Improving lifestyle obesity treatment during the COVID‐19 pandemic and beyond: New challenges for weight management

Abstract Objective The COVID‐19 pandemic has resulted in significant changes to daily life and many health‐related behaviors. The objective of this study was to examine how the stay‐at‐home/safer‐at‐home mandates issued in Colorado (March 2020–May 2020) impacted lifestyle behaviors and mental health among individuals with overweight or obesity participating in two separate behavioral weight loss trials (n = 82). Methods Questionnaires were used to collect qualitative and quantitative data on challenges to weight loss presented by the COVID‐19 pandemic, including changes in dietary intake, physical activity, sedentary behavior, and mental health during the stay‐at‐home/safer‐at‐home mandates. Results Using a convergent mixed method approach integrating qualitative and quantitative data, the greatest challenge experienced by participants was increased stress and anxiety, which led to more unhealthy behaviors. The majority perceived it to be harder to adhere to the prescribed diet (81%) and recommended physical activity (68%); however, self‐reported exercise on weekdays increased significantly and 92% of participants lost weight or maintained weight within ±1% 5–6 weeks following the stay‐at‐home mandate. Conclusion Study results suggest that obesity treatment programs should consider and attempt to address the burden of stress and anxiety stemming from the COVID‐19 pandemic and other sources due to the negative effects they can have on weight management and associated behaviors.


| INTRODUCTION
Obesity significantly increases the risk for severe COVID-19 illness, hospitalization, and mortality. [1][2][3] Thus, the COVID-19 pandemic has increased the urgency to effectively reduce obesity and related comorbidities (e.g., type 2 diabetes, hypertension) while concurrently presenting substantial challenges to behaviors that facilitate weight management, 4-6 particularly for adults with overweight or obesity. 3,7,8 It is important to understand how the unprecedented social and institutional efforts to mitigate the spread of the virus have affected diet and physical activity behavior in persons with overweight or obesity, in order to apply insights gained from this collective stress and trauma to inform obesity treatment moving forward.
Within the past year, several recent studies have been published that report on the impact of the COVID-19 outbreak on adults with overweight or obesity. 3,[9][10][11][12][13][14] The majority of these publications highlight the importance of mental health as well as the challenges in adhering to behavioral recommendations for achieving weight loss. For example, a recent paper examining the effects of the COVID-19 outbreak on health behaviors among a large international sample (n = 7,753) found that individuals with obesity reported higher incidence of weight gain and sharper declines in mental health following the COVID-19 outbreak compared to respondents with healthy weight or overweight. 9 In addition, an online survey of 250 people enrolled in health-related interventions observed high rates of moderate to severe symptoms of anxiety/depression (30%) and symptoms of moderate to severe post-traumatic stress disorder (68%) that influenced respondents' ability to adhere to behavioral recommendations in their intervention. 10 Another study surveyed 123 patients with obesity from an obesity medicine clinic and bariatric surgery practice during the stay-at-home order, and found that 73% reported increased anxiety, 84% reported increased depression, and 70% reported more difficulty achieving weight loss goals. 11 Lastly, among participants of an internet-based behavioral weight loss program in the Northeastern United States, 77% reported experiencing moderate to extreme stress during the stay-at-home orders, and stress levels were significantly associated with having less time to spend on weight-loss efforts. 12 However, no known study has specifically reported the impact of the COVID-19 pandemic on individuals with overweight or obesity enrolled in in-person behavioral weight loss interventions.
The associations between obesity and stress, anxiety, and mental health were recognized prior to the COVID-19 pandemic. 15,16 However, the widespread and dramatic increases in stress and anxiety during the pandemic further highlighted the importance of mental health for weight management and present a natural experiment that provides important insights on how to improve behavioral obesity treatment moving forward.
The objective of this study was to examine the acute effects of the COVID-19 pandemic and stay-at-home/safer-at-home orders issued in Colorado (26 March 2020/26 April 2020) on adults with overweight or obesity actively participating in an in-person behavioral weight loss trials to inform potential adaptations that may be necessary to effectively treat obesity in the wake of the COVID-19 -33 including a dietary prescription, goal to increase physical activity, and group-based behavioral support classes lasting 60-75 min. Classes were shifted from in-person to video conference on March 24 th in DRIFT-2 and March 18 th in TRE-Study. Both trials were ongoing, thus, comparisons between intervention arms were not conducted.
DRIFT-2 is a 12-month interventional trial designed to compare the weight loss efficacy of weekly energy restriction (∼34% from baseline energy requirements) through daily caloric restriction (DCR) or intermittent fasting (IMF; 3 non-consecutive days/week of 80% energy restriction from baseline requirements). Participants receive a recommendation to increase physical activity gradually up to 300 min/week by month 6, and to maintain this level for the duration of the trial. Group-based classes take place weekly (months 0-3) and bi-weekly (months 4-12), with no classes in the follow-up phase (months [13][14][15][16][17][18]. The TRE-Study is a 9-month pilot study designed to examine the feasibility and acceptability of DCR (∼35% energy deficit from baseline energy requirements) plus time-restricted eating (TRE; instructions to eat within a 10-h window starting within 3 h of waking) compared to DCR alone. Participants receive a physical activity recommendation of 150 min/week of moderate intensity physical activity for the full 9-month intervention. Group-based classes take place weekly (months 0-3) and monthly (months 4-9).
Cohort 1 of DRIFT-2 and cohorts 1 and 2 of TRE-Study had already completed the study when the stay-at-home order was issued. Cohort 2 of DRIFT-2 was in the follow-up phase (intervention week 57), while cohort 3 was attending bi-weekly classes (week 18) and cohort 3 of TRE-Study was attending weekly classes (week 7). This analysis was therefore limited to cohorts 2-3 of DRIFT-2 and cohort 3 of TRE-Study ( Figure 1).

This study was approved by the Colorado Multiple Institutional
Review Board at the CU-AMC.

| Diet behaviors
The internal survey asked participants to rate the extent to which the COVID-19 pandemic impacted their ability to adhere to the prescribed study diet during the past 30 days on a scale from 1 = much easier to 7 = much harder. To assess changes in dietary behaviors, the PBCS included a modified Rapid Eating Assessment (REAP-s) 20 scale to assess unhealthy dietary patterns. Participants are asked the frequency of engaging in eight unhealthy dietary behaviors during an average week 'prior to' and 'following the pandemic outbreak' (e.g., breakfast skipping, consuming <2 servings of fruits and vegetables, eating fast food, etc.) on a scale from 1 = usually/often, 2 = sometimes, 3 = rarely/never. Five of the eight items were used to compute scale totals while three questions on skipping breakfast and inadequate intake of fruits/vegetables each day were omitted due to the fasting components of the intervention designs. Scores ranged from 5 to 15, with higher scores indicating healthier diets.

| Physical activity and sedentary behavior
The internal questionnaire asked participants to rate the extent to which the COVID-19 pandemic impacted their ability to adhere to the prescribed physical activity targets during the past 30 days on a scale from 1 = much easier to 7 = much harder. The PBCS assessed changes in physical activity and sedentary behaviors on weekdays and weekends "prior to" and "following the pandemic outbreak."

| Mental health
The PCBS asked participants to rate the extent to which they generally felt more stressed, anxious, and sad following the COVID-19 outbreak on a 5-point Likert scale from 1 = strongly disagree to 5 = strongly agree, and included the 7-item generalized anxiety disorder (GAD-7) 21 scale. Participants rated anxiety symptoms "prior to" and "following the COVID-19 outbreak." Scores ranged from 0 to 21 with higher scores indicating greater anxiety.

| Measured weight change
Pre-COVID-19 weight was measured by study staff at the last inperson class prior to the stay-at-home order in DRIFT-2 and TRE-Study (intervention weeks 18 and 7, respectively). During-COVID-19, weights were measured by participants using home scales due to University-wide restrictions on clinical research. DRIFT-2 participants used ©BodyTrace smart scales (Palo Alto, CA) that wirelessly transmit weight data to a secure website accessible by researchers 6 weeks following the stay-at-home order (intervention week 24). TRE-Study participants used their own home scales and sent a photograph of the scale to researchers 5 weeks after the stay-at-home order was issued (intervention week 12).

| Quantitative analysis
Baseline demographic and clinical characteristics were summarized using descriptive statistics. Descriptive statistics were also performed for the single-item survey questions assessing perceptions of study participation, adherence, and mental health following the COVID-19 outbreak and stay-at-home order. To assess changes in anxiety (GAD-7), unhealthy eating (REAP-s), and physical activity/ sedentary behaviors reported both pre-and during-COVID-19, Wilcoxon signed-rank tests were performed due to the nonnormality of score distributions and/or the ordinal scales used.
To compare self-reported versus measured weight change, a paired-samples t-test was performed on weights prior to and 5-6 weeks after the stay-at-home order issuance. Median (interquartile range) and mean (standard deviation) are reported where relevant. In post-hoc analyses, linear regression models were used to determine if anxiety levels during COVID-19 were associated with participant characteristics, changes in stress or sadness, and weight loss behaviors (Model 1). In Model 2, analyses were adjusted for education and BMI. The alpha level was set at p ≤ 0.05.

| Participant characteristics
Participant characteristics are summarized in Table 1. Response rates for each measure by study and cohort are detailed in Figure 1.

| Overall impact of the COVID-19 outbreak and stay-at-home order
Participants were affected by transitioning to working from home (63%), increased childcare responsibilities with school and daycare closures (39%), and increased work hours (26%). A smaller portion of the sample was furloughed or had reduced work hours due to COVID-19 (16%), and 6% became unemployed. Within the first month following the stay-at-home mandate, none of the participants reported COVID-19 infection, however 4% had a family member sick CALDWELL ET AL.

| Overall qualitative results on challenges to weight loss progress and study participation
In reviewing all data from both open-ended questions (122 responses), three primary categories of difficulties related to study participation and achieving weight loss goals became apparent: (1) Mental health, specifically stress and anxiety, (2) dietary intakespecifically access to food and food consumption, and (3) changes in physical activity. A fourth category was noted, in which participants shared comments and observations on how the guidance and requirements associated with study participation influenced their behavior and experience with the COVID-19 pandemic. Each category is examined more thoroughly below through presentation of both qualitative and quantitative results.

| Quantitative results
Among respondents of the PBCS, the majority either somewhat or strongly agreed that following the COVID-19 outbreak they felt more stressed, anxious, or sad ( Figure 2). Generalized anxiety scores increased significantly from 3.0 (1.0-6.0) pre-COVID-19 to 6.5 outbreak, no participants met the criteria for high anxiety (GAD score ≥ 15), and 35% reported that their anxiety symptoms made their ability to work, take care of things, or get along with other people somewhat difficult. Following the COVID-19 pandemic, five respondents (11%) met criteria for high anxiety and most (76%) reported difficulty with work, taking care of things, or getting along with people due to anxiety symptoms ( Figure 2).

| Qualitative results
The greatest challenge to weight loss identified by participants was a significant increase in stress and anxiety as they adjusted to living in pandemic conditions (84 mentions). In addition to a general anxiety regarding COVID-19 itself, many participants identified the stay-athome order as a root cause of their stress. The stay-at-home order greatly impacted participants' daily routines, causing stress and anxiety as they tried to adjust to new circumstances. As one partic-

| Quantitative results
Among respondents to the internal survey, the overwhelming majority (81%) perceived that it was harder to adhere to their prescribed diet following the stay-at-home order, while 9% said there was no change, and 10% said it was easier (Figure 3).

| Quantitative results
Half of participants reported that changing classes from in-person to video conference negatively impacted their enjoyment of classes, while the other half reported enjoying video conference classes the same or more. Among respondents to the PBCS, 40% perceived that their weight stayed the same following the COVID-19 outbreak, while 25% perceived that they had lost weight and 34% perceived they had gained weight. However, these perceptions were not supported by the measured weight data which showed that the overwhelming majority lost >1% of their body weight (68%) or maintained weight within ±1% (24%), with just 8% (n = 5) gaining >1% body weight during the strictest stay-at-home/safer-at-home period of the pandemic (

| Qualitative results
Several participants reported finding participation in the study helpful during this difficult time (14 mentions 18 and 7, respectively). During COVID-19, weights were measured by participants using home scales due to University-wide restrictions on clinical research. DRIFT-2 participants used ©BodyTrace smart scales (Palo Alto, CA) that wirelessly transmit weight data to a secure website accessible by researchers 6 weeks following the stay-at-home order (intervention week 24). TRE-Study participants used their own home scales and sent a photograph of the scale to researchers 5 weeks after the stay-at-home order was issued (intervention week 12). CALDWELL ET AL. -39

| Association of anxiety with mental health, weight loss behaviors and weight change
Liner regression models were used to examine the relationship between GAD-7 anxiety levels during the COVID-19 pandemic with demographic variables, mental health variables, diet, physical activity and sedentary variables, as well as weight change (Table 3) The lower anxiety scores in the present sample likely reflect screening for major depression and other significant mental health disorders prior to study participation. With the added stress and anxiety came a loss of time and energy, and a corresponding increase in difficulty performing weight loss behaviors.
Dietary intake was more negatively impacted by the COVID-19 outbreak and stay-at-home order than physical activity in both the qualitative and quantitative results. The qualitative analysis revealed that easy access to food at home led many participants to partake in Notwithstanding the many barriers and hardships related to the COVID-19 pandemic and resultant stay-at-home order, several participants found study participation helpful during this time. It helped to provide a sense of structure and accountability. Though nearly all participants experienced negative impacts on their weight-loss journey due to the COVID-19 pandemic, most seemed hopeful that they would be able to get back on track once the worst of it was over.
Changing the format of classes from in-person to video conference was perceived as neutral or increased enjoyment for half of the sample, while the other half enjoyed in-person classes more. This suggests that in-person group-based support is valued for many individuals but offering a video conference option will appeal to others.
Despite study participation, 34% of the sample perceived they had gained weight following the stay-at-home order, a similar proportion to that observed in a large, international sample of individuals with obesity not participating in a weight loss study. 9 However, measured weights contradicted this finding, showing that a smaller proportion actually gained weight and among them, the magnitude of weight gain was small. The effects of the pandemic may have slowed the weight loss participants were experiencing prior to the pandemic, leading to higher perceptions of weight gain, but mean weight was significantly reduced over the 5-6 weeks following the stay-at-home order, and just 5 participants (8%) experienced weight gain >1%.
Qualitative and quantitative findings from the current study complement those of a recently published study examining stress and weight loss behaviors among adults enrolled in online weight loss program during the same time-frame. 12  pandemic may be different than others not attempting to lose weight, across race and ethnic subgroups, or across various regions in the US or around the world. This is particularly important because obesity disproportionately affects racial/ethnic minority groups, 25,26 and Black and Latino individuals bear a disproportionate burden of COVID-19 infection and poorer outcomes resulting from COVID-19. 27,28 It is well-recognized that the racial and ethnic groups represented in lifestyle weight loss studies are often not representative of the U.S. population. 29 However, the present findings on the pervasive impact of the COVID-19 pandemic on mental health are in line with studies with larger samples, 9,10 and those in similar sized samples in different regions (12,23), and those who were not necessarily trying to lose weight. 23 Nonetheless, future behavioral weight loss trials should include strategies to improve recruitment of individuals from minority groups to better determine the potential public health impact of these interventions in groups that might benefit most from weight loss. In addition, data on pre-COVID-19 behaviors and mental health were collected retrospectively. However, the changes in daily life in response to the stay-at-home order were both significant and rapid, which should have allowed partici- may similarly affect individuals' daily behaviors and bandwidth to focus on weight management and health. The economic recession resulting from the COVID-19 pandemic is expected to continue for the foreseeable future and increased financial stress will be common for years to come. Among those who worked from home during the stay-athome/safer-at-home orders in the present study sample, increased access to food throughout the day presented a major challenge to weight management. It has been estimated that 25%-30% of the workforce will continue to work from home following the pandemic. 30 Therefore, specific modules on managing stress, stress eating, and how to make the home environment more conducive to healthy eating may need to be emphasized in lifestyle obesity treatment programs during the ongoing COVID-19 pandemic and beyond.