Effects of In-Person and Online Chronic Disease Self-Management Education (CDSME) Programs on Reducing Loneliness: A Research Proposal and Overview
Alexander Fitzgerald
Introduction
Problem and Impacts
Loneliness refers to a discouraging emotion arising from lacking meaningful social relationships and it is a public health problem that interferes with quality of life among adults with chronic disease. Chronic disease is accompanied by a series of physical and mental health issues which may make it more difficult to interact with others and maintain social relationships (Christiansen et al., 2020; Smith et al., 2022). The burden of disease may also increase difficulty with tasks of daily living, decreasing available time to foster social interaction. This leads to social isolation which may exacerbate existing or cause new physical and mental health problems (Christiansen et al., 2020; Douma et al., 2021). According to social cognitive theory, feelings of social isolation are one of the most significant risk factors that influence behavior which, in turn, diminishes the ability to maintain healthy behaviors within the environmental context. Loneliness can also lead to other consequences for people with chronic diseases such as higher cost of care, diminished sense of identity and purpose, low self-efficacy, and inability to manage one’s own health (Chen et al., 2018; Lenferink et al., 2018).
Currently, around 51.8% of adults in the U.S.A have a diagnosis of at least one chronic disease with 27% having two or more comorbid chronic conditions (Boersma et al., 2020). Medical costs for chronic disease may vary by disease but the overall cost exceeds $1 trillion per year (Hacker, 2024). Additionally, the elderly population is projected to reach nearly double the current population by 2050. As elderly people face the highest burden of chronic disease, the increasing elderly population means significant increases in the medical costs for chronic disease care (Maresova et al., 2019).
Due to the consequences of loneliness on health outcomes in adults with chronic diseases, as well as the prevalence and cost of chronic disease today, it is important to advance research regarding understanding how interventions for adults with chronic disease can address loneliness as a route for maximizing ability to self-manage individual health needs.
Importance to Social Work
The proposed research study is important to the field of social work by upholding ethical values in line with the NASW Code of Ethics (2021) and advancing the mission of social workers. In exploring the connection between loneliness and chronic disease, the study directly promotes the importance of human relationships and recognizes them as a pathway for change. The results will inform future practice around addressing loneliness and social support allowing for more effective support provided to clients dealing with these issues. Competence is also promoted by expanding knowledge in this area of practice for more application in the field. Knowledge on chronic disease and how to improve social support and health management in this population will put social workers in a position to increase quality of life for clients. Better health outcomes will allow social worker and client teams to focus on other issues requiring support. Finally, this study promotes social justice and equity by developing interventions for improved quality of life for the vulnerable population of adults with chronic diseases. Analyzing virtual and in-person interventions may also highlight methods of minimizing costs, time commitments, and mobility issues in administering such treatments. This information on delivery method effectiveness will help social workers make educated decisions about the best method of intervention administration for a client’s specific needs and obstacles, expanding accessibility of support.
Literature Review
Prior to the development of the research question, a thorough literature review was conducted to inform the direction of this study. Several studies were found to be relevant.
A study conducted by Smith et al. (2022) examines the effectiveness of Chronic Disease Self-Management Education (CDMSE) programs on reducing loneliness in people with chronic disease. Researcher’s hypothesized that if a person who participated in in-person CDSME workshops will have lower levels of loneliness. The study employed a Classic Pre-test Post-test design and the sample consisted of adults over age 50 with a chronic disease. They used a convenience sample recruited at multiple sites like libraries and senior centers. The findings demonstrated significant improvement in loneliness scale scores from baseline to six weeks. In an adjusted model, improvement was found to be more significant for participants who were younger, Asian, and/or had a higher number of chronic diseases.
This study also had some limitations. Firstly, the study relied on self-report data which could be impacted by recall or social desirability bias. Secondly, the study was conducted over six weeks with no comparison group used which limits the external validity of the results. Recommendations for future research are to examine the effects on loneliness of virtual CDSME programs as well as incorporating comparison groups into future studies. (Smith, 2022)
A second study investigated the potential benefits of group identification with a chronic disease self-management group on improving self-efficacy, and correspondingly, self-management skills. Researcher’s hypothesized that if a person identifies strongly with the group, then they will have better self-efficacy and self-management ability. A Single Group Pre-Test Post-Test design was used with secondary data from a previous study. The sample consisted of 213 adults aged 44-88 with a diagnosis of a chronic condition. The results suggest that social identification in a group-based self-management program predicts improvement of self-efficacy, and thus, self-management skills. Improved self-efficacy predicted better mental and physical health state at a four month follow-up (Cameron et al. 2018).
Some limitations in this study included a sample size below the optimal limit for evaluation as well as an inability to determine causality due to the correlational nature of the data used. Researchers recommend future studies explore benefits beyond four months and interventions should promote self-identification and group membership for positive health outcomes (Cameron et al, 2018).
In 2021, Douma et al. conducted a study to evaluate the effectiveness of Op Koers Online group interventions on improving health coping strategies in adolescents with chronic disease. The hypothesis stated if adolescents participate in Op Koers Online program then they will demonstrate increased self-management abilities. Using a classic pre-test post-test design, an availability sample was gathered from local hospitals consisting of adolescents ages 12-18 who had a current chronic disease diagnosis. 89% of subjects completed a majority of the sessions and findings demonstrated improvement in health-related coping skills for the treatment group while no improvement was seen in the control group.
This study had a few limitations including a lower response rate leading to a sample size below the desired number of participants. Response rate could also not be calculated due to the online recruitment. The results may be beneficial in Outcomes were also self-reported and have risk of recall or social desirability bias. For the future, researchers advise programs to increase parental involvement and work to maintain benefits at one-year follow-up. Research studies should further evaluate the connection between quality of life and disease-related coping skills (Douma et al., 2021).
Another study, carried out by Christiansen et al., (2020), aimed to investigate the association between social isolation (SI) and loneliness as predictors of chronic disease. Researchers hypothesized that if social isolation and loneliness interact, then an individual is at higher risk for developing chronic disease. Using a cross-sectional cohort study, secondary data from a 2013 study with a sample of adults aged 35-79, was analyzed. The results of this study suggest that the interaction between SI and loneliness increases the risk of developing chronic disease, more than the presence of either SI or loneliness alone. Findings indicated that both psychological and behavioral pathways can partially explain the association of SI and loneliness with chronic disease.
This study faced several limitations. Firstly, causality could not be established. Additionally, the presence of a confounding variable could not be ruled out. Finally, care should be taken with analyzing data as the mediation model used for analysis was not a perfect fit. Future research should be conducted on the relationship of loneliness and SI with specific chronic diseases as well as aim to establish a causal relationship (Christiansen et al., 2020).
Finally, Chen et al. (2018) designed a study to explore the relationship between social support and health literacy on self-management skills in people with Chronic Kidney Disease (CKD). They hypothesize that if a person has higher levels of social support and health literacy then they will have better self-management skills. Quota sampling was applied to secondary data to control the proportion of participants at each disease stage. The final sample included 410 patients. The results suggest that both social support and health literacy are predictors of self-management skills, but social support is the strongest indicator of better self-management skills, health outcomes, and quality of life. Social support from healthcare providers and family were found to be significant.
The first limitation of this study is that participants tended to be older, less-educated, and dealing with later-stage CKD which may decrease generalizability. Another limitation is that self-report information may have recall or social desirability bias. Finally, no causal relationship could be established. In future research, a system-approach should be used to evaluate the social support pathway as well as identify other explanatory factors. Additionally, research should examine the link between health literacy and possible mediating variables for self-management (Chen et al, 2018).
Previous studies have provided a lot of direction for future research that has inspired this study. Research has shown self-efficacy to be a mediating variable between health and social support, but more research is needed in this area. By studying loneliness and self-efficacy as predictor variables of self-management skills, this study will help further explain this relationship. Furthermore, past studies have demonstrated that in-person workshops may improve loneliness, but this was not examined virtually (Cameron et al., 2018; Smith et al., 2022). A separate study found positive effects on loneliness and self-management for chronic disease via online intervention, but with an adolescent population (Douma et al, 2021). This study aims to fill this gap in the research by exploring the complex relationship between self-management, loneliness, and self-efficacy in adults with chronic disease by comparing in-person and virtual interventions in the same study.
Hypothesis
The finalized study will aim to examine the effectiveness of in-person as opposed to virtual self-management interventions in impacting loneliness and self-efficacy. The researcher hypothesizes that if participants are in the in-person intervention then they will score lower loneliness and higher self-efficacy levels than participants in the virtual intervention.
Theoretical Background
This research idea and hypothesis have also been informed by social cognitive theory. Social cognitive theory states that personal and environmental factors can interact to shape behaviors, as well as the idea that behavior is partially learned through observation and social interaction. Further, the theory looks at self-efficacy as an important factor for behavior changes (Sell et al., 2016). Loneliness or lack of social interaction are factors which could impact behavior change. Increased loneliness can hinder behavior changes by diminishing an individual’s sense of self-efficacy. Self-efficacy is essential to proper self-management of chronic disease. As a lot of adults with chronic disease struggle with loneliness, this could impact their self-efficacy and self-management skills. This study explores reducing loneliness in adults with chronic disease as a route to behavior change for improved health management. The in-person intervention has already been shown to reduce loneliness. The researcher hypothesis suggests that the in-person sessions will produce more group interaction and social engagement to encourage meaningful relationships. Higher volume of significant social interactions will be more effective at reducing loneliness, and, following social cognitive theory, then lead to more significant increases in self-efficacy and self-management skills.
METHODOLOGY
Research Design
The experiment will follow a Classic Experimental Pre-test Post-test research design.
This design involves two groups, a treatment group and a control group. The first step is to randomly assign the participants to either the control or treatment group. Before the treatment is administered both the DV for both groups will be measured, allowing for additional measurements to be collected. The two groups will then undergo their intervention before having the DV measured again. This design allows comparison of DV before and after experiencing a treatment and also allows a causal relationship to be determined.
Researchers chose this design for the study for several reasons. Firstly, this design was selected in order to have a comparison group to improve internal validity. Additionally, this design involves random assignment which is important to eliminate selection bias and ensure higher internal validity. A design involving both a pre-test and post-test measure was also used because it can better establish and measure a causal relationship (Engel & Schutt, 2017).
Sampling Method
For this study, researchers will use snowball sampling. Snowball sampling begins with identifying a few individuals in the population to participate. These individuals will be given the information about the study and asked to share this study with other individuals they know who may meet the study criteria. The sample is built from participant connections (Engel & Schutt, 2017). The criteria will require participants to be adults aged fifty and older with at least one chronic disease. Chronic disease will be operationalized as a disease lasting one year or more involving physical or mental symptoms that interfere with regular daily living activities. Participants must be able to access both the in-person and virtual settings to be considered.
The initial participants for this study will be recruited via medical facilities. As individuals with chronic disease require regular follow-up, local primary care physicians will be informed about this study and its requirements and asked to encourage eligible patients to participate. Additional participants will be recruited by the initial participants.
There is no sampling frame for individuals with chronic disease which makes it very difficult to obtain a randomized sample from this population. Because this would be nearly impossible, an availability method was preferred. Snowball sampling was the preferred method of availability sampling because it will allow quicker recruitment of participants and also help to recruit participants who do not need to see their physician as regularly.
Advantages of using this method are that it is quicker and more flexible, allows collection of a sample without an available sampling frame, and low-cost. A major disadvantage of this method is that the sample will not be representative of the population so there is very low ability to generalize. Additionally, there can be issues with data analysis because many statistical tests assume a random sample so the results may be inaccurate (Engel & Schutt, 2017).
Measurement Strategies
The independent variable is the treatment type which may be online or in-person. The dependent variables will be level of loneliness and level of self-efficacy. The independent variable is a nominal variable and both dependent variables are measured on a ratio level.
Loneliness will be assessed using the Three-Item Loneliness Scale. Participants can answer on a three point scoring scale with 1 meaning hardly ever and 3 meaning often. Higher scores indicate a higher level of loneliness. The Three-Item Loneliness Scale was chosen because it has high alternate forms reliability with a correlation of 0.82. This scale also had good convergent validity with other scales for assessing loneliness. Additionally, this scale only had three questions to avoid participants tiring out during the survey (Hughes et al., 2004).
For self-efficacy the scale used would be the Self-Efficacy for Disease Management Measure. This scale is a three-item scale and scored from 1-10 with 1 being not at all confident and 10 being very confident. Similarly, the Self-Efficacy for Disease Management Measure also had high reliability with a Cronbach’s alpha of 0.91. This measure was designed specifically for use with CDSME programs making this scale the most appropriate to use for this study design. Finally, this survey also only has three items ensuring participants do not tire out and can answer all questions. (Melchior et al., 2014).
Both loneliness and self-efficacy will be assessed at baseline, immediately after the intervention, and at the six week point. The surveys will be administered virtually for the virtual intervention group and by paper for the in-person intervention group.
Data Analysis
Both descriptive and inferential statistics will be used to analyze the data and explain the results. Demographic data, such as race or gender, will be described using frequencies. The nominal nature of these variables makes a frequency table the most effective method to display this data. The independent variable is also nominal with only two categories, so frequency can also be used to indicate the distribution of participants across these categories. Both dependent variables are measured as ratio level data making measures of central tendency more appropriate descriptors. Specifically, median and range will be used to share results on loneliness and self-efficacy in order to avoid any inaccuracies that may come from skewness if mean was used.
For inferential statistics, comparative statistics are the most appropriate because this design involves contrasting two different treatment groups by method of intervention delivery. Because this study has identified a categorical independent variable and two continuous dependent variables, MANOVA is the most appropriate method of analysis as it is the only one meant for multiple dependent variables (Engel & Schutt, 2017).
Ethical Considerations
There are many ethical considerations to consider in regards to this research study. First, it is essential that participants provide voluntary, informed consent before any involvement in the study. Packets detailing the purpose of the study, the methodology, risks, and other relevant information will be provided to each participant. Participants will also be provided researcher contact information in case there are any questions. Each participant must return a signed form acknowledging they have read and understand the informed consent document.
A second significant consideration is privacy. Because one group is in-person and one is virtual, there are unique challenges for maintaining anonymity. Virtual groups will be required to have their cameras turned on in order to maintain equal anonymity between groups. Snowball sampling makes it likely that participants will know others in their treatment group but to preserve more anonymity only first names will be used in sessions. No names or identifying information will be published. Participants' names will be removed from their data and coded as a number only in order to preserve confidentiality. As this data is not particularly sensitive, Certificates of Confidentiality will not be used.
Because this is human subjects research, this study also requires review by the IRB for additional assessment of ethical needs. This study would be classified as expedited because it is not anonymous but involves minimal risk to the participants (Engel & Schutt, 2017). At this time the researcher has no conflicts of interest which may impact the study.
Study Limitations
There are a few limitations with this study’s methodology that are important to be aware of. Firstly, the use of snowball sampling worked best for this population, but it will likely only yield a small sample. This means the study will be heavily impacted by any participant attrition or incompletion of the program and surveys. Secondly, the study results will have low external validity because snowball sampling does not produce a random sample. The sample will not be generalizable because it is not representative of the population.
Another limitation is the appropriateness of the measures selected. While both measures have high reliability and the Three-Item Loneliness Scale also had good convergent validity, validity was not reported for the Self-Efficacy for Disease Management Measure (Melchior et al., 2014). The lack of assessment of the validity of this measure may lead to inaccuracies in the results. Finally, MANOVA was selected for the comparative analysis in this study because two dependent variables are being examined. While MANOVA is the most appropriate measure for multiple dependent variables, it also requires a categorical independent variable with at least three categories, and the independent variable of this study only has two. This may mean the analysis will not entirely fit the data and produce some inaccuracies in results.
Monitoring and Implementation Guidelines
To ensure the intervention is carried out and delivered as planned, the inputs were reviewed. Sampling strategy for obtaining participants was already established, but trained program facilitators are also needed to lead the programs. A description of the research project and position and conduct thorough interviews of applicants to determine their fit. To ensure the program leaders are properly trained, the researcher will provide a required training, whether or not they have had previous training. The materials for the CDSME program will be obtained directly from organizations administering the program. Researchers will reach out to the National Council on Aging, Administration for Community Living, and other organizations which run CDSME programs in order to obtain the necessary materials. The experiment will also require both an in-person and virtual space to conduct sessions. The in-person space will be a room rented from an accessible local library or community center nearby the original recruitment site. The virtual sessions will be run from a computer within the designated research lab space or office in order to protect privacy and ensure reliable internet access. A recurring zoom link will be emailed to participants to access the session. Taking these steps will ensure the intervention delivered is high-quality and delivered effectively to participants.
In implementing the intervention, the first aspect to monitor ensuring participants are aware of all the details regarding location and timing. The in-person location and virtual link will be shared with the relevant participants two weeks prior to the first session. A reminder will be sent three days prior. For every session including the first, a reminder with the information will be sent out the morning of the session. An RSVP form will be included in the two week and three day notice emails to help verify participants have received and seen the information. Regarding the virtual sessions, to avoid technical difficulties eating into session time, researchers will test both the instructor and participant side of the zoom link to confirm it is functional. Instructors will be provided with lesson plans and training well in advance, but to ensure the lesson plans are followed researchers will supervise sessions. For the virtual sessions, the researcher will be in the room with the instructor to listen to the session, but out of view of the participants in order not to interfere with the experiment. For the in-person session, the researcher will set up and watch a live stream of the classroom. This will only be viewed by the researcher for monitoring and will not be saved or recorded after the session to ensure confidentiality.
Additional outcomes such as increased level of social support, and increased knowledge and ability to self manage and engage in positive health behaviors will be evaluated. A survey asking about perceived social support, self-management skills, and health behaviors will be administered at the same points the DV measures will be administered, pre-intervention, immediately post-intervention, and six weeks after, for outcome evaluation.
Finally, in order to assess the effectiveness of this intervention, a cost benefit analysis can be run after completion comparing the costs of conducting the intervention to the costs of hospitalization and healthcare that would be needed if participants did not receive the intervention.
Application of Evaluation Research
Cost benefit analysis will demonstrate whether this intervention successfully addresses the problem in a cost-effective way (Engel & Schutt, 2017) . The implications of these findings are primarily applicable to macro society. If the program saves on costs, the amount of healthcare spending required by the government would decrease allowing extra money to be used for other programs like SNAP. This would improve practice effectiveness by allowing more individuals in need of these programs to access them, increasing delivery speed and efficacy, and prioritizing the most-needed programs.
From the opposite side, if the program loses money it would not be worth it for the government to implement as it would be a waste of money and resources. This would mean no additional resources would be freed up preventing any improvement in services or delivery. While it would not negatively impact practice because the government budget would not decrease, it would also mean there would be no increase in practice efficiency.
Overall, the proposed research will impact multiple systems levels by advancing understanding of the relationship between loneliness and chronic disease, the benefits of virtual compared to in-person interventions for loneliness, and informing future research and practice for adults with chronic disease and beyond.
References
Boersma, P., Black, L. I., & Ward, B. W. (2020). Prevalence of multiple chronic conditions among US adults, 2018. Preventing Chronic Disease, 17. http://dx.doi.org/10.5888/pcd17.200130
Cameron, J., Voth, J., Jaglal, S., Guilder, S., Hawker, G., & Salbach, N. (2018). “In this together”: Social identification predicts health outcomes (via self-efficacy) in a chronic disease self-management program. Social Science & Medicine, 208, 172-179. https://doi.org/10.1016/j.socscimed.2018.03.007
Chen, Y., Chang, L., Liu, C., Ho, Y., Weng. S., & Tsai, T. (2018) The roles of social support and health literacy in self-management among patients with chronic kidney disease. Journal of Nursing Scholarship, 50(3), 265-275. https://doi.org/10.1111/jnu.12377
Christiansen, J., Lund, R., Qualter, P., Andersen, C., Pedersen, S., & Lasgaard, M. (2020). Loneliness, social isolation, and chronic disease outcomes. Annals of Behavioral Medicine, 55(3), 203-215. https://doi.org/10.1093/abm/kaaa044
Douma, M., Maurice-Stam, H., Gorter, B., Houtzager, B., Vreugdenhil, H., Waaldjik, M., Wiltink, L., Grootenhuis, M., & Scholten, L. (2021). Online psychosocial group intervention for adolescents with a chronic illness: A randomized controlled trial. Internet Interventions, 26. https://doi.org/10.1016/j.invent.2021.100447
Engel, R. J. & Schutt, R. K. (2017). The Practice of Research in Social Work. Thousand Oaks, CA: Sage Publications
Hacker, K. (2024). The burden of chronic disease. Lifestyle Medicine, 8(1), 112-119. 10.1016/j.mayocpiqo.2023.08.005
Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. (2004). Three-Item Loneliness Scale. PsycTESTS. https://doi-org.proxy-bc.researchport.umd.edu/10.1037/t29584-000
Lenferink, A., Van der Palen, J., & Effing, T. (2018) The role of social support in improving chronic obstructive pulmonary disease self-management. Expert Review of Respiratory Medicine, 12(8), 623-626. https://doi.org/10.1080/17476348.2018.1489723
Maresova, P., Javanmardi, E., Barakovic, S., Husic, J. B., Tomsone, S., Krejcar, O., & Kuca, K. (2019). Consequences of chronic diseases and other limitations associated with old age - A scoping review. BMC Public Health, 19. https://doi.org/10.1186/s12889-019-7762-5
Melchior, M. A., Seff, L. R., Albatineh, A. N., McCoy, H. V., & Page, T. F. (2014). Self-Efficacy for Disease Management Measure. PsycTESTS. https://doi-org.proxy-bc.researchport.umd.edu/10.1037/t33758-000
National Association of Social Workers. (2021). NASW Code of Ethics. https://www.socialworkers.org/About/Ethics/Code-of-Ethics/Code-of-Ethics-English
Sell, K., Amella, E., Mueller, M., Andrews, J., & Wachs, J. (2016). Use of Social Cognitive Theory to assess salient clinical research in chronic disease self-management for older adults: An integrative review. Open Journal of Nursing, 6(3), 213-228. DOI: 10.4236/ojn.2016.63022
Smith, M., Chen, E., Lau, C., Davis, D., Simmons, J., & Merianos, A. (2022). Effectiveness of chronic disease self-management education (CDSME) programs to reduce loneliness. Chronic Illness, 19(3), 646-664. https://doi.org/10.1177/17423953221113604
