This study uses a randomized controlled design (intervention vs. usual care) to test the efficacy of Women of Color and Asthma Control, a telephone-based asthma management intervention for African American women. All study procedures have been reviewed and approved by the Institutional Review Board at the University of Michigan (IRB Study HUM00033784).
Compared to women in the usual care group, women in the intervention group will: 1) use emergency department (ED) services for asthma less frequently and be hospitalized for asthma less frequently; 2) need urgent care in a physician's office less often; 3) experience fewer symptoms of asthma; and 4) have higher levels of asthma-related quality of life. We expect to see between-group differences at both 12 and 24 months from baseline.
Sample size determination
The sample size of 420 at baseline (210 women in each treatment condition) was determined by power calculations using the primary study outcomes of ED visits and hospitalizations in the last 12 months, taking into account 30% attrition rate over 24 months and a loss of 5 women per group due to pregnancy. In previous work we observed that for ED visits and hospitalizations the treatment takes at least 12 months to reach full effect; thus, the sample size calculations for these outcomes were performed at 24 months. However, for the other outcomes, e.g. asthma symptoms and QOL, the intervention effect occurred during the first 12 months and the power calculations for these were performed at 12 months, which is the same or more conservative than 24 months.
The relatively rare occurrence of ED vists and hospitalizations for asthma requires a large sample size to detect clinical improvement over time. In previous work by the authors based on 364 participants with asthma, belonging to the intervention group was associated with a 75% reduction for hospitalization and 53% for ED visits with at least 80% power (alpha = 5%) at 24-month follow-up. We conservatively estimate a smaller reduction for the same two outcomes with a similar sample size and the same statistical power. An initial total sample of 420 and a final available sample of 286 will have sufficient power to detect a 32% improvement in the proportion of participants having at least 1 ED visit (42% in mean ED visits) and a 41% improvement in the proportion having at least one hospitalization (46% in mean hospitalizations). Note that these power calculations also ensure our ability to assess other important outcomes such as asthma symptoms and asthma quality of life.
We are recruiting a total of 420 participants for this study over an approximately two-year period. To be eligible, participants must: 1) be a woman 18 years of age or older and self-identify as African American, 2) have access to a landline or mobile telephone, 3) not be pregnant, and 4) not reside in an institution.
Primary recruitment is through the University of Michigan Health System (UMHS), with eligibility requirements for participants including: 1) have at least one outpatient visit at the University of Michigan Health System (UMHS) in the past year; and 2) be listed on the UMHS Asthma Patient Registry, a validated all-payer registry of patients with persistent asthma cared for within UMHS. Lists of potential participants from the UMHS Asthma Registry are periodically generated and reviewed by a research nurse. Eligible patients are then mailed a packet including consent forms, a study overview page and a letter from the project director inviting them to enroll. The letter is followed by a recruitment phone call, which includes information about the study's purpose, randomization, and data collection. Upon returning a signed consent form, participants are enrolled in the study and randomized after baseline data collection.
Secondary recruitment methods include posting on the University of Michigan's clinical trial research database, http://UMClinicalStudies.org
; partnering with the Blue Cross Blue Shield of Michigan (BCBSM) to identify members in the area who meet study criteria; and instituting a community recruitment plan that includes posting flyers in local health centers, hair salons, community organizations, and churches. All patients recruited outside of the Registry will have patient records in the UMHS or BCBSM systems that can be used for analysis of outcomes related to health care utilization. Any potential participant contacting study staff after learning about the study through one of these secondary recruitment methods will be screened for eligibility and then mailed the study packet as described above.
To enhance recruitment and retention of participants, a number of strategies are employed. Strategies include: 1) providing a $20 gift card upon completion of each data-collection interview; 2) mailing periodic greeting cards to establish a personal connection; 3) providing each intervention-group participant with her counselor's photograph and a personalized letter; 4) utilizing as recruiters, interviewers, and counselors individuals who share a cultural understanding and who have worked with and in the African American population; and 5) attempting contacts over a period of several weeks since disconnected phone lines are sometimes reconnected after a period of time.
After the baseline interview, participants are assigned to either treatment or control groups using a restricted randomization scheme. This scheme achieves balance between the two groups at all points during the study period. Prior to the beginning of the study, a permuted block randomization schedule for 420 subjects was created using the RANTBL and RANUNI functions in SAS to generate blocks with varying sizes of 8, 10, or 12 and random numbers for group assignment. After every block, there will be an equal number of intervention and control participants. This process ensures that over the course of the study, 210 women will be assigned to each of the two study conditions. It also reduces potential confounding effects (e.g., seasonal variation) which might occur due to the rolling recruitment process.
Theory and content of intervention
is based on the Center for Managing Chronic Disease model of self-regulation for disease management, a framework that has been informed by social cognitive theory [18
]. Participants are introduced to a self-regulatory problem-solving process, which is designed to help them engage more effectively in their asthma management. This process, which reflects the key components of the self-regulation model, include the following steps: 1) Identifying a problem
in asthma management (participants are offered a list of potential asthma-management problems, including those associated with their sex, gender roles, culture and other relevant challenges; recommendations from the participant's physician are also reviewed at this stage); 2) Observing
oneself, one's environment, and one's pattern of symptoms so as to understand both the influences on the problem and the way those influences might be ameliorated; 3) Setting an asthma-management goal
; 4) Developing a plan or strategy
to achieve the goal; and 5) Tracking progress, reacting appropriately, and establishing suitable rewards
for success. Participants are guided by the counselor through a period of self-observation using a peak-flow meter and symptom diary, along with a checklist of physical activity, environmental factors, and other potential precipitants to asthma exacerbations. This observation period enables them to see the barriers and facilitators to achieving desired management practices and outcomes. Telephone sessions last 45 minutes to 1 hour and take place approximately every two weeks. Table provides an overview of the content of each session. Importantly, the content of the intervention also reinforces the priority messages put forth by the National Asthma Education and Prevention Program's Guidelines Implementation Panel [19
]: appropriate use of inhaled corticosteroids, reviewing asthma control, having an asthma action plan, periodic follow-up visits with clinician as appropriate, and allergen and irritant exposure control.
Women of Color and Asthma Control intervention: Overview of program session content
The control condition will consist of usual care from the UMHS or a BCBSM provider. At the UMHS, this care consists of evidence-based clinical practice and adherence to National Asthma Education and Prevention Program (NAEPP) guidelines for asthma diagnosis and treatment, and development of asthma action plans. Emphasis is also placed on patient education about disease processes, importance of trigger avoidance, and the development of a partnership for asthma care. Standardized asthma education kits are available in all UMHS clinic sites, and clinic personnel have been trained in their use by a certified asthma educator. Of note, sex-specific, gender-role, or culturally-related issues are not systematically or routinely addressed during the clinical encounter or in educational materials. All WCAC participants are sent basic asthma education materials ("Controlling Your Asthma" booklet from the American College of Chest Physicians) upon completion of the baseline interview.
Outcomes and measures
Data is collected from participants via telephone interviews with trained research assistants. Data is collected on paper and then entered twice into Qualtrics, a secure online survey program, to ensure accuracy. With the participants' permission, an audio recording of the interview is also made to verify data accuracy and completion. Data collection takes place at three time points: baseline (before randomization), 12 months, and 24 months.
The primary study outcomes are the number of asthma-related Emergency Department (ED) visits and the number of overnight hospitalizations in the last 12 months. These will be collected via self-report on telephone interviews, as well as from medical and billing records.
Other outcomes of interest include the following:
Frequency of daytime and nighttime asthma symptoms
(coughing, wheezing, shortness of breath, and chest tightness), over the past month, seasonally, and in the last 12 months.
This validated, five-item questionnaire assesses the effect of asthma on daily functioning, shortness of breath, nighttime asthma symptoms, rescue medication use, and overall self-rated asthma control over the last 4 weeks.
Mini Asthma Quality of Life Questionnaire [21
This validated 15-item scale addresses the following four domains specific to adult asthma patients: 1) activity limitations; 2) symptoms; 3) emotional function; and 4) environmental stimuli.
Absenteeism activity days
In the previous 12 months, 1) total days when physical activity was limited because of asthma; 2) total days of missed work or school because of asthma.
Other health care utilization
In the previous 12 months, unscheduled clinic visits for urgent asthma treatment and scheduled clinic visits for asthma care.
Asthma self-regulation skills and self-efficacy
A scale was developed by the authors that measures the frequency with which participants engage in the self-regulatory behaviors of observations, judgments, and reactions as applied to asthma management; and also measures self-efficacy for each of these behaviors.
Sex/gender-specific management problems
This scale, used in previous work by the authors [15
], measures the frequency of problems of asthma management related to hormonal cycles, sexual activity, urinary incontinence, and triggers associated with gender roles.
African-American-specific management problems
This scale, developed by the authors, measures the frequency with which participants experience asthma-management challenges that may be common among African American women (e.g., having numerous family responsibilities, physician communication, money concerns or worries).
Data is also collected on a wide array of demographic, health and psychosocial characteristics at all three time points.
Statistical testing will evaluate changes in the primary and secondary outcomes, and other outcomes of interest, over the course of the study, and assess differences in these changes as a function of participation in the intervention. Descriptive statistics will be computed for all outcomes, as well as sociodemographic and health characteristics, overall and for each study group (treatment and control). This will help determine whether the treatment and usual-care groups are equivalent prior to intervention, and how these groups differ at specific follow-up points (12 and 24 months after baseline). We will use mixed-effects models for the analysis of continuous outcome variables and generalized estimating equations (GEEs) for analyzing discrete outcomes. Pre-planned contrasts will examine the nature of change over time, including the timing and longevity of intervention effects. SAS Version 9.2 and IVEware (to multiply impute missing data) will be used for analysis [22
Data will also be collected regarding key aspects of the intervention process. First, participants will be asked to complete a brief online or mail survey regarding their experience with the WCAC program (e.g., satisfaction with counselor, perceived improvement in asthma-management skills) when they have completed the series of telephone counseling sessions. This survey also includes items designed to assess participants' satisfaction with the culturally-relevant elements of the program (e.g., "If you were to participate in a program like this one again, how important would it be to you that your health educator be African American?"). Second, telephone counselors will keep logs of each phone session to track elements such as participants' goals and progress in the self-regulation process. Finally, all telephone counseling sessions are audio-recorded and a random subset are reviewed for their fidelity to the planned program content. This process data will be used to enhance understanding of program outcomes, and to inform modifications to future versions of the program.