ORIGINAL ARTICLE: Clinical Endoscopy
Prevalence and risk factors for musculoskeletal injuries related to endoscopy Wiriyaporn Ridtitid, MD,1,2 Gregory A. Coté, MD, MS,1 Wesley Leung, MD,1 Ralph Buschbacher, MD,1 Sheryl Lynch, RN,1 Evan L. Fogel, MD,1 James L. Watkins, MD,1 Glen A. Lehman, MD,1 Stuart Sherman, MD,1 Lee McHenry, MD1 Indianapolis, Indiana, USA
Background: There are limited data regarding work-related injury among endoscopists. Objective: To define the prevalence of endoscopy-related musculoskeletal injuries and their impact on clinical practice and to identify physician and practice characteristics associated with their development. Design: Survey. Setting: Electronic survey of active members of the American Society for Gastrointestinal Endoscopy with registered e-mail addresses. Participants: Physicians who currently or ever performed endoscopy and responded to the survey between February 2013 and November 2013. Intervention: A 25-question, self-administered, electronic survey. Main Outcome Measurements: Prevalence, location, and ramifications of work-related injuries and endoscopist characteristics and workload parameters associated with endoscopy-related injury. Results: The survey was completed by 684 endoscopists. Of those, 362 (53%) experienced a musculoskeletal injury perceived definitely (n Z 204) or possibly (n Z 158) related to endoscopy. Factors associated with a higher rate of endoscopy-related injury included higher procedure volume (O20 cases/week; P! .001), greater number of hours per week spent performing endoscopy (O16 hours/week; P! .001), and total number of years performing endoscopy (P Z .004). The most common sites of injury were neck and/or upper back (29%) and thumb (28%). Only 55% of injured endoscopists used practice modifications in response to injuries. Specific treatments included medications (57%), steroid injection (27%), physiotherapy (45%), rest (34%), splinting (23%), and surgery (13%). Limitations: Self-reported data of endoscopy-related injury. Conclusion: Among endoscopists there is a high prevalence of injuries definitely or potentially related to endoscopy. Higher procedure volume, more time doing endoscopy per week, and cumulative years performing endoscopy are associated with more work-related injuries. (Gastrointest Endosc 2015;81:294-302.)
Several studies have suggested a high prevalence of musculoskeletal injuries among endoscopists.1-6 However, these studies were conducted when procedure
volumes and the complexity of endoscopic therapies were typically lower.7,8 Further, data from Western populations of endoscopists are limited, and a robust analysis
Abbreviations: ACGIH, American Conference of Industrial Hygienists; ASGE, American Society for Gastrointestinal Endoscopy.
Current affiliations: Indiana University School of Medicine, Indianapolis, Indiana, USA (1), Chulalongkorn University, King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand (2).
DISCLOSURE: All authors disclosed no financial relationships relevant to this article. Copyright ª 2015 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 http://dx.doi.org/10.1016/j.gie.2014.06.036 Received April 17, 2014. Accepted June 16, 2014.
294 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
Reprint requests: Lee McHenry, MD, Professor of Medicine, Indiana University School of Medicine, 550 North University Boulevard, UH 2300, Indianapolis, IN 46202. If you would like to chat with an author of this article, you may contact Dr McHenry at [email protected].
www.giejournal.org
Ridtitid et al
to identify risk factors (endoscopist and practice) associated with endoscopy-related injury is lacking. Endoscopists are presumably at risk for overuse syndromes because of the repetitive movements and potentially awkward posture associated with endoscopy. Based on earlier reports,1-3,6,9-11 a systemic review estimated that 37% to 89% of endoscopists develop work-related injuries.12 Moreover, suspected risk factors included repetitive hand motion, high hand forces, and awkward wrist, shoulder, and neck postures.12 However, previous studies of work-related injury among endoscopists have been limited by small sample sizes and limited analysis of risk factors. The present widespread use of EGD and colonoscopy implies that endoscopists may perform more procedures on a daily basis than in the past. Furthermore, the burden and performance of more technically challenging procedures such as ERCP, EUS, device-assisted enteroscopy, and others may predispose endoscopists to higher rates of repetitive stress injuries than previously reported. Therefore, our primary aim was to define the prevalence and types of endoscopy-related musculoskeletal injuries in the current era of high-volume endoscopy with advanced therapeutics. Second, we evaluated endoscopist and practice characteristics associated with these injuries.
METHODS Survey sampling We conducted an electronic survey of endoscopists who were active members of the American Society for Gastrointestinal Endoscopy (ASGE) with registered e-mail addresses (n Z 5239) between February 2013 and November 2013. Members who currently perform or ever have performed endoscopy (by self-report) were eligible to participate. Informed consent was implied by response to the survey. Before dissemination of the survey, our institutional review board approved the study.
Survey instrument The survey was a 25-question, self-administered, electronic survey that was developed by 2 endoscopists (L.M., W.L.) and 1 physiatrist (R.B.) (Supplemental Table 1, available online at www.giejournal.org). The survey instrument was a modified version of a previous survey6 conducted by our group in 1994 after being pilot tested by a small group of endoscopists. The final survey measured endoscopist characteristics, workload parameters, and experience during and after participants endured an injury. Endoscopist characteristics included age, sex, height, weight, hand dominance, physical activity level, main avocational activities, and practice setting. Workload parameters included number of years in practice, hours and number and/or www.giejournal.org
Musculoskeletal injuries related to endoscopy
type of endoscopies per week, and proportion of time spent performing procedures. Injury experiences included location of pain or injury, the effect of the injury on work, modifications of practice, and required treatments. If the respondent had a current or prior musculoskeletal injury, its relationship to endoscopy was further characterized as definitely, possibly, or not related.
Survey data collection Responders were invited to participate via e-mail. The introductory e-mail described the study and included a direct Web link to the online survey instrument (SurveyMonkey, Palo Alto, Calif). A first reminder e-mail was sent to participants who did not respond within 2 weeks after the initial e-mail. A second reminder e-mail was sent to those having no response within 4 weeks. All answers remained anonymous to minimize the potential for response bias. Responders were not remunerated for their participation.
Statistical analysis We used descriptive statistics to define the prevalence of injuries definitely or potentially related to endoscopy. Responders were classified into 3 groups for analysis: those reporting an injury definitely related to endoscopy, those reporting an injury potentially related to endoscopy, and those having no injury or an injury definitely not related to endoscopy. To explore potential differences among these groups, we applied comparative statistics (the Fisher exact test for proportions and Kruskal-Wallis equality test or analysis of variance for continuous variables). Differences across groups were considered significant if the associated P value was ! .05. In addition, we performed ordered logistic regression (because variables of interest, such as years in practice, were ordinal) to evaluate for differences in provider characteristics that may have been associated with a greater frequency of endoscopic injuries. Each variable having a P value ! .05 on univariate analysis was incorporated into a multivariate regression model to confirm its independent association with endoscopic injury. Consistent with the practices of survey research, no correction for multiple testing was made to the P values arising from the various comparisons of data from the respondents’ questionnaires. Analysis was performed by using Stata version 11.2 (StataCorp LP, College Station, Tex).
RESULTS Prevalence and location of musculoskeletal injury A total of 5239 e-mails were sent, and 684 individuals (13%) completed the survey and were included in the Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 295
Musculoskeletal injuries related to endoscopy
Ridtitid et al
Figure 1. Subject cohort: musculoskeletal injury among endoscopists. *Subjects who did not respond to question 17 of our survey were assumed to be non-injured physicians.
analysis (Fig. 1). All respondents reported performing endoscopy ever, and 96.8% currently. Excluding 34 individuals who did not respond to whether or not they had developed an injury related to endoscopy, 464 of 650 respondents (71.4%) experienced an injury related or unrelated to endoscopy in their necks, backs, or upper or lower limbs. The majority (362/684) of individuals had an injury definitely (n Z 204, 29.8%) or possibly (n Z 158, 23.1%) related to endoscopy, whereas 322 of 684 (47.1%) had no injury or an injury unrelated to endoscopy (Table 1). In those who reported pain definitely or probably associated with endoscopy, the pain was evident during endoscopy with or without outside work. Of those having an injury possibly or definitely related to endoscopy (n Z 362), 67 (18.5%) required time off from endoscopy because of occupational injury, ranging from 1 to 21 days. The most common sites of pain were neck and/or upper back (29.3%) and thumb (27.6%). Other areas included lower back (18.8%), elbow (10.5%),
hand (10.2%), and shoulder (10.2%). Hand numbness and carpal tunnel syndrome were found in 4.4% and 5.8%, respectively.
296 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
Factors associated with endoscopy-related injury Baseline characteristics of endoscopists were similar among physicians having injury definitely related to endoscopy versus possibly related to endoscopy versus no injury or having an injury unrelated to endoscopy, including age (P Z .23), sex (P Z .71), height (P Z .43), weight (P Z .05), hand dominance (P Z .32), physical activity level (P Z .70), and main avocational activities and/or hobbies (P Z .54) (Table 2). Compared with physicians in an academic center, physicians working in community practices reported a higher rate of injury definitely related to endoscopy (55; 28% vs 149; 73%) or potentially related to endoscopy (50; 32% vs 108; 68%), as opposed to those having no injury or an injury unrelated to endoscopy (138; 43% vs 184; 57%; P Z .006).
Ridtitid et al
Musculoskeletal injuries related to endoscopy
TABLE 1. Musculoskeletal injuries perceived related to endoscopy (n [ 684 survey respondents)* Variable Experienced injury perceived related to endoscopy
No. (%) 362 (52.9)
Required time off from endoscopy
67 (18.5)
Required modifications of practice
198 (54.7)
Required treatment
248 (68.5)
Type of injuryy Thumb pain
100 (27.6)
Shoulder pain
37 (10.2)
Hand pain
37 (10.2)
Neck/upper back pain
106 (29.3)
Lower back pain
68 (18.8)
Elbow pain
38 (10.5)
Hand numbness
16 (4.4)
Carpal tunnel syndrome
21 (5.8)
Other
15 (4.1)
*Participants who did not respond to question 17 of our survey were assumed to be non-injured physicians. yMore than 1 injury was reported by some individuals.
We compared workload parameters among groups (Table 3). Higher procedure volume was significantly associated with a higher rate of injury (P! .001). Of those with endoscopy-related injuries, 82% performed O20 endoscopies per week, compared with 62% of those having no related injury. Endoscopists who spent more hours per week performing procedures had a significantly higher proportion of injury (P! .001). Of those, 74% performed procedures O16 hours per week, compared with 53% of those having no related injury. Similarly, those who spent a greater proportion of their time performing endoscopy were significantly associated with a higher rate of injury (P! .001). Of those injured, 71% spent O40% of their time doing endoscopy, compared with 49% of those having no related injury. Physicians performing endoscopies for more years had a significantly higher percentage of injury (P Z .007). Based on the types of procedure, there was no statistically significant difference in the proportion of time spent doing colonoscopy (P Z .08), ERCP (P Z .27), EUS (P Z .83), and enteroscopy (P Z .93). However, those who spent a lower proportion of their time doing EGD had a higher rate of injury (P Z .02). There was no significant difference in www.giejournal.org
the frequency of breaks between procedures among groups (P Z .28). On univariate ordered logistic regression (Table 3), higher procedure volume (P ! .001), more hours doing endoscopy per week (P ! .001), greater proportion of time spent performing endoscopy (P ! .001), and cumulative years performing endoscopy (P Z .004) were associated with more work-related injuries. Combining these factors into a multivariate logistic regression model confirmed that physicians spending a greater proportion of their time performing endoscopy (P Z .002) and those with more years of experience (P Z .005) were strongly associated with work-related injuries. After we controlled for other variables, more hours of endoscopy per week (P Z .065) and having a higher weekly procedure volume (P Z .890) were no longer statistically significant.
Environments, modifications, and treatments At their workplaces, most physicians reported using a height-adjustable examination table (77.3%), positioning the video monitor at eye level (61.5%), and sitting during colonoscopy (7.9%). Of those having injury, 198 (54.7%) made practice modifications for the injury, and 248 (68.5%) required specific treatments (Table 4). Modifications, listed as options in the questionnaire, included stretching (56.6%), taking breaks (20.7%), using an adjustable table (20.7%), decreasing the procedure volume of endoscopy (14.4%), standing on a rubber mat (32.6%), and wearing orthopedic shoes (24.9%). Specific treatments included medications (56.6%), steroid injection (26.8%), physiotherapy (45.3%), rest (33.7%), splinting (22.7%), and surgery (13.3%). Of those requiring steroid injection (n Z 97) and surgery (n Z 48), the most common type of injury was thumb (36.1%) and neck and/or upper back (37.5%), respectively.
Trends of musculoskeletal injury among endoscopists We informally compared self-reported endoscopy workload and musculoskeletal injury by using data derived from our 1994 survey with the current sampling frame (Supplemental Table 2, available online at www. giejournal.org). Despite similarities in respondent demographics and practice settings, there were consistent increases in workload parameters and work-related injuries between 1994 and 2013.6
DISCUSSION Previously we reported survey data from 1994 evaluating endoscopist injuries.6 At that time, 265 respondents (43.8%) had experienced injuries for O6 months. The present study confirmed a high prevalence Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 297
Musculoskeletal injuries related to endoscopy
Ridtitid et al
TABLE 2. Baseline characteristics of endoscopists (n [ 684)
Variable
Overall (n [ 684)
Injury definitely related to endoscopy (n [ 204)
Injury possibly related to endoscopy (n [ 158)
No injury/injury not related to endoscopy (n [ 322)
P value
Age, mean ( SD), y
50.8 (11.0)
52.0 (10.8)
50.5 (10.2)
50.2 (11.6)
.23
Male sex, no. (%)
603 (88.1)
177 (86.8)
139 (88.0)
287 (95.1)
.69
Height, m
1.76 (0.10)
1.76 (0.08)
1.76 (0.08)
1.76 (0.12)
.43
Weight, kg
82.8 (34.8)
80.2 (13.3)
81.3 (15.0)
82.7(14.9)
.05
Hand dominance, no. (%)
.34
Left
72 (10.5)
21 (10.3)
12 (7.6)
39 (12.1)
Right
612 (89.5)
183 (89.7)
146 (92.4)
283 (87.9)
Activity level, no. (%)
.70
Mild
104 (15.2)
30 (14.7)
29 (18.4)
45 (14.0)
Moderate
299 (43.7)
90 (44.1)
70 (44.3)
139 (43.2)
Very active
281 (41.1)
84 (41.2)
59 (37.3)
138 (42.9)
Main avocational activities/hobbies,* no. (%)
.55
Light
98 (14.3)
32 (16.2)
22 (14.8)
44 (15.0)
Moderate
322 (47.1)
103 (52.3)
81 (54.4)
138 (47.1)
Vigorous
168 (24.6)
44 (22.3)
36 (24.2)
88 (30.0)
51 (7.4)
18 (9.1)
10 (6.7)
23 (7.9)
Muscle strengthening Practice setting, no. (%)
.005
Small community
165 (24.1)
56 (27.5)
37 (23.4)
72 (22.4)
Large community
276 (40.3)
93 (45.6)
71 (45.0)
112 (34.8)
Small academic center
25 (3.6)
7 (3.4)
2 (1.2)
16 (5.0)
Large academic center
218 (31.9)
48 (23.5)
48 (30.3)
122 (37.9)
*Classified according to Physical Activities Guideline 2008 by Centers for Disease Control and Prevention (http://www.health.gov/paguidelines/guidelines/ default.aspx).
of work-related injury among the Western population of endoscopists (52.9%) despite a diverse set of practice settings and endoscopy patterns. There was a higher rate of injury definitely related to endoscopy (149; 73% vs 55; 27%) or potentially related to endoscopy (108; 68% vs 50; 32%) in physicians working in community practices, compared with those in academic centers (P Z .006). This may be explained by a greater number of endoscopies (O20 endoscopies/week) (84% vs 61%) and more hours spent performing endoscopy (O16 hours/week) (74% vs 56%) among physicians working in the community setting, when compared with those in academic centers. However, there may have been a response bias favoring respondents from academic centers and/or those having prior work-related injuries. With regard to the types of endoscopy, there was a trend
to higher rates of injury among endoscopists performing more colonoscopy than other procedures (P Z .08). Compared with EGD, colonoscopy requires more instrument manipulation and torque, which may lead to thumb and wrist injury.5,13-15 Although more advanced endoscopic procedures such as ERCP and EUS are more likely to inflict musculoskeletal injury than other examinations,16 our study showed no statistically significant difference in proportion of time spent doing ERCP (P Z .27) and EUS (P Z .83) between those having definite and/or possible injury and no injury and/or injury not related to endoscopy. Because we had only a small number of respondents who spent R50% of their time performing ERCP (21; 3.1%), EUS (13; 2.0%), or some combination, our sample was inadequate to detect such a difference.
298 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
Ridtitid et al
Musculoskeletal injuries related to endoscopy
TABLE 3. Endoscopy workload parameters (n [ 684)
Variable
Overall (n [ 684)
Injury definitely related to endoscopy (n [ 204)
Injury possibly related to endoscopy (n [ 158)
No injury/injury not related to endoscopy (n [ 322)
Time performing endoscopy, no. (%), y 0-15
293 (42.8)
69 (33.7)
64 (40.5)
160 (49.7)
16-30
250 (36.6)
88 (43.1)
62 (39.2)
100 (31.1)
O30
141 (20.6)
47 (23.0)
32 (20.3)
62 (19.3)
Endoscopies performed per week, no. (%) 0-20
187 (27.3)
36 (17.6)
29 (18.4)
122 (37.9)
21-40
352 (51.5)
125 (61.3)
91 (57.6)
136 (42.2)
41-60
122 (17.8)
33 (16.2)
34 (21.5)
55 (17.1)
O60
23 (3.4)
10 (4.9)
4 (2.5)
9 (2.8)
Time spent performing endoscopy, no. (%), h/wk 0-15
245 (35.8)
49 (24.0)
44 (27.9)
152 (47.2)
16-30
374 (54.7)
130 (63.7)
98 (62.0)
146 (45.3)
O30
65 (9.5)
25 (12.3)
16 (10.1)
24 (7.5)
Proportion of time (%) spent performing endoscopy, no. (%)
P value
Univariate P value*
Multivariate P valuey
.007
.004
.005
! .001
! .001
.890
! .001
! .001
.065
! .001
! .001
.002
0-20
85 (12.4)
6 (2.9)
9 (5.7)
70 (21.7)
21-40
182 (26.6)
47 (23.1)
42 (26.6)
93 (28.9)
41-60
242 (35.4)
85 (41.7)
72 (45.6)
85 (26.4)
61-80
144 (21.1)
57 (27.9)
28 (17.7)
59 (18.3)
81-100
31 (4.5)
9 (4.4)
7 (4.4)
15 (4.7)
EGD
25 (20-40)
25 (20-30)
30 (20-30)
30 (20-40)
.02
Colonoscopy
60 (40-70)
60 (50-70)
60 (40-70)
60 (40-70)
.08
ERCP
10 (0-10)
10 (0-10)
10 (0-10)
10 (0-10)
.27
EUS
!10
!10
!10
!10
.83
Enteroscopy
!10
!10
!10
!10
.93
Proportion of time (%) spent doing procedure, median (range)
Break frequency, no. (%)
www.giejournal.org
.28
Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 299
Musculoskeletal injuries related to endoscopy
Ridtitid et al
TABLE 3. Continued
Overall (n [ 684)
Injury definitely related to endoscopy (n [ 204)
Injury possibly related to endoscopy (n [ 158)
No injury/injury not related to endoscopy (n [ 322)
Regular
74 (10.8)
21 (10.4)
17 (10.8)
36 (13.9)
Occasional
310 (45.3)
113 (55.9)
74 (47.1)
123 (47.3)
None
235 (34.4)
68 (33.7)
66 (42.0)
101 (38.9)
Variable
P value
Univariate P value*
Multivariate P valuey
*Ordered logistic regression (univariate). yOrdered logistic regression, inclusion of all variables having a P value ! .05 on univariate analysis (multivariate).
In contrast to studies from the United States (n Z 176)4 and Japan (n Z 311),5 on univariate analysis we observed a significantly higher rate of injury among those performing a greater volume of endoscopy (P ! .001), more hours doing endoscopy per week (P ! .001), greater proportion of time spent performing endoscopy (P! .001), and more years of experience (P Z .004). In our study, the majority of respondents (76%) performed O20 endoscopies per week, and most of them (67%) spent O16 hours per week performing endoscopy. In the prior U.S. study, 37 of 71 endoscopists performed procedures %10 half days per month, whereas 17 of 71 spent R21 half days per month performing endoscopy. The mean ( standard deviation [SD]) number of EGD and colonoscopy procedures per week was !20 procedures (13.08 9.89 and 15.85 11.17, respectively).4 The Japanese study reported 161 of 190 endoscopists performing colonoscopies; physicians performing O6.7 colonoscopies per week was not associated with the development of injury, compared with those performing %6.7 procedures per week (P Z .091).5 In our survey, the definition of high volume was higher than the other studies. Thus, in recent studies, there is a greater discrepancy between endoscopists performing high and low procedure volumes when compared with studies in the past. This may be the explanation for the strong association between procedure volume and injury found in our analysis. Multivariate analysis confirmed an association between more years in practice (experience) and having a procedure-dominant practice (greater proportion of time performing endoscopy) with a greater number of endoscopy-related injuries. Our previous study reported that physicians who performed more endoscopy (measured in terms of hours per week, number per week, or percentage of working time) had a significant predisposition to developing the conditions of any overuse syndrome. However, thumb pain, hand pain, elbow pain, low back pain, carpal tunnel syndrome, and neck pain were not more likely in those who performed more endoscopy.6 Although our previous study showed that the majority of injured endoscopists (67%) averaged !16 hours per
week in endoscopy, a precise number of endoscopies per week was not queried. The present study demonstrated that physicians currently spend more time in endoscopy compared with the population surveyed 20 years ago (Supplemental Fig. 1, available online at www. giejournal.org). It appears that the incidence of neck and/or upper back and thumb pain has increased over time, whereas the incidence of low back pain has decreased. A potential explanation for this may be increased procedure volume juxtaposed with lighter and more flexible equipment, potentially reducing injuries to the back. Additional studies exploring these trends would be of interest. In either case, the number of injured endoscopists requiring treatment for workrelated ailments has increased over time. Although our survey demonstrated more frequent musculoskeletal injury among endoscopists who have greater procedure volume, only 54.7% of endoscopists modified their practices as a result of their injuries; this low rate is similar to that of previous reports.3-6 Workplace environment is an important factor for determining the upper body posture of the endoscopist.16 However, only 20.7% of physicians in our survey reported using an adjustable table in response to injury. A minority of injured physicians (18.5%) required time off from endoscopy. We did not query the reason for the low rate of modifications to practice or environment in response to an injury. This may relate to lack of knowledge or influence on endoscopy unit design and practice workflow. A pilot study of musculoskeletal load during colonoscopy reported exceeding the injury threshold for the highest mean ( SD) right-thumb peak pinch forces developing during left (10.4 [4.1] N) and right (10.1 [4.5] N) colon insertion.15 Furthermore, activity of the left-wrist extensors, the left-thumb extensors, and the right-wrist extensors exceeded the American Conference of Industrial Hygienists (ACGIH) hand activity level action limit during routine colonoscopy.15 The ACGIH recommended task modification to reduce the risk of repetitive injury if the activities exceed the hand activity level action limit.17
300 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
Ridtitid et al
Musculoskeletal injuries related to endoscopy
TABLE 4. Modification of practice and treatments for endoscopy-related injury (n [ 362) Variable
No. (%)
Modification of practice* None
164 (45.3)
Stretch
205 (56.6)
Breaks
75 (20.7)
Adjustable table
75 (20.7)
Less endoscopy
52 (14.4)
Stand on rubber mat
118 (32.6)
Orthopedic shoes
90 (24.9)
(O16 hours), and more years in practice are strongly associated with a higher rate of endoscopy-related injury. However, only half of injured physicians modified their practices to facilitate healing of injuries or to prevent further injury. Future studies are needed to identify occupational interventions to minimize injury, including practice arrangement (limits for procedure volume and/or time spent performing endoscopies), technical modifications to endoscopes, and modifications to the endoscopy suite. REFERENCES
Therefore, more attention to injury prevention would be helpful among endoscopists. The ASGE recommends optimizing the ergonomics of endoscopy.16 However, these recommendations are based on laparoscopic surgery and general ergonomic studies.18-23 Thus, further studies of optimizing the ergonomics of endoscopy are needed, with attention to form being a component of endoscopic training. Our study is limited by a low response rate and potential response bias; because the survey was designed to query endoscopic injuries, respondents may have been more likely to have experienced an injury in the past (overestimating the prevalence of injury among endoscopists). Also, injuries definitely or potentially related to endoscopy were classified by self-report. It is possible that some injuries were in fact unrelated to endoscopy (false positives), and others were unknowingly related to endoscopy (false negatives). Nevertheless, this is the largest sample size reported in the medical literature. Our sample size permitted a robust analysis for endoscopist and practice characteristics associated with injury. In conclusion, the present study confirms a high prevalence of work-related injuries among endoscopists. Higher procedure volume (O20 endoscopies per week), more hours per week performing endoscopy
1. O'Sullivan S, Bridge G, Ponich T. Musculoskeletal injuries among ERCP endoscopists in Canada. Can J Gastroenterol 2002;16:369-74. 2. Liberman AS, Shrier I, Gordon PH. Injuries sustained by colorectal surgeons performing colonoscopy. Surg Endosc 2005;19:1606-9. 3. Byun YH, Lee JH, Park MK, et al. Procedure-related musculoskeletal symptoms in gastrointestinal endoscopists in Korea. World J Gastroenterol 2008;14:4359-64. 4. Hansel SL, Crowell MD, Pardi DS, et al. Prevalence and impact of musculoskeletal injury among endoscopists: a controlled pilot study. J Clin Gastroenterol 2009;43:399-404. 5. Kuwabara T, Urabe Y, Hiyama T, et al. Prevalence and impact of musculoskeletal pain in Japanese gastrointestinal endoscopists: a controlled study. World J Gastroenterol 2011;17:1488-93. 6. Buschbacher R. Overuse syndromes among endoscopists. Endoscopy 1994;26:539-44. 7. Everhart JE, Ruhl CE. Burden of digestive diseases in the United States part I: overall and upper gastrointestinal diseases. Gastroenterology 2009;136:376-86. 8. Everhart JE, Ruhl CE. Burden of digestive diseases in the United States part II: lower gastrointestinal diseases. Gastroenterology 2009;136:741-54. 9. Hansel SL, Di Baise JK, Crowell MD, et al. Prevalence and impact of musculoskeletal injury among endoscopists: a controlled study [abstract]. Gastrointest Endosc 2007;65:AB104. 10. Keate RF, Dryden GW, Wang K, et al. Occupational injuries to endoscopists: report from the ASGE Web survey [abstract]. Gastrointest Endosc 2006;63:AB111. 11. Raftopoulos SC, Segarajamingams DS, Yusoff IF, et al; Endoscopy. J Gastroenterol Hepatol 2007;22:A325-43. 12. Shergill AK, McQuaid KR, Rempel D. Ergonomics and GI endoscopy. Gastrointest Endosc 2009;70:145-53. 13. Cappell MS. Colonoscopist’s thumb: DeQuervains’s syndrome (tenosynovitis of the left thumb) associated with overuse during endoscopy. Gastrointest Endosc 2006;64:841-3. 14. Mohankumar D, Garner H, Ruff K, et al. Characterization of right wrist posture during simulated colonoscopy: an application of kinematic analysis to the study of endoscopic maneuvers. Gastrointest Endosc 2014;79:480-9. 15. Shergill AK, Asundi KR, Barr A, et al. Pinch force and forearm-muscle load during routine colonoscopy: a pilot study. Gastrointest Endosc 2009;69:142-6. 16. Pedrosa MC, Farraye FA, Shergill AK, et al. Minimizing occupational hazards in endoscopy: personal protective equipment, radiation safety, and ergonomics. Gastrointest Endosc 2010;72:227-35. 17. ACGIH. TLVs and BEIs: threshold limit values for chemical substances and physical agents and biological exposure indices. Cincinnati (Ohio): Industrial Hygienists, 2008. 18. Haveran LA, Novitsky YW, Czerniach DR, et al. Optimizing laparoscopic task efficiency: the role of camera and monitor positions. Surg Endosc 2007;21:980-4. 19. Matern U, Faist M, Kehl K, et al. Monitor position in laparoscopic surgery. Surg Endosc 2005;19:436-40.
www.giejournal.org
Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 301
Treatment* None
114 (31.5)
Medications
205 (56.6)
Steroid injection
97 (26.8)
Physiotherapy
164 (45.3)
Rest
122 (33.7)
Splinting
82 (22.7)
Surgery
48 (13.3)
*More than 1 modification/treatment was reported by some individuals.
Musculoskeletal injuries related to endoscopy 20. Zehetner J, Kaltenbacher A, Wayand W, et al. Screen height as an ergonomic factor in laparoscopic surgery. Surg Endosc 2006;20: 139-41. 21. Manasnayakorn S, Cuschieri A, Hanna GB. Ergonomic assessment of optimum operating table height for hand-assisted laparoscopic surgery. Surg Endosc 2009;23:783-9.
302 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
Ridtitid et al 22. Berquer R, Smith WD, Davis S. An ergonomic study of the optimum operating table height for laparoscopic surgery. Surg Endosc 2002;16:416-21. 23. van Det MJ, Meijerink WJ, Hoff C, et al. Optimal ergonomics for laparoscopic surgery in minimally invasive surgery suites: a review and guidelines. Surg Endosc 2009;23:1279-85.
www.giejournal.org
Ridtitid et al
Musculoskeletal injuries related to endoscopy
APPENDIX
Supplemental Figure 1. X axis: proportion of time (%) spent performing endoscopies. Y axis: proportion of endoscopists.
www.giejournal.org
Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 302.e1
Musculoskeletal injuries related to endoscopy
Ridtitid et al
SUPPLEMENTAL TABLE 1. A 25-question, self-administered, electronic survey* Question
Answer
1. Sex
O Male O Female
2. Age
____ years
3. Height
____ feet ____ inches
4. Weight
____ pounds
5. Hand dominance 6. What is your activity level?
7. How would you describe your practice?
8. List 2 main avocational activities or hobbies (ie, golf, tennis, video games) 9. How many hours per week do you spend performing them?
O Left O Right O Mildly active O Moderately active O Very active O Small community practice (!5 specialists performing endoscopy) O Large community practice (R5 specialists performing endoscopy) O Small academic practice (University-affiliated group with !5 specialists performing endoscopy) O Large academic practice (university-affiliated group with R5 specialists performing endoscopy) 1)____________________________ 2)____________________________ Activity 1 _______ hours/week Activity 2 _______ hours/week
10. Do you currently perform endoscopy?
O Yes O No
11. Have you ever in your lifetime performed GI endoscopy?
O Yes O No
12. No. of years performing endoscopic procedures 13. Approximate no. of endoscopies performed per week 14. Hours/week usually spent performing endoscopies
O 0-5 O 6-10 O 11-15 O 16-20 O 21-25 O 26-30 O 31-35 O 36-40 O O40 O 0-5 O 6-10 O 11-15 O 16-20 O 21-25 O 26-30 O 31-35 O 36-40 O 41-45 O 46-50 O 51-55 O 56-60 O 61-65 O 66-70 O 71-75 O 76-80 O O80 O 0-5 O 6-10 O 11-15 O 16-20 O 21-25 O 26-30 O 31-35 O 36-40 O O40
15. Endoscopies make up what percentage of your work time?
O 0-10 O 11-20 O 21-30 O 31-40 O 41-50 O 51-60 O 61-70 O 71-80 O 81-90 O 91-100
16. Of the time you spend doing endoscopy, what percentage do you spend doing the following:
EGD O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100 Colonoscopy O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100 ERCP O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100 EUS O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100 Enteroscopy O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100
302.e2 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
Ridtitid et al
Musculoskeletal injuries related to endoscopy
SUPPLEMENTAL TABLE 1. Continued Question 17. Have you ever experienced injury (pain or numbness) in your neck, back, or upper or lower limbs? 18. Details about injury (click all that apply)
Answer O Yes O No Type of injury: O Thumb pain O Shoulder pain O Hand pain O Neck/upper back pain O Lower back pain O Elbow pain O Hand numbness O Carpal tunnel syndrome O Other Side affected: O Left O Right O Both Was/is pain caused by endoscopy? O Yes O No O Maybe When is/was pain evident? O At work performing endoscopy O At work performing endoscopy/clinic O At work/outside work O Outside work only How much does/did this bother you? O 1 (least) O 2 O 3 O 4 O 5 (most) How long have you had these symptoms? _____ years
19. Have you ever had to take time off from performing endoscopy because of musculoskeletal injury perceived related to endoscopy?
O Yes O No
20. If Yes to above, what is the longest consecutive amount of time you have taken off work because of a musculoskeletal injury perceived to be related to endoscopy?
______ days/months/years
21. Have you ever had to modify your practice and/or shorten your endoscopic case load due to occupational injury?
O Yes O No
22. If you experienced musculoskeletal pain/injury related to endoscopy, what modifications have you made to your endoscopic practice? (click all that apply)
O None O Stretch O Breaks O Adjustable bed O Less endoscopy O Stand on rubber mat O Orthopedic shoes/sneakers
23. What treatment(s) have you received for your condition? (click all that apply)
O None O Medications O Steroid injections O Physiotherapy (P.T.) O Rest O Splinting O Surgery
24. How often do you take breaks?
25. Do you perform any of these environmental modifications during endoscopy? (click all that apply)
O Regularly O Occasionally O Never O None O Height-adjustable examination table O Position of monitor in front at eye level O Stopped helping to move patients after procedures O Sit when you perform colonoscopy
*SurveyMonkey, Palo Alto, Calif.
www.giejournal.org
Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 302.e3
Musculoskeletal injuries related to endoscopy
Ridtitid et al
SUPPLEMENTAL TABLE 2. Trends over time (1994 results compared with 2013 results) 1994 (n [ 256)
2013 (n [ 684)
47.8 (8.6)
50.8 (11.0)
95.1
88.1
Community
84.2
64.5
Academic center
15.5
35.5
0-20
35
12.4
21-40
43
26.6
41-60
18
35.4
61-80
2
21.1
81-100
0
4.5
0-15
67
35.8
16-30
32
54.7
O30
2
9.5
Percentage of endoscopists who experienced injury related/unrelated to endoscopy
57
71.4
Percentage of endoscopists who experienced injury definitely/possibly related to endoscopy
46
52.9
Neck pain
13
29.3
Thumb pain
19
27.6
Lower back pain
27
18.8
Elbow pain
15
10.5
Shoulder pain
19
10.2
Hand pain
14
10.2
Carpal tunnel syndrome
6
5.8
Hand numbness
12
4.4
None
49.7
31.5
Medications
17.8
56.6
Physiotherapy
16.5
45.3
Not reported
26.8
Rest
8.3
33.7
Splinting
4.5
22.7
Surgery
3.2
13.3
Variable Age, mean ( SD), y Male sex (%) Practice setting (%)
Proportion of time (%) spent performing endoscopy
Hours/week spent performing endoscopy (%)
Percentage of endoscopists with various injuries,* %
Treatment,* %
Steroid injection
*More than 1 injury/treatment was reported by some individuals.
302.e4 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
Prevalence and risk factors for musculoskeletal injuries related to endoscopy Wiriyaporn Ridtitid, MD,1,2 Gregory A. Coté, MD, MS,1 Wesley Leung, MD,1 Ralph Buschbacher, MD,1 Sheryl Lynch, RN,1 Evan L. Fogel, MD,1 James L. Watkins, MD,1 Glen A. Lehman, MD,1 Stuart Sherman, MD,1 Lee McHenry, MD1 Indianapolis, Indiana, USA
Background: There are limited data regarding work-related injury among endoscopists. Objective: To define the prevalence of endoscopy-related musculoskeletal injuries and their impact on clinical practice and to identify physician and practice characteristics associated with their development. Design: Survey. Setting: Electronic survey of active members of the American Society for Gastrointestinal Endoscopy with registered e-mail addresses. Participants: Physicians who currently or ever performed endoscopy and responded to the survey between February 2013 and November 2013. Intervention: A 25-question, self-administered, electronic survey. Main Outcome Measurements: Prevalence, location, and ramifications of work-related injuries and endoscopist characteristics and workload parameters associated with endoscopy-related injury. Results: The survey was completed by 684 endoscopists. Of those, 362 (53%) experienced a musculoskeletal injury perceived definitely (n Z 204) or possibly (n Z 158) related to endoscopy. Factors associated with a higher rate of endoscopy-related injury included higher procedure volume (O20 cases/week; P! .001), greater number of hours per week spent performing endoscopy (O16 hours/week; P! .001), and total number of years performing endoscopy (P Z .004). The most common sites of injury were neck and/or upper back (29%) and thumb (28%). Only 55% of injured endoscopists used practice modifications in response to injuries. Specific treatments included medications (57%), steroid injection (27%), physiotherapy (45%), rest (34%), splinting (23%), and surgery (13%). Limitations: Self-reported data of endoscopy-related injury. Conclusion: Among endoscopists there is a high prevalence of injuries definitely or potentially related to endoscopy. Higher procedure volume, more time doing endoscopy per week, and cumulative years performing endoscopy are associated with more work-related injuries. (Gastrointest Endosc 2015;81:294-302.)
Several studies have suggested a high prevalence of musculoskeletal injuries among endoscopists.1-6 However, these studies were conducted when procedure
volumes and the complexity of endoscopic therapies were typically lower.7,8 Further, data from Western populations of endoscopists are limited, and a robust analysis
Abbreviations: ACGIH, American Conference of Industrial Hygienists; ASGE, American Society for Gastrointestinal Endoscopy.
Current affiliations: Indiana University School of Medicine, Indianapolis, Indiana, USA (1), Chulalongkorn University, King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand (2).
DISCLOSURE: All authors disclosed no financial relationships relevant to this article. Copyright ª 2015 by the American Society for Gastrointestinal Endoscopy 0016-5107/$36.00 http://dx.doi.org/10.1016/j.gie.2014.06.036 Received April 17, 2014. Accepted June 16, 2014.
294 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
Reprint requests: Lee McHenry, MD, Professor of Medicine, Indiana University School of Medicine, 550 North University Boulevard, UH 2300, Indianapolis, IN 46202. If you would like to chat with an author of this article, you may contact Dr McHenry at [email protected].
www.giejournal.org
Ridtitid et al
to identify risk factors (endoscopist and practice) associated with endoscopy-related injury is lacking. Endoscopists are presumably at risk for overuse syndromes because of the repetitive movements and potentially awkward posture associated with endoscopy. Based on earlier reports,1-3,6,9-11 a systemic review estimated that 37% to 89% of endoscopists develop work-related injuries.12 Moreover, suspected risk factors included repetitive hand motion, high hand forces, and awkward wrist, shoulder, and neck postures.12 However, previous studies of work-related injury among endoscopists have been limited by small sample sizes and limited analysis of risk factors. The present widespread use of EGD and colonoscopy implies that endoscopists may perform more procedures on a daily basis than in the past. Furthermore, the burden and performance of more technically challenging procedures such as ERCP, EUS, device-assisted enteroscopy, and others may predispose endoscopists to higher rates of repetitive stress injuries than previously reported. Therefore, our primary aim was to define the prevalence and types of endoscopy-related musculoskeletal injuries in the current era of high-volume endoscopy with advanced therapeutics. Second, we evaluated endoscopist and practice characteristics associated with these injuries.
METHODS Survey sampling We conducted an electronic survey of endoscopists who were active members of the American Society for Gastrointestinal Endoscopy (ASGE) with registered e-mail addresses (n Z 5239) between February 2013 and November 2013. Members who currently perform or ever have performed endoscopy (by self-report) were eligible to participate. Informed consent was implied by response to the survey. Before dissemination of the survey, our institutional review board approved the study.
Survey instrument The survey was a 25-question, self-administered, electronic survey that was developed by 2 endoscopists (L.M., W.L.) and 1 physiatrist (R.B.) (Supplemental Table 1, available online at www.giejournal.org). The survey instrument was a modified version of a previous survey6 conducted by our group in 1994 after being pilot tested by a small group of endoscopists. The final survey measured endoscopist characteristics, workload parameters, and experience during and after participants endured an injury. Endoscopist characteristics included age, sex, height, weight, hand dominance, physical activity level, main avocational activities, and practice setting. Workload parameters included number of years in practice, hours and number and/or www.giejournal.org
Musculoskeletal injuries related to endoscopy
type of endoscopies per week, and proportion of time spent performing procedures. Injury experiences included location of pain or injury, the effect of the injury on work, modifications of practice, and required treatments. If the respondent had a current or prior musculoskeletal injury, its relationship to endoscopy was further characterized as definitely, possibly, or not related.
Survey data collection Responders were invited to participate via e-mail. The introductory e-mail described the study and included a direct Web link to the online survey instrument (SurveyMonkey, Palo Alto, Calif). A first reminder e-mail was sent to participants who did not respond within 2 weeks after the initial e-mail. A second reminder e-mail was sent to those having no response within 4 weeks. All answers remained anonymous to minimize the potential for response bias. Responders were not remunerated for their participation.
Statistical analysis We used descriptive statistics to define the prevalence of injuries definitely or potentially related to endoscopy. Responders were classified into 3 groups for analysis: those reporting an injury definitely related to endoscopy, those reporting an injury potentially related to endoscopy, and those having no injury or an injury definitely not related to endoscopy. To explore potential differences among these groups, we applied comparative statistics (the Fisher exact test for proportions and Kruskal-Wallis equality test or analysis of variance for continuous variables). Differences across groups were considered significant if the associated P value was ! .05. In addition, we performed ordered logistic regression (because variables of interest, such as years in practice, were ordinal) to evaluate for differences in provider characteristics that may have been associated with a greater frequency of endoscopic injuries. Each variable having a P value ! .05 on univariate analysis was incorporated into a multivariate regression model to confirm its independent association with endoscopic injury. Consistent with the practices of survey research, no correction for multiple testing was made to the P values arising from the various comparisons of data from the respondents’ questionnaires. Analysis was performed by using Stata version 11.2 (StataCorp LP, College Station, Tex).
RESULTS Prevalence and location of musculoskeletal injury A total of 5239 e-mails were sent, and 684 individuals (13%) completed the survey and were included in the Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 295
Musculoskeletal injuries related to endoscopy
Ridtitid et al
Figure 1. Subject cohort: musculoskeletal injury among endoscopists. *Subjects who did not respond to question 17 of our survey were assumed to be non-injured physicians.
analysis (Fig. 1). All respondents reported performing endoscopy ever, and 96.8% currently. Excluding 34 individuals who did not respond to whether or not they had developed an injury related to endoscopy, 464 of 650 respondents (71.4%) experienced an injury related or unrelated to endoscopy in their necks, backs, or upper or lower limbs. The majority (362/684) of individuals had an injury definitely (n Z 204, 29.8%) or possibly (n Z 158, 23.1%) related to endoscopy, whereas 322 of 684 (47.1%) had no injury or an injury unrelated to endoscopy (Table 1). In those who reported pain definitely or probably associated with endoscopy, the pain was evident during endoscopy with or without outside work. Of those having an injury possibly or definitely related to endoscopy (n Z 362), 67 (18.5%) required time off from endoscopy because of occupational injury, ranging from 1 to 21 days. The most common sites of pain were neck and/or upper back (29.3%) and thumb (27.6%). Other areas included lower back (18.8%), elbow (10.5%),
hand (10.2%), and shoulder (10.2%). Hand numbness and carpal tunnel syndrome were found in 4.4% and 5.8%, respectively.
296 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
Factors associated with endoscopy-related injury Baseline characteristics of endoscopists were similar among physicians having injury definitely related to endoscopy versus possibly related to endoscopy versus no injury or having an injury unrelated to endoscopy, including age (P Z .23), sex (P Z .71), height (P Z .43), weight (P Z .05), hand dominance (P Z .32), physical activity level (P Z .70), and main avocational activities and/or hobbies (P Z .54) (Table 2). Compared with physicians in an academic center, physicians working in community practices reported a higher rate of injury definitely related to endoscopy (55; 28% vs 149; 73%) or potentially related to endoscopy (50; 32% vs 108; 68%), as opposed to those having no injury or an injury unrelated to endoscopy (138; 43% vs 184; 57%; P Z .006).
Ridtitid et al
Musculoskeletal injuries related to endoscopy
TABLE 1. Musculoskeletal injuries perceived related to endoscopy (n [ 684 survey respondents)* Variable Experienced injury perceived related to endoscopy
No. (%) 362 (52.9)
Required time off from endoscopy
67 (18.5)
Required modifications of practice
198 (54.7)
Required treatment
248 (68.5)
Type of injuryy Thumb pain
100 (27.6)
Shoulder pain
37 (10.2)
Hand pain
37 (10.2)
Neck/upper back pain
106 (29.3)
Lower back pain
68 (18.8)
Elbow pain
38 (10.5)
Hand numbness
16 (4.4)
Carpal tunnel syndrome
21 (5.8)
Other
15 (4.1)
*Participants who did not respond to question 17 of our survey were assumed to be non-injured physicians. yMore than 1 injury was reported by some individuals.
We compared workload parameters among groups (Table 3). Higher procedure volume was significantly associated with a higher rate of injury (P! .001). Of those with endoscopy-related injuries, 82% performed O20 endoscopies per week, compared with 62% of those having no related injury. Endoscopists who spent more hours per week performing procedures had a significantly higher proportion of injury (P! .001). Of those, 74% performed procedures O16 hours per week, compared with 53% of those having no related injury. Similarly, those who spent a greater proportion of their time performing endoscopy were significantly associated with a higher rate of injury (P! .001). Of those injured, 71% spent O40% of their time doing endoscopy, compared with 49% of those having no related injury. Physicians performing endoscopies for more years had a significantly higher percentage of injury (P Z .007). Based on the types of procedure, there was no statistically significant difference in the proportion of time spent doing colonoscopy (P Z .08), ERCP (P Z .27), EUS (P Z .83), and enteroscopy (P Z .93). However, those who spent a lower proportion of their time doing EGD had a higher rate of injury (P Z .02). There was no significant difference in www.giejournal.org
the frequency of breaks between procedures among groups (P Z .28). On univariate ordered logistic regression (Table 3), higher procedure volume (P ! .001), more hours doing endoscopy per week (P ! .001), greater proportion of time spent performing endoscopy (P ! .001), and cumulative years performing endoscopy (P Z .004) were associated with more work-related injuries. Combining these factors into a multivariate logistic regression model confirmed that physicians spending a greater proportion of their time performing endoscopy (P Z .002) and those with more years of experience (P Z .005) were strongly associated with work-related injuries. After we controlled for other variables, more hours of endoscopy per week (P Z .065) and having a higher weekly procedure volume (P Z .890) were no longer statistically significant.
Environments, modifications, and treatments At their workplaces, most physicians reported using a height-adjustable examination table (77.3%), positioning the video monitor at eye level (61.5%), and sitting during colonoscopy (7.9%). Of those having injury, 198 (54.7%) made practice modifications for the injury, and 248 (68.5%) required specific treatments (Table 4). Modifications, listed as options in the questionnaire, included stretching (56.6%), taking breaks (20.7%), using an adjustable table (20.7%), decreasing the procedure volume of endoscopy (14.4%), standing on a rubber mat (32.6%), and wearing orthopedic shoes (24.9%). Specific treatments included medications (56.6%), steroid injection (26.8%), physiotherapy (45.3%), rest (33.7%), splinting (22.7%), and surgery (13.3%). Of those requiring steroid injection (n Z 97) and surgery (n Z 48), the most common type of injury was thumb (36.1%) and neck and/or upper back (37.5%), respectively.
Trends of musculoskeletal injury among endoscopists We informally compared self-reported endoscopy workload and musculoskeletal injury by using data derived from our 1994 survey with the current sampling frame (Supplemental Table 2, available online at www. giejournal.org). Despite similarities in respondent demographics and practice settings, there were consistent increases in workload parameters and work-related injuries between 1994 and 2013.6
DISCUSSION Previously we reported survey data from 1994 evaluating endoscopist injuries.6 At that time, 265 respondents (43.8%) had experienced injuries for O6 months. The present study confirmed a high prevalence Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 297
Musculoskeletal injuries related to endoscopy
Ridtitid et al
TABLE 2. Baseline characteristics of endoscopists (n [ 684)
Variable
Overall (n [ 684)
Injury definitely related to endoscopy (n [ 204)
Injury possibly related to endoscopy (n [ 158)
No injury/injury not related to endoscopy (n [ 322)
P value
Age, mean ( SD), y
50.8 (11.0)
52.0 (10.8)
50.5 (10.2)
50.2 (11.6)
.23
Male sex, no. (%)
603 (88.1)
177 (86.8)
139 (88.0)
287 (95.1)
.69
Height, m
1.76 (0.10)
1.76 (0.08)
1.76 (0.08)
1.76 (0.12)
.43
Weight, kg
82.8 (34.8)
80.2 (13.3)
81.3 (15.0)
82.7(14.9)
.05
Hand dominance, no. (%)
.34
Left
72 (10.5)
21 (10.3)
12 (7.6)
39 (12.1)
Right
612 (89.5)
183 (89.7)
146 (92.4)
283 (87.9)
Activity level, no. (%)
.70
Mild
104 (15.2)
30 (14.7)
29 (18.4)
45 (14.0)
Moderate
299 (43.7)
90 (44.1)
70 (44.3)
139 (43.2)
Very active
281 (41.1)
84 (41.2)
59 (37.3)
138 (42.9)
Main avocational activities/hobbies,* no. (%)
.55
Light
98 (14.3)
32 (16.2)
22 (14.8)
44 (15.0)
Moderate
322 (47.1)
103 (52.3)
81 (54.4)
138 (47.1)
Vigorous
168 (24.6)
44 (22.3)
36 (24.2)
88 (30.0)
51 (7.4)
18 (9.1)
10 (6.7)
23 (7.9)
Muscle strengthening Practice setting, no. (%)
.005
Small community
165 (24.1)
56 (27.5)
37 (23.4)
72 (22.4)
Large community
276 (40.3)
93 (45.6)
71 (45.0)
112 (34.8)
Small academic center
25 (3.6)
7 (3.4)
2 (1.2)
16 (5.0)
Large academic center
218 (31.9)
48 (23.5)
48 (30.3)
122 (37.9)
*Classified according to Physical Activities Guideline 2008 by Centers for Disease Control and Prevention (http://www.health.gov/paguidelines/guidelines/ default.aspx).
of work-related injury among the Western population of endoscopists (52.9%) despite a diverse set of practice settings and endoscopy patterns. There was a higher rate of injury definitely related to endoscopy (149; 73% vs 55; 27%) or potentially related to endoscopy (108; 68% vs 50; 32%) in physicians working in community practices, compared with those in academic centers (P Z .006). This may be explained by a greater number of endoscopies (O20 endoscopies/week) (84% vs 61%) and more hours spent performing endoscopy (O16 hours/week) (74% vs 56%) among physicians working in the community setting, when compared with those in academic centers. However, there may have been a response bias favoring respondents from academic centers and/or those having prior work-related injuries. With regard to the types of endoscopy, there was a trend
to higher rates of injury among endoscopists performing more colonoscopy than other procedures (P Z .08). Compared with EGD, colonoscopy requires more instrument manipulation and torque, which may lead to thumb and wrist injury.5,13-15 Although more advanced endoscopic procedures such as ERCP and EUS are more likely to inflict musculoskeletal injury than other examinations,16 our study showed no statistically significant difference in proportion of time spent doing ERCP (P Z .27) and EUS (P Z .83) between those having definite and/or possible injury and no injury and/or injury not related to endoscopy. Because we had only a small number of respondents who spent R50% of their time performing ERCP (21; 3.1%), EUS (13; 2.0%), or some combination, our sample was inadequate to detect such a difference.
298 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
Ridtitid et al
Musculoskeletal injuries related to endoscopy
TABLE 3. Endoscopy workload parameters (n [ 684)
Variable
Overall (n [ 684)
Injury definitely related to endoscopy (n [ 204)
Injury possibly related to endoscopy (n [ 158)
No injury/injury not related to endoscopy (n [ 322)
Time performing endoscopy, no. (%), y 0-15
293 (42.8)
69 (33.7)
64 (40.5)
160 (49.7)
16-30
250 (36.6)
88 (43.1)
62 (39.2)
100 (31.1)
O30
141 (20.6)
47 (23.0)
32 (20.3)
62 (19.3)
Endoscopies performed per week, no. (%) 0-20
187 (27.3)
36 (17.6)
29 (18.4)
122 (37.9)
21-40
352 (51.5)
125 (61.3)
91 (57.6)
136 (42.2)
41-60
122 (17.8)
33 (16.2)
34 (21.5)
55 (17.1)
O60
23 (3.4)
10 (4.9)
4 (2.5)
9 (2.8)
Time spent performing endoscopy, no. (%), h/wk 0-15
245 (35.8)
49 (24.0)
44 (27.9)
152 (47.2)
16-30
374 (54.7)
130 (63.7)
98 (62.0)
146 (45.3)
O30
65 (9.5)
25 (12.3)
16 (10.1)
24 (7.5)
Proportion of time (%) spent performing endoscopy, no. (%)
P value
Univariate P value*
Multivariate P valuey
.007
.004
.005
! .001
! .001
.890
! .001
! .001
.065
! .001
! .001
.002
0-20
85 (12.4)
6 (2.9)
9 (5.7)
70 (21.7)
21-40
182 (26.6)
47 (23.1)
42 (26.6)
93 (28.9)
41-60
242 (35.4)
85 (41.7)
72 (45.6)
85 (26.4)
61-80
144 (21.1)
57 (27.9)
28 (17.7)
59 (18.3)
81-100
31 (4.5)
9 (4.4)
7 (4.4)
15 (4.7)
EGD
25 (20-40)
25 (20-30)
30 (20-30)
30 (20-40)
.02
Colonoscopy
60 (40-70)
60 (50-70)
60 (40-70)
60 (40-70)
.08
ERCP
10 (0-10)
10 (0-10)
10 (0-10)
10 (0-10)
.27
EUS
!10
!10
!10
!10
.83
Enteroscopy
!10
!10
!10
!10
.93
Proportion of time (%) spent doing procedure, median (range)
Break frequency, no. (%)
www.giejournal.org
.28
Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 299
Musculoskeletal injuries related to endoscopy
Ridtitid et al
TABLE 3. Continued
Overall (n [ 684)
Injury definitely related to endoscopy (n [ 204)
Injury possibly related to endoscopy (n [ 158)
No injury/injury not related to endoscopy (n [ 322)
Regular
74 (10.8)
21 (10.4)
17 (10.8)
36 (13.9)
Occasional
310 (45.3)
113 (55.9)
74 (47.1)
123 (47.3)
None
235 (34.4)
68 (33.7)
66 (42.0)
101 (38.9)
Variable
P value
Univariate P value*
Multivariate P valuey
*Ordered logistic regression (univariate). yOrdered logistic regression, inclusion of all variables having a P value ! .05 on univariate analysis (multivariate).
In contrast to studies from the United States (n Z 176)4 and Japan (n Z 311),5 on univariate analysis we observed a significantly higher rate of injury among those performing a greater volume of endoscopy (P ! .001), more hours doing endoscopy per week (P ! .001), greater proportion of time spent performing endoscopy (P! .001), and more years of experience (P Z .004). In our study, the majority of respondents (76%) performed O20 endoscopies per week, and most of them (67%) spent O16 hours per week performing endoscopy. In the prior U.S. study, 37 of 71 endoscopists performed procedures %10 half days per month, whereas 17 of 71 spent R21 half days per month performing endoscopy. The mean ( standard deviation [SD]) number of EGD and colonoscopy procedures per week was !20 procedures (13.08 9.89 and 15.85 11.17, respectively).4 The Japanese study reported 161 of 190 endoscopists performing colonoscopies; physicians performing O6.7 colonoscopies per week was not associated with the development of injury, compared with those performing %6.7 procedures per week (P Z .091).5 In our survey, the definition of high volume was higher than the other studies. Thus, in recent studies, there is a greater discrepancy between endoscopists performing high and low procedure volumes when compared with studies in the past. This may be the explanation for the strong association between procedure volume and injury found in our analysis. Multivariate analysis confirmed an association between more years in practice (experience) and having a procedure-dominant practice (greater proportion of time performing endoscopy) with a greater number of endoscopy-related injuries. Our previous study reported that physicians who performed more endoscopy (measured in terms of hours per week, number per week, or percentage of working time) had a significant predisposition to developing the conditions of any overuse syndrome. However, thumb pain, hand pain, elbow pain, low back pain, carpal tunnel syndrome, and neck pain were not more likely in those who performed more endoscopy.6 Although our previous study showed that the majority of injured endoscopists (67%) averaged !16 hours per
week in endoscopy, a precise number of endoscopies per week was not queried. The present study demonstrated that physicians currently spend more time in endoscopy compared with the population surveyed 20 years ago (Supplemental Fig. 1, available online at www. giejournal.org). It appears that the incidence of neck and/or upper back and thumb pain has increased over time, whereas the incidence of low back pain has decreased. A potential explanation for this may be increased procedure volume juxtaposed with lighter and more flexible equipment, potentially reducing injuries to the back. Additional studies exploring these trends would be of interest. In either case, the number of injured endoscopists requiring treatment for workrelated ailments has increased over time. Although our survey demonstrated more frequent musculoskeletal injury among endoscopists who have greater procedure volume, only 54.7% of endoscopists modified their practices as a result of their injuries; this low rate is similar to that of previous reports.3-6 Workplace environment is an important factor for determining the upper body posture of the endoscopist.16 However, only 20.7% of physicians in our survey reported using an adjustable table in response to injury. A minority of injured physicians (18.5%) required time off from endoscopy. We did not query the reason for the low rate of modifications to practice or environment in response to an injury. This may relate to lack of knowledge or influence on endoscopy unit design and practice workflow. A pilot study of musculoskeletal load during colonoscopy reported exceeding the injury threshold for the highest mean ( SD) right-thumb peak pinch forces developing during left (10.4 [4.1] N) and right (10.1 [4.5] N) colon insertion.15 Furthermore, activity of the left-wrist extensors, the left-thumb extensors, and the right-wrist extensors exceeded the American Conference of Industrial Hygienists (ACGIH) hand activity level action limit during routine colonoscopy.15 The ACGIH recommended task modification to reduce the risk of repetitive injury if the activities exceed the hand activity level action limit.17
300 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
Ridtitid et al
Musculoskeletal injuries related to endoscopy
TABLE 4. Modification of practice and treatments for endoscopy-related injury (n [ 362) Variable
No. (%)
Modification of practice* None
164 (45.3)
Stretch
205 (56.6)
Breaks
75 (20.7)
Adjustable table
75 (20.7)
Less endoscopy
52 (14.4)
Stand on rubber mat
118 (32.6)
Orthopedic shoes
90 (24.9)
(O16 hours), and more years in practice are strongly associated with a higher rate of endoscopy-related injury. However, only half of injured physicians modified their practices to facilitate healing of injuries or to prevent further injury. Future studies are needed to identify occupational interventions to minimize injury, including practice arrangement (limits for procedure volume and/or time spent performing endoscopies), technical modifications to endoscopes, and modifications to the endoscopy suite. REFERENCES
Therefore, more attention to injury prevention would be helpful among endoscopists. The ASGE recommends optimizing the ergonomics of endoscopy.16 However, these recommendations are based on laparoscopic surgery and general ergonomic studies.18-23 Thus, further studies of optimizing the ergonomics of endoscopy are needed, with attention to form being a component of endoscopic training. Our study is limited by a low response rate and potential response bias; because the survey was designed to query endoscopic injuries, respondents may have been more likely to have experienced an injury in the past (overestimating the prevalence of injury among endoscopists). Also, injuries definitely or potentially related to endoscopy were classified by self-report. It is possible that some injuries were in fact unrelated to endoscopy (false positives), and others were unknowingly related to endoscopy (false negatives). Nevertheless, this is the largest sample size reported in the medical literature. Our sample size permitted a robust analysis for endoscopist and practice characteristics associated with injury. In conclusion, the present study confirms a high prevalence of work-related injuries among endoscopists. Higher procedure volume (O20 endoscopies per week), more hours per week performing endoscopy
1. O'Sullivan S, Bridge G, Ponich T. Musculoskeletal injuries among ERCP endoscopists in Canada. Can J Gastroenterol 2002;16:369-74. 2. Liberman AS, Shrier I, Gordon PH. Injuries sustained by colorectal surgeons performing colonoscopy. Surg Endosc 2005;19:1606-9. 3. Byun YH, Lee JH, Park MK, et al. Procedure-related musculoskeletal symptoms in gastrointestinal endoscopists in Korea. World J Gastroenterol 2008;14:4359-64. 4. Hansel SL, Crowell MD, Pardi DS, et al. Prevalence and impact of musculoskeletal injury among endoscopists: a controlled pilot study. J Clin Gastroenterol 2009;43:399-404. 5. Kuwabara T, Urabe Y, Hiyama T, et al. Prevalence and impact of musculoskeletal pain in Japanese gastrointestinal endoscopists: a controlled study. World J Gastroenterol 2011;17:1488-93. 6. Buschbacher R. Overuse syndromes among endoscopists. Endoscopy 1994;26:539-44. 7. Everhart JE, Ruhl CE. Burden of digestive diseases in the United States part I: overall and upper gastrointestinal diseases. Gastroenterology 2009;136:376-86. 8. Everhart JE, Ruhl CE. Burden of digestive diseases in the United States part II: lower gastrointestinal diseases. Gastroenterology 2009;136:741-54. 9. Hansel SL, Di Baise JK, Crowell MD, et al. Prevalence and impact of musculoskeletal injury among endoscopists: a controlled study [abstract]. Gastrointest Endosc 2007;65:AB104. 10. Keate RF, Dryden GW, Wang K, et al. Occupational injuries to endoscopists: report from the ASGE Web survey [abstract]. Gastrointest Endosc 2006;63:AB111. 11. Raftopoulos SC, Segarajamingams DS, Yusoff IF, et al; Endoscopy. J Gastroenterol Hepatol 2007;22:A325-43. 12. Shergill AK, McQuaid KR, Rempel D. Ergonomics and GI endoscopy. Gastrointest Endosc 2009;70:145-53. 13. Cappell MS. Colonoscopist’s thumb: DeQuervains’s syndrome (tenosynovitis of the left thumb) associated with overuse during endoscopy. Gastrointest Endosc 2006;64:841-3. 14. Mohankumar D, Garner H, Ruff K, et al. Characterization of right wrist posture during simulated colonoscopy: an application of kinematic analysis to the study of endoscopic maneuvers. Gastrointest Endosc 2014;79:480-9. 15. Shergill AK, Asundi KR, Barr A, et al. Pinch force and forearm-muscle load during routine colonoscopy: a pilot study. Gastrointest Endosc 2009;69:142-6. 16. Pedrosa MC, Farraye FA, Shergill AK, et al. Minimizing occupational hazards in endoscopy: personal protective equipment, radiation safety, and ergonomics. Gastrointest Endosc 2010;72:227-35. 17. ACGIH. TLVs and BEIs: threshold limit values for chemical substances and physical agents and biological exposure indices. Cincinnati (Ohio): Industrial Hygienists, 2008. 18. Haveran LA, Novitsky YW, Czerniach DR, et al. Optimizing laparoscopic task efficiency: the role of camera and monitor positions. Surg Endosc 2007;21:980-4. 19. Matern U, Faist M, Kehl K, et al. Monitor position in laparoscopic surgery. Surg Endosc 2005;19:436-40.
www.giejournal.org
Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 301
Treatment* None
114 (31.5)
Medications
205 (56.6)
Steroid injection
97 (26.8)
Physiotherapy
164 (45.3)
Rest
122 (33.7)
Splinting
82 (22.7)
Surgery
48 (13.3)
*More than 1 modification/treatment was reported by some individuals.
Musculoskeletal injuries related to endoscopy 20. Zehetner J, Kaltenbacher A, Wayand W, et al. Screen height as an ergonomic factor in laparoscopic surgery. Surg Endosc 2006;20: 139-41. 21. Manasnayakorn S, Cuschieri A, Hanna GB. Ergonomic assessment of optimum operating table height for hand-assisted laparoscopic surgery. Surg Endosc 2009;23:783-9.
302 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
Ridtitid et al 22. Berquer R, Smith WD, Davis S. An ergonomic study of the optimum operating table height for laparoscopic surgery. Surg Endosc 2002;16:416-21. 23. van Det MJ, Meijerink WJ, Hoff C, et al. Optimal ergonomics for laparoscopic surgery in minimally invasive surgery suites: a review and guidelines. Surg Endosc 2009;23:1279-85.
www.giejournal.org
Ridtitid et al
Musculoskeletal injuries related to endoscopy
APPENDIX
Supplemental Figure 1. X axis: proportion of time (%) spent performing endoscopies. Y axis: proportion of endoscopists.
www.giejournal.org
Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 302.e1
Musculoskeletal injuries related to endoscopy
Ridtitid et al
SUPPLEMENTAL TABLE 1. A 25-question, self-administered, electronic survey* Question
Answer
1. Sex
O Male O Female
2. Age
____ years
3. Height
____ feet ____ inches
4. Weight
____ pounds
5. Hand dominance 6. What is your activity level?
7. How would you describe your practice?
8. List 2 main avocational activities or hobbies (ie, golf, tennis, video games) 9. How many hours per week do you spend performing them?
O Left O Right O Mildly active O Moderately active O Very active O Small community practice (!5 specialists performing endoscopy) O Large community practice (R5 specialists performing endoscopy) O Small academic practice (University-affiliated group with !5 specialists performing endoscopy) O Large academic practice (university-affiliated group with R5 specialists performing endoscopy) 1)____________________________ 2)____________________________ Activity 1 _______ hours/week Activity 2 _______ hours/week
10. Do you currently perform endoscopy?
O Yes O No
11. Have you ever in your lifetime performed GI endoscopy?
O Yes O No
12. No. of years performing endoscopic procedures 13. Approximate no. of endoscopies performed per week 14. Hours/week usually spent performing endoscopies
O 0-5 O 6-10 O 11-15 O 16-20 O 21-25 O 26-30 O 31-35 O 36-40 O O40 O 0-5 O 6-10 O 11-15 O 16-20 O 21-25 O 26-30 O 31-35 O 36-40 O 41-45 O 46-50 O 51-55 O 56-60 O 61-65 O 66-70 O 71-75 O 76-80 O O80 O 0-5 O 6-10 O 11-15 O 16-20 O 21-25 O 26-30 O 31-35 O 36-40 O O40
15. Endoscopies make up what percentage of your work time?
O 0-10 O 11-20 O 21-30 O 31-40 O 41-50 O 51-60 O 61-70 O 71-80 O 81-90 O 91-100
16. Of the time you spend doing endoscopy, what percentage do you spend doing the following:
EGD O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100 Colonoscopy O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100 ERCP O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100 EUS O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100 Enteroscopy O 0 O 10 O 20 O 30 O 40 O 50 O 60 O 70 O 80 O 90 O 100
302.e2 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
Ridtitid et al
Musculoskeletal injuries related to endoscopy
SUPPLEMENTAL TABLE 1. Continued Question 17. Have you ever experienced injury (pain or numbness) in your neck, back, or upper or lower limbs? 18. Details about injury (click all that apply)
Answer O Yes O No Type of injury: O Thumb pain O Shoulder pain O Hand pain O Neck/upper back pain O Lower back pain O Elbow pain O Hand numbness O Carpal tunnel syndrome O Other Side affected: O Left O Right O Both Was/is pain caused by endoscopy? O Yes O No O Maybe When is/was pain evident? O At work performing endoscopy O At work performing endoscopy/clinic O At work/outside work O Outside work only How much does/did this bother you? O 1 (least) O 2 O 3 O 4 O 5 (most) How long have you had these symptoms? _____ years
19. Have you ever had to take time off from performing endoscopy because of musculoskeletal injury perceived related to endoscopy?
O Yes O No
20. If Yes to above, what is the longest consecutive amount of time you have taken off work because of a musculoskeletal injury perceived to be related to endoscopy?
______ days/months/years
21. Have you ever had to modify your practice and/or shorten your endoscopic case load due to occupational injury?
O Yes O No
22. If you experienced musculoskeletal pain/injury related to endoscopy, what modifications have you made to your endoscopic practice? (click all that apply)
O None O Stretch O Breaks O Adjustable bed O Less endoscopy O Stand on rubber mat O Orthopedic shoes/sneakers
23. What treatment(s) have you received for your condition? (click all that apply)
O None O Medications O Steroid injections O Physiotherapy (P.T.) O Rest O Splinting O Surgery
24. How often do you take breaks?
25. Do you perform any of these environmental modifications during endoscopy? (click all that apply)
O Regularly O Occasionally O Never O None O Height-adjustable examination table O Position of monitor in front at eye level O Stopped helping to move patients after procedures O Sit when you perform colonoscopy
*SurveyMonkey, Palo Alto, Calif.
www.giejournal.org
Volume 81, No. 2 : 2015 GASTROINTESTINAL ENDOSCOPY 302.e3
Musculoskeletal injuries related to endoscopy
Ridtitid et al
SUPPLEMENTAL TABLE 2. Trends over time (1994 results compared with 2013 results) 1994 (n [ 256)
2013 (n [ 684)
47.8 (8.6)
50.8 (11.0)
95.1
88.1
Community
84.2
64.5
Academic center
15.5
35.5
0-20
35
12.4
21-40
43
26.6
41-60
18
35.4
61-80
2
21.1
81-100
0
4.5
0-15
67
35.8
16-30
32
54.7
O30
2
9.5
Percentage of endoscopists who experienced injury related/unrelated to endoscopy
57
71.4
Percentage of endoscopists who experienced injury definitely/possibly related to endoscopy
46
52.9
Neck pain
13
29.3
Thumb pain
19
27.6
Lower back pain
27
18.8
Elbow pain
15
10.5
Shoulder pain
19
10.2
Hand pain
14
10.2
Carpal tunnel syndrome
6
5.8
Hand numbness
12
4.4
None
49.7
31.5
Medications
17.8
56.6
Physiotherapy
16.5
45.3
Not reported
26.8
Rest
8.3
33.7
Splinting
4.5
22.7
Surgery
3.2
13.3
Variable Age, mean ( SD), y Male sex (%) Practice setting (%)
Proportion of time (%) spent performing endoscopy
Hours/week spent performing endoscopy (%)
Percentage of endoscopists with various injuries,* %
Treatment,* %
Steroid injection
*More than 1 injury/treatment was reported by some individuals.
302.e4 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 2 : 2015
www.giejournal.org
