The phenomenon of loneliness in old age is currently receiving more attention from service developers and the research community than in the past, due both to its prevalence and to its connection to the health status and emotional well-being of older adults. In 2014, a request for proposals was issued for the development of programs to reduce loneliness in the framework of a national initiative known as the Project to Reduce and Cope with Feelings of Loneliness among Older Adults (in short, the Loneliness Project). It was headed by the National Insurance Institute’s Fund for Demonstration Projects, JDC- ESHEL and the Ministry of Labor and Social Affairs and Services (MOLSA). The project included 23 different programs operated by 13 organizations (for example, associations for older adults and social service departments) dispersed widely throughout Israel. The directors of the Loneliness Project commissioned the Myers-JDC-Brookdale Institute (MJB) to evaluate the project. The current study was conducted by MJB over three years.
The study had three main goals: 1. To examine the contribution of each program implemented in a locality to the older adults participating (the individual level); 2. To examine each program’s contribution to further addressing the issue of loneliness in every locality where it was conducted (the field level); 3. To formulate insights and recommendations about the programs to reduce loneliness (the policy level).
The study included: Analysis of the intake records of participants in 21 programs; 2 interviews with the participants prior to the program (751 interviews) and 685 interviews six to nine months after the start of the program; 14 observations of activities and 14 focus groups with participants; 36 interviews with program coordinators and professionals in the localities where programs were conducted; 60 self-report questionnaires were sent to program implementers (group leaders, therapists, etc.). A comparison was made between the programs according to their intervention strategies; likewise, a comparison was made between the program participants and a control group comprising people who were meant to participate in a certain program but did not actually do so. The findings were analyzed using quantitative tools of descriptive and inferential statistics using bivariate analysis (x2 , Wilcoxon for paired samples, and t-test for dependent and independent samples) and multivariate analysis (probit regression, logistic regression and linear regression). Qualitative tools for thematic analysis were also used.
The contribution of the programs to the participants (individual level) was examined through the interviews with participants, program coordinators and implementers, and the observations and focus groups with the participants. The interviews with the professionals, coordinators and implementers were also used to evaluate the contribution of the program and the overall project to addressing the issue of loneliness in old age at the local level. Insights about strategies to address the issue of loneliness in old age at the national level were gleaned from all the sources of information.
Seventy-three percent of the people enrolled in programs reported during intake that they frequently or sometimes felt lonely, compared with 34% of all people aged 65+ in Israel. Overall, 40% of the respondents dropped out of the programs (n=271), either at the beginning (23%) or after one or two meetings (17%). Among those who participated in all, or almost all, of the programs (n=414), there was a high level of satisfaction.
The respondents were asked to what extent they felt that the program had helped them in various aspects of their lives. Fifty-seven percent noted that it had helped reduce their sense of loneliness to a great or very great extent; 71% reported improvement in their general mood to a great or very great extent; 48% reported an improved feeling in their general health; 45% reported that the program had helped their ability to cope with difficulties and unpleasant feelings to a great or very great extent.
The various programs were based on two intervention strategies: an opportunity for social integration and the strengthening of social competence. The characteristics of the participants were different in each of the strategies. The participants were matched to take account of their background characteristics and selection bias – selection bias of the program and self-selection bias. This matching made it possible to compare the strategies. The comparison did not find that either of the strategies had a significant advantage over the other in reducing loneliness or depression. In other words, for the “average participant” – an individual with average levels of loneliness and depression – no significant difference was found between the strategies for reducing loneliness and depression.
As to the association between the participants’ background characteristics and the contribution of the
intervention strategy to reducing the levels of loneliness and depression, we found that the social interaction strategy was better suited to people characterized by a relatively low level of initial
loneliness, who are socially active and have no financial difficulties.
A comparison between the level of loneliness and depression among participants before and after
joining the program found an average decline in the UCLA Loneliness Scale and the PHQ-2 depression
measure, in contrast to respondents who did not participate in the programs, where these measures
increased over the same period. A multivariate linear regression analysis found that beyond all other
explanatory variables, participation in the programs of both strategies significantly reduced the level
of the indicator of depression of the participants compared with non-participants. As for reducing
loneliness, participation in programs to strengthen social competence significantly reduced the level of
loneliness of participants compared to those who did not participate, but in social interaction programs
no significant difference was found between participants and non-participants in reducing loneliness.
Furthermore, the project helped to raise awareness about loneliness in old age among professionals in
the localities, to dispel taboos and to professionalize the field. It raised the matter on the social agenda
and led to assimilation of programs and activities at the locality level.
Conclusions and Recommendations
The study findings show that the various programs were able to reduce loneliness in different ways
and that for the “average participant,” neither intervention strategy was preferable to the other.
It is recommended that the program strategy be chosen according to the characteristics of the target
population and that the particular contents of the program be determined together with the participants
according to their areas of interest. It is also recommended that in each locality at least two programs
based on different strategies to reduce loneliness that can respond to different needs, be implemented.
We recommend that the programs be implemented according to the principles set out in the full report.
It is further recommended that the subject of reducing loneliness among the older adults be introduced
in in-service and other training programs for professionals. It is recommended to structure cooperation among the care agencies in order to identify older adults living alone, particularly those who are not
socially active, and reintegrate them into the community and to implement a structured intake process,
while developing different versions suitable for the needs of the different professionals. We recommend
preparing a compendium of programs for reducing loneliness, based on the knowledge accumulated
during the project. We also advise allocating resources and bolstering the projects in the geographic and
social peripheries and in small localities.