- Top of page
Aims: The purpose of this study was to investigate the relationship between inappropriate views on suicide, such as it being a personal choice, inevitable, unpreventable, and permissible, with demographic variables and the feeling of shame in seeking help among the general population.
Methods: A self-administered questionnaire on mental health and suicide was distributed to all residents aged 40–74 in four areas in Oita Prefecture, Japan, and 4487 responded. The association of seven inappropriate views on suicide with demographic variables was examined by multiple logistic analyses. The association between feeling shame in seeking help with demographic variables and the above views on suicide was similarly analyzed.
Results: Inappropriate views on suicide were associated with gender (i.e. men). Some of these views also correlated with age, never having been married, and living in rural areas or areas with high suicide mortality rates. Multivariate analysis revealed that feeling shame in seeking help when distressed was associated with being aged 70–74, living in rural areas or areas with high suicide mortality rates, the view on suicide as a matter of self-choice, and a pessimistic view toward life.
Conclusion: These findings suggest that inappropriate views on suicide adversely affect coping strategies and mental health. Suicide prevention programs aimed at improving mental health literacy in a community should take into consideration the characteristics of elderly male residents.
SUICIDE IS SURROUNDED by feelings of shame, fear, guilt, and uneasiness.1 These feelings and the related defense mechanisms give rise to various incorrect or inappropriate views on suicide, such as suicide being inevitable, permissible, and unpreventable. These inappropriate views on suicide possibly promote maladaptive coping strategies such as emotional inexpressiveness, reluctance to seek help, or alcohol abuse.2 Hesitation in seeking help from anyone when distressed may compromise an individual's mental health.3 Having correct and adequate knowledge and beliefs on suicide is part of mental health literacy, which enables individuals to gain access to, understand, and use information in ways that promote and maintain good mental health.4 It should also be noted that inappropriate views on suicide such as ‘suicidal behaviors are due to a weak mind’ often hurt those who are bereft over the loss of loved ones to suicide. Thus, taboos regarding suicide and emotions evoked by it in individuals likely hinder the implementation of suicide prevention programs.1
It is therefore important to examine how views on suicide are distributed in the general population to establish strategies for suicide prevention in a community.3 However, only a few studies have focused on this, and they are mainly from Japan, where the suicide mortality rate has been very high since 1998.5 Nishi et al. reported that views on suicide as being permissible and unpreventable despite community effort correlated with both gender and the standardized mortality ratio (SMR) for suicide in the area where subjects lived.6 Kaneko and Motohashi found that the view on suicide as being inevitable was frequent among men and those with a low educational level.7 They discussed the gender difference in views on suicide in relation to poor mental health literacy in men. Yamaji et al. further reported that some views on suicide were related to age, educational background, and occupational status.8 Results of a nationwide random sample study in Japan also suggested gender and age group differences in views on suicide, although its sample size was relatively small and statistical analysis was insufficient.9 Another preliminary study suggested that unmarried workers were more sympathetic to suicidal individuals than married workers.10
In this study, a population survey in Oita Prefecture, in western Japan, was conducted to determine possible risk factors for suicide, including inappropriate views on suicide. Middle-aged and elderly residents were chosen as the study population because suicide mortality rates are higher among them than among the younger population in Japan. The objectives of this study were to investigate the relationship of inappropriate views on suicide with demographic variables (gender, age, marital status, and area of residence), and given these demographic variables, to examine the association of these views with feeling shame in seeking help when distressed.
- Top of page
Responses to the itemized views on suicide are shown in Figure 1. The percentages of questions without an answer varied from 29 to 37% across the seven items, and among all the respondents, 24% provided no answer to all seven items. The prevalence of no responses to all seven items was high among the elderly group and widowed individuals, and low in Area Q. Items A–D exhibited moderately positive correlations with one another (Spearman's rank correlation coefficients 0.34–0.54, P < 0.001), whereas other correlations were weak (coefficients, −0.24–0.09), although most of the correlations were statistically significant because of the large number of respondents.
Figure 1. Simple distribution for views on suicide. Note that in the original questionnaire Pessimism was referred to as Optimism and Not so hard was referred to as Particularly hard time.
Download figure to PowerPoint
The prevalence of inappropriate views on suicide was recalculated by gender and age group (Table 2). For all the items, inappropriate views were more prevalent in men than in women. The prevalence for items A and E were high in the middle-aged group, whereas those for items C and G were high in the elderly group. Marital status or area of residence did not correlate with the prevalence of any views on suicide.
Table 2. Percentages of inappropriate views on suicide in relation to gender and age
|Variable||Category||A) Self-choice||B) Inevitable||C) Unpreventable||D) Permissible||E) Not so hard||F) Pessimism||G) Weak mind|
|(Fisher's exact test)||***||***||***||***||*||**||***|
The results of multiple logistic analyses of inappropriate views on suicide are shown in Table 3, in which respondents who did not answer questions on their views on suicide were categorized into those who did not manifest inappropriate views on suicide. Male participants exhibited a significantly high OR for all the inappropriate views on suicide. The elderly group exhibited significantly high OR for items C and G, whereas the middle-aged group exhibited significantly high OR for items A, D, and E. These results confirm the results of cross-tabulation. The OR for item F was highest among those aged 50–64. Those who had never married exhibited a significantly low OR for item E. Area Q exhibited the lowest OR for most of the inappropriate views on suicide, and the other three areas exhibited significantly high OR for item B. However, when respondents who did not answer questions on their views on suicide were excluded from analysis, the results differed. The elderly group showed significantly high OR for item B and G, and those aged 60–64 showed the highest OR for item F, whereas the OR for item A did not correlate with age. The divorced group showed high OR for item D. The three areas other than Area Q exhibited significantly high OR for items A–D and G.
Table 3. Multiple logistic analysis of inappropriate views on suicide
|Variable||Category||A) Self-choice||B) Inevitable||C) Unpreventable||D) Permissible||E) Not so hard||F) Pessimism||G) Weak mind|
|Those with inappropriate views [n (%)]||903 (20.1)||301 (6.7)||359 (8.0)||291 (6.5)||342 (7.6)||177 (3.9)||790 (17.6)|
|Independent variables||OR (95% CL)||OR (95% CL)||OR (95% CL)||OR (95% CL)||OR (95% CL)||OR (95% CL)||OR (95% CL)|
|Gender||Men||1.75 (1.50–2.03)||1.56 (1.21–1.98)||1.60 (1.28–2.00)||1.57 (1.23–2.01)||1.35 (1.08–1.69)||1.61 (1.18–2.20)||1.39 (1.19–1.63)|
|Age (years)||40–44||1.55 (1.11–2.14)||1.23 (0.73–2.05)||1.00||1.16 (1.23–2.01)||1.80 (1.05–3.09)||1.99 (0.97–4.07)||1.00|
|45–49||1.93 (1.44–2.61)||1.11 (0.68–1.82)||1.18 (0.64–2.18)||1.20 (0.70–2.06)||2.15 (1.31–3.52)||1.58 (0.76–3.25)||1.20 (0.78–1.83)|
|50–54||1.68 (1.28–2.22)||1.04 (0.66–1.64)||1.51 (0.87–2.63)||1.39 (0.86–2.25)||2.31 (1.47–3.64)||2.33 (1.27–4.31)||1.04 (0.69–1.57)|
|55–59||1.77 (1.38–2.28)||1.32 (0.90–1.95)||1.91 (1.14–3.20)||1.72 (1.12–2.63)||2.58 (1.70–3.90)||2.32 (1.32–4.09)||1.67 (1.16–2.40)|
|60–64||1.50 (1.17–1.93)||1.40 (0.96–2.06)||1.72 (1.02–2.88)||1.62 (1.06–2.49)||1.90 (1.24–1.92)||2.47 (1.42–4.30)||1.69 (1.18–2.43)|
|65–69||1.10 (0.83–1.44)||1.03 (0.68–1.57)||1.52 (0.89–2.59)||1.00||1.00||2.02 (1.13–3.61)||2.13 (1.48–3.06)|
|70–74||1.00||1.00||1.75 (1.04–2.93).58)||1.30 (0.84–2.01)||1.16 (0.74–1.83)||1.00||2.18 (1.53–3.12)|
|Never||1.04 (0.75–1.43)||0.87 (0.51–1.48)||0.88 (0.52–1.49)||0.97 (0.57–1.66)||0.50 (0.27–0.93)||1.39 (0.78–2.49)||0.94 (0.64–1.37)|
|Widowed||1.07 (0.78–1.47)||0.69 (0.39–1.22)||1.16 (0.76–1.76)||0.91 (0.54–1.54)||1.30 (0.83–2.05)||1.54 (0.86–2.78)||1.01 (0.75–1.37)|
|Divorced||1.27 (0.92–1.75)||0.73 (0.40–1.33)||0.98 (0.59–1.64)||1.00 (0.58–1.73)||1.28 (0.81–2.01)||0.65 (0.28–1.49)||1.15 (0.80–1.65)|
|Area||P||0.91 (0.74–1.13)||1.44 (1.01–2.05)||1.10 (0.80–1.51)||1.18 (0.83–1.68)||1.02 (0.74–1.39)||1.18 (0.77–1.81)||1.18 (0.94–1.47)|
|R||1.22 (0.97–1.53)||1.51 (1.03–2.01)||1.37 (0.98–1.92)||1.24 (0.84–1.81)||1.05 (0.75–1.48)||1.02 (0.63–1.65)||1.07 (0.84–1.37)|
|S||1.11 (0.90–1.36)||1.44 (1.01–2.04)||1.17 (0.86–1.60)||1.30 (0.92–1.83)||1.00 (0.73–1.37)||1.07 (0.69–1.65)||1.09 (0.87–1.35)|
Regarding feeling shame in seeking help when distressed, 6.4% answered ‘Yes’, 12.3% ‘Fairly yes’, 15.5% ‘Fairly no’, 42.1% ‘No’, 6.9% ‘I don't know’, and 16.8% gave no answer. According to the results of cross-tabulation, the prevalence of feeling shame in seeking help did not correlate with gender, age, marital status, or living area (Table 4). However, the prevalence positively correlated with the tendency to agree with items A–D (χ2MH = 53.7, 62.3, 58.1, and 64.8; P < 0.001 for all) and also with the tendency to disagree with item F (χ2MH = 34.9, P < 0.001).
Table 4. Multiple logistic analysis of feeling of shame in seeking help
|Variable||Category||%†||Model 1¶||Model 2¶|
|Gender‡||Men||7.1||1.19 (0.93–1.52)||0.95 (0.74–1.22)|
|Fisher's exact test||NS|| || |
|Age (years)‡||40–44||5.8||1.11 (0.60–2.06)||1.13 (0.60–2.11)|
|50–54||5.9||1.15 (0.66–2.00)||1.13 (0.65–1.97)|
|55–59||6.5||1.29 (0.78–2.16)||1.22 (0.73–2.04)|
|60–64||6.8||1.39 (0.84–2.30)||1.35 (0.81–2.24)|
|65–69||5.9||1.21 (0.71–2.05)||1.24 (0.72–2.11)|
|70–74||7.7||1.63 (0.99–2.68)||1.71 (1.03–2.83)|
|χ2MH||2.5NS|| || |
|Never||7.5||1.41 (0.85–2.38)||1.28 (0.76–2.16)|
|Widowed||7.7||0.89 (0.55–1.45)||0.94 (0.58–1.53)|
|Divorced||7.3||1.59 (0.97–2.56)||1.59 (0.97–2.60)|
|χ2||4.1NS|| || |
|Area‡||P||6.3||1.43 (0.99–2.81)||1.30 (0.90–1.89)|
|R||7.7||1.79 (1.22–2.63)||1.56 (1.06–2.31)|
|S||7.0||1.59 (1.10–2.27)||1.42 (0.99–2.03)|
|χ2||7.2NS|| || |
|Inappropriate views on suicide§|
|A) Self-choice||–||–||–||1.51 (1.10–2.07)|
|B) Inevitable||–||–||–||1.46 (0.94–2.28)|
|C) Unpreventable||–||–||–||1.40 (0.93–2.11)|
|D) Permissible||–||–||–||1.53 (0.97–2.42)|
|E) Difficult time||–||–||–||0.91 (0.57–1.45)|
|F) Pessimism||–||–||–||2.09 (1.32–3.28)|
|G) Weak mind||–||–||–||1.04 (0.76–1.41)|
The correlates of feeling shame were further examined by multiple logistic analyses (Table 4), in which the respondents who answered ‘I don't know’ or gave no response for each view on suicide or feeling shame were categorized into those who did not manifest inappropriate views on suicide or those who did not feel shame, respectively. The results obtained using model 1 showed that living in the high-SMR area correlated with feeling shame. In model 2, in which views on suicide were taken into account, feeling shame did not correlate with gender or marital status, but it was associated with being elderly (i.e. age 70–74), living in area R, viewing suicide as a matter of self-choice (item A), and having pessimism toward life (item F). When respondents who did not answer questions on their views on suicide were compared to those who manifested inappropriate or adequate views on suicide, no significant difference in OR was found. However, if respondents who did not answer questions on their views on suicide or feeling shame in seeking help were excluded from the analysis, this slightly changed the results obtained using model 2. Namely, the inappropriate view on suicide for item C (unpreventable) was associated with the feeling of shame, in addition to the correlates shown in Table 4.
- Top of page
Compared with the results of a previous nationwide survey in Japan,9 the views on suicide as being inevitable and unpreventable appear to be more prevalent in the present study. The respondents with no answer to these items also appear to be more frequent in this study than in the previous survey.9 The reason for this difference is unclear, although this may reflect something characteristic in our study area. The views on suicide as being inevitable, permissible, and unpreventable correlated with one another, but were independent of pessimism toward life and the idea attributing suicide to individuals' characteristics such as having a weak mind.
All of the inappropriate views on suicide were associated with gender (i.e. men), which is consistent with previous surveys.7,9,12 This is consistent with a report from the USA that indicated that male students were more likely than female students to think that anyone should be allowed to commit suicide.13 Previous researchers have discussed these views in connection with traditional gender roles.2,6 Although there have been few empirical investigations of gender difference as risk factors for suicide,14–16 these findings provide a clue to understanding that suicide is more prevalent in men than in women in most countries, including Japan.5,16
Various correlations between age and views on suicide were observed in the present study, which are inconsistent with the report by Kaneko and Motohashi that the frequency for acceptance of suicide is not associated with age.7 Among our respondents, the views of suicide as being unpreventable, permissible, and a matter of self-choice were frequent among the middle aged. These findings are consistent with a report by Sakamoto et al. that indicated that those aged 40–59 were more sympathetic to suicidal individuals than those aged 60–79.17 It can be hypothesized that these views of the middle aged are associated with their high mortality rate for suicide in Japan.5 In addition, views of suicide as being unpreventable and a choice by a weak mind were prevalent among the elderly (i.e. aged 70–74). This is consistent with the previous nationwide survey.9 There are at least two reasons for this association of age and these views on suicide – specifically, a developmental change and a cohort effect.17 It is also probable that there are regional differences in the relationship between age and views on suicide. However, in the present study, nonrespondents with regard to views on suicide were prevalent among the elderly group, and excluding them from the analyses affected the results. This means that we should be careful in comparing the above results with those of previous reports.
Frequently, those who had never been married believed that those who killed themselves did not experience a particularly hard time. This appears to disagree with an early report that unmarried workers were more sympathetic to suicidal individuals.10 Because being unmarried is a risk factor for suicide in Japan,18 the relationship between marital status and suicidal thoughts or views on suicide should be examined in detail. It seems important to distinguish never married, divorced, and widowed individuals in future studies.
As for regional differences in views on suicide, the results of previous studies are inconsistent.3,6,18 In the present study, the view on suicide as being inevitable in some cases was significantly associated with living in the high-SMR area. This supports the hypothesis that the prevalent attitude toward suicide in an area affects the SMR in the area. It is also interesting that the nonresponse rates for key questions in this study were low in the relatively urbanized area (area Q), showing that the respondents in this area are less likely consider it taboo to talk about suicide. This attitude will be helpful in obtaining correct knowledge and appropriate views on suicide. However, regional variations in views on suicide should not be exaggerated and sensationalized, as these differences are no larger than the gender or age group differences.
Feeling shame in seeking help when distressed possibly isolates individuals, which, in turn, may increase the risk of suicide.2,4 The prevalence of this feeling shame in the present study appears higher than that reported in the nationwide survey in Japan.9 There was no gender difference in feeling shame, although a previous study showed gender difference in help-seeking behaviors for depression among the elderly population in a rural area in Japan.19 In contrast, the OR for feeling shame were significantly higher among those aged 70–74 and those living in the rural or high-SMR areas, suggesting a need for suicide prevention programs for these individuals (e.g. public awareness campaigns about distress, depression, suicide, and the importance of seeking help).20,21 Furthermore, feeling shame was independently associated with the view of suicide as being a matter of self-choice and pessimism toward life. It should be noted that the former view on suicide showed a weak internal correlation with items B–D. Namely, the views on suicide as being inevitable, permissible, and unpreventable may be associated with hesitation in seeking help. This finding indicates how inappropriate views on suicide adversely affect individuals' mental health, as pointed out by previous researchers.2,3,6,7,22 These inappropriate views on suicide may promote maladaptive coping profiles such as emotional inexpressiveness, reluctance to seek help, and avoidance of making changes. As far as the present study is concerned, no relationship was found between feeling shame in seeking help and not providing answers to the questions on views on suicide.
There were several limitations to the present study. First, the causal relationship among views on suicide, feeling shame in seeking help, and demographic variables cannot be determined by this cross-sectional study. Second, the difference in the views on suicide between respondents who completed the questionnaire and those who did not is unclear, and the background characteristics of respondents who did not answer the key questions in this study remain unclear. Future research should gather more complete background information. Third, the occupational status and educational background of the respondents were not a part of this study, because local volunteers did not agree with including these items in the questionnaire. This may have affected our results, as these variables have correlated with views on suicide in previous studies.7,8
In conclusion, it was shown that inappropriate views on suicide were associated with gender (i.e. men). Some of these views also correlated with being middle aged, never having been married, and living in a rural or high-SMR area. These inappropriate views also correlated with feeling shame in seeking help when distressed, indicating how inappropriate views on suicide can adversely affect individuals' mental health. Suicide prevention programs aimed at improving mental health literacy should take the characteristics of elderly male residents into consideration.