The formation and design of the 'Acute Admission Database'- a database including a prospective, observational cohort of 6279 patients triaged in the emergency department in a larger Danish hospital

Abstract

BackgroundManagement and care of the acutely ill patient has improved over the last years due to introduction of systematic assessment and accelerated treatment protocols. We have, however, sparse knowledge of the association between patient status at admission to hospital and patient outcome. A likely explanation is the difficulty in retrieving all relevant information from one database. The objective of this article was 1) to describe the formation and design of the 'Acute Admission Database', and 2) to characterize the cohort included.MethodsAll adult patients triaged at the Emergency Department at Hillerød Hospital and admitted either to the observationary unit or to a general ward in-hospital were prospectively included during a period of 22 weeks. The triage system used was a Danish adaptation of the Swedish triage system, ADAPT. Data from 3 different data sources was merged using a unique identifier, the Central Personal Registry number; 1) Data from patient admission; time and date, vital signs, presenting complaint and triage category, 2) Blood sample results taken at admission, including a venous acid-base status, and 3) Outcome measures, e.g. length of stay, admission to Intensive Care Unit, and mortality within 7 and 28 days after admission.ResultsIn primary triage, patients were categorized as red (4.4%), orange (25.2%), yellow (38.7%) and green (31.7%). Abnormal vital signs were present at admission in 25% of the patients, most often temperature (10.5%), saturation of peripheral oxygen (9.2%), Glasgow Coma Score (6.6%) and respiratory rate (4.8%). A venous acid-base status was obtained in 43% of all patients. The majority (78%) had a pH within the normal range (7.35-7.45), 15% had acidosis (pH < 7.35) and 7% had alkalosis (pH > 7.45). Median length of stay was 2 days (range 1-123). The proportion of patients admitted to Intensive Care Unit was 1.6% (95% CI 1.2-2.0), 1.8% (95% CI 1.5-2.2) died within 7 days, and 4.2% (95% CI 3.7-4.7) died within 28 days after admission.ConclusionsDespite challenges of data registration, we succeeded in creating a database of adequate size and data quality. Future studies will focus on the association between patient status at admission and patient outcome, e.g. admission to Intensive Care Unit or in-hospital mortality.

Patient visits to the emergency department at a Norwegian university hospital: variations in patient gender and age, timing of visits, and patient acuity

“…Though, considering 11% red priority patients, a higher level of ICU utilisation would be expected. A paper from Denmark shows an ICU admission rate of 1.6% 21. In a similar study, 4.4% patients were triaged as red and only 10.8% of these patients were admitted to the ICU 22.…”

Peripheral venous lactate at admission is associated with in‐hospital mortality, a prospective cohort study

“…Data cover the initial assessment upon arrival to the ED as well as blood sample results taken at admission and defined outcome measures. The formation and content of the database is described in more details in Barfod et al . Nordsjælland Hospital is a 24/7 acute care hospital that offers emergency, level 2 trauma center, medical, surgical, and intensive care services for 310,000 citizens in the North Zealand.…”

“…We therefore recognized the need for formation of an acute cohort covering all diagnoses and tracking the patient from admission to discharge or death. We have previously described the formation of this acute cohort including patients with all diagnoses and the variables included . Furthermore we have analyzed vital signs as predictor for adverse outcome within this cohort .…”

Improvement of clinical quality indicators through reorganization of the acute care by establishing an emergency department-a register study based on data from national indicators

“…Although a few clinical databases related to Emergency Medicine were under development in 2013 [11,12] the usual quality measurements of choice have been to monitor different time intervals, e.g. door to treatment, which are easy accessible data for administrative use.…”