Sage Journals: Discover world-class research

Abstract

Objectives:

Financial strain has important consequences for patients, providers, and health care systems. However, there is currently no gold standard measure to screen for financial strain. This study compared the performance of 3 single-item screeners using a composite measure of financial strain as a “gold standard.”

Methods:

We conducted a secondary analysis of unweighted data from a 2021 survey of Kaiser Permanente Northern California health plan members comparing the percentages of adults who experienced financial strain based on 3 general single-item screeners, a screener specific to medical and dental health care use, and a composite financial strain measure. The study sample was comprised of 2734 non-Medicaid insured adults who answered all financial strain questions. Kappa statistics evaluating agreement of the 3 general screeners with the composite measure were calculated for the sample overall, by age group, and within age group, by 4 levels of income and 4 racial/ethnic subgroups.

Results:

Among 947 adults aged 35 to 65, 30.7% had just enough money or not enough money to make ends meet, 23.3% had a somewhat hard or hard time paying for basics, 18.8% had trouble paying for ≥1 type of expense, 20.5% had delayed/used less medical/dental care, and 41.5% had experienced financial strain based on the composite measure. Among 1787 adults aged 66 to 85, the percentages who screened positive on these measures were 22.7%, 19.4%, 12.9%, 19.8%, and 34.4%, respectively. Across the sample, by income categories and racial/ethnic groups, the making ends meet screener identified higher percentages of adults experiencing financial strain and performed better when compared with the composite measure than the hard to pay for the very basics and trouble paying for expenses screeners. Overall, substantial decreases in the percentages of adults who screened positive on the financial strain measures were seen as level of income increased. Within income categories, middle-aged adults were more likely than older adults to have experienced financial strain based on the composite and general single-item screeners.

Conclusions:

As social risk screening becomes part of the standard of care, it will be important to assess how well different brief screeners for financial strain perform with diverse patient populations.

Keywords

social determinants of health social risk screening financial strain health care affordability single-item screeners

Introduction

Financial strain, which broadly refers to a person’s ability to cover expenses with available assets, is the most prevalent social risk in healthcare.^1,2 It goes beyond income and reflects a person’s perception of spending proportionate to available resources with the goal of assessing the competing needs for financial resources.^3
-5 Patients experiencing financial strain are at higher risk for decreased medication adherence, worse self-rated health, poor sleep quality, depression, exacerbation of chronic conditions, high blood pressure, negative cardiovascular outcomes, malnutrition risk, early disability, and increased risk of death, independent of income.^3,5
-16 Health care systems receiving federal funding for underserved and vulnerable populations are now required to screen for and document financial strain in the electronic health record (EHR), along with food, housing, and transportation insecurity, for annual reporting of social risks to the U.S. Uniform Data System.¹⁷ Given that individuals with similar incomes can vary widely in the demands on their financial resources and their perceptions of financial strain, it is important to learn how well different brief screeners that might be adopted in healthcare settings perform in identifying people experiencing financial strain.^3,5,11,18

There is currently no “gold standard” measure to screen patients for financial strain. In their 2014 report Capturing Social and Behavioral Domains and Measures in Electronic Health Records, the Institute of Medicine (IOM) Subcommittee on Social and Behavioral Determinants of Health recommended using a question which asks individuals how hard it is for them to pay for the very basics.¹⁹ The American Academy of Family Physicians social needs screening tools asks individuals to assess how often they have enough money to pay their bills.^20,21 The National Health Interview Survey (NHIS), an annual representative survey of U.S. adults, asks about health care affordability.²² The National Association of Community Health Centers and partners ask patients if they were unable to get food, utilities, medical or health care, phone, clothing, childcare, or other items when it was really needed.²³ Initial research showed how patients were asked about financial strain led to differences in the proportion identified with financial strain.^24,25 One study found that the percent of financial strain in an adult primary care sample was nearly triple when individuals were asked about how well they were getting by versus how hard it was to pay for the very basics (10.4% vs 33.2%, P < .001).²⁵

This study had 4 main objectives. First, we wanted to compare the percentages of adults, overall and within demographic subgroups, identified as experiencing financial strain using 3 single-item general screeners that could be readily incorporated into brief social risk screening tools for use in healthcare settings. Second, we wanted to evaluate how each single-item screener performed in identifying financial strain compared to a composite measure created based on a positive screen on any of the 3 screeners or report of reduced or delayed use of medical or dental care or prescription medications due to cost. Third, we wanted to learn whether middle-aged and older adults differed in perceived financial strain within the same income categories based on these measures. Fourth, we wanted to learn the extent to which adults who were having trouble paying for different types of expenses or foregoing health-related care due to costs might be missed by general single-item screeners depending on criteria used to assign risk.

Methods

Setting and Study Population

The study sample was comprised of Kaiser Permanente Northern California (KPNC) health plan members who were aged 35 to 85 years at the time of the survey, had English listed in the EHR as their preferred spoken and written language, were not insured through Medi-Cal (California’s Medicaid program), and had been health plan members for ≥2 years. KPNC is an integrated healthcare delivery system with more than 3 million adult members whose non-Medicaid membership is demographically representative of the broader Northern California population.²⁶ KPNC provides primary and specialty care, hospitalization, laboratory and other diagnostic services, and pharmacy services to health plan members. Most members pay a set monthly premium and have a variable copay or coinsurance for medical services and prescription medications; members covered through a deductible plan pay a set monthly premium and then pay for medical services and prescription medications at a pre-set rate. Dental coverage is not included in the basic plan but is available for purchase through the health plan.

Data Source

Data for this study came from an English-only self-administered (print and online questionnaire) survey that was fielded with a stratified (age group × sex × race/ethnicity) random sample of KPNC members during April to October 2021, approximately 1 year into the COVID-19 pandemic. Details about the survey, which collected information about sociodemographic and health characteristics and social health risks and needs, can be found in an earlier publication.²⁷ After excluding people who could not be reached, who had died, were no longer KPNC members, or could not complete the English-only questionnaire, the overall survey response rate was 29.0% (n = 2870). Our analyses were restricted to the 2734 adults (95.3%) aged 35 to 85 years who answered all 4 financial strain items.

Study Variables

We examined 4 single-item measures of financial strain and a composite measure (Table 1):

• Hard to pay for the very basics [IOM definition¹⁹]: “In general, how hard is it for you to pay for the very basics like food, housing, medical care, and heating? [Not hard, somewhat hard, hard, or very hard]” [Positive screen for financial strain: Somewhat hard or very hard]

• Making Ends Meet [Conger et al’s²⁸ definition modified with a 3-month recall period]: “Thinking about the past 3 months, at the end of each month you generally ended up with: more than enough money left over, some money left over, just enough money left over, almost enough money left over, or not enough money left over to make ends meet.” [Positive screen for financial strain: Had just enough or not enough money]

• Trouble paying for expenses: “In the past 3 months, did you have trouble paying for any of the following: food; housing (rent or mortgage); utilities (eg, heat and electricity); medical needs; dental care; transportation; childcare or helping care for an adult; phone; internet; debts.” [Positive screen for financial strain: Trouble paying for ≥1 of these 10 expenses]

• Delayed/foregone care²²: “In the past 3 months, because of the cost, did you: delay or not get medical care you thought you needed; delay or not get dental care; take medicine in smaller doses, less frequently than prescribed, or decide not to fill a prescription.” [Positive screen for healthcare-related financial strain: Yes to any of these three actions]

• Composite financial strain measure: A positive screen on any of the 4 individual measures.

Table 1.

Financial Strain Measures Used in Study.

Financial strain measure	Question
Hard to pay for the basics [IOM^a]	In general, how hard is it for you to pay for the very basics like food, housing, medical care, and heating? Very hard; Hard; Somewhat hard; Not hard at all
How well made ends meet during the past 3 months [Conger et al^b]	Thinking about the past 3 months, at the end of each month you generally ended up with: More than enough money left over; Some money left over; Just enough to make ends meet; Almost enough to make ends meet; Not enough to make ends meet
Trouble paying for different expenses during the past 3 months	In the past 3 months, did you have trouble paying for any of the following? Food; Housing (rent or mortgage); Utilities (eg, heat and electricity); Medical needs; Dental care; Transportation; Childcare or helping care for an adult; Phone; Internet; Debts
Delayed/used less medical or dental care or prescription medications due to cost during the past 3 months [NHIS^c]	[Medical Care]: During the past 3 months, did you delay or not get medical care you thought you needed because of the cost? Yes/No
	[Dental Care]: During the past 3 months, did you delay or not get dental care because of the cost? Yes/No
	[Prescription Medications]: During the past 3 months, did you take a medicine in smaller doses or less frequently than prescribed, or decide not to fill a prescription because of the cost? Yes/No
Composite financial strain measure	Just making ends meet or not making ends meet OR somewhat hard or hard to pay for the very basics OR had trouble paying for ≥1 types of expense OR delayed/used less medical or dental care or took less prescription medication because of the cost

Committee on the Recommended Social and Behavioral Domains and Measures for Electronic Health Records, Board on Population Health and Public Health Practice, Institute of Medicine. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. National Academies Press (US); 2015. Accessed May 3, 2024. http://www.ncbi.nlm.nih.gov/books/NBK268995/.

Conger RD, Conger KJ, Matthews LS, Elder GH. Pathways of economic influence on adolescent adjustment. Am J Community Psychol. 1999;27(4):519-541. doi:10.1023/A:1022133228206.

Lu K, Xiong X, Horras A, Jiang B, Li M. Impact of financial barriers on health status, healthcare utilization and economic burden among individuals with cognitive impairment: a national cross-sectional survey. BMJ Open. 2022;12(5):e056466. doi:10.1136/bmjopen-2021-056466.

To describe the study sample, we used self-reported data for age, sex (male, female), race/ethnicity (White, Black, Hispanic/Latino, Asian/Pacific Islander [Asian/PI], and Other), educational attainment (high school education or less; technical/trade school, some college with no degree, or associate’s degree; bachelor’s or postgraduate degree), household income in 2020 (≤$35 000, $35 001-$50 000, $50 001-$80 000, $80 001-$100 000, >$100 000), employment status (employed for pay or self-employed; unemployed or on leave; other [not working due to health or disability; retired; not employed as a full time homemaker, caregiver, or student]), and overall rating of physical health and emotional or mental health (excellent, very good, good, fair, or poor).

Statistical Analyses

All analyses were conducted using SAS v9.4 (SAS Institute, Cary, NC, 2014) and Stata Version 18 (StataCorps LLC). We produced descriptive statistics by age group (35-65 and 66-85 years) and financial strain status based on the composite measure. We next calculated the percentages of adults within these age groups who had experienced financial strain based on the four single-item measures and the composite measure, overall and by four household income categories (<$50 000, $50 001-$80 000, $80 001-$100 000, and >$100 000) and 4 racial/ethnic groups (White, Black, Latino, Asian/PI) and compared percentages within and across these groups using McNemar chi-square tests and modified log-Poisson regression models to control for demographic factors. Using the composite financial strain measure as the comparator, we calculated kappa statistics for 3 general single-item screeners (making ends meet, hard to pay for the very basics, and trouble paying for expenses measures) for the full sample and by age and racial/ethnic groups and identified significant differences in kappa statistics across single-item screeners based on non-overlapping 95% confidence intervals (CI). Then, we restricted the analyses to the subgroup of adults identified as experiencing financial strain based on the composite measure and compared the percentages of adults who screened positive for financial strain on the 3 general screeners in the full sample, by age group, and by racial/ethnic group, testing for significant differences in percentages identified using McNemar chi-square tests. Finally, we calculated the percentages of adults who had experienced trouble paying for different types of expenses or had delayed/used less medical care, dental care, and prescription medications in the prior 3 months due to cost who would not have been identified as experiencing financial strain using the making ends meet or hard to pay for the very basics screeners. All comparisons between subgroups were performed using chi-square tests with statistical significance set at P < .05.

This secondary data analysis was conducted in accordance with the procedures approved by the KPNC Institutional Review Board (IRB). The KPNC IRB had waived requirements to obtain informed consent and Privacy Rule Authorization for the survey.

Results

Among the 947 middle-aged adults (ages 35-65) and 1787 older-aged adults (ages 66-85), 41.5% and 34.4%, respectively, screened positive for financial strain using the composite measure (Table 2). In both age groups, adults experiencing financial strain were less likely than those who were not to be non-Hispanic White, to have a bachelor’s degree or higher, and to have a 2020 household income ≤$35 000, and were more likely to have fair or poor self-rated physical and mental/emotional health.

Table 2.

Sample Characteristics by Age Group and Financial Strain Status.

	Ages 35-65 years			Ages 66-85 years
Characteristic	All	Financial strain^a	No financial strain^c	All	Financial strain^a	No financial strain^c
Sample, N (%)	947	393 (41.5)	554 (58.5)	1787	615 (34.4)	1172 (65.6)
Age, mean (standard deviation)
	53.5 (9.3)	52.2 (9.2)	54.4 (9.3)	73.1 (5.1)	72.9 (5.0)	73.2 (5.1)
Sex, %	^b
Male	44.8	39.2	48.7	46.5	45.0	47.2
Women	55.2	60.8	51.3^b	53.5	55.0	52.8
Race/ethnicity, %				^b
White, non-Hispanic	29.9	24.2	33.9^b	31.6	25.0	35.0^b
Black	21.3	27.5	17.0	23.1	26.7	21.3
Hispanic/Latino	25.1	26.6	24.0	21.5	23.9	20.3
Asian/Pacific Islander	23.4	21.4	24.7	23.1	23.2	22.9
Other	0.3	0.3	0.4	0.7	1.2	0.5^b
Educational attainment, %	^b			^b
≤High school graduate	17.0	22.6	12.9	21.3	26.7	18.5^b
Some college/associate degree/trade school	29.1	33.9	25.7^b	35.9	40.7	33.4
≥Bachelor’s degree	53.9	43.5	61.4^b	42.8	32.6	48.1^b
Household income in 2020, %	^b			^b
≤$35 001	7.8	15.2	2.5^b	20.9	42.0	9.3^b
$35 001-$50 000	9.7	15.4	5.6	14.9	21.2	11.3
$50 001-$80 000	15.6	26.3	13.1	27.5	23.4	29.7
$80 001-$100 000	15.7	16.0	15.6	11.8	5.7	15.3
≥$100 001	48.2	27.1	63.2^b	24.9	7.7	34.4^b
Employment Status	^b
Employed for pay or self-employed	74.7	75.8	73.8	15.3	15.5	15.2
Unemployed or on leave	3.9	6.5	2.2	1.8	3.8	0.8
Other^d	21.4	17.7	24.0^b	82.9	80.7	84.0
Overall rating of physical health	^b			^b
Very good/excellent	45.1	30.9	55.2	35.5	22.6	42.4
Good	43.1	51.4	37.3	44.5	49.3	41.9
Fair/poor	11.8	17.7	7.6^b	20.0	28.1	15.7^b
Overall rating of mental health	^b			^b
Very good/excellent	52.9	39.8	62.2	53.3	37.6	61.4
Good	33.1	37.2	30.2	34.4	42.4	30.2
Fair/poor	14.0	23.0	7.6^b	12.3	20.0	8.4^b

Individuals were classified as financially strained if in the prior 3 months, they had difficulty making ends meet (ie, just made ends meet or did not make ends meet), found it somewhat hard or hard to pay for the very basics, had trouble paying for ≥1 of 11 types of expenses, or delayed/used less medical or dental care or took less prescription medication because of the cost.

Significant difference between financially strained versus not financially strained subgroups at P < .05.

No financial strain is defined as lacking any of these prespecified factors.

Other employment refers to not working due to health or disability, retirement, being a full-time homemaker, caregiver, or student.

Among middle-aged adults, 30.7% had just enough money or not enough money to make ends meet, 23.3% had a somewhat hard or hard time paying for basics, 18.8% had trouble paying for ≥1 type of expense, 20.5% had delayed/used less medical/dental care, and 41.5% had experienced financial strain based on the composite measure (Figure 1). In the older age group, 22.7%, 19.4%, 12.9%, 19.8%, and 34.4%, respectively, screened positive on these measures. In both age groups, higher percentages of adults were identified as financially strained by the making ends meet screener than the hard to pay for the very basics screener, and lower percentages were identified by the trouble paying expenses screener compared to both of these measures, with percentages identified by all 3 screeners substantially lower than identified using the composite measure. Middle-aged adults were more likely than older adults to screen positive for financial strain on all 3 general screeners but did not differ regarding delayed/reduced medical/dental care.

Figure 1.

Percentages of adults experiencing financial strain based on 4 measures, by age group. ¹In past 3 months, had trouble paying for any of the following: Food; Housing (rent or mortgage); Utilities (e.g., heat and electricity); Medical needs; Dental care; Transportation; Childcare or helping care for an adult; Phone; Internet; Debts. ²In past 3 months, because of the cost, delayed or did not get medical care you thought you needed, delayed or did not get dental care, or took medicine in smaller doses, less frequently than prescribed, or decided not to fill a prescription. ³In past 3 months, had just enough or not enough money to make ends meet, had a hard time paying for basic expenses, had trouble paying for ≥1 or 10 types of expense; or delayed/used less medical or dental care due to cost.

Examining performance of the general screeners by income level, within all 4 income levels, higher percentages of adults screened positive on the making ends meet than hard to pay for the very basics screener, and percentages for both measures were higher than for the trouble paying expenses screener (Figure 2). In both age groups, substantial decreases in the percentages of adults who screened positive on the financial strain measures were seen as level of income increased; within income levels, older adults were less likely than middle-aged adults to have experienced financial strain across measures. In the older age group, the percentage who screened positive on the making ends meet screener was higher than that for the hard to pay for the very basics screener only in the $50 001 to $80 000 income level, and the percentage identified by the trouble paying expenses screener was lower than these 2 measures only in the 2 lower income levels. Similarly, the percentages of adults identified by all 3 general measures were substantially lower than the percentage identified by the composite measure.

Figure 2.

Percentages of adults experiencing financial strain based on 4 measures, by household income and age group. HHI: Household income; Numbers above each bar indicate the total percentages who were experiencing financial strain based on this measure. ¹In past 3 months, had trouble paying for any of the following: Food; Housing (rent or mortgage); Utilities (e.g., heat and electricity); Medical needs; Dental care; Transportation; Childcare or helping care for an adult; Phone; Internet; Debts. ²In past 3 months, because of the cost, delayed or did not get medical care you thought you needed, delayed or did not get dental care, or took medicine in smaller doses, less frequently than prescribed, or decided not to fill a prescription. ³In past 3 months, had just enough or not enough money to make ends meet, had a hard time paying for basic expenses, had trouble paying for ≥1 or 10 types of expense; or delayed/used less medical or dental care due to cost.

As expected, in the middle-aged group, the percentages of adults identified as experiencing financial strain based on the three general screeners, reduced/delayed medical/dental care, and composite measure all decreased as level of income increased. However, in the older adult group, the percentages identified by the different measures were lower in the $50 001 to $80 000 versus <$50 000 income categories but did not significantly differ between the 2 upper income categories. A comparison of older-aged adults who had a household income ≤ $35 000 (n = 341) to those with an income of $35 001 to $50 000 (n = 242) found that those in the ≤$35 000 category were more likely to screen positive on the making ends meet (54.6% vs 33.5%), hard to pay for the very basics (43.0% vs 30.6%), trouble paying expenses (39.5% vs 20.2%), reduced/delayed medical/dental care (43.0% vs 28.9%), and composite (66.3% vs 50.8%) measures. The middle-aged subgroup in this income level was too small to perform similar analysis. Also seen in Figure 2, across all 4 income levels, higher percentages of adults in the middle-aged versus older age group screened positive on the general screeners and composite measure. Due to the differences found for older adults in the ≤$35 000 versus $35 001 to $50 000 income ranges and because a higher percentage of older versus middle-aged adults in the ≤ $50 000 income category had an income of ≤ $35 000 (58.5% vs 44.5%), we used a modified log-Poisson regression model that controlled for ≤$35 000 versus $35 001 to $50 000 income to compare middle-aged and older adults in the ≤$50 000 category. The results confirmed significant age group differences on all financial strain measures within this income level after adjusting for income distribution.

A comparison of the percentages of White, Black, Latino, and Asian/PI adults identified as experiencing financial strain is found in Figure 3. In the middle-aged group, higher percentages of adults in all 4 racial/ethnic groups screened positive on the making ends meet screener than the hard to pay for the very basics screener, but in the older age group, this was only observed for Asian/PI adults. Similar to the overall analysis, in both age groups the trouble paying expenses screener consistently identified lower percentages of adults than the other 2 general screeners, and all 3 general screeners identified lower percentages than the composite measure.

Figure 3.

Percentages of adults experiencing financial strain based on 4 measures, by race and age group. HHI: Household income; Numbers above each bar indicate the total percentages who were experiencing financial strain based on this measure. ¹In past 3 months, had trouble paying for any of the following: Food; Housing (rent or mortgage); Utilities (e.g., heat and electricity); Medical needs; Dental care; Transportation; Childcare or helping care for an adult; Phone; Internet; Debts. ²In past 3 months, because of the cost, delayed or did not get medical care you thought you needed, delayed or did not get dental care, or took medicine in smaller doses, less frequently than prescribed, or decided not to fill a prescription. ³In past 3 months, had just enough or not enough money to make ends meet, had a hard time paying for basic expenses, had trouble paying for ≥1 or 10 types of expense; or delayed/used less medical or dental care due to cost.

The kappa statistics, which indicates the extent of agreement of a positive screen using the making ends meet, hard to pay for the very basics, and trouble paying expenses screeners with a positive screen on the composite measure were 0.74, 0.62, and 0.46, respectively (Table 3). The first 2 kappa statistics reflect substantial agreement with the comparator and the last only low to moderate agreement. In the overall sample, in both age groups and across White, Black, and Asian/PI racial/ethnic groups, the kappa statistic for the making ends meet screener was significantly better than the kappa statistics for the hard to pay for the very basics and trouble paying expenses screeners; among Latino adults, kappa statistics for the making ends meet and hard to pay for the very basics screeners were similar, but the trouble paying for expenses screener was lower. We found no statistically significant age group difference in kappa statistics within most racial/ethnic groups, but among Latino adults, kappa statistics were lower among older versus middle-aged adults for the making ends meet (0.63 [0.56-0.71] vs 0.77 [0.69-0.85]) and trouble paying expenses (0.40 [0.31-0.48] vs 0.57 [0.47-0.67]) screeners.

Table 3.

Comparison of Three Single-Item Screeners by Age Group and Race/Ethnicity.

	Just enough/not enough money to make ends meet	Somewhat hard or hard to pay for very basics	Had trouble paying for ≥1 expense type	Just enough/not enough money to make ends meet (%)	Somewhat hard or hard to pay for very basics (%)	Had trouble paying for ≥1 expense type (%)
All	0.74 (0.71-0.70)	0.62 (0.59-0.65)^a	0.46 (0.43-0.49)^a,b	69.1	56.2^c	40.5^c,d
Ages 35-65 years	0.77 (0.73-0.81)	0.60 (0.55-0.65)^a	0.49 (0.44-0.54)^a,b	74.0	56.2^c	45.3^c,d
Ages 66-85 years	0.72 (0.68-0.75)	0.63 (0.59-0.66)^a	0.44 (0.40-0.48)^a,b	66.0^c,e	56.3^c	37.4^c,d
White 35-85 years	0.70 (0.64-0.75)	0.58 (0.52-0.64^a*	0.43 (0.36-0.49)^a,b	61.8	49.8^c	34.5^c,d
Black 35-85 years	0.75 (0.70-0.80)	0.62 (0.56-0.68)^a	0.57 (0.64-0.75)^a	73.2^f	59.6^c,f	54.8^c,f
Latino 35-85 years	0.69 (0.64-0.75)	0.67 (0.61-0.73)	0.47 (0.41-0.54)^a,b	65.5	63.1^f	42.9^c,d,f
Asian/PI 35-85 years	0.80 (0.75-0.85)	0.57 (0.51-0.64)^a	0.32 (0.26-0.39)^a,b	75.8^f	51.1^c	27.3^c,d

Kappa statistic is significantly lower than that for just/not making ends meet based on non-overlapping 95% confidence intervals. ^a*Kappa statistic is significantly lower than that for just/not making ends meet based on chi-square test for differences in percentages correctly and incorrectly identified by both measures.

Kappa statistic is significantly lower than that for somewhat hard/hard to pay for very basics based on non-overlapping 95% confidence intervals.

Percentage is significantly lower than that for just/not making ends meet by McNemar test at P < .05.

Percentage is significantly lower than that for somewhat hard/hard to pay for very basics by McNemar test at P < .05.

Percentage identified by this single-item screener is significantly lower in the 66 to 85 than 35 to 65 age group by chi-square test at P < .05.

Percentage identified by this single-item screener is significantly higher in this racial/ethnic group than in White adults by chi-square test at P < .05.

We also examined the performance of the general screeners by calculating the percentage of adults identified by the composite measure who were also identified by each single-item screener. Of the 1008 adults who screened positive for financial strain based on the composite measure in the full sample, 69.1% screened positive on the making ends meet, 56.2% on the hard to pay for the very basics, and 40.5% on the trouble paying expenses screeners (Table 3). Again, significantly higher percentages of financially strained adults were identified using the making ends meet screener than with the hard to pay for the very basics screener and trouble paying expenses in the full sample, both age groups, and the White, Black, and Asian/PI racial/ethnic groups. Significantly higher percentages of middle-aged than older aged adults were identified using the making ends meet (74.0% vs 66.0%) and trouble paying expenses (45.3% vs 37.4%) screeners, with no age group difference seen for the hard to pay for the very basics screener. Significantly higher percentages of Black and Asian/PI adults than White adults were identified using the making ends meet screener (73.2% and 75.8% vs 61.8%, respectively). As was reflected by the kappa statistics for Latino adults, lower percentages of older Latino adults with financial strain based on the composite measure were identified by the making ends meet (58.5% vs 75.2%) and trouble paying expenses (34.7% vs 54.2%) screeners.

Very small percentages (<1%-3%) of adults who indicated that they had more than enough money to make ends meet or that it was not hard for them to pay for the very basics had trouble paying for any of the 10 expenses in the trouble paying expenses checklist. However, while only 2% of these adults who would not have screened positive for financial strain on either general screener indicated having had trouble paying for dental expenses, approximately 10% said that they had delayed or foregone dental care in the prior 3 months due to the cost, suggesting that individuals may not be taking delayed or foregone expenditures into consideration when reporting challenges paying for that type of expense. We calculated the percentages of adults who reported having had trouble paying for different types of expenses that would not have been picked up as experiencing financial strain by the making ends meet and hard to pay for the very basics screeners when responses of ≤ just enough money to make ends meet and ≥ somewhat hard to pay for the very basics were used as positive screens versus not enough money and hard/very hard to pay for the very basics (Table 4). Across both screeners, approximately 10% of those who had trouble paying for utilities, 20% who had trouble paying for dental care, 41% who had delayed or foregone dental care, and 24% who had delayed or reduced use of medical care or prescription medications would not have been identified as experiencing financial strain based on the broader risk levels, along with 24% and 11% who had a hard time paying for debt based on the hard to pay for the very basics and making ends meet screeners, respectively. Substantially higher percentages of adults having difficulty covering these expenses would have been missed using the more restrictive response levels (hard or very hard time paying for basics and not enough money to make ends meet), and across all expense types, more adults would not have been identified as experiencing these financial strains using the hard/very hard time paying for very basics risk category.

Table 4.

Percentages of Adults Who Experienced Financial Strain (Trouble Paying for Different Basic Expenses or Who Delayed/Reduced Use of Health-Related Care) Who Potentially Would Have Been Missed by the Making Ends Meet and Hard to Pay for the Very Basics Screeners.

	Had trouble paying for this expense^a						Delayed or reduced use due to the cost
	Food(n = 87)	Housing (n = 130)	Utilities (n = 163)	Debt (n = 170)	Dental (n = 187)	Medical(n = 68)	Dentalcare^b (n = 499)	Medical/Rx costs^c (n = 167)
Making ends meet at end of month ^d
Has just enough or not enough money	6.9	7.7	10.7	10.6	25.1	7.3	41.5	26.3
Does not have enough money	37.9	45.4	50.9	52.9	64.7	51.5	78.0	67.1
Hard to pay for the very basics ^e
Somewhat hard, hard, or very hard	4.6	5.4	8.6	24.1	20.3	5.9	41.5	23.3
Hard or very hard	50.6	63.8	66.3	72.9	77.0	63.2	84.8	80.8

In the past 3 months, had trouble paying for any of these expenses.

During the past 3 months, delayed or did not get dental care because of the cost.

During the past 3 months, delayed, or used less, or did not use needed medical care or prescription medication due to the cost.

“Thinking about the past 3 months, at the end of each month you generally ended up with: More than enough money left over; Some money left over; Just enough to make ends meet; Almost enough to make ends meet; Not enough to make ends meet.”

“In general, how hard is it for you to pay for the very basics like food, housing, medical care, and heating? Very hard; Hard; Somewhat hard; Not hard at all.”

Discussion

Financial strain is associated with downstream negative health outcomes and can influence the effectiveness of medical interventions or treatments.^29,30 Financial strain can also impact health-related social risks such as malnutrition.³ The overarching goal of our study was to identify which of 3 single-item screeners would best identify adults experiencing financial strain across age groups, different incomes, and racial/ethnic groups. The 3 screeners were derived from established measures of financial strain, and we found that the making ends meet screener identified more adults who experienced financial strain and performed better in comparison to our composite financial strain measure than the IOM-recommended hard to pay for the very basics screener in the full sample, the 2 age groups, and in the 4 income categories and racial/ethnic subgroups within the 2 age groups. The trouble paying for different expenses checklist question performed worse as a general screener for financial strain than the other measures.

Consistent with previous research, we found that financial strain based on the 3 general screeners and composite financial strain measure was present across all levels of income, with the percentage of adults experiencing financial strain decreasing as level of income increased.^3,11,14,31 We also showed that across all income categories, the percentage of adults experiencing financial strain was significantly higher among middle-aged versus older adults, possibly reflecting lower demands on income and/or adjustment of financial outlays to a smaller budget among retirement-age adults. This highlights the value of using a subjective self-reported measure of financial strain alongside income when screening for financial vulnerability in the healthcare setting and adjusting for financial circumstances in healthcare research. An additional advantage is that people may be more willing to affirm financial strain rather than provide specific information about income. For example, in the starting dataset from which our study was drawn, 9.8% of survey respondents did not answer the question about household income, whereas only 4.7% did not answer the making ends meet and 1.9% did not answer the hard to pay for the very basics questions, with no significant difference by age group.

Our finding that the percentage of adults experiencing financial strain varied based on the measure is consistent with the developing literature on this issue.^1,24,25 One study where the authors compared EHR-based financial strain (very hard or hard to pay for the very basics) versus a financial well-being scale found significant differences in the detected rate of financial strain (10.4% vs 33.2%).²⁵ De Marchis et al also found differences, albeit not statistically significant, in percentages of adults who screened positive for financial strain using different measures (57% CMS tool vs 52% National Academies of Medicine tool, P = 0.36).²⁴ In our study, we find that particularly at higher incomes, more restrictive definitions of financial strain (ie, not enough money and hard/very hard to pay for the very basics) fail to identify large subsets of patients who are struggling to afford food, housing, utilities, and debt, or who have delayed or used less medical care, dental care, and prescription medications. A more expansive definition, which incorporates just enough money to make ends meet or somewhat hard to pay for the very basics, captures a broader swath of the population with financial strain, including individuals who report difficulties in affording and using appropriate levels of health care. Future research should evaluate the impact of various cut points for financial strain on downstream health utilization and outcomes to validate these measures in clinical practice.

Limitations

Our analysis has several limitations that impact the generalizability of our results to other settings. The survey was of members of a single Northern California health plan that excluded young adults and adults who were not English proficient or who lacked a stable mailing address. The survey also excluded very low-income adults who were enrolled in the health plan through Medi-Cal.²⁷ Nonetheless, 16.4% of our study sample reported household incomes below $35 000, which is considered low income for the geographic area served by this health plan. The survey data were collected approximately one and a half years after the start of the COVID-19 pandemic, a time during which higher than normal numbers of health plan members were temporarily out of work and some financially eligible adults were receiving government payments to help them cope with the financial strains imposed by the pandemic. The survey had a relatively low response rate, and by design, we further restricted the sample to adults who had answered all relevant financial strain questions to allow comparison of the performance of the financial strain screeners within the same population. As such, the percentages presented in this paper should not be used to represent prevalence of financial strain in this population, but rather a comparison of prevalence in the same population based on different measures. Finally, we used an unvalidated composite measure of financial strain based on data available from the survey as a comparator for the 3 single-item screeners.

Conclusions

The making ends meet screener identified a larger percentage of adults experiencing financial strain than the IOM-recommended hard to pay for the very basics screener and a checklist question asking about trouble paying for various expenses. However, all 3 single-item screeners identified substantially fewer adults experiencing financial strain than a composite measure which also included delayed or foregone medical and dental expenditures. It is imperative that governmental agencies and healthcare systems assess the performance of different brief screeners for financial strain within diverse patient populations as social risk screening increasingly becomes part of the standard of care in healthcare settings. Additionally, the higher proportions of middle-aged versus older adults who reported financial strain within income categories suggests that household income should be supplemented with subjective measures of financial strain if used as an indicator of financial security for screening and research. Further research into the cultural acceptability and performance of these single-item financial strain screeners for use with demographically and linguistically diverse populations in different settings is necessary to determine the generalizability of our findings.

Footnotes

Acknowledgements

The authors wish to acknowledge our appreciation of the Kaiser Permanente Northern California members who contributed the survey data used for this study. We would also like to acknowledge Ashley Aller, MD and Margae Knox, PhD, MPH, co-fellows in Dr. Tucher’s Delivery Science Fellowship for their generous feedback and survey team members Teresa Y. Lin and Pete Bogdanos.

Authors’ Contributions

NG conceived the study design. NG and ET analyzed and interpreted the data. ET wrote the first draft of the manuscript and NG contributed equally to the writing of subsequent drafts. RG reviewed and commented on drafts of the manuscript. All authors approved the final version of this manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Nancy P. Gordon received support for her work on this study from the Kaiser Permanente Northern California Community Health Program. Emma L. Tucher received support for her work on this study from the Kaiser Permanente Northern California Delivery Science Fellowship.

Availability of Data and Materials

The Kaiser Permanente Northern California Institutional Review Board has not provided approval for these survey data to be placed in a public access repository. However, researchers can request access to use this study data by contacting the corresponding author (ET) or the DOR Data Sharing Workgroup at DOR-.

ORCID iDs

Emma L. Tucher

Nancy P. Gordon

References

Ettman

Fan

Philips

, et al. Financial strain and depression in the U.S.: a scoping review. Transl Psychiatry. 2023;13(1):168. doi:10.1038/s41398-023-02460-z

Long

Franklin

Hagan

, et al. Health-related social needs among older adults enrolled in Medicare Advantage. Health Aff. 2022;41(4):557-562. doi:10.1377/hlthaff.2021.01547

Szanton

Allen

Thorpe

Jr Seeman

Bandeen-Roche

Fried

LP.

Effect of financial strain on mortality in community-dwelling older women. J Gerontol B Psychol Sci Soc Sci. 2008;63(6):S369-S374. doi:10.1093/geronb/63.6.s369

Shavers

VL.

Measurement of socioeconomic status in health disparities research. J Natl Med Assoc. 2007;99(9):1013-1023.

Kahn

Pearlin

LI.

Financial strain over the life course and health among older adults. J Health Soc Behav. 2006;47(1):17-31. doi:10.1177/002214650604700102

Ferrie

Martikainen

Shipley

Marmot

MG.

Self-reported economic difficulties and coronary events in men: evidence from the Whitehall II study. Int J Epidemiol. 2005;34(3):640-648. doi:10.1093/ije/dyi063

Georgiades

Janszky

Blom

László

Ahnve

Financial strain predicts recurrent events among women with coronary artery disease. Int J Cardiol. 2009;135(2):175-183. doi:10.1016/j.ijcard.2008.03.093

Krause

Chronic financial strain, social support, and depressive symptoms among older adults. Psychol Aging. 1987;2(2):185-192. doi:10.1037//0882-7974.2.2.185

Steptoe

Brydon

Kunz-Ebrecht

Changes in financial strain over three years, ambulatory blood pressure, and cortisol responses to awakening. Psychosom Med. 2005;67(2):281-287. doi:10.1097/01.psy.0000156932.96261.d2

10.

Pearlin

Lieberman

Menaghan

Mullan

JT.

The stress process. J Health Soc Behav. 1981;22(4):337-356.

11.

Szanton

Thorpe

Whitfield

Life-course financial strain and health in African-Americans. Soc Sci Med. 2010;71(2):259-265. doi:10.1016/j.socscimed.2010.04.001

12.

Hall

Matthews

Kravitz

, et al. Race and financial strain are independent correlates of sleep in midlife women: the SWAN sleep study. Sleep. 2009;32(1):73-82.

13.

Matthews

Smith

Hancock

Jagger

Spiers

NA.

Socioeconomic factors associated with the onset of disability in older age: a longitudinal study of people aged 75 years and over. Soc Sci Med. 2005;61(7):1567-1575. doi:10.1016/j.socscimed.2005.02.007

14.

Angel

Frisco

Angel

Chiriboga

DA.

Financial strain and health among elderly Mexican-origin individuals. J Health Soc Behav. 2003;44(4):536-551.

15.

Khera

Valero-Elizondo

Das

, et al. Cost-related medication nonadherence in adults with atherosclerotic cardiovascular disease in the United States, 2013 to 2017. Circulation. 2019;140(25):2067-2075. doi:10.1161/CIRCULATIONAHA.119.041974

16.

Canterberry

Figueroa

Long

, et al. Association between self-reported health-related social needs and acute care utilization among older adults enrolled in Medicare Advantage. JAMA Heal Forum. 2022;3(7):e221874-e221874. doi:10.1001/jamahealthforum.2022.1874

17.

2023 Crosswalk of Standardized Social Risk Factor Screeners and UDS Appendix D. Published online 2023. https://bphc.hrsa.gov/sites/default/files/bphc/data-reporting/uds-crosswalk-standardized-screeners.pdf

18.

Oakes

Rossi

PH.

The measurement of SES in health research: current practice and steps toward a new approach. Soc Sci Med. 2003;56(4):769-784. doi:10.1016/s0277-9536(02)00073-4

19.

Committee on the Recommended Social and Behavioral Domains and Measures forElectronic Health Records, Board on Population Health and Public Health Practice, Institute of Medicine. Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. National Academies Press (US); 2015. Accessed May 3, 2024. http://www.ncbi.nlm.nih.gov/books/NBK268995/

20.

American Academy of Family Physicians. Social needs screening tool. Published 2018. https://www.aafp.org/dam/AAFP/documents/patient_care/everyone_project/hops19-physician-form-sdoh.pdf

21.

Aldana

Liljenquist

Validity and reliability of a financial strain survey. J Financ Couns Plan. 1998;9(2):11.

22.

Xiong

Horras

, et al. Impact of financial barriers on health status, healthcare utilisation and economic burden among individuals with cognitive impairment: a national cross-sectional survey. BMJ Open. 2022;12(5):e056466. doi:10.1136/bmjopen-2021-056466

23.

National Association of Community Health Centers; Association of Asian Pacific Community Health Organizations. The PRAPARE screening tool. Published 2019. Accessed May 8, 2024. https://prapare.org/the-prapare-screening-tool/

24.

De Marchis

Fleegler

Cohen

, et al. Screening for financial hardship: comparing patient survey responses using two different screening tools. J Gen Intern Med. 2024;39(1):120-127. doi:10.1007/s11606-023-08437-4

25.

Harle

Vest

. Accuracy of electronic health record food insecurity, housing instability, and financial strain screening in adult primary care. JAMA. 2023;329(5):423-424. doi:10.1001/jama.2022.23631

26.

Davis

Voelkel

Remmers

Adams

McGlynn

EA.

Comparing Kaiser Permanente members to the general population: implications for generalizability of research. Perm J. 2023;27(2):87-98. doi:10.7812/TPP/22.172

27.

Gordon

Shanks

Grant

RW.

Social risks, social needs, and attitudes toward social health screening 1 year into the COVID-19 pandemic: survey of adults in an integrated health care delivery system. Perm J. 2023;27(2):61-74. doi:10.7812/TPP/22.142

28.

Conger

Matthews

Elder

Jr.

Pathways of economic influence on adolescent adjustment. Am J Community Psychol. 1999;27(4):519-541. doi:10.1023/A:1022133228206

29.

Sivapalasingam

Wangechi

Marshed

, et al. Monitoring virologic responses to antiretroviral therapy in HIV-infected adults in Kenya: evaluation of a low-cost viral load assay. PloS One. 2009;4(8):e6828. doi:10.1371/journal.pone.0006828

30.

Sivapalasingam

Klein

Howard

Qin

Tseng

Gourevitch

MN.

Housing insecurity and lack of public assistance are risk factors for tuberculin skin test conversion among persons who use illicit drugs in New York City. J Addict Med. 2009;3(3):172-177. doi:10.1097/ADM.0b013e31819eac4c

31.

Grundy

Holt

The socioeconomic status of older adults: how should we measure it in studies of health inequalities?

J Epidemiol Community Health. 2001;55(12):895-904. doi:10.1136/jech.55.12.895

The Percentage of Patients Experiencing Financial Strain Depends on the Screening Measure: Evidence From a Cross-Sectional Survey of Adult Members of an Integrated Healthcare Delivery System

Abstract

Objectives:

Methods:

Results:

Conclusions:

Keywords

Introduction

Methods

Setting and Study Population

Data Source

Study Variables

Statistical Analyses

Results

Discussion

Limitations

Conclusions

Footnotes

Acknowledgements

Authors’ Contributions

Declaration of Conflicting Interests

Funding

Availability of Data and Materials

ORCID iDs

References