Sage Journals: Discover world-class research

Abstract

Background:

Due to minimal effects, new directions for behavioral (non-pharmacologic/non-surgical) obesity interventions are required. Previous research suggests that physical activity and exercise (PA/exercise) might have considerable merit for weight reduction via impacts on mood and other psychosocial correlates of controlled eating. Although as few as 3 bouts/week of low-moderate intensity PA/exercise are associated with improved mood—with no dose–response effect beyond that frequency—it is unknown if mood improvements may be leveraged when participants already complete ⩾3 bouts/week by treatment start.

Objectives:

The present research addressed gaps in the available PA/exercise–psychosocial change–weight management research.

Design:

This study combined group contrasts with mediation analyses.

Methods:

Women participating in a community-based cognitive-behavioral obesity treatment emphasizing the increase of PA/exercise and exercise-related self-regulation and self-efficacy (N = 99) were divided for separate analyses into groupings of <3 bouts/week, and ⩾3 bouts/week, of light (e.g., easy walking) and moderate (e.g., fast walking) PA/exercise regularly completed prior to treatment start. The participants were assessed on psychosocial and PA/exercise changes from baseline to month 3.

Results:

Reductions in total mood disturbance (TMD), depression, and anxiety scores were overall significant, with no significant difference by exercise-frequency grouping. PA/exercise increases were significantly greater in the <3 bouts/week groupings. Change in self-regulatory skills usage mediated relations between groupings and changes in TMD, depression, and anxiety. In subsequent serial mediation models, significant paths from grouping → self-regulation change → self-efficacy change → TMD and depression changes were found.

Conclusion:

Whether or not ⩾3 bouts/week of PA/exercise are completed at intervention start, TMD, depression, and anxiety may be reduced through treatment foci on increasing PA/exercise, self-regulation, and self-efficacy. Findings might be useful for informing scalable obesity treatments.

Keywords

self-regulation self-efficacy mood cognitive-behavioral exercise physical activity

Introduction

Approximately 42% of adults in the United States have a weight-to-height ratio that classifies them as having obesity (i.e., body mass index (BMI) ⩾30 kg/m²).¹ That medical issue, which is associated with numerous health risks,² has been consistently increasing over many decades in wealthy nations.³ Although controlled eating and increased physical activity/exercise (PA/exercise) will reliably reduce weight in almost all adults with obesity, those behaviors have consistently been resistant to change—especially beyond the initial weeks after their initiation.^4,5 First-line approaches aiming to simply inform affected individuals on methods to improve their diet and increase PA/exercise⁶ have been particularly unsuccessful.^7,8

Adjusting behavioral obesity-treatment contents

More recent theory- and evidence-grounded approaches have incorporated weight loss behavior-change methods based on their verified psychosocial correlates.⁹ Those include participants’ use of self-regulatory methods (e.g., goal-setting, cognitive restructuring/productive self-talk) and their development of more favorable mood (e.g., reduced anxiety and feelings of depression) and self-efficacy (i.e., feelings of ability/mastery). Treatment foci on such psychological constructs (and their interactions) is consistent with social cognitive theory^10,11 and theories that support more granular aspects of that broad-reaching paradigm such as self-regulation theory,¹² self-efficacy theory,¹³ and the mood-behavior model.¹⁴ Although some improvements were found via earlier cognitive-behavioral approaches, their overall modest outcomes^15,16 stimulated increased use of more invasive processes such as weight loss (bariatric) surgery¹⁷ and anti-obesity medications.¹⁸ Although those medical interventions are often promoted as remedies for obesity requiring minimal effort on the part of patients,^19,20 each has notable health risks²¹ and can also be expensive and difficult to disseminate in a scalable manner.²² Even given clear challenges,²³ some researchers have persisted at seeking innovative methods to address the obesity problem through behavioral/psychosocial means in a large-scale and cost-effective manner.^24,25

An extended role for physical activity/exercise

In reviews of over 100 related studies, the use of PA/exercise was considered uniquely important for weight loss attainment.^26–29 Increased PA/exercise might be a vehicle for improvements in psychosocial correlates of improved eating behaviors such as mood,³⁰ self-regulation, and self-efficacy.^26,28,29 However, even theory-driven cognitive-behavioral weight loss treatments have often addressed PA/exercise as merely an optional adjunct to eating changes because of its minimal (direct) contribution to energy deficits.^15,30 Research indicates that improvements in both PA/exercise and the aforementioned psychological constructs are most pronounced in the initial 3 months of treatment, but they reliably carry-over to later and longer-term changes.²⁹ Both theory^{10,12–14,31} and research^{26,28,3233,34} suggest that (1) early changes in mood facilitate increases in both self-regulation and self-efficacy, (2) exercise-focused self-regulation and self-efficacy generalize to eating-related self-regulation and self-efficacy (and sustained weight loss via more controlled eating), and (3) use of newly learned self-regulation skills facilitate increased self-efficacy via feelings of mastery over goal-directed behaviors that are considered challenging (e.g., maintaining regular exercise).

Implications for physical activity/exercise amounts

Although less clear for those with a mental health disorder,³⁵ it was further indicated that ⩾3 light-moderate bouts of PA/exercise per week induce significant improvements in mood in sedentary and low-active adults, with no further improvements associated with more frequent bouts and greater intensities.²⁵ A synthesis of 30 related studies also called into question a PA/exercise dose–mood change relationship.³⁶ This favors the “psychological hypothesis” of the physical activity–mood change relationship over biochemically based hypotheses (e.g., “serotonin hypothesis,” “norepinephrine hypothesis”³⁰), which would require the presence of a dose–response relationship.³⁷ Rather, the psychological hypothesis suggests that factors such as self-efficacy and its interrelation with self-regulation mediate the PA/exercise–mood improvement relationship (i.e., feelings of ability and accomplishment are associated with a positive psychological/emotional/mental outlook^30,38,39).

Thus, evaluating the lower end of PA/exercise outputs for mood improvement is important because greater amounts, possibly impelled by physiologically driven guidelines for high energy expenditures for weight loss,⁴⁰ are associated with reduced adherence—a considerable problem most prevalent in women with obesity.^25,41 Based on the aforementioned and other findings,⁴² it is presently unclear whether increased PA/exercise can be leveraged for mood and other relevant psychosocial improvements when program participants regularly complete ⩾3 bouts/week of PA/exercise prior to starting their treatment program. It is possible that associated reductions in negative mood had already been maximized (i.e., floor effect), and thus, treatments might not be able to depend on the valuable mood improvements derived from increased PA/exercise in those already moderately (or greater) active. Such a finding would inform behavioral treatment content.

Research questions/hypotheses

In consideration of the aforementioned theories and studies, a research question was whether reduction in negative mood over 3 months would significantly differ among participant groupings where (the equivalent of) ⩾3 bouts/week of PA/exercise were/were not already regularly completed at treatment start (i.e., baseline). It was expected that PA/exercise would significantly increase across participants, and those changes would be significantly greater in groupings where <3 bouts/week were completed at baseline (as that group would have more available room for improvements). Additionally, it was expected that the 3-month increase in self-regulation skills usage would significantly mediate relations between participant groupings of completing/not completing 3 PA/exercise bouts at baseline and reduction in negative mood. Given expected relationships between changes in self-regulation and self-efficacy,^25,26,28 it was also posited that paths from grouping → change in self-regulation → change in self-efficacy → mood changes would be significant.

Women with obesity were incorporated because research suggests that females are more affected by mood in weight-management contexts than males.⁴³ Also, women are the more frequent enrollees in obesity treatments.⁴⁴ Thus, findings with them could be the most relevant. However, it was also previously indicated that outcomes in the present area of inquiry will likely generalize to men.⁹ It was hoped that the present field-based research of 3 months would help clarify appropriate usages and mechanisms of the exercise–mood improvement relationship, and help determine how they could best be leveraged within efficiently disseminated behavioral obesity treatments.

Materials and methods

Participants

Data were from a continuing research program in the United States focused primarily on behavioral obesity treatment outcomes (e.g., weight and body composition changes). Because the present aims and analyses were unique to this research, the reuse of participant data was justified. Women participants of ⩾21 years of age with obesity at baseline (confirmed by study staff measurement of their BMI at that time) were included. Each had a goal of weight loss. Additional inclusion criteria were: (1) no known contraindication for safe and complete participation within a local community-based health promotion center-based program, (2) no participation in another weight-management program within the previous 12 months, (3) no change in a psychotropic medication (including dosage) within the previous 6 months, and (4) no current/soon-planned pregnancy. Data were collected prior to restrictions associated with the COVID-19 pandemic. Demographic data on the participants are given in Table 1. For the current study, each participant was grouped, post hoc, based on whether she reported completing the equivalent of <3 light PA/exercise bouts/week at baseline (n = 53) or ⩾3 light PA/exercise bouts/week (n = 46), and (separately) the same for moderate PA/exercise bouts/week at baseline (i.e., n = 73 and n = 26, respectively). To facilitate the dichotomization of groupings, each participant’s Leisure-Time Physical Activity Questionnaire score⁴⁵ (described in the “Measures” section below) at baseline was used to separately assess the equivalent of 3 light bouts of PA/exercise/week (i.e., a score of 9) and the equivalent of 3 moderate bouts of PA/exercise/week (i.e., a score of 15) as cut points.

Table 1.

Demographic characteristic of the study sample (N = 99).

Age (years), M (SD)	46.4 (9.2)
BMI (kg/m²), M (SD)	35.0 (3.1)
Racial/ethnic group, n (%)
White	62 (62.6)
Black	32 (32.3)
Hispanic	5 (5.1)
Yearly family income, n (%)
<U.S.$25,000	15 (15.2)
U.S.$25,000–U.S.$49,999	43 (43.4)
U.S.$50,000–U.S.$99,999	39 (39.4)
⩾U.S.$100,000	2 (2.0)
Education level, n (%)
High school/GED	21 (21.2)
Some college	21 (21.2)
Bachelor’s degree	32 (32.3)
More than bachelor’s degree	25 (25.3)

GED: general education degree; BMI: body mass index; SD: standard deviation.

Ethical requirements of the World Medical Association Helsinki Declaration and the American Psychological Association were maintained. The Kennesaw State University Institutional Review Board approved the protocol (study 17173) and the informed consent form signed by each participant prior to their start of any study process.

Measures

Brief measures possessing adequate validity were sought and incorporated to minimize inaccuracies due to participant burden associated with completing multiple lengthier inventories.⁴⁶

PA/exercise was measured using the Leisure-Time Physical Activity Questionnaire.⁴⁵ Bouts of ⩾15 min completed during the “previous 7 days” (one word was altered from the original “typical ⁴⁷ 7 days” for purposes of this study) were recalled. Response options were “mild exercise (minimal effort; e.g., easy walking),” “moderate exercise (heartbeat faster than resting, but not exhausting; e.g., fast walking),” and “strenuous exercise (heart beats rapidly; e.g., running).” Those responses were assigned scores of 3, 5, and 9, respectively, based on their associated metabolic equivalent of task (MET; a measure of physical intensity). That score was multiplied by the corresponding number of bouts of that intensity completed within the previous 7 days, then summed. As examples: 3 bouts of mild exercise (3 × 3 METs) = 9; and 3 bouts of moderate exercise (3 × 5 METs) = 15. Concurrent validity of the Leisure-Time Physical Activity Questionnaire was indicated in previous research through its score correspondences with accelerometry and treadmill stress-test scores with test–retest reliability over 2 weeks at 0.74.^48–52 A recent review also deemed the Leisure-Time Physical Activity Questionnaire sensitive to cases where adverse physical conditions foster minimal levels of PA/exercise.⁵³

Total mood disturbance (TMD) was measured through use of the 30-item Profile of Mood States-B (brief form).⁵⁴ Reflecting over the previous 7 days, response options for each item ranged from 0 (not at all) to 4 (extremely) for each of its one- to three-word items (e.g., “dejected,” “anxious,” “vigorous”), which centered around dimensions of depression, anxiety, fatigue, anger, confusion, and vigor. Those were equally weighted, with the total of vigor item response scores subtracted from the sum of the other five proposed dimensions of overall negative mood. Thus, the possible score range for TMD was −20 to 100. Scores for the depression and anxiety scales of five items each were also separately calculated. Their possible score ranges were both 0–20. For each of the Profile of Mood States-B scores used in this research, a higher score indicated more negative mood. In women, earlier research indicated acceptable internal validity (mean Cronbach’s α = 0.90), with test–retest reliabilities over 3 weeks averaging 0.70.⁵⁴ Previous research also supported concurrent validity through Profile of Mood States score correspondences with well-validated, but lengthier, mood measures such as the Beck Depression Inventory (for depression), the Manifest Anxiety Scale (for anxiety), and the Minnesota Multiphasic Personality Inventory-2.⁵⁵ Associations between scores from the present brief form B and full 65-item Profile of Mood States were strong at β = 0.93–0.98.⁵⁶ Within the present sample, the mean Cronbach’s α = 0.83. A review indicated that Profile of Mood States score changes are sensitive to exercise stimuli across sample types.⁵⁷

Exercise-related self-regulatory skill usage was measured using the 10-item Exercise-Related Self-Regulation Scale (e.g., “I make formal agreements with myself to be physically active”).^21,58 Items reflected self-regulatory skills proposed by Michie et al.⁵⁹ as being consistent with social cognitive theory. Response options for each item ranged from 1 (never) to 4 (often), and were summed. The possible score range was 10–40, with a higher respondent score indicating their more frequent use of self-regulatory skills. Previous research reported an acceptable internal validity (Cronbach’s α = 0.81), with test–retest reliability over 2 weeks at 0.74.⁵⁸ Predictive validity was supported by associations between score increases and increased exercise.⁵⁸ Within the present sample, Cronbach’s α = 0.74.

Exercise-related self-efficacy was measured using the five-item Exercise Self-Efficacy Scale.⁶⁰ It measured confidence at persisting with PA/exercise under challenging conditions (e.g., “I feel I don’t have the time,” “I am tired”). Response options for each item ranged from 1 (not at all confident) to 11 (very confident) and were summed. The possible score range was 5–55, with a higher score indicating more self-efficacy in the respondent. Previous research reported acceptable internal consistencies of Cronbach’s α = 0.76–0.82, and test–retest reliabilities over 2 weeks at 0.74–0.78.⁶⁰ For the present sample, Cronbach’s α = 0.76.

Procedure

The present cognitive-behavioral obesity treatment consisted of components of PA/exercise support (baseline to month 6), and support of controlled eating (months 3–12). Because the focus of this investigation was baseline to month 3, the period where most PA/exercise gains are typically made,³⁰ the PA/exercise support processes within that time frame are given. Subsequent procedures not accounted for in this report are described elsewhere.⁶¹ The treatment content was based on tenets of social cognitive theory,^10,62 but the supporting paradigms of self-regulation theory¹² and self-efficacy theory¹³ additionally informed the treatment content.

After obtaining training in administration of the study treatment by research staff, sessions were administered by existing staff members of the community health promotion facilities agreeing to participate in the present research. There were four 50-min meetings (one instructor/one participant) conducted over the initial 3 months (weeks 1, 3, 7, and 11) held in private office settings. Although safe and acceptable PA/exercise was discussed, most session time was spent on the development of each participant’s self-regulatory skills, which included relapse prevention, cognitive restructuring, dissociation from discomfort, stimulus control, behavioral contracting, and short-term goal progress monitoring. There was instructor–participant collaboration on PA/exercise goal tracking during each session, and even small amounts of progress were highlighted there. The relationship between increased perceptions of competence toward adhering to regular exercise (i.e., increased self-efficacy) through regular usage of newly developed self-regulatory skills was a consistent treatment focus. Regarding PA/exercise amounts, although the governmental recommendation of at least 150 min/week of moderate or higher intensity bouts for health promotion was mentioned, it was suggested that any increase that was maintained could be productive. Thus, no specific exercise amount was assigned. The need for consistent PA/exercise was emphasized because its adherence, especially for women with obesity, has been challenging.^41,63

Non-instructional research staff conducted structured protocol fidelity checks on 15% of instructor-administered treatment sessions. Compliance was strong, and only minor adjustments were required to be presented to the instructors. Non-instructional study staff also administered the study measures to participants in a private area. Based on the ratio of intervention sessions attended/total number of possible sessions, participants’ mean adherence was 82%.

Data analyses

Based on established criteria,⁶⁴ there was no systematic bias in participants with versus without missing data (i.e., missing-at-random in the 9% of overall missing values that were each beyond baseline). That satisfied the established criterion for imputation using the expectation–maximization algorithm,^65,66 and facilitated an intention-to-treat format.^67,68 Variance inflation factor scores of <2.0 indicated acceptable multicollinearity,^69,70 and neither floor nor ceiling effects were observed prior to the analyses.

Mixed-model repeated measures analysis of variance (ANOVA) first assessed significance in mood changes from baseline to month 3 (separately for TMD, depression, and anxiety) and then evaluated whether those changes significantly differed by grouping (i.e., time × grouping effects). Separate analyses were conducted for groupings based on whether light PA/exercise at baseline or moderate PA/exercise at baseline was being considered. Coding for each was no = 0 (at least three sessions/week of PA/exercise was not completed at baseline), yes = 1 (at least three sessions/week of PA/exercise was completed at baseline). A similar analysis assessed PA/exercise change from baseline to month 3. Supplementary within-group (dependent) t tests were also calculated.

Using aggregated data, separate models that accounted for each mood measure and both PA/exercise intensity level grouping were fit. Specifically, significance of mediation of the association between the dichotomously coded PA/exercise groupings and the mood changes, by changes in exercise-related self-regulation, was assessed. Finally, change in exercise self-efficacy was added to the previous models as a second mediator in serial multiple mediation analyses. Based on recent suggestions,⁷¹ a significant relationship between the independent variable (grouping) and the dependent variable (mood change) was not first required to conduct those regression analyses.

Sensitivity analyses assessed the relationship of baseline PA/exercise and baseline moods, and the prediction of mood changes by initial moods, also accounting for light and moderate PA/exercise groupings. SPSS Statistics version 28.0.1.0 (IBM, Armonk, NY, USA) was used for statistical testing, incorporating the PROCESS 4.2 macroinstruction mediation models 4 and 6, with 10,000 percentile-based bootstrap resamples of the data.⁷¹ Statistical significance was set at α ⩽ 0.05 (two-tailed), and one-tailed where relationships had previously been established between PA/exercise and the tested psychosocial variables in similar contexts.⁷² Where bootstrapping was employed, a 95% confidence interval (95% CI) was used to assess significance. The Transparent Reporting of Evaluations with Nonrandomized Designs (TREND) guidelines⁷³ informed the reporting format.

Results

Contrasting baseline physical activity/exercise frequency-based groupings on changes in mood and physical activity/exercise

Table 2 provides descriptive statistics and ANOVA results for each mood and PA/exercise measure change from baseline to month 3 within the groupings, and for aggregated data. There was no significant difference in any mood measure at baseline, by PA/exercise intensity/frequency grouping. For TMD, depression, and anxiety, improvements from baseline to month 3 were significant overall, with large effect sizes in both the light and moderate PA/exercise groupings. None of the associated time × grouping effects were significant. For PA/exercise, improvements were also significant overall, with large effect sizes for both groupings. There was a significant time × grouping effect for only the moderate PA/exercise groupings, with greater increase found in the grouping reporting completion of <3 PA/exercise bouts/week at baseline. In supplementary within-grouping analyses, dependent t tests yielded significant improvements throughout (ps <0.001, except p = 0.011 on depression change and p = 0.020 on anxiety change for the PA/exercise grouping participants who reported ⩾3 moderate-intensity PA/exercise bouts/week at baseline). Effect sizes were large for all changes in TMD and PA/exercise (Cohen’s ds = 0.99–1.66) and moderate for depression and anxiety changes (Cohen’s ds = 0.49–0.76).

Table 2.

Mixed-model repeated measures analysis of variance results.

Measure Effect for time Baseline Month 3 ΔBaseline to month 3 Time group × effect
Grouping	F(1, 97)	p	η²_p	M	SD	M	SD	M	SD	F(1, 97)	p	η²_p
TMD
Light exercise (LTex) <3				25.81	15.63	7.34	12.88	−18.47	16.16
Light exercise (LTex) ⩾3	134.17	<0.001	0.58	21.52	14.19	3.83	11.93	−17.70	14.68	0.06	0.804	0.00
Moderate exercise (MODex) <3				23.96	15.31	5.05	12.03	−18.90	15.20
Moderate exercise (MODex) ⩾ 3	97.31	<0.001	0.50	23.42	14.61	7.54	13.86	−15.88	16.11	0.73	0.394	0.01
Aggregated data				23.82	15.06	5.71	12.51	−18.11	15.42
Depression
Light exercise (LTex) <3				4.58	3.63	2.28	2.52	−2.30	3.10
Light exercise (LTex) ⩾3	46.88	<0.001	0.33	3.67	3.15	2.04	2.55	−1.63	2.53	1.37	0.245	0.01
Moderate exercise (MODex) <3				4.42	3.65	2.14	2.32	−2.29	3.03
Moderate exercise (MODex) ⩾3	28.45	<0.001	0.23	3.42	2.63	2.27	3.07	−1.15	2.15	3.09	0.082	0.03
Aggregated data				4.16	3.43	2.17	2.52	−1.99	2.86
Anxiety
Light exercise (LTex) <3				5.28	4.24	2.70	3.24	−2.58	4.44
Light exercise (LTex) ⩾3	34.16	<0.001	0.26	4.78	2.97	2.89	2.41	−1.89	2.88	0.82	0.367	0.01
Moderate exercise (MODex) <3				5.07	3.94	2.51	2.93	−2.56	4.04
Moderate exercise (MODex) ⩾3	21.27	<0.001	0.18	5.00	2.95	3.58	2.61	−1.42	2.93	1.74	0.191	0.02
Aggregated data				5.05	3.69	2.79	2.87	−2.26	3.80
PA/ex
Light exercise (LTex) <3				2.13	2.67	30.79	16.90	28.66	17.29
Light exercise (LTex) ⩾3	205.19	<0.001	0.68	15.32	5.80	37.98	16.93	22.66	18.33	2.80	0.097	0.03
Moderate exercise (MODex) <3				4.23	4.17	33.09	17.43	28.86	17.42
Moderate exercise (MODex) ⩾3	137.43	<0.001	0.59	19.56	4.05	37.06	16.52	17.50	16.99	8.25	0.005	0.08
Aggregated data				8.26	7.93	34.13	17.21	25.87	17.94

Light exercise (LTex) <3 = baseline physical activity/exercise of less than three light-intensity sessions/week (a baseline score of <9 on the LTPAQ), n = 53. Light exercise (LTex) ⩾3 = baseline physical activity/exercise of at least three light-intensity sessions/week (a baseline score of ⩾9 on the LTPAQ), n = 46. Moderate exercise (MODex) <3 = baseline physical activity/exercise of less than three moderate-intensity sessions/week (a baseline score of <15 on the LTPAQ), n = 73. Moderate exercise (MODex) ⩾3 = baseline physical activity/exercise of at least three moderate-intensity sessions/week (a baseline score of ⩾15 on the LTPAQ), n = 26. Aggregated data, N = 99. Δ = change in score during the designated period. η²_p = partial eta-squared (SS_Effect/[SS_Effect + SS_Error]), where 0.01, 0.06, and 0.14 represent small, medium, and large effects, respectively. TMD: total mood disturbance; PA/ex: physical activity/exercise; LTPAQ: Leisure-Time Physical Activity Questionnaire; SD: standard deviation.

Mediations of relations between baseline physical activity/exercise frequency grouping and mood changes through changes in self-regulation

For the light PA/exercise groupings, 3-month change in (Δ) exercise-related self-regulation significantly mediated relations between grouping and ΔTMD, B = 2.56, SE_B = 1.64, 95% CI [0.025, 5.413]; Δ_depression, B = 0.37, SE_B = 0.25, 95% CI [0.008, 0.807]; and Δ_anxiety, B = 0.32, SE_B = 0.26, 95% CI [0.000, 0.765]. Each of those overall predictive models was significant, R² = 0.27, p < 0.001; R² = 0.18, p < 0.001; and R² = 0.08, p = 0.018, respectively (see Figure 1(a)–(c) for relationship data within the embedded paths).

Figure 1.

Mediation of self-regulation change in predictions of changes in (A) total mood disturbance, (B) depression, and (C) anxiety by light physical activity/exercise grouping (N = 99).

For the moderate PA/exercise groupings, Δ_exercise-related self-regulation significantly mediated relations between grouping and ΔTMD, B = 4.31, SE_B = 1.99, 95% CI [1.167, 7.781]; Δ_depression, B = 0.61, SE_B = 0.32, 95% CI [0.127, 1.173]; and Δ_anxiety, B = 0.52, SE_B = 0.37, 95% CI [0.076, 1.184]. Each of those predictive models was significant, R² = 0.27, p < 0.001; R² = 0.18, p < 0.001; and R² = 0.08, p = 0.016, respectively (see Figure 2(a)–(c) for embedded path data).

Figure 2.

Mediation of self-regulation change in predictions of changes in (A) total mood disturbance, (B) depression, and (C) anxiety by moderate physical activity/exercise grouping (N = 99).

Within each model (Figures 1 and 2), the relationship between Δ_self-regulation and Δ_mood was significant (after controlling for each of the other relationships across the included variables).

Serial mediation analyses of relations between baseline physical activity/exercise frequency grouping and mood changes through changes in self-regulation and self-efficacy

For the light PA/exercise groupings, the following paths were significant: grouping → Δ_exercise-related self-regulation → Δ_exercise self-efficacy → ΔTMD, B = 0.46, SE_B = 0.36, 95% CI [0.016, 1.153]; and grouping → Δ_exercise-related self-regulation → Δ_exercise self-efficacy → Δ_depression, B = 0.07, SE_B = 0.06, 95% CI [0.000, 0.187]. The following path was not significant: grouping → Δ_exercise-related self-regulation → Δ_exercise self-efficacy → Δ_anxiety, B = 0.03, SE_B = 0.06, 95% CI [−0.037, 0.138]. Each of those overall predictive models was significant, R² = 0.34, p < 0.001; R² = 0.22, p < 0.001; and R² = .09, p = 0.036, respectively (see Figure 3(a)–(c) for embedded path data).

Figure 3.

Serial mediation analyses of paths from light physical activity/exercise grouping to changes in (A) total mood disturbance, (B) depression, and (C) anxiety through changes in self-regulation and self-efficacy (N = 99).

For the moderate PA/exercise groupings, the following paths were significant: grouping → Δ_exercise-related self-regulation → Δ_exercise self-efficacy → ΔTMD, B = 0.79, SE_B = 0.45, 95% CI [0.161, 1.591]; and grouping → Δ_exercise-related self-regulation → Δ_exercise self-efficacy → Δ_depression, B = 0.12, SE_B = 0.09, 95% CI [0.016, 0.309]. The following path was not significant: grouping → Δ_exercise-related self-regulation → Δexercise self-efficacy → Δanxiety, B = 0.06, SE_B = 0.09, 95% CI [−0.055, 0.219]. Each of those overall predictive models was significant, R² = 0.34, p < 0.001; R² = 0.23, p < 0.001; and R² = 0.09, p = 0.032, respectively (see Figure 4(a)–(c) for embedded path data).

Figure 4.

Serial mediation analyses of paths from moderate physical activity/exercise grouping to changes in (A) total mood disturbance, (B) depression, and (C) anxiety through changes in self-regulation and self-efficacy (N = 99).

Within each model (Figures 3 and 4), the relationship between Δ_exercise-related self-regulation and Δ_exercise self-efficacy was significant (after controlling for each of the other relationships across the included variables).

Sensitivity analyses

There was no significant relationship between PA/exercise at baseline and baseline TMD, β = −0.16, p = 0.124; baseline depression, β = −0.18, p = 0.079; or baseline anxiety, β = −0.10, p = 0.352. There were significant relationships between TMD at baseline and ΔTMD, β = −0.66; baseline depression and Δ_depression, β = −0.69; and anxiety at baseline and Δ_anxiety, β = −0.71, ps < 0.001. The addition of grouping into step 2 of the above equations did not significantly contribute to the explained variance in ΔTMD, ps >0.300; Δ_depression, ps >0.200; or Δanxiety, ps >0.070.

Discussion

The present results informed behavioral obesity treatments on their usage of PA/exercise-associated mood improvements. Because intensities of completed PA/exercise were sometimes unclear in previous related research, analyses separately accounted for both light and moderate intensities. Supporting the corresponding hypothesis, PA/exercise of a moderate intensity increased more in the grouping that started initially lowest on that measure. That result is consistent with a key target of the present cognitive-behavioral treatment, which was increasing PA/exercise in sedentary and low-active adults.⁷² In evaluating the primary research question, it was found that the significant baseline to month 3 score reductions in TMD, depression, and anxiety did not significantly differ by grouping (nor did they significantly differ at baseline). This suggests that, although attaining three bouts/week of PA/exercise appeared to demonstrate a floor effect in regard to reducing negative mood in some previous research,²⁵ meaningful mood improvements still were observed across both tested groupings. Findings also indicated that those significant mood improvements might similarly occur whether the initial PA/exercise amounts are of a light or moderate intensity. Although exercise amounts of up to 420 min/week have been suggested within the scientific literature in the hopes of expediting weight loss via associated energy expenditures,⁷⁴ those directives might challenge adherence, especially in adults with obesity.^41,75 Thus, if PA/exercise is incorporated to foster improvements in psychosocial correlates of controlled eating—as has been indicated as advantageous⁷⁶—lesser amounts might also be valuable. This is consistent with suggestions that the greatest value of PA/exercise for weight loss is in its ability to induce psychosocial changes that are predictive of controlled eating.^9,25 Because the PA/exercise amount–mood change association was indirectly negated here, the psychological hypothesis of the effects of exercise on mood was supported.^30,37

Based on social cognitive theory,^10,62 and more specifically self-regulation theory¹² and associated research,^25,26,28 the hypothesis that changes in exercise-related self-regulation would mediate the exercise frequency-based grouping-mood change relationships was supported. This further bolstered the present treatment’s focus on building an array of self-regulatory skills across participants. Although requiring distinct testing, based on coaction theory³¹ and previous research,^33,77,78 those self-regulatory skills might also transfer (i.e., generalize) to supporting the eating behavior changes that are essential for weight loss and reductions in obesity. Also, both improved mood and increased self-regulation have been shown to be favorably associated with emotional eating reduction and overall control over eating during the long-term.⁷⁹

The final hypothesis was partially supported in that the significant prediction of heightened self-efficacy by increased self-regulation mediated the grouping → ΔTMD, and grouping → Δ_depression, relationships. However, the path from grouping → Δ_{self-regulation} → Δ_{self-efficacy} → Δ_anxiety was not significant. Although each of the overall predictive models were significant, it is unclear why anxiety had effects that differed somewhat from those of TMD and depression. It is possible that for anxiety change, a longer period is required to observe a significant path, as the present data indicated that directionality was consistent with the models incorporating TMD and depression changes. That finding should be further-explored in extensions of this research because anxiety is an important psychological predictor of controlled eating and sustained weight loss.⁸⁰ Notwithstanding that, findings of the significant paths support self-regulation theory,¹³ and how self-efficacy may be increased via participants’ enhanced mastery over their personal environmental challenges through usage of newly learned self-management skills.

Although various psychosocial constructs have been associated with adherence to PA/exercise and controlled eating, and their interrelations; improvements in mood, self-regulation, and self-efficacy have been primary.^9,81 They each, individually, have significantly contributed to the explained variance in both of those behavioral changes. Although it is presently unclear what psychological factor(s) holds the most value within behavioral obesity treatments, the present focus on self-regulatory skills development appeared to possess considerable merit for improving mood and increasing exercise-related self-efficacy. The findings also indicated prospects for further research regarding their carry over for sustaining improvements in eating behaviors.^25,31,78 Further research should aim to clarify precisely what self-regulation skill or skills are the most salient so that treatment time might be most efficiently allocated. It should be noted, however, that attention to any of the considered psychological variables assessed within this report could represent an advancement over the continued atheoretical reliance on providing suggestions to eat in a controlled manner (or maintain a specific “diet”), to exercise at as high a level as possible, and expect meaningful improvements to emerge from solely those directives.⁸²

Limitations

Although contributions were made within this study related to the development and improvement of the treatment of obesity through behavioral means, limitations should also be acknowledged. For example, given that the present study partitioned the full sample for some analyses and incorporated assessments of both group equivalence and mediation of effects, extensions/replications using similar analytic formats should better-account for sample size and power to increase confidence in the findings. Also, although exercise-related psychological changes over the short term have carried over to longer-term changes in eating behaviors and weight loss,^33,78,83 that was not directly tested within the confines of the present investigation. Thus, related studies of lengthier durations, that adequately address dietary changes via exercise-associated mood improvement, are required. Furthermore, the measurement of PA/exercise was reliant on participants’ recall and self-report and lacked a specific assessment of duration. More objective measurement, such as through accelerometry, might enhance accuracy in extensions of this research. Additionally, analyses of generalizations of findings to men—and across ages, ethnic/racial groups, socioeconomic strata, and medical issues (e.g., diabetes, reduced mobility)—are needed. Moreover, because groupings were established post hoc, this research was essentially a single-group design with all participants receiving the same standardized cognitive-behavioral treatment. In consideration of that, control conditions are needed to account for expectation, social support effects, and other possible experimental confounds associated with an interpersonally administered treatment,⁸⁴ and analyses should separately consider both the earlier stages where weight loss is expected, and beyond 6–9 months where regain of weight is projected.^4,5 Such longitudinal analyses conducted within field environments would also increase relevance regarding the role of PA/exercise in the psychological aspects of obesity treatment and serve to further inform scalable referral possibilities for the medical community.

Conclusion

Given the presently high prevalence of obesity,¹ and that the typical informational approaches (possibly lasting only several minutes once or twice a year in a medical professional’s office) fail,^7,8 health behavior-change researchers should seek to “package” protocols for their widespread application. As in this investigation, community health centers, YMCAs, and fitness/wellness centers provide an opportunity for that. Thus, although staffing has become minimal in such settings, we advocate for further research within those and similar environments. Possibly, referrals by physicians, nurse practitioners, and other medical professionals to such longitudinal programs will allow the present findings to help better address needed mechanisms for large-scale change. Based on the extant results, even manageable amounts of PA/exercise can become a basis for psychosocial changes that help to foster success with weight loss. Findings here also provide a rationale for why PA/exercise is one of the strongest predictors of sustained weight loss,^85,86 but this typically cannot be reconciled by its associated energy expenditures.

Footnotes

Acknowledgements

The authors acknowledge the participants and instructors involved in the present research.

ORCID iD

James J. Annesi

Ethical considerations

The Kennesaw State University Institutional Review Board approved the protocol (study 13173).

Consent to participate

The informed consent process required a signature from each participant prior to any study activity on their part.

Consent for publication

Not applicable.

Author contributions

James J. Annesi: Conceptualization; Writing – original draft; Formal analysis; Project administration.

Maliheh Bakhshi: Conceptualization; Writing – original draft.

Patricia G. Sevene: Conceptualization; Writing – original draft.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The dataset analyzed during the current study will be made available from the corresponding author upon reasonable request.

References

Stierman

Afful

Carroll

, et al. National Health and Nutrition Examination Survey 2017-March 2020 prepandemic data files-development of files and prevalence estimates for selected health outcomes. Natl Health Stat Report 2021; 2021(158): 10.15620/cdc:106273.

Pi-Sunyer

The medical risks of obesity. Postgrad Med 2009; 121(6): 21–33.

Fox

Feng

Asal

What is driving global obesity trends? Globalization or “modernization”?

Global Health 2019; 15(1): 32.

Dombrowski

Knittle

Avenell

, et al. Long term maintenance of weight loss with non-surgical interventions in obese adults: systematic review and meta-analyses of randomised controlled trials. BMJ 2014; 348: g2646.

MacLean

Wing

Davidson

, et al. NIH working group report: innovative research to improve maintenance of weight loss. Obesity (Silver Spring) 2015; 23(1): 7–15.

Snow

Barry

Fitterman

, et al. Pharmacologic and surgical management of obesity in primary care: a clinical practice guideline from the American College of Physicians. Ann Intern Med 2005; 142(7): 525–531.

Hartmann-Boyce

Theodoulou

Oke

, et al. Association between characteristics of behavioural weight loss programmes and weight change after programme end: systematic review and meta-analysis. BMJ 2021; 374: n1840.

Mann

Tomiyama

Westling

, et al. Medicare’s search for effective obesity treatments: diets are not the answer. Am Psychol 2007; 62(3): 220–233.

Greaves

Sheppard

Abraham

, et al. Systematic review of reviews of intervention components associated with increased effectiveness in dietary and physical activity interventions. BMC Public Health 2011; 11: 119.

10.

Bandura

Social foundations of thought and action: a social cognitive theory. Prentice Hall, 1986.

11.

Bandura

The primacy of self-regulation in health promotion. Appl Psychol 2005; 54(2): 245–254.

12.

Vohs

Baumeister

(eds). Handbook of self-regulation: research, theory, and application. 3rd ed. Guilford, 2017.

13.

Bandura

Self-efficacy: the exercise of control. Freeman, 1997.

14.

Gendolla

GHE

Brinkmann

. The role of mood states in self-regulation: effects on action preferences and resource mobilization. Eur Psychol 2005; 10(3): 187–198.

15.

Cooper

Fairburn

Hawker

DM.

Cognitive-behavioral treatment of obesity: a clinician’s guide. Guilford; 2003.

16.

LeBlanc

Patnode

Webber

, et al. Behavioral and pharmacotherapy weight loss interventions to prevent obesity-related morbidity and mortality in adults: updated evidence report and systematic review for the US Preventive Services Task Force. JAMA 2018; 320(11): 1172–1191.

17.

Alalwan

Friedman

Park

, et al. US national trends in bariatric surgery: a decade of study. Surgery 2021; 170(1): 13–17.

18.

Lewis

Moore

Ard

JD.

Game changers: do new medications make lifestyle-based treatment of obesity obsolete?

Obesity (Silver Spring) 2024; 32(2): 237–239.

19.

Hoek

Gendall

Advertising and obesity: a behavioral perspective. J Health Commun 2006; 11(4): 409–423.

20.

Klitzman

Greenberg

Anti-obesity medications: ethical, policy, and public health concerns. Hastings Cent Rep 2024; 54(3): 6–10.

21.

Cannon

Kumar

Treatment of overweight and obesity: lifestyle, pharmacologic, and surgical options. Clin Cornerstone 2009; 9(4): 55–71.

22.

Annesi

JJ.

Generalizability of an adult-validated model for predicting increased physical activity in after-school program children: effects of treatment formats on fitness indicators. Eval Program Plann 2019; 73: 33–43.

23.

Brownell

KD.

The humbling experience of treating obesity: should we persist or desist?

Behav Res Ther 2010; 48(8): 717–719.

24.

Teixeira

Carraça

Marques

, et al. Successful behavior change in obesity interventions in adults: a systematic review of self-regulation mediators. BMC Med 2015; 13: 84.

25.

Annesi

JJ.

Supported exercise improves controlled eating and weight through its effects on psychosocial factors: extending a systematic research program toward treatment development. Perm J 2012; 16(1): 7–18.

26.

Carraça

Silva

Coutinho

, et al. The association between physical activity and eating self-regulation in overweight and obese women. Obes Facts 2013; 6(6): 493–506.

27.

Dashti

Esfehani

Leonard Joseph

HJ.

The preventive effect of physical activity on weight maintenance in overweight and obese women. Clin Ter 2014; 165(3): 155–161.

28.

Teixeira

Silva

Coutinho

, et al. Mediators of weight loss and weight loss maintenance in middle-aged women. Obesity (Silver Spring) 2010; 18: 725–735.

29.

Annesi

Marti

CN.

Path analysis of exercise treatment-induced changes in psychological factors leading to weight loss. Psychol Health 2011; 26(8): 1081–1098.

30.

Arent

Walker

Arent

(2020). The effects of exercise on anxiety and depression. In: Tennenbaum

Eklund

(eds). Handbook of sport psychology. 4th ed. Wiley, 2020, pp.872–890.

31.

Johnson

Paiva

Mauriello

, et al. Coaction in multiple behavior change interventions: consistency across multiple studies on weight management and obesity prevention. Health Psychol 2014; 33(5): 475–480.

32.

Cooper

Doll

Hawker

, et al. Testing a new cognitive behavioural treatment for obesity: a randomized controlled trial with three-year follow-up. Behav Res Ther 2010; 48(8): 706–713.

33.

Mata

Silva

Vieira

, et al. Motivational “spill-over” during weight control: increased self-determination and exercise intrinsic motivation predict eating self-regulation. Health Psychol 2009; 28(6): 709–716.

34.

Annesi

Johnson

Porter

KJ.

Bi-directional relationship between self-regulation and improved eating: temporal associations with exercise, reduced fatigue, and weight loss. J Psychol 2015; 149(6): 535–553.

35.

Dunn

Trivedi

Kampert

, et al. Exercise treatment for depression: efficacy and dose response. Am J Prev Med 2005; 28(1): 1–8.

36.

Mammen

Faulkner

Physical activity and the prevention of depression: a systematic review of prospective studies. Am J Prev Med 2013; 45(5): 649–657.

37.

Dishman

RK.

Physical activity and public health: mental health. Quest 1995; 47(3): 362–385.

38.

Dishman

RK.

Psychological factors and physical activity level. In: Bouchard

Katzmarzyk

(eds). Physical activity and obesity. 2nd ed. Human Kinetics, 2010, pp.89–92.

39.

Yuan

Liwei

Zhixiong

Physical exercise and mental health: the effect of emotion regulation self-efficacy and emotion regulation strategy. Studies Psychol Behav 2018; 16(4): 570–576.

40.

Donnelly

Blair

Jakicic

, et al. American College of Sports Medicine Position Stand. Appropriate physical activity intervention strategies for weight loss and prevention of weight regain for adults Med Sci Sports Exerc 2009; 41(2): 459–471.

41.

Trost

Owen

Bauman

, et al. Correlates of adults’ participation in physical activity: review and update. Med Sci Sports Exerc 2002; 34(12): 1996–2001.

42.

Magnan

Kwan

Bryan

AD.

Effects of current physical activity on affective response to exercise: physical and social-cognitive mechanisms. Psychol Health 2013; 28(4): 418–433.

43.

Péneau

Ménard

Méjean

, et al. Sex and dieting modify the association between emotional eating and weight status. Am J Clin Nutr 2013; 97(6): 1307–1313.

44.

Crane

Jeffery

Sherwood

NE.

Exploring gender differences in a randomized trial of weight loss maintenance. Am J Mens Health 2017; 11(2): 369–375.

45.

Godin

The Godin-Shephard Leisure-Time Physical Activity Questionnaire. Health Fitness J Can 2011; 4(1): 18–22.

46.

Galesic

Bosnjak

Effects of questionnaire length on participation and indicators of response quality in a web survey. Public Opin Q 2009; 73(2): 349–360.

47.

Ainsworth

Haskell

Herrmann

, et al. 2011 Compendium of Physical Activities: a second update of codes and MET values. Med Sci Sports Exerc 2011; 43(8): 1575–1581.

48.

Amireault

Godin

The Godin-Shephard Leisure-Time Physical Activity Questionnaire: validity evidence supporting its use for classifying healthy adults into active and insufficiently active categories. Percept Mot Skills 2015; 120(2) :604–622.

49.

Amireault

Godin

Lacombe

, et al. Validation of the Godin-Shephard Leisure-Time Physical Activity Questionnaire classification coding system using accelerometer assessment among breast cancer survivors. J Cancer Surviv 2015; 9(3): 532–540.

50.

Jacobs

Ainsworth

Hartman

, et al. A simultaneous evaluation of 10 commonly used physical activity questionnaires. Med Sci Sports Exerc 1993; 25(1): 81–91.

51.

Miller

Freedson

Kline

GM.

Comparison of activity levels using the Caltrac accelerometer and five questionnaires. Med Sci Sports Exerc 1994; 26(3): 376–382.

52.

Pereira

FitzerGerald

Gregg

, et al. A collection of physical activity questionnaires for health-related research. Med Sci Sports Exerc 1997; 29(6 Suppl): S1–S205.

53.

Sikes

Richardson

Cederberg

, et al. Use of the Godin Leisure-Time Exercise Questionnaire in multiple sclerosis research: a comprehensive narrative review. Disabil Rehabil 2019; 41(11): 1243–1267.

54.

McNair

Heuchert

JWP

. Profile of Mood States technical update. Multi-Health Systems, 2009.

55.

McNair

Lorr

Droppleman

LF.

Manual for the profile of mood states. Education and Industrial Testing Service, 1992.

56.

Jackson

Which mood state assessments? Cross validation of short mood measurement techniques. Med Sci Sports Exerc 1997; 29(Suppl 5): 188.

57.

Berger

Motl

RW.

Exercise and mood: a selective review and synthesis of research employing the Profile of Mood States. J Appl Sport Psychol 2000; 12(1), 69–92.

58.

Annesi

JJ.

Relationships between self-regulation skills and physical activity and fruit and vegetable consumption in obese adults: mediation of mood and self-efficacy. Psychol Rep 2011; 108(1): 95–103.

59.

Michie

Ashford

Sniehotta

, et al. A refined taxonomy of behaviour change techniques to help people change their physical activity and healthy eating behaviours: the CALO-RE taxonomy. Psychol Health 2011; 26(11): 1479–1498.

60.

Marcus

Selby

Niaura

, et al. Self-efficacy and the stages of exercise behavior change. Res Q Exerc Sport 1992; 63(1): 60–66.

61.

National Institutes of Health/National Cancer Institute. Evidence-Based Cancer Control Programs: obesity management, https://ebccp.cancercontrol.cancer.gov/archive.do (2025, accessed 22, September 2025).

62.

Bandura

Health promotion by social cognitive means. Health Educ Behav 2004; 31(2): 143–164.

63.

del Rey-Moya

Castilla-Álvarez

Pichiule-Castañeda

, et al. Effect of a group intervention in the primary healthcare setting on continuing adherence to physical exercise routines in obese women. J Clin Nurs 2013; 22(15–16): 2114–2121.

64.

White

Horton

Carpenter

, et al. Strategy for intention to treat analysis in randomised trials with missing outcome data. BMJ 2011; 342: d40.

65.

Ding

Song

PX-K

. EM algorithm in Gaussian copula with missing data. Comput Stat Data Anal 2016; 101(C): 1–11.

66.

Little

Rubin

DB.

Statistical analysis with missing data. 2nd ed. Wiley, 2014.

67.

Gupta

SK.

Intention-to-treat concept: a review. Perspect Clin Res 2011; 2(3): 109–112.

68.

Montori

Guyatt

GH.

Intention-to-treat principle. CMAJ 2001; 165(10): 1339–1341.

69.

Cohen

West

, et al. Applied multiple regression/correlation analysis for the behavioral sciences. 3rd ed. Erlbaum; 2003.

70.

Thompson

Kim

Aloe

, et al. Extracting the variance inflation factor and other multicollinearity diagnostics from typical regression results. Basic Appl Soc Psych 2017; 39(2): 81–90.

71.

Hayes

AF.

Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. 3rd ed. Guilford, 2022.

72.

Annesi

JJ.

Physical activity in the treatment of obesity: a marker of psychosocial predictors of controlled eating, or facilitator of their improvements in women with differing body images. Minerva Psichiatrica 2017; 58(3): 144–155.

73.

Des Jarlais

Lyles

Crepaz

, et al. Improving the reporting quality of nonrandomized evaluations of behavioral and public health interventions: the TREND statement. Am J Public Health 2004; 94(3): 361–366.

74.

Swift

McGee

Earnest

, et al. The effects of exercise and physical activity on weight loss and maintenance. Prog Cardiovasc Dis 2018; 61(2): 206–213.

75.

Dalle Grave

Calugi

Centis

, et al. Cognitive-behavioral strategies to increase the adherence to exercise in the management of obesity. J Obes 2011; 2011: 348293.

76.

Joseph

Alonso-Alonso

Bond

, et al. The neurocognitive connection between physical activity and eating behaviour. Obes Rev 2011; 12(10): 800–812.

77.

Fleig

Lippke

Pomp

, et al. Intervention effects of exercise self-regulation on physical exercise and eating fruits and vegetables: a longitudinal study in orthopedic and cardiac rehabilitation. Prev Med 2011; 53(3): 182–187.

78.

Oaten

Cheng

Longitudinal gains in self-regulation from regular physical exercise. Br J Health Psychol 2006; 11(Pt 4): 717–733.

79.

Ling

Zahry

NR.

Relationships among perceived stress, emotional eating, and dietary intake in college students: eating self-regulation as a mediator. Appetite 2021; 163: 105215.

80.

Cifuentes

Campos

Silgado

MLR

, et al. Association between anxiety and eating behaviors in patients with obesity. Obes Pillars 2022; 3: 100021.

81.

Baker

Brownell

KD.

Physical activity and maintenance of weight loss: physiological and psychological mechanisms. In: Bouchard

(ed). Physical activity and obesity. Human Kinetics, 2000, pp.311–328.

82.

Kim

TN.

Barriers to obesity management: patient and physician factors. J Obes Metab Syndr 2020; 29(4): 244–247.

83.

Annesi

Mareno

Psychosocial changes as correlates of weight regain vs. continued loss within 2-year trials of a self-regulation-focused community-based intervention. Clin Obes 2017; 7(1): 22–33.

84.

Rosenthal

Experimenter and clinician effects in scientific inquiry and clinical practice. Prev Treat 2002; 5(1): 38.

85.

Fogelholm

Kukkonen-Harjula

Does physical activity prevent weight gain—a systematic review. Obes Rev 2000; 1(2): 95–111.

86.

Svetkey

Stevens

Brantley

, et al. Comparison of strategies for sustaining weight loss: the weight loss maintenance randomized controlled trial. JAMA 2008; 299(10): 1139–1148.

Reductions in negative mood in minimally versus moderately active women at behavioral obesity intervention start: Assessing floor effects and internal mechanisms of change to inform future treatments

Abstract

Background:

Objectives:

Design:

Methods:

Results:

Conclusion:

Keywords

Introduction

Adjusting behavioral obesity-treatment contents

An extended role for physical activity/exercise

Implications for physical activity/exercise amounts

Research questions/hypotheses

Materials and methods

Participants

Measures

Procedure

Data analyses

Results

Contrasting baseline physical activity/exercise frequency-based groupings on changes in mood and physical activity/exercise

Mediations of relations between baseline physical activity/exercise frequency grouping and mood changes through changes in self-regulation

Serial mediation analyses of relations between baseline physical activity/exercise frequency grouping and mood changes through changes in self-regulation and self-efficacy

Sensitivity analyses

Discussion

Limitations

Conclusion

Footnotes

Acknowledgements

ORCID iD

Ethical considerations

Consent to participate

Consent for publication

Author contributions

Funding

Declaration of conflicting interests

Data availability statement

References