Skip to main content

Table 3 Estimated mean changea from baseline at end of treatment (12 months) and follow-up (18 months)

From: A pragmatic randomised multi-centre trial of multifamily and single family therapy for adolescent anorexia nervosa

 

End of treatment (12 months post randomisation)

Follow-up (18 months post randomisation)

 

Group

Observed: n (%)

Mean Change (95% CI)

Test

Standardised Coefficient

12m Predicted Mean (95% CI)

Observed: n (%)

Mean Change (95% CI)

Test

Standardised Coefficientb

18 m predicted mean (95% CI)

%mBMI

Baseline = 77.95

FT-AN

72 (88%)

10.84 (8.64, 13.04)

t = 9.66 p < 0.001

1.79

88.78 (86.59, 90.98)

55 (67%)

9.63 (7.18, 12.08)

t = 7.71 p < 0.001

1.59

87.58 (85.13, 90.03)

 

MFT-AN

75 (88%)

13.08 (10.98, 15.18)

t = 12.22 p < 0.001

2.16

91.03 (88.94, 91.13)

60 (71%)

13.74 (11.41, 16.08)

t = 11.53 p < 0.001

2.27

91.68 (89.33, 94.03)

BMI

Baseline = 15.7

FT-AN

72 (88%)

2.61 (2.16, 3.06)

t = 11.31 p < 0.001

2.09

18.31 (17.86, 18.76)

55 (67%)

2.59 (2.10, 3.08)

t =10.39 p < 0.001

2.07

18.29 (17.80, 18.78)

 

MFT-AN

75 (88%)

3.01 (2.59, 3.42)

t =14.14 p < 0.001

2.41

18.71 (18.29, 19.12)

60 (71%)

3.33 (2.87, 3.79)

t = 14.24 p < 0.001

2.66

19.03 (18.57, 19.49)

EDE (restraint)

Baseline = 2.78

FT-AN

38 (46%)

−1.38 (−1.90, −0.86)

t = −5.23 p < 0.001

−0.80

1.39 (0.88, 1.92)

32 (39%)

−1.59 (−2.12 −1.04)

t = −5.77 p < 0.001

−0.91

1.20 (0.66, 1.74)

 

MFT-AN

44 (52%)

−0.93 (−1.42 −0.44)

t = −3.76 p = 0.005

−0.52

1.85 (1.36, 2.34)

38 (45%)

−1.14 (−1.64 −0.63)

t = −4.45 p < 0.001

−0.75

1.64 (1.14, 2.15)

EDE (eating concerns)

Baseline = 2.01

FT-AN

38 (46%)

−0.83 (−1.30 −0.35)

t = −3.46 p < 0.001

−0.54

1.18 (0.71, 1.65)

32 (39%)

−1.10 (−1.57 −0.64)

t = −4.25 p < 0.001

−0.72

0.91 (0.44, 1.37)

 

MFT-AN

44 (52%)

−0.70 (−1.13 −0.25)

t = −3.10 p = 0.002

−0.45

1.32 (0.88, 1.76)

38 (45%)

−0.92 (−1.35 −0.50)

t = −4.25 p < 0.001

−0.605

1.09 (0.66, 1.51)

EDE (shape concerns)

Baseline = 1.83

FT-AN

37 (45%)

−1.28 (−1.85 −0.71)

t = −4.44 p < 0.001

−0.70

1.75 (1.18, 2.32)

32 (39%)

−1.25 (−1.84 −0.66)

t = −4.16 p < 0.001

−0.68

1.78 (1.19, 2.37)

 

MFT-AN

44 (52%)

−0.74 (−1.28 −0.20)

t = −2.70 p = 0.0072

−0.41

2.29 (1.75, 2.82)

38 (45%)

−0.69 (−1.24 −0.15)

t = −2.491 p < 0.013

−0.38

2.34 (1.79, 2.88)

EDE (weight concerns)

Baseline = 2.64

FT-AN

38 (46%)

−1.03 (−1.57 −0.48)

t = −3.67 p < 0.001

−0.54

1.62 (1.07, 2.16)

31 (38%)

−0.88 (−1.50 −0.26)

t = −2.80 p = 0.006

−0.47

1.76 (1.14, 2.38)

 

MFT-AN

44 (52%)

−0.59 (−1.12 −0.06)

t = −2.18 p = 0.029

−0.31

2.05 (1.52, 2.58)

37 (44%)

−0.39 (−0.96, 0.18)

t = −1.34 p = 0.181

−0.21

2.25 (1.68, 2.82)

BDI

Baseline = 24.47

FT-AN

35 (43%)

−8.82 (−13.31 −4.33)

t = −3.86 p < 0.001

−0.62

15.65 (11.2, 20.14)

37 (45%)

−7.40 (−12.04 −2.75)

t = - 3.13 p = 0.002

−0.52

17.07 (12.43, 21.72)

 

MFT-AN

41 (48%)

−5.91 (−10.20 −1.63

t = −2.71 p = 0.007

−0.41

18.55 (14.27, 22.84)

33 (39%)

−5.26 (−9.89 −0.62)

t = −2.23 p = 0.026

−0.37

19.22 (14.6, 23.85)

RSE

Baseline = 25.68

FT-AN

36 (44%)

−0.51 (-2.74, 1.72)

t = −0.45 p = 0.653

–0.08

25.19 (22.96, 27.42)

37 (45%)

–1.50 (–3.80, 0.79)

t = –1.29 p = 0.199

–0.51

24.20 (21.90, 26.49)

 

MFT-AN

42 (49%)

–1.60 (–3.65, 0.44)

t =–1.54 p = 0.124

–0.24

24.10 (22.05, 26.14)

35 (41%)

–1.00 (–3.11, 1.11)

t = –0.93 p = 0.352

–0.15

24.70 (22.58, 26.18)

ECI (Negative)

Baseline = 85.56

FT-AN

44 (54%)

–13.59 (–24.58,–2.59)

t = −2.47 p = 0.016

−0.45

71.97 (60.98, 82.97)

     
 

MFT-AN

44 (52%)

−15.81 (−26.88 −4.72)

t = −2.86 p = 0.006

−0.52

69.75 (58.68, 80.83)

     

ECI (Positive)

Baseline = 28.27

FT-AN

47 (57%)

0.42 (−1.61, 2.45)

t−0.41 p = 0.682

0.05

28.67 (26.66, 30.72)

     
 

MFT-AN

46 (54%)

1.13 (−0.99, 3.25)

t−1.07 p = 0.291

0.13

29.40 (27.28, 31.52)

     
  1. The number of observations is presented with percentage of observations possible in each treatment arm. Estimates are derived using multiple imputation with100 imputations
  2. Abbreviations: %mBMI a percentage of median Body Mass Index for young people of the same height, age and sex, EDE Eating Disorder Examination, BDI Beck Depression Inventory, RSE Rosenberg Self Esteem Scale, ECI Experience of Caregiving Inventory
  3. aTo derive predicted mean change scores, continuous covariates were set at the mean of the sample at baseline and categorical values at the most frequent value: For each outcome, at the baseline mean of that outcome and the following additional estimates: Eating disorder in months; mean 9.34, previous Eating Disorder patient: no, family history of Eating Disorder: birth order in family; eldest
  4. bstandardised coefficients were calculated by dividing estimated mean change by the standard deviation of the outcome variable at baseline
Back to article page

BMC Psychiatry

ISSN: 1471-244X

Contact us