Abstract
This study aimed to investigate the association between charcoal-burning suicide attempts and the risk of developing dementia. A nationwide, matched cohort, population-based study enrolled a total of 4103 patients with newly diagnosed charcoal-burning suicide attempts, between 2000 and 2010, which were selected from the National Health Insurance Research Database of Taiwan, along with 12,309 controls matched for sex and age. After adjusting for confounding factors, Fine and Gray’s competing risk analysis was used to compare the risk of developing dementia during the 10-year follow-up period. Of the enrolled patients (n=16,412), dementia developed in 303 (1.85%), including 2.56% in the study group (105 in 4103) and 1.61% (198 in 12,309) in the control group. The Fine and Gray’s survival analysis revealed that the patients with charcoal-burning suicide attempts were likely to develop dementia, with a crude HR of 5.170 (95% CI 4.022 to 6.644, p<0.001). After adjusting for age, sex, comorbidity, geographic area and urbanization level of residence, and monthly insured premium, the adjusted HR was 4.220 (95% CI 3.188 to 5.586, p<0.001). Suicide attempts were associated with an increased risk of degenerative dementia in this study. Patients with charcoal-burning suicide attempts had a fourfold risk of dementia than the control group.
Significance of this study
What is already known about this subject?
Previous studies have found that charcoal-burning survivors would suffer from delayed neurologic sequelae, involving diffuse demyelination in the brain accompanied by lethargy, behavioral changes, forgetfulness, memory loss, and parkinsonian features.
The association between charcoal-burning suicide attempts and the risk of dementia has not yet been studied.
What are the new findings?
Comparing with previous research on the association between carbon monoxide intoxication and the risk of dementia, this population-based study focused on the association between charcoal-burning suicidal attempts and the risk of dementia.
The charcoal-burning suicide attempters were more likely to develop dementia (HR 4.22, 95% CI 3.19 to 5.59, p<0.001), when adjusting for sex, age, monthly income, urbanization level, geographic region, and comorbidities.
Charcoal-burning suicide attempts were associated with an increased risk of degenerative dementia in this study.
How might these results change the focus of research or clinical practice?
If the association between charcoal-burning suicide attempts and the risk of dementia is causal, then the results would remind clinicians who care for charcoal-burning suicide survivors to provide careful monitoring of these patients’ cognitive function in clinical practice.
Introduction
Charcoal-burning suicides became an epidemic in Hong Kong,1 Taiwan,1–4 China,5–7 Korea,8 Japan,9 Singapore,10 and some Western countries such as the UK11 and the USA12 over the last two decades, which were precipitated by wide media reports13 or by internet browsing.14 Charcoal-burning survivors would suffer from delayed neurologic sequelae, with diffuse demyelination in the brain accompanied by lethargy, behavioral changes, forgetfulness, memory loss, and parkinsonian features, and additionally sometimes toxic or ischemic peripheral neuropathies,15 which could impair attempters’ cognitive and motor functions.
Between 2011 and 2012, 130,000 people, or 4.97%, of those aged 65 years and over in Taiwan had dementia,16 which is a heavy burden for the patients and their caregivers, community, or society.17–20 Injuries on the brain such as traumatic brain injury (TBI),21 stroke,22 or even attention deficit hyperactivity disorder and related brain injury23 24 would also contribute to the development of dementia. Studies on delayed neurologic or neuropsychiatric syndrome revealed that carbon monoxide (CO) intoxication from charcoal-burning resulted in the diffuse demyelination in the brain.15 Another study found that the fronto-insular-caudate areas represented the target degenerative network in the CO intoxication.25 Therefore, we hypothesize that CO intoxication in patients with suicide attempts would also be associated with dementia, and we conducted this study to clarify whether coal-burning suicide attempts would be associated with subsequent dementia.
Methods
Data sources
The National Health Insurance (NHI) program was launched in Taiwan in 1995, and as of June 2009 it has included contracts with 97% of medical providers with approximately 23 million beneficiaries, or more than 99% of the entire population.26 The National Health Insurance Research Database (NHIRD), which contains all claims data of the beneficiaries, uses the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes to record diagnoses.27 All the diagnoses of dementia in Taiwan are made by board-certified psychiatrists or neurologists. All patients with suicide attempts by charcoal-burning would be attended and treated by emergency physicians, chest medicine specialists, intensive care physicians, or other medical experts, and the diagnosis would then be confirmed from their medical history and arterial blood gas tests. The NHI Administration randomly reviews the records of ambulatory care visits and inpatient claims to verify the accuracy of the diagnoses.28 Several studies have demonstrated the accuracy and validity of several diagnoses in the NHIRD, including diabetes mellitus (DM),29 30 cancer,31–33 myocardial infarction,29 34 35 central nervous system diseases such as Tourette syndrome,36 stroke,29 37–40 outcomes,33 mortality,29 41 or comorbidity.33 41 In a wide spectrum of conditions, some studies also demonstrated concordance between Taiwan’s National Health Survey and the NHIRD on a variety of diagnoses,42 medication use,42 and health system utilizations.42 43 In this study, we used data from the Longitudinal Health Insurance Database (2000–2010), a subset of the NHIRD, to investigate the association between charcoal-burning suicide attempts and dementia over a 10-year period.
Study design
This study has a population-based, matched-cohort design. Patients with newly diagnosed charcoal-burning suicide attempts were selected from the hospitalization data set from January 1, 2000 to December 31, 2010, according to the diagnosis of the charcoal-burning suicide attempts (ICD-9-CM code: E952). Patients before 2000 were excluded. In addition, patients diagnosed with dementia before 2000 or before the first visit for charcoal-burning suicide attempts were also excluded. All patients aged <20 years were also excluded. All patients with other suicide attempts (ICD-9-CM codes: E950-E951, E953-E959) were also excluded. A total of 16,412 patients were enrolled, including 4103 subjects with charcoal-burning suicide attempts and 12,309 randomly selected sex-matched, age-matched, and index year-matched controls without charcoal-burning suicide attempts in a ratio of 1:3, with a statistical power of up to 0.875, similar to the design of our previous studies (online supplementary figure 1).23 44 Age was grouped into 20–29, 30–39, 40–49, 50–59, 60–69, and ≥70 years.
Supplementary file 1
Covariates
The covariates included geographic area of residence (north, center, south, west, and east of Taiwan), urbanization level of residence (levels 1–4) and monthly income (in New Taiwan Dollars (NT$): <18,000, 18,000–34,999, ≥35,000).8 Other covariates such as complications from acute respiratory failure, rhabdomyolysis, acidosis, encephalopathy, procedures of hyperbaric oxygenation (HBO) therapy, and mechanical ventilations were also recorded.
Comorbidity
In this study, the comorbidity was used with reference to previous cohort studies on the risk factor of dementia.44 45 In these two previous studies, the comorbidity was according to the RxDx-Dementia Risk Index, a useful tool to identify the risk of dementia, which has a c-statistics value of 0.806 (95% CI 0.799 to 0.812).45 Therefore, the baseline comorbidities (in ICD-9-CM codes) included DM (250), hypertension (401.1, 401.9, 402.10, 402.90, 404.10, 404.90, 405.1, 405.9), hyperlipidemia (272.x), coronary artery disease (411, 413, 414), myocardial infarction (410, 412), peripheral vascular disease (444-449), heart failure (428), obesity (278), chronic kidney disease (580, 581–589, 753, 403, 404, 250.4, 274.1, 440.1, 442.1, 447.3, 572.4, 642.1, 646.2), cerebrovascular disease (433, 434, 436), TBI (800–804, 850–854, 905.0, 950.1, 950.3, 907.0, 959.01, 959.9, 310.2, V15.52), primary cancer (140–239, with the exceptions of 190–199 and 210–229), metastatic tumor (190-199), chronic pulmonary diseases (490-496), rheumatologic disease (725-729), peptic ulcer disease (533), hemiplegia or paraplegia (342, 344), liver disease (571), AIDS (042), epilepsy (345), tuberculosis (010–018), Parkinson’s disease (332), cardiac atherosclerotic cardiovascular disease (440, 441, 442, 443), glaucoma (365), cystic fibrosis (277.0x), transplantation (V42), thyroid disorder (240-246), gout (274), Crohn’s disease and ulcerative colitis (555, 556), pain (338), pain and inflammations (710-719), depression (296.2, 296.3, 296.82, 300.4, 311), bipolar disorders (296.4x, 296.5x, 296.6x, 296.89), anxiety disorders (300.00, 300.01, 300.02), and psychotic illness (295.xx, 298.x).
Main outcome measures
All of the study subjects were followed from the index date until the onset of dementia (ICD-9-CM codes: 290.0, 290.10, 290.11, 290.12, 290.13, 290.20, 290.21, 290.3, 290.41, 290.42, 290.43, 290.8, 290.9, and 331.0), withdrawal from the NHI program, or at the end of year 2010. Dementia was divided into three subgroups: Alzheimer-type dementia (331.0), vascular dementia (ICD-9-CM codes: 290.4x), and other degenerative dementia (ICD-9-CM codes: 290.x, with the exception of 290.4x). As mentioned above, all the diagnoses of dementia in Taiwan are made by board-certified psychiatrists or neurologists, and for the types of dementia the NHI Administration also randomly reviews the records of ambulatory care visits and inpatient claims to verify the accuracy of the diagnoses.28
Statistical analysis
All statistical analyses were performed using SPSS for Windows, V.22.0. χ2 and t-tests were used to evaluate the distribution of categorical and continuous variables, respectively, with the Fisher’s exact examination. Fine and Gray’s competing risk analysis was used to determine the risk of dementia, since death can act as a competing risk factor,44 46 47 and the results were presented as HR with 95% CI. Differences in the risk of dementia between the study and control groups were estimated using the Kaplan-Meier method with the log-rank test. A two-tailed p value <0.05 was considered to indicate statistical significance.
Results
Sample characteristics
Table 1 shows the proportion of outcomes (with or without suicide) among exposure groups of sex, age, comorbidities, urbanization, and area of residence, and the income of the charcoal-burning cohort and controls. For most of the exposure groups with comorbidities, the proportions of outcome with charcoal-burning suicides were lower than the proportions of outcome without suicide, except DM. For the exposure groups with HBO therapy, the proportions of outcome with charcoal-burning suicide were higher than the proportions of outcome without suicide. With regard to the exposure groups in different urbanization levels and location of residence, the patients in urbanization level 2, middle, southern and eastern Taiwan showed higher proportions of outcome with charcoal-burning suicide. The exposure group with monthly insured premium lower than NT$18,000 revealed higher proportion of outcome with suicide.
Kaplan-Meier model for the cumulative risk of dementia
At the end of follow-up, 303 patients out of a total of 16,412 enrolled subjects (1.85%) had developed dementia, including 2.6% in the study group (105 in 4103) and 1.6% (198 in 12,309) in the control group, and the Kaplan-Meier analysis for the cumulative incidence of dementia in the study and control groups is shown in figure 1 (log-rank test <0.001). In addition, the percentage of use of HBO therapy in the charcoal-burning cohort was 19.0% (779 in 4103). Table 2 shows the results of Fine and Gray’s competing risk analysis of the factors associated with the risk of developing dementia. The crude HR was 5.17 (95% CI 4.00 to 6.64, p<0.001). After adjusting for age, sex, comorbidities, geographic area of residence, urbanization level of residence, and monthly income, the adjusted HR was 4.22 (95% CI 3.19 to 5.87, p<0.001). In addition, male subjects were at a 1.57-fold risk than female subjects (p<0.001). With regard to age, those who were aged 60–69 and ≥70 years old had 1.66-fold and 2.88-fold risk in comparison with the reference group, that is, those aged 20–29. In addition, a 1-year increase in age increases the additional risk of dementia by 1.03%. Similarly, the charcoal-burning cohort with complications such as acute respiratory failure and encephalopathy was associated with higher risk of developing dementia than those without these complications. The adjusted HR was 2.51 (95% CI 1.20 to 3.68, p<0.001) and 2.99 (95% CI 1.30 to 7.76, p<0.001) in the charcoal-burning cohort who had received HBO therapy for 1–3 times and >3 times, respectively, in contrast to the control group.
Sensitivity analysis for the risk of dementia
We have conducted two types of sensitivity analysis to evaluate the risk of dementia. First, we excluded patients diagnosed with dementia within the first 2 years after the charcoal-burning suicide attempts, and the adjusted HR was 3.22 (95% CI 1.96 to 5.29, p<0.001) for patients with charcoal-burning suicide attempts (table 3). Second, we analyzed the risk of patients who attempted suicide by other methods, and the results showed that suicide attempts by solid or liquid ingestions, gases in domestic use, hanging, drowning, cutting or piercing implements, and jumping from building were not associated with the risk of dementia. However, the subgroup of other methods of suicides (ICD-9-CM code E958) has elevated risk of dementia (adjusted HR=3.30, 95% CI 2.44 to 6.04, p<0.001) (table 4).
Types of dementia in patients after charcoal-burning suicide attempts
With regard to the types of dementia, the study group was associated with a 4.28-fold (p<0.001) risk in developing degenerative dementia than the control group of non-attempters, and most were of the degenerative types, but not significantly associated with Alzheimer dementia (AD) or vascular dementia (VaD) (table 5).
Discussion
Association between charcoal-burning suicide attempts and the risk of dementia
In this study, we found that in the 10-year follow-up of the subject group and the control group, the charcoal suicide attempts were associated with a higher risk of developing dementia. The log-rank of the Fine and Gray’s competing risks regression model was significant (p<0.001). The crude HR of the subject group was 5.17 (95% CI 4.02 to 6.64, p<0.001), and the adjusted HR was 4.22 (95% CI 3.19 to 5.87, p<0.001). We have also conducted two sensitivity analyses to evaluate the influences from protopathic bias. First, even though the patients with the diagnosis of dementia within the first 2 years were excluded, the charcoal-burning suicide attempts were still associated with increased risk of dementia. Second, suicide attempts by solid or liquid ingestions, gases in domestic use, hanging, drowning, cutting or piercing implements, and jumping from building were not associated with the risk of dementia. H owever, the subgroup of other methods of suicides (ICD-9-CM code E958) was associated with a 3.3-fold risk of dementia (p < 0.001), and we speculate that some patients with charcoal-burning suicide attempts might have been classified into this category. Comparing with previous research about the association between CO intoxication and the risk of dementia,48 49 this study focused on charcoal-burning suicide attempts and the risk of dementia. To our best knowledge, this is the first nationwide, population-based cohort study that focused on the association between charcoal-burning suicide attempts and the risk of dementia.
Types of dementia in this study
In this sample, 303 patients out of a total of 16,412 enrolled subjects (1.9%) had developed dementia, including 2.6% in the study group (105 in 4103) and 1.6% (198 in 12,309) in the control group, and the percentage was close to the prevalence of 2%–5% for the population aged ≥65 in community studies.50 51 In Taiwan, several community studies revealed that AD was the most common type of dementia (40%–60% of all dementias), followed by VaD (20%–30% of all dementias), and mixed or other dementias (7%–15%).50 52 53 This finding reflects the fact that patients with dementia tended to be younger in the study subject group: 95 of the 105 (90.47%) patients who developed degenerative dementia were <60 years old and related to the sequelae of charcoal-burning. Nevertheless, the charcoal-burning cohort aged >60 were associated with an increased risk of dementia: adjusted HR 1.66 (95% CI 1.01 to 2.68, p=0.045) in the charcoal-burning cohort aged 60–69, and 2.88 (95% CI 1.73 to 4.70, p<0.001) in comparison with the control group.
Possible mechanisms for the increased risk of dementia in charcoal-burning suicide attempters
Studies on delayed neurologic or neuropsychiatric syndrome revealed that CO intoxication from charcoal-burning resulted in the diffuse demyelination in the brain.15 One study found that the fronto-insular-caudate areas represented the target degenerative network in CO intoxication.25 The decrease in the gray matter volume in the bilateral basal ganglia, left postcentral gyrus, and left hippocampus is also correlated with the decreased perceptual organization and processing speed function in these patients.54 55 Therefore, these wide varieties of brain damage could well play an important role in the development of dementia. In our study, the subjects with stroke, TBI, epilepsy, PD, depression, and bipolar disorder showed a higher risk in developing dementia, and these findings hint that the neurologic or psychiatric disorders were at a higher vulnerability to develop a CO-induced brain damage.
HBO therapy and dementia after charcoal-burning suicide attempts
In our study, the usage of HBO therapy in the charcoal-burning cohort was 19.0% (779 in 4103), which was compatible with another study in Taiwan in which hyperbaric oxygen therapy was only used in 18.8% of patients.14 Evidence of HBO therapy for CO intoxication from charcoal-burning is yet to be established.56–58 However, the charcoal-burning cohort who had received HBO therapy still had an increased risk of dementia in contrast to the control group, with an adjusted HR of 2.80 (95% CI 1.15 to 6.77, p=0.023), and with an adjusted HR of 2.51 (95% CI 1.20 to 3.68, p<0.001) and 2.99 (95% CI 1.60 to 7.76, p<0.001) in the charcoal-burning cohort who had received the HBO therapy for 1–3 times and >3 times, respectively, in contrast to the control group, which were lower than the overall adjusted HR of as high as 4.28. However, further studies were indicated for the association among charcoal-burning attempts, HBO therapy and the risk of dementia.
Psychiatric disorders and risk of dementia in charcoal-burning suicide attempters
In the charcoal-burning cohort, the comorbidities of depression, bipolar disorders, and anxiety disorders were higher than the control groups. This observation indicates that the prevention of charcoal-burning as a suicide method in some populations is, indeed, important. Furthermore, in the charcoal-burning cohort with stroke, TBI, epilepsy, PD, depression and bipolar disorder, the risk of developing dementia was higher than those without these comorbidities.These findings revealed that the prevention of the charcoal-burning suicide in these patients could be important. Since one study showed in its preliminary results that a charcoal-restriction program reduced the method-specific and overall suicides,59 a larger scale and longer follow-up study is needed to evaluate the results of charcoal restriction and their effects on the prevention of suicide.
Limitations
There are several limitations to this study. First, patients with dementia could be identified using the insurance claims data; however, data on severity, stage, and impact on their caregivers were not available. The types of dementia were also identified from the ICD codes in these claims data. Second, even though only newly diagnosed dementia would be included in the follow-up period, a protopathic bias, in which the initiation of an exposure occurs in response to an undiagnosed disease (outcome) under study,60 should also be considered since some of the subjects with charcoal-burning suicide attempts suffered cognitive decline before their suicide attempts. Third, in this study, we identified patients with charcoal-burning suicide attempts by ICD-9-CM code, E952. Even though there are some debates on using this code for charcoal-burning suicide,61 62 charcoal-burning suicide deaths increased drastically from less than 2% before 1998, to 20%–30% of all suicide deaths in Hong Kong and Taiwan within the following decade,63 64 and several authors therefore have used E952 code as charcoal-burning suicides in their studies.1 65 Hence, we assumed that using E952 code to identify charcoal-burning suicide attempts could be a rational choice in Taiwan, even though no previous study has examined the operating characteristics of this code for charcoal-burning suicide attempts.
Conclusions
The patients with charcoal-burning suicide attempts had a nearly fourfold risk of dementia than the control group. Therefore, further studies are needed to elucidate the underlying mechanisms. If the association reflects a causal effect, this finding would remind clinicians who oversee care for charcoal-burning suicide survivors to make careful evaluations and follow-up for cognitive and behavioral manifestations.
Acknowledgments
The authors thank Professors Chang-Huei Tsao, Yung-Fu Wu, Fu-Huang Lin, Chin-Bin Yeh, San-Yuan Huang, and Ru-band Lu for their help in the administrative work and academic guidance in supporting the completion of this study.
Footnotes
Contributors S-YC and N-ST conceived of the study, participated in its design and coordination, data interpretation, performed the statistical analysis, and drafted the manuscript. W-CC, C-HC, and H-WY participated in the design of the study and data interpretation. H-AC and Y-CK participated in the design of the study and data interpretation. C-KP, C-HS, and Y-CC participated in the statistical analysis and data interpretation. S-YC wrote the paper. All authors have read and approved the final manuscript as submitted.
Funding This work was supported by Tri-Service General Hospital Research Foundation grant numbers TSGH-C105-130, TSGH-C106-002, and TSGH-C107-004. Data for this study were based on the National Health Insurance Research Database provided by the National Health Insurance Administration of the Ministry of Health and Welfare, Taiwan, and managed by the National Health Research Institutes, Taiwan. The interpretation and conclusions contained in this article do not represent those of the National Health Insurance Administration, the Ministry of Health and Welfare, or the National Health Research Institutes.
Competing interests None declared.
Patient consent Not required.
Ethics approval This study was conducted in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki). The Institutional Review Board of the Tri-Service General Hospital approved this study and waived the need for individual consents since all the identification data were encrypted in the NHIRD (IRB No 1-104-05-145).
Provenance and peer review Not commissioned; externally peer reviewed.