PREDICTING FUTURE-TRENDS IN THE NUMBER OF PATIENTS ON RENAL REPLACEMENT THERAPY IN DENMARK

Objectives. To predict the future prevalence of patients on renal repl acement therapy due to chronic renal failure in Denmark. Subjects and methods. Four thousand and nine terminal uraemic patients (median age 50.0 years, 15.2% diabetic) were treated in Denmark with renal replace ment therapy in the period 1 January 1991 to 31 December 1995. Inciden ce rates and rates of transition between the treatment modalities (hae modialysis, peritoneal dialysis, and renal transplantation) were calcu lated. The prediction was made using a Markov model in three ways: (1) using the average rates (deterministic model), (2) using rates simula ted with pseudorandom numbers based on the average rates (stochastic m odel), and (3) using increasing incidence rates in a deterministic mod el. Results. Using present rates both model types predicted a signific ant increase in the prevalence of renal transplant recipients < 60 yea rs (from 1003 in 1995 to about 1465 in 2006) and the prevalence of hae modialysis patients greater than or equal to 60 years (from 456 in 199 5 to about 903 in 2006) while the prevalence of other treatment modali ties would change less dramatically. The overall prevalence proportion would increase from 539 patients per million population (p.m.p.) in 1 995 to about 777 p.m.p. in 2006. The stochastic model clearly demonstr ated the uncertainties linked to the prognosis in contrast to the dete rministic model. The deterministic model with increasing rates predict ed a prevalence proportion of 1162 p.m.p. in 2006. Conclusion, Even wi th present rates the prevalence of haemodialysis patients in Denmark w ill continue to increase. Mathematical models offers a good tool to st udy future trends and to plan future capacity.