Psychological well-being and depressive symptoms in individuals who have lost a family member by suicide

Abstract

Background and objectives

The impact of death by suicide represents an important public health issue. The current study reports data relating depressive symptoms and psychological well-being to exposure to suicide in the family.

Methods

A final Portuguese community sample of 266 participants answered a protocol presented on a Google platform. Two groups were defined: a suicide exposure group (n = 45) and a control group (n = 210) who responded to the Échelle de Mesure des Manifestations du Bien-Être Psychologique (Psychological Well-being Manifestation Measure Scale) and to the Center for Epidemiologic Studies Scale to access psychological well-being and depressive symptoms, respectively.

Results

Groups differ significantly on psychological well-being, even when contorting for unemployment and having physical disease, differed on depressive symptoms and tended to differ on depressive symptoms when contorting for unemployment and chronic physical disease.

Conclusion

Based on results, we suggest that therapeutic interventions should focus not only on reducing depressive symptoms in exposed individuals, but as well as promoting well-being. Limitations include the size of the sample and the cross-sectional nature of the design.

The impact of death by suicide represents an important public health issue.¹ Loss by suicide requires more time to grieve than other types of bereavement² and exposure to suicide results in a number of adverse consequences, namely distress and the increased likelihood of one's own suicide.³ There is some evidence relating exposure to suicide in the family and depression,^4,5 but little data has been reported looking at suicide exposure in relation to low positive emotions and low levels of psychological well-being^6. The absence of psychopathological symptoms, namely depressive symptoms does imply well-being. In fact mental health is not the opposite, or simply the absence of psychopathology. In Portugal, Santos, Campos and Tavares⁷ examined whether several indicators of distress, namely depression, related to suicidal ideation in individuals exposed to suicide in the family and just one study⁸ compared a group of individuals exposed to family suicide with a control group, demonstrating that those college students in the exposure group presented significantly higher levels of depressive symptoms than the controls. No study has looked into whether exposed individuals presented lower levels of psychological well-being. The current study aimed to contribute to the understanding of the effect of being exposed to a suicide in the family and reports data relating depressive symptoms and psychological well-being to exposure to suicide in the family.

Materials and methods

A final Portuguese community sample of 266 participants (205, 80.4% women) answered a protocol presented on a Google platform. It related to psychological variables, thoughts about death, and risky behaviors. Participants were not compensated for their participation. Participant mean age was 35.89 years (SD = 12.42, Median = 32.00) and ranged from 18 to 64. Only nine participants were students, and mean years of education was 13.95 (SD = 2.57, Median = 15.00), ranging from 6 to 19. Fifty-two (20.4%) were unemployed. Twenty-seven (10.6%) of the participants had received a psychiatric diagnosis (44.4% with a depressive disorder, 29.6% with an anxiety disorder 22.2% with a mixed anxiety and depressive disorder, and 3.7% with another diagnosis) and 53 (20.8%) had been diagnosed with a chronic physical disease.

Individuals provided informed consent on the first page of the online protocol and then were asked to click “continue” and received a brief socio-demographic and several clinical questionnaires including the Psychological Well-being Manifestation Measure Scale and the Center for Epidemiologic Studies Depression Scale. All participants were provided with telephone numbers of available mental health services. Participation was voluntary and anonymous and could be discontinued at any time. Three individuals were excluded because they reported residing outside of Portugal.

Participants were also asked if any family member had died by suicide and, if so, information was requested about the family member, when, what suicide method was used, and the degree of impact (using a 1-minimal impact to 5-high impact scale). Sixty-nine participants answered yes, however ten of these reported a date for the suicide that was before the participant's birth, and another 11 failed to report any date. Consequently, we defined a suicide exposure group (n = 48) as those individuals who had lost a family member by suicide during their lifetime with an impact different than minimal in addition to a control group (n = 210). The 11 individuals who did not report the date of the suicide were eliminated from the analysis. Of the 48 members of the exposed group, 15 were grandchildren of the deceased, one was a brother, one was a great grandchild; four were children; two were daughters; eighteen were cousins; thirteen were nieces or nephews; one was a father-in-law; and three reported other kinship. For ten exposed individuals (22.2%), two or more family members had died from suicide. Hanging was the most frequent method (approximately 40%) followed by intoxication (approximately 24%), using medication, a drug or poison. Mean time since the suicide was 16.64 years (SD = 10.89).

Participants responded to the Échelle de Mesure des Manifestations du Bien-Être Psychologique (Psychological Well-being Manifestation Measure Scale),⁹ a 25-item five-point Likert scale (1-never to 5-always), originally developed in Canada, that access manifestoes of well-being in the previous month. Examples of items are: "I felt confident"; "I felt good, at peace with myself." The Cronbach alpha of the original version of the questionnaire was .93. In the Portuguese adaptation,¹⁰ used in the present study, a Cronbach alpha of .93 was also obtained. In the present study Cronbach alpha was .97.

Participants also responded to the Center for Epidemiologic Studies Depression Scale - CES-D¹¹ is a 20-item measure that assessed the frequency of depressive symptoms in the week prior the assessment. Items (e.g., “I felt depressed”) are answered on 4-point Likert scales from (0 - “Never or very rarely - less than 1 day” to 3 - “Very frequently or always - 5-7 days”). Radloff ¹¹ reported scale score coefficient alphas of .85 and .90 for community and clinical samples, respectively. In the Portuguese version¹² internal consistency coefficient alpha values have ranged from .85 to .92 in various samples. In the current study, the coefficient alpha was .91.

Results

Table 1 presents descriptive results as a function of group membership (suicide exposure or control) for the socio-demographic variables of age, gender, years of education, unemployment and marital status, and whether participants had a psychiatric diagnosis and/or chronic physical disease. Descriptive statistics by group for psychological well-being and depressive symptoms are also displayed. For all variables, suicide exposed and control groups were compared, using t tests, ANOVA or χ² tests. Each group's psychological well-being and depressive symptoms were also compared, controlling for other relevant variables using ANCOVA.

Table 1.

Descriptive and inferential statistics for control (n = 210) and suicide exposed (n = 48)) groups.

Variables		n	%		Mean (SD)	n	%	Mean (SD)	t / χ^2
Socio-demographic and clinical variables		Control Group				Exposed Group
Age in years					35.79 (12.36)			36.33 (12.85)	0.27
Gender	FemaleMale	16740	80.7%19.3%			3810	79.2%20.8%		0.06
Years of education					14.06 (2.21)			13.45 (2.78)	1.41
Unemployed	NoYes	17136	82.6%17.4%			3216	66.6%33.3%		6.10 *
Marital status	Not married or living togetherMarried or living together	12790	62.4%38.6%			3018	62.5%37.5%		0.02
Having a chronic physical disease	NoYes	17037	82.1%17.9%			3216	66.6%33.3%		5.66 *
Having had a psychiatric diagnosis	NoYes	18621	89.9%10.1%			426	87.5%12.5%		0.23
Psychological well-being					87.74 (21.05)			79.11 (19.97)	2.57 *
Depressive symptoms					20.70 (10.61)			24.77 (12.91)	2.30 *

Note.^. * p < .05.

As displayed in Table 1 suicide exposed individuals did not differ from controls regarding age, gender, years of education, marital status, or having a psychiatric diagnosis. However, the groups did differ significantly on psychological well-being (t = 2.83, p <0.01; F = 6.59, p <0.05; partial η² = 0.03) even when contorting for unemployment and having chronic physical disease (F = 5.28, p <0.05; partial η² = 0.02). Groups also differed regarding depressive symptoms (t = 2.24, p < .05; F = 4.17, p < .05; partial η² = 0.02) and tended to differ when contorting for unemployment and chronic physical disease (F = 3.31, p <0.10; partial η² = 0.01).

Discussion

Results demonstrated that exposed individuals not only presented significantly higher levels of depressive symptoms, confirming previous results, but also reported lower levels of psychological well-being, suggesting that therapeutic interventions should focus on reducing depressive symptoms in exposed individuals, as well as promoting well-being. Often, an excessive focus on psychopathology neglects important dimensions of the individual, such as well-being.¹³ Stigma related to death by suicide may be one of the factors that impact survivors' well-being.⁶ Suicide prevention programs should include strategies targeting broad-based populations aiming to reduce the stigma associated to death by suicide.

The primary limitation of the present study was the size of the sample and the cross-sectional nature of the design. Particularly, the group of participants whose family member died before his/her birth was too small (just 10 participants) to be compared to individuals exposed during life time and to the control group. Secondly, we relied solely on self-report measures to assess depressive symptoms and psychological well-being. Also, the obtained effect sizes were small. Finally, it should be noted that the Google platform does not reregister individuals that start responding but discontinue their participation before the end of the protocol.

Ethical considerations

Authors declare that the research meet all the guidelines of the Portuguese Psychologists Board

Funding

Authors declare that the paper was partially supported by FCT Fundação Para a Ciência e a Tecnologia, I.P., by the project UIDB/04312/2020