Risk of permanent medical impairment after road traffic crashes: A systematic review

Abstract

Purpose

To systematically review the risk of permanent disability related to road traffic injuries (RTIs) and to determine the implications for future research regarding permanent impairment following road traffic crashes.

Methods

We conducted this systematic review according to the preferred reporting items for systematic reviews and meta-analysis statement. An extended search of the literature was carried out in 4 major electronic databases for scientific research papers published from January 1980 to February 2020. Two teams include 2 reviewers each, screened independently the titles/abstracts, and after that, reviewed the full text of the included studies. The quality of the studies was assessed using the strengthening the reporting of observational studies in epidemiology (STROBE) checklist. A third reviewer was assessed any discrepancy and all data of included studies were extracted. Finally, the data were systematically analyzed, and the related data were interpreted.

Results

Five out of 16 studies were evaluated as high-quality according to the STROBE checklist. Fifteen studies ranked the initial injuries according to the abbreviated injury scale 2005. Five studies reported the total risk of permanent medical impairment following RTIs which varied from 2% to 23% for car occupants and 2.8% to 46% for cyclists. Seven studies reported the risk of permanent medical impairment of the different body regions. Eleven studies stated the most common body region to develop permanent impairment, of which 6 studies demonstrated that injuries of the cervical spine and neck were at the highest risk of becoming permanent injured.

Conclusion

The finding of this review revealed the necessity of providing a globally validated method to evaluate permanent medical impairment following RTIs across the world. This would facilitate decision-making about traffic injuries and efficient management to reduce the financial and psychological burdens for individuals and communities.

Introduction

Despite prominent declines in traffic-related fatality rates in many industrialized countries, road traffic injuries (RTIs)-associated mortality, the loss of healthy years of life, and the temporary or permanent disabilities remain controversial.¹ This reflects a lack of reliable data on non-fatal outcomes especially permanent disabilities after RTIs even in the countries where mortality statistics are assembled annually.² Globally, routinely collected data on crash-related disability are non-existent or inaccessible.³ World Health Organization reports about 1.24 million annual traffic-related fatalities, and 20 – 50 million sustained non-fatal injuries.^4,5 World Health Organization also reports that RTI is the leading cause of mortality in young individuals (15 – 29 years old) and is among the top 3 causes of death at ages of 15 – 44 years.^4,5 Institute for Health Metrics and Evaluation estimated approximately 0.9 million, 1.3 million, and 1.4 million fatalities due to RTIs in 1990, 2010, and 2013, respectively.⁶

In addition to mortality rates, studies have reported that a significant proportion of RTI survivors experienced long-term morbidity, but there is little agreement about the magnitude of this risk.³ The previous commentaries have primarily discussed psychiatric outcomes, whiplash-related injuries, or injuries to the head and extremities.⁷ With the global burden of RTIs projected to rank as the third leading cause of disability-adjusted life years by 2020¹, it is necessary to review the epidemiological evidence and identify gaps in knowledge regarding the magnitude and importance of disability following RTIs. This will help inform public policy, health organizations, and preventive interventions. It may also change our perception of priorities.^1,3

To calculate the global risk of RTIs related to permanent impairments, we need to review systematically the epidemiological literature investigating permanent impairments due to RTIs including motor vehicle, car, bike, bicycle occupants, and public transport and related injuries.^8,9 The specific objectives of the present study are to systematically review the risk of permanent disability related RTIs, critically appraise the evidence estimating the risk of permanent disabilities due to road traffic crashes, examine the methodological strengths and weaknesses of relevant studies, and determine the implications for future research regarding permanent impairments following road traffic crashes.

Methods

We conducted this systematic review based on preferred reporting items for systematic reviews and meta-analysis (PRISMA)-2020 statement.¹⁰ The review protocol was registered in International Prospective Register of Systematic Reviews, (registration number: CRD42020190684).

Literature search

An extended search of the literature was conducted in the electronic databases of PubMed, EMBASE, SCOPUS, and the Web of Science, from January 1980 till the end of February 2020. The search strategy was written according to the MeSH terms and related keywords (Appendix 1).

Study selection

All citations were imported to a library using Endnote software (version X8), and the duplicate titles in the databases were removed before reading. Two independent reviewers screened the articles using titles and abstracts of the studies. To access articles not available on the internet, we contacted the corresponding author of each paper. Hand-searching of related journals and references of related articles was also performed to include additional relevant studies.

After the selection of eligible abstracts, 2 teams of reviewers screened articles independently using full texts to finalize the eligible studies. Details are available in the PRISMA flow chart (Fig. 1).

Fig. 1.

PRISMA 2020 flow diagram for new systematic reviews which included searches of databases and registers only. ∗Consider, if feasible to do so, reporting the number of records identified from each database or register searched (rather than the total number across all databases/registers). ∗∗If automation tools were used, indicate how many records were excluded by a human and how many were excluded by automation tools.

Eligible articles were included where they met the following criteria: (1) available full text, (2) original paper, (3) observational studies, (4) permanent impairment or disability risk after road traffic crashes included car occupants, cyclists, and pedestrians, (5) methods of the study were explained clearly, (6) population of the study was injured individuals following crashes, (7) the follow-up time was at least 3 years.

Exclusion criteria were any of (1) contained no clear definition of disability, (2) no clear method of data collection existed, (3) descriptive studies which contain no quantitative variable, (4) evaluate limited body injuries such as studies on brain injuries, whiplash injuries, chest trauma, etc., (5) clinical trials, (6) animal studies and in-vitro studies, (7) case studies and case series, or (8) review articles.

There were no limitations for sample size, age group and language of full texts to include our review. In case of reviewer discrepancies, a third reviewer investigated and decided to select the most eligible studies.

Data extraction

We designed an extraction sheet according to studies’ data which included study characteristics, location, study design, follow-up time, sample size and sampling methods, selection criteria and definition of permanent disability, mean age, the total risk of disability, and the risk of disability in different body regions. Data extraction was performed independently by 2 reviewers. After data extraction was completed, the data were checked for accuracy and merged and completed by 1 of the researchers.

Data syntheses

After overviewing of articles, the quantitative analysis of variables was not indicated, due to the heterogeneity of data and considerable differences in the methods of studies, data collection and reported outcomes. Therefore, it was not possible to carry out a meta-analysis on the extracted data.

Validity assessment

We performed the methodological quality assessment of the included studies using the strengthening the reporting of observational studies in epidemiology (STROBE) statement¹¹ which is an appropriate tool to assess the methodological quality of observational studies while it covers various items in each section of an observational studies.12, 13, 14, 15, 16, 17 This tool evaluates 22 key points that should be provided in the title, abstract, introduction, methods, results, and discussion of the observational investigations including cross-sectional, case-control, and cohort studies. We obtained the STROBE results for each study (Table 1). Each item got 1 (yes) or 0 (no, not applicable) point and the sum of the points was categorized as follows: 0 – 7 points were considered as low quality, 8 – 14 points as intermediate quality, and 15 – 22 points as high-quality articles.¹⁸

Table 1.

Characteristics of the included studies and summary of results.

No.	References	Year	Study method	Country	Sample size	Data collect	Follow-up (year)	Reference for permanent impairment criteria	Injury assessment scale	The most common body region resulting PMI	Permanent impairment in different AIS/ISS
1	Bull et al.33	1985	Retrospective cohort	United Kingdom	2502 Injured from crashes (pedestrians, pedal cyclists, motor cyclists, vehicle occupants) (Children and adults)	Medical records, physical examination	3	(1) Impairment: the loss of structure or function; (2) disablement: the loss of ability to perform certain tasks, and (3) incapacity: the loss of social function. the American Medical Association Guides to the evaluation of permanent impairment (1977) disability score to each injury and overall disability.	AIS MAIS ISS	NA	AIS 1:2% AIS 2: 12% AIS 3: 41% AIS 4: 55% AIS 5: 75%
2	Gustafsson et al.26	1986	Prospective cohort	Sweden	12,007 Injured car occupants (Children and adults)	Medical records, physical examination	4 – 5	Folksam (a Swedish insurance company)	AIS ISS	Neck	NA
3	Gustafsson et al.27	1987	Retrospective cohort	Sweden	13,552 Injured car occupants (Children and adults)	Medical records, physical examination	5	Folksam (a Swedish insurance company)	AIS ISS	Neck	NA
4	Nakayama et al.30	1990	Retrospective cohort	United States	372 Children injured from bicycle crash	Medical records, telephone interviews, personal interviews, and physical examination to confirm answers.	3	Guides to evaluation of permanent impairment (Chicago, American Medical Association,1984)	–	NA	NA
5	Olkkonen et al.22	1993	Retrospective cohort	Finland	542 Injured cyclists (Children and adults)	Medical records, questionnaires	6	Disability lasting at least 6 months was considered a persistent disability; 365 calendar days were used as the maximum duration of work disability for the 2 responders with permanent work disability	AIS MAIS ISS	NA	NA
6	Norin et al.31	1997	Prospective cohort	Sweden	14,470 Adults injured car occupants	Medical records, physical examination	5	Swedish insurance companies Försäkringshranchens Serviceaktiebolag; 1981. (In Swedish)	AIS	Neck	NA
7	Bylund et al.24	1997	Retrospective cohort	Sweden	506 All injured car occupants (Children and adults)	Medical records, physical examination	3.5 – 6.5	Classification manual used by the insurance industry (Personskadekommitten inom Fijrs'akringsbranschens Serviceaktiebolag, 1989).	AIS MAIS	Neck	NA
8	Maraste et al.29	2003	Prospective cohort	Sweden	230 Injured car occupants (Children and adults)	Questionnaire, examination	3.5 – 4	Rosser's three-dimensional index (the injured is asked to describe how he or she feels in terms of functional disability, pain, and distress at different times after the accident)	AIS ISS	NA	ISS 1 – 3: 1% (adult) ISS 4 – 8: 13.5% ISS 9 – 15: 7.5% ISS 16 – 24: 1% ISS > 25: 0%
9	Malm et al.7	2008	Prospective cohort	Sweden	20,484 All injured car occupants	Medical records, physical examination	5	Grading medical impairment (Sveriges Försäkringsförbund, 2004)	AIS ISS	NA	AIS 1: 9.7% AIS 2: 20% AIS 3: 50% AIS 4: 50% AIS 5: 100%
10	Fredriksson et al.20	2010	Retrospective cohort	Germany	1030 Pedestrians stuck by the front of passenger cars or vans (Children and adults)	Medical records, physical examination	5	Grading medical impairment (Sveriges Försäkringsförbund, 2004)	AIS	Leg	NA
11	Rizzi et al.32	2013	Retrospective Cohort	Sweden	55,220 Injured cyclists (Children and adults)	Medical records, physical examination	5	Grading medical impairment (Sveriges Försäkringsförbund, 2004)	AIS ISS	Upper extremity	NA
12	Bohman et al.19	2014	Retrospective cohort	Sweden	2619 Injured children car occupants (0 – 12 years)	Medical records, physical examination	5	Grading medical impairment (Sveriges Försäkringsförbund, 2004)	AIS	Head	AIS 1: 1% AIS 2: 4% AIS 3: 13% AIS 4: 33% AIS 5: 100%
13	Stigson et al.23	2015	Retrospective cohort	Sweden	36,743 All injured car occupants (Children and adults)	Medical records, physical examination	≥ 3	Grading medical impairment (Sveriges Försäkringsförbund, 2013)	AIS	Lower extremity	NA
14	Gustafsson et al.28	2015	Retrospective cohort	Sweden	36,744 All injured car occupants	Medical records, physical examination	3 – 5	Grading medical impairment (Sveriges Försäkringsförbund, 2013).	AIS MAIS	Cervical spine	AIS 1–2: 16.1% AIS 3+: 71.6%
15	Doud et al.25	2017	Retrospective cohort	United states	49,833 Injured children car occupants (7 – 18 years)	Medical records	4	FIM scores within the National Trauma Data Bank Research Data Set	AIS MAIS	Head	AIS 2: 23.4% AIS 3: 29.7% AIS 4: 34.8% AIS 5: 45.8% AIS 6: 70.7%
16	Ohlin et al.21	2019	Cross-sectional	Sweden	964 Injured cyclists >15 years of age.	Hospital records, questionnaire	≥ 3	Grading medical impairment (Sveriges Försäkringsförbund, 2004)	AIS	Hip and upper leg	NA

AIS: abbreviated injury scale; ISS: injury severity scale; MAIS: maximum abbreviated injury scale; FIM: functional independence measure; PMI: permanent medical impairment; NA: not available.

Results

Study selection

Of the total initial 28,474 studies identified from the search strategy, 15,934 studies remained after removing all duplicates. After titles and abstracts screening by separate reviewers, there were 112 abstracts met the inclusion criteria, of which 16 full text papers remained for data extraction (Fig. 1).

Quality assessment

Considering the quality scoring, 5 studies belonged to the high-quality category.19, 20, 21, 22, 23 Ten studies were evaluated as intermediate quality^7,24, 25, 26, 27, 28, 29, 30, 31, 32 and 1 study had low quality³³. The highest score was 18²³ and the lowest was 7.³³ Only 1 out of 16 studies indicated the study's design with a commonly used term in title or abstract.²⁹ Fifteen out of 16 studies provided the appropriate introduction based on scientific background and stated their specific objectives.^7,19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 Thirteen studies did not consider addressing the potential source of bias.^7,19,21,23, 24, 25, 26, 27, 28, 29, 30^,32,33 Most of the studies clearly reported the number of retrieved individuals, defined the variables, but 14 studies did not use flow diagram to indicate the number of participants at each stage of study^7,19,20,23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, Six studies thoroughly discussed the limitations of the study19, 20, 21, 22^,25,26, and 16 of the studies summarized key results. The studies' STROBE quality assessment and the sum of scores are presented in detail in Appendix 2.

Characteristics of the studies

Studies were performed in Sweden^{7,19,21,23,24,}26, 27, 28, 29^,31,32, the United States^25,30, the United Kingdom³³, Finland²², and Germany²⁰ (Table 1) from 1985 to 2019. Study designs included 12 retrospective cohorts^7,19,20,22, 23, 24, 25^{,27,28,30,32,33}, 3 prospective cohorts^26,29,31, and 1 cross-sectional²¹.

Table 2.

Summary of the studies estimated the risk of disability following road traffic injury.

Study	Summary of investigation	Estimated disability risk
Malm et al.7	Investigate the mortality risk due to road injuries. Patients were followed for more than 5 years to evaluate the RPMI. The importance of including minor injuries was noticed. The highest risk of enduring a PMI from an AIS 1 injury was in association with cervical and lower limb injuries. On the other hand, thorax and abdomen injuries were suspected to give the lowest risk for PMI.	RPMI 1% +: 11.2% RPMI 5% +: 5.7% RPMI 10%+: 1.6%
Maraste et al.29	An investigation took place to equip data from 2 long-term prospective for serious road traffic injuries, treatment costs, and health impairments in Sweden in the 1960s and 1990s. The outcomes demonstrated that the long-term consequences were not as serious nowadays as they were 25 years ago.	Long-term consequences in adults: 23%, children: 10%.
Nakayama et al.30	A study was purposed to distinguish the long-term disabilities in children caused by bicycle crashes. Bicycle-related injuries could result in both short-term and long-term disabilities. It was shown that a small percentage of children obtained rehabilitation courses either during their admission period or after their discharge. Since neglecting the rehabilitative care could possibly prevent full recovery, it was suggested that trauma centers should provide rehabilitative care for all major injuries despite their causes.	Long-term disabilities in children cyclists: 2.8%
Norin et al.31	The severity assessments for occupants with frontal impacts were discussed. It was emphasized that neck and limb injuries are the main focus in lower levels of disabilities while brain and skull injuries play an important role in higher levels of disabilities.	Not mentioned directly
Ohlin et al.21	Analysis bicycle crashes resulting in injuries with a high risk of health loss. About 44% of people involved claimed health decay after the crashes. The end results stated that the types of crashes weren't specifically different regarding the health loss results.	RPMI 1%+ cyclist: 46%
Olkkonen et al.22	Medical and social consequences of nonfatal bicycle crashes were assessed in this study. The most severe consequences were because of intracranial injuries in motor vehicle and bicycle crashes.	Persistent disability in cyclists: 32% of inpatients and 5% of outpatients
Rizzi et al.32	This study aimed to use the impairment scaling system for reflecting the long-term complications that were due to cyclist's injuries. RPMI was weighted up against the ISS. 70% of the injuries caused by bicycle crashes were to upper and lower limbs. The need for new protective strategies along with using helmets, for preventing limb injuries, especially upper extremities, was suggested.	Percentage of PMI 1% + in cyclists: 16.8% PMI 10%+ in cyclists: 2.29%
Stigson et al.23	The differences in long-term medical outcomes depending on the collision's direction were evaluated. Each individual person was followed for at least 3 years after the crashes to estimate PMI. It was noted that 12% of all car crash victims underwent a permanent medical impairment and 50% of head traumas resulted in long-term consequences were mostly because of frontal traumas. Most of the injuries with long-term consequences were categorized as minor injuries by AIS.	Percentage of PMI 1% +: 8.3% PMI 10% +: 1%
Gustafsson et al.27	Injured children as car passengers were compared with adults to estimate the different risk factors, years lost, impairment patterns, and severity and the injury's risks. Some of the populations were followed up for 5 years to evaluate the long-term complications. The end results suggested that the epidemiology patterns were different in the two groups, but the method used for injury assessments in adults could also be used in children. Since most of the disabilities were due to head traumas, the use of safety measures was encouraged.	Percentage of PMI 10% +: total 5.5% children 0.6%
Fredriksson et al.20	The article indicated that advancing countermeasures for justifying head injuries in windscreen areas, along with emphasizing the structural parts of the windscreen areas, a special focus on the brain injury is much needed.	Percentage of PMI 1% +: 31% PMI 10% +: 4.6%
Bohman et al.19	The long-term medical consequences for children, from age 0 to 12, injured in a car crash were evaluated. In conclusion it was noticed that the injury patterns were different for children and adults therefore it was suggested to use the child data for children's safety priorities.	Percentage of PMI: 2%
Bull et al.33	Disabilities caused by road traffic crashes and their relation to severity score was investigated the study sample of 500 of 2502 patients manifested the fact that they mostly suffered from single injuries such as contusion or minor fractures which could be classified as AIS 1 or 2. Serious and significant disabilities only occur in about 3% of inpatient hospital cases.	OD 1 (slight): 11.9% OD 2 (moderate): 7.8% OD 3 (severe): 2.79% OD 4 (the most severe): 0.27%
Bylund et al.24	The medical impairment occupational injuries that were obtained in traffic collisions were assessed. 50% of the PMI was because of the damages that minor injuries had caused. It was also indicated that one-third of the study population had to change their jobs due to medical impairments. More severe damages had been done in professional drivers rather than the other occupational groups which could have been in association with the low use of seat belts.	Not mentioned directly
Doud et al.25	This research was done to create a pediatric-specific disability risk metric in order to distinguish the severity of nonfatal motor vehicle crash injuries. The disability risk metric could possibly improve the ability to determine the severity of every individual nonfatal injury.	Disability risk: 0.29%
Gustafsson et al.26	This study was done to evaluate the RPMI and their complications after RTI. Although ASI and ISS cannot properly describe the impairment risks, it was recommended to use the ASI to body regions for better understanding the impairments. The age seemed to be an important factor in long-term complications.	RPMI 10% +: 4.9%
Gustafsson et al.28	This article aimed to discuss whether the RPMI is dependent on the patient's gender or age. The long-term results were dependent on gender and age, although the differences between the age groups were much more significant than between genders.	Percentage of permanent impaired individuals: 11.8%, percentage of injuries resulting in PMI 1% +: 8.37% PMI 10% +: 1.59%

RPMI: risk of permanent medical impairment; PMI: permanent medical impairment; AIS: abbreviated injury scale; ISS: injury severity score; OD: overall disability.

The sample size of studies varied from 230 to 55,220. Participants in these studies included all injured individuals following crashes (pedestrians, pedal cyclists, motorcyclists, vehicle occupants)³³, injured car occupants^7,19,23, 24, 25, 26, 27, 28, 29^,31, injured cyclists^21,22,30,32, and pedestrians injured by cars or vans²⁰. Three studies investigated injury in children^19,25,30, 2 studies on adults^21,31, and 11 studies on all injured individuals regardless of their age^7,20,22, 23, 24^,26, 27, 28, 29^,32,33. During follow-up time, studies used medical records with periodic physical examination,^{7,19,20,23,24,}26, 27, 28, 29, 30, 31, 32, 33, questionnaires^21,22,29, and medical records²⁵ to collect data. One study also used telephone interviews in addition to the physical examination for data collection.³⁰

The diagnostic system of permanent impairment for each study was defined by various references including the definition presented and used by Folksam insurance company in Sweden^26,27, the definition proposed in grading medical impairment (Sveriges Försäkringsförbund, 2004)^7,19, 20, 21^,32, grading medical impairment (Sveriges Försäkringsförbund, 2013)²³,²⁸, Swedish insurance companies Försäkringshranchens Serviceaktiebolag 1981³¹, guides to evaluation of permanent impairment (Chicago, American Medical Association, 1984)³⁰, Rosser's three-dimensional index²⁹, the guides to the evaluation of permanent impairment (American Medical Association, 1977)³³, classification manual used by the insurance industry (Personskadekommittén inom Försäkringsbranschens Serviceaktiebolag, 1989)²⁴, functional independence measure (FIM) scores within the National Trauma Data Bank Research Data Set of the United States²⁵, or some manual criteria defined by studies²². (Table 1)

Outcome measurement of the studies

The outcome measures in these studies evaluated and classified injuries based on different scales. Fifteen studies^7,19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29^,31, 32, 33 assessed injuries according to the abbreviated injury scale (AIS).³⁴ Five studies estimated the risk of permanent medical impairment (PMI) according to the different AIS levels of the different body regions.^{7,19,26,27,31} Seven studies classified the severity of the multiple injuries according to the injury severity score (ISS).^{7,22,26,27,29,32,33} ISS is a combination of the squared AIS from the 3 regions with the highest AIS.³⁵

Total risk of permanent medical impairment (RPMI) was calculated and reported in 11 studies^7,19, 20, 21^,23,25, 26, 27, 28, 29, 30, but 5 remaining studies did not mention directly the total RPMI^22,24,31, 32, 33.

PMI

The summary of the findings of the included studies is available in Table 2. The total RPMI among car occupants following a crash varied from 2% to 23%, and among cyclists varied from 2.8% to 46%. The RPMI among pedestrians was estimated separately as 31% in 1 study.²⁰ Seven studies defined the proportion of persons who sustained at least 10% of initial injury as RPMI 10% +.^{7,20,23,26,27,31,32}

RPMI 1% + (the proportion of persons who sustained at least 1% of initial injury) was estimated at 11.2%⁷, and 8.3%²³ among car occupants, 16.8%³² among cyclists, and was 31%²⁰ among pedestrians. RPMI 10% + was estimated at 1.6%⁷, 5.5%²⁷, and 4.9%²⁶ among injured car occupants, 2.29%³² among cyclists, and was 4.6%²⁰ in pedestrians.

The RPMI following the injuries of different body regions (regardless of AIS levels) was demonstrated in 7 studies.19, 20, 21^,23,25,27,31 Five studies estimated permanent impairment in different AIS levels of different body regions.^{7,19,26,27,31} Injuries were assessed in body regions including head, face, thorax, abdomen, upper extremity, lower extremity, cervical spine, thoracic spine, lumbar spine, pelvis, and external (skin) and thermal injuries. In 6 studies, the neck and cervical spine was the most common body region resulting in PMI after injuries.^7,24,26, 27, 28^,31 Head^19,25, lower extremity^20,21,23, and upper extremity³² were also reported as the common body region to develop PMI (Table 1).

Discussion

Besides a decrease in the rate of mortalities from traffic crashes, management planning to control and reduce long-term outcomes becomes essential for countries, because of its extremely high psychological and financial burden for disabled individuals and society.

Many countries applied AIS and its derivatives, such as ISS, to assess injuries severity of patients. AIS is a measure used by health care professionals to describe, assess, and classified injuries based on body region and severity.⁷ AIS classified injuries by 9 body regions according to 6 points of scale: AIS 1 for minor injuries, AIS 2 for moderate, AIS 3 for serious, AIS 4 for severe, AIS 5 for critical, and AIS 6 for currently untreatable injuries.³⁴ Fifteen studies included in this review ranked injuries based on AIS.^7,19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29^,31, 32, 33 In addition to this, 7 studies assessed the severity of multiple injuries based on ISS scoring.^{7,22,26,27,29,32,33}

Although, these scoring systems were accepted and available in many health care systems around the world, there is still a lack of a globally accepted scale for evaluation of the long-term outcomes of crash injuries. Within the studies included in this review, there have been a few assessment models across the countries to determine the long-term consequences involved injured person after a collision to develop disability.

One of the studies determined permanent impairment according to only the loss of health measured in Rosser's three-dimensional index (the index of health-related quality of life).³⁶ They assess injured individuals by asking them to describe how they feel in terms of functional disability, pain, and distress at different times of follow-up after their crash. The individuals who presented a loss of health after 1-year post-crash determined as a patient with long-term consequences and follow-up visits were continued for 3.5 – 4 years after the crash.²⁹ Telephone interviews and using questionnaires were also applied in some other studies.^22,30

The American Medical Association's guide to the evaluation of permanent impairment³⁷ is used by 1 of the studies and define “impairment” as loss of any body structure or body function, “disablement” as the loss of ability to perform certain tasks and “incapacity” as the loss of social functions.³³ This guide was applied widely in countries to define and assess impairments and disability.³⁸ The strength of this method is evaluation of injury-induced mental and physical impairments. However, this method has less concern about assessment of injury in different body regions and their risk of impairment.

The other model which was used to calculate disability risk in studies is the FIM scores. The FIM contained 18 items, which evaluated physical abilities (including multiple items for self-care, sphincter control, and movement), and cognitive abilities (including communication and social behaviors). Each item was scaled from 1 (means completely task-dependent and disabled) to 7 (means completely independent in function).³⁹ One of our included studies used the truncated FIM score which contained grades 1 to 4 of only 3 items of FIM including the function of self-feed, locomotor function, and verbal functions in injured children. The disability was defined as FIM of 1 or 2 in any of these 3 items. The disability risk was calculated from the ratio of the number of children who sustained a specified injury and classified as disabled according to FIM to the total number of children who sustained a specified injury.²⁵ The strength of FIM method is to consider cognitive impairment in addition to physical impairment, which is less evaluated by other methods of PMI assessment.

Most of the studies in this review were performed in Sweden which has an integrated system for identification, diagnosis, and management of PMI among injured individuals. Swedish insurance companies record all data from traffic crashes. All initial injuries, even minor injuries reported by individuals, were classified based on AIS-2005. Patients were followed 6 – 12 months after the crash, and if the injured patient did not recover, they assessed medical impairment based on the grading rules used by all Swedish insurance companies called grading medical impairment (Sveriges Försäkringsförbund).⁷ In this setting medical impairment was confirmed when physical and/or mental function was reduced, independent of cause and irrespective of the occupation, hobbies, or other special conditions of the injured individual. Follow-up continued for 3 – 5 years after the crash and medical impairment was accounted as permanent if no further recovery was seen despite treatments. Injured persons were assessed to give a PMI degree of 1% – 99%.^7,19, 20, 21^,23,24,26, 27, 28^,31,32 This system seems to be an appropriate method of evaluation and reporting PMI, while it identified patients by the insurance company and the traffic information reporting immediately after accident and then examined and followed up for 5 years by the medical team to find permanent impairment. Therefore, the missed data is greatly reduced and reliable data is obtained. However, it would be better to modify this system by adding the psychological impairment items and criteria.

Despite this same definition of impairments in Swedish studies, reported statistics were not comparable due to different population characteristics (i.e., group of age), different types of vehicles (bicycle or car), different inclusion/exclusion criteria, and different degrees of PMI reported (PMI 1% +, PMI 10% +, etc.). Therefore, the existence of an integrated evaluation method plays an important role in accurate estimation of prevalence of PMI. We should analyze systematically and compare the details of different studies over the years.

Some of the studies assessed PMI of different body regions in different AIS levels and report the RPMI separately. We found that the cervical spine is more delicate to trauma in comparison to other body regions, since data showed that injuries to the neck and cervical spine are more likely to be permanent (Table 1).^7,24,26, 27, 28^,31 Other injuries which make up a large percent of permanent impairment are those involved in the lower and upper extremities and after that thoracic and lumbar spine, face, and head. 19, 21^,23,²⁵,³². These results should take into consideration public health and traffic system and their management plans.

The risk for the more severe injuries according to AIS scoring, such as AIS 5, was more likely to become permanent, compared with lower levels of AIS, such as AIS 1 and AIS 2 injuries. However, generally, lower levels of AIS injuries account for a higher percentage of the total rate of permanent impairment, which is due to the higher initial frequency of these injuries.^7,19 Since traffic crash is one of the most considerable causes of trauma among children and their subsequent loss of health and disability, it is necessary to study the population of children injured in road traffic crashes.²⁵ Four studies included in the present review investigated persistent impairment in children following vehicle crashes.^19,25,27,30 PMI among children had a range of 0.29% – 10.0%.

There were some limitations to this review. Only studies with at least 3-year follow-up were included in this review and, likely, further studies with lesser follow-up time or studies that did not indicate the follow-up time clearly were thereby excluded. Another limitation was the great heterogeneity of data, method of studies, and their outcome measures, which limited us to pool result statistics.

STROBE score is recommended to better evaluate observational studies. STROBE score tries to find potential confounders and effect modifiers in the observational studies based on a checklist of 22 items on the title, abstract, introduction, methods, results, and discussion of each paper. Another strength of STROBE is the capability of this checklist in evaluation of all 3 major types of observational studies: cohort, case-control, and cross-sectional studies. The quality of the included studies assessed by STROBE revealed that the most of the studies had intermediate quality. All these 22 sections scored as 0 – 1 (as yes or no/not applicable), while each of them evaluates different items, that are not scored separately in the scoring method. In the “Variables” section, all the included studies received no score due to not defining potential confounders and effect modifiers. Most of the studies received no point in the “Participant” section which contain 3 different items, while they did not consider using flow diagram. Furthermore, no study got a point in the “Descriptive data” section because of item “(a) Give characteristics of study participants (e.g. demographic, clinical, social) and information on exposures and potential confounders” which evaluates the information on potential confounder in studies. Most of the studies scored 0 in the “Main results” section due to lack of report category boundaries of continuous variables.

Despite the various models of determination of PMI across the world, no globally accepted method has been validated relative to PMI after RTIs. Furthermore, measuring more subjective outcomes such as pain and mental functional reduction should be considered in the present methods of evaluation. Further studies in the road traffic crash field are better to be performed according to a fully integrated valid assessment tool for impairment rating. This would facilitate comparing risks of impairment among countries and during the time, help to make better decisions for the management of traffic crashes, and thereby reduce permanent impairments and related financial and psychological burden for individuals and communities.

Funding

This work was funded by Sina Trauma and Surgery Research Center, Tehran University of Medical Sciences (grant number is 98-01-38-41758).

Ethical statement

The Ethics Committee of Sina Trauma and Surgery Research Center, Tehran University of Medical Sciences, approved the study, and the reference number is 98-01-38-357.

Declaration of competing interest

The authors announce that there is no conflict of interest in this research.

Author contributions

Study conception and design: Mahla Babaie, Mohammadamin Joulani and Vafa Rahimi-Movaghar; Search strategy: Morteza Gholami; Methodological and epidemiological consults: Marzieh Nojomi; Title/abstract screening: Mahla Babaie, Mohammadamin Joulani, Mohammad Hossein Asgardoon and Mohammad Hosein Ranjbar Hameghavandi; Full text screening and data extraction: Mahla Babaie, Mohammadamin Joulani, Mohammad Hossein Asgardoon and Mohammad Hosein Ranjbar Hameghavandi; Quality assessment: Mahla Babaie and Vafa Rahimi-Movaghar; Interpretation of results: Mahla Babaie and Mohammadamin Joulani; Draft manuscript preparation: Mahla Babaie, Mohammad Hossein Asgardoon and Zahra Ghodsi; Revision and structural correction: Gerard M O'Reilly, Mahla Babaie, Mohammadamin Joulani and Vafa Rahimi-Movaghar.

All authors reviewed the results and approved the final version of the manuscript.

Appendix A. Supplementary data

The following is the Supplementary data to this article: