Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jan 1.
Published in final edited form as: Contemp Clin Trials. 2022 Nov 19;124:107016. doi: 10.1016/j.cct.2022.107016

A Digital Health Intervention to Stabilize the 24-hour Rhythm of Sleep, Meals, and Physical Activity for Reducing Depression among Older Bereaved Spouses: Protocol for a Randomized Controlled Trial

Joseph Kazan 1, Thandi Lyew 1, Emilee Croswell 1, Daniel J Buysse 1, Marie Anne Gebara 1, Jordan F Karp 2, Robert T Krafty 3, Ammar A Rashied 3, Charles F Reynolds III 1, Bruce L Rollman 4, Stephen F Smagula 1, Sarah T Stahl 1
PMCID: PMC9839623  NIHMSID: NIHMS1852831  PMID: 36414207

Abstract

Background:

Despite the high prevalence of depression and disruption to 24-hour sleep-wake routines following the death of a spouse in late-life, no bereavement interventions have been developed to re-entrain a regular sleep-wake routine among older widow(er)s. We describe the rationale and methodology of the NIH-funded WELL Study (Widowed Elders’ Lifestyle after Loss), a randomized controlled trial (RCT) comparing the efficacy of a digital health intervention (DHI) to enhanced usual care (EUC) arm for reducing depression symptoms in older spousally-bereaved adults.

Methods:

We will randomize approximately 200 recently bereaved (<12 months) adults aged 60+ years to one of two 12-week interventions: digital monitoring of the timing and regularity of sleep, meals, and physical activity plus weekly motivational health coaching; or enhanced usual care consisting of weekly telephone calls and similar assessment schedules. Participants will complete self-report and clinical assessments at baseline, post-intervention, and 3-, 6-, and 12-months post-intervention, and objective actigraphic assessments of their 24-hour rest-activity rhythm (RAR) at baseline and 1-, 2-, and 3-months during the intervention. The primary outcome is change in depression symptoms burden (using the Hamilton Rating Scale for Depression) from pre- to post-intervention and over 12 months of follow-up.

Discussion:

WELL Study findings will inform the development of widely generalizable and scalable technology-based interventions to support bereaved spouses in community-based settings.

Keywords: aging, internet-based, online intervention, circadian rhythms

1. Background/Introduction

Experiencing the death of a spouse or life partner is a life altering event associated with emotional distress, poor physical health, and feelings of loneliness. In the first year following spousal death, approximately 20-30% of older adults meet criteria for major depressive disorder (MDD).1 As a result, geriatric psychiatry prioritizes reducing depression symptom burden in older adults because of the associated physical illness burden, disability, and excess mortality. One potential mechanism by which depression increases morbidity is through sleep disruption, which includes below average sleep efficiency, difficulty falling and staying asleep, and poor subjective sleep quality. These sleep disruption issues are frequently experienced among the bereaved.2,3 Meals and daytime activity are also disrupted following spousal death. Indeed, compared to married couples, bereaved spouses tend to skip more meals, prepare fewer home-cooked meals, and experience involuntary weight loss,4,5 and bereaved spouses, particularly men, tend to engage in irregular patterns of physical activity post-death.5-7

Health behaviors like sleep, meals, and physical activity are behavioral/social zeitgebers or “time cues” that help keep the body’s biological clock, or circadian timing system, synchronized to the environmental light-dark cycle. Disruptions to the timing and regularity of routine health behaviors place older adults at high risk for disorders related to poor synchronization of the biological clock with the external environment, such as, depression insomnia, diabetes, cardiovascular disease, and early mortality, among others. Studies show that bereaved individuals with MDD have unstable circadian rhythms compared to healthy controls2 and these unstable rhythms are predictive of future increases in depression symptoms.8,9

The objectively assessed rest-activity rhythm (RAR) is a manifestation of the circadian timing system and provides an estimate of the timing and regularity of sleep-wake patterns over a 24-hour period. Disruptions to RAR timing and regularity are associated with depressive symptoms among subgroups of older adults at high-risk for depression.10-12 For example, interdaily stability of the RAR is a measure of how stable the sleep-wake rhythm is over multiple days, indicating how synchronized the internal RAR is to external zeitgebers like the 24h light-dark cycle, food intake, and physical activity. Our previous work found that lower inter-daily stability is associated with a greater burden of depressive symptoms in older bereaved adults.13,14 Lower interdaily stability is also associated with higher risk of developing Alzheimer’s dementia and early mortality.15,16

Despite the high prevalence of depression following spousal death, the evidence based for efficacious interventions for reducing depression symptoms are lacking. Interventions for bereaved adults tend to focus on complicated or pathological grief responses.17 This is the first study with a strong theoretical foundation in circadian science that aims to reduce depression symptom burden in older acutely-bereaved adults by targeting the regularity of routine health behaviors. With the emergence of depression as a significant public health issue, along with data showing the association between lower inter-daily stability and greater depression symptom burden in the spousal bereavement period, we developed the protocol for the WELL (or Widowed Elders’ Lifestyle after Loss) study. The purpose of this report is to describe the protocol for this study in accordance with SPIRIT (Standard Protocol Items: Recommendations or Interventional Trials) guidelines.

1.1. Formative Research

We are using the National Institute of Health’s (NIH) Stage Model for behavioral intervention development to develop a potent and implementable intervention. In stages 1A/1B, we tested the feasibility, acceptability, and preliminary efficacy of our intervention and refined intervention components based upon feedback from 57 adults 60+ years of age who lost their spouse or life partner within the previous 12 months.13 Participants were randomized to 3 months of digital monitoring of sleep, meals, and physical activity (n=18); digital monitoring plus motivational health coaching (n=18); or enhanced usual care (n=9) and followed for 9 months for new-episode depression. We observed high levels of adherence in both digital monitoring (90%) and health coaching (92%) over the 3-month intervention period. The digital health intervention was highly acceptable with 88% of participants retained over 12 months of follow-up. Acceptability of intervention components, based on written feedback, was high. One comment speaks for many: “the WELL study helped organize my life.” Preliminary efficacy data show that lower inter-daily stability of the RAR (representing the inconsistency of activity patterns across days) was significantly associated with more depression symptoms. Modifications to the intervention made in response to participant feedback included (1) sending daily reminders to engage in intervention components via text or email; (2) editing the digital tools to function as a web-based survey and personalized feedback website so that the intervention is functional on participants' own device and across platforms (smartphone, tablet, and/or pc); and (3) offering unlimited technology "check-in" calls to enhance digital literacy. We provided tablet computers to participants (n=2 out of 57) who lacked access to WiFi and/or a mobile device.

2. Study Objectives and Hypotheses

The primary aim of the WELL study is to test the efficacy of a digital health intervention (DHI) for reducing symptoms of depression among older spousally-bereaved adults (NIH Stage II) .The secondary aims are to test whether the DHI improves inter-daily stability of the 24-hour rest-activity rhythm; and if change in inter-daily stability mediates the relationship between intervention participation and depression symptom reduction.

The DHI involves behavioral self-monitoring of sleep, meals, and physical activity, personalized feedback via a Lifestyle Log, and motivational health coaching calls for 3 months. The enhanced usual care (EUC) condition involves attention control calls for 3 months and recommendations for grief and/or mental health support. Our primary hypothesis is that participants randomized to the DHI will experience a greater decrease in depression symptoms from pre-to post-intervention and over 12-months of follow-up, relative to those randomized to EUC (see Figure 1). The primary innovation of the WELL study is that we are addressing multiple behaviors in a unified, scientifically-valid manner by using the circadian timing system as our theoretical framework. Secondarily, we are the first to intervene among a high-risk sample of acutely-bereaved older adults using a daily technology-based monitoring tool. Naturalistically occurring rhythms have been examined in older bereaved adults,18,19 but no one has tried to modify sleep, meals, and physical activity concurrently.

Figure 1. Study Design.

Figure 1.

Open in a new tab

Notes. MDD = Major Depressive Disorder; HDRS = Hamilton Rating Scale for Depression

3. Materials and Methods

3.1. Overview of study design

The WELL study is a randomized controlled trial (RCT) comparing digital monitoring of sleep, meals, and physical activity with an EUC arm (described below) for reducing depression symptoms among older spousally-bereaved adults. Participants are randomized to receive the DHI or EUC over a 12-week intervention period. Objective actigraphic measures of the 24-hour pattern of sleep and daytime activity are measured to evaluate inter-daily stability as a potential mediator of treatment outcomes. Self-report and clinical mood assessments are conducted at 5 time points: baseline (T1), post-intervention (T2) and 3- (T3), 6- (T4), and 12-months (T5) after the intervention period. Our study design is depicted in Figure 1. Participants receive a total of $250 for completing all research activities. They receive $75 after completing T1 and T2; $40 after completing T3; $60 after completing T4; and $75 after completing T5. This study is approved by the Institutional Review Board at the University of Pittsburgh (STUDY19080030).

3.2. Conceptual Model

Our intervention and associated mediators (i.e., regularity and timing of the 24-hour pattern of sleep and daytime activity) are grounded in the social rhythm hypothesis of depression.20-23 In this model, stressful life events like the death of a spouse interrupt an individual’s daily routine or exposure to social zeitgebers (e.g., bedtime and meal time). These disruptions, in turn, place substantial stress on individuals’ circadian or biological clock, thereby increasing the risk of MDD (see Figure 2). In addition to the social rhythm hypothesis, the presence of an attachment figure (e.g., spouse or life partner) promotes the social regulation of health. Repeated social contact with a spouse results in the social regulation of routine daily activities.24 Individuals who become bereaved report a decline in the frequency of health reminders and assistance previously received from their spouse or partner.25 The WELL study is the first of its kind to target the regularity of time cues (sleep, meals, and physical activity), thereby stabilizing circadian rest-activity rhythms and presumably reducing symptoms of depression.

Figure 2.

Figure 2.

Open in a new tab

Proposed mechanism linking spousal death to depression in older adults.

3.3. Randomized controlled trial (RCT)

3.3.1. Study groups

3.3.1.1. Digital monitoring of sleep, meals, and physical activity:

The digital health intervention (DHI) is a multicomponent behavioral intervention that uses digital monitoring, performance feedback, and motivational health coaching to encourage a regular routine of sleep, meals, and physical activity. The multicomponent intervention is described in detail in Table 1. All participants will receive a printed infographic (developed by STS) about the importance of schedules and daily routines for health and wellbeing based on recommendations from the National Sleep Foundation and American Academy of Sleep Medicine. Using this infographic as a guide, research staff will help all participants to choose a personally-meaningful health goal on which to focus during the study. Examples include wake up at the same time every morning, eat 3 meals every day, and walk with my neighbor after work. Most participants can self-select a health goal. For individuals who need assistance, research staff will use summarizing statements based on information collected from the clinical mood assessment, for e.g., “Based on our previous conversation, it sounds like you have difficulty staying asleep. Do you want to focus on a sleep goal?”

Table 1.

Description of Multicomponent Intervention

Digital Health Intervention
Component #1: Digital Monitoring of Sleep, Meals, and Physical Activity. Participants record the timing of sleep, meals, and physical activity in a diary like format via Qualtrics twice daily, for 3 months.
Morning Report: previous night's sleep routine including:

  • time got into bed (HH:MM)

  • time tried to fall asleep (HH:MM)

  • sleep latency (minutes)

  • number of nighttime awakenings

  • mood at bedtime (alert, sleepy, worried, anxious: scale 1-10)

  • quality of sleep (scale 1-10, very bad to very good)

  • time of morning awakening (HH:MM)

  • time out of bed to start daily activities (HH:MM)

  • mood at morning awakening (scale 1-10: very bad to very good)

 Evening Report: routines throughout the day including:

  • time daily activities started (HH:MM)

  • time of first contact with another person (HH:MM)

  • time of breakfast (HH:MM)

  • time of lunch (HH:MM)

  • time of dinner (HH:MM)

  • number of whole grains, fruits, vegetables, and protein (servings)

  • time of naps (HH:MM)

  • time of light-intensity activity (HH:MM)

  • time of moderate-intensity activity (HH:MM)

  • time of vigorous-intensity activity (HH:MM)

  • mood throughout the day (scale 1-10: very bad to very good)
Component #2: Performance Feedback. User data from Qualtrics are automatically fed to a personalized webpage or “Lifestyle Log” that shows progress toward the timing and regularity of participants’ 24-hour sleep-wake routine.
Feedback components that target a robust rest activity rhythm:

  • timing goals and progress towards timing goals over the previous 7 days

  • 24-hr regularity of sleep/naps, meals, and activity over the previous 7 days

Feedback components requested by participants in our pilot RCT:

  • average "daily plate"

  • average sleep duration, latency, and efficiency

  • average mood

  • comprehensive list of grief and mental health support services (local and national)

Component #3: Motivational Health Coaching. A health coach uses motivational interviewing techniques to strengthen participants’ intrinsic motivation to engage in a regular routine of sleep, meals, and physical activity.
During weekly conversations, the health coach:

  • expresses empathy through reflective listening

  • develops discrepancy between participants’ current routine and future goals

  • avoids arguing

  • supports self-efficacy

  • emphasizes the importance of social interactions in helping to keep behaviors entrained
Open in a new tab

The digital diary is based on the Social Rhythm Metric26 and Pittsburgh Sleep Diary27 and asks about the timing and duration of nighttime sleep, timing of breakfast, lunch, and dinner, and timing and duration of physical activity/movement during the day. Participants will monitor their sleep, meals, and activity and view their personalized performance feedback using their own smartphone, tablet, or pc. We will provide WIFI enabled tablets to participants who do not own their own device. The health coaches are certified in the use of motivational interviewing for behavior change via the Motivational Interviewing Network of Trainers.28 Health coaches use open-ended questions, reflective listening, and summarizing statements to promote participants’ intrinsic motivation to engage in a regular behavioral routine. Sessions are conducted weekly for 3 months over the phone.

3.3.1.2. Enhanced usual care:

Participants randomized to the enhanced usual care (EUC) arm will receive the same educational information as the DHI arm, but without the digital monitoring or motivational health coaching that participants in the DHI arm receive. EUC participants will have the same assessment schedule as DHI participants. We match for the non-specific effects of time and attention inherent in health coaching via weekly telephone calls and similar assessment schedules. Staff administering calls encourage continued participation and refer participants to the educational brochures for health-related questions. This condition is considered “enhanced” because we offer education on sleep-wake routines and assessment-based care. These enhancements address a potential confound (between-group differences in depression prevention education) and an ethical problem (identifying but not treating clinical depression and other medical events in the control group).

3.3.1.3. Intervention fidelity.

Fidelity to digital monitoring will be calculated by summing the number of times participants fill out digital diaries: 2 diaries per day for 3 months = 168 possible diaries. Fidelity to personalized feedback will be calculated by summing the number of times participants view the Lifestyle Log. Using the gold-standard for coding fidelity of motivational interviewing – the Motivational Interviewing (MI) Treatment Integrity Coding Manual – four components of MI will be coded during each health coaching session: open-ended questions, affirmations, reflective listening, and summarizing statements.28 All MI sessions will be audio recorded and 20% of sessions (from early, middle, and late intervention) will be randomly selected for fidelity ratings (by a MI expert) to ensure maintenance of treatment specificity and integrity. The MI expert will suggest therapeutic strategies over the course of contact with participants to ensure fidelity to the desired target – regularity of sleep, meals, and physical activity.

3.3.2. Recruitment and screening

Participants are recruited from local hospice organizations, palliative and critical care medicine at the University of Pittsburgh, grief support centers, churches, university-affiliated research registries, and social media ads on Facebook, Twitter, Instagram, and Google. To assure equitable inclusion of minoritized participants, we partner with the Special Populations Community Core at the University of Pittsburgh’s Clinical and Translational Science Institute. We recognize that efforts to recruit adults who are not seeking treatment for mood disorders like depression may be difficult. Recruitment materials emphasize healthy aging, thereby encouraging participation by individuals who are interested in health, rather than illness-related research. Research assistants screen individuals for tentative eligibility over-the-phone and describe the study components and assessment schedule to interested participants. Participants who are potentially eligible for the RCT (defined by a 9-item Patient Health Questionnaire score ≥5)29 are scheduled for a virtual or phone baseline clinical interview to review informed consent and assess inclusion/exclusion criteria.

3.3.3. Participants

Our target population consists of men and women ≥ 60 years of age who experienced the death of a spouse or life partner within the last 12 months. Participants are at high risk for major depressive disorder due to subthreshold symptoms of depression defined by a Hamilton Rating Scale for Depression score ≥ 9,30 together with absence of current major depression or post-traumatic stress disorder. We exclude participants with current DSM-5 diagnosis of syndromal mood or psychosis within the last 12 months (using the PRIME-MD); dementia (defined by a Telephone Interview for Cognitive Status score < 19);31 and acute suicide risk. We also exclude participants taking new psychotropic medications after spousal death to stabilize depression and/or grief, including antidepressants and benzodiazepines. Individuals who have been on a stable dose for at least 1 month and agree not to change during participation, unless it is medically necessary, will be included.

3.3.4. Randomization procedures

Groups are randomized using a 1:1 randomization scheme to minimize threats to internal validity.32 We use permutated-block randomization where subjects are allocated randomly within blocks known only to our biostatisticians. Given the possibility of insomnia as an effect modifier, we stratify randomization based on presence (or absence) of insomnia (defined by an Insomnia Severity Index score ≥10).33 Groups are also stratified by (1) participant sex because women are more likely to experience the death of their spouse compared to men; and (2) death by COVID-19 because individuals who are grieving the unexpected loss of a loved one due to COVID-19 are at higher risk for poor bereavement adjustment compared to individuals bereaved by other causes of death.34

3.3.5. Data collection procedures

Independent evaluators assess participants by phone or by Zoom 5 times over a 15-month period: baseline, post-intervention (3 months from baseline), and 3-, 6, and 12-months post-intervention. All assessments include a clinical mood evaluation and self-report measures conducted by phone or by custom survey link in REDCap (see Table 2). Participants receive a monetary incentive after the intervention period and at each follow-up assessment that increases in value during follow-up to promote participant retention.

Table 2.

Baseline and follow-up data collection methods and frequency

Construct Measure/Source Timepoint
Primary Outcome
Depression symptoms Hamilton Rating Scale for Depression30 T1 - T5
Secondary Outcomes
Major depressive disorder PRIME MD/Mini46 T1 - T5
Prolonged grief disorder Prolonged Grief Disorder 13 - Revised47 T1 - T5
Anxiety Generalized Anxiety Disorder-738 T1 - T5
Complicated grief
Suicidal ideation/behavior		 Inventory of Complicated Grief39;
Beck Suicidal Ideation Scale 41 		T1 - T5
T1 - T5
Post-traumatic stress Post-traumatic Stress Disorder Checklist for DSM-V48 T1 - T5
Personal growth Post-traumatic Growth Inventory49 T1 - T5
Hypothesized Target
Social/behavioral rhythm Social Rhythm Metric26 T1 - T5
24-hr sleep/wake rhythm Rest activity rhythm regularity (inter-daily stability) T1, T1a, T1b, T2
Correlates of depression
Medical comorbidity Cumulative Illness Rating Scale-Geriatrics50 T1, T2
Disability Late-life Function and Disability Instrument51 T1, T2
Insomnia Insomnia Severity Index33 T1, T2
Interpersonal support Interpersonal Support Evaluation List52 T1, T2
Negative life events Elders’ Life Stress Inventory53 T1, T2
Exploratory outcomes
Physical activity Physical Activity Scale for the Elderly54 T1 - T5
Nutrition Rapid Eating Assessment for Participants55 T1 - T5
Sleep Pittsburgh Sleep Quality Index56; Sleep Duration and Sleep Efficiency via actigraphy T1 - T5
Alcohol use Short Michigan Alcoholism Screening Instrument – Geriatric Version57 T1 - T5
Social connectedness UCLA Loneliness Scale58; Social Connectedness and Assurances Scale59; Social Network Index60 T1 - T5
Intervention engagement
Technology acceptance Technology Acceptance Measure61 T1, T2
Technology usability Post-Study Usability Questionnaire62 T1, T2
Open in a new tab

Notes. T1 = baseline assessment; T1a = RAR and depression assessment at month 1 of the intervention period; T1b = RAR and depression assessment at month 2 of the intervention period; T2 = post-intervention assessment; T3 = 3 months post-intervention; T4 = 6 months post-intervention; and T5 = 12 months post-intervention.

After baseline, the research coordinator randomizes the participant and mails a study welcome packet that includes an actigraphy device to be worn for 7 days, during the day and at night (except for showering or swimming). All participants (DHI and EUC) wear an Actiwatch Spectrum PRO (Phillips Respironics Inc.) on their non-dominant wrist for 7 days to generate reliable estimates of their 24-hour rest-activity rhythm (or RAR).35-37 At the end of the 7-day period, participants return the actigraphy device. Once received, the research coordinator performs a data quality check to ensure that 4+ days of continuous accelerometry data are recorded. If the participant’s actigraphy data passes the data quality check, they can start the 12-week intervention period. During the intervention period, additional assessments of depression severity and actigraphy are collected at months 1, 2, and 3. For those randomized to the DHI, the research coordinator conducts a technology coaching session to train participants how to use the digital tools required for monitoring sleep, meals, and physical activity and for viewing performance feedback. Participants randomized to the DHI also receive a booster session of motivational health coaching at 6 months after the end of the intervention to sustain intervention effects on depression.

3.3.5.1. Primary outcome.

The primary outcome is the severity of depression symptoms using the clinician-rated Hamilton Rating Scale for Depression,30 from pre to post-intervention and from pre-intervention to 12-month follow-up.

3.3.5.2. Exploratory outcomes.

Secondary outcomes include incident major depression (using the PRIME-MD) and associated psychopathological conditions including symptom levels of anxiety,38 complicated grief,39 post-traumatic stress,40 and suicidal ideation,41 over 12 months of follow-up.

3.3.5.3. Mediator.

Our proposed mediator (or mechanism of action) is inter-daily stability of the circadian RAR. RAR metrics will be obtained by extracting the raw activity count time-series data from Actiware software and applying non-parametric methods using R code. Interdaily stability is defined as the mean stability of activity within 30-minute bins across the 24-h profile across successive days. Inter-daily stability shows coupling of the RAR with zeitgebers with higher values indicating a stable sleep/wake rhythm across days (range = 0–1).

3.3.6. Power Analyses

Power calculations for each aim were calculated assuming effect sizes as estimated from the preliminary data collected in 55 subjects in the PI’s pilot RCT.13 Calculations were conducted in R using the pwr and powerMediation packages. All tests were conducted at level α=0.05. There is an estimated 96.3% power to detect differences in changes in depression among treatment groups in Aim 1. For Aim 2, there is an estimated 90.4% and 89.4% power to detect differences in changes in intra-daily variability and inter-daily stability of the RAR among treatment groups, respectively. For Aim 3, there is an estimated 81.1% power of the Sobel test to detect mediation by change in RAR stability in the relationship between treatment groups and change in depression symptoms.

3.3.7. Statistical analyses

For Aim 1 (efficacy) we will employ repeated-measures mixed-effects ANCOVA, which extend the standard repeated measures ANCOVA to allow for missing values, error structures other than compound symmetry, and measurements taken non-equal intervals. The test of efficacy for the DHI will be analyzed under an 'intent-to-treat' criteria.42 We will control for important baseline variables, including baseline levels of depression, and, to extend the 'intent to treat' mechanism, all participants randomized into the study are included in all analyses. Controlling for baseline variables will allow us to assess both group differences (main effect of group) and for differential change over time for depression symptoms repeatedly over the course of the study (group X time interaction). If the group X time interaction is significant, we will assess in post hoc tests, where the two groups differ. If no group X time interaction, will test main effects only. In all analyses, the time (in days) since spousal death will be included as a covariate.

For Aim 2 (mechanism of action), we will use linear regression to test whether intervention participation (compared to EUC) is associated with change in the robustness of rest-activity rhythms (RAR) by post-intervention. RARs will be measured prior to randomization and then again at months 1, 2, and 3 of the intervention and change scores in the regularity and timing of sleep-wake activity in determining robustness of the RAR.

For Aim 3 (mediation), we will examine RAR as a mechanism by which the intervention elicits depression symptom reduction, by performing a mediation analysis based on Baron and Kenny’s framework. We will add the change in RAR robustness as an additional fixed effect in the Aim 1 statistical model and note the absolute and relative reductions in the intervention effect to quantify the role that RAR plays in the causal pathway from intervention to depression symptom reduction. Our hypothesis is that our intervention improves robustness of RAR, which in turn will lead to decreased depression symptoms. We will consider the Sobel test and associated multiple mediation and bootstrap methods implemented in %SOBEL and %INDIRECT effect to obtain statistical significance of the mediation model.

Based on theory and prior evidence we believe there is a temporal precedence among variables in that changes in RAR results in change in depression symptoms. However, it is possible that individuals with more depression symptoms at enrollment have more RAR disturbances subsequently. Randomization should help make sure that depression symptom level is balanced across study groups, but it does not rule out the possibility that depression symptoms at baseline themselves predict RAR. We will run analyses while controlling for baseline depression symptoms. We also recognize that the bidirectional association between sleep disturbance (a component of the RAR) and depression increases the difficulty in differentiating the directionality between them. We will examine sleep variables (e.g., sleep duration and efficiency measured via actigraphy), in addition to RAR robustness, as potential explanatory factors that explain depression symptoms.

3.3.8. Qualitative procedures

After the 6-month follow-up visit, we will invite 12-15 participants from the DHI only, who experienced a clinically-significant reduction in depression symptoms (defined as a Hamilton Rating Scale for Depression score < 9), to a semi-structured interview. The interviewer will ask questions about 24-hour sleep-wake routines before and after spousal death and intervention acceptability. Data will be transcribed and coded. Themes will be summarized to understand the unexpected benefits of reducing depression symptoms in the spousal bereavement period.

4. Discussion

The WELL study is a RCT examining the efficacy of a combined digital monitoring plus motivational health coaching intervention versus enhanced usual care for reducing depression symptoms among older adults who recently experienced the death of a spouse/life partner. A major strength of this study is the implementation of a digital/online intervention which will inform the development of widely generalizable and scalable technology-based interventions to support bereaved spouses in community-based settings. Given the proliferation and acceptance of virtual care due to the COVID-19 pandemic, the WELL study paves the way for advances in the delivery of digital health interventions for geriatric mental health. Other strengths include a community-based sample, blinded assessments, objective assessments of 24-hour sleep-wake rhythms, and statistical power for mediation analyses. WELL study results will also advance our understanding of adaptation after spousal death by assessing other sequelae of bereavement including complicated grief, suicidal ideation, post-traumatic stress, and loneliness. We will also test whether reductions in depression symptom burden facilitates psychological adjustment to spousal bereavement in late life.

4.1. Limitations

Our primary limitation is that our digital tools may not be acceptable to older individuals low digital and/or technology literacy. To address this, we offer alternative and hybrid strategies for participation (e.g., telephone services and/or hard-copy assessments) for participants who are not able to use our digital platforms, even when supported by research staff. Most digital health applications and interventions are not designed for older adults with low technology and/or digital literacy.43 Designing and using DHIs requires action by researchers to offer opportunities for inclusion to meet older adults’ age-related learning and usability needs. For instance, applying a user-centered design approach to intervention development ensures that end users’ needs for an accessible and useful application are considered before developing a new DHI.44,45 After developing the DHI, it is crucial to invest time in training older adults to use the DHI to promote their self-efficacy and digital literacy skills.

5. Conclusions and Future Directions

Results of the WELL Study will further our understanding of digital monitoring of sleep-wake routines for reducing depression symptoms among aging widows and widowers who are high-risk for clinical depression following the death of a spouse or life partner. If the intervention effect on depression symptoms is deemed significant, we will deconstruct the multicomponent intervention and examine the impact of individual components on depression symptoms to enhance dissemination potential of our DHI in community settings where bereaved individuals seek support. By examining the efficacy of a highly acceptable and non-pharmacologic intervention for depression, the WELL study has the potential to improve the quality of care for a large and growing population of aging widows and widowers.

Funding sources:

This study was funded by the National Institute of Mental Health R01 MH118270 to STS (Principal Investigator)

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Disclosures and Potential Conflicts of Interest: JFK receives honoraria from American Journal of Geriatric Psychiatry and Journal of Clinical Psychiatry. Within the past two years he received honoraria from NightWare and Biogen for scientific advising, has potential for equity for scientific advising from Aifred Health, and was compensated for preparing a disease-state webinar from Otsuka. Over the past 3 years, DJB has served as a paid consultant to National Cancer Institute, Pear Therapeutics, Sleep Number, Idorsia, Eisai, and Weight Watchers International. DJB is an author of the Pittsburgh Sleep Quality Index, Pittsburgh Sleep Quality Index Addendum for PTSD (PSQI-A), Brief Pittsburgh Sleep Quality Index (B-PSQI), Daytime Insomnia Symptoms Scale, Pittsburgh Sleep Diary, Insomnia Symptom Questionnaire, and RU_SATED (copyrights held by University of Pittsburgh). These instruments have been licensed to commercial entities for fees. He is also co-author of the Consensus Sleep Diary (copyright held by Ryerson University), which is licensed to commercial entities for a fee. He has received grant support from NIH, PCORI, AHRQ, and the VA. MAG was compensated for preparing a disease-state webinar from Otsuka. CFR receives royal income as co-inventor of the PSQI; and an honorarium for editing the American Journal for Geriatric Psychiatry The other authors have nothing to disclose.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

REFERENCES

  • 1.Allen JY, Haley WE, Small BJ, Schonwetter RS, McMillan SC. Bereavement among hospice caregivers of cancer patients one year following loss: predictors of grief, complicated grief, and symptoms of depression. J Palliat Med. 2013;16(7):745–751. doi: 10.1089/jpm.2012.0450 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Brown LF, Reynolds CF, Monk TH, et al. Social rhythm stability following late-life spousal bereavement: associations with depression and sleep impairment. Psychiatry Res. 1996;62(2):161–169. Accessed September 29, 2017. http://www.ncbi.nlm.nih.gov/pubmed/8771613 [DOI] [PubMed] [Google Scholar]
  • 3.Pfoff MK, Zarotney JR, Monk TH. Can a Function-Based Therapy for Spousally Bereaved Seniors Accrue Benefits in Both Functional and Emotional Domains? Death Stud. 2014;38(6):381–386. doi: 10.1080/07481187.2013.766658 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Shahar DR, Schultz R, Shahar A, Wing RR. The Effect of Widowhood on Weight Change, Dietary Intake, and Eating Behavior in the Elderly Population. J Aging Health. 2001;13(2):186–199. doi: 10.1177/089826430101300202 [DOI] [PubMed] [Google Scholar]
  • 5.Stahl ST, Schulz R. The effect of widowhood on husbands’ and wives' physical activity: the Cardiovascular Health Study. J Behav Med. 2014;37(4):806–817. doi: 10.1007/s10865-013-9532-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Janke MC, Nimrod G, Kleiber DA. Reduction in Leisure Activity and Well-Being During the Transition to Widowhood. J Women Aging. 2008;20(1-2):83–98. doi: 10.1300/J074v20n01_07 [DOI] [PubMed] [Google Scholar]
  • 7.Janke MC, Nimrod G, Kleiber DA. Leisure Patterns and Health Among Recently Widowed Adults. Act Adapt Aging. 2008;32(1):19–39. doi: 10.1080/01924780802039238 [DOI] [Google Scholar]
  • 8.Smagula SF, Krafty RT, Taylor BJ, Martire LM, Schulz R, Hall MH. Rest-activity rhythm and sleep characteristics associated with depression symptom severity in strained dementia caregivers. J Sleep Res. Published online May 10, 2017. doi: 10.1111/jsr.12549 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Smagula SF, Ancoli-Israel S, Blackwell T, et al. Circadian rest-activity rhythms predict future increases in depressive symptoms among community-dwelling older men. Am J Geriatr Psychiatry. 2015;23(5):495–505. doi: 10.1016/j.jagp.2014.06.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Smagula SF, Hasler BP, Schulz R, et al. Activity patterns related to depression symptoms in stressed dementia caregivers. Int psychogeriatrics. Published online October 29, 2019:1–8. doi: 10.1017/S1041610219001601 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Smagula SF. Opportunities for clinical applications of rest-activity rhythms in detecting and preventing mood disorders. Curr Opin Psychiatry. 2016;29(6):389–396. doi: 10.1097/YCO.0000000000000283 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Smagula SF, Krafty RT, Thayer JF, Buysse DJ, Hall MH. Rest-activity rhythm profiles associated with manic-hypomanic and depressive symptoms. J Psychiatr Res. 2018;102:238–244. doi: 10.1016/j.jpsychires.2018.04.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Stahl ST, Smagula SF, Dew MA, Schulz R, Albert SM, Reynolds CF. Digital Monitoring of Sleep, Meals, and Physical Activity for Reducing Depression in Older Spousally-Bereaved Adults: A Pilot Randomized Controlled Trial. Am J Geriatr Psychiatry. 2020;28(10):1102–1106. doi: 10.1016/j.jagp.2020.02.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Smagula SF, Hall MH, Stahl ST. Rest-activity rhythms and depression symptoms in older bereaved adults. Int Psychogeriatrics. Published online February 14, 2019:1–2. doi: 10.1017/S1041610218002181 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.AI Luik, Zuurbier LA, Hofman A, Van Someren EJW, Tiemeier H. Stability and Fragmentation of the Activity Rhythm Across the Sleep-Wake Cycle: The Importance of Age, Lifestyle, and Mental Health. Chronobiol Int. 2013;30(10):1223–1230. doi: 10.3109/07420528.2013.813528 [DOI] [PubMed] [Google Scholar]
  • 16.Li P, Gao L, Gaba A, et al. Circadian disturbances in Alzheimer’s disease progression: a prospective observational cohort study of community-based older adults. Lancet Heal Longev. 2020;1(3):e96–e105. doi: 10.1016/S2666-7568(20)30015-5/ATTACHMENT/8ACF05A3-B55B-464C-90D9-09F54A288C64/MMC1.PDF [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Roberts KE, Walsh LE, Saracino RM, et al. A Systematic Review of Treatment Options for Grieving Older Adults. Curr Treat options psychiatry. 2019;6(4):422. doi: 10.1007/S40501-019-00191-X [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Monk TH, Germain A, Reynolds CF. Sleep Disturbance in Bereavement. Psychiatr Ann. 2008;38(10):671–675. doi: 10.3928/00485713-20081001-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Monk TH, Begley AE, Billy BD, et al. Sleep and circadian rhythms in spusally breaved seniors. Chronobiol Int. 2008;25(1):83–98. doi: 10.1080/07420520801909320.Sleep [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Ehlers CL, Frank E, Kupfer DJ. Social zeitgebers and biological rhythms. A unified approach to understanding the etiology of depression. Arch Gen Psychiatry. 1988;45(10):948–952. Accessed September 29, 2017. http://www.ncbi.nlm.nih.gov/pubmed/3048226 [DOI] [PubMed] [Google Scholar]
  • 21.Grandin LD, Alloy LB, Abramson LY. The social zeitgeber theory, circadian rhythms, and mood disorders: Review and evaluation. Clin Psychol Rev. 2006;26(6):679–694. doi: 10.1016/j.cpr.2006.07.001 [DOI] [PubMed] [Google Scholar]
  • 22.Frank E, Swartz HA, Boland E. Interpersonal and social rhythm therapy: an intervention addressing rhythm dysregulation in bipolar disorder. Dialogues Clin Neurosci. 2007;9(3):325–332. Accessed September 29, 2017. http://www.ncbi.nlm.nih.gov/pubmed/17969869 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Haynes PL, Gengler D, Kelly M. Social Rhythm Therapies for Mood Disorders: an Update. Curr Psychiatry Rep. 2016;18(8):75. doi: 10.1007/s11920-016-0712-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Umberson D. Gender, marital status and the social control of health behavior. Soc Sci Med. 1992;34(8):907–917. Accessed September 29, 2017. http://www.ncbi.nlm.nih.gov/pubmed/1604380 [DOI] [PubMed] [Google Scholar]
  • 25.Williams K The Transition to Widowhood and the Social Regulation of Health: Consequences for Health and Health Risk Behavior. Journals Gerontol Ser B Psychol Sci Soc Sci. 2004;59(6):S343–S349. doi: 10.1093/geronb/59.6.S343 [DOI] [PubMed] [Google Scholar]
  • 26.Monk TH, Flaherty JF, Frank E, Hoskinson K, Kupfer DJ. The Social Rhythm Metric. An instrument to quantify the daily rhythms of life. J Nerv Ment Dis. 1990;178(2):120–126. doi: 10.1097/00005053-199002000-00007 [DOI] [PubMed] [Google Scholar]
  • 27.Monk TH, Reynolds CF, Kupfer DJ, et al. The Pittsburgh Sleep Diary. J Sleep Res. 1994;3:111–120. Accessed July 11, 2017. http://www.ncbi.nlm.nih.gov/pubmed/11537903 [PubMed] [Google Scholar]
  • 28.Moyers TB, Manuel JK, Ernst D. Motivational Interviewing Treatment Integrity Coding Manual 4.1 (MITI 4.2). Unpubl Man. 2015;(June). [Google Scholar]
  • 29.Kroenke K, Spitzer RL, Williams JBW. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–613. doi: 10.1046/J.1525-1497.2001.016009606.X [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.HAMILTON M A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960;23:56–62. Accessed July 5, 2017. http://www.ncbi.nlm.nih.gov/pubmed/14399272 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Fong TG, Fearing MA, Jones RN, et al. Telephone Interview for Cognitive Status: Creating a crosswalk with the Mini-Mental State Examination. Alzheimer’s Dement. 2009;5(6):492–497. doi: 10.1016/j.jalz.2009.02.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Hey SP, Kimmelman J. The questionable use of unequal allocation in confirmatory trials. Neurology. 2014;82(1):77–79. doi: 10.1212/01.wnl.0000438226.10353.1c [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Morin CM, Belleville G, Bélanger L, Ivers H. The Insomnia Severity Index: psychometric indicators to detect insomnia cases and evaluate treatment response. Sleep. 2011;34(5):601–608. Accessed August 10, 2018. http://www.ncbi.nlm.nih.gov/pubmed/21532953 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Eisma MC, Tamminga A, Smid GE, Boelen PA. Acute grief after deaths due to COVID-19, natural causes and unnatural causes: An empirical comparison. J Affect Disord. 2021;278:54–56. doi: 10.1016/J.JAD.2020.09.049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ancoli-Israel S, Cole R, Alessi C, Chambers M, Moorcroft W, Pollak CP. The role of actigraphy in the study of sleep and circadian rhythms. Sleep. 2003;26(3):342–392. Accessed July 5, 2017. http://www.ncbi.nlm.nih.gov/pubmed/12749557 [DOI] [PubMed] [Google Scholar]
  • 36.Pollak CP, Tryon WW, Nagaraja H, Dzwonczyk R. How accurately does wrist actigraphy identify the states of sleep and wakefulness? Sleep. 2001;24(8):957–965. Accessed July 5, 2017. http://www.ncbi.nlm.nih.gov/pubmed/11766166 [DOI] [PubMed] [Google Scholar]
  • 37.Ancoli-Israel S, Martin JL, Blackwell T, et al. The SBSM Guide to Actigraphy Monitoring: Clinical and Research Applications. Behav Sleep Med. 2015;13(sup1):S4–S38. doi: 10.1080/15402002.2015.1046356 [DOI] [PubMed] [Google Scholar]
  • 38.Spitzer RL, Kroenke K, Williams JBW, Löwe B. A Brief Measure for Assessing Generalized Anxiety Disorder. Arch Intern Med. 2006;166(10):1092. doi: 10.1001/archinte.166.10.1092 [DOI] [PubMed] [Google Scholar]
  • 39.Prigerson HG, Maciejewski PK, Reynolds CF, et al. Inventory of Complicated Grief: a scale to measure maladaptive symptoms of loss. Psychiatry Res. 1995;59(1-2):65–79. Accessed July 5, 2017. http://www.ncbi.nlm.nih.gov/pubmed/8771222 [DOI] [PubMed] [Google Scholar]
  • 40.Kaltman S, Bonanno GA. Trauma and bereavement: examining the impact of sudden and violent deaths. J Anxiety Disord. 2003;17(2):131–147. Accessed September 28, 2017. http://www.ncbi.nlm.nih.gov/pubmed/12614658 [DOI] [PubMed] [Google Scholar]
  • 41.Beck AT, Steer RA, Ranieri WF. Scale for Suicide Ideation: psychometric properties of a self-report version. J Clin Psychol. 1988;44(4):499–505. Accessed August 15, 2017. http://www.ncbi.nlm.nih.gov/pubmed/3170753 [DOI] [PubMed] [Google Scholar]
  • 42.Gupta SK. Intention-to-treat concept: A review. Perspect Clin Res. 2011;2(3):109–112. doi: 10.4103/2229-3485.83221 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Singh K, Drouin K, Newmark LP, et al. Many Mobile Health Apps Target High-Need, High-Cost Populations, But Gaps Remain. Health Aff (Millwood). 2016;35(12):2310–2318. doi: 10.1377/HLTHAFF.2016.0578 [DOI] [PubMed] [Google Scholar]
  • 44.Dabbs ADV, Myers BA, Mc Curry KR, et al. User-Centered Design and Interactive Health Technologies for Patients. Comput Inform Nurs. 2009;27(3):175. doi: 10.1097/NCN.0B013E31819F7C7C [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Schulz R, Beach SR, Matthews JT, Courtney KL, De Vito Dabbs AJ. Designing and evaluating quality of life technologies: An interdisciplinary approach. Proc IEEE. 2012;100(8):2397–2409. doi: 10.1109/JPROC.2012.2200558 [DOI] [Google Scholar]
  • 46.Tamburrino MB, Lynch DJ, Nagel RW, Smith MK. Primary Care Evaluation of Mental Disorders (PRIME-MD) Screening for Minor Depressive Disorder in Primary Care. Prim Care Companion J Clin Psychiatry. 2009;11(6):339. doi: 10.4088/PCC.08.M00711 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Prigerson HG, Boelen PA, Xu J, Smith KV, Maciejewski PK Validation of the new DSM-5-TR criteria for prolonged grief disorder and the PG-13-Revised (PG-13-R) scale. World Psychiatry. 2021;20(1):96–106. doi: 10.1002/WPS.20823 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Blevins CA, Weathers FW, Davis MT, Witte TK, Domino JL. The Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5): Development and Initial Psychometric Evaluation. J Trauma Stress. 2015;28(6):489–498. doi: 10.1002/jts.22059 [DOI] [PubMed] [Google Scholar]
  • 49.Tedeschi RG, Calhoun LG. The {Posttraumatic} {Growth} {Inventory}. 1996;96(3):455–471. [DOI] [PubMed] [Google Scholar]
  • 50.Miller MD, Paradis CF, Houck PR, et al. Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale. Psychiatry Res. 1992;41(3):237–248. Accessed July 10, 2017. http://www.ncbi.nlm.nih.gov/pubmed/1594710 [DOI] [PubMed] [Google Scholar]
  • 51.McDonough CM, Tian F, Ni P, et al. Development of the computer-adaptive version of the Late-Life Function and Disability Instrument. J Gerontol A Biol Sci Med Sci. 2012;67(12):1427–1438. doi: 10.1093/gerona/gls108 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Merz EL, Roesch SC, Malcarne VL, et al. Validation of Interpersonal Support Evaluation List-12 (ISEL-12) scores among English- and Spanish-speaking Hispanics/Latinos from the HCHS/SOL Sociocultural Ancillary Study. Psychol Assess. 2014;26(2):384–394. doi: 10.1037/a0035248 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.The Elders Life Stress Inventory (ELSI): Research and clinical applications. - PsycNET. Accessed July 18, 2022. https://psycnet.apa.org/record/1991-98927-021 [Google Scholar]
  • 54.Washburn RA, Smith KW, Jette AM, Janney CA. The Physical Activity Scale for the Elderly (PASE): development and evaluation. J Clin Epidemiol. 1993;46(2):153–162. Accessed August 15, 2017. http://www.ncbi.nlm.nih.gov/pubmed/8437031 [DOI] [PubMed] [Google Scholar]
  • 55.Segal-Isaacson CJ, Wylie-Rosett J, Gans KM. Validation of a Short Dietary Assessment Questionnaire: The Rapid Eating and Activity Assessment for Participants Short Version (REAP-S). Diabetes Educ. 2004;30(5):774–781. doi: 10.1177/014572170403000512 [DOI] [PubMed] [Google Scholar]
  • 56.Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193–213. Accessed July 11, 2017. http://www.ncbi.nlm.nih.gov/pubmed/2748771 [DOI] [PubMed] [Google Scholar]
  • 57.Conigliaro J, Kraemer K, McNeil M. Screening and identification of older adults with alcohol problems in primary care. J Geriatr Psychiatry Neurol. 2000;13(3):106–114. http://www.ncbi.nlm.nih.gov/pubmed/11001132 [DOI] [PubMed] [Google Scholar]
  • 58.Russell D, Peplau LA, Ferguson ML. Developing a Measure of Loneliness. J Pers Assess. 1978;42(3):290–294. doi: 10.1207/s15327752jpa4203_11 [DOI] [PubMed] [Google Scholar]
  • 59.Lee RM, Robbins SB. Measuring Belongingness: The Social Connectedness and the Social Assurance Scales. J Couns Psychol. 1995;42(2):232–241. doi: 10.1037/0022-0167.42.2.232 [DOI] [Google Scholar]
  • 60.Edmunds WJ. Social ties and susceptibility to the common cold. JAMA J Am Med Assoc. 1997;278(15):1231b–1231. doi: 10.1001/JAMA.278.15.1231B [DOI] [PubMed] [Google Scholar]
  • 61.Davis FD. Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology. MIS Q. 1989;13(3):319. doi: 10.2307/249008 [DOI] [Google Scholar]
  • 62.Lewis JR. Psychometric Evaluation of the Post-Study System Usability Questionnaire: The PSSUQ. Proc Hum Factors Ergon Soc Annu Meet. 1992;36(16):1259–1260. doi: 10.1177/154193129203601617 [DOI] [Google Scholar]

RESOURCES