Nurses’ knowledge and self-assessment of their clinical experiences of intraosseous access: A multicentre cross-sectional study

Background: Intraosseous access is an effective and safe option when difficult vascular access occurs. The knowledge, competence, and clinical experience of nurses are collectively essential for the successful implementation of this approach in clinical practice. Education and clinical learning are the main pillars supporting this new practice to ensure patient safety. The aim of this study was to identify the nurses’ knowledge and clinical experience of intraosseous access and the factors associated with the adoption of this procedure. Methods: A cross-sectional study was carried out from October to December 2020. A convenience sample of 432 nurses from four Italian hospitals were involved. A structured questionnaire was used to assess the nurses’ knowledge of the intraosseous access guidelines and their clinical experience. Results: Most participants were female (71.5%) with more than 10 years of experience (63.7%) working in an emergency (38.9%) and medical (37.7%) setting. Most of the participants demonstrated their knowledge of the use of a device e.g., it is used if vascular access is not rapidly achieved in a child (83.1%) and the boluses of liquids required in the intraosseous procedure (72.7%). A few participants reported having placed intraosseous access (3.5%). A higher level of educational preparation and working in emergency and paediatric settings were associated with increased knowledge. Conclusions: Our findings highlighted a sub-optimal level of knowledge of the IO procedure, little experience of this practice in clinical contexts, also associated with a lack of adequate protocols and devices available to nurses. Nurses need to develop their knowledge and practice the skill clinically to embed this practice. University and nurse educators should emphasise the relevance of this practice in nursing education and training, so as to improve the nursing care practice and level of patient safety.


Introduction
Intraosseous (IO) access relates to the placement of a specialised hollow bore needled through the cortex of the bone into the medullary space for the infusion of fluids and medication or for blood sampling [1]. This practice ensures a fast, safe, reliable, effective, and economic route to administer medications and perform laboratory tests in both hospital and pre-hospital settings [2,3] or in emergencies [4]. Moreover, it can be considered if attempts at IV access are unsuccessful or not feasible [5]. Also, the International Liaison Committee on Resuscitation (ILCOR) recently reported that IO access is a reasonable option for drug administration during a cardiac arrest when IV access is difficult [6].
IO access could be placed, maintained, and removed by physicians, nurses, and paramedics [7] so long as they are proven to have the appropriate knowledge and competency to ensure patient safety [8]. The Infusion Nurses Society [9] has indicated that specifically trained nurses are well placed to insert intraosseous devices, thereby facilitating timely, safe, and effective care. The research highlighted the higher success rates of IO insertion compared to vascular access [10] even in paediatric patients [10,11] and neonates [12]. However, the updated guidelines highlighted to consider intraosseous (IO) access if intravenous cannulation is unsuccessful or not feasible [13].
Despite IO access having gained considerable attention during the last decade and having been recommended as a safe and effective approach [3,14], it remains still underused in clinical practice [15].
Central venous catheterisation is the prevailing alternative to peripheral IV access, and IO insertion is only considered as the last option [15]. Insufficient knowledge of IO access by healthcare professionals [16], poor skills and limited experience in clinical practice [17], a lack of devices [18], or a lack of protocols to support clinical decision-making are the main reasons for not adopting this technique.

Background
The healthcare professionals' knowledge, skills and education related to IO access are crucial to fostering its more widespread use in clinical practice [17]. Educational interventions are the key facilitators associated with the healthcare professionals' use of IO access [19]. Theoretical knowledge provides the foundation for practice implementation, and at the same time, it is crucial to provide a clinical learning experience for competency development and the development of practitioner capability [20]. To increase the adoption of IO in clinical practice, healthcare professionals need to combine adequate theoretical training with practical experience of IO access [21]. Also, to maximise the knowledge and skills retention among healthcare professionals, it is essential to update the knowledge of practitioners in accordance with the latest guidelines [22] and to plan periodic clinical re-training [23]. The knowledge and competence implementation should be focused on evidence-based findings, on supporting self-directed learning [24], and on embedding them in clinical practice [25]. However, there are still gaps when it comes to translating the evidence into practice, such as a lack of knowledge and skills, healthcare policies, workforce shortage, and equipment availability. On the other hand, having the possibility to refer to experts, supportive leadership, clinical learning opportunities, and easy access to the protocols facilitates the implementation of a new procedure into practice [26]. Supporting the practitioners' confidence and self-efficacy alongside the knowledge and skills to perform intraosseous access will assist in the development of the practitioners' capability, thus enabling the nurses to take a key role in utilising IO in their clinical practice. Capability is defined as the combination of skills, knowledge, and self-efficacy which enables individuals to manage change, to be flexible, and to move beyond their current competency [27].
The guideline adoption requires constant monitoring, organisational support, and feedback regarding the healthcare professionals' clinical practice [28]. Therefore, to facilitate the adoption of IO practice, it is important to support the nurses' education, especially in the critical care departments where this practice could improve patient safety. For this purpose, the IO procedure should become a core element of nursing education and post-graduation training, especially in the intensive care, paediatric and emergency departments [8].
To date, there has been a dearth of evidence examining healthcare workers' knowledge and competence around IO access. Most studies involving medical staff have highlighted those major improvements are needed to implement this practice on a more widespread basis [29]. To our knowledge, no evidence exists about nurses' knowledge or clinical experience in reference to this practice [30].
This study aims to assess the sampled nurses' knowledge and clinical experiences of IO access and the background characteristics that support them in gaining the appropriate knowledge and clinical experience.

Design and setting
A multicentre cross-sectional study was conducted from October to December 2020 in four Italian hospitals [blinded for referee].

Study population
Using a convenience sample, all nurses employed in a medical, surgical, emergency, and paediatric setting who agreed to participate by signing the informed consent form were enrolled.

Study procedures and data collection
In each hospital, a trained researcher recruited the participants and delivered the questionnaire.
The participants were asked to complete the questionnaire and to submit it in a sealed envelope into the designated response box.

Instrument description
A questionnaire based on the European Resuscitation Council Guidelines [31] and the American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care [3,32] for the placement, management, and removal of IO access as designed and adopted. It included 35 multiple choice questions across the following sections: i. Sociodemographic information: gender; age; level of education, years of experience; healthcare setting. ii. Knowledge assessment: 10 multiple-choice items on IO access management selected from the Guidelines and exploring the following issues: , an incorrect answer, a distractor, and an "I don't know" option to prevent random responses from the respondents. The score range was 0-10 with a higher score indicating a higher level of knowledge. iii. Clinical experiences self-assessment: 19 multiple-choice items were used to assess the nurses' perception of their own experience (4 items), skills (7 items), and training needs (8 items) regarding IO access.
A panel of 20 expert nurses evaluated the questionnaire's face and content validity. The expert panel were asked to score the clarity and relevance of each question using a 3-point Likert scale (1 = not relevant, 2 = relevant but not necessary, 3 = absolutely necessary). The panel expressed overall agreement with the questionnaire's clarity and relevance. The questionnaire was then pilot tested on a sample of 50 nurses. Item analysis was conducted to determine the difficulty level and the discriminating power of the items [33]. The item analysis was conducted using SITA v1.0 [34]. Overall, the difficulty index ranged from 0.5 to 0.8, showing satisfactory values (acceptability range was 0.0 to 1.0), except for item 3 (time to reach adequate plasma concentration in drug administration via IO access versus IV access) and for item 4 (in a critical child, the time within which IO is used if IV access cannot be found) (value = 0.24 and value = 0.30, respectively). Although these items were difficult for the responders, they were included in the final version of the instrument because they explored fundamental issues. Moreover, they provide useful insights into core areas of the current guidelines. All items reported an excellent discrimination index with a value ≥0.4 (the acceptability range was − 1.0 to +1.0). Item 5 (administrable infusions through IO access), item no. 6 (indications for clinical use of IO), and item 8 (advantages IO access versus CVC in CPR) reported a high capability to discriminate between groups of subjects (value = 0.9, value = 0.8, and value = 0.8, respectively).

Sample size
The sample size calculation was based on the data analyses performed (Mann-Whitney U test and Kruskal-Wallis tests) and on the expected effect size by considering the maximum number of categories involved, 4, after performing association analysis. Given a p-value of 0.05, a power of 0.80, and an effect size of 3, it was expected to involve at least 122 participants. When performing the multiple regression analysis model, we considered 4 predictors, a p-value of 0.05, and a power of 0.80. According to these parameters, the expected sample size was at least 84 participants.

Data analysis
The categorical variables were presented as frequencies and percentages. For the continuous variables, the mean, median, standard deviation (SD), and interquartile range (IQ) were calculated. The Mann-Whitney U test and Kruskal-Wallis tests were performed to assess the statistical differences. After the Kruskal-Wallis test, post hoc analysis was performed using the Mann-Whitney U test on each pair of categories by adjusting the p-value using the Bonferroni method. After testing that the distribution of the total score was not skewed, the relationship with the data collected through the questionnaire was assessed using multiple linear regression analysis, assuming that the dependent variable was knowledge's total score. The independent variables were level of education, years of experience, and the healthcare setting. All tests were considered statistically significant with a p-value <0.05.

Characteristics of the sample
The final sample consisted of 432 nurses (78.11% response rate), equally distributed among Region Marche (50.9%) and Region Puglia (49.1%). The mean age of the participants was 41.1 (SD = 10.8) years; most were female (71.5%). Table 1 summarises the sample's characteristics.

Nurses' knowledge of IO access management
There was a high rate of incorrect answers on the time to reach adequate plasma concentration versus IV access (82.9%), on the time within which IO is used if IV access is not available in a critical child (83.1%), on the boluses of liquids required in the procedure (72.7%), on the advantages of IO versus CVC in CPR (60.4%), and on main contraindications (62%). More than half of the respondents correctly answered the questions on the most used clinical situations of IO access (52.8%) and on the fluids that can be administered via IO (56.3%) ( Table 2). There were no statistically significant differences for years of experience (p = 0.292) ( Table 3). The median score increased with level of education: 3 for nurses with pre-academic education, 4.5 for those with a Bachelor's in Nursing, and 5 for those with a postgraduate education (p = 0.011). Regarding the healthcare setting, the nurses from emergency and paediatric areas scored significantly higher than nurses from other areas (p < 0.001) ( Table 3).

Factors related to the Nurses' knowledge
The multiple linear regression analysis revealed that 13.1% of the variance in the respondents' score was explained by the nurses' characteristics (R 2 = 0.131, F 5,426 = 12.85, p < 0.001). Level of education is positively and significantly related to the nurses' knowledge (β = 0.636, p < 0.001), while working in a medical healthcare setting is negatively related to the dependent variable (β = − 0.830, p = 0.002) ( Table 4).

Clinical experience
Less than half of the participants reported a lack of knowledge of the IO guidelines (41.7%), did not attend a training course (63%), and did not have any re-training (60.6%). In total, 15 participants had placed an IO access previous (3.5%), of which, 11 were via the proximal tibia access point. Almost the entire sample had never removed an IO access (91%) and had never been in situations where they could have used IO access (83.1%). When they could have, they did not mainly due to an awareness of not having adequate enough knowledge (48%), a lack of devices (31.5%), and a lack of practice (16.4%). Specifically, 96.3% of the sample reported the need for simulation training to enhance their theoretical-practical knowledge (86.3%) and 68.6% reported the need for specific devices (68.6%) and protocols for IO access management (74%) ( Table 5).

Discussion
This study aimed to assess the level of knowledge and clinical experience of the sampled Italian nurses on IO access and to investigate the factors associated with the nurses' capability to perform the procedure. Our results highlight the need for improvements in the nurses' knowledge of the guidelines and more practical experience of managing IO access. The participants indicated their willingness to further deepen their knowledge and clinical competence. IO access is the second line of approach if attempts at IV access are difficult or not successful and it is considered a valid option in emergency situations [6]. For this reason, this study provided useful insights to enhance nurse education and patient safety, by highlighting the gaps in implementing this procedure in clinical practice.

Nurses' knowledge of the IO guidelines
The nurses' guidelines knowledge is a necessary, but not sufficient, condition to implement clinical practice [35]. However, the low levels of knowledge suggest that in everyday clinical practice, the lack of information could be translated into a partial adherence to guidelines, resulting in potential harm and lower level of patient safety [35,36]. Overall, the findings of this study revealed that the level of Italian nurses' knowledge of IO techniques is suboptimal. However, it is important to pointed out that the theoretical knowledge underpins the clinical practice, while the work environment represents the

Table 4
Multiple linear regression analysis to assess the relationship between total score on ten multiple-choice questions and nurses' characteristics. circumstances in which theoretical knowledge are applied. It is therefore necessary to reflect on this aspect, probably associated with the still current Italian debate on the development and recognition of advanced nursing knowledge and competences. In fact, at national level there is no formal recognition of nurses' advanced competences; although the contribution of the academic courses in increasing the theoretical knowledge is well established, continuing clinical education is poorly defined [37]. Therefore, the need to acquire advanced skills, such as those relating to the IO procedure, may seem unlikely for nurses, considering also that our findings showed the poor applicability of the technique in the clinical contexts and the few continuing education opportunities specifically designed to increase nurses' practical skills of the IO technique.
Regarding the specific knowledge of the IO procedure, our results revealed important gaps on the time required to reach plasma concentrations: only 17.1% knew that drug administration by the IO route reaches adequate plasma concentrations in a time comparable to that of the IV access. These findings echo what was found by Smereka and colleagues [38] who reported that 60.5% of the nurses find that the fluid infusion is much slower through the IO access than through the IV access. However, this item demonstrated critical values when testing the content validity and the difficulty index, and it should be considered with caution.
The nurses acknowledged IO as the most effective approach in emergencies. However, only 16.9% were aware that in a critical child, the IO route can be used within one minute if IV access cannot be found. Regarding the main complications of IO access, less than half of the nurses correctly indicated infection and compartment syndrome as the most important problems, while about one-third admitted that they did not know what they were. These results contrast with those of Smereka and colleagues [38] where 87.1% of the respondents correctly indicated Table 5 Nurses' answers to 19 multiple-choice questions related to self-assess the nurse's perception of own experience, skills and training needs regarding IO vascular access.  infection as a possible complication. A small portion of the sample knew the main contraindications to the IO procedure being a bone fracture, infection, and a previous attempt at the same site. Probably, there is a lack of knowledge about the complications and contraindications associated with poor knowledge of the recommended sites for placing IO access. More detailed and practical training is necessary based on innovative teaching methods such as simulation technology-enhanced learning [39]. About half of the nurses did not know that during resuscitation, IO offers the same advantages. This result deserves attention considering that the insertion of a CVC requires a high level of expertise and can be time-consuming and difficult during resuscitation [23], whereas IO is a faster and safer way. Finally, higher levels of knowledge can be detected in two aspects: the IO route as a suitable option for the administration of fluids, drugs, and blood products, and the clinical situations in which IO access could be considered.

Nurses' perception of their experience, skills and training needs regarding IO access
Although the efficacy of IO access is well known, it is rarely adopted [40], as confirmed by our results. Only 3.5% of the sample had placed an IO, mainly in the emergency (53.3%) and surgical (26.7%) settings.
Academic-clinical partnerships are crucial for promoting the nurses' knowledge and clinical competencies [41]. The results show that approximately half of the sample had never participated in an IO training course and were not familiar with IO, similar to the findings of Smereka and colleagues [38]. Only a few respondents considered their knowledge as adequate.
In our sample, about half of those who have attended an IO training course did not carry out retraining. In contrast with this data, research shows that a training course should be consolidated by regular retraining [23].
Almost our entire sample stated that they had never placed an IO access (96.5%) in agreement with the work of Smereka and colleagues [38]. A possible explanation for this lack of experience could be due to the nurses' inadequate self-perceived knowledge, and their scarce awareness of the possibility of a nurse being independent when positioning an IO access. Moreover, more than half of the participants reported that in their clinical settings, IO access insertion is considered to be a medical prescription. Therefore, the position of IO access remains predominantly a medical intervention. This could be explained by considering the sociological theories on the professional boundaries in healthcare. In fact, it is recognised that healthcare professionals play a pivotal role in the development, negotiation, and maintenance of boundaries and in implementing various strategies for claiming specific knowledge and asserting their ownership over certain tasks [41].
However, the Infusion Nurses Society [9] clearly states that a qualified nurse, who is proficient in infusion therapy and who has been appropriately trained for the IO procedure, may insert, maintain, and remove these devices. In accordance with these statements, organisational policies should acknowledge the IO access as a nursing competence when properly documented.
The educational background of healthcare professionals should be characterised by the acquisition of knowledge and skills in IO positioning and management [8]. Continuing education provides the opportunity to learn and advance safe and effective nursing skills, especially those that are useful in life-threatening situations [9]. Given the importance of the IO technique for emergency, medical and surgical care, protocols should be implemented to ensure their effectiveness, appropriateness, and efficiency in nursing care practice [42]. Instead, our results show a lack of protocols and of adequate devices, while the nurses considered the introduction of protocols and devices to be necessary elements for safe and effective patient care [43].

Factors associated with the nurses' knowledge
Our results also highlighted some of the nurses' characteristics related to their level of knowledge of IO access. A higher level of education was predictive of a better total test score: nurses with postgraduate education scored significantly higher than those with a Bachelor's in Nursing and pre-academic education. Evidence suggests that nurses with a postgraduate education are more likely to perform critical thinking and have the right decision-making competencies, making it useful to implement advanced clinical competencies [44].
Furthermore, the nurses working in an emergency or paediatric clinical setting demonstrated a higher knowledge of the IO guidelines. Both areas are not significant predictors in the regression model. This could be due to IO access representing a faster and more effective infusion route in situations where peripheral IV access is difficult to find and therefore it is predominantly used in emergencies such as shock, severe dehydration, cardiac arrest, major trauma, or airway compromise [2], as well as in paediatric patients [10]. In this way, nurses working in those areas have a higher experiential knowledge of the procedure but not necessarily the associated knowledge of the guidelines.

Limitations
This study presents some limitations that need to be acknowledged. First, convenience sampling was utilised and therefore generalisation of our results should be considered with caution.
Secondly, we do not know whether the protocols adopted in the healthcare facilities were based on the updated guidelines and whether periodic retraining programmes were available. The results could therefore be affected by heterogeneous work and training contexts.
Also, the instrument is based on the participants' self-assessment, and due to this, the measures could be affected by a social desirability bias. Two items of the scale were rated as "difficult" by the participants, and this could also affect the outcome measures for those areas of assessment. It has not been possible to delete or reword those items because they covered two specific core areas of the guidelines. It is possible also that the items were scored as "difficult" due to the lack of exposure to IO insertion. There is little established correlation between self-assessment and independently rated competence [45,46,47]. However, self-assessment is widely used in assessing the need for further training [48].
Finally, our sample's characteristics and the cross-sectional design, while supporting reliable associations between the variables, suggests the need for caution when interpreting the regression model. Moreover, the regression model includes also non statistically significant parameters: while this could negatively affect the model's statistical fit, this approach better highlights the core areas to focus on for improving nurses' knowledge.
Paediatric nurses are less represented and further studies should focus more on this clinical setting and recruit a wider and more balanced sample. Furthermore, this study focused specifically on the nursing profession, as IO access is internationally recognised as a nursing competence. In different national settings, other healthcare profiles could be eligible for performing this procedure.

Relevance to clinical practice
This study contributes to strengthening the debate on the role of specialised nursing education and the importance of identifying strategies to expand advanced skills to increase the quality of care and patient safety outcomes [49,50].
The questionnaire used in the study could be periodically administered in clinical settings to assess the nurses' knowledge of IO access and to plan future educational and re-training programmes based on the target groups' characteristics and the nurses' working areas.
Nurse managers can find, in our findings, support for claiming the importance of continuing education and training in the work environment to enhance patient safety. Managers need to convey the current evidence to nurses and inspire and support their adherence to the guidelines to improve the quality and safety of nursing care. Therefore, within multidisciplinary teams, it is necessary to both enhance the nurses' role in this area and to introduce a specific knowledge and set of skills into clear and context-based nursing curricula. This nursing skill adoption is currently suboptimal in terms of clinical practice. Promoting opportunities for education, training, and simulation are crucial for a safe and effective advancement of clinical practice. Moreover, the implementation of an organisational policy for IO access would further promote the adoption of advanced nursing competencies and skills. The process of continuing professional development and skills acquisition is closely associated with practice development and defines the advancement of nursing practice.
Future research should focus on the implementation and evaluation of evidence-based educational interventions to determine their effect on increasing the appropriate and risk-aware use of IO access by nurses. Moreover, further studies might explore the IO access implementation amongst other healthcare workers and paramedics involved in emergency settings, if eligible for this procedure according to the specific healthcare and professional regulations in their own country.

Conclusions
Our study identified a low level of knowledge of the IO procedure, little experience of this practice among Italian nurses and a lack of protocols and of adequate devices in clinical settings. However, the participants showed a proactive approach by acknowledging the need to acquire advanced skills on IO access. A higher level of nurse's education is a predictor of an overall higher level of knowledge of the IO guidelines. Moreover, although we did not demonstrate evidence for working areas as predictors of better nurse's knowledge of IO guidelines, our results suggest that nurses working in paediatric and emergency settings showed higher level of knowledge. Further studies, with larger sample size, are required to better explore these preliminary findings.
However, the findings of this study are a starting point to further plan nursing education, advanced skills implementation, and continued training. The results, therefore, aim to highlight the relevance of academic, clinical, and organisational partnerships to invest in specialised academic courses and continuing nursing education to develop advanced nursing competencies, which positively affect nursing care quality and patient safety, by providing a wider set of competencies and clinical options in life-threatening situations.
Funding source: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Ethical statement: National and European laws (EU, 2016/679) have been adopted to ensure data confidentiality, in accordance with the Personal Data Act [Personal Data Act 523/1999. https://www.finlex.fi/en/laki/kaannok set/1999/en19990523.pdf (accessed October 3, 2020)]. The participation was voluntary to all participants and in compliance with the standards of informed consent, data confidentiality and anonymity (EU, 2016/679). Due to the type of data collected, and the descriptive design of the study, Research Ethics approval was not required. The administrative authorizations were requested and obtained from the Health Directors of the Local Health Providers [blinded for referee].

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.