This study evaluated real-world antidiabetic treatment patterns, therapy modifications, medication adherence, and guideline-recommended glycaemic monitoring among privately insured patients with Type 2 diabetes mellitus (T2DM) in the Kingdom of Saudi Arabia (KSA). The analysis addresses gaps in evidence for privately insured beneficiaries despite national diabetes initiatives under Vision 2030 and recent guideline implementation.
A longitudinal retrospective cohort design was applied using national claims data from the National Platform for Health Information Exchange Services (NPHIES). The study window spanned 1 September 2022 to 31 December 2024. The dataset covered privately insured beneficiaries across KSA and leveraged standardized coding to enable a national-level real-world assessment of antidiabetic care.
Eligible participants were privately insured individuals aged 15 years or older with a recorded diagnosis of T2DM, defined by at least one medical claim coded as E11 in the International Classification of Diseases, Tenth Revision, Australian Modification (ICD-10-AM). Continuous insurance enrollment of at least 12 months was required, with a minimum of 180 days pre-index and 180 days post-index. The index date was the earliest antidiabetic prescription observed within the study period. A total of 187,797 patients met these inclusion criteria and were included in the final analysis.
Claims were linked to the Council of Health Insurance's internal product dictionary via the NPHIES product/service code to retrieve standardized product names and activity types. Data extraction was performed using Structured Query Language (SQL)-based queries, and subsequent transformation used Python. These steps supported consistent identification of drug classes, treatment episodes, and laboratory testing events across the national claims dataset.
The mean age of included patients was 53.7 years. Within the insured population captured by the claims platform, T2DM rates were 47.4 per 1,000 insured men and 43.8 per 1,000 insured women. These descriptive measures characterize the study cohort and contextualize treatment and monitoring patterns reported in subsequent sections.
At treatment initiation, combination therapy constituted 29.4% of initial regimens among eligible patients with T2DM. The study describes initial use patterns by drug class and regimen type, enabling observation of first-line choices in this privately insured population. Specific proportions for other monotherapy classes at initiation were reported in the source data, with combination therapy being a notable share.
Among users of specific drug classes, the study quantified common therapy modifications. For patients starting on biguanide monotherapy, switching occurred in 9% of users and escalation in 7%. Users of sulfonylureas showed a higher escalation rate at 35%. Treatment discontinuation was observed in 22% of users of combination therapies, 22% of SGLT2 inhibitor users, and 19% of sulfonylurea users. These metrics capture real-world dynamics of regimen adjustment in response to clinical need, tolerability, access, or other factors recorded in claims.
Medication adherence was measured using the medication possession ratio (MPR) for at least one treatment category reported. For insulin users, the MPR was 0.87. This indicates relatively high medication availability by claims-based refill metrics among insulin-treated patients in this privately insured cohort.
The study assessed laboratory testing consistent with guideline recommendations for glycaemic control, including haemoglobin A1c (HbA1c), plasma glucose tests, and oral glucose tolerance tests. In the 6 months prior to therapy initiation, 61% of patients had at least one recorded glycaemic test. Following therapy initiation, the proportion of patients with laboratory testing increased to 73%. These findings reflect the frequency of biomarker monitoring captured in claims around treatment initiation.
The analysis demonstrates that an integrated national claims platform with standardized coding and e-prescribing can support large-scale real-world evidence generation for diabetes care. The authors conclude such infrastructure can inform diabetes management strategies in KSA and across MENA health systems by describing treatment patterns, therapy modifications, adherence, and laboratory monitoring at scale. Specific drivers behind treatment changes, patient-level clinical outcomes, and causal inferences were not reported in the source and would require additional clinical data or prospective study designs.