- Email: [email protected]
Evaluating the Precision of Clinical Assessments for Feeding Intolerance
Newborn & Infant Nursing Reviews 13 (2013) 184–188
Contents lists available at ScienceDirect
Newborn & Infant Nursing Reviews journal homepage: www.nainr.com
Evaluating the Precision of Clinical Assessments for Feeding Intolerance Tiffany A. Moore, PhD, RN a,⁎, Rita H. Pickler, PhD, RN, PNP-BC, FAAN b a b
College of Nursing, University of Nebraska Medical Center, Omaha, NE Cincinnati Children’s Hospital Medical Center, Cincinnati, OH
a r t i c l e
i n f o
Keywords: Feeding intolerance Clinimetrics Neonate Very low birth weight infant Neonatal intensive care Gastric residual Nutrition Necrotizing enterocolitis
a b s t r a c t Feeding intolerance is a common occurrence in preterm infants, yet there are no precise measures for clinically assessing this potentially serious manifestation. This article reports the results of a study designed to evaluate neonatal intensive care (NICU) nurses’ precision in abdominal and emesis assessments, considered the most objective, observable signs of feeding intolerance. Forty-six NICU nurses participated in the study by observing pictures of preterm infant abdomens and pictures of “staged” emesis. There was little agreement among the participants regarding the infant abdomen pictures or the amount of emesis observed in the pictures. There was no relationship between years of NICU experience nurses’ assessments. The ability of nurses to assess clinical signs of potentially serious complications in preterm infants is critically important. Standardized education and training as well as precise assessment tools are needed. © 2013 Elsevier Inc. All rights reserved.
Definition and Significance of Feeding Intolerance Preterm infants are unable to tolerate full enteral feedings immediately after birth. Instead, these newborns are placed on trophic feedings with a gradual increase to full enteral feeding status administered through a nasogastric tube. This process is individualized based on the infant’s gestational age, weight, clinical status, and enteral tolerance. The healthcare team in the newborn intensive care unit (NICU) monitors the infant’s tolerance to the enteral feedings for any suspicion of feeding intolerance (FI). Clinical manifestations and symptoms of FI include large gastric residuals, abdominal distention, emesis, bloody stools, apnea, bradycardia, hypotension, and temperature instability. 1,2 The incidence of FI in preterm infants is estimated from 16–29% depending on the definition used. 3,4 Previous studies have defined FI using measurable outcomes (i.e., days to reach full enteral feedings) while other studies have defined the concept using one or more of the clinical manifestations listed above (i.e., number of emesis in a 24hour period). For the purpose of this article, the operational definition of FI will comprise of the final definition described in a recent concept analysis, “FI in the premature infant is the inability to digest enteral feedings presented as GRV more than 50%, abdominal distention or emesis or both, and the disruption of the patient’s feeding plan.” 2 Work completed at The Nebraska Medical Center/University of Nebraska Medical Center. Supported in part by the Nellie House Craven Scholarship from the University of Nebraska-Medical Center, College of Nursing. ⁎ Address correspondence to. Tiffany A. Moore, PhD, RN, Assistant Professor, College of Nursing, Room 5034, University of Nebraska Medical Center, 985330 Nebraska Medical Center, Omaha, NE 68198-5330. Tel.: +1 402 559 6613; fax: +1 402 559 4303. E-mail addresses: [email protected] (T.A. Moore), [email protected] (R.H. Pickler). 1527-3369/1304-0532$36.00/0 – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1053/j.nainr.2013.09.005
FI is a clinically significant clinical phenomenon because of its known association with necrotizing enterocolitis (NEC), a gastrointestinal emergency that is the primary source of neonatal morbidity and mortality. 5 Although the exact pathogenesis remains unknown, NEC is believed to be associated with an unknown cause of dysregulation within the inflammatory cascade causing damage to the intestinal mucosa. 6 Precipitating events and risk factors for NEC often are generalized and systemic such as hypoxia and sepsis. FI also has been theoretically associated with a disruption in the gastrointestinal homeostasis through the brain-gut axis and enteric nervous system. 7,8 Stress, which includes the physiologic stress from disease processes, is thought to affect gastrointestinal function because of the complex and intricate processes within the enteric nervous system. A recent study confirmed this hypothesis because of an association found between levels of stress biomarkers and FI in preterm infants suggesting physiologic dysregulation may play a role in FI. 9 Both NEC and FI are multifactorial and likely reflect the multiple body systems involved in the etiologies. Since the exact etiology of NEC and its link to FI remains unclear, clinicians typically respond to clinical manifestations of FI by changing the infant’s feeding care plan. Changes in the feeding plan include decreasing, withholding, or discontinuing enteral feedings which then prolong the use of intravascular access, Total Parental Nutrition (TPN), and enteral fasting and ultimately increases the risk for further complications. 10,11 Accuracy and Precision of Feeding Intolerance Clinicians and researchers in the NICU rely on the assessment and critical thinking skills of the bedside nurse to prevent and treat complications of prematurity, specifically FI. Appropriately
T.A. Moore, R.H. Pickler / Newborn & Infant Nursing Reviews 13 (2013) 184–188
recognizing and measuring visual assessments associated with FI relies on the observational skills and clinical judgment of the bedside nurse. It also requires standardized policies and procedures provided by the institution. The institution is responsible for educating and maintaining competent assessment skills for the nurses as well as staying updated on the current research evidence to provide the safest and best practices for the patients. Carter has provided comprehensive recommendations for universal standard of care assessment guidelines for nurses to use when assessing for FI in preterm infants. 1 These guidelines discuss the nursing assessment techniques, nursing interventions, and anticipated practitioner orders for gastric residuals, abdominal distention, emesis, stool, apnea, bradycardia, and temperature instability. However, psychometric testing for the measurements related to FI has not been explored. Thus, institutions lack substantiated evidence-based practice guidelines for implementation. Psychometric testing, also called clinimetrics, is used to test the “validity” and “reliability” of measurements used to operationally define a phenomenon. 12 In clinimetrics, validity, which is referred to as accuracy, provides evidence that the measure used (e.g. abdominal distention) is a true reflection of the phenomenon (e.g. FI). Reliability in clinimetrics refers to the measure’s precision and provides evidence that the measure (e.g. abdominal distention) is consistent every time (e.g. same interpretation and documentation between nurses). A literature search failed to reveal any reports of the clinimetric testing of physiologic measurements associated with FI. Research on the accuracy and precision of the clinical measurements associated with FI is needed in order to help establish a universal definition of FI and to advance the science and practice for NICU practice guidelines. The present study was designed to examine the clinimetrics of the final definition stated above from the concept analysis publication. 2 Of the symptoms associated with FI in this definition, abdominal distention and emesis are physiological clinical signs that are considered concrete and objective because they can be physically seen and measured. However, actually measuring and interpreting these clinical signs may be more indeterminate and subjective. Both abdominal distention and emesis have been vaguely defined in previous studies. 2 Specifically, abdominal distention rarely has been defined as a change in abdominal girth using an objective measuring system as suggested by Carter. 1 Instead, abdominal distention has been a subjective observation of the healthcare team. Similarly, emesis most often has been defined as “severe” without specification of volume or color. This definition again must rely on the subjective observation of the healthcare team to define “severe”. Therefore, the purpose of this study was to: 1) evaluate the precision of abdominal assessments (flat, round, full, distended) and emesis assessments (small, moderate, large) by NICU nurses; and 2) identify relationships between assessment interpretation and years of clinical experience.
185
notice of the study’s purpose and protocol, an explanation of their human subject rights, and a statement that participation in the study implied informed consent via email. These same documents were displayed in the NICU nurse’s lounge and nursing stations prior to the annual, institutional proctored competency exams. Photographs After consent and with assistance from the bedside nurse, the lead study investigator took de-identified photographs of the infants’ abdominal area during routine caregiving. For the emesis pictures, pre-measured volumes of formula (5 mL, 10 mL, etc.) were dispersed on separate white diaper clothes. A tape measure was then placed on the diaper cloths after dispersion of the varying amounts of formula to be used as a reference point within the photograph. Procedures to Evaluate Precision Abdominal and emesis photographs were printed on an 8x11 piece of photo paper and given a specific number. All photographs were compiled in a notebook. At the conclusion of each nurse’s annual competency exams, the nurse had the opportunity to participate in the study. A poster about the study, the book of the assessment photographs, a locked ballot box, and a paper copy of the IRB documents (described above) was located in the same area as the competency examinations. Participation in the study was voluntary. The nurses viewed the book of the de-identified abdominal assessment photographs of NICU patients and the emesis photographs. For each assessment photograph, the nurse was instructed to complete survey questions by selecting responses reflecting their perception of the photographs and consistent with their usual charting preferences. The description options of the survey questions were multiple-choice to avoid penmanship recognition errors following data collection and to be consistent with the institution’s current computer documentation format. This anonymous survey also provided a space for the nurse to record his or her years of experience (categorized into 0–5 years, 5–10 years, N10 years). When the nurses completed the survey, they placed their form into the locked ballot box available for that purpose. The descriptive options for abdominal assessment photographs were distended, full, round, and flat which is consistent with the institution’s charting. The descriptive options for emesis photographs were small, moderate, large, x1 (non-specific documentation that the infant had one episode of emesis), 5 mL, 10 mL, 15 mL and 20 mL. Nurses were able to choose more than one answer for the emesis options to remain consistent with usual charting preferences. Statistics
Methods A prospective, descriptive, correlational design was used. Using photographs of the abdomens of preterm infants and photographs of “staged” emesis on cloth diapers that are typically used as “burp” cloths, NICU nurses were asked to complete a survey regarding their abdominal assessment and emesis amount. Approval from the institutional review board was obtained. A convenience sample of NICU patients was recruited in order to obtain abdominal photographs. Research personnel identified NICU patients of varying gestational ages and approached a parent for consent to photograph the infant’s abdomen. A second convenience sample of nurses working in the level III NICU in a Midwestern US tertiary medical center that served as the study setting was recruited to evaluate the precision of abdominal and emesis assessments for the study. The IRB approved the study as exempt from written consent for the nurses while requiring written consent from the infants’ parents. As written consent was not required for the nurses, they were sent a
Descriptive statistics and ANOVA were performed using SPSS, version 19 (SPSS Inc., Chicago, IL). Results The parents of 6 NICU patients consented to the study. These infants were photographed during routine caregiving; 10 abdominal assessment photographs were used. The second convenience sample included 46 (73%) of the setting’s NICU nurses who completed the study. Nurses displayed wide variability in their perceptions of the abdominal and emesis assessment photographs. Table 1 shows the results of the nurses’ assessments. Abdominal Distention Examples of the abdominal assessment photographs are see in Figs 1, 2, and 3. Fig 1 is an abdominal photograph of a preterm infant at
186
T.A. Moore, R.H. Pickler / Newborn & Infant Nursing Reviews 13 (2013) 184–188
Table 1 Assessment Results. Abdominal assessment
Distended
Photograph #1⁎ Photograph #2 Photograph #3⁎
15% 0% 30%
“Staged” emesis† Actual amount = 5 mL Actual amount = 10 mL Actual amount = 20 mL Years of clinical experience
Small 72% 17% 26% b5 years 35%
Full 52% 0% 59%
Round 0% 46% 11%
Flat 33% 54% 0%
Moderate
Large
X1
5 mL
10 mL
15 mL
20 mL
2% 33% 28% 5–10 years 15%
0% 13% 4% N10 years 30%
54% 8% 6% No response 20%
24% 59% 65%
0% 15% 15%
0% 2% 0%
0% 0% 0%
⁎ Photograph #1 and #3 are of the same patient, but at different angles. † Subjects were allowed to choose more than one response option, which is consistent with the study unit’s charting in the medical record.
approximately 34 weeks gestational age. Fifteen percent (15%) of the nurses interpreted this assessment photograph as distended; 52% chose full; 0% chose round; and 33% chose flat. Fig 2 is an abdominal assessment photograph of a preterm infant at approximately 29 weeks gestational age with an umbilical line in place. No nurses interpreted this photograph as distended or full, but 46% chose round and 54% chose flat. Fig 3 is an abdominal photograph of a preterm infant at approximately 34 weeks gestational age and is the same patient as Fig 1 but the photograph is at a different angle. Thirty percent (30%) of nurses interpreted this as distended; 59% chose full; 11% chose round; and none chose flat.
Emesis For the emesis photographs, nurses were allowed to choose more than one response option, which is consistent with the study unit’s charting in the medical record. Fig 4 is the photograph of a diaper cloth saturated with 5 mL of formula. Seventy-two percent (72%) of the nurses chose small; 2% chose moderate; none chose large. Additionally, 54% would have indicated an emesis of x1 in the medical record while 24% would have charted 5 mL; and none would have charted 10 mL, 15 mL, or 20 mL in the medical record as output. Fig 5 is the photograph of a diaper cloth saturated with 10 mL of formula. Seventeen percent interpreted the amount as small emesis while 33% chose moderate, and 13% chose large. Eight percent (8%) would have indicated emesis x1 in the medical record while 59% would have charted 5 mL, 15% would have charted 10 mL, 2% would have charted 15 mL, and none would have charted 20 mL in the medical record.
Fig 1. Preterm infant abdomen at approximately 34 weeks gestational age. Same patient as Fig. 3 but photograph taken at a different angle. Nurse responses: 15% = distended; 52% = full; 33% = flat. (Color version of figure is available online.)
Fig 6 is the photograph of a diaper cloth saturated with 20 ml of formula. Twenty-six percent (26%) interpreted as small while 28% chose moderate, and 4% chose large. Six percent (6%) would have indicated an emesis x1 in the medical record while 65% would have charted 5 mL, 15% would have charted 10 ml, and none would have charted 15 ml or 20 ml in the medical record.
Years of Clinical Experience Twenty percent (20%, n = 9) of the nurses did not complete the question regarding their years of experience in the NICU. For the remaining nurses who completed the study, 35% (n = 16) identified themselves as having b5 years of experience, 15% (n = 7) had 5– 10 years of experience, and 30% (n = 14) had N10 years of experience. Comparison analyses showed no differences in the interpretation of assessments based on years of clinical experience in the NICU.
Discussion This study is the first to examine the precision of the physiologic clinical signs associated with FI in preterm infants. Specifically, interpretation of abdominal and emesis photographs by NICU nurses were measured. Variability of the assessments based on years of clinical experience also was examined. NICU nurses displayed a wide variability in the interpretation of abdominal and emesis assessment photographs of preterm infants. Nurses also displayed a wide variation in how they charted these assessments in the medical
Fig 2. Preterm infant abdomen at approximately 29 weeks gestational age with an umbilical line in place. Nurse responses: 46% = round; 54% = flat. (Color version of figure is available online.)
T.A. Moore, R.H. Pickler / Newborn & Infant Nursing Reviews 13 (2013) 184–188
187
Fig 5. Diaper cloth saturated with 10 mL of formula. Nurse responses: 17% = small; 33% = moderate; 13% = large; 8% = emesis x1; 59% = 5 mL; 15% = 10 mL; 2% =15 mL. (Color version of figure is available online.) Fig 3. Preterm infant abdomen of a at approximately 34 weeks gestational age. Same patient as Fig. 1 but photograph taken at a different angle. Nurse responses: 30% = distended; 59% = full; 11% = round. (Color version of figure is available online.)
record. No relationships between assessment interpretation and charting were found based on years of clinical experience. As discussed by Carter, the bedside nurse’s assessments of the clinical signs and symptoms of FI in preterm infants is essential 1. The results of this study suggest that there is a great deal of variability in these assessments as well as in the way nurses record their assessments in the medical record. In the first case, the lack of consistency across nurses in judging abdominal distention may result in one of the earliest warning signs of FI going undetected, potentially resulting in dangerous medical situations. In the second case, lack of consistency in recording assessments in the medical record may result in further failure of communication within the health care team to act on abnormal physical assessments. In either case, these inconsistencies may disrupt and delay the critical process of identification and prevention for major complications such NEC. No prior studies evaluating either the accuracy or precision of nursing assessments of abdominal distention were found in the literature. Thus, it is not possible to know if the study findings reflect a new phenomenon. One study authored by Craft was found where nursing students and practicing pediatric nurses (n = 109) were presented with visual displays of formula on receiving blankets and asked to determine the volume. 13 Findings showed that few displays were assessed accurately and that error increased with the increase in
Fig 4. Diaper cloth saturated with 5 mL of formula. Nurse responses: 72% = small; 2% = moderate; 54% = emesis of x1; 24% = 5 mL. (Color version of figure is available online.)
displayed volume which is consistent with the findings in the current study. Additional analyses by Craft and colleagues revealed that practice role, nature of clinical practice, and number of displays assessed accounted for significant proportions of variance in the relative error, suggesting that visual assessment techniques require a mental frame of reference with which they compare the observed volume. This is also consistent with a more recent cross-sectional study comparing a visual dental examination method with assessment of intra-oral photographs as means of detecting dental caries in children. 14 The researchers found good intra-examiner reliability for both the visual and the photographic methods with no clinically significant differences between the photographic scores and the visual assessments. The researchers concluded that the photographic approach was equivalent in diagnostic utility to the visual system in a group of well-trained clinicians.
Implications for Nursing Practice NICU educators need to develop policies that support standardized interpretation and recording visual assessments, such as the guidelines provided by Carter.1 Few studies have used routine abdominal girth measurements as an objective measure of abdominal distention. This simple evaluation technique may provide individualized, yet standardized, patient assessments. Another potential standardization method may be the estimation of emesis. Often times, emesis is difficult to
Fig 6. Diaper cloth saturated with 20 mL of formula. Nurse responses: 26% = small; 28% =moderate; 4% = large; 6% = emesis x1; 65% = 5 mL; 15% = 10 mL. (Color version of figure is available online.)
188
T.A. Moore, R.H. Pickler / Newborn & Infant Nursing Reviews 13 (2013) 184–188
measure because of varying and numerous weights and sizes of blankets involved. 15 Whether the NICU identifies a method for object comparison (i.e., less than a half-dollar is small) or education for estimations, thoughtful and strategic discussions of these basic assessments need to occur to guide the standardization policies and procedures. In addition to education, practitioners need to be aware of the variability of visual assessments associated with FI when considering the feeding care plan. Certainly any inconsistencies in charting or in their own assessments as compared to the assessments recorded in the medical record should be questioned. The clinicians making the feeding care decisions also need to be involved in the discussion of standardizing assessments and the entire team would likely benefit from interdisciplinary training.
Implications for Nursing Research Previous studies have demonstrated the benefits of decision trees and clinical guidelines to help bedside nurses evaluate and interpret assessments for early diagnosis and interventions to prevent major complications. 16,17 Although these clinical tools are helpful, psychometric testing of the tools and of the clinical assessments associated with FI is needed. Additionally, and perhaps more fundamentally, more research about FI in preterm infants is needed. First, researchers need to develop methods to increase reliability and validity of visual assessments used as measures of FI. The reliability and precision of the physiologic measurements believed to be associated with FI needs to be established. This will allow researchers to identify valid and accurate methods of measurement for FI. Second, more research is needed to conceptually and operationally define FI in this population. To advance the science for the phenomenon of FI, a universal definition for FI in preterm infants is necessary. However, a reliable and valid measurement tool is needed before a universal definition is identified and tested. Once these fundamental steps have been explored, promotion and intervention studies can be researched.
References 1. Carter BM. Feeding intolerance in preterm infants and standard of care guidelines for nursing assessments. Newborn Infant Nurs Rev. 2012;12:187-201. 2. Moore TA, Wilson ME. Feeding intolerance: a concept analysis. Adv Neonat Care. 2011;11:149-154. 3. Krishnamurthy S, Gupta P, Debnath S, Gomber S. Slow versus rapid enteral feeding advancement in preterm newborn infants 1000-1499 g: a randomized controlled trial. Acta Paediatr. 2010;99:42-46. 4. Leaf A, Dorling J, Kempley S, et al. Early or delayed enteral feeding for preterm growth-restricted infants: a randomized trial. Pediatrics. 2012;129:e1260-1268. 5. Horbar JD, Carpenter JH, Badger GJ, et al. Mortality and neonatal morbidity among infants 501 to 1500 grams from 2000 to 2009. Pediatrics. 2012;129:1019-1026. 6. Gephart SM, McGrath JM, Effken JA, Halpern MD. Necrotizing enterocolitis risk: state of the science. Adv Neonat Care. 2012;12:77-87. 7. Indrio F, Neu J. The intestinal microbiome of infants and the use of probiotics. Curr Opin Pediatr. 2011;23:145-150. 8. Moore TA, Berger AM, Wilson ME. A new way of thinking about complications of prematurity. Biol Res for Nurs. 2012. [Epub ahead of print]. 9. Moore TA, Wilson ME, Schmid KK, French JA, Anderson-Berry A, Berger AM. Relationships between feeding intolerance and stress biomarkers in preterm infants. J Pediatr Gastroenterol Nutr. 2013;57:356-362. 10. Rangel SJ, Calkins CM, Cowles RA, Barnhart DC, Huang EY, Abdullah F, et al. Parenteral nutrition-associated cholestasis: An American Pediatric Surgical Association outcomes and clinical trials committee systematic review. J Pediatr Surg. 2012 01;47:225-240. 11. Hemels MAC, van dH, Verboon-Maciolek M, Fleer A, Krediet TG. Prevention of neonatal late-onset sepsis associated with the removal of percutaneously inserted central venous catheters in preterm infants. Pediatr Crit Care Med. 2011;12: 445-448. 12. Fava GA, Tomba E, Sonino N. Clinimetrics: the science of clinical measurements. Int J Clin Pract. 2012;66:11-15. 13. Craft M, Moss J. Accuracy of infant emesis volume assessment. Appl Nurs Res. 1996;9:2-8. 14. Boye U, Willasey A, Walsh T, Tickle M, Pretty IA. Comparison of an intra-oral photographic caries assessment with an established visual caries assessment method for use in dental epidemiological studies of children [published ahead of print April 9, 2013]. Community Dent Oral Epidemiol. 2013, http://dx.doi.org/ 10.1111/cdoe.12049. 15. Dowling DA, Madigan E, Siripul P. The effect of fluid density and volume on the accuracy of test weighing in a simulated oral feeding situation. Adv Neonatal Care. 2004;4:158-165. 16. Hanson C, Sundermeier J, Dugick L, Lyden E, Anderson-Berry AL. Implementation, process, and outcomes of nutrition best practices for infants b1500 g. Nutr Clin Pract. 2011;26:614-624. 17. Wiedmeier SE, Henry E, Baer VL, et al. Center differences in NEC within one healthcare system may depend on feeding protocol. Am J Perinatol. 2008;25:5-11.
Contents lists available at ScienceDirect
Newborn & Infant Nursing Reviews journal homepage: www.nainr.com
Evaluating the Precision of Clinical Assessments for Feeding Intolerance Tiffany A. Moore, PhD, RN a,⁎, Rita H. Pickler, PhD, RN, PNP-BC, FAAN b a b
College of Nursing, University of Nebraska Medical Center, Omaha, NE Cincinnati Children’s Hospital Medical Center, Cincinnati, OH
a r t i c l e
i n f o
Keywords: Feeding intolerance Clinimetrics Neonate Very low birth weight infant Neonatal intensive care Gastric residual Nutrition Necrotizing enterocolitis
a b s t r a c t Feeding intolerance is a common occurrence in preterm infants, yet there are no precise measures for clinically assessing this potentially serious manifestation. This article reports the results of a study designed to evaluate neonatal intensive care (NICU) nurses’ precision in abdominal and emesis assessments, considered the most objective, observable signs of feeding intolerance. Forty-six NICU nurses participated in the study by observing pictures of preterm infant abdomens and pictures of “staged” emesis. There was little agreement among the participants regarding the infant abdomen pictures or the amount of emesis observed in the pictures. There was no relationship between years of NICU experience nurses’ assessments. The ability of nurses to assess clinical signs of potentially serious complications in preterm infants is critically important. Standardized education and training as well as precise assessment tools are needed. © 2013 Elsevier Inc. All rights reserved.
Definition and Significance of Feeding Intolerance Preterm infants are unable to tolerate full enteral feedings immediately after birth. Instead, these newborns are placed on trophic feedings with a gradual increase to full enteral feeding status administered through a nasogastric tube. This process is individualized based on the infant’s gestational age, weight, clinical status, and enteral tolerance. The healthcare team in the newborn intensive care unit (NICU) monitors the infant’s tolerance to the enteral feedings for any suspicion of feeding intolerance (FI). Clinical manifestations and symptoms of FI include large gastric residuals, abdominal distention, emesis, bloody stools, apnea, bradycardia, hypotension, and temperature instability. 1,2 The incidence of FI in preterm infants is estimated from 16–29% depending on the definition used. 3,4 Previous studies have defined FI using measurable outcomes (i.e., days to reach full enteral feedings) while other studies have defined the concept using one or more of the clinical manifestations listed above (i.e., number of emesis in a 24hour period). For the purpose of this article, the operational definition of FI will comprise of the final definition described in a recent concept analysis, “FI in the premature infant is the inability to digest enteral feedings presented as GRV more than 50%, abdominal distention or emesis or both, and the disruption of the patient’s feeding plan.” 2 Work completed at The Nebraska Medical Center/University of Nebraska Medical Center. Supported in part by the Nellie House Craven Scholarship from the University of Nebraska-Medical Center, College of Nursing. ⁎ Address correspondence to. Tiffany A. Moore, PhD, RN, Assistant Professor, College of Nursing, Room 5034, University of Nebraska Medical Center, 985330 Nebraska Medical Center, Omaha, NE 68198-5330. Tel.: +1 402 559 6613; fax: +1 402 559 4303. E-mail addresses: [email protected] (T.A. Moore), [email protected] (R.H. Pickler). 1527-3369/1304-0532$36.00/0 – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1053/j.nainr.2013.09.005
FI is a clinically significant clinical phenomenon because of its known association with necrotizing enterocolitis (NEC), a gastrointestinal emergency that is the primary source of neonatal morbidity and mortality. 5 Although the exact pathogenesis remains unknown, NEC is believed to be associated with an unknown cause of dysregulation within the inflammatory cascade causing damage to the intestinal mucosa. 6 Precipitating events and risk factors for NEC often are generalized and systemic such as hypoxia and sepsis. FI also has been theoretically associated with a disruption in the gastrointestinal homeostasis through the brain-gut axis and enteric nervous system. 7,8 Stress, which includes the physiologic stress from disease processes, is thought to affect gastrointestinal function because of the complex and intricate processes within the enteric nervous system. A recent study confirmed this hypothesis because of an association found between levels of stress biomarkers and FI in preterm infants suggesting physiologic dysregulation may play a role in FI. 9 Both NEC and FI are multifactorial and likely reflect the multiple body systems involved in the etiologies. Since the exact etiology of NEC and its link to FI remains unclear, clinicians typically respond to clinical manifestations of FI by changing the infant’s feeding care plan. Changes in the feeding plan include decreasing, withholding, or discontinuing enteral feedings which then prolong the use of intravascular access, Total Parental Nutrition (TPN), and enteral fasting and ultimately increases the risk for further complications. 10,11 Accuracy and Precision of Feeding Intolerance Clinicians and researchers in the NICU rely on the assessment and critical thinking skills of the bedside nurse to prevent and treat complications of prematurity, specifically FI. Appropriately
T.A. Moore, R.H. Pickler / Newborn & Infant Nursing Reviews 13 (2013) 184–188
recognizing and measuring visual assessments associated with FI relies on the observational skills and clinical judgment of the bedside nurse. It also requires standardized policies and procedures provided by the institution. The institution is responsible for educating and maintaining competent assessment skills for the nurses as well as staying updated on the current research evidence to provide the safest and best practices for the patients. Carter has provided comprehensive recommendations for universal standard of care assessment guidelines for nurses to use when assessing for FI in preterm infants. 1 These guidelines discuss the nursing assessment techniques, nursing interventions, and anticipated practitioner orders for gastric residuals, abdominal distention, emesis, stool, apnea, bradycardia, and temperature instability. However, psychometric testing for the measurements related to FI has not been explored. Thus, institutions lack substantiated evidence-based practice guidelines for implementation. Psychometric testing, also called clinimetrics, is used to test the “validity” and “reliability” of measurements used to operationally define a phenomenon. 12 In clinimetrics, validity, which is referred to as accuracy, provides evidence that the measure used (e.g. abdominal distention) is a true reflection of the phenomenon (e.g. FI). Reliability in clinimetrics refers to the measure’s precision and provides evidence that the measure (e.g. abdominal distention) is consistent every time (e.g. same interpretation and documentation between nurses). A literature search failed to reveal any reports of the clinimetric testing of physiologic measurements associated with FI. Research on the accuracy and precision of the clinical measurements associated with FI is needed in order to help establish a universal definition of FI and to advance the science and practice for NICU practice guidelines. The present study was designed to examine the clinimetrics of the final definition stated above from the concept analysis publication. 2 Of the symptoms associated with FI in this definition, abdominal distention and emesis are physiological clinical signs that are considered concrete and objective because they can be physically seen and measured. However, actually measuring and interpreting these clinical signs may be more indeterminate and subjective. Both abdominal distention and emesis have been vaguely defined in previous studies. 2 Specifically, abdominal distention rarely has been defined as a change in abdominal girth using an objective measuring system as suggested by Carter. 1 Instead, abdominal distention has been a subjective observation of the healthcare team. Similarly, emesis most often has been defined as “severe” without specification of volume or color. This definition again must rely on the subjective observation of the healthcare team to define “severe”. Therefore, the purpose of this study was to: 1) evaluate the precision of abdominal assessments (flat, round, full, distended) and emesis assessments (small, moderate, large) by NICU nurses; and 2) identify relationships between assessment interpretation and years of clinical experience.
185
notice of the study’s purpose and protocol, an explanation of their human subject rights, and a statement that participation in the study implied informed consent via email. These same documents were displayed in the NICU nurse’s lounge and nursing stations prior to the annual, institutional proctored competency exams. Photographs After consent and with assistance from the bedside nurse, the lead study investigator took de-identified photographs of the infants’ abdominal area during routine caregiving. For the emesis pictures, pre-measured volumes of formula (5 mL, 10 mL, etc.) were dispersed on separate white diaper clothes. A tape measure was then placed on the diaper cloths after dispersion of the varying amounts of formula to be used as a reference point within the photograph. Procedures to Evaluate Precision Abdominal and emesis photographs were printed on an 8x11 piece of photo paper and given a specific number. All photographs were compiled in a notebook. At the conclusion of each nurse’s annual competency exams, the nurse had the opportunity to participate in the study. A poster about the study, the book of the assessment photographs, a locked ballot box, and a paper copy of the IRB documents (described above) was located in the same area as the competency examinations. Participation in the study was voluntary. The nurses viewed the book of the de-identified abdominal assessment photographs of NICU patients and the emesis photographs. For each assessment photograph, the nurse was instructed to complete survey questions by selecting responses reflecting their perception of the photographs and consistent with their usual charting preferences. The description options of the survey questions were multiple-choice to avoid penmanship recognition errors following data collection and to be consistent with the institution’s current computer documentation format. This anonymous survey also provided a space for the nurse to record his or her years of experience (categorized into 0–5 years, 5–10 years, N10 years). When the nurses completed the survey, they placed their form into the locked ballot box available for that purpose. The descriptive options for abdominal assessment photographs were distended, full, round, and flat which is consistent with the institution’s charting. The descriptive options for emesis photographs were small, moderate, large, x1 (non-specific documentation that the infant had one episode of emesis), 5 mL, 10 mL, 15 mL and 20 mL. Nurses were able to choose more than one answer for the emesis options to remain consistent with usual charting preferences. Statistics
Methods A prospective, descriptive, correlational design was used. Using photographs of the abdomens of preterm infants and photographs of “staged” emesis on cloth diapers that are typically used as “burp” cloths, NICU nurses were asked to complete a survey regarding their abdominal assessment and emesis amount. Approval from the institutional review board was obtained. A convenience sample of NICU patients was recruited in order to obtain abdominal photographs. Research personnel identified NICU patients of varying gestational ages and approached a parent for consent to photograph the infant’s abdomen. A second convenience sample of nurses working in the level III NICU in a Midwestern US tertiary medical center that served as the study setting was recruited to evaluate the precision of abdominal and emesis assessments for the study. The IRB approved the study as exempt from written consent for the nurses while requiring written consent from the infants’ parents. As written consent was not required for the nurses, they were sent a
Descriptive statistics and ANOVA were performed using SPSS, version 19 (SPSS Inc., Chicago, IL). Results The parents of 6 NICU patients consented to the study. These infants were photographed during routine caregiving; 10 abdominal assessment photographs were used. The second convenience sample included 46 (73%) of the setting’s NICU nurses who completed the study. Nurses displayed wide variability in their perceptions of the abdominal and emesis assessment photographs. Table 1 shows the results of the nurses’ assessments. Abdominal Distention Examples of the abdominal assessment photographs are see in Figs 1, 2, and 3. Fig 1 is an abdominal photograph of a preterm infant at
186
T.A. Moore, R.H. Pickler / Newborn & Infant Nursing Reviews 13 (2013) 184–188
Table 1 Assessment Results. Abdominal assessment
Distended
Photograph #1⁎ Photograph #2 Photograph #3⁎
15% 0% 30%
“Staged” emesis† Actual amount = 5 mL Actual amount = 10 mL Actual amount = 20 mL Years of clinical experience
Small 72% 17% 26% b5 years 35%
Full 52% 0% 59%
Round 0% 46% 11%
Flat 33% 54% 0%
Moderate
Large
X1
5 mL
10 mL
15 mL
20 mL
2% 33% 28% 5–10 years 15%
0% 13% 4% N10 years 30%
54% 8% 6% No response 20%
24% 59% 65%
0% 15% 15%
0% 2% 0%
0% 0% 0%
⁎ Photograph #1 and #3 are of the same patient, but at different angles. † Subjects were allowed to choose more than one response option, which is consistent with the study unit’s charting in the medical record.
approximately 34 weeks gestational age. Fifteen percent (15%) of the nurses interpreted this assessment photograph as distended; 52% chose full; 0% chose round; and 33% chose flat. Fig 2 is an abdominal assessment photograph of a preterm infant at approximately 29 weeks gestational age with an umbilical line in place. No nurses interpreted this photograph as distended or full, but 46% chose round and 54% chose flat. Fig 3 is an abdominal photograph of a preterm infant at approximately 34 weeks gestational age and is the same patient as Fig 1 but the photograph is at a different angle. Thirty percent (30%) of nurses interpreted this as distended; 59% chose full; 11% chose round; and none chose flat.
Emesis For the emesis photographs, nurses were allowed to choose more than one response option, which is consistent with the study unit’s charting in the medical record. Fig 4 is the photograph of a diaper cloth saturated with 5 mL of formula. Seventy-two percent (72%) of the nurses chose small; 2% chose moderate; none chose large. Additionally, 54% would have indicated an emesis of x1 in the medical record while 24% would have charted 5 mL; and none would have charted 10 mL, 15 mL, or 20 mL in the medical record as output. Fig 5 is the photograph of a diaper cloth saturated with 10 mL of formula. Seventeen percent interpreted the amount as small emesis while 33% chose moderate, and 13% chose large. Eight percent (8%) would have indicated emesis x1 in the medical record while 59% would have charted 5 mL, 15% would have charted 10 mL, 2% would have charted 15 mL, and none would have charted 20 mL in the medical record.
Fig 1. Preterm infant abdomen at approximately 34 weeks gestational age. Same patient as Fig. 3 but photograph taken at a different angle. Nurse responses: 15% = distended; 52% = full; 33% = flat. (Color version of figure is available online.)
Fig 6 is the photograph of a diaper cloth saturated with 20 ml of formula. Twenty-six percent (26%) interpreted as small while 28% chose moderate, and 4% chose large. Six percent (6%) would have indicated an emesis x1 in the medical record while 65% would have charted 5 mL, 15% would have charted 10 ml, and none would have charted 15 ml or 20 ml in the medical record.
Years of Clinical Experience Twenty percent (20%, n = 9) of the nurses did not complete the question regarding their years of experience in the NICU. For the remaining nurses who completed the study, 35% (n = 16) identified themselves as having b5 years of experience, 15% (n = 7) had 5– 10 years of experience, and 30% (n = 14) had N10 years of experience. Comparison analyses showed no differences in the interpretation of assessments based on years of clinical experience in the NICU.
Discussion This study is the first to examine the precision of the physiologic clinical signs associated with FI in preterm infants. Specifically, interpretation of abdominal and emesis photographs by NICU nurses were measured. Variability of the assessments based on years of clinical experience also was examined. NICU nurses displayed a wide variability in the interpretation of abdominal and emesis assessment photographs of preterm infants. Nurses also displayed a wide variation in how they charted these assessments in the medical
Fig 2. Preterm infant abdomen at approximately 29 weeks gestational age with an umbilical line in place. Nurse responses: 46% = round; 54% = flat. (Color version of figure is available online.)
T.A. Moore, R.H. Pickler / Newborn & Infant Nursing Reviews 13 (2013) 184–188
187
Fig 5. Diaper cloth saturated with 10 mL of formula. Nurse responses: 17% = small; 33% = moderate; 13% = large; 8% = emesis x1; 59% = 5 mL; 15% = 10 mL; 2% =15 mL. (Color version of figure is available online.) Fig 3. Preterm infant abdomen of a at approximately 34 weeks gestational age. Same patient as Fig. 1 but photograph taken at a different angle. Nurse responses: 30% = distended; 59% = full; 11% = round. (Color version of figure is available online.)
record. No relationships between assessment interpretation and charting were found based on years of clinical experience. As discussed by Carter, the bedside nurse’s assessments of the clinical signs and symptoms of FI in preterm infants is essential 1. The results of this study suggest that there is a great deal of variability in these assessments as well as in the way nurses record their assessments in the medical record. In the first case, the lack of consistency across nurses in judging abdominal distention may result in one of the earliest warning signs of FI going undetected, potentially resulting in dangerous medical situations. In the second case, lack of consistency in recording assessments in the medical record may result in further failure of communication within the health care team to act on abnormal physical assessments. In either case, these inconsistencies may disrupt and delay the critical process of identification and prevention for major complications such NEC. No prior studies evaluating either the accuracy or precision of nursing assessments of abdominal distention were found in the literature. Thus, it is not possible to know if the study findings reflect a new phenomenon. One study authored by Craft was found where nursing students and practicing pediatric nurses (n = 109) were presented with visual displays of formula on receiving blankets and asked to determine the volume. 13 Findings showed that few displays were assessed accurately and that error increased with the increase in
Fig 4. Diaper cloth saturated with 5 mL of formula. Nurse responses: 72% = small; 2% = moderate; 54% = emesis of x1; 24% = 5 mL. (Color version of figure is available online.)
displayed volume which is consistent with the findings in the current study. Additional analyses by Craft and colleagues revealed that practice role, nature of clinical practice, and number of displays assessed accounted for significant proportions of variance in the relative error, suggesting that visual assessment techniques require a mental frame of reference with which they compare the observed volume. This is also consistent with a more recent cross-sectional study comparing a visual dental examination method with assessment of intra-oral photographs as means of detecting dental caries in children. 14 The researchers found good intra-examiner reliability for both the visual and the photographic methods with no clinically significant differences between the photographic scores and the visual assessments. The researchers concluded that the photographic approach was equivalent in diagnostic utility to the visual system in a group of well-trained clinicians.
Implications for Nursing Practice NICU educators need to develop policies that support standardized interpretation and recording visual assessments, such as the guidelines provided by Carter.1 Few studies have used routine abdominal girth measurements as an objective measure of abdominal distention. This simple evaluation technique may provide individualized, yet standardized, patient assessments. Another potential standardization method may be the estimation of emesis. Often times, emesis is difficult to
Fig 6. Diaper cloth saturated with 20 mL of formula. Nurse responses: 26% = small; 28% =moderate; 4% = large; 6% = emesis x1; 65% = 5 mL; 15% = 10 mL. (Color version of figure is available online.)
188
T.A. Moore, R.H. Pickler / Newborn & Infant Nursing Reviews 13 (2013) 184–188
measure because of varying and numerous weights and sizes of blankets involved. 15 Whether the NICU identifies a method for object comparison (i.e., less than a half-dollar is small) or education for estimations, thoughtful and strategic discussions of these basic assessments need to occur to guide the standardization policies and procedures. In addition to education, practitioners need to be aware of the variability of visual assessments associated with FI when considering the feeding care plan. Certainly any inconsistencies in charting or in their own assessments as compared to the assessments recorded in the medical record should be questioned. The clinicians making the feeding care decisions also need to be involved in the discussion of standardizing assessments and the entire team would likely benefit from interdisciplinary training.
Implications for Nursing Research Previous studies have demonstrated the benefits of decision trees and clinical guidelines to help bedside nurses evaluate and interpret assessments for early diagnosis and interventions to prevent major complications. 16,17 Although these clinical tools are helpful, psychometric testing of the tools and of the clinical assessments associated with FI is needed. Additionally, and perhaps more fundamentally, more research about FI in preterm infants is needed. First, researchers need to develop methods to increase reliability and validity of visual assessments used as measures of FI. The reliability and precision of the physiologic measurements believed to be associated with FI needs to be established. This will allow researchers to identify valid and accurate methods of measurement for FI. Second, more research is needed to conceptually and operationally define FI in this population. To advance the science for the phenomenon of FI, a universal definition for FI in preterm infants is necessary. However, a reliable and valid measurement tool is needed before a universal definition is identified and tested. Once these fundamental steps have been explored, promotion and intervention studies can be researched.
References 1. Carter BM. Feeding intolerance in preterm infants and standard of care guidelines for nursing assessments. Newborn Infant Nurs Rev. 2012;12:187-201. 2. Moore TA, Wilson ME. Feeding intolerance: a concept analysis. Adv Neonat Care. 2011;11:149-154. 3. Krishnamurthy S, Gupta P, Debnath S, Gomber S. Slow versus rapid enteral feeding advancement in preterm newborn infants 1000-1499 g: a randomized controlled trial. Acta Paediatr. 2010;99:42-46. 4. Leaf A, Dorling J, Kempley S, et al. Early or delayed enteral feeding for preterm growth-restricted infants: a randomized trial. Pediatrics. 2012;129:e1260-1268. 5. Horbar JD, Carpenter JH, Badger GJ, et al. Mortality and neonatal morbidity among infants 501 to 1500 grams from 2000 to 2009. Pediatrics. 2012;129:1019-1026. 6. Gephart SM, McGrath JM, Effken JA, Halpern MD. Necrotizing enterocolitis risk: state of the science. Adv Neonat Care. 2012;12:77-87. 7. Indrio F, Neu J. The intestinal microbiome of infants and the use of probiotics. Curr Opin Pediatr. 2011;23:145-150. 8. Moore TA, Berger AM, Wilson ME. A new way of thinking about complications of prematurity. Biol Res for Nurs. 2012. [Epub ahead of print]. 9. Moore TA, Wilson ME, Schmid KK, French JA, Anderson-Berry A, Berger AM. Relationships between feeding intolerance and stress biomarkers in preterm infants. J Pediatr Gastroenterol Nutr. 2013;57:356-362. 10. Rangel SJ, Calkins CM, Cowles RA, Barnhart DC, Huang EY, Abdullah F, et al. Parenteral nutrition-associated cholestasis: An American Pediatric Surgical Association outcomes and clinical trials committee systematic review. J Pediatr Surg. 2012 01;47:225-240. 11. Hemels MAC, van dH, Verboon-Maciolek M, Fleer A, Krediet TG. Prevention of neonatal late-onset sepsis associated with the removal of percutaneously inserted central venous catheters in preterm infants. Pediatr Crit Care Med. 2011;12: 445-448. 12. Fava GA, Tomba E, Sonino N. Clinimetrics: the science of clinical measurements. Int J Clin Pract. 2012;66:11-15. 13. Craft M, Moss J. Accuracy of infant emesis volume assessment. Appl Nurs Res. 1996;9:2-8. 14. Boye U, Willasey A, Walsh T, Tickle M, Pretty IA. Comparison of an intra-oral photographic caries assessment with an established visual caries assessment method for use in dental epidemiological studies of children [published ahead of print April 9, 2013]. Community Dent Oral Epidemiol. 2013, http://dx.doi.org/ 10.1111/cdoe.12049. 15. Dowling DA, Madigan E, Siripul P. The effect of fluid density and volume on the accuracy of test weighing in a simulated oral feeding situation. Adv Neonatal Care. 2004;4:158-165. 16. Hanson C, Sundermeier J, Dugick L, Lyden E, Anderson-Berry AL. Implementation, process, and outcomes of nutrition best practices for infants b1500 g. Nutr Clin Pract. 2011;26:614-624. 17. Wiedmeier SE, Henry E, Baer VL, et al. Center differences in NEC within one healthcare system may depend on feeding protocol. Am J Perinatol. 2008;25:5-11.