The Science of Diabetes Self-Management and Care2023, Vol. 49(2) 136 –149© The Author(s) 2023Article reuse guidelines:sagepub.com/journals-permissionsDOI: 10.1177/26350106221149665journals.sagepub.com/home/tde
AbstractPurpose: The purpose of this study was to explore how treatment adherence and lifestyle changes required for glycemic control in type 2 diabetes (T2D) are related to quality of life (QoL) among predominantly ethnic minority and socioeconomically disadvantaged adults engaged in making changes to improve T2D self-management.
Methods: Adults with T2D in New York City were recruited for the parent study based on recent A1C (≥7.5%) and randomly assigned to 1 of 2 arms, receiving educational materials and additional self-management support calls, respectively. Substudy participants were recruited from both arms after study completion. Participants (N = 50; 62% Spanish speaking) were interviewed by phone using a semistructured guide and were asked to define QoL and share ways that T2D, treatment, self-management, and study participation influenced their QoL. Interviews were analyzed using thematic analysis.
Results: QoL was described as a multidimensional health-related construct with detracting and enhancing factors related to T2D. Detracting factors included financial strain, symptom progression and burden, perceived necessity to change cultural and lifestyle traditions, and dietary and medical limitations. Enhancing factors included social support, diabetes education, health behavior change, sociocultural connection.
Conclusion: QoL for diverse and socioeconomically disadvantaged adults with T2D is multifaceted and includes aspects of health, independence, social support, culture, and lifestyle, which may not be captured by existing QoL measures. Findings may inform the development of a novel QoL measure for T2D.
Diabetes prevalence has increased significantly over the past 2 decades in New York City, disproportionately affecting the city’s low-income residents and people of color. Several efforts have been made to optimize treatment modalities to address disparities in priority populations involving regular medical monitoring and patient involvement in diabetes self-management through education and support.1,2 The diagnosis, progression, and symptom burden of type 2 diabetes mellitus (T2D) have significant impacts on quality of life (QoL) because many facets of disease experience involve daily maintenance and engagement. These impacts increase throughout disease progression, often due to burden of treatment and self-management regimens.3-5 The demands of intensive treatment of diabetes are thought to be offset by the preservation of health and prevention of complications of diabetes; however, major clinical trials of intensive treatment approaches have found no benefit for QoL.6-8 Adults with T2D often view treatment intensification with insulin negatively and anticipate a negative impact on their QoL that is similar in magnitude to that expected from the experience of diabetes complications.9-11 Similar patterns have been found in the association of blood glucose self-monitoring increasing psychological stress, which in turn affects QoL.12 Furthermore, QoL is infrequently accounted for when planning and selecting interventions for individuals within neighborhoods that are disproportionately impacted by diabetes and systematic inequities. Ideally, interventions designed to improve overall health would also improve or sustain the individual’s QoL, prioritizing patientcentered care.
Patient-reported outcomes (PROs) provide an opportunity to focus on aspects of the patient experience that can provide evidence of treatment benefits and burdens that are not captured by blood tests and clinical measures.13-16 For this reason, the US Food and Drug Administration (FDA) recommends inclusion of PROs in the evaluation of treatments and interventions.17 Current treatment recommendations from the American Diabetes Association emphasize that treatment decisions should be based on a patient’s “preferences, needs, and values,” set improvement of QoL and health outcomes on equal footing,18,19 and facilitate adherence and long-term glycemic control.20 Shared decision-making and minimally disruptive approaches to diabetes care are increasingly valued for their patient-centeredness and ability to prioritize QoL as a goal of person-centered treatment.21,22 However, little is known about how individuals with diabetes conceptualize their QoL or how diabetes self-management and lifestyle changes influence QoL.
Despite the clear importance of QoL as a PRO, construct definitions and measurement approaches have been inconsistent. The World Health Organization defines QoL as an individual’s perception of their “position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns.”23 In contrast, the FDA defines QoL as a multidimensional concept with physical, psychological, and social components. Additionally, the Centers for Disease Control and Prevention focuses on the specific impact of illness (or the lack of illness) on QoL, which is often referred to as health-related quality of life (HRQOL). HRQOL is defined as “an individual’s or a group’s perceived physical and mental health over time.”24 Diabetesrelated QoL literature suggests including the effects of the experience of diabetes and its treatment on a full range of personal life experience.25 Experts have defined diabetes-related QoL as a multidimensional concept of an individual’s subjective evaluations of physical, social, and emotional well-being26; as a multifaceted, dynamic element that is known purely from the individual’s perspective27; as a subjective perception a person has of their position in life28; and as physical and social functioning and perceived physical and mental well-being.9 Across definitions, there is a trend to prioritize the individual’s perception. Beyond this point of consensus, there is a disagreement on how to define and measure QoL. Rigorous thematic analyses that explore individuals’ definitions of QoL were generally not used in the studies that have informed these various definitions or in the development of the measures that were informed by these definitions. Fundamental disagreements about whether available measures adequately capture QoL in diabetes remain unresolved, especially with respect to how often QoL measures are contaminated by item content that is more reflective of health status than QoL.29
Previous research indicated a possible negative influence of diabetes treatment intensity on QoL because insulin-treated adults with T2D reported worse functional status and other aspects of QoL compared to those taking oral medication.4,9 Furthermore, the development of diabetes-related complications has been described as one of the most significant diabetes-related impacts on QoL.9,11 As indicated by the results of a review on QoL measurement, currently used QoL scales have tended to focus on items that involve evaluations of health status, symptoms, and changes in functioning.29-32 The Diabetes 39 Scale,33 the Audit of Diabetes-Dependent Quality of Life, and the Diabetes Quality of Life scale34 all serve as assessments of diabetes QoL. However, these measures were not derived directly from the responses of patients from minority backgrounds and communities impacted by structural and systemic inequities. When examining patterns specifically among individuals with T2D and low-incomes, QoL is often impacted35 because multiple elements of socioeconomic vulnerability (education opportunity, income, employment) are related to reduced QoL and tend to cluster within the population of focus.36 This highlights the importance of understanding factors that influence QoL among socioeconomically disadvantaged adults with T2D. The current study aimed to better understand how changes in treatment adherence and lifestyle modifications made to better control T2D may, in turn, impact QoL. This information may guide future development of diabetes-related QoL measures that can better capture the experience ethnic minority and socioeconomically disadvantaged populations impacted by diabetes and systematic inequities.
Qualitative study participants were drawn from a parent trial evaluating different approaches to diabetes care management. The NYC Care Calls Study was designed as a 12-month prospective randomized trial that examined the effectiveness of telephonic self-management support (Tele-SMS) delivered by health educators. Participants were randomized to either enhanced usual care, which provided print materials on diabetes self-management, depression, and diabetes-related distress, or Tele-SMS in addition to the same print materials.37 The current qualitative substudy was conducted after active recruitment of the parent study had concluded. Individuals that participated in the qualitative interview element were not approached until they had completed the parent study.
During the final 6 months of the parent trial, participants were approached for optional participation in the current substudy via telephone outreach. The interview opportunity was only discussed with participants after they completed their final survey for the parent study. All participants that had completed study participation received a phone call with the opportunity to participate. Participants received up to 3 outreach calls to introduce the qualitative opportunity and schedule. Interviews were offered in both Spanish and English. As such, the substudy followed the same inclusion and exclusion criteria as the parent study.
For those willing to participate, a phone interview was scheduled at their convenience. Qualitative interviewing was used to understand how these participants thought about their QoL, how diabetes affects QoL, and how the changes they made through the parent study program were related to their QoL. A gift card valued at $20 was mailed after the interview. Interviews were audio recorded, uploaded to a secure network drive, and deleted from the recorder afterward. Oral consent via telephone was obtained. All recordings were identified using the unique participant ID, and no names or identifying information were used during the interview. Interviews were then transcribed by GMR Transcription services (GMR Transcriptions Services, Inc, Tusin, CA). The phone interview consisted of 5 questions and took approximately 10 to 30 minutes to complete.
Qualitative research generally requires a sample size that permits a thorough and diverse collection of experiences, focusing specifically on the quality and saturation of responses regarding the research goal.38 A total of 50 participants were interviewed and asked to describe QoL in their own words after parent study completion (N = 50; 72% Latino, 20% non-Latino Black, 4% non-Latino White; 62% Spanish speaking; see Table 1). Approximately half of the interviews were conducted with those randomized to enhanced usual care and half to those randomized to Tele-SMS.
A team composed of a clinical health psychologist, an anthropologist, and a psychology graduate student developed interview questions based on clinical and research experiences and existing literature. The 5 open-ended questions and follow-up probes are provided in Figure 1. The focus of the interview development and structure was to provide an opportunity for patients to define QoL in their own words, avoiding questions that may have dichotomous answers. Probes and examples were used to aid patients in instances that they reported that they would benefit from examples or guidance. The interview script was approved by both the New York City Department of Health and Mental Hygiene and Albert Einstein College of Medicine Institutional Review Boards.
This thematic analysis involved multiple phases to ensure accurate and reliable result reporting.39 In the first phase, 3 coders (clinical health psychology graduate students with qualitative research training and experience) used independent open-coding to read 5 transcripts and highlighted relevant text. Coders then met to create an initial codebook to serve as a flexible guide for coding the next sequence of transcripts. Coders agreed on edits to wording and themes before beginning the next phase of coding—applying the edited codebook to 10 additional transcripts. After completion of this phase, the codebook was finalized and used to code the remainder of the transcripts. The coding team completed coding of each transcript individually, meeting weekly to discuss codes, resolve discrepancies, and finalize documents to be evaluated for saturation. Coders were blinded to participant randomization to ensure unbiased coding, and all interviews were coded using the same codebook regardless of intervention assignment. Finalized coding of each transcript was entered into NVivo 12 qualitative data analysis software (NVivo, qualitative data analysis software; QSR International Pty Ltd. Version 12, 2018). The first aim of qualitative analysis was to create a codebook through themes and subthemes across the full sample of transcripts. Next, patientdefined QoL was explored and further evaluated for thematic saturation within definitions.
Thematic saturation was achieved after coding 10 interviews. Eight codes emerged, including subcodes to illuminate patient-reported details and areas of focus, and reached saturation during the initial coding process of the first 10 interviews. Codes ranged in percentage of interview citation frequency (see Table 2). Codes emerged following trends of diabetes-related experiences enhancing (social support, diabetes education, health behavior change, sociocultural connection) and/or detracting (financial strain, symptom progression and burden, perceived necessity to change cultural and lifestyle traditions, dietary and medical limitations) from QoL. Differences between conditions were not specifically evaluated and were not detectable at a thematic level.
Participants were asked about their own definition of QoL (Figure 1) and provided a range of responses. QoL definitions tended to be multidimensional with enhancing and detracting factors interspersed throughout interviews, illustrating the fluidity and diversity in topics that are subsumed under QoL from the patient perspective. Enhancing and detracting elements of diabetes on QoL were identified as themes and subthemes, often appearing as a part of QoL definitions and expanding beyond to illustrate T2D patient experience.
Given the broad scope of self-management practices in T2D care, this theme involved subthemes that spanned across enhancing and detracting factors of QoL. Participants reported A1C and glycemic control, exercise, and weight management as factors predominately enhancing QoL. This trend supports the importance of self-management practices. Furthermore, self-management engagement provides an opportunity for overall health improvement, which in turn positively impacts QoL. Dietary changes and medication adherence concerns were reported to negatively impact QoL. Dietary restrictions were sometimes described as preventing individuals from enjoying preferred or culturally meaningful foods. Medication adherence was often linked to side effects, and modifications to medication regimens were often perceived as necessary to avoid side effects. Overall, 98% of participants included elements of self-management within their QoL definitions.
Participants cited social support as a positive influence on their QoL, including support from NYC Care Calls staff, health care providers, family, and friends. All participants (100%) included elements of social support as part of their QoL definitions. Elements of social support are enhancing factors to QoL because patients often cited these relationships as providing methods of accountability, engagement, emotional support, and caregiving.
Participants cited disease symptoms and functional limitations as detracting from QoL. This was included in 98% of participant definitions, illustrating the necessity to continue to evaluate symptom impact and functionality as important monitoring factors within T2D. Individuals cited detracting factors such as diabetes burden; awareness of mortality; pain; disease progression; functional, psychological, and emotional impact; and side effects and comorbidities. All factors within this theme detracted from QoL because experiencing symptoms and functional limitations was shown to impede independence and ability.
Participants cited parent-study provided education and information about diabetes as an enhancing factor of QoL (eg, print materials, surveys, check-ins, and intervention calls). Overall, 92% of participants included elements of diabetes-related education and information within QoL definitions. Education and information serve as an enhancing factor because individuals reported benefit from receiving information through study calls, print materials, and routine health care visits while also citing the motivation to pursue their own research, relationships, and support with other individuals that have T2D.
Participants cited economic conditions, employment, and health-care-related costs as detracting influences on QoL. This was included in 56% of definitions. Specific detracting factors included expressing necessity for increased income to maintain behavioral health changes, strain due to health care costs, and occupation-related barriers. These factors illustrate financial challenges incurred by T2D disease experience and detract from QoL because individuals are faced with budgetary strain that may limit their ability to maintain dietary or lifestyle changes and complete health-care-related payments. Some individuals noted inability to work due to progression of T2D, therefore experiencing financial strain due to employment and income changes.
Participants described acceptance of changes made to life with diabetes as a factor enhancing QoL, specifically referring to adaptation in reference to health behavior changes (and the benefits to follow) related to diabetes selfmanagement. Elements of adaptation were included in 98% of participant QoL definitions, demonstrating the importance of further exploring this topic. These elements are enhancing factors of QoL in T2D because they illustrate acceptance, implementation of behavioral health change, and the ability to adjust, maintain, and control new lifestyle changes.
Participants cited sociocultural elements as part of their QoL definition, often using cultural connections, religion, and faith as a method of coping with illness, serving as an enhancing element to QoL. This was referenced by 60% of participants, highlighting the importance of understanding the diabetes-related sociocultural implications of QoL. Individuals cited cultural and religious connections as elements of coping (eg, family traditions, faith, food, prayer), often seeking to maintain elements of comfort and integrate them into their new lifestyles.
Results indicate that adults aiming to improve T2D self-management have a multifaceted definition of QoL that includes physical health, independence, social support, and engaging in cultural and lifestyle habits. These findings revealed aspects of QoL not currently captured by available T2D-related QoL measures: sociocultural impact and diabetes education. As illustrated by the resulting themes, patients defined QoL with overlapping terms, using a combination of all themes (see Table 2). The implications of the variation in patient definitions speaks to the unique and multifaceted nature of QoL—although patterns are evident as shown by early thematic saturation, QoL as a concept is truly multidimensional. By adding aspects of sociocultural impact, lifestyle, culturally related traditions, and social-support-related education into QoL evaluations, the accuracy of QoL measurement in T2D may be improved and may better capture the effects of intensive treatment and self-management support interventions.
The consistent reports of importance of social support, accountability, and education through the NYC Care Calls program illustrate the impact of education and accountability throughout the self-management process. Patients consistently reported that they felt supported and cared for through the experience of completing the educational calls in the intervention, and similar sentiments were echoed by individuals in the control group regarding calls for study surveys. Results also describe effects on QoL that may not be typically captured by outcome measures of trials aiming to improve self-management in diverse populations with T2D, including the participants in this study.37 These findings may inform the selection of measures that better capture potential effects on QoL in populations excluded from opportunity and resources.
In this qualitative evaluation, results showed consistent themes depicting both enhancing and detracting factors. Some of the reported themes were consistent with current areas of measurement in quantitative QoL measures, whereas others (ie, diabetes education, financial changes, sociocultural implications) appear to be novel or underutilized and highlight room for improvement in measure development (see Table 3). Regardless of measurement in previously established QoL-related scales, the results of this study and comparison to current measures highlight the shortcomings for intervention research within people of color living in neighborhoods that are disproportionately impacted by diabetes and systematic inequities.
Within this patient population specifically, sociocultural aspects were particularly salient and stood out as a thematic trend that had not previously been accounted for in QoL assessment. Ethnicity, race, socioeconomic status (SES), gender, and age have previously been linked to risk for development of diabetes and treatment outcomes in those living with diabetes.40,41 Given the risk factors associated with people of color,24 it would be beneficial to include sociocultural context in the evaluation of QoL. Ethnic minorities and lower SES individuals report decreased functional status and higher emotional distress compared to White individuals and those of higher SES.31,42-44 Including all facets of the patient’s culture, SES, and resources is a necessary step to improve clinical care and shared decision making for this population.45 As such, the evaluation of QoL must account for variation in patient preferences and expectations.25 These efforts should be at the core of clinical evaluation, engaging the patient in responsibility and education and making the incorporation of disease management natural and the experience of care minimally disruptive.46,47
The results of this study support the natural combination of biological, physiological, and environmental factors in a patient’s QoL definition. This was previously proposed by Wilson and Cleary48 in 1995 to demonstrate that QoL is influenced by micro and macro factors—the individual’s QoL experience is shaped by their own values, preferences, and experiences, just as it is shaped by the influences of their surroundings, including both healthrelated and non-health-related factors (eg, SES, support, employment).31,48 As such, individuals possess their own judgement about their functionality and QoL, making the relationship between health status and QoL a challenge to understand.32 Previous studies have illustrated that physical health is cited as a factor of QoL less frequently among older adults with higher levels of health-related functional impairment compared to the general population.49,50 Additionally, previous studies have shown that there is more focus placed on physical functioning when reporting health status and more focus of emotional functioning when reporting QoL.51
Results indicated overlap in reporting some elements of health status as important facets of QoL. Glycemic control was discussed in terms that suggest it was an important goal of improved QoL and is often evaluated as a primary outcome and predictor of QoL; however, efforts to prevent diabetes complications and intensive treatment regimens often overshadow the impact that current treatment has on QoL.29 Here, there is an opportunity to focus on lifestyle and variables of patient importance in addition to medically based outcomes.52 Self-management and health behavior changes recommended to achieve better glycemic control were described as enhancing QoL, with some caveats around diet, cultural values, and financial pressures. This highlights the importance of evaluating QoL using both medically oriented and lifestyle-focused avenues, creating the opportunity for a holistic evaluation. This distinction underscores the importance of measuring QoL in an all-encompassing way—including elements of physical functioning to accommodate for the physicality of disease impact while also covering aspects of emotional, individual, cultural, and environmental factors as experienced by patients.
Although this sample is not representative of all individuals with T2D, it highlights important discrepancies and inequities in access to care and health-related outcomes. Despite the limitations of this sample, this evaluation may still be applicable across racial, ethnic, and socioeconomic groups. Many of the sociocultural trends reported are likely salient to individuals from ethnic minority backgrounds. Future comparative evaluation of themes in Spanish and English interviews would be beneficial to further explore sociocultural impact. Additionally, the individuals that participated in this substudy had already completed a self-management intervention that may have impacted their selfmanagement routine—individuals that opted out of participation or discontinued enrollment may have had different views. Therefore, it is likely that individuals associated their self-management experience with QoL and were prompted to immediately reflect on this topic rather than others. It would be beneficial to gather QoL definitions at baseline or from individuals that have not already completed an intervention because this may provide a more organic and candid patient-reported definition. Further evaluation to affirm these findings are consistent in different populations is necessary. Results will inform the development of a new, patient-centered measure of T2D-related QoL.
The authors declare that there is no conflict of interest.
This work was supported by Grant R18 DK098742 from the National Institutes of Health. This work was also in part supported by the Einstein–Mount Sinai Diabetes Research Center (P30 DK020541) and the New York Regional Center for Diabetes Translation Research (P30 DK111022). Additionally, JSG is supported by grants R01 DK104845, R01 DK121298, and R01 DK121896 from the National Institutes of Health. CJH is supported by the Drs. David and Jane Willner Bloomgarden Family Fellowship Fund.
Sarah R. Fishman https://orcid.org/0000-0002-9455-9600
Jeffrey S. Gonzalez https://orcid.org/0000-0002-8252-2077
From Ferkauf Graduate School of Psychology, Yeshiva University, Bronx, New York (Dr Fishman, Dr Jonas, Dr Gittleman, Dr Crespo-Ramos, Dr Hoogendoorn, Dr Gonzalez); New York City Department of Health & Mental Hygiene, Queens, New York (Ms Fernandez Galvis, Ms Linnell, Ms Iribarren, Dr Pham-Singer, Dr Wu); Stanford University School of Medicine, Stanford, California (Dr Tanenbaum); New York Academy of Medicine, New York, New York (Ms Scherer, Dr Weiss); Department of Medicine, Albert Einstein College of Medicine, Bronx, New York (Dr Walker, Dr Hoogendoorn, Dr Gonzalez); and Miller School of Medicine, University of Miami Health System, Miami, Florida.
Corresponding Author:Jeffrey S. Gonzalez, Ferkauf Graduate School of Psychology, 1165 Morris Park Avenue, Bronx, NY 10461, USA.Email: jeffrey.gonzalez@yu.edu
