Abstract
Objectives:
Look AHEAD previously reported that intensive lifestyle intervention (ILI) reduced incident depressive symptoms and improved health-related quality of life (HRQOL) over nearly 10 years of intervention, compared with a control group (diabetes support and education, DSE) in participants with type 2 diabetes and overweight or obesity. The present study compared incident depressive symptoms and changes in HRQOL in these groups for an additional 6 years following termination of the ILI in September 2012.
Methods:
1945 ILI and 1900 DSE participants completed at least one of four planned post-intervention assessments at which weight, mood (Patient Health Questionnaire-9), anti-depressant medication (ADM) use, and HRQOL (SF-36) were measured.
Results:
ILI and DSE participants lost (mean±SE) 3.1±0.3 and 3.8±0.3 kg (p=0.10), respectively, during the 6-year post-intervention follow-up. No significant differences were observed between groups during this time in incident mild or greater symptoms of depression or ADM use, or in changes on the physical component (PCS) or mental component summary (MCS) scores of the SF-36. In both groups, MCS scores were higher than PCS scores.
Conclusions:
Prior participation in ILI, compared with DSE, did not appear to improve subsequent mood or HRQOL during 6 years of post-intervention follow-up.
Individuals with type 2 diabetes are at risk of depression and impaired health-related quality of life (HRQOL) (1–4). In addition to the personal suffering it imposes, depression in persons with type 2 diabetes is associated with reduced disease self-management behaviors, higher rates of diabetes-related complications, and increased health-care costs (5–9). Physical and psychosocial complications of diabetes limit individuals’ vocational and recreational opportunities, as well as activities of daily living (3,6). Obesity also is common in persons with type 2 diabetes and is independently associated with increased risks of cardiovascular disease, depression, and impaired quality of life (10–14). Thus, weight management is strongly recommended for persons with type 2 diabetes and overweight/obesity (15–16).
The Look AHEAD (Action for Health in Diabetes) study has previously demonstrated the benefits of an intensive lifestyle intervention (ILI), designed to induce moderate weight loss and increase physical activity, in reducing the risk of symptoms of depression and in maintaining HRQOL in this population (17–18). Adults with type 2 diabetes and overweight/obesity who participated in an ILI for a median of 9.6 years had a 15% lower incidence of mild or greater symptoms of depression over this time and maintained significantly better physical HRQOL than participants randomly assigned to a control group, referred to as diabetes support and education (DSE). The greatest benefits of the ILI on depressive symptoms and HRQOL were observed in the first year when participants in this group lost 8.6% of baseline weight, compared with 0.7% for DSE (19).
Look AHEAD’s randomized intervention was stopped in September 2012, based on a futility analysis that found no significant differences between the ILI and DSE groups on the study’s primary outcome, a composite of death from cardiovascular causes; nonfatal myocardial infarction or stroke; or hospitalization for angina (20). With the termination of the ILI, Look AHEAD was converted to an observational cohort study to assess the potential long-term or late-emerging benefits of ILI, compared to DSE, on numerous outcomes, including all-cause mortality and cognitive function. Favorable legacy effects have been observed in prior studies following treatment termination, including the Diabetes Prevention Program (DPP) and the combined Diabetes Control and Complications Trial (DCCT)/Epidemiology of Diabetes Incidence and Complications (EDIC) (21–24).
The present study investigated whether there were persistent effects of Look AHEAD’s ILI on symptoms of depression and HRQOL, as assessed for up to 6 years after termination of the intervention. The study focused exclusively on post-intervention changes in these variables to assess whether prior participation in the ILI protected participants against impairments in mood and HRQOL going forward. We hypothesized that, compared to DSE, participants originally assigned to ILI would have a lower incidence of symptoms of mild or greater depression, as well as better self-reported HRQOL, during the observational follow-up period. We also examined the effects of sex, age, and race/ethnicity on these outcomes.
METHODS
Participants and Procedures
A total of 5,145 men and women enrolled in Look AHEAD from 2001–2004 at 16 centers across the U.S. (20). Eligible individuals had type 2 diabetes, were 45–76 years of age, and had a BMI ≥25 kg/m2 (or ≥27 kg/m2 if taking insulin). Additional eligibility criteria, study methods, and the primary results have been reported previously (17,20). Participants signed a consent form approved by their center’s institutional review board and were randomly assigned with equal probability to ILI or DSE.
Intervention.
LI participants received a comprehensive lifestyle intervention that provided one individual and three group (in-person) visits per month for the first 6 months, followed by one individual and two group visits monthly in months 7–12 (25). Participants were prescribed a diet of 1200–1800 kcal/d, based on body weight, which included the use of meal replacements. The initial activity goal was 175 minutes per week of moderately vigorous physical activity, typically brisk walking. In years 2–4, participants were provided one individual, in-person visit per month, with a second monthly contact by telephone, e-mail, or mail. They also were offered optional monthly group meetings, as well as two to three annual, multi-week “refresher” groups that focused on maintaining the weight losses and high levels of physical activity achieved in year 1 (19). After year 4, until the ILI was terminated, participants were provided at least one individual in-person visit per month and the other optional, yearly group meetings described previously. Participants in the DSE were invited to three group sessions per year for the first 4 years, with one annual meeting thereafter (17).
Intervention termination.
When the ILI was terminated in September 2012, all participants were informed of the study’s lack of differences in cardiovascular outcomes and of Look AHEAD’s transition to an observational cohort study. All participants were invited to continue in the observational study, which was supported by two additional awards from the National Institutes of Health – the Look AHEAD Continuation study (from late 2013 to 2014) and the Look AHEAD Extension study (from 2016–2020). Participants provided written informed consent to enroll in these follow-up assessments. No intervention materials or counseling visits were provided to either group after September 2012.
Study Assessments
In the original randomized trial, participants attended yearly in-clinic assessments at which body weight, height, and other measures were collected (20). Immediately following termination of the ILI, these annual assessments continued while Look AHEAD was formally reorganized as an observational study. Measurements collected during this interim period are referred to as first post-intervention visits (and occurred a mean of 10.2 years post-randomization) (see Figure 1). The Look AHEAD Continuation study (i.e., Continuation) provided one additional in-person study assessment for all participants (a mean of 11.3 years post-randomization), and the Look AHEAD Extension (i.e., Extension) study subsequently provided two additional such assessments, scheduled approximately 2 years apart (at a mean of 14.4 and 16.4 years post-randomization, respectively). Throughout the randomized trial and the observational follow-up, participants also reported their health status on bi-annual phone calls with staff (20). (All study assessments were completed before the COVID-19 pandemic began.)
Figure 1:
Timeline for randomization, intervention, end of intervention, and post-intervention assessments in the Look AHEAD (LA) study; atime by which the last randomized participants completed a given phase of treatment; bmeasurements included weight, antidepressant medications (ADMs), and SF-36; cmeasurements included weight, ADMs, SF-36, BDI-1A, and PHQ-9; dmeasurements included weight, ADMs, SF-36, and PHQ-9.
Depressed mood.
Symptoms of depression were assessed using two self-report inventories. At year 8 of the randomized trial and again at the Continuation assessment, symptoms were measured using the Beck Depression Inventory (BDI-1A) (26). Total scores range from 0–63, with higher scores indicating greater symptoms of depression. For the present study, likely mild or greater depressive symptoms were defined by a score ≥10 (18,26). At the Continuation visit, depressive symptoms also were assessed by the Patient Health Questionnaire (PHQ-9) (27). This inventory contains 9 items (each scored on a 0–3 scale) which map directly on the diagnostic criteria for major depressive episode specified in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) (28). [We added the PHQ-9 to the Continuation assessment because of these characteristics and have previously reported on the agreement between the BDI-1A and PHQ-9 in identifying depression (29).] Scores ≥5 on the PHQ-9 suggest likely mild or greater symptoms of depression, with higher scores indicating greater depression (27). A diagnosis of major depressive episode requires endorsement of five of nine symptoms of depression listed in the DSM-5, one of which must include experiencing either depressed mood or anhedonia (i.e., lack of pleasure) for most of the day, most days of the week, for the past 2 weeks (the window of time for responding to the PHQ-9). The PHQ-9 was administered at both Extension visits; the BDI-1A was not administered at either visit.
Participants were asked to bring all prescription medications to the assessment visits. Study staff recorded the names of all medications, and we examined the use of anti-depressant medications as another indication of possible depression.
Quality of life.
HRQOL was assessed annually during the randomized trial and at all four post-intervention assessments using the Medical Outcomes Scale, Short Form-36 (SF-36) (30). This self-report inventory yields scores on eight domains including: general health perceptions; physical functioning; role limitations due to physical problems; bodily pain; mental health; role limitations due to emotional problems; vitality; and social functioning. Overall functioning is captured by two values -- a physical component summary (PCS) score and a mental component summary (MCS) score. Values are normed as T scores (mean = 50, SD = 10), and higher scores indicate more favorable HRQOL. The PCS score is derived from the first four subscales described above and the MCS from the last four. The PCS and MCS scores were of primary interest in the present study.
Statistical Analyses
Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC) and included all participants who completed at least one in-clinic post-intervention visit and had not undergone bariatric surgery. Differences in weight change between the two groups from the first post-intervention assessment (mean of 10.2 years post-randomization) to the second Extension visit (mean of 16.4 years post-randomization) were examined using a linear mixed model adjusting for race/ethnicity, sex, cardiovascular disease (CVD) history, age, post-intervention visit weight, and correlation between repeated measures.
Chi-square analyses were used to examine differences at the Continuation visit (11.3 years post-randomization) in symptoms of mild or greater depression (i.e., PHQ-9 score of ≥5), according to treatment group, sex, age, and race/ethnicity. Differences between groups in major depressive episode also were examined. To examine the incidence of mild or greater symptoms of depression during the post-intervention follow-up, participants had to be free of depressive symptoms at the Continuation assessment (PHQ-9 <5). Hazard ratios and 95% confidence intervals (CIs) from Cox proportional hazard models, adjusting for sex, race/ethnicity, age and BMI (at the Continuation visit), were compared between the two treatment groups. Time to event was defined as the time from the Continuation assessment until the first occurrence of the outcome of interest (at the first or second Extension visits). For those with mild or greater depressive symptoms at the Continuation visit, remission during the Extension period was analyzed using logistic regression. Using the BDI-1A, we also examined the incidence of mild or greater symptoms of depression from year 8 of the randomized trial to the Continuation visit. This time period provided an assessment of mood within approximately 12–24 months of the ILI’s termination. These analyses were conducted using a binary logistic regression model with randomization arm, sex, age, and race/ethnicity.
Chi-square analyses were used to examine differences at the immediate post-intervention assessment in the use of anti-depressant medications (ADMs) according to treatment group, sex, age, and race/ethnicity. Incident ADM use during post-intervention follow-up was analyzed in the same manner as the PHQ-9 data using Cox proportional hazard models.
Changes in PCS and MCS scores, according to treatment group, sex, age group, and race/ethnicity, during the post-intervention period were assessed using linear mixed models adjusted for treatment group, sex, age, race/ethnicity, CVD history, BMI, and repeated measures. Year–by-year differences in explanatory variable effects were examined only when a significant treatment-by-time interaction existed and were adjusted for multiple comparisons using Bonferroni corrected p-values. Exploratory analyses were conducted on the SF-36 subscales using a similar approach but without corrected p-values.
RESULTS
Participants’ Post-Intervention Characteristics
Figure 2 shows the progression of the 5145 randomized ILI and DSE participants, all of whom contributed to the trial’s primary (cardiovascular) outcome. Of the 2260 ILI participants who were alive and assessed before the end of the intervention (EOI), 2066 completed one or more post-intervention assessments, with corresponding values of 2204 and 1988 for DSE participants. A total of 1945 ILI and 1900 DSE participants, respectively were included in the post-intervention analyses.
Figure 2:
CONSORT diagram showing the flow of participants through the study beginning at randomization.
Table 1 presents demographic and other characteristics for the 3288 total participants who completed a clinic visit in the first post-intervention period (10.2 years post-randomization). At this time, participants had a mean age of 69.5±6.5 years, weight of 94.7±19.3 kg, body mass index (BMI) of 33.9±6.1 kg/m2, and HbA1c of 7.4±1.5%. Sixty percent of participants were female, and the sample was racially and ethnically diverse. There were no significant differences between the ILI and DSE groups on these or others measures with the exception of weight; compared to DSE, ILI participants had a significantly lower mean body weight and BMI, as a result of prior participation in the lifestyle intervention.
Table 1.
Participants’ characteristics at the immediate post-intervention assessment (a mean of 10.2 years post-randomization).
| Overall N=3288 | ILI N=1646 | DSE N=1642 | P-value | |
|---|---|---|---|---|
| Gender, No. (%) | 0.35 | |||
| Male | 1314 (40.0%) | 671 (40.8%) | 643 (39.2%) | |
| Female | 1974 (60.0%) | 975 (59.2%) | 999 (60.8%) | |
| Race/Ethnicity, No. (%) | 0.69 | |||
| African American/Black | 530 (16.1%) | 251 (15.2%) | 279 (17.0%) | |
| Hispanic | 416 (12.7%) | 206 (12.5%) | 210 (12.8%) | |
| White | 2033 (61.8%) | 1032 (62.7%) | 1001 (61.0%) | |
| American Indian/Native American/ Alaskan Native | 206 (6.3%) | 103 (6.3%) | 103 (6.3%) | |
| Other/Mixed | 103 (3.1%) | 54 (3.3%) | 49 (3.0%) | |
| Cardiovascular history, No. (%)a | 0.21 | |||
| Yes | 410 (12.5%) | 217 (13.2%) | 193 (11.8%) | |
| No | 2878 (87.5%) | 1429 (86.8%) | 1449 (88.2%) | |
| Insulin use, No. (%)a | 0.92 | |||
| Yes | 473 (14.9%) | 236 (14.9%) | 237 (15.0%) | |
| No | 2697 (85.1%) | 1352 (85.1%) | 1345 (85.0%) | |
| Age, mean ± SD, years | 69.54 ± 6.5 | 69.37 ± 6.5 | 69.70 ± 6.59 | 0.15 |
| Diabetes duration, mean ± SD, years | 16.90 ± 6.5 | 16.90 ± 6.6 | 16.91 ± 6.31 | 0.98 |
| Weight, mean ± SD, kg | 94.69 ± 19.3 | 93.24 ± 19.2 | 96.14 ± 19.23 | <.001 |
| BMI, mean ± SD, kg/m2 | 33.92 ± 6.1 | 33.38 ± 6.0 | 34.46 ± 6.05 | <.001 |
| HbA1c, mean ± SD, % | 7.38 ± 1.5 | 7.35 ± 1.5 | 7.41 ± 1.54 | 0.54 |
| PHQ-9 score, mean ± SDb | 2.65 ± 3.5 | 2.65 ± 3.5 | 2.65 ± 3.43 | 0.96 |
| PHQ-9 category, No. (%)b | 0.58 | |||
| 0–4 no or minimal symptoms | 2781 (79.8%) | 1398 (79.5%) | 1383 (80.0%) | |
| 5–9 mild symptoms | 540 (15.5%) | 271 (15.4%) | 269 (15.6%) | |
| 10–14 moderate | 105 (3.0%) | 60 (3.4%) | 45 (2.6%) | |
| 15–19 moderately severe or >20 severe | 61 (1.7%) | 30 (1.7%) | 31 (1.8%) | |
| History of depression, No. (%)a | 0.53 | |||
| Yes | 637 (19.4%) | 326 (19.8%) | 311 (19.0%) | |
| No | 2649 (80.6%) | 1319 (80.2%) | 1330 (81.0%) | |
| Antidepressant use, No. (%) | 0.88 | |||
| Yes | 669 (23.5%) | 338 (23.6%) | 331 (23.4%) | |
| No | 2176 (76.5%) | 1092 (76.4%) | 1084 (76.6%) | |
| SF-36, MCS score, mean ± SD | 54.28 ± 9.3 | 54.17 ± 9.3 | 54.39 ± 9.30 | 0.51 |
| SF-36, PCS score, mean ± SD | 43.50 ± 10.2 | 43.84 ± 10.2 | 43.16 ± 10.21 | 0.06 |
Note. Values are from post-intervention assessment unless otherwise indicated.
Values are from randomization.
Values are from the Continuation visit.
BMI = body mass index. PHQ-9 = Patient Health Questionnaire-9. SF-36 = Short-Form 36. MCS = mental component summary; PCS = physical component summary.
Post-Intervention Weight Change
Participants in the ILI and DSE groups lost a mean (±SE) of 3.1±0.3 and 3.8±0.3 kg, respectively, from the first post-intervention visit to the end of the study (Extension visit 2, mean 16.4 years post-randomization; p=0.10; see Figure 2a). This reduction represented a continuation of the weight loss observed in both groups beginning at year 7 of the randomized trial (presented in Figure 2b as percent reduction in baseline weight).
Symptoms of Depression
Table 1 shows the distribution of PHQ-9 scores at the scale’s first administration (Continuation visit). Approximately 80% of participants in both groups reported no or minimal symptoms of depression (scores of 0–4), with an additional 15% endorsing mild symptoms (scores of 5–9). Less than 2% of participants in both groups scored ≥15, indicative of moderately severe to severe symptoms of depression. Only 1.9% of ILI and 1.5% of DSE participants provided responses consistent with a diagnosis of major depressive episode. Across the two groups, significantly more women than men (22.0 vs 17.6%, p=0.002) reported symptoms of mild or greater depression. Rates of depression did not vary significantly based on age or race/ethnicity.
There were no significant differences between the ILI and DSE groups in incident cases of mild or greater depression (i.e., PHQ ≥5; 16.2% vs 17.1%) from the Continuation visit through Extension visit 2 (HR=0.94, 95% CI=0.78–1.14) (Figure 4a). (Participants had to have scored <5 on the PHQ-9 at the Continuation visit to be included in the incident analyses.) In addition, incident cases did not vary significantly according to sex, age, or race/ethnicity. The two groups also did not differ significantly in the prevalence of depressive symptoms at either Extension visit. Across the two groups, significantly more women than men continued to report mild or greater symptoms of depression at Extension visit 1 (22.6% vs 17.4%, p<0.001) and at visit 2 (22.0% vs 17.6%, p=0.008). No significant differences also were observed at the study’s end (Extension visit 2) in the percentage of ILI and DSE participants who met criteria for major depressive episode (1.0 vs 1.4%). Additional analyses revealed no significant differences at this time between the ILI and DSE participants (29.4 vs 30.9%) in remission of symptoms of mild or greater depressive symptoms originally reported at the Continuation visit.
Figure 4, Panel A:
Cumulative hazard ratio (HR) for incidence of mild or greater depressive symptoms (PHQ-9 score ≥5), as measured from the Continuation visit through Extension visit 2, in participants originally assigned to the intensive lifestyle intervention (ILI) or diabetes support and education (DSE) groups.
In the total sample of participants, mean PHQ-9 scores did not change appreciably over time from their low baseline levels (i.e., Continuation visit) in either the ILI or DSE group (Figure 5). However, in patients who reported mild or greater depressive symptoms at baseline, mean±SE values in the ILI group (N= 285) fell from 9.2±0.2 at the Continuation visit to 6.3±0.3 at the study’s end and in DSE participants (N=259) from 9.2±0.2 to 6.3±0.4. Differences between groups were not statistically significant, but scores of both groups declined significantly over time (ps<0.05).
Figure 5:
Mean PHQ-9 scores over time for the full sample of participants originally assigned to the intensive lifestyle intervention (ILI) or diabetes support and education (DSE) groups, as well as scores for those with a PHQ-9 score ≥5 at the Continuation visit. Asterisks show significant changes over time within both groups (P < 0.005).
Changes in depressive symptoms, as assessed by the BDI-1A, revealed findings similar to those obtained with the PHQ-9. At the last administration of the BDI-1A during Look AHEAD’s intervention phase (Year 8), 1457 ILI (85.3%) and 1450 DSE (86.5%) participants scored in the non-depressed range (score <10). When first re-assessed after the intervention’s termination (i.e., Continuation visit), 7.3% and 7.8% of these participants, respectively, reported incident mild or greater symptoms of depression (i.e., score ≥10). Differences between groups in incident depressive symptoms were not statistically significant [OR = 0.94, 95% CI = 0.71–1.23]. Similarly, there were no significant differences between groups (41.2% and 46.0%, respectively) in the remission of prior mild or greater symptoms of depression originally reported at the year-8 assessment.
ADM Use
At the first post-intervention visit, there were no significant differences between ILI and DSE participants in self-reported use of ADMs (23.6 vs 23.4%, respectively). Across the two groups, a greater percentage of women than men reported taking ADMs (25.3% vs 20.8%; p=0.01). Participants who were ≥75 years were less likely to use ADMs than those 65–74 years or 55–64 years (18.4%, 24.1%, and 26.8%, respectively; p=0.002). Participants who identified as non-Hispanic white were more likely to take ADMs (28.2%) than were African-Americans (12.5%), American Indian/Native American/Alaskan Natives (16.7%), persons of other or mixed races/ethnicities (16.7%), or Hispanics (19.1%) (p<0.001).
Of the 2176 participants who did not use ADMs at the first post-intervention visit, 10.5% of ILI and 10.5% of DSE participants did so by the end of the study (Extension visit 2), with no differences between groups in incident ADM use (HR=1.00; 95% CI = 0.76, 1.33; p=0.97; Figure 4b). Women were more likely than men to report incident ADM use (HR=1.44, 95% CI = 1.06– 1.97). Compared to non-Hispanic white participants, African Americans (HR=0.59; 95% CI = 0.38, 0.92; p=0.02) were less likely to report such use. Age was not associated with incident ADM use.
Figure 4, Panel B:
Cumulative hazard ratio (HR) for incident use of antidepressant medications (ADM), as measured from the first post-intervention visit to Extension visit 2, in participants originally assigned to the intensive lifestyle intervention (ILI) or diabetes support and education (DSE) groups.
Of the 669 total participants who took ADMs at the first post-intervention visit, 92.8% continued to do so at the study’s end; 9.2% of ILI participants discontinued ADMs compared to 5.1% of DSE (p=0.06). Sex, age, and race/ethnicity were not significantly associated with discontinuing ADMs.
Quality of Life
Mean PCS scores declined (i.e., worsened) by approximately 3 T-score points in both the ILI and DSE groups over the post-intervention follow-up (p <0.001 for both within-group changes), with no statistically significant differences between groups (see Table 2 and Figure 6a). Mean scores on the MCS declined by about 1.5 T score points (p <0.001 for both within-group changes), but again without significant differences between groups (see Table 2 and Figure 6A). [Supplementary Figures 1a-c show changes in PCS and MCS scores according to sex, age, and race-ethnicity, with the clearest effects for age; the youngest participants (45–55 years) reported significantly smaller declines on both the PCS and MCS than the oldest participants.] Figure 6b shows the substantially greater decline that both ILI and DSE participants reported in their physical as compared with mental functioning over the earlier course of the study, from randomization to the first post-intervention visit, a trend that continued through Extension visit 2.
Table 2.
Mean (±SE) scores on the Short-Form 36 summary scales and subscales at the first post-intervention visit (mean of 10.2 years post-randomization) and at Extension visit 2 (mean of 16.4 years post randomization).
| Variable | 10.2 Years | 16.4 Years | Change | p-value |
|---|---|---|---|---|
| SUMMARY SCORES | ||||
| PCS | ||||
| ILI | 43.3±0.4 | 40.4±0.4 | −2.9±0.2 | 0.36 |
| DSE | 42.9±0.4 | 39.6±0.4 | −3.3±0.2 | |
| MCS | ||||
| ILI | 53.6±0.3 | 52.2±0.4 | −1.4±0.3 | 0.18 |
| DSE | 53.7±0.3 | 51.9±0.4 | −1.8±0.3 | |
| SUBSCALE SCORES | ||||
| Physical functioning | ||||
| ILI | 43.3±0.4 | 39.1±0.4 | −4.2 ± 0.3 | 0.36 |
| DSE | 42.8±0.4 | 38.6±0.4 | −4.3±0.3 | |
| Pain | ||||
| ILI | 47.6±0.4 | 45.9±0.4 | −1.6±0.3 | 0.32 |
| DSE | 47.6±0.4 | 45.7±0.4 | −1.9±0.3 | |
| General health | ||||
| ILI | 46.5±0.4 | 44.6±0.4 | −1.9±0.2 | 0.81 |
| DSE | 45.6±0.4 | 43.5±0.4 | −2.1±0.2 | |
| Vitality | ||||
| ILI | 52.8±0.4 | 50.9±0.4 | −1.8±0.2 | 0.53 |
| DSE | 52.5±0.4 | 50.3±0.4 | −2.2±0.2 | |
| Social functioning | ||||
| ILI | 49.0±0.3 | 46.9±0.4 | −2.0±0.3 | 0.30 |
| DSE | 48.9±0.4 | 46.5±0.4 | −2.5±0.3 | |
| Mental health | ||||
| ILI | 53.5±0.3 | 52.3±0.3 | −1.3±0.2 | 0.88 |
| DSE | 53.5±0.3 | 52.2±0.4 | −1.3±0.2 | |
| Role physical | ||||
| ILI | 44.6±0.4 | 41.9±0.4 | −2.7±0.3 | 0.02 |
| DSE | 44.7±0.4 | 41.1±0.4 | −3.6±0.3 | |
| Role emotional | ||||
| ILI | 47.7±0.4 | 45.0±0.4 | −2.7±0.3 | 0.04 |
| DSE | 48.0±0.4 | 44.3±0.4 | −3.7±0.3 | |
Note. Values shown are least squares means; PCS = physical component summary; MCS = mental component summary; p value is for difference between groups in change from year 10.2 to year 16.4. ILI = intensive lifestyle intervention; DSE = diabetes support and education.
Figure 6, Panel A:
Mean MCS and PCS scores during the post-intervention follow-up for participants originally assigned to the intensive lifestyle intervention (ILI) or diabetes support and education (DSE) groups.
Figure 6, Panel B:
Mean MCS and PCS scores from randomization through the first post-intervention visit (10.2 years post-randomization) in participants assigned to the intensive lifestyle intervention (ILI) or diabetes support and education (DSE) groups.
Consistent with post-intervention declines on the PCS and MCS scales, mean scores on the eight SF-36 subscales also declined in both groups by 1–4 T score points, as shown in Table 2. The largest decline in both groups (of approximately 4.3 T score points) was on the physical functioning subscale. Compared with DSE, ILI participants reported significantly better preservation of their immediate, post-intervention status on two of the eight subscales – role physical and role emotional functioning (see Table 2).
DISCUSSION
Participants originally assigned to ILI, compared with DSE, experienced a 15% reduction in the risk of developing mild or greater symptoms of depression over the nearly decade-long lifestyle intervention (19). However, the present study found no significant differences between ILI and DSE participants in new cases of mild or greater symptoms of depression in the 6 years following termination of the lifestyle intervention. At the Continuation assessment, for example, a comparable 7.3% of ILI and 7.8% of DSE participants reported incident mild or greater depressive symptoms, as measured by the BDI-1A. Analysis of the PHQ-9 data revealed a similar lack of significant differences between groups in incident depressive symptoms from the Continuation visit to the study’s end. There also were no significant differences between groups in incident ADM use during post-intervention follow-up.
Several factors may explain the lack of the anticipated effect. The first is that the most robust differences between groups in depressive symptoms occurred during the first 4 years of treatment, and particularly in Year 1, when ILI participants achieved their largest weight loss of the trial (8.6% vs 0.7% for DSE). Mean BDI scores were significantly lower in the ILI than DSE group during the first 4 years of the intervention but did not differ at year 8 (the last time assessed in the randomized trial). The intervention’s apparently fading effect on depressive symptoms accompanied a convergence in weight losses in the ILI and DSE groups, which at year 8 were 4.7% and 2.1%, respectively. As reported previously, both groups lost weight during the first 2 years of post-intervention follow-up, and mean weight losses converged further at the end of the present study (31). If weight loss protected participants from depressive symptoms during the intervention, the marked narrowing of differences between groups in weight change over the post-intervention follow-up may well have contributed to the lack of differences during this later time. A similar previously reported convergence in physical activity in the two groups (despite remaining greater in ILI participants) may also have contributed to the lack of differences in depressive symptoms (31). Greater leisure-time physical activity may reduce the risk of depression (32).
The ILI participants’ loss of behavioral and social support, following treatment termination, also likely decreased the intervention’s protective effects on mood. The monthly individual counseling visits that ILI participants received throughout the intervention focused principally on the management of eating and activity behaviors, but also addressed coping with stress, personal relationships, and adverse life events. Similarly, ILI participants lost the support of group behavioral weight loss sessions that provided members an opportunity to discuss shared weight management challenges, as well as aspects of their personal lives. Faulconbridge et al showed that such group behavioral weight loss was effective in reducing clinically significant symptoms of depression, even when treatment did not explicitly focus on this problem (33).
The present data suggest that participants in both groups were largely free of clinically significant depressive symptoms during the post-intervention period. At each of the three PHQ-9 administrations, fewer than 2% of participants in either group reported symptoms consistent with a diagnosis of likely major depressive episode, in contrast to 4.2% of persons with type 2 diabetes sampled (using the PHQ-9) in the National Health and Nutrition Examination Survey (2). Past 12-month prevalence rates of major depression are approximately 7–10% in the general population (34–35). Rates are generally lower in older persons, and women consistently report higher rates than men, both findings observed in the present study (34–36). Our low rates of major depressive episode also likely were attributable to the use of ADMs by approximately 25% of participants in both groups. We do not know what percentage of participants took ADMs to manage conditions other than mood -- including sleep, pain, neuropathy, tiredness, and anxiety disorders, which have been responsible for up to 50% of prescriptions in some surveys (37). ADM use in our participants was somewhat higher than the approximately 20% of older adults in the general population who use them (38,39). ADM use was highest in non-Hispanic whites, as observed previously (34,39).
General mental health, as assessed by the SF-36 MCS score, also appeared to be generally favorable in both groups as indicated by the mean T score of approximately 52 at the final visit, which declined by only about 1.5 points during the post-intervention follow-up. MCS scores were remarkably stable over the course of the entire 16-year study and were not as responsive as symptoms of depression to changes in body weight. We cannot rule out the possibility, however, that depression and other mental health concerns were greater in persons who did not participate in the post-intervention follow-up assessments, as compared with those who did.
During the original intervention, the ILI, compared with DSE, had significantly slowed the decline in self-reported physical function, as measured by the PCS score, particularly during the first year of intervention. This benefit was not apparent during post-intervention follow-up, during which both groups registered mean declines of approximately 3 T-score points on the PCS. [This decline occurred despite ILI participants performing better than DSE on some objective measures of physical functioning (e.g., walking speed) during the initial years of the post-intervention period (40).] Moreover, examination of Figure 6b reveals that participants in both groups began Look AHEAD with lower PCS than MCS scores and reported substantially greater declines in their physical than mental function over the course of the 16-year study. Similar long-term declines in PCS scores have been reported by other investigators (41–43) and are consistent with participants increased frailty and disease burden with aging. The study’s oldest participants reported significantly greater declines in physical function than did the youngest participants. We note that ILI participants reported better function on the role emotional and role function subscales of the SF-36 but these results were not predicted or corrected for multiple comparisons.
Strengths of this study include the 6 years of observation following termination of the randomized trial. Limitations, in addition to those previously noted, include our decision to switch from the BDI-1A to the PHQ-9 to measure depressive symptoms, after the Continuation visit, which prevented us from assessing incident depressive symptoms with the BDI-1A over the entire 16-year study. These two instruments have strong diagnostic agreement, although the BDI-1A appears to be more influenced than the PHQ-9 by somatic symptoms (e.g., weight loss, bodily pain) in diagnosing mild or greater symptoms of depression. Despite these limitations, neither instrument detected significant differences between groups in incident depressive symptoms at either the first post-intervention assessment of mood (using the BDI-1A) or at the last two evaluations (which used the PHQ-9).
In summary, the original favorable effects of the nearly 10-year lifestyle intervention in reducing incident mild or greater symptoms of depression and in preventing declines in physical function were not observed during an approximately 6-year post-intervention follow-up. It is possible that benefits would have been observed in a younger cohort of participants, as compared with our older participants, who at the final study visit had a mean age of approximately 75 years and had lived with type 2 diabetes for a mean 23 years. In such individuals, continued provision of lifestyle intervention, following initial treatment, is likely needed to maintain observed improvements in mood and HRQOL.
Supplementary Material
Figure 3, Panel A:
Weight change (in kg) from the first post-intervention assessment (10.2 years post-randomization) to the last study visit (i.e., Extension visit 2, 16.4 years post-randomization) in participants originally assigned to the intensive lifestyle intervention (ILI) or diabetes support and education (DSE) groups.
Figure 3, Panel B.
Percentage reduction in randomization weight through the first post-intervention visit (10.2 years post-randomization) in participants assigned to the intensive lifestyle intervention (ILI) or diabetes support and education (DSE) groups.
Study Importance.
What is already known about this subject?
The Look AHEAD study previously reported that persons with type 2 diabetes and overweight/obesity who received an intensive lifestyle intervention (ILI) for a median of 9.6 years had a 15% reduction in the risk of incident mild or greater symptoms of depression during this time, compared with persons assigned to a control group, referred to as diabetes support and education (DSE). Better preservation of physical function (but not mental) also was observed in ILI vs DSE participants.
What are the new findings in this manuscript?
The present study examined incident depressive symptoms and quality of life over a 6-year period following termination of the ILI in September 2012. We wished to determine whether prior participation in the ILI, compared with DSE, would protect ILI participants going forward against incident depression and deterioration in physical function; neither hypothesis was supported.
How might these results change the direction of research or the focus of clinical practice?
The present findings suggest that older individuals (i.e., mean age of nearly 75 years at the study’s end) with type 2 diabetes and overweight/obesity will require ongoing lifestyle intervention to maintain benefits in mood and physical function initially observed.
Funding
This study was funded by the National Institutes of Health through cooperative agreements with the National Institute on Aging: AG058571 and National Institute of Diabetes and Digestive and Kidney Diseases: DK57136, DK57149, DK56990, DK57177, DK57171, DK57151, DK57182, DK57131, DK57002, DK57078, DK57154, DK57178, DK57219, DK57008, DK57135, and DK56992. Additional funding was provided by the National Heart, Lung, and Blood Institute; National Institute of Nursing Research; National Center on Minority Health and Health Disparities; NIH Office of Research on Women’s Health; and the Centers for Disease Control and Prevention. This research was supported in part by the Intramural Research Program of the National Institute of Diabetes and Digestive and Kidney Diseases. The Indian Health Service (I.H.S.) provided personnel, medical oversight, and use of facilities. The opinions expressed in this paper are those of the authors and do not necessarily reflect the views of the I.H.S. or other funding sources.
Federal sponsors:
National Institute of Diabetes and Digestive and Kidney Diseases (to Mary Evans, PhD; Van S. Hubbard, MD, PhD; Susan Z. Yanovski, MD); Centers for Disease Control and Prevention (to Edward W. Gregg, PhD; Ping Zhang, PhD).
Additional support was received from The Johns Hopkins Medical Institutions Bayview General Clinical Research Center (M01RR02719); the Massachusetts General Hospital Mallinckrodt General Clinical Research Center and the Massachusetts Institute of Technology General Clinical Research Center (M01RR01066); the Harvard Clinical and Translational Science Center (RR025758-04); the University of Colorado Health Sciences Center General Clinical Research Center (M01RR00051) and Clinical Nutrition Research Unit (P30 DK48520); the University of Tennessee at Memphis General Clinical Research Center (M01RR0021140); the University of Pittsburgh General Clinical Research Center (GCRC) (M01RR000056), the Clinical Translational Research Center (CTRC) funded by the Clinical & Translational Science Award (UL1 RR 024153) and NIH grant (DK 046204); the VA Puget Sound Health Care System Medical Research Service, Department of Veterans Affairs; and the Frederic C. Bartter General Clinical Research Center (M01RR01346).
The following organizations have made major contributions to Look AHEAD: FedEx Corporation; Health Management Resources; LifeScan, Inc., a Johnson & Johnson Company; OPTIFAST® of Nestle HealthCare Nutrition, Inc.; Hoffmann-La Roche Inc.; Abbott Nutrition; and Slim-Fast Brand of Unilever North America.
Some of the information contained herein was derived from data provided by the Bureau of Vital Statistics, New York City Department of Health and Mental Hygiene.
Footnotes
Disclosures: Thomas Wadden serves on advisory boards for Novo Nordisk and WW (formerly Weight Watchers). Medha N. Munshi is a consultant for Sanofi and Lilly. Anne Peters serves on advisory boards for Abbott Diabetes Care, Merck, Lilly, Novo Nordisk, Zealand, Mannkind, BioRad, and Medscape, and receives research funding from Dexcom, vTv, and Abbott Diabetes Care. The other authors report no disclosures.
REFERENCES
- 1.Mezuk B, Eaton WW, Albrecht S, Golden SH. Depression and type 2 diabetes over the lifespan: a meta-analysis. Diabetes Care 2008;31:2383–2390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wang Y, Lopez JM, Bolge SC, Zhu VJ, Stang PE. Depression among people with type 2 diabetes mellitus, US National Health and Nutrition Examination Survey (NHANES), 2005–2012. BMC Psychiatry. 2016;16:88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zurita-Cruz JN, Manuel-Apolinar L, Arellano-Flores ML, et al. Health and quality of life outcomes in type 2 diabetes mellitus: a cross-sectional study. Health Qual Life Outcomes 2018;16:94–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Goldney RD, Phillips PJ, Fisher LJ, Wilson DH. Diabetes, depression, and quality of life: a population study. Diabetes Care 2004;27: 1066–1070. [DOI] [PubMed] [Google Scholar]
- 5.Lustman PJ, Clouse RE. Depression in diabetic patients: the relationship between mood and glycemic control. J Diabetes Complications 2005;19:113–122. [DOI] [PubMed] [Google Scholar]
- 6.Manuel DG, Schultz SE. Health-related quality of life and health-adjusted expectancy of people with diabetes in Ontario, Canada, 1996–1997. Diabetes Care 2004;27:407–414. [DOI] [PubMed] [Google Scholar]
- 7.Egede LE, Nietart PJ, Zheng D. Depression and all-cause and coronary heart disease mortality among adults with and without diabetes. Diabetes Care 2005;28:1339–1345. [DOI] [PubMed] [Google Scholar]
- 8.Egede LE, Zheng D, Simpson K. Comorbid depression is associated with increased health care use and expenditures in individuals with diabetes. Diabetes Care 2002;25:464–470. [DOI] [PubMed] [Google Scholar]
- 9.Hazuda HP, Gaussoin SA, Wing RR, et al. Long-term association of depression symptoms and antidepressant medication use with incident cardiovascular events in the Look AHEAD (Action for Health in Diabetes) clinical trial of weight loss in type 2 diabetes. Diabetes Care. 2019;42(5):910–918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Simon GE, Ludman EJ, Linde JA, et al. Association between obesity and depression in middle-aged women. Gen Hosp Psychiatry 2008;30:32–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Onyike CU, Crum RM, Lee HB, et al. Is obesity associated with major depression? Results from the Third National Health and Nutrition Examination Survey. Am J Epidemiol. 2003;15:1139–1147. [DOI] [PubMed] [Google Scholar]
- 12.Fontaine KR, Barofsky I. Obesity and health‐related quality of life. Obesity Reviews. 2001;2:173–182. [DOI] [PubMed] [Google Scholar]
- 13.Jia H, Lubetkin EI. The impact of obesity on health-related quality-of-life in the general adult US population. J Public Health (Oxf) 2005;27:156–164. [DOI] [PubMed] [Google Scholar]
- 14.Flegal KM, Kit BK, Orpana H, Graubard BI. Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA 2013;309(1):71–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Garvey WT, Mechanick JI, Brett EM, et al. American Association of Clinical Endocrinologists and American College of Endocrinology comprehensive clinical practice guidelines for medical care of patients with obesity. Endocr Pract 2016;22:842–884. [DOI] [PubMed] [Google Scholar]
- 16.American Diabetes Association. Obesity management for the treatment of type 2 diabetes: Standards of medical care in diabetes—2020. Diabetes Care 2020;43 (Suppl 1) S89–S97. [DOI] [PubMed] [Google Scholar]
- 17.Ryan DH, Espeland MA, Foster GD, et al. Look AHEAD (Action for Health in Diabetes): design and methods for a clinical trial of weight loss for the prevention of cardiovascular disease in type 2 diabetes. Control Clin Trials 2003;24:610–628. [DOI] [PubMed] [Google Scholar]
- 18.Look AHEAD Research Group. Impact of intensive lifestyle intervention on depression and health-related quality of life in type 2 diabetes: The Look AHEAD trial. Diabetes Care. 2014;37:1544–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Look AHEAD Research Group. Eight-year weight losses with an intensive lifestyle intervention: the Look AHEAD study. Obesity (Silver Spring). 2014;22:5–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Look AHEAD Research Group. Intensive lifestyle intervention in type 2 diabetes. N Engl J Med 2013;369:145–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med. 2002;346(6):393–403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Diabetes Prevention Program Research Group, Knowler WC, Fowler SE, et al. 10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study. Lancet 2009;374:1677–1686. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Nathan DM, Cleary PA, Backlund JY, et al. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. N Engl J Med 2005;353(25):2643–2653. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Diabetes Control and Complications Trial Research Group, Nathan DM, Genuth S, et al. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329(14):977–986. [DOI] [PubMed] [Google Scholar]
- 25.Look AHEAD Research Group, Wadden T, West DS, et al. The Look AHEAD study: a description of the lifestyle intervention and the evidence supporting it. Obesity (Silver Spring). 2006;14:737–752. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Beck AT, Steer RA. Manual for the Beck Depression Inventory. San Antonio (TX): The Psychological Corporation; 1987. [Google Scholar]
- 27.Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16:606–613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fifth edition. Arlington, VA: American Psychiatric Association, 2013. [Google Scholar]
- 29.Vaughan EM, Johnston CA, Moreno JP, et al. Symptom prevalence differences of depression as measured by the BDI and PHQ scales in the Look AHEAD study. Obes Sci Pract 2019;6:28–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ware JE, Kosinski M, Keller SD. SF-36 Physical and Mental Health Summary Scales: A User’s Manual. Boston, MA: New England Medical Center, 1994. [Google Scholar]
- 31.Look AHEAD Research Group, Chao AM, Wadden TA, et al. Weight change 2 years after termination of the intensive lifestyle intervention in the Look AHEAD Study. Obesity (Silver Spring). 2020;28(5):893–901. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Harvey SB, Øverland S, Hatch SL, Wessely S, Mykletun A, Hotopf M. Exercise and the prevention of depression: results of the HUNT cohort study. Am J Psychiatry. 2018;175(1):28–36. [DOI] [PubMed] [Google Scholar]
- 33.Faulconbridge LF, Wadden TA, Rubin RR, et al. One-year changes in symptoms of depression and weight in overweight/obese individuals with type 2 diabetes in the Look AHEAD Study. Obesity. 2011;20:783–793. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hasin DS, Sarvet AL, Meyers JL, et al. Epidemiology of adult DSM-5 major depressive disorder and its specifiers in the United States. JAMA Psychiatry 2018;75(4):336–346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Centers for Disease Control and Prevention (CDC). Current depression among adults---United States, 2006 and 2008. MMWR Morb Mortal Wkly Rep. 2010;59(38):1229–1235. [PubMed] [Google Scholar]
- 36.Blazer DG. Depression in late life: review and commentary. J Gerontol A Biol Sci Med Sci. 2003;58(3):249–265. [DOI] [PubMed] [Google Scholar]
- 37.Wong J, Motulsky A, Abrahamowicz M, et al. Off-label indications for antidepressants in primary care: descriptive study of prescriptions from an indication based electronic prescribing system. BMJ. 2017;356:j603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Tamblyn R, Bates DW, Buckeridge DL, et al. Multinational comparison of new antidepressant use in older adults: a cohort study. BMJ Open. 2019;9:e027663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Pratt LA, Brody DJ, Gu Q. Antidepressant use among persons aged 12 and over: United States, 2011–2014. NCHS Data Brief. 2017. Aug;(283):1–8. [PubMed] [Google Scholar]
- 40.Houston DK, Miller ME, Kitzman DW, et al. Long-term effects of randomization to a weight loss intervention in older adults: a pilot study. J Nutr Gerontol Geriatr. 2019. Jan-Mar;38(1):83–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Schunk M, Reitmeir P, Schipf S, et al. Health-related quality of life in subjects with and without Type 2 diabetes: pooled analysis of five population-based surveys in Germany. Diabet Med. 2012;29(5):646–653. [DOI] [PubMed] [Google Scholar]
- 42.Buttery AK, Du Y, Busch MA, et al. Changes in physical functioning among men and women aged 50–79 years in Germany: an analysis of National Health Interview and Examination Surveys, 1997–1999 and 2008–2011. BMC Geriatr. 2016;16(1):205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Hopman WM, Berger C, Joseph L, et al. The natural progression of health-related quality of life: results of a five-year prospective study of SF-36 scores in a normative population. Qual Life Res. 2006;15(3):527–536. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.