• Change from baseline psychological resilience at 3 months [ Time Frame: baseline and 3 month follow-up ]
  We measured the psychological resilience of the seniors by the Resilience Appraisal Scale (RAS) which contains 12 questions through which seniors indicated the degree of applicability of each statement to them using a five-point Likert scale ("strongly disagree" to "strongly agree"). The total RAS score ranges from 12 to 60, with a higher score indicating higher perceived psychological resilience.
• Change from baseline perceived social support at 3 months [ Time Frame: baseline and 3 month follow-up ]
  We assessed perceived social support using the 10-item Duke Social Support Index (DSSI). The possible score ranges from 10 to 30. Higher scores indicated a higher level of perceived social support among participants.
• Change from baseline loneliness at 3 months [ Time Frame: baseline and 3 month follow-up ]
  We measured loneliness by the 8-item UCLA Loneliness Scale (ULS-8). The scale employed a 4-point Likert scale with values ranging from "never" to "always" and the total score ranges from 8 to 32. A higher score on this scale indicated more intense feelings of loneliness.
• Change from baseline working alliance at 3 months [ Time Frame: baseline and 3 month follow-up ]
  We measured their working alliance using the Working Alliance Inventory-Short Form (WAI-SF). The WAI-SF scores ranged from 12-84, with higher scores indicating a stronger bond and agreement on tasks and goals.

Poor mental health is getting more common in low- and middle-income countries than in high-income countries due to lack of available resources and access to health services. In these countries, there is a large treatment gap for mental health care, with the majority of people with mental disorders receiving no or inadequate care. Depression, for instance, is one of the most common mental disorders and it affects physical health, social activities, and quality of life of senior citizens. Despite being a commonly studied mental disorder, very little is known about depression interventions conducted in low resource settings.

Recently, Filipinos' mental illness has been increasing and it affects around 10-15% of children and 17-20% of adults. Their major symptoms include excessive sadness, delusion, confusion, and forgetfulness. Additionally, more Filipino senior citizens are committing suicide due to depression. This is associated with their inability to adapt to rapid social and economic developments. In this study, the investigators aimed to assess the efficacy of 3-month-duration interventions with peer counseling, social engagement, and combined intervention vs. control in improving depressive symptoms among community-dwelling Filipino senior citizens.

Study design

The study design was an open (non-blind), non-randomized trial. Senior citizens who got a score suggestive of depression were divided into four groups. The first group (n = 60) only received peer counseling; the second group (n = 60) joined community activities; the third group (n = 60) joined both social engagement and peer counseling, and the last group (n = 60) neither took part in any activities and served as the control group. The investigators analyzed all the participants and assessed the impact of the inventions after three months.

Study area

The study was conducted in Muntinlupa City which is the southernmost city in the National Capital Region, the most populated region in the Philippines. Muntinlupa is classified as highly urbanized city with a poverty incidence of 1.9% in 2012. The city had one of the highest records of senior citizens, which account for 5.63% of its population.

Participants and selection criteria

Participants of this study were community-dwelling seniors in the City of Muntinlupa. Those aged 60 years old and above are regarded as senior citizens in the Philippines. Therefore, the investigators selected Filipino senior citizens who are registered members of the Office of Senior Citizen Affairs (OSCA). Participants were recruited purposively by the primary investigator and trained BHWs through home visits. Each BHW had their own list of senior citizens in their catchment area. The investigators used the list for house-to-house recruitment. The primary investigator and BHWs invited the senior citizens to participate in the study. All participation by seniors was voluntary and participants gave their informed consent prior to participation. Participants must possess a valid senior citizen's identification card to be eligible. The investigators excluded those elderly people in long-term care, with terminal diseases, or with moderate/ severe cognitive impairment and currently suffering from deafness, aphasia or other communication disorders.

The investigators included senior citizens who got a score suggestive of depression (GDS score of 5 and above) in this study. The investigators allocated the senior citizens into four groups: (1) peer counseling, (2) social engagement, (3) combination of peer counseling and social engagement, and (4) control. For sample size computation, the investigators used Open Epi version 3.01 and based the following parameters from a meta-analysis of the effects of outreach programs to depressed senior citizens in the community: effect size of 0.77, power of 90%, alpha set at 0.05 (two-sided) and expected dropout rate of 25%. At least 40 senior citizens per group were calculated. Considering the small sample size, the investigators decided to increase the sample size to at least 60 senior citizens per group.

Each senior peer counselor was in charge of two clients. One client was assigned to group A (peer counseling) and another client was assigned to group C (social engagement + peer counseling). The number of participants for social engagement (n = 60) was decided on the basis of location/ space, manageability, and financial resources. There were two batches for the social engagement group. Each batch consisted of 30 participants.

Data collection and study tools

Peer Counselor - Client Meetings - Peer counselors did home visits to their assigned clients for 1-hour every week for three months. The goals of the meetings were to establish a strong working alliance, identify a client-defined problem, encourage behavior change, and facilitate engagement with the community. At the initial visit, the peer counselor asked what the client would like to get out of the meetings in order to establish a client-identified goal that both can work on together. Peer counselors accomplished weekly reports for documentation purposes. Both the peer counselors and clients completed the Working Alliance Inventory-Short Form (WAI-SF) pre- and post-intervention. Clients answered the same set of questionnaires for depression and psychosocial risk factors after study completion. Peer counselors and clients were interviewed to assess their experiences of the intervention. The investigators conducted separate FGDs (5 members per group) with the clients and peer counselors using a semi-structured questionnaire.

Peer Counselor-Health Provider Supervision Meetings - The health providers (psychologist, physician, pharmacist, BHWs) met with the peer counselors once a month for an hour for supervision and collaboration. During meetings, the peer counselors reported on the client's progress and shared impression and insights. The health professionals provided guidance, reinforcement, and constructive feedback to continue skills development of the peer counselors.

Social Engagement Activities - The trained senior volunteers and health providers facilitated the 3-hour weekly social events for three months. The investigators then collected the same set of data for depression and psychosocial risk factors after the intervention. The investigators conducted FGDs consisting of five members per group for both senior participants and trained senior volunteers. An interview guide was used to explore the trained senior volunteers' and senior participants' experience and personal growth after the intervention. The investigators explored their acceptability and motivation to continue the activities. The investigators encouraged the participants to express their views and opinions without confining to the questions being asked.

Data analysis

The investigators carried out pre- and post-intervention comparisons followed by an analysis of semi-structured interview data. The level of significance was set to 0.05 (two-tailed) and statistical analyses were performed using Stata 13.1 (StataCorp, College Station, TX, USA). For semi-structured interviews, the investigators transcribed the recorded notes verbatim, analyzed, and translated the themes and quotes into English. The investigators analyzed with the aid of NVivo using combinations of inductive and analytical approaches. Five investigators were involved in this process and every transcribed interview was analyzed by at least two investigators to test the reliability of the interpretations. Findings were presented as themes that emerged from the analysis of transcripts for peer counseling and social engagement activities.

Ethical considerations

Ethical approval was sought from the Research Ethics Committee of the University of Tokyo and the University of the Philippines Research Ethics Board. The investigators obtained the permission of community-dwelling seniors by giving a letter of consent/permission to conduct research. All participation by seniors was voluntary, and participants gave their informed consent prior to participation. In addition, participants were allowed to withdraw from the study at any time without penalty and had the right to obtain the results of the study if participants so wish. The investigators were aware not to cause emotional harm to the participants by being careful and sensitive during the interviews and activities. Personal biases and opinions did not get in the way of the research.

• Experimental: Peer counseling group
  Peer counselors performed 1-hour home visits weekly to their assigned clients for three months.
  Intervention: Behavioral: Community-based depression interventions
• Experimental: Social engagement group
  Senior citizens joined 3-hour weekly social events held at the OSCA Center for three months.
  Intervention: Behavioral: Community-based depression interventions
• Experimental: Combination group
  Senior citizens in this group underwent both peer counseling and social engagement interventions mentioned above.
  Intervention: Behavioral: Community-based depression interventions
• No Intervention: Control group
  Senior citizens in this group had access to usual or standard care from health and aged care services that were usually available.

Inclusion Criteria:

• Senior citizens aged 60 years old and above with a valid senior citizen's identification card provided by the OSCA.
• Senior citizens who reported a depression score of 5 or more which indicated a tendency towards depression based on the 15-item Geriatric Depression Scale (GDS-15).

Exclusion Criteria:

• Those elderly people in long-term care, with terminal diseases, or with moderate/ severe cognitive impairment and currently suffering from deafness, aphasia or other communication disorders.