Co-Bedding Twins

Methods Co-Bedding Twins How Pilot Study Findings Guided Improvements in Planning a Larger Multicenter Trial Kathryn Hayward 4 Marsha Campbell-Yeo 4 Sheri Price 4 Della Morrison 4 Robin Whyte Heather Cake 4 Jocelyn Vine Editor’s Note Materials documenting the review process for this article are posted at http://www.nursing-research-editor.com. b Background: Co-bedding, a developmental care practice for twins and multiples, has been theorized as a strategy to decrease the adverse neurodevelopmental effects that may be associated with hospitalization. b Objective: The aim of this study was to determine the feasibility of study design, methods, and the measurement of desired outcomes for the development of a larger multicentered study. b Results: Study findings were used to estimate effect size, determine staff and bedside care organization, evaluate feasibility of data collection measures, and identify issues related to recruitment and follow-up. Results were incorporated in the development of a larger multicentered trial grant proposal. b Discussion: Pilot studies can play an important role in the development of a competitive grant proposal and efficient conduct of a research trial. Pilot studies strengthen a proposal by providing essential baseline information. A general overview of the purpose of pilot studies is provided here, along with a description of the process of using findings from a pilot study to inform the development of a larger multisite trial. Findings from this pilot study examining the effects of co-bedding on twins and their parents are used to revise the research protocol for a larger multisite trial. These changes, which lead to improvement to the protocol, and the rational for these changes are highlighted. b Key Words: co-bedding & feasibility study & pilot study A pproximately 50% of twins and 90% of triplets are born prematurely, often facing medical and developmental challenges. Co-bedding of twins and multiples is a developmental initiative. Its purpose is to minimize neurodevelopmental sequelae (Lutes & Altimier, 2001). Co- bedding is believed to enhance twin coregulation, improve respiratory status, decrease oxygen requirements, increase weight gain, and facilitate mutuality in their circadian rhythms and sleepYawake patterns (Fowler Byers, Yoviaish, Lowman, & Francis, 2003; Hayward, 2003; Lutes & Altimier, 2001; Nyqvist & Lutes, 1998). In 1998, Nyqvist and Lutes proposed that the synactive theory of development (Als, 1986) could provide an explanatory model for understanding how co-bedding may assist preterm twins in coping with the Kathryn Hayward, RN, BScN, IBCLC, is Assistant Professor and Masters Student, Dalhousie University School of Nursing; and Staff Nurse, IWK Health Centre, Halifax, Nova Scotia, Canada. Marsha Campbell-Yeo, MN, is Neonatal Nurse Practitioner, IWK Health Centre, Halifax, Nova Scotia; Adjunct Professor, Dalhousie University School of Nursing, Halifax, Nova Scotia; and Doctoral Student, McGill University, Montreal, Quebec, Canada. Sheri Price, MN, is Research Associate, Interdisciplinary Research, IWK Health Centre, Halifax, Nova Scotia; Adjunct Faculty, Dalhousie University School of Nursing, Halifax, Nova Scotia; and Doctoral Student, University of Toronto, Toronto, Ontario, Canada. Della Morrison, MN, is Neonatal Nurse Practitioner; Heather Cake, MA, is Psychologist; and Jocelyn Vine, MN, is Vice President, Patient Care, IWK Health Centre, Halifax, Nova Scotia, Canada. Robin Whyte, MB, is Associate Professor and Neonatalogist, IWK Health Centre and Dalhousie University, Halifax, Nova Scotia, Canada. Nursing Research March/April 2007 Vol 56, No 2 Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 137 138 Co-Bedding Twins: A Pilot Study extrauterine environment. Maintaining continued physical contact by swaddling twins in the same blanket and boundary, provides them with the opportunity to coregulate and to continue to progress in their unique interactive development (Fowler Byers et al.; Hayward; Lutes & Altimier; Nyqvist & Lutes). Despite the theorized benefits, due to a paucity of research on the practice of co-bedding, neonatal care providers do not have adequate information to make evidencebased policy decisions (National Association of Neonatal Nurses, 2001). The term pilot study is often used interchangeably with feasibility study. In actuality, pilot studies refer to two very different types of studies: feasibility studies conducted to test the research plan and studies conducted to pretest a particular instrument (Burns & Grove, 2005; van Teijlingen & Hundley, 2001). Roberts and Burke (1989) define a pilot study as ‘‘a small field test, often done at the end of the planning phase, to test methods or feasibility for fine tuning the full proposal’’ (p. 366). The purpose of this pilot study was to determine the feasibility of study design, methods, and the measurement of desired outcomes for the development of a larger multicentered study to examine the effects of cobedding on twins and their parents. Feasibility studies are preparatory studies designed to assess study design, study methods, and outcome measurement, which can ultimately provide ‘‘convincing evidence that a largescaleclinical trial is feasible and well worth the effort’’ (Melnyk & FineoutOverholt, 2005, p. 267). A pilot study provides an opportunity for the researcher to obtain valuable information related to the actual conduct of the trial and to assess the adequacy of the data collection plan (Burns & Grove, 2003; Polit & Hungler, 1999). It is not unusual for problems to arise in the course of a project that result in a delay in the study while changes are made to the research design. In quantitatively designed studies, data collected prior to implementation of any major alterations in the protocol often have to be discarded from the final analysis (Polit, Beck, & Hungler, 2001). ‘‘A pilot study is critical in determining the feasibility of subject Nursing Research March/April 2007 Vol 56, No 2 enrollment, the intervention, the protocol or data collection plan for the study, and the likelihood that subjects will complete follow-up measures’’ (Melnyk & Fineout-Overholdt, 2005, p. 266). In a pilot study, the basic mechanics of the trial can be assessed for their ease of administration within the clinical environment. Properly conducted randomized control trials can be quite lengthy and expensive. Funding agencies generally include feasibility as part of their scoring template prior to the allocation of any funding. Scores reflect the agency’s level of assurance that the study can be completed in the proposed timeframe and that all conditions ensuring minimal loss of data have been addressed (Lancaster, Dodd, & Williamson, 2004). A pilot study strengthens a proposal by providing essential baseline information related to the development and justification of the trial, data to determine an appropriate power analysis, and an opportunity to carry out a dry run of the study to work out any kinks in the proposal (Durso & Mellgren, 1989; Lancaster et al., 2004). ‘‘A wellconducted pilot study, giving a clear list of aims and objectives within a formal framework will encourage methodological rigor, ensure that the work is scientifically valid and publishable, and will lead to higher quality RCT’s’’ (Lancaster et al., 2004, p. 307). A high-quality clinical trial proposal increases the likelihood of success from funding agencies. The design of a pilot project does not negate the need for scientific rigor or ethical protocol. Adherence to research standards is just as important in a pilot study as it is in a larger trial. Maintaining accuracy within a pilot project ensures the likelihood that any weaknesses or inadequacies intrinsic to the proposed larger trial will be uncovered (Polit & Hungler, 1999). Within a pilot project, participants should be representative of the proposed population for the larger trial. This will provide the researchers the ability to assess the acceptability of the trial to potential research participants (Wittes & Brittain, 1990). Furthermore, ethical approval and informed consent is required for any research involving human participants. Pilot Project: Methods Design A randomized control design was utilized in this pilot study to compare the feasibility of data collection related to parental self-efficacy, parental anxiety, infant coregulation (defined as synchrony of infant state, heart rate, respirations, temperature, and oxygen saturation), infection rate, and incidence of caregiver error when twins are co-bedded (cared for in the same incubator) versus standard care (cared for in separate incubators). Ethical approval was obtained through the IWK Health Centre Research Ethics Board. Twins were randomized into two groups. Stratification of eligible twins to less than 2,000 g and to greater than and equal to 2,000 g was the initial plan for the larger study. Given the increased risks associated with the lower weight group, enrollment was limited to twins weighing less than 2,000 g. This ensured that all safety issues for the smaller infants at the greatest risk would be addressed. Dyads in the experimental group were placed in a common twin (Giraffe, GE Medical, Fairfield, CT) incubator for care. Twins in the control group received standard neonatal care in separate incubators. To decrease possible bias related to nursing care practices, twin dyads, regardless of group allocation, were assigned each shift to be cared for by the same nurse. Sample Sample size was time specific as opposed to number specific. The study took place over a 3-month span of time and eligible sets of twins being cared for in a tertiary level neonatal intensive care unit were recruited. It was anticipated that 6Y12 twin dyads would meet inclusion criteria. After ethical approval, parents of twins who met the inclusion criteria (G2,000 g and stable, defined as no endotracheal tubes, no umbilical lines, no symptoms of sepsis, and no chest tubes) were approached by a research assistant. The study was explained, and consent for participation was obtained. Group allocation was assigned randomly using opaque envelopes. During the study period, eight sets of twins were admitted to the neonatal intensive care Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Co-Bedding Twins: A Pilot Study 139 Nursing Research March/April 2007 Vol 56, No 2 M = 47.8 days; standard: M = 52.7 days). However, there was a wider variation in the length of the hospital stay for the co-bedding group (co-bed: M = 57 days; standard: M = 21 days; Table 2). q TABLE 1. Frequency Distribution Sample characteristics Marital/Common-law status Male twins Female twins Maternal age 930 years Paternal age 930 years Vaginal delivery Caucasian Group (n) Co-bed (n) Standard (n) 6 8 4 3 3 8 12 3 4 2 3 3 4 6 3 4 2 0 0 4 6 and met the study inclusion criteria. Six families were recruited. The primary hypothesis of the pilot study is that co-bedding has a positive effect on parental self-efficacy. The secondary hypothesis is that there are differences between groups with relation to infant coregulatory behavior, parental anxiety, incidents of infection, and frequency of caregiver errors. When completing a pilot study, it is important to have clearly identified outcome measures, with reliable methods of measurement for each outcome. Sample Characteristics The frequency distribution of the sample is shown in Table 1. All participants in the study were Caucasian. There was a difference in the distribution by age, education, and employment status. All parents in the co-bedding group were older than 30 years and all parents in the standard care group were younger than 30 years. Only two parents in the standard care group had greater than Grade 12 educations compared with five parents in the co-bedding group. More parents in the standard care group were employed (five parents as opposed to three parents). Both groups were equal in the number of cesarean sections compared with vaginal births. Two out of three sets of parents in each group were first-time parents. Although the groups differed in gestational age at birth (co-bed: M = 31.8 weeks; standard: M = 32.6 weeks; Table 2), gestational age at entry to the study was similar for both groups (co-bed: M = 33.6 weeks; standard: M = 33.3 weeks). Both groups had equal numbers of male and female infants (male = 66.67%; Table 1). Birth weight for the two groups were different (co-bed: M = 1,717 g; standard: M = 1,520 g; p = .002; Table 2). Overall, the co-bedding group had a shorter length of hospital stay (co-bed: Data Collection Parental self-efficacy was measured using the Infant Care Survey (ICS; Froman & Owen, 1989, 1990) and parental anxiety was measured using the Speilberger State-Trait-Anxiety Inventory (STAI; Speilberger, Gorsuch, Lushene, Vagg, & Jacobbs, 1983). Parents were given the ICS and STAI to complete upon entry to the study, before discharge, and at 1-month postdischarge. The ICS has been shown to have reliability (internal consistency " = .975) and validity (F = 9.069, p G .001, df = 5,136) for measurement of self-efficacy in maternal knowledge and skills related to infant care. The STAI is a proven tool, with a Cronbach’s " of .93 for State (Byers, Yovaish, Lowmand, & Francis, 2003) and .92 for Trait (Speilberger et al., 1983), for measuring anxiety and has been used extensively in research and clinical practice. Both State and Trait subscales were utilized in the pilot study as neonatal intensive care unit parental stress would be expected to be associated with both State and Trait Anxiety (Franck, Cox, Allen, & Winter, 2005). The tool has demonstrated high testYretest reliability, with a Pearson coefficient of .96 for the State subscale q TABLE 2. Descriptive Statistics Variable Sample M SD Range SE Length of hospital stay (days) Group Co-bed Standard Group Co-bed Standard Group Co-bed Standard Group Co-bed Standard 50.25 47.8 52.67 32.214 31.89 32.619 1,619 1,717 1,520 33.428 33.57 33.285 18.28 25.7 7.63 1.283 1.155 1.514 410 410 423 1.525 1.87 1.505 17Y74 days 17Y74 days 46Y67 days 30 4/7Y34 wks 31Y34 days 30 4/7Y32 6/7 wks 1025Y2380 gms 1380Y2380 gms 1025Y2190 gms 31 3/7Y34 6/7 wks 31 3/7Y34 6/7 wks 31 3/7Y34 6/7 wks 5.28 10.5 3.12 0.524 0.667 0.874 118 167 173 .623 1.08 .869 Gestational age at birth (weeks) Infant birth weight (g) Gestational age at entry to study (weeks) Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 140 Co-Bedding Twins: A Pilot Study Nursing Research March/April 2007 Vol 56, No 2 q TABLE 3. Descriptive Statistics Variable Quiet sleep Crying Coregulated Study group M Co-bedding care 5.357 Standard care 0.780 Co-bedding care 5.341 Standard care 0.1944 Co-bedding care 14.895 Standard care 13.941 and .98 for the Trait subscale (Speilberger et al., 1983). The Nursing Child Assessment Sleep/ Activity Record (NCASAR; Barnard, 1999) was utilized to capture coregulatory behavior. A minimum of eight 5-hr NCASARs were completed by the research assistant on each set of twins on a biweekly basis. All activities (feedings, blood work, x-rays, etc.), physiological parameters, and the infant’s state were tracked on the form. The NCASARs for each dyad were compared for synchrony of activity (coregulatory behavior) by recording infant state, heart rate, temperature, respirations, and oxygen saturation every 15 min for 5 hr. Physiological parameters were captured noninvasively through the use of the Hewlett Packard monitoring system. Interrater reliability was assessed by having both the research assistant and a certified Nursing Child Assessment Satellite Training provider assess the infant state simultaneously for the first two 5-hr observation periods. Interrater reliability was scored using Cronbach’s " at .98. The research assistant, using ongoing chart review, collected data on heart rate, respiration, temperature, oxygen saturation, infection rates, caregiver error, infant feeding, and infant demographics using two separate data collection tools developed for this study. Infection rate was determined by three separate measures: incidence of septic workup, treatment with antibiotics, and confirmed incidence of sepsis. Caregiver errors were collected from readily available quarterly reports. All data collected were kept confidential and individual identities were protected throughout the analysis process. SD SE Minimum Maximum 2.33 1.27 2.352 0.4672 3.315 2.045 0.359 0.199 0.367 0.078 0.760 0.496 1 0 1 0 10 9 12 4 12 2 20 19 Pilot Study: Results Data were insufficient to analyze parental self-efficacy and parental stress. The co-bedding group spent more time in quiet sleep (co-bedding group: M = 5.36, SD = 2.33; standard care group: M = 0.78, SD = 1.27; p = .00) and more time crying (cobedding group: M = 5.34, SD = 2.35; standard care group: M = 0.91, SD = 0.47; p = .00; Table 3). Although the co-bedding group appeared to spend more time coregulated, significant findings were not found (co-bedding group: M = 14.90, SD = 3.32; standard care group: M = 13.94, SD = 2.05; p = .31; Table 3). No caregiver errors or infections occurred with either group. Staff and Bedside Issues The clinical application of any research protocol is paramount to its success. Plans for the setup of the site had to be organized ahead of time and well considered. A single twins, incubator was set up between two infant care sites. The Giraffe incubator has a large mattress surface and permits access from both sides (Figure 1). To facilitate safety, each twin and his or her equipment was assigned a side of the incubator and a specific color code. All cables and tubing for each infant exited on his or her assigned side. Color coding decreased the risk of treatment errors and cross contamination from shared equipment (i.e., soother, temperature probe, etc.) by providing an easily distinguishable marker for infant identification. A second incubator was readily available for separation of the twins, if necessary. Implications for Larger Multisite Trial Through the provision of information for the planning and justification of randomized controlled trials, pilot studies play a key role in health research (Anderson & Prentice, 1999). Findings from this pilot study proved to be essential in the development of a larger research proposal. Alterations to the methods sections, sources of FIGURE 1. Co-bedding site using twin Giraffe incubator. Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Co-Bedding Twins: A Pilot Study 141 Nursing Research March/April 2007 Vol 56, No 2 potential bias, data collection and recruitment and follow-up were incorporated into the design of a larger multisite trial that was successful in receiving provincial funding. Revised Protocol: Revisions to Methods Sample One of the first steps in study design is to calculate the threshold effect size required for detecting appropriate statistical power. The effect size provides the researcher with some assurance that the study will possess adequate sensitivity to the treatment effects under investigation (Lipsey, 1990). In the absence of a priori data, data derived from a pilot study can provide the researcher with valuable information regarding effect size. Based on coregulatory data collected in the pilot study (specifically, time spent in quiet sleep), the population effect was set at 0.05. The multisite trial significance was set at .05, and power was set at 0.80, resulting in a sample size of 152 sets of twins (76 sets co-bedding and 76 sets receiving standard care). Because the effect size was based on coregulatory data, the primary hypothesis in the multisite trial was changed to ‘‘cobedding twins will increase twin coregulatory behavior.’’ Inclusion criteria for the pilot study was based on weight (e2,000 g). Weight was initially believed to be an important factor because it generally directed whether infants would be cared for in incubators or open cribs. In this unit, the transition of infants to cribs upon reaching 2,000 g was considered to be standard practice (Figure 2). It is believed that an infant’s ability to alter state and physiological status is correlated more closely to gestational age and maturation than infant weight (Anders, Keener, & Kraemer, 1985; Holditch-Davis, Scher, Schwartz, & Hudson-Barr, 2004). Coregulation is defined as newborn twins’ activities for supporting one another in the physiologic transition to postnatal life and in the achievement of stable sleepYawake states through activities mediated by physical contact (Als, 1986). With the change in the primary hypothesis, it was believed that the inclusion criteria should reflect gestational age as opposed to infant FIGURE 2. Co-bedding twins in crib. weight. Therefore, in the larger study, infants were stratified by gestational age (G31 6/7 weeks or Q32 weeks) and site location. Exclusion criteria for the pilot study included a weight less than 1,000 g, ventilatory assistance, sepsis, chest tubes, umbilical lines, and congenital anomalies. When one twin of a set was placed on phototherapy, questions related to the unnecessary treatment for the second twin arose. Given that a moderate level of unconjugated bilirubin is not considered to be harmful, and given that the antioxidant properties of bilirubin may actually provide some protective qualities, it was determined that ‘‘current treatment with phototherapy’’ be added as an exclusion criteria for the larger study (Figure 3). Protecting Against Possible Bias The research assistant’s awareness of the randomization allocation provided the opportunity for bias. To eliminate any potential bias related to randomization, a 24-hr accessible computergenerated randomization site will be utilized for the larger study. The larger study was also strengthened by blinding the data collection procedure. This was accomplished by using videotaped data collection for infant state, as well as continuous recording using Somte sleep screen for collection of physiological parameters. One research nurse will approach the parents to introduce the study, obtain consent, and film the infants for coding. Two separate research assistants, blinded off-site and who would not have access to the unit or familiarity with the study, will code either the study group or the control group for infant state. Data Collection Because baseline data were collected after randomization, it was believed that knowledge of randomization allocation may have affected the results of the parental questionnaires, and although the data on the twins were collected within 24 hours, even this short duration of co-bedding appeared to have a possible effect on their state regulation. It was determined that baseline data on all twins and parents needed to be collected prior to randomization. At the time of data entry, it was determined that a portion of the data were redundant. The data collection procedure was simplified for the larger study. Eight 5-hr data collection periods were reduced to three 3-hr data collection sets on study days 1, 3, and 7. Data were analyzed using repeatedmeasure ANOVA and Pearson’s r. This remained unchanged for the larger study. Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 142 Co-Bedding Twins: A Pilot Study Nursing Research March/April 2007 Vol 56, No 2 study prior to developing a proposal for a larger multisite trial. Information derived from pilot studies can determine feasibility, strengthen project design, improve funding success, and ultimately enhance the outcomes of a larger study. Hopefully, other researchers will recognize the important role that pilot studies play in health research and consider the use of a pilot study in preparation for larger trials. q FIGURE 3. Co-bedding twins receiving phototherapy in incubator. Recruitment and Follow-up The pilot study yielded a poor parental questionnaire response: 100% completion of questionnaires at enrollment, 37.5% at discharge, and 0% at 1 month. This low-level rate of return is believed to be due to the busy schedule of parents of multiples (Holditch-Davis, Roberts, & Sandelowski, 1999). Followup is important to determine infant sleep practices after discharge and to assess the impact of sleep decisions on parental self-efficacy and stress. As a result, the budget was adjusted to include a small compensation to be mailed with the final questionnaires. In addition, the research nurse will conduct a 1-month discharge telephone interview to complete the questionnaires and to assess current infant sleep arrangements. A community partnership was established with the Parents of Multiple Birth Association. It is believed that this collaboration would promote awareness of the study prior to birth and encourage ownership of the study by the study participants, which will hopefully encourage completion of the questionnaires. Including key stakeholders was anticipated to improve follow-up and dissemination of findings. Determining the recruitment rate is an important aspect of the pilot study because it permits the research- ers to assess accurately an enrollment rate for the larger study and to determine the yearly budget breakdown when applying for funding (Lancaster et al., 2004). The pilot study had a 70% recruitment rate. Based on this rate, it is anticipated that it would take 2.5 years to complete data collection in the multisite trial. Revised Protocol: Anticipated Outcomes and Significance This pilot study was developed to assess the feasibility of conducting a multicenter trial examining parental self-efficacy, parental anxiety, infant coregulation, infant safety, infant physiological data, and caregiver error with twins who are assigned randomly to an experimental group who are co-bedding and to a control group of twins who are receiving standard care. In the pilot study, the use of the three chosen data collection techniques for examining the relationship between these variables was assessed. The information obtained during this pilot was used to develop a proposal that was successful in securing external grant support for a larger multisite trial. This article demonstrates the considerable benefits of conducting a pilot Accepted for publication November 6, 2006. Thank you to Kelly Robson for her assistance with data collection; Colleen O’Connell for her support with data analysis and completion of the power analysis for sample size for the multisite trial; and Celeste Johnston for her ongoing mentorship throughout the development of this pilot study and the larger multisite trial. Corresponding author: Kathryn Hayward, RN, BScN, IBCLC, Dalhousie University School of Nursing, 5869 University Avenue, Halifax, Nova Scotia, Canada B3H 3J5; or PO Box 9700 IWK Health Centre, 5850/ 5980 University Avenue, Halifax, Nova Scotia, Canada B3K 6R8 (e-mail: kathryn. hayward@dal.ca). References Als, H. (1986). A synactive model of neonatal behavioral organization: Framework for the assessment of neurobehavioral development in the premature infants and parents in the neonatal intensive care environment. Physical and Occupational Therapy in Pediatrics, 6(3/4), 3Y53. Anders, T. F., Keener, M. A., & Kraemer, H. (1985). SleepYwake state organization, neonatal assessment and development in premature infants during the first year of life. II. Sleep, 8(3), 193Y206. Anderson, G. L., & Prentice, R. L. (1999). Individually randomized intervention trials for prevention and control. Statistical Methods in Medical Research, 8, 287Y309. Barnard, K. E. (1999). Beginning rhythms: The emerging process of sleep wake behaviors and self-regulation. Seattle: NCAST, University of Washington. Burns, N., & Grove, S. K. (2003). Understanding nursing research (3rd ed.). Philadelphia: W. B. Saunders. Burns, N., & Grove, S. K. (2005). The practice of nursing research: Conduct, critique and utilization. St. Louis, MO: Elsevier Saunders. Byers, J. F., Yovaish, W., Lowmand, L. B., & Francis, J. D. (2003). Co-bedding versus single-bedding premature multiplegestation infants in incubators. Journal of Obstetrical, Gynecologic, and Neonatal Nursing, 32(3), 340Y347. Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Co-Bedding Twins: A Pilot Study 143 Nursing Research March/April 2007 Vol 56, No 2 Durso, F. T., & Mellgren, R. L. (1989). Thinking about research. St. Paul, MN: West Publishing. Fowler Byers, J., Yovaish, W., Lowman, L. B., & Francis, J. D. (2003). Cobedding versus single-bedding premature multiple-gestation infants in incubators. Journal of Obstetrical, Gynecologic, and Neonatal Nursing, 32(3), 340Y347. Franck, L. S., Cox, S., Allen, A., & Winter, I. (2005). Measuring neonatal intensive care unit-related parental stress. Journal of Advanced Nursing, 49(6), 608Y615. Froman, R. D., & Owen, S. V. (1989). Infant care self-efficacy. Scholarly Inquiry for Nursing Practice, 3(3), 199Y211. Froman, R. D., & Owen, S. V. (1990). Mothers’ and nurses’ perceptions of infant care skills. Research in Nursing & Health, 13, 247Y253. Hayward, K. M. (2003). Cobedding of twins: A natural extension of the socialization process? MCN: The American Journal of Maternal Child Nursing, 28(4), 260Y263. Holditch-Davis, D., Roberts, D., & Sandelowski, M. (1999). Early parental interactions with and perceptions of multiple birth infants. Journal of Advanced Nursing, 30(1), 200Y210. Holditch-Davis, D., Scher, M., Schwartz, T., & Hudson-Barr, D. (2004). Sleeping and waking state development in preterm infants. Early Human Development, 80(1), 43Y64. Lancaster, G. A., Dodd, S., & Williamson, P. R. (2004). Design and analysis of pilot studies: Recommendations for good practice. Journal of Evaluation in Clinical Practice, 10(2), 307Y312. Lipsey, M. W. (1990). Design sensitivity: Statistical power for experimental research. Newbury Park, CA: Sage. Lutes, L., & Altimier, L. (2001). Co-bedding multiples. Newborn and infant nursing reviews, 1(4). Retrieved July 12, 2006, from http://www.nainr.com/scripts/om. dll/serve. Melnyk, B. M., & Fineout-Overholt, E. (2005). Generating evidence through quantitative research. Evidence-based practice in nursing & healthcare: A guide to best practice. Philadelphia: Lippincott, Williams & Wilkins. National Association of Neonatal Nurses. (2001). Co-bedding twins or higher order multiples. Retrieved July 12, 2006, from http://www.nann.org/public/ articles/details.cfm?id=42. Nyqvist, K. H., & Lutes, L. M. (1998). Co-bedding twins: A developmentally supportive care strategy. Journal of Obstetrical, Gynecologic, and Neonatal Nursing, 27(4), 450Y456. Polit, D. F., Beck, C. T., & Hungler, B. P. (2001). Essentials of nursing research: Methods, appraisal, and utilization (5th ed.). Philadelphia: Lippincott. Polit, D. F., & Hungler, F. P. (1999). Overview of the research process. Nursing research: Principles and methods (6th ed.). Chestnut Hill, MA: Lippincott. Roberts, C. A., & Burke, S. O. (1989). Nursing research: A quantitative and qualitative approach. Boston: Jones and Bartlett. Speilberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). State trait anxiety inventory for adults. Redwood City, CA: Mind Garden Publishing. Van Teijlingen, E. R., & Hundley, V. (2001) The importance of pilot studies. Sociology at Surrey, (35). Retrieved July 12, 2006, from http://www.soc.surrey. ac.uk/sru/SRU35.html. Wittes, J., & Brittain, E. (1990). The role of internal pilot studies in increasing the efficiency of clinical trials. Statistics in Medicine, 9(1Y2), 65Y72. Call for Manuscripts on Violence, Injury and Human Safety Nursing Research and Advances in Nursing Science are pleased to announce a collaborative initiative that will feature articles focused on violence, injury and human safety. Nursing Research invites manuscripts that report research findings that contribute to knowledge about violence, injury and human safety including methodologic concerns specific to cross cultural and international research. Advances in Nursing Science invites manuscripts that address a broad range of theoretical, methodologic, ethical and philosophic issues related to violence, injury and human safety knowledge development. The focus of each journal is not exclusive; research findings and underlying conceptual or philosophic issues can be included in manuscripts submitted for either journal. We encourage research teams to submit complimentary manuscripts to each journal simultaneously, but complimentary submissions are not required. The Editors welcome e-mail or telephone discussions with potential authors. Due date: October 1, 2007. The companion issue of each journal will be published in May/June 2008. Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.