Sedation scoring and managing abilities of intensive care nurses post educational intervention

Sedation scoring and managing abilities of intensive care nurses post educational intervention Vimala Ramoo, Khatijah Lim Abdullah, Patrick SK Tan, Li Ping Wong, Yan Piaw Chua and Li Yoong Tang RESEARCH doi: 10.1111/nicc.12180 ABSTRACT Background: Inappropriate sedation assessment can jeopardize patient comfort and safety. Therefore, nurses abilities in assessing and managing sedation are vital for effective care of mechanically ventilated patients. Aims and objectives: This study assessed nurses sedation scoring and management abilities as primary outcomes following educational interventions. Nurses perceived self-confidence and barriers to effective sedation management were assessed as secondary outcomes. Design: A post-test-only quasi-experimental design was used. Data were collected at 3 and 9 months post-intervention. Methods: A total of 66 nurses from a 14-bed intensive care unit of a Malaysian teaching hospital participated. The educational interventions included theoretical sessions, hands-on sedation assessment practice using the Richmond Agitation Sedation Scale, and a brief sedation assessment tool. Nurses sedation scoring and management abilities and perceived self-confidence level were assessed at both time points using self-administered questionnaires with case scenarios. Sedation assessment and management barriers were assessed once at 9 months post-intervention. Results: Median scores for overall accurate sedation scoring (9 months: 4 00; 3 months: 2 00, p = 0 0001) and overall sedation management (9 months: 14 0; 3 months: 7 0, p = 0 0001) were significantly higher at 9 months compared to 3 months post-intervention. There were no significant differences in the perceived self-confidence level for rating sedation level. Overall perceived barrier scores were low (M = 27 78, SD = 6 26, possible range = 11 0 55 0). Patient conditions (M = 3 68, SD = 1 13) and nurses workload (M = 3 54, SD = 0 95) were the greatest barriers to effective sedation assessment and management. Demographic variables did not affect sedation scoring or management abilities. Conclusions: Positive changes in nurses sedation assessment and management abilities were observed, indicating that adequate hands-on clinical practice following educational interventions can improve nurses knowledge and skills. Relevance to clinical practice: Educational initiatives are necessary to improve ICU practice, particularly in ICUs with inexperienced nurses. Key words: Case scenario assessment Clinical decision-making Intensive care nurses Perceived barriers Sedation assessment and management BACKGROUND Patient comfort and safety, specifically related to sedation therapy for critically ill patients, has recently received particular attention in critical care practice. Critical care practitioners are urged to standardize sedation management practices by adhering to specific strategies that could optimize patient comfort and safety (Patel and Kress, 2012; Riggi and Glass, Authors: V Ramoo, Lecturer, MEd (Mal.), BNSc (Mal.), RN, Department of Nursing Science, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia; KL Abdullah, Associate Professor, DClinP Nursing (UK), MSc in HSM (UK), BSc (Hons.) Nursing (UK), RN, RM, Department of Nursing Science, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia; PSK Tan, Consultant Intensivist, MBBS (Lon.), FRCA, EDIC, Intensive Care Unit, University of Malaya Medical Centre & Department of Anaesthesiology, Faculty of Medicine, 50603 University of Malaya, Kuala Lumpur, Malaysia; LP Wong, Associate Professor, PhD (Mal.), MSc (Mal.), BSc (Hons.) (Mal.), Department of Social and Preventive Medicine, Faculty of Medicine, 50603, University of Malaya, Kuala Lumpur, Malaysia; YP Chua, Professor, PhD (Mal.), MSc (Mal.), BSc (Mal.), Institute of Educational Leadership & Unit for the Enhancement of Academic Performance (ULPA), 50603, University of Malaya, Kuala Lumpur,, Malaysia; LY Tang, Lecturer, PhD (Mal), MNSc (Aust.), BNSc (Mal.), RN, Department of Nursing Science, Faculty of Medicine, University of Malaya, 50603, Kuala Lumpur, Malaysia Address for correspondence: V Ramoo, Department of Nursing Science, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia E-mail: vimala@ummc.edu.my; vimala@um.edu.my 2015 British Association of Critical Care Nurses 1

2013). The Society of Critical Care Medicine stresses the importance of patient-focused sedation-monitoring tools and protocols for effective intensive care unit (ICU) sedation management (Barr et al., 2013). The tools enable nurses to rapidly assess patients sedation levels which then provides direction for sedation adequacy assessment and therapy titration according to patients needs (Patel and Kress, 2012; Riggi and Glass, 2013). In addition, the use of such tools and protocols in critical care practices is effective in decreasing duration of mechanical ventilation, length of ICU stay and duration of sedation infusion (Treggiari et al., 2009; O Connor et al., 2010; Strøm et al., 2010). Although sedation guideline and assessment tool use has long been common practice in western ICUs (O Connor et al., 2010), their usage in Malaysian ICUs remains limited. Within the sparse published local research on sedation management, a study surveying 37 Malaysian public hospitals reported that only 13 (35%) ICUs routinely assessed sedation levels and only 14 (38%) utilized written protocols (Ahmad et al., 2007). Although 36 (95%) ICUs routinely adjusted sedation levels by patients clinical progress, only 10 (27%) interrupted sedation on a daily basis. Ahmad et al. (2007) recommended that routine assessment of sedation levels and utilizing a written protocol for sedation and analgesia could improve ICU standard practices. Another recent survey of 11 Malaysian hospitals revealed that 182 (71%) of 257 ICU patients were deeply sedated at the first assessment (Richmond Agitation Sedation Scale [RASS] scores of 3 to 5 on a scale of 5 to +4, where lower scores represent higher levels of sedation and 5 indicates no response to vocal or physical stimulation). At the 48-h assessment, 159 (61%) were deeply sedated (Shehabi et al., 2013). The study also noted that 58% (1658 of 2859 RASS assessments) and 34% (4528 of 13 319 RASS assessments) of patients were deeply sedated during the first 48 h and after 48 h, respectively. These findings indicated that many sedated ICU patients might be at-risk of oversedation. However, oversedation can only be identified if an appropriate sedation scale is used to monitor the degree of sedation. Study rationale Consistent with calls to improve local sedation practice, we felt it was necessary to standardize the sedation monitoring process. Sedation scales or guidelines were not widely used in ICU practice at the study setting before this study. Thus, we introduced a validated sedation assessment tool as part of an educational initiative to standardize sedation practices and promote continuity of patient care. Although guidelines or assessment tools might aid sedation delivery, the effectiveness of sedation administration still depends largely on ICU nurses abilities. Nurses independent assessments and skills are vital because they are expected to notice changes in patients sedation levels (Walker and Gillen, 2006) as they are at the patients bedside constantly and are responsible for ensuring that patients are safely and optimally sedated. Therefore, nurses knowledge, skills (Pun and Dunn, 2007; Patel and Kress, 2012), attitudes, experience, confidence (Walker and Gillen, 2006; Tanios et al., 2009) and clinical judgment (Aitken et al., 2009; Walker and Gillen, 2006) are important for safe sedative administration. In addition, studies have shown that a lack of knowledge and skills, fear of oversedation, lack of confidence, and a lack of support and acceptance are among nurse-related contributing factors to poor sedation protocol implementation. These factors have adversely affected adherence to evidence-based sedation practices (Walker and Gillen, 2006; Tanios et al., 2009). Thus, initiatives to improve ICU nurses awareness of and skills in sedation administration are crucial to improve evidence-based sedation therapy implementation and adherence (Tanios et al., 2009). In a previous study, nurses knowledge of sedation assessment and management was assessed before and after educational interventions. The nurses had limited knowledge of sedation assessment and management practice before the educational interventions. The nurses indicated that they had no exposure to or prior knowledge of RASS. Their role in sedation management before the study was limited to sedation delivery as prescribed with minimal emphasis on continuous monitoring (Ramoo et al., 2014). In this study, we evaluated nurses abilities to assess and manage sedation at 3 and 9 months after the educational interventions. In addition, nurses perceived confidence levels in assessing and managing sedation were measured at the two time points. As potential barriers might hinder the success of sedation practice (Tanios et al., 2009), perceived barriers were assessed at 9 months. METHODS Study design, setting, participants and procedure A total of 66 intensive care registered nurses (RNs; henceforth referred to as nurses) from a 14-bed general adult ICU (97% of all nurses in the unit) of a 920-bed teaching hospital in Kuala Lumpur, Malaysia, participated in this study between December 2010 and April 2012. Nurses worked full-time with three 2 2015 British Association of Critical Care Nurses

rotation-based shifts per day and a nurse-to-patient ratio of 1:1 or 1:2. Patients admitted to this unit were primarily medical and surgical cases. In this post-test-only quasi-experimental study, a questionnaire with case scenarios was used to assess nurses sedation scoring and management abilities post-educational interventions at two time points. Nurses pre-intervention abilities were not assessed because they had no prior exposure to the sedation assessment tool. However, to strengthen the study design, the nurses abilities were assessed twice to determine the changes over time. The educational interventions included theoretical sessions, introduction of a validated sedation assessment tool (RASS) into daily ICU practice and a brief algorithm-based sedation protocol to guide nurses sedation management practices. All nurses were trained in assessing, documenting and managing sedation levels using these tools. The processes and procedures involved in the educational interventions were described in a previous study (Ramoo et al., 2014). Following the educational interventions, nurses were required to assess and document patients sedation levels in the observation charts as part of ICU practice. Each nurse s sedation assessment method and documentation of sedation scores were randomly checked and audited at least twice by the main researcher and a trained research assistant during the pilot period. Coaching and bedside teaching were provided as necessary based on checking and audit feedback. Nurses documentation of sedation scores in patient records was also checked randomly in the post-intervention period. Nurses compliance with documentation of sedation scores was audited for 15 months in the post-intervention phase and the results will be reported in another paper. Research instrument The research instrument was a self-administered questionnaire with two sections. The first section, designed to facilitate sub-group analyses, assessed demographic information on age, level of nursing education and years of work experience. The second section presented four case scenarios. Each case scenario had a set of six questions: four regarding assessment and management of sedation and two assessing nurses perceived self-confidence levels based on the case scenario. The researchers developed case scenarios using realtime patient situations observed at the study setting. Each case differed in terms of descriptions of sedation level, sedative agents, dosage and patients haemodynamic status. Each correct answer received one point. Of the four questions on sedation assessment and management, one assessed the ability to score sedation level accurately using the RASS score (one point), and three assessed the sedation management ability based on the determined sedation level. Nurses determination of (i) adequacy of sedation (one point), (ii) titration of sedation infusion (one point) and calculation of drug dosage (one point) and (iii) subsequent nursing actions (two or three points, depending on the answers required) were the components evaluated for sedation management skills. The total score for overall accuracy in sedation scoring was four points and there were 21 points for the overall sedation management score. Perceived self-confidence level was assessed using a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree) from two self-confidence perspectives, specifically, (i) rating the sedation level accurately and (ii) making correct decisions for sedation management based on the case scenarios. The perceived self-confidence scores for each case scenario were computed to represent overall self-confidence level for sedation rating and for making decisions regarding sedation management, respectively. The levels of confidence were defined as follows: 1 00 3 00, low self-confidence; 3 01 4 00, moderate self-confidence; and 4 01 5 00, high self-confidence. At the 9-month post-intervention data collection, items to assess nurses barriers to effective sedation management and assessment were added to the questionnaire. This section had 11 items formulated based on previous studies (Walker and Gillen, 2006; Tanios et al., 2009) and the researchers clinical experiences. A 5-point frequency scale (0% = never to >75% = routinely) was used to rate the items. A higher percentage represented more barriers. Linguistic experts utilized a forward backward translation technique to translate the English version of the questionnaire into the Malaysian national language (Bahasa Malaysia). Instrument reliability was piloted with 20 post-registration critical care nursing students at two time points 4 weeks apart. The test retest results produced a Pearson s reliability coefficient greater than 0 80 for the two sets of tests, indicating good test retest reliability. Cronbach s alphas for the perceived barriers to sedation assessment and management scale were 0 86 (test 1) and 0 84 (test 2), showing a high level of internal consistency. Item clarity, comprehensiveness and content validity were assessed and critiqued by 10 critical care experts, including anaesthetists and senior nursing staff members. Minor corrections were made to the instrument based on their feedback. Data collection and ethical considerations Ethical approval from the hospital s Medical Ethics Committee and consent from nursing management 2015 British Association of Critical Care Nurses 3

were obtained prior to the study. Data were collected at two time points using the same set of questionnaires. After the conclusion of the 3-month educational interventions in July 2011, data were collected in October 2011 (at 3 months) and April 2012 (at 9 months). Codes were used in both sets of questionnaires for matching and comparing the nurses responses; however, these codes were kept confidential by the main researcher. Nurses voluntarily participated in the study following an explanation of the purpose and nature of the study. Nurses were assured of the anonymity and confidentiality of their survey responses. They completed the questionnaire in 35 45 min in a private room at both time points without referring to any resources during their off-duty hours. The researcher was present to answer any of their inquiries. A completed questionnaire implied consent. Data analysis Data were analysed using SPSS. Paired-samples t-tests (for normally distributed data) and Wilcoxon signed-rank tests (for non-normally distributed data) compared the changes in sedation management and self-confidence levels between 3 and 9 months post-intervention. The scores for overall sedation scoring, overall sedation management and perceived self-confidence levels were not normally distributed (Shapiro Wilk tests, p-values < 0 05); therefore, non-parametric tests were used for these variables. Fisher s exact tests analysed the association between changes in overall sedation scoring and nurses demographic characteristics; chi-square was not suitable because more than one cell had expected frequencies below 5. Mann Whitney U-tests and Kruskal Wallis tests were used to examine the differences in the overall sedation management scores and self-confidence scores by demographic characteristics. Statistical significance was determined at p < 0 05. RESULTS Demographic characteristics The same 66 nurses completed the questionnaire at both time points (response rate: 100%). More than one third of the nurses were 21 25 years old (40%, n = 26) and approximately two thirds had less than 5 years of work experience as registered (66%, n = 44) and critical care (68%, n = 45) nurses. Only 18% of the nurses had a post-registration certificate in critical care nursing; the rest had not completed any specialized nursing course. Accurate sedation scoring At 3 months post-interventions, only 6 (9 1%) nurses had correctly rated the sedation level in all four case scenarios and four nurses did not rate sedation level accurately for any of the case scenarios. In contrast, at 9 months, 37 (56 1%) nurses had rated all case scenarios correctly and 27 (40 9%) nurses had rated sedation level correctly for three case scenarios. There was a significant increase in sedation scoring accuracy at 9 months post-interventions [3-month median score: 2 00 (interquartile range; IQR= 1 75 3 00) versus 9-month median score: 4 00 (IQR = 3 00 4 00), z (64) = 6 04, p = 0 0001] (Table 1). There were no significant associations between changes in sedation scoring and demographic characteristics (Table 2). Table 1 Changes in sedation management and assessment scores at 3-month and 9-month post-educational intervention 3-month 9-month 3-versus 9-month Sedation management n Range M (SD) Range M (SD) t df p Adequacy of sedation 66 1 0 4 0 1 84 (0 91) 1 0 4 0 3 18 (0 71) 9 58 61 0 0001**, Titration of sedation 66 0 4 0 2 48 (1 25) 0 4 0 3 66 (2 14) 6 99 61 61 0 0001**, Subsequent actions 66 0 7 0 3 39 (1 89) 0 7 0 6 03 (1 82) 8 57 61 61 0 0001**, Sedation management for: Range Mdn Range Mdn z p Case scenario 1 66 0 4 0 0 00 1 0 5 0 4 00 6 58 0 0001** Case scenario 2 66 0 5 0 0 00 0 6 0 5 00 6 19 0 0001** Case scenario 3 66 0 4 0 2 00 0 5 0 4 00 6 36 0 0001** Case scenario 4 66 0 4 0 1 00 0 5 0 4 00 6 01 0 0001** Overall sedation management 66 1 0 14 0 7 00 3 0 21 0 14 00 5 31 0 0001** Sedation scoring Overall accurate sedation scoring 66 0 4 0 2 00 1 0 4 0 4 00 6 04 0 0001** M, mean; Mdn, median. Analysis based paired t-test. Analysis based on Wilcoxon signed-rank test. **Significant at p value < 0 05. 4 2015 British Association of Critical Care Nurses

Table 2 Changes in sedation scoring according to nurses demographic characteristic No difference Reduced Improved Demographic characteristics n X 2 df p Years of experience as RN 6 months to 5 years 44 4 (9 1) 3 (6 8) 37 (84 1) More than 5 years 22 5 (22 7) 3 (13 3) 14 (63 6) 3 494 1 0 117 Years of experience in CCU 6 months to 5 years 45 5 (11 1) 3 (6 7) 37 (84 4) More than 5 years 21 4 (19 0) 3 (14 3) 14 (66 7) 1 973 1 0 160 Level of nursing education Basic diploma 54 4 (7 4) 4 (7 4) 46 (85 2) Post basic education 12 5 (41 7) 2 (16 6) 5 (41 7) 3 482 1 0 056 Age groups 21 to 25 years old 26 3 (11 5) 2 (7 7) 21 (80 8) 26 to 30 years old 25 3 (12 0) 1 (4 0) 21 (84 0) 31 years and above 15 3 (20 0) 3 (20 0) 9 (60 0) 3 373 2 0 185 Total 66 9 (13 6) 6 (9 1) 51(77 3) CCU, critical care unit; RN, registered nurse. Fisher s exact test. Sedation management ability Data on sedation adequacy, sedation infusion titration and subsequent nursing actions were distributed normally with skewness and kurtosis between 1 96 and +1 96. Hence, paired-samples t-tests analysed the differences between scores at 3 and 9 months. Nurses ability to determine sedation adequacy (p = 0 0001), sedation infusion titration (p = 0 0001) and subsequent nursing actions (p = 0 0001) were significantly higher at 9 months than they were at 3 months post-intervention (Table 1). Comparisons between 3- and 9-month sedation management scores for each of the case scenarios and overall case scenario scores were conducted using Wilcoxon signed-rank tests. Significant increases were observed at 9 months post-intervention (Table 1) indicating that nurses sedation management abilities had improved over time. Further analyses revealed that among the demographic characteristics tested, significant differences in sedation management scores were noted only with years of work experience as RNs, and this was only observed at 3 months post-intervention (Table 3). Nurses with more than 5 years of work experience as an RN (mean rank= 40 16) had significantly higher sedation management scores than did nurses with less than 5 years of work experience (mean rank: 30 17), U = 337 50, z = 2 001, p = 0 045. Perceived self-confidence level Overall, the nurses indicated high confidence for accurately rating sedation levels for all the case scenarios at both time points and there were no significant differences between the two time points. However, significant differences in perceived self-confidence level were observed for making correct decisions regarding sedation management, specifically for case scenario 3 (z = 2 403, p = 0 016), case scenario 4 (z = 2 268, p = 0 023), and overall perceived self-confidence level (z = 3 471, p = 0 001). Table 4 displays the results in detail. Perceived barriers for sedation assessment and management The mean score for the overall barriers scale was 27 78 (SD: 6 26; possible range: 11 55) indicating that relatively few barriers were perceived (10 00 30 00, low hindrance; 30 01 40 00, moderate hindrance; and 40 01 50 00, high hindrance). Patient s critical condition (M: 3 68, SD: 1 13), haemodynamic stability (M: 3 59, SD: 1 17) and nursing workload (M: 3 54, SD: 95) were the most prevalent barriers to effective sedation assessment and management (Table 5). However, no significant differences were observed between demographic characteristics on overall perceived barrier scores. DISCUSSION This study found significant increases in sedation scoring accuracy and sedation management ability among the nurses from 3 to 9 months post-intervention. Although in the initial stage, the nurses had no prior knowledge of the RASS tool, their accuracy in scoring sedation was about 50% at 3 months and increased to about 88% at 9 months post-interventions. To our knowledge, this is the first study to assess nurses abilities in scoring sedation levels using case 2015 British Association of Critical Care Nurses 5

Table 3 Overall sedation management score by nurses demographic characteristics at 3-month and 9-month post-intervention (n = 66) 3-month 9-month Demographic characteristics n Mdn Mean rank Sum of ranks p Mdn Mean rank Sum of ranks p Years of experience as RN <1 years to 5 years 44 6 00 30 17 1327 50 13 74 34 41 1514 00 More than 5 years 22 7 50 40 16 883 50 0 045** 13 54 31 68 697 00 0 585 Years of experience in CCU <1 years to 5 years 45 6 00 30 87 1389 00 14 00 33 67 1515 00 More than 5 years 21 7 00 39 14 822 00 0 101 14 00 33 14 696 00 0 917 Level of nursing education Basic diploma 54 6 00 31 92 1755 50 14 00 33 29 1831 00 Post basic education 12 8 50 41 41 455 50 113 14 00 34 55 380 00 843 Age groups 21 to 25 years old 26 4 50 28 29 14 50 35 42 26 to 30 years old 25 7 00 38 46 13 00 33 10 31 years and above 15 6 00 34 27 0 162 14 00 30 83 0 754 CCU, critical care unit; Mdn, median; RN, registered nurse. Analysis based on Mann Whitney U-test. Analysis based on Kruskal Wallis test. **Significant at the 0 05 level (2-tailed). Table 4 Perceived self-confidence level at 3-month and 9-month post-educational intervention (n = 66) Self-confidence on rating sedation level Self-confidence on decision making 3-month 9-month 3- versus 9-month 3-month 9-month 3- versus 9-month Self-confidence level Mean (SD) Mdn Mean (SD) Mdn z p Mean (SD) Mdn Mean (SD) Mdn z p Case scenario 1 4 04 (0 59) 4 00 4 14 (0 55) 4 00 1 000 317 4 04 (0 51) 4 00 4 14 (0 55) 4 00 1 025 0 305 Case scenario 2 4 07 (0 61) 4 00 4 03 (0 58) 4 00 0 354 0 723 4 07 (0 53) 4 00 4 09 (0 55) 4 00 0 200 0 841 Case scenario 3 4 10 (0 61) 4 00 4 12 (0 59) 4 00 0 170 0 865 3 92 (0 48) 4 00 4 17 (0 54) 4 00 2 403 0 016** Case scenario 4 4 11 (0 38) 4 00 4 16 (0 57) 4 00 0 726 0 468 4 02 (0 41) 4 00 4 20 (0 50) 4 00 2 268 0 023** Overall self-confidence level 4 09 (0 33) 4 00 4 11 (0 52) 4 00 1 890 0 059 4 01 (0 49) 4 00 4 15 (0 45) 4 00 3 471 0 001** Mdn, median. Analysis based on Wilcoxon signed-rank test. Three-month post-educational intervention. Nine-month post-educational intervention. **Significant at p value < 0 05. scenarios. Thus far, only a few studies have examined practitioners accuracy in scoring sedation after the implementation of a sedation scale or algorithm. Among these studies, a prospective observational cohort study examined nurses accuracy in rating sedation depth (besides detecting delirium) after the implementation of a validated tool for monitoring sedation and delirium at two medical ICUs. Using an expert reference-standard rater and random spot-check method, 40% of nurses assessments of sedation level (RASS) were checked for accuracy and the agreement rates were high at both centres (Pun et al., 2005). A similar study by Vasilevskis et al. (2011) found that medical and surgical ICU nurses showed substantial and stable agreement with researchers in the assessment of both delirium (CAM-ICU weighted kappa: 0 67) and sedation (RASS weighted kappa = 66) in typical ICU practice despite prolonged time lapses from the initial implementation of the sedation improvement project. Such studies, if conducted on actual patients, would require considerable time- and cost-related budgets. Although this study assessed nurses sedation scoring accuracy using case scenarios, the main researcher and trained research assistant randomly spot-checked all of the bedside nurses sedation scores in the first 2 months post-intervention. The spot checks were intended for re-education and coaching rather than for statistical analysis or reporting. Although the nurses knowledge increased at 3 months post-intervention (Ramoo et al., 2014), their ability to 6 2015 British Association of Critical Care Nurses

Table 5 Perceived barriers to sedation assessment and management practice (n = 66) Items Never 1 Seldom 2 Sometimes 3 Often 4 Routinely 5 Mean (SD) Nursing workload 2 (3 0) 5 (7 6) 24 (36 4) 25 (37 9) 10 (15 2) 3 54 ( 95) Lack of clarity sedation tool 16 (24 2) 26 (39 4) 18 (27 3) 5 (7 6) 1 (1 5) 2 22 ( 96) Lack of familiarity sedation tool 15 (22 7) 21 (31 8) 23 (34 8) 6 (8 1) 1 (1 5) 2 34 ( 98) Patient hemodynamic instability 5 (7 6) 7 (10 6) 13 (19 7) 26 (39 4) 15 (22 7) 3 59 (1 17) Patient s critical condition 4 (6 1) 6 (9 1) 14 (21 2) 25 (37 9) 17 (25 8) 3 68 (1 13) Lack of protocols/guidelines 11 (16 7) 21 (31 8) 21 (31 8) 1 1(16 7) 2 (3 0) 2 57 (1 05) Low priority 17 (25 8) 22 (33 3) 17 (25 8) 7 (10 6) 3 (4 5) 2 35 (1 11) Poor documentation 11 (16 7) 22 (33 3) 25 (37 9) 7 (10 6) 1 (1 5) 2 47 ( 95) Poor team communication 17 (25 8) 19 (28 8) 19 (28 8) 11 (16 7) 0 (0) 2 36 (1 05) Lack of knowledge 8 (12 1) 29 (43 9) 20 (30 3) 9 (13 6) 0 (0) 2 45 ( 88) accurately determine sedation levels and make clinical decisions regarding sedation management were lower at the 3-month assessment. However, significant increases were noted after 9 months of actual clinical practice. The 6-month time interval may have allowed them to integrate and reinforce theoretical knowledge into practice, gain practical experience and develop critical thinking skills necessary for assessing and managing patients on sedation therapy. Therefore, the nurses might have gained experience and maturity. In other words, improvement in assessment and management abilities could have occurred based on a combination of knowledge enhancement from theoretical learning and practical experience. As clinical practice is fundamental to creating competent and highly skilled critical care practitioners, the findings should stimulate more studies to confirm these results. One previous study showed that experienced nurses were more likely to provide higher sedation quality than junior nurses (Aitken et al., 2009). This current study, however, found that demographic characteristics did not affect nurses abilities to assess and manage sedation, particularly at 9 months. This implies that there may not be a relationship between improvements in accuracy of sedation scoring and management and nurses demographic characteristics although differences in nursing experience might explain initial discrepancies in sedation management ability. The nurses in this study showed high levels of confidence for rating sedation levels for all case scenarios at both time points. Although confidence levels were high at 3 months, they actually scored quite poorly on sedation scoring accuracy for most of the case scenarios. Nurses may have either overestimated their ability to correctly assess sedation levels or were unwilling to acknowledge their uncertainty. Nurses should be able to reflect more objectively on their own capability and to identify their weaknesses more rationally; this would provide them insight to develop their skills and capabilities further. Qualitative studies exploring the reasons behind their high confidence levels despite low accuracy in sedation scoring would be worthwhile. Overall, nurses indicated minimal effects of barriers to effective sedation management. However, critically ill patients conditions, haemodynamic instability and nurses workloads were identified as the main barriers to effective sedation management. Volatile patient conditions and haemodynamic instability are expected phenomena in critically ill patients and their constantly changing parameters require frequent reassessment and refinement of therapeutic goals. Accommodating patients conditions and prioritizing tasks during a hectic situation is not simple even for an experienced nurse. Moreover, in this sample, the majority of nurses were juniors and lacked experience. In addition, the shortage of nursing staff leading to an increased patient-to-nurse ratio makes patient care more complex and challenging. There were some limitations to this study. First, the educational intervention was conducted in a single ICU with a relatively small sample. Thus, the findings might not be generalizable to other local nurses or to international contexts. Second, the same questionnaire was used to collect data at two time points, which might threaten the validity of the findings. Although no feedback or discussion on correct answers was provided after the first round of data collection and the nurses were reminded not to discuss the case scenarios or answers among them after the test, they could have been sensitized to the case situations or found answers through other routes of learning. However, the interval of 6 months between the two time points of data collection probably reduced this threat. Third, data collection using case scenarios might not be sensitive enough to detect nurses actual abilities 2015 British Association of Critical Care Nurses 7

in scoring and managing sedation. Sedation management in real-life circumstances requires clinical assessment and decision-making based on extremely precise nursing judgments as patients conditions might vary considerably, be ambiguous or exist in a highly complex, fluid and critical situation. Furthermore, context and culture surrounding the patient might influence nurses decisions on sedation management. The dynamic view of cultural mind-sets implies that decision making is affected by cultural norms and unique interpretations of cultural and contextual influences (Oyserman, 2011), explorations of which are outside the scope of this paper. Further, valid baseline measurements could not be established because the nurses had no prior exposure to the RASS as indicated in a prior study (Ramoo et al., 2014) and the pilot study; thus, pre- and post-intervention effects could not be measured. Therefore, further studies to assess nurses ability to assess and manage sedation based on actual patient situations are recommended using a larger sample size, multiple study settings, and pre-post methods to enhance the generalizability and credibility of these findings. IMPLICATIONS FOR NURSING PRACTICE AND EDUCATION The introduction of a standardized sedation assessment scale as part of educational interventions to optimize patient comfort facilitated significant changes in nurses abilities to assess depth of sedation and manage sedation based on clinical decisions. Adequate and appropriate sedation is a marker of quality care; therefore, dose and sedation administration must be monitored vigilantly using a standardized assessment tool. Furthermore, the administration of ideal and safe levels of sedatives can only be achieved with an appropriate assessment of actual need (Mehta et al., 2009). Clinical evaluation based on sedation scales allows bedside nurses to make rough distinctions between adequate, inadequate and excessive sedation. This together with sedation protocols could further guide management of patients who are on sedation therapy. Therefore, sedation scales and protocols are necessary for the management of patients on sedation therapy and should be a standard component of ICU practice (Patel and Kress, 2012). For any new practice to gain success, time and sufficient clinical practice are required in addition to knowledge acquisition. Thus, future studies assessing the success of innovations in practice should monitor longitudinal effects to produce reliable results. Although it is not easy to provide additional nursing staff to ease ICU workloads, particularly given the current nursing shortage and need for efficient use of human resources due to global economic concerns, the development of new strategies to offset bedside nurses workloads might improve nurses concentration on patient care. Nurses demographic characteristics did not significantly affect their abilities in assessing and managing sedation following educational interventions. These results imply that educational initiatives could successfully improve and develop the skills of all nurses regardless of their demographic background. Alternatively, this finding might suggest that there were no effects for demographic factors because of an overrepresentation of junior nurses. Fewer than 14% of the nurses had over 8 years of work experience. As a result, we could not meaningfully investigate differences across more than two levels of experience. Therefore, educational programs should be planned to enhance nurses knowledge and skills with consideration of the inexperienced nature of the nurses studied in this ICU. Finally, observed mismatches between nurses levels of self-confidence and actual abilities in scoring sedation suggest that initiatives to develop nurses self-awareness are necessary for better reflection and judgment of their own ability. Acknowledgement of one s limitations is necessary for personal and professional development. Although basic strategies to develop self-awareness have been integrated into the nursing curriculum, it should be further fostered in nurses at the beginning of their careers through reflective and experiential learning. Conclusion The integration of theoretical knowledge with adequate hands-on clinical practice might facilitate improvements in nurses assessment and management abilities. Assessment of sedation level is fundamental to ensure that patients receive adequate and appropriate sedation. Therefore, this study emphasized the importance of standardizing sedation assessment practices using sedation scales and protocols validated by knowledgeable and competent nurses to improve ICU practice and quality of patient care. ACKNOWLEDGEMENTS This study was funded by the University of Malaya (RG176HTM10). The authors would like to thank all registered nurses involved in this project. Special thanks are extended to all doctors and nursing managers from the participating study unit for their kind support and assistance throughout the study. 8 2015 British Association of Critical Care Nurses

WHAT IS KNOWN ABOUT THIS TOPIC Recent studies have shown that sedation scales and protocols could lead to better patient outcomes. Administration of sedation therapy based on individual patient needs is crucial for safe and effective management of critically ill patients. Nurses have significant roles in sedation assessment and management and therefore their competency in assessing and managing sedation is pivotal for patient safety. WHAT THIS PAPER ADDS This paper highlights the importance of adequate hands-on clinical practice post-educational intervention to improve nurses sedation assessment and management abilities. This paper also emphasizes that assessment of the effectiveness of educational interventions should be conducted with a longitudinal perspective as integration of theoretical knowledge with clinical experience would result in better development of understanding and skills. REFERENCES Ahmad N, Tan CC, Balan S. (2007). The current practice of sedation and analgesia in intensive care units in Malaysian public hospitals. The Medical Journal of Malaysia; 62: 122 126. Aitken LM, Marshall AP, Elliott R, McKinley S. (2009). Critical care nurses decision making: sedation assessment and management in intensive care. Journal of Clinical Nursing; 18: 36 45. Barr J, Fraser GL, Puntillo K, Ely EW, Gélinas C, Dasta JF, Davidson J, Devlin JW, Kress J, Joffe AM, Coursin DB, Herr DL, Tung A, Robinson BR, Fontaine DK, Ramsay MS, Riker RR, Sessler CN, Pun BT, Skrobik Y, Jaeschke R, American College of Critical Care Medicine. (2013). Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Critical Care Medicine; 41: 263 306. Mehta S, McCullagh I, Burry L. (2009). Current sedation practices: lessons learned from international surveys. Critical Care Clinics; 25: 471 488. O Connor M, Bucknall T, Manias E. (2010). International variations in outcomes from sedation protocol research: where are we at and where do we go from here? Intensive & Critical Care Nursing; 26: 189 195. Oyserman D. (2011). Culture as situated cognition: cultural mindsets, cultural fluency, and meaning making. European Review of Social Psychology; 22: 164 214. Patel SB, Kress JP. (2012). Sedation and analgesia in the mechanically ventilated patient. American Journal of Respiratory and Critical Care Medicine; 185: 486 497. Pun BT, Dunn J. (2007). The sedation of critically ill adults: part 2: management. The American Journal of Nursing; 107: 40 49. Pun BT, Gordon SM, Peterson JF, Shintani AK, Jackson JC, Foss J, Harding SD, Bernard GR, Dittus RS, Ely EW. (2005). Large-scale implementation of sedation and delirium monitoring in the intensive care unit: a report from two medical centers. Critical Care Medicine; 33: 1199 1205. Ramoo V, Abdullah KL, Tan PSK, Wong LP, Chua YP. (2014). Intervention to improve intensive care nurses knowledge of sedation assessment and management. Nursing in Critical Care. Advance online publication. DOI: 10.1111/nicc.12105. Riggi G, Glass M. (2013). Update on the management and monitoring of deep analgesia and sedation in the intensive care unit. AACN Advanced Critical Care; 24: 101 107. Shehabi Y, Chan L, Kadiman S, Alias A, Ismail WN, Tan MA, Khoo TM, Ali SB, Saman MA, Shaltut A, Tan CC, Yong CY, Bailey M, Sedation Practice in Intensive Care Evaluation (SPICE) Study Group Investigators. (2013). Sedation depth and long-term mortality in mechanically ventilated critically ill adults: a prospective longitudinal multicentre cohort study. Intensive Care Medicine; 39: 910 918. Strøm T, Martinussen T, Toft P. (2010). A protocol of no sedation for critically ill patients receiving mechanical ventilation: a randomized trial. Lancet; 375: 475 480. Tanios MA, de Wit M, Epstein SK, Devlin JW. (2009). Perceived barriers to the use of sedation protocols and daily sedation interruption: a multidisciplinary survey. Journal of Critical Care; 24: 66 73. Treggiari M, Romand JA, Yanez ND, Deem SA, Goldberg J, Hudson L, Heidegger CP, Weiss NS. (2009). Randomized trial of light versus deep sedation on mental health after critical illness. Critical Care Medicine; 37: 2527 2534. Vasilevskis EE, Morandi A, Boehm L, Pandharipande PP, Girard TD, Jackson JC, Thompson JL, Shintani A, Gordon SM, Pun BT, Wesley EE. (2011). Delirium and sedation recognition using validated instruments: reliability of bedside intensive care unit nursing assessments from 2007 to 2010. Journal of the American Geriatrics Society; 59: S249 S255. DOI: 10.1111/j.1532-5415.2011.03673.x. Walker N, Gillen P. (2006). Investigating nurses perceptions of their role in managing sedation in intensive care: an exploratory study. Intensive & Critical Care Nursing; 22: 338 345. 2015 British Association of Critical Care Nurses 9