Exploring the Link Between Job Mobility, Health Insurance, and Health-related Financial Outcomes in People with Cystic Fibrosis

To the Editor:

Cystic fibrosis (CF) is a multisystem genetic disease that affects over 160,000 people worldwide (1). Over the last decade, outcomes for people with CF have dramatically improved (2, 3). This improvement is driven predominantly by the CF care model, which includes active monitoring through frequent clinical assessments (4–8), and by advances in pharmacotherapy, including cystic fibrosis transmembrane regulator (CFTR) modulators (9–12). Maintaining optimal health outcomes requires people with CF to frequently access a highly specialized healthcare system and use expensive pharmacotherapy. As such, the costs associated with CF care are high ($160,000 per person in 2022) (13–15), and adequate health insurance coverage for people with CF is paramount.

Nearly two-thirds (64%) of people with CF in the United States are now participating in the workforce (3) and thus acquire insurance coverage through their employers (16). As a result of this employer-sponsored insurance arrangement, individuals are at risk of experiencing “job lock,” or remaining in a job they would otherwise leave because of concerns about losing insurance coverage (17). Individuals (18–20) or caregivers (21) of those with chronic medical conditions such as CF may be at even greater risk of experiencing job lock, incentivized by high medical care cost and the need to consistently access the healthcare system.

Job lock has adverse impacts on the economy by increasing inefficiency (22); lowering productivity (22); and, at the individual level, by lowering life satisfaction (23). Little is known, however, about how job lock financially impacts people with CF directly. We assessed whether job lock was associated with adverse financial outcomes or unmet medical needs in a large cohort of people with CF. We hypothesized that people with CF experiencing job lock would have greater financial toxicity and delayed medical care than people with CF not experiencing job lock.

Methods

We performed a secondary data analysis of the cross-sectional, U.S.-administered 2019 CF Health Insurance survey. The survey used previously validated questions from U.S. national surveys to explore characteristics of health insurance, financial burden, and unmet medical needs in people with CF or caregivers of people with CF (24). CF-specific questions were created with feedback from the Cystic Fibrosis Foundation staff and advisers (24). The survey was distributed through CF community listservs, CF care centers, and social media from July to December 2019 (24). This survey was performed before the release of the CFTR modulator elexacaftor/tezacaftor/ivacaftor and the COVID-19 pandemic (24). Respondents with job lock were identified on the basis of a question asking “if the respondent had ever stayed in a job longer than they otherwise would to keep health insurance coverage.”

Responses were dichotomized on the basis of their reported job lock status. We performed adjusted multivariable logistic regression to evaluate the association of sociodemographic factors with job lock. We then examined for associations between job lock and other survey outcomes assessing financial toxicity (i.e., paying for healthcare-created financial problems, presence of medical debt), trouble paying for CF-related therapies, adverse financial complications because of medical bills (e.g., filed for bankruptcy, increased credit card debt, used up all or most savings), and unmet medical needs (e.g., skipped CF care visits, did not receive test/treatment recommended by CF provider, delayed filling prescriptions). The measures of adverse financial outcomes were evaluated in pooled responses by category (paying for healthcare-created financial problems, delays in medical care, adverse financial complications) and by each question individually. All logistic regression models were adjusted for age category, health insurance type (i.e., private insurance with and without a high deductible health plan [HDHP], Medicaid/Children’s Health Insurance Program [CHIP], Medicare, other), respondent-reported sex, U.S. Census region, categorized household income as a percentage of federal poverty level (% FPL), and categorized residence in an urban/suburban/rural location. Reference groups were selected on the basis of the group anticipated to have the lowest reporting of job lock. For example, individuals with income, 100% FPL would qualify for government insurance programs, and individuals aged 18–25 years would have the option to remain on a parent’s insurance plan. All statistical analysis was performed using Stata 18 (StataCorp LLC). This study was deemed non–human subject research by the Boston Children’s Hospital Institutional Review Board.

Results

A total of 1,284 respondents (57% people with CF, 41% parents of persons with CF, and 2% partners of persons with CF) completed the survey. A majority (61%) of respondents endorsed having a private insurance plan, and 21% reported being covered by Medicaid/CHIP/state health insurance. Fifty-two percent of respondents had a household income that was 400% or greater than the federal poverty level. Other demographic characteristics of the sample are included in Table 1.

Table 1.

Demographic characteristics of respondents stratified by reported job lock status

	n (%)
	No Job Lock (n = 723)	Job Lock* (n = 557)	OR (95% CI)†	P Value
Job lock
Entire cohort	723 (56.5)	557 (43.5)	—	—
Person with CF	391 (53.6)	332 (46.4)	—	—
Parent of person with CF	317 (59.9)	212 (40.1)	—	—
Spouse/partner of person with CF	15 (68.2)	7 (31.8)	—	—
Age 17 yr and younger	281 (62.6)	168 (37.4)	—	—
Age 18 yr and older	442 (53.2)	389 (46.8)	—	—
Age group, yr
0–17	281 (62.6)	168 (37.4)	0.91 (0.64–1.31)	0.63
18–25	122 (60.4)	80 (39.6)	Reference	NA
26–35	149 (51.4)	141 (48.6)	1.35 (0.92–2.00)	0.13
36–64	149 (51.4)	162 (51.4)	1.49 (1.01–2.21)	0.046
65 and older	18 (75.0)	6 (25.0)	0.54 (0.19–1.53)	0.25
Health insurance group
Private insurance (low deductible)‡	153 (54.1)	130 (45.9)	1.70 (1.14–2.52)	0.009
Private insurance (high deductible)§	189 (46.3)	219 (53.7)	2.32 (1.61–3.35)	<0.001
Private insurance (unknown deductible)||	58 (65.9)	30 (34.1)	1.15 (0.67–1.97)	0.62
Medicaid/CHIP/state insurance	186 (70.5)	78 (29.6)	Reference	NA
Medicare	88 (57.1)	66 (42.9)	1.48 (0.91–2.39)	0.11
Other¶	49 (59.0)	34 (41.0)	1.36 (0.79–2.34)	0.26
Patient-reported sex
Male	297 (57.5)	220 (42.6)	Reference	NA
Female	421 (55.9)	332 (44.1)	1.06 (0.83–1.35)	0.64
Other	0	2 (100)	NA	NA
U.S. Census region
Northeast	122 (60.7)	79 (39.3)	Reference	NA
Midwest	219 (55.6)	175 (44.4)	1.18 (0.82–1.70)	0.37
South	253 (54.1)	215 (45.9)	1.19 (0.83–1.69)	0.34
West	129 (59.5)	88 (40.6)	1.00 (0.66–1.52)	0.99
Residential category**
Urban	140 (56.7)	107 (43.3)	Reference	NA
Suburban	260 (54.5)	217 (45.5)	1.04 (0.75–1.44)	0.81
Small town	242 (59.3)	166 (40.7)	0.97 (0.69–1.36)	0.85
Rural	81 (54.7)	67 (45.3)	1.16 (0.75–1.78)	0.51
Income, % federal poverty level
<100%	61 (72.6)	23 (27.4)	Reference	NA
101–199%	99 (62.3)	60 (37.7)	1.48 (0.82–2.69)	0.2
200–399%	193 (51.2)	184 (48.8)	1.89 (1.08–3.28)	0.025
400% or greater	370 (56.1)	290 (43.9)	1.40 (0.81–2.43)	0.23

Definition of abbreviations: CF = cystic fibrosis; CHIP = Children’s Health Insurance Program; CI = confidence interval; NA = not applicable.

^*Defined as answering yes to question asking if the respondent stayed in a job longer than they otherwise would have in order to keep health insurance coverage.

^†Obtained from logistic regression model, including age, health insurance, sex, U.S. Census region, residential category, and income.

^‡Defined by answering less than $1,500 to a question asking about annual individual deductible.

^§Defined by answering greater than or equal to $1,500 to a question about annual individual deductible.

^||Defined by answering "do not know” to a question asking about annual individual deductible.

^¶Includes those who reported having dual eligibility for Medicaid and Medicare, TRICARE, or were uninsured.

^**Self-reported by answering which best describes the place where you now live.

Job lock was reported by 44% (n = 557) of respondents. The proportion of people with CF who reported job lock (46%; n = 332) was greater than parents of people with CF (40%; n = 212). The adjusted odds ratio (aOR) of experiencing job lock in all respondents was greater for respondents between ages 36 and 64 years (aOR, 1.49; 95% confidence interval [CI], 1.01–2.21; referent 18–25 yr), those privately ensured with and without HDHP (with HDHP aOR, 2.32; 95% CI, 1.61–3.35; without HDHP aOR, 1.70; 95% CI, 1.14–2.42; referent Medicaid/CHIP/state health insurance), and those with a household income between 200% and 399% FPL (aOR, 1.89; 95% CI, 1.08–3.28; referent <100% FPL) (Table 1).

The aOR of reporting any financial toxicity was greater among respondents who experienced job lock (Table 2). Notably, those with job lock had a greater aOR of having medical debt (aOR, 1.69; 95% CI, 1.33–2.15; referent no job lock) and reporting that paying for care created financial problems (aOR, 2.74; 95% CI, 2.11–3.55; referent no job lock).

Table 2.

Adjusted odds of experiencing financial toxicity, delays in medical care, problems paying for treatment, or adverse financial complications among those with job lock compared with those without job lock

	aOR*	95% CI	n
Experienced financial toxicity
Presence of medical debt	1.69	1.33–2.15	1,268
Paying for healthcare created financial problems	2.74	2.11–3.55	1,211
Cost of care caused you to delay any services/care/treatment in last 12 mo (composite)†	2.41	1.87–3.11	1,270
Took less medications	2.19	1.48–3.25	1,266
Delayed filling a prescription	2.10	1.57–2.80	1,262
Skipped one of your routine CF care center visits	1.92	1.28–2.90	1,243
Delayed a CF care center visit when you didn’t feel well	2.83	1.88–4.26	1,245
Didn’t get a medical test/treatment that was recommend by a doctor	2.37	1.60–3.50	1,244
Experienced problems paying medical bills related to any service/treatment in last 12 mo (composite)‡	1.93	1.52–2.46	1,270
Cystic fibrosis transmembrane regulator modulators	2.08	1.32–3.28	759
Airway clearance devices (vest, PEP, Acapella, Flutter devices)	1.75	1.24–2.47	1,116
CF routine visit	2.01	1.43–2.82	1,240
CF sick visit	2.90	1.82–4.60	965
Hospital (inpatient) intravenous treatment for pulmonary exacerbations	1.92	1.22–3.04	599
Ever faced with any issue because of medical bills (composite)§	3.25	2.43–4.35	1,270
Been contacted by a collection agency	2.83	2.09–3.84	855
Increased credit card debt	1.97	1.55–2.50	1,253
Took out another type of loan	2.05	1.41 –2.96	1,238
Took money out of retirement, college, or other long-term savings accounts	3.02	2.32–3.95	1,239
Used up all or most of savings	2.21	1.73–2.84	1,232
Filed for bankruptcy	2.07	1.13–3.80	794
Had difficulty paying for food	3.21	2.05–5.01	808
Had difficulty paying for rent/mortgage payment or utilities	2.87	2.13–3.85	1,248
Forced to move or evicted	2.65	1.30–5.41	819
Put off major purchases such as a new home, car, or vacations	3.27	2.53–4.21	1,243

Definition of abbreviations: aOR = adjusted odds ratio; CF = cystic fibrosis; CI = confidence interval; PEP = positive expiratory pressure.

^*Obtained from logistic regression model including age, health insurance, sex, U.S. Census region, residential category, and income.

^†Composite variable identified by 11 questions addressing delayed or forgone CF care.

^‡Composite variable identified by 23 questions addressing problems paying for CF-specific medications and treatments.

^§Composite variable identified by 12 questions addressing adverse financial complications.

Job lock was also associated with differences in accessing and paying for care. Those who experienced job lock had a greater aOR of reporting delaying care due to cost (aOR, 2.41; 95% CI, 1.87–3.11; referent no job lock) and endorsing problems paying for CF medications or treatments because of cost (aOR, 1.93; 95% CI, 1.52–2.46; referent no job lock) (Table 2). Specifically, those reporting job lock were more likely to delay filling a prescription (aOR, 2.10; 95% CI, 1.57–2.80), to delay a CF care center visit when not feeling well (aOR, 2.83; 95% CI, 1.88–4.26), and to experience problems paying for CFTR modulators (aOR, 2.08; 95% CI, 1.32–3.28) and CF sick visits (aOR, 2.90; 95% CI, 1.82–4.60). Individuals with job lock also had a greater aOR of experiencing any adverse financial complication because of medical bills (aOR, 3.22; 95% CI, 2.40–4.30) (Table 2). Notably, compared with those without job lock, those who reported job lock were more likely to increase credit card debt (aOR, 1.97; 95% CI, 1.55–2.50); to take money out of retirement, college, or long-term savings accounts (aOR, 3.02; 95% CI, 2.32–3.95); to be forced to move or evicted (aOR, 2.65; 95% CI, 1.30–5.41); to have difficulty paying rent/mortgage or utilities (aOR, 2.87; 95% CI, 2.13–3.85); and to report difficulty paying for food (aOR, 3.21; 95% CI, 2.05–5.01).

Discussion

In this secondary data analysis of a large, representative U.S. CF population, we found that nearly half of respondents endorsed job lock, with more people with CF than parents of children with CF reporting job lock. Job lock was associated with reporting limitations in the ability of people with CF to access and pay for care. Notably, job lock was also associated with food and housing insecurity and adverse health-related financial outcomes even after controlling for overall income.

In a recent study using pooled 2016–2022 National Survey of Children’s Health data, caregivers of children with CF (n = 175) in the previous 12 months had a greater aOR of endorsing job lock (aOR, 3.43), spending ⩾$1,000 on medical care (aOR, 2.25) and self-reporting problems paying for medical or health care (aOR, 3.20) when compared with families without children with CF (n = 278,654) (25). Notably, our study found that job lock prevalence was higher for adults with CF than for caregivers of people with CF and also revealed that people who experienced job lock were at even greater risk of developing financial challenges than people without job lock. This is important, given the increasing workforce participation of people with CF as effective treatments have improved health and extended lifespan.

Of greater concern is our finding that those who experienced job lock reported greater challenges in accessing and paying for care. Delays in accessing medical care or CFTR modulator medications would increase the risk of developing pulmonary exacerbations and requiring hospitalization. These are known drivers of CF cost and expense for individuals with CF and the health system (14).

As the population of people with CF ages and increasingly participates in the workforce, the ability to participate in other age-congruent activities is likely impaired in those reporting job lock. For example, activities such as increasing credit card debt or using all long-term savings impair the ability to generate wealth or save for retirement and likely heighten the challenges associated with job lock. Furthermore, this may have adverse impacts on life satisfaction, which increase the risk of developing anxiety or depression, a comorbidity known to negatively impact CF health outcomes (23, 26).

Our study found that people with CF insured through private HDHPs had greater odds of experiencing job lock than those insured through lower-deductible private plans. Thirty percent of workers with employer-sponsored plans have HDHPs with linked savings accounts, which have the highest deductibles of any other plan type (16). As more people with CF join the workforce, knowledge of insurance design and outcomes will be important to inform insurance selection. Potential solutions include selection of a lower-deductible plan in which choices of employer-sponsored plans are available and employers could consider offering a lower-deductible health plan.

Our study has several limitations. Because respondents were asked about financial and health experiences in the prior year, our survey results could be subject to recall bias. The survey was administered before the widespread introduction of the highly effective CFTR modulator elexacaftor-tezacaftor-ivacaftor, the COVID-19 pandemic, and general hyperinflationary pressures. It is likely that the introduction of very expensive, highly effective treatments may exacerbate the adverse financial outcomes observed in our study. Finally, the job lock question in the survey asked if the respondent had ever experienced job lock, whereas several of the health-related financial outcomes asked if events were experienced over the last 12 months. Future surveys should ask the job lock question in the same time horizon using the last 12 months as framing so that researchers can track job lock trends longitudinally.

In summary, our analysis reveals the high prevalence of job lock within a representative CF population and the adverse impacts on financial outcomes and the ability to access clinical care. It is unclear how job lock impacts clinical outcomes, and future studies should evaluate for disparities in clinical outcomes or healthcare use. Monitoring job lock longitudinally will be important to ensure that financial burden does not grow or access to care does not worsen as more people with CF enter the workforce.