Health Insurance, Health Literacy, Health Services Accessibility, Informed Decision Making
Conditions
Keywords
Health insurance, Health services accessibility, Decision making, Health literacy
Brief summary
The overall purpose of the study is to better understand how the investigators previously developed decision support (DS) tool can help people make decisions about health insurance plans available through the federal exchanges created by the Affordable Care Act (ACA). The investigators will evaluate the DS tool compared to the federal government website. The investigators will also evaluate the feasibility of disseminating this tool. There are two primary aims to be completed in this project: (1) examine the reach and effectiveness of the health insurance DS tool; and (2) collect stakeholders' feedback to improve the likelihood of implementation of the DS tool.
Detailed description
First, for Aim 1 part 1, the investigators will recruit 40 key stakeholders (uninsured participants, health providers, community advisors, and health policy experts) to refine the DS tool through a series of individual semi-structured interviews. Next, the investigators will use their feedback to program the DS strategies into an online DS tool. After its initial production, the tool will be pilot tested with 30 individuals to assess readability, message clarity, format, and function of the tool as well as to test the randomized trial study procedures. Then, for Aim 1 part 2, the investigators will test the DS tool in a randomized trial with 362 participants eligible for the Affordable Care Act (ACA) exchanges. Half will use the DS tool and the other half will use the federal government site to learn about the exchanges. The investigators will use computerized random assignment to assign participants to study condition, after which they will all complete the same short survey. For Aim 2, the investigators will collect 40 stakeholders' feedback on likelihood of adoption and implementation of the DS tool to plan for dissemination and implementation. Stakeholders will be asked both open-ended and closed-ended questions in order to gather feedback about delivering the DS tool. These stakeholders will be different from those who were interviewed in Aim 1 as to ensure broader applicability of the DS tool.
Interventions
BEHAVIORALDecision Aid (DA)
Participants will be shown (on a computer) a targeted, web-based decision aid focused on the topic of health insurance plans in addition to usual care.
Sponsors
National Institutes of Health (NIH)
National Institute on Minority Health and Health Disparities (NIMHD)
Washington University School of Medicine
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
PREVENTION
Masking
SINGLE (Subject)
Eligibility
Sex/Gender
ALL
Age
18 Years to 64 Years
Healthy volunteers
Yes
Inclusion criteria
* Must be without health insurance currently; or * Must be enrolled in one of the ACA exchange plans in the past 1 year; and * Must speak English
Exclusion criteria
* Currently has health insurance that was not provided by being enrolled in one of the ACA exchange plans in the past year * Currently has health insurance that is provided through public insurance or private employer-based insurance * Does not speak English
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Knowledge Score (% Correct) | Completed immediately after reviewing Decision Aid tool, taking about 5 minutes to complete. | 8 questions developed in researchers' past work. Assessed health insurance knowledge. |
| Decision Self-efficacy | Completed immediately after reviewing Decision Aid tool, taking about 5 minutes to complete. | The decision self-efficacy (DSE) scale measured participants' perceived ability to understand insurance info and resist unwanted decision pressure. The 6 items on the DSE scale were each rated on a 3-point scale (0=Not confident; 2=A little confident; 4=Very confident). The sum of the DSE items was divided by 6 and multiplied by 25 to obtain a score on a 0-100 scale. Higher values indicate more confidence in one's decision-making ability. |
| Confidence in Choice | Completed immediately after reviewing Decision Aid tool, taking about 5 minutes to complete. | The 4-item SURE (Sure of myself; Understand information; Risk-benefit ratio; Encouragement) decisional conflict scale assessed confidence in plan choice. Each item could be answered dichotomously (1=yes; 0=no). Responses were summed and a group average was obtained. Higher SURE values indicate more confidence in choice. The scale ranged from 0-4. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Improvements in HILM 1 (Health Insurance Literacy Measure) | Pre-intervention and post-intervention | Assessed confidence estimating costs of care.Improvement in HILM was defined as moving from not confident pre-intervention to a little confident or very confident post-intervention, or from a little confident pre-intervention to very confident post-intervention. |
| Improvements in HILM 2 (Health Insurance Literacy Measure) | Pre-intervention and post-intervention | Assessed confidence understanding terms.Improvement in HILM was defined as moving from not confident pre-intervention to a little confident or very confident post-intervention, or from a little confident pre-intervention to very confident post-intervention. |
| Intended Choice Metal Level | Completed immediately after reviewing Decision Aid tool, taking about 5 minutes to complete. | Participants indicated the plan they would choose that day. We categorized plans by governmental classifications of metal level (catastrophic, bronze, silver, gold). |
Countries
United States
Participant flow
Recruitment details
Recruitment occurred between 10/2015 - 1/2016 via community events, online advertisements, social service organizations, and the recruitment/retention arm of our Center for Community-Engaged Research.
Pre-assignment details
482 participants were assessed for eligibility. Participants were excluded (n=154) if they did not meet the inclusion criteria (n=82) or could not be reached (n=65).
Participants by arm
| Arm | Count |
|---|---|
| Decision Aid (DA) The decision aid (DA) will be provided to participants randomized to the experimental/intervention group.
Decision Aid (DA): Participants will be shown (on a computer) a targeted, web-based decision aid focused on the topic of health insurance plans in addition to usual care. | 164 |
| Control Participants randomized to the control group will receive usual care. They will be directed to the healthcare.gov website and asked to follow the prompts to view and select (if applicable) a health insurance plan. | 163 |
| Total | 327 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Technical failure | 1 | 0 |
Baseline characteristics
| Characteristic | Control | Total | Decision Aid (DA) |
|---|---|---|---|
| Age, Continuous | 41.4 years STANDARD_DEVIATION 12.5 | 42.3 years STANDARD_DEVIATION 12.9 | 43.1 years STANDARD_DEVIATION 13.2 |
| Chronic Conditions - Participant + Dependents (n=107) | NA Number chronic conditions | NA Number chronic conditions | 2.69 Number chronic conditions STANDARD_DEVIATION 2.35 |
| Chronic Conditions - Participant (n=207) | 2.17 Number chronic conditions STANDARD_DEVIATION 1.35 | 2.22 Number chronic conditions STANDARD_DEVIATION 1.43 | 2.27 Number chronic conditions STANDARD_DEVIATION 1.53 |
| Education College + | 49 participants | 90 participants | 41 participants |
| Education HS or GED | 99 participants | 210 participants | 111 participants |
| Education Less than HS | 15 participants | 27 participants | 12 participants |
| % Federal Poverty Line < 100 | 94 participants | 166 participants | 72 participants |
| % Federal Poverty Line 100-249 | 47 participants | 111 participants | 64 participants |
| % Federal Poverty Line 250-399 | 10 participants | 25 participants | 15 participants |
| % Federal Poverty Line 400+ | 8 participants | 21 participants | 13 participants |
| Health Literacy (SILS) Adequate | 128 participants | 267 participants | 139 participants |
| Health Literacy (SILS) Limited | 34 participants | 59 participants | 25 participants |
| Insurance Status (Uninsured) Insured | 70 participants | 149 participants | 79 participants |
| Insurance Status (Uninsured) Uninsured | 93 participants | 178 participants | 85 participants |
| Objective Numeracy (Lipkus) | 2.3 units on a scale | 2.4 units on a scale | 2.6 units on a scale |
| Race/Ethnicity, Customized African American only | 107 participants | 203 participants | 96 participants |
| Race/Ethnicity, Customized Caucasian only | 41 participants | 87 participants | 46 participants |
| Race/Ethnicity, Customized Hispanic | 4 participants | 12 participants | 8 participants |
| Race/Ethnicity, Customized Not Hispanic | 159 participants | 315 participants | 156 participants |
| Race/Ethnicity, Customized Other (Including Mixed) | 15 participants | 37 participants | 22 participants |
| Sex: Female, Male Female | 94 Participants | 191 Participants | 97 Participants |
| Sex: Female, Male Male | 69 Participants | 136 Participants | 67 Participants |
| Smoking status - Dependent (n=124) No | 48 participants | 105 participants | 57 participants |
| Smoking status - Dependent (n=124) Yes | 9 participants | 19 participants | 10 participants |
| Smoking status - Participant No | 116 participants | 225 participants | 109 participants |
| Smoking status - Participant Yes | 47 participants | 102 participants | 55 participants |
| Subjective Numeracy (SNS) | 3.6 units on a scale | 3.8 units on a scale | 3.9 units on a scale |
| Total number covering 1 | 106 participants | 203 participants | 97 participants |
| Total number covering 2 | 21 participants | 53 participants | 32 participants |
| Total number covering 3+ | 36 participants | 71 participants | 35 participants |
| Urbanicity Rural | 11 participants | 20 participants | 9 participants |
| Urbanicity Suburban | 99 participants | 196 participants | 97 participants |
| Urbanicity Urban | 53 participants | 111 participants | 58 participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 0 / 165 | 0 / 163 |
| serious Total, serious adverse events | 0 / 165 | 0 / 163 |
Outcome results
Primary
Confidence in Choice
The 4-item SURE (Sure of myself; Understand information; Risk-benefit ratio; Encouragement) decisional conflict scale assessed confidence in plan choice. Each item could be answered dichotomously (1=yes; 0=no). Responses were summed and a group average was obtained. Higher SURE values indicate more confidence in choice. The scale ranged from 0-4.
Time frame: Completed immediately after reviewing Decision Aid tool, taking about 5 minutes to complete.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Decision Aid (DA) | Confidence in Choice | 3.5 units on a scale | Standard Deviation 0.9 |
| Control | Confidence in Choice | 2.9 units on a scale | Standard Deviation 1.3 |
p-value: <0.001Regression, Linear
Comparison: Examined the relation between group and categorical outcomes, controlling for % Federal Poverty Level (400 or greater, 250-399, 100-249, less than 100).p-value: 0.82Regression, Linear
Comparison: Controlling for Numeracy.p-value: 0.15Regression, Linear
Primary
Decision Self-efficacy
The decision self-efficacy (DSE) scale measured participants' perceived ability to understand insurance info and resist unwanted decision pressure. The 6 items on the DSE scale were each rated on a 3-point scale (0=Not confident; 2=A little confident; 4=Very confident). The sum of the DSE items was divided by 6 and multiplied by 25 to obtain a score on a 0-100 scale. Higher values indicate more confidence in one's decision-making ability.
Time frame: Completed immediately after reviewing Decision Aid tool, taking about 5 minutes to complete.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Decision Aid (DA) | Decision Self-efficacy | 81.5 units on a scale | Standard Deviation 22.5 |
| Control | Decision Self-efficacy | 73.0 units on a scale | Standard Deviation 20.4 |
p-value: 0.002Regression, Linear
Comparison: Examined the relation between group and categorical outcomes, controlling for % Federal Poverty Level (400 or greater, 250-399, 100-249, less than 100).p-value: 0.3Regression, Linear
Comparison: Controlling for Numeracy.p-value: 0.82Regression, Linear
Primary
Knowledge Score (% Correct)
8 questions developed in researchers' past work. Assessed health insurance knowledge.
Time frame: Completed immediately after reviewing Decision Aid tool, taking about 5 minutes to complete.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Decision Aid (DA) | Knowledge Score (% Correct) | 77.6 percentage of 8 items correctly answered | Standard Deviation 18.2 |
| Control | Knowledge Score (% Correct) | 58.8 percentage of 8 items correctly answered | Standard Deviation 21 |
p-value: <0.001Regression, Linear
Comparison: Examined the relation between group and categorical outcomes, controlling for % Federal Poverty Level (400 or greater, 250-399, 100-249, less than 100).p-value: 0.16Regression, Linear
Comparison: Controlling for Numeracy.p-value: <0.001Regression, Linear
Secondary
Improvements in HILM 1 (Health Insurance Literacy Measure)
Assessed confidence estimating costs of care.Improvement in HILM was defined as moving from not confident pre-intervention to a little confident or very confident post-intervention, or from a little confident pre-intervention to very confident post-intervention.
Time frame: Pre-intervention and post-intervention
Population: One participant in healthcare.gov (Control) condition did not finish HILM (n=162)
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Decision Aid (DA) | Improvements in HILM 1 (Health Insurance Literacy Measure) | No | 98 participants |
| Decision Aid (DA) | Improvements in HILM 1 (Health Insurance Literacy Measure) | Yes | 66 participants |
| Control | Improvements in HILM 1 (Health Insurance Literacy Measure) | Yes | 49 participants |
| Control | Improvements in HILM 1 (Health Insurance Literacy Measure) | No | 113 participants |
p-value: 0.195% CI: [0.92, 2.36]Regression, Logistic
Comparison: Examined the relation between group and categorical outcomes, controlling for % Federal Poverty Level (400 or greater, 250-399, 100-249, less than 100).p-value: 0.26Regression, Logistic
Comparison: Controlling for Numeracy.p-value: 0.58Regression, Linear
Secondary
Improvements in HILM 2 (Health Insurance Literacy Measure)
Assessed confidence understanding terms.Improvement in HILM was defined as moving from not confident pre-intervention to a little confident or very confident post-intervention, or from a little confident pre-intervention to very confident post-intervention.
Time frame: Pre-intervention and post-intervention
Population: One participant in healthcare.gov (Control) condition did not finish HILM (n=162)
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Decision Aid (DA) | Improvements in HILM 2 (Health Insurance Literacy Measure) | Yes | 88 participants |
| Decision Aid (DA) | Improvements in HILM 2 (Health Insurance Literacy Measure) | No | 76 participants |
| Control | Improvements in HILM 2 (Health Insurance Literacy Measure) | Yes | 51 participants |
| Control | Improvements in HILM 2 (Health Insurance Literacy Measure) | No | 111 participants |
p-value: <0.00195% CI: [1.58, 4.01]Regression, Logistic
Comparison: Examined the relation between group and categorical outcomes, controlling for % Federal Poverty Level (400 or greater, 250-399, 100-249, less than 100).p-value: 0.39Regression, Logistic
Comparison: Controlling for Numeracy.p-value: 0.57Regression, Linear
Secondary
Intended Choice Metal Level
Participants indicated the plan they would choose that day. We categorized plans by governmental classifications of metal level (catastrophic, bronze, silver, gold).
Time frame: Completed immediately after reviewing Decision Aid tool, taking about 5 minutes to complete.
Population: Intended plan choice metal level data were not available for 2 participants in SMHP (Show Me My Health Plans decision aid) condition (n=162) and for 11 participants in healthcare.gov (Control) condition (n=152)
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Decision Aid (DA) | Intended Choice Metal Level | Bronze | 41 participants |
| Decision Aid (DA) | Intended Choice Metal Level | Gold | 0 participants |
| Decision Aid (DA) | Intended Choice Metal Level | Silver | 121 participants |
| Decision Aid (DA) | Intended Choice Metal Level | Platinum | 0 participants |
| Decision Aid (DA) | Intended Choice Metal Level | Catastrophic | 0 participants |
| Control | Intended Choice Metal Level | Platinum | 0 participants |
| Control | Intended Choice Metal Level | Catastrophic | 5 participants |
| Control | Intended Choice Metal Level | Bronze | 108 participants |
| Control | Intended Choice Metal Level | Silver | 30 participants |
| Control | Intended Choice Metal Level | Gold | 9 participants |
p-value: <0.00195% CI: [5.94, 17.95]Regression, Logistic
Comparison: Examined the relation between group and categorical outcomes, controlling for % Federal Poverty Level (400 or greater, 250-399, 100-249, less than 100).p-value: 0.001Regression, Logistic
Comparison: Controlling for numeracy.p-value: 0.46Regression, Linear