Skip to main content
. 2019 Oct 9;9(10):e033831. doi: 10.1136/bmjopen-2019-033831

Table 2.

Proposed variables and data sources for WS2

Variable Type of variable Proposed measure Source Timings of data collection (months)
0 9 18
Participant characteristics/predictors
 Age Predictor Date of birth Parent
 Ethnicity Census groups Parent
 Deprivation Index of Multiple Deprivation (based on postcode) Parent
 Parental educational attainment Census groups Parent
 Household composition Number of children, marital/living status Parent
 Diagnosis Paediatrician
 Comorbidities Paediatrician
 All medications Paediatrician
 Complexity Disability Complexity Scale23 24 Paediatrician
 Length of time gastrostomy fed at T0 Months/years Parent
Comparator
 Diet: formula/blended Main grouping variable Parent
Outcomes
 Nutritional content of feeds Outcome Commercial for formula Dietitian/parent
myfood2415 Parent
 Anthropometric data Outcome Height or length
Weight
Triceps skinfold thickness or mid-arm circumference
Dietician
Paediatrician
 Gastrointestinal symptoms Outcome PedsQL gastrointestinal19 Parent/child
 Child quality of life Outcome PedsQL generic module18 Parent/child
 Parental quality of life Outcome EQ-5D-5L25
Parenting Morale Index26
Parent
 Healthcare use Outcome Client service receipt inventory27 Parent
Outcome Number of hospital admissions
Number of accident and emergency (A&E) attendances
HES data
 Safety Outcome Tube blockages
Number of stoma infections
Number of gastrointestinal infections
Parent
Outcome Number of hospital admissions and A&E attendance associated with child’s gastrostomy/diet HES and parent
 Family resource use Outcome Time preparing feeds; impact on other caring/parenting
Financial costs
Parent
 Non-staff NHS resource use Outcome Cost of formula and packaging
Dietetic resources
Dietitian

EQ-5D-5L, 5-level version of EuroQol-5 Dimension; HES, Hospital Episodes Statistics; NHS, National Health Service; PedsQL, Pediatric Quality of Life Inventory; WS2, workstream 2.