- Email: [email protected]
Persistent medical problems and permanent impairment: Injuries associated with work, vehicles, and sports
Arci&. Anal. and Preu. Vol. 26, No. 1. pp. 41-48. Prinki in the U.S.A.
~1~57S/~4 $6.00 0 1993 Pergamon Pres.
1994
.oO Ltd.
PERSISTENT MEDICAL PROBLEMS AND PERMANENT IMPAIRMENT: INJURIES ASSOCIATED WITH WORK, VEHICLES, AND SPORTS ULF ‘The Umed Accident
BJ~RNSTIG’
and TORE J. LARSSON~
Analysis Group, University Hospital; S-901 85 Umei, Sweden; ‘IPSO Australia, 41 Koornalla Crescent, Mt. Eliza 3930, Victoria, Australia (Received 22 June 1992;in revisedform IO Februury
1993)
Abstract-Among injuries treated in one year at the University Hospital in Umel, Sweden, work- and sportsrelated accidents caused 16% each and vehicle-related accidents caused 12% of all injuries treated. Most fatalities and severe injuries were associated with vehicles. The proportion of victims with permanent medical impairment was highest among vehicle- and work-related injuries- 6%. The proporiion of persons who reported some form of persistent medical problem five years after the accident was similar in the three groups-between 21% and 24%. Among those injured in work- and vehicle-related accidents, one percent had been retired early or were on tong-term sick feave. In all three groups, between 1% and 3% had changed jobs because of the injury, and between 3% and 6% had changed their leisure time activities to a large degree. Twelve percent of the work-related injuries and 8% of the vehicie-related injuries, but only 3% of the sportsrelated injuries, were claimed under some sort of insurance scheme. Pain was the most common persistent medical problem. Pain, together with serious brain and spinal injuries, is rather poorly defined in the impairment classification presently used by Swedish insurers to determine degree of permanent medical impairment. Improved and clarified criteria would be of great value.
aktieboiag 1989). In this guide, containing arollnd 1,000 classifications, percentages are set to rey,-esent specific anatomical, physical and mental ilefects. We have shown, in another study of regiolral injury material, collected from three differ nt sources, that among persons injured at work. I% suffered a permanent medical impairment (Lars:,#jn and BjGrnstig 1993). However, every fifth injur Ed person had some form of persistent medical probl, m five years after the accident (Larsson and Bjiirn: ig 1990; 1993). Nygren, Hansson, and Tingvall (19.~6) and Clay, Van Kampen, and Hogerzeil (1987) hz ge reported on residual medical problems after rc ,td traffic accidents and have shown that “minor” inj iries also can cause permanent disability. Vehicles not used in road traffic, like snowmobiles and ;~llterrain vehicles (frequently used in our area of study), have been shown to be associated with p rmanent disabilities and fatalities (Allan et al. 19X3; Bjiirnstig, Eriksson, and Mellbring, 1984; Bjdrns, ig and Bylund 1986; Eriksson and BjGrnstig 1982). Residual medical problems after sports injuries have been reported (Lanese et al. 1990), but we have fourtd no studies on permanent medical impairment afr,x
BACKGROUND Permanent medical impairment and disability resulting from accidents constitute a major social, medical, and economic problem (Jonsson 1973). The proportions of injured persons with persistent medical problems and permanent medical impairment after different types of injury events are, however, virtually unknown. A few studies report proportions of remaining problems, persistent loss offunction, or simil2.r descriptions, not always unambiguously defined (Allan, Reid, and Saboe 1988; Holman, Olszewski, and Maier 1987; Jansson and Jacobsson 1988; Lanese et al. 1990; Lings et al. 1984; MacKenzie et al. 1988; Maimaris, Barnes, and Allen 1988; Nakayama, Gardner, and Rogers 1990; WalIer, Payne, and Skelly 1989). A few studies describe the proportion of trauma patients returning to work one year after injury (MacKenzie et al. 1987; 1988). Some recent studies contemplate the measurement of disability and suggest methods for improving the description of results of impairment (States and Viano 1990; Yates et al, 1991). In Sweden, permanent medical impairment is defined in a classification manual used by the Insurance Industry (Personskadekonlmitt~n inom F~rs~kringsbranchens Service41
42
U. BJ~RNSTIG and T. J. LARSSON
Table
1. Subpopulations
= I = 2 2 3
Total
to severity
Table
Work
Vehicle
Sports
1292 (86%) 203 (13%) 11 (0.7%)
668 (61%) 376 (34%) 47 (4%)
962 (66%) 494 (34%) 5 (0.3%)
Severity MAIS MAIS MAIS
according
I506
1461
1091
sports-related injuries from an area representing the Nordic spectrum of sports. The purpose of this study was to compare and analyse permanent medical impairment, persistent medical problems, and social consequences of injuries from three different areas-work, vehicles, and sports-treated at a large regional hospital in Sweden. MATERIAL During one year (1 April 1985 to 31 March 1986), 9,172 injured patients were primarily treated at the University Hospital in Umei. The hospital serves an area with 118,000 inhabitants. From an epidemiological point of view, this hospital has the great advantage that nearly all the trauma cases from the area are treated at the hospital, either as out- or inpatients. Only 10% of all injuries in the area, mostly minor injuries, are treated by general practitioners at local outpatient clinics. The analysed population were persons who had sustained injuries while working and persons with injuries associated with the use of vehicles or injuries sustained during sports activities. To map persistent medical problems, permanent medical impairment, and social consequences of the injury, a sample was drawn from the three subpopulations. Since severe injuries are probably associated with a higher incidence of persistent medical problems and permanent medical impairment, the sample was stratified according to Maximum AIS. According to the Abbreviated Injury Scale (Committee on Injury Scaling 1985), AIS = 1 refers to a minor injury and AIS = 6 to a virtually unsurvivable injury. The proportions selected were for MAIS = 1, lo%, for MAIS = 2, 25%, and for MAIS 2 3, 100%. The number of cases in the subpopulations and the stratified samples are shown in Tables 1 and 2. The amount of missing data in the Table
2. Stratified
Severity MAIS MAIS MAIS Total
sample Work
= 1 = 2 2 3
according Vehicle
to severity
3. Missing
cases
in the sample
according
to category
Permanent impairment Category Work Vehicle sport
Missing cases
Fatal injury
2 34 3
Known
Unknown
1 22 3
1 10 0
0 2 0
strata is shown in Table 3. Among those not reached for interview, information about medical impairment in cases with known disability have been gleaned from medical records. The most common injury events among the work accidents were foreign body in eye, 20%; sawn/cut/grazed, 18%; hit by/against moving object, 13%; and jammed/squeezed between hard objects, 9%. The different types of road users among the vehicle-related accidents in the research population were: Bicyclists Drivers/passengers Motorcyclists Pedestrians Drivers/passengers mobiles Others
in cars
on snow-
46% 28% 7% 4% 4% 10%
The persons with sports related injuries had been injured in the following activities: Team sports (55%) football (soccer) handball, basketball, ball ice hockey indoor bandy other team sports
and volley-
24% 11% 8% 6% 6%
Individual sports (48%) down hill skiing gymnastics equestrian cross-country running badminton other ind. sports
11% 7% 7% 3% 3% 14%
sports
129 51 II
66 95 47
96 125 5
I91
208
226
METHOD All injured persons were interviewed about the origin of their injuries at the initial visit to the emergency department of the hospital. Information on
Persistent
medical
problems
sick leave from the local branch of the National Social Insurance Office was added to all relevant cases. A survey was distributed by mail to all persons in the sample about five years after the accident. Persons not responding were reminded by mail and t’zlephone until contact was established with all who t-easonably could be reached. Medical records and i~lsurance records were analysed. Complementary telephone interviews were conducted with all persons who had indicated persistent medical problems in the survey, in order to specify the nature of the problems. Clinical examinations were undertaken in cases where permanent medical impairment was unassessed, but could be suspected from the interview. Impairment was assessed according to the guidelines presently used by Swedish insurance companies. In cases where compensation for permanent disability had been awarded, information was procured from the relevant insurer. The term medical impairment implies an alteration of health status assessed by medical means, whereas the term disability denotes an alteration of a person’s capacity to meet personal, social, or occupational demands assessed by nonmedical mean’s (AMA 1984). In this study,persistent medical prohkms, are self-reported medical problems not always possible to describe as impairing, according to present guidelines. Statistical methods were used to control whether the samples were representative of the subpopulations from which they were drawn. The scale level of the variables and the number of observations in each cell decided what tests were used; z test, x2 test, or Kolmogorov-Smirnov’s test (Eilertsson 1984; Henriksson 1990). The estimate of total p-vallies of each variable in the samples was based
and permanent
43
impairments
on compounded p-values and their variances (MI :ndelhall, Ott, and Schaefer 1971). Estimates on true values of relevant variables in the subpopulatic’ns are given as percentages and integers. Representativity was checked in the sample of work injuries for the four most common injl,ry events mentioned above, in the sample of vehi :le associated injuries for car occupants, bicyclists, IInd others, and in the sample of sports-related injuries for team and individual sports. P < 0.05 was u*‘ed as the statistically significant probability level. 1 he studied stratified samples were in all these aspei?s representative of their subpopulations. Work-related injuries were defined as injuries contracted during working hours. Two persons in the sample were injured in vehicle accidents dur ng working hours (MAIS = 2), and one person v as injured during sports activities while at w( rk (MAIS = 2). These cases have been allocated to vehicle and sports samples, respectively. Vehicle-related injuries were defined as injure i’s contracted in association with moving vehicles, neluding not only those injured in traffic accidents m public roads, but also moto-cross riders, sno’vmobile riders, and riders of other vehicles “,bff road”. Sports-related injuries were defined as injur 2s contracted during some form of systematic spot ts activity, including equestrian activity.
RESULT Sex and age The proportion of females was lowest among persons injured at work-12%-while the tort z-
%
60 m
Work Vehicle
j
Sports
Age -19 Fig.
20-29
30-39
1. Age distribution
40-49 in the three
50-59
60-
subpopulations
U. BJ~RNSTIG and T. J. LARSSON
44
Eye Head
Trunk,
back
Upper
extr.
Lower
extr.
0
20
Fig. 2. Injury
locations
40
in the three
60
subpopulations
The three subpopulations (Table 1) showed completely different injury patterns. The work injuries contained a large proportion of minor eye injuries and superficial injuries to hands, while the vehicle injuries had a large number of head and face injuries. The sports injuries most often affected the lower extremities in the form of sprains and fractures (Figs. 2 and 3).
sponding figures were 32% among sports-related injuries and 38% among vehicle related injuries. There were great differences between the three subpopulations regarding age distribution. More than 80% of those injured in sports were under 30 years of age, and 50% were under 20. Among the vehicle-related injuries, there was also a large proportion of young people-one in three was under 20-but this subpopulation also contained a large proportion of persons aged 60 or more (Fig. 1). Severity
%
U
I
Sick leave
The proportion of persons reported to have been on sick leave due to the injury was about onethird in all subpopulations. Those who had suffered a vehicle-related injury had a significantly longer average period on sick leave than the others. The follow-up period for sick leave was limited to one year after the event. Among those injured in work related accidents, eight persons were listed sick for more than one year, six with MAIS = 2 injuries and two with MAIS 2 3 injuries. Among the vehiclerelated accidents the corresponding figure was nine persons, three at each severity level indicated in Table 5, and among the sports-related accidents one person, with an MAIS = 2 injury. Slightly higher
of the injuries
The work-related injuries contained a much larger proportion of minor injuries (MAIS = 1) than the other categories (Table 1). Among persons injured in vehicle- and sports-related accidents the proportion with moderate or more severe injuries (MAIS 2 2)-most often a fracture or a concussion-was one-third. Injuries associated with vehicles had the largest proportion of serious injuries (MAIS 2 3), and ten persons in this subpopulation (1%) died from their injuries. One person died in a work accident. No person was fatally injured in any sporting activity.
Wound/contusion Sprain/strain Fracture/h&.
I
Concussion Foreign
body
Vehicle
I
+
$
I
0 Fig. 3. Injury
20 types
in the three
I
40 subpopulations.
/
60
%
45
Persistent medical problems and permanent impairments
Table 4. Proportion listed sick (%) and average length of sick listing period for the three subpopulations, among persons 15-65 years of age
Table 6. Proportion listed sick (%) over and under 40 years of, ge, among persons 15-65 years of age, for the three subpopulatl 3ns -
(n = 1492)
Vehicle (n = 794)
sport (n = l(177) -
33% 45%
21% 37%
20% 42%
Work Work (n = 1492)
Vehicle (n = 794)
Sports (n = 1077)
37%
35%
30%
34
55
31
Proportion listeli sick Averag:e time (days)
<40 years of age >40 years of age
propcrtions of younger and older persons among those injured at work were listed sick, compared with the other two subpopulations. The vehicle and sported populations, however, probably contain a number of people outside the workforce and thus without sick benefits (Tables 4, 5, 6, 7). Persistent the elent
medical
problems
jive years
after
About the same proportion of persons in the three subpopulations reported some form of persistent medical problem five years after the event. This was tlue in spite of large variations in sex and age distrib’ution, injury localization and type, and injury severity between the subpopulations. Among workand vehicle-related injuries, 21% reported persistent medic,21 problems, and among sports injuries 24%. Tape of medical problem. The most common problem was pain. One in five persons injured in a sport activity and one in ten among those injured at work, reported some type of pain five years after the accident. In Tables 8 and 9 are shown the problems most commonly reported. Those interviewed were allowed to name up to three different complaints. Permanent
medical
impairment
Among those reporting persistent medical problems, all cases where permanent medical impairment could be suspected were called in for clinical examination. The number of persons with a permanent medical impairment was six (+2)% among the work-
Table 5. Proportion listed sick (%) and median sick listing period (m) among persons 15-65 years of age, by MAIS for the three subpopulations --
MAIS = 1 MAIS = 2 MAIS = 3
Age
Work (n = 1492)
Vehicle (n = 794)
Sports (n = 1077)
%
m
%
m
%
m
30 76 91
12 42 121
17 39 36
5 35 45
17 32 40
11 32 130
-
related injuries, six (+-3)% among the vehicle-relal ed injuries, and one (* 1%) among the sports-relal ed injuries. One percent of the vehicle-related injurres were fatal. Among the persons with permanent rn’,dical impairment, the number of minor injurles (MAIS = 1) was highest among the work-relay :d injuries-l/3-and only l/6 among the vehicle- ‘elated injuries. There were no cases of permanc nt impairment after a minor injury among the spor ~srelated injuries. However, cases of medical impLlirment among persons who have not reported ally medical problems remain unknown and unassesst d. Reports
to insurance
Persons injured at work had the largest propl’rtion of injuries reported to an Insurance or Compcmnsation scheme. Among the persons with a permanc’nt impairment, the reporting frequency was highc,st among those injured at work or in vehicle-relatl.:d accidents. Only a small number of sports relat;:d injuries had been reported (Table 10). Change
of work
Among persons with permanent medical impal rment, one in five who had been injured in a workor vehicle-related accident had changed jobs or WOIk tasks because of the injury. Sports injury cases II:ported a considerably lower rate of job change. C: Iculations were based on living persons (1990) agcmd 15 to 65 at the time of injury (Table 11). Change
in leisure-time
activities
More than one in three persons with a perm ~Inent medical impairment after a work- or vehicl;.‘related accident reported that they had changed the1 r Table 7. Proportion listed sick (%) and average sick listing perild (.?) among persons 15-65 years of age, by injury location for ttle three subpopulations Work (n = 1492)
Vehicle (n = 794)
Injury location
%
x
%
x
%
i
Head, Trunk, Upper Lower Eyes
8 8 40 22 22
23 38 33 41 17
28 23 23 25 1
44 69 55 56 3
12 5 26 54 2
(1 1‘ 3t1 3.’ .’
face back extr. extr.
sports (n = 107,‘) ___.
U. BJ~RNSTIGand T. .I. LARSON
46
Table 8. Proportion of persons with different persistent medical problems in the three subpopulations (95% confidence interval) Work (n = 191)
Problem
Vehicle (n = 208)
Table 10. Proportion of persons who had reported the injury to insurance or compensation scheme in the three subpopulations (95% confidence interval)
Sports (n = 226) Group
Pain Loss of strength or sensibility, sense of cold Functional defect to joint Anatomic defect Headache
10 (?4)% 6 (?3)%
1.5 (?S)% 3 (t-3)%
20 (+-S)% 2 (-+l)%
5 (?2)%
2 (&2)%
6 (-+3)%
2 (?2)% -
1 (+l)%
-
4(C3)%
-
leisure-time activities to a considerable degree. The corresponding figure among sports-related injuries of course was much higher-persistent medical problems after a sports accident often renders the victim unable to continue that particular activity (Table 12). Early retirement Early retirement or Iong-term sick leave (full or part time) was most frequent among the work-related permanent medical impairments. The proportion was almost as large among the vehicle-related injuries, while the corresponding figure among the sports-related injuries was very small (Table 13).
DISCUSSION Of the 11 fatahties in the study, ten were associated with vehicle accidents and one with work. This illustrates where the fatality risks are the highest. Among the work- and vehicle-related injuries, one person among 17 surviving suffered a permanent medical impairment, while the corresponding figure among sports injuries was considerably lower. Low average age and, therefore, good rehabilitation capacity among the sports-injury cases might explain the difference. Comparable results have not been found in the scientific literature.
Table 9. Proportion (%f of persons (aged 15-65) with a MAIS 2 2 injury and average sick listing period (f), among persons with medical impairment, persistent problems and among recovered in the three subpopulations Work (n = 1492)
Impaired Persistent problems Recovered
Vehicle (n = 794)
sports (n = 1077)
%
x
%
.?
%
x
67 38
109 24
85 50
68 9
100 60
34 31
8
9
34
9
26
7
All injured Persons with persistent medical problems Persons with permanent medical impairment
Work (n = 191)
Vehicle (n = 208)
Sports (n = 226)
12 (?4)% 57%
8 (t4)% 39%
3 (+-I)% 12%
89%
88%
33%,*
* In this table four persons with MAIS = 2 injuries and persistent medical problems, not yet at a medically stationary level, have been excluded. If these four are classified as permanent disabilities > 0 the figure will be 46%. Percentages given without confidence intervals above represent only the samples interviewed.
Sports injuries could imply long-term problems, however. In a popuIation of predominantly young people, with a rather high proportion of persistent medical problems after injury-often in the form of pain-these injuries could constitute problems over long periods in life. The medical problems would presumably interact with future exposures in working life, with unknown effects. In spite of the fact that the proportion of medically significant (MAIS 2 2) injuries was almost three times larger among the vehicle- and sportsrelated injuries compared to the work related injuries, one person in five in all three injury groups reported some form of persistent medical problem five years after the injury. The present study indicates that nine out of ten permanently impairing work- and vehicle-related injuries were reported to an insurance or compensation scheme, while the proportion of thus reported sports injuries was considerably lower. For the work injury cases, this is in line with earlier findings (Larsson and Bjornstig 1993). The inclination to report
Table 11. Proportion of persons who had changed jobs or tasks because of the injury in the three subpopulations (95% confidence interval) Group All injured Persons with persistent medical problems Persons with permanent medical impairment
Work (n = 191)
Vehicle (n = 208)
sports (n = 226)
2 (?2)% 9%
3 (231% 11%’
I (?2)%
20%
18%
-
Percentages without confidence only the samples interviewed.
intervals
5%
given represent
Persistent
medical
problems
Table 12. Proportion of persons who reported major change of leisur: time activity because of the injury in the three subpopulations (95% confidence interval) Work (n = 191)
Group All in, ured Perso IS with persistent medical pros,lems Perso’x with permanent medical impair ment Pxcentages only tile samples
Vehicle (n = 208)
Sports (n = 226)
3 (?2)% 14%
4 (?2)% 19%
6 (?2)% 52%
40%
37%
76%
given without interviewed.
confidence
intervals
represent
the injury to an insurance or compensation scheme of ccurse varies with the quality of the insurance covei-age for the population in question and the inclination to file a claim for compensation on the part of thr: individual. Our results indicate, however, that injur!/ data collected for insurance purposes can yield relevant information about permanently impairing work- and vehicle-related injuries, but is of less value in evaluating permanently impairing sports injuries and injuries that are not followed by permanent medical impairment. Early retirement and long-term sick leave was the consequence for one person in five suffering a perm,&nently impairing work-related injury and one person in seven suffering a permanently impairing vehicle-related injury. This indicates considerable social costs (Jonsson 1973). The aetiology of permanentl;i impairing injuries is probably different to the aetiology of fatal and nonimpairing injuries (Bj(irnstig et -11. 1984; Erikssonand Bjiirnstig 1982; Bjiirnstig et al. 1985; Bjiirnstig and Bylund 1986; Larsson 1990) In view of the present results, the aetiology of permanently impairing injuries stands out as an area in need of further scrutiny. I,? the stratified samples of this study, the medi-
Table 13. Proportion of persons (aged 15-65) in early retirement or still five years after the injury, on (full- or part-time) sick leave (95% confidence interval) Work (n = 191)
Group All injured Person!, with persistent medical problems Person! with permanent medical impairment Percentages only the samples
1 (?l)% 6%
22%
given without interviewed.
Vehicle (n = 208) 1 (?l)% 4%
0.3 (?0.6)% 2%
-
14%
confidence
sports (n = 226)
intervals
represent
and permanent
47
impairments
tally serious and more severe injuries were \I/ell represented. That minor injuries can cause pervnanent medical impairment and disability has bi:en shown elsewhere (Nygren et al. 1986; Larsson ;tnd Bjijrnstig 1993). Some types of medically less se\ :re injuries, thus, cannot be disregarded in the stud:, of permanent medical impairment and disability. The present Swedish classification guide to llermanent medical impairment (Personskadekomr,littCn inom F6rsZkringsbranschens Serviceaktiebc! lag 1989) seems to focus quite disproportionally on c Iassifications of injuries to hand and arm (90%) .lnd mainly describes injuries with some measurable a Inatomical defect. Pain was the most common per’ istent medical problem in this study. Pain, togel ler with serious brain and spinal injuries, is ratller poorly defined in the classification presently used by Swedish insurers to determine degree of permanl.:nt medical impairment. Persons suffering persistl.:nt pain after injury are thus disadvantaged by the l;lck of well-defined evaluation criteria. Improved ;I nd clarified criteria would be valuable. A more detai ed application, similar to the Guide to the Evaluat on of Permanent Impairment (AMA 1984), is recc’nmended. Acknowledgement-The
study was financed by the Swet ish Work Related No-Fault Liability Insurance (TFA) through agl mt to IPSO Sweden. We thank Anna Maria Backlund and Per C ov Bylund warmly for their hard and scrupulous assistance and ‘IVidar Henriksson for statistical advice.
REFERENCES Allan, D. G.; Reid, D. C.; Saboe, L. Off-road recreation lal motor vehicle accidents: Hospitalizations and deat IIS. Can. J. Surg. 31:233-236;
1988.
AMA (American Medical Association). Guide to the et IIuation of permanent impairment. 2nd edition. Chica o; American Medical Association; 1984. Bjarnstig, U.; Eriksson, A.; Mellbring G. Snowmobile ng injuries. Mechanisms and consequences. Acta C~Iir. &and. 150:619-624; 1984. Bjiirnstig, I-J.; Bylund, P.-O.; Lekander, T.; Brorsson, B. Motorcycle fatalities in Sweden. Acta Chir. Scarld. 151577-581; 1985. Bjiirnstig, U.: Bylund, P.-O. Motorcycle injuries in nor1 hern Sweden. Travel Medicine International 4: 16- ‘1: 1986. Clay, W.; van Kampen, L. T. B.; Hogerzeil, H. H. IV. Injury and disability effects of motor car acciden’s. Int. Disabil. Stud. 9:145-148; 1987. Committee on Injury Scaling. The abbreviated injl ~1 scale. 1985. Revision. Morton Grove, IL: Americ tn Association for Automotive Medicine; 1985. Eilertsson, G. Grundlgggande statistik med till%mpning ar inom sjukvlrden. Lund, Sweden: Studentlitteratllr; 1984. Eriksson, A.; Bjiirnstig, U. Fatal snowmobile acciderlts in northern Sweden. J. Trauma 22:977-982; 1982.
48
U. BJ~RNSTIG and T. J. LARSSON
Henriksson, W. Kompendium i statistik. Icke-parametriska metoder. UmeH, Sweden: Pedagogiska institutionen, Umel Universitet; 1990. Holman, R. G.; Olszewski, A.; Maier, R. V. The epidemiology of logging injuries in the Northwest. J. Trauma 27:1044-1050; 1987. Jansson, B. R.; Jacobsson, B. S. Medical consequences of work-related accidents on 2,454 Swedish farms. Stand. J. Work Environ. Health 14:21-26; 1988. Jonsson, E. Samhallskostnader for trafikolyckor, yrkesskador och sjukdomar orsakade av tobaksrok. Stockholm: Liber; 1973. Lanese, R. R.; Strauss, R. H.; Leitzman, D. J.; Rotondi, A. M. Injury and disability in matched men’s and women’s intercollegiate sports. Am. J. Public Health 80: 1459-1462; 1990. Larsson, T. J.: Accident information and priorities for prevention. Thesis. Stockholm: Royal Institute of Technology, IPSO Factum 21; 1990. Larsson, T. J.; Bjiirnstig, U. The epidemiology of occupational accidents: Three ways to measure the problem in a Swedish municipality. J Occup. Health Safety-Aust NZ; 6( 1):39-52; 1990. Larsson, T. J.; Bjornstig, U. Persistent medical problems and permanent impairment five years after occupational injury. Accepted: Stand. J. Sot. Med. 1993. Lings, S.; Jensen, J.; Christensen, S.; Mgller, J. T. The consequences to the injured of occupational accidents. Stand. J. Sot. Med. 12:26-29; 1984. MacKenzie, E. J.; Shapiro, S.; Smith, R. T.; Siegel, J. H.; Moody, M.; Pitt, A. Factors influencing return
to work following hospitalization for traumatic injury. Am. J. Public Health 77:329-334; 1987. MacKenzie, E. J.; Siegel, J. H.; Shapiro, S.; Moody, M.; Smith, R. T. Functional recovery and medical costs of trauma: An analysis by type and severity of injury. J. Trauma 28:281-297; 1988. Maimaris, C.; Barnes, M. R.; Allen, M. J. “Whip lash injuries” of the neck: A retro-spective study. Injury 19:393-396; 1988. Mendelhall, W.; Ott, L.; Schaeffer, R. L. Elementary survey sampling. Belmont, CA: Wadsworth Publishing Comp Inc., 1971. Nakayama, D. K.; Gardner, M. J.; Rogers, K. D. Disability from bicycle-related injuries in children. J. Trauma 30:1392-1394; 1990. Nygren, A; Hansson, P. G.; Tingvall, C. Acute injury scaling related to residual disability. Acta Neurochir. Suppl. 36:25-27; 1986. Personskadekommitten inom Forsakringsbranschens Serviceaktiebolag. Grunder for gradering av kvarstaende men efter skador (medicinsk invaliditet). Stockholm: 1989. States, J. D.; Viano, D. C. Injury impairment and disability scales to assess the permanent consequences of trauma. Accid. Anal. Prev. 22:151-160; 1990. Waller, J. A.; Payne, S. R.: Skelly, J. M. Injuries to carpenters. J. Occup. Med. 31:687-692; 1989. Yates, D. W.; Heath, D. F.; Mars, E.; Taylor, R. J. A system for measuring the severity of temporary and permanent disability after injury. Accid. Anal. Prev. 23:323-329; 1991.
~1~57S/~4 $6.00 0 1993 Pergamon Pres.
1994
.oO Ltd.
PERSISTENT MEDICAL PROBLEMS AND PERMANENT IMPAIRMENT: INJURIES ASSOCIATED WITH WORK, VEHICLES, AND SPORTS ULF ‘The Umed Accident
BJ~RNSTIG’
and TORE J. LARSSON~
Analysis Group, University Hospital; S-901 85 Umei, Sweden; ‘IPSO Australia, 41 Koornalla Crescent, Mt. Eliza 3930, Victoria, Australia (Received 22 June 1992;in revisedform IO Februury
1993)
Abstract-Among injuries treated in one year at the University Hospital in Umel, Sweden, work- and sportsrelated accidents caused 16% each and vehicle-related accidents caused 12% of all injuries treated. Most fatalities and severe injuries were associated with vehicles. The proportion of victims with permanent medical impairment was highest among vehicle- and work-related injuries- 6%. The proporiion of persons who reported some form of persistent medical problem five years after the accident was similar in the three groups-between 21% and 24%. Among those injured in work- and vehicle-related accidents, one percent had been retired early or were on tong-term sick feave. In all three groups, between 1% and 3% had changed jobs because of the injury, and between 3% and 6% had changed their leisure time activities to a large degree. Twelve percent of the work-related injuries and 8% of the vehicie-related injuries, but only 3% of the sportsrelated injuries, were claimed under some sort of insurance scheme. Pain was the most common persistent medical problem. Pain, together with serious brain and spinal injuries, is rather poorly defined in the impairment classification presently used by Swedish insurers to determine degree of permanent medical impairment. Improved and clarified criteria would be of great value.
aktieboiag 1989). In this guide, containing arollnd 1,000 classifications, percentages are set to rey,-esent specific anatomical, physical and mental ilefects. We have shown, in another study of regiolral injury material, collected from three differ nt sources, that among persons injured at work. I% suffered a permanent medical impairment (Lars:,#jn and BjGrnstig 1993). However, every fifth injur Ed person had some form of persistent medical probl, m five years after the accident (Larsson and Bjiirn: ig 1990; 1993). Nygren, Hansson, and Tingvall (19.~6) and Clay, Van Kampen, and Hogerzeil (1987) hz ge reported on residual medical problems after rc ,td traffic accidents and have shown that “minor” inj iries also can cause permanent disability. Vehicles not used in road traffic, like snowmobiles and ;~llterrain vehicles (frequently used in our area of study), have been shown to be associated with p rmanent disabilities and fatalities (Allan et al. 19X3; Bjiirnstig, Eriksson, and Mellbring, 1984; Bjdrns, ig and Bylund 1986; Eriksson and BjGrnstig 1982). Residual medical problems after sports injuries have been reported (Lanese et al. 1990), but we have fourtd no studies on permanent medical impairment afr,x
BACKGROUND Permanent medical impairment and disability resulting from accidents constitute a major social, medical, and economic problem (Jonsson 1973). The proportions of injured persons with persistent medical problems and permanent medical impairment after different types of injury events are, however, virtually unknown. A few studies report proportions of remaining problems, persistent loss offunction, or simil2.r descriptions, not always unambiguously defined (Allan, Reid, and Saboe 1988; Holman, Olszewski, and Maier 1987; Jansson and Jacobsson 1988; Lanese et al. 1990; Lings et al. 1984; MacKenzie et al. 1988; Maimaris, Barnes, and Allen 1988; Nakayama, Gardner, and Rogers 1990; WalIer, Payne, and Skelly 1989). A few studies describe the proportion of trauma patients returning to work one year after injury (MacKenzie et al. 1987; 1988). Some recent studies contemplate the measurement of disability and suggest methods for improving the description of results of impairment (States and Viano 1990; Yates et al, 1991). In Sweden, permanent medical impairment is defined in a classification manual used by the Insurance Industry (Personskadekonlmitt~n inom F~rs~kringsbranchens Service41
42
U. BJ~RNSTIG and T. J. LARSSON
Table
1. Subpopulations
= I = 2 2 3
Total
to severity
Table
Work
Vehicle
Sports
1292 (86%) 203 (13%) 11 (0.7%)
668 (61%) 376 (34%) 47 (4%)
962 (66%) 494 (34%) 5 (0.3%)
Severity MAIS MAIS MAIS
according
I506
1461
1091
sports-related injuries from an area representing the Nordic spectrum of sports. The purpose of this study was to compare and analyse permanent medical impairment, persistent medical problems, and social consequences of injuries from three different areas-work, vehicles, and sports-treated at a large regional hospital in Sweden. MATERIAL During one year (1 April 1985 to 31 March 1986), 9,172 injured patients were primarily treated at the University Hospital in Umei. The hospital serves an area with 118,000 inhabitants. From an epidemiological point of view, this hospital has the great advantage that nearly all the trauma cases from the area are treated at the hospital, either as out- or inpatients. Only 10% of all injuries in the area, mostly minor injuries, are treated by general practitioners at local outpatient clinics. The analysed population were persons who had sustained injuries while working and persons with injuries associated with the use of vehicles or injuries sustained during sports activities. To map persistent medical problems, permanent medical impairment, and social consequences of the injury, a sample was drawn from the three subpopulations. Since severe injuries are probably associated with a higher incidence of persistent medical problems and permanent medical impairment, the sample was stratified according to Maximum AIS. According to the Abbreviated Injury Scale (Committee on Injury Scaling 1985), AIS = 1 refers to a minor injury and AIS = 6 to a virtually unsurvivable injury. The proportions selected were for MAIS = 1, lo%, for MAIS = 2, 25%, and for MAIS 2 3, 100%. The number of cases in the subpopulations and the stratified samples are shown in Tables 1 and 2. The amount of missing data in the Table
2. Stratified
Severity MAIS MAIS MAIS Total
sample Work
= 1 = 2 2 3
according Vehicle
to severity
3. Missing
cases
in the sample
according
to category
Permanent impairment Category Work Vehicle sport
Missing cases
Fatal injury
2 34 3
Known
Unknown
1 22 3
1 10 0
0 2 0
strata is shown in Table 3. Among those not reached for interview, information about medical impairment in cases with known disability have been gleaned from medical records. The most common injury events among the work accidents were foreign body in eye, 20%; sawn/cut/grazed, 18%; hit by/against moving object, 13%; and jammed/squeezed between hard objects, 9%. The different types of road users among the vehicle-related accidents in the research population were: Bicyclists Drivers/passengers Motorcyclists Pedestrians Drivers/passengers mobiles Others
in cars
on snow-
46% 28% 7% 4% 4% 10%
The persons with sports related injuries had been injured in the following activities: Team sports (55%) football (soccer) handball, basketball, ball ice hockey indoor bandy other team sports
and volley-
24% 11% 8% 6% 6%
Individual sports (48%) down hill skiing gymnastics equestrian cross-country running badminton other ind. sports
11% 7% 7% 3% 3% 14%
sports
129 51 II
66 95 47
96 125 5
I91
208
226
METHOD All injured persons were interviewed about the origin of their injuries at the initial visit to the emergency department of the hospital. Information on
Persistent
medical
problems
sick leave from the local branch of the National Social Insurance Office was added to all relevant cases. A survey was distributed by mail to all persons in the sample about five years after the accident. Persons not responding were reminded by mail and t’zlephone until contact was established with all who t-easonably could be reached. Medical records and i~lsurance records were analysed. Complementary telephone interviews were conducted with all persons who had indicated persistent medical problems in the survey, in order to specify the nature of the problems. Clinical examinations were undertaken in cases where permanent medical impairment was unassessed, but could be suspected from the interview. Impairment was assessed according to the guidelines presently used by Swedish insurance companies. In cases where compensation for permanent disability had been awarded, information was procured from the relevant insurer. The term medical impairment implies an alteration of health status assessed by medical means, whereas the term disability denotes an alteration of a person’s capacity to meet personal, social, or occupational demands assessed by nonmedical mean’s (AMA 1984). In this study,persistent medical prohkms, are self-reported medical problems not always possible to describe as impairing, according to present guidelines. Statistical methods were used to control whether the samples were representative of the subpopulations from which they were drawn. The scale level of the variables and the number of observations in each cell decided what tests were used; z test, x2 test, or Kolmogorov-Smirnov’s test (Eilertsson 1984; Henriksson 1990). The estimate of total p-vallies of each variable in the samples was based
and permanent
43
impairments
on compounded p-values and their variances (MI :ndelhall, Ott, and Schaefer 1971). Estimates on true values of relevant variables in the subpopulatic’ns are given as percentages and integers. Representativity was checked in the sample of work injuries for the four most common injl,ry events mentioned above, in the sample of vehi :le associated injuries for car occupants, bicyclists, IInd others, and in the sample of sports-related injuries for team and individual sports. P < 0.05 was u*‘ed as the statistically significant probability level. 1 he studied stratified samples were in all these aspei?s representative of their subpopulations. Work-related injuries were defined as injuries contracted during working hours. Two persons in the sample were injured in vehicle accidents dur ng working hours (MAIS = 2), and one person v as injured during sports activities while at w( rk (MAIS = 2). These cases have been allocated to vehicle and sports samples, respectively. Vehicle-related injuries were defined as injure i’s contracted in association with moving vehicles, neluding not only those injured in traffic accidents m public roads, but also moto-cross riders, sno’vmobile riders, and riders of other vehicles “,bff road”. Sports-related injuries were defined as injur 2s contracted during some form of systematic spot ts activity, including equestrian activity.
RESULT Sex and age The proportion of females was lowest among persons injured at work-12%-while the tort z-
%
60 m
Work Vehicle
j
Sports
Age -19 Fig.
20-29
30-39
1. Age distribution
40-49 in the three
50-59
60-
subpopulations
U. BJ~RNSTIG and T. J. LARSSON
44
Eye Head
Trunk,
back
Upper
extr.
Lower
extr.
0
20
Fig. 2. Injury
locations
40
in the three
60
subpopulations
The three subpopulations (Table 1) showed completely different injury patterns. The work injuries contained a large proportion of minor eye injuries and superficial injuries to hands, while the vehicle injuries had a large number of head and face injuries. The sports injuries most often affected the lower extremities in the form of sprains and fractures (Figs. 2 and 3).
sponding figures were 32% among sports-related injuries and 38% among vehicle related injuries. There were great differences between the three subpopulations regarding age distribution. More than 80% of those injured in sports were under 30 years of age, and 50% were under 20. Among the vehicle-related injuries, there was also a large proportion of young people-one in three was under 20-but this subpopulation also contained a large proportion of persons aged 60 or more (Fig. 1). Severity
%
U
I
Sick leave
The proportion of persons reported to have been on sick leave due to the injury was about onethird in all subpopulations. Those who had suffered a vehicle-related injury had a significantly longer average period on sick leave than the others. The follow-up period for sick leave was limited to one year after the event. Among those injured in work related accidents, eight persons were listed sick for more than one year, six with MAIS = 2 injuries and two with MAIS 2 3 injuries. Among the vehiclerelated accidents the corresponding figure was nine persons, three at each severity level indicated in Table 5, and among the sports-related accidents one person, with an MAIS = 2 injury. Slightly higher
of the injuries
The work-related injuries contained a much larger proportion of minor injuries (MAIS = 1) than the other categories (Table 1). Among persons injured in vehicle- and sports-related accidents the proportion with moderate or more severe injuries (MAIS 2 2)-most often a fracture or a concussion-was one-third. Injuries associated with vehicles had the largest proportion of serious injuries (MAIS 2 3), and ten persons in this subpopulation (1%) died from their injuries. One person died in a work accident. No person was fatally injured in any sporting activity.
Wound/contusion Sprain/strain Fracture/h&.
I
Concussion Foreign
body
Vehicle
I
+
$
I
0 Fig. 3. Injury
20 types
in the three
I
40 subpopulations.
/
60
%
45
Persistent medical problems and permanent impairments
Table 4. Proportion listed sick (%) and average length of sick listing period for the three subpopulations, among persons 15-65 years of age
Table 6. Proportion listed sick (%) over and under 40 years of, ge, among persons 15-65 years of age, for the three subpopulatl 3ns -
(n = 1492)
Vehicle (n = 794)
sport (n = l(177) -
33% 45%
21% 37%
20% 42%
Work Work (n = 1492)
Vehicle (n = 794)
Sports (n = 1077)
37%
35%
30%
34
55
31
Proportion listeli sick Averag:e time (days)
<40 years of age >40 years of age
propcrtions of younger and older persons among those injured at work were listed sick, compared with the other two subpopulations. The vehicle and sported populations, however, probably contain a number of people outside the workforce and thus without sick benefits (Tables 4, 5, 6, 7). Persistent the elent
medical
problems
jive years
after
About the same proportion of persons in the three subpopulations reported some form of persistent medical problem five years after the event. This was tlue in spite of large variations in sex and age distrib’ution, injury localization and type, and injury severity between the subpopulations. Among workand vehicle-related injuries, 21% reported persistent medic,21 problems, and among sports injuries 24%. Tape of medical problem. The most common problem was pain. One in five persons injured in a sport activity and one in ten among those injured at work, reported some type of pain five years after the accident. In Tables 8 and 9 are shown the problems most commonly reported. Those interviewed were allowed to name up to three different complaints. Permanent
medical
impairment
Among those reporting persistent medical problems, all cases where permanent medical impairment could be suspected were called in for clinical examination. The number of persons with a permanent medical impairment was six (+2)% among the work-
Table 5. Proportion listed sick (%) and median sick listing period (m) among persons 15-65 years of age, by MAIS for the three subpopulations --
MAIS = 1 MAIS = 2 MAIS = 3
Age
Work (n = 1492)
Vehicle (n = 794)
Sports (n = 1077)
%
m
%
m
%
m
30 76 91
12 42 121
17 39 36
5 35 45
17 32 40
11 32 130
-
related injuries, six (+-3)% among the vehicle-relal ed injuries, and one (* 1%) among the sports-relal ed injuries. One percent of the vehicle-related injurres were fatal. Among the persons with permanent rn’,dical impairment, the number of minor injurles (MAIS = 1) was highest among the work-relay :d injuries-l/3-and only l/6 among the vehicle- ‘elated injuries. There were no cases of permanc nt impairment after a minor injury among the spor ~srelated injuries. However, cases of medical impLlirment among persons who have not reported ally medical problems remain unknown and unassesst d. Reports
to insurance
Persons injured at work had the largest propl’rtion of injuries reported to an Insurance or Compcmnsation scheme. Among the persons with a permanc’nt impairment, the reporting frequency was highc,st among those injured at work or in vehicle-relatl.:d accidents. Only a small number of sports relat;:d injuries had been reported (Table 10). Change
of work
Among persons with permanent medical impal rment, one in five who had been injured in a workor vehicle-related accident had changed jobs or WOIk tasks because of the injury. Sports injury cases II:ported a considerably lower rate of job change. C: Iculations were based on living persons (1990) agcmd 15 to 65 at the time of injury (Table 11). Change
in leisure-time
activities
More than one in three persons with a perm ~Inent medical impairment after a work- or vehicl;.‘related accident reported that they had changed the1 r Table 7. Proportion listed sick (%) and average sick listing perild (.?) among persons 15-65 years of age, by injury location for ttle three subpopulations Work (n = 1492)
Vehicle (n = 794)
Injury location
%
x
%
x
%
i
Head, Trunk, Upper Lower Eyes
8 8 40 22 22
23 38 33 41 17
28 23 23 25 1
44 69 55 56 3
12 5 26 54 2
(1 1‘ 3t1 3.’ .’
face back extr. extr.
sports (n = 107,‘) ___.
U. BJ~RNSTIGand T. .I. LARSON
46
Table 8. Proportion of persons with different persistent medical problems in the three subpopulations (95% confidence interval) Work (n = 191)
Problem
Vehicle (n = 208)
Table 10. Proportion of persons who had reported the injury to insurance or compensation scheme in the three subpopulations (95% confidence interval)
Sports (n = 226) Group
Pain Loss of strength or sensibility, sense of cold Functional defect to joint Anatomic defect Headache
10 (?4)% 6 (?3)%
1.5 (?S)% 3 (t-3)%
20 (+-S)% 2 (-+l)%
5 (?2)%
2 (&2)%
6 (-+3)%
2 (?2)% -
1 (+l)%
-
4(C3)%
-
leisure-time activities to a considerable degree. The corresponding figure among sports-related injuries of course was much higher-persistent medical problems after a sports accident often renders the victim unable to continue that particular activity (Table 12). Early retirement Early retirement or Iong-term sick leave (full or part time) was most frequent among the work-related permanent medical impairments. The proportion was almost as large among the vehicle-related injuries, while the corresponding figure among the sports-related injuries was very small (Table 13).
DISCUSSION Of the 11 fatahties in the study, ten were associated with vehicle accidents and one with work. This illustrates where the fatality risks are the highest. Among the work- and vehicle-related injuries, one person among 17 surviving suffered a permanent medical impairment, while the corresponding figure among sports injuries was considerably lower. Low average age and, therefore, good rehabilitation capacity among the sports-injury cases might explain the difference. Comparable results have not been found in the scientific literature.
Table 9. Proportion (%f of persons (aged 15-65) with a MAIS 2 2 injury and average sick listing period (f), among persons with medical impairment, persistent problems and among recovered in the three subpopulations Work (n = 1492)
Impaired Persistent problems Recovered
Vehicle (n = 794)
sports (n = 1077)
%
x
%
.?
%
x
67 38
109 24
85 50
68 9
100 60
34 31
8
9
34
9
26
7
All injured Persons with persistent medical problems Persons with permanent medical impairment
Work (n = 191)
Vehicle (n = 208)
Sports (n = 226)
12 (?4)% 57%
8 (t4)% 39%
3 (+-I)% 12%
89%
88%
33%,*
* In this table four persons with MAIS = 2 injuries and persistent medical problems, not yet at a medically stationary level, have been excluded. If these four are classified as permanent disabilities > 0 the figure will be 46%. Percentages given without confidence intervals above represent only the samples interviewed.
Sports injuries could imply long-term problems, however. In a popuIation of predominantly young people, with a rather high proportion of persistent medical problems after injury-often in the form of pain-these injuries could constitute problems over long periods in life. The medical problems would presumably interact with future exposures in working life, with unknown effects. In spite of the fact that the proportion of medically significant (MAIS 2 2) injuries was almost three times larger among the vehicle- and sportsrelated injuries compared to the work related injuries, one person in five in all three injury groups reported some form of persistent medical problem five years after the injury. The present study indicates that nine out of ten permanently impairing work- and vehicle-related injuries were reported to an insurance or compensation scheme, while the proportion of thus reported sports injuries was considerably lower. For the work injury cases, this is in line with earlier findings (Larsson and Bjornstig 1993). The inclination to report
Table 11. Proportion of persons who had changed jobs or tasks because of the injury in the three subpopulations (95% confidence interval) Group All injured Persons with persistent medical problems Persons with permanent medical impairment
Work (n = 191)
Vehicle (n = 208)
sports (n = 226)
2 (?2)% 9%
3 (231% 11%’
I (?2)%
20%
18%
-
Percentages without confidence only the samples interviewed.
intervals
5%
given represent
Persistent
medical
problems
Table 12. Proportion of persons who reported major change of leisur: time activity because of the injury in the three subpopulations (95% confidence interval) Work (n = 191)
Group All in, ured Perso IS with persistent medical pros,lems Perso’x with permanent medical impair ment Pxcentages only tile samples
Vehicle (n = 208)
Sports (n = 226)
3 (?2)% 14%
4 (?2)% 19%
6 (?2)% 52%
40%
37%
76%
given without interviewed.
confidence
intervals
represent
the injury to an insurance or compensation scheme of ccurse varies with the quality of the insurance covei-age for the population in question and the inclination to file a claim for compensation on the part of thr: individual. Our results indicate, however, that injur!/ data collected for insurance purposes can yield relevant information about permanently impairing work- and vehicle-related injuries, but is of less value in evaluating permanently impairing sports injuries and injuries that are not followed by permanent medical impairment. Early retirement and long-term sick leave was the consequence for one person in five suffering a perm,&nently impairing work-related injury and one person in seven suffering a permanently impairing vehicle-related injury. This indicates considerable social costs (Jonsson 1973). The aetiology of permanentl;i impairing injuries is probably different to the aetiology of fatal and nonimpairing injuries (Bj(irnstig et -11. 1984; Erikssonand Bjiirnstig 1982; Bjiirnstig et al. 1985; Bjiirnstig and Bylund 1986; Larsson 1990) In view of the present results, the aetiology of permanently impairing injuries stands out as an area in need of further scrutiny. I,? the stratified samples of this study, the medi-
Table 13. Proportion of persons (aged 15-65) in early retirement or still five years after the injury, on (full- or part-time) sick leave (95% confidence interval) Work (n = 191)
Group All injured Person!, with persistent medical problems Person! with permanent medical impairment Percentages only the samples
1 (?l)% 6%
22%
given without interviewed.
Vehicle (n = 208) 1 (?l)% 4%
0.3 (?0.6)% 2%
-
14%
confidence
sports (n = 226)
intervals
represent
and permanent
47
impairments
tally serious and more severe injuries were \I/ell represented. That minor injuries can cause pervnanent medical impairment and disability has bi:en shown elsewhere (Nygren et al. 1986; Larsson ;tnd Bjijrnstig 1993). Some types of medically less se\ :re injuries, thus, cannot be disregarded in the stud:, of permanent medical impairment and disability. The present Swedish classification guide to llermanent medical impairment (Personskadekomr,littCn inom F6rsZkringsbranschens Serviceaktiebc! lag 1989) seems to focus quite disproportionally on c Iassifications of injuries to hand and arm (90%) .lnd mainly describes injuries with some measurable a Inatomical defect. Pain was the most common per’ istent medical problem in this study. Pain, togel ler with serious brain and spinal injuries, is ratller poorly defined in the classification presently used by Swedish insurers to determine degree of permanl.:nt medical impairment. Persons suffering persistl.:nt pain after injury are thus disadvantaged by the l;lck of well-defined evaluation criteria. Improved ;I nd clarified criteria would be valuable. A more detai ed application, similar to the Guide to the Evaluat on of Permanent Impairment (AMA 1984), is recc’nmended. Acknowledgement-The
study was financed by the Swet ish Work Related No-Fault Liability Insurance (TFA) through agl mt to IPSO Sweden. We thank Anna Maria Backlund and Per C ov Bylund warmly for their hard and scrupulous assistance and ‘IVidar Henriksson for statistical advice.
REFERENCES Allan, D. G.; Reid, D. C.; Saboe, L. Off-road recreation lal motor vehicle accidents: Hospitalizations and deat IIS. Can. J. Surg. 31:233-236;
1988.
AMA (American Medical Association). Guide to the et IIuation of permanent impairment. 2nd edition. Chica o; American Medical Association; 1984. Bjarnstig, U.; Eriksson, A.; Mellbring G. Snowmobile ng injuries. Mechanisms and consequences. Acta C~Iir. &and. 150:619-624; 1984. Bjiirnstig, I-J.; Bylund, P.-O.; Lekander, T.; Brorsson, B. Motorcycle fatalities in Sweden. Acta Chir. Scarld. 151577-581; 1985. Bjiirnstig, U.: Bylund, P.-O. Motorcycle injuries in nor1 hern Sweden. Travel Medicine International 4: 16- ‘1: 1986. Clay, W.; van Kampen, L. T. B.; Hogerzeil, H. H. IV. Injury and disability effects of motor car acciden’s. Int. Disabil. Stud. 9:145-148; 1987. Committee on Injury Scaling. The abbreviated injl ~1 scale. 1985. Revision. Morton Grove, IL: Americ tn Association for Automotive Medicine; 1985. Eilertsson, G. Grundlgggande statistik med till%mpning ar inom sjukvlrden. Lund, Sweden: Studentlitteratllr; 1984. Eriksson, A.; Bjiirnstig, U. Fatal snowmobile acciderlts in northern Sweden. J. Trauma 22:977-982; 1982.
48
U. BJ~RNSTIG and T. J. LARSSON
Henriksson, W. Kompendium i statistik. Icke-parametriska metoder. UmeH, Sweden: Pedagogiska institutionen, Umel Universitet; 1990. Holman, R. G.; Olszewski, A.; Maier, R. V. The epidemiology of logging injuries in the Northwest. J. Trauma 27:1044-1050; 1987. Jansson, B. R.; Jacobsson, B. S. Medical consequences of work-related accidents on 2,454 Swedish farms. Stand. J. Work Environ. Health 14:21-26; 1988. Jonsson, E. Samhallskostnader for trafikolyckor, yrkesskador och sjukdomar orsakade av tobaksrok. Stockholm: Liber; 1973. Lanese, R. R.; Strauss, R. H.; Leitzman, D. J.; Rotondi, A. M. Injury and disability in matched men’s and women’s intercollegiate sports. Am. J. Public Health 80: 1459-1462; 1990. Larsson, T. J.: Accident information and priorities for prevention. Thesis. Stockholm: Royal Institute of Technology, IPSO Factum 21; 1990. Larsson, T. J.; Bjiirnstig, U. The epidemiology of occupational accidents: Three ways to measure the problem in a Swedish municipality. J Occup. Health Safety-Aust NZ; 6( 1):39-52; 1990. Larsson, T. J.; Bjornstig, U. Persistent medical problems and permanent impairment five years after occupational injury. Accepted: Stand. J. Sot. Med. 1993. Lings, S.; Jensen, J.; Christensen, S.; Mgller, J. T. The consequences to the injured of occupational accidents. Stand. J. Sot. Med. 12:26-29; 1984. MacKenzie, E. J.; Shapiro, S.; Smith, R. T.; Siegel, J. H.; Moody, M.; Pitt, A. Factors influencing return
to work following hospitalization for traumatic injury. Am. J. Public Health 77:329-334; 1987. MacKenzie, E. J.; Siegel, J. H.; Shapiro, S.; Moody, M.; Smith, R. T. Functional recovery and medical costs of trauma: An analysis by type and severity of injury. J. Trauma 28:281-297; 1988. Maimaris, C.; Barnes, M. R.; Allen, M. J. “Whip lash injuries” of the neck: A retro-spective study. Injury 19:393-396; 1988. Mendelhall, W.; Ott, L.; Schaeffer, R. L. Elementary survey sampling. Belmont, CA: Wadsworth Publishing Comp Inc., 1971. Nakayama, D. K.; Gardner, M. J.; Rogers, K. D. Disability from bicycle-related injuries in children. J. Trauma 30:1392-1394; 1990. Nygren, A; Hansson, P. G.; Tingvall, C. Acute injury scaling related to residual disability. Acta Neurochir. Suppl. 36:25-27; 1986. Personskadekommitten inom Forsakringsbranschens Serviceaktiebolag. Grunder for gradering av kvarstaende men efter skador (medicinsk invaliditet). Stockholm: 1989. States, J. D.; Viano, D. C. Injury impairment and disability scales to assess the permanent consequences of trauma. Accid. Anal. Prev. 22:151-160; 1990. Waller, J. A.; Payne, S. R.: Skelly, J. M. Injuries to carpenters. J. Occup. Med. 31:687-692; 1989. Yates, D. W.; Heath, D. F.; Mars, E.; Taylor, R. J. A system for measuring the severity of temporary and permanent disability after injury. Accid. Anal. Prev. 23:323-329; 1991.