Table 2.
Detailed Description of Included Studies
| Study | Year | Study type; variable adjustment in analysis | Population (including N) | HIL assessment | Type of utilization outcome | Mode of utilization assessment | Study findings | Magnitude of association/effect |
|---|---|---|---|---|---|---|---|---|
| Alesci et al.17 | 2004 | Intervention; multivariable adjustment | Smokers in Minnesota insurance plan (n=1930) | Self-reported knowledge of insurance plan’s smoking cessation benefit and 3-item questionnaire about benefit | Tobacco cessation treatment; tobacco cessation | Survey | -Knowledge of plan benefit greater in control vs. intervention group with targeted mailed communication but no significant differences in tobacco benefit utilization between two groups | Bupropion treatment in past 12 months 23.1% in control vs. 24.6% in intervention group (p=0.92); any nicotine replacement treatment use in past 12 months 25.9% vs. 26.9% in control vs. intervention group (p=0.26) |
| Fox et al.18 | 2001 | Intervention; multivariable regression | Women age > 65 in southern California (n=917, 922) | Knowledge that screening mammograms are covered by Medicare | Mammography in last 2 years | Survey | -Mailing to increase knowledge that Medicare pays for breast cancer screening led to increased mammogram use among minorities who received the intervention relative to control group. | -Black women (OR 1.97) |
| -Hispanic women (OR 2.33) | ||||||||
| -White women (OR 1.04) | ||||||||
| -However, the intervention did not increase screening amongst white women | ||||||||
| Kneipp et al.19 | 2011 | Intervention; multivariable regression | Women with chronic health conditions receiving TANF benefits in Florida (n=285) | 20-item questionnaire related to Medicaid coverage | New mental health visit, primary care routine, or preventive visit | Survey | -Subjects receiving intervention with public health nurse case manager who taught HIL and other topics were more likely to have a new mental health visit but not more likely to have preventive care visit | -New mental health visit (OR 1.92) |
| -Preventative care visit (OR 1.50) | ||||||||
| Ehresmann et al.20 | 2001 | Cross-sectional, quantitative; multivariable regression | Adults age > 65 in Minnesota (n=353) | Survey assessing awareness of Medicare coverage for pneumococcal vaccine | Pneumococcal vaccination | Survey | -Awareness that Medicare covers pneumococcal vaccine associated with receipt of pneumococcal vaccine | OR 5.1 |
| Hsu et al.24 | 2008 | Cross-sectional, quantitative; multivariable regression | Medicare Advantage beneficiaries (age>65) in Kaiser Permanente system, California (n=1040) | After defining coverage gap, participants were asked whether their drug plan included such a gap, at what amount their gap began and ended, and how much they paid before, during, and after the gap | Set of medication utilization behaviors, including cost-coping behaviors (e.g., switch to cheaper med) and decreased adherence (e.g., skip pills, didn’t fill) | Interviews | -Compared with beneficiaries unaware of having a Medicare prescription drug coverage gap, those who were aware more frequently reported any behavior change, including switching to a cheaper drug | -Any behavioral change: difference of 11.3% |
| -Switching to cheaper drug: difference of 7.4% | ||||||||
| -No significant association with decreased med adherence | ||||||||
| James et al.32 | 2018 | Cross-sectional, quantitative; multivariable regression | College students at a public university in Florida (n=1450) | KFF knowledge scale and HILM | Number of visits to student health center, number of visits to a doctor’s office | Survey | -Knowledge (KFF) not significantly associated with utilization including the use of student health services | HILM score associated with overall utilization: OR 1.91 |
| -Higher insurance self-efficacy (HILM) was associated with greater probability of overall utilization but not with student health services | ||||||||
| McDonnell et al.30 | 2013 | Cross-sectional, quantitative; multivariable regression | ED visitors (or parents of pediatric patients) in Utah (n=4136) | Survey asking if respondents were aware of a law that ED must examine and treat, regardless of insurance status or ability to pay | ED visits in prior year | Survey | -Knowledge of the Emergency Medical Treatment and Labor Act associated with any ED utilization | -Any ED utilization: OR 1.44 |
| -High-frequency ED utilization: OR 1.69 | ||||||||
| -Knowledge of the Emergency Medical Treatment and Labor Act also associated with high-frequency ED utilization of at least 5 visits in last year | ||||||||
| McMenamin et al.36 | 2006 | Cross-sectional, quantitative; multivariable regression | Current smokers or recent quitters, ages 18–64 with Medicaid in the USA (n=820) | Questions regarding knowledge of coverage for several tobacco dependency treatments under their state Medicaid program | Use of tobacco dependency treatment | Survey | -Knowledge of Medicaid coverage associated with greater use of tobacco dependency treatments, including any medication and use of quitline | -Use of tobacco dependency treatments: OR 3.0 |
| -Use of quitline: OR 3.5 | ||||||||
| Morgan et al.22 | 2008 | Cross-sectional, quantitative; multivariable regression | Medicare beneficiaries in the USA (n=2997) | Subjects asked how familiar they were with Medicare and Medicare Advantage | Clinic visits, ED visits, hospital admissions in the past year | Survey | Lower familiarity with Medicare associated with: | -Clinic visits: OR 0.67 |
| -Prescription drug use: OR 0.58 | ||||||||
| -More frequent ED visits: 2.88 | ||||||||
| -Lower likelihood of clinic visits | ||||||||
| -Delayed clinic visits: OR 1.72 | ||||||||
| -Lower likelihood of prescription drug use | ||||||||
| -Delayed ED visits: OR 2.07 | ||||||||
| -Higher likelihood of more frequent ED visits | ||||||||
| -Delayed inpatient care: OR 2.60 | ||||||||
| -Non-significant association with greater inpatient care | ||||||||
| -Higher likelihood of delays due to cost for clinic visits, ED visits, and inpatient care | ||||||||
| Obrist et al.37 | 2014 | Cross-sectional, quantitative; multivariable regression | Breast cancer patients at medical center in Ghana (n=117) | Interview | Completion of medically recommended breast cancer treatment | Medical records | -Patients who completed treatment were significantly more likely to understand what their insurance covered regarding surgery, radiation, chemotherapy, and other medications than those who did not complete treatment | -89.4% of patients who completed treatment understood coverage |
| -67.74% of those who did not complete treatment understood coverage | ||||||||
| -Awareness of coverage associated with completion of treatment: OR 11.859 | ||||||||
| -Those who were unaware of their insurance coverage policy for breast care had higher odds of not completing their prescribed breast cancer treatment protocol | ||||||||
| Parente et al.38 | 2005 | Cross-sectional, quantitative; multivariable regression | Medicare beneficiaries (age>65) in the USA (n=7473; with n=4296 women for mammogram analysis) | Medicare beneficiary survey with test of knowledge of Medicare coverage for flu shot and mammography | Obtaining flu shot, mammogram | Medicare claims data | -In both analytic models, individuals who had knowledge of the Medicare flu shot benefit had more flu shots in the 12-month period studied | -Model 1: 0.092 more flu shots per year |
| -Model 2: 0.182 more flu shots per year | ||||||||
| Piette and Heisler23 | 2006 | Cross-sectional, quantitative; multivariable regression | Adults age > 50 in the USA who had prescription drug coverage and at least one chronic condition (n=3119) | Survey questions assessing understanding of usual cost per prescription and knowledge about drug coverage's spending limits | Medication adherence (more specifically, cost-related) | Survey | -Low HIL regarding drug coverage caps associated with cost-related medication nonadherence | -Low HIL regarding drug coverage gaps: OR 1.7 |
| -Low HIL regarding usual out-of-pocket costs: OR 1.0 | ||||||||
| -Low HIL regarding usual out-of-pocket costs for medication not associated with cost-related medication nonadherence | ||||||||
| Reed et al.39 | 2012 | Cross-sectional, quantitative; multivariable regression | Adults (ages 18–64) with a high-deductible health plan/health savings account through Kaiser Permanente in California (n=456) | Questions assessing whether preventive office visits (e.g., annual physicals); non-preventive doctor’s office visits; preventive medical tests; and non-preventive medical tests applied toward deductible; general knowledge of deductible | Whether the amount they would have to pay caused them to delay or avoid any preventive office visits or screening tests | Survey | -Those who mistakenly thought that the deductible applied to all office visits were more likely to delay or avoid a preventive office visit because of cost than those who correctly understood the cost-sharing scheme | -23.8% of those who mistakenly thought that deductibles applied to all office visits delayed/avoided a preventative office visit, and |
| -18.1% of those who mistakenly thought that the deductible did not apply to either preventive or non-preventive visits were more likely to delay or avoid a preventive office visit, compared to | ||||||||
| -7.8% of those who knew that preventative office visits had no out-of-pocket costs delayed/avoided care | ||||||||
| -Those who mistakenly thought that the deductible did not apply to either preventive or non-preventive visits were more likely to delay or avoid a preventive office visit because of cost (18.1%) compared to those who correctly understood the cost-sharing scheme | ||||||||
| -No significant association between HIL and delay or avoidance of tests | ||||||||
| Sawyer et al.40 | 2018 | Cross-sectional, quantitative; multivariable regression | Women (ages 18–44) MTurk online survey takers in the USA (n=1083) | Knowledge questions regarding covered essential health benefits under the Affordable Care Act, and level of certainty that each response was correct | Preventive reproductive health services | Surveys | Knowledge of Affordable Care Act mandated coverage was associated with greater utilization of: | -Well-woman exams: OR 1.109 |
| -Pelvic exams: OR 1.128 | ||||||||
| -Breast exams: 1.075 | ||||||||
| -STI testing: OR 1.106 | ||||||||
| -Well-woman exams | ||||||||
| -HPV vaccination: 1.088 | ||||||||
| -Pelvic exams | ||||||||
| -Breast exams | ||||||||
| -STI testing | ||||||||
| -HPV vaccination | ||||||||
| -No significant association with receiving a Pap smear | ||||||||
| Smith et al.25 | 2018 | Cross-sectional, quantitative; multivariable regression | MTurk online survey takers in the USA (n=470) | Knowledge of health insurance terms: true/ false questions; single item regarding Affordable Care Act coverage of preventive services without out-of-pocket costs | Delaying/avoiding any care, delaying/avoiding common health care services (3 preventive and 3 non-preventive services) in the past 12 months | Survey | -Those who delayed/avoided preventive care had less general knowledge about health insurance | -General knowledge about health insurance: 67% (those who delayed care) vs. 75% (those who did not delay care) |
| -Knowledge | ||||||||
| that preventative care is covered with no out-of-pocket costs: 24% (those who delayed care) vs. 42% (those who did not delay care) | ||||||||
| -Those who delayed/avoided care were less likely to know that preventive care is covered at no out-of-pocket cost | ||||||||
| -Knowledge that preventative care is covered with no out-of-pocket costs associated with less delaying/avoiding care: OR 0.444 | ||||||||
| -General knowledge about health insurance associated with less avoidance/delay in care: OR 0.989 | ||||||||
| -Those who knew that preventive care was covered at no out-of-pocket cost were less likely to delay/avoid any care | ||||||||
| -Individuals were more likely to avoid/delay preventive care if they had lower health insurance knowledge or did not know that preventive care is covered at no out-of-pocket cost | ||||||||
| Tipirneni et al.26 | 2018 | Cross-sectional, quantitative; multivariable regression | Adults with health insurance, MTurk online survey takers in the USA (n=506) | HILM | Use of preventive and non-preventive services; delayed or forgone care owing to perceived costs (questions about specific services) | Survey | -Those with lower HILM were more likely than those with higher HILM to avoid preventive services | -23.8% of those with lower HILM avoided preventative care vs. 11.4% of those with higher HILM avoided preventative care |
| -Those with lower HILM were more likely than those with higher HILM to avoid non-preventative services | ||||||||
| -19.3% of those with lower HILM avoided non-preventative care vs. 12.6% of those with higher HILM avoided non-preventative care -Each SD increase in HILM associated with less | ||||||||
| delayed/foregone preventative care due to cost: OR 0.61 | ||||||||
| -Each SD increase in HILM associated with less delayed/foregone non-preventative care: OR 0.71 | ||||||||
| -HILM score associated with utilizing preventative services: OR 1.57 | ||||||||
| -Each 12-point increase in HILM score (~1 SD) was associated with lower likelihood of delayed or forgone care owing to cost for preventive care | ||||||||
| -Each 12-point increase in HILM score (~1 SD) was associated with lower likelihood of delayed or forgone care owing to cost for non-preventive care | ||||||||
| -HILM score was associated with a higher likelihood of preventive services use, but not with non-preventive services use | ||||||||
| Webster27 | 2011 | Cross-sectional, quantitative; multivariable regression | Adults age >65 in the USA (N=30,002) from the National Health Interview Survey | Knowledge questions related to type of Medicare coverage, whether enrolled in Medicare Advantage or HMO, whether referrals needed for specialty care, and whether paying for supplemental coverage | Number of medical office visits, ED visits, time speaking with health professional, and surgeries in past 12 months | Survey | -Low HIL associated with: | -Number of office visits (3.6 vs. 3.3) |
| ---Greater number of medical office visits | -Time since last talked with health professional: 1.3 vs. 1.2 on time scale ranging from 6 months or less to never | |||||||
| -Likelihood of talking with a health professional: 44.1% vs. 47.1% | ||||||||
| ---More time since last talked with health professional | ||||||||
| -Likelihood of surgery: 18.0% vs. 20.5% | ||||||||
| ---Lower likelihood of talking with health professional | ||||||||
| ---Lower likelihood of surgery | ||||||||
| ---No association with ED visits in the past twelve months | ||||||||
| Burns et al.31 | 2005 | Cross-sectional, quantitative; adjusted for survey weights, no covariates | Wisconsin state employees in state-sponsored health plan (2001 n=5609; 2002 n=6518) | Yes no question on whether insurance covers list of specific tobacco cessation therapies | Use of tobacco cessation medications | Survey | -HIL related to coverage of tobacco cessation therapies associated with utilization of tobacco cessation therapies | -39.6% utilization among those aware vs. 3.5% among those unaware of benefit |
| Lischko and Burgess29 | 2010 | Cross-sectional, quantitative; multivariable regression | Massachusetts state employees (age <65) continuously enrolled in health plan for 3 years (n=1322) | Knowledge questions regarding co-pays for different services | ED or office visits | Claims data, Survey | -Greater knowledge of costs was associated with utilization of office visits | -0.0923 more office visits for those with the highest level of cost-sharing knowledge vs. no knowledge |
| -Those who overestimated or accurately knew co-pays were more likely to delay/avoid care than those who underestimated co-pays | ||||||||
| -Those who overestimated (OR 2.47) or accurately knew co-pays (OR 1.87) were more likely to delay/avoid office visits | ||||||||
| -Knowledge of specific co-pays had no association with office visits or ED utilization | ||||||||
| Edward et al.21 | 2018 | Cross-sectional, mixed methods; unadjusted | Latinx (primarily Spanish speaking) adults attending health insurance enrollment event (n=139) | Subjects asked to define health insurance terms (copay, premium, deductible) | Whether participants had accessed health care in the USA | Surveys, semi-structured interviews | No association between HIL and time since last accessed health care | -N/A |
| Nobles et al.28 | 2019 | Cross-sectional, mixed methods; unadjusted | Undergraduate and graduate students at a single university in Virginia (n=455) | Knowledge of health insurance vocabulary and ability to apply knowledge to determine cost-sharing, self-rated understanding of insurance terminology | Delayed/forgone medical care because of confusion about health insurance plan | Survey | -Low HIL associated with delayed or forgone care | -24.4% indicated that lack of understanding of their health insurance stopped or delayed them from seeking medical care in the past |
Abbreviations: HIL, health insurance literacy; OR, odds ratio; TANF, Temporary Assistance for Needy Families; HILM, Health Insurance Literacy Measure, a subjective measure of confidence in health insurance decision-making; KFF, Kaiser Family Foundation objective measure of health insurance knowledge; MTurk, Amazon Mechanical Turk; STI, sexually transmitted infection; HMO, health maintenance organization; ED, Emergency Department
Included studies are sorted by study type and presented in the same order as in Table 1