Introduction
Procedural sedation and analgesia (PSA) has been widely used since its origin in dental anaesthesia in the 1970s. Since that time, advancements in the pharmaceutical industry have led to the discovery and evolution of many short-acting drugs and refined knowledge of pharmacokinetic principles. Consequently, PSA has become a widely practised alternative to general anaesthesia but has, simultaneously, required the formulation of specific guidelines to ensure safe practice.(1) Sedation has traditionally formed part of the discipline of anaesthesia, with international and South African societies having published guidelines.(2–4,6) Other disciplines that use PSA have also published speciality-specific guidelines.(7–10)
The multiple guideline sources serve to emphasise the importance of following policies in order to avoid adverse patient outcomes. The incidence of adverse events is difficult to assess due to inconsistent research findings.(11) The American Society of Anesthesiologists (ASA) liability claims for 2009 show that procedures performed outside the operating room had a higher incidence of death compared to operating room procedures, with 50% of remote location deaths involving PSA. Respiratory dysfunction was found to be the most common complication.(12,13) Complication rates in South Africa show a mortality rate of 1 per 7500–11,000 endoscopic procedures and 0.03% of all procedures using PSA.(14) These figures may appear low but are of concern as they are higher than those for general anaesthesia for outpatient procedures.(1) Anecdotally it is also known that adverse events are under-reported.
The demand for health services is ever-growing and is not a uniquely South African phenomenon.(1) PSA offers an attractive economic alternative to general anaesthesia, reducing hospital length of stay, operating room time and procedure costs.(15–17) Furthermore, trained non-anaesthesiologists are able to provide PSA thereby reducing the workload on a diminishing number of anaesthesiologists.(15)
Due to the growing use of PSA and concerns about guideline knowledge and use, this study aimed to assess the profile of non-anaesthesiologist PSA providers at a 3200-bedded Johannesburg hospital, the awareness of the South African Society of Anaesthesiologists (SASA) Sedation Guidelines 2010 (5) and the level of comfort when performing PSA. With this information, inadequacies found can be addressed.
Methods
A cross-sectional prospective, descriptive, contextual study design was used. Approval for the study was obtained from the WITS Human Research Ethics Committee.
Doctors practising in the general surgery and trauma, radiology, emergency medicine, orthopaedic surgery and internal medicine departments and belonging to the professional levels of second year intern, community service doctor, senior house officer, specialist registrar or consultant formed the study group. A convenience sampling method was used for consultants, registrars, community service doctors and senior house officers. All interns in their second year of internship were purposively sampled as they would have had sufficient exposure to PSA during their clinical rotations in their first year.
A total of 359 doctors were identified. However, due to sick and annual leave as well as emergency duties, only 60% were accessible thereby making a sample of 215 doctors.
Data collection was done from September to November 2012 with the use of a self-administered questionnaire (Appendix 1). An extensive literature review was done with the search terms ‘awareness’, ‘knowledge’ and ‘comfort’. The questionnaire developed by Fanning (17) was deemed the most appropriate.
In addition, a detailed review of the 2010 SASA Guidelines for the provision of PSA to adults was done.(5) These guidelines were the most recently updated PSA guideline published by SASA at the time. The reasons for choosing the SASA Guidelines were that these guidelines were the most comprehensive guidelines available at the time of questionnaire development and were specifically developed for use by non-anaesthesiologists. The questionnaire consisted of four sections, namely, demographics, professional level and training, PSA procedures and locations, and guideline and pharmacology knowledge.
All medical doctors that met the inclusion criteria within the departments were identified with the assistance of the departmental secretary and an indication of appropriate times to approach these medical doctors was sought, e.g., departmental meetings. The questionnaires were distributed during departmental meetings, with consent being implied by agreement to participate. Questionnaires were completed anonymously and were placed in a sealed box, and the authors had access to the raw data only, thereby ensuring confidentiality and anonymity.
Data was analysed using descriptive and inferential statistics using Microsoft Excel 2010®. For normally distributed data, mean and standard deviation (SD) were used. The assumptions for ANOVA (equal variance and normality) were tested and met. Bonferroni testing and correction procedure was used for post-testing to identify where the significant differences occurred. A p-value of less than 0.05 was considered to be statistically significant.
Results
One hundred and sixty of the total accessible population of 215 doctors (74.42%) agreed to participate and completed questionnaires. Of the questionnaires returned, one questionnaire was excluded as the respondent did not perform PSA making a total of 159 respondents.
The departmental breakdown and the professional levels of respondents are shown in Table 1 and Figure 1, respectively. The post-graduate training completed by respondents showed that 113 (70.63%) had completed Basic Life Support, 97 (60.63%) had completed Advanced Cardiac Life Support, 25 (15.63%) PSA training/lectures/workshops and 40 (25%) indicated they had done other forms of training. One hundred and twenty-seven (79.87%) respondents indicated that they would benefit from PSA training.
| Clinical departments | Completed questionnaires Total = 159, n (%) |
|---|---|
| Radiology | 14 (8.81%) |
| Orthopaedics | 19 (11.95%) |
| General surgery/trauma | 37 (23.27%) |
| Internal medicine | 22 (13.84%) |
| Emergency medicine | 20 (12.58%) |
| Blank* | 47 (29.56%) |
*Questionnaires of interns: these doctors rotate among many departments
Respondents were questioned about their awareness of PSA guidelines and results are shown in Figure 2. One hundred and seventeen respondents (73.59%) indicated they were not at all aware of the PSA guidelines.
The mean scores for guideline and pharmacology knowledge were 11.54 (2.58) out of 16 and 8.01 (3.14) out of 15, respectively. The combined mean score was 19.55 (4.75) out of 31. PSA knowledge between the different professional levels was compared by dividing respondents into three groups. Consultants and specialist registrars were grouped separately, and senior house officers, community service doctors and interns (SHO/CS/I) were combined to form the third group. For guideline knowledge, a statistically significant difference was found between the groups (p = 0.0054). The SHO/CS/I group performed better than the consultant group (p = 0.008) with mean scores of 12.26 (2.18) and 10.71 (2.92), respectively. For total knowledge, a statistically significant difference was found between the groups (p = 0.027) with the SHO/CS/I performing better than the consultant group (p = 0.02). Mean scores for the SHO/CS/I and consultant groups were 20.50 (4.16) and 17.92 (5.34), respectively.
Departmental differences are shown in Table 2. For guideline knowledge, a statistically significant difference was found between the emergency medicine department and the orthopaedic surgery department (p = 0.03). Mean scores were 12.45 (1.50) and 9.89 (3.41), respectively. For pharmacology knowledge, no difference was found (p-value).
Guideline and pharmacology knowledge per department.
| Clinical departments | Knowledge (mean (SD) %) | ||
|---|---|---|---|
| Guideline | Pharmacology | Total | |
| Radiology (n = 14) | 10.21 (2.55) 63.81 | 6.21 (3.36) 41.40 | 16.43 (4.80) 53.00 |
| Orthopaedics (n = 19) | 9.89 (3.41) 61.81 | 6.58 (2.59) 43.87 | 16.47 (5.28) 53.13 |
| General surgery/trauma (n = 37) | 11.05 (2.86) 69.06 | 8.19 (3.22) 54.60 | 19.24 (4.77) 62.06 |
| Internal medicine (n = 22) | 11.45 (2.42) 71.56 | 7.86 (3.23) 52.40 | 19.32 (4.58) 62.32 |
| Emergency medicine (n = 20) | 12.45 (1.50) 77.81 | 9.20 (3.37) 61.33 | 21.65 (4.25) 69.84 |
The total knowledge score showed emergency medicine outperforming both radiology (p = 0.02) and orthopaedic surgery (p = 0.009). Mean scores for radiology and emergency medicine were 16.43 (4.80) and 21.65 (4.25), respectively, and mean scores for orthopaedic surgery and emergency medicine were 16.47 (5.28) and 21.65 (4.25), respectively.
The levels of comfort when performing PSA were compared among the different professional levels and consisted of three categories, namely, comfortable, neutral and uncomfortable. Table 3 shows the number of respondents per professional level and their respective levels of comfort when administering drugs for PSA and managing complications related to PSA. Three respondents did not indicate their professional level and so they were excluded from the analysis. A statistically significant difference (p = 0.031) existed between the groups. This difference was found between the MO/CS/I group and the consultant group, where 36.76% of the MO/CS/I group, in contrast to 19.44% of the consultant group, indicated they felt uncomfortable administering PSA. In addition, for managing complications related to PSA, a statistically significant difference (p = 0.008) between the groups was found. In the consultant group, 55.6% felt comfortable managing complications related to PSA in contrast to only 22% of the MO/CS/I group.
Level of comfort per professional level when administering drugs for PSA.
| Professional level (n, %) | |||
|---|---|---|---|
| Levels of comfort administering PSA drugs | Consultant (36) | Registrar (52) | MO/CS/I (68) |
| Comfortable | 19 (52.78) | 16 (30.77) | 16 (23.53) |
| Neutral | 10 (27.78) | 23 (44.23) | 27 (39.71) |
| Uncomfortable | 7 (19.44) | 13 (25.00) | 25 (36.76) |
| Levels of comfort managing PSA complications | Consultant (36) | Registrar (52) | MO/CS/I (68) |
| Comfortable | 20 (55.56) | 24 (46.15) | 15 (22.06) |
| Neutral | 8 (22.22) | 17 (32.69) | 31 (45.59) |
| Uncomfortable | 8 (22.22) | 11 (21.15) | 22 (32.35) |
Finally, a relationship between knowledge and level of comfort was examined. No statistically significant relationship was found between guideline knowledge (p = 0.50), pharmacology knowledge (p = 0.44), or the overall knowledge (p = 0.41), and the level of comfort when administering drugs for PSA. A statistically significant relationship was found between pharmacology knowledge and the level of comfort identifying complications related to PSA, with the comfortable group scoring 8.63 (57.55%) and the uncomfortable group scoring 7.03 (46.87%) (p = 0.036).
Discussion
The study was conducted at an academic hospital and included doctors from various professional designations. The proportion of doctors for each professional level is a reflection of this. Specialist registrars constituted the greatest proportion of participants (32.70%). This mirrored the study by Fanning,(18) which was conducted at the university teaching hospitals in Dublin, Ireland.
The proportion of doctors with PSA training was very low (15.63%). The Royal College of Anaesthetists (19) and Leroy et al.(20) have suggested that PSA training within the specialist disciplines is overlooked but needs to be incorporated into specialist training in order to reduce PSA-related adverse events.(13,21)
While the number of respondents with PSA training was low, the perceived benefit of PSA training was high (79.87%). Similar results were obtained in a Canadian study among the radiologists.(22) Studies have shown that the incorporation of formalised PSA teaching was thought to be useful and that training was indeed able to address PSA knowledge gaps among emergency medicine practitioners and those working in the endoscopy suite.(21–24) For this reason, the introduction of formal PSA training within specialist departments has the potential to contribute to practitioners’ PSA knowledge.
The development of guidelines was done in order to improve standards of care and safety; however, adherence to guidelines is a complex process. Most respondents (73.58%) indicated that they were not at all aware of a protocol for sedation practice in their department and only 9.43% of respondents used them. While these results are lower than those found in other studies,(18,20,25–27) the successful implementation of guidelines remains a challenging task across specialities. One of the reasons for this may be the tendency to follow the guidelines developed by ones’ own speciality.(27,28) This was the reason for enquiring about the use of guidelines other than those developed by SASA. Specialist affiliation was, however, not observed in this study. Other reasons for poor guideline adherence are discussed in Pathman et al.(29) awareness – agreement – adoption model. The impact of these factors was not explored in this study but may serve as a guide for future research.
The assessment of guideline knowledge showed low levels of awareness and the use of PSA guidelines. The mean score of 72.13% suggests that knowledge is acquired through means other than guideline use. When analysing the results further it was found that the SHO/CS/I group performed better than the consultant group. This was a surprising result as consultants have completed their specialist training and would thus be expected to have greater knowledge. The factors discussed by Cabana et al.(30) may provide an explanation for this. Attitude, which comprises self-efficacy, motivation and degree of agreement, all influence guideline knowledge and use.(30) Thus, consultants may feel reduced motivation to acquire and maintain knowledge once their training is completely leading to an eventual decline in knowledge. Another reason may be that PSA is assigned to more junior doctors, with the result being a decline in PSA knowledge and skill among consultants.
Pharmacology knowledge was also assessed. The mean score was only 53.40% (8.01 out of 15), which demonstrates the large pharmacology knowledge gap. Comparison between professional levels and departments showed no statistically significant difference. Fanning (18) found similarly poor results with a mean pharmacology score of 48.75%. Furthermore, 19.82% of respondents reported the occurrence of adverse events while administering PSA.
Anecdotal evidence suggested that PSA practitioners did not feel comfortable performing PSA due to the lack of training in this area. The results were confounding as most respondents felt neutral with regard to the levels of comfort when administering drugs for PSA and managing PSA-related complications, yet they felt comfortable identifying complications. An encouraging finding, however, was respondents’ perceived ability to object should they feel uncomfortable administering PSA.
Analysis of levels of comfort according to professional level revealed the SHO/CS/I group to be the most uncomfortable when administering PSA, whereas the consultant group felt the most comfortable managing PSA-related complications. Ones’ amount of experience would appear to be a reasonable explanation for this result, with the SHO/CS/I group having the least experience and years of training.
These results give some insights to the level of knowledge and comfort among PSA practitioners; however, the limitation of the study is its contextual nature, which infers the limited ability to generalise the results to other departments and hospitals in the country. Convenience sampling was used, particularly within the specialist registrar group, and this may contribute to bias. Further limitations include the ability of respondents to check for correct answers before submitting questionnaires and certain departments obstructing access to their staff. This may explain the poor response rate.
In conclusion, despite these limitations, to the best of our knowledge, this is the first study to be conducted at an academic hospital in South Africa examining PSA practice among doctors. The low levels of knowledge demonstrate a need to address this problem in order to improve patient safety as the amount of sedation performed is only expected to increase with the increased demand for health-care services.