Study Design Overview
This pilot trial uses a pragmatic, randomized 2-group parallel design to examine the feasibility, acceptability, appropriateness, and preliminary efficacy of a 6-week telemedicine group self-management intervention for adults with long COVID. The intervention is aimed at improving symptom management and coping in adults with long COVID. Participants are randomized to the intervention group or a wait-list control group. All study methods and procedures have been approved by the Institutional Review Board at MASKED. The trial is registered on ClinicalTrials.gov with the identifier: NCT05658536. See Table 1 for SPIRIT Flow Diagram for participant timeline. See supplementary table for SPIRIT Checklist.
Study Setting
This trial aims to enroll and randomize 50 participants at the MASKED Post-COVID Rehabilitation and Recovery Clinic (“PCRRC”) within the MASKED. Assuming a 10% dropout rate, a total of 50 participants will yield 20 completers in each study condition. This sample size was selected to characterize the intervention’s feasibility, acceptability, appropriateness, and preliminary efficacy to inform intervention improvement. A sample size of 20 participants in each condition (accounting for attrition) is appropriate given the specificity of our qualitative research questions (16,17). Likewise, this sample size is appropriate for estimating the proportion of participants who rate the intervention as feasible, acceptable, and appropriate. The lower bound of the 95% CI will be above 60% if at least 17 out of 20 (85%) of intervention participants rate the intervention as feasible, acceptable, or appropriate (Supplemental Figure 1), which is a conservative estimate given satisfaction ratings for similar rehabilitation interventions (18).
Participant Eligibility and Recruitment Procedures
For this pragmatic pilot trial, inclusion and exclusion criteria were selected to reflect clinical care in the MASKED PCRRC and to maximize generalizability such that a broad range of patients with long COVID engaged in care in a multidisciplinary clinic like the MASKED PCRRC would be represented. Inclusion criteria are that participants must be at least 18 years of age, reside in MASKED state, have a long COVID diagnosis, be able to read, speak, and understand English, and have reliable telephone and internet access. Exclusion criteria are severe cognitive impairment as evidenced by two or more errors on the Six-Item Cognitive Screener (19) and current participation in other psychosocial treatment primarily for a long COVID symptom(s) or problem(s).
The MASKED PCRRC serves as the recruitment site. Prospective participants are identified through the wait-list for the existing telemedicine group self-management intervention at the MASKED PCRRC. Prospective participants are sent an approach letter via email or physical mailing that indicates why they are being approached and how study staff received their name and contact information. Study staff call prospective participants within one week after sending the approach letter. Interested patients may also contact research staff directly after receiving the approach letter. Prospective participants may also be identified and referred to the study by providers within the MASKED PCRRC who are already referring patients to the telemedicine group-based self-management intervention.
Screening and Baseline Assessments
The study coordinator provides a brief description of the study to prospective participants over the phone. Prospective participants who are interested in enrolling in the study are then screened for eligibility (see inclusion/exclusion criteria above). Eligible prospective participants complete the informed consent process to enroll into the study. Following enrollment, participants are asked to confirm their availability for upcoming self-management groups which are scheduled based on provider availability. If participants are available for the upcoming group schedule, participants complete a brief re-assessment of eligibility, confirming that they have not participated in any other similar self-management programs targeting Post-COVID symptoms. Within two weeks of the group’s Session 1 date, participants are randomly assigned to the intervention group or the wait-list control group. If availability is not confirmed, the study coordinator contacts the participants to confirm availability for the next group(s), as soon as that schedule is available from the MASKED PCRRC. Following randomization, participants complete a battery of self-report baseline measures. See Table 1 for study schedule and outcome measures. Study data are collected and managed using REDCap electronic data capture tools hosted at the MASKED (20,21).
Table 1. Schedule of enrollment, interventions and assessments (SPIRIT 2013 guidelines)
|
STUDY PERIOD
|
|
Enrollment
|
Allocation
|
Post-allocation
6 weeks (post-treatment/wait-list)
|
TIMEPOINT
|
-t1
|
0
|
t1
|
ENROLLMENT
|
Eligibility screen
|
X
|
|
|
Informed consent
|
X
|
|
|
Allocation
|
|
X
|
|
CONDITIONS
|
Telemedicine Self-management Group
|
|
|
X
|
Wait-list Control
|
|
|
X
|
ASSESSMENTS
|
Primary Outcomes
|
Acceptability of Intervention Measure (22)
|
|
|
X*
|
Intervention Appropriateness Measure (22)
|
|
|
X*
|
Feasibility of Intervention Measure (22)
|
|
|
X*
|
Secondary Outcomes
|
Depression Short Form 8b** (23)
|
|
X
|
X
|
Anxiety Short Form 8a** (23)
|
|
X
|
X
|
Fatigue Short Form 8a** (23)
|
|
X
|
X
|
Sleep Disturbance Short Form 8a** (23)
|
|
X
|
X
|
Cognitive Function – Abilities Short Form 8a** (23)
|
|
X
|
X
|
Self-Efficacy for Managing Chronic Conditions - Managing Symptoms Short Form 8a** (23)
|
|
X
|
X
|
Ability to Participate in Social Roles and Activities Short Form 8a** (23)
|
|
X
|
X
|
Pain Interference Short Form 8a** (23)
|
|
X
|
X
|
Long COVID Symptom and Impact Tools (24)
|
|
X
|
X
|
Credibility/Expectancy Questionnaire (25)
|
|
X*
|
|
Patient Global Impression of Change scale (26)
|
|
|
X*
|
Note: *Completed by intervention participants only. **Patient-Reported Outcomes Measurement Information System (PROMIS).
Self-Management Intervention
The MASKED and MASKED developed a 6-session group self-management intervention for patients with long COVID. This intervention teaches participants behavioral strategies known to be helpful in managing symptoms in patient populations with similar symptom profiles as long COVID (e.g., myalgic encephalomyelitis, multiple sclerosis), including fatigue, memory and attention problems, insomnia, dyspnea, coping with anxiety and uncertainty, and stress management (5,6). See Table 2 for a description of intervention content by session. The intervention is delivered via telemedicine by a licensed rehabilitation psychologist at MASKED. The rehabilitation psychologists follow the treatment manual and cover the major components of each session, which are highlighted with fidelity statements. The rehabilitation psychologists meet monthly for case consultation via video conference.
Table 2. Major components of the manualized self-management intervention for long COVID
Session
|
Session Content
|
1
|
Introduction to Long COVID and the Biopsychosocial Model
Information and discussion on: long COVID and related symptoms, biopsychosocial model and self-management philosophy, and dyspnea.
Skill: diaphragmatic breathing
|
2
|
Energy Management
Information and discussion on: long COVID fatigue, cycle of deconditioning, assessing and reestablishing priorities.
Skill: pacing for energy management
|
3
|
Enhancing Sleep, Memory and Attention
Information and discussion on: sleep problems, memory and attention difficulties.
Skill: sleep hygiene and strategies to improve attention
|
4
|
Anxiety and Coping with Uncertainty
Information and discussion on: anxiety and emotional responses to COVID-19 and long COVID, discernment for effective coping, mindfulness.
Skill: coping with controllable and uncontrollable stressors, grounding mindfulness exercise
|
5
|
Stress Management
Information and discussion on: stress associated with long COVID, cognitive behavioral model and related strategies to enhance coping with stress.
Skill: thought challenging, coping thoughts
|
6
|
Enhancing Communication
Information and discussion on: effective communication with healthcare providers for long COVID care, communication with family and friends about long COVID and symptom management, effective goal setting.
Skill: communication strategies, goal setting that is specific, measurable, achievable, relevant, time-bound (SMART)
|
Randomization and Procedures to Minimize Bias
Participants are randomized using a 1:1 randomization to treatment versus wait-list control condition system. When participants confirm availability for an upcoming clinic group schedule and are confirmed eligible following the Re-Assessment of Eligibility, participants are 1:1 randomized to either the treatment experimental condition or wait-list control condition. The order of randomization is determined by the order in which participants chronologically completed the Re-Assessment of Eligibility. To minimize bias, the clinicians delivering the intervention are blinded to the research participation status of their group members. In addition, the REDCap system automatically sends participants baseline and post-treatment/wait-list survey links, where the participant can directly enter their data. As a result, the potential for study staff biases to interfere with data collection are avoided.
Data Collection Procedures and Measures
Participants complete self-report measures via REDCap at baseline (prior to starting the treatment/wait-list period) and at post-treatment/wait-list after completing the 6-session intervention group or 6-week wait-period. For data capture and storage, this study uses REDCap, a secure, password protected, and HIPAA compliant web-based data platform hosted by the MASKED Institute of Translational Health Sciences (20,21).
Primary outcomes include the Feasibility of Intervention Measure (FIM), Acceptability of Intervention Measure (AIM), and Intervention Appropriateness Measure (IAM) (22). Each measure consists of 4 items that are rated on a 5-point ordinal scale assessing the degree to which an intervention is feasible (e.g.,“this intervention seems doable”), agreeable (e.g., “I like this intervention”), or appropriate (e.g., “this intervention seems fitting”). A score greater than 3 (“neither agree nor disagree”) will be used as the cut-off for a dichotomous rating of feasibility, acceptability, and appropriateness. The FIM, AIM, and IAM have shown excellent reliability as well as substantive and discriminant content validity (22). Secondary outcomes are listed in Table 1.
In addition, participants randomized to the self-management intervention (n=25) complete a 45-minute semi-structured interview with research staff within one month after the final session. The interviewer queries participants about the feasibility, acceptability, and appropriateness of the intervention as well as suggestions for improving the intervention across these areas. Participants are also asked about the perceived efficacy of the intervention and how the intervention could be improved to increase efficacy. Last, the interviewer queries participants’ impressions of the intervention workbook and elicit suggestions for improvement. Data will be transcribed via Rev, a HIPAA-compliant online transcription service (27), and analyzed using Dedoose (28).
Data and Safety Monitoring
The Principal Investigators provide oversight of the trial in collaboration with study staff to assure adequate protection of the rights of human subjects. They monitor progress of the project, protocol adherence, substantial protocol amendments, and participant safety including adverse events on an ongoing basis in weekly meetings and submit a report to the MASKED Institutional Review Board twice annually. Because this study does not use investigational drugs or devices and poses minimal risk to participants, a data and safety monitoring board, which is required for most FDA studies, was not deemed necessary. Instead, the above process serves two purposes: it accommodates the expected low-base rate of problems and allows for the earlier identification of issues more appropriately than a data and safety monitoring board that would meet at an arbitrary regular interval. Likewise, there is no auditing process in place beyond the MASKED IRB standard procedures of potentially conducting an audit during the course of the research or after the project is completed. Regarding provisions for ancillary and post-trial care, participants in the self-management group are passively monitored for suicidal ideation. If a group member reports suicidal ideation, the licensed psychologist leading the group provides appropriate follow up including risk assessment, safety planning, and provision of resources depending on level of risk.
We will make the data and associated documentation available to users only under a data-sharing agreement that provides for: (1) a commitment to using the data only for research purposes and not to identify any individual participant; (2) a commitment to securing the data using appropriate computer technology; and (3) a commitment to destroying or returning the data after analyses are completed.
Statistical Analysis Plan
An intent-to-treat (ITT) approach will be used for all quantitative analyses. If participants are missing data at the individual-item level of a multi-item scale, we will impute the missing data using the within-group mean scores for the scale; however, a score will not be imputed if the participant is missing ≥40% data in a scale.
First, we will compute descriptive statistics to characterize the sample in terms of demographic and condition-related factors. To assess the first study aim, we will calculate the means and standard deviations of the AIM, IAM, and FIM and quantify the proportion of participants who rate the intervention as feasible, acceptable, and appropriate (average score greater than 3). We will quantify adherence as the percentage of sessions attended out of 6 sessions. These statistics will be used to contextualize the qualitative results and help inform the extent of modifications necessary to improve the feasibility, acceptability, and appropriateness of the intervention for adults with long COVID.
The second study aim to examine intervention effects is exploratory. We will determine whether randomization was successful by using t tests and chi-square tests to compare the intervention and wait-list groups on a variety of demographic and baseline outcome variables. Any variables that differ between the groups will be included as covariates in the subsequent regression models. After examining randomization, we will explore intervention effects. For the second aim, we will conduct linear regression to assess intervention effects on each outcome (9 models) and will covary for the baseline level of the outcome variable, intervention credibility/expectancy (25), age, sex, and race.
For the third aim, we will analyze qualitative data using the rigorous and accelerated data reduction (RADaR) technique (29). The RADaR technique involves identifying an overarching research question, developing an all-inclusive data table from interview transcripts, and revising the table to reduce the data by retaining only the data that can answer the research question. This overall process is completed for each research question. The overarching research questions for this qualitative analysis are: what do participants like about the long COVID group intervention and the participant workbook, what do they dislike about the long COVID group intervention and the participant workbook, and what are their suggestions to improve the intervention’s feasibility, acceptability, appropriateness, and efficacy? Qualitative findings will contextualize the primary outcomes data to inform specific modifications to improve the feasibility, acceptability, and appropriateness of the intervention.
We will independently and collectively review interview transcripts. Step 1 of the RADaR technique (transcription) will be completed by Rev, a HIPAA-compliant online transcription service (27). Then a research team member will complete Step 2 of combining the individual transcripts to create an all-inclusive data table. We will complete Step 3 individually to reduce the data to a more concise data table that answers a single research question and includes notes and preliminary codes. We will then review, compare and discuss the Step 3 data tables to select which text chunks will move forward in the analysis of the particular research question and to develop consensus on codes that will likely remain in the code column throughout subsequent steps and inform subsequent theme discussion and identification. A research team member will produce a revised data table based on the team consensus from the meeting. This process will be repeated (i.e., repeat Step 3 individually and discuss the data tables to achieve consensus) and generate and refine themes related to the research question as many times as necessary to agree the remaining data is sufficient for answering the particular research question. We will repeat the RADaR technique for each research question. We will then produce a report for each question that contains themes and exemplar quotes to help guide intervention refinement. Rigor for the RADaR technique is achieved by mapping the information in each data table phase between the individual team members and across the data table reduction phases. Thus, the iterative process involves close assessment of each analyst’s reviewing, reducing, and coding styles. Additionally, we will adhere to the standards for reporting qualitative research throughout data collection, analysis, synthesis and interpretation and report these standards in the resultant manuscript (30). Trial results will be disseminated via presentation at local and national conferences, peer-reviewed manuscript publications, and reporting on ClincalTrials.gov.
Trial Status
The MASKED Institutional Review Board approved the study protocol on October 12, 2022. The trial was posted to clinicaltrials.gov on December 20th, 2022 with the identifier: NCT05658536. Recruitment started in January 2023. We expect to complete the treatment phase of the study in January 2024 and to complete assessments and qualitative interviews by February 2024.