Increased Mortality Risk for Workers With a Compensated, Permanent Occupational Disability of the Upper or Lower Extremities: A 21-Year Follow-up Study

(1)

March 17, 2010

Original Contribution

Increased Mortality Risk for Workers With a Compensated, Permanent

Occupational Disability of the Upper or Lower Extremities: A 21-Year Follow-up

Study

Sheng-Hsuan Lin, Hsin-Yi Lee, Yu-Yin Chang, Yuh Jang, Pau-Chung Chen, and Jung-Der Wang*

*Correspondence to Prof. Jung-Der Wang, Institute of Occupational Medicine and Industrial Hygiene, College of Public Health, National Taiwan University, Room 719, No. 17, Xu-Zhou Road, Taipei City 100, Taiwan (e-mail: jdwang@ntu.edu.tw).

Initially submitted May 4, 2009; accepted for publication January 5, 2010.

This 1986–2006 study sought to determine whether specific causes led to increased mortality risks for Taiwa-nese workers with an approved compensation claim for permanent occupational disability (ACCPOD) of the upper or lower extremities. All cases of ACCPOD between 1986 and 2006 were collected from the database of com-pensation claims at the Bureau of Labor Insurance. Standardized mortality ratios and 95% confidence intervals were calculated for different causes of death among workers with an ACCPOD of the upper or lower extremities. A total of 800,047 person-years were accrued for 71,001 workers with a single type of disability. Standardized mortality ratios were significantly increased for all causes, including liver cirrhosis, injuries, and intentional self-harm. Standardized mortality ratios for workers with amputations of the lower extremities increased to 7.66 (95% confidence interval (CI): 5.36, 10.61), 2.40 (95% CI: 1.44, 3.75), 2.07 (95% CI: 1.03, 3.70), and 5.09 (95% CI: 2.20, 10.03) for those with diabetes mellitus, cerebrovascular disease, liver cirrhosis, and chronic renal failure, respec-tively. The authors concluded that workers with occupational disabilities involving an upper or lower extremity should be assisted to prevent further injuries or intentional self-harm, whereas those with lower limb amputations should be provided care related to proactive control of diabetes mellitus and associated complications during rehabilitation.

amputees; diabetes mellitus; insurance, disability; lower extremity; mobility limitation; occupations; upper extremity

Abbreviation: ICD-9-CM, International Classiﬁcation of Diseases, Ninth Revision, Clinical Modiﬁcation.

Occupational injuries impose both health-related and financial burdens on individuals and society as a whole. Because of impaired mobility in daily life, workers with permanent impairment of a lower extremity are more likely to have worse survival rates than those with an impairment of an upper extremity (1, 2). Although amputees returning from war have been found to have higher mortality rates than the general population from poisoning, intentional self-harm, alcoholic liver cirrhosis, and acute pancreatitis (3), the detailed causes of premature deaths among workers with an occupational disability of the lower extremities have yet to be determined.

Many studies have found an increase in morbidity and mortality from cardiovascular disease among veterans with posttraumatic lower extremity amputations (3–7). However,

few reports have focused on occupational injuries incurred by workers. Therefore, the purpose of this study was to determine whether there were increased mortality risks for workers with an approved compensation claim for permanent occupational disability of the upper or lower extremities.

MATERIALS AND METHODS Study population

In Taiwan, the Bureau of Labor Insurance has adminis-tered the compensation scheme of workers since 1950. A retrospective study was carried out based on the computer-ized registry established by the Bureau of Labor Insurance, which included all compensated cases of permanent

at National Taiwan University Library on September 2, 2010

http://aje.oxfordjournals.org

(2)

disability from 1986 to 2006. According to statements re-leased by the Bureau of Labor Insurance in Taiwan, the term ‘‘approved compensation claim for permanent occupational disability’’ indicated that the disabilities of affected workers were caused by occupational injuries or diseases and that workers were unable to recover under medical care for at least one year, regardless of whether they were able to return to work.

In total, the computerized registry consisted of 106,437 compensation claims for permanent occupational disability resulting from work-related injuries with details of workers’ gender, age at date of injury, types of disability and accident, body part affected, and monthly insured salary. Initially, 4,745 cases were excluded because of incomplete informa-tion related to any one of the following items: sex, age at date of injury, and body part affected. Because of the occur-rence of multiple injuries at different sites and difficulty in determining etiologic factors, another 18,605 cases with more than 2 types of disability were also excluded to simplify the analysis. Among these cases, 12,237 were com-pensated in a single application, and 6,368 suffered from recurrent injuries contributed by 3,049 subjects. Cases for whom there were disparities regarding injury type and

com-pensation (n¼ 179) that might imply incomplete

informa-tion or double registry were deleted as well. There were also 11,907 cases involving injury sites other than the upper or lower extremities that were not included. As summarized in Figure 1, 71,001 cases remained in the final analysis.

Among those workers included in this study, 8,017 suf-fered from injuries of their lower extremities and 62,984 of their upper extremities. These cases were further organized

into those with injuries of the toes only (n¼ 2,683), the leg

or foot only (n¼ 5,334), the fingers only (n¼ 59,545), and

the hand or arm only (n¼ 3,439). The survival status of each

case was followed through the end of December 2006 and was verified with the database of the National Mortality Registry in Taiwan. This registry, which was established and has been maintained by the Department of Health, Executive Yuan since 1950, is the most reliable official database for certificates of death in Taiwan. It has been computerized since 1981. In this study, survival status of those with compensated cases from the Bureau of Labor Insurance registry was verified by using the personal iden-tification number as the matched variable to cross-link the mortality registry. If a worker was found to have a record from the mortality registry under the same identification number, then the date and cause of death could be accurately assured. If this verification was not possible, the person was considered to be alive or censored at least prior to December 31, 2006. Foreign workers were not included in the study because they were not given identification numbers and hence their survival status could not be verified.

Standardized mortality ratios and Life Table Analysis System

The Life Table Analysis System, developed by the US National Institute for Occupational Safety and Health, was created to analyze cohorts defined by occupational expo-sures and may be downloaded from the following website:

http://www.cdc.gov/niosh/LTAS/. The Life Table Analysis System constructs rates according to age, sex, race, calendar time, and duration or level of exposure and then compares the observed rates for the cohort being studied with rates from a reference population, which, for this study, was the general population of Taiwan. Rates are calculated as ob-served events—typically cause of death—divided by person-time at risk in the interval. Therefore, this process requires the input of a work history file in addition to a de-mographic and outcome file. We used the Life Table Anal-ysis System to calculate the standardized mortality ratios of the population with upper or lower extremity injuries.

The starting point of follow-up was January 1, 1986. Person-years at risk were organized by gender, age, and calendar time (in 5-year intervals). The causes of death in-cluded cancer (International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes 140–208), benign tumors (ICD-9-CM codes 224–225, 210–223, 226–239, 237.0–237.4, 239.0–239.5, 239.8– 239.9), alcoholism (ICD-9-CM codes 305, 291, 303), dia-betes mellitus CM code 250), heart disease (ICD-9-CM codes 390–398, 402, 404, 410–414, 420–429), cerebro-vascular diseases (ICD-9-CM codes 430–438), respiratory diseases (ICD-9-CM codes 460–519), liver cirrhosis and

Incomplete data (n = 4,745) or

disparity in data (n = 179) Total number of subjects in the

database (N = 106,437) Toe(s) only (n = 2,683) Foot or leg (n = 5,334) Finger(s) only (n = 59,545) Hand or arm (n = 3,439) More than 2 types of disabilities at one time (n = 12,237) or resulted from recurrence of injuries

(n = 6,368)

Subjects included in the final analysis (N = 71,001)

Disability of lower extremity

(n = 8,017) Disability of upper extremity (n = 62,984)

Cases with injury sites other than upper and lower extremities (n = 11,907)

Figure 1. Flowchart of the selection procedure for the population included in the study of risk of mortality for Taiwanese workers with an occupational disability of the upper or lower extremities using data derived from the Bureau of Labor Insurance database of workers’ compensation, 1986–2006.

(3)

other chronic liver disease (ICD-9-CM code 571), chronic nephritis and renal failure (ICD-9-CM codes 582–583, 585– 587), transportation injuries (ICD-9-CM codes E800–849), falls (ICD-9-CM codes E880–888), intentional self-harm CM codes E950–959), and other injuries (ICD-9-CM codes E850–879, E890–949, E980–999).

The reference rates of standardized mortality ratios used in the analysis were derived from the Taiwanese mortality rates for 119 underlying causes of death from 1986 until 2006. Ninety-five percent confidence intervals of the re-ported standardized mortality ratios were computed by as-suming the Poisson distribution model.

RESULTS

A total of 800,047 person-years were accrued for all workers; the median years of follow-up and total cumulative person-years at risk, together with the frequency distribution of demographic characteristics, are summarized for 4 different groups in Table 1. Subjects with disabilities of the leg or foot appeared to have the shortest median follow-up period of 5.9 years because 56.2% of their injuries occurred during 2001–2006. The mean age of each group was between 36.6 years and 44.7 years. Almost half of the occupational injuries among workers with a hand or arm injury or a foot or leg injury occurred after December 31, 2000. Furthermore, about three-fourths of the workers with finger injuries were trapped or caught in machinery, whereas 37.7% of the foot or leg injuries were due to transportation accidents. The adjusted monthly insured salary among each group was $23,211–$26,935. This salary was comparable to the average salary of the employed population of Taiwan (27,404 Taiwanese new dollars), and this information may be downloaded from the following website: http://www.bli. gov.tw/en/. Workers with finger disabilities seemed to have a lower average adjusted monthly insurance salary, whereas that for the other 3 groups seemed much more similar.

Compared with the general population, workers with all types of occupational disabilities of the upper or lower extremities showed increased standardized mortality ratios for all causes, as well as for injuries and intentional self-harm (Table 2). Specifically, workers with toe injuries had approx-imately a 3 times higher risk of mortality caused by falls, whereas workers with amputations of a lower limb had 4.5 and 2.7 times higher standardized mortality ratios for inten-tional self-harm and other injuries, respectively (Table 3).

In addition, workers with an occupational disability of the foot or leg, when coupled with diabetes mellitus, respiratory diseases, chronic nephritis or renal failure, intentional self-harm, falls or other injuries, cerebrovascular disease, liver cirrhosis, or cancer, had a significantly higher mortality rate (Table 2). For workers with toe disabilities, standardized mortality ratios were increased for all causes, diabetes mel-litus, liver cirrhosis, and falls, whereas workers with hand or arm disabilities primarily had more significant increases in standardized mortality ratios for liver cirrhosis and inten-tional self-harm. Furthermore, there was a significant rise in mortality due to alcoholism, intentional self-harm, and liver cirrhosis among workers with approved compensation claims for permanent finger disabilities.

The population of workers with foot or leg disabilities was organized into nonamputated versus amputated, and the latter category was further divided into those with below-the-knee and above-the-knee amputations (Table 3). For workers both with and without amputations, the standardized mortality ra-tios for all causes were significantly higher than those for the general population. The standardized mortality ratios for workers with amputations increased significantly when they also suffered from diabetes mellitus, cerebrovascular diseases, liver cirrhosis, chronic nephritis or renal failure, other injuries, or intentional self-harm. On the other hand, the standardized mortality ratios for those who had cancers, benign neoplasms, diseases of the respiratory system, and injuries increased sig-nificantly among workers without amputations. Furthermore, 7 of the 46 workers who died from respiratory diseases suf-fered from silicosis or pneumoconiosis.

DISCUSSION

To the best of our knowledge, this study was the first long-term follow-up to investigate the causes of death for workers suffering from permanent disabilities of the upper or lower extremities as a result of occupational injuries. Because the nationwide database was used, the sample size for the

pres-ent study was considered substantial (N¼ 71,001), and, in

contrast to most of the previous studies conducted among the veteran population, all cases were certified as occupa-tional. Overall, because a general tendency toward an in-crease in mortality risks for all types of injuries and intentional self-harm among workers with different types of occupational disabilities of the upper or lower extremities was found, more attention must be focused on psychosocial care and safety issues throughout the process of rehabilita-tion (8, 9).

In this study, we found an increase in the mortality risk from diabetes mellitus for workers with lower extremity injuries (Tables 2 and 3), but this finding has rarely been reported in any previous studies. Hrubec and Ryder (3) re-ported a slight increase in the mortality risk from diabetes mellitus for veterans with distal amputations, but that study could not make any strong inference because of the limited number of subjects. One potential explanation might be that workers with diabetes mellitus were more likely to receive amputations after sustaining an injury to the lower extremity because of poor circulation. However, since the average age of this group was 39.5–44.7 years (Table 1), most of them were probably too young to have developed significant complications from diabetes mellitus–related peripheral neuropathy or angiopathy (10). Furthermore, diabetes mellitus–related amputations usually occur for the distal parts of lower extremities, whereas our study found that 6 of 36 deceased cases, certified as dying from diabetes mel-litus, were above-the-knee amputees. Thus, the likelihood of diabetes mellitus causing a lower extremity amputation can explain, at most, a minor portion of the 7.7-fold increase in risk (Table 3).

Another possible explanation might be that workers of lower socioeconomic status and living deleterious lifestyles were more likely to die from diabetes mellitus, cardiovas-cular disease, respiratory diseases, or liver cirrhosis.

(4)

However, the insured monthly salaries of workers with foot or leg amputations were slightly higher than for those in the other 3 groups (Table 1). After adjustment for gender and age, insured salaries of workers with finger disabilities were about 2,000 Taiwanese new dollars (1 US dollar was equiv-alent to 32.5 Taiwanese new dollars in 2006) lower than for

the other 3 groups. Above all, the insured salaries among the studied groups were comparable with the national average for the employed population in 2006. Because people with similar incomes usually have similar health behaviors (11–13), socioeconomic differences do not seem to explain the major mortality differences between the 4 subcohorts

Table 1. Frequency Distribution of Taiwanese Subjects by Demographic Factors and Injury Types, 1986–2006

Disability of Upper Extremities Disability of Lower Extremities Finger(s) Only Hand or Arm Toe(s) Only Foot or Leg No. of cases 59,545 3,439 2,683 5,334 No. of cumulative person-years at risk 701,417 30,437 28,280 39,914 Median follow-up, years 12.5 7.2 10.5 5.9 Deceased cases, % 6.8 8.4 9.2 10.6 Male gender, % 76.3 76.6 86.1 79.9 Mean (SD) age at injury, years 36.6 (12.1) 42.2 (12.1) 39.5 (12.3) 44.7 (11.2) Calendar period of injury, % 1986–1990 27.7 20.5 21.1 16.2 1991–1995 27.1 15.6 25.9 9.9 1996–2000 22.0 16.5 24.3 17.7 2001–2006 23.2 47.4 28.7 56.2 Insured salary (NT$a) at the time of injury, % 20,000 52.1 35.3 36.7 29.0 20,001–30,000 29.2 35.3 35.0 37.0 30,001–40,000 10.1 17.4 14.8 18.9 40,001–50,000 7.8 11.9 12.3 15.0 >50,000 0.8 0.1 1.2 0.1 Average insured salary (SD) 23,211 (8,936) 25,816 (9,303) 25,858 (9,897) 26,935 (9,603) Injury type, % Fall to a lower level 0.2 9.4 1.5 12.8 Slip/trip/stumble 0.2 9.0 0.8 9.6 Struck by a sliding object or knocked down 1.5 9.7 18.8 18.6 Trapped/caught in machinery 74.9 34.5 47.7 10.0 Cuts, lacerations, punctures 21.1 8.5 14.7 4.3 Transportation incidents 0.7 22.0 10.5 37.7 Not elsewhere classiﬁed 1.4 7.0 6.1 7.0

Abbreviations: NT$, Taiwanese new dollar; SD, standard deviation.

a

1 US dollar¼ 32.5 NT$ in 2006.

(5)

(Table 2). Therefore, we concluded that mortality risks from diabetes mellitus, cerebrovascular disease, respiratory diseases, and chronic renal failure are increased for workers with disabilities of the lower extremities. Future rehabilita-tion for workers with a disability of the lower extremities should be proactive to prevent diabetes mellitus and/or met-abolic syndrome and its associated diseases, paying special attention to diseases of the respiratory system for workers without amputations (Tables 2 and 3).

Although the underlying mechanism of this trend is not yet fully clarified, the inconvenience of regular exercise and hemodynamic changes caused by lower extremity disabil-ities may elevate the risks of hyperinsulinemia, obesity, and hypertension, which commonly occur in lower extremity amputees (14). Moreover, Peles et al. (15) observed that lower extremity amputees exhibited insulin resistance in tests of fasting plasma insulin levels and the oral glucose tolerance test. The increased prevalence of diabetes mellitus

may accelerate the atherosclerotic process in vessels, gen-erally leading to an increase in the incidence of strokes and chronic kidney disease (16), as well as elevated mortality from end-stage renal disease (Table 3). Furthermore, the actual amputation of a lower extremity might increase blood coagulability (5) and elevate inflammatory factors, such as interleukin-6 or C-reactive protein (17), further contributing to development of atherosclerosis and stroke.

In this study, lower extremity amputees were found to have about a 4-fold greater risk of intentional self-harm, and it corroborated previous reports suggesting that people with severe lower extremity injuries and/or amputations usually suffer considerable psychological distress often leading to alcohol addiction and substance abuse (16, 18, 19). Such a tendency may explain the increase in mortality from liver cirrhosis, other injuries, and intentional self-harm in this group of workers (Table 3). In contrast, workers with finger disabilities tended to die from transportation injuries,

Table 2. Standardized Mortality Ratios for Major Causes of Death Among Taiwanese Workers With an Approved Compensation Claim for Permanent Occupational Disability of the Upper or Lower Extremities, 1986–2006

Disability of Upper Extremities Disability of Lower Extremities Finger(s)

Only Hand or Arm Toe(s) Only Foot or Leg

Observed No. SMR 95% CI Observed No. SMR 95% CI Observed No. SMR 95% CI Observed No. SMR 95% CI No. of cases 59,545 3,439 2,683 5,334 All causes 4,047 1.13 1.09, 1.16** 288 1.25 1.11, 1.41** 247 1.33 1.17, 1.51** 567 1.57 1.44, 1.70** All cancers 1,098 1.01 0.95, 1.07 73 1.02 0.80, 1.28 70 1.21 0.95, 1.53 154 1.33 1.13, 1.56** Benign and unspeciﬁed neoplasms 28 1.01 0.67, 1.46 0 0.00 0.00, 1.96 1 0.69 0.02, 3.85 9 3.02 1.38, 5.74** Diabetes mellitus 174 0.92 0.78, 1.06 18 1.30 0.77, 2.05 24 2.35 1.50, 3.49** 57 2.59 1.96, 3.36** Diseases of the heart 274 1.02 0.90, 1.15 17 0.90 0.52, 1.44 18 1.23 0.73, 1.94 38 1.24 0.88, 1.70 Cerebrovascular disease 301 0.97 0.86, 1.09 27 1.21 0.80, 1.76 14 0.82 0.45, 1.38 56 1.54 1.16, 2.00** Diseases of the respiratory system 185 1.08 0.93, 1.24 17 1.31 0.76, 2.10 15 1.58 0.88, 2.61 53 2.46 1.84, 3.22** Alcoholism 13 2.30 1.23, 3.94* 0 0.00 0.00, 13.53 0 0.00 0.00, 13.51 0 0.00 0.00, 9.72

Liver cirrhosis and other chronic liver disease 300 1.20 1.07, 1.35** 25 1.74 1.13, 2.57* 16 1.24 0.71, 2.01 36 1.63 1.14, 2.26** Diseases of the musculoskeletal system and connective tissue 27 2.01 1.32, 2.92** 0 0.00 0.00, 4.28 0 0.00 0.00, 5.44 1 0.74 0.02, 4.12 Chronic nephritis and renal failure

59 0.91 0.69, 1.18 7 1.49 0.60, 3.07 8 2.35 1.01, 4.62* 18 2.42 1.44, 3.83** Transportation injuries 439 1.64 1.49, 1.80** 19 1.45 0.87, 2.26 15 1.24 0.70, 2.05 17 0.93 0.54, 1.49 Falls 87 1.57 1.26, 1.94** 5 1.67 0.54, 3.91 9 3.29 1.51, 6.25** 7 1.57 0.63, 3.24 Other injuries 288 1.54 1.37, 1.73** 14 1.57 0.86, 2.63 14 1.66 0.91, 2.79 25 2.03 1.31, 2.99** Intentional self-harm 179 1.40 1.20, 1.62** 13 2.02 1.08, 3.46* 11 1.88 0.94, 3.37 23 2.46 1.56, 3.69**

Abbreviations: CI, conﬁdence interval; SMR, standardized mortality ratio. *P < 0.05; **P < 0.01.

(6)

alcoholism, and diseases of the musculoskeletal system (Table 2). This trend might partially indicate the disadvan-tage of driving motorcycles or scooters after injury; these motor vehicles were the most common commuting tools for workers and people in Taiwan. As stated previously, these potential problems should be dealt with in the long-term care of these workers.

A few limitations of our study should be recognized. First, prior to injury, all individuals were actively employed, healthy workers who would have a better baseline health status and reduced mortality compared with the general population in the absence of work-related disabilities (20, 21). Thus, the decrease in standardized mortality ratios for diabetes mellitus and cerebrovascular diseases (0.92 and 0.97, respectively; Table 2) among workers with finger disabilities might still indicate underestimation of the mortality risks. Second, the coding for injuries as the single underlying cause of death would generally reduce the likeli-hood of attributing death to other chronic diseases and

consequences, which further underestimates the potential contribution of diabetes mellitus, cerebrovascular and cardiovascular diseases, and cancer. Third, although we tried to show that the insured monthly salaries for the 4 groups were comparable, we were unable to control for any residual confounding factors, other than gender, age, and income, because our data did not record every detail of health behaviors for each worker. Finally, it has been reported that number of deaths attributed to diabetes melli-tus in Taiwan was higher than that in Australia and Sweden (22, 23). However, since the process of coding the underly-ing cause of death is comparable in workers and the general population of Taiwan, the possible differential bias, if pres-ent, should be minimal.

In conclusion, workers with an occupational disability of the upper or lower extremities seem to have increased mor-tality risks for more injuries and intentional self-harm later in life, indicating the need for more comprehensive psycho-logical care and safety precautions during the process of

Table 3. Standardized Mortality Ratios for Major Causes of Death Among Taiwanese Workers With and Without Amputations With an Approved Compensation Claim for Permanent Occupational Disability of the Lower Extremity, 1986–2006

Workers Without Amputations

Workers With Amputations

Total Below-Knee and

Foot Amputees Above-Knee Amputees Observed No. SMR 95% CI Observed No. SMR 95% CI Observed No. SMR 95% CI Observed No. SMR 95% CI No. of cases 4,339 995 570 425 All causes 392 1.40 1.27, 1.55** 175 2.14 1.83, 2.48** 119 2.37 1.96, 2.84** 56 1.77 1.34, 2.30** All cancers 124 1.38 1.15, 1.64** 30 1.17 0.79, 1.68 16 1.01 0.58, 1.64 14 1.44 0.79, 2.42 Digestive organs and peritoneum 70 1.51 1.17, 1.90** 13 0.98 0.52, 1.68 7 0.85 0.34, 1.75 6 1.19 0.44, 2.60 Benign and unspeciﬁed neoplasms 8 3.45 1.49, 6.79** 1 1.52 0.04, 8.49 1 2.49 0.06, 13.85 0 0.00 0.00, 14.51 Diabetes mellitus 21 1.21 0.75, 1.86 36 7.66 5.36, 10.61** 30 9.91 6.68, 14.15** 6 3.59 1.32, 7.81* Diseases of the heart 26 1.09 0.71, 1.60 12 1.78 0.92, 3.11 7 1.66 0.67, 3.42 5 1.98 0.64, 4.63 Cerebrovascular disease 37 1.30 0.92, 1.80 19 2.40 1.44, 3.75** 9 1.81 0.83, 3.43 10 3.40 1.63, 6.24** Diseases of the respiratory system 46 2.72 1.99, 3.63** 7 1.51 0.61, 3.12 4 1.35 0.37, 3.45 3 1.81 0.37, 5.30 Asthma 4 1.96 0.53, 5.01 2 3.49 0.42, 12.62 0 0.00 0.00, 10.26 2 9.40 1.14, 33.96* Liver cirrhosis and other chronic liver disease 25 1.49 0.97, 2.20 11 2.07 1.03, 3.70* 10 3.20 1.53, 5.88** 1 0.45 0.01, 2.53 Chronic nephritis and renal failure 10 1.71 0.82, 3.14 8 5.09 2.20, 10.03** 6 5.96 2.19, 12.96** 2 3.54 0.43, 12.80 Transportation injuries 13 0.96 0.51, 1.65 4 0.84 0.23, 2.15 3 1.10 0.23, 3.22 1 0.49 0.01, 2.73 Other injuries 16 1.77 1.01, 2.87* 9 2.73 1.25, 5.19* 8 4.30 1.86, 8.47** 1 0.70 0.02, 3.88 Intentional self-harm 13 1.83 0.97, 3.13 10 4.48 2.15, 8.23** 7 5.51 2.22, 11.35** 3 3.11 0.64, 9.09

Abbreviations: CI, conﬁdence interval; SMR, standardized mortality ratio. *P < 0.05; **P < 0.01.

(7)

rehabilitation. In addition, workers with lower limb ampu-tations should receive more attention with regard to the pro-active control of diabetes mellitus and associated vascular and kidney complications.

ACKNOWLEDGMENTS

Author affiliations: Department of Medicine, College of Medicine, National Taiwan University, Taipei, Taiwan (Sheng-Hsuan Lin); Institute of Occupational Medicine and Industrial Hygiene, College of Public Health, National Taiwan University, Taipei, Taiwan (Hsin-Yi Lee, Yu-Yin Chang, Pau-Chung Chen, Jung-Der Wang); School of Occupational Therapy, College of Medicine, National Taiwan University, Taipei, Taiwan (Yuh Jang); Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan (Jung-Der Wang); and Department of Envi-ronmental and Occupational Medicine, National Taiwan University Hospital, Taipei, Taiwan (Jung-Der Wang).

This study was supported by a grant from the Institute of Occupational Safety and Health of the Council of Labor Affairs, Executive Yuan of Taiwan (no. IOSH97-M101). The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the funding agency.

The authors are indebted to the Bureau of Labor Insur-ance of Taiwan for providing the database of workers’ com-pensation from 1986 to 2006.

Conflict of interest: none declared.

REFERENCES

1. Ho JJ, Hwang JS, Wang JD. Life-expectancy estimations and the determinants of survival after 15 years of follow-up for 81 249 workers with permanent occupational disabilities. Scand J Work Environ Health. 2006;32(2):91–98.

2. Hashemi L, Webster BS, Clancy EA, et al. Length of disability and cost of workers’ compensation low back pain claims. J Occup Environ Med. 1997;39(10):937–945.

3. Hrubec Z, Ryder RA. Traumatic limb amputations and subse-quent mortality from cardiovascular disease and other causes. J Chronic Dis. 1980;33(4):239–250.

4. Vollmar JF, Paes E, Pauschinger P, et al. Aortic aneurysms as late sequelae of above-knee amputation. Lancet. 1989;2(8667): 834–835.

5. Modan M, Peles E, Halkin H, et al. Increased cardiovascular disease mortality rates in traumatic lower limb amputees. Am J Cardiol. 1998;82(10):1242–1247.

6. Lorenz M, Panitz K, Grosse-Furtner C, et al. Lower-limb am-putation, prevalence of abdominal aortic aneurysm and athero-sclerotic risk factors. Br J Surg. 1994;81(6):839–840. 7. Paes E, Schmid-Scho¨nbein H. Pathogenesis of arteriosclerosis

and aneurysm development of the infrarenal aorta and pelvic circulation after unilateral leg amputation [in German]. Vasa. 1999;28(3):215.

8. Jan Y, Hwang MT, Hsieh CL, et al. Factors influencing return to work after occupational hand injuries. J Occup Ther Assoc ROC. 1996;14(2):115–125.

9. Flinn-Wagner S, Mladonicky A, Goodman G. Characteristics of workers with upper extremity injuries who make a suc-cessful transition to work. J Hand Ther. 1990;2(2):51–55. 10. Powers AC. Diabetes mellitus. In: Kasper KL, Braunwald E,

Fauci AS, et al, eds. Harrison’s Principles of Internal Medi-cine. 16th ed. New York, NY: McGraw-Hill; 2005:2152–2179. 11. Wang JD, Wegman DH, Smith TJ. Cancer risks in the optical manufacturing industry. Br J Ind Med. 1983;40(2):177–181. 12. Sterling TD, Weinkam JJ. Smoking characteristics by type of

employment. J Occup Med. 1976;18(11):743–754. 13. Covey LS, Wynder EL. Smoking habits and occupational

status. J Occup Med. 1981;23(8):537–542.

14. Rose HG, Schweitzer P, Charoenkul V, et al. Cardiovascular disease risk factors in combat veterans after traumatic leg amputations. Arch Phys Med Rehabil. 1987;68(1):20–23. 15. Peles E, Akselrod S, Goldstein DS, et al. Insulin resistance and

autonomic function in traumatic lower limb amputees. Clin Auton Res. 1995;5(5):279–288.

16. Yang HY, Wang JD, Lo TC, et al. Increased mortality risk for cancers of the kidney and other urinary organs among Chinese herbalists. J Epidemiol. 2009;19(1):17–23.

17. Ranjit N, Diez-Roux AV, Shea S, et al. Psychosocial factors and inflammation in the Multi-Ethnic Study of Atherosclero-sis. Arch Intern Med. 2007;167(2):174–181.

18. McCarthy ML, MacKenzie EJ, Edwin D, et al. Psychological distress associated with severe lower-limb injury. J Bone Joint Surg Am. 2003;85-A(9):1689–1697.

19. Gunawardena N, Senevirathne Rde A, Athauda T. Mental health outcome of unilateral lower limb amputee soldiers in two districts of Sri Lanka. Int J Soc Psychiatry. 2007;53(2): 135–147.

20. Wang JD, Miettinen OS. Occupational mortality studies. Princi-ples of validity. Scand J Work Environ Health. 1982;8(3):153–158. 21. Radon K, Goldberg M, Becklake M. Healthy worker effect in cohort studies on chronic bronchitis. Scand J Work Environ Health. 2002;28(5):328–332.

22. Lu TH, Shau WY, Shih TP, et al. Factors associated with errors in death certificate completion. A national study in Taiwan. J Clin Epidemiol. 2001;54(3):232–238.

23. Lu TH, Walker S, Johansson LA, et al. An international comparison study indicated physicians’ habits in reporting diabetes in part I of death certificate affected reported national diabetes mortality. J Clin Epidemiol. 2005;58(11): 1150–1157.