Multi-Dimensional Barriers for Work Productivity and Possible Solutions to Overcome Presenteeism in Young-Onset Diabetes: A Mixed-Methods Study.

Multi‐Dimensional Barriers for Work Productivity and Possible Solutions to Overcome Presenteeism in Young‐Onset Diabetes: A Mixed‐Methods Study

ABSTRACT

Aims

Young‐onset non‐type 1 diabetes (YOD, age‐at‐diagnosis ≤ 40) is associated with poor physical and mental health, presenteeism and productivity loss. We examined patient‐perceived barriers to work productivity and key enablers to reduce presenteeism.

Materials and Methods

In this mixed‐methods study, we utilized validated questionnaires to measure presenteeism (World Health Organization Health and Performance Questionnaire); self‐care activities (Summary of Diabetes Self‐Care Activities; Compliance Questionnaire for Medication); health‐related quality of life (HRQoL, EQ‐5D‐3L) and psychological distress (Depression Anxiety Stress Scale; Chinese Diabetes Distress Scale) among participants with YOD in the Precision Medicine to Redefine Insulin Secretion and Monogenic Diabetes‐Randomised Controlled Trial (PRISM‐RCT). We conducted four semi‐structured focus group discussions (FGDs) for inductive thematic analysis.

Results

Employees with YOD (n = 465, mean age: 43.4 ± 6.4 years, disease duration: 10.9 ± 6.0 years, 59.8% male) had better cardiometabolic, psychological and HRQoL profiles than non‐employees (n = 103). Among the employees, 67.4% had at least one sickness presenteeism episode in the preceding year, with a mean presenteeism score of 7.1 ± 1.6 (out of 10). Increased presenteeism was independently associated with diabetes‐related emotional burden (β = −0.08, p = 0.029) and suboptimal HRQoL (β = 3.30, p = 0.041). [Correction added on 11 September 2026, after first online publication: In the preceding sentence, the beta and p values have been corrected.] Participants of FGDs (n = 16) reported distress contributors related to inter‐personal (psychological/physical status, family collectivism), work (schedules/types, absence justification, colleagues' attitudes) and societal factors (relationships with providers, out‐of‐pocket payment, stigmatisation). Enablers included public awareness, early diagnosis (genetic testing for offspring, regular health‐check), peer support, treatment choices autonomy and insurance coverage.

Conclusion

In YOD, presenteeism is associated with interpersonal, work and societal factors accompanied by psychological distress, suboptimal diabetes care, self‐management challenges, calling for holistic assessment and person‐centred management.

Abbreviations

FGDfocus group discussion

HRQoLhealth‐related quality of life

YODYoung‐onset non‐Type 1 diabetes

1Introduction

The prevalence of young‐onset non‐type 1 diabetes (YOD) diagnosed before age 40 is increasing [1]. China has the largest number of people with diabetes estimated at 148 million, partly due to economic growth with shifts in demographics, environment, lifestyle and healthcare needs [2]. Patients with YOD experience accelerated decline in β‐cell function often driven by increased adiposity‐associated insulin resistance leading to premature complications and mortality [3]. These young individuals face psychological‐behavioural challenges [4] in adjusting lifestyle and adhering to the medical care plan and often experience diabetes‐related distresses due to conflicts from work and family caregiving duties [5]. Social determinants also impede annual review and attainment of target glycated hemoglobin (HbA1c) [6]. These multidimensional factors pertinent to personal, work, healthcare and societal factors were associated with a two‐ to sixfold increased risk of cardiovascular‐renal events and recurrent hospitalisations, notably related to mental illnesses, in YOD [7, 8, 9].

Poor quality of life, complications and impaired physical functioning co‐occurred with absenteeism (absent from work) and presenteeism (physically present at work but with reduced work performance or productivity) [10, 11]. Persistence with self‐care practice from the time of diagnosis is essential to sustain their well‐being and productivity [9, 12]. Young adulthood, particularly between the ages of 30 and 40 years, represents a critical life stage marked by career advancement and increasing family responsibilities [13]. Integrating complex self‐care routines into daily life may create diabetes distress and/or negative emotions, affecting one's daily life and their work productivity [9, 12]. Existing presenteeism‐related research focused on quantifying healthcare costs and risk factors. In the United Kingdom, psychosocial distress and impaired health‐related quality of life (HRQoL) were associated with presenteeism [14]. Sociocultural dynamics, workplace environments and health systems may also shape the experiences of people with YOD undergoing rapid socioeconomic transition [15].

In the Hong Kong Diabetes Register comprising 19 440 Chinese patients with type‐2 diabetes (T2D), onset of complications was associated with an average incremental healthcare cost of USD$2990 per patient per year, with hospitalisations accounting for the majority of the excess expenditure [16]. In an earlier report including 444 Hong Kong Chinese patients with YOD in the Precision Medicine to Redefine Insulin Secretion and Monogenic Diabetes randomised controlled trial (PRISM‐RCT), 93.9% of the participants experienced productivity loss, with a projected annual loss of USD$1.12 billion (37% from presenteeism, 63% from absenteeism). Female employees, higher health locus of control, higher salary, flexible work schedules and caregiving responsibilities for children were associated with lower presenteeism [10]. In this analysis, we further evaluated the associations of patient‐reported outcomes (PROMs) including self‐care activities, HRQoL and psychological distress with presenteeism and work productivity. We also conducted focused group discussions (FGDs) to explore possible barriers faced by people with YOD on self‐care and work that may affect productivity and enablers to overcome presenteeism loss.

2Methods and Materials

2.1Study Design

Between June 2023 and December 2024, we conducted a convergent‐parallel mixed‐methods study, comprising cross‐sectional analysis of factors associated with presenteeism. Nested online FGDs were conducted as an exploratory, hypothesis‐generating complement to the quantitative analysis to identify barriers, mechanisms and potential enablers of presenteeism not captured quantitatively.

2.2Sample Selection and Recruitment

Participants were recruited from the PRISM‐RCT (NCT04049149) [17, 18] at the Prince of Wales Hospital, Hong Kong (Supporting Information Appendix S1). This mixed‐methods study was embedded within the ‘Economic Impact of Diabetes and Related Complications (EIDC)’ substudy of the PRISM study during Year‐3 follow‐up visits. Inclusion criteria included Chinese ethnicity, ages 18–50 with age‐at‐diabetes‐diagnosis ≤ 40 years. Patients with physician‐diagnosed type‐1 diabetes (T1D) defined as presentation with diabetic ketoacidosis and continuous insulin requirement within 12 months of diagnosis were excluded. Upon completion of the quantitative survey [10], participants who expressed interest in the follow‐up FGDs were enrolled into the recruitment pool. Participants from the pool were then purposively selected using a maximum variation sampling strategy to ensure heterogeneity of perspectives, guided by four pre‐defined diversity criteria: employment types, work schedules, presence of complications and use of insulin.

2.3Measures and Procedures

2.3.1Study Assessments

Presenteeism was measured by the World Health Organization Health and Performance Questionnaire (WHO‐HPQ) [19] using a self‐assessed item ‘How would you rate your overall job performance on the days you worked during the past 4 weeks (28 days)?’ expressed on a 10‐point Likert scale (0—worst performance; 10—top performance), with lower scores indicating greater presenteeism (i.e., productivity loss). A cut‐off score of ≤ 40 (after multiplying the presenteeism score by 10) was used to indicate high‐presenteeism [20].

We used the following PROMs to measure psychological distress, HRQoL and self‐care activities: Depression Anxiety Stress Scales (DASS‐21) [21]; EuroQoL 5‐Dimension 3‐Level (EQ‐5D‐3L) [22, 23]; Chinese version of the Diabetes Distress Scale (CDDS‐15) [24]; Summary of Diabetes Self‐Care Activities (SDSCA‐15) [25] and Compliance Questionnaire (CQ‐4) [26] along with documentation of demographic and socioeconomic characteristics, medical history and medication use. Cardiometabolic risk factors included fasting plasma glucose, HbA1c, high‐density lipoprotein cholesterol (HDL‐C), low‐density lipoprotein cholesterol (LDL‐C), triglycerides (TG), systolic (SBP) and diastolic blood pressure (DBP) and markers of kidney function (spot urine albumin‐to‐creatinine ratio [uACR] and estimated glomerular filtration rate[eGFR]) [27].

2.3.2Qualitative Semi‐Structured FGD Study Design and Procedures

We designed a question guide covering decision‐making processes underlying presenteeism, the influence of workplace and human‐resource policies on employees' ability to manage their health and the socioeconomic impact of diabetes and its complications on employees living with the condition (Supporting Information Appendix S2). We conducted the semi‐structured FGD by Zoom meeting. Each FGD comprised three to five participants and lasted 60–90 min, conducted sequentially by the lead researcher and a trained research assistant. Chinese verbatim transcripts were preliminarily coded after each session. Data collection ceased upon unanimous agreement on thematic saturation when no new codes emerged from subsequent FGDs. Saturation was achieved after four FGDs with themes redundancy which fell within a recommended range of 4–8 FGDs [28].

2.4Statistical Analysis

Data were summarised as mean ± standard deviation (SD), median (interquartile range, IQR) or percentages. Group comparisons were performed using independent t‐tests and Mann–Whitney U tests, as appropriate. We used hierarchical linear regression, adjusted for covariates, to evaluate associations of psychological health, HRQoL and self‐care activities with presenteeism. Model‐1 included demographics; Model‐2 additionally adjusted for cardiometabolic risk factors and Model‐3 further added medications (Supporting Information Appendix S1).

Thematic analysis was conducted using NVivo14. Four researchers inductively coded transcripts following Braun and Clarke's six‐step framework (Supporting Information Appendix S1), allowing codes to emerge directly from the data [29]. The researchers' gender, roles, qualifications and reflexivity are described in Supporting Information Appendix S1. The team met after the interview to compare, discuss and refine codes to reach consensus. Emergent themes were organised into a multi‐level framework derived inductively from the data rather than defined a priori. Joint display diagrams were constructed and meta‐inferences were generated to integrate the quantitative and qualitative findings.

TABLE 1Mixed‐methods joint display of findings and meta‐inferences.

Quantitative findingsa
Qualitative findings Meta‐inferences

Comparisons Contributing factors Healthcare enablers

↑ Presenteeism group had …

↑ Stress Psychological and physical strain

Peer support group

The high‐presenteeism group experienced higher overall stress, which may come from work‐life imbalance, self‐care demands and family responsibilities. Peer support might alleviate loneliness and isolation while strengthen their capacity for diabetes management.

↓ Dietary adherence Cultural collectivism

Psychological and physical strain

Work types and schedules

Concerns about the attitude of employers and colleagues

Healthy lifestyle promotion

Peer support group

Participant perceived psychological and physical strain could trigger emotional eating; while family and workplace environments or schedules could lead to irregular meal times and limited food choices. Healthy lifestyle promotion not only expands food choices, but also equips individuals with healthier dietary adherence.

↑ Working hours; ↑ sickness attendance Concerns about the attitude of employers and colleagues

Societal perception

Genetic screening and regular body check

Educational programme

Long working hours and sickness attendance were prominent among the high‐presenteeism group. This pattern is likely driven by fears of workplace stigmatisation (e.g., diabetes signals incompetence), compounded by widespread misunderstandings of the disease. Educational programmes and genetic screening/regular health checks could raise public awareness, correct misconceptions and enhance self‐management.

↑ Regimen‐and‐social‐support‐related distress Medical costs

Relationships with care providers

Patient involvement in shared decision‐making

Conditional insurance coverage

Genetic screening and regular body check

The regimen‐ and social support‐related distress may arise from effective yet high‐cost treatment, lack of autonomy treatment planning and insufficient trust in providers. Involving patients in shared decision‐making and offering genetic screening may empower them over their condition and treatment; while conditional insurance could help ease their financial burden, thus, reduce their distress.

Associations

↑ Diabetes‐related emotional burden predicted ↑ presenteeism Challenges in applying for leaves

Work types and schedules

Peer support group

Healthy lifestyle promotion

Educational programme

Participants reported significant challenges in balancing work and self‐care. Inflexible company policies and demanding work schedules were cited as key barriers, resulting in missed medical follow‐ups and increased presenteeism. Some participants changed their job to better manage their health. Peer support, healthy lifestyle promotion and educational programmes could raise awareness on health priorities, reduce emotional burden and improve HRQoL, thereby lowering presenteeism.

↓ HRQoL predicted ↑ presenteeism Psychological and physical strain

Medical costs

Work types and schedules

a↑ = higher level, ↓ = lower level.

3Results

3.1Quantitative Findings

In this EIDC substudy, among 568 Chinese participants with YOD, 82% (n = 465) were employees (Figure 1). Employees were more likely to be men, older at diagnosis, carers of child(ren) under 18‐year‐old and had better cardiometabolic risk factors and lower prevalence of all‐site cancer than non‐employees (Table S1). Among the employees, the weekly working hours were 48.9 ± 14.0. One‐third of them earned HKD18 700–29 499/month (USD$1 = HKD$7.8). 67.4% reported at least one episode of sickness presenteeism last year, with a mean presenteeism score (i.e., productivity loss severity) of 7.1 ± 1.6 (out of 10) (Table S1). Overall, one in three had diabetes‐related distress (CDSS‐15 ≥ 45).

Among these participants, 23 (5.0%) reported presenteeism score ≤ 40 (range 0–100). The high‐presenteeism group reported longer working hours of 55.4 ± 14.8 per week, a higher rate of sickness attendance in the past year (91.3%), poorer HRQoL, higher stress, higher regimen‐and‐social‐support‐related distress and poorer dietary adherence than the low‐presenteeism group (Table 2).

TABLE 2Demographic characteristics of employed participants in the substudy of the PRISM‐RCT stratified by high‐ versus low‐presenteeism.

Frequency (%)/mean ± SD/median (interquartile range) p

N Low‐presenteeism (N = 438) N High‐presenteeism (N = 23)

Demographic characteristics

Age (years) 438 43.4 ± 6.4 23 42.9 ± 6.4 0.678

Men (n, %) 438 259 (59.1) 23 17 (73.9) 0.233

College or above (n, %) 438 170 (38.8) 23 10 (43.5) 0.820

Taking care of children < 18‐year (n, %) 438 145 (33.1) 23 9 (39.1) 0.711

Current or former tobacco use (n, %) 438 143 (32.6) 23 8 (34.8) 1.000

Never drank alcohol (n, %) 438 254 (58.0) 23 10 (43.5) 0.248

Family history of diabetes (n, %) 432 344 (79.6) 22 19 (86.4) 0.619

Metabolic risk factors

Age at diagnosis (years) 438 33.0 ± 6.4 23 33.0 ± 6.4 0.993

Duration of diabetes (years) 438 10.9 ± 6.0 23 10.4 ± 5.7 0.706

Fasting plasma glucose (mmol/L) 436 7.4 ± 2.4 23 7.4 ± 1.6 0.992

HbA1c (%) 436 7.3 ± 1.5 23 7.3 ± 1.0 0.983

HbA1c < 7% (n, %) 436 208 (47.7) 23 14 (60.9) 0.309

Body mass index (kg/m2) 435 28.1 ± 5.3 23 28.0 ± 3.6 0.953

Waist circumference (men, cm) 203 100.0 ± 12.4 12 96.8 ± 7.4 0.390

Waist circumference (women, cm) 146 93.5 ± 12.9 5 103.2 ± 6.9 0.094

Systolic BP (mmHg) 435 125.6 ± 15.1 23 121.9 ± 13.2 0.241

Diastolic BP (mmHg) 435 75.9 ± 10.3 23 74.1 ± 11.5 0.420

LDL‐cholesterol (mmol/L) 426 2.1 ± 0.8 23 2.0 ± 0.9 0.663

Triglycerides (mmol/L) 436 1.3 (0.9, 2.0) 23 1.6 (1.0, 2.0) 0.460

eGFR (mL/min/1.73 m2) 436 104.6 (89.1, 111.3) 22 104.7 (93.7, 110.2) 0.953

Albuminuria (n, %) 438 148 (33.8) 23 9 (39.1) 0.763

Diabetes self‐management and hypoglycaemia

Self‐reported hypoglycaemia in last 3 months (n, %) 438 109 (24.9) 23 3 (13.0) 0.298

Self‐monitoring of blood glucose/week (n, %) 438 146 (33.3) 23 7 (30.4) 0.952

Physical activity ≥ 5 times/week (n, %) 438 84 (23.9) 23 4 (25.0) 1.000

Good adherence to balanced diet (n, %) 438 98 (22.4) 23 2 (8.7) 0.196

Medication

Oral glucose‐lowering drugs (n, %) 438 412 (94.1) 23 22 (95.7) 1.000

Insulin (n, %) 438 124 (28.3) 23 3 (13.0) 0.174

GLP‐1 RA (n, %) 433 113 (26.1) 21 5 (23.8) 1.000

Blood pressure‐lowering drugs (n, %) 438 343 (78.3) 23 19 (82.6) 0.819

Renin angiotensin system inhibitor (n, %) 430 310 (72.1) 21 15 (71.4) 1.000

Lipid‐regulating drugs (n, %) 438 360 (82.2) 23 22 (95.7) 0.166

Statin (n, %) 430 344 (80.0) 20 17 (85.0) 0.794

History of complications and comorbidities

Cardiovascular disease (n, %) 438 44 (10.0) 23 0 (0.0) 0.217

Coronary heart disease (n, %) 438 29 (6.6) 23 0 (0.0) 0.404

Stroke (n, %) 438 11 (2.5) 23 0 (0.0) 0.945

Peripheral vascular disease (n, %) 438 6 (1.4) 23 0 (0.0) 1.000

Congestive heart failure (n, %) 438 2 (0.5) 23 0 (0.0) 1.000

Chronic kidney disease (n, %) 438 31 (7.1) 23 1 (4.3) 0.935

End‐stage kidney disease (n, %) 438 4 (0.9) 23 1 (4.3) 0.605

All‐site cancer (n, %) 438 15 (3.4) 23 1 (4.3) 1.000

Presenteeism

Working hours per week (hours) 435 48.6 ± 13.7 23 55.4 ± 14.8 0.023

Presenteeism score (0–10)a
438 7.3 ± 1.3 23 3.4 ± 0.9 < 0.001

Patient‐reported outcomes

EQ‐5D‐3L utility score 429 0.9 ± 0.1 23 0.8 ± 0.2 0.001

EQ‐5D‐3L VAS (0–100) 433 69.3 ± 16.8 23 64.0 ± 13.2 0.136

DASS‐21

Depression (0–21) 432 1.5 (0.0, 5.0) 23 2.0 (1.5, 6.0) 0.056

Anxiety (0–21) 433 2.0 (1.0, 5.0) 23 3.0 (2.0, 5.5) 0.058

Stress (0–21) 433 4.0 (0.0, 8.0) 23 6.0 (2.0, 8.5) 0.033

C‐DDS‐15

Emotional burden (6–36) 431 15.9 ± 6.5 23 18.1 ± 6.4 0.112

Regimen‐ and social‐related distress (6–36) 431 16.8 ± 5.8 23 19.6 ± 5.7 0.025

Physical‐related distress (3–18) 430 5.8 ± 2.7 23 5.3 ± 2.1 0.399

C‐DDS‐15 ≥ 45 428 125 (29.2) 23 9 (39.1) 0.435

SDSCA‐15

Diet (0–7) 433 3.5 (2.8, 4.5) 23 3.0 (2.0, 3.5) 0.004

Exercise (0–7) 433 2.0 (1.0, 3.5) 23 1.5 (0.5, 2.5) 0.101

Blood sugar testing (0–7) 431 1.0 (0.0, 3.0) 23 1.0 (0.0, 2.0) 0.818

Foot care (0–7) 433 3.3 (2.3, 5.0) 23 3.0 (2.3, 3.8) 0.116

Adherence to medical advice (0–7) 174 4.7 (2.3, 6.3) 6 3.8 (2.8, 4.7) 0.511

CQ‐4 Total (0–4) 427 4.0 (3.0, 4.0) 23 3.0 (3.0, 4.0) 0.165

Note: p‐values less than 0.05 were shown in bold.

Abbreviations: C‐DDS‐15, Chinese version of Diabetes Distress Scale—15 items; CQ‐4, Compliance Questionnaire—4 items; DASS‐21, Depression, Anxiety and Stress Scales—21 items; EQ‐5D‐3L, EuroQol 5 Dimension 3 Level Version; SDSCA‐15, Summary of Diabetes Self‐Care Activities—15 items; VAS, Visual Analogue Scale.

p‐values indicate the comparison between the high‐ and low‐presenteeism groups.

In the multivariate analysis including all PROMs, higher HRQoL and lower emotional distress (CDDS‐15) were independently associated with lower presenteeism. These associations remained significant after adjusting for demographics, cardiometabolic risks and medications (Table 3). Consistent findings were observed in separate models examining each PROM in relation to presenteeism (Table S2).

TABLE 3Multivariable linear regression analysis of the associations of presenteeism with psychological health and self‐care activities.

Model‐1 Model‐2 Model‐3

β SE 95% CI VIF p β SE 95% CI VIF p β SE 95% CI VIF p

DASS‐21

Depression −0.06 0.07 −0.19, 0.08 5.82 0.392 −0.08 0.08 −0.23, 0.07 6.55 0.288 −0.05 0.08 −0.21, 0.10 6.10 0.501

Anxiety 0.01 0.06 −0.11, 0.13 3.55 0.893 0.05 0.07 −0.09, 0.18 4.24 0.507 0.06 0.07 −0.08, 0.20 4.08 0.393

Stress 0.06 0.05 −0.04, 0.16 4.71 0.276 0.06 0.05 −0.05, 0.17 5.06 0.264 0.03 0.06 −0.08, 0.15 5.54 0.578

C‐DDS‐15

Emotional burden −0.06 0.03 −0.12, 0.00 3.12 0.056 −0.08 0.03 −0.14, −0.01 3.41 0.019 −0.08 0.04 −0.15, −0.01 3.83 0.029

Regimen‐and‐social‐related distress 0.04 0.03 −0.03, 0.10 2.84 0.298 0.04 0.04 −0.04, 0.11 3.04 0.334 0.05 0.04 −0.03, 0.13 3.36 0.190

Physician‐related distress 0.05 0.05 −0.05, 0.14 1.43 0.330 0.04 0.05 −0.06, 0.13 1.39 0.480 0.05 0.05 −0.05, 0.15 1.38 0.339

EQ‐5D‐3L utility score 3.54 1.27 1.03, 6.04 2.36 0.006 3.66 1.34 1.02, 6.31 2.49 0.007 3.30 1.60 0.14, 6.46 2.61 0.041

SDSCA

Diet 0.11 0.11 −0.12, 0.33 1.75 0.342 0.12 0.12 −0.12, 0.36 1.92 0.331 0.18 0.13 −0.08, 0.44 2.15 0.177

Exercises −0.04 0.08 −0.20, 0.11 1.38 0.560 −0.08 0.08 −0.24, 0.09 1.53 0.351 −0.11 0.09 −0.28, 0.06 1.64 0.188

Blood sugar testing −0.08 0.07 −0.22, 0.06 1.50 0.256 −0.11 0.08 −0.26, 0.04 1.58 0.144 −0.13 0.08 −0.29, 0.03 1.79 0.111

Foot care 0.22 0.08 0.07, 0.37 1.41 0.004 0.21 0.08 0.04, 0.37 1.62 0.013 0.16 0.09 −0.01, 0.34 1.79 0.063

Medication adherence 0.04 0.07 −0.09, 0.17 1.49 0.574 0.01 0.07 −0.13, 0.16 1.70 0.851 0.03 0.09 −0.14, 0.21 2.29 0.713

CQ‐4 −0.02 0.11 −0.25, 0.20 1.53 0.836 −0.06 0.13 −0.31, 0.19 1.85 0.632 0.00 0.14 −0.27, 0.27 2.03 0.998

Note: Presenteeism, “How would you rate your overall job performance on the days you worked during the past 4 weeks (28 days)?” from World Health Organization‐Health and Work Performance Questionnaire (WHO‐HPQ); Model‐1: Demographics (age, sex, duration of diabetes, college or above, taking care of child, group assignment); Model‐2: Model‐1 + Cardiometabolic risk factors (BMI, HDL‐C, LDL‐C, SBP, DBP, TG, eGFR, HbA1c, fasting plasma glucose); Model‐3: Model‐2 + Medications (antidiabetic drugs, insulin, GLP‐1, antihypertensive drugs, RASi, hyperlipidaemia agents, statin) were adjusted. Regression was conducted with all the PROMs (i.e., DASS‐21, C‐DDS‐15, EQ‐5D‐3L utility score, SDSCA and CQ‐4) entered to the model simultaneously. p‐values indicate the association between study variables and presenteeism, p‐values less than 0.05 were shown in bold.

Abbreviations: C‐DDS‐15, Chinese version of Diabetes Distress Scale—15 items; CQ‐4, Compliance Questionnaire—4 items; DASS‐21, Depression, Anxiety and Stress Scales—21 items; EQ‐5D‐3L, EuroQol 5 Dimension 3 Level Version; SDSCA‐15, Summary of Diabetes Self‐Care Activities—15 items; VIF, variance inflation factor.

3.2Qualitative Findings

Of the 465 employees invited, 115 (24.7%) agreed to participate in FGDs, of whom 16 were purposively selected and assigned to four FGDs (mean age: 45.1 ± 5.9 years, age‐at‐diagnosis: 33.6 ± 6.8 years, diabetes duration: 2.1 ± 7.1 years). The clinical characteristics of FGD participants and non‐participants had comparable working profiles and presenteeism score (7.1 ± 1.6 vs. 7.1 ± 1.5 respectively) (Table S3). We performed comparative analysis between the employed and unemployed, high‐ and low‐presenteeism and FGD and non‐FGD groups using p‐values with false discovery rate (FDR) adjusted for multiple comparisons (Table S4).

The qualitative analysis generated 15 codes, which were organised into two central categories: contributing factors and healthcare policies, each with three sub‐themes. Thematic prevalence table and example quotes under each subtheme were included in Tables 4 and S5, respectively.

3.2.1Contributing Factors

The contributing factors included personal factors, workplace environment and societal influences (Figure 2a).

TABLE 4Prevalence of themes identified from focus group discussions.

Theme Number of participants (n = 16) Frequency (%) Prevalence categorya

Contributing factors

Personal factors

Cultural collectivism 5 (31.3) Common

Psychological and physical strain 7 (43.8) Common

Working environment

Concerns about the attitude of employers and colleagues 11 (68.8) Majority

Frequent work adjustment 9 (56.3) Common

Challenges in applying for leave 11 (68.8) Majority

Work types and schedules 8 (50.0) Common

Societal influences

Relationships with care providers 7 (43.8) Common

Medical costs 10 (62.5) Majority

Societal perception 3 (18.8) Minority

Healthcare policies

Prevention

Educational programme 7 (43.8) Common

Healthy lifestyle promotion 3 (18.8) Minority

Early detection for family members of patients with YOD

Genetic screening and regular body check 4 (25.0) Minority

Ongoing support

Patient involvement in shared decision‐making 3 (18.8) Minority

Conditional insurance coverage 8 (50.0) Common

Peer support group 3 (18.8) Minority

aMajority theme = ≥ 60% of participants; Common theme = 35%–59%; minority theme = < 35%.

3.2.1.1Personal Factors

3.2.1.1.1Family Collectivism

Commensality and shared meals are norms in many Asian families. Waiting for other family members to return from work often delayed meals. Sharing meals with children and older relatives without chronic medical conditions posed barriers to healthy food choices. Desire to encourage a nutrient‐balanced diet within the households could be met by resistance from other family members. Caregiving responsibilities could increase stress and fatigue, which in turn impeded effective health management (respondent‐10‐FGD‐2). Family reciprocity and support, especially spouses with similar medical conditions, reinforced adherence to self‐monitoring and medication (respondent‐14‐FGD‐4).

3.2.1.1.2Psychological and Physical Strain

Work‐related stress contributed to binge eating or increased food cravings with high content of sugar or sodium glutamate (respondent‐15‐FGD‐4; respondent‐16‐FGD‐4). In Hong Kong, eating is a convenient way to relieve stress (respondent‐11‐FGD‐3), posing challenges in adhering to a balanced diet despite having the appropriate knowledge (respondent‐4‐FGD‐1).

3.2.1.2Working Environment

3.2.1.2.1Concerns About the Attitude of Employers and Colleagues

Colleagues were perceived to view diabetes as a reflection of ineffective self‐management and vulnerability. The was in contrast to greater societal acceptance for hypertension with more stigmatisation for diabetes, obesity and associated treatments (e.g., injections) (respondent‐14‐FGD‐2). Disclosure of diabetes diagnosis might negatively influence employers' and colleagues' perceptions of competence, performance evaluation and career advancement (respondent‐4‐FGD‐1; respondent‐7‐FGD‐2). Participants experienced ambivalence when faced with decision of declining meal invitations or refusing certain foods during significant team events (respondent‐6‐FGD‐2). Supporting workplace cultures (respondent‐16‐FGD‐4) and health‐literate colleagues facilitated acceptance, often with friendly reminders on food choices/portions and health tips (respondent‐4‐FGD‐1).

3.2.1.2.2Frequent Work Adjustment

Regular medical visits (e.g., half‐day hospital visits every 3 months), biennial complication assessments and education sessions disrupted work schedule requiring temporary staff substitution (respondent‐4‐FGD‐1). These challenges sometimes prompted the acceptance of a lower‐intensity position with reduced earnings (respondent‐7‐FGD‐2).

3.2.1.2.3Challenges in Applying for Leave

Some supervisors only accepted sick leave certificates issued by a doctor, posing challenges for assessments or education by non‐medical staff, who could only issue certificates of attendance in Hong Kong public healthcare settings (respondent‐5‐FGD‐1).

3.2.1.2.4Work Types and Schedules

Shift‐work hindered adherence to a healthy lifestyle due to irregular mealtimes (respondent‐15‐FGD‐4) and late working hours. These non‐routine working hours reduced frequencies of home cooking with frequent dining out with unbalanced diets (respondent‐1‐FGD‐1; respondent‐9‐FGD‐2). These challenges were often underappreciated by clinicians (respondent‐1‐FGD‐1).

3.2.1.3Societal Influences

3.2.1.3.1Relationships With Care Providers

Many respondents were knowledgeable about advanced diabetes technologies, such as continuous glucose monitoring (CGM) devices and used them to support self‐management (respondent‐9‐FGD‐2). These efforts were not fully recognised by doctors who preferred using medications to achieve control (respondent‐4‐FGD‐1). Limited personalisation, brief consultations and fragmented care in public settings reduced perceived usefulness of advice, contributing to disengagement and mistrust (respondent‐14‐FGD‐4, respondent‐6‐FGD‐2).

3.2.1.3.2Medical Costs

High costs of innovative medications (e.g., GLP1‐RA) and technologies (e.g., CGM) were major barriers. In Hong Kong. In public hospitals/clinics, coverage for GLP1‐RA subsidies was restricted to patients with obesity who had complications or multiple risk factors. Many young patients with overweight but without complications had to pay out of pocket with frequent discontinuation (respondent‐6‐FGD‐2, respondent‐14‐FGD‐4). Lack of or insufficient insurance coverage for diabetes‐related expenses, disqualification due to the ‘inherited’ nature of diabetes (respondent‐5‐FGD‐1) and additional premium loadings (respondent‐4‐FGD‐1) further constrained access.

3.2.1.3.3Societal Perception

Limited health literacy of the public on YOD, misconceptions that diabetes is a disease of the elderly and that young age is a disease‐free state hindered discussion between individuals with YOD and those without diabetes (respondent‐15‐FGD‐4).

3.2.2Healthcare Policies

Several potential enablers proposed by the participants were listed in the healthcare policies category, including prevention, early detection and personalised treatment (Figure 2b).

3.2.2.1Prevention

3.2.2.1.1Educational Programme

Preventive efforts should incorporate early public education on diabetes, targeting children and their parents to increase awareness and promote healthy behaviours from a young age. Curricula should include explanation on diabetes subtypes, healthy lifestyle practices, correction of misconceptions that diabetes is a ‘geriatric disease’ and de‐stigmatisation of diabetes‐related practices (e.g., injections, CGM, self‐monitoring of blood glucose) (respondent‐11‐FGD‐3). Early life education that promotes regular meals, physical activity and healthy nutrition (respondent‐16‐FGD‐4) was viewed as a long‐term strategy to improve individual and societal well‐being (respondent‐3‐FGD‐1).

3.2.2.1.2Healthy Lifestyle Promotion

Other suggestions included government policies aimed at encouraging restaurants and caterers to offer healthier meal choices, especially for individuals with limited cooking time (respondent‐12‐FGD‐3).

3.2.2.2Early Detection for Family Members of Patients With YOD

3.2.2.2.1Genetic Screening and Regular Body Check

Genetic screening and regular health check‐ups for complications were suggested to enable early detection, raise awareness and motivate healthier lifestyles (respondent‐11‐FGD‐3). Addressing concerns about inherited risk and improving access to genetic testing with clear action plans for offspring to mitigate risk were highlighted (respondent‐13‐FGD‐3).

3.2.2.3Ongoing Support

3.2.2.3.1Patient Involvement in Shared Decision‐Making

Insufficient explanation by physicians regarding medication options, benefits and risks hindered informed decision‐making. Clearer personalised information could increase willingness to pay out‐of‐pocket for organ‐protective therapies as a long‐term health investment (respondent‐11‐FGD‐3; respondent‐12‐FGD‐3).

3.2.2.3.2Conditional Insurance Coverage

Conditional insurance incentives for sustained glycaemic control (HbA1c < 7%) were suggested to encourage self‐care and reduce financial burden (respondent‐8‐FGD‐2).

3.2.2.3.3Peer Support Group

Feelings of loneliness and isolation might be relieved through peer knowledge‐sharing and learning on self‐care (respondent‐7‐FGD‐2), preferably via online platforms or chat groups that do not require immediate response due to different work schedules (respondent‐9‐FGD‐2).

3.3Mixed‐Methods Integration

Table 1 presents a joint display diagram with meta‐inferences. Quantitative analysis on diabetes‐related emotional burden and HRQoL with presenteeism converged with qualitative themes. The latter included challenges in submitting leave applications, psychological and physical strain, financial burden of medical care and variation in work types and schedules. Inflexible company policies and demanding schedules were identified as barriers to attending medical follow‐ups and contributors to work–health conflicts.

Concerns over employers' and colleagues' attitudes, alongside societal perceptions, influenced decisions related to presenteeism. Financial strain on medical care and patient‐provider relationships underlie the heightened regimen‐and‐social support‐related distress among participants with high presenteeism. Cultural norms emphasising collectivism also appeared to influence self‐care practices.

4Discussion

In this study, we applied a convergent parallel mixed‐methods design, utilising validated questionnaires to explore the associations of psychological distress, HRQoL and self‐care activities with presenteeism. This was complemented by semi‐structured interviews to elucidate underlying barriers and enablers.

4.1Vicious Cycle of Work, Stress and Collectivism Challenging Self‐Care Adherence

Work stress and caregiving responsibilities were weighted heavily by our employed interviewees. Both high‐ and low‐presenteeism groups had long working hours (48–55 h/week). In Hong Kong, these long working hours had been associated with cardiovascular diseases [30]. Consistent with prior evidence, working hours ≥ 48/week were accompanied by frequent fatigue and exhaustion, especially for those with childcare responsibilities. Some participants reported binge eating as a stress‐coping strategy, a behaviour linked to obesity [31] and depression [32, 33].

In China, the prevalence of overweight/obesity had risen from 3.1% in 2004 to 16.4% in 2023 [34]. Our qualitative findings suggest a potential sequential pathway comprising work‐related stress precipitating emotional eating with consequent weight gain, deterioration in glycaemic control and heightened presenteeism. Other researchers had reported that obesity and stress‐related eating behaviours were associated with nearly a twofold increase in workplace presenteeism. Excess adiposity may exacerbate insulin resistance and impair β‐cell function [35] with ineffective energy metabolism which may diminish work capacity and performance [36]. This stress–eating–weight cycle warrants rigorous investigation, as it may perpetuate a self‐reinforcing pattern of weight gain, psychological distress, suboptimal health and increased presenteeism.

Work‐health conflicts arose when occupational demands impeded diabetes self‐management [37]. Within Chinese collectivist culture, shared meals are central to family and workplace relationships [38, 39]. Finishing a meal is etiquette, while offering food expresses hospitality [40]. Consistent with prior reports [41], our participants reported pressure and embarrassment with food choices as a major barrier to dietary adherence. Work schedules also exerted complex effects. Shift work may disrupt circadian rhythms [42] although in our prior analysis, this was linked with lower presenteeism, possibly reflecting greater flexibility and reduced sick‐leave‐related stress [10]. Participants also cited Hong Kong‐specific administrative barriers where non‐acceptance of sick‐leave certificates issued by non‐physician care providers might limit access to multidisciplinary care. These constraints may coincide with frequent defaults for educational programmes and suboptimal diabetes control [37]. These practical barriers highlight the sociocultural work‐health conflicts and system‐level gaps calling for flexible policies, such as acceptance of leave for educational or assessment sessions in addition to medical consultations.

Diabetes‐related emotional distress and HRQoL remained significantly associated with presenteeism after covariate adjustment. These findings suggest that diabetes‐specific psychological factors may affect work performance, independent of physical illness burden and mechanistically, may be mediated by a pathway distinct from cardiometabolic indices. In keeping with the negative emotions associated with pill burden, depressive symptoms were attenuated after adjustment for medication use [43]. Despite the high prevalence of stress among those with high presenteeism, there was no clear relationship between stress and depression. These dissociative patterns may reflect differences in how condition‐specific versus general mood constructs may have different effects on work functioning.

Overall, work‐ and diabetes‐specific demands may represent more proximal influences on presenteeism than general mood symptoms in employees with YOD. This interpretation aligns with the notion that the day‐to‐day burden of diabetes management, including sustained behavioural adaptation, self‐monitoring and treatment adherence, may exert more immediate effects on occupational performance than generalised psychological distress.

Our participants identified peer support as a potentially valuable strategy. Peer‐based interventions have been associated with improvements in cardiometabolic outcomes, self‐care behaviours, emotional well‐being and healthcare utilisation [44, 45, 46]. Future research should evaluate such interventions in our workforce with diabetes using larger, representative samples with real‐time assessment and prospective follow up for clinical outcomes. This holistic approach incorporating biomedical‐cognitive‐psychological‐behavioural intervention will better characterise the temporal and causal relationships among psychological burden, work functioning, clinical outcomes and societal productivity.

4.2Health Literacy and Diabetes Stigmatisation in Presenteeism

Several participants noted a lay misconception that diabetes exclusively affects older individuals. Stigma thrives in power‐laden social contexts [47, 48], such as workplace, where hierarchies, scrutiny and professional consequences converge. Diabetes‐related attributes, notably obesity and visible self‐care activities (e.g., fingerpricks, insulin injections), are often misattributed to personal failings [48, 49]. These negative perceptions foster ‘ingroup’ versus ‘outgroup’ comparisons, creating automatic stereotype that culminate status loss and discrimination [47, 48]. Experiences of felt stigma (i.e., shame and anticipation of judgement) [50] coincided with presenteeism as reported by the participants. These included concealment of their condition, avoidance of taking sick leave for fear of jeopardising job security and embarrassment when declining food at social occasions. These complex internal and external experiences can adversely affect physical and mental health as well as productivity, with particular relevance to YOD [51]. Similarly, low public awareness of diabetes risk factors, subtypes, pathophysiology and treatment has been documented in a Singapore study [52]. Enhancing public awareness that diabetes can affect individuals of any age, body weight or ethnicity may support prevention and reduce stigma, as echoed by our participants.

4.3Care Gaps, Patient Autonomy, Trust and Financial Barriers

High regimen‐and‐social‐support‐related distress was observed in the high‐presenteeism group with insufficient patient‐provider engagement and variations in health knowledge. Consistent with prior findings [53], some respondents recognised the potential of diabetes technologies to improve health behaviours but perceived insufficient clinical support to utilise them effectively. This misalignment may undermine patient‐provider engagement and trust. Genetic testing was raised during focus group discussions. Some participants suggested that such information could enhance health awareness and support preventive efforts. In the PRISM‐RCT study where we recruited this sub‐cohort, genetic testing was associated with improvements in metabolic control, β‐cell function indices and patient‐provider communications [17]. Shifting from paternalistic doctor–patient relationships towards promoting health literacy and patient autonomy could enhance HRQoL and positive behaviour changes [54] which may reduce presenteeism. The growing trend towards a collaborative model [55] demands healthcare system reform to empower patients' decision‐making. This approach is particularly relevant to YOD, where complex and fragmented care prevail. These achievable factors, perceived disease autonomy and empowerment, may mitigate presenteeism.

Financial barriers can lead to suboptimal care with direct and indirect costs due to poor health outcomes and continuing presenteeism. Studies estimating healthcare costs [16, 56, 57] reported diabetes‐related costs to be 2.6 times higher than non‐diabetes conditions, excluding hidden self‐care costs [57]. Some technologies (e.g., CGM, insulin pumps) and newer glucose lowering drugs (e.g., SGLT‐2, GLP‐1) were considered cost‐effective with reduced acute hyper/hypoglycaemic episodes and hospitalisations [58, 59, 60, 61]. In Hong Kong, many patients have to pay out‐of‐pocket for these technologies and therapeutics due to insufficient insurance coverage [53]. Indeed, in many jurisdictions, diabetes is excluded from private insurance or covered with high premium. Our younger employees yearned for affordable insurance for better access to specialists, technologies and therapeutics. A few suggested conditional insurance schemes (e.g., linked to HbA1c targets) as a potential model to reduce insurer risk while incentivising adherence. In the United States, uninsured adults with diabetes receive fewer recommended services and delayed care [62, 63]. Reducing financial strain may reduce cognitive bandwidth, enabling individuals to focus on proactive health behaviours. Whether these additional financial support would improve health outcomes and reduce presenteeism warrants further investigations.

4.4Strengths and Limitations

Notable strengths include the involvement of a multi‐disciplinary expert panel (psychologist, health economist, public health specialist and clinicians) and the utilisation of methodological triangulation (integration of quantitative and qualitative data), which enhanced the credibility and validity of the findings. The high questionnaire completion rate (94%) supports the reliability of the survey data [10]. Qualitative analysis was conducted by trained researchers following Braun and Clarke's (2006) six‐step framework [29], with reporting aligned to the Consolidated Criteria for Reporting Qualitative Research (Table S6) to ensure methodological rigour. Participants exhibited diverse clinical characteristics, minimising selection bias and improving generalizability. Neither interviewer had clinical or personal ties to respondents and confidentiality reassurances promoted open discussion.

This study is not without limitations. Hospital‐based recruitment may over‐sample participants with complex disease than those from primary care. The qualitative subsample demonstrated higher distress and lower HRQoL than the broader YOD cohort. This may be attributable to maximum variation sampling that balanced participants by complications and insulin use, shifting the group averages in a generally low‐burden sample. As distress and perceived barriers were central outcomes, findings reflect more burdened individuals rather than being representative of the entire YOD population. Qualitative insights are hypothesis‐generating and intended to inform the development of systematic measurement approaches in future studies involving larger, representative populations. The single‐item WHO‐HPQ may be prone to social desirability bias. However, it remains a validated, widely used and pragmatic measure of presenteeism in clinical research [19]. The effect size between emotional burden and presenteeism was only modest, calling for careful interpretation. However, in a population with high prevalence of YOD and distress, the cumulative impact of emotional strain on productivity can become significant.

5Conclusion

Among Chinese individuals with YOD, long working hours, inflexible workplace policies, entrenched presenteeism norms and insurance exclusions create a challenging context for optimal care. One in three participants experienced diabetes distress, which was independently associated with higher presenteeism. There are multiple personal, family, occupational and societal barriers for having access to optimal care. These were compounded by stigma, cultural expectations and suboptimal patient–provider communication. Participants proposed several mitigation strategies, including strengthening public awareness through early education, promoting earlier detection (e.g., through genetic testing where appropriate) and supporting informed decision‐making. Collectively, these findings underscore the need for a patient‐centred, multidimensional and multi‐stakeholder approach to address unmet needs and improve health and work‐related outcomes. The policy and intervention strategies suggested by participants warrant further evaluation in larger, representative samples.

Funding

The study was supported by the Health and Medical Research Fund Commissioned Research Grant (CFS‐CUHK2).

Ethics Statement

This study was approved by the Chinese University of Hong Kong—New Territories East Cluster Clinical Research Ethics Committee (reference number: 2019‐080‐T) and complied with the Declaration of Helsinki. Participants gave informed consent to participate in the study before taking part.

Conflicts of Interest

J.C.N.C. reported receiving grants (through institutions) and/or honoraria for providing consultancy or giving lectures from Abbott, Applied Therapeutics, AstraZeneca, Bayer, Biomea, Boehringer Ingelheim, Eli Lilly, Hua Medicine, Merck, Novo Nordisk, Powder Pharmaceuticals, Roche, Sanofi, Servier and Zuellig Pharma Commercialisations. She is the Chief Executive Officer (pro bono) of the Asia Diabetes Foundation that developed the web‐based JADE platform for implementation of data‐driven diabetes management. She is a co‐inventor of patents owned by the Chinese University of Hong Kong with claims of using biomarkers and mathematical models to predict diabetes and its complications. J.C.N.C. and R.C.W.M. are co‐founders of GemVCare, a biotechnology company, with partial support from the Hong Kong Government, which uses biogenetic markers and information technology to implement precision diabetes care and prevention through partnerships. J.N.M.L. received research funding from Roche Diagnostics and Johnson and Johnson Ltd. for projects that are unrelated to the work presented in this manuscript. A.P.S.K. has received research grants and/or speaker honoraria from Abbott, Astra Zene ca, Bayer, Biomea, Boehringer Ingelheim, Dexcom, Eli‐Lilly, Kyowa Kirin, Merck Serono, Merck Sharp & Dohme, Nestle, Novo‐Nordisk, Pfizer, Sanofi and Zuellig Pharma.

Supporting information

References

1J. C. Chan , C. K. O , and A. O. Luk , “Young‐Onset Diabetes in East Asians: From Epidemiology to Precision Medicine ,” Endocrinology and Metabolism
39 , no. 2 (2024 ): 239 –254 .10.3803/EnM.2024.1968 PMC11066447 38626908

2Y. Xu , J. Lu , M. Li , et al., “Diabetes in China Part 1: Epidemiology and Risk Factors ,” Lancet Public Health
9 , no. 12 (2024 ): e1089‐1097.10.1016/S2468-2667(24)00250-0 39579774

3The Lancet Diabetes & Endocrinology
, “Alarming Rise in Young‐Onset Type 2 Diabetes ,” Lancet Diabetes & Endocrinology
12 , no. 7 (2024 ): 433 .10.1016/S2213-8587(24)00161-X 38852596

4A. P. Kong , T. Lew , E. S. Lau , et al., “Real‐World Data Reveal Unmet Clinical Needs in Insulin Treatment in Asian People With Type 2 Diabetes: The Joint Asia Diabetes Evaluation (JADE) Register ,” Diabetes, Obesity & Metabolism
22 , no. 4 (2020 ): 669 –679 .10.1111/dom.13950 PMC7540442 31903728

5M. M. Barker , M. J. Davies , F. Zaccardi , et al., “Age at Diagnosis of Type 2 Diabetes and Depressive Symptoms, Diabetes‐Specific Distress, and Self‐Compassion ,” Diabetes Care
46 , no. 3 (2023 ): 579 –586 .10.2337/dc22-1237 PMC10020022 36630531

6S. Misra , N. Holman , E. Barron , et al., “Characteristics and Care of Young People With Type 2 Diabetes Included in the National Diabetes Audit Datasets for England ,” Diabetic Medicine
40 , no. 1 (2023 ): e14940.10.1111/dme.14940 PMC10087129 36054265

7H. Wu , E. S. Lau , A. Yang , et al., “Young Age at Diabetes Diagnosis Amplifies the Effect of Diabetes Duration on Risk of Chronic Kidney Disease: A Prospective Cohort Study ,” Diabetologia
64 , no. 9 (2021 ): 1990 –2000 .10.1007/s00125-021-05494-4 34121143

8H. Wu , A. Yang , E. S. Lau , et al., “Age‐ and Sex‐Specific Hospital Bed‐Day Rates in People With and Without Type 2 Diabetes: A Territory‐Wide Population‐Based Cohort Study of 1.5 Million People in Hong Kong ,” PLoS Medicine
20 , no. 8 (2023 ): e1004261.10.1371/journal.pmed.1004261 PMC10403124 37540646

9Y.‐L. Yeung , K.‐L. Lee , E. S. Lau , et al., “Associations of Comorbid Depression With Cardiovascular‐Renal Events and All‐Cause Mortality Accounting for Patient Reported Outcomes in Individuals With Type 2 Diabetes: A 6‐Year Prospective Analysis of the Hong Kong Diabetes Register ,” Frontiers in Endocrinology
15 (2024 ): 1284799.10.3389/fendo.2024.1284799 PMC10999250 38586459

10J. N. Lui , K. T. Wong , E. S. Lau , et al., “Work Productivity, Associated Risk Factors and Costs on Presenteeism and Absenteeism in Chinese Patients With Young‐Onset Type‐2 Diabetes in Hong Kong ,” Diabetes, Obesity & Metabolism
28 (2026 ): 1730 –1740 .10.1111/dom.70352 PMC12890741 41402136

11H. Wu , H. Zhou , C. Huang , et al., “Identifying and Visualising Temporal Trajectories of Hospitalisations for Traditional and Non‐Traditional Complications in People With Type 2 Diabetes: A Population‐Based Study ,” Lancet Regional Health–Western Pacific
57 (2025 ): 101532.10.1016/j.lanwpc.2025.101532 PMC11999218 40236266

12S. Y. Jessica , T. Xu , R. A. James , W. Lu , and J. E. Hoffman , “Relationship Between Diabetes, Stress, and Self‐Management to Inform Chronic Disease Product Development: Retrospective Cross‐Sectional Study ,” JMIR Diabetes
5 , no. 4 (2020 ): e20888.10.2196/20888 PMC7787890 33355538

13J. J. Arnett , “New Horizons in Research on Emerging and Young Adulthood,” in Early Adulthood in a Family Context, National Symposium on Family Issues , vol. 2 , eds. Booth , A. , S. Brown , N. Landale , W. Manning , and S. McHale , (Springer, 2012 ), 231 –244 .

14D. Johnston , S. Harvey , N. Glozier , R. Calvo , H. Christensen , and M. Deady , “The Relationship Between Depression Symptoms, Absenteeism and Presenteeism ,” Journal of Affective Disorders
256 (2019 ): 536 –540 .10.1016/j.jad.2019.06.041 31280078

15A. Luk , S. H. Wild , S. Jones , et al., “Early‐Onset Type 2 Diabetes: The Next Major Diabetes Transition ,” Lancet
405 , no. 10497 (2025 ): 2313 –2326 .10.1016/S0140-6736(25)00830-X 40570862

16J. N. Lui , E. S. Lau , A. Q. Li , et al., “Temporal Incremental Healthcare Costs Associated With Complications in Hong Kong Chinese Patients With Type 2 Diabetes: A Prospective Study in Joint Asia Diabetes Evaluation (JADE) Register (2007–2019) ,” Diabetes Research and Clinical Practice
219 (2025 ): 111961.10.1016/j.diabres.2024.111961 39701541

17A. O. Luk , C. K. O , B. Fan , et al., “Precision Medicine to Redefine Insulin Secretion and Monogenic Diabetes (PRISM): A Randomized Controlled Trial in Chinese Patients With Young‐Onset Diabetes ,” Diabetes Care
49 , no. 7 (2026 ): 1202 –1212 .10.2337/dc26-0114 PMC13294787 42118608

18C. K. O , Y. N. Fan , B. Fan , et al., “Precision Medicine to Redefine Insulin Secretion and Monogenic Diabetes‐Randomized Controlled Trial (PRISM‐RCT) in Chinese Patients With Young‐Onset Diabetes: Design, Methods and Baseline Characteristics ,” BMJ Open Diabetes Research & Care
12 , no. 3 (2024 ): e004120.10.1136/bmjdrc-2024-004120 PMC11212116 38901858

19R. C. Kessler , C. Barber , A. Beck , et al., “The World Health Organization Health and Work Performance Questionnaire (HPQ) ,” Journal of Occupational and Environmental Medicine
45 , no. 2 (2003 ): 156 –174 .10.1097/01.jom.0000052967.43131.51 12625231

20T. Suzuki , K. Miyaki , Y. Sasaki , et al., “Optimal Cutoff Values of WHO‐HPQ Presenteeism Scores by ROC Analysis for Preventing Mental Sickness Absence in Japanese Prospective Cohort ,” PLoS One
9 , no. 10 (2014 ): e111191.10.1371/journal.pone.0111191 PMC4207778 25340520

21C. Zanon , R. E. Brenner , M. N. Baptista , et al., “Examining the Dimensionality, Reliability, and Invariance of the Depression, Anxiety, and Stress Scale–21 (DASS‐21) Across Eight Countries ,” Assessment
28 , no. 6 (2021 ): 1531 –1544 .10.1177/1073191119887449 31916468

22EuroQol Group
, “EuroQol—A New Facility for the Measurement of Health‐Related Quality of Life ,” Health Policy
16 , no. 3 (1990 ): 199 –208 .10.1016/0168-8510(90)90421-9 10109801

23G. G. Liu , H. Wu , M. Li , C. Gao , and N. Luo , “Chinese Time Trade‐Off Values for EQ‐5D Health States ,” Value in Health
17 , no. 5 (2014 ): 597 –604 .10.1016/j.jval.2014.05.007 25128053

24R. Z. Ting , H. Nan , M. W. Yu , et al., “Diabetes‐Related Distress and Physical and Psychological Health in Chinese Type 2 Diabetic Patients ,” Diabetes Care
34 , no. 5 (2011 ): 1094 –1096 .10.2337/dc10-1612 PMC3114496 21398526

25D. J. Toobert , S. E. Hampson , and R. E. Glasgow , “The Summary of Diabetes Self‐Care Activities Measure: Results From 7 Studies and a Revised Scale ,” Diabetes Care
23 , no. 7 (2000 ): 943 –950 .10.2337/diacare.23.7.943 10895844

26D. E. Morisky , L. W. Green , and D. M. Levine , “Concurrent and Predictive Validity of a Self‐Reported Measure of Medication Adherence ,” Medical Care
24 , no. 1 (1986 ): 67 –74 .10.1097/00005650-198601000-00007 3945130

27A. S. Levey , L. A. Stevens , C. H. Schmid , et al., “A New Equation to Estimate Glomerular Filtration Rate ,” Annals of Internal Medicine
150 , no. 9 (2009 ): 604 –612 .10.7326/0003-4819-150-9-200905050-00006 PMC2763564 19414839

28M. Hennink and B. N. Kaiser , “Sample Sizes for Saturation in Qualitative Research: A Systematic Review of Empirical Tests ,” Social Science & Medicine
292 (2022 ): 114523.10.1016/j.socscimed.2021.114523 34785096

29V. Braun and V. Clarke , “Using thematic analysis in psychology ,” Qualitative Research in Psychology
3 , no. 2 (2006 ): 77 –101 , 10.1191/1478088706qp063oa .

30M. Kivimäki , M. Jokela , S. T. Nyberg , et al., “Long Working Hours and Risk of Coronary Heart Disease and Stroke: A Systematic Review and Meta‐Analysis of Published and Unpublished Data for 603 838 Individuals ,” Lancet
386 , no. 10005 (2015 ): 1739 –1746 .10.1016/S0140-6736(15)60295-1 26298822

31S. ZelitchYanovski , “Binge Eating Disorder: Current Knowledge and Future Directions ,” Obesity Research
1 , no. 4 (1993 ): 306 –324 .10.1002/j.1550-8528.1993.tb00626.x 16350580

32G. W. Alpers and B. Tuschen‐Caffier , “Negative Feelings and the Desire to Eat in Bulimia Nervosa ,” Eating Behaviors
2 , no. 4 (2001 ): 339 –352 .10.1016/s1471-0153(01)00040-x 15001027

33R. H. Striegel , R. Bedrosian , and C. Wang , “Comparing Work Productivity in Obesity and Binge Eating ,” International Journal of Eating Disorders
45 , no. 8 (2012 ): 995 –998 .10.1002/eat.22069 23044632

34National Health Commission
, Report on Nutrition and Chronic Disease Status of Chinese Residents (People's Medical Publishing House, 2020 ).

35M. P. Czech , “Insulin Action and Resistance in Obesity and Type 2 Diabetes ,” Nature Medicine
23 , no. 7 (2017 ): 804 –814 .10.1038/nm.4350 PMC6048953 28697184

36D. M. Gates , P. Succop , B. J. Brehm , G. L. Gillespie , and B. D. Sommers , “Obesity and Presenteeism: The Impact of Body Mass Index on Workplace Productivity ,” Journal of Occupational and Environmental Medicine
50 , no. 1 (2008 ): 39 –45 .10.1097/JOM.0b013e31815d8db2 18188080

37A. G. Dugan , R. E. Decker , H. L. Austin , et al., “Work‐Health Conflict Among Breast Cancer Survivors: Associations With Cancer Self‐Management, Quality of Life, and Anticipated Turnover ,” Occupational Health Science
8 , no. 2 (2024 ): 315 –335 .

38H. C. Ho , M. Mui , A. Wan , et al., “Family Meal Practices and Well‐Being in Hong Kong: The Mediating Effect of Family Communication ,” Journal of Family Issues
39 , no. 16 (2018 ): 3835 –3856 .

39G. Ma , “Food, Eating Behavior, and Culture in Chinese Society ,” Journal of Ethnic Foods
2 , no. 4 (2015 ): 195 –199 .

40E. Cooper , “Chinese Table Manners: You Are How You Eat ,” Human Organization
45 , no. 2 (1986 ): 179 –184 .

41S. J. Wellard , S. Rennie , and R. King , “Perceptions of People With Type 2 Diabetes About Self‐Management and the Efficacy of Community Based Services ,” Contemporary Nurse
29 , no. 2 (2008 ): 218 –226 .10.5172/conu.673.29.2.218 18844535

42G. Costa , “Shift Work and Health: Current Problems and Preventive Actions ,” Safety and Health at Work
1 , no. 2 (2010 ): 112 –123 .10.5491/SHAW.2010.1.2.112 PMC3430894 22953171

43J. C. Chan , J. C. Mbanya , J. M. Chantelot , et al., “Patient‐Reported Outcomes and Treatment Adherence in Type 2 Diabetes Using Natural Language Processing: Wave 8 of the Observational International Diabetes Management Practices Study ,” Journal of Diabetes Investigation
15 , no. 9 (2024 ): 1306 –1316 .10.1111/jdi.14228 PMC11363103 38840439

44J. C. Chan , Y. Sui , B. Oldenburg , et al., “Effects of Telephone‐Based Peer Support in Patients With Type 2 Diabetes Mellitus Receiving Integrated Care: A Randomized Clinical Trial ,” JAMA Internal Medicine
174 , no. 6 (2014 ): 972 –981 .10.1001/jamainternmed.2014.655 24781960

45S. M. Smith , G. Paul , A. Kelly , D. L. Whitford , E. O'Shea , and T. O'Dowd , “Peer Support for Patients With Type 2 Diabetes: Cluster Randomised Controlled Trial ,” BMJ
342 (2011 ): d715.10.1136/bmj.d715 PMC3039437 21324992

46J. Yin , R. Wong , S. Au , et al., “Effects of Providing Peer Support on Diabetes Management in People With Type 2 Diabetes ,” Annals of Family Medicine
13 , no. Suppl 1 (2015 ): S42 –S49 .10.1370/afm.1853 PMC4648131 26304971

47B. G. Link and J. C. Phelan , “Conceptualizing Stigma ,” Annual Review of Sociology
27 (2001 ): 363 –385 .

48J. Schabert , J. L. Browne , K. Mosely , and J. Speight , “Social Stigma in Diabetes: A Framework to Understand a Growing Problem for an Increasing Epidemic ,” Patient‐Patient‐Centered Outcomes Research
6 , no. 1 (2013 ): 1 –10 .10.1007/s40271-012-0001-0 23322536

49B. Major and L. T. O'brien , “The Social Psychology of Stigma ,” Annual Review of Psychology
56 (2005 ): 393 –421 .10.1146/annurev.psych.56.091103.070137 15709941

50A. Jacoby , “Felt Versus Enacted Stigma: A Concept Revisited: Evidence From a Study of People With Epilepsy in Remission ,” Social Science & Medicine
38 , no. 2 (1994 ): 269 –274 .10.1016/0277-9536(94)90396-4 8140453

51S. Misra , K. Khunti , A. Goyal , et al., “Managing Early‐Onset Type 2 Diabetes in the Individual and at the Population Level ,” Lancet
405 (2025 ): 2341 –2354 .10.1016/S0140-6736(25)01067-0 40570864

52H. L. Wee , H. K. Ho , and S. C. Li , “Public Awareness of Diabetes Mellitus in Singapore ,” Singapore Medical Journal
43 , no. 3 (2002 ): 128 –134 .12005338

53T. Hall , M. K. Warman , T. Oser , et al., “Clinician‐Reported Barriers and Needs for Implementation of Continuous Glucose Monitoring ,” Journal of the American Board of Family Medicine
37 , no. 4 (2024 ): 671 –679 .10.3122/jabfm.2024.240049R1 39455273

54D. Cosme and E. T. Berkman , “Autonomy Can Support Affect Regulation During Illness and in Health ,” Journal of Health Psychology
25 , no. 1 (2020 ): 31 –37 .10.1177/1359105318787013 PMC6933086 30062919

55T. E. Quill and H. Brody , “Physician Recommendations and Patient Autonomy: Finding a Balance Between Physician Power and Patient Choice ,” Annals of Internal Medicine
125 , no. 9 (1996 ): 763 –769 .10.7326/0003-4819-125-9-199611010-00010 8929011

56T. P. Gilmer , P. J. O'Connor , W. A. Rush , et al., “Predictors of Health Care Costs in Adults With Diabetes ,” Diabetes Care
28 , no. 1 (2005 ): 59 –64 .10.2337/diacare.28.1.59 15616234

57E. D. Parker , J. Lin , T. Mahoney , et al., “Economic Costs of Diabetes in the U.S. in 2022 ,” Diabetes Care
47 , no. 1 (2023 ): 26 –43 .10.2337/dci23-0085 37909353

58J. G. Choi , A. N. Winn , M. R. Skandari , et al., “First‐Line Therapy for Type 2 Diabetes With Sodium–Glucose Cotransporter‐2 Inhibitors and Glucagon‐Like Peptide‐1 Receptor Agonists: A Cost‐Effectiveness Study ,” Annals of Internal Medicine
175 , no. 10 (2022 ): 1392 –1400 .10.7326/M21-2941 PMC10155215 36191315

59S. J. Fonda , C. Graham , J. Munakata , J. M. Powers , D. Price , and R. A. Vigersky , “The Cost‐Effectiveness of Real‐Time Continuous Glucose Monitoring (RT‐CGM) in Type 2 Diabetes ,” Journal of Diabetes Science and Technology
10 , no. 4 (2016 ): 898 –904 .10.1177/1932296816628547 PMC4928220 26843480

60R. Lin , F. Brown , S. James , J. Jones , and E. Ekinci , “Continuous Glucose Monitoring: A Review of the Evidence in Type 1 and 2 Diabetes Mellitus ,” Diabetic Medicine
38 , no. 5 (2021 ): e14528.10.1111/dme.14528 33496979

61I. P. Smith , C. L. Whichello , J. Veldwijk , et al., “Diabetes Patient Preferences for Glucose‐Monitoring Technologies: Results From a Discrete Choice Experiment in Poland and the Netherlands ,” BMJ Open Diabetes Research & Care
11 , no. 1 (2023 ): e003025.10.1136/bmjdrc-2022-003025 PMC9853131 36649973

62S. S. Casagrande , J. Park , W. H. Herman , and K. M. Bullard , “Health Insurance and Diabetes,” in Diabetes in America , eds. Lawrence JM , SS Casagrande , WH Herman , et al. (National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), 2023 ), https://www.ncbi.nlm.nih.gov/books/NBK597725/.38117923

63E. D. Doucette , J. Salas , J. Wang , and J. F. Scherrer , “Insurance Coverage and Diabetes Quality Indicators Among Patients With Diabetes in the US General Population ,” Primary Care Diabetes
11 , no. 6 (2017 ): 515 –521 .10.1016/j.pcd.2017.05.007 28619242