The health impact of the typhoon (tropical cyclone) on emergency department visits and disease patterns -A study in eastern Taiwan (Ilan, Hualian, and Taitung)-

Background and objectives A tropical cyclone (typhoon) is a storm system characterized by a large low-pressure center, which could produce strong winds, and heavy rain. Previous studies in the United States have shown that tropical cyclones would increase the healthcare demand, shown by the increase of the emergency department(ED) visits after the landfall day. People would possibly get injured more easily during the impact, and also the recovery period. It was suggested that more manpower should be prepared in the EDs to meet the healthcare needs. Taiwan, located at the west Pacific Ocean, is one of the areas frequently impacted by tropical cyclones with an average of 3 to 4 tropical cyclones per year. From this study, we’d like to know the healthcare impact of tropical cyclones and prepare for the possible increased needs in the EDs of Taiwan. By using the different characteristics of each tropical cyclone, we’d also like to show which the strong predictors of ED visits were and the major diseases presented.

Methods This was a retrospective observational study performed in three counties in east Taiwan, Ilan, Hualiang, and Taitung. After excluding those EDs with yearly average daily ED visits less than 10, total twenty hospitals (1 medical center, 7 regional hospitals, and others were area hospitals) were included. We used the National Health Insurance Research Database (NHIRD) provided by National Health Research Institutes. The claimed data, including sex, birthday, visiting date, three codes of “The International Classification of Diseases, 9th Revision, Clinical Modification" (ICD-9 codes) and treatment codes, of all the patients applied for emergency services in the three counties from June to November from 2000 A.D. to 2008 A.D. were collected. ICD-9 codes were further used for identifying trauma and non-trauma patients and different types of trauma. Hospital profiles, including location, contract types (medical center, regional hospital or area hospital) were retrieved from NHIRD, too. All tropical cyclones’ profiles were subtracted from the Tropical Cyclones Database provided by Central Weather Bureau (CWB) of the Republic of China. Total twenty-two tropical cyclones which came from the eastern of Taiwan and had covered any of the three counties were included in the study. The characteristics of those tropical cyclones were retrieved, including the radius, the minimum pressure recorded, maximum windspeed, grade, landing area and total alarm days. The daily rainfall amounts were also retrieved from CWB. The primary outcome was the daily ED visits of each hospital. Different types of diseases, age group, triage level, sex group were the secondary outcomes. Linear regression was used to evaluate the univariate or multivariate associations between the characteristics of tropical cyclones and all primary and secondary outcomes. The regression models were further adjusted by yearly ED visits, holidays. The stepwise variable selection procedure (with iterations between the forward and backward steps) was applied to obtain the candidate final regression model. A two-tailed p-value, 0.05, was considered significant.

Results In the areas the severe typhoon landed directly, the ED visits would increase for two days since the landfall day. For a hospital with a yearly average ED visit of 100 per day, there would be 18.1 visits more than the non-typhoon day (not affected by typhoon and the daily rain amount <50 mm) (p<0.001) on the day of landfall and 24.9 visits more (p<0.001) on the next day after landfall day. In the areas, where the severe typhoon didn’t land directly but brought heavy rain (daily rain amount >50 mm), the ED visits would increase by 6.3 on the day of landfall (p<0.001). The percentage of trauma ED visits would increase in the area landed by severe typhoon on the landfall day. For a hospital with a yearly average ED visit of 100 per day, the trauma ED visits increased 4.5% compared with the non-typhoon day (p<0.001). Among the traumatic ED visits, patients visited ER for skin laceration increased 4.1% (p<0.001). There were no significant differences of trauma ED visits on the other days after landfall day. However, patients with skin laceration still increased 6.9% (p=0.02) in the areas where the severe typhoon brought heavy rain without direct landing. In the areas where tropical storm passed, the ED visits were not significantly different on the day of landfall, but decreased on the next day of landfall day. In the areas where moderate typhoons passed, the ED visits decreased on the day before landfall day, but there were 7.2 more visits per day in the direct landing area on the landfall day, for a hospital with a yearly average ED visit of 100 per day (p=0.002).

Discussion The emergency department managers and health authorities should consider preparing more manpower to meet the increased ED visits, for two days in the severe typhoon landfall area since the landfall day of and for one day in the non-landfall area with heavy rain. Doctors and nursing staffs that specialized in trauma care, especially wound management were needed preferentially. Future emergency response planning and prepare of both hospitals and health authorities for tropical cyclones could based on the current findings, in order to meet the health care need.