OBJECTIVE: Despite evidence supporting its use, nasogastric rehydration is rarely used in North America. We conducted a prospective, cross-sectional, 3-phase study to evaluate current perspectives.
METHODS: We compared the proportions of respondents in favor of nasogastric (as opposed to intravenous) rehydration, should oral rehydration fail, between clinicians and caregivers. Phase 1: caregivers of children aged 3 to 48 months, who presented to a Canadian pediatric emergency department with symptoms of gastroenteritis, were invited to complete a survey. Phase 2: phase 1 participants administered intravenous or nasogastric rehydration had the procedure observed and outcome data recorded. Phase 3: pediatric emergency medicine physicians, fellows, and nurses completed a survey.
RESULTS: Four hundred thirty-five children-parent dyads and 113 health care providers participated. If oral rehydration were to fail, 10% (47 of 435) of caregivers and 14% (16 of 113) of clinicians would choose nasogastric rehydration (difference = 3.4%; 95% confidence interval: −2.8 to 11.4). Caregivers were more familiar with the term intravenous than nasogastric rehydration (80% vs 20%; P < .001). Sixty-four children (15%) received intravenous rehydration; none received nasogastric rehydration. Participating nurses have inserted 90 (interquartile range: 25–150) intravenous cannulas compared with 4 (interquartile range: 2–10) nasogastric tubes during the preceding 6 months (P < .001). After a brief educational intervention, the proportion recommending nasogastric rehydration increased to 27% (117 of 435) among caregivers (P < .001) and 43% (49 of 113) among health care providers (P < .001).
CONCLUSIONS: In keeping with caregiver desires, health care providers in a Canadian emergency department employ intravenous rehydration when oral rehydration fails. Enhanced change management strategies will be required for nasogastric rehydration to become adopted in this environment.
- CI —
- confidence interval
- ED —
- emergency department
- IQR —
- interquartile range
- ORT —
- oral rehydration therapy
What's Known on This Subject:
Some children with gastroenteritis fail to respond to oral rehydration. Subsequent interventions are dictated by regional preference. In North America, nasogastric rehydration is rarely administered. Caregiver and health care providers’ perspectives regarding its use have not been described previously.
What This Study Adds:
Both caregivers and health care providers would select intravenous rehydration instead of nasogastric rehydration when oral rehydration fails. Greater knowledge mobilization efforts will be required for nasogastric rehydration to be adopted into clinical practice.
Despite evidence in favor of oral rehydration therapy (ORT), some children with gastroenteritis fail to respond to therapy because of persistent vomiting, inadequate oral intake, or excessive diarrhea. The choice of subsequent interventions is dictated by regional preference. In Europe1 and Oceania,2 nasogastric rehydration predominates; in North America, intravenous rehydration is administered routinely.3 Despite endorsement by the American Academy of Pediatrics, nasogastric rehydration is administered to <1% of North American children with gastroenteritis.4
Although authors of numerous studies have evaluated the role of ORT,5 few have examined nasogastric rehydration as the sole rehydration intervention6–8 or in combination with ORT.9–12 A study of children with moderate to severe dehydration found that those administered nasogastric rehydration had a greater intake of fluids and electrolytes, fewer electrolyte disturbances, less vomiting, shorter duration of diarrhea, and lower mortality compared with those administered intravenous rehydration.8 Similar findings have been reported in children with lesser degrees of dehydration.6,7,11–13
Because adoption into clinical practice is highly variable,14 the identification of barriers to nasogastric rehydration use in North America merits evaluation. We conducted a multiphase study to evaluate caregiver and health care practitioner perspectives on the use of nasogastric and intravenous rehydration. We sought to determine, in a hypothetical model that includes the failure of ORT, whether the proportion of respondents in favor of nasogastric rehydration (as opposed to intravenous rehydration) would differ between health care providers and caregivers and whether an educational intervention could alter this outcome.
Two sample populations were evaluated: (1) caregivers of eligible children were enrolled in the emergency department (ED) of The Hospital for Sick Children, Toronto, between August 2010 and May 2011. Eligible children were aged 3 to 48 months and presented with gastroenteritis, defined by ≥3 episodes of vomiting or watery stools within the preceding 24 hours.15,16 Children were excluded if they had symptoms for ≥7 days, had previously been enrolled, or if caregivers were non–English-speaking. (2) Health care providers were eligible if they were pediatric emergency medicine physicians, fellows, or nurses. Research assistants were present 15 hours per day, 7 days per week to enroll participants. The study was approved by The Hospital for Sick Children’s research ethics board.
Survey Content and Administration
The surveys were developed in accordance with guidelines17 and included a literature review, expert panel item generation, reduction, and pretesting. Both surveys employed dichotomous and frequency-of-endorsement-question formats and required 10 minutes to complete. The caregiver survey (Supplemental Information 1) was administered by the research assistants and was used to collect data on demographics, symptoms, experiences, beliefs, knowledge, and desires regarding rehydration methods. The self-administered health care provider survey (Supplemental Information 2) was used to collect demographic and experiential data along with perceptions, knowledge, beliefs, and treatment patterns in children with gastroenteritis. It employed advance mailings and reminder e-mails and was distributed in electronic and paper formats. Reminders were sent to nonresponders at predetermined 1- to 2-week intervals, up to a maximum of 4 additional e-mails. To preserve anonymity while preventing multiple responses, each participant was assigned a study identification number.
This prospective, cross-sectional, 3-phase study was conducted as follows:
Phase 1: Caregiver Survey
After triage, the caregivers of consecutive, potentially eligible children presenting between 8:00 am and 11:00 pm were assessed for eligibility. Those meeting eligibility criteria and providing informed consent were invited to complete the survey. The final survey question was, “If you were given the choice, would you prefer that your child receive nasogastric or intravenous rehydration?” Subsequently, the caregiver was provided with a 1-page document outlining the advantages and disadvantages of nasogastric and intravenous rehydration (Supplemental Information 3), and then provided the opportunity to modify his or her answer to the final question.
Phase 2: Rehydration Data Collection
All treatment decisions (ie, oral, nasogastric, or intravenous rehydration) and the application of topical analgesics and other pain reduction modalities were unrelated to the study and at the discretion of the supervising physician. Participants administered intravenous or nasogastric rehydration had the procedure observed by a research assistant who documented the number of insertion attempts. Caregivers were asked to indicate their satisfaction with the insertion process on a 5-point Likert scale. Although agreement in pain scores assigned by children, parents, and practitioners is not perfect,18 given the importance of the topic, we asked parents to provide an assessment of their child’s pain during the procedure on a 100-mm Visual Analog Scale.19 The Visual Analog Scale allowed parents to indicate, along a 100-mm line, the intensity of their child’s pain with 0 mm representing “no pain” and 100 mm representing the “worst possible pain.”
Phase 3: Health Care Provider Survey
All full-time and part-time physicians, nurses, and pediatric emergency medicine fellows working in the ED that provided informed consent were eligible. As with the caregiver survey, the last question recorded the preferred method of treatment of children who fail to respond to ORT. Again, respondents were provided with a document outlining the pros and cons of nasogastric and intravenous rehydration, and the final question was repeated.
Sample Size and Data Analysis
Based on local expert opinion, health care providers were expected to be more aware of the reasons to use nasogastric rehydration, so we a priori estimated that in our hypothetical model, the proportions of responses in favor of nasogastric rehydration would be 15% for caregivers and 30% for health care providers. Assuming a 1:4 recruitment ratio for health care providers to caregivers, we estimated that responses from 96 health care providers and 384 caregivers would provide 90% power, with a type I error probability of .05 to reject the null hypothesis (PASS 2008, Version 08.0.2; NCSS Statistical Software, Kaysville, UT).
Survey data were entered into Microsoft Excel (Microsoft Corp, Redmond, WA), and then into SPSS, version 16.0 (SPSS, Inc, Chicago, IL). Responses were analyzed by using descriptive statistics to describe demographic variables and percentages to summarize categorical data. For the primary outcome, preference of rehydration method, the χ2 test was used. Other categorical response variables were tested by using the χ2 test or Fisher’s exact test, as appropriate. Continuous variables were compared by using the 2-sample t test. When paired responses were evaluated, McNemar’s test was used for categorical variables, the paired t test was used for continuous parametric variables, and the Wilcoxon signed rank test was used for nonparametric continuous variables. Subanalyses were used to explore the preferences based on previous intravenous rehydration episode, time of presentation (8:00 am–8:00 pm vs 8:00 pm–8:00 am), intravenous rehydration status at the current visit, and distance to hospital (>25 km = rural). Results were calculated based on the number of respondents to a particular question. A Bonferroni correction for multiple comparisons was applied to the analyses presented in each table.20 The Bonferroni correction divides the type I (α) error (.05) by the number of comparisons to yield a more conservative P value that is then used to assign statistical significance.
A total of 1103 children were screened; 435 were enrolled, 193 caregivers did not consent, and 475 were ineligible (317 had symptoms for ≥7 days, 141 did not meet vomiting or diarrhea frequency criteria, 12 had a language barrier, and 5 were too old or too young). Sixty-six percent (260 of 435) of participants were male, the mean age was 1.8 ± 1.0 years, 65% (283 of 435) had diarrhea, 90% (393 of 435) had vomiting, and 60% (263 of 435) had a history of fever. Forty-nine percent (211 of 435) had seen a physician during the current illness, and 26% (115 of 435) had caregivers who were instructed to bring their child for ED care.
Primary Outcome (Before Information Leaflet)
Given the choice, 10% (47 of 435) of the caregivers would choose nasogastric rehydration compared with 14% (16 of 113) of health care providers (difference = 3.4%; 95% confidence interval [CI]: −2.8% to 11.4%; P = .41). This finding was similar for those presenting during daytime (36 of 356; 10%) and evening (11 of 79; 14%) hours (P = .37); by intravenous fluid administration at the index visit (10 of 64; 16%, yes; 38 of 371; 10%, no; P = .20); and by distance to the hospital (19 of 135; 14% rural; 45 of 300; 15% urban; P = .80).
Caregiver Survey: Phase 1
Caregivers report being more familiar with the concept of intravenous rehydration, and they believe this route to be easier, less painful, and more effective than nasogastric rehydration (Table 1). Similar proportions of caregivers of children who have (61 of 174; 35%) and have not (80 of 261; 31%) previously received intravenous rehydration reported that intravenous insertion is easy (difference = 4.4%; 95% CI: −4.5% to 13.5%; P = .09). Forty-five percent (18 of 40) of the caregivers of children who previously experienced nasogastric tube insertion indicated that tube insertion is easy. This proportion was greater than the proportion among those who had never experienced nasogastric tube insertion, in which only 51 of 395 (12%) believed this task to be easy (difference = 32%; 95% CI: 17% to 48%; P < .001). Only 13 of 40 (33%) of the caregivers of children who previously had a nasogastric tube believe it is as effective as intravenous rehydration; by comparison, 50% (197 of 395) of those who never experienced nasogastric tube insertion believe it is as effective (difference = 17.4%; 95% CI: −0.1% to 31.7%; P = .06).
At baseline, 10% of caregivers favored nasogastric rather than intravenous rehydration. After reading the information leaflet (Supplemental Information Appendix 3), the proportion in favor increased to 27% (117 of 435; difference = 17%; 95% CI: 11% to 21%; P < .001). Caregivers of children with previous intravenous experience indicated they would opt for intravenous rehydration as often as caregivers without previous intravenous experience (155 of 174 [89%] vs 233 of 261 [89%]; difference = 0.2%; 95% CI: −5.6% to 6.6%; P = .95). After reviewing the information leaflet, fewer caregivers of children without previous intravenous experience chose intravenous rehydration: 79% (137 of 174) vs 69% (181 of 261) (difference = 9.4%; 95% CI: 0.9% to 17.4%; P = .03). Lastly, if ORT were to fail, the caregivers of children who previously experienced the insertion of a nasogastric tube were more likely to opt for nasogastric rehydration than those without a previous experience (9 of 40 [23%] vs 38 of 395 [10%]; difference = 12.9%; 95% CI: 2.2% to 28.1%; P = .01). After reading the information leaflet (Supplemental Information 3), the difference was no longer significant: 13 of 40 (33%) vs 104 of 395 (26%) (difference = 6.2%; 95% CI: −7.1% to 22.2%; P = .40).
Nasogastric/Intravenous Rehydration Data: Phase 2
Three hundred seventy-one children (85%) were successfully treated with ORT before discharge; 64 (15%) failed to improve and were administered intravenous rehydration. No child received nasogastric rehydration. The distance from home to the ED was similar between those administered intravenous fluids (24 ± 27 km) and those successfully treated with ORT (23 ± 21 km; P = .56). The number of intravenous insertion attempts was available for 52 children. The median number of attempts was 1 (interquartile range [IQR]: 1–2). Nine children required 2 attempts, 4 required 3 attempts, 2 required 4 attempts, and 1 required 8 attempts. The median caregiver-reported (N = 33) perception of pain during intravenous insertion was 70 out of 100 (IQR: 50–90), and median parental satisfaction with the insertion process was 1 (“very satisfied”; IQR: 1–1) on a 5-point Likert scale. Scores of 3 (neither satisfied nor dissatisfied) to 5 (very dissatisfied) were reported by 3 parents: those whose children required 2, 3, and 8 attempts.
Health Care Provider Survey: Phase 3
Survey responses were received from 57 physicians (response rate = 67%; 19 full-time, 22 part-time, 16 fellows); 2 declined, and 26 (25 part-time, 1 fellow) did not respond to the invitation (Tables 2 and 3). Fifty-six nurses (71%) completed the survey, and 23 did not respond. The reported rehydration methods used (oral, nasogastric, and intravenous) in children with moderate dehydration differed significantly among groups (Fig 1; P = .001). At baseline, 14% of health care providers favored nasogastric over intravenous rehydration. After reading the information leaflet (Supplemental Information 3), the number in favor of nasogastric rehydration increased to 43% (49 of 113) (difference = 29%; 95% CI: 18% to 40%; P < .001).
Nurses reported performing a significantly greater number of intravenous and nasogastric insertions than physicians; however, the former group performs a ratio of 22:1 intravenous-to-nasogastric insertions. Although both physicians and nurses estimated high success rates with nasogastric tube insertion (80%–90%), nurses estimate greater success with nurse-executed intravenous insertion (85%) compared with the physician group (60%–70%; P < .001). Compared with physicians, nurses are also more likely to estimate that intravenous insertion is “easy” (79% vs 32% ED staff vs 31% ED fellows; difference = 47%; 95% CI: 29% to 61%; P < .001), whereas estimates for ease of nasogastric insertion were similar (P = .12; Table 3). Although most physicians believe that nasogastric rehydration results in shorter ED length of stay and is less painful to insert, nurses do not share those beliefs (Table 4). Lastly, although our educational intervention altered some perceptions by physicians, a relative paucity of nurses were willing to recommend nasogastric rehydration after the intervention (Table 4).
Our results suggest that despite evidence supporting the use of nasogastric rehydration, it is rarely favored by parents or health care providers at our center. The proportions of health care providers and caregivers who would recommend and choose nasogastric rehydration were similar. Consequently, nasogastric rehydration was not administered to any children in our cohort, which may reflect the fact that nurses at our institution have become confident and skilled at intravenous insertion while being less familiar, experienced, and confident with nasogastric tube insertion. Given their skill set and the fact that nurses are the primary individuals responsible for intravenous catheter and nasogastric tube insertion in our institution, the use of intravenous rehydration has become deep-rooted. In such a scenario, it is logical that nurses estimate their success rate with intravenous insertion as very high and view the task as being “easy.” It appears that current practice begets current practice.
The choice of rehydration method is a regional one, with nasogastric rehydration being preferred in Australia and New Zealand,2 where a survey of ED physicians revealed that 82% use nasogastric rehydration in children with moderate dehydration. By comparison, only 14% of health care providers in our survey selected nasogastric rehydration. In actual practice, none of the 64 children who failed to respond to ORT were administered nasogastric rehydration. Our fellows were almost 4 times as likely to select nasogastric over intravenous rehydration (Fig 1), a finding that may reflect the geographical diversity of our fellows; however, health care providers in North America must be cognizant of the implications that our excessive use of intravenous rehydration may have on our trainees, particularly those from developing nations. Trainees returning to these countries and those working in small, remote hospitals in which expertise in intravenous insertion in dehydrated infants may be lacking must feel comfortable with ORT and nasogastric rehydration. Clinicians caring for children in North America must consider the impact our practice patterns might have on children around the world.21
The preference for intravenous treatment in our cohort was driven by both caregiver and health care provider comfort and familiarity. Although only a small number of children were ineligible owing to language barriers, it is possible that the multinational composition of the Toronto population may have impacted the results. Our caregiver preference for intravenous rehydration is in keeping with a study from the United States, however, in which the authors provided the options of oral or intravenous rehydration to parents,22 62% of whom, even after education, preferred intravenous rehydration. In our study, even the caregivers of children with previous intravenous experience predominantly opted for intravenous rehydration; they were also less influenced by our information leaflet. Their preference for intravenous rehydration likely reflects a high satisfaction rate with the intravenous rehydration process and a lack of familiarity with the use of nasogastric rehydration. Most dissatisfied caregivers witnessed their children undergo multiple intravenous insertion attempts. Perhaps such children could be targeted for an alternative rehydration method (eg, nasogastric) through the use of intravenous access scores such as the Difficult IntraVenous Access (DIVA) score.23 On the other hand, the lack of adoption of nasogastric rehydration by health care providers is likely related to the multitude of challenges involved in implementing knowledge translation.24 Moreover, although health care providers seem to underestimate the discomfort of nasogastric tube insertion,25,26 they overestimate its complication rate (eg, 3%–6% rate of intrapulmonary infusion). The reasons for clinicians’ choice of intravenous rehydration are likely multifactorial, which is most clearly evidenced by the failure to integrate nasogastric rehydration into our department despite our having a gastroenteritis pathway that includes the use of nasogastric rehydration in children in whom intravenous insertion is unsuccessful after 2 attempts.
Translating evidence to practice is not easy, and a variety of strategies targeting barriers to change at different levels is required.27 Although we were able to document changes in perception, changes in clinical practice are more challenging and only partly within doctors’ control, with prevailing professional and organizational cultures playing significant roles.28
Professional development needs to be incorporated into patient care, employing both clinical decision support tools and real-time, patient-specific reminders to help physicians make the best decisions. Additional barriers include the patient, the care process, available resources, and physician and nursing leadership.29 Intervention at the team level is required to develop and implement a transition from intravenous to nasogastric rehydration.
Our study has several limitations, the most significant of which is that the opinions expressed by clinicians are reflective of our local context and are not generalizable beyond North America. Our findings must be interpreted in the context of the study milieu, which lacks equipoise in the usage and experience with nasogastric and intravenous rehydration. In addition, our data describe a convenience sample of children-caregiver dyads presenting to a single ED, with caregivers bringing different experiences, resources, and preferences than those presenting to other health care facilities. As such, our findings should be limited to environments in which nasogastric tube rehydration is not the standard of care. Furthermore, the educational material was not designed to reflect a full knowledge translation approach; rather, it was a practical method of informing participants that nasogastric rehydration may be a reasonable alternative to intravenous rehydration. Although designed to be unbiased and understood by English-speaking families, the leaflet remains unvalidated.
In keeping with current perception, most health care providers in North American EDs are relatively unfamiliar with the use of nasogastric rehydration, so many prefer intravenous rehydration when ORT fails. This treatment choice is in keeping with caregiver desires. Future research should include a detailed study to better clarify barriers to nasogastric rehydration use in North America. Additionally, high-quality research, in a region where nasogastric tube insertion is commonplace, should assess the discomforts associated with this procedure. Given current knowledge and perceptions, extensive knowledge translation efforts are required for the nasogastric route to become a routine treatment option in North America.
The authors gratefully acknowledge the assistance of the Paediatric Research Academic Initiative in SickKids Emergency (PRAISE) volunteers with the conduct of the study. We also thank Johanna Crudden, Clinical Research Manager, for supervising the work of the PRAISE volunteers, Catherine Wong for her administrative assistance, Tuula Kalliomaki for preparing the required study materials, and Monica Monchis for assisting with data entry.
- Accepted July 25, 2012.
- Address correspondence to Stephen B. Freedman, MDCM, MSc, Division of Pediatric Emergency Medicine, Alberta Children's Hospital, 2888 Shaganappi Trail NW, Calgary, AB, Canada, T3B 6A8. E-mail:
Dr Freedman wrote the first draft of the article.
Dr Freedman's current affiliation is Division of Pediatric Emergency Medicine and Gastroenterology, Alberta Children's Hospital Research Institute, Alberta Children's Hospital, Calgary, Alberta, Canada.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
- Reis EC,
- Goepp JG,
- Katz S,
- Santosham M
- Ozuah PO,
- Avner JR,
- Stein REK
- Nager AL,
- Wang VJ
- Sharifi J,
- Ghavami F,
- Nowrouzi Z,
- et al
- Mackenzie A,
- Barnes G
- ↵World Health Organization. The Treatment of Diarrhoea: A Manual for Physicians and Other Senior Health Workers. 4th revision. Geneva, Switzerland: WHO Press; 2005
- Vesikari T,
- Van Damme P,
- Giaquinto C,
- et al.,
- Expert Working Group,
- European Society for Paediatric Infectious Diseases,
- European Society for Paediatric Gastroenterology, Hepatology, and Nutrition
- Burns KE,
- Duffett M,
- Kho ME,
- et al.,
- ACCADEMY Group
- Babl FE,
- Goldfinch C,
- Mandrawa C,
- Crellin D,
- O’Sullivan R,
- Donath S
- Copyright © 2012 by the American Academy of Pediatrics