Evaluation of a hospital diabetes specialist nursing service: a randomizedcontrolled trial

Aims To evaluate the effectiveness and cost implications of a hospital diab etes specialist nursing service. Methods We conducted a prospective, open, randomized, controlled trial of s tandard in-patient care for adults with diabetes, with and without the inte rvention of a diabetes specialist nursing (DSN) service. The setting was a single UK university hospital. Subjects were unselected patients referred t o the hospital DSN service. Primary out come measures were length of hospit al stay and patterns of readmission (frequency and time to first readmissio n). Secondary outcome measures were subjects' diabetes-related quality of l ife, diabetes knowledge score, satisfaction with treatment, and GP and comm unity care contacts following discharge. Costs were estimated from the hosp ital and published sources. Results Median length of stay was lower in the intervention group (11.0 vs. 8.0 days, P < 0.01). Readmission rates were the same in the two groups (25 %), and mean time to readmission was similar in the two groups, although sl ightly less in the control group (278 vs. 283 days, P = 0.80). The cost per patient for nursing input was <pound>38.94. However, when the reduced leng th of stay was accounted for, the intervention produced a mean cost per adm ission of pound 436 lower than that of the control group (P= 0.19). Patient s in the intervention group were more knowledgeable regarding their diabete s and more satisfied with their care. Conclusions Diabetes specialist nurses are potentially cost saving by reduc ing hospital length of stay (LOS). There was no evidence of an adverse effe ct of reduced LOS on re-admissions, use of community resources, or patient perception of quality of care.