Use of protective hand shielding by veterinary workers during small animal radiography

Abstract

Federal government guidelines recommend wearing hand shielding that provides full protection for the entire hand during manual restraint of animals for radiography. The primary objective of this cross-sectional survey of 143 Saskatchewan veterinary workers was to describe behaviors of workers who do not follow guidelines for effective hand shielding, and to examine the factors associated with these behaviors. An electronic invitation to complete an online questionnaire was sent to 1261 members of the provincial veterinary medical and veterinary technologist associations. More than half of the workers reported that their hand protection was visible on a radiograph at least once a month, and 1/5 reported visible unshielded body parts at least once a month. More than 1/3 of workers never used shielding that fully enclosed their hands. Use of fully enclosing gloves or mittens was more likely for workers in academic workplaces (P < 0.001).

Introduction

In Canada, federal government guidelines state that a worker restraining an animal for radiography must wear protective gloves that provide protection throughout the glove, including fingers and wrist, and that irradiation by the X-ray beam must be avoided, whether or not gloves are worn (1). These are recommendations only; radiation protection for occupational exposure to X-rays is legislated at the provincial or territorial level. Current Saskatchewan legislation requires that employers ensure that the radiation dose to a worker is as low as reasonably achievable, and that workers use personal protective equipment provided by an employer, although no reference is made in the provincial legislation to the specific type of leaded hand shielding to be used (2,3).

Despite these guidelines and legislation, previous studies have found that Canadian veterinary workers do not consistently use leaded hand protection when restraining animals during small animal radiography (4–6). A 2017 survey of 331 Saskatchewan veterinary workers also reported inconsistent use of leaded gloves; only 6% of workers reported that they always wore gloves on both hands, while 35% of workers reported that they never wore gloves on both hands (7). Of the workers who did wear gloves, 34% reported never wearing gloves that fully enclosed their fingers and hands (7). Leaded hand protection that does not fully enclose the hand includes mittens with single or multiple openings on the palmar surface and hand shields with a fully open palm, and is intended to facilitate manual restraint of animals by allowing workers to grasp patients directly. These types of hand shielding do not provide effective protection from radiation. Most radiation dose to workers is from X-rays scattered from the patient, and therefore any part of the worker that is not shielded from the patient will not be protected (Figure 1) (8). To the authors’ knowledge, there are no published data on the workplace availability and factors affecting use of different types of hand shielding during small animal radiography.

Figure 1.

A digital radiograph of the left hind limb of a fox. The darker gray square (black arrow) shows the primary beam field margins, which coincide with the light field. The worker’s gloved hands and the anatomy of the fox outside the primary beam field are visible mainly due to attenuation of X-rays that have been scattered from the fox’s body. These scattered X-rays are the main source of dose to the hands of a worker.

In addition to use of hand shielding that does not fully protect the fingers and wrist, other behaviors previously described in veterinary workplaces do not adhere to regulatory requirements to keep the dose as low as reasonably achievable. In a 2017 observational study of workers in a veterinary teaching hospital that used motion-triggered video cameras in a radiology room to observe use of personal protective equipment during manual restraint for small animal radiography, workers were observed laying gloves over the top of their hands during exposures, and also placing their gloved hands into the primary beam (6). Gloves laid over the top of hands will not protect a worker from X-rays scattered from the patient, and leaded hand protection does not effectively protect against primary beam radiation, which has a higher average energy than scattered radiation (9). In the authors’ experience, unshielded worker body parts are also included in the radiograph at times. To our knowledge, the frequency of shielded and unshielded body parts in the primary beam has not been described in private veterinary practice.

Hand doses are not monitored for most veterinary workers involved in diagnostic radiography. Whole body doses of ionizing radiation are monitored for some veterinary workers (37% of Canadian veterinarians and veterinary technicians in 2006); however, the whole body dose is usually measured using a dosimeter worn underneath a leaded apron, and the dose to unshielded hands would be expected to be much higher (10). The limit on dose from occupational exposure to ionizing radiation for extremities is based on the estimated amount of radiation that is required to cause skin reaction in 1% of individuals (11). This dose limitation also provides what is judged to be sufficient protection for the skin against cancer. In contrast to the risk of skin reaction, however, the risk of stochastic effects such as cancer has no dose threshold, meaning that with any dose some finite risk exists (11). This is the reason that radiation dose should be kept as low as reasonably achievable, even when worker doses are expected to be well below dose limits. These studies have identified a wide gap between the recommended and actual worker use of leaded hand shielding in veterinary workplaces, and the aim of this cross-sectional survey study was to answer questions raised by these previous studies (6,7).

The first objective of this study was to determine availability and use of different types of hand shielding, and factors that affect worker use of these types of hand shielding. The second objective was to describe the frequency of behaviors, such as placing shielded or unshielded body parts into the primary beam, which would result in higher worker dose. Finally, we planned to determine the most important reasons individual workers decide not to wear leaded hand shielding.

Materials and methods

The study protocol was submitted to the University of Saskatchewan’s Behavioral Research Ethics Board and was determined to be exempt from review as per Article 2.5 of the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans, December 2014 (BEH 16-127).

An electronic invitation to complete the online questionnaire was sent to all members of the Saskatchewan Veterinary Medical Association with an electronic mail address (750 veterinarians), and to all members of the Saskatchewan Association of Veterinary Technologists (458 veterinary technologists, 89 veterinary clinics, and 53 student members). Membership in these professional associations is mandatory for all veterinarians and technologists practicing in the province. An initial invitation to complete the questionnaire and a second invitation 10 d later were sent by electronic mail. Recipients were asked to share the electronic survey link with any workers involved in taking radiographs at their clinic who were not veterinarians or veterinary technologists.

The questionnaire was developed by 4 authors (MM, NK, TG, CW). The questionnaire first asked workers if they had been involved in taking a radiograph of a small animal (including companion animals, exotics, and wildlife) in the last year, and only workers who answered yes were asked to complete the remainder of the questionnaire. Workers were asked how many X-ray exposures on average they were involved in acquiring in 1 month, as well as for how many X-ray exposures on average the glove(s) they wore or an unshielded body part was visible on the radiograph in 1 month. They were also asked if leaded hand shielding was available at their practice, and what type of leaded hand shielding was available (gloves that fully enclose hands and fingers, mittens that fully enclose hands and fingers, mittens with a slit or small opening in the palm, hand shields with a fully open palm, other) (Figure 2).

Figure 2.

Examples of leaded hand shielding used by veterinary workers. A — a glove that fully encloses the hand; B — a mitten with openings in the palm; and C — a hand shield with a fully open palm.

Workers were then asked how often they wore different types of hand shielding (gloves or mittens that fully enclose hands and fingers, mittens with a slit or small opening in the palm, hand shields with a fully open palm, other) and how often they used no hand shielding.

Workers were also asked to describe their knowledge of the risks of ionizing radiation, and to select the 3 most important reasons they sometimes did not use hand shielding when taking an X-ray, with reason 1 being most important. Seven reasons were available: i) gloves interfere with restraint of the animal; ii) I am not concerned about adverse health effects; iii) gloves are unhygienic; iv) gloves are not required by my employer; v) gloves do not fit properly; vi) not enough gloves for all workers; and vii) my coworkers do not wear gloves. The available reasons were developed from open text suggestions to increase use of leaded gloves made by veterinary workers in 2 previous studies (6,7).

The unconditional associations between potential risk factors of interest and i) the percentage of time fully enclosing gloves or mittens with no openings were used (for 61 workers with access to both fully enclosing and other types of hand shielding); and ii) the percentage of time any type of hand shielding was used (all workers with access to hand shielding) was examined with linear regression as these values were reported on the survey as a continuous outcome. Model residuals were graphed and checked for homogeneity of variance and normality. A multivariable model was built using stepwise, manual backwards elimination. Only variables that were significant independent risk factors (P < 0.05) were retained in the model. Age, gender, and worker position, and practice type were evaluated as potential confounders. As this was an exploratory analysis no interactions were examined. Categorical risk factors considered included: how many X-ray exposures the worker acquired per month, employer requirement that leaded hand shielding be worn, worker knowledge of the risks of exposure to ionizing radiation, practice type, and worker position, age and gender. Employer requirement that leaded hand shielding be worn was examined only for the percentage of time any type of hand shielding was used.

The association between the type of hand shielding workers had access to (only fully enclosing hand shielding or both fully enclosing and other types) and the percentage of time no hand shielding was used was examined with linear regression.

Results

Fifteen percent (183/1261) of workers who were sent an e-mail invitation completed a questionnaire. The overall response rate is an approximation as the number of non-veterinarian, non-technologist workers who were forwarded the questionnaire is unknown, and the questionnaire was sent to 89 veterinary clinic electronic mail addresses in addition to the individual members of the 2 professional associations.

Of the workers who completed a questionnaire, 22% (40/183) had not been involved in taking a radiograph of a small animal in the last year, and their questionnaire was terminated after the first question. Seventy-eight percent (143/183) of workers who completed a questionnaire had been involved in taking a radiograph of a small animal in the last year, and these workers completed the remainder of the questionnaire. Characteristics of these 143 workers are described in Table 1. Not all respondents completed every question and this is reflected in the denominator of the reported fractions.

Table 1.

Characteristics of Saskatchewan veterinary workers involved in taking small animal radiographs who completed a questionnaire on radiation safety practices (n = 143).

Practice type
Academic institution	10% (15/143)
Small animal exclusive private	48% (68/143)
Mixed animal private	38% (55/143)
Other	3% (5/143)
Position
Veterinarian	31% (44/143)
Technologist	63% (90/143)
Veterinary student	0
Technologist student	< 1% (1/143)
Other	6% (8/143)
Age (years)
Greater than 65	< 1% (1/143)
45 to 65	20% (28/143)
25 to 44	66% (94/143)
18 to 24	14% (20/143)
< 18	0
Gender
Female	94% (134/142)
Male	6% (8/142)

Thirty-eight percent (55/143) of workers were involved in acquiring 10 or less X-ray exposures of small animals a month, 38% (55/143) 11–30 X-ray exposures a month, 11% (16/143) 31–50 X-ray exposures a month, 8% (12/143) 51–70 X-ray exposures a month, and 3% (5/143) > 70 X-ray exposures a month.

Ninety-eight percent (140/143) of workers reported that leaded gloves were available at their practice, 1% (2/143) did not have leaded gloves available, and 1% (1/143) did not know if leaded gloves were available. Seventy percent (97/139) of workers reported that hand protection that did not fully enclose the hand was available at their workplace; mittens with a slit or small opening in the palm were available to 45% (63/139) of workers, and hand shields with a fully open palm were available to 32% (44/139) of workers. Twenty-seven percent (37/139) of workers reported that the only hand protection available at their workplace were types that did not fully enclose the hand.

Excluding workers who did not have leaded gloves available and workers who were the employer, 47% (62/133) of workers reported that their employer required them to wear leaded gloves when taking radiographs of an animal, 50% (67/133) reported that their employer did not require them to wear leaded gloves, and 3% (4/133) did not know if their employer required them to wear leaded gloves.

Reported use of different types of leaded hand shielding is presented in Table 2. Twenty-eight percent (38/137) of workers used mittens with a slit or small opening in the palm at least 50% of the time, and 14% (19/137) used hand shields with an open palm at least 50% of the time. Thirty-six percent (49/137) of workers never used gloves that fully enclosed the hand and fingers.

Table 2.

Types of leaded hand shielding used by Saskatchewan veterinary workers during small animal radiography.

	Always	> 75%	50% to 75%	> 50%	Never
Gloves or mittens that fully enclose hands with no openings	29% (40/137)	7% (10/137)	11% (15/137)	17% (23/137)	36% (49/137)
Mittens with a slit opening	15% (21/137)	4% (6/137)	8% (11/137)	11% (15/137)	61% (84/137)
Hand shields with open palm	6% (8/137)	6% (8/137)	2% (3/137)	10% (14/137)	76% (104/137)
Use no gloves	9% (12/140)	26% (36/140)	16% (22/140)	13% (18/140)	37% (52/140)

The number of radiographs, on average, on which workers reported visible leaded gloves or unshielded body parts are presented in Table 3.

Table 3.

Report by Saskatchewan veterinary workers of average number of radiographs per month on which lead gloves and unshielded body parts were visible during small animal radiography.

Number of radiographs/month	Lead glove visible	Body part visible
< 1	69/141 (49%)	114/143 (80%)
1 to 5	54/141 (38%)	26/143 (18%)
6 to 10	11/141 (8%)	1/143 (1%)
11 to 15	2/141 (1%)	0/143 (0%)
16 to 20	2/141 (1%)	0/143 (0%)
> 20	3/141 (2%)	2/143 (1%)

Eleven percent (16/143) of workers described their knowledge of the risks of exposure to ionizing radiation as poor, 29% (41/143) described their knowledge as fair, 46% (66/143) described their knowledge as good, and 14% (20/143) described their knowledge as excellent.

Factors associated with use of fully enclosing hand shielding

In the final multivariable model, use of fully enclosing gloves or mittens was significantly more likely for workers in academic workplaces than in small animal or mixed private practice (P < 0.001) (Table 4). Risk factors that were not associated with the use of fully enclosing gloves or mittens included the number of exposures acquired per month (P = 0.80), a knowledge of risks of ionizing radiation (P = 0.31), position (P = 0.80), age (P = 0.07), and gender (P = 0.20).

Table 4.

Unconditional associations between significant risk factors of interest and whether fully enclosing hand shielding was used by veterinary workers (n = 61).

	Relative frequency of workers in category	Mean	95% CI	P-value
Practice type
Academic institution	0.11 (7/61)	0.95	0.69 to 1.21	reference
Small animal exclusive private	0.34 (21/61)	0.41	0.25 to 0.56	0.001
Mixed animal private	0.49 (30/61)	0.35	0.22 to 0.49	< 0.001
Other	0.05 (3/61)	0.77	0.37 to 1.16	0.45

CI — Confidence interval

Factors associated with use of any type of hand shielding

In unconditional analysis, risk factors that were significantly associated with use of any type of hand shielding included employer requirement that hand shielding be worn (P < 0.001), knowledge of risks of ionizing radiation (P = 0.009), practice type (P < 0.001), and age (P = 0.02). Risk factors that were not associated with the use of hand shielding included the number of exposures acquired per month (P = 0.46), position (P = 0.44), and gender (P = 0.33).

The risk factors that were significantly associated with use of any type of hand shielding in the final multivariable analysis are presented in Table 5. Practice type was included in the multivariable analysis and was a significant risk factor (P = 0.005). However, significant differences were found only between practice type “other” and the other categories. For all respondents selecting “other,” the practice type consisted of > 1 of academic, small, and mixed private practice; therefore, it was felt that this finding was not important.

Table 5.

Final multivariable model of the associations between significant risk factors of interest and whether any type of hand shielding was used by veterinary workers.

	Relative frequency of workers in category	Mean	95% CI	P-value
Employer requiresa
Yes	44% (62/140)	0.89	0.82 to 0.95	reference
No	48% (67/140)	0.27	0.21 to 0.33	< 0.001
I don’t know	3% (4/140)	0.45	0.19 to 0.70	0.003
I am the employer	5% (7/140)	0.48	0.29 to 0.67	< 0.001
Knowledge of risksa
Poor	11% (16/143)	0.51	0.38 to 0.64	0.761
Fair	29% (41/143)	0.49	0.41 to 0.56	reference
Good	46% (66/143)	0.58	0.52 to 0.65	0.058
Excellent	14% (20/143)	0.67	0.56 to 0.79	0.006

^aResults adjusted for confounding by practice type.

CI — Confidence interval.

There was no difference in the use of any type of hand shielding between workers with access to only fully enclosing hand shielding and workers with access to both fully enclosing and other types of hand shielding (P = 0.86).

The most important reasons selected by workers for not wearing leaded gloves when taking a radiograph are presented in Table 6. Ninety-three percent (113/122) of workers selected “gloves interfere with restraint of the animal” as the most important reason they do not wear gloves.

Table 6.

Three most important reasons (with reason 1 being most important) Saskatchewan veterinary workers do not wear leaded gloves when taking a radiograph.^a

	Number of workers selecting the reason
Reason 1 (n = 122 respondents)	Gloves interfere with restraint of the animal (n = 113)
Reason 2 (n = 89 respondents)	Gloves do not fit properly (n = 43) Gloves interfere with restraint of the animal (n = 16)
Reason 3 (n = 83 respondents)	Gloves are not required by my employer (n = 22) My coworkers do not wear gloves (n = 18) Not enough gloves for all workers (n = 12) Gloves interfere with restraint of the animal (n = 11)

^aOnly reasons selected by > 10 respondents are presented.

Discussion

Health Canada and the National Council on Radiation Protection and Measurements have developed guidelines for radiation protection in veterinary medicine (1,12). While both organizations recommend that veterinary workers always use gloves that fully enclose the hand when holding patients during radiographic imaging, our results demonstrate that this does not happen consistently in clinical practice. When workers do use hand protection, more than 1/3 never use shielding that fully encloses the hand. Forty percent of workers use hand shields with an open palm, or mittens with an opening for fingers, at least half of the time. Mittens with an opening to allow fingers to grasp the patient are marketed to veterinary employers to facilitate manual restraint. However, if portions of the hand or fingers extend through openings during the exposure, they will not be shielded from scattered radiation. In order for mittens with an opening to provide full protection, the worker would need to pull the portion of the animal that they are holding with their fingers through the opening back into the mitten. This could work for the lower portion of a leg or the foot of a smaller sized animal; however, in the authors’ experience, workers do not use mittens with openings in this manner.

Of workers with access to both fully enclosing hand shielding and hand shielding with openings, workers at an academic institution reported using fully enclosing hand shielding more often than workers in small animal and mixed animal private practice. It is possible that this difference was due to the type of hand shielding that is readily available to academic workers. For example, at the veterinary teaching hospital in Saskatchewan, radiology technologists store hand shielding with openings out of sight of workers in an attempt to limit use, while fully enclosing hand shielding is stored in open view next to the radiology table.

While a greater knowledge of risks of ionizing radiation was associated with more frequent use of any type of hand shielding, knowledge of risks was not associated with use of fully enclosing gloves or mittens. It may be that workers with more knowledge understand the reduction in dose that leaded shielding provides, but are unaware that the main source of dose is scattered X-rays from the patient, and that the level of protection is therefore not equivalent between the different types of hand shielding. If multiple types of hand shielding are made available by their employer, workers may infer that all types provide equal protection, and there is no impetus to wear fully enclosing hand shielding. In addition, veterinary supply companies offer hand shielding with openings specifically for use with veterinary patients, and this may suggest efficacy to workers.

An obvious recommendation to decrease use of hand shielding that does not provide full protection to fingers and wrist would be for employers to only make fully enclosing gloves or mittens available to employees. However, in contrast to hand shielding with openings for fingers, fully enclosing gloves or mittens do not allow workers to use their fingers to grasp patients, and workers in previous studies have identified interference with restraint as a reason they choose not to wear gloves (6,7). For this reason, we examined whether workers with access to only fully enclosing gloves or mittens were less likely to wear hand shielding than workers who also had access to hand shielding that did not fully enclose the hand. Our results showed no difference in use of hand shielding between these 2 groups, and we therefore recommend that employers make only fully enclosing gloves or mittens readily accessible to workers.

Half of veterinary workers reported that the glove they were wearing was visible on at least 1 radiograph a month, and 1 in 20 workers reported that a glove was visible on > 11 radiographs a month. Leaded apparel is designed to protect workers from the energies composing scattered radiation; the primary beam has a higher average energy and the effectiveness of protection is much lower (9). One-fifth of workers reported unshielded body parts visible in the field at least once a month, while 2 workers reported unshielded body parts visible in the field > 20 times a month. Our findings may underestimate the true frequency of gloves and unshielded body parts on radiographs as not all workers view the radiographs for which they restrain animals, and therefore may be unaware that gloves or body parts are visible. A study using a canine cadaver and an anthropomorphic phantom found that the dose within the primary beam was on the order of 100 times higher than the scattered radiation dose measured at the position of the restrainer’s hands (13). We therefore recommend that both unshielded and shielded hands never be placed in the primary beam. Employers are legally responsible for this safety behavior; an employer must ensure that the dose to a worker is kept as low as reasonably achievable (3). If a worker is still likely to engage in this practice, we recommend that a ring dosimeter be worn to monitor dose.

Based on our findings, employers can increase worker use of any type of hand shielding by requiring that hand shielding be worn when manual restraint is used for radiographs, and by educating their employees on the risks of exposure to ionizing radiation. Employers should ensure that all workers are aware that hand shielding is required, that shielding with openings does not provide equal protection to shielding that fully encloses the hand, and that hands must not be placed in the primary beam even when shielding is worn. The probability of harmful effects caused by exposure to ionizing radiation associated with diagnostic radiography has been described for veterinary workers, and educational materials designed specifically for veterinary workers are available to employers (14–16).

Workers identified hand shielding interfering with restraint of an animal as the most important reason they chose to sometimes not wear hand shielding. This is consistent with the findings of a study carried out in 2015 in a veterinary teaching hospital in which 73% of workers reported that glove use would increase if they could use less cumbersome gloves that would allow them to grasp and position animals more easily (6). The lack of flexibility is due to the minimum lead attenuation equivalent that is required to reduce the percentage of X-rays passing through the material. Reduction in the lead attenuation equivalent could increase flexibility; however, the percentage of X-rays reaching the worker’s hands would increase. The second most important reason that workers identified for not wearing hand shielding was fit. In the past, personal protective equipment has been sized based on measurements for men, and this equipment may not provide a good fit for women, who on average have smaller and narrower hands than men (17). Employers can address this reason by providing leaded gloves or mittens in different sizes for workers, and ensuring that leaded gloves or mittens are available that fit all employees.

In addition to unwieldiness and poor fit, workers reported that their decision to use hand shielding was influenced by their coworkers, suggesting that modelling of appropriate radiation safety practices by employers may increase employee’s use of hand shielding. As well, employers should ensure that adequate numbers of gloves or mittens are available, as some workers reported that there were not enough gloves or mittens for all workers.

A limitation of this study is that an exact response rate cannot be reported, as we asked recipients of the e-mailed questionnaire to share it with other workers who were not technologists or veterinarians. There is also the potential for reporting bias, or selective suppression of information about behaviors that do not follow radiation safety guidelines by respondents. This could result in an underestimation of behaviors that increase dose to workers.

In conclusion, less than 1/3 of veterinary workers who do use hand shielding during small animal radiography always use fully enclosing gloves or mittens. Hand shielding that does not fully enclose the hands does not provide effective protection against scatter radiation, and therefore we recommend that employers supply only fully enclosing hand shielding to veterinary workers. Employers can increase the use of any type of hand shielding by requiring and modelling use, by educating workers on the risks of ionizing radiation exposure, and by providing well-fitting gloves or mittens for all employees. CVJ