Association of chronic disease prevalence and quality of life with suicide-related ideation and suicide attempt among Korean adults.

AIMS: The aim of this study is to find the association of chronic disease prevalence (CDP) with suicide-related ideation (SI) and suicide attempt (SA) and to determine the combined effect of CDP and quality of life (QoL) with SI or SA.
DESIGN: This was a cross-sectional study.
MATERIALS AND METHODS: The data were collected from the nationally representative Korea National Health and Nutrition Examination Survey IV and V (2007-2012). For the analysis, a total of 35,075 adult participants were selected as the final sample, which included 5773 participants with SI and 331 with SA. STATISTICAL ANALYSIS: Multiple logistic regression models were used to examine the odds ratio after adjusting for age, sex, marital status, education, occupation, and household income. RESULTS AND
CONCLUSION: SI was positively associated with selected CDP, such as cardiovascular disease (CVD), stroke, ischemic heart disease (IHD), cancer, diabetes, renal failure, and depression, except hypertension. Subjects with CVD, IHD, renal failure, and depression were found likely to have increased odds for SA as compared to non-SA controls. Lower QoL strongly affected SI and SA. Furthermore, the likelihood of SI increased for depressed and cancer subjects who had low QoL in comparison to subjects with high QoL and without chronic disease. Similarly, statistically, significant interaction was observed between lower QoL and depression in relation to SA compared to non-SA controls. These data suggest that suicide-related behavior could be predicted by the prevalence of chronic disease and low QoL.

INTRODUCTION

Suicide-related behaviors, suicide-related ideation (SI) and suicide attempt (SA), are predictive of suicide and are associated with high levels of burden to an individual and society as a whole.[1] Furthermore, SA is up to 20 times more frequent than suicide.[2] The suicide rate in Korea is the highest among the member countries of the Organization for Economic Cooperation and Development,[3] which is about 33.3 per 100,000 people in 2014.[4] The lifetime prevalence of SI and SA was found to be 15.2% and 3.2%, respectively, which is higher than the rate of Japan, China, and Western countries.[56] As the suicide rate had more than tripled between 1990 and 2010,[4] it is a serious health problem in Korea.

Druss et al. reported a significant association of general medical illness with SI and SA.[7] In schizophrenic patients, major medical conditions have been associated with SA.[8] Many researchers have reported depression as an important predictor for SI[910] and SA.[1112] Bernal et al. showed that the relative risk for depression was 2.9 and 3.91 for SI and SA, respectively.[11] Jeon et al., based on nationwide sample of Korea, observed that mood disorders, obsessive–compulsive disorder, and posttraumatic stress disorder were significant risks of SA.[6] Although mental illness is responsible for a large proportion of suicide,[1314] number of chronic diseases were shown to be associated with elevated risk of suicide in elderly.[15]

Several physical diseases are independently associated with an increased risk of suicide,[16] and the presence of chronic physical conditions is a risk factor for suicide-related behavior.[17] Chronic diseases are the independent risk factors for SI.[18] In addition, chronic diseases were found associated with elevated risk of suicide.[15] Artero et al. reported that ischemic heart disease (IHD) is a risk factor of suicide.[19] The World Mental Health Survey of 14 countries showed that stroke and cancer are associated with planned SA.[17] Several studies have reported that risk of suicide is higher in cancer patients than that in general population, and depression, isolation, and fear of burden increase the vulnerability in cancer patients.[202122] Chung et al. showed that diabetes mellitus was associated with increase in SI and SA.[23] Diabetes and heart disease were strongly associated with SI as reported by Chan et al.[24] A study conducted among Korean adult population reported that renal failure was significantly associated with a higher risk of SI and SA.[25]

Kim et al. reported that the decrease in health-related quality of life (QoL) strongly affected SI; particularly, older adults with lower QoL (EQ-5D <0.7) had higher odds for SI and SA than those at higher QoL.[26] Chronic disease may result in lower perceived health status and decrease in QoL.[27]

Thus, the present study aimed to find the association of chronic disease prevalence (CDP) with SI and SA and to evaluate the combined effect of selected CDP and QoL on SI and SA.

SUBJECTS AND METHODS

Study population

The present study used the nationally representative cross-sectional survey data from the Fourth and Fifth Korea National Health and Nutrition Examination Survey (KNHANES IV and V, 2007–2012) conducted by the Korean Centers for Disease Control and Prevention (KCDC) and the Ministry of Health and Welfare, Korea. KNHANES study is composed of four component surveys: health interview, health behavior survey, nutrition survey, and health examination. For the selection of a representative sample, a multi-stage clustered probability design based on administrative district was adopted. Sampling frame for KNHANES IV and KNHANES V was noninstitutionalized civilians aged 1 year or older in the Korean population based on the 2005 population census and 2009 registration population, respectively. A total of 13,800 sample households from 600 sample statistical units and 11,400 sample households from 576 sample statistical units were primarily selected for the KNHANES IV and V, respectively. Among the initial study samples for the KNHANES IV and V (N = 50,405), we excluded individuals under 19 years of age (N = 12,400) and who did not give information to assess SI (N = 2,930). A total of 35,075 participants including 5773 subjects who had SI and 331 with SA were selected as the final sample to be analyzed. Written statement of informed consent was signed by subjects or their legal guardians, and the subjects were informed about the study protocol and importance. The study protocol was approved by the KCDC review board (nos. 2007-02CON-04-P, 2008-04EXP-01-C, 2009-01CON-03-2C, 2010-02CON-21-C, 2011-02CON-06-C, and 2012-01EXP-01-2C).

Assessment of suicide-related behaviors

Information of SI and SA was collected by asking specific questions for the evaluation of suicide-related behaviors. SI was evaluated by asking following question to the participants: “In the last 12 months, did you ever think about committing suicide?” Subjects who denied were assigned to non-SI controls and those who concurred with the question were determined to have SI. For evaluating SA, participants who had SI were further asked question as: “In the last 12 months, did you ever attempt suicide?” The subjects who answered “yes” were grouped under SA. All the subjects without SA were taken as non-SA controls. Terminologies related to suicide were used as described in Silverman et al.[28]

General factors

Gender, age, marital status, education level, occupation, and household income were included as general characteristics in the study. Marital status was categorized as married or single (i.e., never married or divorced, separated, and bereaved). Educational attainment was classified by the years of education as elementary school, middle school, high school, and college. Categories of occupational status were adopted from the Korean Standard Classification of Occupations (KSCO, 2007) of Korea National Statistical Office. Household income (calculated by dividing monthly income by the square root of the number of household members) was categorized into quartile groups according to the equivalent income of household.

Chronic disease

For the presence of chronic disease, subjects were asked whether they had any of the following chronic diseases: cardiovascular disease (CVD) (myocardial infarction and angina), stroke, IHD, cancer, renal failure, hypertension, diabetes, and depression. These chronic diseases were selected with respect to the prevalence and their association with morbidity in the Korean population.[29]

Quality of life

QoL of the subjects was evaluated by EQ-5D index.[30] Classification of the category was done by EQ-5D index score of 1 versus lesser than 1, which was the median of the index score (mean: 0.804, standard deviation: 0.153). Participants were asked to rate their QoL in terms of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression on a Korean version scale as: 1 – no problems; 2 – some or moderate problems; and 3 – extreme problems. EQ-5D has been established as a validated tool to measure QoL.[313233]

Statistical analysis

Multiple logistic regression models were employed to assess the risk of each candidate variables using odd ratios (ORs) and 95% confidence intervals (CI) after adjusting for age, gender, marital status, education level, occupation, and household income. In addition, the combined effect of chronic disease and QoL on the SI and SA was evaluated. For this analysis, EQ-5D index was measured with each chronic disease and categorized into high and low QoL score based on the median score of EQ-5D index. The likelihood ratio test was used to assess potential interaction of each study variables by comparing the models with and without the cross-product terms of these variables. All statistical tests were based on two-sided probability. All statistical analyses were performed using SAS statistical software version 9.4 (SAS Institute Inc., Chicago, Illinois, USA).

RESULTS

The general characteristics of subjects in relation to chronic disease are described in Table 1. Female was found to have risk of chronic disease (OR = 1.06). Increase in age presented a positive association with chronic disease (P < 0.001), whereby above 70 years subjects were at highest risk (OR = 41.7). Education showed a significant negative association with chronic disease (P < 0.001). Categories of occupation were seen to be associated with chronic disease in reference to white collar jobs, and joblessness was at highest risk of being diseased. Household income was observed to have significant positive association with chronic disease (P < 0.001).

Table 1

Distribution of general characteristics of subjects with chronic disease prevalence

Table 2 presents the association of CDP and QoL with the risk of SI and SA. In this representative study population, 15.1% and 0.85% of the Korean adults were reported to have SI and SA during the last 12 months of interview, respectively (data not shown). Subjects with the selected chronic diseases including CVD, stroke, IHD, cancer, diabetes mellitus, renal failure, and depression were likely to have SI, except hypertension. However, subjects who were diagnosed with chronic diseases including CVD, IHD, renal failure, and depression were at greater risk of SA. Particularly, subjects with depression were at greatest odds of SI (OR = 3.60, 95% CI: 3.19–4.06) and SA (OR = 8.15, 95% CI: 6.17–10.8). Among the subjects who endorsed SI, 35.6% were in high QoL and 64.4% were in low QoL. Similarly, for SA, 26.7% had high QoL and 73.3% had low QoL. Low QoL presented risk of SI (OR = 3.31, 95% CI: 3.10–3.54) and SA (OR = 4.18, 95% CI: 3.19–5.48) with reference to high QoL.

Table 2

Association of chronic disease prevalence and quality of life with suicide-related ideation and suicide attempt

In addition, advanced analyses of the combined effect of chronic disease and QoL on SI are presented in Table 3.

Table 3

Combined effect of quality of life and chronic disease prevalence on suicide-related ideation

Lifetime cancer or depression combined with low QoL was found to have significant interactive effect on SI (Pfor interaction cancer: 0.040; depression: 0.006). Compared to the subjects without depression and high QoL, the subjects with depression and high QoL had an OR of 3.85 (95% CI: 3.11–4.77). Notably, subjects with depression and low QoL combined presented synergistic increase of risk on SI (OR = 8.70, 95% CI: 7.51–10.1) in comparison with those with high QoL without depression. Although subjects with cancer and high QoL had statistically nonsignificant reduced pattern (OR = 0.85, 95% CI: 0.62–1.17), subjects with cancer and low QoL showed significant combined effect on SI (OR = 4.06, 95% CI: 3.37–4.90) as compared to noncancer and high QoL group. On the other hand, irrespective of no interaction, subjects with other selected chronic disease and low QoL showed elevated ORs for SI compared with subjects without chronic disease and high QoL.

The combined effect of chronic disease and QoL on SA is presented in Table 4.

Table 4

Combined effect of quality of life and chronic disease prevalence on suicide attempt

Depression combined with low QoL has significant interactive effect on SA (Pfor interaction0.008). Subjects suffering from depression and having low QoL were more likely to SA (OR = 23.3, 95% CI: 15.9–34.2) in comparison with those without depression and high QoL. No interactive effect was seen for other chronic disease; however, subjects with the selected chronic disease and low QoL showed elevated ORs for SA as compared to subjects without chronic disease and high QoL.

DISCUSSION

The present nationally representative study examined the association of CDP and QoL on SI and SA. Age, gender, education, occupation, and household income were found associated with chronic disease. SI was found to be higher among subjects with CVD, stroke, IHD, cancer, diabetes, renal failure, and depression. Subjects with selected chronic disease were at increased risk of SA compared to the non-SA controls; they were CVD, IHD, renal failure, and depression. Moreover, we demonstrated the combined effect of chronic disease and QoL with SI and SA, the first study of its kind. Among selected chronic disease, depression and cancer were observed to have combined effect with QoL on SI and only depression on SA.

In the present study, elderly age was significantly associated with chronic disease consistent with other studies.[3435] In consistence with other studies, lower education attainment was found associated with chronic disease.[3637] Occupation was associated with the risk of chronic disease. Similar to the finding of other study, jobless subjects were found to be at higher risk of chronic disease.[36] Economic status plays an important role in the health of individuals.[38] Similarly, we found significantly positive association of household income with the risk of chronic disease, and upper household income showed threefold greater risk of chronic disease than low household income.

We found a significant positive association of CVD with SI and SA. As far as our knowledge, none of the research attempted to study CVD, SI, and SA; however, CVD was found associated with suicide,[39] and Jee et al. reported, in Korean population, the association of biological and behavioral CVD risk factors with suicide.[40] Our results showed that stroke was associated with SI whereas IHD had a significant higher OR for SI and SA. Similar to our result, Pohjasvaara et al. reported positive correlation of stroke with SI.[41]

In our finding, diabetes mellitus was associated with SI but not with SA. Similar to the results of Pompili et al., diabetic patients usually presented with SI but not with SA even though our sample had six times more SA subjects. In our study, we found significant association of cancer with SI but not with SA. In contrast to our finding, Druss et al. reported association of cancer with more than fourfold increased likelihood of SA.[7] This difference in result may be due to less number of SA in our sample compared to Druss et al. Although we could not find literature to explain the relationship, majority of studies have shown cancer as a known risk of suicide.[4243] Patients with renal failure were at increased risk of SI and SA which was similar to another study.[25] Patients with chronic renal disease have a higher prevalence of psychiatric disorder than the general population and they often experience higher levels of anxiety and depression.[4445] Thus, lack of coping strategies to reduce stress during the dialysis[4647] and the pain of calciphylaxis[48] may prompt patients toward SI and SA. In addition, renal failure with dialysis is known risk factor of suicide in earlier study.[49] Of the chronic diseases we studied, depression emerged as the most strongly associated with SI and SA, which is consistent with other studies.[5051] Interestingly, we found that depressed subjects had higher odds for SA than for SI.

As reported by Kim et al., lower QoL was correlated with SI and SA.[26] Similarly, we found significant association of QoL with SI and SA. Subjects with low QoL were at increased risk of SI and SA as compared to subjects with high QoL. Among selected chronic disease, cancer and depression were seen to have significant interactive effect of QoL on SI, and depression showed significantly interactive effect combined with QoL on SA. Cancer is a known disease of physical difficulties and psychosocial problems, thus having substantial influence on QoL.[52] In our results, subjects with high QoL and cancer did not have significantly higher association with SI than who had high QoL without cancer. However, in additional analysis, we found that the subjects with low QoL and cancer had significantly higher association with SI than those who had low QoL without cancer (results not presented in a table). In this result, we may assume that cancer does not affect the subject in high QoL to have SI, but state of low QoL and cancer combined increases the association than the subjects only in low QoL. Chronic diseases usually result in low QoL. The particular strong relationship between depressive condition and health-related QoL[53] could be the result of interactive relation between them. Future studies should be carried out to clarify the combined effect of QoL and depression on suicide-related behaviors.

This paper should be considered in light of some limitations. First, the KNHANES is a cross-sectional survey in which all information is based on retrospective nature of data collection and thus subject to recall bias; however, retrospectively, reported results have valuable place in research.[54] Second, small sample of suicide attempters (331) may have resulted in insufficient statistical power and precision; however, this study may provide relevant information regarding SI and SA for prevention program among Koreans. Third, this study was unable to consider some other relevant variables such as history of SA, number of times of SA, severity of chronic diseases, and inclusion of institutionalized subjects. Despite these limitations, this study also has several strengths. First, our study is proper weighted nationwide sampling, which represents the national estimations of Koreans. Second, this study extends the previous researches by observing novel interactive relationship between comorbidity and QoL on the prevalence of SI and SA. Third, because of the large sample size, we were able to include well-known confounding factors in the analysis such as age, gender, marital status, education, occupation, and household income. Further prospective studies are warranted to support our results.

CONCLUSION

Selected chronic diseases, except hypertension, were found associated with SI, and CVD, cancer, renal failure, and depression were associated with SA. Low QoL was positively associated with SI and SA. Depressed and cancer patients who had low QoL were found associated with elevated risk of SI whereas depression and low QoL increased the likelihood of SA. Hence, the current study supports the recommendation that presence of chronic disease and low QoL might be reliable predictor for suicide-related behaviors (SI and SA).

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.