Skip to main content
. Author manuscript; available in PMC: 2019 Apr 16.
Published in final edited form as: JAMA. 2018 Oct 16;320(15):1560–1569. doi: 10.1001/jama.2018.14433

Table 3.

Associations between common measures of weight regain and concurrent declines in select health outcomes among adults who underwent Roux-en-Y gastric bypass.

Progression of diabetesa
n=689
Clinically important decline
in physical HRQoLb
n=903
Clinically important decline
in mental HRQoLb
n=903
Clinically important decline
in satisfactionc
n=272

RRd (95% CI) P value BICe RRd (95% CI) P value BICe RRd (95% CI) P value BICe RRd (95% CI) P value BICe
Weight regain, continuousf
Weight in kg, per 9.1 kg 1.40 (1.18–1.67) <.001 1543.2 1.21 (1.12–1.31) <.001 3541.9 1.08 (1.00–1.17) 0.050 3739.2 1.41 (1.14–1.74) 0.002 584.4
BMI, per 3.2 kg/m² 1.41 (1.19–1.68) <.001 1543.2 1.21 (1.12–1.31) <.001 3542.8 1.09 (1.01–1.18) 0.03 3738.5 1.44 (1.16–1.78) <.001 583.3
% pre-surgery weight, per 6.9% 1.45 (1.22–1.73) <.001 1541.9 1.22 (1.12–1.32) <.001 3544.1 1.11 (1.02–1.20) 0.02 3737.7 1.45 (1.12–1.87) 0.005 585.5
% nadir weight, per 11.0% 1.36 (1.15–1.60) <.001 1545.5 1.16 (1.07–1.25) <.001 3551.2 1.08 (1.00–1.17) 0.04 3738.8 1.35 (1.07–1.70) 0.011 586.9
% max weight lost, per 18.9% 1.51 (1.27–1.78) <.001 1535.8 1.28 (1.18–1.38) <.001 3532.8 1.11 (1.02–1.20) 0.02 3737.6 1.56 (1.19–2.04) 0.001 582.6

Clinically important weight regaing
≥10 kg 1.56 (1.15–2.11) 0.004 1549.4 1.39 (1.19–1.63) <.001 3548.0 1.12 (0.96–1.30) 0.15 3741.9 1.91 (1.21–3.01) 0.006 584.3
≥ 5 BMI points 1.66 (1.19–2.30) 0.002 1548.6 1.40 (1.18–1.65) <.001 3550.2 1.23 (1.04–1.45) 0.014 3737.3 1.84 (1.13–3.01) 0.02 586.9
≥10% of pre-surgery weight 1.63 (1.19–2.25) 0.003 1549.5 1.45 (1.23–1.71) <.001 3544.4 1.17 (1.00–1.38) 0.054 3739.6 1.73 (1.05–2.84) 0.03 587.1
≥10% of nadir weight 1.37 (1.02–1.84) 0.03 1553.2 1.36 (1.15–1.59) <.001 3552.0 1.11 (0.95–1.29) 0.19 3741.0 1.82 (1.14–2.91) 0.013 586.5
≥15% of nadir weight 1.41 (1.03–1.92) 0.03 1554.0 1.36 (1.16–1.59) <.001 3550.3 1.19 (1.01–1.39) 0.03 3739.6 1.51 (0.93–2.46) 0.09 589.8
≥10% of max weight lost 1.42 (1.02–1.98) 0.04 1554.0 1.42 (1.17–1.72) <.001 3550.4 1.09 (0.92–1.29) 0.31 3742.2 2.54 (1.34–4.82) 0.005 582.9
≥20% of max weight lost 1.64 (1.22–2.19) <.001 1547.2 1.55 (1.33–1.82) <.001 3535.5 1.23 (1.06–1.43) 0.008 3735.8 2.33 (1.50–3.63) <.001 579.4
≥25% of max weight lost 1.64 (1.21–2.20) 0.001 1547.3 1.43 (1.22–1.68) <.001 3546.5 1.16 (0.99–1.36) 0.06 3739.9 2.22 (1.44–3.42) <.001 575.9

Abbreviations: BIC=Bayesian Information Criterion; HRQoL=health-related quality of life; max=maximum; RR= Relative Risk.

a

A change from not taking diabetes medication to taking diabetes medication, change from not taking insulin to taking insulin, or an increase of glycated hemoglobin (HbA1c) by at least 0.5% plus a post-weight nadir value of 5.7% or greater.

b

A decrease of ≥5 points on the Short-Form 36-item Physical Component Summary (PCS) or Mental Component Summary (MCS) score, respectively.

c

An increase of ≥1 on a 7 point scale to a rating of at least 3 (i.e., “somewhat satisfied” – “very dissatisfied”).

d

Adjusted for pre-surgery factors related to missing data (i.e., site, age and smoking status), MCS, PCS or HbA1c at time of weight loss nadir, respectively, and time since weight nadir as a continuous fixed effect (only linear term retained). Sample size for each model is based on n=1286 minus those with missing outcome data.

e

A difference of more than 2 is good evidence that the model with the lower BIC has better fit.

f

RR is reported per median weight regain across post-nadir time points.

g

As previously defined in the scientific literature11,19,20,24,2628.