Characteristics of the eligible sample (respondents, n = 384), and those who attempted the diabetes-specific QoL questionnaires but did not provide qualitative feedback (non-respondents, n = 1194), are shown in Table 2.

Table 2 Sample characteristics of the final eligible sample and non-responders

Most respondents attempted all five diabetes-specific QoL measures of interest (n = 344, 90%) and provided qualitative feedback on one questionnaire (n = 220, 57%), most commonly for the DSQoLS (see Table 3). Across the five measures, a total of 711 qualitative responses were provided. A minority of participant feedback included general comments (e.g. all five questionnaires were perceived by some as “good”, “fine”, “liked”, or described as “enjoyable” and “interesting”) with no further detail. Specific feedback was organised within four main themes: (1) clarity and ease of completion; (2) relevance and comprehensiveness; (3) length and repetition, and (4) preferences and impact of questionnaire wording and tone. Overall themes and sub-themes are described below, and quotes relevant to the five questionnaires are shown in Table 3.

Table 3 Themes, sub-themes and proportion of participants attempting and providing feedback for each questionnaireQuestionnaire clarity and ease of completion

In addition to general endorsements that questionnaires are “simple”, “easy”, and “straightforward”, participants provided specific feedback relating to the clarity of instructions and questionnaire wording; suitability of response options, and difficulty isolating the impact of diabetes on QoL.

Instructions and question wording

Differing views were offered about whether questionnaire instructions or questions were clear and easy to understand, as well as preferred questionnaire attributes. For instance, some participants praised the inclusion of examples in the Diabetes QOL-Q (‘I can go out or socialise as I would like, e.g., cinema, concerts…’), suggesting that it helped to clarify what is being asked, while another participant reported the inclusion of such examples as “patronising”.

Consistently, respondents indicated that the if/then wording used in the ADDQoL-19 was “confusing” and questions containing double negatives (i.e. item 17: “If I did not have diabetes, I would have to depend on others when I do not want to”) were “illogical” and “incomprehensible”. They also indicated that the broad question wording that encompassed diverse situations made it difficult to answer the questions. For example, item 3 of the DIDP incorporates three separate concepts (i.e. ‘relationship with your family, friends and peers’) and item 8 of the ADDQoL-19 combines current experience of or wish for a ‘close personal relationship’ within a single question. Participants also found overly-specific wording did not capture their actual experience. For example, two DCP items assess the impact of diabetes on food intake, but not “when I would like” to eat. One respondent provided feedback on the seventh item of the modified DIDP (‘Your freedom to eat as you wish’), reporting that this question stood out and was “much more specific than the other[s]”.

For the DCP and DSQoLS, participants also indicated that the timeframe referred to in the instructions (e.g. “in the past 4 weeks”, “in the past year”) was confusing, “easily forgotten”, or “too long”.

Response options

There was mixed feedback regarding the preferred number of response options and phrasing. Some participants reported liking the bi-directional response scale used in the DIDP, while others stated that they found it confusing or that it did not account for the potential combined positive and negative impacts of diabetes on particular aspects of life, such as physical health. In general, participants favoured inclusion of a “not applicable” response option, and some reported that this option was missing from the DCP and DSQoLS. Across measures, several participants indicated a desire to explain the underlying reasons for their response and participants positively reviewed the inclusion of an open-ended question inviting a free-text response in the ADDQoL-19.

Difficulty assessing the impact of diabetes

In rating the impact of diabetes on a particular aspect of life, the ADDQoL-19 asks participants to imagine their life without diabetes (e.g. ‘If I did not have diabetes, my quality of life would be…’). Though some participants reported that the opportunity to reflect on their life without diabetes was “an interesting concept”, others disliked this hypothetical phrasing and found the questions “impossible” to answer. While the other four measures of interest do not instruct participants to compare or rate their QoL against a life without diabetes, some participants reported difficultly responding as it’s “the only life I know” and that they may have over- or underestimated the impact of diabetes on their QoL. Participants also reported difficulty isolating diabetes from the impact of other life factors (e.g., other health condition, responsibilities, financial challenges). Some indicated a preference for a generic measure, which would allow participants to rate their QoL overall.

Questionnaire comprehensiveness and relevance

Where participants reported questionnaires as irrelevant, some indicated an inability to identify with the questionnaire, or that the questionnaire assumed a view of diabetes that did not reflect (their) reality. The lengthy DSQoLS (57 items) had the most references discussing perceptions of its (ir)relevance, while feedback on the shorter DIDP (7 items) indicated that it was “simplistic”.

Relevance of specific aspects of life and perceived omissions

Conflicting feedback was identified regarding the relevance of measured domains, particularly related to the impact of diabetes on finances. More typically, UK participants perceived such questions as irrelevant and suggested their removal, while Australian respondents endorsed such questions as relevant. Other specific questionnaire items reported as relevant included future health and development of diabetes complications (DSQoLS); emotional well-being (DIDP and DSQoLS, ); body image and family relationships (Diabetes QOL-Q). In contrast, irrelevant items commonly related to romantic partners and intimacy (DSQoLS, ADDQoL-19); driving (Diabetes QOL-Q); eating as you wish (ADDQoL-19); treatment modality-specific questions (DSQoLS), and experience of hypoglycaemia (DSQoLS).

Table 4 lists topics, or aspects of life, which participants perceived as missing from a questionnaire or not adequately assessed. Across questionnaires, commonly perceived omissions related to the impact of diabetes on mental / emotional health, and on finances, as well as the impact of diabetes management activities (e.g. insulin administration modality; diet; glucose monitoring) and extreme glucose levels on overall QoL. The Diabetes QOL-Q had the fewest references to omissions overall (n = 22), while the DSQoLS had the most reported omissions (n = 69), of which most referred to the impact of hyperglycaemia.

Table 4 List of domains perceived as omitted or inadequately assessed in one or more questionnaireQuestionnaire length and repetition

Related to, but discrete from comprehensiveness, participants commented on the length of measures, and were divided in their preference for brevity versus breadth. For example, the 7-item DIDP was described both as “brief and direct” and “so brief it was almost offensive”. Similarly, the 57-item DSQoLS was reported by some to be “too long” and by others as allowing for a detailed or comprehensive assessment (Theme 2). Item repetition was reported for the DSQoLS, to a lesser degree for the DCP and Diabetes QOL-Q and not at all for the ADDQoL-19 or the DIDP. Participants reported that repetitive questioning was “depressing”, overstated the relevance of certain topics/life aspects (i.e. the 11 DSQoLS items examining the impact of hypoglycaemia), and left them feeling more “worried”.

Questionnaire framing

Some respondents reported disliking the negative framing used by the ADDQoL-19, DCP and DSQoLS because they perceived it as placing a focus on the limitations of diabetes. In contrast, others appreciated the ADDQoL-19’s recognition of the “negative aspects of life”. Feedback about the positively-framed Diabetes QOL-Q was similarly divided between those who liked the “more positive manner” and those who felt it painted a “falsely positive picture”. Participants also reflected on the specific words and phrases used, reporting that certain terms (e.g. ‘normal’ and ‘control’ in the Diabetes QOoL-Q; ‘burden’, ‘worry’, ‘bother’ in the DSQoLS) made them feel like a “victim” or implied that diabetes controlled them. In contrast, the language used in the ADDQoL-19 was described as “extremely respectful”.