‘Your Tube’: the role of different diets in children who are gastrostomy fed

BACKGROUND:

There are increasing numbers of children with complex health care needs that require having all, or part, of their nutritional intake via gastrostomy feeds. The recommended feed for children via gastrostomy is a commercially produced formula. However, there is a growing body of parents who are interested in and/or choosing to feed their children home-blended meals. These parents often report benefits such as improved gastro-oesophageal reflux symptoms, less constipation and less distress in their child.

The need for further research in this area has come from a research prioritisation exercise, recent review of the literature and professional organisations e.g. British Dietetic Association.

The ‘Your Tube’ study is a consented cohort study which has recruited 180 children (aged between 6 months – 18 years) who are fed via a gastrostomy tube and follow them up for an 18 month period. 15 of the 180 participants recruited have been consented via consultee advice. The study participants have been consented for transfer of personal data to NHS England for the purposes of linking with their healthcare data. NHS England data will be used to assess healthcare resource use and complications from the children's gastrostomy and to assess agreement between parent reported healthcare use and HES data. This is important to assess the feasibility of a longer-term follow-up study using only HES data.

Please note that for some of the cohort (15 children), the parents of the child have consented. For example, consent forms completed by the parent for children aged 7-11 years and 12-15 years. A patient information sheet was provided in lay person terms to all participants regardless of age. Additional information on the project can be found here - https://www.york.ac.uk/healthsciences/research/public-health/projects/yourtube/

COHORT

Children who are gastrostomy fed and recruited in the Your Tube study (180 children). For analyses purposes the cohort will be split into two groups:

1. those who are fed formula feeds

2. those who are fed a home-blended diet.

The University of York are relying on GDPR article 6 (1)(e) and 9 (2) (j):

Article 6(1)(e) – as per the Charter of Incorporation, power is conferred upon the University of York “to provide instruction in such branches of learning as the University may think fit and to make provision for research and for the advancement and dissemination of knowledge in such manner as the University may determine”.

Article 9(2)(j) – this research is in the public interest as there are increasing numbers of children with complex health care needs that require having all, or part, of their nutritional intake via gastrostomy feeds. The outcomes of this research into gastronomy feeding will be of significant application to parents, young people and health professionals – particularly its focus on the use of home-blended diets.

The main research question for this study is:

What are the risks, benefits and resource implications for using home-blended food for children with gastrostomy tubes compared to currently recommended formula feeds?

Other objectives are:

1. To identify the important outcomes of gastrostomy feeding for parents, young people and health professionals.

2. To assess the safety of home-blended diets for children who are gastrostomy fed compared to liquid formula diets.

3. To identify and quantify the benefits of home-blended diets compared to liquid formula diets for children who are gastrostomy fed and their parents.

4. To identify and quantify the resources (family and statutory services) required to support home-blended diets compared to liquid formula diets.

5. To assess whether long-term follow-up of children who are gastrostomy fed is feasible using routine data sources.

DATA REQUEST:

The linked Hospital Episode Statistics (HES) data will be used to:

- assess healthcare resource use and complications from the child's gastrostomy

- assess agreement between parent reported healthcare use and HES data. This is important to assess the feasibility of a longer-term follow-up study using only HES data.

Individual linkage to HES inpatient care, emergency and outpatient data for each child in this cohort (180 children) from 2 years prior to recruitment to the study and for the 12-18 months duration of the study is needed to fulfil the study aims. Some children will have a lot of hospital use whilst others will have little, hence the need for the 2 years prior to the start of the study.

HES Admitted Patient Care (APC) and Emergency Care Dataset (ECDS) data are required to assess any planned or unplanned hospital use in relation to their gastrostomy, complications from their gastrostomy or their underlying condition. Although parents are being asked to report any hospital admission for their children they are likely not to recall the detail of reason for admission and length of stay which are important for assessing healthcare resources use.

The minimum amount of data is being requested that is required to achieve the stated study aims. i.e. only asking for data on the 180 children recruited to this study and only for the 2 years prior to recruitment to the study and the 12-18 month duration that they are in the study.

The University of York is the sole data controller who will also process the data. There are co-investigators on this study from the University of Leeds, Sunderland Hospitals Trust and South Gloucestershire. The co-investigators will not have any access to the NHS England data nor have any role in data processing and as such are not considered data processors. The co-investigators are not determining the means and purpose of the processing of the data and as such are not listed as data controllers.

The NIHR Health Technology Assessment programme fund this study but have no role in data analyses, data processing or data controllership.

The study protocol is published and available here https://bmjopen.bmj.com/content/9/10/e033831.

DARS-NIC-657422-S1K1C is a sister application to DARS-NIC-334459-R9H4C. These applications have been split, as a subset of the cohort are recruited under consultee advice. As such, opt outs will be applied to the those recruited under consultee advice (under NIC-657422) and not to those recruited under explicit consent (under NIC-334459).

The record-level cohort data sent to NHS England from the University of York contained the following identifiers:

Date of birth,

NHS number,

Gender,

Postcode,

Study ID,

Recruitment date.

Data flow from NHS England to the University of York will be identifiable healthcare data with Study ID as the link to the 'Your Tube' cohort data. All other identifiers, barring Study ID will be removed. Although the data which NHS England will disseminate will not contain any identifiers, the University of York hold identifiers for the cohort and the data is therefore deemed identifiable.

Once received at the University of York, there will be no onward sharing of NHS England data. All data processing will happen at NHS England or the University of York by University of York employees who have undertaken University data protection training.

At the University of York, NHS England data will be linked to the clinical and demographic data collected from the cohort participants. University of York already have personal identifiable data on these cohort participants as this is a consented cohort. These data will not be linked with any other data sources.

DATA ACCESS:

Data will be stored securely on the Department of Health Sciences server. Access will either be via on site PC or via secure remote VPN from home which does not allow any download of data.

The drive is only accessible to devices on the Health Sciences subnets which are behind a separate firewall from the rest of the university, or to devices on the University's network that have been whitelisted specifically. People are accessing those devices remotely using the University's virtual private network (VPN) which requires multifactor authentication for all users. After accessing the VPN separate credentials are required to connect to Health Sciences.

An accurate assessment of the healthcare use and safety cannot be undertaken without access to the requested data from NHS England. The data derived from the HES data will be added to the data collected from the parents/child/clinician for analyses to assess the following:

1. SAFETY OUTCOMES

The data on safety outcomes will be provided via parent questionnaires and via NHS England data (see next section).

• No. hospital admissions

• No. accident and emergency attendances

• Tube blockages

• No. of infections

• No. of gastrointestinal infections

Appropriate statistical models will be used to compare these outcomes between the two groups of interest at baseline at the end of the study. All models will adjust for the propensity score and any additional confounding factors in this population (age, underlying diagnoses, co-morbidities, outpatient attendance, parental factors, socioeconomic status). Propensity scores are a statistical model used to estimate the probability of participants receiving treatment/intervention by taking into account their individual characteristics. Study site will be added as a random effect to the models to allow for site-level variation.

2. HEALTHCARE USE OF DATA

The number of hospital admissions (emergency and planned) and length of stay will be calculated for each child. A comparison between the two analytical groups of interest at baseline will be undertaken using appropriate statistical models.

After final data collection at 18 months the number of hospital admissions and length of stay will be compared between the two analytical groups of interest using appropriate statistical models.

Appropriate statistical models will account for confounding factors in this population (age, underlying diagnoses, co-morbidities, outpatient attendance, parental factors, socioeconomic status).

An analyses of accuracy of reporting between parent report and HES data will be undertaken to assess whether long term follow-up of this cohort using routine data would be feasible and accurate.

The data processing will only be carried out by substantive employees of the University of York only (who are the Data Controller who also process the data) and who have been trained in data protection and confidentiality.

The outputs from this study will include:

1. Final report for the funder National Institute for Health Research (NIHR) Health Technology Assessment. This will be published on the NIHR journals website – target date August 2023 (submit to change due to COVID-19).

2. An academic paper will be submitted to Developmental Medicine and Child Neurology - target date August 2023 (submit to change due to COVID-19)

3. An abstract for conference presentation at the European Association for Childhood Disability – target date July 2023

4. Short Plain English summary that will be shared with policy makers (NHS England) commissioners and parents via professional and parent networks - target date August 2023

5. A plain English summary for parents in collaboration with the parent advisory panel, which will be made available here - see here https://www.york.ac.uk/healthsciences/research/public-health/projects/martinhouse/mh-publications/.

All outputs will be published at aggregate level with small number suppression rules applied in line with the HES analysis guide.

All outputs will be disseminated via twitter @UoYMHRC, professional email lists and the University of York website.

The dissemination of these findings hope to inform national clinical guidelines including those of the British Dietetic Association.

At present there is little research evidence on which to base these guidelines, but more parents are feeding their children a home-blended diet. It is of public interest to have an evidence base on which to base clinical guidelines on the feeding of children with a gastrostomy.

Impact of children and families:

There are more than 10000 children in England who are currently fed via a gastrostomy. Understanding the safety and effectiveness of a home-blended diet may benefit children and their parents in terms of making an evidence-based decision about their child’s health. This may enable children and their family to have adequate support from their clinical team.

Impact of Healthcare professionals:

Understanding the role, safety and effectiveness of home-blended diet for these children is important in terms of supporting parents choices, ensuring that healthcare staff are aware of the study findings. These findings hope to enable clinical staff to support these children and families appropriately.

Impact on national guidelines:

At present there is little research on which to base clinical guidelines for home blended diet in children with gastrostomies. This study could provide evidence regarding safety that could enable these guidelines to be evidence based and robust.

Impact of NHS and commissioners:

This study could provide information on resource use and resource requirements for home-blended diets. This is important in terms of commissioning services for these children and families.

Data is still being analysed and yielded benefits have therefore not yet been realised,