Basic Life Support for Nurses, KAREN MARDEGAN

Chapter 2 — Review of the Effectiveness of BLS Training Methods 62 | P a g e In health professionals seven days post training, an Internet program (that included instructor-led manikin practice) produced very low skill competence as did a Traditional instructor-led program (32% vs 36%, Moule et al., 2008a). Skill competence for health professionals 10 days post training was 44% for an Internet program and 73% for a Traditional program (Roppolo et al., 2011). For university students 12 weeks post training, 18% competence for compressions and seven percent for ventilations for the Internet program was reported compared with 80% and 37% for the Traditional program (Sarac & Ok, 2010). These differences between the Internet and Traditional programs 10 days and 12 weeks post training in health professional and lay people were statistically significant. Therefore skill competence at 10 days post training in medical students (Roppolo et al., 2011) and 12 weeks post training in one study of university students (Sarac & Ok, 2010) was significantly better with a standard Traditional program (Sarac & Ok, 2010) and a Traditional AHA program which includes a skill Video (Roppolo et al., 2011) compared to Internet programs which either included a Video manikin kit (Sarac & Ok, 2010) or the AHA Internet program plus a facilitator and a VAM manikin (Roppolo et al., 2011). Furthermore there was significant skill decline between 12 and 18 weeks post training in both the Internet and Traditional programs in university students (Sarac & Ok. 2010). BLS knowledge competence following Internet BLS programs have been evaluated in one health professional and one lay study of high school students, immediately post training (Moule et al., 2008a; Teague & Riley, 2006). In contrast to BLS skill, comparably high BLS knowledge (85%) in health professionals and (63%) in high school students was reported immediately post Internet training. Studies reporting on retention of knowledge following Internet BLS programs have not been found. High BLS knowledge competence immediately post training is consistent with CD and DVD studies (Cason et al., 2006; Clark et al., 2000; Fabius et al., 1994; Moule, 2002; Reder et al., 2006) and higher than in Video and the Traditional programs evaluated in these studies (Clark et al., 2000; Fabius et al., 1994; Reder, et al., 2006; Todd et al., 1998, 1999 [see Table 2.2 and Table 2.5]). These studies of BLS Internet programs therefore appear to suggest that skill competence with Internet BLS programs are low immediately and seven days to 12 weeks post training (Moule et al., 2008a; Roppolo et al., 2011; Sarac & Ok, 2010; Teague & Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 63 | P a g e Riley, 2006). Although the Internet program reported significantly better (but low) competence immediately post training in nursing students (Kardong-Edgren et al., 2010), Traditional programs appear to produce significantly better skill competence with higher competence rates (73% - 80%) 10 days to 12 weeks post training in both health professional and lay people (Roppolo et al., 2011; Sarac & Ok, 2010). There also appears to be significant skill decline with both the Internet and Traditional programs evaluated. BLS knowledge appears high immediately post training in Internet and the Traditional programs in health professional and lay groups (Moule et al., 2008a; Teague & Riley, 2006). It is also noteworthy that additional insights into Traditional plus skill Video programs, Video/DVD manikin kits, VAM manikins and the role of feedback can also be drawn from these Internet studies. These are discussed below. It is interesting that the Traditional plus Video programs in the Kardong-Edgren et al. (2010) and Roppolo et al. (2011) studies were the same AHA program, yet skill competence in this Traditional program in the Kardong-Edgren et al. (2010) study was low, where as skill competence for the same program in the Roppolo et al. (2011) study was high. Additionally, high skill competence has also been reported in other Traditional with Video RCT studies (Christenson et al., 2007; Riegel et al., 2006) and Traditional with CD studies (Clark et al., 2000; Moule & Gilchrist 2001), suggesting that there is most likely some benefit in combining a skill Video/DVD or CD with Traditional programs. The low Video/DVD manikin kit skill competency at 10 days and 12 weeks when used with Internet programs by medical and university students (Roppolo et al., 2010; Sarac & Ok, 2010) is consistent with some lay (Braslow et al., 1997; Isbye et al., 2006; Neilson et al., 2010; Todd et al., 1999) Video/DVD manikin kit studies but not all. Other lay (Chung et al., 2010; Mancini et al., 2009; Roppolo et al., 2007) and health professional (Bjorshol et al., 2009; Cason et al., 2009) Video /DVD manikin kit studies found high skill competence up to a year post training. The skill decline post training noted in the Sarac & Ok (2010) RCT in university students is also consistent with some (Einspruch et al., 2007; Roppolo et al., 2007) but not all (Chung et al., 2010) Video and DVD manikin kits in lay people. Therefore, there is inconsistent skill competence and retention particularly in lay groups when Video manikin kits are combined with Internet programs or used in isolation. The need for continued exploration to identify modes of Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 64 | P a g e training that can improve the outcomes seen with Video/DVD manikin kits therefore continue to be warranted. Enquiry into the VAM is only very recent, however from the two studies in health professionals in Table 2.6, it appears that when a VAM is added to the design of an Internet program to allow for manikin practice and feedback by the manikin, there was low (46% - 53%) but significantly better skill competence for the Internet plus VAM design when compared with the Traditional AHA program immediately post training with nursing students (Kardong-Edgren et al., 2010). Interestingly, at 10 days post training medical students using the same Internet plus VAM program demonstrated similar low competence levels (44%) to that seen immediately post training in the Kardong-Edgren et al. (2010) study, but the Traditional AHA program (also the same as in Kardong-Edgren et al., 2010 study), was in this case significantly better than the Internet VAM program (Roppolo et al., 2011). Considering that the VAM skill outcome in both these studies is comparable (44% - 53%), difference in skill competence between medical and nursing students appear unlikely, therefore this variation in the Traditional AHA program outcome is potentially variation in instructor delivery. The role of feedback in BLS training is also of interest. There is low skill competence in health professional samples (Kardong-Edgren et al., 2010; Roppolo et al., 2010) when a VAM feedback manikin is used in conjunction with Internet programs. There are also many instances, particularly in the CD literature where instructor feedback has not appeared to make any difference to skill outcome (Fabius et al., 1994; Moule, 2002; Reder et al., 2006). High skill competence is also seen in some instances, particularly in health professional samples, with the Video/DVD kit studies where there is no feedback (Cason et al., 2009; Chung et al., 2010; Mancini et al., 2009; Roppolo et al., 2007; Todd et al., 1998). This suggests that feedback is possibly not as crucial for BLS training as suggested in the broader training design and delivery (Aguinis & Kraiger, 2009; Salas & Cannon-Bowers, 2001) and resuscitation literature (ARC & NZRC, 2010a; Hazinski et al., 2010; Mancini et al., 2010). However, as this is the beginning of enquiry in the VAM, further evaluation of the VAM in particularly lay people is needed before definitive conclusions can be reached on VAM-provided feedback and feedback overall. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 65 | P a g e BLS animations. The one BLS animation study presented in Table 2.6 was with university students (n = 40) and compared a two minute animation (where viewing time was one hour and no manikin practice was provided) with a Traditional (one hour) instructor-led program. Skill competence immediately post training was low for both the animation (with 53% ventilations and 65% compressions performed competently) and Traditional groups (44% ventilations and 67% compressions performed competently). This low but comparable competency rate between the animation and Traditional program in Choa et al. (2006) implies that BLS animations as a BLS training method is comparable to Traditional instructor-led methods, and that CPR skills are not necessarily dependent on manikin practice. The comparable skill competence with and without manikin practice reported in the Bobrow et al. (2011) and Reder et al., 2006 RCT studies in lay adults supports this finding. However, skill competence is variable and more likely to be low in studies where manikin practice is not provided (Bobrow et al., 2011; Choa et al., 2006; Monsieurs et al., 2004; Reder et al., 2006; Teague & Riley, 2006). Additional research into lay adults, children and health professional groups would therefore assist in clarifying the role of manikin practice. Virtual world BLS training. The three virtual world BLS training programs in Table 2.6 were conducted by the Creutzfeldt team. Two of the studies were with medical students (Creutzfeldt et al., 2008, 2010) and one was in high school students (Creutzfeldt et al., 2009). These pilot studies (n < 16) evaluated BLS training using virtual world simulated BLS scenarios in addition to Traditional BLS training. There was a decreased mean time to the commencement of CPR, 10% improvement in the performance of compressions at nine months, and good BLS knowledge (80%) at three months post virtual world scenario training, with a decline (62%) in BLS knowledge from three to nine months post training for medical students (Creutzfeldt et al., 2008, 2010). In the 2009 study, which compared Traditional BLS training with and without virtual world training, less variation from resuscitation guidelines for the virtual training group and comparable BLS knowledge 18 months post training was reported for a high school student sample (Creutzfeldt et al., 2009). This recent evaluation of virtual world BLS scenario training suggests the potential for modest Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 66 | P a g e improvements in BLS skill and comparable outcomes in BLS knowledge to Traditional training with this form of training. Conclusions from BLS Internet studies. It is envisaged that the continual development and expansion of the Internet will foster the availability of various combinations of BLS DVD and CD training programs, simulations and virtual-world training programs through this medium. Research into the effectiveness of this medium as a vehicle for BLS training is beginning to be published in both the health professional and lay populations. The current research, presented in Table 2.6, where the BLS Internet program is an adjunct to Traditional instructor training (Creutzfeldt et al., 2008, 2009, 2010; Moule et al., 2008a), compared with no training (Teague & Riley, 2006) or used without concurrent manikin practice (Teague & Riley, 2006) limits determination of the efficiency of the training method for both BLS skill and knowledge attainment. However, at this time, skill competence post training with Internet programs appear to be comparably below skill mastery levels with Traditional methods (Kardong-Edgren et al., 2010; Moule et al., 2008a; Roppolo et al., 2011). There are, however, potential access advantages, good BLS knowledge achieved with Internet training programs (Moule et al., 2008a; Teague & Riley, 2006) and modest improvements in skill with virtual world training (Creutzfeldt et al., 2008, 2009, 2010). Furthermore, BLS skill competence and retention of skill with the BLS Video/DVD self-instructional kits, provided with Internet program in the Sarac and Ok (2010) and Roppolo et al. (2011) studies provided further example of where skill competence with these kits can be low, even in health professionals. Skill competence when Video manikin kits are combined with Internet programs or used in isolation have therefore not consistently demonstrated high skill competence with consistently good retention particularly in lay groups (Batcheller et al., 2000; Braslow et al., 1997; Einspruch et al., 2007; Lynch et al., 2005; Roppolo et al., 2007; Todd et al., 1999). The need for continued exploration to identify modes of training that can improve upon the outcomes seen with Video/DVD manikin kits therefore continues to be needed. The findings from the BLS animation study conducted by Choa et al. (2006) imply that manikin practice is not essential to the development of CPR skills. This is consistent with the Bobrow et al. (2011) Video and Monsieurs et al. (2004) and Reder et al. (2001) Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 67 | P a g e CD studies and the recent evaluation of the Internet AHA program accompanied by a VAM (Kardong-Edgren et al., 2010; Roppolo et al., 2011). This implies that ‗practice‘ and ‗feedback‘ is potentially not as crucial for BLS training as suggested in the training design and delivery (Aguinis & Kraiger, 2009; Salas & Cannon-Bowers, 2001) and resuscitation practice recommendations (ARC & NZRC, 2010a; Hazinski et al., 2010; Mancini et al., 2010). However, if this is the case then how best can BLS training programs be developed to significantly improve upon the BLS training outcomes from Traditional programs? The literature reviewed suggests variable but overall positive findings from Video/DVD kits particularly with health professionals (Bjorshol et al., 2009; Cason et al., 2009; Todd et al., 1998), a lack of comparisons available with independent CD training formats with and without practice (see Table 2.5) and limited outcomes from VAM in Internet studies (Kardong-Edgren et al., 2010; Roppolo et al., 2011). It is therefore proposed that enquiry firstly needs to concentrate on comparing the effectiveness of a CD-based manikin BLS training program with a Traditional BLS training program both initially post training and in retention of BLS skill and knowledge using standard manikins. This will have cost and access advantages for the user. CD programs which are not dependent on Internet access yet can also be provided via the Internet offer increased flexibility when compared with programs which are only available via the Internet. The Video/DVD kits have inflatable or cardboard manikins which are a compact low-cost form of manikin which can easily and cheaply accompany a CD program. The VAM at this stage is a cumbersome and expensive manikin that is most suited to large organisations. Wide spread access to BLS training that includes manikin practice is therefore more easily achieved in both the health professional and lay populations using standard manikins and CD rather than Internet programs. There is also the potential for improved outcomes with the unevaluated CD standard manikin model. The minimal evaluation of BLS CD programs generally and the lack of evaluation of CD designs which incorporate independent manikin practice coupled with their cost and access advantages substantiate priority being given to the development and evaluation of CD-based training methods rather than programs exclusively available via the internet or VAM incorporated designs. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 68 | P a g e BLS Training Design, Delivery and Outcomes BLS training is delivered using predominately Traditional classroom approaches. However, the presented research suggests that there is a building interest in Multimedia approaches to training. The skill and knowledge outcomes from these approaches to training vary within each method and also to some degree are training method dependent, making it difficult to determine the best current method, particularly when the development and evaluation of methods like CD programs have been so limited. Examination of how effectively the various modes of BLS training can address the prescribed principles of training design and delivery, and overall BLS skill and knowledge outcomes for these modes of BLS training potentially can assist in the drawing of conclusions from this literature. BLS training design and delivery. The review of the general training design and delivery literature suggests that training programs which meet the four training and delivery principles of relevance, demonstration, practice and feedback should produce the best training outcomes. Table 2.7 summarizes and compares the features of Traditional and Multimedia BLS approaches to training in relation to these prescribed principles of training design and delivery. Table 2.7 illustrates that both the Traditional and Multimedia BLS training methods have the capacity to comply with the four principles of training design and delivery. However, in Traditional approaches, standardisation of the delivery of the training is difficult to guarantee because it is dependent upon the quality of the trainers delivering the program (see Table 2.7). Multimedia approaches are standardised self-paced approaches which facilitate incorporation of the principles of relevant information, demonstration and practice; however their ability to provide feedback is dependent on the availability of an instrumented (rather than the standard) manikin (Dine et al., 2008; Edelson et al., 2008; Kardong-Edgren et al., 2010). Multimedia approaches are also potentially more cost-effective than Traditional methods because they do not rely on a human instructor, and allow for unlimited use for the individual and across organisations (Todd et al., 1999). Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 69 | P a g e Table 2.7: The design and delivery principles in relation to BLS training methods. Principle Traditional BLS Training Methods Relevant information Detailed information Multimedia (Video, DVD, CD, Internet + manikin) Simplified approach, only most relevant information provided Demonstration Potential for variability in Standardised information and delivery information and delivery Recorded demonstration Human instructor Potential variability in Standardised quality quality Instructor-led Self-paced Practice Limited opportunity for Opportunity to review as often as required. ongoing review Independent manikin practice Supervised manikin practice Shared manikins Manikin per trainee limited opportunity to Unlimited opportunity to practise during practise during and after and after training training Feedback Via human instructor Via instrumented manikin (when available) Potential for variability Standardised Potentially immediate Immediate Potentially individualised Individualised Feedback on entire BLS Feedback only on ventilation and process compression quality References: Braslow et al. (1997); Dine et al. (2008); Edelson et al. (2008); Kardong-Edgren et al. (2010); Todd et al. (1998, 1999). Analysis of the design and delivery of Traditional and Multimedia approaches to BLS training (in Table 2.7), in relation to the principles of training design and delivery (Table 1.5), indicate that Multimedia (Video, DVD, CD and Internet) training methods comply with the established principles of training design and delivery and, as such, are worthy of serious consideration and full evaluation. The variation in outcomes seen on review of particularly the CD BLS training programs could potentially be the result of the inadequate evaluation of these training methods presently, and as such there is also the potential for particularly CD to help to address BLS skill and retention concerns. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 70 | P a g e Basic life support training outcomes. The individual research pertaining to each mode of BLS training has been presented above. However, it is also useful when trying to determine the overall effectiveness of the various training method to examine the overall BLS skill and knowledge results for each method of training. The mean percentages competent for BLS skill and knowledge initially post training and for retention for the various methods of training have been calculated from the studies which have evaluated these methods. These mean percentages for each method of training are presented in Table 2.8 and Table 2.9. Table 2.8: A summary of BLS skill and knowledge competency achieved initially post training with the Traditional, Video, DVD, CD and Internet training methods which included manikin practice. BLS Training Mean % (SD) Mean % (SD) Studies Method Skill Adequate Knowledge Skill: Andresen et al., 2008; Batcheller et (including practice) Competence al., 2000; Braslow et al., 1997; Brennan & 63.5% (23.5) Braslow, 1998; Cason et al., 2009; Chung et TRADITIONAL 65.4% (37.6) al., 2010; Clark et al., 2000; Fabius et al., 1994; Jones et al., 2007; Lynch et al., 2005; VIDEO 56.3% (24.5) 77%* Madden 2006; Mancini et al., 2009; Mellor & Woollard 2004; Miyadahira et al., 2008 ; DVD 87.9% (15.7) 84%* Moule et al., 2008a; Roppolo et al., 2007 CD & Internet 41.3% (36.4) 86.7% (5.88) Knowledge: Cason et al., 2009; Clark et al., 2000; Kallestedt et al., 2010; Miyadahira et al., 2008; Moule et al., 2008a Skill: Batcheller et al., 2000; Bobrow et al., 2011; Braslow et al., 1997; Lynch et al., 2005; Reder et al., 2006 Knowledge: Reder et al., 2006 Skill: Cason et al., 2009; Chung et al., 2010; Jones et al., 2007; Mancini et al., 2009; Roppolo et al., 2007 Knowledge: Cason et al., 2009 Skill: Clark et al., 2000; Fabius et al., 1994; Kardong Edgren et al., 2010 ; Moule et al., 2008a; Reder et al., 2006 Knowledge: Clark et al., 2000; Fabius et al., 1994; Moule 2002; Moule et al., 2008a; Reder et al., 2006 * = one study Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 71 | P a g e Table 2.9: A summary of retention of BLS skill and knowledge competency post training with the Traditional, Video, DVD, CD and Internet training methods which included manikin practice. BLS Training Mean % (SD) Mean % (SD) Studies Method (including Skill Competence Adequate Knowledge Skill: Andresen et al., 2008; practice) Braslow et al., 1997; Chung et ≤ 6 month ≤ 6 month al., 2010; Einspruch et al., 2007; TRADITIONAL 44.8% (28.3) 62.5% (16.3) Fabius et al., 1994; Gasco et al., 2000; Isbye et al., 2006; Madden VIDEO ≤ 12 month ≤ 18 month* 2006; Mahony et al., 2008; 59.8% (30.9) 60% Roppolo et al., 2007; Sarac & Ok DVD 2010; Todd et al., 1998, 1999; Overall Overall Wollard et al., 2004 50.3% (29.4) 61.9% (13.3) Knowledge: Creutzfeldt et al., ≤ 6 month ≤ 6 month 2009; Kallestedt et al., 2010; 48.3% (23.1) 71.3% (5.06) Madden 2006; Todd et al., 1998, 1999 ≤ 6 month 55.0% (37.5) Skill: Bobrow et al., 2011; Braslow et al., 1997; Einspruch ≤ 12 month* et al., 2007; Reder et al., 2006; 100% Todd et al., 1998, 1999 Knowledge: Reder et al., 2006; Todd et al., 1998, 1999 Skill: Bjorshol et al., 2009; Chung et al., 2010; Isbye et al., 2006; Nielson et al., 2010; Roppolo et al., 2007; Sarac & Ok 2010 CD & Internet Overall ≤ 6 month Skill: Creutzfeldt et al., 2008; 61.4% (38.2) 73.0% (14.1) Fabius et al., 1994; Moule, ≤ 6 month 2002; Moule & Gilchrist 2001; 46.9% (38.7) ≤ 12 month* Reder et al., 2006; Roppolo et 63% al., 2011; Sarac & Ok, 2010. ≤ 12 month* 52% ≤ 18 month* Knowledge: Creutzfeldt et al., 62% 2009, 2010; Reder et al., 2006 Overall Overall 49.6% (38.8) 69.3% (11.8) * = one study Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 72 | P a g e When examining the effectiveness of these various BLS training methods the mean percentage for BLS skill and knowledge post training with Traditional methods are 65.4% (SD = 37.6) and 63.5% (SD = 23.5) respectively, Video is 56.3% (SD = 24.5) and 77% (one study only), DVD is 87.9% (SD = 15.7) and 84% (one study only), and CD and Internet programs are 41.3% (SD = 36.4) and 86.7% (SD = 5.88) respectively (see Table 2.8). This illustrates that BLS skill initially post training is well below skill mastery standards (80%) for all training methods accept for DVD manikin kits, and DVD, CD and Internet programs produce adequate BLS knowledge (above 80%) post training. For retention, BLS skill and knowledge by six months with Traditional methods is 44.8% (SD = 28.3) and 62.5% (SD = 6.4) respectively, Video is 48.3% (SD = 23.1) and 71.3% (SD = 5.06), DVD is 55.0% (SD = 37.5), with no studies found which examined knowledge, and CD and Internet programs are 46.9% (SD = 38.7) for skill and 73.0% (SD = 14.1) for knowledge (see Table 2.9). This illustrates decay of skill and knowledge with all BLS training methods of at least 20% for skill and 10% for knowledge competency by six months post training (≤ 6 month overall all methods: Skill decline 22%, Knowledge decline 12.6% [see Table 2.8 and Table 2.9]). Video and particularly DVD programs have therefore demonstrated improved BLS skill (DVD: mean skill competence initially post training 87.9% [see Table 2.8]) and comparable outcomes to other methods at follow-up (≤ 6 month DVD: Skill decline = 32.9% [see Table 2.8 and Table 2.9]), which suggests partial improvement in program effectiveness when compared to other methods for skill acquisition. BLS knowledge initially post training and over time, in the small number of available studies, appears to be consistently higher than skill competency rates, regardless of training method. CD and Internet programs appear to have marginally better initial and retention of knowledge rates when compared with Traditional and Video studies (see Table 2.8 and Table 2.9). It would therefore theoretically follow from this evaluation of training design and delivery principles and overall training outcomes for the methods, that CD programs that include independent manikin practice have the capacity to produce at least the overall skill outcomes seen with DVD (because they are able to include more indepth information then Video or DVD, in addition to the video footage of the BLS sequence), and improved Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 73 | P a g e BLS knowledge illustrated in the CD and Internet studies evaluated above. There is therefore also the potential for improved skill and knowledge retention. Summary of BLS skill, knowledge and current modes of training. The poor BLS skills and retention of skill in the health professional and lay populations, illustrated in the Traditional BLS training summaries Table 1.2, Table 1.3 and Table 2.1, suggest that there are limitations to the Traditional approaches to BLS training using an instructor-led BLS presentation/demonstration/practice format. The Videotape and its modern equivalent, the DVD and manikin model, have been heralded as a suitable alternative to Traditional BLS training methods (Hazinski et al., 2010; Mancini et al., 2010). However, variable BLS skill outcomes and poor retention are still a concern with these two methods. CD programs have not been evaluated sufficiently to determine their capabilities. However, particularly if independent manikin practice is incorporated, there is a potential for them to improve upon the currently variable BLS skill and poor retention of skill and knowledge reported in the literature. The above literature review illustrates that both the development and evaluation of Videotape and DVD BLS training programs have been far greater than the corresponding development and evaluation of BLS CD (see Table 2.2, Table 2.3, Table 2.4, and Table 2.5). The scarcity of literature (see Table 2.5) evaluating BLS CD training programs suggests that there may be little pressure for further evaluating BLS CD technology, if Videotape/DVD technology is providing potential improvement at least in BLS skill immediately following training. However, studies evaluating and comparing Traditional BLS training approaches with a BLS CD-based approach have not been found, either with or without a non- integrated manikin. This suggests that the actual potential for this method of training is still unknown. Providing the opportunity for independent manikin practice while viewing a BLS CD could be potentially as effective, flexible and cost effective as the Video manikin formula of Braslow and his contemporaries in Video and DVD (Batcheller et al., 2000; Braslow et al., 1997; Chung et al., 2010; Einspruch et al., 2007; Lynch et al., 2005; Todd et al., 1998, 1999 [see Table 2.2, Table 2.3, and Table 2.4]). Furthermore, retention of BLS skill over time continues to be a concern regardless of training method (Table 2.9). Video and DVD have not consistently demonstrated any Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 74 | P a g e considerable improvement in retention of skill over time when compared with outcomes from Traditionally-trained participants (Braslow et al, 1997; Einspruch et al., 2007; Roppolo et al., 2007; Todd et al., 1999). Training methods which maximise skill retention are required. The potential of CD-based and Internet BLS training programs have not yet been established in the literature. It is also noteworthy that many previous studies have evaluated BLS skills where participants perform direct mouth-to-mouth ventilation and one-rescuer BLS. Health professionals worldwide are currently required to be able to competently perform not only BLS using mouth-to-mouth ventilation, but also two-rescuer BLS using ventilation apparatus such as the one-way valve mask and bag-mask devices (ARC, 2004a; Henderson, 1992; Hurst, West, Austin, Branson, & Beck, 2007; Kardong-Edgren et al., 2010; Lee, Cho, Choi, & Yoon, 2008; Osterwalder & Schuhwerk, 1998; Paal, et al., 2010; Salas, Wisor, Agazio, Branson, & Austin, 2007). Studies which evaluate participants‘ ability to perform BLS using required equipment are needed to assist understanding particularly of health professionals‘ BLS ventilation skills and the proficiency of various methods in teaching the use of ventilation equipment and two-rescuer BLS. BLS knowledge has also been relatively under-evaluated (Cason et al., 2009; Creutzfeldt et al., 2010; Khan et al., 2010; Moule & Gilchrist, 2001; Moule, 2002; Moule et al., 2008a; Reder et al., 2006; Teague & Riley, 2006; Todd et al., 1998, 1999 [see Table 2.8 and Table 2.9]). Very few of these studies have provided comparison of acquired knowledge across two or more BLS training methods. Where training methods have been compared, findings appear to indicate that acquisition of BLS knowledge is comparable across the various methods of BLS training, with the potential for improved knowledge with CD programs (Cason et al., 2009; Reder et al., 2006; Todd et al., 1998, 1999). Further evaluation of BLS knowledge acquisition with the various training methods is needed to determine methods which maximise participants‘ BLS knowledge. Participant rating of the BLS program undertaken has been evaluated in a small number of studies, particularly BLS CD studies (Moule & Gilchrist, 2001; Moule 2002; Monsieurs et al., 2004). The feedback from participants who have trained using Multimedia programs is positive to date (Moule & Gilchrist, 2001; Moule 2002; Monsieurs et al., 2004; Shindo, Goto, Shibano, Okabe, & Inaoka, 2009). Further Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 2 — Review of the Effectiveness of BLS Training Methods 75 | P a g e evaluation of participants‘ perception of the various modes of BLS training is additionally needed to aid understanding of the relative merits of all forms of BLS training. A further point is that the studies discussed above have not been conducted in Australia and have also not examined whether findings for novices can be translated to those who have previously learnt the skills, and are re-accrediting in BLS. Re- accreditation is required for most groups of health professionals. Digital BLS programs, like BLS CDs, could be more suited to either beginners or to those reaccrediting in BLS. Therefore, to comprehensively evaluate the effectiveness of BLS CD studies which compare CD-based BLS training programs where a manikin is provided for independent practice with Traditional approaches to BLS training, in groups of varying experience, within the Australian experience are needed. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 Method Aims Primary aim. The primary aim of this doctoral study was: 1. To compare the BLS skill of Novice and Practising Nurses who trained via a CD- based (unsupervised practice) BLS training program or a Traditional (presentation/demonstration/supervised practice) program. Outcome measures were taken at one week and at two months post-training. Secondary aims. The secondary aims were: 1. To compare participants‘ knowledge of BLS at one week and at two months post- training. 2. To compare participants‘ rating (in terms of satisfaction with training program content, structure and assessment) of the CD and Traditional BLS training program undertaken. Hypothesis. It was hypothesised that the CD-based BLS training program would result in a higher rate of overall BLS skill competence and knowledge than would the Traditional BLS instruction program among both Novice and Practising Nurses. Research Design The project undertaken was a quasi-experimental study which compared two modes of BLS training in two cohorts: a Novice sample of 187 nursing students and an experienced sample of 107 Practising Nurses undertaking reaccreditation. Engaging a range of participant expertise from novice to practising professional was thought to Basic life support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 77 | P a g e promote relevance for the participant, assist participation and broaden the evaluation capacity of the study. The two training groups were a BLS CD training program which included a manikin for unsupervised practice and a Traditional BLS (presentation demonstration/supervised practice) program. Participants within the two groups were allocated to the two modes of training. There was no pre-test, but two post-tests, one conducted one week after training to determine skill and knowledge acquisition, and the second post-test conducted two months after training to assess skill and knowledge retention. BLS skill competence was determined using an assessor-graded BLS skills assessment form and automated manikin. BLS knowledge was established through answers given to BLS knowledge questions contained within a questionnaire. Participants‘ ratings of the CD and Traditional BLS training program (content, structure, assessment, and overall quality and satisfaction) were assessed via responses to a program evaluation form at one week post-training (see Figure 3.1). The post-test only comparative group research design of the project is similar to that used in foundation BLS Video studies conducted by Braslow et al. (1997) and Todd et al. (1998, 1999). This design was modelled on Campbell and Stanley‘s (1966) experimental design number six (post–test only control group design), except the present design (and the Braslow et al. 1997 study), did not use random allocation to groups. Additionally, the present design incorporates a two month follow-up period to test for retention of intervention effects. Design six was considered by Campbell and Stanley (1966) to be a robust design with a high degree of internal validity. The feasibility of including a pre-test into the design of this study was considered. Pre-tests had not been routinely employed in the founding studies (Braslow et al., 1997; Todd et al., 1998, 1999) nor in many of the studies that comprise the subsequent comparative literature (Batcheller et al., 2000; Einspruch et al., 2007; Jones et al., 2007; Lynch et al., 2005; Mancini et al., 2009; Roppolo et al., 2007). In the end, access constraints and concern that this additional assessment would increase participant burden and therefore could negatively impact on recruitment to the study led to the decision for a pre-test not to be included in the research design. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 78 | P a g e Figure 3.1 Study Design Karen Mardegan Basic life Support training for nurses: evaluating an alternative CD-based approach

Chapter 3 — Method 79 | P a g e To test the adequacy of measures, a pilot feasibility study with a small sample of 20 novice nursing students was undertaken prior to commencing the doctoral study. The pilot study results are described later in this chapter. Setting. The settings for the study were a large university which provides undergraduate training in nursing (novice nursing student sample), and a large tertiary hospital (reaccrediting graduate year nurse sample). The university was a multi-campus facility with a Faculty of Health Sciences and School of Nursing where undergraduate through to doctoral studies in nursing were offered. The hospital was a multi-campus tertiary health service with an established reputation for teaching and research. These organisations were deemed of sufficient size to accommodate the requirements of the study. Sampling frame. Three hundred and ten participants comprising 200 second year nursing students and 110 graduate year nurses, were invited to participate in the study. In total, there were two hundred and ninety four participants (94.8%). One hundred and eighty seven novice nursing students comprised the Novice cohort, and 107 graduate nurses comprised the Practising Nurses cohort (see Figure 3.1). Participants Recruitment of organisations. When the design of the research program had been finalised, the Dean of the School of Nursing from the university and the Director of Nursing and Ambulatory Services at the hospital were approached. Permission to access participants and support for each organisation‘s participation in the study was obtained (see Appendix A1 and Appendix A2). As the BLS CD and BLS assessment form in the current study was the product of Austin Health (Austin &Repatriation Medical Centre [A&RMC], 1999, 2000), permission from the Director of Nursing and Patient Support Services at Austin Health to use these tools was also obtained. Additionally, permission to use the questions from the Wilkinson and Chu (1999) surveys was also granted at this time (see Appendix A3). The research protocol of the study was then approved by the relevant Ethics Committees of the two organisations participating in the study (Research & Graduate Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 80 | P a g e Studies Human Research Ethics Committee of La Trobe University and the Austin Health Human Research Ethics Committee respectively [see Appendix A1 and Appendix A2]). The study commenced in January 2003 and was completed by December 2003. Recruitment of participants. Participants were approached as a group by the researcher during a scheduled university or hospital class to participate in the study. Participation in the study included: agreement to be allocated to a CD or Traditional instruction group, to undertake the BLS training and to complete a questionnaire, BLS skills assessment and program evaluation one week after the training (Post Test 1) and repeat completion of the questionnaire and BLS skills assessment two months after the training (Post Test 2). Once the requirements of participation were explained, and participants had an opportunity to review the written information provided (see Appendix B1 and Appendix B2), participants were formally invited to participate. Those who wished to take part completed the study consent form (see Appendix B1 and Appendix B3). Participant assignment. A pragmatic method was used to allocate participants into the CD and Traditional training groups. The student and graduate nurses had been arbitrarily allocated to class groups on enrolment in their respective university and hospital programs. The co- ordinators of the university and hospital programs randomly assigned equal numbers of these previously determined groups to the CD-based BLS training and Traditional (presentation /demonstration/practice) training methods. The training schedule was based on the students‘ pre-existing university/hospital group allocations, the previously scheduled group class timetable (which included BLS training) and the availability of the computer lab (for CD viewing). Participant allocation to a BLS training method by the university and hospital program co-ordinators ensured that the researcher, who assisted in the BLS assessments, was unaware of which training method was used with which participants. Participant characteristics. Two hundred and ninety four consenting second year nursing student and practising nurses participated in the study (see Figure 3.1). From the 231 participants who Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 81 | P a g e completed the Post Test 1 questionnaire, the majority were females (91%) aged 18 - 30 years (90%). All participants were working toward or held a degree. Most (91.8%) were computer literate and therefore capable of navigating the CD BLS training program irrespective of their allocation to the CD or Traditional group, and BLS was a required skill for all participants in this study (see Table 3.1). Table 3.1: Baseline characteristics for age, gender and computer literacy by cohort. CHARACTERISTICS COHORTS Novice Practising Nurses Combined % (n = 159) (n = 72) (n =231 ) n% n% n AGE 143 89.9 65 90.3 208 90.0 18 -30yrs 9.7 23 10.0 31+ 16 10.1 7 88.9 211 91.3 11.1 20 8.7 GENDER 147 92.5 64 Female 94.4 212 91.8 5.6 19 8.2 Male 12 7.5 8 COMPUTER LITERACY 144 90.6 68 Competent a Not Competent 15 9.4 4 a = competence represents 82% or above (9/11) score on 11 computer literacy questions The second year nursing students (n = 159) who comprised the Novice cohort, were assumed to have had no previous BLS training and were considered to be novices. However, on analysis of the demographic data in the questionnaire, (presented in Table 3.2), 59.7% of the Novice cohort had undertaken previous BLS training (separate to their university program) prior to participating in the study. There was no statistically significant difference in prior experience between those allocated to the CD and Traditional BLS training programs (Novices: CD 56%, Traditional 64%, χ2= 0.938, p = 0.333). Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 82 | P a g e Table 3.2: Chi-square tests of difference in previous BLS training between the CD and Traditional training groups. PREVIOUS BLS TRAINING COHORT TRAINING GROUPS NOVICE Total CD Traditional yes n % n %n% χ2 p 0.333 (n =159) (n = 87) (n = 72 ) NA 0.188 95 59.7 49 56.3 46 63.9 no 64 40.3 38 43.7 26 36.1 0.938 PRACTISING NURSES (n =72) (n = 34) (n = 38) yes 72 100 34 100 38 100 no 0 0 0 0 00 - COMBINED (n =231 ) (n = 121) (n =110) 84 76.4 yes 167 72.3 83 68.6 26 23.6 no 64 27.7 38 31.4 1.736 p ≤ 0.05; df = 1; NA = not applicable due to small cell sizes The Practising Nurses (n = 72) were in their first year of practice. They had previously learnt BLS skills as part of their undergraduate degree (see Table 3.2) and were therefore re-accrediting in BLS. Cohorts in this study were therefore essentially homogeneous in many variables (age, gender, computer literacy, and previous BLS training). Therefore, participant characteristics in this study were unlikely to influence comparisons between the BLS training methods. Procedure Training procedures. The BLS training programs were conducted, during allocated class time, at the participating university and hospital. In the Novice cohort (n= 187), 91 consenting participants received the BLS CD training program and 96 received the Traditional BLS training program. In the Practising Nurses cohort (n= 107), 53 consenting participants received the BLS CD training program and 54 received the Traditional BLS training program (see Figure 3.1). These numbers per cohort gave the study sufficient power to detect reliable differences between groups (see calculations in Appendix C). The BLS CD training program. During allocated class time, university and hospital program staff supervised the participants receiving the CD BLS training program. Participants were provided with a Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 83 | P a g e computer, the BLS CD (A&RMC, 1999) and a Laerdal Resusci Anne® manikin. The only instruction given by the university/hospital supervising staff was to work through the whole program practising on the manikin as they worked through each section of the CD. Participants had the opportunity to view the CD while practising alone and in pairs (for two-person BLS practice). Each participant documented the time spent in the training room (Appendix G). No further access to the CD or manikin was permitted after the training session, which ensured a controlled training time for the group. The basic life support CD. The BLS CD used in the study was developed in 1999 by Austin Health, a major metropolitan hospital in Melbourne Australia. (The development team included the current study‘s researcher, Karen Mardegan). The CD was an interactive multimedia program which uses voice, text, animated graphic images and video in an integrated way to provide viewers with the information traditionally covered in BLS training. The content of the BLS CD Program was divided into nine main sections and topics. The nine main sections covered were: 1. What is BLS? 2. What is an Emergency? 3. CPR; 4. Emergency Response Steps; 5. Anatomical Differences: infant/child/adult; 6. Emergencies in Health Care Settings; 7. Defibrillation; 8. Frequently Asked Questions; and 9. Self Check. Although the program was suitable for the general population, the CD also covered the broader issues required by health professionals when performing BLS, such as the use of the one-way valve mask (e.g. concord mask), bag-mask device (e.g. air-viva), defibrillation, and minijets for drug administration. The CD program was structured so Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 84 | P a g e that BLS beginners could complete each section in a recommended order. Viewers wishing to revise specific topics could also move around quickly using the section map. The two sections of CPR and Emergency Response Steps provided the complete instruction on how to perform BLS. The later sections build on this knowledge by introducing adjustments required for children and instruction on the equipment used by health professionals. In the CPR section of the CD program, viewers were trained in how to perform CPR. This was achieved by a combination of text, and short video sequences with voice- over to illustrate hand position, compression and ventilation techniques. This was followed by a longer video sequence with voice-over which illustrates the technique of CPR in its entirety. In the emergency response step section, viewers were then instructed on the whole BLS sequence commencing with checking for Danger, then for Response, Airway, Breathing and Circulation. The training of this information was also achieved by a combination of text, diagrams and video sequences with voice-over. The BLS CD was designed to teach the viewer BLS without manikin practice. It therefore did not prompt the viewer to practise on a manikin while working through the CD program. Practice has however been recommended in training design and delivery and resuscitation practice literature (Aguinis & Kraiger, 2009; ARC & NZRC, 2010a; Hazinski et al., 2010; Mancini et al., 2010; Salas & Cannon-Bowers, 2001). A number of studies which have examined BLS instruction with and without manikin practice have also found that when the opportunity for manikin practice was not provided as part of BLS training programs, adequate BLS skill levels were not acquired (Choa et al., 2006; Reder et al., 2006; Teague & Riley, 2006). Therefore, as it appears useful to provide manikin practice in all forms of BLS training, this study‘s CD group had access to both the CD and a BLS Laerdal Resusci Anne® manikin for unsupervised practice. In contrast to Traditional BLS training, the CD program focused only on the critical elements of BLS training for lay people and health professionals. Therefore, more time for hands-on practice and performance of BLS skills was available. In Traditional classroom courses trainees often share a manikin and have minimal practice time. With the CD-based program a trainee can practise on their own manikin while working through the CD and in pairs for two-person BLS practice. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 85 | P a g e Traditional BLS program. The Traditional BLS program comprised the BLS teaching program used by the respective participating organisations at the time of the data collection. These two Traditional programs were therefore not standardised across the two cohorts and there may have therefore been some instructor delivery differences from within each cohort and across the cohorts. However, BLS certified instructors (from the participating university and hospital) delivered the Traditional program during allotted class time. This ensured that the researcher and supervisors of the research study had no involvement in the delivery of the Traditional program. The training consisted of a BLS presentation and a practical demonstration of BLS followed by instructor-supervised manikin practice. The instructor outlined the emergency response steps, potential complications, health professional responsibilities and the use of equipment (bag-mask device, one-way valve mask, and artificial airway). The emergency response steps and the use of any equipment were demonstrated by the instructor using a BLS Laerdal Resusci Anne® manikin. Participants were then supervised practising BLS on a manikin. The participants in the Traditional group were divided into groups of 8-10 participants per instructor. One Laerdal Resusci Anne® manikin per 4-5 participants was available for the manikin practice component of the program. Participants had the opportunity to practise both alone and in pairs (for two-person BLS training). CD and Traditional BLS program content and length. All materials and apparatus used in the CD and Traditional programs were designed to comply with the ARC guidelines at the time the study was conceptualised (ARC, 1997), not the most recently advocated changes to practice (ILCOR, 2000b, 2005; Hazinski et al., 2010; Sayre et al., 2010). The BLS algorithm and compression- ventilation ratio taught and assessed was DRABC with a ratio of 1:5 ventilation to compressions for one operator and 2:15 for two operators (ARC, 1997; see Figure 3.2). Practising health professionals are required to use one-way valve masks and bag- mask devices, such as concord masks and air-vivas to administer the ventilation component of BLS. Therefore, although many studies evaluate BLS programs using mouth-to-mouth techniques, the training programs evaluated in this study assessed participants‘ ability to perform BLS using the equipment required in practice by health professionals. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 86 | P a g e The Novice cohort (cohort A) who were learning BLS for the first time received two hours of BLS training time. The Practising Nurses (Cohort B) who were re- accrediting in BLS, having some experience in BLS, received one hour of BLS training time. These (one or two hour) training times were based on the length of the existing Traditional programs at the two participating organisations. Post Test procedures. Two post tests were conducted at the participating university and hospital to compare acquisition of BLS skills and knowledge and retention over two months. As shown in Figure 3.1, Post Test 1, which was conducted one week after the completion of the BLS skills training program, comprised: 1. Completion of the questionnaire (which contained participant demographic, computer literacy, BLS experience and BLS knowledge questions), immediately prior to the 1st BLS skills assessment. 2. Completion of the 1st assessor-conducted BLS skills assessment. 3. Completion of the program evaluation form immediately after the 1st BLS skills assessment. To assess BLS skills, each participant was invited into a room with a trained assessor (who was blind to the training program completed by the participant), and a Laerdal Skill Reporter™ Resusci Anne® manikin (Laerdal, 2002). The assessment followed the standardised procedures used at Austin Health to assess BLS skills of health professionals. Participants were asked to imagine that the manikin was a person who had just collapsed, and to perform exactly as they would in real life. After one-person BLS was demonstrated, the assessor joined the participant in two-person BLS. After two minutes, participants were told to cease resuscitation efforts. The assessor graded each individual BLS skill step as competent or not competent. After the assessor had completed the skill assessment rating, the printed readout of CPR performance from the automated manikin used in the assessment, was obtained. Certified Austin Health BLS assessors conducted the BLS assessments in the post tests. A minimum of four certified assessors (the researcher and three research assistants) per cohort (therefore seven assessors in total) were necessary to ensure that the number of Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 87 | P a g e post test skill assessments could be completed within time and access constraints. To ensure that the inter-rater reliability of the assessors participating in the current study was suitable high, a proportion (17%) of BLS assessments were simultaneously assessed by the researcher and assessors. These dual assessments were performed after the training component of the programs, at the commencement of the post test assessments. The respective course co-ordinators for each cohort assigned participants to an assessor based on student availability. There was 100% agreement in the competent/not competent rating and ordinal scale grading (1=not competent to 5 = outstanding competence) of the dual assessments (see Appendix D). Assessor inter-rater reliability was therefore judged to be suitably high. One hundred and eighty seven consenting participants from the Novice cohort, (91 from the CD and 96 from the Traditional BLS training group), attended Post Test 1. One hundred and seven consenting participants in the Practising Nurses cohort, (53 from the CD and 54 from the Traditional BLS training groups), attended Post Test 1 (see Figure 3.1). There was a longer latency between training and Post Test 1 for the CD versus the Traditional groups (Combined: CD Group M = 6.34 days, SD = 3.19; Traditional Group M= 5.06, SD = 1.25) which was statistically significant (t(286) = 5.74, p < 0.001 [see Appendix E]), however in practical terms the difference between five and seven days was judged to be not so large as to impact on recall at the post-test assessments. At Post Test 2, which was conducted two months after the completion of the BLS skills training program, participants were required to: 1. Complete a shortened questionnaire (containing only the BLS experience and BLS knowledge questions), immediately prior to the 2nd BLS skills assessment. 2. Complete the 2nd assessor conducted BLS skills assessment. Based on feedback from the pilot study (presented later in this chapter), the background questions which were essentially stable over time (demographics and computer literacy sections), were removed from the questionnaire when administered at Post Test 2 so as to reduce the time required to complete the questionnaire (see Appendix F1 and Appendix F2). The BLS experience and knowledge sections of the questionnaire were re-administered at Post Test 2 to track any changes in BLS experience and knowledge over time. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 88 | P a g e Within the Novice cohort (n=106), 55 participants from the CD and 51 from the Traditional BLS training group attended Post Test 2 (60% of CD and 53% of Traditional Post Test 1 cohorts respectively). Within the Practising Nurses cohort (n= 35), only 23 from the CD and 12 from the Traditional BLS training groups attended Post Test 2 (43% of CD and 22% of Traditional Post Test 1 cohorts respectively, see Figure 3.1). There was no statistically significant difference in the number of days between training and Post Test 2 for the CD and Traditional groups (Combined: CD M = 59.08 days, SD = 2.88, Traditional M = 59.07 days, SD = 1.86, t(133) = 1.32, p > 0.180 [see Appendix E]). Measures There were four measurement tools used for the post tests. These were in order of administration: (a) BLS knowledge and participant characteristics were measured via a questionnaire, (b) Overall BLS skill competence was measured by an assessor using a BLS assessment form, (c) Specific CPR skills, were measured in conjunction with the assessor rating using an automated manikin, and (d) Program evaluation forms measured participants‘ rating of the BLS training program undertaken (see Figure 3.1). Questionnaire. The questionnaire used in Post Test 1 and 2 has been included as Appendix H. It was constructed by the researcher and contains four sections: demographic, computer literacy, BLS experience and BLS knowledge. The demographic and computer literacy sections of the questionnaire were drawn from the surveys designed by Wilkinson and Chu (1999) for CD-based and face- to-face delivery of education. The questions in the BLS experience and knowledge sections of the questionnaire were based on those in founding BLS Video studies (Braslow et al., 1997; Todd et al., 1998, 1999) and the content of the Traditional and CD programs in the current study. The resultant questionnaire was a broadly focused tool aimed at capturing participant characteristics Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 89 | P a g e (age, gender, educational level, employment), computer literacy, BLS experience and BLS knowledge. Demographic and computer literacy sections of the questionnaire. The demographic and computer literacy sections of the Wilkinson and Chu (1999) surveys were the same for the CD-based and face-to-face program surveys. These included six questions on the demographics of age, gender, highest educational level, study status, current employment and employment type; three questions on current need for BLS skills; and 15 questions on computer literacy issues such as access, frequency of use and experience, and a single question on preferred education mode (face to face lectures, paper based self-directed learning packages, packages on CD [see Appendix F1]). The questions in the demographic and computer literacy sections of the questionnaire which have been analysed and presented as part of the thesis are: the question on Age group which was divided into five categories (18 - 20, 21 - 30, 31 - 40, 41 - 50, 51 and above years of age); and gender (male or female). Participants were asked to tick the most appropriate box for both these questions (see Appendix F1). The computer literacy questions (see Appendix F1) required participants to tick the most appropriate box for: the frequency of computer use (not at all, occasionally [less than once a week], once a week, variable number of times per week, once a day, all the time); access to a computer at home and 11 computer usage experience questions (yes/no). The computer usage questions were: computer experience, beginning exploration of computers, can use one computer program for general purpose, can use one computer program proficiently, can use two or more programs for general purpose, can use two or more programs proficiently, can transfer data, can use internet for general purpose, can use internet proficiently, can design computer applications and expertise in computer application design. Responses to 11 computer usage experience questions were summed to obtain an overall computer literacy score out of eleven (see Appendix F1). The remaining demographic (and computer literacy sections of the questionnaire questions), which were not analysed as part of this thesis, were either defined by the recruitment process (i.e. education level, current studies/employment, and BLS required for employment/studies), or did not identify any differences between the groups (i.e. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 90 | P a g e whether they were living with a person at high risk of needing BLS, and performance of BLS in an emergency). BLS experience and knowledge sections of the questionnaire. BLS experience questions. The eight questions in the BLS experience section which have been analysed and presented as part of the thesis include: previous BLS training (yes/no) response from participants; and the BLS skill post training question which required participants to rate their skill on a five point ordinal scale (very low, low, neutral, high and very high [see Appendix F1]). These participant gradings were analysed to determine if there was significant difference in participants‘ rating of their BLS skill post training. The remaining questions in the BLS experience section of the questionnaire were not presented as part of this thesis because they were not central to the aims of the current study and did not identify any differences between the groups (i.e. performance of BLS in an emergency during and after project, confidence to perform BLS in an emergency, and practice prior to assessment). They are however available in Appendix F1. BLS knowledge questions. As the BLS knowledge questions in the Todd et al. (1998, 1999) studies were not specified, therefore the six BLS knowledge questions were derived by the researcher from the content covered in the Traditional and CD programs in the current study (see Appendix D1). The content validity of the developed questions were supported by an independent BLS training expert and the six BLS skill assessors who agreed that all important areas of knowledge were addressed by the set of questions developed. The six BLS knowledge questions were: the defining of respiratory and cardiac arrest, causes of these, the potential complications of CPR and the most common drugs administered in arrests (see Appendix F1). Participants were required to provide written answers to all six BLS knowledge questions. These participant answers were matched against the correct answer (see Appendix F3). Replies were graded as correct or incorrect by the researcher and summed to give an overall BLS knowledge score out of six (see Appendix F3). Internal consistency of the questionnaire. No validity and reliability information was provided by Wilkinson and Chu (1999) in respect to their surveys, so the internal consistency for the demographic and computer Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 91 | P a g e literacy questions was unknown. The Cronbach‘s alpha test of scale internal consistency for the BLS experience questions was reported by the Todd team as α = 0.77 (Todd et al., 1998). This suggests adequate internal consistency for the BLS experience questions. Considering that the questionnaire was constructed from the tools used in other studies (Braslow et al., 1997; Todd et al., 1998, 1999; Wilkinson & Chu, 1999) and that these tools were not fully validated, it was deemed necessary to evaluate all the tools compiled for the current project prior to embarking on the current study. To do so a pilot feasibility study was conducted. Results from this pilot study are presented in the later part of this chapter. BLS assessment form. The BLS skill assessment form used in the current study was the standard form used to assess health professionals at Austin Health (A&RMC, 2000) at the time of the data collection (see Figure 3.2). The broad categories of Danger, Response, Airway, Breathing, and Circulation (DRABC) and the specific skills within these categories were listed in the form. The BLS skill steps illustrated in the form were consistent with ARC guidelines at the time of the data collection (ARC, 1997), and therefore similar to those described in previous key studies in this area (Braslow et al., 1997; Brennan et al., 1996; Todd et al., 1998, 1999), with the addition of extra health-professional-related skill steps, e.g. features and demonstration of the use of the one-way valve mask (i.e. concord), demonstration of use of the bag-mask device (i.e. air-viva), insertion of a guedel airway (see Figure 3.2), which are the pieces of equipment required for health professional practice in Australia. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 92 | P a g e Figure 3.2. Nurses – Basic Life Support assessment form Karen Mardegan Basic life Support training for nurses: evaluating an alternative CD-based approach

Chapter 3 — Method 93 | P a g e The bolded skills in the assessment form were the skills considered mandatory for competent performance of BLS (Austin & Repatriation Medical Centre [A&RMC], 2000; see Figure 3.2). Non-mandatory (unbolded in Figure 3.2) skills were considered desirable but not essential to the competent performance of BLS. Participants who demonstrated all bolded (mandatory) skills correctly were graded by an assessor as achieving overall BLS skill competence. Overall BLS skill competence was then rated on a five point ordinal scale (5 outstanding = all bolded and all unbolded skills correct; 4 very good = all bolded skills and the majority of unbolded skills correct; 3 competent = all bolded skills correct but not the majority of unbolded skills correct; 2 = questionable competence = majority of bolded skills not achieved but the majority of unbolded skills correct; 1 not competent = majority of bolded and majority of unbolded skills not correct). This ordinal scale which replicated the grading scale in the Brennan et al. (1996) BLS assessment tool, was added to the assessment form used in the current study by the researcher to allow for a finer grading of overall competence. Participants who rated 3 or above (therefore all mandatory skills correct) on this five point scale achieved overall skill competence in the BLS skills assessment (see Figure 3.2). Laerdal Skill Reporter™ Resusci Anne®. The Laerdal Skill Reporter™ Resusci Anne® (automated) manikin was used as an independent measure of CPR skills. It recorded the average frequency and depth of chest compressions and the average frequency and volume of ventilation during CPR. In addition the manikin also documented the compression/ventilation ratio and any improper hand positioning during compressions i.e. wrong hand position, hand position too low and incomplete release. A printed readout of (one-operator) CPR performance was obtained from this automated manikin, after the assessor had completed the assessment rating on the BLS assessment form. Program evaluation forms. Two program evaluation forms were compiled, one specifically for those who undertook the Traditional BLS program and another form for those who undertook the CD BLS training program. The program evaluation questions were drawn from the Wilkinson and Chu (1999) staff education surveys for ―face to face‖ and ―CD-based‖ programs. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 94 | P a g e Questions common to both program evaluation forms. The questions which were common to both the Traditional BLS program and CD BLS program evaluation forms were questions on participant views about the program content, structure, assessment component and overall quality and satisfaction (see Appendix H). The content section included eight questions relating to whether the content of the respective programs was considered to be up-to-date, relevant, of an appropriate breadth, and complexity, and that simulations and scenarios were useful. The structure section contained six questions about the organisation and structure of the topics as well as the sequencing of the information. Two questions relating to the overall quality and participant satisfaction with the program and two questions related to the assessment components (i.e. format and appropriateness of the practical assessment) were also asked of both groups (see Appendix H). The questions in these sections of the evaluation forms required a five point ordinal scale grading (strongly disagree, disagree, neutral, agree, strongly agree) from participants, which was consistent with that used in comparable studies (Moule et. al., 2001; Moule & Gilchrist, 2001; Moule, 2002). For some analyses, the strongly agree/agree and disagree/strongly disagree/ neutral responses were grouped into two categories, so that a dichotomous rating per question was able to be obtained (see Appendix H). Additional questions. The Traditional and CD program evaluation forms each contained five additional unique questions. For the Traditional program, evaluation form, the additional questions were related to the content, pace and usefulness of the instructor-supervised manikin practice (see Appendix H1). For the CD program, the additional questions related to the adequacy of access and viewing time, comparison between CD and face to face program quality and effectiveness, and the potential need for additional support with CD programs (see Appendix H2). These additional questions in the program evaluation forms were not central to the study aims because they did not enable comparison across training methods. Therefore they were not reported in the current study, but are available in Appendix H. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 95 | P a g e Internal consistency of the program evaluation tools. Cronbach‘s alpha test of scale internal consistency was calculated for the CD and Traditional groups‘ program evaluation forms. Good internal consistency was demonstrated (CD program evaluation form α = 0.905 and Traditional program evaluation form α = 0.954, see Appendix H3). To assist further evaluation of all the tools compiled for the current project, including the program evaluation forms, a pilot study (presented in the later part of this chapter) was conducted prior to embarking on the main study. Data Analysis The data generated from this quantitative study were entered into a computer database and then analysed using the Statistical Package for the Social Sciences (SPSS) 15.0 for Windows Graduate Pack. To reduce errors the data were entered twice and compared. Using this approach, four errors in data entry were identified and corrected. Data analysis was performed on the Traditional and CD group data within the Novice cohort, Practising Nurses cohort, and the cohorts combined to provide a comprehensive set of findings. Exploration of the data associated with the main study variables indicated that the data generally was not normally distributed, and therefore it was decided not to use the planned parametric tests for identifying group differences (t test, F test). Consequently, the main analysis performed was to identify associations between categorical variables using the chi-square test. Sample size calculation and power analysis. Sample size calculations for the study were determined by applying Cohen‘s procedures (Cohen, 1988), and on the basis of being comparable to those used in similar studies where statistically significant effects have been demonstrated (Braslow et al., 1997; Todd et al., 1998, 1999). Sample size calculations were conducted based on a power of 0.8 (Cohen, 1988) with an effect size of 0.6, and setting alpha at 0.05. This corresponds to a total sample size of 88 for each cohort or 44 per group within cohorts. To achieve 44 per group at the data analysis stage, it was necessary to recruit at least 50 per group initially (i.e. a total of at least 100 per cohort to allow for drop outs, unusable data etc [see Appendix C]). These participant numbers were achieved on recruitment to the study (see Figure 3.1). The effect size of 0.6 was based on the related studies at the Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 96 | P a g e time the project was planned (Batcheller et al., 2000; Braslow et al., 1997; Todd et al., 1998, 1999) which found significant difference between training methods at comparative sample sizes. More recent studies (Cason et al., 2009; Choa et al., 2006; Chung et al., 2010; Isbye et al., 2006; Jones et al., 2007; Kardong-Edgren et al., 2010; Moule et al., 2008a; Reder et al., 2006; Roppolo et al., 2007) have not been able to replicate this statistical difference suggesting that the chosen effect size of 0.6 may in hindsight have been too high, and the sample sizes needed to be greater. Questionnaire. The questionnaire responses for the two groups within each cohort were compared. Frequencies, percentages and chi-square analyses were performed. This allowed for a test of statistical difference for each question and each section of the questionnaire (demographics, computer literacy, BLS experience, BLS knowledge sections). Analysis of participants‘ age, gender, computer literacy, previous BLS training, and participant rating of BLS skill post training, which was related to overall BLS skill, were analysed because they were relevant to the key aims of this study. Age group. Age group was divided into categories and coded (1 = 18 - 20, 2 = 21 - 30, 3 = 31 - 40, 4 = 41 - 50, 5 = 51 and above years of age). Preliminary descriptive analysis was performed, and age categories were collapsed to two groups, 18 - 30 and above 31 years to produce more adequate cell sizes. These categories were compared by performing a chi-square analysis to determine if there was significant difference in participants‘ age between the groups, and cohorts. Gender and previous BLS training. The questions on gender (female/male) and previous BLS training responses were coded 1 = yes and 0 = no (see Appendix F1). Frequencies, percentages and chi-square analyses were performed to identify any significant difference between the training groups for these measures. Overall computer literacy. To determine the overall level of computer literacy in the groups and any significant differences between them an overall score was calculated from summing the 11 computer usage experience questions (1= yes and 0 = no, see Appendix F1). Overall computer Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 97 | P a g e literacy was set at 82% (a score of 9/11 or above), reflecting generally proficient computer skill. The last two questions on ability to design computer programs was an advanced computer skill well above the level of computer literacy required to navigate the CD BLS program (see Appendix F1). The scores were coded 1 = ≥ 9/11, 0 = ≤ 8/11, and frequencies, percentages and chi-square analysis was performed to determine if there was significant difference in computer literacy across the groups and cohorts. Participants’ self-rating of BLS skill post training. Participants‘ ratings (on an ordinal scale from very low to very high) of their BLS skill post training (in the BLS experience section of the questionnaire) were analysed and found to be not normally distributed, which prohibited parametric (t-test) analysis of responses. Consequently responses were coded as 1 if participant‘s rated their skill as very high or high and 0 if they rated their skill as very low, low or neutral. Frequencies, percentages, and chi-square analyses were applied to these gradings to provide an indication of program effectiveness, and determine if there were significant differences between the CD and Traditional groups in participants‘ rating of their BLS skills post training. BLS Knowledge. The BLS knowledge data and the BLS skill data have been analysed similarly. Overall BLS knowledge. The six BLS knowledge questions were graded as correct (coded 1) or incorrect (coded 0). An overall score out of 6 was calculated. Participants were coded as 1 (adequate knowledge) if they scored 66% (4/6) or above and 0 (inadequate knowledge) if they scored below 66%. To determine program effectiveness and identify any significant difference in overall BLS knowledge between the CD and Traditional groups, chi-square analyses on group frequencies were applied for Post Test 1 and Post Test 2, for each of the Novice, Practising Nurses, and the Combined cohorts. BLS knowledge of each question. Frequencies, percentages, and chi-square tests were also calculated on the correct/incorrect grading for each of the six BLS knowledge questions to identify any significant differences between the groups and cohorts for each specific question. As small cell sizes limited the ability to statistically analyse many BLS questions, only the Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 98 | P a g e Combined Novice and Practising Nurses cohort results for each knowledge question were presented in the results chapter. There was a potential for a type 1 error in the analysis of the number of BLS knowledge questions. Bonferroni adjustment was therefore calculated, and a consistent p value of 0.001 was applied. Retention of BLS knowledge. Retention of overall BLS knowledge was determined by coding each participant as 1 if their overall score at Post Test 2 was the same as or better than at Post Test 1, and 0 if their overall score at Post Test 2 was below their Post Test 1 score. Percentages, frequencies and chi-square analyses of difference were calculated for the Novice, Practising nurses and the Combined cohorts to determine retention of overall BLS knowledge and therefore an indication of program effectiveness. Approximately 1% of questions in the questionnaire were not answered by participants. There are a number of established techniques for managing missing responses e.g. listwise deletion, pairwise deletion, mean imputation, full analysis (Pallant, 2007). As there are a number of possible explanations for a question not being answered (e.g. answer unknown, question accidentally overlooked), and only a small number of responses were missing in the questionnaire data, these missing responses were replaced with the mean response for that question. BLS skills assessment form. Overall BLS skill competence. The mandatory (bolded) skills in the BLS assessment form were analysed to determine BLS skill competence. Firstly, an overall BLS skill score (out of a maximum of 32 mandatory BLS skills) were calculated for each participant for Post Test 1 and Post Test 2 from the assessor grading of each individual skill on the assessment form (coded 1 = competent, 0 = not competent). Overall BLS skill competence was set at 100% performance on mandatory skills (a score of 32/32). To determine program effectiveness and identify any significant difference for overall BLS skill competence between the CD and Traditional groups‘ frequencies, and chi-square analyses were applied for Post Test 1 and Post Test 2 for the Novice, Practising Nurses and the Combined cohorts. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 99 | P a g e BLS skill categories and specific BLS skills. Frequencies, percentages, and chi-square tests were also calculated on the competent/not competent grading for each BLS skill within the BLS procedure. Both mandatory and non-mandatory BLS skills were included in this analysis. The specific skills were grouped under the headings of initial response steps, ventilation, circulation and Health Professional skills. Where cell sizes allowed (> 5), overall frequencies, and chi-square analyses were calculated to identify any significant differences between the groups and cohorts for these skill categories and each specific skill. As small cell sizes limited the ability to statistically analyse many of the skill categories and specific BLS skill steps, only the Combined Novice and Practising Nurses cohort results for the BLS skill categories and specific skills were presented. Bonferroni adjustment was calculated to control for type 1 error in these measures and a p value of 0.001 was applied. Retention of BLS skill level and competence. Frequencies and chi-square analyses of difference were calculated for the Novice, Practising Nurses and Combined cohorts, to determine retention of overall BLS skill level and therefore an indication of program effectiveness. To determine retention of overall BLS skill competence, those who were competent at both Post Test 1 and Post Test 2, and those who were competent at Post Test 1 but not competent at Post Test 2 were then compared by performing a chi-square analyses to determine if there were significant differences in retention of overall BLS skill competence over time. Those who were not competent at Post Test 1 (and consequently received additional teaching) were excluded from the analysis of retention of skill competence. Results are presented for each of the Novice, Practising Nurses and the Combined cohorts. There were small cell sizes (< 5) in the retention data for the Practising Nurses cohort, so retention results were interpreted with caution. Laerdal Skill Reporter™ Resusci Anne® Printout. Although it was the original intention to include the data received from the automated manikin, it became apparent that a number of factors had compromised the validity of this data. On many occasions the manikins did not consistently produce a read-out or produced a read-out that did not correspond to the assessors‘ observations and therefore implied inadequate sensitivity of the manikin. These limitations in the manikin Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 100 | P a g e data resulted in only a small amount of data being available for analysis which was not convincingly representative of the population‘s CPR performance. On this basis it was decided not to use this data in any of the analyses. Program evaluation forms. Participants‘ ratings of their BLS training program in the program evaluation form, were found to be not normally distributed. This prohibited parametric (t-test) analysis of responses. Consequently, for the questions that were common to both the CD and Traditional program evaluation forms (i.e. content, structure, assessment and overall quality and satisfaction questions), participants‘ ratings were coded as 1 if participants strongly agreed/agreed and 0 if they strongly disagree/disagreed/neutral with these questions. Each question, each group of questions, and replies overall were summed and compared by performing chi-square analyses to determine significant differences in participants‘ rating of their BLS training program between the groups and cohorts. As some small cell sizes (< 5) limited the ability to statistically analyse many responses (particularly the Practising Nurses cohort), only the Combined Novice and Practising Nurses cohort replies for the specific and grouped questions were presented. Bonferroni adjustment was applied to control for type 1 error in these measures and a p value of 0.001 was applied. Any unanswered questions (< 1%) were treated as missing responses and replaced with the mean response, as previously described. The additional questions in the program evaluation forms which were not common to both program evaluation forms were not addressed in this study (see Appendix H). Ethical Considerations There was an ethical responsibility to ensure that participants met their respective (university/hospital) program requirement and corresponding patient care requirement of BLS accreditation. Those who chose to participate in the project met the university / hospital program requirement of BLS accreditation by virtue of their study participation. Those who chose not to participate in the study (Novices n = 13, Practising nurses n = 3) completed the standard BLS training (Traditional method) and were assessed as per program requirements along with study participants. Those in either instructional method who were not competent at the BLS assessment in the week immediately following Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 101 | P a g e training (Post Test 1) were given further instruction using the Traditional method and reassessed, to ensure that every participant demonstrated adequate BLS skills and met the university/hospital program requirement. Record of participants‘ BLS competence was provided to the university / hospital on successful completion of the BLS assessment. These strategies ensured that withdrawal from the study at any time was possible without fear of negative consequence. The additional strategies (i.e. coding of participant names, secure data storage, and anonymity on publication) necessary to ensure maintenance of participant confidentiality during and following completion of the study have been outlined in Appendix A4. Trialing of Materials Prior to embarking on the main study, it was judged useful to conduct a pilot feasibility study for a number of reasons. Firstly, BLS is a life saving skill and the available literature on the effectiveness of CD BLS training programs is limited. If the pilot study, demonstrated that the BLS CD training program was not able to produce adequate BLS skills, then it would be unethical to embark on a larger scale study, especially when those partaking in the study require this skill to safely perform their work role. Second, the questionnaire and program evaluations planned to be used in the study were compiled primarily from surveys developed for a study proposed but not conducted by Wilkinson and Chu (1999). The BLS assessment form which had been widely used to assess staff at the participating hospital had not been previously used for research purposes. The tools were therefore considered to be lacking validation. A pilot study provided the opportunity to initially evaluate and validate test procedures and tools prior to use in the main study. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 102 | P a g e Figure 3.3 Pilot Study Design Karen Mardegan Basic life Support training for nurses: evaluating an alternative CD-based approach

Chapter 3 — Method 103 | P a g e Design of pilot study. The pilot study compared two modes of BLS training in a small novice sample of twenty nursing students, using a similar design and novice cohort as the main study. There was no-pre-test, but two post-tests, one conducted two weeks after training, and the second post-test conducted ten weeks after training. The two training groups were a BLS CD training program which included a manikin for unsupervised practice and a Traditional BLS (presentation demonstration/practice) program. Key outcome measures were as planned for the main study. The research design of this pilot study, described above, has been summarised in Figure 3.3. The pilot study procedures have been outlined in Appendix I. It commenced in March 2002 and was completed by July 2002. Pilot study results. Measures. The BLS assessment form, questionnaire and program evaluation questions used in the pilot study were completed correctly and questions answered appropriately, which implied that the forms were clear and understood by participants. Based on feedback from the pilot study, the background questions (demographics and computer literacy sections) in the questionnaire were removed when administered for the second time in the main study, to reduce the time required to complete the questionnaire in the main study (see Appendix F1 and Appendix F2). Removal of these questions in the Post Test 2 questionnaire in the main study was deemed reasonable because the information relevant to these particular questions would not change over the short time period involved. On analysis of the Laerdal Skill Reporter™ Resusci Anne® manikin recordings, (which were taken in conjunction with the BLS assessors grading) of participant CPR performance, many of the printouts were incomplete or produced results which appeared incorrect, when compared with the human-assessor grading. This suggested that the skill recording manikins were potentially inaccurate. However, the manikins used in the pilot were the most sophisticated performance recording manikins available at the time of the data collection. It was therefore decided that the manikins would still be used as planned in the main study, despite these accuracy concerns having been raised. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 3 — Method 104 | P a g e BLS competence. Twelve second year nursing students attended the pilot post test assessments (see Figure 3.3). On testing of the training programs and assessment procedure, 75% of the Pilot CD BLS group and 50.0% of the Pilot Traditional BLS group demonstrated competent BLS skills at 2 weeks post training. The degree of BLS skill competence of participants implies that both training programs were potentially able to produce BLS skills in the majority of trainees. This outcome was also comparable with the outcomes of other Traditional BLS training programs which typically produced from 45% - 74% competent trainees (Gasco et al., 2000; Kallestedt et al., 2010; Woollard et al., 2004). It therefore appeared feasible to attempt to train nurses in BLS via a CD BLS training method, thus justifying investigation of training technique differences via the planned larger main study. Implications of the pilot study. The pilot study provided information on the utility of the BLS training methods and measures. It also provided an indication of the potential utility of CD BLS training. When the CD BLS training method was evaluated in the 12 second year nursing students who attended the pilot post test assessments, the small sample sizes precluded statistical analysis. Nevertheless, overall judgments on the feasibility and required adjustments to the proposed method, measures and training programs were possible from the pilot study. Only minor adjustments to the planned procedure were required. Presentation of the findings of the main study now follows. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 4 Results The results from this study which evaluated a CD-based and Traditional BLS training program in Novice and Practising Nurses will now be presented. The primary aim of this study was to compare the BLS skill of Novice and Practising Nurses in the two groups at one week and again at two months post training. The secondary aims were to evaluate participants‘ BLS knowledge and participants‘ rating of the CD and Traditional BLS training programs undertaken. Therefore firstly, overall effectiveness of the BLS training (regardless of training method), in Novice and Practising Nurses will be presented. Then the results from the comparison of the CD and Traditional training in respect to BLS skill competence and adequacy of BLS knowledge at one week, and two months, post training are presented. This is followed by the presentation of skill and knowledge retention results, and participants‘ self-rating of their BLS skill post training. Finally are presented the participants‘ evaluative rating of the CD and Traditional training programs. As the main statistical test used in the results following was a non-parametric one (chi-square test), descriptive statistics for each group‘s scores on the main study variables are not presented here but are included in Appendix J. The Effectiveness of the BLS Training for Novice and Practising Nurses Firstly, to gain an understanding of the effectiveness of BLS training by experience level, the percentage competent in terms of adequate BLS skill and knowledge for the Novice, Practising Nurses and Combined cohorts are described in Table 4.1. When applying the skill mastery and program effectiveness standard of at least 80% competence following training (Fabius et al., 1994; Frieson & Stotts, 1984; Marzooq & Lyneham, 2009; Morrison et. al.,1996; Wayne et al., 2005, 2006), the low BLS skill and knowledge competency rates presented in Table 4.1 illustrate overall low training effectiveness and poor retention. Basic life support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 4 — Results 106 | P a g e Table 4.1: The percentage competent for the performance of BLS skill and knowledge of the Novice, Practising Nurses and Combined cohorts overall. BLS SKILL & KNOWLEDGE COMPETENCE Novice COHORTS Combined n% Practising Nurses n% n% (n = 294) 174 59.2 BLS SKILL (n = 187) (n = 107) Post Test 1 101 54.0 73 68.2 (n = 141) 58 41.1 Competent (n = 35 ) 15 42.9 (n = 141) Post Test 2 (n = 106 ) 76 54.0 Competent 43 40.6 (n = 35) 18 51.4 (n = 193) Retention (n = 106) 51 26.4 Same or Better 58 54.7 (n = 66) 39 59.1 (n = 112) BLS KNOWLEDGE (n = 127) 20 17.9 Post Test 1 12 9.4 (n = 31) 19 61.3 (n = 112) Adequate 61 54.5 (n = 31) Post Test 2 (n = 81) 19 61.3 Adequate 1 1.2 Retention (n = 81) Same or Better 42 51.9 Note: retention numbers based on raw scores not competence. For BLS skill, at Post Test 1, 54.0% of the Novices, 68.2% of the Practising Nurses, and therefore 59.2% overall for the cohorts, were competent at Post Test 1 (one week after training). At Post Test 2, 40.6% of the Novices, 42.9% of the Practising Nurses, and 41.1% overall, were competent at Post Test 2 (two months post training). For retention of BLS skill, 54.7% of the Novices, 51.4% of the Practising Nurses, and 54.0% overall, were graded the same or better (at Post Test 2). For BLS knowledge, at Post Test 1, 9.4% of the Novices, 59.1% of the Practising Nurses, and therefore 26.4% overall for the cohorts, were able to answer at least 4 out of the 6 (66%) BLS knowledge questions correctly. At Post Test 2, 1.2% of the Novices, 61.3% of the Practising Nurses, and 17.9% overall, were able to answer at least 4 out of the 6 (66%) BLS knowledge questions correctly. The results were therefore particularly poor for the Novice student nurses‘ knowledge at both Post Test 1 and Post Test 2. For retention of BLS knowledge, 51.9% of the Novices, 61.3% of the Practising Nurses, and 54.5% overall, were graded same or better (at Post Test 2). Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 4 — Results 107 | P a g e Overall these results indicate that there was low BLS training effectiveness for the Novice, Practising Nurses and Combined cohorts overall, with only about 50% of trainees reaching and retaining BLS skill and knowledge competence post training. Evaluation of BLS Skill for the Two Training Methods The primary aim of the study compared the BLS skill of the Novice and Practising Nurses who undertook the CD and Traditional BLS training programs when skill was assessed at one week and again at two months post training (to evaluate retention of skill). Overall BLS skill competence. The overall BLS skill competence at Post Test 1 (one week after training) and at Post Test 2 (two months after training) for the two training methods was therefore examined, with competence defined as 100% performance of the 32 mandatory skills within the 49 skill BLS assessment form used (see Figure 3.2). Overall BLS skill competence at Post Test 1. Overall BLS skill competence at Post Test 1 is presented in Table 4.2. There was low overall BLS skill competence with no statistically significant differences (at p ≤ 0.05) between those who undertook the CD program and those who undertook the Traditional program for the Novice cohort, Practising Nurses cohort and Combined cohorts. Table 4.2: Chi-square tests of difference between the CD and Traditional training methods in BLS skill competence at Post Test 1 for the Novice, Practising Nurses and Combined cohorts. BLS SKILL COMPETENCE POST TEST 1 COHORT TRAINING GROUPS CD Traditional n%n% χ2 p NOVICE (n = 91) (n = 96 ) Competent 48 52.7 53 55.2 Not Competent 43 47.3 43 44.8 0.114 0.736 PRACTISING NURSES (n = 53) (n = 54) Competent 33 62.3 40 74.1 Not Competent 20 37.7 14 25.9 1.721 0.190 COMBINED (n = 144) (n = 150) Competent 81 56.3 93 62.0 Not Competent 63 43.8 57 38.0 1.006 0.316 Note: competence represents 100% performance on 32 mandatory skills; df =1; p ≤ 0.05 Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 4 — Results 108 | P a g e For the Novices, 52.7% of the CD group and 55.2% of the Traditional group were competent at Post Test 1 (χ2 = 0.114, p = 0.736). For the Practising Nurses, 62.3% of the CD group and 74.1% of the Traditional group were competent at Post Test 1 (χ2 = 1.72, p = 0.190). When the cohorts were combined, 56.3% of the CD group and 62% of the Traditional group were competent at Post Test 1 (χ2 = 1.006, p = 0.316). Overall BLS skill competence at Post Test 2. Overall BLS skill competence at Post Test 2 is presented in Table 4.3. There was very low overall BLS skill competence with no statistically significant differences (at p ≤ 0.05), between those who undertook the CD program and those who undertook the Traditional program for the Novice, Practising Nurses and Combined cohorts. Table 4.3: Chi-square tests of difference between the CD and Traditional training methods in BLS skill competence at Post Test 2 for the Novice, Practising Nurses and Combined cohorts. BLS SKILL COMPETENCE POST TEST 2 COHORT TRAINING GROUPS CD Traditional n % n % χ2 p NOVICE (n = 55) (n = 51 ) Competent 23 41.8 20 39.2 Not Competent 32 58.2 31 60.8 0.074 0.785 PRACTISING NURSES (n = 23) (n = 12) Competent 11 47.8 4 33.3 Not Competent 12 52.2 8 66.7 0.676 0.411 COMBINED (n=78) (n=63) Competent 34 43.6 24 38.1 Not Competent 44 56.4 39 61.9 0.435 0.510 Note: Competence represents 100% performance on 32 mandatory skills; df = 1; p ≤ 0.05 For the Novices at Post Test 2, 41.8% of the CD group and 39.2% of the Traditional group were competent (χ2 = 0.074 p = 0.785). For the Practising Nurses at Post Test 2, 47.8% of the CD group and 33.3% of the Traditional group were competent (χ2 = 0.676 p = 0.411), however the small sample size necessitates interpreting this result with caution. When the cohorts were combined, 43.6% of the CD group and 38.1% of the Traditional group were competent at Post Test 2 (χ2 = 0.435, p = 0.510). Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 4 — Results 109 | P a g e Evaluation of competence in BLS skill categories and specific BLS skills. Each of the specific 32 mandatory and 17 non-mandatory skills within the BLS procedure were then grouped into initial response skills, ventilation skills, circulation skills and health professional skills. Competence for these skill categories were the sum of the correct performance of each specific skill within each category. The results of the combined cohort of Novices and Practising nurses for the mandatory and non-mandatory skills within these four skill categories and each specific skill within these categories at Post Test 1 and Post Test 2 are presented next for the training groups. (Small cell counts in the Practising Nurses cohort prohibited statistical analysis and therefore presentation of the cohorts separately). Competence in BLS skill categories at Post Test 1. Competence in the four BLS skill categories at Post Test 1 are presented in Table 4.4. A higher proportion of the sample achieved competence for the skill categories, than seen in the overall BLS competency scores (Table 4.2), due to both mandatory and non- mandatory scores being combined to form these categories. However, once again no statistically significant differences were found at p ≤ 0.001, between those who undertook the CD program and those who undertook the Traditional program in any of the four BLS skill categories and overall at Post Test 1 for the Combined Novice and Practising Nurses cohort. For initial response skills, when the Novice and Practising Nurses were combined, 92.4% of the CD group and 94.7% of the Traditional group were competent at Post Test 1 (χ2 = 0.646, p = 0.422). For ventilation skills, 94.4% of the CD group and 95.3% of the Traditional group were competent at Post Test 1 (χ2 = 0.120, p = 0.729). For circulation skills, 93.1% of the CD group and 95.3% of the Traditional group were competent at Post Test 1 (χ2 = 0.700, p = 0.403). For Health Professional skills, 79.2% of the CD group and 78.7% of the Traditional group were competent at Post Test 1 (χ2 = 0.011, p = 0.916). Overall for these four BLS skill categories, 89.6% of the CD group and 90.7% of the Traditional group were competent at Post Test 1 (χ2 = 0.097, p = 0.755 [see Table 4.4]). Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 4 — Results 110 | P a g e Table 4.4: Chi-square tests of difference between the CD and Traditional training methods in the competent performance of BLS skill categories at Post Test 1 for the Combined Novice and Practising Nurses cohort. BLS SKILL COMPETENCE POST TEST 1 SKILL CATEGORIES TRAINING GROUPS CD (n = 144) Traditional (n = 150) n% n% χ2 p Initial Response Skills Competent 133 92.4 142 94.7 Not Competent 11 7.6 8 5.3 0.646 0.422 Ventilation Skills Competent 136 94.4 143 95.3 Not Competent 8 5.6 7 4.7 0.120 0.729 Circulation Skills Competent 134 93.1 143 95.3 Not Competent 10 6.9 7 4.7 0.700 0.403 Health Professional Skills 114 79.2 118 78.7 Competent Not Competent 30 20.8 32 21.3 0.011 0.916 Overall Competent 129 89.6 136 90.7 Not Competent 15 10.4 14 9.3 0.097 0.755 Note: 1. Inability to perform analysis of difference (due to small cell counts) in the Practising Nurses cohort necessitated only combined results being presented; Note: 2. Competence represents the sum of the correct performance of all the skills in the category; Bonferroni adjustment p ≤ 0.001; df = 1 Competence in BLS skill categories at Post Test 2. Competence in the four BLS skill categories at Post Test 2 are presented in Table 4.5. There were once again higher percentages of those who were competent for the skill categories, than seen in the overall BLS skill competency scores (see Table 4.3). Also no statistically significant differences (at p ≤ 0.001), between those who undertook the CD program and those who undertook the Traditional program in any of the four BLS skill categories and overall for the categories at Post Test 2 for the Combined Novice and Practising Nurses cohort were identified. Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach

Chapter 4 — Results 111 | P a g e Table 4.5: Chi-square tests of difference between the CD and Traditional training methods in the competent performance of BLS skill categories at Post Test 2 for the Combined Novice and Practising Nurses cohort. BLS SKILL CATEGORY COMPETENCE POST TEST 2 SKILL CATEGORIES TRAINING GROUPS Initial Response Skills CD (n = 78) Traditional (n = 63) Competent Not Competent n% n % χ2 p Ventilation Skills 71 91.0 55 87.3 Competent 7 9.0 8 12.7 0.508 0.476 Not Competent 71 91.0 56 88.9 Circulation Skills 7 9.0 7 11.1 0.178 0.673 Competent Not Competent 70 89.7 57 90.5 8 10.3 6 9.5 0.021 0.885 Health Professional Skills Competent 60 76.9 45 71.4 Not Competent 18 23.1 18 28.6 0.553 0.457 Overall Competent 68 87.2 54 85.7 Not Competent 10 12.8 9 14.3 0.064 0.800 Note: 1: Inability to perform analysis of difference (due to small cell counts) in the Practising Nurses cohort necessitated only combined results being presented; Note 2: Competence represents the sum of the correct performance of all the skills in the category; Bonferroni adjustment p ≤ 0.001; df = 1 For initial response skills, when the Novice and Practising Nurses were combined, 91.0% of the CD group and 87.3% of the Traditional group were competent at Post Test 2 (χ2 = 0.508, p = 0.476). For ventilation skills, 91.0% of the CD group and 88.9% of the Traditional group were competent at Post Test 2 (χ2 = 0.178, p = 0.673). For circulation skills, 89.7% of the CD group and 90.5% of the Traditional group were competent at Post Test 2 (χ2 = 0.021, p = 0.885). For health professional skills, 76.9% of the CD group and 71.4% of the Traditional group were competent at Post Test 2 (χ2 = 0.553, p = 0.457). Overall for these four BLS skill categories, 87.2% of the CD group and 85.7% of the Traditional group were competent at Post Test 2 (χ2 = 0.064, p = 0.800). Including all 49 skills in the overall score (in Table 4.4 and Table 4.5) has resulted in a higher overall skill competency rate (percentage) than the main overall BLS skill Basic life Support training for nurses: Karen Mardegan evaluating an alternative CD-based approach