Skip to main content

Table 2 Characteristics of included studies (N = 64)

From: Characteristics and practices of school-based cluster randomised controlled trials for improving health outcomes in pupils in the United Kingdom: a methodological systematic review

Characteristic

N

Statistics

Setting

Country

64

 

                England, n (%)

 

47 (73)

                Scotland, n (%)

 

5 (8)

                Wales, n (%)

 

3 (5)

                Northern Ireland, n (%)

 

3 (5)

                More than one countrya, n (%)

 

6 (9)

Number of regions from which schools were drawnb

64

 

                One

 

40 (62)

                Two

 

10 (16)

                Three

 

1 (2)

                Four

 

1 (2)

                Unclear

 

12 (19)

School level

64

 

                Preschool only, n (%)

 

2 (3)

                Primary only, n (%)

 

36 (56)

                Secondary only, n (%)

 

24 (38)

                Primary and Secondary, n (%)

 

2 (3)

School types that were included [83]c

44

 

                State, n (%)

 

41 (93)

                Independent, n (%)

 

6 (14)

                Academies, n (%)

 

2 (5)

                Grammar, n (%)

 

2 (5)

                Special, n (%)

 

2 (5)

                Voluntary aided, n (%)

 

2 (5)

                Foundation, n (%)

 

1 (2)

                Faith, n (%)

 

1 (2)

Intervention

Health area of interventiond

64

 

                Nutrition, n (%)

 

18 (28)

                Physical activity, n (%)

 

15 (23)

                Socioemotional function and its influencese, n (%)

 

15 (23)

                Dental health, n (%)

 

7 (11)

                Smoking, n (%)

 

5 (8)

                Injury, n (%)

 

5 (8)

                Sexual health, n (%)

 

3 (5)

                Alcohol misuse, n (%)

 

2 (3)

                Cancer, n (%)

 

1 (2)

                Communication skills (for children with autism), n (%)

 

1 (2)

                Health attitudes (breast feeding), n (%)

 

1 (2)

Level of prevention

64

 

                Primary prevention, n (%)

 

60 (94)

                Secondary prevention, n (%)

 

4 (6)

Type of intervention [1]f

64

 

                Individual-cluster, n (%)

 

11 (17)

                Professional-cluster, n (%)

 

33 (52)

                External-cluster, n (%)

 

32 (50)

                Cluster–cluster, n (%)

 

53 (83)

                Multifaceted, n (%)

 

53 (83)

Primary outcome

Primary outcome health area

64

 

                Socioemotional function and its influencesg, n (%)

 

15 (23)

                Nutrition, n (%)

 

10 (16)

                Dental health, n (%)

 

7 (11)

                Physical activity, n (%)

 

7 (11)

                Obesity, n (%)

 

7 (11)

                Smoking, n (%)

 

5 (8)

                Injury, n (%)

 

3 (5)

                Sexual health, n (%)

 

2 (3)

                Obstetrics, n (%)

 

2 (3)

                Alcohol misuse, n (%)

 

2 (3)

Cancer, n (%)

 

1 (2)

                Communication skills (for children with autism), n (%)

 

1 (2)

                Gross motor skills, n (%)

 

1 (2)

                Safety, n (%)

 

1 (2)

Main reporter of primary outcome

64

 

                Pupil, n (%)

 

34 (53)

                Researcher, n (%)

 

12 (19)

                Dentist, n (%)

 

6 (9)

                Teacher, n (%)

 

5 (8)

                Parent, n (%)

 

4 (6)

                Routine data, n (%)

 

2 (3)

                Researcher and parent, n (%)

 

1 (2)

Primary outcome reporter blind to allocation status

64

 

                Yes, n (%)

 

18 (28)

                No, n (%)

 

46 (72)

Primary outcome measurement was objective

64

 

                Yes, n (%)

 

14 (22)

                No, n (%)

 

50 (78)

Study design and analysis methods

Justification provided for randomising clusters

64

 

                Yes, n (%)

 

17 (27)

                No, n (%)

 

47 (73)

Reason for randomising clusters

17

 

                To avoid contamination, n (%)

 

9 (53)

                Intervention was delivered at the cluster level, n (%)

 

4 (24)

                To avoid contamination and for logistical reasons, n (%)

 

2 (12)

                To avoid contamination and avoid “selection bias”, n (%)

 

1 (6)

                To avoid contamination and because intervention was delivered at the cluster level, n (%)

 

1 (6)

Unit of randomisation

64

 

                Schools, n (%)

 

56 (88)

                Classes, n (%)

 

6 (9)

                Year groups, n (%)

 

2 (3)

Number of trial arms

64

 

                Two, n (%)

 

55 (86)

                Three, n (%)

 

5 (8)

                Four, n (%)

 

4 (6)

Study design

64

 

                Parallel group, n (%)

 

61 (95)

                Factorial, n (%)

 

3 (5)

Method used to sample schools

64

 

                All potentially eligible schools invited, n (%)

 

33 (52)

                Random sample, n (%)

 

5 (8)

                Purposive sample, n (%)

 

4 (6)

                Convenience sample, n (%)

 

3 (5)

                Mixed random/convenience sample, n (%)

 

1 (2)

                Unclear, n (%)

 

18 (28)

Type of randomisation

64

 

                Completely randomised, n (%)

 

13 (20)

                Stratified, n (%)

 

29 (45)

                Matched, n (%)

 

8 (13)

                Minimisation, n (%)

 

8 (13)

                Constrained [84, 85], n (%)

 

6 (9)

Type of follow-up

64

 

                Cohort, n (%)

 

62 (97)

                Repeated cross-sectional, n (%)

 

1 (2)

                Mixed, n (%)

 

1 (2)

Number of follow-ups

64

 

                1, n (%)

 

32 (50)

                2, n (%)

 

21 (33)

                3, n (%)

 

6 (9)

                4, n (%)

 

5 (8)

Length of follow-up

64

 

                Up to 6 months, n (%)

 

22 (34)

                7 to 12 months, n (%)

 

19 (30)

                13 to 18 months, n (%)

 

6 (9)

                19 to 24 months, n (%)

 

8 (13)

                25 to 36 months, n (%)

 

7 (11)

                More than 36 months, n (%)

 

2 (3)

Participants recruited before clusters were randomised

64

 

                Yes, n (%)

 

21 (33)

                No, n (%)

 

17 (27)

                Unclear, n (%)

 

26 (41)

Baseline data collected before clusters were randomised

64

 

                Yes, n (%)

 

16 (25)

                No, n (%)

 

27 (42)

                Unclear, n (%)

 

21 (33)

Method of analysis

64

 

                Individual-level analysis that allows for clustering, n (%)

 

46 (72)

                Cluster-level analysis, n (%)

 

10 (16)

                Did not allow for clustering, n (%)

 

8 (12)

Sample size calculation

  

Assumed school-level intra-cluster correlation coefficient of outcome, median (IQR; range)

37

0.05 (0.02 to 0.1;

0.005 to 0.175)

Assumed design effect, median (IQR; range)

36

2.21 (1.98 to 3.53;

1.22 to 8.11)

Study allowed for drop-out at cluster level

64

 

                Yes, n (%)

 

4 (6)

                Not stated, n (%)

 

60 (94)

Study allowed for drop-out at individual levelh

62

 

                Yes, n (%)

 

18 (29)

                Not stated, n (%)

 

44 (71)

Target number of clusters, median (IQR; range)

46

30 (20 to 40;

4 to 160)

Target number of schools, median (IQR; range)

41

30 (20 to 42;

4 to 160)

Target number of individuals, median (IQR; range)i

45

964 (498 to 2000;

90 to 9000)

Ethics and consent procedures

From whom was consent/assent sought for pupil participation?

64

 

                Parents and pupils, n (%)

 

40 (63)

                Parents only, n (%)

 

15 (23)

                Pupils only, n (%)

 

2 (3)

                Not stated / Neither parent nor pupil, n (%)

 

7 (11)

Opt-out consent/assent procedure used for either parent/guardian or pupils

64

 

                Yes, n (%)

 

29 (45)

                Not stated / No, n (%)

 

35 (55)

Other study characteristics of methodological interest

Ethnicity: percentage of pupils that are White, median (IQR; range)

33

76.8 (51.5 to 86.2; 24 to 95.3)

Total number of clusters recruited, median (IQR; range)

62

31.5 (21 to 50;

4 to 486)

Total number of schools recruited, median (IQR; range)

63

29 (15 to 50;

4 to 486)

Total number of pupils recruited, median (IQR; range)j

60

1308 (604 to 3201;

17 to 27,435)

Percentage of clusters followed-up for primary outcome, median (IQR; range)

62

100 (92.5 to 100;

60.5 to 100)

Percentage of pupils followed-up for primary outcome, median (IQR; range)k

58

79.9 (64.1 to 87.5;

7.7 to 100)

Observed school-level intra-cluster correlation coefficient of primary outcome, median (IQR; range)

23

0.028 (0.017 to 0.12;

0.0005 to 0.21)

  1. aStudies that included schools from more than one country in the United Kingdom
  2. bEnglish regions included: South West, South East (including Greater London), East of England, West Midlands, East Midlands, North West, North East, Yorkshire and The Humber, “Southern England”, “Central England” and “West of England”. Scottish regions included: Glasgow, Inverclyde, Tayside, Grampian, Lanarkshire, Lothian and Fife. Welsh regions included: North Wales, South West Wales and South East Wales. Northern Irish regions included: South Belfast, East Belfast, Ulster, Leinster, Connacht and Munster
  3. cSome studies included more than one school type. This is the number of studies that included specific types of school. State schools receive funding through their local authority or directly from the government. The most common ones are local authority, foundation and voluntary aided school which are all funded by the local authority. Academies are run by government and not-for-profit trusts, and are independent of local authority. Grammar schools are run by local authorities but intake is based on assessment of the pupils’ academic ability. Special schools cater for pupils with special educational needs. Faith schools follow the national curriculum but can decide what they teach in religious studies. Independent schools follow the national curriculum but charge fees for attending pupils
  4. dSome interventions targeted more than one health area
  5. eIncludes mental health, behaviour, ADHD, wellbeing, quality of life, bullying, social and emotional learning, and self-esteem
  6. fIntervention type was summarised based on the typology described by Eldridge and colleagues [1]. ‘Individual-cluster’ interventions include components that are directed at individual participants (e.g. pupils) on whom outcomes are measured. ‘Professional-cluster’ interventions include components for training professionals in the cluster (e.g. teachers in schools) to deliver the intervention. ‘External-cluster’ interventions involve additional staff outside the cluster to deliver the intervention (e.g. researchers, trained facilitators). ‘Cluster–cluster’ interventions include components that necessarily have to be administered to entire clusters (e.g., school policy). ‘Multifaceted’ interventions include components across more than one of the ‘individual-cluster’, ‘professional-cluster’, ‘external-cluster’ and ‘cluster–cluster’ categories
  7. gIncludes mental health, behaviour, hyperactivity/inattention (ADHD), wellbeing, quality of life, bullying, social and emotional learning, and self-esteem (body image)
  8. hSummary excludes the two CRTs that did not use the cohort design
  9. iSummary excludes the two CRTs that did not use the cohort design
  10. jSummary excludes the two CRTs that did not use the cohort design
  11. kSummary excludes the two CRTs that did not use the cohort design
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us