Results tables from the main trial

Elizabeth Goyder; Daniel Hind; Jeff Breckon; Munyaradzi Dimairo; Jonathan Minton; Emma Everson-Hock; Simon Read; Robert Copeland; Helen Crank; Kimberly Horspool; Liam Humphreys; Andrew Hutchison; Sue Kesterton; Nicolas Latimer; Emma Scott; Peter Swaile; Stephen J Walters; Rebecca Wood; Karen Collins; Cindy Cooper

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Goyder E, Hind D, Breckon J, et al. A randomised controlled trial and cost-effectiveness evaluation of ‘booster’ interventions to sustain increases in physical activity in middle-aged adults in deprived urban neighbourhoods. Southampton (UK): NIHR Journals Library; 2014 Feb. (Health Technology Assessment, No. 18.13.)

Cover of A randomised controlled trial and cost-effectiveness evaluation of ‘booster’ interventions to sustain increases in physical activity in middle-aged adults in deprived urban neighbourhoods

A randomised controlled trial and cost-effectiveness evaluation of ‘booster’ interventions to sustain increases in physical activity in middle-aged adults in deprived urban neighbourhoods.

Show details

Contents

< Prev Next >

Appendix 6Results tables from the main trial

TABLE 16

Mail-out areas

Mail-out	Neighbourhoods	Letters sent, n	Replies received, n (%)	Randomised, n (%)
1	High Green	3300	329 (10.0)	48 (1.5)
2	Southey Green, Longley, New Parsons Cross, Old Parsons Cross, Shirecliffe, Tinsley, Darnall, Flower, Shiregreen, Stubbin, Brushes, Acres Hill, Winn Gardens	15,366	1045 (6.8)	88 (0.6)
3	Gleadless Valley, Hemsworth, Lowedges, Batemoor, Jordanthorpe, Manor, Woodthorpe, Park Hill, Norfolk Park, Arbourthorne, Tinsley, Darnall, Acres Hill	18,784	974 (5.2)	60 (0.3)
4	Broomhall, Sharrow (including the Abbeydale corridor), Highfield, Burngreave, Abbeyfield, Fir Vale, Firshill, Woodside, Netherthorpe, Upperthorpe, Langsett, Wybourn, City Centre	16,072	1185 (7.4)	52 (0.3)
5	Brightside, Firth Park, Fox Hill, Richmond, Wincobank	6862	385 (5.6)	8 (0.1)
6	Base Green, Beighton, Charnock, Gleadless, Halfway, Handsworth, Hollins End, Mosborough, Owlthorpe, Sothall	10,000	836 (8.4)	26 (0.3)

TABLE 17

Baseline characteristics of all randomised study participants (n = 282)

Variable	Control (n = 96)	Booster (n = 186)	Total (n = 282)
Gender, n (%)
Male	37 (38.5)	93 (50.0)	130 (46.1)
Female	59 (61.5)	93 (50.0)	152 (53.9)
Employment status, n (%)
Part-time	18 (18.8)	34 (18.3)	52 (18.4)
Full-time	34 (35.4)	59 (31.7)	93 (33.0)
Not employed	44 (45.8)	90 (48.4)	134 (47.5)
Missing	0 (0.0)	3 (1.6)	3 (1.1)
Ethnicity, n (%)
White British	83 (86.5)	163 (87.6)	246 (87.2)
Other	13 (13.5)	20 (10.8)	33 (11.7)
Missing	0 (0.0)	3 (1.6)	3 (1.1)
Marital status, n (%)
Single	16 (16.7)	29 (15.6)	45 (16.0)
Married	50 (52.1)	101 (54.3)	151 (53.5)
Co-habiting	5 (5.2)	15 (8.1)	20 (7.1)
Divorced/separated	20 (20.8)	35 (18.8)	55 (19.5)
Widowed	5 (5.2)	6 (3.2)	11 (3.9)
Stage of change, n (%)
Contemplation	1 (1.0)	11 (5.9)	12 (4.3)
Preparation	39 (40.6)	86 (46.2)	125 (44.3)
Action	36 (37.5)	55 (29.6)	91 (32.3)
Maintenance	18 (18.8)	32 (17.2)	50 (17.7)
Missing	2 (2.1)	2 (1.1)	4 (1.4)
Age (years)
n (%)	96 (100.0)	186 (100.0)	282 (100.0)
Mean (SD)	54.5 (6.8)	54.6 (7.6)	54.6 (7.3)
Median (IQR)	54.9 (50.3 to 60.0)	55.6 (48.1 to 61.8)	55.3 (48.8 to 61.4)
Min. to max.	40.5 to 65.5	40.4 to 65.1	40.4 to 65.5
Height (m)
n (%)	95 (99.0)	186 (100.0)	281 (99.6)
Mean (SD)	1.7 (0.1)	1.7 (0.1)	1.7 (0.1)
Median (IQR)	1.7 (1.6–1.7)	1.7 (1.6–1.8)	1.7 (1.6–1.8)
Min. to max.	1.5 to 1.9	1.5 to 1.9	1.5 to 1.9
Weight (kg)
n (%)	96 (100.0)	186 (100.0)	282 (100.0)
Mean (SD)	84.5 (19.2)	85.5 (18.5)	85.2 (18.7)
Median (IQR)	82.8 (70.1 to 95.0)	83.0 (73.0 to 96.8)	82.9 (72.5 to 96.6)
Min. to max.	50.8 to 160.0	46.9 to 145.8	46.9 to 160.0
BMI (kg/m²)
n (%)	95 (99.0)	186 (100.0)	281 (99.6)
Mean (SD)	30.3 (6.3)	30.3 (5.7)	30.3 (5.9)
Median (IQR)	28.9 (25.7 to 33.0)	29.8 (26.4 to33.2)	29.8 (26.3 to 33.0)
Min. to max.	20.3 to 53.4	17.1 to 49.4	17.1 to 53.4
SPAQ change^a (3 months post randomisation)
n (%)	96 (100.0)	186 (100.0)	282 (100.0)
Mean (SD)	199.8 (261.6)	215.3 (416.1)	210.0 (370.4)
Median (IQR)	120.0 (75.0 to 255.0)	120.0 (50.0 to 255.0)	120.0 (60.0 to 255.0)
Min. to max.	−1010.0 to 1240.0	−1840.0 to 3360.0	−1840.0 to 3360.0
SF-12v2 plus 4 (PCS)
n (%)	90 (93.8)	180 (96.8)	270 (95.7)
Mean (SD)	46.8 (11.2)	46.2 (10.6)	46.4 (10.8)
Median (IQR)	51.8 (39.9 to 54.8)	49.1 (39.6 to 53.9)	49.9 (39.7 to 54.5)
Min. to max.	13.9 to 60.7	13.0 to 66.7	13.0 to 66.7
SF-12v2 plus 4 (MCS)
n (%)	90 (93.8)	180 (96.8)	270 (95.7)
Mean (SD)	47.8 (10.2)	49.2 (9.6)	48.7 (9.8)
Median (IQR)	50.4 (41.7 to 54.9)	51.0 (41.9 to 57.1)	50.8 (41.9 to 56.2)
Min. to max.	20.8 to 65.8	15.9 to 68.5	15.9 to 68.5
BREQ-2 (RAI)
n (%)	93 (96.9)	181 (97.3)	274 (97.2)
Mean (SD)	5.4 (3.7)	5.2 (3.7)	5.3 (3.7)
Median (IQR)	5.8 (3.3 to 7.9)	6.3 (3.3 to 8.0)	6.0 (3.3 to 8.0)
Min. to max.	−3.3 to 12.0	−9.1 to 11.5	−9.1 to 12.0

: max., maximum; min., minimum.
a: Twenty-eight participants reported an increase in physical activity of < 30 minutes at pretrial screening (23 reported atypical activity either at brief intervention or pretrial screening; five reported typical activity on both occasions and a change in activity less than the required amount of at least 30 minutes).

TABLE 18

Baseline characteristics of ITT study participants (n = 160)

Variable	Control (n = 61)	Booster (n = 99)	Total (n = 160)
Gender, n (%)
Male	26 (42.6)	42 (42.4)	68 (42.5)
Female	35 (57.4)	57 (57.6)	92 (57.5)
Employment status, n (%)
Part-time	14 (23.0)	19 (19.2)	33 (20.6)
Not employed	27 (44.3)	55 (55.6)	82 (51.3)
Ethnicity, n (%)
White British	54 (88.5)	89 (89.9)	143 (89.4)
Any other	7 (11.5)	8 (8.1)	15 (19.4)
Missing	0 (0.0)	2 (2.0)	2 (1.3)
Marital status, n (%)
Single	11 (18.0)	13 (13.1)	24 (15.0)
Married	30 (49.2)	55 (55.6)	85 (53.1)
Co-habiting	5 (8.2)	6 (6.1)	11 (6.9)
Divorced/separated	11 (18.0)	20 (20.2)	31 (19.4)
Widowed	4 (6.6)	5 (5.1)	9 (5.6)
Stage of change, n (%)
Contemplation	0 (0.0)	5 (5.1)	5 (3.1)
Preparation	24 (39.3)	40 (40.4)	64 (40.0)
Action	23 (37.7)	34 (34.3)	57 (35.6)
Maintenance	13 (21.3)	19 (19.2)	32 (20.0)
Missing	1 (1.6)	1 (1.0)	2 (1.3)
Age (years)
n (%)	61 (100.0)	99 (100.0)	160 (100.0)
Mean (SD)	54.3 (7.0)	55.3 (7.7)	54.9 (7.4)
Median (IQR)	54.0 (50.4 to 60.6)	57.0 (48.5 to 62.3)	56.0 (49.4 to 61.8)
Min. to max.	40.5 to 65.5	40.5 to 65.1	40.5 to 65.5
Height (m)
n (%)	61 (100.0)	99 (100.0)	160 (100.0)
Mean (SD)	1.7 (0.1)	1.7 (0.1)	1.7 (0.1)
Median (IQR)	1.7 (1.6 to 1.7)	1.6 (1.6 to 1.7)	1.7 (1.6 to 1.7)
Min. to max.	1.5 to 1.9	1.5 to 1.9	1.5 to 1.9
Weight (kg)
n (%)	61 (100.0)	99 (100.0)	160 (100.0)
Mean (SD)	81.7 (15.1)	82.1 (17.9)	82.0 (16.9)
Median (IQR)	82.0 (70.1 to 92.0)	80.6 (70.1 to 91.1)	80.6 (70.1 to 91.5)
Min. to max.	50.8 to 115.0	47.1 to 142.2	47.1 to 142.2
BMI (kg/m²)
n (%)	61 (100.0)	99 (100.0)	160 (100.0)
Mean (SD)	29.2 (5.0)	29.6 (5.7)	29.4 (5.4)
Median (IQR)	27.9 (25.6 to 32.2)	29.2 (25.5 to 32.2)	28.9 (25.6 to 32.2)
Min. to max.	20.3 to 43.7	17.1 to 49.4	17.1 to 49.4
SPAQ change (3 months post randomisation)
n (%)	61 (100.0)	99 (100.0)	160 (100.0)
Mean (SD)	216.9 (243.1)	188.6 (316.4)	199.4 (290.2)
Median (IQR)	120.0 (90.0 to 255.0)	110.0 (45.0 to 225.0)	120.0 (60.0 to 237.5)
Min. to max.	−210.0 to 1240.0	−480.0 to 1740.0	−480.0 to 1740.0
SF-12v2 plus 4 (PCS)
n (%)	58 (95.1)	95 (96.0)	153 (95.6)
Mean (SD)	48.4 (9.3)	47.3 (9.8)	47.7 (9.6)
Median (IQR)	52.0 (42.0 to 54.8)	49.9 (42.8 to 54.3)	50.7 (42.6 to 54.5)
Min. to max.	17.5 to 60.6	20.7 to 60.9	17.5 to 60.9
SF-12v2 plus 4 (MCS)
n (%)	58 (95.1)	95 (96.0)	153 (95.6)
Mean (SD)	47.7 (9.8)	50.0 (8.7)	49.1 (9.2)
Median (IQR)	50.1 (41.1 to 54.9)	51.2 (44.2 to 57.1)	51.0 (43.2 to 56.3)
Min. to max.	20.8 to 63.8	21.7 to 65.5	20.8 to 65.5
BREQ-2 (RAI)
n (%)	59 (96.7)	96 (97.0)	155 (96.9)
Mean (SD)	5.6 (3.6)	5.6 (3.6)	5.6 (3.6)
Median (IQR)	6.1 (3.4 to 8.0)	6.4 (3.7 to 8.3)	6.3 (3.5 to 8.3)
Min. to max.	−3.3 to 12.0	−9.1 to 11.5	−9.1 to 12.0

: max., maximum; min., minimum.
: Note: the ITT set is defined as randomised participants with at least 4 complete days on the Actiheart device as measured by lost minutes during the day of not more than 1000 minutes at 3 months of follow-up.

TABLE 19

Comparability of baseline characteristics of completers and non-completers

Variable	Non-completers			Completers
Variable	Control	Booster	All	Control	Booster	All
Age (years)
n	35	87	122	61	99	160
Mean (SD)	55.4 (6.6)	54.3 (7.4)	54.6 (7.2)	54.7 (7.0)	55.7 (7.7)	55.3 (7.4)
Median (IQR)	58 (50.5 to 60.3)	52.9 (48.4 to 61.9)	54.6 (48.9 to 60.7)	54.7 (50.7 to 61.0)	57.8 (49.0 to 62.7)	56.4 (49.8 to 62.3)
Height (m)
n	34	87	121	61	99	160
Mean (SD)	1.7 (0.1)	1.7 (0.1)	1.7 (0.1)	1.7 (0.1)	1.7 (0.1)	1.7 (0.1)
Median (IQR)	1.6 (1.6 to 1.7)	1.7 (1.6 to 1.8)	1.7 (1.6 to 1.8)	1.7 (1.6 to 1.7)	1.6 (1.6 to 1.7)	1.7 (1.6 to 1.7)
Weight (kg)
n	35	87	122	61	99	160
Mean (SD)	89.4 (24.1)	89.4 (18.4)	89.4 (20.1)	81.7 (15.1)	82.1 (17.9)	82.0 (16.9)
Median (IQR)	82.8 (69.0 to 105.0)	87.7 (76.0 to 99.5)	87.5 (75.9 to 101.9)	82 (70.1 to 92.0)	80.6 (70.1 to 91.1)	80.6 (70.1 to 91.5)
BMI (kg/m²)
n	34	87	121	61	99	160
Mean (SD)	32.2 (7.8)	31.2 (5.7)	31.5 (6.4)	29.2 (5.0)	29.6 (5.7)	29.4 (5.4)
Median (IQR)	30 (26.3 to 37.9)	30.9 (26.9 to 34.0)	30.9 (26.9 to 34.4)	27.9 (25.6 to 32.2)	29.2 (25.5 to 32.2)	28.9 (25.6 to 32.2)
SPAQ change (3 months post randomisation)
n	35	87	122	61	99	160
Mean (SD)	170.0 (292.2)	245.7 (506.6)	224.0 (455.6)	216.9 (243.1)	188.6 (316.4)	199.4 (290.2)
Median (IQR)	110 (60.0 to 270.0)	120 (60.0 to 300.0)	120 (60.0 to 275.0)	120 (90.0 to 255.0)	110 (45.0 to 225.0)	120 (60.0 to 237.5)
SF-12v2 plus 4 (PCS)
n	32	85	117	58	95	153
Mean (SD)	43.9 (13.6)	45.0 (11.4)	44.7 (12.0)	48.4 (9.3)	47.3 (9.8)	47.7 (9.6)
Median (IQR)	51.4 (32.1 to 54.2)	48.0 (37.3 to 53.6)	48.7 (35.8 to 53.9)	52.0 (42.0 to 54.8)	49.9 (42.8 to 54.3)	50.7 (42.6 to 54.5)
SF-12v2 plus 4 (MCS)
n	32	85	117	58	95	153
Mean (SD)	48.1 (11.1)	48.3 (10.5)	48.3 (10.6)	47.7 (9.8)	50.0 (8.7)	49.1 (9.2)
Median (IQR)	51.2 (43.1 to 55.5)	49.8 (41.2 to 56.2)	49.9 (41.2 to 56.2)	50.1 (41.1 to 54.9)	51.2 (44.2 to 57.1)	51.0 (43.2 to 56.3)
BREQ-2 (RAI)
n	34	85	119	59	96	155
Mean (SD)	5.0 (3.9)	4.8 (3.9)	4.8 (3.9)	5.6 (3.6)	5.6 (3.6)	5.6 (3.6)
Median (IQR)	5.5 (2.5 to 7.4)	5.5 (2.0 to 7.8)	5.5 (2.2 to 7.8)	6.1 (3.4 to 8.0)	6.4 (3.7 to 8.3)	6.3 (3.5 to 8.3)
Gender
n	35	87	122	61	99	160
Male, n (%)	11 (31.4)	51 (58.6)	62 (50.8)	26 (42.6)	42 (42.4)	68 (42.5)
Female, n (%)	24 (68.6)	36 (41.4)	60 (49.2)	35 (57.4)	57 (57.6)	92 (57.5)
Ethnicity, n (%)
White British	29 (82.9)	74 (85.1)	103 (84.4)	54 (88.5)	89 (89.9)	143 (89.4)
Other	6 (17.1)	12 (13.8)	18 (14.8)	7 (11.5)	8 (8.1)	15 (9.4)
Missing	0 (0.0)	1 (1.1)	1 (0.8	0 (0.0)	2 (2.0)	2 (1.3)
Employment status, n (%)
Part-time	4 (11.4)	15 (17.2)	19 (15.6)	14 (23.0)	19 (19.2)	33 (20.6)
Full-time	14 (40.0)	34 (39.1)	48 (39.3)	20 (32.8)	25 (25.3)	45 (28.1)
Not employed	17 (48.6)	35 (40.2)	52 (42.6)	27 (44.3)	55 (55.6)	82 (51.2)
Missing	0 (0.0)	3 (3.4)	3 (2.5)	0 (0.0)	0 (0.0)	0 (0.0)

: Note: completers and non-completers are those with and without evaluable data at 3 months for any reason respectively.

TABLE 20

Exploratory analysis of the risk factors (categorical) associated with completing evaluable data at 3 months

Variable	Completers (n = 160), n (%)	Non-completers (n = 122), n (%)	OR (95% CI)^a	p-value	Interaction
Variable	Completers (n = 160), n (%)	Non-completers (n = 122), n (%)	OR (95% CI)^a	p-value	OR (95% CI)^b	p-value
Gender
Female	92 (57.5)	60 (49.2)	1.40 (0.87 to 2.24)	0.165	0.3 (0.1 to 0.9)	0.034
Male	68 (42.5)	62 (50.8)	1.00		0.3 (0.1 to 0.9)	0.034
Treatment group
Booster	99 (61.9)	87 (71.3)	0.65 (0.39 to 1.08)	0.098
Control	61 (38.1)	35 (28.7)	1.00		NA
Treatment group
Full	52 (33)	42 (34)	0.71 (0.40 to 1.27)	0.249
Mini	47 (29)	45 (37)	0.60 (0.33 to 1.07)	0.085	NA
Control	61 (38)	25 (27)	1.00
Employment status
Yes	78 (48.8)	70 (57.4)	0.71 (0.44 to 1.14)	0.151	0.5 (0.2 to 1.3)	0.127
No	82 (51.3)	52 (42.3)	1.00
Ethnicity^c
n	158	121
White British	143 (90.5)	103 (85.1)	1.67 (0.80 to 3.46)	0.171	NA
Any other	15 (9.5)	18 (14.9)	1.00

: NA, not applicable.
a: The odds of completing a 3-months assessment with evaluable data compared with not completing a 3-months assessment with evaluable data for any other reasons using logistic regression.
b: Interaction effect between treatment (booster and control) and completers group (completer or non-completer) using logistic regression.
c: Excludes three participants with missing ethnicity data at baseline (no interaction test because of small sample within subgroups).

TABLE 21

Exploratory analysis of the risk factors (continuous) associated with completing evaluable data at 3 months

Variable	Non-completers		Completers		Mean difference (95% CI)^a	p-value^a	Interaction
Variable	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^a	Effect (95% CI)^b	p-value^b
Age (years)	122	54.6 (7.2)	160	55.3 (7.4)	0.7 (−1.0 to 2.4)	0.426	2.1 (−1.6 to 5.8)	0.267
BMI (kg/m²)	121	31.5 (6.4)	160	29.4 (5.4)	−2.0 (−3.4 to −0.6)	0.005	1.5 (−1.5 to 4.5)	0.320
Weight (kg)	122	89.4 (20.1)	160	82.0 (16.9)	−7.4 (−11.7 to −3.1)	0.001	0.5 (−8.9 to 9.8)	0.924
SF-12v2 plus 4 (PCS)	117	44.7 (12.0)	153	47.7 (9.6)	3.1 (0.5 to 5.7)	0.019	−2.2 (−7.8 to 3.5)	0.452
SF-12v2 plus 4 (MCS)	117	48.3 (10.6)	153	49.1 (9.2)	0.8 (−1.6 to 3.2)	0.508	2.1 (−3.1 to 7.2)	0.430
Height (m)	121	1.7 (0.1)	160	1.7 (0.1)	−0.0 (−0.0 to 0.0)	1.000	−0.0 (−0.1 to 0.0)	0.080
SPAQ change^c	122	199.8 (261.6)	160	215.3 (416.1)	−15.5 (−95.3 to 64.3)	0.703	−104.0 (−292.6 to 84.6)	0.279
BREQ-2 (RAI)	119	4.8 (3.9)	155	5.6 (3.6)	0.8 (−0.1 to 1.7)	0.077	0.2 (−1.7 to 2.1)	0.843

a: Using a two independent sample t-test.
b: Interaction effect between treatment (control or booster) and completers group (completer or non-completer) using logistic regression.
c: Using independent two-sample t-test with unequal variances.

TABLE 22

Difference in mean TEE per day between the booster intervention group (mini plus full) and the control group at 3 months (n = 160)

Outcome	Control (n = 61), mean (SD)	Booster (n = 99), mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Mean TEE per day (kcal)	2265.9 (410.8)	2226.9 (422.6)	−39.0 (−173.4 to 95.4)	0.567	−40.3 (−117.2 to 36.6)	0.302

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
: Note: a negative mean difference favours the control over the booster interventions and indicates decreased physical activity in the booster intervention group compared with the control group.

TABLE 23

Sensitivity analysis: difference in mean TEE per day between the booster intervention group (mini plus full) and the control group at 3 months

Mean TEE per day (kcal)	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Multiple imputation (≥ 4 days)	(n = 61)	(n = 99)
Multiple imputation (≥ 4 days)	2252.3 (393.6)	2215.1 (366.0)	−37.2 (−174.9 to 100.6)	0.595	−34.5 (−118.1 to 49.0)	0.415
Regression imputation (≥ 4 days)	(n = 58)	(n = 95)
Regression imputation (≥ 4 days)	2263.1 (398.6)	2245.2 (409.4)	−18.0 (−151.5 to 115.5)	0.790	−27.0 (−97.3 to 43.3)	0.449
Complete cases (7 complete days)	(n = 44)	(n = 68)
Complete cases (7 complete days)	2241.7 (393.6)	2222.5 (366.0)	−19.2 (−163.8 to 125.4)	0.793	−31.2 (−114.3 to 51.9)	0.458
Per protocol^d	(n = 61)	(n = 96)
Per protocol^d	2265.9 (410.8)	2231.3 (426.6)	−34.6 (−170.6 to 101.4)	0.616	−35.6 (−112.5 to 41.3)	0.362
Multiple imputation (≥ 1 day)	(n = 70)	(n = 113)
Multiple imputation (≥ 1 day)	2194.0 (393.6)	2215.3 (366.0)	21.3 (−133.1 to 175.7)	0.786	9.9 (−111.3 to 131.1)	0.872
Mean TEE per day (kcal)^e	(n = 61)	(n = 98)
Mean TEE per day (kcal)^e	2265.9 (410.8)	2211.4 (395.5)	−54.5 (−183.8 to 74.8)	0.406	−43.6 (−119.9 to 32.7)	0.260

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: ‘Per protocol’ defined as received intended intervention at 1 and 2 months post brief intervention.
e: Excluding an outlier (only for sensitivity analysis).
: Note: a positive mean difference favours the booster interventions over the control and indicates increased physical activity in the booster intervention group compared with the control group.

TABLE 24

Difference in mean TEE per day between the full booster intervention group and the control group at 3 months

Outcome	Control (n = 61), mean (SD)	Full booster (n = 52), mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Mean TEE per day (kcal)	2265.9 (410.8)	2279.9 (425.6)	14.0 (−142.2 to 170.2)	0.859	−6.7 (−96.6 to 83.2)	0.883

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
: Note: a positive mean difference favours the full booster intervention over the control and indicates increased physical activity in the full booster intervention group compared with the control group.

TABLE 25

Difference in mean TEE per day between the mini booster intervention group and the control group at 3 months

Outcome	Control (n = 61), mean (SD)	Mini booster (n = 47), mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Mean TEE per day (kcal)	2265.9 (410.8)	2168.2 (415.8)	−97.7 (−256.6 to 61.2)	0.226	−69.6 (−165.4 to 26.2)	0.153

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
: Note: a negative mean difference favours the control over the mini booster intervention and indicates decreased physical activity in the mini booster intervention group compared with the control group.

TABLE 26

Subgroup evaluation: difference in mean TEE per day between the mini booster intervention group and the full booster intervention group at 3 months

Outcome	Mini booster, mean (SD)	Full booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Mean TEE per day (kcal)	(n = 47)	(n = 52)
Mean TEE per day (kcal)	2168.2 (415.8)	2279.9 (425.6)	111.7 (−56.5 to 279.9)	0.190	55.5 (−38.2 to 149.2)	0.242
Mean TEE per day (kcal)^d	(n = 47)	(n = 51)
Mean TEE per day (kcal)^d	2168.2 (415.8)	2251.2 (375.6)	82.9 (−75.8 to 241.6)	0.302	47.0 (−45.2 to 139.2)	0.314

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: Sensitivity analysis excluding an outlying participant.
: Note: a positive mean difference favours the full booster intervention over the mini booster intervention and indicates increased physical activity in the full booster intervention group compared with the mini booster intervention group.

TABLE 27

Difference in mean TEE per day between the booster intervention group (mini plus full) and the control group at 9 months

Outcome	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Mean TEE per day (kcal)	(n = 36)	(n = 55)
Mean TEE per day (kcal)	2177.2 (390.7)	2308.2 (646.3)	131.0 (−107.5 to 369.5)	0.278	51.5 (−137.2 to 240.2)	0.589
Mean TEE per day (kcal)^d	(n = 36)	(n = 54)
Mean TEE per day (kcal)^d	2177.2 (390.7)	2239.1 (397.1)	61.9 (−106.8 to 230.6)	0.468	−7.1 (−115.8 to 101.6)	0.897

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: Sensitivity analysis excluding one participant with extreme mean TEE per day of 6041.6 kcal (see Figure 16) (data were queried but record assumed to be correct).
: Note: a positive mean difference favours the booster interventions over the control and indicates increased physical activity in the booster intervention group compared with the control group.

TABLE 28

Sensitivity analysis: difference in mean TEE per day between the booster intervention group (mini plus full) and the control group at 9 months

Mean TEE per day (kcal)	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Multiple imputation (≥ 4 days)	(n = 36)	(n = 55)
Multiple imputation (≥ 4 days)	2163.0 (298.9)	2235.2 (395.5)	72.3 (−101.8 to 246.3)	0.411	18.1 (−102.9 to 139.1)	0.766
Regression imputation (≥ 4 days)	(n = 34)	(n = 52)
Regression imputation (≥ 4 days)	2202.0 (371.3)	2281.7 (379.8)	79.6 (−85.5 to 244.7)	0.341	13.9 (−80.1 to 107.9)	0.769
Complete cases	(n = 21)	(n = 39)
Complete cases	2118.1 (298.9)	2315.5 (726.2)	197.5 (−134.8 to 529.8)	0.239	118.6 (−152.7 to 389.9)	0.384
Complete cases^d	(n = 21)	(n = 38)
Complete cases^d	2118.1 (298.9)	2217.5 (395.5)	99.4 (−99.1 to 297.9)	0.320	31.7 (−88.7 to 152.1)	0.599
Multiple imputation (≥ 1 days)	(n = 37)	(n = 61)
Multiple imputation (≥ 1 days)	2168.4 (298.9)	2215.9 (395.5)	47.5 (−122.5 to 217.5)	0.581	14.5 (−105.6 to 134.6)	0.811
Per protocol	(n = 36)	(n = 55)
Per protocol	2177.2 (390.7)	2308.2 (646.3)	131.0 (−107.5 to 369.5)	0.278	51.5 (−137.2 to 240.2)	0.589
Per protocol^d	(n = 36)	(n = 54)
Per protocol^d	2177.2 (390.7)	2239.1 (397.1)	61.9 (−106.8 to 230.6)	0.468	−7.1 (−115.8 to 101.6)	0.897

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: Sensitivity analysis excluding one participant with extreme mean TEE per day of 6041.6 kcal (see Figure 16) (data were queried but record assumed to be correct).
: Note: a positive mean difference favours the booster interventions over the control and indicates increased physical activity in the booster intervention group compared with the control group.

TABLE 29

Subgroup evaluation: difference in mean TEE per day between the mini booster intervention group and the full booster intervention group at 9 months

Outcome	Mini booster, mean (SD)	Full booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Mean TEE per day (kcal)	(n = 25)	(n = 30)
Mean TEE per day (kcal)	2204.0 (415.9)	2395.1 (785.7)	191.1 (−159.3 to 541.5)	0.279	100.9 (−203.8 to 405.6)	0.508
Mean TEE per day (kcal)^d	(n = 25)	(n = 29)
Mean TEE per day (kcal)^d	2204.0 (415.9)	2269.4 (384.9)	65.4 (−153.4 to 284.2)	0.551	−9.3 (−175.2 to 156.6)	0.911

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: Sensitivity analysis excluding one participant with extreme mean TEE per day of 6041.6 kcal (see Figure 16) (data were queried but record assumed to be correct).
: Note: a positive mean difference favours the full booster intervention over the mini booster intervention and indicates increased physical activity in the full booster intervention group compared with the mini booster intervention group.

TABLE 30

Difference in PACs per week between the booster intervention group (mini plus full) and the control group at 3 months

Outcome	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
PACs per week (×1000)	(n = 61)	(n = 99)
PACs per week (×1000)	339.0 (146.0)	331.7 (169.4)	−7.2 (−59.0 to 44.5)	0.783	0.4 (−49.9 to 50.6)	0.988
PACs per week (×1000)^d	(n = 61)	(n = 98)
PACs per week (×1000)^d	339.0 (146.0)	324.4 (154.0)	−14.5 (−63.1 to 34.1)	0.556	−7.1 (−53.3 to 39.2)	0.763

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: Sensitivity analysis excluding one participant with extreme PACs per week.
: Note: a positive mean difference favours the booster interventions over the control and indicates increased physical activity in the booster intervention group compared with the control group; a negative difference favours the control over the booster interventions.

TABLE 31

Sensitivity analysis: difference in PACs per week between the booster intervention group (mini plus full) and the control group at 3 months

PACs per week (×1000)	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Multiple imputation (≥ 4 days)	(n = 61)	(n = 99)
Multiple imputation (≥ 4 days)	379.4 (135.0)	368.5 (155.6)	–10.9 (–77.6 to 55.8)	0.747	–2.7 (–67.1 to 61.8)	0.935
Regression imputation (≥ 4 days)	(n = 58)	(n = 95)
Regression imputation (≥ 4 days)	332.5 (133.8)	328.1 (157.1)	–4.4 (–53.4 to 44.6)	0.859	–1.6 (–48.0 to 44.9)	0.946
Complete cases (7 complete days)	(n = 44)	(n = 68)
Complete cases (7 complete days)	325.1 (134.9)	322.6 (155.6)	–2.6 (–59.3 to 54.1)	0.929	–12.3 (–65.1 to 40.4)	0.644
Per protocol	(n = 61)	(n = 96)
Per protocol	339.0 (146.0)	335.6 (170.5)	–3.3 (–55.6 to 48.9)	0.900	2.9 (–47.8 to 53.6)	0.910
Per protocol^d	(n = 61)	(n = 95)
Per protocol^d	339.0 (146.0)	328.2 (154.9)	–10.8 (–59.9 to 38.3)	0.665	–4.9 (–51.6 to 41.8)	0.836

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: Sensitivity analysis excluding one participant with extreme PACs per week.
: Note: a positive mean difference favours the booster interventions over the control and indicates increased physical activity in the booster intervention group compared with the control group.

TABLE 32

Difference in PACs per week between the booster intervention group (mini plus full) and the control group at 9 months

Outcome	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
PACs per week (×1000)	(n = 36)	(n = 55)
PACs per week (×1000)	319.0 (158.9)	315.4 (202.1)	–3.6 (–83.0 to 75.7)	0.928	–14.5 (–90.8 to 61.7)	0.705
PACs per week (×1000)^d	(n = 36)	(n = 54)
PACs per week (×1000)^d	319.0 (158.9)	302.5 (179.6)	–16.6 (–90.0 to 56.8)	0.655	–29.4 (–100.0 to 41.4)	0.411

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: Sensitivity analysis excluding one participant with extreme PACs per week.
: Note: a positive mean difference favours the booster interventions over the control and indicates increased physical activity in the booster intervention group compared with the control group.

TABLE 33

Sensitivity analysis: difference in PACs per week between the booster intervention group (mini plus full) and the control group at 9 months

PACs per week (×1000)	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Multiple imputation (≥ 4 days)	(n = 36)	(n = 55)
Multiple imputation (≥ 4 days)	351.2 (102.0)	353.2 (176.0)	19.8 (–120.0 to 123.8)	0.974	–1.3 (–120.0 to 117.5)	0.982
Regression imputation (≥ 4 days)	(n = 34)	(n = 52)
Regression imputation (≥ 4 days)	323.0 (151.6)	319.8 (188.2)	–3.3 (–79.9 to 73.4)	0.933	–13.5 (–83.3 to 56.3)	0.702
Complete cases (7 complete days)	(n = 21)	(n = 39)
Complete cases (7 complete days)	299.4 (102.0)	295.0 (176.0)	–4.4 (–88.1 to 79.3)	0.916	–15.2 (–110.0 to 74.9)	0.735
Complete cases (7 complete days)^d	(n = 21)	(n = 37)
Complete cases (7 complete days)^d	299.4 (102.0)	265.8 (123.7)	–33.7 (–97.4 to 30.0)	0.294	–37.3 (–110.0 to 31.6)	0.282
Per protocol	(n = 36)	(n = 55)
Per protocol	319.0 (158.9)	315.4 (202.1)	–3.6 (–83.0 to 75.7)	0.928	–14.5 (–90.8 to 61.7)	0.705
Per protocol^d	(n = 36)	(n = 54)
Per protocol^d	319.0 (158.9)	302.5 (179.6)	–16.6 (–90.0 to 56.8)	0.655	–29.4 (–100.0 to 41.4)	0.411

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: Sensitivity analysis excluding one participant with extreme PACs per week.
: Note: a positive mean difference favours the booster interventions over the control and indicates increased physical activity in the booster intervention group compared with the control group.

TABLE 34

Difference in SF-12v2 plus 4 dimension scores between the booster intervention group (mini plus full) and the control group at 3 and 9 months (ITT set)

Follow-up (months)	Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Follow-up (months)	Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
3	PCS score	57	49.3 (8.9)	89	48.8 (9.9)	0.7 (–1.4 to 2.7)	0.509	0.9 (–1.2 to 2.9)	0.394
	MCS score	57	47.2 (9.5)	89	49.5 (9.1)	0.8 (–1.5 to 3.1)	0.499	0.8 (–1.6 to 3.1)	0.525
	SF-6D utility score	56	0.644 (0.064)	88	0.648 (0.070)	0.002 (–0.017 to 0.021)	0.826	0.005 (–0.014 to 0.024)	0.623
9	PCS score	38	46.1 (9.4)	62	48.3 (10.4)	1.8 (–1.2 to 4.7)	0.244	1.7 (–1.2 to 4.6)	0.260
	MCS score	38	48.3 (10.4)	62	52.3 (8.7)	2.0 (–0.7 to 4.8)	0.149	1.7 (–1.1 to 4.5)	0.237
	SF-6D utility score	38	0.651 (0.083)	62	0.665 (0.080)	–0.000 (–0.028 to 0.028)	0.989	0.001 (–0.027 to 0.030)	0.921

a: Adjusted for baseline using the ANCOVA model.
b: Adjusted for age, gender, BMI, total minutes of physical activity at –3 months and –1 week before randomisation, HRQoL (SF-12v2 plus 4 total score) and baseline outcome measurement.
c: Using multiple regression.
: Note: higher PCS and MCS scores indicate a better physical and mental health status respectively; higher SF-6D utility scores indicate a better health status; a positive mean difference indicates improved health status in favour of the booster intervention.

TABLE 35

Difference in SF-12v2 plus 4 dimension scores between the booster intervention group (mini plus full) and the control group at 3 and 9 months (for all available data)

Follow-up (months)	Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Follow-up (months)	Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
3	PCS score	70	48.6 (9.7)	118	48.8 (9.6)	0.7 (–1.1 to 2.5)	0.449	0.8 (–0.9 to 2.6)	0.354
	MCS score	70	47.9 (9.4)	118	48.8 (10.2)	–0.2 (–2.4 to 2.0)	0.849	–0.3 (–2.6 to 1.9)	0.770
	SF-6D utility score	71	0.645 (0.071)	117	0.649 (0.075)	–0.001 (–0.017 to 0.016)	0.920	0.001 (–0.016 to 0.018)	0.934
9	PCS score	46	46.2 (9.0)	81	48.6 (10.1)	2.2 (–0.5 to 4.9)	0.102	1.8 (–0.9 to 4.5)	0.191
	MCS score	46	49.1 (10.2)	81	50.4 (9.3)	–0.2 (–3.0 to 2.6)	0.892	–0.4 (–3.3 to 2.6)	0.804
	SF-6D utility score	47	0.651 (0.083)	81	0.655 (0.077)	–0.004 (–0.028 to 0.021)	0.755	–0.005 (–0.030 to 0.020)	0.709

a: Adjusted for baseline using the ANCOVA model.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, HRQoL (SF-12v2 plus 4 total score) and baseline outcome measurement.
c: Using multiple regression.
: Note: higher PCS and MCS scores indicate a better physical and mental health status respectively; higher SF-6D utility scores indicate a better health status; a positive mean difference indicates improved health status in favour of the booster intervention.

TABLE 36

Difference in the average number of minutes spent on type of activity per day (over a week) between the booster intervention group (mini plus full) and the control group at 3 months based on an objective measure (ITT set; n = 160)

Outcome	Control (n = 61), mean (SD)	Booster (n = 99), mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
At least moderate activity (minutes)	48.4 (42.9)	46.8 (39.6)	–1.6 (–14.7 to 11.5)	0.810	–1.9 (–15.1 to 11.3)	0.776
Moderate activity (minutes)	47.1 (42.1)	44.7 (37.9)	–2.3 (–15.0 to 10.4)	0.721	–2.4 (–15.1 to 10.3)	0.710
Vigorous activity (minutes)	1.3 (2.4)	2.0 (4.2)	0.7 (–0.5 to 1.9)	0.236	0.5 (–0.7 to 1.7)	0.409

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at –3 months and -1 week prior to randomisation, and HRQoL (SF12v2 plus 4 total score).
c: Using multiple regression.
: Note: at least moderate, moderate and vigorous activity were defined as having at least 3 METs, 3–6 METs and > 6 METs respectively.

TABLE 37

Difference in the average number of minutes spent on type of activity per day (over a week) between the booster intervention group (mini plus full) and the control group at 9 months based on an objective measure (ITT set; n = 91)

Outcome	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Main analysis	(n = 36)	(n = 55)
At least moderate activity (minutes)	36.6 (22.7)	50.8 (71.9)	14.2 (–10.4 to 38.8)	0.255	17.0 (–7.9 to 41.9)	0.177
Moderate activity (minutes)	35.7 (22.2)	44.9 (47.9)	9.1 (–7.9 to 26.1)	0.289	11.8 (–5.1 to 28.7)	0.168
Vigorous activity (minutes)	0.9 (2.2)	6.0 (29.0)	5.1 (–4.6 to 14.8)	0.297	5.2 (–4.8 to 15.2)	0.306
Sensitivity analysis^d	(n = 36)	(n=54)
At least moderate activity (minutes)	36.6 (22.7)	42.5 (36.7)	5.8 (–7.8 to 19.4)	0.400	8.9 (–5.0 to 22.8)	0.208
Moderate activity (minutes)	35.7 (22.2)	40.3 (34.6)	4.6 (–8.3 to 17.5)	0.482	7.6 (–5.5 to 20.7)	0.252
Vigorous activity (minutes)	0.9 (2.2)	2.1 (4.7)	1.2 (–0.5 to 2.9)	0.159	1.3 (–0.5 to 3.1)	0.150

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.
d: Sensitivity analysis excluding a participant with an extreme value.
: Note: at least moderate, moderate and vigorous activity were defined as having at least 3 METs, 3–6 METs and > 6 METs respectively.

TABLE 38

Difference in the proportions of participants who maintained or increased physical activity at 3 and 9 months from baseline [based on self-reported measure of physical activity (SPAQ) for ITT participants]

Follow-up (months)	Outcome	Control, n (%)	Booster, n (%)	Difference in proportion maintaining or increasing physical activity (95% CI) (%)^a	p-value^a
3	Maintained or increased physical activity	(n = 60)	(n = 96)
3	Maintained or increased physical activity	30 (50.0)	55 (57.3)	7.3 (–8.8 to 23.3)	0.374
9	Maintained or increased physical activity	(n = 44)	(n = 66)
9	Maintained or increased physical activity	23 (52.3)	41 (62.1)	9.8 (–9.0 to 28.7)	0.305

a: Based on a two independent sample proportion test with continuity correction.
: Note: a positive difference in proportion indicates increased self-reported physical activity in the booster intervention group compared with the control group.

TABLE 39

Follow-up (months)	Outcome	Control, n (%)	Booster, n (%)	Difference in proportion maintaining or increasing physical activity (95% CI) (%)^a	p-value^a
3	Maintained or increased activity	(n = 76)	(n = 125)
3	Maintained or increased activity	41 (53.9)	71 (56.8)	2.9 (–11.0 to 17.1)	0.693
9	Maintained or increased activity	(n = 54)	(n = 85)
9	Maintained or increased activity	25 (46.3)	49 (57.6)	11.4 (–5.6 to 28.3)	0.191

a: Based on a two independent sample proportion test with continuity correction.
: Note: a positive difference in proportion indicates increased self-reported physical activity in the booster intervention group compared with the control group.

TABLE 40

Differences in change in the BREQ-2 multidimensions between the booster intervention group (mini plus full) and the control group at 3 months (all available data)

Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Amotivation	75	0.3 (0.5)	123	0.2 (0.5)	–0.0 (–0.2 to 0.1)	0.515	–0.0 (–0.2 to 0.1)	0.762
External regulation	75	0.3 (0.5)	123	0.4 (0.6)	0.1 (–0.1 to 0.2)	0.406	0.1 (–0.1 to 0.2)	0.408
Introjected regulation	76	1.1 (1.0)	123	1.4 (1.2)	0.3 (0.1 to 0.6)	0.018	0.3 (–0.0 to 0.6)	0.053
Identified regulation	75	2.9 (0.7)	123	3.0 (0.6)	0.2 (0.0 to 0.3)	0.047	0.1 (–0.0 to 0.3)	0.092
Intrinsic regulation	73	2.8 (1.0)	120	2.9 (1.0)	0.2 (–0.0 to 0.4)	0.077	0.2 (–0.0 to 0.4)	0.103
RAI	73	6.2 (3.4)	120	6.2 (3.3)	0.3 (–0.5 to 1.0)	0.478	0.2 (–0.5 to 0.9)	0.606

a: Adjusted for baseline.
b: Using the ANCOVA model.
c: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, HRQoL (SF-12v2 plus 4 total score) and baseline outcome measurement.
d: Using multiple regression.
: Note: a higher RAI score indicates higher self-determination and a positive mean difference favours the booster interventions over the control.

TABLE 41

Differences in change in the BREQ 2 multidimensions between the booster intervention group (mini plus full) and the control group at 9 months (all available data)

Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Amotivation	54	0.3 (0.6)	84	0.2 (0.5)	−0.0 (−0.2 to 0.1)	0.751	−0.0 (−0.2 to 0.1)	0.785
External regulation	54	0.3 (0.6)	84	0.3 (0.5)	0.0 (−0.2 to 0.2)	0.895	0.0 (−0.2 to 0.2)	0.939
Introjected regulation	54	1.2 (1.1)	84	1.4 (1.1)	0.1 (−0.2 to 0.4)	0.479	0.1 (−0.2 to 0.4)	0.657
Identified regulation	55	2.9 (0.7)	84	3.0 (0.6)	0.1 (−0.1 to 0.3)	0.412	0.1 (−0.1 to 0.3)	0.515
Intrinsic regulation	54	2.8 (1.1)	81	2.8 (1.0)	0.2 (−0.1 to 0.4)	0.239	0.2 (−0.1 to 0.4)	0.210
RAI	53	6.0 (3.8)	81	6.1 (3.5)	0.2 (−0.8 to 1.2)	0.711	0.2 (−0.8 to 1.3)	0.649

a: Adjusted for baseline.
b: Using the ANCOVA model.
c: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, HRQoL (SF-12v2 plus 4 total score) and baseline outcome measurement.
d: Using multiple regression.
: Note: a higher RAI score indicates higher self-determination and a positive mean difference favours the booster interventions over the control.

TABLE 42

Differences in change in the BREQ-2 multidimensions between the booster intervention group (mini plus full) and the control group at 3 months (ITT set)

Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Amotivation	59	0.3 (0.5)	95	0.2 (0.5)	−0.1 (−0.2 to 0.1)	0.510	−0.0 (−0.2 to 0.1)	0.735
External regulation	59	0.3 (0.5)	95	0.4 (0.6)	0.0 (−0.1 to 0.2)	0.514	0.0 (−0.1 to 0.2)	0.565
Introjected regulation	60	1.1 (1.1)	95	1.4 (1.2)	0.2 (−0.1 to 0.5)	0.175	0.1 (−0.2 to 0.5)	0.434
Identified regulation	59	2.9 (0.8)	95	3.1 (0.6)	0.2 (0.1 to 0.4)	0.011	0.2 (0.0 to 0.4)	0.020
Intrinsic regulation	57	2.8 (1.0)	92	3.0 (0.9)	0.3 (0.0 to 0.5)	0.028	0.2 (0.0 to 0.5)	0.049
RAI	57	5.8 (3.4)	92	6.4 (3.4)	0.6 (−0.3 to 1.5)	0.168	0.6 (−0.3 to 1.4)	0.192

a: Adjusted for baseline.
b: Using the ANCOVA model.
c: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, HRQoL (SF-12v2 plus 4 total score) and baseline outcome measurement.
d: Using multiple regression.
: Note: a higher RAI score indicates higher self-determination and a positive mean difference favours the booster interventions over the control.

TABLE 43

Differences in change in the BREQ-2 multidimensions between the booster intervention group (mini plus full) and the control group at 9 months (ITT set)

Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Amotivation	44	0.3 (0.6)	65	0.2 (0.3)	−0.1 (−0.2 to 0.1)	0.386	−0.1 (−0.2 to 0.1)	0.454
External regulation	44	0.3 (0.6)	65	0.2 (0.4)	−0.1 (−0.2 to 0.1)	0.328	−0.1 (−0.3 to 0.1)	0.340
Introjected regulation	44	1.3 (1.1)	65	1.4 (1.1)	−0.0 (−0.4 to 0.3)	0.894	−0.1 (−0.4 to 0.2)	0.527
Identified regulation	44	2.9 (0.7)	65	3.0 (0.6)	0.1 (−0.1 to 0.3)	0.483	0.1 (−0.2 to 0.3)	0.558
Intrinsic regulation	43	2.8 (1.1)	62	3.0 (0.9)	0.2 (−0.0 to 0.5)	0.072	0.3 (0.0 to 0.6)	0.043
RAI	43	5.9 (3.9)	62	6.8 (2.6)	0.7 (−0.4 to 1.7)	0.218	0.8 (−0.3 to 1.9)	0.169

a: Adjusted for baseline.
b: Using the ANCOVA model.
c: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, HRQoL (SF-12v2 plus 4 total score) and baseline outcome measurement.
d: Using multiple regression.
: Note: a higher RAI score indicates higher self-determination and a positive mean difference favours the booster interventions over the control.

TABLE 44

Difference in BMI between the booster intervention group (mini plus full) and the control group at 3 and 9 months (all available data)

Follow-up (months)	Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Follow-up (months)	Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
3	BMI (kg/m²)	75	29.6 (5.9)	124	30.1 (5.5)	−0.2 (−0.5 to 0.1)	0.194	−0.1 (−0.4 to 0.2)	0.385
9	BMI (kg/m²)	55	28.4 (5.4)	85	29.1 (5.1)	0.1 (−0.3 to 0.6)	0.601	0.1 (−0.4 to 0.6)	0.594

a: Adjusted for baseline BMI.
b: Using the ANCOVA model.
c: Adjusted for age, gender, total minutes of physical activity at 3 months and 1 week before randomisation, HRQoL (SF-12v2 plus 4 total score) and baseline BMI.
d: Using multiple regression.
: Note: a negative mean difference in BMI favours the booster interventions over the control.

TABLE 45

Difference in BMI between the booster intervention group (mini plus full) and the control group at 3 and 9 months (ITT set)

Follow-up (months)	Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
Follow-up (months)	Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^b	Adjusted mean difference (95% CI)^c	p-value^d
3	BMI (kg/m²)	59	29.0 (4.5)	95	29.6 (5.5)	−0.2 (−0.6 to 0.1)	0.160	−0.2 (−0.5 to 0.1)	0.300
9	BMI (kg/m²)	44	28.4 (4.9)	66	28.6 (5.2)	0.1 (−0.5 to 0.6)	0.844	0.1 (−0.5 to 0.6)	0.839

a: Adjusted for baseline BMI.
b: Using the ANCOVA model.
c: Adjusted for age, gender, total minutes of physical activity at 3 months and 1 week before randomisation, HRQoL (SF-12v2 plus 4 total score) and baseline BMI.
d: Using multiple regression.
: Note: a negative mean difference in BMI favours the booster interventions over the control.

TABLE 46

Difference in total minutes of self-reported physical activity over a week period between the booster intervention group (mini plus full) and the control group at 3 and 9 months (ITT set)

Follow-up (months)	Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Follow-up (months)	Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
3	Total minutes of physical activity	60	383.8 (404.5)	96	419.8 (527.9)	49.5 (−109.4 to 206.3)	0.545	32.1 (−131.1 to 195.3)	0.698
9	Total minutes of physical activity	44	461.0 (457.4)	66	535.0 (560.2)	74.0 (−128.2 to 276.2)	0.470	120.0 (−92.1 to 332.1)	0.264

a: Adjusted for baseline total minutes of physical activity using the ANCOVA model.
b: Adjusted for age, gender, BMI, HRQoL (SF-12v2 plus 4 total score) and total minutes of physical activity.
c: Using multiple regression.
: Note: a positive mean difference in self-reported total minutes of physical activity favours the booster interventions over the control.

TABLE 47

Difference in total minutes of self-reported physical activity over a week period between the booster intervention group (mini plus full) and the control group at 3 and 9 months (all available data)

Follow-up (months)	Outcome	Control		Booster		Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
Follow-up (months)	Outcome	n	Mean (SD)	n	Mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
3	Total minutes of physical activity	76	458.0 (681.2)	125	411.2 (498.3)	−35.4 (−200.0 to 129.2)	0.672	−52.5 (−222.6 to 117.6)	0.543
9	Total minutes of physical activity	54	446.1 (472.4)	81	508.6 (597.4)	62.4 (−128.4 to 253.3)	0.519	88.4 (−114.3 to 291.1)	0.390

a: Adjusted for baseline total minutes of physical activity using the ANCOVA model.
b: Adjusted for age, gender, BMI, HRQoL (SF-12v2 plus 4 total score) and total minutes of physical activity.
c: Using multiple regression.
: Note: a positive mean difference in self-reported total minutes of physical activity favours the booster interventions over the control.

TABLE 48

Difference in distance walked between the booster intervention group (mini plus full) and the control group at 3 and 9 months (all available data)

Follow-up (months)	Outcome	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
3	Distance walked (m) on 12-minute walk test	(n = 75)	(n = 118)
3	Distance walked (m) on 12-minute walk test	898.9 (257.5)	962.4 (227.5)	63.5 (−6.3 to 133.3)	0.074	80.3 (13.8 to 146.8)	0.018
9	Distance walked (m) on 12-minute walk test	(n = 51)	(n = 82)
9	Distance walked (m) on 12-minute walk test	992.2 (292.7)	1077.4 (301.0)	85.2 (−19.9 to 190.3)	0.111	80.6 (−26.8 to 188.0)	0.140

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.

TABLE 49

Difference in distance walked between the booster intervention group (mini plus full) and the control group at 3 and 9 months (ITT set)

Follow-up (months)	Outcome	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)^a	p-value^a	Adjusted mean difference (95% CI)^b	p-value^c
3	Distance walked (m) on 12-minute walk test	(n = 59)	(n = 91)
3	Distance walked (m) on 12-minute walk test	917.6 (254.9)	984.2 (226.2)	66.6 (−12.0 to 145.2)	0.096	90.8 (14.5 to 167.1)	0.020
9	Distance walked (m) on 12-minute walk test	(n = 42)	(n = 64)
9	Distance walked (m) on 12-minute walk test	1000.2 (308.8)	1102.4 (285.4)	102.2 (−13.9 to 218.3)	0.084	115.9 (1.1 to 230.7)	0.048

a: Using a two independent sample t-test.
b: Adjusted for age, gender, BMI, total minutes of physical activity at 3 months and 1 week before randomisation, and HRQoL (SF-12v2 plus 4 total score).
c: Using multiple regression.

TABLE 50

Proportions of participants meeting current recommendations of at least 30 minutes of at least moderate physical activity for at least 5 days a week at 3 and 9 months

Follow-up (months)	Outcome	Control, n (%)	Mini booster, n (%)	Full booster, n (%)	Mini + full booster, n (%)	Difference in proportion maintaining or increasing physical activity (95% CI) (%)	p-value
3	Meeting current physical activity recommendations	(n = 73)	(n = 57)	(n = 59)	(n = 116)
3	Meeting current physical activity recommendations	25 (34.2)	12 (21.1)	27 (45.8)	39 (33.6)	−0.6 (−14.5 to 13.2)	0.929
9	Meeting current physical activity recommendations	(n = 39)	(n = 27)	(n = 35)	(n = 62)
9	Meeting current physical activity recommendations	11 (28.2)	7 (25.9)	12 (34.3)	19 (30.6)	2.4 (−15.7 to 20.6)	0.794

: Note: at least moderate physical activity was defined as having at least 3 METs.

TABLE 51

Subgroup evaluation: interaction effect at 3 months between gender and intervention

Outcome measure	Subgroup	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)	p-value
Mean TEE per day (kcal)	Male	(n = 26)	(n = 42)
	Male	2483.0 (421.8)	2455.2 (416.3)	–27.8 (–236.3 to 180.6)	0.791
	Female	(n = 35)	(n = 57)
	Female	2104.6 (322.7)	2058.6 (343.0)	–46.0 (–189.1 to 97.1)	0.525
	All	(n = 61)	(n = 99)
	All	2265.9 (410.8)	2226.9 (422.6)	–39.0 (–173.4 to 95.4)	0.567
	Interaction test	NA	NA	–18.2 (–260.6 to 224.3)^a	0.882

: NA, not applicable.
a: Interaction test mean difference is the difference in the mean difference in treatment effect between men and women.
: Note: positive and negative mean difference in mean TEE favours the booster group and the control group respectively.

TABLE 52

Subgroup evaluation: interaction effect at 9 months between gender and intervention

Outcome measure	Subgroup	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)	p-value
Mean TEE per day (kcal)	Male	(n = 13)	(n = 27)
	Male	2425.9 (402.6)	2409.2 (382.1)	–16.8 (–282.4 to 248.9)	0.899
	Female	(n = 23)	(n = 27)
	Female	2036.6 (311.9)	2069.1 (339.6)	32.4 (–154.2 to 219.1)	0.728
	All^a	(n = 36)	(n = 54)
	All^a	2177.2 (390.7)	2239.1 (397.1)	61.9 (–106.8 to 230.6)	0.468
	Interaction test	NA	NA	49.2 (–262.6 to 361.0)^b	0.755

: NA, not applicable.
a: Excluding a participant with an extreme value (mean TEE per day of 6047.6 kcal).
b: Interaction mean difference is the difference in the mean difference in treatment effect between men and women.
: Note: positive and negative mean difference in mean TEE favours the booster group and the control group respectively.

TABLE 53

Subgroup evaluation: interaction effect at 3 months between use of community facilities in the last month (yes or no) and intervention

Outcome measure	Subgroup	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)	p-value
Mean TEE per day (kcal)	Yes	(n = 16)	(n = 22)
	Yes	2097.5 (244.0)	2181.1 (358.8)	83.6 (–127.0 to 294.2)	0.426
	No	(n = 45)	(n = 76)
	No	2325.8 (442.5)	2249.2 (435.7)	–76.6 (–239.8 to 86.6)	0.355
	All	(n = 61)	(n = 98)
	All	2265.9 (410.8)	2233.9 (418.9)	–32.0 (–166.0 to 101.9)	0.638
	Interaction test	NA	NA	–160.2 (–469.1 to 148.7)^a	0.307

: NA, not applicable.
a: Interaction test mean difference is the difference in the mean difference in treatment effect between men and women.
: Note: positive and negative mean difference in mean TEE favours the booster group and the control group respectively.

TABLE 54

Subgroup evaluation: interaction effect at 9 months between use of community facilities in the last month (yes or no) and intervention

Outcome measure	Subgroup	Control, mean (SD)	Booster, mean (SD)	Mean difference (95% CI)	p-value
Mean TEE per day (kcal)	Yes	(n = 11)	(n = 14)
	Yes	2084.8 (398.5)	2079.7 (230.0)	–5.1 (–267.2 to 257.1)	0.968
	No	(n = 25)	(n = 39)
	No	2217.9 (388.4)	2313.1 (419.0)	95.2 (–113.4 to 303.9)	0.365
	All^a	(n = 36)	(n = 53)
	All^a	2177.2 (390.7)	2251.5 (390.3)	74.2 (–93.4 to 241.9)	0.381
	Interaction test	NA	NA	100.3 (–264.8 to 465.3)^b	0.586

: NA, not applicable.
a: Excluding a participant with an extreme value (mean TEE per day of 6047.6 kcal).
b: Interaction mean difference is the difference in the mean difference in treatment effect between men and women.
: Note: positive and negative mean difference in mean TEE favours the booster group and the control group respectively.

TABLE 55

Subgroup evaluation: interaction effect at 3 months between seasonality (time of initial contact) and intervention (full booster vs. control)

Outcome measure	Subgroup	Control, mean (SD)	Full booster, mean (SD)	Mean difference (95% CI)	p-value
Mean TEE per day (kcal)	Summer/spring	(n = 48)	(n = 33)
	Summer/spring	2275.8 (417.6)	2315.2 (444.9)	39.4 (–153.7 to 232.4)	0.686
	Winter/autumn	(n = 13)	(n = 19)
	Winter/autumn	2229.3 (398.3)	2218.6 (394.0)	–10.8 (–301.6 to 280.1)	0.940
	All	(n = 61)	(n = 52)
	All	2265.9 (410.8)	2279.9 (425.6)	14.0 (–142.2 to 170.2)	0.860
	Interaction test	NA	NA	–50.1 (–404.0 to 303.7)^a	0.779

: NA, not applicable.
a: Interaction test mean difference is the difference in the mean difference in treatment effect between those approached in summer/spring and those approached in winter/autumn.
: Note: positive and negative mean difference in mean TEE favours the full booster group and the control group respectively.

TABLE 56

Subgroup evaluation: interaction effect at 3 months between seasonality (time of initial contact) and intervention (mini booster vs. control)

Outcome measure	Subgroup	Control, mean (SD)	Mini booster, mean (SD)	Mean difference (95% CI)	p-value
Mean TEE per day (kcal)	Summer/spring	(n = 48)	(n = 35)
	Summer/spring	2275.8 (417.6)	2154.5 (411.9)	–121.4 (–305.0 to 62.3)	0.192
	Winter/autumn	(n = 13)	(n = 12)
	Winter/autumn	2229.3 (398.3)	2208.4 (442.8)	–20.9 (–368.8 to 327.0)	0.902
	All	(n = 61)	(n = 47)
	All	2265.9 (410.8)	2168.2 (415.8)	–97.7 (–256.6 to 61.2)	0.226
	Interaction test	NA	NA	100.4 (–277.6 to 478.5)^a	0.599

: NA, not applicable.
a: Interaction test mean difference is the difference in the mean difference in treatment effect between those approached in summer/spring and those approached in winter/autumn.
: Note: positive and negative mean difference in mean TEE favours the full booster group and the control group respectively.

Copyright © Queen’s Printer and Controller of HMSO 2014. This work was produced by Goyder et al. under the terms of a commissioning contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

Included under terms of UK Non-commercial Government License.

Bookshelf ID: NBK261665

Contents