Regional Variation in Diabetes Clinic Utilisation in Saudi Arabia: An Analysis of Ministry of Health Statistical Yearbook Data

Abstract

Background: Saudi Arabia has one of the highest diabetes burdens in the world, yet regional monitoring relies mainly on periodic surveys. The Ministry of Health (MOH) Statistical Yearbook reports annual encounters at hospital diabetes clinics by region, a routinely published but rarely analysed indicator.

Objective: To describe regional variation in MOH hospital diabetes clinic utilisation across Saudi Arabia's 13 administrative regions from 2022 to 2024, and to examine its association with health system supply and population composition.

Methods: Encounters at MOH hospital diabetes clinics by region, sex and nationality were extracted from the 2022 to 2024 Statistical Yearbooks and harmonised to the 13 administrative regions. Rates per 10,000 population used 2022 census denominators. Variation was summarised with the extremal quotient (EQ), coefficient of variation (CV) and Gini coefficient, rank stability with Spearman correlation, and associations with negative binomial regression (incidence rate ratios, IRRs).

Results: Encounters rose from 593,157 in 2022 to 620,742 in 2024, a national rate of 184.4 and 175.8 per 10,000. Saudi rates were about ten times non-Saudi rates, and women had higher rates than men. Regional rates varied 6.2-fold in 2022 and 9.3-fold in 2024, from 75.1 per 10,000 in Riyadh to 699.4 in Al-Baha (CV 64.5%, Gini 0.37), with highly stable rankings (Spearman rho 0.87 to 0.88). Utilisation correlated with MOH beds per 10,000 (rho 0.80) and inversely with the non-Saudi share (rho -0.81). Among Saudis, PHC centre density was associated with higher use (IRR 1.32 per SD, 95% CI 1.00 to 1.73).

Conclusions: Hospital diabetes clinic use varies widely and persistently between regions and tracks MOH service supply and population composition more closely than survey-based prevalence. The indicator measures where diabetes care is delivered, not disease burden, and would be more useful for monitoring health sector transformation if linked to prevalence, age structure and non-MOH providers.