A Systematic Review of Self-Management Interventions Conducted Across Global Settings for Depressive Symptoms in Persons with HIV

Abstract

Depressive symptoms can affect health outcomes in people living with HIV (PLWH) including adherence to treatment and disease prognosis. Self-management interventions targeting depressive symptoms have been effective in preventing these negative sequelae of depressive symptoms. The processes of self-management include learning skills related to living with the illness needs, accessing resources to manage the illness, and coping with the illness. A systematic literature review was conducted to appraise and synthesize the current evidence of self-management interventions targeting depressive symptoms in PLWH. Following the PRISMA guidelines, an electronic search of 4 databases was conducted. Original studies written in English that used a randomized controlled trial design to test the effect of self-management intervention on depressive symptoms were included. Studies were selected that were published on/before April 19, 2022, thus yielding 13 relevant articles. Risk of bias was assessed using the NIH Quality Assessment Tool for Controlled Intervention Studies and narrative synthesis was used to synthesize the results. 40 to 755 participants were included in the studies, with each using various measures to assess depressive symptoms pre-and post-intervention, and timepoints for assessing depressive symptoms post-intervention varied. While 12 studies showed a significant reduction in depressive symptoms post-intervention, only 4 studies that used individual coaching or technology showed lower depressive symptoms in intervention groups in comparison to the control groups. This review can be used to inform scale-up and dissemination of these interventions to improve depressive symptoms in PLWH.

Introduction

Depression is a common and serious medical illness that negatively affects behavior, thoughts, mood, and overall health outcomes. Persons suffering from depression can experience a constellation of symptoms including but not limited to: feeling sad, worthless, or guilty, loss of interest or pleasure in activities once enjoyed, difficulty thinking, concentrating or making decisions, trouble sleeping or sleeping too much, loss of energy or increased fatigue, changes in appetite, and thoughts of death or suicide [1]. Clinically, a person must have depressive symptoms most of the day, nearly every day, for at least 2 weeks to be diagnosed with depression [2].

While there is evidence that many people living with HIV (PLWH) may have a co-morbid diagnosis of depression with rates of depression two to four times higher in PLWH than the general population [3], a more recent study that focused on depressive symptoms reported by PLWH found that as many as 40% of PLWH experience depressive symptoms during their lifetime [4].

Depressive symptoms affect adherence to medication regimens and disease prognosis [5]. Prior research has revealed that PLWH who are experiencing depressive symptoms are at increased risk of non-adherence to antiretroviral therapy (ART) and causation is not known [6]). Depressive symptoms are linked to AIDS diagnosis and an elevated risk of mortality [7]. Thus, identifying effective interventions to manage depressive symptoms is an urgent priority.

Self-management interventions can help people with chronic diseases achieve higher health-related quality of life, lower levels of depression, and better health behaviors and health outcomes [8–11]. The term self-management refers to “the individual’s ability to manage the symptoms, treatment, physical and psychological consequences, and lifestyle changes inherent in living with a chronic condition. Efficacious self-management encompasses the ability to monitor one’s condition and to affect the cognitive, behavioral, and emotional responses necessary to maintain a satisfactory quality of life” [12]. The processes of self-management include (a) learning skills [13, 14] focusing on the illness needs (e.g., decision-making, goal-setting, self-monitoring, problem-solving, emotional control, self-evaluation), (b) accessing resources (e.g., coordinating health care and social services, identifying psychological resources, communicating with health care providers overcoming social and environmental challenges being part of a spiritual community, seeking social support, and (c) coping with the illness (adjustment, processing emotions, and integrating illness into daily life). Additionally, self-management interventions focused on depressive symptoms in the general population of adults with a chronic condition experiencing depressive symptomology have been shown in a systematic review and meta-analysis to have a moderate effect on depressive symptoms [15]. This is especially timely given the increased burden on the healthcare system and dearth of healthcare professionals amidst the decrease in workforce post-COVID-19 [16].

Despite the evidence to support the efficacy of self-management interventions for improving depressive symptoms in general, no review has been conducted to assess and synthesize the evidence amongst PLWH. Our goal was to review the existing self-management interventions targeting depressive symptoms in PLWH and to assess the effectiveness of self-management interventions on reducing depressive symptoms.

Methods

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols (PRISMA-P) 2020 [16].

Inclusion/Exclusion Criteria

We included peer-reviewed articles that focused on persons living with HIV ages 18 or older. The interventions were focused on the process of self-management of depression or depressive symptoms or HIV more generally. A study was deemed eligible for this review if the intervention focused on at least one process of self-management (needs, resources, and coping). Interventions that solely utilized antidepressants or psychotherapies, such as cognitive behavioral therapy (CBT), were excluded. The outcomes of interest were depressive symptoms or clinical depression. All included studies used a randomized control trial (RCT) design. Studies without results (i.e., protocol papers) were excluded.

Search Strategy

One author conducted a literature search of the following electronic databases for articles published from inception to April 19, 2022: PubMed, CINAHL, PsycInfo, and EMBASE. A range of Medical Subject Headings (MeSH), keywords, and indexed terms related to self-management, depressive symptoms, depression, and HIV were searched. Boolean operators and wildcards were utilized to define relationships between the terms. The reference lists of studies meeting the inclusion criteria were hand-searched to identify additional relevant studies.

Literature search strategies and search results were tracked in a Microsoft Excel document. The search results from each database were exported into EndNote, whereby duplicates were removed. Articles were imported along with their full-texts into Covidence©, an online systematic review management platform that allows multiple users to review the publications and make decisions on whether to include or exclude the publication.

Study Selection, Data Extraction, and Synthesis

Two reviewers (MYJ, GA) independently reviewed the titles and abstracts, followed by full-text review over the course of 3 months. When either of the two reviewers considered a study potentially eligible or uncertain, its full-text was reviewed for further assessment. A third reviewer (RS) resolved any disagreements or discrepancies. One reviewer (GA) extracted the following data from included studies and synthesized the results using narrative synthesis [17]: study characteristics (sample size & setting/geographic location); intervention description (duration, frequency, dose, type, moderator); self-management process (needs, resources, coping); measurements for depressive symptoms; and any intervention effect on depressive symptoms.

Quality Appraisal

The National Institutes of Health National Heart, Lung, and Blood Institute’s (NIH/NHLBI) Quality Assessment Tool of Controlled Intervention Studies was used to assess the methodological quality of the included studies [18]. This tool was selected because of its comprehensiveness in assessing the internal validity of the study and its rigor of the methods of published studies based on certain study designs. The assessment tool for RCT design has 14 items where each item can be answered with a “Yes,” “No,” or “Other.” “Other” can have three responses (cannot determine, not applicable, or not reported). Two reviewers (MYJ, GA) independently rated the studies and any discrepancies between raters were resolved via inter-rater discussion or by involving a third reviewer (RS). Studies were not excluded based on the quality appraisal; rather, the quality appraisal was used to evaluate and discuss the rigor of available evidence.

Results

Figure 1 provides a flow chart of the literature search process. The database search identified 6,061 publications published up to April 19, 2022, resulting in a total of 3,272 unique titles after removing duplicates. We screened the full text of 27 publications against the inclusion and exclusion criteria. A total of 13 studies were included in this review and details of each study are described in Table 1.

Fig. 1.

PRISMA Diagram. (Note: PLWH = People living with HIV)

Table 1.

Study Characteristics

	Patient Population Sample Size & Setting	Intervention	Comparator	Randomization	Outcome: Measures of Depressive Symptoms and Administration	Timing	Setting	Effect of Intervention on Depressive Symptoms
Jones et al. (2010)	N = 451, USA Intervention = 212 Control = 239	Based on the SMART/EST (Stress Management And Relaxation Training/Expressive-Supportive Therapy) Women’s Project Weekly 120-minute cognitive behavioral stress management /expressive supportive (CBSM+) group therapy sessions	Individual information conditions (“low intensity experimental control”): 10-weekly individual 120-minute information sessions around stress management/relaxation training and coping with HIV/AIDS, equal in session length to intervention group Active control Blinding not mentioned in manuscript	Randomized using a table of random numbers	Beck Depression Inventory (BDI)	10 weeks	Participants recruited from hospitals, community health centers/agencies, and via participant referrals in Miami-Dade County, FL, NYC, and New Jersey metropolitan area Specific care setting where intervention occurred not described in manuscript Interventions can be useful adjuncts for therapy and clinic waiting areas or adapted in clinical practice and mental health settings as short-term interventions	Those in the CBSM + group who reported increases in self efficacy saw significant improvements in anxiety and depression at post-intervention and 12 months follow-up. Those in the control group did not report any significant changes in depression status.
Laperriere et al. (2005)	N = 154, USA Intervention = 80 Control = 74	Based on The SMART/EST Women’s Project (subgroup analysis) Weekly 120-minute CBSM + group therapy sessions led by doctoral and post-doctoral level psychologists	Individual information conditions (“low intensity experimental control”): 10-weekly individual 120-minute information sessions around stress management/relaxation training and coping with HIV/AIDS, equal in session length to intervention group Active control Blinding not mentioned in manuscript	Randomization method not mentioned in manuscript, but based on same project as above, therefore can infer randomization using a table of random numbers	BDI	10 weeks	Participants recruited from hospitals, community health centers/agencies, and via participant referrals in Miami-Dade County, FL, NYC, and New Jersey metropolitan area Specific care setting where intervention occurred not described in manuscript Interventions can be useful adjuncts for therapy and clinic waiting areas or adapted in clinical practice and mental health settings as short-term interventions	Participants in both the intervention and control groups saw significant improvements in depressive symptoms at post-intervention. At 12 months follow up, participants in the intervention group scored an average BDI score of 7.4 which is below the scale’s range for depression categorization, while those in the control group scored an average of 10.2 (mild depression category). These results were significantly different from one another (p < 0.001).
Hecht et al. (2018)	N = 177, USA Intervention = 89 Control = 88	Weekly, in-person Mindfulness-Based Stress Reduction (MBSR) group sessions	HIV disease self-management skills group (controlled for the effects of being in a group program): 8 weekly group sessions that covered variety of educational topics abut managing HIV infection) Active control Participants and staff were aware of group assignments. Personnel performing laboratory assays masked to group assignment	Randomized using computer-generated randomization list with random assignment of participants in a 1:1 ratio using block sizes. A database manager generated randomization sequence with guidance from a study statistician. No other study staff had access to randomization sequence file	BDI	8 weeks	Specific care setting where intervention occurred not described in manuscript Recruitment conducted via broad advertising across United States	Those in the intervention saw a statistically significant within group improvement in BDI scores for depressive symptoms from baseline to 3-months post-intervention (p < 0.05). BDI scores did not improve significantly within the control group from baseline to any point post-intervention.
Guo et al. (2020)	N = 300, China Intervention = 150 Control = 150	WeChat (internet)-based culturally adapted CBSM course	Usual care waitlist control group: received a brochure on nutrition in addition to usual care for HIV treatment (i.e. ART as soon as feasible) Active control By nature of the trial design, the research staff nor the participants were blinded to allocation	Randomized via computer-generated randomization list with block size of 4 using SAS	Center for Epidemiologic Studies Depression Scale (CES-D)	3 months	Specific care setting where intervention occurred not described in manuscript Recruited from the outpatient clinic of the only hospital designated for HIV treatment in Guangzhou, China.	Those in the intervention group saw statistically significant improvements in depressive symptoms at all follow-up points compared to baseline (p < 0.001 at 3,6,& 9 months).
Heckman & Carlson (2007)	N = 299, USA Intervention (coping) = 108 Intervention (support) = 84 Control = 107	Teleconference coping or information support group led by a professional	Usual care condition: no active intervention, but had access to all services provided by their ASO (i.e. case management, support groups, and assistance obtaining housing, employment, and child care) Active control Blinding not discussed in manuscript	Randomization methods not mentioned in manuscript	BDI and 20-item Hopkins Symptom Checklist (SCL-20)	8 weeks	Intervention delivered via reserved conference calls using the participants’ standard telephones. Recruited through 27 different AIDS service organizations (ASOs) in Ohio, Pennsylvania, Virginia, New York, Wisconsin, West Virginia, Montana, Arizona, Indiana, and Alaska.	Participants who received either form of the intervention in all conditions/subgroups did not report statistically significant improvements in depressive symptoms compared to the usual care group.
Miles et al. (2003)	N = 109, USA Intervention = 59 Control = 50	Biweekly emotional processing strategies and cognitive reframing sessions led by a registered nurse	Usual care: health visits for primary care and specialty visits for HIV-related care. Most women got their care at the tertiary HIV clinics, where focus of care was on health problems rather than self-care symptom management. Active control Data collectors were blinded to the woman’s status.	Randomized using a table of random numbers (performed by project manager who was part of intervention team)	CES-D and Profile of Mood States (POMS)	3 months	Intervention was conducted in the homes of the women by 3 nurses (2 African American nurses and one White nurse). Visits were arranged when home alone or with others who were aware of dx Recruited from two tertiary care university-based infectious disease clinics in the Southeast	Those in the intervention group saw statistically significant reductions in feelings of Depression-Dejection (measured by POMS), from baseline to 1-month post-intervention (p < 0.003). Those in the control group saw no significant changes in feelings of Depression-Dejection at any follow-up point. The depressive symptoms measured with the CES-D scale did not change significantly in the intervention group at any time point.
Inouye et al. (2001)	N = 40, USA Intervention = 20 Control = 20	Individually administered self-management and coping skills training program	Usual care: standard treatment of their primary care providers, an educational videotape about nutrition and health in relation to HIV/AIDS, wellness education, resource materials to take home. Invited to a shorter version of the protocol in a group setting after final session Active control Blinding not discussed in manuscript	Randomization methods not mentioned in manuscript	POMS	7 weeks	Specific care setting where intervention occurred not described in manuscript Recruited from advertisements in newspapers, private physicians, hospital flyers, AIDS organization, and word-of-mouth communication.	The intervention significantly improved participants’ mood as measured by the POMS, with a 27% improvement in Depression-Dejection scores for the intervention group from baseline (p < 0.05). Those in the control group did not show significant improvements in Depression-Dejections scores from baseline.
van Luenen et al. (2018)	N = 188, Netherlands Intervention = 97 Control = 91	Internet-based program adapted from a previously evaluated self-help booklet + individualized coaching session led by a master’s level graduate student in the field of psychology	Control condition: put on a waiting list and received attention only via telephone calls from a coach. Invited to start the intervention after the second post-test; Wait List control; Participants, researchers, and coachers were not masked to the participant’s assigned treatment condition.	Randomized using random number tables to generation randomization sequence with block sizes of 12, stratified by treatment center and sex, and concealed from the main researcher.	Patient Health Questionnaire (PHQ-9) and CES-D	8 weeks	All assessments were completed online via a secure website, with the exception of screening with the PHQ-9, which was completed via telephone conversation Recruited from 23 HIV treatment centers in the Netherlands	A large proportion (64% - PHQ-9 and 67% - CES-D) of those in the intervention group had significant decreases in their depressive symptoms. These improvements were significantly different than those in the control group.
Wantland et al. (2008)	N = 755, Puerto Rico, USA, South Africa & Kenya Intervention = 426 Control = 349	Individually administered HIV/AIDS Symptom Management Manual	Basic nutrition manual: of similar size as compared to intervention and at an eighth-grade reading level. Modified from the WHO HIV/AIDS Nutrition Guide; Active control Blinding not discussed in manuscript	Randomization methods not discussed in manuscript	Revised Sign and Symptom Checklist for Persons with HIV Disease (SSC-HIVrev)	2 months	Specific care setting where intervention occurred not described in manuscript Recruited from clinics and HIV-focused community settings in Africa, Puerto Rico, and from 10 sites across the US.	Those in the intervention group with comorbid depression showed a statistically significantly decrease mean symptom intensity post-intervention compared to those in the control group (p < 0.018).
Schnall et al. (2018)	N = 80, USA Intervention = 40 Control = 40	mHealth-app on HIV/AIDS Symptom Management Manual	Control app without self-care strategies Placebo control Single-blinded study	Randomized using an allocation sequence developed by the PI prior to the start of trial via a computerized random number generator.	SSC-HIVrev	12 weeks	Took place at study site, Columbia University School of Nursing. Research setting—care setting not applicable Recruited through flyers at a local HIV clinic and community-based organizations, and through e-mail invitations	The intervention group participants showed a significantly greater improvement than the control group in depression (p = 0.001).
Chen et al. (2018)	N = 41, China Intervention = 21 Control = 20	Self-and family management intervention led by a nurse	Usual care at the clinic as well as medical care as usual. Study sessions involved only assessment survey, no counseling; Placebo control	Randomized using a computer generated number in recruitment sequence	CES-D	1 month	Intervention was conducted within clinic, and session times coordinated with patients’ clinic visits to reduce participant burden Recruited from Beijing Ditan Hospital and Shanghai Public Health Clinic Center (SPHCC)	Those in the intervention group had a significant decrease in their depressive symptoms at both 1 and 3-months post-intervention, compared to the control group.
Nyamathi et al. (2012)	N = 68, India Intervention = 34 Control = 34	Six sessions of HIV-specific and life management skills led by a lay village Asha (health activist)	Usual care: 6 group classes matched in terms of number and length of time to intervention. Included content from Ashas with the goal of monitoring barrier to ART adherence, inquire about side effects, and provide basic education. The women received basic nutrition supplementation to address malnourishment. Ashas did not assist with linkage to care. Did not fulfill the same supportive role as intervention Ashas. Placebo control	Randomization methods not mentioned in manuscript	CES-D	6 weeks	Interventions involved visits to women performed by Ashas, assumed in private home setting Recruited from 2 high prevalence major HIV/AIDS villages or mandals in rural Andhra Pradesh were selected randomly from 16 major villages that were demographically alike in terms of HIV prevalence of 2% and served by a distinct Public Health Center.	The intervention group had a much greater reduction in depressive symptom scores than those in the usual care group (B = 22.89, p = 0.001). Those with higher depressive symptom scores at baseline also showed greater reduction in symptoms than those with lower baseline depressive symptom scores.
Eller et al. (2013)	N = 222, Puerto Rico, USA, South Africa Intervention = 124 Control = 98	HIV/AIDS Symptom Management Manual led by a nurse	Basic nutrition manual: of similar size as compared to intervention and at an eighth-grade reading level. Modified from the WHO HIV/AIDS Nutrition Guide. Reviewed for 30-minute sessions with a research nurse to mask treatment assignment Active control Additional blinding methods not mentioned in manuscript.	Randomization methods not discussed in manuscript	CES-D; Single item asking frequency, intensity, and impact of depressive symptoms	One 30-minute session	Specific care setting where intervention occurred not described in manuscript Recruited from clinics and HIV-focused community settings in Africa, Puerto Rico, and from 10 sites across the US.	No significant differences between the intervention and control groups. Within the intervention group, there was significant reduction in depressive symptom frequency [F(2, 207) = 3.27, p = 0.05], intensity [F(2, 91) = 4.6, p = 0.01], and impact [F(2, 252) = 2.92, p = 0.05), at one month but not at two months.

Study Characteristics

The total number of participants ranged from 40 to 755 PLWH across all studies. One study focused specifically on low-income, racial/ethnic minority individuals living with HIV [19]. Four studies included only women living with HIV [19–22], one study included only women with HIV and their caregivers [23], and the remainder included both men and women living with HIV. The studies were conducted both in the United States (US) and internationally in China [23, 24], Netherlands [25], and India [22]. Two studies were from the same multi-site study comprised of sites in the US, Puerto Rico, Kenya, and South Africa [26, 27]. Two studies [20, 21] were from the same multi-site study based in the US using the Stress Management And Relaxation Training/Expressive-Supportive Therapy (The SMART/EST)- Women’s Project [28]. Participants were recruited from multiple settings including but not limited to community sites, AIDS service organizations, and HIV treatment centers/clinics.

Quality of Included Studies

Detailed results of the quality assessment can be found in Fig. 2; Table 2. No studies met all assessment items. Many of the studies lacked information on whether the assignment to the control/intervention group was concealed (n = 11; item 3), whether participants were blinded to the intervention group (n = 12; item 4), and whether individuals assessing the outcomes were blinded to the group assignments (n = 12; item 5). Nine studies had more than 20% drop out rates in the intervention group (item 7) and seven studies failed to report the adequacy of the sample size to detect a difference in the main outcome between groups with power (item 12).

Fig. 2.

Quality assessment of self-management intervention on depressive symptoms. (Quality of the selected observational study was assessed using the National Institutes of Health National Heart, Lung, and Blood Institute’s (NIH/NHLBI) Quality Assessment tool for Controlled Intervention Studies. Item (1) Was the study described as randomized, a randomized trial, a randomized clinical trial, or an RCT? Item (2) Was the method of randomization adequate (i.e., use of randomly generated assignment)? Item (3) Was the treatment allocation concealed (so that assignments could not be predicted)? Item (4) Were study participants and providers blinded to treatment group assignment? Item (5) Were the people assessing the outcomes blinded to the participants’ group assignments? Item (6) Were the groups similar at baseline on important characteristics that could affect outcomes (e.g., demographics, risk factors, co-morbid conditions)? Item (7) Was the overall drop-out rate from the study at endpoint 20% or lower of the number allocated to treatment? Item (8) Was the differential drop-out rate (between treatment groups) at endpoint 15% points or lower? Item (9) Was there high adherence to the intervention protocols for each treatment group? Item10. Were other interventions avoided or similar in the groups (e.g., similar background treatments) Item 11. Were outcomes assessed using valid and reliable measures, implemented consistently across all study participants? Item 12. Did the authors report that the sample size was sufficiently large to be able to detect a difference in the main outcome between groups with at least 80% power? Item 13. Were outcomes reported or subgroups analyzed prespecified (i.e., identified before analyses were conducted)? Item 14. Were all randomized participants analyzed in the group to which they were originally assigned, i.e., did they use an intention-to-treat analysis?; CD, cannot determine; NA, not applicable; NR, not reported.)

Table 2.

Results of the quality assessment of self-management intervention on depressive symptoms

	Jones et al. (2010)	Laperriere et al. (2005)	Hecht et al. (2018)	Guo et al. (2020)	Heckman & Carlson (2007)	Miles et al. (2003)	Inouye et al. (2001)	van Luenen et al. (2018)	Wantland et al. (2008)	Schnall et al. (2018)	Chen et al. (2018)	Nyamathi et al. (2012)	Eller et al. (2013)
Item 1	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Item 2	Yes	CD	Yes	Yes	NR	Yes	NR	Yes	NR	Yes	Yes	No	Yes
Item 3	CD	No	No	No	NR	CD	No	Yes	NR	Yes	NR	No	CD
Item 4	No	No	No	No	NR	Yes	NR	No	NR	No	NR	No	No
Item 5	No	No	Yes	No	NR	CD	NR	No	NR	No	NR	CD	CD
Item 6	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No
Item 7	No	No	No	No	No	No	Yes	No	No	Yes	Yes	Yes	No
Item 8	No	No	No	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes	No
Item 9	CD	CD	No	Yes	CD	CD	CD	Yes	Yes	Yes	Yes	NR	CD
Item 10	Yes	Yes	No	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes	No	No
Item 11	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Item 12	No	Yes	Yes	Yes	No	Yes	No	Yes	Yes	No	No	No	No
Item 13	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Item 14	Yes	Yes	Yes	Yes	Yes	Yes	CD	Yes	CD	Yes	Yes	NR	Yes

CD = cannot determine; NR=not reported; Quality of the selected observational study was assessed using the National Institutes of Health National Heart, Lung, and Blood Institute’s (NIH/NHLBI) Quality Assessment tool for Controlled Intervention Studies. Item (1) Was the study described as randomized, a randomized trial, a randomized clinical trial, or an RCT? Item (2) Was the method of randomization adequate (i.e., use of randomly generated assignment)? Item (3) Was the treatment allocation concealed (so that assignments could not be predicted)? Item (4) Were study participants and providers blinded to treatment group assignment? Item (5) Were the people assessing the outcomes blinded to the participants’ group assignments? Item (6) Were the groups similar at baseline on important characteristics that could affect outcomes (e.g., demographics, risk factors, co-morbid conditions)? Item (7) Was the overall drop-out rate from the study at endpoint 20% or lower of the number allocated to treatment? Item (8) Was the differential drop-out rate (between treatment groups) at endpoint 15% points or lower? Item (9) Was there high adherence to the intervention protocols for each treatment group? Item10. Were other interventions avoided or similar in the groups (e.g., similar background treatments) Item 11. Were outcomes assessed using valid and reliable measures, implemented consistently across all study participants? Item 12. Did the authors report that the sample size was sufficiently large to be able to detect a difference in the main outcome between groups with at least 80% power? Item 13. Were outcomes reported or subgroups analyzed prespecified (i.e., identified before analyses were conducted)? Item 14. Were all randomized participants analyzed in the group to which they were originally assigned, i.e., did they use an intention-to-treat analysis?

Measures of Depressive Symptoms

Various measures were used to assess depressive symptoms across the studies, with Beck Depression Inventory (BDI) [20, 21, 29, 30] and Center for Epidemiologic Studies Depression Scale (CES-D) [19, 22–26] being the measures most widely used by the study investigators. Other measures that were used included the 90-item Hopkins Symptom Checklist (SCL-90-R) [30]; Profile of Mood States (POMS) Depression-Dejection subscale [19, 31]; and 64-item Revised Sign and Symptom Checklist for Persons with HIV Disease (SSC-HIVrev) [27, 32].

Four studies used two different methods for measuring depressive symptoms and CES-D was the measure that was consistently used along with another measurement. Among the four, three studies used reliable and validated scales for depressive symptoms: POMS [19]; Veterans Short Form-12 [23]; Patient Health Questionnaire (PHQ-9) [25]. One study used CES-D and a single-item Depressive Symptom Self-Report (DSSR) that asked whether a respondent has had depressive symptoms in the past week (Yes/No) with subsequent items asking for the frequency and the impact of such symptoms [26].

The internal reliability of these depression measures range from good to excellent with Cronbach α = 0.90 for BDI [30], Cronbach alphas ranging from 0.88 to 0.94 across several studies using CES-D [19, 22, 26], Cronbach α = 0.91 for SCL-90-R [30], Cronbach α = 0.89 for PHQ-9, [33] and Cronbach α = 0.92 for SSC-HIVrev [27]. Internal reliability for POMS was calculated across the six subscales (including Depression-Dejection) with Cronbach alphas ranging from 0.78 to 0.95 over time [19]. The Veterans Short Form-12 is an abbreviated version of the the 36-item Veterans Short Form (VSF-36), with the 36-item having Cronbach alphas ranging from 0.80 to 0.95 in veteran study populations [34].

Validity for BDI and CES-D is also high when comparing the BDI to the Hamilton Rating Scale for Depression (r = 0.73), [21] when comparing the CES-D to Lubin’s Depression Adjective Check (r = 0.70), [35] and when comparing the single item DSSR with CES-D (r = 0.48) [26]. The POMS Depression-Dejection subscale is correlated with BDI (r = 0.69), [36] PHQ-9 highly correlates to the Short Form 20 Mental Health subscale (r = 0.73), [33] and the SCL-90-R also has demonstrated validity within psychiatric populations. [37, 38]

Assessment Time Points

Time points for assessing depressive symptoms pre- or post-intervention varied across studies (Fig. 3). All studies assessed depressive symptoms at baseline before any intervention was given. Most studies (n = 8) assessed depressive symptoms right after the completion of the intervention, while the remaining studies relied on follow-up assessments between 1 and 12 months post-intervention [19; 22; 26; 27; 29; 32]. Only one study assessed depressive symptoms each week during the intervention period [4]. All but one [26] studies assessed the depressive symptoms using the same validated measure at each time point. Eller et al. (2013) assessed baseline depressive symptoms using the CES-D and a single item. However, in this study, the CES-D was not used in their post-intervention follow-up assessment and only the single item was used.

Fig. 3.

Time Points for Assessment of Depressive Symptoms

Description of Control Groups

Participants randomized to the control arm of the included studies received usual care. However, the definition of the usual care was not described in most studies. When explicitly written, usual care was an education on management of HIV or in general health. Usual care in the control groups was delivered via face-to-face or paper-based forms and sometimes deviated from the delivery method of the intervention. For example, a study by Guo et al. (2020) delivered intervention content via web-based platform during a period of 3 months, while the control group only received a single brochure on nutrition guide.

Description of Self-Management Interventions

Interventions were delivered through technology (web-based or smartphones) [24, 25, 30, 32], interactive group sessions with professionals [20, 21, 23, 29], individual session(s) with professionals or lay workers [19, 22, 26], or a combination of these different approaches [27, 31]. For interactive group sessions, the description about the context/setting where intervention was delivered was not captured for all studies.

The duration of the intervention varied substantially across the studies, ranging from a single session [26] to 12 weeks [32] or 3 months [24]. Most of the studies did not explicitly report on the exact length and frequency of each session within an intervention.

Based on our operationalization and categories of self-management interventions, there were twelve that were categorized under coping, ten that were categorized under needs, and one that was categorized under resources. Among twelve interventions that addressed coping, three specifically focused on coping skills [19, 30, 31] either on an individual basis or in a group setting. Only one study addressed resources that fall into self-management process, and this study utilized a lay worker to provide skills and guidance on where individuals can find additional support and resources [22].

Three studies used Cognitive Behavioral Stress Management (CBSM) [20, 21, 24], which is a short-term self-management approach focused on emotional/behavioral regulation [39]. This approach aligned with needs and coping under self-management process. Whereas two studies that used CBSM occurred in person as a group [20, 21] Guo et al. (2020) adapted the CBSM into a web-based program (i.e., WeChat) where it allowed participants in the intervention group to self-educate on the intervention content on a weekly basis.

Three studies used the HIV/AIDS Symptom Management Manual, which is a paper-based symptom management manual with self-care strategies for 21 common HIV/AIDS symptoms developed by a team of researchers at the UCSF School of Nursing [27]. However, the mode of delivery of the intervention differed, where Wantland et al. (2008) used a paper-based format that was administered at their participants’ discretion (self-teach), Eller et al. (2013) used the paper-based format delivered by a nurse, and Schnall et al. (2018) used a mobile platform to deliver the intervention content.

Individualized coaching sessions were used in two studies. van Luenen et al. (2018) adapted a self-help booklet into a web-based program, followed by an individualized coaching session based on motivational interviewing with a master’s level graduate student in Psychology. Similarly, in a study led by Nyamathi et al. (2012), participants in the intervention group received the Asha Life intervention and six individualized coaching sessions at individuals’ homes with a lay community health worker who offered lessons in HIV, nutrition, and life skills.

Effects of Intervention on Depressive Symptoms

Among the included studies, one study that utilized clinician-delivered teleconferencing in group settings did not show any effect of the intervention on depressive symptoms [30]. Of the remaining 12 studies with statistically significant reduction in depressive symptoms in the intervention group, four studies [22, 23, 25, 32] showed significant differences in depressive symptoms between intervention and its control groups. These four studies focused on individualized/personalized content either via personal coaching [22, 23, 25] or selecting content based on participants’ needs via mHealth format [32].

Discussion

In our review, we identified 13 articles focused on self-management of depressive symptoms in PLWH using an RCT design. Many of the interventions showed initial efficacy in reduction of depressive symptoms, providing evidence for the use of self-management interventions for ameliorating depressive symptoms in PLWH. Comparatively, there is substantial literature documenting the efficacy of psychostimulants, conventional antidepressants (e.g. selective serotonin reuptake inhibitors), and dehydroepiandrosterone, for the treatment of depression in HIV [39]. Nonetheless, many of these studies lack rigor (i.e. comparative studies, no follow-up data) making it difficult to draw a firm consensus [39]. Therefore, there remains a need for self-management interventions, such as those detailed in this review, as well as consideration of psychotherapy which also has demonstrated efficacy for treatment of depression [39].

Differences in depressive symptoms were noted between the intervention and control groups depending on the method of content delivery. The use of individualized/personalized approaches leveraging technology or face-to-face time with a coach/professional [22, 23, 25, 32] yielded significant decreases in depressive symptoms compared to those conducted in a manualized format or a group setting. Personalized and regular engagement with the intervention content over time may have provided reinforcement of the behaviors, and therefore may have contributed to the significant reduction in depressive symptoms. Traditional manualized self-care interventions did not reduce depressive symptoms beyond the initial study period. These findings support the notion that knowledge of self-management strategies alone does not translate into behavior change [40].

Most studies emphasized coping as the primary form of self-management through their intervention contents. Moreover, the largest proportion of significant decreases in depressive symptoms came from studies centered around needs identification and coping skills development and may suggest the high efficacy of needs and coping-based self-management strategies for targeting depressive symptoms. This presumption aligns with current research which cites the effectiveness of HIV-related depression management strategies [41].

Our review identified only one study that used resource utilization for the management of depression symptoms. A study by Nyamathi et al. focused on resource-based self-management through the incorporation of linkages to community resources for depression management within the intervention group [22]. Apart from this study, we found no self-management intervention focusing on resource referrals amongst PLWH. However, the nature of referrals to other resources including professional mental health support is separate from the concept of “self-management” and may not be within the scope of this review. Given that only one study focused on resources, it is premature to conclude that self- management interventions incorporating resources are effective and therefore future studies should assess how to include resource-based strategies in self-management interventions.

Notably, none of the studies presented findings related to sex differences in depressive symptoms. This is noteworthy since, in our own work, we noted significant differences in the symptom experience of women as compared to men living with HIV [41]. On the other hand, menopausal status in women living with HIV has not been shown to have a relationship with depressive symptoms [41] pointing to the complex sex-based differences in the symptom experience of PLWH.

There were several limitations of this review. First, the standard of care/control arm was poorly described in most studies making it difficult to fully identify the intervention effects. Second, there was heterogeneity in the measurement of depressive symptoms across studies. Future studies would be strengthened by comparing the effectiveness of these interventions and using the same standardized measurement tools for evaluation. It is especially important to use comparable validated measures, studies in this review did not. Thirdly, there were risks of bias that were inherent in the studies due to lack of reporting on whether the study was able to conceal group assignments. This is due to the nature of the behavioral intervention where assignment to intervention/control groups can be difficult to conceal in comparison to drug trials. However, studies should mention strategies that they used to conceal assignments so that there is no risk of bias from the perspective of the participants. Additionally, many studies had high dropout rates in the intervention group, indicating that the strategies for retaining participants and designing recruitment to mitigate attrition are critical to further develop before study enrollment.

Finally, many of the studies did not fully describe the duration and frequency of the intervention sessions.

Despite these limitations, most of the included studies showed a positive reduction in depressive symptoms in PLWH who were in the intervention group. Given these findings and the need to address this disproportionately prevalent and burdensome issue of depressive symptoms among PLWH, future work should consider how to best disseminate these interventions as well as the best approach for their implementation in clinical and community settings. The findings in this review attest to the challenges in dissemination and implementation of interventions, despite positive findings from clinical trials [42].

Conclusion

Given the increased rates of depressive symptoms in PLWH [43] and the negative health impacts of these symptoms, including treatment non-adherence and poor disease prognosis [44], there is a critical need for identifying efficacious interventions for addressing depressive symptoms in PLWH. Opportunities for an individualized approach and personalization through individual coaching or technology can hold promise for reducing depressive symptoms. This review synthesized the evidence of self-management interventions and can be used to inform scale-up and dissemination of interventions that effectively improve depressive symptoms in PLWH.

Authors’ Contributions

All authors on this paper have contributed to the conception and design of the study, drafted or have been involved in revising this manuscript, reviewed the final version of this manuscript before submission, and agree to be accountable for all aspects of the work. Using the CRediT taxonomy, the specific contributions of each author are as follows: Conceptualization: M. Yoo Jeong, J. Barroso, R. Schnall; Methodology: M. Yoo Jeong; Data Curation: M. Yoo Jeong, G. Alvarez; Formal Analysis: M. Yoo Jeong, G. Alvarez; Resources: G. Alvarez; Supervision: R. Schnall; Writing – original draft: M. Yoo Jeong, R. Schnall; Writing – review and editing: G. Alvarez, G. Khawly, J. Voss, T. Wang, J. Barroso.

Funding

Research reported in this publication was supported by the National Institute of Nursing Research through the National Institute of Health under award numbers: R01NR015737 and K24NR018621. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The findings and conclusions in this review are those of the authors and do not necessarily represent the official positions of the official views of the National Institutes of Health.

Data Availability (data transparency)

Not applicable.

Code Availability (software application or custom code)

Not applicable.

Declarations

Conflicts of interest/Competing Interests

The authors report no real or perceived vested interests related to this article that could be construed as a conflict of interest. All authors declare that they have no conflicts of interest.

Ethics Approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Consent to Participate

Not applicable.

Consent for Publication

Not applicable.