Asthma in Families Facing Out-of-pocket Requirements Due to COVID-19

In addition to its impact on health, the COVID-19 pandemic has led to increased unemployment
and loss of employer-sponsored insurance coverage. Obtaining coverage can be challenging and
eligibility for public programs and subsidies can be limited, and those who do not qualify
can face steep premiums, high-deductibles, and high out-of-pocket costs. Disruptions to
employment and insurance coverage during the pandemic threaten to negatively affect asthma
care and outcomes.

Our parent project, Asthma in Families Facing Out-of-pocket Requirements with Deductibles
(AFFORD), found that patients with asthma may be particularly vulnerable to insurance-related
cost barriers and challenges navigating health insurance. Together with the Asthma and
Allergy Foundation of America (AAFA), investigators developed an asthma chat bot to help
patients with asthma navigate insurance benefits and optimize health care decisions. The chat
bot is an artificial intelligence-enabled interactive online tool that can answer
insurance-related questions and provide information on coverage and how to find lower-cost
alternatives for asthma care.

In this supplement, the investigators propose to capitalize on the work of AFFORD to
understand and address the insurance and health care cost challenges faced by patients with
asthma who lose employer-sponsored coverage due to COVID-19. Specifically, our Aims are: 1)
to conduct a pilot randomized controlled trial to evaluate the effectiveness and
implementation outcomes (e.g. feasibility, acceptability) of an existing insurance navigation
intervention, including the chat bot, to help patients with asthma regain coverage after the
loss of job-related insurance during the COVID-19 pandemic; 2) to qualitatively explore the
experiences of Aim 1 participants to understand barriers and facilitators to accessing
coverage and asthma care more broadly during the COVID-19 pandemic

Findings will provide evidence about the effectiveness of patient-centered technological
strategies to obtain coverage and maintain access to affordable asthma care and can inform
ongoing and future decision making in response to the COVID-19 pandemic and other public
health and economic threats.

Study Setting

Study participants will be recruited through the AAFA and Harvard Pilgrim Health Care. As in
AFFORD, the proposed study will take advantage of AAFA's wide national reach and role as a
trusted information source to recruit a national sample of patients with asthma or parents of
children with asthma. We will also recruit through Harvard Pilgrim Health Care, a non-profit
health plan that offers insurance plans in Massachusetts, Maine, New Hampshire, and
Connecticut

Aim 1

In this aim, the investigators will conduct a pilot randomized controlled trial of patients
with asthma or parents of children with asthma who lost employer-sponsored coverage during
the COVID-19 pandemic. Potential study subjects will be recruited through AAFA and Harvard
Pilgrim. Eligible participants will be randomized into one of two study groups: 1) an
intervention group that receives access to an asthma chat bot and personalized insurance
navigation (n=145), or 2) a wait-list control group (n=285). At study entry, participants
will receive an online baseline survey to assess outcomes related to insurance coverage,
asthma severity and control, medication use and adherence, asthma care access and
affordability, and COVID-19 related illness. The intervention group will be offered the chat
bot and AAFA navigation services. One month later, the investigators will field a follow-up
survey to measure short-term changes in outcomes; for the intervention group, this survey
will also measure implementation outcomes including uptake, feasibility, acceptability, and
usability. Four months later, the investigators will field a second follow-up survey to
measure longer-term changes in outcomes. Control subjects will be offered the chat bot after
completion of the follow-up surveys.

Eligible subjects will include adults aged 18-64 years who have asthma or have a child with
asthma aged 4-17 years, and lost employer-sponsored health insurance after February 2020.
Potential subjects will be recruited through AAFA, who will send out a description of the
study to potential subjects through its online community, email listserv, website, Facebook
page, and newsletter. Potential subjects will also be recruited through Harvard Pilgrim.
Members who disenrolled from employer-sponsored plans after February 2020 will be identified,
and those aged 4-64 with an International Classification of Diseases (ICD)-10 diagnosis code
for asthma in claims data after January 1, 2019 will be contacted by email and invited to
participate. The investigators will collect information on race/ethnicity and select a sample
representative of the racial/ethnic distribution of patients with asthma, with no more than
60% non-Hispanic White participants selected for the sample.

The investigators will conduct a pilot randomized controlled trial where they will randomize
participants who lost coverage to: 1) be offered a combination of automated chat bot and
personalized outreach provided by AAFA-affiliated navigators (n=145), or 2) be a wait-list
control (n=285). The investigators will use a wait-list, usual care control group as the
comparator given that there is no other standard treatment or practice for people with asthma
who lose insurance coverage to which to compare our navigation intervention. After
randomization, participants will be given a RedCap link for a baseline survey.

After completion of the baseline survey, intervention subjects will be sent an email with
information about AAFA's asthma chat bot and a link to access it and information about AAFA's
insurance navigation program, how it can help with finding coverage and managing asthma
costs, and how to access it via group and private messaging within AAFA's asthma community
platform with the availability of telephonic follow-up from AAFA navigators. Subjects can
join the private group where they can ask questions and share resources. They will be able to
send messages to the AAFA navigator to communicate about insurance issues, access to asthma
care, and assistance with asthma costs. The navigator will offer telephonic follow-up as
needed.

One month after receiving the intervention, subjects will be emailed a RedCap link to a
follow-up survey to measure changes in insurance coverage and other outcomes. The follow-up
survey will be structured similarly to the baseline survey, with closed-ended questions about
insurance coverage, asthma care and control, health care access and affordability. For
intervention subjects, the follow-up survey will also ask closed and open-ended questions
about implementation outcomes such as uptake, acceptability, feasibility, usability, and
fidelity.

Four months later, intervention and control subjects will be emailed a RedCap link to a
similar follow-up survey to measure changes in insurance coverage, asthma care and control,
health care access, and affordability. Survey subjects will be asked for permission to be
contacted again for participation in a telephone interview for Aim 2.

The primary outcome will be whether the participant with asthma (or their child with asthma)
has insurance coverage. Secondary outcomes will include whether the person with asthma used
less asthma medication than prescribed because of cost, and whether they experienced health
care-related financial burden. Other outcomes will assess asthma care use, asthma control
using the Asthma Control Test (ACT), health status, mental health/worry, and material
hardships such as food or housing insecurity.

The study hypothesis is that participants receiving the intervention will be more likely to
have coverage at four-month follow up (primary outcome) and will be less likely to report
non-adherence to asthma medications, delayed/forgone asthma care, and financial burden than
those receiving usual care (secondary outcomes).

The investigators will compare unadjusted outcomes between the intervention and control
groups using chi square and t tests; because of the randomized study design this difference
measures the causal impact of the intervention. They will conduct post hoc exploratory
analysis to assess differences in intervention effectiveness by race/ethnicity to assess
whether the intervention reduces racial/ethnic coverage disparities. Given the small pilot
nature of this study, these analyses will be hypothesis-generating rather than
hypothesis-testing. In this pilot study, efficacy testing will be limited but the
investigators expect to have 100 intervention and 200 control subjects with follow-up data to
be able to detect absolute differences of 10% or more between study groups in changes in
uninsurance rates with 80% power with a two-tailed alpha of 0.05. The investigators assume
30% loss to follow-up from baseline to four-month follow-up. Given the small pilot nature of
this study, dropout will be carefully tracked in order to inform feasibility assessment and
planning for a larger trial. The investigators will closely monitor timing of dropout, impact
of strategies such as telephone outreach, and characteristics of those dropping out compared
to those retained, and will use multiple imputation methods to conduct a sensitivity analysis
to test the sensitivity of findings analyzed with and without imputed data. Findings will be
reported using CONSORT reporting guidelines.

Aim 2

In this aim, the investigators will conduct in-depth qualitative interviews to better
understand barriers and facilitators to accessing coverage and asthma care more broadly
during the COVID-19 pandemic. Study subjects will be selected from among those who completed
the follow-up survey and gave permission to be re-contacted for a telephone interview. The
investigators will approach 100 eligible survey participants by email to invite them to
participate in an interview. They expect to complete 30 phone interviews, half from the
intervention group and half from the control group. Interviews will use open-ended questions
to explore patients' experiences related to obtaining coverage after job loss and associated
asthma care access and affordability challenges and facilitators. Interviews will be
conducted by the study investigators and will be audio recorded and the recordings will be
transcribed.

Eligible participants will be intervention and control participants who complete the
four-month follow-up survey, give permission to be re-contacted for an interview, and provide
an email and/or phone number to use to contact them. The investigators will seek to maximize
variation in our sample by using survey data to identify potential subjects with a range of
races/ethnicities, incomes, geographic residences, ages (child vs adult), and asthma
severity.

The investigators will collect data from 30 study participants through in-depth qualitative
interviews. They will approach 100 eligible survey participants by email to invite them to
participate in an interview. They will then follow up by phone to schedule an interview.
After scheduling a telephone interview and describing the study, they will obtain verbal
informed consent. Interviews will last for approximately 45 minutes and will be audio
recorded with permission and transcribed verbatim. They estimate completing 30 phone
interviews, half with participants from the intervention group and half from the control
group.

The investigators will transcribe recordings of the interviews and analyze these data via
thematic content analysis. Using a codebook, 2 trained coders will independently code
transcripts using NVivo, resolving disagreements via discussion. They will describe data
thematically to probe quantitative findings and describe strengths, weaknesses, and perceived
intervention impact.