2.1. Study Design and Setting
The present study was a (medical administrative) trial in an ED of a teaching (central) hospital with an annual input of more than 100,000 patients. This study was also a case-control research conducted in Golestan Teaching Hospital located in the city of Ahvaz, the capital city of Khuzestan Province in southwestern Iran. At the recall time, the patients or the companions or both might be unavailable or out of the ED for imaging or requesting the results. Pressing the repeat bottom, the physicians could recall the patients up to three times. If no one appeared over these three recalls, physicians could set the absent number to be recalled after five new patients in the queue by pressing the absent bottom. Patients’ triage levels might also change on different reasons while they were residing in the ED. To prevent any disturbances in the DQS functioning, all levels of adjustments at all times were possible just through a request form sealed by the physician in charge. Based on this request, the receptionist would issue a new number of the new level. The panel used in station 3 also had a numerical key in addition to the ones explained in Fig. 2, for stations 1 and 2. This could make the DQS manual feeding possible for ticket number of the patients whose files had been completed and formed according to the time that files were ready for re-evaluation and not based on the sequence numbers in stations 1 and 2. Once reaching the number 200 in station 1 and number 500 in station 2, the DQS needed to be reset to launch the ticket number 001 in station 1 and 201 in station 2. The recalls of the tickets with an unexpectedly smaller number could thus cause confusion among companions, especially during ED overcrowding peak hours. Therefore, system-resets were left to the least crowded hours (namely, 5 a.m.). At this time, emergency physicians could discard tickets of the few present patients in the ED and request new tickets for them, according to their triage levels. Totally, in this DQS, tickets numbered 001 to 200 were devoted to patients who had been labeled as level 3 and tickets numbered 201 to 500 were assigned to those labeled as level 4.
2.2. Participants
The triage algorithm used in this center was the Emergency Severity Index (ESI) (Version 4) in which the patients were classified based on the severity of their conditions and the assumed number of resources they needed. Patients labeled as levels 1 or 2 had accordingly critical conditions mandating rapid assessment and treatment in the cardiopulmonary resuscitation (CPR) room. On the other side, patients of triage level 5 were the least urgent groups who could wait as outpatients for their first visits in the fast track area. Thus, in this study, levels 1, 2, or 5 were excluded and patients labeled as levels 3 or 4 were included in the DQS. In practice, triage levels 3 and 4 accounted for the majority of patients contributing to ED overcrowding and chaos. Of note, work shifts were mainly 12 hours in this center. Therefore, information regarding 30 (12-hour) shifts was collected as the control group before installing the DQS in the ED and after installing the DQS, the information about 50 shifts was collected as the case group. The study protocol was further approved by the Ethics Committee of Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran.
2.3. Data Collection
The information about work shifts was entirely extracted from questionnaires filled by the medical staff, including emergency medicine residents, interns, and nurses. In this questionnaire, they were asked to fill in the form with a number from 1 to 10, based on the most chaotic shifts they could remember in their career as 10 and the least chaotic ones as 1. Then, scores 10 − 7 were considered as the worst chaotic shift, scores 6 − 3 indicated medium chaos, and scores 3 − 1 were taken as best shifts with the least chaos. This variable was named the Relative Chaos Scale (RCS) and was implemented to quantify tension and chaos ratio during work shifts. During each shift, levels of satisfaction from five randomly selected patients were recorded in a scale with items of “not acceptable, indifferent, and acceptable”. As shown in the flowcharts of Figs. 1 and 2, the DQS attempts to keep the entire companions out of the ED treatment area, in a waiting room or similar places, until the exact time they are recalled to collaborate in fetching a blood bag or transferring patients for imaging, ultimate disposition, or so on. Later, laboratory and imaging results could be delivered by the companions into a box, previously applied out of the ED treatment area. This box was also emptied by the staff in regular intervals and the results were enclosed to the related files.
2.4. Statistical Analysis
The sample size in this study was calculated by considering a confidence interval (CI) of 95% and a test power of 80%. With regard to 20% exclusion, 30 and 50 patients were respectively enrolled in the case and control groups to increase the power of the study. Statistical analyses were further performed using the SPSS Statistics software (version 18) (SPSS Inc., Chicago, IL, USA). The categorical data were accordingly reported by frequency and percentage and the quantitative continuous data were indicated by mean ± standard deviation (SD). Moreover, analysis of variance (ANOVA) was employed to compare the significant difference between work shifts and to adjust the number of patients as confounders. P-values less than 0.05 were considered as the significance level.