The Earned Income Tax Credit and short-term changes in financial strain and drug use.

Abstract

Introduction

Financial strain is common in the United States and associated with substance use. We examined whether eligibility for the federal Earned Income Tax Credit (EITC), the largest US anti-poverty program, affected financial strain and drug use.

Methods

Using a difference-in-difference design with data from the Population Assessment on Tobacco and Health Wave 1 Adult Survey, a nationally representative survey of US adults, we estimated short-term EITC-associated changes in past-month financial strain, cannabis use, and central nervous system (CNS) depressant use for EITC-eligible people during the EITC disbursement period. Interview timing during/outside the disbursement period was independent of individual characteristics.

Results

Approximately 15.4% of adults were EITC-eligible with refunds ≥$500. Unadjusted prevalences of financial strain among EITC-eligible persons were 35.2% outside and 32.2% during the disbursement period. Refunds were associated with significantly lower financial strain (β = −4.5% [−8.9%, −0.1%]) vs EITC-ineligible individuals. Unadjusted prevalences of cannabis use in both periods were 10.7% and 9.8% in EITC-eligible vs 7.8% and 7.6% among EITC-ineligible; corresponding CNS depressant unadjusted prevalences were 6.3% and 6.6% in EITC-eligible and 4.9% and 4.7% among EITC-ineligible. There were no significant EITC-associated differences in drug use.

Conclusion

Findings support generous EITC refunds, with no evidence that financially supporting low-income people increased drug use.

Untitled section

Keywords: cannabis, CNS depressants, earned income tax credit, financial strain

Article notes

Untitled section

Received 2025 Oct 20; Revised 2025 Dec 31; Accepted 2026 Jan 6; Collection date 2026 Feb.

Author notes

Introduction

Financial strain (ie, the inability to meet financial needs including bills, housing, healthcare, or food)^1,2 is associated with adverse physical and mental health consequences. Consequences potentially include increased blood pressure,³ sleep problems,⁴ depression,^4-7 anxiety,⁶ and substance use,^8-12 suggesting that interventions to reduce financial strain may contribute to changes in drug use. Income support policies aiming to alleviate financial strain¹³ could alter drug use in two potential ways: (1) drug use may increase if individuals have more disposable income that they spend on drugs^14,15 or (2) drug use may decrease¹⁶ among individuals who had been using drugs to relieve stress related to financial strain. Policymaker concerns about drug use among welfare recipients has led many states to implement restrictions and drug testing policies among beneficiaries.^17,18 However, policies aiming to reduce financial strain may decrease drug use by addressing a potential cause. Studying income support policies provides the opportunity to understand whether intervening on financial strain may alter drug use behavior.

The federal Earned Income Tax Credit (EITC), an expansive United States (US) anti-poverty initiative,^19,20 is a policy designed to alleviate financial strain among people with low incomes.^19,20 The federal EITC is a refundable tax credit providing financial support to people with annual earned incomes below certain thresholds.^19,20 When the EITC exceeds the amount the taxpayer owes in taxes, the difference is refunded as a lump sum payment.^19,20 Taxpayers receiving EITC refunds often have immediate short-term changes in spending,^21-23 particularly for debt repayment, utilities, housing, and food,^23-30 suggesting that refunds can provide a short-term intervention to improve a household's ability to meet financial needs and lower the risk of financial strain.³¹ The federal EITC began in 1975 to support low-income taxpayers with children¹⁹ and was extended to taxpayers without children in 1993.¹⁹ EITC eligibility was expanded again for tax year 2021.^32,33 Although expansions expired for tax year 2022, ongoing discussions regarding EITC policy changes^34,35 indicate the need to examine EITC as a potential strategy to reduce financial strain.

Defraying costs of basic needs through EITC support is associated with reduced stress and improved mental health,^36-38 and similar effects could be expected for drug use outcomes. Despite political rhetoric that anti-poverty disbursements will be used for drug purchases, available evidence indicates the contrary. Extant studies examining relationships between the federal EITC and alcohol or drug use have reported no or modest decreases in alcohol^16,39 and cannabis use.¹⁶ However, the relationship between federal EITC receipt and cannabis has not been evaluated in the 21st century¹⁶ and studies using more recent data are needed to understand potential changes in this relationship. Additionally, evidence is needed regarding EITC and other drug use.⁴⁰ To date, no study has examined whether federal EITC affects short-term drug use, particularly non-medical use of painkillers, sedatives, or tranquilizers (ie, central nervous system [CNS] depressants) that may be used to dampen financial strain-related stress⁴¹ and that are widely used in the United States.⁴²Thus, further research is needed to examine the relationship between the federal EITC, cannabis use, and CNS depressant use.

This study therefore examined whether federal EITC eligibility affects financial strain or drug use. Using data from the Population Assessment on Tobacco and Health (PATH), we leveraged variation in the EITC disbursement period^{16,20,21,23,43} relative to PATH interview dates to assess the association between estimated EITC refunds and short-term financial strain, cannabis use, and CNS depressant use among US adults. Findings can provide insights into important consequences of the largest US income support policy.

Methods

Data

We used restricted data from Wave 1 of the PATH Adult Longitudinal Survey,^44-46 a nationally representative household survey of non-institutionalized US adults ages ≥18 in 2013.^44-46 Participants were selected for Wave 1 in 2013 using a multi-stage probability sampling design^44,45 and sampled using a household and adult screener (74% overall weighted response rate),⁴⁴ oversampling key demographic subgroups, including young adults ages 18-24, Black Americans, and people who used tobacco.^44,45 Participants completed computer-assisted personal interviews and audio computer-assisted survey instruments at each wave of data collection.^44,45 Survey weights and balanced repeated replicate weights accounted for stratification, clustering, oversampling, and survey non-response.^44,45 Our analytic sample was limited to people with complete information on income and employment at Wave 1 (n = 32 307) and specific analyses were limited to the ∼93% of people with complete information on each outcome. The PATH study protocol was approved by the Westat Institutional Review Board (IRB). Our protocol was considered Non-Human Subjects Research by the Columbia University IRB.

Measures

Financial strain (yes/no)

Financial strain was defined using a single-item measure asking whether the participant was unable to pay important expenses due to a shortage of money.⁴⁷ Responses included yes, no, and not applicable. We combined responses of “not applicable” and “no” into a dichotomous variable based on information from with the Westat study team indicating that individuals who responded “not applicable” may have thought that the question did not apply to them for various reasons, eg, young adults not responsible for bills, or people who were missing payments for reasons other than a shortage of money.⁴⁸

Past-month cannabis or CNS depressant use (yes/no)

Past-month cannabis use was assessed by combining responses to the questions of lifetime blunt⁴⁹ or cannabis use⁵⁰ and timing since last use, including the past month (yes/no). Past-month CNS depressant use (yes/no) was assessed using questions regarding non-medical use of “painkiller, sedatives or tranquilizers”⁵¹ and timing since last use.⁵²

Estimated federal EITC refund (yes/no)

We estimated an individual's household EITC refund at Wave 1 using a respondent's age, annual household income category (<$10 000, $10 000-$14 999, $15 000-$24 999, $25 000-$34 999, $35 000-$49 999, $50 000-$74 999, $75 000-$99 999, $100 000-$149 999, $150 000-$199 999, ≥$200 000), and the number of adults and children living in the household. We assigned the highest end of an individual's household income category (ie, $299 999 for people earning ≥$200 000) as their income to estimate EITC refunds. Individuals failing to report household income were asked whether they earned above or below $50 000,⁵³ and assigned $100 000 or $50 000 depending on their response (above/below). Because marital status was not measured at Wave 1, if more than one adult lived in the household, we assumed that one of the additional adults was the respondent's spouse and they filed taxes jointly. Estimated refunds were computed using the usincometaxes package for R⁵⁴ based on the National Bureau of Economic Research internet TAXSIM35 package.^55,56 EITC refunds do not reflect an individual's current employment status as households may be eligible for the EITC if at least one household member was employed in the past tax year, with an annual household income >$0 excluding government assistance. Depending on an individual's annual household income, marital status, and the number of qualifying children in the household, refunds could range from $2 to approximately $5900.⁵⁷ Individuals with an estimated refund ≥$500 were considered EITC eligible, consistent with prior research.⁵⁸

Federal EITC disbursement period (yes/no)

Approximately 90% of federal EITC refunds are disbursed in February-April.^16,20-23,43 Leveraging participants’ interview dates, eligible participants selected to be interviewed in February-April were considered “exposed” to federal EITC disbursement. May-January interviews were considered unexposed.¹⁶ Whether a participant was interviewed during or outside the disbursement period was independent of individual characteristics.

Demographic characteristics

EITC participation varies by race and ethnicity, and education about EITC eligibility.^59-61 Consistent with prior research,⁵⁸ we adjusted for sex (male, female), race and ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, non-Hispanic other), and educational attainment at Wave 1 (high school degree or equivalent, less than high school, some college, bachelors, or more).

Statistical analysis

PATH participant interview dates were independent of individual characteristics, suggesting that whether an EITC eligible person was interviewed during/outside the disbursement period could be considered randomly assigned. To examine whether demographic characteristics varied by disbursement period, we applied cross-sectional survey weights with balanced repeated replicate weights to estimate the survey-weighted prevalence of demographic characteristics and drug use outcomes among EITC-eligible and ineligible people during and outside the disbursement period.

Consistent with studies of different health outcomes,^16,39,58 we leveraged random interview dates to estimate short-term effects of EITC eligibility during the disbursement period (ie, assumed refund receipt) on past-month financial strain, cannabis use, and CNS depressant use. We controlled for sociodemographic differences in sex, race and ethnicity, and educational attainment. We used a difference-in-differences design estimator from adjusted survey-weighted binomial models with an identity link to estimate beta coefficients (β) and 95% CIs of two-way interaction terms between EITC eligibility and disbursement period, representing the difference in risk differences of our binary outcomes between the EITC-eligible and ineligible participants during the disbursement period.^16,39,58,62 A binomial distribution was used for the outcome to account for non-normally distributed errors.⁶³ Individuals in EITC ineligible households were used to control for secular changes in outcomes in both time periods. Given that 80% of eligible households receive the federal EITC,^59,64-67 our analysis mimicked an “intent-to-treat” approach because it assumed all eligible individual people would receive the refunds.⁶⁸ All analyses were conducted in Stata Version 18.⁶⁹ Results were approved for disclosure by the ICPSR.

Sensitivity analyses

We tested robustness of study findings by varying EITC refund size thresholds, eligibility criteria, and study sample. Consistent with prior research,¹⁶ we varied the refund size by examining refunds of ≥$1000 because greater refunds could produce larger changes in financial strain and drug use. We then tested eligibility criteria by varying our income assumption and marital status assumptions. To vary our income assumption, we assigned an individual's income to the lowest end of each income category to estimate their EITC refund. We expected that this would increase the number of people who are eligible for any EITC refund but lower the number of people eligible for refunds of ≥$500 or ≥$1000. To vary our marital status assumption, we used marital status information from Wave 2 of the PATH study data to impute marital status at Wave 1. Participants recruited in Wave 1 were re-interviewed for Wave 2 in 2014,⁴⁴ at which time marital status was collected. We varied our study sample by restricting to people who reported lifetime or past-month use of each drug because individuals who have never used cannabis or CNS depressants would likely not change their use following EITC disbursement. Finally, we restricted our sample to people with household incomes of ≤$100 000/year because EITC ineligible individuals may have much higher incomes than people who are eligible for EITC, making the populations less comparable.

Limitations

These analyses had several limitations. First, our “intent-to-treat” approach may overestimate EITC exposure, as 20% of EITC eligible people during the disbursement period likely did not receive refunds.^65,66 This could underestimate measured associations, and future studies should use tax information to identify refund receipt. Second, we used a single-item measure of financial strain, which could contribute to measurement error. The single-item measure in our study is similar to measures in prior research^10,70-75 and validated measures,^1,76,77 but future studies should consider multiple dimensions of financial strain. Third, our EITC eligibility indicator required assumptions regarding household income and marital status. We tested the robustness of our findings by varying these assumptions in sensitivity analyses. An individual's marital status could have changed between waves, and changes in marital status have contributed to differences in EITC refunds.⁷⁸ However, <1% of our sample was reclassified when using Wave 2 marital status and results were consistent, supporting the robustness of results from our main analyses. Fourth, the PATH study does not collect information on frequency of cannabis or CNS depressant use, which could change following EITC refunds. Future research should explore whether EITC refunds are associated with changes in frequency of use among people reporting past-month use. Fifth, there were differences in insurance status between EITC eligible individuals during/outside the disbursement period. Insurance benefits are not considered for EITC eligibility or refunds but could have contributed to reductions in financial strain. Differences could be due to changes in Medicaid that coincided with our study period, and future studies should evaluate whether Medicaid expansion altered this relationship. Sixth, seasonal variations in spending^79,80 and employment⁸¹ particularly among EITC recipients and people in low-income households may also affect financial strain over time.

Results

Sample description

Approximately 15.4% of the sample was eligible for EITC refunds of $500 or more (Table 1). EITC-eligible participants were more likely to be female, Hispanic, have a high school degree or less, and report financial strain compared with EITC-ineligible participants (Table 1). Demographic characteristics, financial strain, and past-month drug use outcomes among EITC-eligible or ineligible people did not vary by disbursement period. The unadjusted prevalence of financial strain among EITC-eligible persons was 35.2% outside the disbursement period and 32.2% during the disbursement period; among EITC-ineligible individuals, financial strain was lower in both periods, 12.6% and 13.9%, respectively (Table 1). The unadjusted prevalence of past-month cannabis use in both periods was 10.7% and 9.8% in EITC-eligible and 7.8% and 7.6% among EITC ineligible. The unadjusted prevalence of past-month CNS depressant use in both periods was 6.3% and 6.6% in EITC-eligible and 4.9% and 4.7% among EITC-ineligible.

Table 1.Survey-weighted prevalence of characteristics and drug use by EITC eligibility and EITC disbursement timing.

	EITC Eligibility High end of income category, $500 Threshold
	EITC Eligible				EITC Ineligible
	Outside of the disbursement period (N = 5210, 11.79%)		During the disbursement period (N = 1506, 3.56%)		Outside of the disbursement period (N = 19822, 65.18%)		During the disbursement period (N = 5769, 19.42%)
	N	%	N	%	N	%	N	%
Sex
Male	2117	40.78	629	40.21	10 485	49.42	3088	49.48
Female	3093	59.22	877	59.79	9337	50.58	2681	50.52
Age
18-20	834	9.17	231	8.45	2070	4.67	614	4.55
21-34	2534	42.61	727	41.20	6510	22.58	1926	21.53
35-49	1331	33.71	388	34.69	4388	23.82	1182	22.66
50-64	415	10.23	132	12.47	4539	28.19	1376	30.28
65+	96	4.27	28	3.20	2315	20.74	671	20.99
Race
NH White	2233	41.92	682	41.94	12 923	70.38	3740	69.28
NH Black	1043	17.27	282	18.49	2601	10.42	752	10.60
Hispanic	1554	34.40	441	33.56	2703	11.06	861	12.91
NH Other	380	6.41	101	6.01	1595	8.14	416	7.20
Education
HS or GED equivalent	1479	30.14	412	27.50	4482	23.93	1226	21.81
< HS	1801	32.63	541	34.51	3107	13.37	1039	15.18
Some college	1690	30.23	493	31.81	7141	31.32	2047	30.53
≥ Bachelors	240	7.00	60	6.19	5092	31.38	1457	32.47
Smoking
Not currently smoking	2982	73.11	852	74.10	13 191	83.14	3806	83.43
Currently smoking	2214	26.48	651	25.75	6584	16.66	1952	16.39
Insurance
Uninsured	1720	32.62	459	28.48	3568	12.83	1065	13.38
Insured	3426	65.60	1027	69.63	15 764	84.44	4563	83.91
Missing	64	1.79	20	1.89	490	2.73	141	2.71
Employment
FT	2011	41.98	555	40.88	9406	47.93	2526	44.53
PT	1177	20.88	339	20.90	3602	14.92	1026	14.35
Student	155	2.32	50	2.30	737	2.24	280	2.62
Retired	86	3.22	32	3.40	2040	17.62	629	18.81
Unemployed	1760	31.14	522	32.12	3856	16.29	1263	18.95
Income
< $10 000	1888	31.47	550	32.30	2885	10.51	955	12.29
$10 000-$24 999	2132	43.07	594	41.13	3114	13.73	927	13.56
$25 000-$49 999	1190	25.47	362	26.56	3970	19.43	1148	19.90
$50 000-$99 999	0	0.00	0	0.00	5151	28.49	1444	27.85
$100 000+	0	0.00	0	0.00	3101	19.23	812	17.83
Missing	0	0.00	0	0.00	1601	8.60	483	8.58
Household Structure
Living alone	0	0.00	0	0.00	2832	16.61	868	17.69
Adults only	0	0.00	0	0.00	11 960	57.17	3525	56.72
Children only	561	11.32	193	13.25	259	1.49	61	1.16
Adults and children	4649	88.68	1313	86.75	4771	24.74	1315	24.42
Financial Strain
No financial strain	3320	64.38	1006	67.46	16 450	86.75	4714	85.71
Financial strain	1873	35.24	496	32.15	3247	12.60	1026	13.87
Missing	17	0.38	4	0.39	125	0.64	29	0.43
PM Cannabis
Never use	2618	64.22	749	63.83	9724	62.39	2693	60.64
PM Use	931	10.74	248	9.79	2988	7.80	888	7.59
Lifetime	1634	24.61	502	25.97	6950	29.00	2139	31.17
PM CNS depressants (ie, painkillers, sedatives, or tranquilizers)
Never use	4240	83.50	1216	83.55	16 268	85.52	4671	85.45
PM Use	360	6.33	112	6.57	1152	4.85	326	4.67
Lifetime	588	9.48	171	8.99	2271	8.80	741	9.26

Note: EITC = Earned Income Tax Credit, defining eligibility by any estimated EITC refund greater than or equal to $500; High end of Income Category = we used the high end of an individual's household income response category to estimate EITC refund size; Outside the disbursement period = participant interview date in May-January; During the disbursement period = participant interview date in February-April; NH = non-Hispanic; HS = high school degree; GED = general educational development; FT = employed full time; PT = employed part time; Children only = 1 adult living with children in the household; Adults and Children = More than one adult and children living in the household; Lifetime = lifetime use but not in the past month; PM = past month; CNS = central nervous system. There were 75 people in our sample who were missing information on current smoking status, 175 people missing information on financial strain, 255 people had other employment status, 243 people who were missing information on cannabis use, and 191 people missing information on CNS depressant use.

Federal EITC eligibility, financial strain, and drug use—difference in difference estimates

The change in the probability of financial strain was 4.51 percentage points lower among people who were eligible for estimated federal EITC refunds of ≥$500 during the disbursement period compared with the change among people in EITC-ineligible households interviewed during the disbursement period (β = −4.51%, 95% CI: −8.87%, −0.14%) (Figure 1A). Refunds of ≥$500 were not associated with changes in past-month cannabis or CNS depressant use (Figure 1A).

Figure 1.Short-term changes in past-month financial strain and drug use associated with EITC eligibility in the disbursement period. (A). Changes in financial strain and drug use with estimated refunds of $500 or more (14.5% of the population). (B). Changes in financial strain and drug use associated with estimated refunds of $1000 or more (13.9% of the population—subset of those in Figure 1A). Source: Adults with complete outcome data in Wave 1 (2013) of the Population Assessment on Tobacco and Health. Note: EITC = Earned Income Tax Credit, defining eligibility by an estimated EITC refund of ≥$500 or ≥$1000; High end of Income Category = we used the high end of an individual's household income response category to estimate EITC refund size; Low end of Income Category = we used the low end of an individual's household income response category to estimate EITC refund size; PM = past month; CNS = central nervous system. We used survey-weighted binomial models with an identity link to estimate adjusted betas representing the short-term change in each outcome associated with EITC eligibility during the EITC disbursement period, adjusting for sex, race and ethnicity, and education. Estimates are represented by a dot and 95% confidence intervals are represented by a bar around the estimate. Confidence intervals that do not cross the null value (0.00) are considered significant at a P-value <0.05. There were 175 people missing information on financial strain, 243 people who were missing information on cannabis use, and 191 people missing information on CNS depressant use.

Sensitivity analyses

Varying federal EITC refund threshold

Approximately 14.4% were eligible for EITC refunds of ≥$1000 (Appendix A). The unadjusted prevalence of financial strain among EITC eligible persons was 36.0% outside the disbursement period and 32.5% during the disbursement period; among EITC ineligible individuals, financial strain was lower in both periods, 12.7% and 14.1%, respectively (Appendix A). The change in the probability of financial strain was 4.91 percentage points lower among people who were eligible for estimated federal EITC refunds of ≥$1000 during the disbursement period compared with the change among people living in EITC-ineligible households interviewed during the disbursement period (β = −4.91%, 95% CI: −9.29%, −0.54%) (Figure 1B). EITC refunds of ≥$1000 remained unassociated with changes in past-month cannabis or CNS depressant use (Figure 1B).

Varying eligibility assumptions

Assigning the low end of the income category decreased the number of people receiving estimated EITC refunds of ≥$500 and ≥$1000; approximately 14.5% and 13.5% of people met EITC eligibility when estimating refund sizes using the low end of the income category and a $500 threshold and $1000 threshold, respectively (Appendix B). Federal EITC refunds of ≥$500 or ≥$1000 were associated with decreased financial strain, but confidence intervals overlapped the null (Figures 1A and 1B). Consistent with main analyses, refunds of ≥$500 or ≥$1000 were not associated with changes in cannabis or CNS depressant use using the low end of the income category (Figures 1A and 1B).

Using an individual's Wave 2 marital status, the percentage of people living in EITC-eligible households with refunds of ≥$500 and ≥$1000 decreased by 0.48% and 0.74%, respectively (Appendix C). Refunds of ≥$1000 were associated with a decreased probability of financial strain, and ≥$500 or ≥$1000 refunds were not associated with cannabis or CNS depressant use (Appendix D).

Varying cannabis and CNS depressant study sample

In separate analyses limiting the sample to people reporting lifetime cannabis or CNS depressants use, federal EITC refunds of ≥$500 and ≥$1000 at the low and high end of the income category remained unassociated with past-month cannabis or CNS depressant use (Table 2).

Table 2.Short-term changes in drug use associated with EITC eligibility during the EITC disbursement period among people reporting lifetime or past-month drug use.

Outcome	β	95% CI	β	95% CI	β	95% CI	β	95% CI
	People with lifetime or past-month drug use
	Low end of the Income Category, $500 Threshold		Low end of the Income Category, $1000 Threshold		High end of the Income Category, $500 Threshold		High end of the Income Category, $1000 Threshold
Past-month cannabis use
Difference-in-differences estimates of past month cannabis use (vs lifetime cannabis use)	0.0148	−0.0372, 0.0668	0.0094	−0.0452, 0.0640	−0.0152	−0.0664, 0.0360	−0.0100	−0.0633, 0.0433
Past-month CNS depressant use
Difference-in-differences estimates of past month CNS depressant use (vs lifetime CNS depressant use)	0.0877	−0.0078, 0.1832	0.0873	−0.0021, 0.1766	0.0758	−0.0078, 0.1595	0.0724	−0.0167, 0.1616

Note: EITC = Earned Income Tax Credit, defining eligibility by different estimated EITC refunds; Low end of income category = we used the low end of an individual's household income response category to estimate EITC refund size; High end of income category = we used the high end of an individual's household income response category to estimate EITC refund size; β = adjusted estimated change associated with EITC eligibility during the EITC disbursement period, adjusted for sex, race and ethnicity, and education; CI = confidence interval. Betas were estimated using separate adjusted survey-weighted binomial models with an identity link for each drug among people reporting lifetime or past-month use of each drug, varying EITC eligibility, high or low ends of the income category and outcomes.

Restricting study sample to people earning ≤$100 000

Results were consistent with overall analyses when restricting the sample to people earning ≤$100 000 (Table 3). Among people eligible for estimated refunds of ≥$500 and ≥$1000 at the high end of the income category interviewed during the disbursement period, the change in the probability of financial strain was 4.54 percentage points lower and 4.94 percentage points lower, respectively, than the change in people in EITC-ineligible households interviewed during the same months (≥$500: β = −4.54%, 95% CI: −8.94%, −0.14%; ≥$1000: β = −4.94%, 95% CI: −9.32%, −0.56%). Estimated refunds of any size at the low and high end of the income category remained unassociated with past-month cannabis or CNS depressant use overall and among people reporting lifetime use (Table 3).

Table 3.Short-term changes in financial strain and drug use associated with EITC eligibility during the EITC disbursement period among people earning $100 000 or less.

Outcome	β	95% CI	β	95% CI	β	95% CI	β	95% CI
	People with household incomes ≤$100 000
	Low end of the Income Category, $500 Threshold		Low end of the Income Category, $1000 Threshold		High end of the Income Category, $500 Threshold		High end of the Income Category, $1000 Threshold
Past-month financial strain
Difference-in-differences estimates of financial strain (vs no financial strain)	−0.0342	−0.0769, 0.0086	−0.0329	−0.0778, 0.0121	−0.0454	−0.0894, −0.0014	−0.0494	−0.0932, −0.0056
Past month cannabis use
Difference-in-differences estimates of cannabis use (vs never or lifetime cannabis use)	0.0011	−0.0175, 0.0198	0.0014	−0.0182, 0.0209	−0.0088	−0.0277, 0.0101	−0.0064	−0.0264, 0.0136
Difference-in-differences estimates of cannabis use (vs lifetime cannabis use)	0.0193	−0.0350, 0.0735	0.0134	−0.0434, 0.0702	−0.0131	−0.0653, 0.0391	−0.0074	−0.0617, 0.0469
Past-month CNS depressant use
Difference-in-differences estimates of CNS depressant use (vs never lifetime CNS depressant use)	0.0194	−0.0037, 0.0426	0.0185	−0.0045, 0.0415	0.0068	−0.0141, 0.0277	0.0040	−0.0177, 0.0257
Difference-in-differences estimates of CNS depressant use (vs lifetime CNS depressant use)	0.0844	−0.0155, 0.1844	0.0835	−0.0100, 0.1769	0.0710	−0.0169, 0.1589	0.0666	−0.0266, 0.1599

Note: EITC = Earned Income Tax Credit, defining eligibility by varying estimated EITC refunds; Low end of income category = I used the low end of an individual's household income response category to estimate EITC refund size; High end of income category = I used the high end of an individual's household income response category to estimate EITC refund size; β = adjusted estimated change associated with EITC eligibility during the EITC disbursement period, adjusted for sex, race and ethnicity, and education; CI = confidence interval. Betas were estimated using separate adjusted survey-weighted binomial models with an identity link for financial strain and each drug among people with household incomes of $100 000 or less, varying EITC eligibility, high or low ends of the income category, outcomes, and sample of people reporting lifetime use of each drug.

Discussion

This study used a nationally representative sample of US adults to examine short-term changes in past-month financial strain, cannabis use, and CNS depressant use attributed to estimated EITC refunds. Results indicate that refunds of ≥$500 were associated with decreased financial strain without altering past-month cannabis or CNS depressant use, including among people reporting lifetime use. Our study builds on current literature by examining associations between the federal EITC, cannabis use, and CNS depressant use in the 21^st century.

Study findings suggest the federal EITC lowers the probability of financial strain when refunds are ≥$500. Given that refunds are often used for debts, bills, or housing,^24-29 we expected that EITC-eligible people would have less trouble paying important bills due to money shortages (ie, experiencing financial strain) during the disbursement period. Findings are consistent with a prior study in the South and Western US showing federal EITC refunds of ≥$1000 during the disbursement period lowered the likelihood of food insecurity, a component of financial strain.¹⁶ Findings in the prior study were no longer significant when examining EITC refunds >$0,¹⁶ suggesting that small federal EITC refunds may not be sufficient to alleviate financial strain³⁰ and supporting more generous refunds. Several states also have state EITCs, structured as a percentage of the federal EITC that may complement federal refunds and modify this relationship.^82-84 This study expands on prior work by using national data and examining a different aspect of financial strain; future studies should evaluate whether this relationship is modified by state EITCs.

EITC refunds were not associated with cannabis or CNS depressant use overall or among people reporting lifetime use of these drugs, respectively. Findings contradict the only published study evaluating EITC and cannabis use, which was published in 2014.¹⁶ Using 1988-1994 data from the South and Western US, this prior study found cannabis use decreased among EITC-eligible people during the disbursement period.¹⁶ Differences in our results may be due to the study period as well as data source, as 19 states passed medical cannabis laws, and two states passed recreational cannabis laws by PATH Wave 1.⁸⁵ Medical and recreational cannabis laws are associated with increased adult cannabis use overall^86-90 and may have contributed to our null findings comparing people living in EITC-eligible and ineligible households. While examination of variation in the relationship between EITC and cannabis use by state cannabis legal status is beyond the scope of the present study, future studies should examine this relationship.

Conclusion

In a nationally representative survey, EITC refunds of ≥$500 were associated with a decreased probability of financial strain. This is likely because individuals receiving EITC refunds often use them to pay bills such as utilities or rent, and re-pay debt.^24-29 Findings suggest EITC refunds may meet their intended purpose, supporting generous refunds to help mitigate financial strain among low-income populations and increased outreach efforts to encourage EITC participation.^91,92 At a time when the US is considering changes in EITC eligibility,³⁵ findings generally refute concerns at the center of the welfare policy debate,^17,18,93 suggesting income support among people with low incomes or people with a history of drug use does not increase drug use.

Contribution statement

S.G. had full access to the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Concept and Design: All authors. Acquisition, analysis, or interpretation of the data: All authors. Drafting of the manuscript: All authors. Critical review of the manuscript for important intellectual content: All authors. Statistical analysis: S.G. Obtained funding: S.G. Administrative, technical, or material support: S.G.

Supplementary Material

Contributor Information

Sarah Gutkind, Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY 10032, United States; Department of Population Health Sciences, Weill Cornell Medicine, New York, NY 10022, United States.

Melanie M Wall, Department of Biostatistics, Columbia University Mailman School of Public Health, New York, NY 10032, United States; New York State Psychiatric Institute, New York, NY 10032, United States.

Katherine M Keyes, Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY 10032, United States.

Silvia S Martins, Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY 10032, United States.

Deborah S Hasin, Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY 10032, United States; New York State Psychiatric Institute, New York, NY 10032, United States.

Supplementary material

Supplementary material is available at Health Affairs Scholar online.

Funding

This work was supported by grants R36DA058180 (PI: Gutkind) and R01DA059376 (PI: Martins) from the National Institute on Drug Abuse. The funding organizations had no further role in designing and conducting the study; analyzing or interpreting the data; reviewing or approving the manuscript; and deciding to submit for publication.

Conflicts of interest

Please see ICMJE form(s) for author conflicts of interest. These have been provided as supplementary materials.

D.S.H. receives funding from Syneos Health for an unrelated project.

Notes

Untitled section

References

1. Institute of Medicine . Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. 2015. Accessed January 20, 2022. https://www.ncbi.nlm.nih.gov/books/NBK269339/
2. Grafova I. Financial strain and health. Oxford Research Encyclopedia of Economics and Finance. 2018:1–17. 10.1093/acrefore/9780190625979.013.379
3. Steptoe A, Brydon L, Kunz-Ebrecht S. Changes in financial strain over three years, ambulatory blood pressure, and cortisol responses to awakening. Psychosom Med. 2005;67(2):281–287. 10.1097/01.psy.0000156932.96261.d2
4. Huang Y, Heflin CM, Validova A. Material hardship, perceived stress, and health in early adulthood. Ann Epidemiol. 2021;53:69–75.e3. 10.1016/j.annepidem.2020.08.017
5. Ettman CK, Fan AY, Philips AP, et al. Financial strain and depression in the U.S.: a scoping review. Transl Psychiatry. 2023;13(1):168. 10.1038/s41398-023-02460-z
6. Katz J, Crean HF, Cerulli C, Poleshuck EL. Material hardship and mental health symptoms among a predominantly low income sample of pregnant women seeking prenatal care. Matern Child Health J. 2018;22(9):1360–1367. 10.1007/s10995-018-2518-x
7. McCarthy B, Carter A, Jansson M, Benoit C. Material Hardship Shows a Clearer Picture of Mental Health among low-Wage Workers. Center for Poverty Research, University of California; 2015:1051–1071.
8. Finch KA, Ramo DE, Delucchi KL, Liu H, Prochaska JJ. Subjective social status and substance use severity in a young adult sample. Psychol Addict Behav. 2013;27(3):901–908. 10.1037/a0032900
9. Vijayasiri G, Richman JA, Rospenda KM. The great recession, somatic symptomatology and alcohol use and abuse. Addict Behav. 2012;37(9):1019–1024. 10.1016/j.addbeh.2012.04.007
10. VanKim NA, Laska MN. Socioeconomic disparities in emerging adult weight and weight behaviors. Am J Health Behav. 2012;36(4):433–445. 10.5993/AJHB.36.4.1
11. Brown E, Wehby GL. Economic conditions and drug and opioid overdose deaths. Med Care Res Rev. 2019;76(4):462–477. 10.1177/1077558717722592
12. Nagata JM, Palar K, Gooding HC, et al. Food insecurity, sexual risk, and substance use in young adults. J Adolesc Health. 2021;68(1):169–177. 10.1016/j.jadohealth.2020.05.038
13. McKernan S-M, Ratcliffe C, Braga B. The effect of the US safety net on material hardship over two decades. J Public Econ. 2021;197:104403. 10.1016/j.jpubeco.2021.104403
14. Rosen MI. The ‘check effect’ reconsidered. Addiction. 2011;106(6):1071–1077. 10.1111/j.1360-0443.2011.03409.x
15. Richardson L, Dong H, Kerr T, Milloy MJ, Hayashi K. Drug-related harm coinciding with income assistance payments: results from a community-based cohort of people who use drugs. Addiction. 2021;116(3):536–545. 10.1111/add.15182
16. Rehkopf DH, Strully KW, Dow WH. The short-term impacts of earned income tax credit disbursement on health. Int J Epidemiol. 2014;43(6):1884–1894. 10.1093/ije/dyu172
17. McCarty M, Falk G, Aussenberg RA, Carpenter DH; Congressional Research Service . Drug testing and crime-related restrictions in TANF, SNAP, and housing assistance. 2016. Accessed June, 2024. https://sgp.fas.org/crs/misc/R42394.pdf
18. Office of the Assistant Secretary for Planning and Evaluation; Office of Human Services Policy; US Department of Health and Human Services . Drug Testing Welfare Recipients: Recent Proposals and Continuing Controversies. 2011. Accessed June, 2024. https://aspe.hhs.gov/reports/drug-testing-welfare-recipients-recent-proposals-continuing-controversies-0
19. Hotz VJ. The earned income tax credit. Means-tested Transfer Programs in the United States. University of Chicago Press; 2003:141–198.
20. Nichols A, Rothstein J. The Earned Income Tax Credit (EITC) (No.w21211). National Bureau of Economic Research; 2015:137–218.
21. Barrow L, McGranahan L. The effects of the earned income credit on the seasonality of household expenditures. Natl Tax J. 2000;53(4.2):1211–1243. 10.17310/ntj.2000.4S1.08
22. Aladangady A, Aron-Dine S, Cashin D, et al. High-frequency spending responses to the Earned Income Tax Credit. Proceedings Annual Conference on Taxation and Minutes of the Annual Meeting of the National Tax Association. 2018;111:1–10.
23. Goodman-Bacon A, McGranahan L. How do EITC recipients spend their refunds? Econ Perspect. 2008;32(2):17–32.
24. Lenhart O. The effects of income on health: new evidence from the Earned Income Tax Credit. Rev Econ Househ. 2019;17(2):377–410. 10.1007/s11150-018-9429-x
25. Despard MR, Perantie DC, Oliphant J, Grinstein-Weiss M. Do EITC Recipients Use Their Tax Refunds to Get Ahead? Evidence From the Refund to Savings Initiative. Center for Social Development (CSD) Research Brief; 2015:15-38.
26. Sykes J, Križ K, Edin K, Halpern-Meekin S. Dignity and dreams: what the Earned Income Tax Credit (EITC) means to low-income families. Am Sociol Rev. 2015;80(2):243–267. 10.1177/0003122414551552
27. Smeeding TM, Phillips KR, O'Connor M. The EITC: expectation, knowledge, use, and economic and social mobility. Natl Tax J. 2000;53(4.2):1187–1209. 10.17310/ntj.2000.4S1.07
28. Mendenhall R, Edin K, Crowley S, et al. The role of earned income tax credit in the budgets of low-income households. Soc Serv Rev. 2012;86(3):367–400. 10.1086/667972
29. Hamad R, Yeb J, Jackson K, Gosliner W, Fernald LC. Potential mechanisms linking poverty alleviation and health: an analysis of benefit spending among recipients of the U.S. Earned income tax credit. BMC Public Health. 2023;23(1):1385. 10.1186/s12889-023-16296-1
30. Fisher J, Rehkopf DH. The Earned Income Tax Credit as supplementary food benefits and savings for durable goods. Contemp Econ Policy. 2022;40(3):439–455. doi: 10.1111/coep.12565
31. Kramer KZ, Andrade FCD, Greenlee AJ, Mendenhall R, Bellisle D, Lemons Blanks R. Periodic Earned Income Tax Credit (EITC) payment, financial stress and wellbeing: a longitudinal study. J Fam Econ Issues. 2019;40(3):511–523. 10.1007/s10834-019-09618-2
32. Internal Revenue Service . Changes to the earned income tax credit for the 2022 filing season. Accessed March 11, 2024. https://www.irs.gov/newsroom/changes-to-the-earned-income-tax-credit-for-the-2022-filing-season
33. Center on Budget and Policy Priorities . Policy Basics: The Earned Income Tax Credit. Accessed March 11, 2024. https://www.cbpp.org/research/policy-basics-the-earned-income-tax-credit
34. Bipartisan Policy Center . A Bipartisan Proposal to Expand the Earned Income Tax Credit. Accessed April 17, 2024. https://bipartisanpolicy.org/explainer/bipartisan-earned-income-tax-credit/
35. Jason S; The United States House Committee on Ways & Means . The One, Big, Beautiful, Bill. 2025. Accessed September 15, 2025. https://waysandmeans.house.gov/wp-content/uploads/2025/05/The-One-Big-Beautiful-Bill-Section-by-Section.pdf
36. Morgan ER, Hill HD, Mooney SJ, Rivara FP, Rowhani-Rahbar A. State earned income tax credits and general health indicators: a quasi-experimental national study 1993-2016. Health Serv Res. 2020;55(Suppl 2):863–872. 10.1111/1475-6773.13307
37. Courtin E, Allen HL, Katz LF, Miller C, Aloisi K, Muennig PA. Effect of expanding the Earned Income Tax Credit to Americans without dependent children on psychological distress. Am J Epidemiol. 2022;191(8):1444–1452. 10.1093/aje/kwab164
38. Lenhart O. The effects of state-level earned income tax credits on suicides. Health Econ. 2019;28(12):1476–1482. 10.1002/hec.3948
39. Collin DF, Shields-Zeeman LS, Batra A, et al. Short-term effects of the earned income tax credit on mental health and health behaviors. Prev Med. 2020;139:106223. 10.1016/j.ypmed.2020.106223
40. van Draanen J, Peng J, Ye T, Williams EC, Hill HD, Rowhani-Rahbar A. No change in substance use disorders or overdose after implementation of state Earned Income Tax Credit (EITC). Drug Alcohol Depend. 2024;260:111344. 10.1016/j.drugalcdep.2024.111344
41. Dollar CB, Hendrix JA. “I’m not a traditional woman”: tranquilizer misuse as self-medication among adult women. Am Behav Sci. 2018;62(11):1562–1585. 10.1177/0002764218787027
42. Substance Abuse and Mental Health Services Administration . 2022 National Survey on Drug Use and Health Detailed Tables: Tables 1.3A and 1.3B. Accessed March 27, 2024. https://www.samhsa.gov/data/sites/default/files/reports/rpt42728/NSDUHDetailedTabs2022/NSDUHDetailedTabs2022/NSDUHDetTabsSect1pe2022.htm
43. LaLumia S. The EITC, tax refunds, and unemployment spells. Am Econ J Econ Policy. 2013;5(2):188–221.
44. Inter-University Consortium for Political and Social Research (ICPSR) . Population Assessment of Tobacco and Health (PATH) Study [United States] Public-Use Files User Guide. 2021. Accessed June 9, 2022. https://www.icpsr.umich.edu/files/NAHDAP/documentation/ug36498-all.pdf
45. United States Department of Health and Human Services; National Institutes of Health; National Institute on Drug Abuse; Food Drug Administration; Center for Tobacco Products . Data from: Population Assessment of Tobacco and Health (PATH) Study [United States] Restricted-Use Files. 2021. Accessed June 9, 2022. 10.3886/ICPSR36231.v29
46. Hyland A, Ambrose BK, Conway KP, et al. Design and methods of the Population Assessment of Tobacco and Health (PATH) study. Tob Control. 2017;26(4):371–378. 10.1136/tobaccocontrol-2016-052934
47. National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AM0015: R01_AM0015: Because of shortage of money were you unable to pay important bills on time in the past 30 days. Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AM0015?archive=nahdap
48. Gutkind S. Westat . PATH Study Data Collection Questions. Personal Communication. 2023.
49. National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AG9105: R01_AG9105: Ever smoked part or all of a cigar, cigarillo or filtered cigar with marijuana in it (blunt). Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AG9105?archive=nahdap
50. National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AX0085: Ever used marijuana hash THC or grass. Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AX0085?archive=nahdap
51. National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AX0089_02: R01_AX0089_02: Ever used prescription drugs not prescribed to you: Painkillers, sedatives or tranquilizers. Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AX0089_02?archive=nahdap
52. National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AX0081_02: R01_AX0081_02: How long since last used: Painkillers, sedatives or tranquilizers. Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AX0081_02?archive=nahdap
53. National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AM0031: R01_AM0031: Nonresponse follow-up probe: Total household income above or below $50,000 during the past 12 months. Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AM0031?archive=nahdap
54. Orr SJ. usincometaxes: Calculate Federal and State Income Taxes in the United States. 2023. Accessed February 4, 2024. https://github.com/shanejorr/usincometaxes, https://shaneorr.io/r/usincometaxes
55. National Bureau of Economic Research . Internet TAXSIM Version 35. Accessed March 11, 2024. https://taxsim.nber.org/taxsim35/
56. Feenberg D, Coutts E. An introduction to the TAXSIM model. J Policy Anal Manag. 1993;12(1):189–194. 10.2307/3325474
57. Internal Revenue Service . Earned Income Tax Credit for 2012; Do I Qualify? IRS Fact Sheet (FS-2013-1). Accessed April 3, 2024. https://www.irs.gov/pub/irs-news/FS-13-01.pdf. 2013.
58. Hamad R, Collin DF, Rehkopf DH. Estimating the short-term effects of the earned income tax credit on child health. Am J Epidemiol. 2018;187(12):2633–2641. 10.1093/aje/kwy179
59. Hamad R, Gosliner W, Brown EM, et al. Understanding take-up of the earned income tax credit among Californians with low income. Health Aff (Millwood). 2022;41(12):1715–1724. 10.1377/hlthaff.2022.00713
60. Iselin J, Mackay T, Unrath M. Measuring take-up of the California EITC with state administrative data. J Public Econ. 2023;227:105002. 10.1016/j.jpubeco.2023.105002
61. Thomson D, Chen Y, Gennetian LA, Basurto LE. Earned income tax credit receipt by hispanic families with children: state outreach and demographic factors. Health Aff (Millwood). 2022;41(12):1725–1734. 10.1377/hlthaff.2022.00725
62. Griffin BA, Schuler MS, Stuart EA, et al. Variation in Performance of Commonly Used Statistical Methods for Estimating Effectiveness of State-Level Opioid Policies on Opioid-Related Mortality. National Bureau of Economic Research; 2020. 10.3386/w27029
63. Holmberg MJ, Andersen LW. Estimating risk ratios and risk differences: alternatives to odds ratios. JAMA. 2020;324(11):1098–1099. 10.1001/jama.2020.12698
64. Internal Revenue Service . EITC Participation Rate by States Tax Years 2012 through 2019. Accessed March 4, 2023. https://www.eitc.irs.gov/eitc-central/participation-rate-by-state/eitc-participation-rate-by-states
65. Internal Revenue Service . EITC Participation Rate by States Tax Years 2011 through 2018. Accessed May 16, 2022. https://www.eitc.irs.gov/eitc-central/participation-rate/eitc-participation-rate-by-states
66. Scholz JK. The earned income tax credit: participation, compliance, and antipoverty effectiveness. Natl Tax J. 1994;47(1):63–87. 10.1086/NTJ41789053
67. Jones MR, Ziliak JP. The antipoverty impact of the EITC: new estimates from survey and administrative tax records. Natl Tax J. 2022;75(3):451–479. 10.1086/720614
68. Hamad R, Niedzwiecki MJ. The short-term effects of the earned income tax credit on health care expenditures among US adults. Health Serv Res. 2019;54(6):1295–1304. 10.1111/1475-6773.13204
69. StataCorp . Stata Statistical Software: Release 18. StataCorp LLC; 2023.
70. Krause N, Pargament KI, Ironson G, Hill P. Religious involvement, financial strain, and poly-drug use: exploring the moderating role of meaning in life. Subst Use Misuse. 2017;52(3):286–293. 10.1080/10826084.2016.1225096
71. Brantley ML, Kerrigan D, German D, Lim S, Sherman SG. Identifying patterns of social and economic hardship among structurally vulnerable women: a latent class analysis of HIV/STI risk. AIDS Behav. 2017;21(10):3047–3056. 10.1007/s10461-017-1673-1
72. Lui CK, Chung PJ, Ford CL, Grella CE, Mulia N. Drinking behaviors and life course socioeconomic status during the transition from adolescence to adulthood among Whites and Blacks. J Stud Alcohol Drugs. 2015;76(1):68–79. 10.15288/jsad.2015.76.68
73. Zemore SE, Mulia N, Jones-Webb RJ, Liu H, Schmidt L. The 2008-2009 recession and alcohol outcomes: differential exposure and vulnerability for Black and Latino populations. J Stud Alcohol Drugs. 2013;74(1):9–20. 10.15288/jsad.2013.74.9
74. Kalousova L, Burgard SA. Unemployment, measured and perceived decline of economic resources: contrasting three measures of recessionary hardships and their implications for adopting negative health behaviors. Soc Sci Med. 2014;106:28–34. 10.1016/j.socscimed.2014.01.007
75. Tucker-Seeley RD, Selk S, Adams I, Allen JD, Sorensen G. Tobacco use among low-income housing residents: does hardship motivate quit attempts? Cancer Causes Control. 2015;26(11):1699–1707. 10.1007/s10552-015-0662-9
76. Kahn JR, Pearlin LI. Financial strain over the life course and health among older adults. J Health Soc Behav. 2006;47(1):17–31. 10.1177/002214650604700102
77. Centers for Medicare and Medicaid Services . A Guide to Using the Accountable Health Communities Health-Related Social Needs Screening Tool: Updated December 2023 Promising Practices and Key Insights. Accessed February 26, 2024. https://www.cms.gov/priorities/innovation/media/document/ahcm-screeningtool-companion
78. Michelmore K. The earned income tax credit and union formation: the impact of expected spouse earnings. Rev Econ Househ. 2018;16(2):377–406. 10.1007/s11150-016-9348-7
79. Doremus JM, Jacqz I, Johnston S. Sweating the energy bill: extreme weather, poor households, and the energy spending gap. J Environ Econ Manag. 2022;112:102609. 10.1016/j.jeem.2022.102609
80. Nord M, Kantor LS. Seasonal variation in food insecurity is associated with heating and cooling costs among low-income elderly Americans. J Nutr. 2006;136(11):2939–2944. 10.1093/jn/136.11.2939
81. Yang T-T. Family labor supply and the timing of cash transfers: evidence from the earned income tax credit. J Hum Resour. 2018;53(2):445–473. 10.3368/jhr.53.2.0115-6857R1
82. Tax Policy Center (TPC) . State EITC as a Percentage of Federal EITC. Accessed May 18, 2022. https://www.taxpolicycenter.org/statistics/state-eitc-percentage-federal-eitc
83. Urban Institute . State Earned Income Tax Credits. Accessed May 18, 2022. https://www.urban.org/policy-centers/cross-center-initiatives/state-and-local-finance-initiative/state-and-local-backgrounders/state-earned-income-tax-credits
84. Komro KA, Dunlap P, Sroczynski N, et al. Anti-poverty policy and health: attributes and diffusion of state earned income tax credits across US states from 1980 to 2020. PLoS One. 2020;15(11):e0242514. 10.1371/journal.pone.0242514
85. Ali MM, McClellan C, West KD, Mutter R. Medical marijuana laws, marijuana use, and opioid-related outcomes among women in the United States. Womens Health Issues. 2021;31(1):24–30. 10.1016/j.whi.2020.09.003
86. Hasin DS, Sarvet AL, Cerdá M, et al. US adult illicit cannabis use, cannabis use disorder, and medical marijuana laws: 1991-1992 to 2012-2013. JAMA Psychiatry. 2017;74(6):579–588. 10.1001/jamapsychiatry.2017.0724
87. Martins SS, Mauro CM, Santaella-Tenorio J, et al. State-level medical marijuana laws, marijuana use and perceived availability of marijuana among the general U.S. population. Drug Alcohol Depend. 2016;169:26–32. 10.1016/j.drugalcdep.2016.10.004
88. Mauro CM, Newswanger P, Santaella-Tenorio J, Mauro PM, Carliner H, Martins SS. Impact of medical marijuana laws on state-level marijuana use by age and gender, 2004–2013. Prev Sci. 2019;20(2):205–214. 10.1007/s11121-017-0848-3
89. Cerdá M, Mauro C, Hamilton A, et al. Association between recreational marijuana legalization in the United States and changes in marijuana use and cannabis use disorder from 2008 to 2016. JAMA Psychiatry. 2020;77(2):165–171. 10.1001/jamapsychiatry.2019.3254
90. Martins SS, Segura LE, Levy NS, et al. Racial and ethnic differences in cannabis use following legalization in US states with medical cannabis laws. JAMA Netw Open. 2021;4(9):e2127002. 10.1001/jamanetworkopen.2021.27002
91. Goldin J, Homonoff T, Javaid R, Schafer B. Tax filing and take-up: experimental evidence on tax preparation outreach and benefit claiming. J Public Econ. 2022;206:104550.
92. Guyton J, Langetieg P, Manoli D, Payne M, Schafer B, Sebastiani M. Reminders and recidivism: using administrative data to characterize nonfilers and conduct EITC outreach. Am Econ Rev. 2017;107(5):471–475. 10.1257/aer.p20171062
93. Metsch LR, Pollack HA. Welfare reform and substance abuse. Milbank Q. 2005;83(1):65–99. 10.1111/j.0887-378X.2005.00336.x

Associated Data

Data Citations

References

National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AM0015: R01_AM0015: Because of shortage of money were you unable to pay important bills on time in the past 30 days. Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AM0015?archive=nahdap
National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AG9105: R01_AG9105: Ever smoked part or all of a cigar, cigarillo or filtered cigar with marijuana in it (blunt). Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AG9105?archive=nahdap
National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AX0089_02: R01_AX0089_02: Ever used prescription drugs not prescribed to you: Painkillers, sedatives or tranquilizers. Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AX0089_02?archive=nahdap
National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AX0081_02: R01_AX0081_02: How long since last used: Painkillers, sedatives or tranquilizers. Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AX0081_02?archive=nahdap
National Addiction & HIV Data Archive Program; Inter-University Consortium for Political and Social Research (ICPSR) . R01_AM0031: R01_AM0031: Nonresponse follow-up probe: Total household income above or below $50,000 during the past 12 months. Accessed March 11, 2024. https://www.icpsr.umich.edu/web/NAHDAP/studies/36231/datasets/1011/variables/R01_AM0031?archive=nahdap

Studies

Conditions

Cannabinoids and active compounds

Related research

Abstract

Introduction

Methods

Results

Conclusion

Untitled section

Article notes

Untitled section

Author notes

Introduction

Methods

Data

Measures

Financial strain (yes/no)

Past-month cannabis or CNS depressant use (yes/no)

Estimated federal EITC refund (yes/no)

Federal EITC disbursement period (yes/no)

Demographic characteristics

Statistical analysis

Sensitivity analyses

Limitations

Results

Sample description

Federal EITC eligibility, financial strain, and drug use—difference in difference estimates

Sensitivity analyses

Varying federal EITC refund threshold

Varying eligibility assumptions

Varying cannabis and CNS depressant study sample

Restricting study sample to people earning ≤$100 000

Discussion

Conclusion

Contribution statement

Supplementary Material

Contributor Information

Supplementary material

Funding

Conflicts of interest

Notes

Untitled section

References

Associated Data

Data Citations

References

Supplementary Materials

Additional materials