Long-term Consequences of Cyclist Injuries

Overview

There is a lack of knowledge about the extent of bicycle injuries in Norway. Among other things, a significant underreporting of bicycle injuries has been found in the official road traffic accident statistics based on police-registered accidents. Furthermore, there is a lack of knowledge about the long-term consequences of serious bicycle injuries. The main purpose of this project is to generate new knowledge about the consequences of bicycle injuries for later disability, employment status and use of prescribed drugs. In addition, the investigators want to study whether such long-term consequences vary according to demographic and socioeconomic characteristics, as well as to assess the degree of overlap between bicycle injuries registered in the official road traffic accident statistics and bicycle injuries registered in the health service, including an assessment of the severity of the injury. The investigators will use a retrospective analysis where a population-based dataset from the National Population Register is linked with information on injured cyclists from Statistics Norway's database on police-reported road traffic accidents, the Norwegian Patient Register, the National Trauma Registry and the Norwegian Cause of Death Registry. In order to study the long-term consequences of bicycle injuries, the investigators will also compile information on social security benefits and employment from Statistics Norway's historical events database (FD-Trygd) and the use of prescribed drugs from the Norwegian Prescribed Drug Registry. The investigators will also collect information on income and educational attainment from registers in Statistics Norway. The project will contribute to increased knowledge about the societal and individual burden of bicycle injuries, knowledge that is crucial for prioritizing and implementing necessary preventive measures.

1. Project organisation The research activity described in this protocol constitutes one work package of the research project ReCyCLIST (Recording Cyclist Crashes and Long-term Injury Consequences by new Smart Tools). The ReCyCLIST project is a collaboration between research institutes from the health and transport sciences, public authorities, medical expertise, web-tool developers, and end-user organisations. The project is coordinated by the Institute of Transport Economics (TØI), with Torkel Bjørnskau as overall project manager. Additional project partners in this work package include the National Institute of Public Health (NIPH) and the Norwegian Trauma Registry (NTR)/Oslo University Hospital (OUS). The project group in this work package, which in the following will be referred to as "the project", consists of the following researchers: * Torkel Bjørnskau (TØI) * Ingeborg Hesjevoll (TØI) * Rikke Ingebrigtsen (TØI) * Katrine Karlsen (TØI) * Eyvind Ohm (NIPH) - project leader * Christian Madsen (NIPH) * Elling Tufte Bere (NIPH) * Yusman Bin Kamaleri (NIPH) * Svetlana Skurtveit (NIPH) * Olav Røise (NTR/OUS) 2. Study objectives This project has three primary objectives: 1. To assess the degree of overlap of cyclist injuries registered in the official road accident statistics and in the health services, and thereby map the level of underreporting of cyclist injuries in the former data source 2. To obtain knowledge of long-term consequences of cyclist injuries with respect to: 1. Disability 2. Employment status 3. Use of prescribed drugs 3. To examine whether such long-term consequences vary by sociodemographic characteristics 3 Methodology 3.1 STUDY DESIGN To address these issues, the investigators aim to construct a population-based dataset containing census data (including sociodemographic information) from the National Population Register for all individuals residing in Norway in the period 2008-2021. This dataset is to be linked with information on injuries registered in four data sources: 1) Statistics Norway's (SSB) database on police-reported road traffic accidents, 2) the National Patient Registry (NPR), 3) the Norwegian Trauma Registry (NTR) and 4) the Norwegian Cause of Death Registry (DÅR). To analyse long-term consequences of cyclist injuries with respect to disability, employment status and use of prescribed drugs, the investigators will further link data for adults aged 18-67 years from SSB's historical events database (FD-Trygd) and the Norwegian Prescribed Drug Registry. Finally, to examine socioeconomic differences in risk of such long-term consequences, the investigators will link data on educational attainment and income, both supplied by SSB. The overall population-based design of the study was chosen to enhance the validity of the estimates and to facilitate the interpretation of the statistical analyses. For instance, when assessing long-term consequences of cyclist injuries and how these vary by sociodemographic characteristics, it is paramount to collect this type of information also for individuals that have not been involved in bicycle crashes. By including the whole population, the investigators will generate a more powerful comparison group than other potential designs and reduce potential sources of bias. Moreover, this design will allow for more robust estimates of incidence rates and risk factors for the outcome variables by including valid estimates of person-time at risk. To control for the occurrence of non-cycle injuries that may influence functional outcomes, the investigators will also collect information on other types of injuries, both for injured cyclists and for the population at large. 3.2 DESCRIPTION OF DATA SOURCES 1. The National Population Register: Census data for all residents of Norway in the period 2008-2021, including the following information: * Month and year of birth * Sex * Immigrant status and country of origin * Marital status * Residential code - classification of individuals as resident, emigrated or dead (with month and year for change of code) * Municipality of residence * Encrypted ID number for individual-level linkage with other registers 2. Police-reported road traffic accidents: Information about all road traffic accidents reported to the police in the period 2008-2021, including the following variables: * Date of crash * Municipality and location of crash (X- and Y coordinates) * Road category (European route, national road, county road etc.) * Type of road (highway, pedestrian zone, walking/cycling path etc.) * Speed limit * Accident type (single accident, rear-end collision, head-on collision etc.) * Type of vehicle (passenger car, heavy goods vehicle, bicycle, motorcycle, bus etc.) * Classification of road user (driver/passenger of car, rider/passenger of motorcycle/moped/bicycle, pedestrian etc.) * Age and sex of injured person * Injury severity (killed, very seriously injured, seriously injured, slightly injured, unspecified) * Suspicion of alcohol intoxication * Protective equipment (seat belt, helmet use) * Counterpart in collision (motor vehicle (light/heavy), bicycle, pedestrian etc.) 3. National Patient Registry: Information about all injury contacts in secondary care registered in NPR in the period 2008-2021, identified by diagnostic codes from chapter XIX and chapter XX of the tenth edition of the International Statistical Classification of Diseases and Health Related Problems (ICD-10). For these patients, the investigators will include the following information: * Date of treatment * Place of treatment (hospital ID) * Injury diagnoses (principal and secondary) In addition, injury data for all patients registered in the common minimum dataset for personal injury (FMDS) will be included. This dataset contains information about the external circumstances of injuries treated in secondary care, including the following variables: * Date and time of injury * Intent (accident, self-harm, assault) * Place of occurrence (road/street, residence, workplace, playground etc.) * Type of transport vehicle * Activity (sport/leisure, work, education etc.) * Injury mechanism * Injury severity * Municipality where injury occurred * Location of road traffic crash (X- and Y-coordinates) 4. Norwegian Trauma Registry: Since 2015, NTR contains nationwide data on patients admitted to hospital with potentially severe injury (i.e., high- and low energy trauma patients). From this register, the investigators aim to use information about all injured patients admitted to trauma-receiving hospitals in the period 2015-2021, including the following variables: * Date and time of injury * Municipality where injury occurred * ASA score before injury (a measure of physiological status/comorbidity developed by the American Society of Anesthesiologists to predict operative risk) * Glasgow Outcome Scale (measure of functional status) before injury and at discharge from hospital * Helmet use * Intent * Injury mechanism * Type of traffic accident * Sport and leisure accident * Dead on arrival * Days of hospital stay * Dead within 30 days of injury * Abbreviated Injury Scale (AIS) * Injury Severity Scale (ISS) * New Injury Severity Scale (NISS) 5. The Cause of Death Registry: Information about all fatal injuries registered in the period 2008-2021, identified by diagnostic codes from chapter XX of ICD-10 as underlying cause of death, including the following variables: • Month and year of death * Resident in Norway at time of death * Age in years * Municipality of residence * Country of death * Municipality where death occurred * Special circumstances * Place where injury occurred * Activity related to injury * Underlying cause of death (chapter XX) * Main injury (codes from chapter XIX) * Contributing causes of death 6. FD-Trygd: FD-Trygd is Statistics Norway's historical events database, which links basic demographic data with employment records and information about services and benefits administered by the Norwegian Labour and Welfare Administration (NAV). The investigators aim to use the following variables from this database: • Sickness benefit * Work assessment allowance * Disability benefit * Old age pension and early retirement pension * Basic benefit and attendance benefit * Employment (including occupation) * Job seeker status 7. The Norwegian Prescribed Drug Registry: Information on analgesics and other psychopharmaceutical drugs dispensed from pharmacies to individual patients living outside institutions in the period 2008-2021, including the following variables: • Prescriber's project ID * Prescriber's profession and specialty * Patient's sex and year/month of birth * Patient's residential municipality * Dispensing number * Dispensing date * Number of packages per filled prescription * Number of DDDs (defined daily dose) per filled prescription * Regulation (legal basis) * Reimbursement Code * Reimbursement Code ICD * Reimbursement Code ICPC * Article number (unique product identifier) * ATC (Anatomical Therapeutic Chemical) code at 5th level for all N codes (anatomical main group: nervous system) apart from N04 (anti-Parkinson drugs), plus codes M01A (anti-inflammatory and antirheumatic products, non-steroids), M02A (topical products for joint and muscular pain), M03BA (muscle relaxants, peripherally acting agents) and H02A (corticosteroids for systemic use, plain) * Prescription category (narcotic drug, other addictive drug, prescription drug) 8. The National Education Database (NUDB): From this database the investigators will include information about highest educational attainment, coded according to the Norwegian Standard Classification of Education, and the year this level was achieved. The database is administered by Statistics Norway. 9. Statistics Norway's income database: Statistics Norway collects income data by linking different administrative registers and statistical data sources, including tax returns. From this database the investigators will include the following variables: • Pensionable income • Total income • Household income • Taxable gross wealth 3.3 STATISTICAL ANALYSIS To map the degree of overlap of cyclist injuries between health registers (NPR, NTR and DÅR) and the official road traffic accident statistics, the investigators will compare the number and characteristics of such injuries (in total and in different sub-populations) registered in these four data sources. To obtain knowledge of long-term consequences of cyclist injuries, the investigators will perform descriptive analyses to examine disability rates, employment status and use of prescription drugs. The investigators will further use Poisson regression to model incidence rates with respect to various injury-related factors (injury severity, type of accident, injury mechanism, helmet use etc.) and to examine the degree to which these long-term consequences vary by sex, age, and other sociodemographic variables. 3.4 REQUIRED PERMISSIONS The project will require permissions from: * The Regional Committees for Medical and Health Research Ethics (REC) for advance ethical approval * Statistics Norway (SSB) for access to microdata from the National Population Register, the police-reported road accident database, FD-Trygd and socioeconomic data (including data from NUDB) * The Norwegian Directorate of Health for access to injury data in NPR * Oslo University Hospital (OUS) for access to trauma data in NTR * The Norwegian Institute of Public Health (NIPH) for approval of Data Protection Impact Assessment (DPIA) and for access to data in DÅR and the Norwegian Prescribed Drug Registry 3.5 PROCEDURE FOR DATA LINKAGE Statistics Norway will define the study population as described in paragraph 4.1 and generate project-specific ID numbers with linkage to the unique personal ID numbers given to all Norwegian residents. Statistics Norway will then link data from the National Population Register, the police-reported road accident database, FD-Trygd and sociodemographic data according to specifications described in paragraph 4.2. Likewise, the Norwegian Directorate of Health, NTR and NIPH will link their respective data. After linkage, each data owner will send these datasets (in which the unique personal ID number is deleted and replaced by the project-specific ID number) to the project. The Norwegian Directorate of Health, NTR and NIPH will subsequently delete the file linking the personal ID numbers and project-specific ID numbers, which will for the duration of the project be stored securely by Statistics Norway. 4 Significance and ethical aspects The study will improve our understanding of the societal and individual burden of bicycle injuries and provide new knowledge of their long-term consequences. Exploring sociodemographic risk factors will further guide policy makers in identifying effective safety measures. The project will follow ethical guidelines for statistical analysis and presentation of results, considering both privacy issues and statistical robustness. A potential risk involves identification of individuals. However, the files received by the project contain no directly identifiable information, as unique personal ID numbers are replaced by project-specific ID numbers. To minimise the risk of backwards identification, analyses will be performed on files containing only the variables needed to address the different research questions. For instance, to assess the degree of overlap between injured cyclists in health registries and the official road traffic accident database, analyses will be restricted to variables common to these data sources, thus omitting information on welfare benefits, use of prescription drugs and socioeconomic variables. Level of detail will also be reduced, for instance by categorising age into 5-year age groups. In addition, analyses of long-term consequences of bicycle injuries (including sociodemographic differences in such consequences) will be restricted to adults aged 18-67 years. Another potential risk is stigma attached to specific groups of individuals, for instance those receiving welfare benefits. However, this project will not present detailed information about such groups. All results will be presented on aggregate level, and results based on few observations (\< 6 individuals) will not be presented or discussed. The project will include sensitive information about individuals (e.g., medical diagnoses, use of prescription drugs, receipt of welfare benefits and measures of comorbidity/functional status). However, the investigators will not collect any new data, but utilise data already registered in other data sources. The use of these data is regulated by law. To ensure secure storing of data, the investigators will use Services for sensitive data (TSD), which is a service where users can collect, store, share and analyse data within a secure environment with a high degree of safety. 5 Project timeline The project will be developed in tasks with the following timeline: • Task 1 Preparation and linkage of registers: January 2022 - February 2023 • Task 2 Main data analysis: February 2023 - May 2023 • Task 3 Estimation of overlap between different registries: March 2023 - June 2023 • Task 4 Dissemination: by December 2026 6 Funding The ReCyCLIST project is funded by the Research Council of Norway (NFR) and Agder County. 7 Dissemination A designated ReCyCLIST project website (https://www.toi.no/recyclist/) will be used for regular communication of project updates and results. For this specific work package, the investigators plan to submit three focused articles to peer-reviewed international journals, for open access publication. In addition, results from this project will be presented at national and international seminars and conferences. Nationally, the most relevant will be the yearly conferences arranged by the Norwegian Public Roads Administration, the Norwegian Safety Forum, and the Norwegian Council for Road Safety (Trygg Trafikk). Internationally, relevant conferences include the biannual Safe Communities Conference, the annual International Cycling Safety Conference, and the biannual Transportation Research Arena.