What is diabetes distress and how can we measure it? A narrative review and conceptual model

Abstract

Background

Diabetes distress is the negative emotional impact of living with diabetes. It has tangible clinical importance, being associated with sub-optimal self-care and glycemic control. Diabetes distress has been operationalized in various ways and several measures exist. Measurement clarity is needed for both scientific and clinical reasons.

Objectives

To clarify the conceptualization and operationalization of diabetes distress, identify and distinguish relevant measures, and evaluate their appropriateness for this purpose.

Results

Six measures were identified: Problem Areas in Diabetes (PAID) scale, Diabetes Distress Scale (DDS); Type I Diabetes Distress Scale (T1-DDS), Diabetes-specific Quality of Life Scale—Revised (DSQoLs-R) ‘Burden and Restrictions—Daily Hassles’ sub-scale, Well-being Questionnaire 28 (W-BQ 28) ‘Diabetes Well-being’ sub-scale, and Illness Perceptions Questionnaire—Revised (IPQ-R) ‘Emotional Representations’ sub-scale. Across these measures a broad spectrum of diabetes distress is captured, including distress associated with treatment regimen, food/eating, future/complications, hypoglycemia, social/interpersonal relationships, and healthcare professionals. No single measure appears fully comprehensive. Limited detail of the qualitative work informing scale design is reported, raising concerns about content validity.

Conclusions

Across the available measures diabetes distress is seemingly comprehensively assessed and measures should be considered in terms of their focus and scope to ensure the foci of interventions are appropriately targeted.

Introduction

Living with diabetes is complex and involves various self-care activities, e.g., medication taking, healthy eating, carbohydrate counting, physical activity, checking blood glucose, and problem solving.¹ These self-care behaviors are required to keep glycated hemoglobin (HbA1c) in target range, to prevent/delay onset of devastating complications.² The burden of self-management, living with diabetes-related complications (or the risk of their development), and managing difficult social situations, has the potential to cause considerable emotional distress. In the mid-1990s, the emotional impact of living with diabetes was brought to the fore, with introduction of the concept of ‘diabetes distress’ (DD).³ DD emerged from research on stress and coping, and emotional regulation in response to specific acute or chronic stressors, which suggests that emotions can emerge from specific situational contexts and that emotional distress is a response to perceptions of health threats balanced against an appraisal of available coping resources.⁴

In recent years, research into DD has gained significant traction. Around one quarter of UK adults with diabetes experience elevated, or severe, DD at any given time.5, 6 Similar rates are reported elsewhere in Europe,⁷ Australia⁸ and the USA.⁹ Almost 50% of people experience elevated DD over an 18-month period.⁹ It should be noted, though not the focus of this review, that DD is additionally well-documented among partners of those with diabetes,¹⁰ children and adolescents with diabetes¹¹ and parents of children and adolescents with diabetes.¹² Indeed, measures of DD specific to the needs of these populations have been developed.10, 13, 14

DD is positively associated with HbA1c, such that fluctuations in each correspond over time,15, 16, 17, 18 and reductions in DD are accompanied by clinically significant improvements in HbA1c.19, 20 DD also impacts upon certain self-management behaviors.15, 17These relationships have primarily been associative, however, and hence do not indicate causality. Individuals with high DD are less likely to participate in educational and self-management interventions,²¹ and exhibit less improvement in HbA1c following such interventions.²² Conversely, when interventions target DD, individuals with elevated DD engage to a greater extent and this results in improved DD, self-management and HbA1c.²³ This emerging evidence has prompted calls for further interventions to target DD.¹⁷

From a clinical perspective, measurement clarity is crucial to ensure appropriate identification of need and tailoring of care. From a scientific perspective, it is necessary to ensure valid operationalization of constructs, maximum responsiveness of measures to enable demonstration of effective interventions, and appropriate interpretation of data.²⁴ US and European regulatory bodies have released guidance on the development and use of patient-reported outcome measures (PROMs), describing the scientific rigor with which such measures must be developed to enable meaningful measurement and outcomes evaluation.24, 25 First among the issues discussed is content validity, i.e., the extent to which a questionnaire measures what is claimed. A ‘conceptual model’ provides a representation of the relevant concepts that comprise the construct, and the relationships among the concepts. It should be developed, following a systematic literature review and qualitative work with patients and health professionals, to inform the structure and content (items) of a new questionnaire.

In the academic literature, the concept of DD has been assumed to be relatively simple. However, definition and measurement have been circular. DD is defined largely with reference to the issues measured by the Problem Areas in Diabetes (PAID) scale,³ which is widely regarded as the first PROM to assess DD in adults. Thus, the PAID is widely considered a suitable measure of DD. The more recently developed Diabetes Distress Scale (DDS)²⁶ is also gaining traction as a measure of DD in adults. Recent research has suggested that there are important content differences between the PAID and the DDS.²⁷ Until the emergence of the DDS, there had been little discussion about what comprises DD or of the rationale for questionnaire selection. Indeed, it is unclear whether other questionnaires might also be suitable for assessing DD.

Furthermore, while many clinicians/researchers refer to the PAID and DDS (seemingly) appropriately as measures of DD, others have used broader terminology, such as (diabetes-specific) quality of life²⁸ and diabetes-dependent impairment.²⁹ Similarly, measures assessing other constructs (e.g., the ATT-39, which assesses diabetes attitudes and beliefs) have been reported as measures of DD.30, 31 Beyond operationalizing DD with the PAID or the DDS, a common understanding of how to conceptualize DD and differentiate it from other commonly assessed constructs has not yet emerged in the literature.

Previous reviews have disentangled the conceptualization and measurement of other diabetes-specific PROMs.32, 33, 34 Recently, researchers have also sought to clarify the conceptual distinction between DD and depression.4, 16, 17, 31 To date there has been no attempt to derive a conceptual model of DD, identify and distinguish measures of DD from the vast array of other diabetes-specific PROMs, and explore their validity for this purpose.

Thus, our overall aims were to: a) conceptualize and operationalize DD; b) identify measures of the broad concept of DD by examining their face validity in terms of measuring DD (i.e., the extent to which a measure looks as though it measures DD); and c) review the content validity of the identified measures in terms of assessing DD (i.e., the aspects of DD covered and the extent to which it is likely that each measure captures DD comprehensively) with a view to offering guidance on the context-specific selection of measures.