JOGNN. Ill and premature infants gain invaluable protection

Transcription

1 JOGNN Success of a Lactation Education Program on NICU Nurses Knowledge and Attitudes Laura W. Bernaix, Cynthia A. Schmidt, Michelle Arrizola, Dina Iovinelli, and Clarisa Medina-Poelinez Correspondence Laura W. Bernaix, PhD, RN, School of Nursing, Southern Illinois University Edwardsville, P.O. Box 1066, Edwardsville, IL lbernai@siue.edu Keywords lactation education nurses neonatal intensive care unit ABSTRACT Objectives: To test an educational intervention designed to improve lactation knowledge, attitudes, and beliefs of NICU nurses and to improve their intentions to provide mothers with lactation support. Design: Quasi-experimental, time-series pretest/posttest. Setting: NICU of a Midwestern, free-standing, tertiary-care children s hospital. Participants: Convenience sample of 64 NICU nurses and 2 separate convenience samples of mothers of infants hospitalized in the NICU (n 5 19 and 13, respectively). Methods: Nurses were measured on study outcomes at multiple time points, beginning with 2 weeks before and ending at 3 months after attendance to a 4-hour educational program. Mothers were sampled before and 3 months after the intervention. Main Outcome Measure(s): Nurses lactation knowledge, attitudes, beliefs, and intentions to support lactation and mothers perceptions of lactation support in the NICU. Results: Findings suggest that this educational intervention was effective for improving NICU nurses lactation knowledge and attitudes, and that these improvements were maintained over time. Further, the supportive atmosphere for lactation in this NICU significantly improved following the implementation of the educational intervention for nurses. Conclusion: Intermittent, short educational programs which include practical how-to s and motivational encouragement for staff may provide the empowerment nurses need in order to be supportive of lactation. JOGNN, 37, ; DOI: /j x Accepted May 2008 Laura W. Bernaix, PhD, RN, is a professor, School of Nursing, Southern Illinois University Edwardsville, IL. Cynthia A. Schmidt, PhD, RN, is an associate professor, School of Nursing, Southern Illinois University Edwardsville, IL. Michelle Arrizola, RN, BSN, IBCLC, is an NICU director, NICU, Children s Memorial Medical Center, Chicago, IL. (Continued) Ill and premature infants gain invaluable protection from serious illnesses and complications speci c to their high-risk status when they receive their own mother s milk (OMM). Immunologic and developmental bene ts, which include lower incidence of nosocomial infections and necrotizing enterocolitis, enhanced brainstem, and neurocognitive development, shorter duration of parenteral nutrition while hospitalized, and shorter NICU hospital stays, make OMM the ideal food choice for these high-risk infants (Amin, Merle, Orlando, Dalzell, & Guillet, 2000; Andorsky et al., 2001; Schanler, 2001). As such, the major focus of several international and national health organizations (American Academy of Pediatrics [AAP], 2005; Association of Women s Health, Obstetric, and Neonatal Nurses, 2007; U.S. Department of Health and Human Services, 2000; World Health Organization [WHO], 2004) has been to encourage the provision of OMM to this special population. In the most recent publication by the AAP, speci c guidelines for this population were included (2005). Despite this documented support and promotion of breastfeeding for these high-risk infants, the current initiation rate for low-birthweight and vulnerable infants is low. Research has indicated that there are a myriad of barriers which discourage mothers of these high-risk infants to either breastfeed or be successful in their attempts. Bernaix, Schmidt, Jamerson, Seiter, and Smith (2006) found that parents perceive the NICU to be a stressful and frightening environment, and as such, nd it hard to establish and sustain lactation. Other barriers identi ed by mothers of preterm and/or verylow-birthweight infants include concern for the 436 & 2008 AWHONN, the Association of Women s Health, Obstetric and Neonatal Nurses

2 Bernaix, L. W., Schmidt, C. A., Arrizola, M., Iovinelli, D., and Medina-Poelinez, C. compromised health and tenuous condition of the infant, low milk volume, di culty with breastfeeding technique, breast and nipple problems, delayed and infrequent pumping, and supplementation issues (Callen, Pinelli, Atkinson, & Saigal, 2005; Hill, Aldag, Chatterton, & Zinaman, 2005). Mothers perceived lack of competence and unrealistic expectations related to the lactation process may also explain variations in lactation success (Pridham, Saxe, & Limbo, 2004; Woolridge & Hall, 2003). Further, limited information-sharing and encouragement by healthcare professionals about the importance and mechanics of providing OMM is believed to in uence lactation initiation rates in this population (Miracle, Meier, & Bennett, 2004; Campbell & Gutman, 2006). Interestingly, recent studies have suggested that when nurses and other health professionals provide facts and counseling about the bene ts and process of providing OMM, mothers are more likely to choose breastfeeding and have successful outcomes (Miracle et al., 2004; Sisk, Lovelady, Dillard, & Gruber, 2006). Nurses and other healthcare professionals, however, may not be adequately prepared to provide this degree of informational and technical support (Spatz, 2005). Improving nurses lactation knowledge and supportive behavior has been the subject of several studies in which nurses lactation support has been examined and interventions developed and tested for their e ect on improving lactation outcomes in the NICU. According to Jaeger, Lawson, and Filteau (1997), 48% of a sample of 44 mothers reported con icting advice from their NICU nurses regarding the frequency and mechanics of pumping and methods for transitioning the infant from bottle to breast. Spatz (2005) reported that a survey completed by 407 nurses working in a free-standing children s hospital also revealed gaps in knowledge regarding breastmilk management even though there had been hospital-wide policies and procedures in place for the preceding 2 years. Merewood, Philipp, Chawla, and Cimo (2003) evaluated a hospital s breastfeeding initiation and duration rates in its NICU after the hospital was awarded Baby-Friendly status; this designation is awarded through the Baby-Friendly Hospital Initiative of UNICEF and the WHO for hospitals who implement evidence-based breastfeeding policies throughout their institution. Over a span of 4 years, which included training NICU physicians and nurses about breastfeeding, initiation rates increased from 34.6% to 74.4% (po.001), and 2-week duration rates increased from 27.9% to NICU nurses may not feel prepared to provide the support needed by mothers who are providing breastmilk to their hospitalized infants. 65.9% (po.001). Gonzalez et al. (2003) evaluated the e ectiveness of providing an International Board Certi ed Lactation Consultant service within the NICU to increase the proportion of infants given OMM. The service not only provided individual instruction to mothers, but also provided consultation and instruction to the NICU physicians and nurses. A pre-intervention/postintervention chart review revealed signi cantly improved lactation incidence (p 5.002). More speci cally, the proportion of infants ever provided OMM increased from a baseline of 31% to 47% after implementing the service. Likewise, a study by Siddell, Marinelli, Froman, and Burke (2003) demonstrated, through a pretest, posttest design, that the provision of an 8-hour educational session for NICU nurses increased their lactation knowledge signi cantly (po.001) and enhanced their attitudes about breastfeeding (p 5.055). The authors however only examined changes in knowledge and attitudes once following the intervention (at 2 weeks) and concluded that knowledge and attitudinal change resulting from an educational intervention should realistically be examined over time. While these e orts have demonstrated some value in helping nurses assist mothers in the NICU to be successful in their lactation e orts, more is needed. Because nurses have the most constant contact with the mother and her infant during the infant s hospitalization in the NICU, interventions should be targeted toward improving the nurse s lactation knowledge and supportive skills/behaviors. This study addressed this need by testing the long-term e ect of an educational intervention designed to improve NICU nurses lactation knowledge, attitudes, and beliefs, and their intentions to provide lactation support to mothers of infants hospitalized in the NICU. The study was guided by the Theory of Reasoned Action (TRA) (Ajzen & Fishbein, 1980), which suggests that a person s attitudes and social pressures (subjective norms) about performing or not performing a particular behavior in uences their intentions toward the behavior and that those intentions are the immediate determinant of the nal behavior. For this study, it was believed that some NICU nurses may have negative attitudes and beliefs, and inaccurate knowledge, about the value of providing breast milk to high-risk infants and/or perceive that their e orts to assist a mother with lactation initiation and maintenance may be futile JOGNN 2008; Vol. 37, Issue 4 437

3 Lactation Education Program Dina Iovinelli, RN, BSN, is an NICU clinical supervisor, NICU, Children s Memorial Medical Center, Chicago, IL. Clarisa Medina-Poelinez, MSN, RN is an NICU staff nurse, NICU, Children s Memorial Medical Center, Chicago, IL. and/or not valued by their peers or supervisors. The intervention for this study was created therefore with the intent of improving NICU nurses lactation knowledge and attitudes, and their intentions to provide lactation support to mothers in the NICU. Based on the literature review and theoretical framework, research questions for this study were: (a) What is the relationship between nurses demographic variables (age, education, years worked in the NICU setting, personal breastfeeding experience, number of previous breastfeeding continuing education courses/sessions) and nurses preeducational intervention knowledge and attitudes regarding lactation in the NICU setting?, (b) Is there a di erence in the lactation knowledge, attitudes, and beliefs of NICU nurses following implementation of an educational intervention?; (c) Is there a di erence in the intentions of NICU nurses to provide lactation support to mothers of hospitalized infants following implementation of an educational intervention?; and (d) Is there a di erence in NICU nurses lactation support as measured by the perceptions of mothers before and after the implementation of an educational intervention? Method Design, Setting, and Participants This quasi-experimental study took place in a NICU at a Midwestern tertiary-care children s hospital. The NICU has a 42-bed capacity, an average monthly admission of 40 infants, and a total of 120 nurses sta ng the unit. A time-series pretest/posttest design was used to address the research questions. Approval to conduct the study was granted by the Institutional Review Board of the study site and the researchers institution. Two sample elements, NICU nurses and mothers of high-risk infants hospitalized in the same NICU, were recruited. Two separate samples of mothers (pre-intervention and 3-month postintervention) provided a cross-sectional sense of the NICU s supportive atmosphere for lactation. NICU registered nurses who consented and who were not part of the research team were included in the study. Mothers were included if they were at least 15 years of age, had a high-risk infant who had been hospitalized in the NICU for no more than 2 weeks at time of entry into the study, and had chosen to provide breast milk to their infant. Measures Both samples of mothers completed a demographic questionnaire which contained 12 items: age, marital status, ethnicity, educational background, prior experience with breastfeeding and with having a child hospitalized in the NICU, their preferred infant feeding method before delivery of their infant, attendance to childbirth classes, infant age in hours upon admission to the NICU, and the length in time (hours) following the delivery when the mother began to pump her breasts. Mothers also completed the Mothers Perceived Support Questionnaire (MPSQ) (Bernaix, 2000). Previously developed for use with breastfeeding mothers of healthy infants during the postpartum hospitalization, this tool was modi ed to t this study s sample population.the tool required the mother to evaluate the lactation support provided by the nurses during her infant s NICU stay and provided a cross-sectional sense of the supportive atmosphere for lactation in the NICU as perceived by the mothers. Each of the 46 items is measured on a 5-point Likert scale, with 1 5strongly disagree and 5 5 strongly agree. Items are designed to di erentiate di erent types of lactation support: informational support (16 items), technical support (15 items), and emotional support (15 items). Previous internal consistency reliability of the original version of the instrument was.97 overall, and between.91 and.95 for each of its three subscales. Overall coe cient a values for this study were.94 (pre-intervention) and.96 (3-month postintervention) and between.79 and.92 for the subscales. Nurses completed a 13 item demographic questionnaire which measured selected demographic data including highest nursing degree, recent attendance (within 2 years) to continuing education courses speci c to lactation, and the nurse s personal breastfeeding history. The Nursing Support for Breastfeeding Questionnaire (NSBQ) (Bernaix, 2000), a 58-item tool measuring thetra constructs and originally developed to measure nurses intentions, attitudes, and beliefs about providing lactation support to new mothers during the immediate postpartum, was also administered. For this study, the tool was modi ed to measure the NICU nurse s intentions, attitudes, and beliefs toward supporting the mother s attempts to provide OMM to her high-risk infant. The tool is comprised of ve subscales; responses to each of the 64 total items are based on a 7-point Likert-scale format. Internal consistency reliability for the original tool is supported by Cronbach s a coe cients ranging from.75 to.93 for the ve subscales. Content validity for the modi ed tool was established based on the review of literature and consensus among a certi ed lactation consultant (LC) and two doctorally prepared nurse researchers. For this study, all 438 JOGNN, 37, ; DOI: /j x

4 Bernaix, L. W., Schmidt, C. A., Arrizola, M., Iovinelli, D., and Medina-Poelinez, C. subscales of the NSBQ had Cronbach s as above.72, except for one. The combined subscale of normative referents and the motivation to comply with those normative referents had a Cronbach s a of.58 at the 3-month postintervention measurement. Because inconsistency in breastfeeding attitudes and values for each referent used by each nurse would be expected, internal consistency of this subscale would not be anticipated. The Nurse Lactation Survey (NLS), which measured nurses knowledge related to basic lactation physiology, pumping and storage of breast milk, and the nutritional needs of the high-risk infant, was constructed based on a review of literature, input from the research team, and four experts (two certi ed LCs, the study consultant, and one additional doctorally prepared nurse researcher). The tool is comprised of 24 items, presented in a forced-choice (true/false/unsure) format. The highest possible score is 24. Cronbach s as of.52 (pre-intervention) and.55 (3-month postintervention), and a Guttman Split-Half coe cient.57 were obtained. The 2-week test-retest reliability correlation for the NLS was.33 (p 5.162). Procedure There were six phases of data collection. In Phase I, which occurred over a span of 2 months,19 mothers were recruited by a research team member within 2 weeks of their infant s admission to the NICU. After providing written consent, the mothers completed the demographic questionnaire and the MPSQ and deposited them into a locked box located in the unit s family lounge. Two weeks after Phase I ended, Phase II began. During this phase, the NICU nurses were made aware of the prospective study during sta meetings and informed that they would receive a packet of materials in their sta mailbox. Nurses who agreed to participate returned their signed consent form and completed questionnaires, the demographic questionnaire, the NLS, and the NSBQ, to a locked box located in the sta lounge. Phase III began 2 weeks after Phase II and involved the delivery of the 4-hour educational program for the nurses. Two sessions of the program on the same day, one in the morning and one in the afternoon, were o ered to ensure attendance. All of the nurses who consented to participate in the study attended one of the sessions, with approximately half of the study sample attending each session. At the end of each session, the nurses were asked to complete the same knowledge survey as was completed in Phase II. The four-hour educational intervention included practical strategies for the nurse to use to help mothers establish and maintain lactation. Two weeks later, Phase IV began. Nurses completed the NLS for measurement of test-retest reliability during this phase. Phase V, which began 3 months following completion of Phase IV, involved the recruitment of a di erent sample of mothers in order to obtain a postintervention, cross-sectional sense of the NICU s supportive atmosphere for lactation. Two weeks after this phase was completed, the nal phase (Phase VI) of data collection began. This phase involved having the nurses who participated in the prior phases complete the NLS and the NSBQ for one last time. Intervention The 4-hour educational intervention was provided using a lecture and discussion format, and included speci c content related to breast anatomy and lactation physiology, lactoengineering, the nutritional needs of the high-risk infant, and the mechanics of pumping and storing breast milk (Table 1). Nurses were provided with practical strategies for introducing mothers to the topic of providing milk for their infant, for helping mothers establish and maintain lactation, and for providing discharge instruction. The importance of kangaroo care and speci c suggestions for transitioning the infant from bottle to breast were also discussed. In addition, motivational encouragement for the purpose of empowering the nurses was incorporated through- Table 1: Content Outline of the Lactation Educational Program Anatomy and physiology of lactation OMM and immunity protection Kangaroo care Lactoengineering Milk expression guidelines Introducing lactation in the NICU Discharge teaching and counseling Postdischarge resources Sharing the science on human milk with NICU mothers Common questions and concerns of NICU nurses Developing a NICU-based lactation resource team Note.OMM5own mother s milk. NICU = neonatal intensive care unit. JOGNN 2008; Vol. 37, Issue 4 439

5 Lactation Education Program out the session. The intervention was developed and taught by a neonatal nurse practitioner/lc; she was employed at a di erent NICU setting and had considerable previous experience assisting mothers and mentoring nurses with the lactation process in the NICU (Rodriquez, Miracle, & Meier, 2005). Data Analysis Data entry and analysis, using SPSS version 14.0, was conducted by research team members not employed at the institution. Analyses included descriptive statistics for the nurse and mother demographic data, distribution of scores and psychometric evaluation of the study instruments, and repeated measure bivariate analyses. A signi cance level of po.05 was set. Results Table 2: Characteristics of Mother Samples Pre-Intervention 3-Month Postintervention Characteristic (n 519) (n 513) Age (years) (M) Married (n) Education High School diploma 1 2 Associate s degree 3 3 Bachelor s degree 5 2 Master s degree 5 3 Ethnicity White, Non-Hispanic 11 9 Hispanic 5 2 Breastfed previous children (yes) Preferred feeding method before delivery (breast) Previous child in the NICU (yes) Sample Characteristics A sample of 19 mothers participated in Phase I (preintervention) and 13 mothers constituted the second sample of mothers (Phase V, postintervention). Both groups of women were similar in age, education, ethnicity, and marital status (Table 2). Fortyseven percent of the multiparas (n 5 9) in Phase I reported that they had breastfed one or more of their other children and 11% (n 5 2) had a previous child who was hospitalized in a NICU; 84% (n 516) had chosen to breastfeed their infant before delivery. While none of the multiparas in Phase V ever had a previous infant hospitalized in the NICU, 31% (n 5 4) of them had breastfed one or more of their other children. Similar to the women in Phase I, 77% (n 510) of the mothers in Phase V had chosen to breastfeed before the delivery of their infant. Chisquare tests for categoric data and t tests for continuous variables did not reveal any statistically signi cant di erences between the mothers characteristics by data collection point. A sample of 64 nurses consented to be in the study and participated in Phase II (Table 3). They ranged in age from 22 to 57 years, with a mean age of 32. Most of them described themselves as White, non- Hispanic (89%), baccalaureate-prepared (78%), and employed full time (77%). Experience as a NICU nurse ranged from less than 1 to 9 years, with a mean of 7 years. Of the 15 nurses who reported having children, 14 (93%) had breastfed. Seventy- ve percent of the nurses (n 5 48) had never attended any continuing education courses speci c to breastfeeding. Expected attrition and attrition of the original nurse sample due to sta turnover and absences due to illness and vacation during one or more of the data collection periods resulted in 31 nurses completing the nal round of questionnaires. No statistically signi cant di erences for sample characteristics were noted between the initial (n 5 64) and remaining (n 5 32) group of nurses. Correlations Among Pre-Intervention NSBQ Variables Pearson s Product correlations performed on preintervention data revealed that nurses intentions to provide lactation support were positively correlated with nurses age (r 5.51, p.01), years of experience working in the NICU (r 5.54, p.01), subjective normative beliefs (r 5.32, p.05), and lactation beliefs (r 5.40, p.01). Beliefs about lactation, namely the importance of providing OMM for the vulnerable infant and the nurses in uence on that outcome, was positively correlated with the nurses subjective normative beliefs (r 5.39, p.01) and pre-intervention lactation knowledge (r 5.32, p.05). This indicated that as nurses perceived greater support for providing lactation support and had greater lactation knowledge, their 440 JOGNN, 37, ; DOI: /j x

6 Bernaix, L. W., Schmidt, C. A., Arrizola, M., Iovinelli, D., and Medina-Poelinez, C. Table 3: Characteristics of Nurse Sample, n 5 64 Pre-Intervention Postintervention (n 5 32) Characteristic M, SD, Range n (%) M, SD, Range n (%) Age (years) , , Ethnicity White, Non-Hispanic 57 (89%) 29 (91%) Nursing education Associate s degree 3 (5%) 2 (6%) Bachelor s degree 50 (78%) 27 (84%) Some graduate courses 7 (11%) 1 (3%) Master s degree 3 (5%) 2 (6%) NICU experience (years) 7 7.4, , 0-27 Employment status Employed, full time 49 (77%) 27 (84%) Employed, part time 15 (23%) 5 (16%) Previous CE courses speci c to breastfeeding None 48 (75%) 25 (78%) One 10 (16%) 4 (13%) Two 3 (5%) 2 (6%) Three 1 (2%) 1 (3%) Have children 15 (23%) 9 (28%) Breastfed own children Yes 14 (93%) 9 (28%) No 1 (7%) beliefs about lactation support for the vulnerable infant were more positive. Nurses attitudes about providing lactation support however were negatively correlated to lactation beliefs and subjective normative beliefs (r 5.42, p.01 and r 5.32, p.05, respectively); while nurses attitudes about providing lactation support were positive, nurses had negative beliefs about their e ect on lactation outcome for mothers in their care and did not perceive support for their attitudes/actions from their normative referents (nurse peers and administrators). Pre-Intervention/Postintervention Bivariate Analyses Paired t tests revealed that all subscale mean scores of the NSBQ improved signi cantly from the pre-intervention measurement to the 3-month postintervention measurement; nurses intentions (M , SD vs. M , SD5 2.67; po.001), subjective normative beliefs (M , SD vs. M , SD ; po.01), attitudes (M , SD vs. M , SD ; po.001), and lactation beliefs (M , SD vs. M , SD ; po.001) each increased. These di erences indicated a move from a moderately positive to a very positive proclivity for providing lactation support to mothers in the NICU (Table 4). Lactation knowledge scores, as measured by the NLS, also improved at each postintervention time point when compared with the pre-intervention measurement. Mean scores increased from (pre-intervention) to (immediate posttest), (2-week posttest), and (3-month posttest). A bivariate analysis for the e ect of time on knowledge scores revealed statistically signi cant JOGNN 2008; Vol. 37, Issue 4 441

7 Lactation Education Program Table 4: Bivariate Analysis of Nursing Support for Breastfeeding Questionnaire Subscale Scores Pre-Intervention and 3-Month Postintervention Subscale n M SD df t p Normative belief/comply (pre) Normative belief/comply (3 post) Pair o.01 Combined belief/outcome (pre) Combined belief/outcome (3 post) Pair Attitude (pre) Attitude (3 post) Pair o.001 Intention (pre) Intention (3 post) Pair o.001 results for the pre-intervention to immediate postintervention score means (po.000), pre-intervention to 2-week postintervention score means (po.000), and immediate postintervention to 3- month postintervention score means (po.000) (Table 5). Although knowledge scores increased from the pre-intervention measurement to the 3-month postintervention measurement, the di erence was not statistically signi cant (p 5.103). To estimate the risk that our negative ndings were due to Type II error, we conducted post hoc power analyses. Comparing the immediate posttest to the 2-week posttest, the attained sample size was 20 patients and the assumptions included an a of.05, a true di erence between treatments of 2.21U, and a within patient standard deviation of The power analysis indicated power of.80 to detect a di erence. Further analyses also indicated power of.80 to detect a di erence when comparing the pretest with the 3-months posttest (n 5 31, p 5.05 and assuming a true di erence between the treatments of 1.92 U and a within-patient standard deviation of the response variable of 2.60). This provides reasonable evidence that the negative results were not the result of Type II error. Mothers Perceived Support Questionnaire Mean Scores Pre-intervention scores for the MPSQ ranged from 122 to 159 (possible score range was ), with a mean of143 and standard deviation of 9.7, thereby indicating a moderately favorable evaluation of the nurses lactation support received by the rst group of mothers. Mothers Perceived Support Questionnaire scores at the 3-month postintervention measurement ranged from 100 to 204, with a mean of 162 and standard deviation of The di erence in MPSQ mean scores was statistically signi cant (t , p 5.019) thereby indicating that the mothers sampled after the intervention perceived greater Table 5: Bivariate Analysis for Effect of Time on Knowledge Scores Knowledge Pair Mean Difference df t p Pretestçimmediate Posttest o.000 Pretestç2-week posttest o.000 Pretestç3-month posttest Immediate posttestç2-week posttest Immediate posttestç3-month posttest o JOGNN, 37, ; DOI: /j x

8 Bernaix, L. W., Schmidt, C. A., Arrizola, M., Iovinelli, D., and Medina-Poelinez, C. lactation support from the nurses than the pre-intervention mothers. Discussion According to the TRA, the provision of education can positively in uence attitudes and beliefs and therefore intentions to perform a particular behavior (Ajzen & Fishbein, 1980). In this study, the short educational program tested was e ective in improving the sampled NICU nurses lactation knowledge, and intentions, attitudes, and beliefs toward providing lactation support to mothers of high-risk infants. Perceptions of the supportive atmosphere for lactation in the NICU, as identi ed by mothers of hospitalized infants in the NICU, also signi cantly improved following the program. The importance of these ndings is twofold. First, the nurses in this study, before the educational intervention, demonstrated moderately positive attitudes, beliefs, and intentions toward providing lactation support, but were not knowledgeable regarding lactation management. In addition, they did not perceive that the time and energy needed to support lactation were valued in their immediate professional circle, and/ or that their supportive e orts would make a di erence in a mother s ability to provide breastmilk to her infant. The results of this study suggest that arming nurses with the necessary facts about lactation management in the NICU, along with speci c strategies for interacting with mothers who want to provide OMM to their infant, may provide the con dence and motivation nurses need to encourage and assist these mothers. Interestingly, the educational intervention was only 4 hours long as compared with the 8- and 16-hour programs developed by Siddell et al. (2003) and Spatz (2005), respectively. This suggests that an educational program like the one in this study could be coste ective and easy to o er amidst complex sta ng constraints and the hectic NICU environment. NICU nurses lactation knowledge and intentions, attitudes, and beliefs toward providing lactation support improved with education. Secondly, this educational intervention appeared to have had a more long-lasting e ect on improving nurses lactation attitudes, beliefs, and intentions than it did on improving their lactation knowledge. By the 3-month postintervention measure, the signi cant improvements in attitude, belief, and intention scores were maintained, but the improvement in knowledge scores tapered o slightly, thereby indicating a loss in retention. Previous studies that have implemented educational interventions designed to improve NICU nurses lactation knowledge and supportive behaviors have not examined if the bene ts of such programs go beyond the immediate postintervention time interval (Siddell et al., 2003; Spatz, 2005). This study suggests that this intervention had a short-term e ect on knowledge, and that an additional reinforcement or intervention booster o ered no later than 3 months may be required for continued retention of facts. It is important to note that the nurses in this study began with modest beliefs that their normative referents (i.e., nursing coworkers, nurse manager) were supportive of their e orts to promote lactation in the NICU. Immediately following the intervention, and 3 months later, those beliefs were signi cantly stronger. It seems logical that for nurses to feel empowered in their role as lactation advocates, that they not only need improved knowledge and positive attitudes, but also support and recognition from peers and superiors that their time and e orts are valued and important. Hiring LCs to work in the NICU would seem to be an important rst step for demonstrating a supportive climate for lactation and indeed its success in improving lactation initiation and duration in the NICU has been demonstrated (Gonzales et al., 2003). While having an LC employed in the NICU is important, as it provides positive role-modeling and on-site resources for nursing sta, it is not su cient. The culture of the NICU must exude support for lactation promotion in all regards. The sta nurse must be empowered with knowledge and skills to provide the ongoing day-to-day lactation support to mothers. O ering opportunities for sta to attend lactation educational programs, either in-house or at professional conferences, therefore should be a necessary unit goal. In doing so, it suggests a budgetary and time commitment from administrators that is value-laden; namely, lactation promotion and assisting mothers in the provision of OMM is an essential nursing behavior. Having administrators attend these very same sessions would also communicate a sense of importance to the sta nurse. This study had several limitations. The small convenience samples of nurses and mothers limit the generalizability of the study ndings. The nurses primarily represented only one ethnic background (White) and only one practice setting; further, only 64 out of a possible 120 nurses participated and a majority of those participating had not attended any other continuing education program speci c JOGNN 2008; Vol. 37, Issue 4 443

9 Lactation Education Program to lactation besides the current intervention. This self-selection may have produced a highly motivated group of nurse participants who both wanted, and would bene t from, such instruction. Further, any di erences between nurses who participated in the study and those who did not relative to demographic characteristics were not measured. However, the nurse manager indicated that the study population was representative of the nursing sta employed in the unit. The small samples of mothers, which were a result of either high/ low patient acuity/census or the lack of eligible mother-infant pairs, also re ected a very de ned, homogenous group, namely, older, well-educated, married, and mostly White, non-hispanic mothers. However, because a majority of both samples of mothers had chosen to breastfeed before giving birth to a sick newborn, these mothers may have been more knowledgeable about the type/amount of lactation support they needed and their perceptions of support received re ected a very accurate measurement of the supportive atmosphere of the NICU. Nevertheless, actual measurement of nurses behavior may provide a more thorough account of their lactation support and should be considered. Other study limitations include the lack of a control group and the rst time use of the NLS. While the less than acceptable psychometric properties of the NLS may be re ective of the small sample size, modi cations to the tool and continued pilot tests are necessary. Future research should also include replication of the study with larger and more diverse samples of nurses and mothers, and the incorporation of an attention-control intervention to strengthen con dence in study ndings. In addition, the e ect of this educational intervention on NICU lactation initiation and duration rates was not determined and should be incorporated into future study. Prior research has indicated that OMM provides invaluable protection from serious and debilitating complications common to the high-risk infant. Nurses in the NICU can play a crucial role in promoting the provision of OMM to vulnerable infants. Studies have demonstrated that mothers will choose to provide OMM over formula to their sick infant, and achieve lactation success, when presented with accurate and consistent information and technical support from their nurses and other health professionals (Jaeger et al., 1997; Merewood et al., 2003; Miracle et al., 2004; Sisk et al., 2006; Spatz, 2005). Critical to these outcomes is the adequacy and availability of education and training of nursing sta regarding the evidence-based facts and speci c skills necessary to help mothers initiate and manage lactation. Studies have indicated that a change in attitude and an improved knowledge base regarding lactation by NICU nurses can positively in uence the supportive atmosphere and culture of a unit or an entire hospital (Merewood et al., 2003; Spatz, 2005). Such positive changes in culture were mirrored in this study even though there was only 50% nurse involvement. Periodic attendance to educational programs like the one tested in this study, therefore, should be mandated for all NICU nurses. REFERENCES Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting behavior. Englewood Cli s, NJ: Prentice-Hall. American Academy of Pediatrics (AAP). (2005). Breastfeeding and the use of human milk. Pediatrics, 115, Amin, S. B., Merle, K. S., Orlando, M. S., Dalzell, L. E., & Guillet, R. (2000). Brainstem maturation in premature infants as a function of enteral feeding type. Pediatrics, 106, Andorsky, D. J, Lund, D. P., Lillehei, C.W., Jaksic,T., Dicanzio, J., Richardson, D. S., et al. (2001). Nutritional and other postoperative management of neonates with short bowel syndrome correlates with clinical outcomes. Journal of Pediatrics, 139, Association of Women s Health, Obstetric, and Neonatal Nursing. (2007). Breastfeeding support: Prenatal care through the rst year: Evidence-based clinical practice guidelines (2nd ed.). AWHONN. Washington, D.C.: Author. Bernaix, L. (2000). Nurses attitudes, subjective norms, and behavioral intentions toward support of breastfeeding mothers. Journal of Human Lactation, 16, Bernaix, L., Schmidt, C., Jamerson, P., Seiter, L., & Smith, J. (2006). The NICU experience of lactation and its relationship to family management style. MCN: The American Journal of Maternal-Child Nursing, 31, Callen, J., Pinelli, J., Atkinson, S., & Saigal, S. (2005). Qualitative analysis of barriers to breastfeeding in very-low-birthweight infants in the hospital and postdischarge. Advances in Neonatal Care, 5, Campbell, S. H., & Gutman, C. (2006). Challenges of breastfeeding preterm infants. AWHONN Lifelines, 10, Gonzalez, K. A., Meinzen-Derr, J., Burke, B. L., Hibler, A. J., Kavinsky, B., Hess, S., et al. (2003). Evaluation of a lactation support service in a children s hospital neonatal intensive care unit. Journal of Human Lactation, 19, Hill, P. D., Aldag, J. C., Chatterton, R. T., & Zinaman, M. (2005). Primary and secondary mediators in uence on milk output in lactating mothers of preterm and term infants. Journal of Human Lactation, 21, Jaeger, M. C., Lawson, M., & Filteau, S. (1997). The impact of prematurity and neonatal illness on the decision to breast-feed. Journal of Advanced Nursing, 25, Merewood, A., Philipp, B. L., Chawla, N., & Cimo, S. (2003). The babyfriendly hospital initiative increases breastfeeding rates in a U.S. neonatal intensive care unit. Journal of Human Lactation, 19, Miracle, D. J., Meier, P. P., & Bennett, P. A. (2004). Mothers decisions to change from formula to mothers milk for very-low-birth-weight infants. Journal of Obstetric, Gynecologic, and Neonatal Nursing, 33, Pridham, K., Saxe, R., & Limbo, R. (2004). Feeding issues for mothers of very low-birth-weight, premature infants through 444 JOGNN, 37, ; DOI: /j x

10 Bernaix, L. W., Schmidt, C. A., Arrizola, M., Iovinelli, D., and Medina-Poelinez, C. the rst year. Journal of Perinatal and Neonatal Nursing, 18, Rodriguez, N. A., Miracle, D. J., & Meier, P. P. (2005). Sharing the science on human milk feedings with mothers of very-low-birth-weight infants. Journal of Obstetric, Gynecologic, and Neonatal Nursing, 34, Schanler, R. J. (2001). The use of human milk for premature infants. Pediatrics Clinics of North America, 48, Siddell, E., Marinelli, K., Froman, R. D., & Burke, G. (2003). Evaluation of an educational intervention on breastfeeding for NICU nurses. Journal of Lactation, 19, Sisk, P. M., Lovelady, C. A., Dillard, R. G., & Gruber, K. J. (2006). Lactation counseling for mothers of very low birth weight infants: E ect on maternal anxiety and infant intake of human milk. Pediatrics, 117, e67-e75. Spatz, D. (2005). Report of a sta program to promote and support breastfeeding in the care of vulnerable infants at a children s hospital. Journal of Perinatal Education, 14, Woolridge, J., & Hall, W. A. (2003). Posthospitalization breastfeeding patterns of moderately preterm infants. Journal of Perinatal Neonatal Nursing, 17, World Health Organization. (2004). Evidence for the ten steps to successful breastfeeding. Geneva: Division of Child Health Organization. U. S. Department of Health and Human Services. (2000). Blueprint for action on breastfeeding. Hyattsville, MD: Public Health Service. JOGNN 2008; Vol. 37, Issue 4 445