Survival outcomes after traumatic brain injury during national academic meeting days in Japan

Surgeons and medical staff attend academic meetings several times a year. However, there is insufficient evidence on the influence of the “meeting effect” on traumatic brain injury (TBI) treatments and outcomes. Using the Japan Trauma Data Bank, we analyzed the data of TBI patients admitted to the hospital from 2004 to 2018 during the national academic meeting days of the Japanese Association for Acute Medicine, the Japanese Society of Intensive Care Medicine, the Japanese Association for the surgery of trauma, the Japan Society of Neurotraumatology and the Japan Neurosurgical Society. The data of these patients were compared with those of TBI patients admitted 1 week before and after the meetings. The primary outcome was in-hospital death. We included 7320 patients in our analyses, with 5139 and 2181 patients admitted during the non-meeting and meeting days, respectively; their in-hospital mortality rates were 15.7% and 14.5%, respectively. No significant differences in in-hospital mortality were found (adjusted odds ratio, 0.93; 95% confidence interval, 0.78–1.11). In addition, there were no significant differences in in-hospital mortality during the meeting and non-meeting days by the type of national meeting. In Japan, it is acceptable for medical professionals involved in TBI treatments to attend national academic meetings without impacting the outcomes of TBI patients.

Society of Intensive Care Medicine (JSICM) 17 , JAST 18 , Japan Society of Neurotraumatology (JSNT) 19 , and Japan Neurosurgical Society (JNS) 20 . This is because we believe that medical professionals, such as physicians, nurses, and medical engineers belonging to these academic societies, may play important roles in treating severe TBI patients after hospital admission. The JAAM, JSICM, JAST, and JNS have a total of 10,150, 9117, 2208, and 9940 members 21 and 5439 16 , 2127 17 , 231 18 , and 7903 19 specialists, respectively. The JNST has 814 members; however, their first examination for selecting specialists commenced in 2021 20 . The national meetings of JAAM, JSICM, and JNS usually last 3 consecutive days every October, March, and September, respectively, while the JAST and JSNT meetings last 2 consecutive days in May and February, respectively. Calendar days at these meetings were obtained for each year during the study period.
For analyses, we used the double-control method based on previous studies 22 , which allows near-perfect temporal symmetry for cases and controls and does not create time imbalance inside each pair to assess outcome differences during the exposure and control periods. In accordance with this method, we identified two groups: the exposure group, which included patients with TBI occurrence on meeting days, and the control group, which included patients with TBI occurrence during the same day of the week, 1 week before and after meetings.
Main outcome measures. The primary outcome of this investigation was in-hospital death, and the secondary outcomes were remergency department (ED) mortality, time from onset to arrival at the hospital, time from arrival to operation commencement, and duration of hospital stay. Statistical analysis. Descriptive data are presented as counts and percentages (categorical variables) or as medians with interquartile ranges (numerical variables). Patients' characteristics and hospital care among the eligible TBI patients were compared between the exposure and control groups using Student's t-test and the χ 2 test for categorical variables. Multivariable analysis with logistic regression models was used to compare differences in mortality outcomes between the two groups; adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. Clinically significant confounders were carefully selected from previous reports and adjusted for analyses [6][7][8][9][23][24][25][26] . In the multivariable logistic regression model, we adjusted for the following 12 variables: age groups (age < 18, 18-64, and ≥ 65 years), sex (male, female), type of injury (blunt, other), mechanism of trauma (traffic accident, fall, and other), transfer system (ambulance, car/helicopter with physician, other), GCS group on arrival (severe: GCS score, 3-8; moderate: GCS score, 9-12; mild: GCS score, 13-15), hypotension (defined as SBP ≤ 90) on ED admission (no, yes), operation indicated for TBI (no, yes), use of anticoagulant or antiplatelet drugs (no, yes), maximum head AIS scores [3][4][5] , multiple trauma (no, yes), and ISS. Additionally, the adjusted ORs were calculated for each meeting.
Statistical significance was defined as two-sided p-values < 0.05 for patient characteristics or was assessed using the 95% CI for mortality in all statistical analyses. Analyses were performed using STATA version 16 (StataCorp, College Station, TX, USA).
The study was conducted in accordance with the Declaration of Helsinki. This manuscript was written based on the STROBE statement for comprehensive reporting of cohort and cross-sectional studies 27 .

Results
Patient characteristics. During the study period, a total of 95,484 cases of TBI were documented (Fig. 1).
After excluding victims during the non-eligible days, the data of 7320 participants (2181 in the exposure group and 5139 in the control group) were included for analyses. www.nature.com/scientificreports/ The characteristics of patients with TBI events during the academic meeting and non-meeting days are presented in Table 1. Among victims in both groups, the proportion of men was approximately 67%, and the mean age was 66 years. There were no differences between meeting and non-meeting days regarding the mechanism of trauma, transfer system, type of TBI, and proportion of surgeries for TBI. The proportion of hypotension on arrival, low GCS score on arrival, high AIS score, and multiple traumas, all of which indicate a severe case, were almost the same between the groups. Table 2 shows the proportion of in-hospital mortality and death in the ER, time from onset to arrival at the hospital, time from arrival to operation commencement, and duration of hospital stay during the meeting and non-meeting days. The proportion of participants experiencing in-hospital and ER mortality after TBI did not differ markedly between the non-meeting and meeting days (   vs. 14 5-31 days, p = 0.066; respectively). Table 3 shows the TBI outcomes during the meeting and non-meeting days as per the type of national meeting. Even after adjusting for potential confounding factors, there were no significant differences in mortality during the meeting and non-meeting days (

Discussion
Contrary to our hypothesis, there were no significant differences in mortality among TBI patients admitted in JTDB-registered hospitals during the national academic meeting and non-meeting days. To the best of our knowledge, the present study is the first to examine the "national meeting effect" on mortality in TBI patients. These results presented unique findings and might help medical professionals attend such academic meetings and learn from them without being concerned regarding this effect.
The "weekend effect" refers to the phenomenon in which patients admitted during weekends may have more fatal outcomes than those admitted during the weekdays 7,8 . Poor performances in hospitals are considered a reason for this effect. In terms of instances requiring time-critical intervention for better outcomes (acute myocardial infarction, cardiac arrest, and ischemic stroke), poor outcomes during off-hours have been observed with fewer aggressive interventional procedures, less subspecialty care, more complications, increased medical errors, and varied staff composition [28][29][30] .
Regarding the "meeting effect" on trauma, it has been reported that adjusted mortality did not increase significantly for patients admitted to trauma centers with American College of Surgeons trauma verification Level 1 during the conference versus non-conference days 6 . Although our study focused on traumatic brain injury, our results are consistent with those from previous studies on the "meeting effect" and trauma mortality using the JTDB database 31 . In a report using the JTDB database, both in-hospital mortality and death in the ER were significantly lower during the day than at night among emergency trauma patients; nevertheless, weekdays/ weekends were not associated with either endpoint 7 . The reason why weekend effects on trauma were not obvious could be explained by the possibility that patients could immediately access the operation room or resources that www.nature.com/scientificreports/ might otherwise be occupied during normal business hours 7 . Similarly, the decline in the volume of scheduled surgery during the meeting periods might have contributed to the unchanged mortality. We further assessed time differences as per the "meeting effect. " In our country, not all institutions have inhouse neurosurgeons or trauma surgeons 32 . Hence, it is unlikely that high-risk patients were directed to such hospitals, possibly resulting in a longer duration from onset to arrival at the appropriate hospital or hospital arrival to operation commencement. However, we did not find any difference. Contrary to the aforementioned hospitals, JTDB-registered hospitals are accountable for "24 h a day, 365 days a year" available resources for injured patients even during national meetings, coupled with a sophisticated prehospital system that preferentially directs patients to the correct facilities. Advance assessment of staff composition and rotation during meeting days may help against the conference effect. Compared to other countries, Japan has a smaller land area and better transportation. The on-call attending surgeons head toward the hospital, responding immediately to trauma calls and provide consistent quality of trauma care even during meetings, reflected in the unchanged time Table 1. Information on patient background between meeting days and non-meeting days. TBI traumatic brain injury, GCS glasgow coma scale, BP blood pressure, AIS abbreviated injury scale, ISS injury severity score, IQR interquartile range. www.nature.com/scientificreports/ from the arrival to operation. Furthermore, no difference in time outcome also contributed to the unchanged mortality during meeting days. In our study, which focused on TBIs, only mortality associated with JAAM attendance tended to decrease on meeting than on non-meeting days; however, unlike previous studies, no significant differences were observed 31 . Although the definitive reasons for this remain unclear, this finding could be attributed to the fact that more ER physicians belong to the JAAM compared to the other four societies. Strong leadership, teamwork, and technical skills are essential for team performance and patient care in initial trauma management 33,34 , and more experienced ER physicians or trauma surgeons would encourage their hospitals to lead the treatment of hospitalized patients during academic national meeting dates 9 .
Considering TBI treatment, factors associated with consistent, high-quality care include appropriate staffing, prompt triage and decision making, and establishing inter-facility consultations. Trauma centers are required to be fully staffed and resourced irrespective of meetings 34,35 . Based on our results, medical professionals involved in TBI treatments in Japan can attend national academic meetings because through these meetings they might learn regarding the latest findings on TBI cases. Importantly, owing to the new coronavirus disease (COVID-19) pandemic, organizers of these conferences may consider offering optional opportunities for Web conferences so that trauma centers or ICUs would maintain their care capacities while medical professionals would benefit from educational opportunities that academic meetings provide. Therefore, it is important to continue monitoring the "meeting effect" concerning the survival of TBI patients.
Our study had several limitations. First, we did not address the influence of the location of the hospital providing the data. For instance, it has been shown that TBI mortality rates are higher in rural areas of the United States 35 . Therefore, our results might be different if we considered information regarding the regionality of hospitals. Second, we did not obtain detailed information on medical staff attending national meetings and non-participants at the hospitals. Additionally, we focused on TBI occurrence during five representative national meetings to simplify our research. The logical next step in this field of research would be the examination of both regional and international meeting periods. Third, the study primarily included major critical care cases; therefore, the results cannot be extended to other institutions. Fourth, this was an observational study, and other unknown confounding factors possibly existed. Finally, the validity and integrity of the data, and ascertainment bias, were potential limitations of our study. However, uniform data collection based on the JTDB registering system and large sample sizes should minimize these potential sources of bias. www.nature.com/scientificreports/