This project was partially funded by The Swedish Council on Technology Assessment in Health Care—SBU Alert. We are thankful to Karin Rydin at SBU for performing the literature search and for valuable comments from the SBU Alert council on the Swedish report (Wallin & Eriksson 2006).
Newborn Individual Development Care and Assessment Program (NIDCAP): A Systematic Review of the Literature
Article first published online: 29 APR 2009
©2009 Sigma Theta Tau International
Worldviews on Evidence-Based Nursing
Volume 6, Issue 2, pages 54–69, June 2009
How to Cite
Wallin, L. and Eriksson, M. (2009), Newborn Individual Development Care and Assessment Program (NIDCAP): A Systematic Review of the Literature. Worldviews on Evidence-Based Nursing, 6: 54–69. doi: 10.1111/j.1741-6787.2009.00150.x
- Issue published online: 11 JUN 2009
- Article first published online: 29 APR 2009
- Accepted 21 October 2008
- systematic review;
- randomized controlled trial;
- neonatal care;
- developmental care;
- cognitive development;
- psychomotor development
Background: Important advancements have been made in the care of preterm infants. Health services have introduced various methods aimed at promoting attachment, breastfeeding, and neurological development. The Newborn Individualized Developmental Care and Assessment Program (NIDCAP), developed to stimulate preterm infants at levels adapted to the child's degree of neurological maturity, is increasingly being used.
Objectives: The aim was to investigate the impact of NIDCAP on the psychomotor development, neurological status, medical/nursing care outcomes, and parental perceptions. A further aim was to evaluate the cost-related effects of NIDCAP.
Data Extraction and Analysis: A literature search up to September 2007 was performed. The reviewed papers were assessed for methodological quality and only statistically significant findings were extracted.
Findings: The evidence compiled on the effects of NIDCAP is based on 12 articles from six randomized controlled trials that included approximately 250 children. Each of the studies was assessed as having medium quality. Most of the studies were small and many investigated a huge number of outcome variables, which decreased their scientific strength. On outcome variables in which a significant difference was found between the intervention (NIDCAP) and control groups, most studies showed better results for the NIDCAP group. This was particularly valid for cognitive and psychomotor development. Four studies also showed a reduced need for respiratory support for the NIDCAP group. No studies were identified that weighed the total cost of NIDCAP against its effects.
Conclusions: Despite promising findings, primarily on cognitive and motor development, the scientific evidence on the effects of NIDCAP is limited. Shortcomings in design and methods in the reviewed studies hamper far-reaching claims on the effectiveness of the method. Scientific grounds for assessing the effects of NIDCAP would be substantially enhanced by a sufficiently comprehensive study with extended follow-up and a clear focus on a few important outcome variables.
The care of sick newborn and premature infants has undergone major developments in recent years. Various methods aimed at promoting bonding, breastfeeding, and neurological development have been introduced by neonatal health care providers. These methods include kangaroo care and family-centered neonatal care (Beal 2005; Griffin 2006). However, scientific evidence for the effects of these methods has often been lacking. One common property of these methods is their focus on the infant as an individual with the ability to interact with its surroundings, which is in sharp contrast to earlier ideas in which it was assumed that the newborn infant was principally controlled by reflexes and innate behavior.
Over the past 20 to 30 years, the incidence of preterm birth in most developed countries has been between 5% and 7% of live births. The incidence in the United States is higher at about 12%. (Tucker & McGuire 2004). Medical and technological advances have led to the survival of an ever greater number of infants with low birth weight and those born after even shorter periods of gestation (Hack & Fanaroff 2000), implying that the infant and the families will be in contact with intensive care facilities for a long period. This is often a distressing and stressful period for the family members (Dudek-Shriber 2004; Franck et al. 2005).
Premature infants have a higher neonatal morbidity than full-term infants, where an increase in risk is inversely correlated with the length of gestation. Problems that may arise include those of the central nervous system, eyes, and lungs. There are also negative long-term effects on school achievement and behavior (Hack & Fanaroff 2000).
The Neonatal Individualized Developmental Care and Assessment Program (NIDCAP) is an interventional model that has been developed with the aim of providing adequate sensorial stimulation to premature infants at a level that is adapted to the degree of neurological maturity of the infant. NIDCAP is a registered trademark, with training offered by certified NIDCAP centers, mainly in the U.S. but also in Argentina, Belgium, France, Sweden, the Netherlands, and United Kingdom (Als 2007).
NIDCAP requires extensive training and education of the NICU staff, which is provided by the NIDCAP centers. The NIDCAP Federation International suggests a 5-year strategically planned process to implement NIDCAP in a neonatal setting. At least two developmental care specialists should be trained at each unit that wants to implement NIDCAP. In addition, a core group of nursing staff should take the introductory course or the full training in NIDCAP, and a multidisciplinary leadership support team should be formed. Protected time for regular observations of targeted preterm infants must be reserved after introduction. Refresher courses are necessary in order to ensure that the method continues to be applied correctly (Als 2007).
The NIDCAP Intervention
NIDCAP has been developed at the Children's Hospital in Boston (Als 2007). Briefly, the method involves premature infants undergoing systematic observation of their behavior at intervals of 7 to 10 days. The results of these observations form the basis for the design of individual care plans that describe how the infant should be cared for and how the care can be designed such that it provides stimuli that he or she can cope with.
The theoretical background of NIDCAP is known as the “synactive theory,” a conceptual foundation that describes the neurological development of the fetus and infant (Als 1982). The theory proposes that neurological subsystems (the autonomic, motor, state organizational, and attentional systems) develop in a certain order. These subsystems interact and influence each other, and the infant attempts to maintain balance within and between the subsystems by self-regulation.
The infant is observed by a specific schedule in 2-minute cycles before, during, and after a care procedure, such as change of diaper (Als et al. 2005). The observer notes the existence of physiological and behavioral parameters grouped into the following areas: respiration, color and visceral parameters, motor functions, facial features, state, and attention. Other factors noted are the posture of the infant, the care manipulations that are performed, and any stimuli (sound, light, activity) that are present in the surroundings. Observation is carried out for approximately 10 minutes before the intervention that will be studied in order to obtain baseline values. The infant is then observed throughout the complete procedure and for about 10 to 15 minutes after the intervention is finished or until the infant has returned to its initial status.
After the observation is complete, the observed neurological signals are summarized and related to what happened with the infant and in surroundings. Observed behavior and physiological reactions are classified as signs either of approach/self-regulation on the one hand, or of avoidance/inadequate self-regulation on the other.
The observer then draws up care recommendations that are based on the assessment of which situations the infant “coped with” and which situations produced signs of avoidance. These recommendations normally deal with:
- • the surroundings in the room (location of the incubator relative to windows, sinks, desks, etc., lighting, sound, and activities),
- • the environment in the incubator or bed (use of nest and incubator cover),
- • aids to self-regulation (positioning, aids for sucking or gripping, and eye protection),
- • timing and coordination of care operations, the daily rhythm, and
- • facilitating the transition between different events.
The care plan should be passed on verbally and in writing to those who care for the infant and the infant's parents. The infant is continuously developing neurologically and thus the observation is repeated at intervals of 7–10 days, at which time a new or modified care plan is drawn up.
Relationship with Other Methods
Conventional neonatal care, that is, care without the use of NIDCAP, has changed since the first NIDCAP study was published in 1986. At that time, premature infants were kept naked in intensely lit incubators, on smooth mattresses, often with the head and extremities in a fixed position. The environment in many neonatal wards has changed since then. Such items as incubator covers, nests, and subdued illumination are now used, even if NIDCAP observations are not carried out.
Kangaroo care (skin-to-skin care; World Health Organisation 2003) is a method of care that, as it is used in the western world, emphasizes contact between the infant and the parents and support for breastfeeding. The kangaroo method can be used on its own, as is done in many parts of the world, or it may form one component of NIDCAP. A further concept that supports the contact between the infant and his/her parents is family-centered care, which is defined by such properties as unlimited visiting hours, participation of the parents in the planning of care, and an adaptation of the surroundings such that they stimulate parent's presence and participation (Harrison 1993; Levin 1999; Sizun et al. 1999).
Previous Reviews of NIDCAP
Two previous reviews on the effects of NIDCAP have been published. Jacobs and colleagues (2002) included five articles from randomized controlled studies (RCT) and three articles from historical phase-lag design studies. A Cochrane report by Symington and Pinelli (2006) that examines a wide spectrum of developmental supportive interventions, including NIDCAP, was first published in 2003. Its latest version included nine articles; all RCTs. Both reviews perform meta-analysis where outcomes and time points were comparable. Compared to these reviews, the current review adds an exclusive focus on NIDCAP (not assessing other interventions). It also includes additional studies on the NIDCAP intervention.
This systematic literature review addresses the question of what effects the use of NIDCAP will have on the premature infant regarding psychomotor development, neurological status, medical/nursing care outcomes, and parental perceptions during the inpatient period. The review, which was originally carried out as a project within The Swedish Council on Technology Assessment in Health Care (SBU; Wallin & Eriksson 2006), also addresses the cost-related effects of the NIDCAP intervention.
Literature searches for the period 1950 to September 2007 were conducted in the databases PubMed, the Cochrane Library, and OHE HEED. Table 1 presents all search terms and limitations used. Additionally, reference lists of relevant articles were screened. Information about unpublished studies is not included in the review but presented in the discussion.
|Search strategy: effects of NIDCAP|
|PubMed 1950–2007 (September)|
|Individualized developmental care (TW)|
|Search strategy: economic aspects|
|PubMed 1950–2007 (September)|
|Individualized developmental care (TW)||And||Economics|
|Search strategy: economic aspects|
|OHE HEED 1967–2007 (September)|
|NIDCAP (all data)|
|“Individualized developmental care” (all data)|
|Search strategy: treatment effects and economic effects|
|Cochrane library version 3—2007|
|NIDCAP (TI, AB, KW)|
|“Individualized developmental care” (TI, AB, KW)|
Inclusion criteria were: (1) the study design was a randomized controlled trial and (2) the intervention group was subject to NIDCAP observation on a regular basis followed by recommendations for the care. No exclusion was defined for type of outcome or for the length of the follow-up period.
The reviewed papers were assessed by using a form originating from SBU's approach to quality assessment in systematic reviews (Britton 2000). It included 12 criteria: power calculation, number of participants according to power calculation, comparable groups at start of the study, intention to treat, dropout rate, blinded assessment, assessment of intervention performance and extension, equal treatment of groups except for the intervention, contamination of intervention to the control group, relevant outcome measures (meaning clinical relevance, an outcome that is likely to affect the infant or the family in a persistent way), relevant number of outcome measures, and ethical approval. Assessment resulted in “yes” or “no” for each criterion. To be classified as having high methodological quality, a paper needed to fulfill 10 to 12 criteria, for medium quality 6 to 9 criteria, and for low quality ≤ 5 criteria. The two authors assessed each paper individually; in case of disagreement, the assessment was discussed until agreement was reached.
Intervention, Patients, and Settings
The NIDCAP intervention aims at providing adequate stimulation to premature infants, which by definition are infants with a gestational age of less than 37 weeks. These infants are cared for in specialized neonatal intensive care units. Information on gestational ages and setting specifications are presented in Table 2 when available, but did not affect inclusion in the review. In all studies the intervention was compared to standard neonatal intensive care (or conventional care). This has never been a steady state but merely been depending on an ongoing development of medical, technical, and nursing practices. Standard care, in this review, is defined as the absence of NIDCAP observations and care planning.
|STUDY DESIGN||STUDY GROUPS||PATIENTS INCLUDED AND WITHDRAWALS||OUTCOME VARIABLES THAT SHOWED SIGNIFICANTLY BETTER RESULTS FOR THE INTERVENTION GROUP THAN FOR THE CONTROL GROUP||COMMENTS|
|Als et al. (1994), Boston, MA, USA|
|RCT||Birth weight: <1,250 g||I: 20||Follow-up after 2 weeks||Medium quality of evidence|
|Follow-up: 2 weeks and 9 months corrected age Independent evaluation of results||Length of gestation: >24 <30 weeks Seriously ill infants, extremely premature||C: 18 Withdrawals: C 5% at 9 months||Lower age at discharge, shorter inpatient period, fewer days with extra O2, fewer days before bottlefeeding, lower score on the Pediatric Complication Scale, lower incidence of BPD, lower IVH incidence, lower costs for care, 3/6 APIB subscales, 4/18 APIB summary variables, 5 EEG measurements (activation of brain regions) Follow-up after 9 months Bayley Scale: MDI and PDI Kangaroo box paradigm: Play episode 17/20 variables, Still face episode 12/19 variables, Interaction 3 variables||C was provided aspects of NIDCAP concerning the environment and this reinforces the result. The influence of IVH and BPD was discussed by Als et al. for certain result parameters; the intervention was judged as having a greater effect. A solid study, but it cannot be assessed as high quality because a great number of outcome variables (>100) were analyzed.|
|Fleisher et al. (1995), article 1 based on the study at Stanford, CA, USA|
|RCT||Birth weight: ≤1,250 g||I: 17||Fewer days with ventilator/CPAP treatment, fewer infants remaining in hospital at 42 weeks of age after conception, 4/6 APIB subscales||Medium quality of evidence|
|Follow-up: 42 weeks after conception Only the APIB evaluation was carried out blind||Length of gestation: ≤30 weeks Seriously ill infants, extremely premature||C: 18||Issues on selection for inclusion because 140 eligible participants were excluded. The control group was provided aspects of NIDCAP concerning the environment, which reinforces the result. A consistent trend of better, but insignificant, results also on other variables for I (p = 0.06–0.09).|
|Ariagno et al. (1997), article 2 based on the study at Stanford, CA, USA|
|RCT||See Fleisher et al. (1995)||I: 14||4/6 APIB subscales (reported in Fleisher et al. (1995))||Medium quality of evidence|
|Follow-up: 36 weeks of age after conception and at age 2 years||C: 14 Withdrawals: 20–40%||Adds nothing new (except for nonsignificant results concerning sleep) in that the APIB result was already reported in Fleisher et al. (1995). Relatively large number of withdrawals.|
|Heller et al. (1997), article 3 based on the study at Stanford, CA, USA|
|RCT||See Fleisher et al. (1995)||I: 17 C: 18||Lower cumulative dose of chloral hydrate during the hospital stay in the group of most seriously ill infants (27/35 infants).||Medium quality of evidence|
|Patient record notes at discharge from hospital||The result is found in a subgroup analysis, but this analysis includes most of the sample.|
|Buehler et al. (1995), Boston, MA, USA|
|RCT||Premature infants||I: 12||2/6 APIB subscales, 3/11 Prechtl scales (no difference after application of the Bonferroni correction), 70/180 EEG factors (with ANCOVA 41/180)||Medium quality of evidence|
|Two groups of premature infants versus one group of full-term (FT) infants (not randomized) Follow-up: 2 weeks corrected age Independent evaluation of results||Birth weight: <2,500 g Length of gestation: 30–34 weeks Relatively well and moderately premature infants Full-term infants (FT) Length of gestation: 38–40 weeks Adequate weight and length||C: 12 FT: 12||Similar results for I and FT on 32 of 180 EEG factors. No difference between I and FT on APIB||No difference between the two premature groups when using the Bonferroni correction. It is a confusing description of significant findings and difficult to interpret the EEG results. In addition, a large number of variables were tested (>200, of which 180 were EEG variables). C, was provided aspects of NIDCAP concerning the environment, and this reinforces the result. I demonstrated a higher risk because of the higher score on the Obstetric Complication Scale and a higher requirement for oxygen on days 1–2.|
|Westrup et al. (2000), article 1 based on the study at Karolinska Hospital, Sweden|
|RCT Follow-up: 35 (growth) and 36 weeks (BPD) of age after conception||<32 weeks age after conception at birth Seriously ill infants, extremely premature||I: 12 C: 13 Withdrawals: I 43%, C 35%||Fewer CPAP days, free of extra oxygen at lower age, lower BPD incidence||Medium quality of evidence The infants in I were somewhat larger and older at birth, but the differences were not significant. A high number of withdrawals that are explained by the application of exclusion criteria and by the parents' decision to take part taken after randomization.|
|Westrup et al. (2002), article 2 based on the study at Karolinska Hospital, Sweden|
|RCT Follow-up: 32 and 36 weeks of age after conception Independent evaluation of results||See Westrup et al. (2000)||I: 11 C: 11 Withdrawals: I 48%, C 45%||No significant differences in the sleep variables||Medium quality of evidence No increase in quiet sleep for I. See also the comments concerning Westrup et al. (2000). Higher number of withdrawals compared to Westrup et al. (2000) because of infants that died later were included in the first study.|
|Kleberg et al. (2002), article 3 based on the study at Karolinska Hospital, Sweden|
|RCT Follow-up: 1 year corrected age Independent evaluation of results||See Westrup et al. (2000)||I: 11 C: 9 Withdrawals: I 48%, C 55%||Bayley Scale higher MDI, odds ratio 18.6 I vs. C for having MDI >80||Medium quality of evidence See also the comments concerning Westrup et al. (2000). Higher number of withdrawals compared to Westrup et al. (2000) because of infants that died later included in the first study.|
|Westrup et al. (2004), article 4 based on the study at Karolinska Hospital, Sweden|
|RCT Follow-up: age 5.5 years||See Westrup et al. (2000)||I: 11 C: 15 Withdrawals: I 48%, C 25%||Odds ratio 19.9 I versus C for surviving, and not experiencing any major behavioral disorder||Medium quality of evidence Follow-up carried out for more infants than in Westrup et al. (2000, 2002) and Kleberg et al. (2002, 2007). See also the comments concerning Westrup et al. (2000).|
|Kleberg et al. (2007), article 5 based on the study at Karolinska Hospital, Sweden|
|RCT Follow-up: 36 weeks of age after conception (all infants still in the neonatal unit)||See Westrup et al. (2000)||I: 10 C: 10 Withdrawals: I 52%, C 50%||Mothers perceived feeling closer to their infant and higher levels of anxiety.||Medium quality of evidence The investigators of the original study interpret the higher levels of anxiety as an indication that the mothers in the NIDCAP group had already bonded to their infants. Such a “positive” conclusion from this finding is questionable. See also the comments concerning Westrup et al. (2000). Higher number of withdrawals compared to Westrup et al. (2000) because of infants that died later included in the first study.|
|Als et al. (2003), Boston, MA and Oakland, CA, USA|
|RCT (3 centers) Follow-up: 2 weeks corrected age Independent evaluation of results||Birth weight: <1,250 g Length of gestation: <28 weeks Seriously ill infants, extremely premature||I: 45 C: 47 Withdrawals: 10–20% for PSI and MVC||Fewer days with parenteral nutrition, fewer days with combined parenteral and feeding by mouth, fewer days with ventilator treatment, lower NEC incidence. Better gain in weight. Greater weight, greater length, and larger head circumference. Younger age at discharge, shorter inpatient period, lower costs for care, 6/6 APIB scales, 4/4 PSI scales, MVC (total score)||Medium quality of evidence A significant improvement in I in a large number of variables (all developmental variables). C was provided aspects of NIDCAP concerning the environment, and this reinforces the result. There is an issue on patient inclusion as 142 of 234 eligible infants were not included.|
|Als et al. (2004) Boston, MA, USA|
|RCT Follow-up: 2 weeks and 9 months corrected age Independent evaluation of results||Length of gestation: 28 + 4 − 33 + 3 Low-risk premature infants||I: 16 C: 14 Withdrawals: 20% at 9 months||2/6 APIB scales, 6/12 Prechtl scales, 3/3 Bayley II scales, 4/40 EEG factors, 1/8 MRI variables||Medium quality of evidence C exhibited significantly higher Apgar scores at 5 minutes (8.5 vs. 7.88), which may have contributed to reducing differences between C and I. C was provided aspects of NIDCAP concerning the environment, and this reinforces the result. However, 60 of 90 eligible participants were not included (50% of the parents declined participation). Interpretation of EEG and MRI variables is difficult. More than 100 outcome variables were analyzed (48 of these being EEG and MRI variables).|
Data Extraction, Analysis, and Outcome Variables
This review presents outcome variables that showed a statistically significant difference between the NIDCAP and the control group, allocated into one of four groups: (1) psychomotor development/neurological status (assessment of preterm infant behavior, Bayley and Bayley II, Kangaroo box, Prechtl, EEG, mortality, and morbidity), (2) medical/nursing care outcome during the inpatient period (days on oxygen, continuous positive airway pressure [CPAP] and mechanical ventilation, duration of nutritional assistance, growth, complications), (3) mothers' perceptions of care, and (4) economy and inpatient duration (age at discharge; see Table 3). Data extraction on these variables was performed independently by the two authors for each study.
|Psychomotor development and neurological status|
|Als et al. (1994)||Better scores on three APIB subscales (autonomic system and motor system, self-regulation) and for four APIB summary variables at age 2 weeks. Differences in activation of five brain regions as shown by EEG with topographical mapping.|
|Better Bayley MDI and PDI at age 9 months. Better values in the Kangaroo box paradigm on three subscales at age 9 months.|
|Buehler et al. (1995)||Better scores on two APIB subscales (autonomic system and motor system) and better scores on three Prechtl scales (truncal and limb posture, syndromes of abnormal reactivity, and total score) at corrected age 2 weeks (no difference in Prechtl scores when the Bonferroni correction is applied). Differences in 70 of 180 EEG factors.|
|Fleisher et al. (1995)a||Better on four APIB subscales (motor system, state regulation, interactive capabilities, and ability to self-regulate) at age 42 weeks after conception.|
|Kleberg et al. (2002)||Better Bayley MDI index at 1-year corrected age. Odds ratio 18.6 I versus C for having an MDI >80.|
|Als et al. (2003)||Better on all six APIB subscales, all four PSI subscales, MVC total score at age 2 weeks.|
|Als et al. (2004)||Better on two APIB subscales (motor system modulation and self-regulation), better scores on six Prechtl scales (trunk and limb posture, pathological movements, intensity of responses, motor response, state stability, and total Prechtl score) at 2 weeks corrected age.|
|Better Bayley II scores on all three subscales at age 9 months. Differences in four EEG factors and one MRI variable.|
|Westrup et al. (2004)||Odds ratio 19.9 I versus C for survival and not experiencing a major behavioral disorder.|
|Medical/nursing care outcomes during the inpatient period|
|Als et al. (1994)||Fewer days with extra oxygen|
|Fleisher et al. (1995)||Fewer days with ventilator/CPAP treatment|
|Westrup et al. (2000)||Fewer days with CPAP. Extra oxygen not required at lower age|
|Als et al. (2003)||Fewer days with ventilator treatment|
|Als et al. (1994)||Fewer days before bottle feeding|
|Als et al. (2003)||Fewer days with parenteral nutrition. Fewer days with combined parenteral and feeding by mouth. Better increase in weight.|
|Higher weight, greater length, and larger head circumference at 2 weeks corrected age.|
|Als et al. (1994)||Lower score on Pediatric Complication Scale. Lower BPD incidence and IVH incidence.|
|Heller et al. (1997)||Lower cumulative dose of chloral hydrate during the hospital stay in the most seriously ill group.|
|Westrup et al. (2000)||Lower BPD incidence|
|Als et al. (2003)||Lower NEC incidence|
|Mothers' perceptions of their role, their infant, and neonatal care|
|Kleberg et al. (2007)||Mothers perceived being closer to their infants and experienced higher levels of anxiety.|
|Economic issues and inpatient period|
|Als et al. (1994)||Lower age at discharge. Shorter inpatient period. Lower costs for care.|
|Fleisher et al. (1995)||Fewer infants remained in hospital at age 42 weeks after conception.|
|Als et al. (2003)||Younger at time of discharge, shorter inpatient period, lower costs for care.|
The 12 papers that were identified that followed the inclusion criteria are presented in Table 2. Studies with various types of historical control group design were also found during the literature search and 11 such articles were excluded (Table 4). Because implementation of NIDCAP occurred at the same time as did many other changes in medical and nursing practice, the historical design results in studies having a low grade of evidence and, thus, not included in the current review.
|AUTHOR (YEAR), LOCATION||STUDY DESIGN, STUDY GROUPS|
|Als et al. (1986)||Historical control group design with prospective data collection in both I and C.|
|Boston, MA, USA||Birth weight <1,250 g|
|Becker et al. (1991, 1993)||Historical control group design with prospective data collection in both I and C.|
|Location not specified||Birth weight <1,500 g|
|Mouradian et al. (1994)||Historical control group design with retrospective data collection in both I and C.|
|Boston, MA, USA||Length of gestation <34 weeks|
|Stevens et al. (1996) Toronto, Canada||Historical control group design with prospective data collection in both I and C. Birth weight 750–1,500 g|
|Petryshen et al. (1997) Toronto, Canada||Historical control group design with retrospective data collection in both I and C. Economic analysis on the basis of the same sample as Stevens et al. (1996). Birth weight 750–1,500 g|
|Brown et al. (1997). Location not specified||Historical control group design with retrospective data collection in both I and C. A matched historical control group. Birth weight <1,500 g|
|Westrup et al. (1997) Falun, Sweden||Historical control group design with retrospective data collection in C and prospective data collection in I. Birth weight <1,500 g|
|Kleberg et al. (2000) Falun, Sweden||Historical control group design with prospective data collection in both I and C. Follow-up of the infants from Westrup et al. (1997). Birth weight <1,500 g|
|Prentice et al. (2003)||Historical control group design with retrospective data collection in both I and C.|
|Sydney, Australia||Birth weight <1,500 g|
|Wielenga et al. (2007)||Prospective phase-lag cohort design.|
|Amsterdam, The Netherlands||Length of gestation <30 weeks|
The following description of the effects of NIDCAP is based on the examination of the 12 articles that come from six randomized controlled studies. The studies were performed in Boston, MA (Als et al. 1994), Stanford, CA (Fleisher et al. 1995; Ariagno et al. 1997; Heller et al. 1997), Boston, MA (Buehler et al. 1995), Stockholm, Sweden (Westrup et al. 2000, 2002, 2004; Kleberg et al. 2002, 2007), Boston, MA and Oakland CA (Als et al. 2003), and Boston, MA (Als et al. 2004; Table 2). All articles were published after 1993. The studies are mainly carried out with patients being randomized within the care unit. Infants in the intervention group were initially observed and then given a care plan based on this observation according to the NIDCAP concept, whereas infants in the control group were not exposed to this. Approximately 250 infants were included in these six studies; either infants with very low birth weight (VLBW), often less than 1,250 grams, corresponding to a gestational age less than 30 weeks (Als et al. 1994, 2003; Fleisher et al. 1995; Ariagno et al. 1997; Heller et al. 1997; Westrup et al. 2000, 2002, 2004; Kleberg et al. 2002, 2007), or infants with low birth weight (LBW), which is less than 2,500 grams, length of gestation 30–34 weeks (Buehler et al. 1995; Als et al. 2004). Follow-up periods varied between the studies, depending on outcome variables. The longest follow-up period was 5.5 years.
All included articles, that is, the six randomized controlled trials were assessed to be of medium methodological quality, fulfilling 6 to 9 of the 12 quality criteria. A common feature of most studies was the low number of participants (mean 32, range 20–92) and thus low statistical power. A high dropout rate (withdrawals are detailed in Table 2; varied between 20% and 55% in six of the studies) and a high risk for contamination of intervention were common weaknesses. In some studies, a very large number of outcome variables were analyzed using bivariate analyses, implying a risk for mass significance. Comments on the assessment of quality are provided in Table 2; the comments' column.
The studies in this review evaluated many outcome variables. To enable comparison we present all significant findings reported in the reviewed studies, grouped into the following categories: psychomotor development and neurological status, medical and nursing care outcome during the inpatient period, mothers' perceptions of care, and economy and inpatient duration (Table 3).
Psychomotor Development and Neurological Status
Based on different outcome measures, positive effects of NIDCAP are reported in seven articles from the six randomized studies (Als et al. 1994, 2003, 2004; Buehler et al. 1995; Fleisher et al. 1995; Kleberg et al. 2002; Westrup et al. 2004). Five of the included studies used the assessment of preterm infants' behavior (APIB) index as the outcome measure (at a corrected age of 2 weeks, i.e., 2 weeks after the estimated date of full term). All five of these studies reported improved psychomotor development for the NIDCAP group within some or all of the six subscales of APIB (Als et al. 1994, 2003, 2004; Buehler et al. 1995; Fleisher et al. 1995). The NIDCAP group had consistently better results on the motor system subscale and the autonomic system scale in three of the five studies (Als et al. 1994, 2003; Buehler et al. 1995). Four studies used the Bayley scale (mental and psychomotor development) as the follow-up method (at ages from 9 months to 2 years): three of these studies reported better results for the NIDCAP group on the Bayley Mental Developmental Index (MDI) scale (Als et al. 1994, 2004; Kleberg et al. 2002) and two of these studies reported better results for the NIDCAP group on the Bayley Psychomotor Development Index (PDI) scale (Als et al. 1994, 2004). The Prechtl instrument, a neurological assessment conducted at a corrected age of 2 weeks, was used in two studies (Buehler et al. 1995; Als et al. 2004), one of which reported better results in six of the 12 subscales (Als et al. 2004). Buehler and colleagues (1995), who compared a NIDCAP group with a group of full-term infants, found no difference between these two groups on most of the tests. Two studies, using other tests than APIB, Prechtl, or Bailey, reported better results for the NIDCAP group; Westrup and coworkers (2004) reported a higher odds ratio for normal behavior at age 5.5 years for the NIDCAP group and Als et al. (1994) reported better results on the Kangaroo box test. The two studies that have assessed the effects of NIDCAP on infant sleep behavior did not show any differences between the intervention group and the control group (Ariagno et al. 1997; Westrup et al. 2002).
Medical and Nursing Care Outcomes During the Inpatient Period
The summary of medical and nursing care outcomes can be grouped into three areas: respiration, feeding/growth, and complications. Treatment variables that can be related to the infant's need of respiratory support have been assessed in all six studies (Als et al. 1994, 2003, 2004; Buehler et al. 1995; Fleisher et al. 1995; Westrup et al. 2000). Respiratory variables have received this high level of attention because both the duration of such support and the need for auxiliary oxygen increase the risk of lung complications in premature infants, especially the risk for bronchopulmonary dysplasia (BPD). A longer period of respiratory support also means that the contact between infant and parent is weakened because of obstacles (both physical and psychological) introduced by the technical equipment. In summary, positive effects (shorter periods of treatment) of the NIDCAP intervention were reported in four of the six randomized studies. Three studies reported fewer days of ventilator or CPAP treatment for the NIDCAP group (Fleisher et al. 1995; Westrup et al. 2000; Als et al. 2003), two studies reported shorter treatment times for extra oxygen (Als et al. 1994; Westrup et al. 2000), and one study reported that the most care-dependent infants in the NIDCAP group required a lower total dose of chloral hydrate, which is a sedative that was given during the ventilator treatment (Heller et al. 1997).
Measures of outcome for feeding/growth have been evaluated in all six randomized controlled studies (Als et al. 1994, 2003, 2004; Buehler et al. 1995; Fleisher et al. 1995; Westrup et al. 2000). Two studies reported better results for the NIDCAP group; it had shorter duration of care until the infant was fully fed by mouth (Als et al. 1994, 2003) and shorter duration of care with complete parenteral feeding (Als et al. 2003). The latter study also showed that the infants in the NIDCAP group had a better development of weight to a corrected age of 2 weeks and that they were longer and had greater head circumference at this age (Als et al. 2003). Medical complications have also been evaluated using various outcomes in all six studies (Als et al. 1994, 2003, 2004; Buehler et al. 1995; Fleisher et al. 1995; Heller et al. 1997; Westrup et al. 2000): three of these studies reported significant results. Als et al. (1994) reported that the NIDCAP group had a lower score on the pediatric complication scale and a lower frequency of BPD and intraventricular hemorrhage, together with a lower frequency of necrotizing enterocolitis (Als et al. 2003). Also Westrup et al. (2000) reported fewer cases of BPD in the NIDCAP group.
One study (Kleberg et al. 2007) reported on mothers' perceptions of NIDCAP. This is one of the papers originating from the Westrup et al. (2000) study. A questionnaire was used to evaluate various aspects of the mothers' apprehension of their maternal role, perception of their infant, and neonatal care. Although the mothers in the NIDCAP group perceived more closeness to their infant, they reported more anxiety than the mothers in the control group.
Economic Issues and Inpatient Period
Data on costs and resources consumed were collected within the framework of all randomized studies evaluating NIDCAP. For example, the length of the inpatient period (or the age of the infant when discharged from the hospital) has been compared between intervention groups and control groups (Als et al. 1994, 2003, 2004; Buehler et al. 1995; Fleisher et al. 1995; Westrup et al. 2000). In addition, three studies collected care costs for the intervention and control group (Buehler et al. 1995; Westrup et al. 2000; Als et al. 2003). Results from two studies indicated that infants treated with NIDCAP could be discharged at a younger age, had a shorter hospital stay, and had lower care costs (Als et al. 1994, 2003). A further study showed that a lower number of the infants in the NIDCAP group were still receiving hospital care at 42 weeks after conception (Fleisher et al. 1995).
Complications and Undesired Effects
In general, complications or undesired effects of the NIDCAP intervention were not reported in any of the reviewed studies. However, Kleberg and coworkers' (2007) finding on more anxiety among mothers in the NIDCAP group might be classified as an undesired effect.
In summary, a number of significant positive results on the infant's cognitive and psychomotor development have been reported as an effect of NIDCAP. The improvements are mostly in relation to higher scoring in the subscales of the APIB, Prechtl, and Bailey scales; however, higher odds ratio for normal behavior at 5.5 years of age and better results on the Kangaroo box test were also noted. Moreover, a reduced need for respiratory support was reported. However, there are some serious weaknesses in design and methods in the studies constituting the evidence base on NIDCAP. In the following section we discuss findings and problems in the studies reviewed.
Consistency in Results
In the six studies reviewed, there is an almost uniform trend showing significantly better results in the NIDCAP group. It must however be noted that we have not performed a meta-analysis to verify such a trend statistically. Meta-analysis was not part of the adopted review approach and in our judgment there were only few variables where meta-analyses would have been appropriate. From the findings it appears that NIDCAP primarily provides positive results in the cognitive and psychomotor development of the infant. Similar results were reported in all six randomized studies. Most of these studies used the Prechtl or the APIB neurobehavioral assessment during the neonatal period. Both of these methods are judged to be valid with high interrater reliability. Hüppi et al. (1996) demonstrated that structural and functional brain development is well correlated with APIB and the Maas group (2000) found that the Prechtl method is the best predictor of normal neurological development at the corrected age of 2 years. It should also be noted that there is an immediate benefit with positive outcomes during the early infant period, independently of subsequent (longer term) outcomes. Higher scores in the assessments indicate that infants have a greater capacity to interact with their parents, which enhances opportunities for bonding. Similarly, the finding that the need for respiratory support is lower with NIDCAP is a promising finding in the sense that this may reduce the risk of lung complications while at the same time providing space for increased contact between parents and infants. The only study that investigated the perceptions of mothers of NIDCAP reported that mothers to NIDCAP-treated infants scored higher on feelings of being close to the infant than mothers in the control group. However, these same mothers reported a greater degree of anxiety as compared with the mothers in the control group. Kleberg and coworkers were puzzled by this finding and suggested it could be interpreted as a sign of earlier bonding, that is, the authors interpreted the anxiety the mothers were experiencing as a positive force. We find this interpretation questionable, however. Rather, we see it as an adverse effect of NIDCAP if it results in mothers feeling more fearful of the technical environment and that the infant is at risk of not surviving.
Design and Methods in the Examined Studies
Current information on the effects of NIDCAP is grounded on a relatively small number of studies, often with few participants (about 250 infants in the six studies). In some studies it is remarkable that many eligible infants were not included in the trials (Fleisher et al. 1995; Als et al. 2003, 2004). There was also a relatively large withdrawal rate in some of the studies (Ariagno et al. 1997; Westrup et al. 2000, 2002, 2004; Kleberg et al. 2002, 2007). The withdrawal rate in the Swedish studies (Westrup et al. 2000, 2002, 2004; Kleberg et al. 2002, 2007) appears to result from the exclusion criteria being applied after randomization, whereas these criteria were applied before randomization in the other studies. Another problem with the sample in these studies is that the weight levels used differ from standard levels, which are <1,000 grams = extremely low birth weight (ELBW), <1,500 grams = VLBW, and <2,500 grams = LBW. Thus, the levels used are different from those in many other neonatal outcome studies making comparisons difficult.
The small number of infants included in the studies introduces a risk that, despite randomization, the baseline variables differ between the intervention and control group. On the other hand, small studies with low statistical power imply that large differences between the groups are required in order to identify significant differences. It is possible that additional differences between the intervention group and the control group would have been observed in larger studies. A very large number of statistical tests have been carried out in some studies, in particular, those by Als et al. (1994, 2004) and Buehler et al. (1995). This, combined with a lack of primary outcome variables, that is, the studies did not have one specific primary hypothesis to test, introduces the problem of mass significance. To a certain extent this problem has been managed by statistical corrections, but designing a study to focus on a few outcome variables would highly increase the internal validity in evaluating effects. The approach of examining many outcome variables might be understandable in an exploratory phase of evaluating an intervention model, but it is not desirable to have this repeated in many studies.
It is clear that the studies included in the current review are marked by methodological shortcomings. The scientific basis for evaluating the effect of NIDCAP would be significantly improved by an appropriately dimensioned study, that is, a study where the number of participants builds on a proper power analysis of the primary outcome measure, with a long follow-up. So far, the longest follow-up period reported is 5.5 years (Westrup et al. 2004). Such a study should have a clear focus on one or a small number of important outcome variables. Previous systematic reviews of NIDCAP research have also emphasized the need of such a research strategy (Jacobs et al. 2002; Symington & Pinelli 2006). A Cochrane review (Symington & Pinelli 2006) that evaluated a number of different interventions intended to support development claimed that the NIDCAP studies that were included (Als et al. 1986, 1994, 2003, 2004; Buehler et al. 1995; Fleisher et al. 1995; Ariagno et al. 1997; Westrup et al. 2000, 2004; Kleberg et al. 2002) showed limited effects of NIDCAP in premature infants on chronic lung disease, necrotizing enterocolitis, and family stress. Furthermore, Symington and Pinelli (2006) found very limited evidence of a NIDCAP effect on behavior and motor skills at the age of 5 years. A meta-analysis carried out by Jacobs et al. (2002), which included five articles based on four randomized controlled studies (Als et al. 1994; Buehler et al. 1995; Fleisher et al. 1995; Ariagno et al. 1997; Westrup et al. 2000) and three studies with historical design of the control group (Als et al. 1986; Becker et al. 1993; Stevens et al. 1996), showed that NIDCAP reduced the requirement for supplemental oxygen and improved neurodevelopmental outcome at an age of 9 to 12 months. Both these reviews revealed methodological shortcomings in the studies reviewed. The authors drew conservative conclusions, which differed slightly from one another. Jacobs et al. (2002) concluded that there is insufficient evidence that NIDCAP improves medical and neurological development of premature infants. The Cochrane report felt that NIDCAP gives limited effects on some medical outcomes, the family situation, and on behavior and motor skills at the age of 5 years (Symington & Pinelli 2006). It is also worth noting that the present report includes two studies that were not included in Jacobs et al. (2002) review (Als et al. 2003, 2004). One of these studies, comprising 92 infants, is the largest carried out so far (Als et al. 2003). Both these studies are reviewed in the latest update of the Cochrane report (Symington & Pinelli 2006).
We could not identify any study in which the cost effectiveness of the method was assessed, that is, a study that calculates the total costs and weighs these against the effects achieved. The results on shorter respiratory support time and shorter hospital stay indicate cost reductions but these are not compared with costs for implementing and sustaining the NIDCAP method. NIDCAP involves an initial investment for training and reorganization, as well as subsequent costs for the time involved in the continuous observations of infant behavior. The course fee to become a certified NIDCAP observer is approximately U.S. $6,000 (Westrup 2007), in addition to costs for working time and travel. It has been estimated that the cost of time involved in observation of behavior is approximately U.S. $700 for 10 observations (Westrup 2007). Furthermore, NIDCAP is a registered trademark for which the Children's Medical Center Corporation in Boston owns all rights and the method is introduced as an “all or nothing” concept, implicating that it is difficult to evaluate whether certain elements of the NIDCAP intervention are more significant than others. For instance, no research is available into how many certified NIDCAP observers are required on a neonatal unit nor is there research suggesting 7 to 10 days is the optimal interval between observations of behavior.
Additional two randomized NIDCAP studies have been carried out but not yet reported in peer-reviewed journals. The Edmonton study from Canada, which includes 110 infants born in the period 1999–2002, investigated the effects of NIDCAP on several factors, including duration of inpatient care, duration of incubator care, morbidity, and psychomotor development. Some preliminary results have been presented that indicate a reduction in the following variables: length of stay in hospital, mechanical ventilation days, incidence of chronic lung disease, and infectious complications (Tyebkhan et al. 2004; Hendson et al. 2005). The Leiden study, which included approximately 340 infants born in the period 2000–2004, evaluated the requirement for mechanical ventilation, duration of intensive care, growth, and morbidity. Preliminary findings show a positive effect on the head circumference growth in a study subgroup (Maguire 2003). Both of these studies are relatively large and may contribute with valid information on the effects of NIDCAP. Discussions are currently under way concerning starting a European multicenter NIDCAP study or introducing benchmarking and qualitative research to study the effects of NIDCAP (Sizun & Westrup 2004; Pierrat et al. 2007).
NIDCAP implementation has latterly been linked with other nursing models and with further questions that illuminate the development of the infant. NIDCAP has, for example, been used as a method for estimating pain (Holsti et al. 2004, 2005), as a method for the quantitative evaluation of the behavior of premature infants (Pressler et al. 2001; Pressler & Hepworth, 2002), and as a model for music therapy (Abromeit 2003). A French research group has studied how NIDCAP techniques (subdued illumination and sound levels, positioning, and the opportunity to grip and suck) applied for short periods of time (minutes to hours) influence signals of pain when weighing premature infants (Catelin et al. 2005) and the duration of sleep (Bertelle et al. 2005, 2007). One single NIDCAP observation with subsequent recommendations for the care resulted in faster recovery after neonatal eye screening examinations (Kleberg et al. 2008). The process of implementing the NIDCAP method has been studied, as well as its effect on the experiences of parents (Wielenga et al. 2006) and health care personnel (Westrup et al. 1997; Sizun & Westrup 2004; Cuttini 2005; van der Pal et al. 2007). Whether NIDCAP observations can be used in an intervention program targeted at mothers has also been studied (Parker et al. 1992).
Despite the relatively consistent trend of significant positive findings reported as an effect of the NIDCAP, particularly the findings on cognitive and motor development in preterm infants, it must be concluded that the scientific evidence concerning the positive effects of NIDCAP is limited. The shortcomings in design and methodological accomplishment of the studies reviewed are too apparent to support far-reaching claims on the effectiveness of the method. Scientific grounds for assessing the effects of NIDCAP would be considerably enhanced by a sufficiently comprehensive study with extended follow-up and a clear focus on one or two important outcome variables, preferably, psychomotor development and neurological status.
- • NIDCAP is safe with no complications or undesired effects on the infant
- • Evidence for NIDCAP's positive effects is limited and potential positive effects are hard to weigh against the costs in implementing and sustaining the method
- • A sufficiently dimensioned study focusing on a few outcome measures, with extended follow-up would enhance the scientific grounds for assessing the effects of NIDCAP.
- 2003). The newborn individualized developmental care and assessment program (NIDCAP) as a model for clinical music therapy interventions with premature infants. Music Therapy Perspectives, 21, 60–68. (
- 1982). Toward a syntactive theory of development: Promise for the assessment and support of infant individuality. Infant Mental Health Journal, 3, 229–243. (
- 2007). NIDCAP program guide. Boston : Children's Hospital. (
- 2005). The assessment of preterm infants' behavior (APIB): Furthering the understanding and measurement of neurodevelopmental competence in preterm and full-term infants. Mental Retardation & Developmental Disabilities Research Reviews, 11, 94–102. , , & (
- 2004). Early experience alters brain function and structure. Pediatrics, 113, 846–857. , , , , , , , , , , & (
- 2003). A three-center, randomized, controlled trial of individualized developmental care for very low birth weight preterm infants: Medical, neurodevelopmental, parenting, and caregiving effects. Journal of Developmental and Behavioral Pediatrics, 24, 399–408. , , , , , , , , , , , & (
- 1986). Individualized behavioral and environmental care for the very low birth weight preterm infant at high risk for bronchopulmonary dysplasia: Neonatal intensive care unit and developmental outcome. Pediatrics, 78, 1123–1132. , , , , , & (
- 1994). Individualized developmental care and the very low-birth-weight preterm infant—medical and neurofunctional effects. JAMA, 272, 853–858. , , , , & (
- 1997). Developmental care does not alter sleep and development of premature infants [in process citation]. Pediatrics, 100, E9. , , , , , & (
- 2005). Evidence for best practices in the neonatal period. MCN, American Journal of Maternal Child Nursing, 30, 397–403. (
- 1991). Outcomes of developmentally supportive nursing care for very low birth weight infants. Nursing Research, 40, 150–155. , , & (
- 1993). Effects of developmental care on behavioral organization in very-low-birth-weight infants. Nursing Research, 42, 214–220. , , & (
- 2005). Sleep of preterm neonates under developmental care or regular environmental conditions. Early Human Development, 81, 595–600. , , & (
- 2007). Sleep in the neonatal intensive care unit. Journal of Perinatal and Neonatal Nursing, 21, 140–150. , , , & (
- 2000). Evidensbaserad medicin. Så graderas en studies vetenskapliga bevisvärde och slutsatsernas styrka. [Evidence-based medicine. Grading the scientific values and strength of conclusions in clinical trials]. Lakartidningen, 97, 4414–4415. (
- 1997). The effect of developmental care on preterm infant outcome. Applied Nursing Research, 10, 190–197. . & (
- 1995). Effectiveness of individualized developmental care for low-risk preterm infants: Behavioral and electrophysiologic evidence. Pediatrics, 96, 923–932. , , , & (
- 2005). Clinical, physiologic, and biologic impact of environmental and behavioral interventions in neonates during a routine nursing procedure. Journal of Pain, 6, 791–797. , , , & (
- 2005). A European developmental care survey. Poster presentation at IDNIC- International Conference on Infant Development in Neonatal Intensive Care 2005. London : IDNIC 2005. (
- 2004). Parent stress in the neonatal intensive care unit and the influence of parent and infant characteristics. American Journal of Occupational Therapy, 58, 509–520. (
- 1995). Individualized developmental care for very-low-birth-weight premature infants. Clinical Pediatrics, 34, 523–529. , , , , , , & (
- 2005). Measuring neonatal intensive care unit-related parental stress. Journal of Advanced Nursing, 49(6), 608–615. , , & (
- 2006). Family-centered care in the NICU. Journal of Perinatal and Neonatal Nursing, 20, 98–102. (
- 2000). Outcomes of children of extremely low birthweight and gestational age in the 1990s. Seminars in Neonatology, 5, 89–106. & (
- 1993). The principles for family-centered neonatal care. Pediatrics, 92, 643–650. (
- 1997). Sedation administered to very low birth weight premature infants. Journal of Perinatology, 17, 107–112. , , , & (
- 2005). NIDCAP-based care reduces late onset sepsis in very low birth weight VLBW infants. Poster presentation at Pediatric Academic Societies' Annual Meeting , Washington , DC . , , , & (
- 2004). Specific Newborn Individualized Developmental Care and Assessment Program movements are associated with acute pain in preterm infants in the neonatal intensive care unit. Pediatrics, 114, 65–72. , , & (
- 2005). Prior pain induces heightened motor responses during clustered care in preterm infants in the NICU. Early Human Development, 81, 293–302. , , , , , , & (
- 1996). Structural and neurobehavioral delay in postnatal brain development of preterm infants. Pediatric Research, 39, 895–901. , , , , , & (
- 2002). The Newborn Individualized Developmental Care and Assessment Program is not supported by meta-analyses of the data. Journal of Pediatrics, 140, 699–706. , & (
- 2007). Mothers' perception of newborn individualized developmental care and assessment program (NIDCAP) as compared to conventional care. Early Human Development, 83, 403–411. , & (
- 2008). Lower stress responses after Newborn Individualized Developmental Care and Assessment Program care during eye screening examinations for retinopathy of prematurity: A randomized study. Pediatrics, 121, e1267–e1278. , , , , , , , , & (
- 2000). Developmental outcome, child behaviour and mother-child interaction at 3 years of age following Newborn Individualized Developmental Care and Intervention Program (NIDCAP) intervention. Early Human Development, 60 123–135. , & (
- 2002). Indications of improved cognitive development at one year of age among infants born very prematurely who received care based on the newborn individualized developmental care and assessment program (NIDCAP). Early Human Development, 68, 83–91. , , & (
- 1999). Humane neonatal care initiative. Acta Paediatrica, 88, 353–355. (
- 2000). Predictive value of neonatal neurological tests for developmental outcome of preterm infants. Journal of Pediatrics, 137, 100–106. , , , , & (
- 2003). The Leiden developmental care study: The effect of developmental care on growth of preterm infants < 32 weeks gestational age. Pediatric Research, 53, 397A. (
- 1994). The influence of neonatal intensive care unit caregiving practices on motor functioning of preterm infants. American Journal of Occupational Therapy, 48 527–533. & (
- 1992). Outcome after developmental intervention in the neonatal intensive care unit for mothers of preterm infants with low socioeconomic status. Journal of Pediatrics, 120, 780–785. , , & (
- 1997). Comparing nursing costs for preterm infants receiving conventional vs. developmental care. Nursing Economics, 15 138–145. , , & (
- 2007). How can we evaluate developmental care practices prior to their implementation in a neonatal intensive care unit Early Human Development, 83, 415–418. , , & (
- 2003). Outcomes of developmental care in an Australian neonatal intensive care nursery. Neonatal Network, 22 17–23. & (
- 2002). A quantitative use of the NIDCAP® tool: The effect of gender and race on very preterm neonates' behavior. Clinical Nursing Research, 11, 89–102. & (
- 2001). Behaviors of very preterm neonates as documented using NIDCAP observations. Neonatal Network, 20, 15–24. , , & (
- 1999). Human neonatal care initiative, NIDCAP and family-centred neonatal intensive care. Acta Paediatrica, 88, 1172. , & (
- 2004). Early developmental care for preterm neonates: A call for more research. Archives of Disease in Childhood Fetal and Neonatal Edition, 89, F384–F389. & (
- 1996). Developmental versus conventional care: A comparison of clinical outcomes for very low birth weight infants. Canadian Journal of Nursing Research, 28, 97–113. , , , & (
- 2006). Developmental care for promoting development and preventing morbidity in preterm infants. Cochrane Database of Systematic Reviews, Issue 2. Art. No.: CD001814. DOI: DOI: 10.1002/14651858.CD001814.pub2. & (
- 2004). Epidemiology of preterm birth. British Medical Journal, 329, 675–678. & (
- 2004). The impact of developmental care in the NICU; The Edmonton randomized controlled trial of NIDCAP®. Pediatric Research, 55, 505A. , , , & (
- 2007). Staff opinions regarding the newborn individualized developmental care and assessment program (NIDCAP). Early Human Development, 83, 425–432. , , , , , & (
- 2006). Newborn individualized developmental care and assessment program – NIDCAP. SBU alert reports. Stockholm : The Swedish Council on Technology Assessment in Health Care . & (
- 2007). Cost-benefit analyses of NIDCAP training. Scandinavian NIDCAP® Center , Stockholm , Sweden . (
- 2004). Preschool outcome in children born very prematurely and cared for according to the newborn individualized developmental care and assessment program (NIDCAP). Acta Paediatrica, 93, 498–507. , , & (
- 2002). No indications of increased quiet sleep in infants receiving care based on the newborn individualized developmental care and assessment program (NIDCAP). Acta Paediatrica, 91, 318–322. , , & (
- 1997). Evaluation of the newborn individualized developmental care and assessment program (NIDCAP) in a Swedish setting. Prenatal and Neonatal Medicine, 2, 366–375. , , , , & (
- 2000). A randomized, controlled trial to evaluate the effects of the Newborn Individualized Developmental Care and Assessment Program in a Swedish setting. Pediatrics, 105, 66–72. , , , & (
- 2007). Individualized developmental care in a Dutch NICU: Short-term clinical outcome. Acta Paediatrica, 96, 1409–1415. , , & (
- 2006). How satisfied are parents supported by nurses with the NIDCAP model of care for their preterm infant Journal of Nursing Care Quality, 21, 41–48. , & (
- World Health Organisation. (2003). Kangaroo mother care: A practical guide. Geneva , Switzerland : Author (Department of Reproductive Health and Research) .