because of costs (Figure). In addition, 7.6% of parents delayed or avoided taking their child to any physician’s office visit and 5.3% to any ED visit because of costs. Among those who reduced medication use, 41.1% of parents reported that it affected their child’s asthma care or overall control. Among those who delayed or avoided any office or ED visit, 38.1% and 27.4%, respectively, reported that it affected their child’s asthma care or control. Overall, 15.6% of parents reported financial stress because of the costs of their child’s asthma care, including borrowing money (10.8%) or cutting back on necessities (10.2%).
Cost-related responses were largely concentrated among commercially insured children with household incomes at or below 250% of the FPL and higher cost-sharing levels (Table 3). After adjusting for covariates, parents with incomes at or below 250% of the FPL with lower vs higher cost-sharing levels were less likely to delay taking their child to an office visit (3.8% vs 31.6%; odds ratio [OR], 0.07 [95% CI, 0.01-0.39]) or an ED visit (1.2% vs 19.4%; 0.05 [0.01-0.25]). Parents of children enrolled in Medicaid or CHIP were also less likely to delay their children’s care compared with parents of commercially insured children with incomes at or below 250% of the FPL with higher cost-sharing levels. For example, 6.9% of parents of children enrolled in Medicaid delayed/avoided taking their child to an office visit (OR, 0.14 [95% CI, 0.03-0.58]). Similarly, parents with higher income levels were less likely to have delayed their children’s care than those with incomes at or below 250% of the FPL. We found no significant differences by income and cost-sharing level in cost-related use of medication behaviors.
Among families with incomes at or below 250% of the FPL, after adjustment, 33.3% of those with a higher level of cost sharing borrowed money to pay for their children’s asthma care (Table 4); in comparison, families with similar incomes but a lower cost-sharing level were less likely to borrow money (12.4%; OR, 0.26 [95% CI, 0.07-0.95]). Parents of children in CHIP were also less likely to borrow money (10.8%; OR, 0.22 [95% CI, 0.05-0.91]), as were higher-income families (eg, 8.0% of parents at >400% of the FPL with higher cost-sharing levels; OR, 0.15 [95% CI, 0.02-0.96]).
We investigated the effect of out-of-pocket costs on care-seeking behaviors for children with asthma. We found that families with incomes at or below 250% of the FPL with higher cost-sharing levels frequently delayed or avoided their children’s care because of costs or borrowed money to pay for their children’s care. Higher-income families and low-income families with lower levels of cost sharing through existing subsidies (Medicaid and CHIP) were less likely to report these behaviors. The insurance subsidies mandated by the ACA starting in 2014 could provide important protection for eligible low-income children with chronic medical needs from cost-related barriers to care and financial stress.
For families purchasing insurance through state marketplaces, the ACA limits their premium contributions to a range of 2% (<133% of the FPL) to 9.5% (300%-400% of the FPL) of income for the second lowest-cost Silver plan in an area. Silver plans are the third of 4 levels of coverage available within the marketplaces (including Platinum, Gold, Silver, and Bronze) and are required to cover 70% of medical costs (actuarial value). The ACA cost-sharing subsidies increase the actuarial value of a Silver plan from 70% to 94% (100%-150% of the FPL), 70% to 87% (150%-200% of the FPL), and 70% to 73% (200%-250% of the FPL); thus, families with lower incomes pay a smaller percentage of their total costs.22
The ACA also funds CHIP through 2015 and provides states the option to expand Medicaid eligibility to 138% of the FPL for all eligible individuals. Twenty-five states decided not to expand their Medicaid eligibility in 201423; however, because all but 2 states cover children to 200% of the FPL or higher through CHIP, these decisions have a more limited effect on children.24,25 In 26 states plus the District of Columbia, CHIP covers all children to 250% of the FPL or higher25; for children living in the remaining states, the availability of cost-sharing subsidies for insurance purchased in state marketplaces is likely to be of greater importance.
Whether these cost-sharing subsidies, particularly the modest subsidies for those at 200% to 250% of the FPL, are sufficient to reduce financial barriers to care is unknown. Notably, the mean benefit generosity of plans in this study, even among those with higher levels of cost sharing, is likely to be greater than that of the Silver plans available in state marketplaces. For example, the standard benefit for a Silver plan in California includes copayments of $45 for primary care visits, $250 for ED visits, and $50 for preferred brand-name drugs, with additional deductibles ($2000 for medical expenses and $250 for brand-name drugs).26 Moreover, cost-sharing subsidies will not be available to families that do not enroll in Silver plans, including those choosing Bronze plans with lower premiums and higher levels of cost sharing.27 Whether families are aware of the details of these provisions remains unknown. Existing surveys suggest that even basic knowledge about the ACA is limited,28 raising concerns that some families may mistakenly miss available subsidies.
Many states also have crowd-out provisions to limit potential substitution of CHIP for private insurance coverage; for example, in California, children must not have had employer-sponsored insurance in the past 3 months. The ACA subsidies are also not available to families with access to affordable employer-sponsored coverage. Current interpretation of the law defines insurance as affordable if the employee’s cost for individual coverage is less than 9.5% of household income. Of concern, it does not apply to the cost of purchasing family coverage; the mean cost of a family plan was more than 4-fold higher than the cost of individual coverage in 2012.29,30
This “family glitch” could result in a substantial number of low-income families facing unaffordable employer-sponsored coverage, with no option to obtain subsidies in state health insurance marketplaces. An estimated 3.9 million dependents in 2008 would have been ineligible for subsidies because the cost of individual but not family coverage was less than 9.5% of household income.31 The US Government Accountability Office estimates that 460 000 uninsured children in 2009 would be ineligible for subsidies because of this rule.32 Although the administration proposed to exempt uninsured dependents from penalties for noncompliance with the individual mandate if the cost of family coverage is more than 8% of income, other plans to address the glitch are lacking.33
These gaps in subsidy availability raise concerns about continued financial barriers to care for many low-income children. In our study, largely among families with employer-sponsored coverage, nearly 1 in 3 parents with incomes at or below 250% of the FPL with higher levels of cost sharing delayed or avoided taking their child to an office visit, and nearly 1 in 5 delayed an ED visit, because of costs. Cost-related reductions in care have been linked with adverse clinical effects.6,19,34- 37
About one-third of the parents in this study reported that foregoing office or ED visits because of costs affected their child’s overall asthma care or control. Although some avoided visits might not have been essential for care management, other studies have found that parents tend to overestimate their children’s asthma control compared with their children’s self-assessments.38- 40 In addition, we did not assess longer-term effects or effects outside of asthma care. These findings highlight the importance of incorporating discussions about costs into clinical encounters, even among commercially insured children, when assessing potential barriers to care.41- 45
This study was conducted within an integrated health care delivery system. Commercially insured children in this study had more generous benefits, on average, than those offered by employer plans nationally29; thus, this study could represent conservative estimates of the frequency of these behaviors more broadly. In this setting, a high proportion of prescribing is for formulary-preferred drugs; thus, opportunities to switch to lower-cost and potentially more cost-effective drugs is likely to be more limited than in less integrated settings. The low rates of switching could also be because of the limited generic asthma controller medication options and switches that occurred before the study. Our measures of financial stress captured only actual behaviors that could be more difficult for lower-income families, such as borrowing money; the measures did not assess the perceived need to borrow money. Moreover, these measures did not include potential indirect costs, such as parental work or absence of a child from school, which can be substantial in pediatric asthma.46
This study excluded non–English-speaking respondents, who could face greater difficulties navigating responses to costs. The study also excluded children with very severe asthma (eg, long-term users of oral corticosteroids) and severe comorbidities (eg, immunodeficiency), who likely have greater health care needs and costs; whether they would be more or less sensitive to cost-sharing is unclear. Income was collected via self-report, which could be subject to reporting error.47 More objective individual-level measures of household income, however, are not available in administrative data or widely available for research. We also have limited ability to identify children who were eligible for public subsidies but not enrolled; however, Medicaid and CHIP take-up rates in California were relatively high (eg, 84.8% in 2010).48 Consistent with this statistic, we found few participants (n = 3) with incomes at or below 250% of the FPL likely to be eligible for Medicaid or CHIP (ie, no employer-sponsored insurance), and sensitivity analyses excluding these subjects from the group at or below 250% of the FPL yielded similar findings. Last, this was a cross-sectional survey and there could be unmeasured differences across the income/subsidy or cost-sharing groups; however, we controlled for a wide range of sociodemographic and clinical characteristics, including asthma severity, using a combination of administrative and survey data.
We found that delaying and avoiding health care because of costs was concentrated among commercially insured children with higher levels of cost sharing and household incomes at or below 250% of the FPL. The ACA will expand cost-sharing subsidies to families with incomes at or below 250% of the FPL, which could reduce cost-related barriers to care, especially for families with children with chronic conditions like asthma and living in states with lower income eligibility limits for CHIP. For families at 200% to 250% of the FPL, however, these cost-sharing subsidies will be modest. Moreover, because of a family glitch, these subsidies will not be available to millions of dependents for whom employer-sponsored family coverage could be unaffordable. Work is needed to evaluate the effects of the ACA and potential unintended gaps in subsidy access to inform ongoing policy refinements.
Accepted for Publication: January 6, 2014.
Corresponding Author: Vicki Fung, PhD, Mongan Institute for Health Policy, Massachusetts General Hospital, 50 Staniford St, Ninth Floor, Boston, MA 02114 (email@example.com).
Published Online: May 19, 2014. doi:10.1001/jamapediatrics.2014.79.
Author Contributions: Drs Fung and Graetz had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Fung, Galbraith, Hsu.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Fung, Hamity, Wu.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Fung, Graetz, Huang, Vollmer.
Obtained funding: Wu.
Administrative, technical, or material support: Galbraith, Hamity, Huang, Vollmer, Hsu, Wu.
Study supervision: Fung, Wu.
Conflict of Interest Disclosures: Dr Fung has a financial interest in Merck. Merck produces a leukotriene inhibitor, which is one of the asthma regimens assessed in the Population-Based Effectiveness in Asthma and Lung Diseases (PEAL) study. Dr Fung’s interests were reviewed and are managed by Massachusetts General Hospital and Partners HealthCare in accordance with their conflict-of-interest policies. No other disclosures were reported.
Funding/Support: This study was funded by grant R01 HS019669-01 from the Agency for Healthcare Research and Quality.
Role of the Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: The other members of the PEAL Network study team and the PEAL Network National Advisory Committee provided valuable input and advice.