Abstract
Ionising radiation is a modality used in diagnostic and therapeutic medicine. The technology has improved and resulted in lower dose exposure but there has been an escalation in the quantity of procedures, their duration and complexity. These factors have meant increased occupational radiation exposure for interventionalists. Ionising radiation exposure can have detrimental health effects and includes radiation skin burns, various carcinomas, genetic and chromosomal aberrations and cataractogenesis of the lenses of the eye. The lenses of the eye are of the most radiosensitive organs and the risk of cataracts is high despite low radiation dose exposures. The use of personal protective equipment (PPE) is a method that can be used to mitigate the risk for developing lens opacifications. The consistent and effective utilisation of PPE is marred by availability, proper fit and ease of use when performing procedures. Radiation safety training is imperative to enforce a culture of radiation safety among interventionalists. The aim of this study was to quantify and describe cataracts among South African interventionalists and to understand their radiation safety practices. For this purpose, a cross sectional study was designed using multiple methods. A survey was conducted to determine the demographics and the risk factors of doctors exposed to radiation to doctors not exposed. The radiation workload and radiation safety practices of interventionalists were explored. Both groups had slit lamp examinations. The data were analysed analytically and a regression model developed looking at the outcomes and the risk factors. Qualitative in-depth interviews and group interviews were conducted to explore the perceptions of interventionalists regarding radiation safety. Deductive and inductive thematic analysis was done. Interdisciplinary research is challenging but offers tremendous opportunity for exploring and tackling complex issues related to securing a safe radiation work environment.
Export citation and abstract BibTeX RIS
1. Introduction
Ionising radiation, although potentially harmful, is used in many modalities in diagnostic and therapeutic medicine. There have been substantive advancements in the technology resulting in reduced doses for performing procedures [1]. In the last 2–3 decades the number of procedures being carried out by interventionalists has however increased together with their complexity [1]. Despite the technological advances resulting in a reduced dose, the cumulative dose may potentially increase because interventionalists spend more time doing more complex procedures. This results in increased occupational radiation exposure for interventionalists.
The effects of radiation on the body are manifold. They include skin damage, haemopoietic carcinomas, thyroid cancer, brain tumours and ophthalmic changes [2, 3]. The eyes are extremely radiosensitive and can be affected in the following ways: radiation retinopathy, neuropathy, papilopathy and lens changes. The most common changes to the eye are the development of cataracts in the lens [4, 5].
Interventional cardiologists and interventional radiologists are the two categories of medical radiation workers that have the highest radiation dose exposure [6]. This places them at increased risk of developing occupational radiation-associated cataracts [7, 8].
Radiation-associated cataracts may develop due to low dose exposure and until recently were thought to be a deterministic effect [9, 10]. The prevalence of cataracts is higher in radiation healthcare workers compared to the average population [11]. The most common site for radiation-associated cataracts to develop is in the posterior sub-capsular area of the lens [5].
The occupational hazard hierarchy of control is a framework that can be applied to the control of occupational radiation exposure in the interventional catheterisation theatre [12]. The least effective control measure in this framework is the utilisation of personal protective equipment (PPE), but it is an important control measure for radiation safety. PPE has to be used properly and consistently for it to be effective. There is inconsistent and improper use of PPE in many occupational settings. The reasons for this include poor availability, ill-fitting PPE, low or no choice in selecting the PPE and hampering the performance of procedures [13].
Effective radiation control measures in the catheterisation theatre require that a culture of radiation safety be developed. The establishment of this culture hinges on the participation of the entire team that works in the catheterisation theatre and this depends on the norms and attitudes within the workplace [7, 13].
Critical to developing this radiation safety culture is ensuring that radiation safety training is core to the training of the end users [14, 15]. In South Africa radiation physics and radiobiology are part of the core curriculum of radiologists, but this is not the case for the training of electro-physiologists, adult cardiologists and paediatric cardiologists. Changes in the training curriculum of the interventionalists will thus be critical to fostering this culture [5, 16].
There is a need to investigate the occurrence of radiation-associated cataracts and radiation safety practices among interventionalists in South Africa. Literature searches of several databases including PubMed and Medline did not yield similar studies for South Africa (and Africa) and this necessitates further investigation to identify and address these shortfalls.
The aim of this study was to compare the occurrence of cataracts in South African interventionalists occupationally exposed to radiation to a group of non-occupationally exposed doctors, to evaluate the risk factors for cataracts and to look at utilisation of PPE. The study qualitatively explored the knowledge, attitudes and practices of interventionalists to radiation safety. The aim of this article is to describe the methodology for this study.
2. Methods, materials and study design
2.1. Study design
This was a cross sectional study with exposed and non-exposed participants using multiple methods (quantitative and qualitative). A nested case control study compared cases to a control group.
2.2. Population and setting
The exposed participants were interventionalists who regularly perform fluoroscopic work and included adult and paediatric interventional cardiologists, and interventional radiologists. The comparison group comprised unexposed doctors who do not do fluoroscopic work and included general practitioners, family physicians, specialist physicians, surgeons and paediatricians. The exposed and unexposed groups were comparable to each other socially and differed chiefly in their occupational exposure to radiation.
The exposed group was recruited at several specialised national conferences and workshops in South Africa and this ensured representativeness of South Africa because they were from across the country. There are approximately 229 adult cardiologists, 41 paediatric cardiologists and 50 interventional radiologists in South Africa and they work in either the public or private sectors or both. The Health Professions Council of South Africa and the Cardiology and Radiology Societies of South Africa supplied these statistics (February 2016). These highly specialised interventionalists are concentrated in the major cities in South Africa viz. Johannesburg, Cape Town, Pretoria, Durban and Bloemfontein.
The unexposed participants were recruited in Bloemfontein, and incidentally at the conferences where the exposed group were recruited. The researchers approached various clinical departments at the public and private hospitals and practices in the city to recruit the unexposed participants.
In the case control analysis, the cases and controls were matched on age and sex. The cases were all participants diagnosed with lens opacities (exposed and unexposed groups). The controls were frequency matched from the cohort.
2.3. Participant enrolment
Participant recruitment began in May 2015. Qualitative data collection was completed July 2016. Ophthalmological screening will continue until February 2017. There was no randomisation and all participants eligible for enrolment were included. Participants with a history of radiotherapy and penetrating eye trauma were excluded from the study. Participants had to be doctors to be included in the study. The participants enrolled to date are tabulated in table 1.
Table 1. Particiants enrolled into the study thus far.
| Completed questionnaire | Had ophthalmological screening | |
|---|---|---|
| Adult cardiologists | 37 | 35 |
| Paediatric cardiologists | 41 | 23 |
| Radiologists | 48 | 13 |
| Not exposed comparative group | 94 | 32 |
| Participants excluded because they were not doctors e.g. radiographers and scrub nurses | 19 | 3 |
2.4. Quantitative component of the study
2.4.1. Quantitative data collection tools
2.4.1.1. Questionnaire
The self-administered questionnaire (table 2) developed collected information on demographics, medical history, occupational history, and radiation workload exposure and radiation safety practices and training. The questionnaire was developed based on literature and existing questionnaires. An expert panel of radiologists, cardiologists (adult and paediatric), a medical physicist, an ophthalmologist, an occupational medicine specialist and a public health medicine specialist assisted with the development of the questionnaire. Experts assisted with developing the section relevant to their discipline. The questionnaire was reviewed several times by this expert panel and once finalised it was piloted and a final version was developed based on the pilot study. The expert panel approved the final version. A biostatistician reviewed the final questionnaire to ensure all the necessary data would be collected in alignment with the aim of the study and an analysis plan was developed. The final questionnaire was developed in electronic survey format using Evasys® (www.evasys.co.uk) and in hard copy format.
Table 2. Information collected in self-administered questionnaire.
| Demographics |
|---|
| Sex |
| Age and date of birth |
| City working in |
| Medical history |
| Weight |
| Height |
| Chronic diseases: diabetes, hypertension, cancer, myopia, glaucoma, congenital cataracts, obesity, rheumatoid arthritis, autoimmune diseases |
| Eye conditions: cataracts, previous surgery, previous trauma, chronic infections |
| Ionising radiation exposure: CT brain, radiotherapy |
| Non-ionising radiation exposure: outdoor activities, hobbies, use of dark glasses |
| Drug history: oral contraception, steroids |
| Smoking history |
| Alcohol usage |
| Occupational history |
| Capacity worked in |
| Location worked and duration worked |
| Fluoroscopic work? |
| Biplane work? |
| Right or left handed? |
| Vascular access? |
| Radiation workload |
| List of procedures, the mean number of procedures per week, the duration of the procedures, the average number of cine runs and frames per procedure? |
| Radiation safety |
| Use of PPE: suspended screen, lead apron, thyroid shield, and lead glasses/goggles? |
| Barriers to utilisation of PPE: availability of PPE, years utilised, fit of PPE, ease of utilisation? |
| Personal radiation dosimetry: utilisation, dosimetry readings, and knowledge of radiation limits? |
| Radiation safety training: trained in radiation safety as a doctor, trained in the use of x-ray equipment, and trained how to protect self and patients from radiation exposure? |
2.4.1.2. Cataract grading and ophthalmological screening tool
There are several tools available for grading cataracts for this purpose and these include the World Health Organisation Simplified Cataract Grading System (WHOSCGS), the Lens Opacities Classification System, the Wisconsin Cataract Grading System (Wisconsin system) and the Oxford Clinical Cataract Classification System [17–19]. A challenge with various grading systems is that they have different cut off points making comparison difficult [19].
The bases of all these grading systems however is that they describe the anatomy of the cataract, the morphology of it, the colouration and the degree of opacification. The WHOSCGS was developed as a simplified classification system to allow for screening of cataracts by observers with minimal ophthalmological training making it user friendly with good user agreement [17]. The grading is comparable to the other systems as it collects similar information. The grading tool is available on open access, which influenced our decision to use this grading system. The WHOSCGS is also the standard grading system used in the Department of Ophthalmology at the University of the Free State (UFS) and the ophthalmic examiner was au fait with it [17].
The ophthalmological screening tool (table 3) is routinely used in the Department of Ophthalmology at the UFS. This tool collected information on external examination of the eye, the pupil colour, tear break up time, visual acuity, examination of the lens of the eye and the macular. The anatomical position (cortical, nuclear and posterior subcapsular) of all opacities was recorded.
Table 3. Ophthalmological screening tool.
| Name | ||
|---|---|---|
| Date of birth | ||
| Medical history | Y/N | Treatment |
| Diabetes | ||
| Hypertension | ||
| Tuberculosis | ||
| Epilepsy | ||
| Asthma | ||
| Other | ||
| Allergies | ||
| Left eye | Examination | Right eye |
| Visual acuity: | ||
| Without glasses | ||
| With glasses | ||
| Lids and tarsus | ||
| Conjunctiva | ||
| Pupil | ||
| Cornea | ||
| Lens: | ||
| Anterior chamber | ||
| Posterior chamber | ||
| Vitreous | ||
| Disc | ||
| Disc: cup ratio | ||
| Macular | ||
| Vessels | ||
| Periphery | ||
| Muscle movements | ||
| Tear break up time | ||
| Pressure | ||
2.4.2. Quantitative data collection
The scientific meetings where data were collected included an interventional radiology workshop, a paediatric cardiology interventional workshop, and the annual radiology and cardiology societies of South Africa conferences. The organisers of the conferences and workshops were approached beforehand to obtain permission to conduct research at these scientific meetings. We requested that a hyperlink to our online questionnaire be sent to prospective conference and workshop delegates. A letter explaining the study was sent with the hyperlink. Reminders were sent prior to the upcoming scientific meetings. At the scientific meetings the delegates were invited to voluntarily complete the questionnaire online or the hard copy version.
An incentive was provided at the scientific meetings to increase participation in the study. All participants at these meetings were included into a lucky draw for a pair of lead glasses. Appropriate banners were placed at the scientific meetings to increase awareness of the study. An informed research assistant was tasked to lobby delegates to participate in the study.
The screening was done by a senior ophthalmology resident trained by the head of department of the Department of Ophthalmology (UFS). All the screening was done by the same clinician/examiner using the same slit lamp. This improved reliability and repeatability of the findings. All delegates were invited to have a free ophthalmological slit lamp examination to screen for cataracts. They were explained that the examination comprises dilatation of their pupils that could cause minor discomfort and that their pupils could remain dilated for up to 2–3 h. A short acting mydriatic drug was used. Participants were asked about drug allergies prior to administration of the mydriatic medication. Ten minutes after the participants' pupils were dilated they had the ophthalmological examination. Participants were advised to avoid direct sunlight and not to drive for at least 2–3 h post dilatation. The eye findings were linked to the questionnaire completed.
2.4.3. Quantitative sampling strategy
Targeted convenience sampling was used for the quantitative component.
2.4.4. Quantitative data analysis
The data were analysed descriptively and analytically using STATA® 14 (http://stata.com). The demographic data were described. The exposed and unexposed groups were compared. We estimated the odds ratios for various risks between the exposed and unexposed groups. We controlled for age and confounders such as diabetes, myopia, and penetrating trauma to the eye, steroid use and radiation exposure.
We estimated the work related radiation workload using information reported by the participants in the questionnaire and scatter radiation dose estimates. The scatter radiation dose estimates were based on work done by Vano [20]. We estimated the workload adapting the methodology used by Ciraj-Bjelact [11]. We estimated workload exposure (WL) per procedure using the following function:
where
- Y is the number of years a procedure was done, P is the number of that procedure/year, D is the dose/procedure and M is the protection modification factor.
The dose per procedure was based on work previously done in South Africa. We used dose reference levels and dose area products estimated in these studies [21, 22]. The modification factor (M) was calculated taking into consideration the use of the ceiling suspended screen and the protective eyewear and radial access and angulation. A protection modification factor of 0.1 for the use of screens and protective eyewear and 2.0 for radial access were applied [23]. We applied a correction factor of 1.8 for angulation [11, 23]. The total workload estimates were the sum of the workloads of all the different procedures over time.
There are many factors that influence workload radiation exposure and affect the estimates [23]. We recognise that we cannot control for all of these factors, but believe that this estimation will provide us with a good estimate of the relative workloads, which we assume to be approximately proportional to the radiation exposure to the eye.
In the nested case control analysis we frequency matched for age and sex to increase precision of the estimates.
2.5. Qualitative component of the study
2.5.1. Qualitative data collection tools
An interview schedule (table 4) was developed to ensure the researcher collected thick data. The interview schedule was designed to ensure uniformity of the questions asked. The schedule was discussed with several qualitative researchers and suggestions were incorporated and the adaptations included accordingly.
Table 4. Interview schedule.
|
|
|
|
|
|
2.5.2. Qualitative data collection
Qualitative research allows the researcher to explore the depth of a subject and to hear the unique voice of the targeted participants [24]. The participants for the qualitative data collection could also be included as participants for the quantitative component.
The researcher conducted 6 group interviews and 30 in-depth individual interviews. An interview schedule (table 4) was used to guide the discussions. The researcher has experience with qualitative research and conducted the interviews. Data collection was continued until data saturation was reached. We decided that we had reached data saturation when we started getting recurring responses and were finding that no new information was emerging. The interviews were audio recorded and transcribed verbatim. All interviews were conducted only in English, which is the lingua franca for the medical fraternity in South Africa.
2.5.3. Qualitative sampling strategy
The researchers aimed to understand how the participants perceived radiation safety in the workplace and the stakeholders responsible for this safety [25]. Purposive sampling (snowball) was used to recruit participants and key informants were contacted directly [26, 27].
2.5.4. Qualitative data analysis
Data were analysed as we received it. We used Braun and Clarke's steps in the analysis process [28, 29]. The researchers independently read the transcripts and coded the data. The codes were organised into categories and the categories grouped into themes. We used a deductive and inductive approach. We discussed the interpretations that emerged. We debated the themes and then reached consensus on the findings, which will be presented elsewhere. The qualitative software package MAXQDA® 12 (http://maxqda.com) was used to analyse the qualitative data.
3. Ethical considerations
The study was approved by the Human Research Ethics Committee of the Faculty of Health Sciences of the UFS (ECUFS 44/2015). (The certificate is obtainable from ethicsFHS@ufs.ac.za.) Written informed consent was obtained from participants for the qualitative and quantitative components. Informed consent was assumed when participants agreed to proceed with the online questionnaire.
4. Strengths and limitations of this study
This is the first study of its type in South Africa. Similar studies have been done elsewhere in the world [30, 31]. We believe we will make a contribution to the epidemiology of occupational radiation-associated cataracts in the African context. Low participation may however introduce uncertainty and bias [32]. Participation was voluntary in the exposed and unexposed groups, which may have introduced selection bias. The self-administered questionnaire may have introduced bias. The ophthalmic examiner was aware of the exposure status of the participants because the delegates at the conferences came for the specified conference e.g. a cardiologist at the cardiology conference. The workload exposure will be estimated and will not be a direct dose measurement.
5. Discussion
This study was conducted in South Africa, which has a different occupational context to other settings where similar studies were completed. Several similar studies looking at radiation induced occupational cataracts were done in the USA, Europe and South America [31]. The social context of these countries and the ethos that underpins their health sectors are often more homogeneous compared to South Africa which has a two-tiered public and private sector. The equity divide between the public and private health care sectors in South Africa is tremendous [33].
The private sector is well developed while the public health sector is ailing. The private healthcare sector is comparable to any developed country, and characterised by profit driven costs. High patient loads, shortages of medical equipment and poor human resource establishments, burden a failing public health system in South Africa. It is not uncommon that South African doctors work in both systems [33, 34]. The working hours in the public sector are long due to low staff complements while the private sector is market driven. The movement of skilled healthcare professionals from the public to private sectors exacerbates the inequity within the health system.
Radiation safety practices differ between South Africa and other settings, which may affect the prevalence of occupational radiation associated cataracts. In South Africa, especially the public sector where the technology may not be as advanced, there may be a resultant increased workload exposure. This study may help us understand the workloads, doses, and safety practices in the private and public sectors.
Interventionalists' training is long and intensive. In an article on cardiology training in South Africa, Sliwa et al (2016) deliberated that there has been an increase in the burden of cardiac diseases in South Africa without a reciprocal increase in trained cardiologists [35]. This contributes to perpetuating the shortage of highly skilled doctors such as interventionalists, which strengthens the need to ensure a safe work environment. This study will assist us to understand the prevalence of radiation-associated cataracts and radiation safety practices in South Africa. This will assist with developing recommendations for improving this workplace environment in South Africa.
6. Conclusion
The occurrence of radiation-associated cataracts and the radiation safety practices among South African interventionalists are poorly understood. To make sense of the complexity of these issues it was necessary to take an interdisciplinary approach. The skills and knowledge of several disciplines are crucial to define the problem and propose plausible solutions for it. We believe that this study offers a novel way of combining the paradigm of several disciplines and a blend of methodologies to address the research problem and offers solutions relevant to the research environment in which the study was conducted.
Acknowledgements
The PhD from which this study emanated was funded by the Medical Research Council of South Africa under SAMRC Clinician Researcher Programme. AR received the Discovery Foundation Scholarship, which funded the data collection of this project.
Authors' contribution
WR and AR conceptualized the study. AR wrote the protocol with input from WR. PC and WM provided technical input for the study protocol. AR wrote the first draft of this paper. All authors gave input to the article. All authors read and approved the final manuscript. We thank Margaret Ann Sweetlove for assisting with the workload exposure calculation.
Conflict of interest
All authors declare that there are no conflicts of interest.