This was a cross-sectional descriptive quantitative study.

The study took place in established EDs in KZN. These EDs belong to referral hospitals and receive paediatric patients referred for trauma, burns, general surgery pathologies, fractures and snake bites.

The study population included MPs working in the described departments. Qualified independent MPs comprising medical officers, Emergency Medicine Registrars and Emergency Medicine Specialists were included. Community service officers and interns were excluded from the study. This study took place between 07 July 2023 and 11 November 2023. The sampling strategy used was a combination of purposive and convenience sampling. To reduce social bias from the limited sampling period, department heads were informed about the study from the outset when granting permission. The heads of each department were asked to identify a day when all the staff would be present, allowing researchers to approach them for the study. On the scheduled day, they asked all department members to be present for data collection and to learn about the study in order to decide whether to participate. On this day, the questionnaire was presented to all staff members present at the time. Those who consented to participate were given the opportunity to take part.

The perceived level of confidence of MPs in performing critical procedures on paediatric patients was investigated using a questionnaire (Online Appendix 1) with a Likert Scale (ranging from having no confidence to being extremely confident). The questionnaire was adapted from a questionnaire used in a similar study after gaining consent from the author of that study.¹¹ The questionnaire that was received from the referenced study was modified and adapted by removing questions that were not pertaining to the research questions in this study as well as by refining the list of critical paediatric procedures that were looked at. This questionnaire was not piloted. The definition of self-reported confidence was explained to them as a group when approaching each department with a pre-scripted introductory speech. It was defined as their own individual assessment of their ability to perform the stated tasks and not on the outcome of those tasks. The questionnaire also included the MPs’ perceived importance of the critical procedures and possible factors that may influence their level of confidence when performing these procedures. All data were collected on days when the departments met for their departmental meetings. The study was explained to the department as a group, and questionnaires were then distributed within the department. Those individuals who consented to participating completed a questionnaire and dropped it inside a sealed box placed outside the meeting venue. This box was then collected by the author at the end of the departmental meeting. The questionnaire did not have the respondent’s name, and so the consent forms were collected separately. All data were then handled solely by the author, entered and stored on a Microsoft Excel spreadsheet (Microsoft Office 2016, Microsoft Corporation) on a password-protected computer.

Descriptive statistics were used to summarise the data. The study data were exported into IBM SPSS version 28 for statistical analysis. Descriptive statistics showed the participants’ demographic and professional profiles as well as details on their practices in the critical procedures. The descriptive statistics were reported in frequencies and percentages for categorical variables and as means with standard deviation (s.d.) for continuous variables. To explore the possible relationships, comparisons were made using Chi-square Statistical tests for categorical data and t-test or the Wilcoxon Rank-Sum test for numerical data. Levels of confidence in performing the critical procedures were compared according to the MPs’ demographics and profiles using one-way analysis of variance (ANOVA). The same method was used to determine if the level of confidence in performing a procedure was associated with the last time that procedure was performed. The Pearson correlation coefficient was used to examine if there existed a linear relationship between the level of confidence in performing the procedures and the perceived importance of performing those procedures. Statistical significance testing was set at the 95% confidence level and thus, P < 0.05 denoted statistical significance.

Ethical approval was granted by a Biostatistics Research Ethics Committee in KwaZulu-Natal on 28 July 2023 (Protocol reference number: BREC/00005328/2023) and the Provincial Department of Health (National Health Research Database [NHRD] reference: KZ_202306_014). A formal detailed consent form was signed by each participant. The data were collected and recorded only by the author of the study to maintain participant confidentiality.

Rank plays a crucial role in perceived confidence. The statistically significant relationship between rank and the mean level of confidence demonstrated that as the rank increased, the mean level of confidence increased. This could imply that the further training that is received to move up in ranks assists in confidence. However, Connick et al. showed a lower level of procedural confidence for consultants compared to specialist registrars stating that this may be because of registrars needing to perform more procedures as a training requirement.²⁴ In addition, the years of experience and seniority in each rank have not been considered in this study. As there are conflicting results, further research is required to assess the true impact of rank on self-assessed confidence in performing procedures.

Confidence in performing procedures increased with the frequency of performance as shown in Figure 1. This suggests that repeated practice leads to higher self-perceived confidence. This resonated with a study done by Connick et al., showing that the more frequently a procedure was performed, the greater the practitioner’s confidence.²⁴ Although studies conducted outside of South Africa indicated that these procedures were performed infrequently, a study in South Africa found that 71% of respondents had participated in resuscitations in paediatric patients over a 2-year period.²⁵ The study indicates a higher frequency of critical procedures performed in South Africa; however, it also identified notable gaps in resuscitation management in paediatric patients, which may involve critical procedures.²⁵ Present research showed that most of the participants performed all procedures apart from procedural sedation, zero to five times in their career. These findings indicate that additional practice may be necessary to improve confidence levels for most critical procedures involving paediatric patients.

In this study, most participants attended more than one emergency course. Paediatric Advanced Life Support and ETAT demonstrated a statistically significant improvement in mean confidence levels; however, these courses did not encompass all of the procedures evaluated in this study. The procedures not covered by PALS are laryngeal mask airway insertion, central line insertion, umbilical vein line insertion and procedural sedation. However, there was no comparable relationship between the confidence level of people who attended PALS and the procedures covered by PALS. The course with the lowest mean level of confidence was ATLS. This is most likely because the course focuses on trauma management in adults. This suggests that the current paediatric courses alone are not enough to maintain one’s confidence levels in performing all critical procedures in paediatric patients.

The study showed that gender may affect confidence levels with men being more confident than women with a P-value of 0.07 (mean level of confidence of men was 3.39 compared to women’s being only 2.8). This is not statistically significant. However, a similar trend was found by Selman et al. when evaluating plastic surgery residents’ operating abilities.²⁶ Madrazo et al. found that women tend to underestimate themselves more than men in self-assessment scores.²⁷ This is possibly because of a sex-based bias, where women unconsciously assume that they will perform poorly because of a stereotypical assumption that men are better than women.²⁶ Although Dufourq et al. in their study did not show a statistically significant difference between genders, Connick et al. showed greater confidence in men compared to women.^17,24 Therefore, gender does not have a statistically significant impact on mean confidence levels when undertaking critical procedures in paediatric patients.

The skills necessary to conduct critical procedures in children are considered extremely important.^11,28 This was echoed in present research showing the overall importance of performing critical procedures in paediatric patients in the ED was 4.35 on the Likert Scale. Comparably, in 2018, Craig et al. did a multi-centre cross-sectional survey of senior emergency medicine MPs working in EDs, and even though the critical procedures were not performed frequently, they were still considered important.¹¹ In this study, most participants were paediatric specialists; therefore, the perceived need for maintaining the skill set for critical procedures in children may have been biased.¹¹ Mittiga et al. reported that over 90% of the 262 participants considered maintaining the studied set of paediatric critical procedures to be highly important.²⁸ However, 68% of these institutions had paediatric emergency fellowship programmes, suggesting greater exposure to paediatric emergency care that may have influenced physicians’ views on maintaining critical procedural skills in children.²⁸ When an MP believes a procedure is important to result in positive patient outcomes, they will be motivated to engage in performing that procedure.²⁵ Vogt et al. showed that general practitioners are motivated to maintain and perform a procedure if they believe that it is effective and important.²⁹ These studies, together with the present research, highlight the extreme importance of efficiently performing critical procedures in paediatric patients. The present research shows a large discrepancy between the mean level of confidence and the perceived importance of the studied critical procedures. The fact that the study showed a large discrepancy when correlating confidence and importance possibly indicates that MPs believe that the procedures are effective and are motivated to receive continued medical education for these procedures.

This study has thus allowed for important considerations. To get the best outcomes for paediatric patients in the ED, one needs to consider improving MPs’ confidence in performing critical procedures in paediatric patients.^14,23,26 It is recommended to increase the practice of these procedures and to pursue ongoing medical education related to them, regardless of participation in standard emergency courses. Previous studies have shown that reinforcement improves confidence and commitment to change.²³ Craig et al. in their study suggested the framework of learn, see, practise, do and maintain to confidently perform procedures.¹¹ This includes ‘rapid cycle deliberate practice’, which involves repeated supervised attempts at resuscitation procedures with specific feedback and coaching.¹¹ Further research is required to observe if this study’s results are reproducible for the rest of South Africa. This includes research regarding the best method in our setting to improve confidence in performing these procedures. However, this study does highlight the need to report back to the participating EDs suggesting further practical training to improve their MPs’ confidence in performing these skills.

The limitation of this study is that, firstly, it is a small study which includes MPs only from predetermined EDs in KZN. Secondly, the majority of these EDs are general EDs and only treat limited paediatric emergencies. The exposure to the number of paediatric patients in these EDs is limited; this may affect the perceived confidence. The lack of an extended sampling period may limit generalisability, as well as the questionnaire, despite being adapted from a prior study, was not piloted. The number of years of experience of the participants was not taken into consideration, and the last time they took part in the training courses mentioned was also not considered, which may affect the results of the study. Finally, this study looks at self-perceived confidence when an MP performs this skill set, which is subjective and may suggest bias.