Neural networks and hospital length of stay: an application to support healthcare management with national benchmarks and thresholds.

BACKGROUND: The problem of correct inpatient scheduling is extremely significant for healthcare management. Extended length of stay can have negative effects on the supply of healthcare treatments, reducing patient accessibility and creating missed opportunities to increase hospital revenues by means of other treatments and additional hospitalizations. METHODS: Adopting available national reference values and focusing on a Department of Internal and Emergency Medicine located in the North-West of Italy, this work assesses prediction models of hospitalizations with length of stay longer than the selected benchmarks and thresholds. The prediction models investigated in this case study are based on Artificial Neural Networks and examine risk factors for prolonged hospitalizations in 2018. With respect current alternative approaches (e.g., logistic models), Artificial Neural Networks give the opportunity to identify whether the model will maximize specificity or sensitivity. RESULTS: Our sample includes administrative data extracted from the hospital database, collecting information on more than 16,000 hospitalizations between January 2018 and December 2019. Considering the overall department in 2018, 40% of the hospitalizations lasted more than the national average, and almost 3.74% were outliers (i.e., they lasted more than the threshold). According to our results, the adoption of the prediction models in 2019 could reduce the average length of stay by up to 2 days, guaranteeing more than 2000 additional hospitalizations in a year. CONCLUSIONS: The proposed models might represent an effective tool for administrators and medical professionals to predict the outcome of hospital admission and design interventions to improve hospital efficiency and effectiveness.

The hospital care of patients with cancer: a retrospective analysis of the characteristics of their hospital stay in comparison with other medical conditions.

PURPOSE: Hospital admission is a frequent occurrence among patients with cancer, and a significant proportion of patients admitted to medical units have cancer. Their hospital stay has features that may be different compared with patients without cancer. We performed a retrospective analysis of the characteristics of patients with cancer admitted for medical conditions. PATIENTS AND METHODS: We studied the administrative data of patients with solid cancer admitted to the medical department of a large referral hospital over a 12-month period and compared them with those of patients without cancer. RESULTS: Seven thousand eight hundred two consecutive admissions were analysed, of which 1099 (14.1%) had a principal or associated diagnosis of cancer. Admissions were distributed across 12 units, with 44% concentrated in the medical oncology unit and 56% in other units. Patients with cancer were more frequently men and were younger than patients without cancer. Admission less frequently involved the emergency department (ED), while discharge was more frequently assisted. The in-hospital death rate was higher, as was the readmission rate. Length of stay was longer (11.3 days vs. 9.8 days; p < 0.0001). Patients with cancer admitted to the medical oncology unit used the ED even less, and their length of stay was shorter than that of patients with cancer admitted in other units. CONCLUSIONS: The in-hospital pathway of patients with cancer displays specific issues and adds complexity to hospital stay of patients with medical conditions. The medical oncology unit plays a role in reducing ED use and in providing efficient care. The evidence gathered should help in shaping new models of care and in improving adequate clinical competencies.

Organizational determinants of hospital stay: establishing the basis of a widespread action on more efficient pathways in medical units.

Given the high hospital costs, the increasing clinical complexity and the overcrowding of emergency departments, it is crucial to improve the efficiency of medical admissions. We aimed at isolating organizational drivers potentially targetable through a widespread improvement action. We studied all medical admissions in a large tertiary referral hospital from January 1st to December 31st, 2018. Data were retrieved from the administrative database. Available information included age, sex, type (urgent or elective) and Unit of admission, number of internal transfers, main ICD-9 diagnosis, presence of cancer among diagnoses, surgical or medical code, type of discharge, month, day and hour of admission and discharge. National Ministry of Health database was used for comparisons. 8099 admissions were analyzed. Urgent admissions (80.5% of the total) were responsible for longer stays and were the object of the multivariate analysis. The variables most influencing length-of-stay (LOS) were internal transfers and assisted discharge: they contributed, respectively, to 62% and 40% prolongation of LOS. Also, the daily and weekly kinetics of admission accounted for a significant amount of variation in LOS. Long admissions (≥ 30 days) accounted for the 15.5% of total bed availability. Type of discharge and internal transfers were again among the major determinants. A few factors involved in LOS strictly depend on the organizational environment and are potentially modifiable. Re-engineering should be focused on making more efficient internal and external transitions and at ensuring continuity of the clinical process throughout the day and the week.

Toward uniform and controlled clinical pathways in cancer care: a qualitative description.

QUALITY ISSUE: The definition of clinical pathways (CPs) and their application are heterogeneous. Each center is used to choose whether to adopt this instrument or not and to variably conceive its features We consider CPs as the necessary description of the cancer patient journey and we emphasize their role as the user view of clinical processes rather than a local translation of guidelines. CHOICE OF SOLUTION: We proposed a unique CPs model for all the centers of our regional network, with the aim of making CPs accountable and comparable. We also established a central quality evaluation. IMPLEMENTATION: Through a multi-step process, the model was proposed to the 22 Regional centers. Landmark characteristics of the project were: the involvement of hospital administrations; reference to a unique set of guidelines; a peer-review and open evaluation. EVALUATION: Of the 374 expected CPs, 253 (68%) were received and evaluated. A median number of 131 items were the object of evaluation in each hub center and 77 in each spoke center. About 79.5% items were considered well described, 15.5% were absent and 5.0% partially described. The median percentage of fulfilled indicators was 85.6% in hub CPs and 82.2% in spoke CPs. Although, not all diseases were equally covered through the territory a high degree of homogeneity and a good quality of compilation were achieved. LESSONS LEARNED: The project was shown to be feasible and achieved its goal. We suggest this process as a functional way for building uniform cancer CPs.

Synthesis and Degradation of Adenosine 5'-Tetraphosphate by Nicotinamide and Nicotinate Phosphoribosyltransferases.

Adenosine 5'-tetraphosphate (Ap4) is a ubiquitous metabolite involved in cell signaling in mammals. Its full physiological significance remains unknown. Here we show that two enzymes committed to NAD biosynthesis, nicotinamide phosphoribosyltransferase (NAMPT) and nicotinate phosphoribosyltransferase (NAPT), can both catalyze the synthesis and degradation of Ap4 through their facultative ATPase activity. We propose a mechanism for this unforeseen additional reaction, and demonstrate its evolutionary conservation in bacterial orthologs of mammalian NAMPT and NAPT. Furthermore, evolutionary distant forms of NAMPT were inhibited in vitro by the FK866 drug but, remarkably, it does not block synthesis of Ap4. In fact, FK866-treated murine cells showed decreased NAD but increased Ap4 levels. Finally, murine cells and plasma with engineered or naturally fluctuating NAMPT levels showed matching Ap4 fluctuations. These results suggest a role of Ap4 in the actions of NAMPT, and prompt to evaluate the role of Ap4 production in the actions of NAMPT inhibitors.