Skip to main content

07-09-2019 | Inpatient diabetes | Highlight | News

Holistic approach can shorten diabetes inpatient stays


medwireNews: A nurse-delivered care program based on a whole-systems approach can lead to shorter hospital stays for patients with diabetes, UK researchers report.

Gerry Rayman (Ipswich Hospital NHS Trust) and colleagues explain that the Diabetes Inpatient Care and Education (DICE) program was initially designed to use technology, education, protocols, and pathways for identifying patients most in need of specialist diabetes input.

In the current analysis, the researchers assessed the impact of the intervention on length of stay, in hospital mortality, and readmissions for people admitted to Ipswich Hospital between January 2008 and June 2016.

In the 6 months before DICE implementation, 2337 patients with diabetes and 13,765 patients without diabetes were admitted to the hospital under specialties routinely involved in the DICE project (ie, excluding pediatrics, neonatology, and obstetrics). In the 6 months after implementation, the corresponding admission numbers were 2433 and 14,290.

Rayman and team report in Diabetic Medicine, that the mean length of stay fell from 7.5 days before the intervention to 6.7 days after the intervention among the people with diabetes and from 5.0 to 4.7 days in those without diabetes. The median reductions were 0.4 and 0.1 days, respectively.

After adjustment for potential confounders such as age, sex, ethnicity, deprivation level, comorbidity, and type of admission (elective vs emergency), before‐and‐after analysis revealed that the intervention was associated with significant 11% and 7% reductions in length of stay for patients with and without diabetes, respectively.

However, when the investigators repeated the analysis using an interrupted time series approach, which used data for the entire 7.5-year time period to estimate an intervention effect whilst adjusting for any underlying secular trend, they found that there was a significant acceleration in the trend for reducing length of stay in people with diabetes after the intervention but not in those without diabetes.

The team also found that, in the 6 months after the intervention, mortality fell by a significant 37% (from 6.4 to 4.4%) in the patients with diabetes and by a significant 19% (from 3.7 to 3.1%) in those without diabetes. But these trends were not significant within the interrupted time series analysis.

The interrupted time series analysis also showed a significant increase in readmission for both people with and without diabetes in the post-intervention period.

Rayman et al conclude that “the DICE project showed that a well-staffed inpatient team, delivering care through a whole-systems approach and weekend working led to a sustained reduction in length of stay for inpatients with diabetes, which has important financial implications.”

The authors also support the use of the “quasi-experimental” interrupted time series analysis for assessing the impact of interventions in the real-world. “With multiple factors influencing patient length of stay and mortality in the inpatient setting, the methodology used in the present study shows the additional information that can be gained by using interrupted time series analysis, and highlights the benefit of using a negative control group comprising patients without a diagnosis of diabetes,” they write.

By Laura Cowen

medwireNews is an independent medical news service provided by Springer Healthcare. © 2019 Springer Healthcare part of the Springer Nature group


Be confident that your patient care is up to date

Medicine Matters is being incorporated into Springer Medicine, our new medical education platform. 

Alongside the news coverage and expert commentary you have come to expect from Medicine Matters diabetes, Springer Medicine's complimentary membership also provides access to articles from renowned journals and a broad range of Continuing Medical Education programs. Create your free account »