Impact of seasonal and meteorological factors on the incidence of adhesive small bowel obstruction: A large‐scale study using a national inpatient database

Abstract Aim Whether seasonal and meteorological factors affect the incidence of adhesive small bowel obstruction (ASBO) remains unclear. This study aimed to clarify the impacts of seasonal and meteorological factors on the occurrence of ASBO. Methods Clinical data of patients with ASBO were acquired from 42 national university hospitals in Japan, using a national inpatient database, between April 2012 and March 2020. Meteorological data were obtained from the Japan Meteorological Agency. The number of monthly admissions of patients with ASBO was compared between each of the 12 months. Daily weather variables were investigated to clarify their association with ASBO patient admissions on a total of 119 802 days (Formula for calculation: study period [2922 days] ×41 cities). Results Overall, 4985 patients with ASBO were admitted. The number of admissions in June was smaller than that in October, November, and December (39 vs 63.5, P = .002, 39 vs 65, P = .004, and 39 vs 59.5, P = .002, respectively). Logistic regression analysis revealed that January, October, November, and December were associated with increased risk of admission compared to June (odds ratio [OR], 1.264; P = .001; OR, 1.454; P < .001; OR, 1.408; P < .001; OR, 1.330; P < .001), respectively. Regarding the weather variables, higher barometric pressure and lower humidity were associated with increased risk of admission (OR, 1.011; P < .001 and OR, 0.995; P < .001), respectively. Conclusion The incidence of ASBO is susceptible to barometric pressure and humidity and varies monthly. These results can contribute to the prevention, early detection, and immediate and appropriate management of ASBO.


| BACKG ROU N D
Adhesive small bowel obstruction (ASBO) is a common disorder of the small intestine. Although the risk of ASBO is decreasing due to the expansion of laparoscopic surgery, patients with this problem comprise a substantial percentage of admissions after intraabdominal surgery. 1 Appropriate and prompt responses, including surgical intervention, are important because some patients with ASBO experience bowel strangulation or ischemia. 2 Primary management of patients with ASBO by a surgical team reduces healthcare utilization and cost and improves outcomes. 3 Therefore, prediction of the number of admissions of patients with ASBO could contribute to better outcomes due to the advantages of appropriate management preparation.
With regard to abdominal diseases, the occurrences of appendicitis, cholecystitis, and diverticulitis are more frequent in summer than in winter. 4,5 In contrast, the winter climate may be associated with a higher incidence of ASBO in Tokyo. 6 However, seasonal variations in the occurrence of ASBO remain unclear. As fluctuations in barometric pressure were observed at the onset, 7 we hypothesized that the incidence of ASBO was susceptible to meteorological factors and showed seasonal variation. Japan is located in the midlatitudes and has a temperate climate in which the weather is quite seasonal. Therefore, the present study aimed to clarify the impact of seasonal and meteorological factors on the incidence of ASBO using a national inpatient database, and meteorological factors measured in multiple places throughout Japan.

| Data source
This retrospective study used the Diagnosis Procedure Combination (DPC) database of 42 National University Hospitals. The DPC includes the discharge summary and administrative claim information, including patient demographics, diagnoses, and comorbidities, using the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) codes for surgeries and procedures performed. 8 This system covers over 1000 hospitals in Japan.
Daily meteorological data were measured using an automated meteorological data acquisition system at each meteorological station in 41 regions in Japan, where the 42 National University Hospitals were located. The mean barometric pressure and temperature readings from continuous measurements (from 0:00 to 24:00) and the cumulative values of daylight hours and precipitation (from 0:00 to 24:00) were regarded as daily weather variables and were obtained from the website of the Japan Meteorological Agency (URL: www.jma.go.jp/jma/index.html). Barometric pressure was estimated based on the altitude of each meteorological station, using the sea level value as a base datum. Day-to-day differences in barometric pressure and air temperature were defined as the mean daily variables on the index day minus those on the previous day. Diurnal variation in air temperature was defined as the variation between the highest and the lowest air temperature that occurred during the index day.  Quan's protocol, and each ICD-10 code for 12 comorbidities was converted into a score and totaled. 9

| Study design
Eight daily weather variables (mean barometric pressure, mean temperature, mean humidity, daylight hours, precipitation, day-to-day differences in mean barometric pressure and mean air temperature, and diurnal variation in air temperature), on a total of 119 802 days (formula: study period [2922 days] ×41 cities), and admission days (days on which patients with ASBO were admitted) were defined.
Next, the numbers of monthly admissions of patients with ASBO were compared between each of the 12 months. The relationships between the months and the admission days were evaluated.
Further, associations between the admission days and the eight daily weather variables were investigated.    Table 2).

| Statistical analysis
There was no strong bivariate correlation between the eight weather variables (| r | > 0.7 (Table S1). Multiple logistic regression analysis revealed that higher mean barometric pressure and lower mean humidity were associated with an increased risk of admission (OR, 1.009; 95% CI, 1.004-1.014, P < .001; OR, 0.995; 95% CI, 0.993-0.997, P < .001), respectively (Table 3). Daily weather variables in June were characterized by lower barometric pressure and higher humidity than those in January, October, November, and December in Japan (Figure 3) (Table S2).
Lastly, meteorological factors on admission days between patients who responded to conservative management and those who required surgical management were compared (Table S3). There were no differences in these variables between the two groups.

| DISCUSS ION
This large-scale study, using a national inpatient database, con- Regarding acute surgical diseases in the abdominal organs, seasonal variations were found in appendicitis, cholecystitis, and diverticulitis, indicating that more admissions occurred during the summer months. 4,5 In contrast, the winter climate has previously

F I G U R E 3
Mean values of five daily weather variables in each month in all 41 regions of Japan during the study period (A, barometric pressure; B, air temperature; C, humidity; D, daylight hours; E, precipitation). Data are presented as mean and standard deviation. P values with an asterisk were determined by the one-way analysis of variance, and P < .05 was considered statistically significant. P values with a double asterisk were determined by two-tailed unpaired t-tests with Bonferroni corrections, and P < .0042 was considered statistically significant been associated with a higher incidence of ASBO in Tokyo. 6 In accordance with this, the present study revealed that the incidence of ASBO in June was significantly lower than that in December, as well as those in October and November. Furthermore, multiple logistic regression analysis indicated that higher mean barometric pressure and lower mean humidity were associated with an increased risk of admission. In Japan, weather variables in summer were characterized by low barometric pressure and high humidity, whereas those in winter were characterized by high barometric pressure and low humidity ( Figure 3). These meteorological data support the seasonal variation in the occurrence of ASBO.
However, the reasons for this variability are unknown because the association between meteorological factors and the etiology of ASBO is unclear. Previously, we reported fluctuations in barometric pressure at the onset of ASBO. 7 Changes in barometric pressure are also associated with the occurrence of pneumothorax, 10,11 intracerebral hemorrhage, 12 myocardial infarction, and brain stroke. 13 Regarding air temperature, the gastrointestinal tract is innervated by the enteric nervous system, which has cold chemosensors that are activated by low temperature. [14][15][16] These studies indicate that changes in barometric pressure and air temperature could be associated with the incidence of ASBO.
In this study, day-to-day differences in mean barometric pressure and mean air temperature, and diurnal variation in air temperature were not associated with the admissions. Further, no weather variable was associated with the need for management (Table S3).  (Table S4). Therefore, some differences in the data may have originated from gaps in altitude, longitude, and latitude between these hospitals and the measuring stations concerned.

| CON CLUS ION
This large-scale study elucidated that the incidence of ASBO varies from month to month. In Japan, June weather is characterized by low barometric pressure and high humidity. In contrast, January, October, November, and December weather is characterized by high barometric pressure and low humidity. Weather variables, namely higher barometric pressure and lower humidity, were associated with an increased risk of admissions of ASBO. These results can contribute to the prevention, early detection, and immediate and appropriate management of ASBO.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author, YY, upon reasonable request.