Improving Healthcare Communication for Caregivers: A Pilot Study

Abstract

Background:

With the growing older adult population, there will also be more informal caregivers assisting friends and family with their healthcare. With the increasing complexity of healthcare, improved caregiver communication skills have the potential to reduce caregiver burden, frustration and improve care recipient health.

Objective:

The primary goal of this project was to develop and refine the content and teaching methods of a small group behavioral change program to improve communication between caregivers of older adults and healthcare professionals.

Methods:

The authors developed the Care Talks program for improving communication between caregivers and healthcare professionals. They conducted a prospective cohort feasibility study of the intervention to assess caregiver communication confidence at baseline and 1 month post-intervention.

Results:

16 participants were enrolled. Of the 15 participants who answered the question, 15 (100%) would recommend this program to a friend. There was significant improvement in a 10 question composite of communication confidence pre/post scores from 74.1 to 79.6 p=0.03.

Conclusion:

This small group behavioral change intervention significantly improved communication confidence for this sample of caregivers. Further research is needed to determine the long-term effects of this program on caregivers and care recipients.

Introduction

Informal caregivers provide unpaid support to family and friends. They are essentially a “shadow workforce,” acting as geriatric case managers, medical record keepers, paramedics, and patient advocates to fill dangerous gaps in a system that is uncoordinated, fragmented, bureaucratic and frustrating. (Bookman 2007) In 2009, there were 61.6 million caregivers in the United States, (Feinberg, 2011) and as the population ages, the number of caregivers will also increase. Since healthcare and the healthcare system are complex with ever increasing demands on caregivers as their care recipients continue their life journey, the demand for programs and training to assist caregivers is also increasing.

One significant issue for caregivers is their ability to find, understand and use health information. These critical components of health literacy (Nielson, 2004) are necessary to navigate the healthcare system, maintain health and address health problems. Unfortunately, as few as 12% of American adults are proficient in health literacy, (Kutner, 2003, 2015) so a very large number of people find it difficult to make healthcare decisions, and develop and complete a plan of action. Improving effective communication between healthcare professionals, patients and caregivers is one proposed method of addressing these issues (U.S. Department of Health and Human Services, 2010) and has the potential to reduce caregiver burden.

Caregiver burden is a significant problem that can result in a variety of adverse outcomes including both physical and psychological conditions such as anxiety, (Cooper, 2007) depression (Van Vliet, 2010) and sleep disturbances. (Lee, 2011) These often accelerate the need for institutional care, and cause economic and social problems such as financial difficulties, family conflict and isolation. (Van Vliet, 2010) Many interventions to reduce caregiver burden have been tested, (Adelman, 2014) yet we found very limited literature describing interventions aimed at improving the communication between caregivers and health care professionals. (Moore, 2008, 2001, Chiah 2012)

We believe that improved communication between caregivers, their care recipients and healthcare professionals has the potential to reduce confusion about care plans, decrease errors during the process of care and reduce caregiver burden. We report on a pilot project to develop and evaluate a series of workshops intended to increase confidence as it relates to communication with healthcare professionals and decrease caregiver burden.

Theoretical Basis for Intervention

This program is designed to increase caregivers’ self-efficacy to communicate effectively with healthcare professionals. Self-efficacy has its roots in Bandura’s Social Cognitive Theory which asserts that adoption of a new behavior is enhanced with increased self-efficacy. (Bandura, 1997) Participants in this project experienced Bandura’s key mechanisms for increasing adoption of new behaviors: 1) trainers modeled the desired communication behavior; 2) learners rehearsed the desired behaviors; 3) learners received constructive, positive feedback on their communication style to enhance their skills and self-efficacy to achieve their desired outcomes.

Methods

This prospective observational cohort study was conducted in Monroe, Wisconsin. Data were collected between April and October 2014. The research was approved by the University of Wisconsin Institutional Review Board and written informed consent was obtained.

Intervention Development

Program content development started with the successful Med Wise program (Martin, 2015) developed by two of the authors (BC, BM) to improve patients’ communication with pharmacists.

Based on the Med Wise content, with modifications for communication with any healthcare professional, a list of potential workshop topics and communication skills were developed by the project team comprised of two pharmacists, one nurse, one family physician, one experienced educator for adults and support staff. Three informal caregivers then joined the project team for discussion during team meetings to help prioritize the areas of most need, suggest additional potential topics and issues, and provide recommendations to maximize effectiveness of the intervention. Following the discussions with the caregivers, a detailed leader guide, key point posters and key point handouts were developed for the first set of workshops and then modified following each workshop based upon caregiver participant verbal comments during the workshops, written participant surveys after each workshop and feedback from workshop leaders and fidelity assessors.

The Care Talks program consisted of four 2-hour workshops (weeks 1, 2, 3, and 7). Three series of workshops were held, with approximately 2 months between series. On-site respite care was offered during workshops, although none of the participants used this service.

Care Talks workshops were a combination of didactic and skill-building exercises including brain-storming, problem-solving and role-playing. Workshop topics included: 1) Identifying facilitators and barriers to effective communication; 2) Caregivers negotiating their role with the care recipient; 3) Developing a list of concerns and questions before a healthcare visit (creating an agenda); 4) Sharing concerns in a positive manner; 5) Sharing concerns and asking questions during a healthcare visit; 6) A primary care physician’s (PS) view of an ideal office visit; 7) Reviewing key points with their care recipient after a healthcare visit; 8) Acting on a disagreement with a healthcare professional; and 9) Confirming understanding using reflective listening. Reflective listening occurs when a listener repeats back key points to the speaker to confirm correct understanding of what was said. Workshops allowed participants to practice communication methods using real-world, challenging examples. Participants were given a folder for storing notes at the first workshop and handouts with key points were distributed at each workshop, including home practice activities.

The first Care Talks workshop series was led by Med Wise creators (BC, BM) and a community research team member (SH). Two community leaders with patient education experience were provided 6 hours of training with a combination of didactics and role playing and they led the 2^nd and 3^rd workshop series with a co-author (SH).

Participants

Participants were a convenience sample living in Monroe, Wisconsin. Inclusion criteria included: 1) Informal caregivers (not paid); 2) Providing daily living skills assistance for a family member or friend and/or accompanying the care recipient to healthcare visits; 3) Verbal fluency in English; 4) Demonstrated cognitive ability to provide informed consent, assessed by the Mini-Cog test; (Borson, 2000) and 5) care recipient age 60 or older. Exclusion criteria included: 1) Inability to attend all workshop sessions; 2) Lack of telephone access for a follow up survey; or 3) Care recipient was the caregiver’s child. We excluded parent caregivers because the role negotiation and communication dynamics are very different between a parent caregiver and child care recipient compared to the interactions of other caregiver/care recipient dyads.

Potential participants were recruited using a variety of direct and indirect methods to promote the program and encourage them to contact community research staff (SH). We developed press releases, newsletter articles, posters and a tri-fold brochure and sent them to the local Green County collaborators to spread the word in their newsletters, facilities and online.

In addition, promotional materials were given to key healthcare professionals at the only local healthcare facility. We did presentations to local support groups and outreach to some major local employers. The most successful strategy was direct mailings and follow up phone calls to caregivers with previous interactions with the local caregiver support organization such as attendees from past classes and home delivered meal recipients. Participants did not receive participation incentives.

Measures and Procedures

A researcher administered the baseline survey by interview at enrollment and follow up survey by telephone approximately one month after Workshop 4. Collected data included age, gender, race, years of education, income level, and current employment. Self-reported health status was assessed using a validated instrument from the Stanford Patient Education Research Center. (Sanford, 2015) Depression was assessed using the Patient Health Questionnaire-9 (PHQ-9). (Kroenke, 2001) Caregiver and care recipient environment was characterized using data about living location of care recipient, caregiving tasks, additional caregivers, frequency of telephone and personal contact, and caregiver/care recipient relationship. Health literacy level was assessed by administering the Newest Vital Sign. (Weiss, 2005)

Unfortunately, appropriate caregiver-specific questions did not exist to measure communication self-efficacy. We adapted questions from several validated instruments, including the Med Wise survey, Communication and Attitudinal Self-Efficacy (CASE) scale, (Wolf, 2005) and Patient Activation Measure (PAM).(Hibbard, 2004) Pertinent questions from the Zarit Caregiver Burden Interview (Ballesteros, 2012) were used to assess caregiver burden. The resulting questions primarily measured caregiver burden and confidence, not self-efficacy, a related, but distinct construct.

Questions intended for patients were modified to be applicable to caregivers. For example the sentence “I am confident I can prepare for a visit with the doctor.” Was changed to “I am confident I can prepare for a visit with the doctor and the person I help.” Since caregivers may not be related to the care recipient, we also modified the Zarit burden questions from using the phrase “your relative” to “the person you help.” We had several very negative reactions to the Zarit question “Overall, how burdened do you feel in caring for your relative?” Therefore, the question was modified to “We know that taking care of other people can become challenging at times. In the past week, how challenging has it been to take care of the person you help?” with a 0 (Not challenging at all) to 4 (Very challenging) scale. Since many of the questions from each of the sources were modified and no longer validated, and different sources used different Likert scales (Med Wise, 0–10; CASE, 1–4; PAM 1–4; Zarit 0–4), we used a 0–9 point scale for all of the confidence and burden questions to avoid participant scale confusion except the burden question noted above. All modified questions were assessed for understandability by cognitive testing with two of the community caregivers involved in development of the workshop materials. Cognitive testing is a process where the caregiver reads the question aloud, answers the question, tells the interviewer what they think the question is asking about and if they were confused by any words or phrases. Questions were modified to improve clarity and understandability based on the caregiver feedback. Communication, confidence and caregiver burden questions were then assessed for understandability by the Wisconsin Community Advisors on Research Design and Strategy (CARDS) team.(Wisconsin, 2015) CARDS teams are trained groups of community members from diverse racial, ethnic, socioeconomic and educational backgrounds who regularly review research questionnaires and give feedback to researchers about understandability and cultural appropriateness for the target audience. Questions were again modified to improve clarity and understandability based on the CARDS team feedback.

Workshop Program Evaluation

Program evaluation included a 4 question survey after each workshop about what the participants enjoyed most, what they learned and intended to use, what they found confusing and general comments. The final workshop survey also included 15 statements that participants rated using a 5 point Likert scale from 0 (strongly disagree) to 4 (strongly agree). Statements included: The workshop venue was: comfortable, well located (2 separate statements); The workshops’ content was: relevant, comprehensive, easy to understand (3 statements); The workshop handouts: supported presentation material, provided useful additional information, were clear and well-organized (3 statements); The workshops were: well-paced, a good mix between listening and activities, had enough breaks (3 statements); The workshop leaders were: knowledgeable, well-prepared, responsive to questions (3 statements); and: The activities were useful learning experiences. The participants were also asked if they would recommend the program to a friend (yes/no/not sure).

Analysis

Demographic and other participant characteristics were summarized with descriptive statistics including means and standard deviations for continuous variables and frequencies for categorical variables. Paired t-tests were used to compare pre- and post-intervention data for caregiver burden (individual variables and 6-item composite), and confidence (individual variables and 10-item composite). Significance was set at p<0.05. Cronbach’s alpha was calculated to determine reliability of the confidence and burden scales. All analyses were conducted using IBM SPSS Statistics Version 22. Participant comments were reviewed by one author (PS) and representative quotes were identified.

Results

Participant description

Of potential participants, 22 expressed interest in participation. Of potential participants, 19 met screening criteria, but 2 chose not to participate due to the time commitment. Of potential participants, 17 met with study personnel and were enrolled. Of the participants, 16 attended at least one class. Of the participants, one did not participate in any classes and her baseline data was excluded from analysis. Of the participants, 16/16 (100%) completed the post-intervention survey, although every participant did not answer every question. Of the participants, 12/16 (75%) attended all four workshops and 4/16 (25%) attended 3 workshops.

Of the participants, 16/16 (100%) were female (see Table 1). The mean age was 71 (SD 9, range 50–83). Of the participants, 15/16 (93.8%) were non-Hispanic whites. Average years of education was 15.3 (SD 3.3, range 12–23). Eight participants had worked in the healthcare field during their lifetime, including 4 nurses. The majority of participants reported the following care activities: 1) Be sure care recipient takes their medication(s) correctly (69%); 2) Attend appointments for medical care (100%); 3) Monitor symptoms such as medication side effects (69%); 4) Speak to doctor about physical or emotional needs (63%); 5) Make health care decisions for/with the care recipient (75%); 6) Plan and prepare meals/monitor diet (88%).

Table 1.

Demographics

Characteristics	N (%)	Mean (SD)
Female	16 (100)
Age (years)		70.8 (9.1) Range: 50–83
Race
White or Caucasian	15 (93.8)
Other	1 (6.3)
Highest education completed
High school	4 (25)
Some college	5 (31.3)
College or higher	7 (43.7)
Average number of paid days work/week outside of home		1.8 (2.9) Range: 0–7
Relationship to care recipient
Spouse/life partner	10 (62.5)
Child	1 (6.3)
Friend	5 (31.3)
Worked in healthcare field	8 (50)
Care recipient home
With caregiver	10 (62.5)
Independently in their own home or    apartment	4 (25)
Assisted living facility	1 (6.3)
Nursing home	1 (6.3)
Number of years helping care recipient (mean, SD)		6.8 (12.7) Range: 0.1–50
Number of additional unpaid people who also help care recipient (mean, SD)		1 (1.5) Range: 0–5
Self-reported health
Excellent	1 (6.3)
Very Good	10 (62.5)
Good	2 (12.5)
Fair	3 (18.8)
Poor	0

Confidence Questions

There was significant improvement in a 10 item composite of communication confidence with pre- and post-intervention scores of : 74.1 and 79.6 (See Table 2). There was also statistically significant improvement in mean scores for questions about use of reflective listening (6.9 and 8.0) and developing a list of concerns to discuss with the doctor (7.8 and 8.5). Internal consistency was good with Cronbach’s alpha of 0.82 for the 10-point confidence scale on the pre survey and 0.84 for the post survey.

Table 2.

Confidence (scale ranged from 0 for “never” to 9 for “nearly always”)

	Pre	Post
Question	Mean, SD	Mean, SD	95% CI	p-value
1) I am confident I can share my opinions when I  disagree with the doctor.	7.6, 2.3	7.9, 1.4	−1.6, 0.8	.48
2) It is easy for me to ask the doctor questions.	7.6, 1.7	8.4, 0.8	−1.8, 0.2	.09
3) I am confident I can prepare for a visit with  the doctor and the person I help.	7.9, 1.1	8.3, 1.1	−1.2, 0.4	.33
4) At the end of the doctor’s visit, I am sure of  what we need to do next.	7.1, 1.6	8.1, 1.2	−2.1, 0.3	.12
5) I am confident I understand my role right now  with the person I help.	8.1, 1.5	8.2, 1.4	−0.9, 0.8	.88
6) I am confident I can use reflective listening to  improve communication with doctors.	6.9, 2.2	8.0, 1.3	−2.1, −0.2	.03*
7) I am confident I can develop a list of concerns  that we should discuss when I go to the visit  with a doctor and the person I help.	7.8, 1.3	8.5, 0.8	−1.3, −0.05	.04*
8) I am confident I can tell the doctor what issues we want to discuss at the visit.	7.9 1.0	8.1, 1.5	−1.1, 0.4	.33
9) After a visit to the doctor, I am confident I can review the main points with the person I help.	7.3, 2.0	7.7, 1.4	−1.4, 0.6	.44
10) I am confident I can work out disagreements with a doctor.	6.8, 1.7	7.5, 1.6	−1.5, 0.2	.12
Composite score	74.1, 11.0	79.6, 10.0	−10.5, −0.6	.03*

^*p <0.05

Burden Questions

There was not a significant change in a mean composite score of the 6 caregiver burden questions (See Table 3). There was significant worsening for the question “How often do you feel strained when you are around the person you help?” from 3.9 to 5.2. The pre- post- change in the mean score of the question “How often do you feel that you will be unable to take care of the person you help much longer?” was not significant, although two caregivers had dramatic improvement of 8 or 9 points. Internal consistency was high with Cronbach’s alpha of 0.94 and 0.92 for the 6 burden questions on the pre and post surveys, respectively.

Table 3.

Burden (scale ranged from 0 for “never” to 9 for “nearly always” unless indicated)

	Pre	Post
Question	Mean, SD	Mean, SD	p-value
1) How often do you feel that, because of the time you spend with the person you help, you don’t have enough time for yourself?	4.8, 3.6	5.5, 3.4	.24
2) How often do you feel stressed between caring for the person you help and trying to meet other responsibilities for your family or work?	4.7, 3.5	5.9, 3.2	.054
3) How often do you feel strained when you are around the person you help?	3.9, 3.0	5.2, 2.9	.03*
4) How often do you feel uncertain about what to do with the person you help?	3.9, 3.2	4.4, 3.2	.47
5) How often do you feel that you will be unable to take care of the person you help much longer?	4.4, 3.7	3.8, 3.7	.52
6) In the past week, how challenging has it been to take care of the person you help?+	3.7, 3.2	3.9, 3.1	.77
Adjusted Composite Score	26.9, 17.6	30.4, 16.2	.11

^*p <0.05

⁺Scores standardized from original scale of 0 (not at all challenging) to 4 (very challenging) to a 0–9 scale for analysis.

At least 80% of participants reported using the following workshop skills: 1) Develop a list of concerns to discuss with the doctor (create agenda); 2) Tell the doctor what issues they want to discuss at the visit (share agenda); 3) Summarize what they heard to confirm understanding;

4) Express concerns in a positive manner; 5) Review with the care recipient the main points after a visit to the doctor; 6) Do something about a conflict with a doctor.

Program evaluation

On a 0–4 point scale, 15 program evaluation questions’ mean was 3.9 (SD 0.11, range 3.7–4.0, see Table 4). Of the participants, 15/16 (94%) rated the statement “The activities were useful learning experiences” a 4.0, with no answer from one participant. Of the participants, 15/16 (94%) would recommend classes to a friend, with no answer from one participant. Qualitative review of workshop comments was overwhelmingly positive. See representative quotes in Table 5.

Table 4.

Program Evaluation (scale: 0 for “strongly disagree” to 4 for “strongly agree”)

Statement	Mean	N	Range	SD
1) The workshop venue was comfortable.	4.0	15	NA	.00
2) The workshop venue was well located.	3.9	15	3–4	.35
3) The workshops’ content was relevant.	3.7	14	2–4	.61
4) The workshops’ content was comprehensive.	3.7	14	3–4	.47
5) The workshops’ content was easy to understand.	3.8	15	3–4	.41
6) The workshop handouts supported presentation material.	3.9	14	3–4	.36
7) The workshop handouts provided useful additional information.	3.7	15	2–4	.62
8) The workshop handouts were clear and well-organized.	3.9	15	3–4	.26
9) The workshops were well-paced.	3.9	14	2–4	.58
10) The workshops were a good mix between listening and activities.	3.8	15	2–4	.56
11) The workshops had enough breaks.	3.9	14	3–4	.36
12) The facilitators were knowledgeable.	4.0	14	NA	.00
13) The facilitators were well prepared.	4.0	15	NA	.00
14) The facilitators were responsive to questions.	3.9	15	3–4	.26
15) The activities were useful learning experiences.	4.0	15	NA	.00
Mean of all Scores	3.9			.11

Table 5.

Qualitative Analysis Representative Quotes

1) Being prepared is so important. I used to just walk in and try to remember everything I wanted to discuss, but would inevitably forget something. This is one part of the caregiving process I feel much more confident about and that reduces my stress immeasurably.

2) I did not think I needed this type of workshop, but my mind has been changed.

3) I can’t think of anything to do to improve the program. I’ve learned several things to help me do a better job.

Discussion

Our primary goal was to develop and refine the content and teaching methods of a small group behavioral change program intended to improve communication between caregivers of older adults and healthcare professionals.

One of the strengths of the Care Talks program is the method of development starting with an established Med Wise program and collaborating with caregivers and representatives from organizations that regularly provide training programs to older adults and caregivers during the development of the content. Another strength of this project is the development process for the evaluation. With exhaustive searching, we were unable to find suitable validated questions that specifically addressed all the key aspects of caregiver communication self-efficacy. We took questions intended for patients from previously published research and adapted the wording to make them appropriate for caregivers resulting in field tested questions that comprehensively evaluated communication confidence and caregiver burden with good levels of reliability.

We accomplished our primary goal of developing a well-accepted program with high attendance rates, many positive participant comments and 100% follow up. All 15 participants who answered the question affirmed they would recommend this program to a friend. With a very small sample size, we report modest, but statistically significant improvement in the mean composite score of communication confidence, use of reflective listening and developing a list of concerns to discuss with the doctor. Whether these improvements are clinically significant is a question to be answered in a future study. Given that healthcare communication is only one of many tasks and stressors that impact caregiver burden, it is not surprising that the changes in the mean composite burden score were not significant. This aligns with findings from many studies show minimal improvement in burden even when there are significant improvements in caregiver burden symptoms. (Adelman, 2014) We did find a significant worsening for the question “How often do you feel strained when you are around the person you help?” This may be a result of our intervention or the normal progression of caregiver feelings over time and can be clarified with a future randomized trial. Although the mean scores for the question about caregivers feeling they will be unable to take care of the recipient much longer were not significant, two caregivers had dramatic improvement of 8 or 9 points on a 0–9 point scale. These outliers deserve further evaluation if they occur again in future studies.

Our study has several limitations. A very small sample size with mostly white, all female participants from rural Wisconsin limits the generalizability of our results. Half of the participants had worked in the healthcare field including four nurses, although we would expect professional healthcare experience might reduce the benefits of this program. Beyond distribution of key point handouts, this project did not include other educational reinforcement activities such as follow up telephone calls or group meetings. In addition, our post-intervention evaluation was only one month after the final workshop, so we do not know if use of communication skills or confidence will change over time.

Unfortunately, we did not systematically gather information about the health or healthcare needs of the care recipients, so it is unclear how these important factors influenced our results. One might anticipate that caregivers for care recipients with multiple medical conditions would benefit more from our program, but they also might find it more difficult to attend four 2 hour workshops as caregiving demands increase. Understanding caregiving demands and ability for the caregiver to safely leave the care recipient unattended for several hours or in the care of others will be important factors to evaluate in future studies as they will impact engagement and feasibility.

Although some questions were unchanged, another limitation is many of our confidence questions were based on questions used in previous research, but had been modified slightly to be appropriate for caregivers and improve the understandability. None of the modified questions have been formally validated. Some questions from the recently published Partnering for Better Health - Living with Chronic Illness: Dementia (Sadak, 2015) may be appropriate for future studies.

There appears to be substantial need and potential benefits for the program we developed. There is much research about effective doctor-patient communication (Stewart, 1995, Fong Ha, 2010) that has shown improvement in a variety of health outcomes.(Zolnierek, 2009) We know that limited patient health literacy is associated with an increased frequency of hospitalizations and Emergency Department visits. (Berkman, 2011) Yet, with a rigorous literature search, we are unable to find any reports of caregiver-physician communication interventions and their impact on care recipient healthcare or health outcomes. Nor are we able to find any evidence-based educational programs targeting caregiver-healthcare professional communication currently in use.

There are significant barriers that will have to be overcome to implement this program on a broader scale. Recruitment was by far our biggest challenge. Some potential barriers to recruitment include use of marginally effective indirect methods such as announcements in newsletters, fliers and posters. Another barrier is people who are serving as informal caregivers often do not identify themselves in that role (O’Connor, 1999, Henderson, 2001, O’Connor, 2007) or see the potential benefit this program can offer them now or in the future. We thought concerns about leaving care recipients home alone might be a barrier, yet none of the participants used the respite care we offered. Why this potential facilitator for attendance was not used will require more research.

The Care Talks program was acceptable to the participants and they used many of the skills covered in the program. We found some promising results with caregiver confidence for communication with healthcare professionals and attitude about their ability to continue caregiving. However, many questions about the scalability and impact on health and healthcare outcomes for caregivers and care recipients remain to be answered. Further research is needed to determine if this program will have long term effects on the health of a more diverse population of informal caregivers and care recipients or the healthcare they receive.