Management of non-traumatic abdominal pain in the emergency department: a multicentre, stepped-wedge, cluster-randomised trial

Abstract

Background: Even though abdominal pain is one of the most frequent chief complaints in emergency medicine, standardized care pathways are still lacking. In this study, a standardized, digitally-supported care pathway for non-traumatic abdominal pain in the emergency department was investigated with regard to emergency department length of treatment, pain score at discharge, and patient satisfaction.

Methods: In a prospective mixed-methods, multicentre, cluster-randomised, controlled stepped wedge trial, adult patients with non-traumatic abdominal pain were enrolled in ten emergency departments across Germany. The new care pathway was implemented at randomly assigned time points (every four months) within a 24-month recruitment period and consisted of a structured care pathway for the management of abdominal pain patients in the emergency department. During the control period, the standard treatment for abdominal pain in the emergency department was administered. The planned sample size was 2000. Primary outcomes were: emergency department length of treatment, pain score (numeric rating scale 0-10), and patient satisfaction score at the end of emergency department treatment. Exploratory safety outcomes were serious adverse events within 30 days. Trial registration: DRKS00021052.

Findings: Of 2119 patients, 1017 were enrolled in the control group, and 1102 in the intervention group. Crude mean emergency department length of treatment was 5.2 h (±3.0) in the control group, and 4.3 h (±2.2) in the intervention group while the adjusted mean difference was -0.31 h (95% confidence interval (CI) -0.70 to 0.07). Mean pain score was 4.3 (±2.5) in the control group, and 3.6 (±2.4) in the intervention group, resulting in an adjusted mean difference of -0.69 (95% CI -1.04 to -0.34). The adjusted mean difference of patient's satisfaction score was 1.54 (95% CI 0.96 to 2.12); mean control group: 26.7 (±4.0); mean intervention group: 27.9 (±3.8)). Serious adverse events were comparable between both groups while 30-day mortality was 2.3% (n = 23) in the control group, and 0.8% (n = 9) in the intervention group (mean difference: -1.4% (95% CI -2.5 to -0.4)).

Interpretation: The APU process is safe and did not increase emergency department length of treatment, while patient-reported outcomes and safety were improved accompanied by an increased use of diagnostic procedures.

Funding: The study was funded by the German Innovations Funds.

Keywords: Abdominal pain; Acute care; Emergency care; Emergency department; Standardised treatment; Stepped wedge design.

Fig. 1

Enrollment and randomization of patients. A total of 45,202 patients were screened for eligibility and 41,902 were excluded. The primary reason for exclusion was that study staff was not available for study inclusion (81.5%). In 10.8% the inclusion and/or exclusion criteria were not met and 3.6% were initially treated by specialists from other departments than the emergency department. Of all patients who did not meet inclusion and/or exclusion criteria, 800 were not able to provide written informed consent, 3689 did not have public health insurance, 505 had trauma-related abdominal pain, 21 were not adults and 2799 had language barriers. Patients who presented to a study centre within the two-week implementation period between control and intervention phase were screened but inclusion stopped during this time. This was the case in 2.1% of all patients. Of all eligible patients who were approached for study participation, 1181 declined participation and 2119 were included in the study. Of enrolled patients 48.0% were enrolled during the control phase and 52.0% during the intervention phase.

Fig. 2

Exploratory subgroup analyses of the primary study endpoints in subgroups of sex, age and nationality. Shown are mean differences between intervention and control group with 95% confidence intervals.