Motivational profiles and change in physical activity during a weight loss intervention: a secondary data analysis

Abstract

Background: High levels of moderate-to-vigorous intensity physical activity (MVPA) are strongly associated with sustained weight loss, however the majority of adults are unsuccessful in maintaining high levels of MVPA long-term. Our goal was to identify profiles based on exercise motives, and examine the association between motivational profile and longitudinal changes in MVPA during a weight loss intervention.

Methods: Adults with overweight or obesity (n = 169, mean ± SE; age 39 ± 0.7 years, BMI 34.4 ± 0.3 kg/m2, 83% female) underwent an 18-month behavioral weight loss program, including 6 months of supervised exercise, followed by 6 months of unsupervised exercise. Participants self-reported behavioral regulations for exercise at baseline (BREQ-2). Latent profile analysis identified subgroups from external, introjected, identified, and intrinsic regulations measured at baseline. Mean differences in device-measured total MVPA were compared across motivational profiles at baseline, after 6 months of supervised exercise and after a subsequent 6 months of unsupervised exercise.

Results: Three motivational profiles emerged: high autonomous (high identified and intrinsic, low external regulations; n = 52), high combined (high scores on all exercise regulations; n = 25), and moderate combined (moderate scores on all exercise regulations; n = 92). Motivational profile was not associated with baseline level of MVPA or the increase in MVPA over the 6-month supervised exercise intervention (high autonomous: 21 ± 6 min/d; high combined: 20 ± 9 min/d; moderate combined: 33 ± 5 min/d; overall P > 0.05). However, during the transition from supervised to unsupervised exercise, MVPA decreased, on average, within all three profiles, but the high autonomous profile demonstrated the least attenuation in MVPA (- 3 ± 6 min/d) compared to the moderate combined profile (- 20 ± 5 min/d; P = 0.043).

Conclusions: Results were in alignment with the Self-Determination Theory. Adults motivated by autonomous reasons (value benefits of exercise, intrinsic enjoyment) may be more likely to sustain increases in MVPA once support is removed, whereas participants with moderate-to-high scores on all types of exercise regulations may need additional long-term support in order to sustain initial increases in MVPA.

Clinical trial registration: NCT01985568. Registered 24 October 2013.

Keywords: Exercise; Latent profile analysis; Motivation; Obesity; Weight loss.

Conflict of interest statement

The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. DC reports consulting income from Gelesis, Inc., a company that has developed a weight loss device. SP has a grant from WW International unrelated to this work.

© 2021. The Author(s).

Figures

Fig. 1

Aligned Assessment Period by Randomized Group. To ensure equal exposure to PA in analyses, PA was captured during two time points: 1) after 6 months of supervised exercise (month 6 for standard, month 12 for sequential), and 2) after 6 months of subsequent unsupervised exercise (month 12 for standard, month 18 for sequential); PA: physical activity

Fig. 2

Consort Diagram. PA: physical activity

Fig. 3

Mean Exercise Regulation Score across Baseline Motivational Profiles. Exercise regulation score (mean) across the four motivational profiles; Exercise regulations included external category (range 0–3; 4 categories include: 0 (score = 0), 1 (score > 0 but ≤ 0.5), 2 (score > 0.5 but ≤ 1.25), and 3 (score > 1.25), introjected (range 0–4), identified (range 0–4), and intrinsic regulations (range 0–4); n = 92 for Moderate Combined; n = 52 for High Autonomous; n = 25 for High Combined

Fig. 4

A-C Change in Mean Total MVPA over time across Baseline Motivational Profiles. Mean difference ( ± SE) in change in total MVPA (min/d) across profiles tested with Wald test and subsequent between group comparisons; * indicates significant difference (P < 0.05) from moderate combined profile; MVPA: minutes of moderate-to-vigorous physical activity. a For Panel A sample sizes are as follows: n = 64 for Moderate Combined; n = 36 for High Autonomous; n = 13 for High Combined. b For Panel B sample sizes are as follows: n = 57 for Moderate Combined; n = 30 for High Autonomous; n = 11 for High Combined. c For Panel C sample sizes are as follows: n = 58 for Moderate Combined; n = 30 for High Autonomous; n = 12 for High Combined