Background and aims: Epidemiological studies have shown that the incidence of type 1 diabetes is bimodal, peaking at puberty and again at age 40. Often this peak in mid-aged individuals is not appreciated because of the nearly 10-fold greater incidence of type 2 as compared to type 1 diabetes in this age group. Moreover, non-obese individuals aged 30-55 years presenting with hyperglycemia and relative insulin deficiency and controlled initially on oral agents are clinically distinct from older, obese, insulin resistant patients with type 2 diabetes. These clinical characteristics are also prognostic of Latent Autoimmune Diabetes in Adults (LADA), which typically exhibits with antibodies characteristic of type 1 diabetes including ICA and GADA and rapidly leads to insulin dependency. The purpose of our cross-sectional study was to evaluate the impact of insulin versus oral therapy on quality of life in subjects with characteristics predictive of LADA.

Materials and methods: We identified 163 individuals with typical characteristics of LADA (BMI < 30 kg/m², age = 30-55 yrs, c-peptide = 0.1-0.6 nmol/l) from a baseline database of 8 pooled clinical trials of type 1 and type 2 diabetes with a total of 3,727 participants. We compared patients using insulin only (INS) to those using oral agents only (OA) by examining Overall QOL (scaled 100 - 600) and its 10 psychological and physical subscales. Between-group differences were evaluated using linear regression models after adjusting for covariates of HbA1c, age, and duration of diabetes.

Results: Patients had a mean ± SD age = 46 ± 6 yrs, HbA1c = 8.9 ± 1.4%, BMI=25.9 ± 2.5 kg/m² and diabetes duration = 8.8 ± 6.7 yrs. 85% had a diagnosis of type 2 diabetes and 15% type 1 diabetes; 43% were on INS (2 or more injections per day) vs. 57% on OA or diet only. The INS group was younger (45 vs. 47 yrs), had lower HbA1c (8.0 vs. 9.6%), both P < 0.001, and longer duration of disease (11 vs. 7 yrs, P = 0.046) compared to the OA group. For both groups combined higher HbA1c was associated with worsened mental health (r = 0.16, p = 0.049), particularly psychological distress (r = 0.19, p = 0.02) and loss of emotional and behavioral control (r = 0.2, p = 0.17), and greater symptom distress (r = 0.17, p = 0.04). After adjustment for covariates, Overall QOL (mean ± SE) was still significantly worse for INS vs. OA (457 ± 9 vs. 483 ± 8, P=0.04). The difference was primarily driven by the QOL scale of General Health Perceptions (vitality, health status, and sleep) where insulin users scored significantly worse than non-users (443 ± 10 vs. 489 ± 9, P=0.003).

Conclusion: In individuals with typical characteristics of LADA, Overall QOL was lower among those receiving insulin, even after controlling for HbA1c, age, and duration of diabetes. The findings suggest an underlying physical as well as psychological differential between the two treatment modalities. Since the majority of LADA patients are initially treated for type 2 diabetes, screening patients for age, BMI, and c-peptide suggestive of LADA, and performing confirmatory tests for ICA or GAD antibodies, might enable clinicians to prepare high risk individuals for the physical, behavioral and psychological adjustments required with eventual insulin therapy.