Aim: Cost-effective arguments are needed to help convince less-resourced governments to provide essential T1D services. We developed two scenarios - for Pakistan (Lower-Middle Income) (S1) and Azerbaijan (Upper-Middle Income) (S2).

Methods: We modelled the complications/costs/mortality/Disability Adjusted Life Years (DALYs) associated with achieving a mean HbA1c of 9.0% (75 mmol/mol) for S1 and 8.5% (69 mmol/mol) for S2 compared with those with a mean HbA1c level of 12.5% (113 mmol/mol). ’Acceptable care’ S1 involved human insulin by multiple daily injections (MDI), two blood glucose tests/day, education and complications screening etc. S2 was S1 with analog rather than human insulin and four tests/day. ’Minimal care’ was non-MDI insulin, syringes and routine clinic reviews. A discrete time Markov illness-death model with diabetes-duration-dependent transition probabilities was developed in R 3.3.1. Input parameters included 30 years (y) of University of Pittsburgh Epidemiology of Diabetes Complications (EDC) cohort statistics for T1D diagnosed <17 y of age and DALYs and background mortality data from WHO. Eleven complications were included. Costs were determined from international and local estimations.

Results: Complications were markedly raised with HbA1c at 12.5% compared to 9.0% and 8.5%. For instance, blindness at 30y was 49.4%, 11.6% and 9.5% respectively. For S1, 30y total costs (maintenance and cost of acute and chronic complications) were $20,919/patient at 9.0% and $20,796 at 12.5%. For S2, costs were $61,429 at 8.5% and $60,826 at 12.5%. Thirty y survival rate for Pakistan was 87% at 9.0% and 64% at 12.5%, and for Azerbaijan 91% at 8.5% and 66% at 12.5%. Cumulative DALYs lost in 30 years in Pakistan were 2.6y at 9.0% and 6.0y (12.5%). Azerbaijan was 2.0y (8.5%) and 5.8y (12.5%).

Conclusion: ’Acceptable care’ is of similar cost to ’Minimal care’ in Pakistan and Azerbaijan, with improved survival and reduced DALYs. This model can be used in other less-resourced countries.


G.D. Ogle: None. J. Guo: None. G.A. Ahmadov: None. A. Fawwad: None. G.A. Gregory: None. E.L. Klatman: None. T.J. Orchard: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at