Comparison of Cardiopulmonary Resuscitation Training Methods for Parents of Infants at High Risk for Cardiopulmonary Arrest

PEDIATRICS/ORIGINAL CONTRIBUTION

Comparison of Cardiopulmonary Resuscitation Training Methods for Parents of Infants at High Risk for Cardiopulmonary Arrest From the University of California at Los Angeles, School of Nursing* and School of Medicine,‡ Los Angeles, CA; and The Ohio State University, College of Nursing,§ Columbus, OH. Received for publication August 6, 1997. Revision received March 30, 1998. Accepted for publication April 18, 1998. Presented at the American Heart Association Scientific Sessions, Anaheim, CA, November 1995. Supported by National Institutes of Nursing Research (R01 NR02434). Copyright © 1998 by the American College of Emergency Physicians.

Kathleen Dracup, RN, DNSc* Debra K Moser, RN, DNSc§ Lynn V Doering, RN, DNSc* Peter M Guzy, MD, PhD‡

Study objectives: To compare three different methods of teaching CPR to parents of infants at high risk for sudden cardiopulmonary arrest and to identify characteristics that predict difficulty in learning CPR. Methods: We conducted a prospective, multicenter clinical trial of 480 parents and other infant caretakers. Subjects were randomly assigned to 1 of 3 CPR training protocols: an instructortaught CPR class, an instructor-taught CPR class followed by a social support intervention, or a self-training video module. CPR proficiency was evaluated with the use of a CPR skills checklist. Results: Of 480 subjects, 301 (63%) were able to demonstrate successful CPR after training. Univariate analysis revealed that unsuccessful learners were likely to be less educated, to have lower incomes, to never have attended a previous CPR class, and to have better psychosocial adjustment to their infant’s illness, compared with successful learners. The proportion of successful learners was significantly higher in the 2 instructor-taught classes than in the self-training video class. Multiple logistic regression analysis was used to develop a predictive profile to describe unsuccessful learners. The following characteristics independently predicted unsuccessful learners: CPR learned in the self-training video group, fewer years of education, and better psychosocial adjustment. Conclusion: Most parents of infants at high risk for sudden death can demonstrate successful CPR skills at the completion of 1 class. However, a significant minority will require extra attention to be successful. Self-training video instruction may not be an adequate substitute for instructor-taught CPR. [Dracup K, Moser DK, Doering LV, Guzy PM: Comparison of cardiopulmonary resuscitation training methods for parents of infants at high risk for cardiopulmonary arrest. Ann Emerg Med August 1998;32:170-177.]

1 7 0

ANNALS OF EMERGENCY MEDICINE

32:2 AUGUST 1998

CPR TRAINING METHODS Dracup et al

INTRODUCTION

Many infants hospitalized in neonatal ICUs (NICUs) are born with a condition that places them at risk for a respiratory or cardiac arrest after discharge from the hospital.1 Recognizing this risk, health care professionals working in NICUs strongly recommend that parents and other caretakers learn how to perform CPR before taking their infant home.2,3 Many hospitals require CPR training as a condition of infant discharge.4 This policy is based on data demonstrating that almost all sudden infant deaths occur in the home5,6 and that an inverse relationship exists between survival and the amount of time that elapses between cardiopulmonary arrest and the initiation of CPR.6,7 Although the need for parents of high-risk infants to learn CPR is undisputed, little research has been performed to document parents’ ability to learn CPR at a time that is inherently stressful.8 Numerous researchers have documented parents’ concerns about the emotional, physical, and financial burdens involved in caring for an infant with special needs, and parents commonly experience both anxiety and depression related to their infant’s health status.9,10 Anxiety and depression adversely affect learning in other educational settings,11 and in a study of family members of adult cardiac patients, those family members who were depressed had greater difficulty learning CPR.12 An additional question relates to the optimal method of teaching CPR. CPR instruction is a time-intensive activity. Community courses taught by the American Heart Association (AHA) and the American Red Cross require successful completion of a multihour program. Most classes provided by NICU staff do not use this extensive protocol but do require several hours of instructor time. As hospitals have reduced the numbers of professional nursing staff available in ICUs,13 instructors have relied increasingly on CPR teaching methods, such as videotapes and computer programs, that reduce the amount of time required to teach parents.14,15 These techniques have been tested in other populations, such as health professionals,16 college students,17 and lay people,15 and have generally been found to result in outcomes equivalent to those obtained with traditional teaching methods. However, the effectiveness of these modalities in highly stressed parents of infants hospitalized in NICUs is unknown. Although several investigators have studied CPR training for parents,14,17-19 all but 2 have focused on normal newborns and in all cases findings have been limited by small, homogeneous samples or the use of written tests rather than demonstration of CPR skills to assess learning. Therefore we conducted a study of CPR skill performance in parents whose

AUGUST 1998

32:2

ANNALS OF EMERGENCY MEDICINE

infants were at risk for sudden death. The study was designed with the following aims: (1) to determine whether parents or other caretakers of infants at risk for sudden death can learn CPR; (2) to identify the characteristics that predict difficulty in learning CPR successfully; and (3) to compare the effectiveness of 3 different methods of teaching CPR. M AT E R I A L S A N D M E T H O D S Subjects

We obtained approval to conduct the study from the appropriate institutional review boards, medical and nursing NICU directors, and physicians and nurses involved in caring for the infants and their families. We recruited 484 parents or other caretakers of high-risk infants from the Level I NICUs of 5 large metropolitan-area hospitals. Participants were told that CPR was being taught as part of a research study to evaluate the optimal way to teach CPR to parents and caretakers of infants. All subjects signed consent forms agreeing to participate in the study. To accommodate the large number of participants who spoke Spanish as their first language, parents were offered their choice of classes conducted in Spanish or in English. Although 484 persons were taught CPR, return demonstration of CPR was not rated on 4 participants. One refused, and data were missing on the remaining 3. Therefore the final sample size reported here is 480. To participate in the study, subjects had to live with the infant and be literate in English or Spanish. Infants were considered at risk for future life-threatening events if they were born prematurely (less than 38 weeks) or with low birth weight (less than 2,500 g); had a documented episode of apnea or bradycardia; had a congenital cardiac or gastrointestinal anomaly; or had respiratory distress syndrome or bronchopulmonary dysplasia. Potential participants were excluded if they had taken a CPR class within the last 2 years. This exclusion was based on the large body of research showing uniformly poor CPR skills retention at 1 year after CPR instruction.20 The majority of participants were parents of the infants (n=441, or 92%); their mean age was 29.9±8.4 years. Most were high school graduates (mean years of education, 12.6±3.5 years). Ethnicity was highly diverse, with 212 subjects (45%) identifying themselves as Hispanic, 162 (34%) as white, 39 (8%) as black, 38 (8%) as Asian, and 29 (5%) as other. Fifty-eight percent of the sample (276 subjects) were mothers, 34% (165) were fathers, and 8% (39) were other relatives or caretakers. Thirty-eight percent (132) of

1 7 1

CPR TRAINING METHODS Dracup et al

the participants reported an annual family income lower than $20,000, whereas 41% (143) reported an income higher than $40,000. Twenty-six percent (122 subjects) had taken a CPR class at some time in the past. Procedures

Subjects were asked to complete a packet of paper-andpencil questionnaires in either English or Spanish just before attending the CPR class. CPR classes were held in a conference room in close proximity to the NICU. Participants were given their choice of attending classes conducted in English or in Spanish. Participants were randomly assigned to 1 of 3 CPR protocols: an instructor-taught class (CPR-Didactic), an instructor-taught class combined with a social support intervention (CPR-Social Support), and a video-only class (CPR-Video). The content of CPR instruction was identical in all 3 conditions, and all 3 used the same videotape, which was available in either Spanish or English. Although all 3 protocols used the same videotape, there were differences in method of instruction. In the CPRDidactic and CPR-Social Support groups, instruction was modeled after standard CPR classes offered in community settings, with class size kept to an average of 6 persons to maximize practice time. The instructor stopped the tape at designated points to answer questions, demonstrate the techniques, and allow the subjects to practice the techniques in each segment and then in total sequence. In both groups, the participants received instructor feedback each time the videotape was stopped and they practiced what they had been shown on the tape. In the CPR-Social Support intervention, participants learned CPR from the instructor and the videotape, using the identical protocol as the CPR-Didactic group. However, in this group the instructor began the class with a brief discussion of the feelings involved in bringing home an infant with special health needs. Instruction was followed by a 30-minute discussion among participants about fears related to learning CPR and about caring for the infant at home. The intervention was designed not necessarily to enhance learning of CPR but to reduce the potential negative emotions evoked by learning of CPR, which have been documented in previous studies of family members of patients with cardiac conditions.21,22 In the CPR-Video class, which had an average class size of 2 persons, the instructor provided information about how to use the videotape and gave the participants a written, self-paced module to guide their learning. Participants were told to read the module, play the video, and practice at the

1 7 2

indicated times. The instructor then left the room, returning only when the subjects said they were ready to demonstrate CPR to the instructor. All CPR instructors were nurses who had successfully completed a cardiac life support course for health care professionals and were bilingual in English and Spanish. At each hospital the same nurses implemented each of the 3 protocols to provide consistency across the 3 conditions, and each site had its own group of nurse-instructors. In general, there were 2 instructors at each site. The content of the CPR class was based on recent recommendations to simplify infant and pediatric CPR so as to optimize retention of knowledge and skills.23 Only 1-person CPR was taught, on the basis that the individual was most likely to be alone with the infant at the time of an arrest. Also, only infant and pediatric CPR was taught; no information was provided about adult CPR. Instruments

Participants were asked to do a complete demonstration of CPR at the end of class or, in the video group, when they indicated they were ready to be tested. The CPR instructor rated their performance with the use of a CPR Skills Checklist developed by Mandel.24 Interrater reliability for this instrument has been documented to be .98, and reliability between an instructor and a printout from a recording resuscitation mannequin was .88.12,25 On this checklist, the skill of CPR is divided into initial assessment, chest compressions, and ventilations. In each of these 3 sections, instructors rate subjects on several discrete components that are the hallmarks of CPR. Examples of individual items include assessment of pulse (initial assessment), landmark location and proper depth of compression (chest compressions), and proper ventilation depth (ventilations). After rating of the individual items, participants are given a score of good, fair, or poor in each section. A final score on overall adequacy of CPR performance then is assigned, based on the performance in the 3 sections. All 3 sections had to be rated good or fair to achieve an overall score of satisfactory. These ratings are described in detail elsewhere.25 Each participant was given an overall CPR performance score of either successful (good or fair) or unsuccessful (poor). For a successful performance, each of the 3 steps had to be performed in proper sequence and according to AHA standards. A performance that did not meet AHA standards and was given a score of fair was rated successful if it was judged adequate and safe.25 An example of a performance that did not meet strict AHA standards but was judged successful was one in which timing, hand position, or

ANNALS OF EMERGENCY MEDICINE

32:2 AUGUST 1998

CPR TRAINING METHODS Dracup et al

mouth and nose seal was incorrect but was corrected by the subject. An unsuccessful performance was one that received a poor rating. Unsuccessful performances included those in which sequencing was incorrect; some maneuver was performed that was potentially injurious; compressions were inadequate; or ineffective ventilations were delivered due to uncorrected failure to open the airway properly, poor volumes, or poor timing. After CPR skills were scored, the instructor worked with the subjects until successful CPR was demonstrated. No subject left the CPR class in any of the 3 groups without being able to demonstrate successful CPR. To assess the potential influence of anxiety and depression on learning, each participant’s emotional state was measured immediately before CPR teaching with the Multiple Affect Adjective Check List.26-27 It consists of 132 alphabetically arranged adjectives with instructions to work rapidly and check the words that describe feelings today. Anxiety, depression, and hostility are scored separately. Higher scores indicate higher levels of anxiety, depression, and hostility, with the range of scores being 0 to 21, 0 to 40, and 0 to 30 respectively. The instrument has been used in numerous clinical populations and has acceptable reliability and validity.26 In this study, reliability was tested with the use of Cronbach’s α. Internal consistency was .80 for the anxiety subscale, .85 for the depression subscale, and .80 for the hostility subscale. The Psychosocial Adjustment to Illness Scale–Family (PAIS)28 was used to assess the general level of adjustment of the parents to the birth of the baby. The PAIS consists of 46 questions in 7 domains: health care orientation, vocational environment, domestic environment, sexual relationships, extended family relationships, social environment, and psychologic distress. Respondents are asked to indicate their response to each question on a 4-point scale of distress ranging from 0 (no distress) to 3 (extreme distress). Scores range from 0 to 138, with higher scores indicating worse psychosocial adjustment. Reliability in the current study was tested by Cronbach’s α and ranged from .53 to .85 for the 7 subscales and the total score. The total score was used in this study. The Perceived Social Support instrument was used to measure subjects’ perceptions of available social support.29 The instrument has 12 items, and subjects indicate their response to each item on a 7-point Likert scale ranging from 1 (very strong disagreement) to 7 (very strong agreement). Because each item is worded positively, higher scores indicate higher levels of social support, with a possible range of 12 to 84. The stability of the instrument has been reported

AUGUST 1998

32:2

ANNALS OF EMERGENCY MEDICINE

as .85 (established by test-retest at 2 months). In the current study, Cronbach’s α was .84. Analysis

Differences in baseline sociodemographic parameters and emotional and psychosocial characteristics between successful and unsuccessful CPR learners were assessed with the use of χ2 or independent t tests. To determine which method of CPR instruction was the most effective, we used the χ2 test to compare CPR performance ratings among the 3 CPR groups. To determine independent predictors of CPR learning, multiple logistic regression was performed. Variables were entered into the regression model in hierarchical fashion, with sociodemographic variables entered first, followed by emotional and psychosocial variables. Examination of residuals, leverage, and Cook’s distance revealed that the assumptions of logistic regression were met and that no values were unduly influential in construction of the model. The following predictor variables were entered: group (CPR-Video, CPR-Social Support, CPR-Didactic); sex (male, female); relationship to infant (mother, father, other); marital status (married, single, divorced/separated/widowed); employment outside home (yes, no); family yearly income (<$20,000, $20,000 to 39,999, >$39,999, declined to answer); previous CPR training (yes, no); age; years of education; anxiety level; hostility level; depression level; degree of perceived social support; and level of psychosocial adjustment to illness. The last 7 variables were treated as continuouslevel variables. The criteria for entry and removal of variables in the model were a significance level lower than .05 for the score statistic (for entry) and a significance lower than .10 for the likelihood-ratio test (for removal). Odds ratios and their 95% confidence intervals (CIs) for the prediction of unsuccessful CPR learners were calculated.

R E S U LT S

Of the 480 subjects enrolled in a CPR class, 301 (63%) demonstrated successful CPR immediately after CPR training, and 179 (37%) did not. Table 1 provides a comparison of the sociodemographic characteristics of subjects by CPR skills rating. Successful and unsuccessful learners were not distinguished by differences in age, sex, relationship to the baby, marital status, or employment status. Subjects who learned CPR successfully were likely to have more years of education and to have a higher income than unsuccessful learners, and they were more likely to have attended a previous CPR class.

1 7 3

CPR TRAINING METHODS Dracup et al

Comparison of emotional and psychosocial characteristics between successful and unsuccessful learners revealed that successful learners had worse overall psychosocial adjustment to the illness of their infant. There were no statistically significant differences in anxiety, depression, hostility, or perceived social support between successful and unsuccessful learners (Table 2). The way in which subjects learned CPR made a significant difference in their ability to demonstrate successful CPR immediately after the class. As summarized in Table 3, a higher percentage of subjects in the instructor-taught classes (ie, CPR-Didactic or CPR-Social Support groups) successfully learned CPR than in the CPR-Video group. Only 38% of subjects in the video group were able to demonstrate all the appropriate components of CPR at a level judged successful, despite the fact that they believed themselves ready to demonstrate their CPR skills to the instructor. In contrast, 80% of the CPR-Social Support group were rated as having successful CPR skills, as were 73% of those in the CPRDidactic group. The proportion of unsuccessful learners

Table 1.

Comparison of sociodemographic characteristics of 301 successful and 179 unsuccessful CPR learners.

Parameter Age, yr (mean±SD) Education, yr (mean±SD) Sex Male Female Relationship to infant Mother Father Other Marital status Married Single Divorced, separated, or widowed Employment outside home No Yes Family yearly income <$20,000 $20,000–$39,999 >$39,999 Declined to answer Previous CPR training No Yes

1 7 4

Successful No. (%)

Unsuccessful No. (%)

Probability Value

29.9±7.8 13.1±3.4

30.0±9.4 11.7±3.5

.89 .0001

106 (35.2) 195 (64.8)

66 (36.9) 113 (63.1)

177 (58.8) 102 (33.9) 22 (7.3)

99 (55.3) 63 (35.2) 17 (9.5)

206 (69.6) 71 (24.0) 19 (6.4)

103 (58.9) 57 (32.5) 15 (8.6)

.72 .62

was significantly higher in the CPR-Video group compared with the other 2 groups. There was no significant difference between the 2 instructor-taught groups in the proportion of unsuccessful learners. Multiple logistic regression revealed a predictive profile for subjects with difficulty learning CPR (Table 4). Unsuccessful learners were likely to have less education, to have learned CPR in the CPR-Video group, and to have better psychosocial adjustment to infant’s illness. None of the other variables tested entered into the final model. To assess the degree to which this model fit our data (goodness of fit), we compared our predictions for categorization of subjects as successful or unsuccessful with the observed outcomes and found that 79% of subjects were correctly categorized. In addition, the goodness of fit of the model was assessed by determining the model χ2 value (P=.0001) and the improvement χ2 value (P=.004). All of these indicate that the model adequately fit the data. DISCUSSION

We conducted this study to determine whether parents of infants at high risk for cardiopulmonary arrest could learn CPR successfully, to identify the parental characteristics that predict difficulty in learning CPR, and to compare results of different CPR training protocols. Our findings indicate that the majority of parents can learn CPR but a significant minority are unsuccessful initially and require special instructor attention. Our findings corroborate those of earlier reports14,17 that were conducted in small, homogeneous samples. Learners were successful regardless of a number of sociodemographic characteristics, including age, sex, relationship to infant, and employment status. These findings are consistent with the work of other researchers who found

.06

Table 2.

Emotional and psychosocial characteristics of 301 successful and 179 unsuccessful CPR learners. .47

120 (41.0) 173 (59.0)

79 (44.4) 99 (55.6)

76 (24.7) 50 (16.2) 104 (33.8) 78 (25.3)

53 (32.9) 25 (15.5) 34 (21.1) 49 (30.4)

213 (70.8) 88 (29.2)

129 (80.1) 32 (19.9)

Parameter .02

.04

Anxiety Hostility Depression Perceived social support Psychosocial adjustment to infant’s illness* *Higher

Successful (mean±SD)

Unsuccessful (mean±SD)

Probability Value

9.8±4.5 10.5±4.1 16.3±6.6 68.6±14.2

9.5±4.0 10.1±4.0 15.3±7.0 68.6±12.9

.41 .33 .13 .90

32.8±16.0

28.5±16.4

.007

scores indicate worse psychosocial adjustment.

ANNALS OF EMERGENCY MEDICINE

32:2 AUGUST 1998

CPR TRAINING METHODS Dracup et al

no differences in CPR learning based on sex or work history.24,30 However, age was identified as a significant predictor of difficulty learning CPR in several studies.12,31 In one of these,12 participants were relatives of cardiac patients and were in their sixth and seventh decades. In the current study the participants were relatively young, so age was not expected to be an important predictor of learning. Level of education and income were significantly different on univariate analyses between participants who did and those who did not demonstrate successful CPR techniques. In the multivariate analysis, of these 2 variables, only education was predictive. This finding suggests that learners with fewer years of education may need additional attention to be successful. Although we had hypothesized that those parents and caretakers who reported greater psychologic distress would have more difficulty learning CPR, the opposite was true. Those participants who demonstrated successful CPR initially had poorer psychosocial adaptation than unsuccessful learners did. Although the differences were not statistically significant, the same trends were noted for anxiety, depression, and hostility. It may be that parents and caretakers who are emotionally distressed about the birth of a premature or high-risk infant are more focused on learning how to deal with potential emergencies. This interpretation is supported by previous findings that anxiety and problemsolving ability were significantly correlated in mothers of preterm infants.32 These findings support the interpretation that psychologic distress may function as a motivator for mobilizing resources and learning to cope, which enhances learning of CPR before infant discharge. With shortened lengths of stay and increasing severity of hospitalized infants, discharge planning has become increasingly challenging in the setting of the NICU.33 Parents’ first concern is knowing what to do in the case of a home emergency.10,34 In contrast, nurses have numerous demands on

their time, which has resulted in the popularity of timesaving teaching methods. Our findings that parents in the video group had significantly lower psychomotor skills at the conclusion of the CPR class is in contrast to the findings of many investigators of adult CPR training. In general, investigators have documented either no significant differences in results between instructor-taught and audiovisual methods of teaching14,16,17,29 or improved learning with video self-teaching compared with traditional CPR training classes.35 In a study by Long14 comparing didactic and audiovisual tape methods of teaching CPR to parents in NICUs, parents in both groups scored similarly in a return demonstration of CPR. Psychomotor skills were rated slightly lower in the audiovisual group compared with the didactic class (70% versus 75%), but the difference was not significant. The differences between our findings and those of previous investigators may reflect the fact that all 3 experimental groups, including the 2 instructor-led groups, used a videotape to provide course content. In his analysis of 68 CPR classes taught by the American Red Cross, Brennan36 found that use of a videotape in a standard, instructor-led class had a significant positive effect on student achievement. Our results may also reflect differences in manner and time of testing. For example, Long’s percentage score of CPR learning combined a written test and skills demonstration, whereas the current study focused on psychomotor skills.14 In all studies to date, average test scores on written tests immediately after CPR training are higher than 90%; therefore the lack of difference between the 2 types of CPR training in Long’s study may have resulted from high written scores. Our finding that video instruction alone was less effective than classroom CPR instruction is different from that of Braslow and colleagues.36 They used a 3-group design to compare 2 different 30-minute video self-instruction methods with a 4-hour traditional classroom instructor method. They found that 54% of video self-instruction sub-

Table 3.

Comparison of 301 successful and 179 unsuccessful CPR learners by method of CPR instruction.

Table 4.

Predictors of unsuccessful CPR learners. Instructional Group CPR-Video* CPR-Social Support CPR-Didactic *P<.001

Successful No. (%)

Unsuccessful No. (%)

65 (38.2) 125 (79.6) 111 (72.5)

105 (61.8) 32 (20.4) 42 (27.5)

compared with both other groups. Comparison of CPR-Social Support and CPR-Didactic:

P>.05.

AUGUST 1998

32:2

ANNALS OF EMERGENCY MEDICINE

Variable CPR-Video group Years of education Score for psychosocial adjustment to infant’s illness

Odds Ratio

95% CI

Probability Value

3.96 .92

2.55–6.15 .87–.98

.00001 .01

.98

.97–.99

.005

1 7 5

CPR TRAINING METHODS Dracup et al

jects and 80% of a group using revised video self-instruction were rated competent or better, compared with 45% of subjects who learned CPR using traditional classroom instruction lasting 4 hours. In contrast, our study was a test of the method of delivery and not of the content. All of our CPR groups received the same content in approximately the same amount of time. Parents in the current study were tested immediately after the CPR class to assess the feasibility of using a video protocol with no instructor. Several investigators who have compared audiovisual and instructor training methods have tested CPR skills after a period of instructor coaching, thereby blurring the differences between video and instructor protocols. We tested parents before coaching. All parents in our study ultimately scored satisfactorily on the skills assessment, regardless of group assignment, but more than 60% of those in the video group required individual coaching by the instructor. Based on their review of more than 40 CPR retention studies, Kaye et al20 concluded that CPR training must include adequate practice time with immediate corrective feedback for skill mastery. Participants in the video group may not have given themselves sufficient practice time, and they did not have corrective feedback. One limitation of this study is that we did not measure the effect of CPR training methods on success of actual resuscitation efforts. Ultimately, the major outcome of interest in any CPR training study is whether the trainee can perform adequate CPR in an actual emergency and whether the victim is successfully resuscitated. However, given the relatively small likelihood of sudden respiratory or cardiac arrest even in a high-risk population, thousands of subjects would have to be trained and followed to assess this effect. Such a project was beyond the scope of the current study. A strength of our study is that it was conducted in an ethnically and economically diverse population of subjects. A second strength lies in the selection of a study population for whom CPR training holds particular salience. The majority of previous CPR studies were conducted among trainees who had a low probability of ever encountering a situation requiring CPR and using their skills. In summary, CPR training is an important aspect of discharge planning in every NICU. Parents are taught how to initiate CPR in case of a respiratory or cardiac arrest in the home, based on compelling data that the most important predictor of infant morbidity and mortality in such events is the amount of time that elapses between cardiopulmonary arrest and resuscitation.6 Our findings suggest that most parents can learn CPR with little difficulty, but that a minority need special attention for successful learning. Parents with less education and better psychosocial adjustment may

1 7 6

require more instructor time. Perhaps most important in these days of hospital restructuring and reduced nursing staff, videotape instruction alone is not an adequate substitute for instructor-taught CPR in the setting of the NICU. In this population, CPR instruction by videotape alone provides similar knowledge about CPR but a lower CPR skills level, compared with instructor-taught courses. Video instruction should be followed by reinforcement of skills by a CPR instructor before discharge of the infant from the hospital.

REFERENCES 1. Kuisma M, Suominen P, Korpela R: Paediatric out-of-hospital cardiac arrest: Epidemiology and outcome. Resuscitation 1995;30:141-150. 2. Chameides L: Cardiopulmonary resuscitation: Standards, guidelines, and education. Pediatrics 1987;79:446-449. 3. Emergency Cardiac Care Committee and Subcommittee of the American Heart Association: Guidelines for CPR and emergency cardiac care. JAMA 1992;268:2172-2298. 4. Higgins SS, Hardy CE, Higashino SM: Should parents of children with congenital heart disease and life-threatening dysrhythmias be taught cardiopulmonary resuscitation? Pediatrics 1989;84: 1102-1104. 5. Hickey RW, Cohen DM, Strausbaugh S, et al: Pediatric patients requiring CPR in the prehospital setting. Ann Emerg Med 1995;25:495-501. 6. Innes PA, Summers CA, Boyd IM, et al: Audit of paediatric cardiopulmonary resuscitation. Arch Dis Child 1993;68:487-491. 7. Eisenberg M, Bergner L, Hallstrom A: Epidemiology of cardiac arrest and resuscitation in children. Ann Emerg Med 1983;12:672-674. 8. Grace JT: Mother’s self-reports of parenthood across the first 6 months postpartum. Res Nurs Health 1993;16:431-439. 9. Goldberg S, Simmons RJ, Newman J, et al: Congenital heart disease, parental stress, and infant-mother relationships. J Pediatr 1991;119:661-666. 10. Drake E: Discharge teaching needs of parents in the NICU. Neonatal Network 1995;14:49-53. 11. DeCecco JP: The Psychology of Learning and Instruction. Englewood Cliffs, NJ: Prentice-Hall, 1968. 12. Dracup K, Heaney DM, Taylor SE, et al: Can family members of high-risk cardiac patients learn cardiopulmonary resuscitation? Arch Intern Med 1989;149:61-64. 13. Buerhaus PI, Staiger DO: Managed care and the nurse workforce. JAMA 1996;276:1487-1493. 14. Long CA: Teaching parents infant CPR: Lecture or audiovisual tape? Matern Child Nurs J 1992;17:30-32. 15. Braslow A, Brennan RT, Newman MM, et al: CPR training without an instructor: Development and evaluation of a video self-instructional system for effective performance of cardiopulmonary resuscitation. Resuscitation 1997;34:207-220. 16. Nelson M, Brown CG: CPR instruction: Modular versus lecture course. Ann Emerg Med 1984; 13:118-120. 17. Coleman S, Dracup K, Moser DK: Comparing methods of cardiopulmonary resuscitation instruction on learning and retention. J Nurs Staff Development 1991;7:82-87. 18. Messmer PM, Meehan R, Gilliam N, et al: Teaching infant CPR to mothers of cocaine-positive infants. J Contin Educ Nurs 1993;24:217-220. 19. Kaiserman K, Martin GI, Sindel BC, et al: The effectiveness of a cardiopulmonary resuscitation program for mothers of newborn infants. J Perinatol 1989;9:49-51. 20. Kaye W, Rallis SF, Mancini ME, et al: The problem of poor retention of cardiopulmonary resuscitation skills may lie with the instructor, not the learner or the curriculum. Resuscitation 1991;21:67-87.

ANNALS OF EMERGENCY MEDICINE

32:2 AUGUST 1998

CPR TRAINING METHODS Dracup et al

21. Dracup K, Guzy PM, Taylor SE, et al: Cardiopulmonary resuscitation (CPR) training: Consequences for family members of high-risk cardiac patients. Arch Intern Med 1986;146:1757-1761. 22. Dracup K, Moser DM, Taylor SE, et al: The psychological consequences of cardiopulmonary resuscitation training for family members of patients at risk for sudden death. Am J Pub Health 1997;87:1434-1439. 23. Seidel JS, Henderson DP, Spencer PE: Education in pediatric basic and advanced life support. Ann Emerg Med 1993;22:489-494. 24. Mandel LP, Cobb LA: Initial and long-term competency of citizens trained in CPR. Emerg Health Serv Q 1982;1:49-63.

Reprint no. 47/1/91518 Address for reprints: Kathleen Dracup, RN, DNSc University of California, Los Angeles School of Nursing Post Office Box 951702 700 Tiverton Avenue

25. Moser DK, Dracup K, Guzy PM, et al: Cardiopulmonary resuscitation skills retention in family members of cardiac patients. Am J Emerg Med 1990;8:498-503.

Los Angeles, CA 90095-1702

26. Zuckerman M, Lubin B: Manual for the Multiple Affect Adjective Check List. San Diego, CA: Educational and Industrial Testing Service, 1965.

Fax 310-794-7482

27. Zuckerman M, Lubin B, Rinck CM: Construction of new scales for the Multiple Affect Adjective Check List. Behav Assessment 1983;5:119-129.

310-206-3831 E-mail [email protected]

28. Derogatis L: Scoring and Procedures Manual for PAIS. Baltimore, MD: Clinical Psychometric Research, 1976. 29. Blumenthal JA, Burg MM, Barefoot J, et al: Social support, type A behavior and coronary artery disease. Psychosom Med 1987;49:331-340. 30. Kalmthout PM, Speth PA, Rutten JR, et al: Evaluation of lay skills in cardiopulmonary resuscitation. Br Heart J 1984;53:562-566. 31. Van Kerschaver E, Delooz HH, Moens GFG: The effectiveness of repeated cardiopulmonary resuscitation training in a school population. Resuscitation 1989;17:211-222. 32. Gennaro S: Maternal anxiety, problem-solving ability, and adaptation to the premature infant. Pediatr Nurs 1985;11:343-348. 33. Arenson J: Discharge teaching in the NICU: The changing needs of NICU graduates and their families. Neonatal Network 1988;6:29-52. 34. Gehl M, Lantzy A: Parents’ needs surrounding discharge from the neonatal intensive care unit. Matern Child Nurs J 1990;19:179-180. 35. Braslow A, Brennan RT, Newman MM, et al: CPR training without an instructor: Development and evaluation of a video self-instructional system for effective performance of cardiopulmonary resuscitation. Resuscitation 1997;34:207-220. 36. Brennan RT: Trainee, instructor, and class factors predicting achievement in CPR training classes. Am J Emerg Med 1991;9:220-224.

AUGUST 1998

32:2

ANNALS OF EMERGENCY MEDICINE

1 7 7