Diagnostic Radiographers' Application of the Radiation Protection Principles of Justification, Optimisation and Dose Limits in Mpumalanga Province, South Africa

ABSTRACT

Introduction

The effective application of radiation protection principles is essential for minimising unnecessary radiation exposure and reducing associated risks to patients, personnel, and the public. This study investigated the application of radiation protection principles by diagnostic radiographers in Mpumalanga province, South Africa.

Methods

The quantitative cross‐sectional study employed stratified random sampling and an online questionnaire to collect data from diagnostic radiographers working in Mpumalanga hospitals. The data were analysed using descriptive and inferential statistics.

Results

The survey achieved a response rate of 85.6% (n = 125, N = 146). Most participants (38.4%, n = 48) had 0–5 years of work experience. Referral forms often lacked sufficient clinical history (64.8%, n = 81), with diagnostic radiographers frequently supplementing by assessing the body region of concern (49.2%, n = 60). Lack of communication between radiologists and diagnostic radiographers was noted when referrals were incomplete (31.3%, n = 36) or when requested examinations did not correspond with the clinical history (32.8%, n = 38). Most participants (66.4%, n = 81) felt undervalued due to hierarchical structures. Most X‐ray rooms had functional warning lights (80%, n = 84). Full lead shields (98.1%, n = 102) were widely accessible, and lead goggles (33.7%, n = 34) were the least accessible. Although most participants were issued with personal radiation monitoring devices (97.1%, n = 99), only 53.5% (n = 53) reported consistent use, and 33.3% (n = 32) received radiation dose readings.

Conclusion

The application of radiation protection principles was inconsistent. Justification was achieved through referral assessment but was compromised by systemic barriers. Optimisation demonstrated suitable room designs but was limited by resource accessibility. Dose limitation was hindered by device usage and feedback despite high issuance.

Summary

This study examined the application of justification, optimisation and dose limits among diagnostic radiographers in Mpumalanga, South Africa. Among the 125 participants, communication gaps and hierarchical barriers limited justification, resource constraints affected optimisation, and inconsistent monitoring and feedback undermined dose limitation. Stronger interprofessional communication, training and monitoring practices are needed to strengthen radiation safety.

The graphic abstract was created with the assistance of the AI tool Claude (Accessed August 2026), using data from the study. The authors reviewed and verified the accuracy of all data and figures presented in the graphic abstract prior to submission.

1. Introduction

The use of ionising radiation plays a critical role in medicine, enabling accurate diagnosis, guiding treatment and follow‐up across a wide range of clinical conditions. Ionising radiation accounts for the vast majority of radiation from artificial sources. The extensive use of ionising radiation has potential health risks if not properly regulated. It has been proven that high doses of radiation and non‐threshold dose values can result in harmful effects characterised as stochastic or deterministic [1, 2].

International and national radiation protection frameworks were established to mitigate these risks [1, 3, 4, 5]. Radiation protection is the safeguarding of personnel, patients, and the public against the adverse effects of radiation and the measures applied to achieve this protection [6, 7]. The International Commission on Radiological Protection (ICRP) stipulated three key principles of radiation protection: Justification, optimisation and dose limits [1]. The principle of justification and optimisation focuses on exposed individuals, whereas the principle of dose limits focuses on occupational and public exposures [8]. The justification principle states that any decision that alters a radiation exposure situation should do more good than harm [1]. Justification is applied at three levels: Level one ensures that the medical use of radiation benefits society more than it harms; level two evaluates whether a specific radiology examination will improve diagnosis, treatment, or provide essential information; and level three requires that an individual's radiology examination be justified [1]. The optimisation principle aims to minimise radiation exposure by reducing the likelihood of exposure, limiting the number of individuals exposed, and keeping doses As Low as Reasonably Achievable (ALARA). Optimisation is applied at two levels: Level one involves the design, selection, and construction of equipment and facilities, while level two focuses on the daily practices of diagnostic radiographers [1]. The dose limit principle states that the total dose to occupational workers and the general public from all regulated sources, in planned situations, should not exceed the set limits. This principle does not apply to medical exposure of patients undergoing a radiology examination [1].

In South Africa, diagnostic radiographers must be registered with the Health Professions Council of South Africa (HPCSA), under the Professional Board for Radiography and Clinical Technology, before they may practice. Registration requires completion of an accredited qualification [9]. Qualifications vary among diagnostic radiographers due to the historic curriculum transition from national diplomas and three‐year bachelor's degrees to a four‐year bachelor's degree to align with international standards [9, 10]. Registration and scope‐of‐practice requirements are standardised nationally through the HPCSA. Diagnostic radiographers' responsibilities include a full patient journey through an examination, from correctly identifying patients, explaining procedures, assessing and validating referral forms for clinical justification, and selecting appropriate protocols based on the patient's clinical history. Moreover, diagnostic radiographers perform radiographic examinations to produce diagnostic images, evaluate for technical and medico‐legal adequacy, and attend to patients' well‐being throughout the process, while collaborating with radiologists and other specialists. Diagnostic radiographers' role also extends to maintaining accurate documentation and confidentiality, and to upholding health, safety and radiation protection standards, including regular equipment checks and quality control testing, which directly relates to the radiation protection principles [11].

Diagnostic radiographers are trained to operate a range of imaging equipment while adhering to the strict radiation protection principles, consequently facilitating the diagnostic use of radiation with minimal harm [1, 11, 12]. The application of radiation protection by diagnostic radiographers involves key strategies to ensure the safety of patients, personnel, and the public. This includes vetting of referrals to ensure radiology examinations are justified, maintaining effective communication with clinicians and patients, implementing quality assurance (QA) programmes, record and safekeeping of patient radiation doses, optimising exposure technique factors, utilising lead shielding apparel for personnel and patients, dose monitoring for radiation personnel, ensuring structural designs comply with regulatory standards, and appropriate placement of radiation and pregnancy warning signs [13, 14, 15, 16, 17, 18, 19]. When effectively applied, these measures protect patients, personnel, and the public from unnecessary radiation exposure.

Despite the well‐established principles, inconsistent implementation of radiation protection remains a concern globally and within South Africa, where challenges such as underfunding, lack of equipment, management inefficiencies, and insufficient training persist [20, 21, 22, 23, 24, 25]. Noncompliance with radiation protection not only endangers patients, personnel, and the public but may also have ethical and legal implications for radiology departments [26]. Therefore, diagnostic radiographers serve as gatekeepers, protecting personnel, patients, and the public from unnecessary radiation and upholding radiation safety standards [27, 28].

South Africa's healthcare system is divided into two sectors: Public and private. In Mpumalanga, the public sector is structured into four tiers: Primary Health Care, District Hospitals, Regional Hospitals, and Tertiary Hospitals, offering services ranging from basic care to specialised interventions [29]. However, primary health care facilities in the province do not have radiology departments [30]. The private sector offers a wide range of services at a higher cost, making it accessible to a minority of the population. Radiology services available in Mpumalanga province include general X‐ray, interventional radiology, mammography, bone densitometry, Computed Tomography (CT), and Magnetic Resonance Imaging (MRI) [31].

Mpumalanga province is ranked as the third most rural province in South Africa and has a history of underperformance in healthcare service delivery, which may have an impact on the consistent implementation of radiation safety practices [32]. There is limited research on radiation protection principles in Mpumalanga, making the province a relevant setting for research. Moreover, the collective application of the three key principles of radiation protection among diagnostic radiographers remains unexplored in the South African context, representing a gap in the literature. Therefore, this study aimed to determine the application of radiation protection principles by diagnostic radiographers in Mpumalanga, South Africa. Understanding their role and compliance is essential for targeted interventions to improve radiation safety.

2. Methods

2.1. Research Design

This study followed a quantitative approach with a descriptive and cross‐sectional design using an online survey (refer to supporting information).

2.2. Research Setting

The study was set in 37 radiology departments encompassing both private and public hospitals within Mpumalanga province.

2.3. Population and Sampling

The estimated target population comprised 230 diagnostic radiographers working in 37 hospitals across Mpumalanga province. The target population number was obtained by contacting the heads of departments (HODs) of the respective hospitals. Stratified random sampling was used to select hospitals. The hospitals were stratified into four strata based on hospital type: Tertiary, regional, district, and private. Mpumalanga's public sector comprises three hospital tiers (tertiary, regional, and district), with private hospitals forming the fourth category; stratifying by hospital type therefore ensured the sample proportionally represented all diagnostic radiographers working in hospitals [29, 33]. Twenty‐three hospitals (two tertiary, two regional, 14 district, five private) were selected from the strata using simple random sampling through the RAND function on Microsoft Excel, with hospitals with the smallest random numbers selected. All the diagnostic radiographers in the selected hospitals were sampled, resulting in a sample size of 146, calculated using Yamane's formula: 𝑛 = 𝑁/1+𝑁(ⅇ)² [34] with a 95% confidence interval and a 5% margin of error. The inclusion criteria were diagnostic radiographers registered with the HPCSA and working in hospitals within Mpumalanga province. The exclusion criteria included diagnostic radiographers working in clinics or private practices and student radiographers.

2.4. Data Collection Process

The study poster containing the link to the study information letter and consent form was shared with the HODs of the selected hospitals via email for distribution to diagnostic radiographers after ethical clearance and gatekeeper permission were sought. Prospective participants clicked on the consent form link to view the consent form, and those who consented to participate were directed to the online questionnaire administered through QuestionPro. The questionnaire was developed and adapted from previous studies [23, 35, 36, 37]. Permission to use the questionnaire was obtained from the authors [23, 35, 36, 37]. The questionnaire consisted of four sections: (i) demographic data, (ii) justification, (iii) optimisation, (iv) dose limits. Data collection took place from 26 January 2024 to 31 May 2024; due to a low response rate, the invitation poster was regularly re‐sent to the HODs as a reminder.

The validity of the study was assessed through content validity. Formal quantitative content validity indices (I‐CVI) were not calculated; instead, content validity was assessed qualitatively through expert review by seven medical imaging academics from various tertiary institutions in South Africa. The reliability of the study was determined through Cronbach's alpha on multi‐item scales to assess internal consistency. Seven of the ten scales demonstrated acceptable internal consistency of α ≥ 0.7. As Cronbach's alpha is sensitive to scales with fewer than ten items, the inter‐item correlation was used to assess internal consistency for the three remaining scales with α < 0.7. The inter‐item correlation values for the three scales were all above 0.2, indicating good internal consistency [38], as shown in Table 1.

TABLE 1.

Cronbach's alpha values and inter‐item correlation means.

Section B	Factors	Cronbach's alpha	Inter‐item correlation means
Question 7	Factor 1: Hindrances to assessing referral forms (items B7.4, B7.3, B7.5, B7.6 & B7.1)	0.784
	Factor 2: Challenges to the application of justification (items B7.9, B7.8 & B7.10)	0.624	0.369
Question 8	Factor 1: Key factors in optimising radiographic imaging (items B8.9, B8.7, B8.10 & B8.8)	0.831
	Factor 2: Prioritising patient well‐being and appropriate modality selection in imaging exams (items B8.5, B8.4 & B8.6)	0.780
	Factor 3: Essential pre‐examination preparation steps (items B8.2 & B8.1)	0.574	0.409
Question 9	Factor 1: Fostering collaboration and responsibility in radiography practices (items B9.3, B9.1 & B9.2)	0.899
	Factor 2: Enhancing efficiency and patient care in radiology practices (B9.5, B9.6 & B9.4)	0.632	0.397
Question 11	Factor 1: Adherence to referral guidelines (items B11.2, B11.3, B11.4 & B11.1)	0.933
Question 12	Factor 1: Medical dominance (B12.4, B12.2, B12.6, B12.5, B12.3 & B12.1)	0.912
Question 13	Factor 1: Private radiology practice (B13.3, B13.2, B13.4, B13.5 & B13.1)	0.865

Note: Inter‐item correlation is only applicable for scales with fewer than 10 items where Cronbach's alpha is < 0.7.

Data were statistically analysed using IBM SPSS v29, with the assistance of an expert statistician. Descriptive statistical analyses included frequencies, means, and standard deviations. Inferential statistical analyses included Chi‐square tests with Cramer's V, t‐tests, Fisher's exact test, and Pearson correlation, with significance set at p‐value < 0.05.

2.5. Ethical Considerations

Ethical clearance was obtained from a higher education Research Ethics Committee (REC‐2289‐2023). Gatekeeper permissions were obtained from the National Health Research Database for public hospitals and from each radiology practice for private hospitals. Participation was voluntary, with electronic informed consent obtained, and participation was anonymous.

3. Results

A total of 125 participants responded to the questionnaire, representing an overall response rate of 85.6% (n = 125, N = 146). Of these, 101 participants completed all sections of the questionnaire. The number of responses (N) varied across the questionnaire sections, as some participants did not complete all questions, resulting in a decline from 125 to 101. Most participants reported having a radiography degree (78.4%, n = 98, N = 125), having less than 10 years of experience as diagnostic radiographers (68.8%, n = 86, N = 125), and working in the public sector (68.8%, n = 86, N = 125). Almost all the participants reported working with general X‐ray (99.2%, n = 124, N = 125), while 61.6% (n = 77, N = 125) reported seeing over 500 patients per month. Just over half of the participants (50.8%, n = 63, N = 124) reported that they did not have a radiologist on site, and most participants reported using digital imaging systems, as shown in Table 2.

TABLE 2.

Demographic data.

Demographic	Options	%(n)
Qualification/level of education	National Diploma	21.6% (n = 27, N = 125)
	Bachelor (3 years)	29.6% (n = 37, N = 125)
	Bachelor (4 years)	28.0% (n = 35, N = 125)
	Honours/Bachelor of Technology (B‐Tech)	20.8% (n = 26, N = 125)
Number of years of experience as a diagnostic radiographer	0–5 years	38.4% (n = 48, N = 125)
	6–10 years	30.4% (n = 38, N = 125)
	11–15 years	16.0% (n = 20, N = 125)
	16–20 years	7.2% (n = 9, N = 125)
	21–25 years	3.2% (n = 4, N = 125)
	26+ years	4.8% (n = 6, N = 125)
District	Ehlanzeni	44.8% (n = 47, N = 105)
	Gert Sibande	23.8% (n = 25, N = 105)
	Nkangala	31.4% (n = 33, N = 105)
Type of hospital	Public: District hospital	37.6% (n = 47, N = 125)
	Public: Regional hospital	16.8% (n = 21, N = 125)
	Public: Tertiary hospital	14.4% (n = 18, N = 125)
	Private hospital	31.2% (n = 39, N = 125)
What is the average number of patients seen in your department per month	Less than 100	4.8% (n = 6, N = 125)
	100–200	11.2% (n = 14, N = 125)
	201–300	8.0% (n = 10, N = 125)
	301–400	8.0% (n = 10, N = 125)
	401–500	6.4% (n = 8, N = 125)
	Above 500	61.6% (n = 77, N = 125)
Which type of imaging modalities do you work with?	General X‐ray	99.2% (n = 124, N = 125)
	CT scan	56.8% (n = 71, N = 125)
	Fluoroscopy	52.8% (n = 66, N = 125)
	Interventional/angiography	10.4% (n = 13, N = 125)
	Mammography	24.0% (n = 30, N = 125)
	Theatre	48.8% (n = 61, N = 125)
	Bone densitometry	24.0% (n = 30, N = 125)
	Cardiac catheterisation	4.0% (n = 5, N = 125)
	Dental	41.6% (n = 52, N = 125)
Which imaging system are you using?	Analogue (conventional) system	14.4% (n = 18, N = 124)
	Digital system: CR	52.0% (n = 65, N = 124)
	Digital system: Digital Radiography (DR)	72.8% (n = 91, N = 124)
Does your department have a radiologist on‐site?	No	50.8% (n = 63, N = 124)
	Yes	49.2% (n = 61, N = 124)

3.1. Justification

Most referrals were issued by general practitioners (74.4%, n = 93, N = 125). Most participants perceived that referral forms (64.8%, n = 81, N = 125) had insufficient clinical history. To address this, most participants (49.2%, n = 60, N = 122) always assessed the body region of concern, while 31.3% (n = 36, N = 115) reported never discussing with the radiologists to supplement information. When the requested examination did not correlate with clinical history, most participants always consulted with the referring clinician (49.2%, n = 61, N = 124), but never consulted with the radiologist (40.5%, n = 49, N = 121), as shown in Table 3. Chi‐square tests demonstrated significant associations between the presence of an on‐site radiologist and communication with a radiologist, both to supplement insufficient referral information (Cramer's V = 0.48) and when the requested examination did not correspond with the clinical history (Cramer's V = 0.530). In both cases, the associations reflected a large effect.

TABLE 3.

Actions taken when the requested examination does not correspond with the clinical history.

	Never	Rarely	Sometimes	Often	Always
I consult with the referring clinician	2.4% (n = 3)	8.1% (n = 10)	12.9% (n = 16)	27.4% (n = 34)	49.2% (n = 61)
I consult with the radiologist	32.8% (n = 38)	7.8% (n = 9)	19.8% (n = 23)	14.7% (n = 17)	25.0% (n = 29)
I consult with a fellow radiographer	5.0% (n = 6)	2.5% (n = 3)	34.7% (n = 42)	29.8% (n = 36)	28.1% (n = 34)
I change the referral to an appropriate examination	29.5% (n = 36)	19.7% (n = 24)	20.5% (n = 25)	18.0% (n = 22)	12.3% (n = 15)
I return the referral along with a reason	6.5% (n = 8)	20.3% (n = 25)	26.0% (n = 32)	27.6% (n = 34)	19.5% (n = 24)
I conduct the examination as requested	40.5% (n = 49)	22.3% (n = 27)	26.4% (n = 32)	8.3% (n = 10)	2.5% (n = 3)

The majority of participants (68.9%, n = 84, N = 122) had referral guidelines in their departments and strongly agreed that referral guidelines significantly reduced the frequency of unnecessary examinations (57.9%, n = 70, N = 121) and improved awareness of appropriate use of radiology imaging (54.9%, n = 67, N = 122).

In terms of perceptions of radiology referrals, most participants (60.8%, n = 76, N = 125) perceived referrals as recommendations for examination rather than a command. The final assessment of referrals before examinations was conducted by diagnostic radiographers for general X‐rays (74.4%, n = 93, N = 125) and dental X‐rays (49.0%, n = 51, N = 104). For more specialised modalities such as Computed Tomography (CT) scans (56.5%, n = 61, N = 108) and fluoroscopy (47.7%, n = 51, N = 107), both diagnostic radiographers and radiologists collaborated in the final assessment, and angiography examinations were solely assessed by radiologists (29.4%, n = 30, N = 102).

Challenges such as ineffective communication channels among health professionals (48.8%, n = 60, N = 123), lack of response from radiologists (39.5%, n = 47, N = 119), and lack of training in systematic assessment of referrals (33.9%, n = 42, N = 124) hindered the participation of diagnostic radiographers in referral assessment. A t‐test revealed a statistically significant difference between the hindrances to assessing referral forms (B7.4, B7.3, B7.5, B7.6, and B7.1) (p = 0.024) between participants who had a radiologist on‐site (mean = 3.70, SD = 0.868) and those who did not (mean = 3.33, SD = 0.894).

Only 63.1% (n = 77, N = 122) of the participants agreed that they had adequate autonomy. Most participants (66.4%, n = 81, N = 122) felt that their opinions and expertise in the justification of examinations were undervalued and overlooked due to hierarchy within the healthcare system.

In private radiology departments, 56.4% (n = 22, N = 39) of participants agreed that financial considerations sometimes took priority over patient welfare, whereas 51% (n = 20, N = 39) disagreed that patient quantity was prioritised over personalised and patient‐centred care.

3.2. Optimisation

Regarding the X‐ray rooms, most participants reported functional warning lights (80%, n = 84, N = 105) and doors (91.4%, n = 96, N = 105) in their departments. Most participants reported that X‐ray rooms did not have windows (67.6%, n = 71, N = 105). Most participants underwent QA training (58.7%, n = 61, N = 104). No significant Pearson correlation was found between hospital type and QA training (p = 0.835).

The majority of participants identified patients by asking multiple personal details, including name, date of birth, or identity number (92.3%, n = 96), and 60.6% (n = 63, N = 104) of participants always provided clear instructions before conducting radiology examinations.

The majority of participants applied exposure technique factors, including minimising exposure time (91.1%, n = 92, N = 101), increased distance (89%, n = 89, N = 100), using high kV technique (85%, n = 85, N = 100) and adhering to the exposure chart (89.9%, n = 89, N = 99). Moreover, most participants agreed that proper collimation minimises radiation exposure and improves image quality (95.2%, n = 99, N = 104) and disagreed that image cropping is the same as collimation (94.2%, n = 98, N = 104).

As shown in Table 4, most participants reported that full lead aprons (98.1%, n = 102, N = 104) were the most accessible radiation safety apparel, while lead goggles (33.7%, n = 34, N = 101) were the least accessible. Fisher's exact test revealed significant relationships between hospital type and the availability of thyroid shield (p = 0.004, Φ = 0.276, small effect), lead gloves (p = 0.000, Φ = 0.436, medium effect), and lead goggles (p = 0.000, Φ = 0.362, medium effect).

TABLE 4.

Availability of radiation safety apparel.

Radiation safety apparel
Full lead aprons	98.1%, n = 102, N = 104
Thyroid shields	87.3%, n = 89, N = 102
Waist lead aprons	81.4%, n = 83, N = 102
Mobile lead shields	81.2%, n = 82, N = 101
Gonad shields	69.9%, n = 72, N = 103
Lead gloves	53.9%, n = 55, N = 102
Lead goggles	33.7%, n = 34, N = 101

3.3. Dose Limits

Most participants were issued with a personal radiation monitoring device (PRMD) (97.1%, n = 99, N = 102) monthly monitoring (81.8%, n = 81, N = 99) and always wore a PRMD (53.5%, n = 53, N = 99). Only 33.3% (n = 32, N = 96) of participants always received radiation monitoring device readings. PRMDs were typically placed underneath lead aprons when wearing a lead apron only (50.6%, n = 40, N = 79), and a lead apron and thyroid shield were worn (51.9%, n = 42, N = 81), as shown in Table 5. A negative weak Pearson correlation was found between years of experience and “Hanging on my torso on top of the apron” placement when only a lead apron was worn (p = 0.031, r = −0.238). The majority of participants always wore the necessary lead‐protective apparel during screening in the theatre (60%, n = 60, N = 100) and angiography examinations (26.9%, n = 25, N = 93). Moreover, most participants never/rarely hold patients as a form of immobilisation (geriatric: 50.5%, n = 51, N = 101; unconscious: 68.0%, n = 68, N = 101), as shown in Table 6.

TABLE 5.

Placement of radiation monitoring devices.

		Never	Rarely	Sometimes	Often	Always
When wearing a lead apron and thyroid shield?	Outside the thyroid shield	52.4% (n = 43)	6.1% (n = 5)	12.2% (n = 10)	6.1% (n = 5)	23.2% (n = 19)
	Hanging on my torso on top of the apron	52.5% (n = 42)	5.0% (n = 4)	13.8% (n = 11)	12.5% (n = 10)	16.3% (n = 13)
	Underneath lead apron	25.9% (n = 21)	6.2% (n = 5)	8.6% (n = 7)	7.4% (n = 6)	51.9% (n = 42)
When wearing a lead apron?	Hanging on my torso on top of the apron	48.8% (n = 40)	8.5% (n = 7)	12.2% (n = 10)	11.0% (n = 9)	19.5% (n = 16)
	Underneath lead apron	26.6% (n = 21)	7.6% (n = 6)	7.6% (n = 6)	7.6% (n = 6)	50.6% (n = 40)

TABLE 6.

Holding of patients during radiology examinations.

	Never	Rarely	Sometimes	Often	Always	Not applicable
Paediatric patients	14.9% (n = 15)	21.8% (n = 22)	34.7% (n = 35)	18.8% (n = 19)	8.9% (n = 9)	1.0% (n = 1)
Geriatric patients	29.7% (n = 30)	20.8% (n = 21)	20.8% (n = 21)	15.8% (n = 16)	7.9% (n = 8)	5.0% (n = 5)
Unconscious patients	47.0% (n = 47)	21.0% (n = 21)	17.0% (n = 17)	8.0% (n = 8)	7.0% (n = 7)	0% (n = 0)
Patients with cognitive or behavioural disorders	30.7% (n = 31)	27.7% (n = 28)	23.8% (n = 24)	8.9% (n = 9)	7.9% (n = 8)	1.0% (n = 1)
Patients with physical disabilities	27.0% (n = 27)	27.0% (n = 27)	30.0% (n = 30)	7.0% (n = 7)	8.0% (n = 8)	1.0% (n = 1)

4. Discussion

This study determined diagnostic radiographers' application of radiation protection principles of justification, optimisation, and dose limits. Participants viewed referral forms as recommendations for examinations to be conducted rather than directives, aligning with the International Atomic Energy Agency [3] guidelines that promote clinical judgement in decision‐making. This permits diagnostic radiographers to assess referrals, particularly general X‐ray, while radiologists focus on specialised modalities, which is consistent with existing literature on task distribution in radiology departments [39, 40]. Moreover, in general X‐rays, diagnostic radiographers serve as key decision‐makers, thereby enhancing radiation protection and operational efficiency [39, 40].

Most referrals were written by medical practitioners, dentists, and nurses, as supported by the National Health Act No. 61 of 2003 of South Africa [41] and the National Health Service Reform and Health Care Professions Act 2002, section 25 of the United Kingdom [42], which states that trained clinicians registered with a regulatory body can refer patients. Consistent with previous studies, participants were frequently confronted with inadequate clinical information on referrals, revealing longstanding concerns of incomplete referral forms and lack of clinical questions [43, 44]. Diagnostic radiographers compensated for inadequate clinical information by assessing the body region, demonstrating clinical assertiveness [45]. Referral information ensures accurate imaging of the correct body region, which is essential for accurate diagnosis and aids in retrieval of previous imaging results, reducing duplication of examinations [14, 19].

Clinicians should rely on referral guidelines to refer patients for clinically justified examinations to improve the quality of referrals [46]. In South Africa, SAHPRA [47] provides a national guideline on requests for radiology examinations to support appropriate requests and minimise unnecessary examinations. However, there are no comprehensive, evidence‐based, modality‐specific imaging guidelines. The Radiological Society of South Africa [48] recommends the use of the American College of Radiology Appropriateness Criteria as an appropriate resource for imaging referral. Regular review and update of referral guidelines may therefore be warranted to ensure that guidelines remain aligned with advances in diagnostic imaging and emerging evidence. Consideration could also be given to strengthening the implementation of referral guidelines with clinical practice rather than relying solely on voluntary adherence, which may improve the referral process and support justification of medical exposure. Corroborating this, participants in the current study expressed strong agreement that referral guidelines effectively reduce unnecessary examinations. In instances where the clinical history did not correspond with the requested examination, most participants consulted referring clinicians, consistent with findings by Chilanga, Olerud and Lysdahl [35].

Notably, unlike previous studies, most diagnostic radiographers in the current study did not engage with radiologists [35, 49]. This limited interaction is attributed to the lack of on‐site radiologists, as significant associations showed a large effect between the presence of on‐site radiologists and communication with radiologists. The lack of on‐site radiologists is consistent with the general shortage of radiologists in South Africa, where there are 17 radiologists per 1 million population [50]. Moreover, in Mpumalanga, radiologists in the public sector are primarily accessible to the two tertiary and three regional hospitals [51, 52].

Barriers such as ineffective communication channels and a lack of training in systematic assessment hindered diagnostic radiographers' ability to assess referrals. Clarke, Akudjedu and Salifu [44], Chilanga, Olerud and Lysdahl [35], and Chilanga, Olerud and Lysdahl [52] reported similar findings, suggesting that effective interprofessional communication and targeted training are important in improving referral assessment competency. A significant comparison indicated that diagnostic radiographers with a radiologist on‐site were impacted to a greater extent than those without a radiologist on‐site by the factor “hindrances to assessing referral forms” (B7.4, B7.3, B7.5, B7.6 & B7.1), indicating that the presence of on‐site radiologists does not translate to improved collaboration. This highlights the need for proactive communication strategies to address interprofessional communication between radiologists and diagnostic radiographers [35, 49].

Previous studies reported long‐lasting challenges of diagnostic radiographers feeling disempowered, undermined, and lacking professional respect [22, 53, 54, 55, 56]. In contrast, most participants in the current study reported having adequate autonomy. Moreover, no significant difference was found between medical dominance and the presence of onsite radiologists, which may indicate a cultural shift towards recognising the contribution of diagnostic radiographers, especially amid radiologist shortages [57]. However, hierarchical constraints still left many participants feeling that their professional opinions were undervalued. These results were comparable to those of previous studies in which diagnostic radiographers experienced inferiority [58, 59].

In the private sector, most participants revealed that financial pressures took precedence over patient welfare. Similarly, a Finnish study revealed that imaging was overused in patients with private health insurance, raising concerns about imaging overuse in the commercialised healthcare system [60, 61, 62]. Nevertheless, most participants continued to uphold ethical standards, prioritising personalised patient care, consistent with professional values [63, 64].

In terms of optimisation principles, most X‐ray rooms complied with functional doors, warning lights, warning signs, and windowless design, in line with South African Health Products Regulatory Authority (SAHPRA) [65] standards. Most participants have undergone QA training, consistent with Maina, Motto and Hazell [53] and Willemse, Williams and Glober [24]. This contrasts with studies from Nigeria, which highlighted poor QA compliance [25, 66]. No significant Pearson correlation was found between hospital type and QA training, suggesting that all the hospitals enforced QA as per the Radiation Control Guidelines [16].

Participants reported utilising various radiation protection techniques that align with established radiography practices and international standards, which are fundamental in reducing radiation exposure [13, 67]. These techniques included minimising exposure time, increasing distance from the ionising radiation source, adherence to exposure charts, and high kV, consistent with previous studies [13, 68, 69, 70, 71]. Most participants acknowledged that collimation is important in minimising radiation exposure and improving image quality. However, previous studies have shown that the actual application of collimation remains inadequate and has been replaced by digital cropping [59, 72, 73]. Although most participants understood the distinction between collimation and cropping, this knowledge did not translate to correct clinical application.

Most participants reported using multiple identifiers to confirm patients, reflecting adherence to safety recommendations, thereby reducing the risk of misidentification, as relying on a single point of verification is prone to errors [19, 74, 75]. Moreover, most participants gave clear pre‐examination instructions, which have been shown to improve patient cooperation and procedural outcomes [76, 77, 78]. Given South Africa's linguistic diversity, improved communication strategies, such as mobile applications that translate radiographic positioning instructions into multiple languages, are necessary to ensure patient understanding and facilitate effective communication [78].

Full aprons and thyroid shields were the most accessible apparel, whereas lead goggles and gloves were less accessible. This corresponds with previous studies identifying disparities in the accessibility of lead protective apparel [11, 53, 69, 79]. Moreover, Lewis, Downing and Hayre [56] found that despite a favourable attitude towards radiation protection, the actual practice was hindered by limited resources and insufficient managerial support. Hospital type influenced access to the least available lead shield apparel, with private hospitals being better equipped, highlighting ongoing inequalities in the South African health system [53].

Dose monitoring practices mostly adhered to South African standards, with the majority of participants issued with PRMDs on a monthly basis in line with SAHPRA [18, 80] and South African Bureau of Standards [81] guidelines. Nevertheless, South African studies revealed that all radiology personnel were issued with PRMDs [37, 82]. Only one‐third of the participants received regular feedback on radiation exposure, a pattern comparable to that reported by Rose, Uebel and Rae [82] and Mohd Ridzwan et al. [83], where feedback was usually only provided for unusually high readings. Limited feedback may hinder continuous learning and the promotion of a radiation safety culture [82].

The correct placement of PRMDs is essential for accurate exposure measurement. Despite guidelines stipulating that PRMDs have to be placed outside the lead apron when only a lead apron is worn and underneath the lead apron at the waist when wearing a lead apron, lead goggles and thyroid shield [17], most participants in the study placed PRMDs underneath the lead apron regardless of the type of apparel worn. Similarly, a South African study of interventionalists presented similar practices [82]. Additionally, less experienced diagnostic radiographers were more likely to place PRMDs on top of the lead apron when only a lead apron was worn, while more experienced diagnostic radiographers were more likely to place their PRMDs underneath the lead apron in all instances. These suggest a potential gap in awareness and might be attributed to the lack of refresher training for long‐serving personnel in departments.

Participants demonstrated high compliance with wearing lead shielding apparel in radiation‐use areas such as theatres, angiography and portable examinations, as well as holding of patients as a form of immobilisation during examinations. These results are consistent with previous studies and guidelines, highlighting a high level of adherence to occupational radiation protection practices among diagnostic radiographers [69, 84].

5. Limitations

Data were self‐reported and may be subject to social desirability and recall bias. Reliance on HODs for study dissemination may have introduced sampling and selection bias. Additionally, restricting the study to hospital‐based diagnostic radiographers may limit the generalisability of the results to other settings.

6. Conclusions

Diagnostic radiographers in Mpumalanga province who responded to this study applied the principles of justification, optimisation, and dose limits to some extent. However, incomplete referral assessment, limited interprofessional communication, hierarchical constraints, incorrect PRMD placement, and inadequate feedback on dose readings hindered optimal radiation protection practices. Local practice could be strengthened through targeted refresher training, structured interprofessional communication, implementation of standardised procedures for reviewing and communicating PRMD readings, and improved access to protective apparel, particularly lead gloves and goggles, through collaboration between radiology management and hospital administration.

Funding

The authors have nothing to report.

Ethics Statement

Ethics approval was obtained from the following:

• The Faculty of Health Sciences Research Ethics Committee of the University of Johannesburg (REC‐2289‐2023).

• Mpumalanga Provincial Health Research and Ethics Committee (MP_202310_010).

Written informed consent was obtained for anonymised information to be published in this article.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supplementary Material.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.