Characterization of pre-transplant psychosocial burden in an integrated national islet transplant program.

The psychological burden experienced by people with diabetes prior to islet transplantation is recognized but has not been studied comprehensively, especially in relation to glycemia. Therefore, we conducted a rigorous pre-operative psychosocial profile of UK islet transplant recipients, and compared groups with higher/lower HbA1 c to test the null hypothesis that pre-transplant hypoglycemia awareness and psychosocial burden would not be related to baseline HbA1 c in this high-risk cohort. Pre-transplant, recipients (n = 44) completed validated hypoglycemia awareness questionnaires and generic/diabetes-specific measures of psychological traits and states. Scores were compared in groups, dichotomized by HbA1 c (≤8% versus >8%). Participants were aged (mean±SD) 53 ± 10 years; 64% were women; with HbA1 c 8.3 ± 1.7%. Median rate of severe hypoglycemia over the preceding 12 months was 13 events/person-year and 90% had impaired awareness of hypoglycemia (Gold/Clarke score ≥4). Participants had elevated fear of hypoglycemia (HFS-II Worry), impaired diabetes-specific quality of life (DQoL) and low generic health status (SF-36; EQ-5D). One quarter reported scores indicating likely anxiety/depression (HAD). Dispositional optimism (LOT-R) and generalized self-efficacy (GSE) were within published 'norms.' Despite negative perceptions of diabetes (including low personal control), participants were confident that islet transplantation would help (BIPQ). Hypoglycemia awareness and psychosocial profile were comparable in lower (n = 24) and higher (n = 20) HbA1 c groups. Islet transplant candidates report sub-optimal generic psychological states (anxiety/depressive symptoms), health status and diabetes-specific psychological states (fear of hypoglycemia, diabetes-specific quality of life). While their generic psychological traits (optimism, self-efficacy) are comparable with the general population, they are highly optimistic about forthcoming transplant. HbA1 c is not a proxy measure of psychosocial burden, which requires the use of validated questionnaires to systematically identify those who may benefit most from psychological assessment and support.

Introduction

Following seminal success with the Edmonton protocol, insulin independence was established as the principal goal for islet transplantation.[1] Edmonton’s success has been replicated internationally but sustained insulin independence for more than 3 years cannot be achieved in more than 50% of recipients.[2] In the UK, islet transplantation is established as a fully reimbursed, equitably available intervention, where the goal is resolution of recurrent severe hypoglycemia (i.e., requiring the assistance of another person for recovery)[3] and not insulin independence per se. This has been demonstrated alongside improved awareness of hypoglycemia and attainment of optimal HbA1 c (<7%).[3]

Simple assessment of severe hypoglycemia frequency and HbA1 c does not illuminate an individual’s experience of living with diabetes. The potential impact of islet transplant on psychological outcomes (for better or worse) has been emphasized.[4-10] Therefore, it is important that pre-transplant psychosocial experience is taken into consideration. Increasingly, patient-reported outcomes (PROs) are considered alongside biomedical outcomes as part of holistic transplant assessment. However, a systematic review showed that PRO assessment in potential islet transplantation recipients has been limited largely to the appraisal of generic health status and/or fear of hypoglycemia.[4] Few studies have conducted a rigorous assessment of psychological burden in this high-risk cohort. The most comprehensive to date found impaired generic mental health status and depressive symptoms, alongside high fear of hypoglycemia and elevated diabetes distress.[11] These factors may predict adaptation to life post-transplant, including the need for lifelong immunosuppression medication.[12] Furthermore, psychological traits, such as dispositional optimism and general self-efficacy, which may also be of consequence, have not been assessed previously in an islet transplantation population.

The inverse relationship between HbA1 c and severe hypoglycemia reported in the landmark Diabetes Control and Complications Trial[13] appears to have driven a belief that sufficiently problematic/frequent severe hypoglycemia to justify islet transplantation occurs only in those with lower HbA1 c. Conversely, it is often considered that suboptimal psychological well-being is only experienced in those with higher HbA1 c. Yet, severe hypoglycemia is associated with considerable psychological burden,[14,15] and recent real-world clinical data demonstrate that severe hypoglycemia occurs in adults with type 1 diabetes regardless of their HbA1 c level.[14,16]

The primary aim of this study was to characterize the pre-transplant self-reported psychological and health burden experienced by recipients in the UK program. We hypothesized that pre-transplant hypoglycemia awareness and psychological burden would not be related to baseline HbA1 c in this high-risk cohort and, therefore, compared those with higher (>8%) and lower (≤8%) HbA1 c at listing for transplantation.[17]

Results

Forty-four consecutive participants consenting to this study completed questionnaires on completion of pre-transplant assessment and activation on the waiting list. All subsequently received one or more islet transplant (70 islet transplants in total). Table 2 summarizes the demographic and clinical characteristics of participants. Sixty-four percent were women. Mean±SD age was 53 ± 10 years, with diabetes duration 34 ± 12 years, BMI 24.9 ± 3.4 kg/m2 and HbA1 c 8.3 ± 1.7% (67 ± 18 mmol/mol). All had experienced at least two severe hypoglycemia events over the preceding 2 years, fulfilling listing criteria. Within the preceding 12 months, 90% had experienced at least one severe hypoglycemic event, with 55% reporting ≥11 events, and 90% reported impaired awareness of hypoglycemia (Gold and/or Clarke score ≥4).

Table 2.

Demographic and clinical characteristics of islet transplant recipients (n = 44)

Gender, woman	28 (64%)
Age, years	53.1 ± 10.0
Diabetes duration, years	34.4 ± 11.5
BMI, kg/m[2]	24.9 ± 3.4
HbA1 c, %mmol/mol	8.3 ± 1.767 ± 18
Insulin dose, Units/kg/day	0.52 ± 0.19
Creatinine, µmol/l	89.8 ± 36.2
Severe hypoglycemia, events per person-year● 0● 1● 2–4● 5–10● 11–20● 21–50● >50	13.0 (3.5–51.0) [0–51.0]4 (10.0%)2 (5.0%)5 (12.5%)7 (17.5%)5 (12.5%)5 (12.5%)12 (20.0%)

Data are n (%) or, depending on distribution, mean±SD or median (IQR) [range].

Missing datapoints were apparent for creatinine (n = 42) and severe hypoglycemia (n = 40).

When demographic and clinical characteristics were compared in the group with higher HbA1 c (>8%: n = 20) versus those with lower HbA1 c (≤8%: n = 24) at listing for transplant, no significant differences were found in severe hypoglycemia rate or other baseline parameters, except a trend toward longer diabetes duration in the lower HbA1 c group (≤8%: 38 ± 12 vs >8%: 30 ± 9 years; p = .028).

Psychosocial outcomes are summarized for the whole cohort and stratified by higher and lower HbA1 c in Table 3. Mean±SD HFS-II Worry score in the overall cohort was 45 ± 15, indicating high fear of hypoglycemia. Mean DQOL subscale scores for Satisfaction, Impact and Diabetes Worry ranged from 47 to 57, indicating impaired diabetes-specific quality of life; while the Social Worry scale had considerable missing data.

Table 3.

Pre-transplant psychosocial characteristics, for whole sample and stratified by HbA1c

Diabetes-specific psychosocial states	n	Whole sample	HbA1c				p-Value
			n	≤8%(n = 24)	n	>8%(n = 20)
Hypoglycemia awareness● Gold score (1–7)● Clarke questionnaire (0–7)	3030	6.0 (6.0–7.0) [1–7]6.0 (5.0–7.0) [1–7]	1716	7.0 (6.0–7.0)5.5 (5.0–7.0)	1314	6.0 (6.0–7.0)6.0 (6.0–7.0)	.234 b.644 b
Fear of hypoglycemia● HFS-II Worry (0–72)● HFS-II Behavior (0–60)	4039	45.2 ± 15.334.4 ± 13.3	2122	45.7 ± 15.431.0 ± 14.4	1917	44.8 ± 15.438.9 ± 10.6	.867 a.066 a
Diabetes-specific quality of life (0–100)● DQoL Satisfaction● DQoL Impact● DQoL Social Worry● DQoL Diabetes Worry	4342844	56.8 ± 16.851.7 ± 13.375.6 ± 15.747.3 ± 22.0	2323524	59.8 ± 17.453.5 ± 13.383.2 ± 13.247.7 ± 18.1	2019320	53.3 ± 15.849.6 ± 13.462.9 ± 11.347.0 ± 26.5	.206 a.359 a.069 a.921 a
Perceptions of diabetes (0–10)● BIPQ Consequences● BIPQ Timeline● BIPQ Personal control● BIPQ Transplant control● BIPQ Identity● BIPQ Illness concern● BIPQ Coherence● BIPQ Emotional representation	4343434241424244	9 (8–10) [4–10]10 (10–10) [0–10]3 (1–6) [0–10]10 (8–10) [7–10]8 (4–9) [0–10]10 (9–10) [3–10]10 (8–10) [0–10]8 (5.5–10) [0–10]	2324242323242324	9.0 (8.0–10)10.0 (10–10)4.0 (0.25–8.0)10 (8.0–10.0)8.0 (5.0–10)10.0 (8.5–10)10 (9.0–10)8.0 (5.0–9.5)	2019191918181920	8.5 (7.5–10.0)10 (10–10)3.0 (1.0–6.0)10 (8.0–10)6.5 (1.0–9.0)10 (9.0–10)9.0 (8–10)8.5 (7.5–10)	.543 b.732 b.413 a.852 b.158 b1.000 b.298 b.290 b
Generic psychosocial states and traits
Anxiety symptoms: HAD-A (0–21)● HAD-A ≥ 8● HAD-A ≥ 11 (‘caseness’)Depressive symptoms: HAD-D (0–21)● HAD-D ≥ 8● HAD-D ≥ 11 (‘caseness’)	43211143209	7.9 ± 4.921 (48.8)11 (26.6)6.9 ± 4.420 (46.5)9 (20.9)	239223115	6.4 ± 4.69 (39.1)2 (8.7)6.1 ± 4.411 (47.8)5 (21.7)	201292094	9.6 ± 4.812 (60.0)9 (45.0)7.8 ± 4.49 (45.0)4 (20.0)	.032 a.221 a
Dispositional optimism: LOT-R (0–24)	44	11.8 ± 4.4	24	12.2 ± 5.0	20	11.3 ± 3.7	.500 a
Generalized self-sfficacy: GSES (10–40)	42	29.2 ± 4.6	22	30.4 ± 4.8	20	27.9 ± 3.9	.080 a
Generic health status
SF-36v2 Summary scores (0–100)● Mental component● Physical component	4141	40.0 ± 13.743.0 ± 8.2	2222	43.2 ± 12.943.8 ± 7.7	1919	36.3 ± 14.142.0 ± 8.8	.110 a.486 a
EQ-5D-3 L (% with some/extreme problems)● Mobility● Self-care● Usual activities● Pain/discomfort● Anxiety/depression	4142424241	22 (53.7%)9 (21.4%)29 (69.0%)29 (69.1%)24 (55.8%)	2323232322	11 (47.8%)3 (13.0%)17 (73.9%)15 (65.2%)12 (54.5%)	1819191919	11 (61.1%)6 (31.6%)12 (63.2%)14 (73.7%)12 (63.2%)	.397 c.257 d.453 c.555 c.577 c

Data are n (%) or, depending on distribution, mean±SD or median (IQR) [range]. Differences analyzed using aStudent’s t-test, bMann-Whitney U test, cChi-square test or dFisher’s exact test.

For each measure the potential range is shown in the left-hand column next to the name of the measure or subscale. Higher scores Gold and Clarke Score indicate likely impaired awareness of hypoglycemia. Higher scores on the HFS-II indicate increased worry about, and behavior to avoid hypoglycemia and its negative consequences. Higher scores on the DQOL indicate better diabetes-specific quality of life, while higher scores on the SF-36 indicate better general physical or mental health status. Higher scores on the BIPQ indicate greater endorsement of the named dimension (perception of diabetes). Higher HAD scores indicate greater anxiety or depression symptomatology. Higher scores on the LOT-R indicate a greater disposition toward optimism. Higher GSES scores indicates greater confidence in one’s capacity for coping.

Regarding diabetes perceptions (Table 3), study participants reported very high ‘Concern’ about their diabetes and its impact on their lives (‘Consequences’). They perceived that diabetes would ‘continue forever’ (‘Timeline’) and reported a negative emotional burden ‘making them angry, scared, upset or depressed’ (‘Emotional representation’). Mid-range scores for ‘experiencing symptoms from your diabetes’ were reported (‘Identity’). While participants reported understanding their diabetes ‘clearly’ (‘Coherence’), they felt little ‘Personal control’ over their condition. However, participants believed resoundingly that a transplant would ‘help their diabetes’ (‘Transplant control’).

Almost half reported clinically relevant (at least mild) depression symptoms (HAD-D ≥ 8: 47%) and/or anxiety symptoms (HAD-A ≥ 8: 49%). SF-36 Summary scores indicated substantial impairment across Mental and Physical components. For the EQ-5D, more than two-thirds reported (some/extreme) problems in Usual Activities and Pain/Discomfort, half reported (some/extreme) problems with Anxiety/Depressive symptoms and Mobility, while one in five had problems with Self-care.

For general psychological traits, the LOT-R and GSES scores indicated that dispositional optimism and generalized self-efficacy were comparable with published norms for the UK general adult population.[18-20]

Impaired awareness of hypoglycemia was comparable in those with higher and lower HbA1 c (Table 3). No significant differences were observed between the two HbA1 c groups for diabetes-specific or generic psychosocial measures, with the exception of a trend toward greater anxiety in those with higher HbA1 c (p = .032). There were no significant differences in self-reported health status for the EQ-5D domains or SF-36 Summary scores.

Discussion

This study provides the most comprehensive psychosocial characterization to date of adults with type 1 diabetes undergoing islet transplantation. It demonstrates that the self-reported psychological and health burden before the procedure is not trivial, with many reporting considerable psychological distress associated with recurrent dangerous hypoglycemia and long disease duration.

Despite problematic hypoglycemia being the major indication for islet transplantation candidacy, fear of hypoglycemia has been assessed in only four previous studies.[6,8,10,21] In the current study, we used the HFS-II, but others have used the original HFS, or used an alternate scoring method for the HFS-II, such that scores are not directly comparable between islet transplantation studies. In the current cohort, mean HFS-II Worry score was twice that of the US validation sample comprising a wide range of individuals with established type 1 diabetes,[22] and comparable with or higher than studies in which participants have reported at least one severe hypoglycemic event within the past 6 months.[14,17] Norms and cut-points to define clinically relevant elevation in fear of hypoglycemia have been established for the HFS-II in adults with type 2[23] but not type 1 diabetes; however, higher scores indicate greater fear and are typically found among those with most frequent and severe hypoglycemia.[22] Listing for islet transplantation in those with recurrent severe hypoglycemia and impaired awareness of hypoglycemia is thus associated with significant pre-transplant fear of hypoglycemia.

Diabetes-specific quality of life was assessed with the DQOL questionnaire in order to enable comparisons with previous islet transplantation studies.[4] Scores suggested lower diabetes-specific quality of life in prospective UK recipients than observed at baseline in other published islet transplant studies,[7,9] and also in a UK study of adults with type 1 diabetes,[24] reflecting the high burden of living with diabetes in the current cohort. These findings are consistent with an in-depth, qualitative study of islet transplant candidates, which demonstrated a high pre-operative psychological burden, predominantly from recurrent, unpredictable severe hypoglycemia, with considerable negative impact on quality of life.[5]

At the point of joining the islet transplantation waiting list, questionnaire screening indicative of ‘caseness’ for anxiety and depression was detected in one quarter and one-fifth of the participants, respectively. The HAD scale can offer only an indicator of symptom severity, which requires confirmation with a clinical diagnostic interview. While these rates appear concerning, they are similar to those reported in epidemiological studies and meta-analyses of the general type 1 diabetes population.[25-27] For example, these pre-transplant rates are highly comparable with those observed in a large cohort (n = 639) – 24% had ‘caseness’ for anxiety and 21% for depression – prior to attending a UK structured type 1 diabetes education program.[28] Importantly, it has been noted elsewhere that questionnaire-based scores of generic anxiety/depression show moderate-to-strong correlations with diabetes-related distress.[23]

The Brief Illness Perceptions Questionnaire has not previously been completed by prospective islet transplant recipients. This confirmed how severely life is affected by diabetes in this select, high-risk cohort, leading to a high level of concern and considerable emotional impact. Participants felt little control over their diabetes, and perceived that islet transplantation would be extremely helpful. Despite being on the islet transplant list they had a clear perception that their diabetes would continue forever. It was notable, however, in the preceding qualitative study, that UK islet transplant recipients hoped for at least ‘a break from diabetes.’[5] They reported realistic expectations, having being counseled that long-term insulin independence was not a goal, but many retained ‘hidden hopes of being among the minority to remain insulin free.’

While participants reported a high level of confidence that the transplant would help their diabetes, their general dispositional optimism was comparable to the general UK adult population[18,19] and somewhat lower than that of Dutch adults with type 1 diabetes.[29] Generalized self-efficacy levels among the current islet transplantation recipients were also comparable with the general UK adult population[20] and with Dutch adults with type 1 diabetes.[29] Many studies have highlighted the positive health benefits of an optimistic outlook, though these have tended to focus on outcomes of cardiovascular disease and cancer. A previous qualitative study of islet transplant candidates demonstrated that they were willing to accept the potential risks associated with transplantation and ongoing immunosuppression.[5] Psychological traits, such as dispositional optimism and general self-efficacy, were also shown to be important for how participants dealt with the uncertainty of ‘life on the list’ for a transplant.[5] This is the first islet transplantation cohort in which these psychological traits have been assessed quantitatively. Given that islet transplantation involves accepting, adapting and coping with the procedure and its consequences, it will be of interest to note in future studies, whether high expectations for transplantation (which may or may not be realized) and/or more generally optimistic beliefs impact upon biomedical outcomes and quality of life following islet transplantation.

Previous studies of conventional non-transplant intervention for severe hypoglycemia[17] have dichotomized participants into groups with HbA1 c ≤ 8% (to capture those with a strong primary motivation to avoid all high glucose excursions) and >8% (characterized by higher glucose variability) and we agreed on a priori to dichotomize the current pre-islet transplant cohort using this HbA1 c value. HbA1 c was >8% in just under half of those studied. All fulfilled the listing criteria of recurrent severe hypoglycemia with comparable impaired awareness of hypoglycemia and fear of hypoglycemia scores to those with lower HbA1 c. All diabetes and generic psychosocial scores in this highly selected cohort were very similar in higher and lower HbA1 c groups, with the exception of a trend toward greater anxiety in those with higher HbA1 c, consistent with studies in a wider spectrum of adults with type 1 diabetes.[30] This underlines the negative psychological impact and burden of severe hypoglycemia and impaired awareness of hypoglycemia, irrespective of baseline HbA1 c.

This study has several clinical implications. Clinicians need to be aware that individuals listed for islet transplantation are characterized by the considerable psychological burden and impact of their problematic hypoglycemia on well-being and quality of life; and that this is irrespective of whether baseline HbA1 c is in a higher or lower range. Furthermore, individuals considering islet transplantation exhibit high levels of confidence that the procedure can alleviate this burden, despite having only an averagely optimistic disposition in comparison to the wider population. As noted previously, these high expectations need to be recognized by transplant teams, and strategies considered for how best to support the individual in coping with disappointments at any stage.[5] Since its inception, the UK national program has included provision for psychological assessment of suitability for islet transplantation, mandated as part of the multidisciplinary team decision for inclusion on the waiting list.[31] The Diabetes UK patient guide ‘Islet Cell Transplant: What You Need to Know’ indicates that a meeting with a clinical psychologist will be arranged to determine whether ‘extra psychological support during or after the procedure’ may be needed.[32] In addition to ruling out major psychological/psychiatric co-morbidity, a particular focus is on assessing the individual’s level of understanding of the potential risks as well as benefits of islet transplantation, and determining whether they have realistic expectations. It is not clear whether expert psychological evaluation is embedded within other programs internationally.[33] Informed by the current study, we recommend formal-structured psychological assessment of all potential islet transplant recipients, to include evaluation of fear of hypoglycemia, diabetes-specific quality of life, anxiety/depressive symptoms and perceptions/expectations of the procedure. This will enable identification and follow-up of individualized psychological needs pre- and post-transplantation, facilitating provision of evidence-based support and intervention as required. This should augment adoption of the evidence-informed clinical pathways, which have been proposed in recent years to guide the therapeutic strategies offered to those with problematic hypoglycemia, including those under assessment for or awaiting islet transplant.[34] The absence of statistically significant differences in psychosocial questionnaire outcomes between those with higher versus lower pre-transplant HbA1 c further underlines the importance of holistic assessment of all potential islet recipients using validated tools and approaches, in parallel with baseline biomedical evaluation.

The current study has several strengths. It is a multi-center, multi-disciplinary study, recruiting a cohort of patients within the first national health service funded program of islet transplantation as an established clinical intervention for prevention of severe hypoglycemia and achievement of HbA1 c < 7%. It is the largest and most comprehensive study to date of the psychosocial characterization of adults undergoing islet transplantation. Furthermore, this study used validated generic and diabetes-specific psychological measures to characterize both psychological traits and states. This study also has several limitations, including a relatively small sample compared with non-islet transplant studies. There is no control group, as it was designed as a cohort study. We have not reported data on diabetes-related complications beyond confirmation of satisfactory renal function through serum creatinine. Despite few missing data overall, the completion rate for DQOL Social Worry subscale was very low, in keeping with the previously observed low face validity of this section of the questionnaire for an adult cohort.[35]

In summary, we have conducted detailed baseline profiling of a relatively large islet transplant cohort. Overall, these adults with long-standing type 1 diabetes exhibit considerable diabetes-specific psychological burden (i.e., lower diabetes-specific quality of life, greater fear of hypoglycemia and equivalent anxiety/depressive symptoms) compared with the wider adult population with type 1 diabetes in the UK. Their psychological traits (e.g., optimistic beliefs) are comparable to the general population, though they exhibit extremely high levels of confidence in the ability of islet transplantation to help with their diabetes. Furthermore, this study highlights that HbA1 c is not a proxy measure for underlying psychological burden. Comprehensive diabetes-specific psychosocial evaluation needs to be integral to the multidisciplinary assessment of whether islet transplantation will be suitable for an individual with type 1 diabetes; and ongoing psychosocial support will be needed throughout the transplant process.

Patients and methods

The integrated UK Islet Transplant Consortium (UKITC) program comprises seven National Health Service commissioned centers: Bristol, Edinburgh, King’s College London, Royal Free London, Manchester, Newcastle and Oxford. Following ethical approval from the NRES Committee North East – Tyne and Wear South (REC reference 07/Q0904/11) and informed written consent, a prospective cohort of consecutive individuals were recruited fulfilling all consensus inclusion criteria for islet transplantation without exclusion criteria (Table 1). In parallel with pre-listing biomedical assessment, all were invited to complete psychological measures at the point of joining the transplant waiting list between April 2008 and March 2011. All subsequently received one or more islet transplant(s).

Table 1.

Inclusion and exclusion criteria for the UK islet transplant program

Inclusion criteria	Exclusion criteria
● Age: ≥18 years● Diagnosis: C-peptide-negative diabetes● History: recurrent severe hypoglycemia (at least two events over the preceding 24 months requiring assistance from another person to administer carbohydrate, glucagon or other resuscitative actions[36]) despite optimized conventional management	● Insulin resistance (insulin requirement >0.7 U/kg)● Any contraindication to immunosuppression, including impaired renal function with isotopic GFR<60 mL/min/1.73 m2 or albumin excretion rate >300 mg/24 h (unless previous renal transplant)

Demographic and clinical data included HbA1 c and number of self-reported severe hypoglycemia events over the preceding 12 months.[36] A comprehensive set of psychological characteristics and outcomes were assessed, as described below.

Hypoglycemia awareness

The Gold Score[37] is a single question (“do you know when your hypos are commencing?”) rated on a 7-point Likert scale, from 1 = ‘always aware’ to 7 = ‘never aware.’ The eight-item Clarke Questionnaire[38] assesses occurrence of moderate and severe hypoglycemia, and experience of hypoglycemic symptoms. Responses to each question are categorized as either ‘aware’ or ‘reduced awareness.’ For both the Gold score and the Clarke questionnaire, a score of ≥4 indicates impaired awareness of hypoglycemia.

Fear of hypoglycemia

The Hypoglycemia Fear Survey-II (HFS-II)[39] includes 33 statements, rated in terms of how often each has been a concern in the last 6 months, from 0 = ‘never’ to 4 = ’almost always.’ Eighteen item scores are summed to create a ‘worry’ score (range: 0–72), where higher scores indicate greater worry about hypoglycemia and its negative consequences. Fifteen item scores are summed to form a ‘behaviour’ score (range: 0–60), where higher scores indicate increased behavior to avoid hypoglycemia and its negative consequences.

Diabetes-specific quality of life

The Diabetes Quality Of Life questionnaire (DQOL)[40] includes 46 items across four subscales: ‘life satisfaction’ (15 items), ‘diabetes impact’ (20 items), ‘worries about diabetes’ (4 items), and ‘social/vocational concerns’ (7 items). All items are scored on a 5-point Likert scale (from 1 (‘very satisfied,’ ‘no impact, or ‘no worry’) to 5 (‘very dissatisfied,’ ‘always impacted,’ or ‘always worried’)). Subscale scores are calculated by taking the sum of items (reversing scores as needed) and converting this raw total to a score out of 100. Higher scores indicate better diabetes-specific quality of life.

Perceptions of diabetes

The Brief Illness Perceptions Questionnaire (BIPQ)[41] includes eight items, each one assessing a different dimension (e.g., perceived control, emotional impact). Each item is rated on an 11-point scale (from 0 to 10), where higher scores indicate greater endorsement of that dimension. According to convention, the questionnaire was made condition-specific for the current study, by replacing ‘illness’ with ‘diabetes,’ and replacing ‘treatment’ with ‘transplant.’ For item 3 (‘How much control, do you have over your diabetes?’) and item 7 (‘How well do you feel you understand your diabetes?’), higher scores indicate more positive perceptions of diabetes. Item 4 is the only treatment-specific question (‘How much do you think your transplant can help your diabetes?’), with higher scores referring to a more positive perception. For all other items, higher scores indicate more negative perception of diabetes impact.

Anxiety and depressive symptoms

The Hospital Anxiety and Depression (HAD) scale[42] includes two 7-item subscales: ‘anxiety’ and ‘depression.’ Respondents rate items using a 4-point scale (from 0 to 3). Item scores are summed to form subscale scores (from 0 to 21), with higher scores indicating greater anxiety or depression symptomatology. A score of ≥8 indicates at least mild anxiety or depression symptoms (borderline ‘caseness’), while a score of ≥11, indicating moderate-to-severe symptoms, is regarded as ‘caseness.’[42]

Optimistic beliefs

Outcome expectancies were measured with the Life Orientation Test – Revised [LOT-R],[43] which includes 10 statements rated on a 5-point Likert scale, where 0 = ‘strongly agree’ and 4 = ‘strongly disagree.’ The three pessimism item scores are reversed and added to the three optimism item scores, while four ‘filler’ items are ignored. Total scores range from 0 to 24, with higher scores indicating a greater disposition toward optimism. Participants also completed the Generalized Self-Efficacy Scale (GSES)[44] includes 10 statements rated on a 4-point scale, where 1 = ‘not at all true’ and 4 = ‘exactly true.’ Item scores are summed to create a total score ranging from 10 to 40, with higher scores indicating greater confidence in one’s capacity for coping.

Generic health status

The SF-36v2[45] is a 36-item questionnaire that yields two summary scores: Physical Component Score and Mental Component Score (each ranging from 0 to 100). Participants also completed the EQ-5D-3 L[46] tariff, which comprises five items or dimensions of general health rated using three response options (‘no problems,’ ‘some problems,’ ‘extreme problems’). For the current study, we report the proportion of participants experiencing ‘some/extreme problems’ for each dimension. For both SF-36 and EQ-5D, higher scores indicate better general health status and lower scores indicate more health limitations.

Statistical analyses were conducted using STATA statistical package version 14.1 (StataCorp, Texas, USA). For all psychological measures, missing data were handled with expectation-maximization imputation for up to 10% missing data, unless the questionnaire guidelines stated otherwise. For the DQOL, subscale scores were not computed if respondents had more missing data than deemed acceptable in the scoring guideline.[40] Categorical data were expressed as n (%). Continuous data were expressed as mean ± standard deviation (SD) or median (interquartile range) [range] as appropriate to the data distribution. Demographic, clinical and psychological outcomes were compared in participants with higher (>8%), and lower (≤8%) HbA1c.[17] Comparisons were conducted using Student’s t-test for parametric data and Mann-Whitney U test for non-parametric data. Categorical variables were compared using the Chi-square test, or Fisher’s exact when the expected values in any of the cells of a contingency table are below 5, or below 10 when there is only one degree of freedom. Given the number of statistical tests, and the exploratory nature of the analysis, significance was set at p < .01.