key: cord-0811155-qsrnvw00
authors: Im, James; Escudero, Carlos; Zhang, Kendra; Choi, Dorothy; Sivakumar, Arani; Booth, Gillian L.; Sale, Joanna; Pritlove, Cheryl; Advani, Andrew; Yu, Catherine H.
title: Perceptions and correlates of distress due to the COVID-19 pandemic and stress management strategies among people with diabetes: a mixed methods study
date: 2021-10-22
journal: Can J Diabetes
DOI: 10.1016/j.jcjd.2021.10.006
sha: 2825b638c55ff5cb69c8d2d82c7d7bc04b86f2b2
doc_id: 811155
cord_uid: qsrnvw00

Introduction Greater risk of adverse health outcomes and public health measures have increased distress among people with diabetes during the COVID-19 pandemic. The objectives of this study were to explore how the experiences of people with diabetes during the COVID-19 pandemic differ according to sociodemographic characteristics and identify diabetes-related psychosocial correlates of COVID distress. Methods Patients with type 1 or 2 diabetes were recruited from clinics and community health centres in Toronto, Ontario as well as patient networks. Participants were interviewed to explore the experiences of people with diabetes with varied sociodemographic and clinical identities, with respect to wellness (physical, emotional, social, financial, occupational), level of stress, and management strategies. Multiple linear regression was used to assess the relationships between diabetes distress, diabetes self-efficacy, and resilient coping with COVID distress. Results Interviews revealed that specific aspects of psychosocial wellness affected by the pandemic, and stress and illness management strategies utilized by people with diabetes differed based on socioeconomic status, sex, type of diabetes, and race. Resilient coping (β = -0.0517; 95% CI: -0.0918,-0.0116; P-value = 0.012), diabetes distress (β = 0.0260; 95% CI: 0.0149,0.0371; P-value < 0.0001), and diabetes self-efficacy (β = -0.0184; 95% CI: -0.0316,-0.0052; P-value = 0.007) were significantly associated with COVID distress. Conclusions Certain subgroups of people with diabetes have experienced a disproportionate amount of COVID distress. Assessing correlates of COVID distress among people with diabetes will help inform interventions such as diabetes self-management education to address the psychosocial distress caused by the pandemic.

The novel coronavirus disease 2019 (COVID- 19) pandemic represents one of the greatest contemporary public health challenges [1] . People with diabetes (PwD) are at greater risk of medical complications and fatality, and have endured strict quarantine measures to minimize their risk [2] . PwD may also have endured worsening glycemic control during lockdown due to reduced capacity to exercise and control their diet, and reduced availability of anti-diabetic medications and medical advice [3] . Despite the many possible challenges of managing diabetes during the pandemic, the lived experiences of PwD and how they have been impacted psychosocially have not been adequately explored.

Additionally, it has been shown that the COVID-19 pandemic negatively impacts certain populations more so than others, depending on factors including gender, race, and socioeconomic status (SES) [4, 5] .

Among PwD, these sociodemographic variables may pose further challenges to coping with the pandemic. Thus, research on the impacts of the pandemic on PwD should consider the experiences of underserved populations so that healthcare providers can address disparities.

Given the high prevalence of psychological distress among PwD during the pandemic [6] , investigating ways to alleviate COVID distress would be beneficial. General self-efficacy has been linked to lower levels of distress and may be a correlate of reduced COVID distress [7] . Another factor potentially associated with COVID distress is diabetes distress, which is associated with negative health outcomes such as reduced adherence to medication regimens [8] . Resilient coping is associated with a greater quality of life and metabolic control in patients with diabetes [9] . Despite their potential implications for psychosocial wellbeing, there is a lack of research into how these factors are associated with COVID distress.

This study aimed to explore the psychosocial experiences of PwD (type 1 and type 2) of different sociodemographic and clinical groups, and the associations between modifiable psychosocial constructs J o u r n a l P r e -p r o o f and COVID distress to inform clinicians of targets for minimizing distress among their patients with diabetes.

This study was a mixed-methods study, involving qualitative interviews and an online cross-sectional survey. Given the disparities in distress experienced by subgroups in our interview sample, we examined, through the cross-sectional survey, certain diabetes-related psychosocial constructs which may be associated with COVID distress.

Community-dwelling adults aged 18 years and older with either type 1 or type 2 diabetes living in Ontario, Canada and those part of patient networks in Canada were included in this study. Pregnant women, adults in long-term care, at end of life or who could not give consent were excluded from this study since their experiences were presumed to be different from PwD in the general population [10] .

Recruitment of participants for the interviews and cross-sectional survey occurred from May 2020 to September 2020 and May 2020 to February 2021, respectively. Recruitment occurred via individuals from participants' circles of care from St. Michael's Hospital specialty clinics, primary care clinics and community health centers in Toronto, Ontario. Professional, research and patient networks such as Diabetes Canada and KT Canada Network were also used to recruit patients. Purposive sampling was used to ensure more demographic diversity among participants. Purposive sampling was used to deliberately select participants for recruitment to ensure demographic diversity and capture underrepresented populations [11] . Participants who were interviewed were also invited to participate in the survey.

J o u r n a l P r e -p r o o f All participants provided verbal informed consent to participate in the study. This study was approved by the Research Ethics Board of St. Michael's Hospital in Toronto, Ontario.

Interviews were conducted, audiotaped, then transcribed verbatim and annotated using field notes for subsequent analysis. Data were collected via 45-60-minute semi-structured telephone interviews with open-ended questions to explore participant experiences [12] . Interview questions were informed by the Wellness Evaluation of Lifestyle Inventory, developed from the wheel of wellness theoretical framework, the gold standard of wellness assessments in clinical settings [13, 14] , and Stress Coping theory, which comprehensively explores psychological and emotional responsiveness and coping with multiple types of stressors [15] . Interview questions were pilot tested with knowledge users. Participants with significant language barriers were interviewed alongside a caregiver.

Data collection and analysis were conducted concurrently until saturation was attained [16, 17, 18] . The interview guide was iteratively refined throughout data collection to capture emerging ideas. Transcripts were coded to develop concise summaries of key themes within and across interviews [19] . Coding was conducted independently by four individuals (J.I., C.E.K., D.C., A.S.), and codes were refined and organized according to Shaw's framework for coping, illness behaviour and outcomes [20] , to provide a greater focus on the impact of the pandemic on PwD. In particular, we focused on participants' appraisal of the situation, coping strategies and health behaviours. Transcripts were critically analyzed using inductive thematic analysis and constant comparative analysis [21] . As an initial exploration, a categorical comparison with SES, type of diabetes, gender and race was used to compare subgroups and identify characteristics within which there appeared to be inequities in COVID-19 burden. NVivo software (v.12) was used to manage data.

Validated psychometric scales were used to measure diabetes distress [19] , resilient coping [20] , diabetes self-efficacy [21] , and COVID distress [22] (Table 1 ). These scales were selected because they demonstrated good internal validity and reliability [19, 20, 21, 22] . The Brief Resilient Coping Scale (BRCS), Problem Areas in Diabetes (PAID), Impact of Event Scale-6 (IES-6), and Self-Efficacy for Managing Chronic Diseases (SEMCD) scales were treated as continuous variables. Validated psychometric scales were used to measure diabetes distress, resilient coping, diabetes self-efficacy and COVID distress ( Table 1 ). The Problem Areas in Diabetes (PAID), a 20-question scale with an internal consistency of 0.92, was used to measure diabetes distress. Each question is scored on a 5-point Likert scale [22] . Resilient coping was measured using the Brief Resilient Coping Scale (BRCS), a 4-question scale with an internal consistency of 0.69. Similarly, questions on the BRCS are also scored using a 5point Likert format [23] . Diabetes self-efficacy was measured using the Self-Efficacy for Managing Chronic Diseases (SEMCD) scale, which has an internal consistency of 0.80 and contains 6 questions measured on a 10-point Likert scale [24] . Lastly, the Impact of Event Scale-6 (IES-6), which assesses post-traumatic stress reaction, was used as a measure of COVID distress. The IES-6 has an internal consistency of 0.93, contains 6 questions, and is measured on a 5-point Likert scale [25] . All scales were treated as continuous variables.

All outcome and covariate data were self-reported by participants. Surveys were conducted over the telephone for participants with low literacy skills and offered in multiple languages for participants who were unable to speak English, although all were conducted in English.

Multiple linear regression was used to assess the associations between psychosocial constructs (diabetes distress, resilient coping, diabetes self-efficacy) and COVID distress, adjusting for age, gender, type of diabetes, race, occupation, and diabetes duration. For all models, β coefficients and associated 95% confidence intervals were reported. Diabetes distress and resilient coping were assessed as potential mediators by including them as covariates for adjustment in the regression model with diabetes selfefficacy as the primary predictor and COVID distress as the outcome [26] . A two-sided P-value < 0.05 was used as the cut-off for statistical significance.

The size of the sample was based on guidelines for regression analyses by Harrell of 10 observations per coefficient [27] . Complete case analysis was performed because the proportion of missing data in included covariates was very small and there was little evidence to suggest the data were not missing at random [28] . Furthermore, all regression analyses included approximately 140 observations and thus were considered to be sufficiently powered. All statistical analyses were conducted using R, version 4.0.2 (R Foundation for Statistical Computing; Vienna, Austria).

Characteristics of the 47 included participants are shown in Table 2 .

Participants' clinical and sociodemographic characteristics seemed to impact their perceptions of the pandemic and stress management strategies. Accordingly, we identified contrasting experiences of: (1) white versus non-white participants; (2) participants of high SES versus low SES; (3) participants with type 1 versus type 2 diabetes; and (4) women versus men, as detailed below. The perceptions of participants with type 1 and type 2 diabetes also seemed to impact the coping strategies employed to manage stressors. Participants with type 1 diabetes seemed to engage in more active problem-focused behaviors to manage stressors and reduce distress: On the other hand, participants with type 2 diabetes reported greater distress from having reduced contact with their support network and made more attempts to interact with their support network to manage stress:

"I made sure that I was contacting people either through Zoom or by phone and I was talking to several people every week. (P47)"

Participants' perceptions of the pandemic and management methods influenced the kinds of healthcare interventions they thought were necessary. Participants with type 1 diabetes indicated that messaging from clinics could supplement current methods to help ascertain other resources:

"The Ontario government has released a lot of stuff but… I don't typically consume that kind of stuff… I think an email or something like that could definitely be really helpful from each of the doctors and the government. (P2)"

On the other hand, because participants with type 2 diabetes were generally more fearful and anxious, many indicated that increasing healthcare resources to accommodate for all participants was one improvement the healthcare system should adopt: "I think it's a concern from everybody, you know? You start wondering, will you ever shake somebody's hand again when you first meet them? (P1)"

The approach to managing stressors appeared to have differed between non-white and white participants.

Many non-white participants reported relying on their support networks to manage stress. For example, when asked about her normal approach to stress, a participant responded:

"I usually talk with my husband. He's my friend and he usually listens to me a lot. That helps me.

Faith and spirituality also appeared to be much more important and prevalent aspects of non-white participants' lives in comparison to reports by white participants. Many non-white participants noted that faith and spirituality helped them manage stress and also provided them with a strong community of support during the pandemic:

On the other hand, white participants typically reported employing more task-based stress management methods. In response to changing schedules, many white participants reported adapting their routines by engaging in more health-related behaviors such as improving their diets, which facilitated the management of their diabetes:

As a whole, perceived instability regarding the future and relationships was a major driver of anxiety and distress, especially among white participants. On the other hand, participants of high SES perceived the pandemic as less negative and something that facilitated the organization of their daily lives. In particular, participants of high SES seemed to be able to work from home more frequently which they perceived as advantageous for their diabetes management: The experiences of women during the pandemic contrasted starkly with those of men. Women reported having much greater burdens during the pandemic than men. In particular, the strain of having to adopt multiple roles within the family was a major driver of distress:

"I feel as a female, because they are more a foundation, they are not just taking care of themselves, they have other people to take care of. I think for females, it is a burden even more.

In addition, another major source of distress among many women was the strain resulting from a mixing of their home and work lives: The stress management strategies of women and men also differed markedly. Although women reported greater distress and burden, women also appeared to have more self-directedness towards their own Additionally, this self-directedness also seemed to enable the utilization of problem-focused management strategies which facilitated greater improvements in their health:

"I was doing some research, reading some books regarding healthy eating so I did achieve a little bit of a lifestyle of homemade food. Now when I go to work… I make sure that I eat it on time. (P13)"

Men appeared to manage stress primarily through methods focused on avoidance. One man found that avoidance was the optimal solution during the pandemic, and this sentiment was echoed by several others: Women reported having greater distress and burdens during the pandemic but seemed to have a sense of self-directedness which facilitated usage of problem-focused management strategies.

Characteristics of the included participants are shown in Table 2 . The distribution of men (47.1%) and women (51.6%) in the sample was approximately equal. Participants aged 65 years and older comprised 39.2% of the sample. The majority of participants were white (65.4%) and had type 2 diabetes (62.7%).

As shown in Table 3 , higher levels of diabetes distress (β = 0.0260; 95% CI: 0.0149,0.0371; P-value < 0.0001) were significantly associated with higher COVID distress while higher levels of resilient coping (β = -0.0517; 95% CI: -0.0918,-0.0116; P-value = 0.012) and diabetes self-efficacy (β = -0.0184; 95% CI: -0.0316,-0.0052; P-value = 0.007) were associated with lower levels of COVID distress, adjusted for all covariates.

This study examined the impacts of the pandemic on diverse PwD and psychosocial factors associated with COVID distress. Our results revealed how sociodemographic characteristics shaped PwD's experiences of the pandemic, and potential avenues for mitigating the negative impacts of the pandemic.

From the qualitative component of our study, we identified several differences in participants' experiences of the pandemic. Firstly, our results demonstrate that anxiety was lower among people with type 1 diabetes; distress may have been heightened in patients with type 2 diabetes due to additional comorbidities, including obesity and hypertension [29] . These findings indicate the need to address the psychosocial wellbeing of PwD uniquely depending on their type of diabetes.

Secondly, we found that the perceptions and coping strategies of white versus non-white PwD differed.

White participants expressed both a more general sense of anxiety and distress as well as distress specifically regarding uncertainties about the future due to the pandemic. In contrast, cross-sectional studies by Özcan et al. and Stoop et al. found that diabetes distress was more prevalent among ethnic minorities [30, 31] ; however, cross-sectional studies by Peyrot et al. and Pouwer et al. found that ethnic minorities had better quality of life and wellbeing [32, 33] . These studies, along with our findings, suggest that white and non-white patients experience different sources of distress and also respond to them differently, which influences their overall levels of distress. These differences in sources of distress and coping strategies may be due to unmeasured factors related to ethnicity, such as migration stress, language barriers, and cultural values [30] .

Thirdly, our participants of lower SES were unable to use the same glycemic control strategies as those of higher SES. Patients of lower SES also reported greater anxious and distressed due to a loss of daily routine and control over their life circumstances and had greater difficulties managing stressors in our study. A cross-sectional study by Fidan et al. found that low education levels were associated with difficulties in coping, which may be due to a lack of resources to buffer the impacts of stressful events and cope adaptively [31] . These findings demonstrate that SES can either be an enabler or a barrier to health and wellbeing, and should therefore be considered when devising management plans for PwD. A cross-sectional study by Fidan et al. found that low education levels were associated with difficulties in coping, which may be due to a lack of resources to buffer the impacts of stressful events [34] . Similarly, our results showed that PwD of higher SES were better able to cope with their distress by finding ways to maintain activities disrupted by the pandemic such as exercising and eating well, which in turn translated into greater mental wellbeing compared to PwD of lower SES.

Fourthly, we found that women with diabetes expressed more anxiety and distress than did men. Women with diabetes may generally experience more anxiety due to an interplay between diabetes-related distress and additional responsibilities related to work and caring for families [32] . With our finding being consistent with several studies involving women from the general population [33, 34, 35] , diabetes may pose yet further challenges for women who are PwD. This finding aligns with several studies which found that women in general have experienced higher levels of psychological distress during the pandemic and are at increased risk for mental health problems [35, 36, 37] . Women with diabetes may experience more anxiety due to an interplay between diabetes-related responsibility and additional responsibilities related to employment and caring for families [38] .

From the quantitative component of our study, we determined that higher levels of resilient coping, diabetes self-efficacy, and lower levels of diabetes distress were associated with lower levels of COVID distress. Thus, utilizing resilient coping behaviours, such as setting behavioural goals, fulfilling healthcare obligations, and increasing motivation, would help reduce PwD's COVID distress [39] . Self-efficacy has been shown to be relevant in adequately managing chronic health conditions, as it heavily influences the pursuit of health behaviours [40] . Diabetes distress may have increased in PwD due to pandemic-induced disruptions in self-management routines (e.g., diet and physical activity) [41, 42] . This was reflected in our interviews, particularly among participants with type 2 diabetes, who expressed worry about their diabetes status as well as more fear and anxiety surrounding the pandemic. Thus, diabetes selfmanagement education and psychological interventions such as cognitive behavioural therapy, which have shown success in reducing diabetes distress [43, 44] , may be useful for PwD for reducing COVID distress as well.

Our study had limitations. Firstly, participants were not completely balanced in terms of sociodemographic characteristics. A higher proportion of participants were from higher SES and thus, J o u r n a l P r e -p r o o f their perspectives may not be representative of all people with diabetes. Secondly, data were collected across different periods of public health restrictions, which may have led to differences in perceptions, experiences, and levels of distress. In the quantitative component, causality cannot be inferred from the relationships examined given the cross-sectional nature of this study. In addition, these data may not be generalizable to the general population of PwD in Ontario. Older adults, women, and those of high SES were over-represented in our sample. In the qualitative component, we were unable to assess whether differences in observed themes were attributable to other, or intersecting, patient characteristics due to sample size limitations.

Nevertheless, our study also had strengths. It is one of the first to explore the multifaceted impact of COVID-19 on PwD and strategies used to manage stressors. It is also among the first to evaluate possible determinants of COVID distress among PwD. In addition, methodological rigour was promoted as coding was performed by four independent coders and qualitative data were interpreted and analyzed by five individuals. Feedback from a team with expertise in qualitative data analysis and diabetes care was incorporated at multiple stages in the study [45] . Moreover, we used validated and reliable psychometric scales to assess all constructs in this study.

Our results suggest that the impacts of the pandemic have varied across sociodemographic and clinical groups, and that, clinicians and educators can target resilient coping, diabetes self-efficacy, and diabetes distress to minimize COVID distress. This study therefore highlights the need to evaluate and contextualize the psychosocial wellbeing of PwD at routine check-ups. 

Evolution of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) as coronavirus disease 2019 (COVID-19) pandemic: A global health emergency

Understanding novel COVID-19: its impact on organ failure and risk assessment for diabetic and cancer patients

Social isolation during Covid-19: Boon or bane to diabetes management

COVID-19 and Black, Asian, and Minority Ethnic Communities: A Complex Relationship Without Just Cause

Gender inequality during the COVID-19 pandemic: Income, expenditure, savings, and job loss

Mental health in the era of COVID-19: prevalence of psychiatric disorders in a cohort of patients with type 1 and type 2 diabetes during the social distancing

Coping Self-Efficacy as a Mediator of Distress Following a Natural Disaster

The Importance of Addressing Depression and Diabetes Distress in Adults with Type 2 Diabetes

Healthy coping, negative emotions, and diabetes management: a systematic review and appraisal

The risk of clinical complications and death among pregnant women with COVID-19 in the Cerner COVID-19 cohort: a retrospective analysis

Purposeful Sampling for Qualitative Data Collection and Analysis in Mixed Method Implementation Research

Qualitative research method-interviewing and observation

The Assessment and Measurement of Wellness in the Clinical Medical Setting: A Systematic Review

The Wheel of Wellness Counseling for Wellness: A Holistic Model for Treatment Planning

The Handbook of Stress and Health: A Guide to Research and Practice

Saturation in qualitative research: exploring its conceptualization and operationalization

Determining Sample Size

Characterising and justifying sample size sufficiency in interviewbased studies: systematic analysis of qualitative health research over a 15-year period

Three approaches to qualitative content analysis

A framework for the study of coping, illness behaviour and outcomes

A Purposeful Approach to the Constant Comparative Method in the Analysis of Qualitative Interviews

The development and psychometric evaluation of the Brief Resilient Coping Scale

Evaluating self-efficacy for managing chronic disease: psychometric properties of the six-item Self-Efficacy Scale in Germany

Brief measure of posttraumatic stress reactions: impact of Event Scale-6

The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations

Regression modeling strategies: with applications to linear models, logistic and ordinal regression, and survival analysis

What is the difference between missing completely at random and missing at random?

Health-related quality of life in type 1 and type 2 diabetic patients in a Portuguese central public hospital

High Diabetes Distress Among Ethnic Minorities Is Not Explained by Metabolic, Cardiovascular, or Lifestyle Factors: Findings From the Dutch Diabetes Pearl Cohort

Diabetes-specific emotional distress in people with Type 2 diabetes: a comparison between primary and secondary care

Ethnic differences in psychological outcomes among people with diabetes: USA results from the second Diabetes Attitudes, Wishes, and Needs (DAWN2) study

Ethnic aspects of emotional distress in patients with diabetes--the Amsterdam Health Monitor Study

Patients With Type 2 Diabetes Mellitus: Obstacles in Coping

Mindfulness, age and gender as protective factors against psychological distress during COVID-19 pandemic

Risk factors for psychological distress during the COVID-19 pandemic in Israel: Loneliness, age, gender, and health status play an important role

Emotional distress in young adults during the COVID-19 pandemic: evidence of risk and resilience from a longitudinal cohort study

Gender Differences in Living with Diabetes Mellitus

Psychological adaptive difficulties and their management during COVID-19 pandemic in people with diabetes mellitus

Self-management education: history, definition, outcomes, and mechanisms

Diabetes and COVID-19: Risks, Management, and Learnings From Other National Disasters

Healthy coping: issues and implications in diabetes education and care

Diabetes distress: understanding the hidden struggles of living with diabetes and exploring intervention strategies

The detection and management of diabetes distress in people with type 1 diabetes

Immersion/crystallisation

The authors thank our knowledge users Pauline Wijeyesekera and Heather Whetstone; and Ariel Yeheskel for interview transcription.