This was a retrospective study conducted at four National Guard Hospitals Affairs (NGHA) hospitals in Saudi Arabia from January 2016 to March 2019. All patients meeting the inclusion criteria were identified. Data captured included four locations: King Abdulaziz Medical City in Riyadh with 394 patients, King Abdulaziz Medical City in Jeddah total number 63, King Abdulaziz Hospital in Al-Ahsa with 39 patients, Al-Imam Abdulrahman bin Faisal Hospital in Dammam included ten patients. Data were ascertained from the trauma registry provided comprehensive information on trauma patients who were admitted to the hospital for at least one day following MVC during the study period.
Next, rehabilitation data were ascertained from the electronic medical records. The study population comprised all patients ≥16 years old who were admitted to KAMC due to MVC between 2016 and 2019. A trained research coordinator reviewed all admitted patients with MVC and verified patients were actually injured in MVC and hospitalized. Patients treated in the ED then immediately discharged were excluded from the study.
Patients were classified as requiring long-term rehabilitation if they were diagnosed with one of the following injuries: spinal cord injury, amputation, traumatic brain injury, lower extremity fracture, multiple fractures, open fracture or polytrauma. This was based on two physicians’ expert opinions and consensus (an orthopedic surgeon and a rehabilitation physiatrist). In addition, long-term rehabilitation users included those discharged and seeking care in outpatient clinics for at least three months. Patients who required short-term rehabilitation were those who sought rehabilitation services for three months or less.
The trauma registry includes demographics, injury type and cause, severity, LOS, and discharge status. To complement information not available in the registry, research coordinators used a pre-designed data collection sheet that includes access to rehabilitation, rehabilitation length of inpatient stay, description of patient’s disability, date of discharge, and discharge status or death. A patient was considered disabled if they suffered from a long-term impairment of function [4]. This includes if the patient suffered any amputation, fracture, traumatic brain injury, paralysis, and limb deformation [6].
Patients were classified based on the site and the severity of the injury. The injuries were classified using the AIS in all five major body regions into minor (1 point), up to severe or maximal (untreatable) (6 points) [16]. Only the three highest AIS scores in each body region are used to calculate the Injury Severity Score (ISS), which ranges between 0 and 75. A higher score indicates a more severe injury (1–8 minor, 9–15 moderate, 16–24 severe, 25–75 very severe) [17]. The hospital LOS was the time from ER admission until hospital discharge. The study was retrospective a chart review without any direct contact with patients and was performed in accordance with the Declarations of Helsinki. The Institutional Review Board (IRB) of King Abdulah International Medical Research Center (KAIMRC) reviewed the protocol, waived informed consent, and approved the study (number: 20/103/R).
Statistical analysis
The trauma registry in KAMC includes over 20,000 patients. In this study, the sample size was calculated using 95% as a confidence interval, 5% margin of error, and assuming p = 0.5, the precision of 5% of the true value. The minimum required sample size is 385 patients to estimate the prevalence [18]. We included all patients in the trauma registry who met the inclusion criteria for a total of 506 patients.
Descriptive statistics were carried out in the form of mean, standard deviation (SD), and frequencies. The outcome is the prevalence of MVC patients who require long-term rehabilitation. It was defined as the proportion of patients with long-term rehabilitation as part of the overall admitted MVC patients. We also estimated the prevalence of patients who received interdisciplinary rehabilitation, defined as receiving all the services including physiotherapy, occupational therapy, psychiatric treatment, and social service support. For inferential statistics, chi-square tests were applied to compare long-term and short-term rehabilitation patients. Continuous variables, including ISS score and LOS, were compared using Student’s t-test. In addition, multivariate logistic regression techniques were used to model the probability of long-term rehabilitation with associated variables. We used multivariate forward selection regression with entry criteria of at least a p-value 0.3 and retained variables at a p-value of 0.05. The following variables were entered into the initial model: age, gender, BMI, mechanism of injury, body region, surgery, and ICU admission. The model included the following variables: ICU admission (“no” as the reference group), surgery (“no” as the reference group), and affected body region (“abdomen” as the reference, head and neck, face, thorax, extremities, and external and other). Statistical analyses were performed using STATA version 15.0 software, and a p-value of < 0.05 was declared as statistically significant.