Improving the timeliness and accuracy of injury severity data in road traffic accidents in an emerging economy setting

Improving the timeliness and accuracy of injury severity data in road traffic accidents in an... Abstract Road traffic injuries (RTIs) are among the leading causes of injury and fatality worldwide. RTI casualties are continually increasing in Taiwan; however, because of a lack of an advanced method for classifying RTI severity data, as well as the fragmentation of data sources, road traffic safety and health agencies encounter difficulties in analyzing RTIs and their burden on the healthcare system and national resources. These difficulties lead to blind spots during policy-making for RTI prevention and control. After compiling classifications applied in various countries, we summarized data sources for RTI severity in Taiwan, through which we identified data fragmentation. Accordingly, we proposed a practical classification for RTI severity, as well as a feasible model for collecting and integrating these data nationwide. This model can provide timely relevant data recorded by medical professionals and is valuable to healthcare providers. The proposed model’s pros and cons are also compared to those of other current models. road traffic injury, injury prevention, injury severity, data quality Introduction Road traffic injuries in Taiwan Road traffic injuries (RTIs) are major causes of injury worldwide [1]. In recent two decades, the number of motor vehicles, particularly motorcycles, has grown considerably in Taiwan. Confusion caused by a mixed traffic flow of cars, motorcycles and trucks, is exacerbated by dangerous driving habits. These factors cause the RTI death rate in Taiwan to substantially exceed that in other developed countries [2]. In 2014, 413 229 RTI were reported, representing a 421.3% increase within 14 years. Among those in 2014, almost 60% were from motorcycle accidents [3]. The annual medical cost of RTI-related orthopedic fractures almost doubled between 2002 and 2011 [4]. In 2002, total years of life lost from RTIs was reported as 70 006, with individuals aged 15–29 years being the largest category [5]. Because of high accessibility to acute care facilities under the Taiwanese National Health Insurance (NHI) system, a majority of RTI casualties visit emergency departments (EDs), including patients with slight injury. However, persistently increasing numbers of injured patients will eventually exhaust police and emergency care providers and consume substantial healthcare resources. Difficulties caused by lack of effective classification Understanding injury severity is crucial for preventing and controlling RTIs. Injury severity measures can provide more detailed information to distinguish varying degrees of RTI severity, as well as their effect on national resources, both of which are essential for policy planning. Furthermore, surveilling RTI severity can assess the likelihood of the consumption of certain medical resources, such as hospitalization or surgery, as well as intervention effectiveness. This feedback can assist healthcare providers in resources allocation and patient management. Currently, all RTIs within police reports are classified into injury or death according to patients’ survival status within 24 h after the RTI. The lack of a more detailed classification system hinders relevant agencies in conducting epidemiological analyses and impact assessments. If RTI severity can be differentiated using a straightforward evidence-based classification system, then agencies can use this information for injury prevention and control. Methods To correct the shortcomings in classifying RTI severity, a task force comprising physicians and academicians researching injury prevention and road traffic safety was assembled by the Institute of Transportation of the Ministry of Transportation and Communication (MOTC). With the Road Safety Annual Report 2017 of the International Traffic Safety Data and Analysis Group (IRTAD Group) as a reference, the task force summarized the RTI severity classification in various countries. This Annual Report used information from many sources such as transportation safety-focused government agencies, non-governmental organizations, along with academic and industrial ones. It detailed the RTI severity classifications in 40 member and observer countries. Upon many group discussions and expert meetings, the task force proposed a feasible classification for application in Taiwan. Furthermore, the task force reported the current situation regarding the fragmentation of domestic RTI severity data and its influences, along with proposals to meet the demand for accurate and timely data monitoring. Results Classification of RTI severity in other countries A common global classification system for RTI severity is not yet available. Most countries divide such injuries into only serious or slight injury for practical convenience. In Switzerland, a subcategory of life-threatening injury is separately recorded, creating a three-category system [6]. Based on the IRTAD annual report and its relevant articles, the task force identified the following criteria used for classifying injury severity. Length of hospitalization Length of hospitalization (LOH) is easily captured in administration databases and used as a proxy of injury severity [7]. Moreover, this indicator reflects the consumption of health resources. However, LOH may be biased by personal factors and healthcare facility policies [8]. Furthermore, the definition of LOH in terms of injury severity may be inconsistent among various health delivery systems. Using the minimum LOH (≥24 h) to define serious injury is not only convenient, but also minimizes the influence of variations in LOH [6, 9]. Table 1 reveals how various countries define serious RTI using LOH. Table 1 Definitions of serious RTI using LOH by various countries or areas   LOH  Argentina, Austria, Belgium, Chile, Germany, Israel, Luxembourg, Portugal, Spain and Switzerland  ≥24 h  France (before 2005), Morocco  ≥6 days  Cambodia  ≥8 days  Hong Kong  ≥12 h    LOH  Argentina, Austria, Belgium, Chile, Germany, Israel, Luxembourg, Portugal, Spain and Switzerland  ≥24 h  France (before 2005), Morocco  ≥6 days  Cambodia  ≥8 days  Hong Kong  ≥12 h  RTIs, road traffic injuries; LOH, length of hospitalization. Table 1 Definitions of serious RTI using LOH by various countries or areas   LOH  Argentina, Austria, Belgium, Chile, Germany, Israel, Luxembourg, Portugal, Spain and Switzerland  ≥24 h  France (before 2005), Morocco  ≥6 days  Cambodia  ≥8 days  Hong Kong  ≥12 h    LOH  Argentina, Austria, Belgium, Chile, Germany, Israel, Luxembourg, Portugal, Spain and Switzerland  ≥24 h  France (before 2005), Morocco  ≥6 days  Cambodia  ≥8 days  Hong Kong  ≥12 h  RTIs, road traffic injuries; LOH, length of hospitalization. Length of treatment Japan and South Korea have adopted length of recovery for classifying injury severity. ED physicians predict the required length of treatment without uniform standards of judgment [6]. Injury type Injury severity may be solely defined by the specific injury type assessed by managing police officers. In the USA, police are required to complete a KABCO scale assessment that features multiple predefined items [10]. Maximum abbreviated injury score In 1969, the Association for the Advancement of Automotive Medicine developed the abbreviated injury score (AIS) to assess RTIs. The human body is divided into six regions: head and neck, face, chest, abdomen, extremities (including pelvis) and external. A 6-point score (1 = minor and 6 = unsurvivable) is used to represent each injury severity level. The maximum AIS (MAIS) is the highest AIS across all body regions [11]. The AIS is calculated according to the treating physician’s diagnosis, reflecting the immediate RTI severity. This ensures that assessments are professional, comparable, and not biased by socioeconomic factors or variations in health delivery systems [10]. In the USA, the MAIS has been reported more consistently between states than the KABCO scale [12]. The IRTAD Group recommended a MAIS of 3 or more (MAIS3+) to define serious injury. Currently, the MAIS is adopted to define serious RTIs in Finland, France and the Netherlands [6]. Data sources of RTI severity in Taiwan Various governmental and non-governmental agencies are responsible for collecting RTI severity data in Taiwan. The injury severity items collected by these agencies were compiled and interpreted by task force according to the members’ experience and knowledge. Upon various expert meetings with other physicians, academics and different government agency representatives, the task force found that little linkage exists among these data sources as will be described below. The injury severity indicators collected by each data sources are illustrated in Fig. 1. Figure 1 View largeDownload slide Sources and contents of RTI severity databases in Taiwan. *Including those who died were discharged or transferred to another hospital. ALS, advanced life support; CPCR, cardiopulmonary-cerebral resuscitation; ED, emergency department; EMS, emergency medical service; GCS, Glasgow Coma Scale; ICD, International Classification of Diseases; ICU, intensive care unit; ISS, injury severity score; LOH, length of hospitalization; NHIRD, National Health Insurance Research Database; RTA, road traffic accident; RTI, road traffic injury. Figure 1 View largeDownload slide Sources and contents of RTI severity databases in Taiwan. *Including those who died were discharged or transferred to another hospital. ALS, advanced life support; CPCR, cardiopulmonary-cerebral resuscitation; ED, emergency department; EMS, emergency medical service; GCS, Glasgow Coma Scale; ICD, International Classification of Diseases; ICU, intensive care unit; ISS, injury severity score; LOH, length of hospitalization; NHIRD, National Health Insurance Research Database; RTA, road traffic accident; RTI, road traffic injury. Police reports Police officers mandatorily complete a report within 3–5 days after an RTI. This report has two parts. The first form records the crash event and the second concerns the involved individuals. Since 1985, all road traffic accidents (RTAs) have been divided into three categories: A1: injured people died at the scene or within 24 h of the RTI; A2: people were injured, including those who died 24 h after the RTI; and A3: only property was damaged. The National Police Agency provides a dataset of these reports to the Department of Statistics of the Ministry of Health and Welfare (MOHW) annually for data deidentification and encryption before releasing the database for academic purposes. Despite the injury severity level (injury or death within 24 h after the accident) being mandatory to record [13], the accuracy of this recorded item has come under question [14, 15]. Ambulance records Emergency medical technicians (EMTs) mandatorily complete the ambulance record [16]. The basic version of the ambulance record was established in 2005 and covers dispatch information, personal information, complaints, rescue procedures and vital signs. Injured patients requiring advanced life support (ALS) are immediately transported to emergent rescuer responsive hospitals. Table 2 lists the ALS criteria. Table 2 Criteria of ALS trauma patientsa Vital signs Consciousness change (GCS<14) Respiratory rate: >29 or <10 per min Pulse rate: >150 or <50 per min Systolic blood pressure: >200 or <90 mmHg Capillary filling time ≥2 s Body temperature >41°C or <32°C Oxygen saturation (SpO2) <90%Trauma mechanism Fall from height >6 m or >2 stories (For children: >3 m or >2 times their body height) Escape time >20 min Crush injury by car over the body trunk Ejected from the vehicle Motor vehicle accident with death in same vehicle Other high-energy mechanisms of injury  Trauma sites Second-degree or third-degree burn injuries on the face, perineum, or >25% of the total body surface area Penetrating injuries or open wounds over the head, neck, trunk, chest, abdomen, groin or the extremities above the elbow or knee joint Massive subcutaneous emphysema Flail chest Exposure of brain tissue or organs Amputation above the wrist or ankle ≥2 Fractures of long bones over thigh or upper arm Pelvic fracture Large and deep wounds Non-palpable pulse over limbs Head or spinal injuries with paralysis Crush injuries or severe lacerations  Vital signs Consciousness change (GCS<14) Respiratory rate: >29 or <10 per min Pulse rate: >150 or <50 per min Systolic blood pressure: >200 or <90 mmHg Capillary filling time ≥2 s Body temperature >41°C or <32°C Oxygen saturation (SpO2) <90%Trauma mechanism Fall from height >6 m or >2 stories (For children: >3 m or >2 times their body height) Escape time >20 min Crush injury by car over the body trunk Ejected from the vehicle Motor vehicle accident with death in same vehicle Other high-energy mechanisms of injury  Trauma sites Second-degree or third-degree burn injuries on the face, perineum, or >25% of the total body surface area Penetrating injuries or open wounds over the head, neck, trunk, chest, abdomen, groin or the extremities above the elbow or knee joint Massive subcutaneous emphysema Flail chest Exposure of brain tissue or organs Amputation above the wrist or ankle ≥2 Fractures of long bones over thigh or upper arm Pelvic fracture Large and deep wounds Non-palpable pulse over limbs Head or spinal injuries with paralysis Crush injuries or severe lacerations  aEMT standard operating procedure manual, National Fire Agency. Table 2 Criteria of ALS trauma patientsa Vital signs Consciousness change (GCS<14) Respiratory rate: >29 or <10 per min Pulse rate: >150 or <50 per min Systolic blood pressure: >200 or <90 mmHg Capillary filling time ≥2 s Body temperature >41°C or <32°C Oxygen saturation (SpO2) <90%Trauma mechanism Fall from height >6 m or >2 stories (For children: >3 m or >2 times their body height) Escape time >20 min Crush injury by car over the body trunk Ejected from the vehicle Motor vehicle accident with death in same vehicle Other high-energy mechanisms of injury  Trauma sites Second-degree or third-degree burn injuries on the face, perineum, or >25% of the total body surface area Penetrating injuries or open wounds over the head, neck, trunk, chest, abdomen, groin or the extremities above the elbow or knee joint Massive subcutaneous emphysema Flail chest Exposure of brain tissue or organs Amputation above the wrist or ankle ≥2 Fractures of long bones over thigh or upper arm Pelvic fracture Large and deep wounds Non-palpable pulse over limbs Head or spinal injuries with paralysis Crush injuries or severe lacerations  Vital signs Consciousness change (GCS<14) Respiratory rate: >29 or <10 per min Pulse rate: >150 or <50 per min Systolic blood pressure: >200 or <90 mmHg Capillary filling time ≥2 s Body temperature >41°C or <32°C Oxygen saturation (SpO2) <90%Trauma mechanism Fall from height >6 m or >2 stories (For children: >3 m or >2 times their body height) Escape time >20 min Crush injury by car over the body trunk Ejected from the vehicle Motor vehicle accident with death in same vehicle Other high-energy mechanisms of injury  Trauma sites Second-degree or third-degree burn injuries on the face, perineum, or >25% of the total body surface area Penetrating injuries or open wounds over the head, neck, trunk, chest, abdomen, groin or the extremities above the elbow or knee joint Massive subcutaneous emphysema Flail chest Exposure of brain tissue or organs Amputation above the wrist or ankle ≥2 Fractures of long bones over thigh or upper arm Pelvic fracture Large and deep wounds Non-palpable pulse over limbs Head or spinal injuries with paralysis Crush injuries or severe lacerations  aEMT standard operating procedure manual, National Fire Agency. ED records Since 2010, all EDs in Taiwan have used the Taiwan Triage and Acuity Scale (TTAS) to calculate patients’ triage level, which is determined using patient complaints, vital signs, pain score and injury mechanism/site [17]. The TTAS score is based on the Canadian Triage and Acuity Scale, and is categorized into five triage levels; the lowest represents the highest acuity of medical attention. The TTAS was reported to have high discriminant validity for patient acuity, and favorable accuracy and efficiency in determining medical priorities [17, 18]. However, the association of the TTAS score with patients’ injury severity is yet to be validated. Hospital medical records These records document a single patient’s medical history and care across time within one particular hospital. Currently, electronic medical records have been widely applied in Taiwan. Discharge diagnoses are mandatorily coded with the International Statistical Classification of Diseases and Related Health Problems (tenth revision; ICD-10-CM) that can be further converted to ISS or MAIS via specific software. Each individual hospital maintains such records. Hospital trauma registries Since 2005, a nationwide trauma registry has been implemented in Taiwan. The registry includes the clinical information of trauma patients hospitalized in EDs or who expired in an ED [19]. In Taiwan, all hospitals classified as moderate and severe emergent rescuer responsive hospitals have mandatorily established their own trauma registries [20]. Currently, nearly 120 hospitals have their own trauma registries; however, these registries are separately stored in various hospitals without effective integration. National Health Insurance Research Database The NHI covers more than 99% of the total population. The National Health Insurance Research Database (NHIRD) is a deidentified database originally designed for recording medical payment claims. Currently, this database is widely used for medical research [21]. Because a 2- to 3-year time lag exists between data upload from hospitals and the academic release of the database, the NHIRD cannot provide timely information on RTI severity. In addition, it does not contain certain indicators of injury severity, such as AIS, injury severity score (ISS), Glasgow Coma Scale (GCS) score, or vital signs. Additionally, the NHIRD only records external codes for describing causes of injury for hospitalized patients; therefore, injured patients solely visiting EDs or outpatient clinics are not identified. Medical expenditure in the NHIRD may be considered a proxy of injury severity. However, this downstream indicator cannot reflect immediate injury severity. Furthermore, it can be influenced by personal characteristics, hospital policies, and healthcare delivery systems; for example, self-paid services are not included in the database. Forensic reports In Taiwan, when a patient’s death is not or suspected not to be from medical or natural causes, the treating physician is obliged to report to the procuratorial authority for forensic investigation before issuing a death certificate. Cause of death data are maintained by the Department of Statistics of the MOHW, and classified according to ICD-10-CM codes. Insurance company records Insurance companies are responsible for paying reparations and insurance benefits for financial loss associated with RTI casualties. The Compulsory Automobile Liability Insurance (CALI) scheme was launched for all motor vehicles in 1999 [22]. The insurance benefits cover personal injury medical expenditures, disabilities, and death. Insurance companies must regularly provide their reparation data to the Financial Supervisory Commission for inclusion in the CALI database. Disability benefits are divided into 15 levels [22]. Disability severity does not correlate well with AIS because a single limb injury may result in a serious disability [23]. Compensation claims and claim amounts have been considered a proxy of injury severity [24, 25]. However, these indicators are indirect and downstream. Data fragmentation in RTI severity Domestic RTI severity data are dispersed among various organizations and hospitals (Fig. 1). They are constructed and maintained according to the requirements of the collecting organizations without horizontal linkage, resulting in data fragmentation that prevents comprehensive understanding and effective utilization. The NHIRD is the only available nationwide database providing hospital information on patients’ RTI severity. However, the lack of RTA records in the NHIRD causes uncertainty regarding the association between accident and hospitalization. Moreover, the NHIRD has a time lag and lacks external codes in EDs or outpatient clinics. It also lacks certain injury severity indicators. Therefore, the task force suggested that it may not be optimal for providing detailed and timely information regarding RTI severity. To reduce fragmentation in RTI severity data, an electronic information platform is being evaluated under the sponsorship of the MOTC, which aggregates information from the police RTI database, CALI database, local governments’ data of accident scene scans, and the MOTC database, including motor vehicle and driver registration, traffic violation records, motor vehicle inspection records, accident assessment reports, and Geographic Information System data [26]. However, linking data of nationwide hospital is crucial to obtain timely medical information on RTI patients. Therefore, an information system to centrally collect nationwide hospital information should be implemented without delay [27]. Discussion Approaches to classifying and collecting RTI severity data A proposed classification of RTI severity Although the IRTAD Group recommends using MAIS3+ for classifying serious RTIs, the MAIS can be obtained only after patients are discharged from hospital, possibly resulting in delayed reporting. Furthermore, the ICD code in hospital medical records or the AIS in trauma registries are all separately stored in various hospitals. An information system can integrate hospital data without increasing labor demand. However, this proposal necessitates not only financing the establishment and maintenance of this system, but also overcoming the process of interdepartmental communication and coordination among different administration systems. By contrast, defining injury severity according to hospitalization for ≥24 h is simple and cost-effective because hospitalized patients consume the majority of medical resources. Because this classification is currently in use in numerous countries, its application will facilitate communication and comparison between countries. Therefore, the task force suggested adopting a binary indicator with hospitalization for ≥24 h to define serious RTIs in the short-term; simultaneously, an information system’s design for integrating hospital data should be started for applying the MAIS [27]. Proposed model for timely collection of RTI severity data In 2010, the World Health Organization (WHO) suggested that an optimal RTI data system should record accurate and detailed information to facilitate timely, evidence-based decisions making [28]. Irrespective of the classification method, obtaining hospital data is essential to achieve this goal. Unlike hospital databases that include only RTI patients who visited hospital, police records include almost all RTIs. High accessibility and affordability of ED services under the NHI mean that the majority of RTI patients visit EDs in the vicinity of the accident. Although patients may be sent to hospital by EMTs before police arrive, the officers routinely interview the patients in the ED. Therefore, the task force proposed an integrated model for timely collecting data on RTI severity. Police officers are at the beginning of this information collection chain. They are responsible only for recording patients’ personal information and crash conditions. The police should not be expected to record injury severity because they lack the requisite professional knowledge; furthermore, they have heavy workloads from numerous light motorcycle accidents. RTA information completed by the police is conveyed to the ED by the managing police officer and passed on to the hospital. The appointed section in the hospital should integrate this information with the hospital’s data, particularly hospitalization status and MAIS score. This integrated information is transferred regularly to road traffic safety agencies for timely analysis of RTI severity. Figure 2 details this proposed model’s flowchart and potential items. Figure 2 View largeDownload slide The proposed model’s flowchart and potential items. AIS, abbreviated injury score; ED, emergency department; ICD-10-CM, International Statistical Classification of Diseases and Related Health Problems (tenth revision); MAIS, maximum abbreviated injury score; RTA, road traffic accident. Figure 2 View largeDownload slide The proposed model’s flowchart and potential items. AIS, abbreviated injury score; ED, emergency department; ICD-10-CM, International Statistical Classification of Diseases and Related Health Problems (tenth revision); MAIS, maximum abbreviated injury score; RTA, road traffic accident. Comparison between proposed and other current models Currently in Taiwan, government RTI statistics are solely derived from police reports. Insufficient training for the classification and heavy workloads markedly impair the accuracy and completeness of RTI severity data within police reports. =The proposed model can be implemented nationwide with limited extra human resources. Furthermore, police officers’ workloads will be decreased by eliminating the burden of differentiating RTI severity. Connecting the hospital information to the victims list provided by police will guarantee that the RTI severity assessment is medically professional. The RTI severity information uploaded from hospitals in a timely manner will provide high-quality data to relevant agencies for resource allocation, public communication, and policy-making. This proposed model can be used for short-term implantable pragmatic solution. Moreover, the practicality of these databases will provide knowledge for establishing a future information system. Due to few national hospital datasets, some European countries use regional trauma registries to estimate national RTI severity [29]. However, the task force believes that information system is the sure method for obtaining accurate and timely data of RTI severity in the medium-term. By building an information system such as the Finnish Hospital Discharge Register to centrally collect national hospital information [26], regular linkage and combination of hospital data with police data can be performed within the government statistical department. Yet, the aforesaid difficulty caused by both interdepartmental communication and coordination might complicate this information system’s construction. Building a shared platform for containing data entered by police and healthcare providers and synchronously combining information concerning the same accident should be long-term goals. This shared platform’s creation provides continuous real-time information of RTI severity to relevant agencies and healthcare providers. Currently, only few developed countries build this platform due to its high resources consumption for daily operations [30]. Table 3 shows these models’ pros and cons. Table 3 The different models’ pros and cons for collecting data of RTI severity Option  Required data  Pro  Cons  Current model in Taiwan  Police data  Avoid interdepartmental communication and coordination Avoid financial burden for building information system  Data accuracy is questionable Lack professional data assessment  Proposed model  Police data Nationwide hospital data  Provide professional data assessment Avoid financial burden for building information system  Need nationwide personnel training Need interdepartmental communication and coordination (minor)  Linkage of police and hospital data  Police data Nationwide hospital data  Provide professional data assessment Avoid extra workload to the first-line personnel  Need financial burden for building information system Need interdepartmental communication and coordination (major)  Synchronized information platform  Police data Nationwide hospital data  Provide professional data assessment Achieve real-time data integration of individual RTI  Need financial burden for building information system Need interdepartmental communication and coordination (major) Need high resources consumption for daily operation  Option  Required data  Pro  Cons  Current model in Taiwan  Police data  Avoid interdepartmental communication and coordination Avoid financial burden for building information system  Data accuracy is questionable Lack professional data assessment  Proposed model  Police data Nationwide hospital data  Provide professional data assessment Avoid financial burden for building information system  Need nationwide personnel training Need interdepartmental communication and coordination (minor)  Linkage of police and hospital data  Police data Nationwide hospital data  Provide professional data assessment Avoid extra workload to the first-line personnel  Need financial burden for building information system Need interdepartmental communication and coordination (major)  Synchronized information platform  Police data Nationwide hospital data  Provide professional data assessment Achieve real-time data integration of individual RTI  Need financial burden for building information system Need interdepartmental communication and coordination (major) Need high resources consumption for daily operation  RTI, road traffic injury. Table 3 The different models’ pros and cons for collecting data of RTI severity Option  Required data  Pro  Cons  Current model in Taiwan  Police data  Avoid interdepartmental communication and coordination Avoid financial burden for building information system  Data accuracy is questionable Lack professional data assessment  Proposed model  Police data Nationwide hospital data  Provide professional data assessment Avoid financial burden for building information system  Need nationwide personnel training Need interdepartmental communication and coordination (minor)  Linkage of police and hospital data  Police data Nationwide hospital data  Provide professional data assessment Avoid extra workload to the first-line personnel  Need financial burden for building information system Need interdepartmental communication and coordination (major)  Synchronized information platform  Police data Nationwide hospital data  Provide professional data assessment Achieve real-time data integration of individual RTI  Need financial burden for building information system Need interdepartmental communication and coordination (major) Need high resources consumption for daily operation  Option  Required data  Pro  Cons  Current model in Taiwan  Police data  Avoid interdepartmental communication and coordination Avoid financial burden for building information system  Data accuracy is questionable Lack professional data assessment  Proposed model  Police data Nationwide hospital data  Provide professional data assessment Avoid financial burden for building information system  Need nationwide personnel training Need interdepartmental communication and coordination (minor)  Linkage of police and hospital data  Police data Nationwide hospital data  Provide professional data assessment Avoid extra workload to the first-line personnel  Need financial burden for building information system Need interdepartmental communication and coordination (major)  Synchronized information platform  Police data Nationwide hospital data  Provide professional data assessment Achieve real-time data integration of individual RTI  Need financial burden for building information system Need interdepartmental communication and coordination (major) Need high resources consumption for daily operation  RTI, road traffic injury. Conclusion An online injury severity data acquisition model that connects RTA and hospital data can improve road traffic safety by instantly providing real-time information about crash patterns, locations, environment, causes and resulting injury severity to road traffic safety agencies. For healthcare providers, this model is vital because it can supply information about the potential injury severity and medical outcomes of patients exposed to various RTA patterns. This model not only alleviates injuries’ damage to the prognosis of RTI patients, but also provides critical references for medical resource allocation in prehospital and ED settings. To achieve police-hospital data linkage, an information system to collect nationwide hospital data should be built without any delay. Our study indicated that the timely integration of concurrent RTA and hospital data will effectively reduce police manpower consumption and provide timely and accurate RTI severity information to road safety agencies and academicians. Furthermore, our study suggested that improving RTI data flow through connecting injury severity and medical outcome information are vital to healthcare providers for improving both the RTI patients’ management and prognoses. Acknowledgements We thank Chu-Hsuan Cheng for assisting with this study. Funding This work was supported by the Institute of Transportation, Ministry of Transportation and Communications, Executive Yuan [grant numbers MOTC-IOT-104-SEB009 and MOTC-IOT-105-SEB010] and Wan Fang Hospital [grant numbers 106-wf-phd-03 and 107-wf-phd-03]. The results and conclusions are solely those of the authors and do not represent the opinions of the Institute of Transportation, Ministry of Transportation and Communications. References 1 World Health Organization. Global status report on road safety 2015 . Geneva, Switzerland: WHO, 2015. 2 National Police Agency, Ministry of the Interior, Taiwan, ROC. International comparison analysis of road traffic fatalities (in Chinese). https://www.npa.gov.tw/NPAGip/wSite/ct?xItem=77625&ctNode=12598&mp=1. Accessed August 26, 2017 3 Ministry of Transportation and Communications. Road Traffic Safety Annual Report 2014 (in Chinese) . Taipei, Taiwan, ROC: MOTC, 2015. 4 Pan RH, Chang NT, Chu D et al.  . Epidemiology of orthopedic fractures and other injuries among inpatients admitted due to traffic accidents: a 10-year nationwide survey in Taiwan. Scientific World Journal  2014; 2014: 637872. Google Scholar PubMed  5 Lai CH, Doong JL, Huang WS et al.  . Mortality burden from motor vehicle accidents in Taiwan. Public Health Nurs  2008; 25: 28– 36. Google Scholar CrossRef Search ADS PubMed  6 OECD/ITF. Road Safety Annual Report 2017 . Paris: OECD Publishing, 2017. 7 Newgard CD, Fleischman R, Choo E et al.  . Validation of length of hospital stay as a surrogate measure for injury severity and resource use among injury survivors. Acad Emerg Med  2010; 17: 142– 50. Google Scholar CrossRef Search ADS PubMed  8 Cryer C, Gulliver P, Langley JD et al.  . Is length of stay in hospital a stable proxy for injury severity? Inj Prev  2010; 16: 254– 60. Google Scholar CrossRef Search ADS PubMed  9 Adminaite D, Jost G, Stipdonk H et al.  . Ranking EU progress on road safety: 11th Road Safety Performance Index Report. etsc.eu/wp-content/uploads/PIN_ANNUAL_REPORT_2017-final.pdf. Accessed September 18, 2017. 10 OECD/ITF. Reporting on serious road traffic casualties: combining and using different data sources to improve understanding of non-fatal road traffic crashes . Paris: International Transport Forum, 2011. 11 Gennarelli TA, Wodzin E. AIS 2005: a contemporary injury scale. Injury  2006; 37: 1083– 91. Google Scholar CrossRef Search ADS PubMed  12 Burch C, Cook L, Dischinger P. A comparison of KABCO and AIS injury severity metrics using CODES linked data. Traffic Inj Prev  2014; 15: 627– 30. Google Scholar CrossRef Search ADS PubMed  13 Institute of Transportation, Ministry of Transportation and Communications, Taiwan, ROC. Road traffic accident report. http://talas-pub.iot.gov.tw/. Accessed August 26, 2017. 14 Chen WH, Huang WS, Lai CH Exploration of the relationship between the drivers’ injury severity and their expenses during hospitalization (in Chinese). In: Proceedings of the Traffic Safety and Law Enforecement Conference, Taiwan, 2004. 15 Lin MR, Kraus JF. Methodological issues in motorcycle injury epidemiology. Accid Anal Prev  2008; 40: 1653– 60. Google Scholar CrossRef Search ADS PubMed  16. Emergency Medical Services Act. Taiwan, ROC: Ministry of Health and Welfare, 2013. 17 Ng CJ, Yen ZS, Tsai JC et al.  . Validation of the Taiwan triage and acuity scale: a new computerised five-level triage system. Emerg Med J  2011; 28: 1026– 31. Google Scholar CrossRef Search ADS PubMed  18 Chi CH, Huang CM. Comparison of the Emergency Severity Index (ESI) and the Taiwan Triage System in predicting resource utilization. J Formos Med Assoc  2006; 105: 617– 25. Google Scholar CrossRef Search ADS PubMed  19 Formosa Association for the Surgery of Trauma. The operating manual for trauma registry system (in Chinese). Taichung, Taiwan, ROC: FAST, 2008. 20 Ministry of Health and Welfare. Accreditation standards for the hierarchical system of emergent rescuer responsive hospitals. Taiwan, ROC: Ministry of Health and Welfare, 2017. 21 Hsing AW, Ioannidis JA. Nationwide population science: lessons from the Taiwan National Health Insurance Research Database. JAMA Intern Med  2015; 175: 1527– 9. Google Scholar CrossRef Search ADS PubMed  22 Taiwan Insurance Institute. Compulsory Automobile Liability Insurance. http://www.cali.org.tw/en/about.aspx. Accessed September 9, 2016. 23 Bull JP. Disabilities caused by road traffic accidents and their relation to severity scores. Accid Anal Prev  1985; 17: 387– 97. Google Scholar CrossRef Search ADS PubMed  24 Smith P, Hogg-Johnson S, Mustard C et al.  . Comparing the risk factors associated with serious versus and less serious work-related injuries in Ontario between 1991 and 2006. Am J Ind Med  2012; 55: 84– 91. Google Scholar CrossRef Search ADS PubMed  25 Ministry of Transportation and Communications, Taiwan, ROC. Developing an information platform for integration and analysis on road traffic safety data (in Chinese). https://www.grb.gov.tw/search/planDetail?id=12080384&docId=502353. Accessed March 29, 2018. 26 Ramaswamy SK, Mosher GA. Using workers’ compensation claims data to characterize occupational injuries in the commercial grain elevator industry. J Agric Saf Health  2017; 33: 203– 17. Google Scholar CrossRef Search ADS   27 Härö AS. The centralized statistics on hospital utilization in Finland. Methods Inf Med Suppl  1969; 4: 277– 87. Google Scholar PubMed  28 World Health Organization. Data systems: a road safety manual for decision-makers and practitioners . Geneva, Switzerland: WHO, 2010. 29 Amoros E, Martin JL, Lafont S et al.  . Actual incidences of road casualties, and their injury severity, modelled from police and hospital data, France. Eur J Public Health  2008; 18: 360– 5. Google Scholar CrossRef Search ADS PubMed  30 Howard C, Linder A. Review of Swedish experiences concerning analysis of people injured in traffic accidents . Linköping, Sweden: Swedish National Road and Transport Research Institute, 2014. © The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal for Quality in Health Care Oxford University Press

Improving the timeliness and accuracy of injury severity data in road traffic accidents in an emerging economy setting

Loading next page...
 
/lp/ou_press/improving-the-timeliness-and-accuracy-of-injury-severity-data-in-road-OzLWqIyDX0
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1353-4505
eISSN
1464-3677
D.O.I.
10.1093/intqhc/mzy115
Publisher site
See Article on Publisher Site

Abstract

Abstract Road traffic injuries (RTIs) are among the leading causes of injury and fatality worldwide. RTI casualties are continually increasing in Taiwan; however, because of a lack of an advanced method for classifying RTI severity data, as well as the fragmentation of data sources, road traffic safety and health agencies encounter difficulties in analyzing RTIs and their burden on the healthcare system and national resources. These difficulties lead to blind spots during policy-making for RTI prevention and control. After compiling classifications applied in various countries, we summarized data sources for RTI severity in Taiwan, through which we identified data fragmentation. Accordingly, we proposed a practical classification for RTI severity, as well as a feasible model for collecting and integrating these data nationwide. This model can provide timely relevant data recorded by medical professionals and is valuable to healthcare providers. The proposed model’s pros and cons are also compared to those of other current models. road traffic injury, injury prevention, injury severity, data quality Introduction Road traffic injuries in Taiwan Road traffic injuries (RTIs) are major causes of injury worldwide [1]. In recent two decades, the number of motor vehicles, particularly motorcycles, has grown considerably in Taiwan. Confusion caused by a mixed traffic flow of cars, motorcycles and trucks, is exacerbated by dangerous driving habits. These factors cause the RTI death rate in Taiwan to substantially exceed that in other developed countries [2]. In 2014, 413 229 RTI were reported, representing a 421.3% increase within 14 years. Among those in 2014, almost 60% were from motorcycle accidents [3]. The annual medical cost of RTI-related orthopedic fractures almost doubled between 2002 and 2011 [4]. In 2002, total years of life lost from RTIs was reported as 70 006, with individuals aged 15–29 years being the largest category [5]. Because of high accessibility to acute care facilities under the Taiwanese National Health Insurance (NHI) system, a majority of RTI casualties visit emergency departments (EDs), including patients with slight injury. However, persistently increasing numbers of injured patients will eventually exhaust police and emergency care providers and consume substantial healthcare resources. Difficulties caused by lack of effective classification Understanding injury severity is crucial for preventing and controlling RTIs. Injury severity measures can provide more detailed information to distinguish varying degrees of RTI severity, as well as their effect on national resources, both of which are essential for policy planning. Furthermore, surveilling RTI severity can assess the likelihood of the consumption of certain medical resources, such as hospitalization or surgery, as well as intervention effectiveness. This feedback can assist healthcare providers in resources allocation and patient management. Currently, all RTIs within police reports are classified into injury or death according to patients’ survival status within 24 h after the RTI. The lack of a more detailed classification system hinders relevant agencies in conducting epidemiological analyses and impact assessments. If RTI severity can be differentiated using a straightforward evidence-based classification system, then agencies can use this information for injury prevention and control. Methods To correct the shortcomings in classifying RTI severity, a task force comprising physicians and academicians researching injury prevention and road traffic safety was assembled by the Institute of Transportation of the Ministry of Transportation and Communication (MOTC). With the Road Safety Annual Report 2017 of the International Traffic Safety Data and Analysis Group (IRTAD Group) as a reference, the task force summarized the RTI severity classification in various countries. This Annual Report used information from many sources such as transportation safety-focused government agencies, non-governmental organizations, along with academic and industrial ones. It detailed the RTI severity classifications in 40 member and observer countries. Upon many group discussions and expert meetings, the task force proposed a feasible classification for application in Taiwan. Furthermore, the task force reported the current situation regarding the fragmentation of domestic RTI severity data and its influences, along with proposals to meet the demand for accurate and timely data monitoring. Results Classification of RTI severity in other countries A common global classification system for RTI severity is not yet available. Most countries divide such injuries into only serious or slight injury for practical convenience. In Switzerland, a subcategory of life-threatening injury is separately recorded, creating a three-category system [6]. Based on the IRTAD annual report and its relevant articles, the task force identified the following criteria used for classifying injury severity. Length of hospitalization Length of hospitalization (LOH) is easily captured in administration databases and used as a proxy of injury severity [7]. Moreover, this indicator reflects the consumption of health resources. However, LOH may be biased by personal factors and healthcare facility policies [8]. Furthermore, the definition of LOH in terms of injury severity may be inconsistent among various health delivery systems. Using the minimum LOH (≥24 h) to define serious injury is not only convenient, but also minimizes the influence of variations in LOH [6, 9]. Table 1 reveals how various countries define serious RTI using LOH. Table 1 Definitions of serious RTI using LOH by various countries or areas   LOH  Argentina, Austria, Belgium, Chile, Germany, Israel, Luxembourg, Portugal, Spain and Switzerland  ≥24 h  France (before 2005), Morocco  ≥6 days  Cambodia  ≥8 days  Hong Kong  ≥12 h    LOH  Argentina, Austria, Belgium, Chile, Germany, Israel, Luxembourg, Portugal, Spain and Switzerland  ≥24 h  France (before 2005), Morocco  ≥6 days  Cambodia  ≥8 days  Hong Kong  ≥12 h  RTIs, road traffic injuries; LOH, length of hospitalization. Table 1 Definitions of serious RTI using LOH by various countries or areas   LOH  Argentina, Austria, Belgium, Chile, Germany, Israel, Luxembourg, Portugal, Spain and Switzerland  ≥24 h  France (before 2005), Morocco  ≥6 days  Cambodia  ≥8 days  Hong Kong  ≥12 h    LOH  Argentina, Austria, Belgium, Chile, Germany, Israel, Luxembourg, Portugal, Spain and Switzerland  ≥24 h  France (before 2005), Morocco  ≥6 days  Cambodia  ≥8 days  Hong Kong  ≥12 h  RTIs, road traffic injuries; LOH, length of hospitalization. Length of treatment Japan and South Korea have adopted length of recovery for classifying injury severity. ED physicians predict the required length of treatment without uniform standards of judgment [6]. Injury type Injury severity may be solely defined by the specific injury type assessed by managing police officers. In the USA, police are required to complete a KABCO scale assessment that features multiple predefined items [10]. Maximum abbreviated injury score In 1969, the Association for the Advancement of Automotive Medicine developed the abbreviated injury score (AIS) to assess RTIs. The human body is divided into six regions: head and neck, face, chest, abdomen, extremities (including pelvis) and external. A 6-point score (1 = minor and 6 = unsurvivable) is used to represent each injury severity level. The maximum AIS (MAIS) is the highest AIS across all body regions [11]. The AIS is calculated according to the treating physician’s diagnosis, reflecting the immediate RTI severity. This ensures that assessments are professional, comparable, and not biased by socioeconomic factors or variations in health delivery systems [10]. In the USA, the MAIS has been reported more consistently between states than the KABCO scale [12]. The IRTAD Group recommended a MAIS of 3 or more (MAIS3+) to define serious injury. Currently, the MAIS is adopted to define serious RTIs in Finland, France and the Netherlands [6]. Data sources of RTI severity in Taiwan Various governmental and non-governmental agencies are responsible for collecting RTI severity data in Taiwan. The injury severity items collected by these agencies were compiled and interpreted by task force according to the members’ experience and knowledge. Upon various expert meetings with other physicians, academics and different government agency representatives, the task force found that little linkage exists among these data sources as will be described below. The injury severity indicators collected by each data sources are illustrated in Fig. 1. Figure 1 View largeDownload slide Sources and contents of RTI severity databases in Taiwan. *Including those who died were discharged or transferred to another hospital. ALS, advanced life support; CPCR, cardiopulmonary-cerebral resuscitation; ED, emergency department; EMS, emergency medical service; GCS, Glasgow Coma Scale; ICD, International Classification of Diseases; ICU, intensive care unit; ISS, injury severity score; LOH, length of hospitalization; NHIRD, National Health Insurance Research Database; RTA, road traffic accident; RTI, road traffic injury. Figure 1 View largeDownload slide Sources and contents of RTI severity databases in Taiwan. *Including those who died were discharged or transferred to another hospital. ALS, advanced life support; CPCR, cardiopulmonary-cerebral resuscitation; ED, emergency department; EMS, emergency medical service; GCS, Glasgow Coma Scale; ICD, International Classification of Diseases; ICU, intensive care unit; ISS, injury severity score; LOH, length of hospitalization; NHIRD, National Health Insurance Research Database; RTA, road traffic accident; RTI, road traffic injury. Police reports Police officers mandatorily complete a report within 3–5 days after an RTI. This report has two parts. The first form records the crash event and the second concerns the involved individuals. Since 1985, all road traffic accidents (RTAs) have been divided into three categories: A1: injured people died at the scene or within 24 h of the RTI; A2: people were injured, including those who died 24 h after the RTI; and A3: only property was damaged. The National Police Agency provides a dataset of these reports to the Department of Statistics of the Ministry of Health and Welfare (MOHW) annually for data deidentification and encryption before releasing the database for academic purposes. Despite the injury severity level (injury or death within 24 h after the accident) being mandatory to record [13], the accuracy of this recorded item has come under question [14, 15]. Ambulance records Emergency medical technicians (EMTs) mandatorily complete the ambulance record [16]. The basic version of the ambulance record was established in 2005 and covers dispatch information, personal information, complaints, rescue procedures and vital signs. Injured patients requiring advanced life support (ALS) are immediately transported to emergent rescuer responsive hospitals. Table 2 lists the ALS criteria. Table 2 Criteria of ALS trauma patientsa Vital signs Consciousness change (GCS<14) Respiratory rate: >29 or <10 per min Pulse rate: >150 or <50 per min Systolic blood pressure: >200 or <90 mmHg Capillary filling time ≥2 s Body temperature >41°C or <32°C Oxygen saturation (SpO2) <90%Trauma mechanism Fall from height >6 m or >2 stories (For children: >3 m or >2 times their body height) Escape time >20 min Crush injury by car over the body trunk Ejected from the vehicle Motor vehicle accident with death in same vehicle Other high-energy mechanisms of injury  Trauma sites Second-degree or third-degree burn injuries on the face, perineum, or >25% of the total body surface area Penetrating injuries or open wounds over the head, neck, trunk, chest, abdomen, groin or the extremities above the elbow or knee joint Massive subcutaneous emphysema Flail chest Exposure of brain tissue or organs Amputation above the wrist or ankle ≥2 Fractures of long bones over thigh or upper arm Pelvic fracture Large and deep wounds Non-palpable pulse over limbs Head or spinal injuries with paralysis Crush injuries or severe lacerations  Vital signs Consciousness change (GCS<14) Respiratory rate: >29 or <10 per min Pulse rate: >150 or <50 per min Systolic blood pressure: >200 or <90 mmHg Capillary filling time ≥2 s Body temperature >41°C or <32°C Oxygen saturation (SpO2) <90%Trauma mechanism Fall from height >6 m or >2 stories (For children: >3 m or >2 times their body height) Escape time >20 min Crush injury by car over the body trunk Ejected from the vehicle Motor vehicle accident with death in same vehicle Other high-energy mechanisms of injury  Trauma sites Second-degree or third-degree burn injuries on the face, perineum, or >25% of the total body surface area Penetrating injuries or open wounds over the head, neck, trunk, chest, abdomen, groin or the extremities above the elbow or knee joint Massive subcutaneous emphysema Flail chest Exposure of brain tissue or organs Amputation above the wrist or ankle ≥2 Fractures of long bones over thigh or upper arm Pelvic fracture Large and deep wounds Non-palpable pulse over limbs Head or spinal injuries with paralysis Crush injuries or severe lacerations  aEMT standard operating procedure manual, National Fire Agency. Table 2 Criteria of ALS trauma patientsa Vital signs Consciousness change (GCS<14) Respiratory rate: >29 or <10 per min Pulse rate: >150 or <50 per min Systolic blood pressure: >200 or <90 mmHg Capillary filling time ≥2 s Body temperature >41°C or <32°C Oxygen saturation (SpO2) <90%Trauma mechanism Fall from height >6 m or >2 stories (For children: >3 m or >2 times their body height) Escape time >20 min Crush injury by car over the body trunk Ejected from the vehicle Motor vehicle accident with death in same vehicle Other high-energy mechanisms of injury  Trauma sites Second-degree or third-degree burn injuries on the face, perineum, or >25% of the total body surface area Penetrating injuries or open wounds over the head, neck, trunk, chest, abdomen, groin or the extremities above the elbow or knee joint Massive subcutaneous emphysema Flail chest Exposure of brain tissue or organs Amputation above the wrist or ankle ≥2 Fractures of long bones over thigh or upper arm Pelvic fracture Large and deep wounds Non-palpable pulse over limbs Head or spinal injuries with paralysis Crush injuries or severe lacerations  Vital signs Consciousness change (GCS<14) Respiratory rate: >29 or <10 per min Pulse rate: >150 or <50 per min Systolic blood pressure: >200 or <90 mmHg Capillary filling time ≥2 s Body temperature >41°C or <32°C Oxygen saturation (SpO2) <90%Trauma mechanism Fall from height >6 m or >2 stories (For children: >3 m or >2 times their body height) Escape time >20 min Crush injury by car over the body trunk Ejected from the vehicle Motor vehicle accident with death in same vehicle Other high-energy mechanisms of injury  Trauma sites Second-degree or third-degree burn injuries on the face, perineum, or >25% of the total body surface area Penetrating injuries or open wounds over the head, neck, trunk, chest, abdomen, groin or the extremities above the elbow or knee joint Massive subcutaneous emphysema Flail chest Exposure of brain tissue or organs Amputation above the wrist or ankle ≥2 Fractures of long bones over thigh or upper arm Pelvic fracture Large and deep wounds Non-palpable pulse over limbs Head or spinal injuries with paralysis Crush injuries or severe lacerations  aEMT standard operating procedure manual, National Fire Agency. ED records Since 2010, all EDs in Taiwan have used the Taiwan Triage and Acuity Scale (TTAS) to calculate patients’ triage level, which is determined using patient complaints, vital signs, pain score and injury mechanism/site [17]. The TTAS score is based on the Canadian Triage and Acuity Scale, and is categorized into five triage levels; the lowest represents the highest acuity of medical attention. The TTAS was reported to have high discriminant validity for patient acuity, and favorable accuracy and efficiency in determining medical priorities [17, 18]. However, the association of the TTAS score with patients’ injury severity is yet to be validated. Hospital medical records These records document a single patient’s medical history and care across time within one particular hospital. Currently, electronic medical records have been widely applied in Taiwan. Discharge diagnoses are mandatorily coded with the International Statistical Classification of Diseases and Related Health Problems (tenth revision; ICD-10-CM) that can be further converted to ISS or MAIS via specific software. Each individual hospital maintains such records. Hospital trauma registries Since 2005, a nationwide trauma registry has been implemented in Taiwan. The registry includes the clinical information of trauma patients hospitalized in EDs or who expired in an ED [19]. In Taiwan, all hospitals classified as moderate and severe emergent rescuer responsive hospitals have mandatorily established their own trauma registries [20]. Currently, nearly 120 hospitals have their own trauma registries; however, these registries are separately stored in various hospitals without effective integration. National Health Insurance Research Database The NHI covers more than 99% of the total population. The National Health Insurance Research Database (NHIRD) is a deidentified database originally designed for recording medical payment claims. Currently, this database is widely used for medical research [21]. Because a 2- to 3-year time lag exists between data upload from hospitals and the academic release of the database, the NHIRD cannot provide timely information on RTI severity. In addition, it does not contain certain indicators of injury severity, such as AIS, injury severity score (ISS), Glasgow Coma Scale (GCS) score, or vital signs. Additionally, the NHIRD only records external codes for describing causes of injury for hospitalized patients; therefore, injured patients solely visiting EDs or outpatient clinics are not identified. Medical expenditure in the NHIRD may be considered a proxy of injury severity. However, this downstream indicator cannot reflect immediate injury severity. Furthermore, it can be influenced by personal characteristics, hospital policies, and healthcare delivery systems; for example, self-paid services are not included in the database. Forensic reports In Taiwan, when a patient’s death is not or suspected not to be from medical or natural causes, the treating physician is obliged to report to the procuratorial authority for forensic investigation before issuing a death certificate. Cause of death data are maintained by the Department of Statistics of the MOHW, and classified according to ICD-10-CM codes. Insurance company records Insurance companies are responsible for paying reparations and insurance benefits for financial loss associated with RTI casualties. The Compulsory Automobile Liability Insurance (CALI) scheme was launched for all motor vehicles in 1999 [22]. The insurance benefits cover personal injury medical expenditures, disabilities, and death. Insurance companies must regularly provide their reparation data to the Financial Supervisory Commission for inclusion in the CALI database. Disability benefits are divided into 15 levels [22]. Disability severity does not correlate well with AIS because a single limb injury may result in a serious disability [23]. Compensation claims and claim amounts have been considered a proxy of injury severity [24, 25]. However, these indicators are indirect and downstream. Data fragmentation in RTI severity Domestic RTI severity data are dispersed among various organizations and hospitals (Fig. 1). They are constructed and maintained according to the requirements of the collecting organizations without horizontal linkage, resulting in data fragmentation that prevents comprehensive understanding and effective utilization. The NHIRD is the only available nationwide database providing hospital information on patients’ RTI severity. However, the lack of RTA records in the NHIRD causes uncertainty regarding the association between accident and hospitalization. Moreover, the NHIRD has a time lag and lacks external codes in EDs or outpatient clinics. It also lacks certain injury severity indicators. Therefore, the task force suggested that it may not be optimal for providing detailed and timely information regarding RTI severity. To reduce fragmentation in RTI severity data, an electronic information platform is being evaluated under the sponsorship of the MOTC, which aggregates information from the police RTI database, CALI database, local governments’ data of accident scene scans, and the MOTC database, including motor vehicle and driver registration, traffic violation records, motor vehicle inspection records, accident assessment reports, and Geographic Information System data [26]. However, linking data of nationwide hospital is crucial to obtain timely medical information on RTI patients. Therefore, an information system to centrally collect nationwide hospital information should be implemented without delay [27]. Discussion Approaches to classifying and collecting RTI severity data A proposed classification of RTI severity Although the IRTAD Group recommends using MAIS3+ for classifying serious RTIs, the MAIS can be obtained only after patients are discharged from hospital, possibly resulting in delayed reporting. Furthermore, the ICD code in hospital medical records or the AIS in trauma registries are all separately stored in various hospitals. An information system can integrate hospital data without increasing labor demand. However, this proposal necessitates not only financing the establishment and maintenance of this system, but also overcoming the process of interdepartmental communication and coordination among different administration systems. By contrast, defining injury severity according to hospitalization for ≥24 h is simple and cost-effective because hospitalized patients consume the majority of medical resources. Because this classification is currently in use in numerous countries, its application will facilitate communication and comparison between countries. Therefore, the task force suggested adopting a binary indicator with hospitalization for ≥24 h to define serious RTIs in the short-term; simultaneously, an information system’s design for integrating hospital data should be started for applying the MAIS [27]. Proposed model for timely collection of RTI severity data In 2010, the World Health Organization (WHO) suggested that an optimal RTI data system should record accurate and detailed information to facilitate timely, evidence-based decisions making [28]. Irrespective of the classification method, obtaining hospital data is essential to achieve this goal. Unlike hospital databases that include only RTI patients who visited hospital, police records include almost all RTIs. High accessibility and affordability of ED services under the NHI mean that the majority of RTI patients visit EDs in the vicinity of the accident. Although patients may be sent to hospital by EMTs before police arrive, the officers routinely interview the patients in the ED. Therefore, the task force proposed an integrated model for timely collecting data on RTI severity. Police officers are at the beginning of this information collection chain. They are responsible only for recording patients’ personal information and crash conditions. The police should not be expected to record injury severity because they lack the requisite professional knowledge; furthermore, they have heavy workloads from numerous light motorcycle accidents. RTA information completed by the police is conveyed to the ED by the managing police officer and passed on to the hospital. The appointed section in the hospital should integrate this information with the hospital’s data, particularly hospitalization status and MAIS score. This integrated information is transferred regularly to road traffic safety agencies for timely analysis of RTI severity. Figure 2 details this proposed model’s flowchart and potential items. Figure 2 View largeDownload slide The proposed model’s flowchart and potential items. AIS, abbreviated injury score; ED, emergency department; ICD-10-CM, International Statistical Classification of Diseases and Related Health Problems (tenth revision); MAIS, maximum abbreviated injury score; RTA, road traffic accident. Figure 2 View largeDownload slide The proposed model’s flowchart and potential items. AIS, abbreviated injury score; ED, emergency department; ICD-10-CM, International Statistical Classification of Diseases and Related Health Problems (tenth revision); MAIS, maximum abbreviated injury score; RTA, road traffic accident. Comparison between proposed and other current models Currently in Taiwan, government RTI statistics are solely derived from police reports. Insufficient training for the classification and heavy workloads markedly impair the accuracy and completeness of RTI severity data within police reports. =The proposed model can be implemented nationwide with limited extra human resources. Furthermore, police officers’ workloads will be decreased by eliminating the burden of differentiating RTI severity. Connecting the hospital information to the victims list provided by police will guarantee that the RTI severity assessment is medically professional. The RTI severity information uploaded from hospitals in a timely manner will provide high-quality data to relevant agencies for resource allocation, public communication, and policy-making. This proposed model can be used for short-term implantable pragmatic solution. Moreover, the practicality of these databases will provide knowledge for establishing a future information system. Due to few national hospital datasets, some European countries use regional trauma registries to estimate national RTI severity [29]. However, the task force believes that information system is the sure method for obtaining accurate and timely data of RTI severity in the medium-term. By building an information system such as the Finnish Hospital Discharge Register to centrally collect national hospital information [26], regular linkage and combination of hospital data with police data can be performed within the government statistical department. Yet, the aforesaid difficulty caused by both interdepartmental communication and coordination might complicate this information system’s construction. Building a shared platform for containing data entered by police and healthcare providers and synchronously combining information concerning the same accident should be long-term goals. This shared platform’s creation provides continuous real-time information of RTI severity to relevant agencies and healthcare providers. Currently, only few developed countries build this platform due to its high resources consumption for daily operations [30]. Table 3 shows these models’ pros and cons. Table 3 The different models’ pros and cons for collecting data of RTI severity Option  Required data  Pro  Cons  Current model in Taiwan  Police data  Avoid interdepartmental communication and coordination Avoid financial burden for building information system  Data accuracy is questionable Lack professional data assessment  Proposed model  Police data Nationwide hospital data  Provide professional data assessment Avoid financial burden for building information system  Need nationwide personnel training Need interdepartmental communication and coordination (minor)  Linkage of police and hospital data  Police data Nationwide hospital data  Provide professional data assessment Avoid extra workload to the first-line personnel  Need financial burden for building information system Need interdepartmental communication and coordination (major)  Synchronized information platform  Police data Nationwide hospital data  Provide professional data assessment Achieve real-time data integration of individual RTI  Need financial burden for building information system Need interdepartmental communication and coordination (major) Need high resources consumption for daily operation  Option  Required data  Pro  Cons  Current model in Taiwan  Police data  Avoid interdepartmental communication and coordination Avoid financial burden for building information system  Data accuracy is questionable Lack professional data assessment  Proposed model  Police data Nationwide hospital data  Provide professional data assessment Avoid financial burden for building information system  Need nationwide personnel training Need interdepartmental communication and coordination (minor)  Linkage of police and hospital data  Police data Nationwide hospital data  Provide professional data assessment Avoid extra workload to the first-line personnel  Need financial burden for building information system Need interdepartmental communication and coordination (major)  Synchronized information platform  Police data Nationwide hospital data  Provide professional data assessment Achieve real-time data integration of individual RTI  Need financial burden for building information system Need interdepartmental communication and coordination (major) Need high resources consumption for daily operation  RTI, road traffic injury. Table 3 The different models’ pros and cons for collecting data of RTI severity Option  Required data  Pro  Cons  Current model in Taiwan  Police data  Avoid interdepartmental communication and coordination Avoid financial burden for building information system  Data accuracy is questionable Lack professional data assessment  Proposed model  Police data Nationwide hospital data  Provide professional data assessment Avoid financial burden for building information system  Need nationwide personnel training Need interdepartmental communication and coordination (minor)  Linkage of police and hospital data  Police data Nationwide hospital data  Provide professional data assessment Avoid extra workload to the first-line personnel  Need financial burden for building information system Need interdepartmental communication and coordination (major)  Synchronized information platform  Police data Nationwide hospital data  Provide professional data assessment Achieve real-time data integration of individual RTI  Need financial burden for building information system Need interdepartmental communication and coordination (major) Need high resources consumption for daily operation  Option  Required data  Pro  Cons  Current model in Taiwan  Police data  Avoid interdepartmental communication and coordination Avoid financial burden for building information system  Data accuracy is questionable Lack professional data assessment  Proposed model  Police data Nationwide hospital data  Provide professional data assessment Avoid financial burden for building information system  Need nationwide personnel training Need interdepartmental communication and coordination (minor)  Linkage of police and hospital data  Police data Nationwide hospital data  Provide professional data assessment Avoid extra workload to the first-line personnel  Need financial burden for building information system Need interdepartmental communication and coordination (major)  Synchronized information platform  Police data Nationwide hospital data  Provide professional data assessment Achieve real-time data integration of individual RTI  Need financial burden for building information system Need interdepartmental communication and coordination (major) Need high resources consumption for daily operation  RTI, road traffic injury. Conclusion An online injury severity data acquisition model that connects RTA and hospital data can improve road traffic safety by instantly providing real-time information about crash patterns, locations, environment, causes and resulting injury severity to road traffic safety agencies. For healthcare providers, this model is vital because it can supply information about the potential injury severity and medical outcomes of patients exposed to various RTA patterns. This model not only alleviates injuries’ damage to the prognosis of RTI patients, but also provides critical references for medical resource allocation in prehospital and ED settings. To achieve police-hospital data linkage, an information system to collect nationwide hospital data should be built without any delay. Our study indicated that the timely integration of concurrent RTA and hospital data will effectively reduce police manpower consumption and provide timely and accurate RTI severity information to road safety agencies and academicians. Furthermore, our study suggested that improving RTI data flow through connecting injury severity and medical outcome information are vital to healthcare providers for improving both the RTI patients’ management and prognoses. Acknowledgements We thank Chu-Hsuan Cheng for assisting with this study. Funding This work was supported by the Institute of Transportation, Ministry of Transportation and Communications, Executive Yuan [grant numbers MOTC-IOT-104-SEB009 and MOTC-IOT-105-SEB010] and Wan Fang Hospital [grant numbers 106-wf-phd-03 and 107-wf-phd-03]. The results and conclusions are solely those of the authors and do not represent the opinions of the Institute of Transportation, Ministry of Transportation and Communications. References 1 World Health Organization. Global status report on road safety 2015 . Geneva, Switzerland: WHO, 2015. 2 National Police Agency, Ministry of the Interior, Taiwan, ROC. International comparison analysis of road traffic fatalities (in Chinese). https://www.npa.gov.tw/NPAGip/wSite/ct?xItem=77625&ctNode=12598&mp=1. Accessed August 26, 2017 3 Ministry of Transportation and Communications. Road Traffic Safety Annual Report 2014 (in Chinese) . Taipei, Taiwan, ROC: MOTC, 2015. 4 Pan RH, Chang NT, Chu D et al.  . Epidemiology of orthopedic fractures and other injuries among inpatients admitted due to traffic accidents: a 10-year nationwide survey in Taiwan. Scientific World Journal  2014; 2014: 637872. Google Scholar PubMed  5 Lai CH, Doong JL, Huang WS et al.  . Mortality burden from motor vehicle accidents in Taiwan. Public Health Nurs  2008; 25: 28– 36. Google Scholar CrossRef Search ADS PubMed  6 OECD/ITF. Road Safety Annual Report 2017 . Paris: OECD Publishing, 2017. 7 Newgard CD, Fleischman R, Choo E et al.  . Validation of length of hospital stay as a surrogate measure for injury severity and resource use among injury survivors. Acad Emerg Med  2010; 17: 142– 50. Google Scholar CrossRef Search ADS PubMed  8 Cryer C, Gulliver P, Langley JD et al.  . Is length of stay in hospital a stable proxy for injury severity? Inj Prev  2010; 16: 254– 60. Google Scholar CrossRef Search ADS PubMed  9 Adminaite D, Jost G, Stipdonk H et al.  . Ranking EU progress on road safety: 11th Road Safety Performance Index Report. etsc.eu/wp-content/uploads/PIN_ANNUAL_REPORT_2017-final.pdf. Accessed September 18, 2017. 10 OECD/ITF. Reporting on serious road traffic casualties: combining and using different data sources to improve understanding of non-fatal road traffic crashes . Paris: International Transport Forum, 2011. 11 Gennarelli TA, Wodzin E. AIS 2005: a contemporary injury scale. Injury  2006; 37: 1083– 91. Google Scholar CrossRef Search ADS PubMed  12 Burch C, Cook L, Dischinger P. A comparison of KABCO and AIS injury severity metrics using CODES linked data. Traffic Inj Prev  2014; 15: 627– 30. Google Scholar CrossRef Search ADS PubMed  13 Institute of Transportation, Ministry of Transportation and Communications, Taiwan, ROC. Road traffic accident report. http://talas-pub.iot.gov.tw/. Accessed August 26, 2017. 14 Chen WH, Huang WS, Lai CH Exploration of the relationship between the drivers’ injury severity and their expenses during hospitalization (in Chinese). In: Proceedings of the Traffic Safety and Law Enforecement Conference, Taiwan, 2004. 15 Lin MR, Kraus JF. Methodological issues in motorcycle injury epidemiology. Accid Anal Prev  2008; 40: 1653– 60. Google Scholar CrossRef Search ADS PubMed  16. Emergency Medical Services Act. Taiwan, ROC: Ministry of Health and Welfare, 2013. 17 Ng CJ, Yen ZS, Tsai JC et al.  . Validation of the Taiwan triage and acuity scale: a new computerised five-level triage system. Emerg Med J  2011; 28: 1026– 31. Google Scholar CrossRef Search ADS PubMed  18 Chi CH, Huang CM. Comparison of the Emergency Severity Index (ESI) and the Taiwan Triage System in predicting resource utilization. J Formos Med Assoc  2006; 105: 617– 25. Google Scholar CrossRef Search ADS PubMed  19 Formosa Association for the Surgery of Trauma. The operating manual for trauma registry system (in Chinese). Taichung, Taiwan, ROC: FAST, 2008. 20 Ministry of Health and Welfare. Accreditation standards for the hierarchical system of emergent rescuer responsive hospitals. Taiwan, ROC: Ministry of Health and Welfare, 2017. 21 Hsing AW, Ioannidis JA. Nationwide population science: lessons from the Taiwan National Health Insurance Research Database. JAMA Intern Med  2015; 175: 1527– 9. Google Scholar CrossRef Search ADS PubMed  22 Taiwan Insurance Institute. Compulsory Automobile Liability Insurance. http://www.cali.org.tw/en/about.aspx. Accessed September 9, 2016. 23 Bull JP. Disabilities caused by road traffic accidents and their relation to severity scores. Accid Anal Prev  1985; 17: 387– 97. Google Scholar CrossRef Search ADS PubMed  24 Smith P, Hogg-Johnson S, Mustard C et al.  . Comparing the risk factors associated with serious versus and less serious work-related injuries in Ontario between 1991 and 2006. Am J Ind Med  2012; 55: 84– 91. Google Scholar CrossRef Search ADS PubMed  25 Ministry of Transportation and Communications, Taiwan, ROC. Developing an information platform for integration and analysis on road traffic safety data (in Chinese). https://www.grb.gov.tw/search/planDetail?id=12080384&docId=502353. Accessed March 29, 2018. 26 Ramaswamy SK, Mosher GA. Using workers’ compensation claims data to characterize occupational injuries in the commercial grain elevator industry. J Agric Saf Health  2017; 33: 203– 17. Google Scholar CrossRef Search ADS   27 Härö AS. The centralized statistics on hospital utilization in Finland. Methods Inf Med Suppl  1969; 4: 277– 87. Google Scholar PubMed  28 World Health Organization. Data systems: a road safety manual for decision-makers and practitioners . Geneva, Switzerland: WHO, 2010. 29 Amoros E, Martin JL, Lafont S et al.  . Actual incidences of road casualties, and their injury severity, modelled from police and hospital data, France. Eur J Public Health  2008; 18: 360– 5. Google Scholar CrossRef Search ADS PubMed  30 Howard C, Linder A. Review of Swedish experiences concerning analysis of people injured in traffic accidents . Linköping, Sweden: Swedish National Road and Transport Research Institute, 2014. © The Author(s) 2018. Published by Oxford University Press in association with the International Society for Quality in Health Care. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

International Journal for Quality in Health CareOxford University Press

Published: May 18, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off