In one of his famous quotes, Albert Einstein once pointed out the potential of goals for the enhancement of well-being: ‘‘If you want to live a happy life, tie it to a goal, not to people or things’’ 131. Since then, research in the field of motivation has presented strong theoretical and empirical foundations for the association between striving for personally relevant intrinsic goals and psychological well-being 2345678. Longitudinal studies and path analyses have indicated that personal commitment to the pursuit of personal goals predicts psychological well-being (PWB) 910. Furthermore, people can achieve sustainable gains in happiness, and generate positive experiences, by setting new goals and intentionally changing their actions (with efforts and engagement), instead of waiting for changes in their life circumstances 11.

However, when important obstacles (real or perceived) make the goal hard to achieve and the person’s efforts are in vain, a state characterized by disorganization, discouragement, and helplessness is observed 12. The duration of this state depends on the importance and the value given to the goal. Brandtstädter 12 found that difficulty in shifting from tenacity (persistent effort to modify current situation to reach a goal) to flexibility (relinquishing blocked goals and changing one’s aspirations) seemed to predict depression. Other studies have also shown that disengagement from unattainable goals and reengagement in more satisfying, feasible, alternative goals were associated with a high level of subjective well-being and attenuated depression when facing negative life events 131415. More precisely, goal disengagement capacities can reduce psychological distress, while goal reengagement capacities are associated with positive affect and purpose in life 16.

Attitude toward the future and the regulation of unattainable goals have been studied in the context of suicide risk. Hopelessness (pessimism toward the future) has been consistently predicting suicidal ideation and behavior 171819, while hope and meaning in life seemed to serve as protective factors 2021. Moreover, low levels of positive future thinking, rather than the preponderance of negative future thinking, were a better predictor of suicidal ideation than hopelessness, in patients who had attempted suicide 2223. Additionally, suicidal ideation were significantly higher 2.5 months after a suicide attempt among persons who reported difficulties to reengage in new goals when those they were pursuing were unobtainable 24. The relationship with suicidal ideation is particularly strong in those who reported low levels of goal reengagement coupled with high levels of goal disengagement 24. Furthermore, difficulty reengaging in new goals independently predicted self-harm (by overdose) two years after a suicide attempt, in another sample of patients 25.

These results provide interesting options for suicide prevention and intervention. Helping people who lost interest in life engage in new personal projects that offer meaning to their life could be an innovative prevention strategy 23242526. This is especially important in old age when illnesses or disability make goals unattainable, and consequently increase the wish to die and suicidal ideation 2728.

Brief interventions that target goal-setting and planning have been created to improve well-being 92930, but those designed to prevent suicidal ideation are uncommon, especially for older adults 31. In fact, most suicide prevention programs are centered on the reduction of risks factors and not on strengthening protective factors 31.

However, one program has emphasized the development of meaning in life by helping participants set, plan, pursue, and realize meaningful, concrete personal goals 26. Although this 11-week group workshop was not designed with suicide prevention in mind, nor was it offered to a suicidal population, the analyses of data from 21 retirees expressing suicidal ideation (from a sample of 354 retirees) showed statistically significant differences between the experimental and control groups. By the end of the program, suicidal retirees in the experimental group (n = 10) had improved significantly more than those in the control group (n = 11) on PWB and their levels of depression and psychological distress significantly decreased. These gains were maintained six months later. Although interesting, the results from this study had some limitations. To be confirmed as an innovative approach for suicide prevention, it needed validated suicidal ideation questionnaires to evaluate the outcome variable. Also, since the group intervention format could be adding some protection against suicide 32, the research should have evaluated the effect of the program on social variables, such as loneliness or relatedness.

The objective of the present research was to replicate the study design of Lapierre et al. 26 with a sample of depressed older adults aged 65 and over, since depression is a high risk factor for suicide in old age 33, and also to extend the list of outcome variables by adding some measures of affiliation and suicidal ideation. Since this type of intervention had never been offered to depressed individuals, we primarily wanted to see if the goal management program could be useful to them.

First, it was hypothesized that PWB (measured by several indicators) and social abilities of the participants in the goal intervention program would improve statistically, compared to pre-test levels, and that their levels of depression and suicidal ideation would decrease statistically.

Second, it was essential to compare the PWB of the participants in the intervention group to that of a control group who did not take part in the program. It was hypothesized that the participants in the goal management workshop would improve beyond the controls on all measures.

The intervention program was offered to 42 older adults. Fourteen dropped out of the program (9 after the first or second meeting; 5 after the third or fourth meeting). Reasons mentioned included that they did not identify with other depressed participants, that the program or group format was not for them, or that they had fallen behind in the “class” after having missed the content of the first meetings. Three participants completed only 6 to 8 meetings (because of illness or conflicting schedule), and 1 did not complete the follow-up questionnaires. Twenty-four (24) participants completed more than 12 meetings and persisted until follow-up. The control group included 29 respondents; 8 dropped off at post-test, and 3 more at follow-up. Eighteen completed all three time measurements.

There were significant difference on age (t(40) = 2.89, p < .01) and social desirability (t(37.9) = 2.00, p < .05) between those who dropped out of the program and those who persisted. Individuals who left the program were significantly older (M = 73.2) than those who persisted (M = 68.6) and they wanted to be seen in a positive light (social desirability: dropped out, M = 9.17 vs persistent, M = 7.71). As for the control group, those who dropped out were also significantly (t(27) = 3.24, p < .01) older (M = 75.6) than those who persisted (M = 69.5), and had a lower level of education (t(27) = 2.18, p < .05). There were no statistically significant differences on any psychological variables, between those who abandoned and those who persisted, in each group.

The final sample included 42 older adults aged from 65 to 84 years (M = 71.9; SD = 5.6), 24 in the intervention group (14 women; 10 men) and 18 in the control group (12 women; 6 men). Most of them (57.1%) were unmarried (33.3% divorced; 16.7% widowed; 7.1% single). The average level of education was 12.1 years (SD = 4). There were no significant differences between groups on age, level of education, marital status, gender, perception of health, or satisfaction with finances.

Four (9.5%) participants (3 in intervention group; 1 in control group) were pursuing an individual therapy, ten (23.8%, 5 in each group) were taking prescribed medication for depression and 17 (40.5%, 12 in the experimental group and 5 in the control group) were taking insomnia medication (information on type and dosage of medications was not available). Seven persons (16.7%; 5 in intervention group, 2 controls) said that they had tried to end their life in the past. No information is available on participants’ past psychiatric history or traumatic events.

Each participant completed 13 questionnaires (14 with the MMSE) during an individual interview that lasted 90 minutes. The interview started with a general information questionnaire, followed by the MMSE, to ensure that participants had no cognitive problems and could take part in the study. Those who had a score below 24 on the MMSE were excluded from the study. Afterwards, they answered the Beck Depression Inventory-II (BDI-II) to confirm their eligibility to the study, and completed the other questionnaires if scores were within the required range. If not, they were thanked for their help and referred to other psychological resources by the research coordinator. Table 1 summarizes the main groups of measures that were used in the study.

In order to verify the hypotheses, repeated measures of variance were done using SPSS Statistics 24. A statistical a priori power analysis was performed for sample size estimation, based on an effect size of 0.25, which is considered moderate using Cohen's (1988) criteria, an alpha of .01, and a power of 0.80. The sample size needed with this effect size (GPower 3.1) was approximately N = 40 for repeated measures between/within group comparisons. Therefore, our projected sample of 66 seemed to be adequate to detect the effect of the program and allowed for the expected attrition which led to a final sample of 42 participants.

Preliminary analyses showed no significant difference between groups at pre-test on the inclusion variable of depression: The mean score of the intervention group was 21.1, and 23.3 in the control group (t(40) = .88, p = ns). In the sample (N = 42), scores on the BDI-II ranged from 9 to 39. According to the norms 36, depression was minimal (scores from 9 to 13) for 16.7% of the participants, mild (14 to 19) for 21.4%, moderate (20 to 28) for 42.9%, and severe (≥ 29) for 19%. Proportions were similar in both groups. It should be noted that 83.4% of the participants also suffered from insomnia, which is also a risk factor for suicide 52.

Other comparative analyses revealed significant differences between groups at pre-test for social inclusion (t(40) = 4.77, p < .001) and autonomy (t(40) = 2.49, p < .05), which were both higher in the intervention group. Also, controls were significantly lower on the GSIS meaning in life subscale (t(40) = 10.48, p < .001), and on the LS (t(40) = 4.15, p < .001). Essentially, at pre-test, depressed participants in the intervention group felt more included and supported by their social environment, and that they could decide and take charge of their life, while controls felt isolated and that their life lacked meaning. Table 2 presents the mean scores of each variable for each group.

The goal of this study was to verify if the personal goal realization program could improve the PWB of depressed participants. Therefore, we first looked at the results of repeated analyses of variance only for those in the intervention group (with the Greenhouse-Geisser correction after testing for Mauchly’s sphericity test). Results showed that the first hypothesis was confirmed: Depressed older adults taking part in the program showed an improvement on most goal variables and PWB indicators at post-test. This improvement was maintained six months later (see Table 2). More precisely, there was a significant increase in participants’ perceived ability to set realistic and concrete goals, to plan efficient means and concrete actions to achieve them, and to identify strategies to overcome obstacles (higher GRP scores). There was also an increase in their level of hope, especially their determination to move into action and their ability to find alternative means to reach their goals. Participants in the goal program also seemed to have satisfied two out of three basic psychological needs: they felt more competent to achieve desired outcomes, and felt related to others. Moreover, results suggest that the participants showed an increase in their level of serenity, indicating that they could accept what couldn’t be changed, and did not let events influence their well-being. Higher scores on the LRI showed that they also felt that there was meaning to their life, that they saw life more positively, had clear objectives for living, and were fulfilling their life goals. Notably, there was a significant decrease in the levels of depression, insomnia, hopelessness, loneliness, and on the GSIS, especially on the sub-scales that depicted suicidal ideation, lack of meaning to life, and loss of personal worth. It should be noted that men seemed to improve as much as women, since there was no significant difference between genders.

The second hypothesis proposed that the participants to the goal intervention program would improve beyond the controls on PWB at post-test and follow-up. Gain scores were chosen to analyse the results because they provide an easy way to test whether a group improves or deteriorates between two measures 53. Ancova (with pre-test scores as covariate) was rejected because it usually requires that covariate does not vary between groups, which is often a problem in non-equivalent group designs such as ours. Therefore, in order to test the second hypothesis, three change scores indicating progress were calculated for each variable: one between pre-test and post-test, the second between post-test and follow-up, and the third between pre-test and follow-up (see Table 3).

Results of the analyses of variance showed that the intervention group made significantly more progress than the control group between pre-test and post-test on the LRI, especially the Framework subscale (having a set of life goals and seeing life more positively). Between pre-test and follow-up, the gains were greater in the experimental group for the Fulfillment subscale (sense of accomplishment) and also for the Serenity scale.

Results also revealed many significant differences between groups in change scores between post-test and follow-up. Serenity decreased significantly in controls, while lack of self-worth increased, compared to those in the intervention group. On the other hand, controls made significant gains, compared to the intervention group, in autonomy and social inclusion. There was also a significant reduction in loneliness, and in the lack of meaning in their life. It should be noted that, at pre-test, the controls were significantly lower than participants in the intervention group on those last four variables. Comparison of the change scores also indicated a significant diminution in tenacity in the control group, again during the six months after post-test. It seems that the controls became less persistent in their pursuit of their goals. Finally, it was observed that controls’ scores (see Table 2) were always poorer than those of the intervention group on all variables.

As stated in a systematic review of suicide prevention programs designed especially for older adults 31, interventions centered on the development of protective factors are extremely rare. Based on theoretical and empirical motivation research, the program “On the road to a better life”, was an attempt at creating an alternative method to improve the state of depressed older adults who lost their interest in life by teaching them how to engage in new goals. This innovative approach seemed beneficial to the participants and could be promising for suicide prevention. Undoubtedly, leading a meaningful life is incompatible with thoughts of self-destruction. In fact, the increases in levels of PWB (hope, competence, positive attitude toward life-framework, sense of accomplishment-fulfillment, serenity) are interesting since suicidal persons are usually characterized by helplessness, negative beliefs about their future and about reaching their aspirations, as well as difficulty in generating alternative solutions when facing obstacles 2125. More importantly, program participants expressed less depressive symptoms and suicidal ideation at post-test and follow-up, six months later. Similar effects were observed in terminally ill older adults who took part in a meaning-centered group psychotherapy: Meaning in life was significantly enhanced, and wish for hasten death was reduced, compared to a support control group 54. The program “On the road to a better life” could provide an interesting addition to the choice of treatments available for depressed suicidal older individuals, especially for men who might prefer a program centered on action and problem solving to a therapy focusing on the expression of emotions and inner conflicts 55.

The group format of the program seemed to have been beneficial to create new relationships since 87.5% of the participants in the intervention group met after the end of the program, and 21% met more than ten times. Social support can act as a protective factor against suicidal ideation 3256. Then again, a complementary regression analysis indicated that perceived ability to realize one’s goals (GRP scores) at post-test significantly predicted the reduction of geriatric suicidal ideation scores (β = -0.47; t =-3.64, p < .001, R²_adjusted = 0.41), while social variables (social inclusion, relatedness, loneliness) did not.

The goal management program has not proven to be more effective compared to the control group, except for LRI, namely “the belief that one is fulfilling life-goals that provide a highly valued understanding of one’s life” 41 ^{p. 410}. Nonetheless, it should be noted that changes in the controls happened during the six months between post-test and follow-up, while the changes in the intervention group were seen immediately after the program and the improvement was maintained until follow-up. Even if participants were asked if they faced any stressful events during that period, there was no question about positive events, new activities, or relationships that might have influenced participants’ perception of life. Also the changes in the controls could simply reflect the effect of summer activities (follow-up was done after summer for the control group and in winter for the intervention group) on the life of Canadian elderly persons who are often trapped at home during a cold and icy winter that lasts six months, a season when they are particularly confronted to their functional limitations. Moreover, the improvements were seen mainly on variables (loneliness, autonomy, social inclusion, and meaning in life) where the controls were significantly lower than the intervention participants at pre-test, suggesting a regression toward the mean. On the other hand, during the same period, controls deteriorated on self-worth and serenity; they also showed a significant decline in tenacity indicating that they weren’t making any efforts to pursue their goals. This last observation could indicate higher disengagement, increased resignation, or merely enjoying the present. New research should look closely at changes in tenacity and flexibility, along with goal disengagement and reengagement, in depressed older individuals.

It has been demonstrated that mental health promotion programs and psychotherapy contribute to the reduction of depression and suicide 5960. Future research should continue looking at interventions that increase protective factors. Medications are often necessary to improve depressed individuals’ mood, but do not give meaning to life. Also, programs enhancing various protective factors might have longer lasting effects, making people happier and actively participating in the community. The program “On the road to a better life” is a type of positive psychology intervention cultivating positive feelings, behaviors, and cognitions through meaningful goal achievement experiences 60. The use of such strategies for suicide prevention could provide an interesting and innovative approach to improve the state of individuals who have lost their interest in life.