Which Infants in a Neonatal Unit Are at Most Risk of Feeding Difficulties?

It is known that the incidence of feeding problems is high in neonatal populations and certain patient groups appear to be particularly at risk. It is not yet known which infant groups admitted to Level 3 Neonatal Units are at most risk of feeding problems and in need of specialist feeding assessment and interventions. While many studies have examined the impact of prematurity and additional medical problems on feeding outcomes, most studies to date are limited to sub-populations, and/or do not investigate how best to identify the babies who are at most risk of feeding difficulties. Neonatal Speech and Language Therapists require information about feeding risks and outcomes for the neonatal population as a whole for caseload prioritization and to target populations most in need of specialist feeding interventions.

Babies born extremely preterm (before 28 weeks gestational age) and very low birth weight (less than 1500 grams at birth) have been identified as slower to achieve full oral feeding than those born at an older gestational age and higher birth weight. Certain medical problems have been found to correlate with feeding outcomes in preterms. Impaired respiratory function, gastrointestinal complications, neurological involvement, and heart defects have all been associated with delayed feeding progression and poor feeding outcomes.

This study aims to determine if gestational age at birth or number and type of medical conditions is the better indicator of risk of feeding disorders in infants admitted to a Level 3 neonatal unit.

Design:

Prognostic study using retrospective case note analysis.

Setting:

A Level 3 Neonatal Unit (including intensive care, high dependency, and special care) at the Royal Preston Hospital.

Procedure:

A retrospective audit of routinely collected data will be undertaken and analysed. The BadgerNet patient record management system is updated daily for all babies in neonatal units around the country. Data will be collected from the BadgerNet daily notes and discharge summary and added to a bespoke spreadsheet prior to statistical analysis. Data collection will be conducted by a member of the Royal Preston Hospital Neonatal Unit clinical team. No patient identifiable data will be collected. Data will be listed on the spreadsheet in a random order to ensure there is no possibility of anyone, including the clinical researcher, tracing coded data back to the infant.

Data storage:

The spreadsheet of coded variables will be stored in a password protected folder accessible only to the clinical researcher. In order to move the spreadsheet to the University of Central Lancashire computer system for statistical analysis, it will be emailed in secure/encrypted form then saved on a restricted drive in a password protected folder that will only be accessible to the clinical researcher and statistician.

Information to be included on spreadsheet includes:

• Gestational age category (coded 1-4)
• Medical status - number of systems involved (coded 1-5) and types (coded 1-27)
• Full oral feeding by 37 weeks and 40 weeks (coded 1-3/1-4)

Coding:

Gestational age category:

1. Extremely preterm: <28 weeks
2. Very preterm: 28-31+6 weeks
3. Moderate to late preterm: 32-36+6 weeks
4. Term: 37 weeks+

Medical status - number of systems involved:

1. None
2. 1
3. 2
4. 3
5. 4 or more

Medical status - types and combinations:

1. No involvement in any of the specified systems
2. Neurological only
3. Respiratory only
4. Gastrointestinal only
5. Cardiac only
6. Craniofacial only
7. Neurological + respiratory
8. Neurological + gastrointestinal
9. Neurological + cardiac
10. Neurological + craniofacial
11. Respiratory + gastrointestinal
12. Respiratory + cardiac
13. Respiratory + craniofacial
14. Gastrointestinal + cardiac
15. Gastrointestinal + craniofacial
16. Cardiac + craniofacial
17. Neurological + respiratory + gastrointestinal
18. Neurological + respiratory + cardiac
19. Neurological + respiratory + craniofacial
20. Neurological + gastrointestinal + cardiac
21. Neurological + gastrointestinal + craniofacial
22. Neurological + cardiac + craniofacial
23. Respiratory + gastrointestinal + cardiac
24. Respiratory + gastrointestinal + craniofacial
25. Respiratory + cardiac + craniofacial
26. Gastrointestinal + cardiac + craniofacial
27. 4 or more systems

Slow attainment of full oral feeding (preterms only):

1. Yes
2. No
3. Information not available (exclude from analysis)

Feeding difficulty (all babies):

1. Yes - achieved oral feeding after 40 weeks but before discharge
2. Yes - discharged home tube feeding
3. No
4. Information not available (exclude from analysis)

Operational Definitions:

Gestational age groups are based on the categories used by the World Health Organisation:

• Extremely preterm: <28 weeks
• Very preterm: 28-31+6 weeks
• Moderate to late preterm: 32-36+6 weeks
• Term: 37 weeks+

Medical status will be categorized by the number of impaired body systems from the following list: neurological, respiratory, gastrointestinal, cardiac, and craniofacial. Following a thorough literature review, these five systems have been identified as the most commonly related to feeding difficulties.

Presence/absence of involvement of these systems will be determined by the patients discharge letter on BadgerNet. This letter is completed by a medical doctor and includes a section for each of the listed systems. If a problem is listed under any of these system headings it will be recorded as a positive, regardless of severity or duration. Exceptions to this include the following: retinopathy of prematurity, transient tachypnea of the newborn, intraventricular hemorrhage grades 1-2, and brachial plexus injury with no respiratory compromise.

Slow attainment of full oral feeding is defined as inability to achieve full oral feeding by:

• 37 weeks post-menstrual age (full term)
• Discharge home (if earlier than 37 weeks) or
• Within one week of birth if born at or over 36 weeks

Feeding difficulty is defined as inability to achieve full oral feeding by:

• 40 weeks post-menstrual age (due date)
• Discharge home if earlier than 40 weeks, or
• Within one week of birth if born at or over 39 weeks.

Analysis:

Descriptive statistics and appropriate statistical tests will be used to determine the presence or absence of correlation between variables. Outcomes with the strongest correlation will be identified and analysed using a logistic regression model.