- Email: [email protected]
hyperactivity disorder prescribed methylphenidate: A Taiwan nationwide population-based cohort study
Research in Developmental Disabilities 72 (2018) 96–105
Contents lists available at ScienceDirect
Research in Developmental Disabilities journal homepage: www.elsevier.com/locate/redevdis
Research Paper
Suicide risk reduction in youths with attention-deficit/ hyperactivity disorder prescribed methylphenidate: A Taiwan nationwide population-based cohort study
MARK
Sophie Hsin-Yi Lianga,b, Yao-Hsu Yangc,d,e, Ting-Yu Kuod, Yin-To Liaof,g, Tzu-Chin Linf,g, Yena Leeh, Roger S. McIntyreh,i, Brent A. Kelsenj, Tsu-Nai Wangk, ⁎ Vincent Chin-Hung Chena,l, a
Department of Medicine, Chang Gung University, Taoyuan, Taiwan, ROC Department of Child Psychiatry, Chang Gung Memorial Hospital at Taoyuan, Taoyuan, Taiwan, ROC c Department of Traditional Chinese Medicine, Chiayi Chang Gung Memorial Hospital, Chiayi, Taiwan, ROC d Health Information and Epidemiology Laboratory of Chang Gung Memorial Hospital, Chiayi, Taiwan, ROC e School of Traditional Chinese Medicine, College of Medicine, Chang Gung University, Taoyuan, Taiwan, ROC f Department of Psychiatry, Chung Shan Medical University Hospital, Taichung, Taiwan, ROC g Department of Psychiatry, School of Medicine, Chung Shan Medical University, Taichung, Taiwan, ROC h Mood Disorders Psychopharmacology Unit, University Health Network, Toronto, Canada i Department of Psychiatry and Pharmacology, University of Toronto, Toronto, Canada j Language Center, National Taipei University, New Taipei City, Taiwan, ROC k Department of Public Health, College of Health Science, Kaohsiung Medical University, Kaohsiung, Taiwan, ROC l Chang Gung Medical Foundation, Chiayi Chang Gung Memorial Hospital, Chiayi, Taiwan, ROC b
AR TI CLE I NF O
AB S T R A CT
Number of reviews completed is 2.
Background: Attention-deficit/hyperactivity disorder (ADHD) youths have increased suicide risk. Nevertheless, the beneficial effects of methylphenidate (MPH) on suicide attempt have received relatively little attention. Aims: To investigate the MPH usage and the risk of suicide attempt among ADHD youths. Methods: We identified 84,898 youths less than 18 years old with ADHD diagnosis between 1997 and 2013 from National Health Insurance, and examined whether MPH use affected suicide attempt risk using Cox proportional-hazards models. Outcome and results: Among ADHD youths, reduction of suicide risk was found in patients prescribed 90–180 days of MPH after adjusting for confounding factors (hazard ratio (HR): 0.41, 95% confidence interval (CI): 0.19–0.90) and a greater reduction in those prescribed more than 180 days of MPH (HR: 0.28, 95% CI: 0.17–0.48). Conclusions and implications: We observed a 59% suicide attempt risk reduction among ADHD youths prescribed between 90 and 180 days and a 72% risk reduction in those prescribed more than 180 days of MPH. The protective benefit observed by the group prescribed MPH for longer duration underscores the importance of psychoeducation and compliance enhancement as part of ADHD management. Indication bias is identified as a limitation of this study, and future self-case control study to investigate the association between suicide attempt and ADHD medication is
Keywords: Attention-deficit/hyperactivity disorder Suicide attempt Methylphenidate
Abbreviations: ADHD, attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; MPH, methylphenidate; NHIRD, National Health Insurance Research Database; HR, hazard ratio; CI, confidence interval ⁎ Corresponding author. Current address: Chang Gung Medical Foundation, Chiayi Chang Gung Memorial Hospital, 6. West Sec. Chiapu Road, Putzu City, Chiayi Hsien, 61363, Taiwan, ROC. E-mail address: [email protected] (V.C.-H. Chen). http://dx.doi.org/10.1016/j.ridd.2017.10.023 Received 1 June 2017; Received in revised form 9 October 2017; Accepted 25 October 2017 0891-4222/ © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
suggested. What this paper adds: This nationwide population-based cohort study showed that among ADHD youths, reduction of suicide risk was observed in patients prescribed MPH for duration 90 days and longer, underscoring the importance of appropriate ADHD pharmacotherapy and enhancing drug compliance.
1. Introduction Attention-deficit/hyperactivity disorder (ADHD) is the most common childhood neurodevelopmental disorder that can affect people across their lifespan and is characterized by persistent hyperactivity, impulsivity, impaired attention and distractibility (Wilens, Biederman, & Spencer, 2002). It is well established that ADHD is significantly associated with impairments in day-to-day life, daily functioning at school or at work, as well as on social and emotional development (Buitelaar & Medori, 2010), family stress and healthcare consumption (Banaschewski et al., 2014), increased risk of accident and injury (Chang, Lichtenstein, D’Onofrio, Sjolander, & Larsson, 2014) and increased suicidality (Impey & Heun, 2012). Pharmacotherapy including MPH and atomoxetine can effectively reduce ADHD symptoms at all ages (Asherson, Chen, Craddock, & Taylor, 2007). Additional benefits of stimulant therapy for core symptoms of ADHD are decreased risk for substance use disorder (Faraone & Wilens, 2007), injury (Dalsgaard, Leckman, Mortensen, Nielsen, & Simonsen, 2015), fracture (Chou, Lin, Sung, & Kao, 2014a) and improved functional outcomes (Subcommittee on Attention-Deficit/Hyperactivity Disorder, 2011). It is estimated that suicide claims approximately 1 million lives worldwide every year, and as many as 60% of incidences occur in Asia (Vijayakumar, 2005). In most of the developed world, suicide is among the top 10 leading causes of death for individuals of all ages, and is the third leading cause of death among adolescents, after motor vehicle accidents and homicide (Bridge, Goldstein, & Brent, 2006). Numerous studies have found that ADHD is associated with suicide-related events (Chronis-Tuscano et al., 2010; Impey & Heun, 2012; James, Lai, & Dahl, 2004) and self-harm (Allely, 2014; Meszaros, Horvath, & Balazs, 2017). In a review of six prospective studies measuring annual suicide rates, ADHD subjects were reported to have 1.3–13.6 times greater annual suicide rates in comparison to the general population (Impey & Heun, 2012). A Finnish study of a treatment-naive 1986 birth cohort found that, when compared with adolescents without ADHD, those with ADHD had more suicidal ideation (57% vs. 28%, p < 0.01) and self-harm (69% vs. 32%, p < 0.01), with a clear conclusion that ADHD is a risk factor for both suicidal ideation and self-harm (Hurtig, Taanila, Moilanen, Nordström, & Ebeling, 2012). In Taiwan, Chou et al. conducted a national population-based study from the National Health Insurance Research Database (NHIRD) and found that ADHD patients were at a 4.65-fold greater risk of developing self-poisoning than the control patients were (0.9% vs 0.1%; P < .001, Odds ratio: 4.65, 95% Confidence interval (CI): 2.4–8.9) (Chou, Lin, Sung, & Kao, 2014b). Evidence has confirmed that the rates of suicide ideation, self-injury and suicide attempts are significantly increased in untreated ADHD populations (James et al., 2004). However, the roles of ADHD treatment across disparate measures of suicidality are mixed (Bangs, Tauscher-Wisniewski et al., 2008; Bangs, Wietecha, Wang, Buchanan, & Kelsey, 2014; Chen et al., 2014; Paternite, Loney, Salisbury, & Whaley, 1999). A longitudinal population-based study using a Swedish national registered database exploring the incidence rate of suicide-related events amongst individuals with ADHD did not identify an increase in risk for suicide-related events with ADHD treatment (hazard ratio (HR) = 0.89, 95% CI 0.79–1.00) (Chen et al., 2014). A meta-analysis of 14 clinical trials of adverse-event data associated with atomoxetine treatment reported a statistically significant elevated risk for suicidal thoughts amongst atomoxetine-treated children and adolescents when compared with patients receiving a placebo (Mantel-Haenszel incidence difference of 0.5, 95% CI 0.1–0.8) (Bangs, Tauscher-Wisniewski et al., 2008). Recently, in a separate meta-analysis, Bangs, Wietecha et al. (2014) reported there was no evidence of increased risk for suicidal behavior in atomoxetine-treated pediatric or adult patients (Bangs, Wietecha et al., 2014). Relatively few studies using data for suicide-related events associated with MPH have been reported (Paternite et al., 1999). Summaries of Product Characteristics (SPCs) for MPH list suicidality as a contraindication for usage, contain a warning relating to the emergence of suicidal behaviours, and list suicidal ideation as a rare and serious adverse event (Savill & Bushe, 2012). The mixed findings regarding the effects of MPH across measures of suicidality provides the impetus for our study herein. Thus, as our starting point, we hypothesize that MPH treatment reduces the risk of suicide attempts and that the beneficial effect is influenced by duration of exposure to treatment amongst younger populations newly diagnosed with ADHD whose data is captured within the nation-wide population database in Taiwan.
2. Methods 2.1. Data source Taiwan’s National Health Insurance (NHI) was implemented in March 1995 and established a single-payer insurance system covering the delivery of health care for 92% of the national population at that time. This government-supported insurer centralized the disbursement of all healthcare funds including outpatient, ambulatory, dental service and hospital inpatient care, and, as of 2009, NHI coverage had grown to 99.5% of medical claims (National Health Insurance Administration, 2013). 97
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
2.2. Identification of ADHD cohort In this study, we used a dataset containing participants with at least one or more diagnoses using the following International Classification Diagnostic [ICD]–9-CM codes (World Health Organization, 1975) 290–319 or A-code (a simplified coding system used mainly for ambulatory care before 2000) from the total population cohort of Taiwan NHIRD, which included all patients who received a diagnosis of ADHD (ICD-9 codes: 314.*). The date of the initial ADHD diagnosis was defined as the index date of ADHD. The definition of this ADHD cohort is based on the patients who received diagnosis of ADHD (ICD-9 codes: 314.*) with at least two records of psychiatric outpatient visits within one year or one admission diagnosis between January 1, 1997, and December 31, 2013. A total of 84,898 patients were included in the study cohort. These patients accrued follow-up time beginning on January 1, 1997, and ended on the date the event happened, death, or withdrawal from the registry or on December 31, 2013. Enrolees having a diagnosis of suicide attempt before a diagnosis of ADHD (n = 524), birth years before January 1, 1996, or later than December 31, 2005, (n = 74,860) and the ADHD diagnosis in 2013 (n = 4662) were excluded (Fig. 1). 2.3. Covariates Covariates considered in this analysis included gender, age, the level of urbanization (most urban, urban, rural, and most rural), physical illness ((seizure (ICD-9 codes 345), asthma (ICD-9 codes 493)), psychiatric comorbidities ((mental retardation (ICD-9 code 317–319), autism (ICD-9 code: 299), depression (ICD-9 codes: 296.2, 296.3, 300.4, 311), anxiety disorder (ICD-9: 300)), usage of
Fig. 1. Flow diagram of criteria for inclusion and exclusion in the present study, 1996–2013.
98
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
medication ((MPH, sedative/hypnotic/anxiolytics (excluding ATC code: N05BB01-hydroxyzine), antidepressant drugs, anticonvulsant drugs)). 2.3.1. MPH usage MPH is the only stimulant approved for the treatment of ADHD in Taiwan. Atomoxetine, a non-stimulant pharmacological agent, was approved and launched in Taiwan in 2007. According to the guideline of the Bureau of NHI in Taiwan, MPH is the first-line treatment for ADHD. Atomoxetine is only approved for usage in individuals with ADHD who have had insufficient outcomes with MPH or who have comorbid tic or anxiety disorders. As only 4.3% of individuals with ADHD in the database were prescribed atomoxetine (Lee et al., 2016), we confined our analysis to MPH in this study. The duration of MPH use was calculated and divided into four groups (0, 1–90, 91–180, > 180 days) with days of exposure defined as cumulative duration of MPH prescription within the follow-up time until suicide attempt or the end of study period. 2.4. Outcome measures Suicide attempt was operationalized based on the following ICD-9 codes: E950–E959 including self-inflicted poisoning Table 1 Characteristics of ADHD youths with and without MPH usage. Variables
Frequency (N = 84,898)
MPH = 0 day (N = 19,181)
MPH > 0 day (N = 65,717)
count
%
count
%
count
%
79.6 20.4
74.8 25.2
< .0001
31.0 63.2 5.8
53,247 12,470 7.6 ± 2.4 10,398 50,637 4682
81.0 19.0
19.3 73.9 6.8
14,339 4842 6.9 ± 2.5 5947 12117 1117
15.8 77.1 7.1
< .0001
95.4 4.7
18,102 1079
94.4 5.6
62,847 2870
95.6 4.4
< .0001
59.6 40.4
10,922 8259
56.9 43.1
39,648 26,069
60.3 39.7
< .0001
86.9 13.1
16,862 2319
87.9 12.1
56,881 8836
86.6 13.5
< .0001
90.6 9.4
16,766 2415
87.4 12.6
60,134 5583
91.5 8.5
< .0001
97.4 2.6
18,827 354
98.2 1.9
63,892 1825
97.2 2.8
< .0001
78.2 21.8
17,068 2113
89.0 11.0
49,332 16,385
75.1 24.9
< .0001
78.7 21.3
15,712 3469
81.9 18.1
51,074 14,643
77.7 22.3
< .0001
86.7 13.3
17,269 1912
90.0 10.0
56,327 9390
85.7 14.3
< .0001
75.3 24.7
14,918 4263
77.8 22.2
49,007 16,710
74.6 25.4
< .0001
35.3 47.9 11.5 5.3
7726 8795 1892 768
40.3 45.9 9.9 4.0
22,278 31,844 7880 3715
33.9 48.5 12.0 5.7
< .0001
99.9 0.1
19,144 37
99.8 0.2
65,654 63
99.9 0.1
0.0006
Gender Male 67,586 Female 17,312 Age 7.4 ± 2.4 0–5 16,345 6–11 62,754 12–18 5799 Covariates Seizure No 80,949 Yes 3949 Asthma No 50,570 Yes 34,328 Mental retardation No 73,743 Yes 11,155 ASD No 76,900 Yes 7998 Depression No 82,719 Yes 2179 Anxiety No 66,400 Yes 18,498 Sedative/hypnotics/anxiolyticsa No 66,786 Yes 18,112 Antidepressant drug No 73,596 Yes 11,302 Antiepileptic drug No 63,925 Yes 20,973 Residenceb 1(City) 30,004 2 40,639 3 9772 4(Villages) 4483 Suicide attempt No 84,798 Yes 100
ASD: autism spectrum disorder. a Sedative/hypnotics/Anxiolytics: excluded ATC code: N05BB01. b Residential areas are subgrouped by urbanized level.
99
P value
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
(E950–E952), self-inflicted injury (E953–E959) and adding E980–E989 (undetermined cause) to broaden the definition of suicide attempts (World Health Organization, 1975).
2.5. Data analysis To describe the distribution of the study population, the first step examined the differences between the without MPH use and with MPH use groups through Chi-square tests for categorical variables. To compare the risk for suicide attempts between the two groups, a multivariate Cox proportional hazards regression was employed to determine the risk of suicide attempt. The results are presented as hazard ratios (HRs) with 95% confidence intervals (CIs). Sensitivity analysis of risk for suicide attempts was performed according to the duration of MPH usage stratified into four groups: 0, 1–90, 91–180, above 180 days. The results of comparisons with a P-value of less than or equal to 0.05 were considered to represent statistically significant differences. All data management and statistical analyses were performed using SAS, Version 9.3 software (SAS Institute Inc., Cary, NC, USA).
Table 2 Cox’s proportional hazards model for suicide attempt in ADHD youths. Variables
Univariate Hazard ratio
Gender Male 0.80 Female(ref.) 1.00 Age 0–5(ref.) 1.00 6–11 1.27 12–18 1.03 MPH usage 0 day(ref.) 1.00 > 0 day 0.51 Covariates Seizure No(ref.) 1.00 Yes 1.14 Asthma No(ref.) 1.00 Yes 1.14 Mental retardation No(ref.) 1.00 Yes 1.11 ASD No(ref.) 1.00 Yes 0.43 Depression No(ref.) 1.00 Yes 1.89 Anxiety No(ref.) 1.00 Yes 0.78 Sedative/hypnotics/anxiolyticsa No(ref.) 1.00 Yes 1.48 Antidepressant drug No(ref.) 1.00 Yes 0.92 Antiepileptic drug No(ref.) 1.00 Yes 2.55 Urbanized levelb 1(City) (ref.) 1.00 2 1.84 3 1.53 4(Villages) 3.11
Multivariate 95%CI
P value
Hazard ratio
95%CI
P value
0.51–1.26
0.337
0.88 1.00
0.55–1.40
0.5877
0.78–2.06 0.30–3.49
0.3358 0.9688
1.00 1.30 0.85
0.79–2.14 0.25–2.95
0.3061 0.8017
0.34–0.76
0.001
1.00 0.45
0.29–0.69
0.0002
0.50–2.61
0.7539
1.00 0.65
0.27–1.55
0.3289
0.77–1.69
0.5127
1.00 1.12
0.76–1.67
0.566
0.65–1.90
0.7041
1.00 1.09
0.62–1.90
0.7652
0.18–1.06
0.0664
1.00 0.45
0.18–1.12
0.0866
0.77–4.65
0.1643
1.00 2.19
0.79–6.04
0.1323
0.47–1.29
0.3282
1.00 0.77
0.45–1.30
0.3233
0.97–2.26
0.0691
1.00 1.32
0.85–2.04
0.2147
0.52–1.63
0.7822
1.00 0.78
0.42–1.47
0.4444
1.72–3.78
< 0.0001
1.00 2.61
1.74–3.93
< 0.0001
1.13–2.99 0.75–3.14 1.48–6.53
0.014 0.2468 0.0028
1.00 1.84 1.49 3.17
1.13–2.99 0.73–3.07 1.50–6.69
0.0141 0.2772 0.0025
ASD: autism spectrum disorder. a Sedative/hypnotics/anxiolytic: excluded ATC code: N05BB01. b Residential areas are subgrouped by urbanized level.
100
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
2.6. Ethics The study data did not indicate any identifying information. Institutional Review Board approval was approved at Chang Gung Memorial Hospital. 3. Results 3.1. Sample characteristics We identified 84,898 youths with an incident ADHD diagnosis from this dataset between 1997 and 2013. Of the total subjects, 100 (0.1%) youths met the criteria for suicide attempts during the surveillance period: 37 (0.2%) in the group without any MPH use and 63 (0.1%) in the group with MPH use (p = 0.0006). Statistically significant differences in demographic data and comorbidities and medication use are reported in Table 1. In the group with MPH use, subjects tended to be more male, older age, less urbanized, have a higher percentage of comorbidity with mental retardation and anxiety disorder, have a lower percentage of seizure, asthma, ASD, and more usage of sedative/hypnotic/anxiolytic medication, antidepressant and antiepileptic medication compared to the group without MPH use (Table 1). 3.2. Adjusted hazard ratio analysis of suicide attempt among ADHD youths In the fully adjusted Cox regression model for competing risk analysis, which included adjustment for sex, age, level of urbanization, psychiatric illness, physical illness and medication use, we found a risk reduction of suicide attempt amongst patients with ADHD prescribed MPH (HR: 0.45, 95% CI: 0.29–0.69, p = 0.0002) (Table 2). Additional factors associated with the incidence of suicide attempt were the usage of antiepileptic drugs (HR: 2.61, 95% CI: 1.74–3.93) and less urbanization (HR: 3.17; 95% CI: 1.50–6.69). 3.3. Sensitivity analysis for the risk of suicide attempt stratified by duration of MPH use Furthermore, in the sensitivity analysis of the Cox’s proportional hazards model for suicide attempt stratified by days of MPH usage, the result showed that this protective factor for suicide attempt in ADHD youths was only found in the group who took MPH 90 days to 180 days (HR: 0.41, 95% CI: 0.19–0.90, p = 0.0252), or above 180 days (HR: 0.28, 95% CI: 0.17–0.48, p < .0001), but not in the group with usage between 1 and 90 days (HR: 0.86, 95% CI: 0.53–1.41) (Table 3). 4. Discussion To our knowledge, this is the first study using a national population-based and longitudinal dataset to investigate the duration of MPH usage in individuals with ADHD on measures of suicide attempt. We found a 59% suicide attempt risk reduction among ADHD youths prescribed between 90 days to 180 days and a 72% risk reduction for those with more than 180 days of MPH compared to those without MPH usage, after adjusting for confounding factors. However, this protective effect was not apparent if the subjects received MPH less than 90 days. 4.1. Reduction of suicide attempt risk and prescribed ADHD medication The suicide attempt in children and adolescents diagnosed with ADHD was found to be higher than the normal population from analysis using the NHIRD in Taiwan. The incidence rate of self-poisoning in sex-matched control children and adolescents was 1.36 per 1000 person-years, while in ADHD youths was 6.13 per 1000 person-years (Chou et al., 2014a, 2014b). However, information was not provided regarding the effect of ADHD medication and their analysis only focused on one type of suicide attempt – selfpoisoning. Our study extends current knowledge on the role of ADHD medication in suicide attempt in ADHD youths. We observed a Table 3 Cox’s proportional hazards model for suicide attempt stratified by duration of MPH usage in ADHD youths. Variables
MPH usage 0 day(ref.) 1–90 days 91–180 days > 180 days
Frequency
Univariate
Multivariate
count
%
Hazard ratio
95%CI
P value
Hazard ratio
95%CI
P value
19181 16778 9305 39634
22.6 19.8 11.0 46.7
1.00 0.99 0.49 0.32
0.61–1.61 0.23–1.05 0.19–0.53
0.9744 0.067 < 0.0001
1.00 0.86 0.41 0.28
0.53–1.41 0.19–0.90 0.17–0.48
0.5476 0.0252 < 0.0001
Adjusted variables: gender, age, medication, seizure, asthma, mental retardation, autism-spectrum disorder, depression, anxiety, sedative/hypnotics/anxiolytic (excluded ATC code: N05BB01), antidepressant drug, antiepileptic drug, urbanized level.
101
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
reduction of risk for suicide attempt in ADHD patients with longer duration of medication treatment. Our finding is similar to that reported by Paternite et al. (1999) who conducted a retrospective study examining the contributions of childhood ADHD symptoms and aspects of MPH treatment (including dosage, response and duration) to the prediction of young adult outcomes in boys in a clinical setting in the United States (Paternite et al., 1999). In the foregoing study, higher dosage of MPH was associated with fewer suicide attempts (r = −0.23, p < 0. 05) (Paternite et al., 1999). A longitudinal population-based study using a Swedish national registered database explored the incidence rate of suicide-related events during ADHD drug treatment periods compared to drug nontreatment periods (Chen et al., 2014). The study reported that the ADHD treatment-emergent risk for suicide-related events among ADHD patients was 0.89 (95% CI 0.79–1.00). In psychostimulant users, a reduced rate of suicide-related events was found during treatment periods (HR = 0.81, 95% CI 0.70–0.94), whereas among non-stimulant/mixed users, no significant increase during nonstimulant treatment periods was seen (HR = 0.96, 95% CI 0.77–1.20). 4.2. Longer duration of prescribed medication reduced the risk of suicide attempt In this current study, the protective effect for suicide attempt was observed only in youths with ADHD prescribed more than 90 days. Our results contrast with previous studies: for example, Paternite et al.’s (1999) study found that there was no correlation between treatment duration of MPH in childhood and suicide attempts amongst adult men with ADHD (Paternite et al., 1999). Chen et al.’s (2014) study conducted sensitivity analysis regarding the duration of ADHD treatment and showed no risk reduction either in the treatment status ended in 30 days (HR = 0.96; 95% CI 0.86–1.08) or 3 months (92 days) within patient groups (HR = 0.90, 95% CI 0.80–1.01) (Chen et al., 2014). However, Chen’s study didn’t conduct further analysis of ADHD treatment into stimulant group or non-stimulant group by the duration of treatment. Studies have revealed that the full therapeutic effect of stimulant treatment on ADHD core symptoms might not be observed until 24 weeks after the initiation of treatment, which provided an adequate duration and dosage of stimulant usage (Clemow & Bushe, 2015; Gazer-Snitovsky, Brand-Gothelf, Dubnov-Raz, Weizman, & Gothelf, 2015). We conducted further sensitivity analysis according to the cumulative defined daily dosage (DDD) to analyze the dosage effect of MPH (Appendix A). A Cox proportional-hazards model estimated the risks for suicide attempt between the group prescribed MPH subgrouped by four cumulative dosages: 0, 0–28 DDDs, 28–84 DDDs, and > 84 DDDs. The result was similar to the present study which showed a significant trend toward decreasing risk of suicide with increasing cumulative DDDs of MPH (p for trend < 0.0001). This protective effect was only observed in patients prescribed more than 84 DDDs of MPH after adjusting for confounding factors (HR: 0.29, 95% CI: 0.17–0.48, p < .0001). 4.3. Possible explanation of the risk reduction of longer duration of treatment on suicide attempt In our study, the protective effect likely stemmed from effective treatment for ADHD, including decreasing the extent of ADHD core symptoms along with improvements in interpersonal, psychosocial, and education/vocational functioning (Bangs, TauscherWisniewski et al., 2008). A neurobiological correlate of the foregoing hypothesis is the result from a separate functional MRI study that reported increased activation in the inferior frontal gyrus after 12 weeks MPH therapy, which was associated with decreased severity of impulsivity (Chou, Chia, Shang, & Gau, 2015). Moreover, a meta-analysis of brain image studies revealed there is increased striatal dopamine transporter density in subjects with ADHD in previously medicated patients and lower density in drugnaïve subjects (Fusar-Poli, Rubia, Rossi, Sartori, & Balottin, 2012). A separate neurobiological mechanism could involve molecular systems implicated in synaptoplasticity. For example, plasma brain-derived neurotrophic factor (BDNF) levels have been reported to be increased in youths with ADHD after 6-weeks MPH treatment (Amiri et al., 2013). Evidence suggests that BDNF is implicated in the pathogenesis, as well as the phenomenology, of mood disorders (e.g. suicidality) (Dwivedi, 2009). It could be hypothesized that the beneficial effects that we have observed on measures of suicide attempt after 180 days of MPH treatment were mediated by increased expression in neurobiological systems involved in synaptoplasticity, learning, and memory. 4.4. Association between antiepileptic drugs and suicide risk among ADHD youths In this study, we found usage of antiepileptic drugs among ADHD youths was associated with a 2.6-fold increased suicide risk (HR: 2.61, 95% CI: 1.74–3.93, p < .0001) after adjusting for confounding factors. Antiepileptic drugs are a heterogeneous pharmacological class, which are widespread and commonly used to treat epilepsy and other conditions in children and adolescents, including aggressive behaviours, pervasive developmental disorder, movement disorder, bipolar disorder, neuralgia and migraine (Golden, Haut, & Moshé, 2006; Landmark, Larsson, Rytter, & Johannessen, 2009). In 2008, the Food and Drug Administration (FDA) in the USA issued a safety warning to antiepileptic drugs prescribers of a 2-fold increase in the risk of suicidal behavior or ideation (odds ratio: 1.80, 95% CI: 1.24–2.66) (U.S. Food and Drug Administarion, 2008). Since the publication of the FDA alert on the risk of suicide associated with antiepileptic drugs, several studies have emerged which appear to validate the previously reported result of a 2- to 4-fold overall increase in suicide risk (Olesen et al., 2010; Patorno et al., 2010). Among the studies related to antiepileptics and suicide risk associated with psychiatric comorbidities, Arana, Wentworth, Ayuso-Mateos, and Arellano (2010) focus on with or without psychiatric comorbid and depression in epilepsy patients (Arana et al., 2010). Studies that were conducted with patients with bipolar disorder showed mixed results (Gibbons, Hur, Brown, & Mann, 2009; Søndergård, Lopez, Andersen, & Kessing, 2008; Leon et al., 2012). As far as we know, there are few studies exploring the association of suicide risk between those subjects comorbid ADHD and prescribed antiepileptic drugs. However, evidence of significant pharmacokinetic or pharmacodynamic interactions between 102
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
MPH and antiepileptic drugs were not found in a systematic review (Markowitz & Patrick, 2001). Future study related to antiepileptic drug and suicide risks among ADHD youths is needed. 4.5. Strengths and limitations 4.5.1. Strengths Key strengths of the study were the nationally representative sample, physician-based diagnosis to identify ADHD cases, longitudinal dimension of the analysis, clear information of medication prescription, and range of covariates considered including sociodemographic, general health and mental health factors. The clinical diagnoses of ADHD in the registration database has been reported to have high validity (Dalsgaard, Hansen, Mortensen, Damm, & Thomsen, 2001). Both recall bias and selection bias are avoided with registration medical claim data from the NHIRD. The reverse causal relationship between ADHD and suicide attempt is eliminated by excluding suicide attempt prior to ADHD diagnosis. 4.5.2. Limitations There are also important limitations encompassing lack of data of subjects’ body weight, medication compliance, effect of behavior therapy or psychoeducation, as well as stressful life event, smoking and family psychopathology. Moreover, we included ADHD patients only using MPH and excluded those receiving atomoxetine or combined atomoxetine and MPH. Atomoxetine use in ADHD patients is rare in Taiwan (4.3%) (Lee et al., 2016), and including atomoxetine use in our study resulted in the similar protective effect. Furthermore, the possible shortcoming relating to indication bias, the severity of ADHD, is overlooked in this study. To avoid indication bias, future study conducted employing a self-case control design to investigate the association between suicide attempt and ADHD medication is suggested. The group under longer medication treatment may reflect higher severity of ADHD, which itself is a risk factor of suicide attempt. Nevertheless, we found a suicide attempt reduction in the group with a longer duration of medication treatment. One possible interpretation of this is that longer duration of medication usage reflects a phenomenon of better compliance and better insight and motivation for treatment in both youths and their parents (Gajria et al., 2014). In addition, studies from other countries have shown that children from more highly educated families tend to have higher medical or health-related service utilization (Fusar-Poli et al., 2012). Moreover, some studies have shown higher youth suicide attempt rates for those from socio-economically disadvantaged and poor parenting and schooling backgrounds (Agerbo, Nordentoft, & Mortensen, 2002). However, Gau et al. (2008) found that higher paternal education level was one of the risk factors of poor drug adherence of MPH usage in a national survey study of adherence in ADHD youths in Taiwan (Gau et al., 2008). 5. Conclusion The current study employing data from the NHIRD found a decreased risk of suicide attempt in youths with ADHD prescribed MPH treatment for greater than 90 days compared to those without any medication use and greater suicide reduction after more than 180 days use. However, this protective effect of MPH treatment was not observed among those prescribed with medication for less than 90 days compared to those without any medication prescription. Our results indicate that providing appropriate ADHD pharmacotherapy and enhancing drug compliance not only decreases ADHD core symptoms but also may decrease suicide attempt. In clinical practice, it is important for healthcare providers to emphasize the importance of duration and adherence to medication in all individuals prescribed stimulants for ADHD. The generalization of the protective effects of MPH on different types of suicide attempts needs to be verified through further studies. Funding source The present study is supported in part by the Ministry of Science and Technology, Taiwan, ROC (MOST 102-2314-B-040-004MY3) and Chang Gung Medical Foundation, Chiayi Chang Gung Memorial Hospital, Taiwan, ROC (CMRPG6E0261). Conflict of interest Conflict of interest Author Vincent Chin-Hung Chen has received speaking honoraria from Pfizer, Eli Lilly, Janssen, Astellas, GlaxoS-mithKline,and AstraZeneca, and has been an investigator in two clini-cal trials from Eli Lilly and Janssen. All other authors declare that they have no conflicts of interest. Acknowledgements The authors would also like to thank the Health Information and Epidemiology Laboratory (CLRPG6G0041) for their comments and assistance in data analysis. This study was supported by a grant from Chang Gung Memorial Hospital, Chia-Yi Branch, and based on the National Health Insurance Research Database provided by the Central Bureau of National Health Insurance, the Department of Health, and managed by the National Health Research Institutes. The interpretation and conclusions contained herein do not represent those of Bureau of National Health Insurance, Department of Health, or National Health Research Institutes.
103
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
Appendix A See Table A1.
Table A1 Cox’s proportional hazards model for suicide attempt stratified by cumulative defined daily dose (DDD) of MPH exposure in ADHD youths. Variables
MPH usage 0 DDD(ref.) 0 –28 DDDs 28–84 DDDs > 84 DDDs
Frequency
Univariate
Multivariate
count
%
Hazard ratio
95%CI
P value
Hazard ratio
95%CI
P value
19,064 13,839 13,145 38,850
22.5 16.3 15.5 45.8
1.00 0.90 0.65 0.33
0.53–1.51 0.36–1.18 0.20–0.54
0.6818 0.1592 < 0.0001
1.00 0.77 0.56 0.29
0.46–1.31 0.30–1.03 0.17–0.48
0.3389 0.0614 < 0.0001
Adjusted variables: gender, age, medication, seizure, asthma, mental retardation, autism-spectrum disorder, depression, anxiety, sedative/hypnotics/anxiolytic (excluded ATC code: N05BB01), antidepressant drug, antiepileptic drug, urbanized level.
References Agerbo, E., Nordentoft, M., & Mortensen, P. B. (2002). Familial, psychiatric, and socioeconomic risk factors for suicide in young people: Nested case-control study. British Medical Journal, 325(7355), 74–77. Allely, C. S. (2014). The association of ADHD symptoms to self-harm behaviours: A systematic PRISMA review. BMC Psychiatry, 14(1), 133. Amiri, A., Torabi Parizi, G., Kousha, M., Saadat, F., Modabbernia, M. J., Najafi, K., & Atrkar Roushan, Z. (2013). Changes in plasma brain-derived neurotrophic factor (BDNF) levels induced by methylphenidate in children with attention deficit-hyperactivity disorder (ADHD). Progress in Neuropsychopharmacology & Biologiical Psychiatry, 47, 20–24. Arana, A., Wentworth, C. E., Ayuso-Mateos, J. L., & Arellano, F. M. (2010). Suicide-related events in patients treated with antiepileptic drugs. New England Journal of Medicine, 363(6), 542–551. Asherson, P., Chen, W., Craddock, B., & Taylor, E. (2007). Adult attention-deficit hyperactivity disorder: Recognition and treatment in general adult psychiatry. The British Journal of Psychiatry, 190(1), 4–5. Banaschewski, T., Johnson, M., Lecendreux, M., Zuddas, A., Adeyi, B., Hodgkins, P., ... Coghill, D. R. (2014). Health-related quality of life and functional outcomes from a randomized-withdrawal study of long-term lisdexamfetamine dimesylate treatment in children and adolescents with attention-deficit/hyperactivity disorder. CNS Drugs, 28(12), 1191–1203. Bangs, M. E., Tauscher-Wisniewski, S., Polzer, J., Zhang, S., Acharya, N., Desaiah, D., ... Allen, A. J. (2008). Meta-analysis of suicide-related behavior events in patients treated with atomoxetine. Journal of the American Academy of Child and Adolescescent Psychiatry, 47(2), 209–218. Bangs, M. E., Wietecha, L. A., Wang, S., Buchanan, A. S., & Kelsey, D. K. (2014). Meta-analysis of suicide-related behavior or ideation in child, adolescent, and adult patients treated with atomoxetine. Journal of Child and Adolescent Psychopharmacology, 24(8), 426–434. Bridge, J. A., Goldstein, T. R., & Brent, D. A. (2006). Adolescent suicide and suicidal behavior. Journal of Child Psychology and Psychiatry, 47(3–4), 372–394. Buitelaar, J., & Medori, R. (2010). Treating attention-deficit/hyperactivity disorder beyond symptom control alone in children and adolescents: A review of the potential benefits of long-acting stimulants. European Child and Adolescent Psychiatry, 19(4), 325–340. Chang, Z., Lichtenstein, P., D'Onofrio, B. M., Sjolander, A., & Larsson, H. (2014). Serious transport accidents in adults with attention-deficit/hyperactivity disorder and the effect of medication: A population-based study. JAMA Psychiatry, 71(3), 319–325. Chen, Q., Sjölander, A., Runeson, B., D’Onofrio, B. M., Lichtenstein, P., & Larsson, H. (2014). Drug treatment for attention-deficit/hyperactivity disorder and suicidal behaviour: Register based study. British Medical Journal, 348. Chou, I. C., Lin, C. C., Sung, F. C., & Kao, C. H. (2014a). Attention-deficit hyperactivity disorder increases the risk of deliberate self-poisoning: A population-based cohort. European Psychiatry, 29(8), 523–527. Chou, I., Lin, C. C., Sung, F. C., & Kao, C. H. (2014b). Attention-deficit-hyperactivity disorder increases risk of bone fracture: A population-based cohort study. Developmental Medicine & Child Neurology, 56(11), 1111–1116. Chou, T.-L., Chia, S., Shang, C.-Y., & Gau, S. S.-F. (2015). Differential therapeutic effects of 12-week treatment of atomoxetine and methylphenidate on drug-naïve children with attention deficit/hyperactivity disorder: A counting stroop functional MRI study. European Neuropsychopharmacology, 25(12), 2300–2310. Chronis-Tuscano, A., Molina, B. S., Pelham, W. E., Applegate, B., Dahlke, A., Overmyer, M., & Lahey, B. B. (2010). Very early predictors of adolescent depression and suicide attempts in children with attention-deficit/hyperactivity disorder. Archives of General Psychiatry, 67(10), 1044–1051. Clemow, D. B., & Bushe, C. J. (2015). Atomoxetine in patients with ADHD: A clinical and pharmacological review of the onset, trajectory, duration of response and implications for patients. Journal of Psychopharmacology, 29(12), 1221–1230. Dalsgaard, S., Hansen, N., Mortensen, P. B., Damm, D., & Thomsen, P. (2001). Reassessment of ADHD in a historical cohort of children treated with stimulants in the period 1969–1989. European Child & Adolescent Psychiatry, 10(4), 230–239. Dalsgaard, S., Leckman, J. F., Mortensen, P. B., Nielsen, H. S., & Simonsen, M. (2015). Effect of drugs on the risk of injuries in children with attention deficit hyperactivity disorder: A prospective cohort study. The Lancet Psychiatry, 2(8), 702–709. Dwivedi, Y. (2009). Brain-derived neurotrophic factor: Role in depression and suicide. Neuropsychiatric Disease and Treatment, 5, 433–449. Faraone, S. V., & Wilens, T. E. (2007). Effect of stimulant medications for attention-deficit/hyperactivity disorder on later substance use and the potential for stimulant misuse, abuse, and diversion. Journal of Clinical Psychiatry, 68(Suppl. 11), 15–22. Fusar-Poli, P., Rubia, K., Rossi, G., Sartori, G., & Balottin, U. (2012). Striatal dopamine transporter alterations in ADHD: pathophysiology or adaptation to psychostimulants? A meta-analysis. American Journal of Psychiatry, 169(3), 264–272. Gajria, K., Lu, M., Sikirica, V., Greven, P., Zhong, Y., Qin, P., & Xie, J. (2014). Adherence, persistence, and medication discontinuation in patients with attentiondeficit/hyperactivity disorder—A systematic literature review. Neuropsychiatric Disease and Treatment, 10, 1543. Gau, S. S.-F., Chen, S.-J., Chou, W.-J., Cheng, H., Tang, C.-S., Chang, H.-L., ... Chou, M.-C. (2008). National survey of adherence, efficacy, and side effects of methylphenidate in children with attention-deficit/hyperactivity disorder in Taiwan. Journal of Clininical Psychiatry, 69(1), 131–140. Gazer-Snitovsky, M., Brand-Gothelf, A., Dubnov-Raz, G., Weizman, A., & Gothelf, D. (2015). High familial correlation in methylphenidate response and side effect profile. Journal of Attention Disorders. http://dx.doi.org/10.1177/1087054715580844. Gibbons, R. D., Hur, K., Brown, C. H., & Mann, J. J. (2009). Relationship between antiepileptic drugs and suicide attempts in patients with bipolar disorder. Archives of General Psychiatry, 66(12), 1354–1360.
104
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
Golden, A. S., Haut, S. R., & Moshé, S. L. (2006). Nonepileptic uses of antiepileptic drugs in children and adolescents. Pediatric Neurology, 34(6), 421–432. Hurtig, T., Taanila, A., Moilanen, I., Nordström, T., & Ebeling, H. (2012). Suicidal and self-harm behaviour associated with adolescent attention deficit hyperactivity disorder—A study in the Northern Finland Birth Cohort 1986. Nordic Journal of Psychiatry, 66(5), 320–328. Impey, M., & Heun, R. (2012). Completed suicide, ideation and attempt in attention deficit hyperactivity disorder. Acta Psychiatrica Scandinavica, 125(2), 93–102. James, A., Lai, F. H., & Dahl, C. (2004). Attention deficit hyperactivity disorder and suicide: A review of possible associations. Acta Psychiatrica Scandinavica, 110(6), 408–415. Landmark, C. J., Larsson, P. G., Rytter, E., & Johannessen, S. I. (2009). Antiepileptic drugs in epilepsy and other disorders—A population-based study of prescriptions. Epilepsy Research, 87(1), 31–39. Lee, M.-J., Yang, K.-C., Shyu, Y.-C., Yuan, S.-S., Yang, C.-J., Lee, S.-Y., ... Wang, L.-J. (2016). Attention-deficit hyperactivity disorder, its treatment with medication and the probability of developing a depressive disorder: A nationwide population-based study in Taiwan. Journal of Affective Disorders, 189, 110–117. Leon, A. C., Solomon, D. A., Li, C., Fiedorowicz, J. G., Coryell, W. H., Endicott, J., & Keller, M. B. (2012). Antiepileptic drugs for bipolar disorder and the risk of suicidal behavior: A 30-year observational study. The American Journal of Psychiatry, 169(3), 285. Markowitz, J. S., & Patrick, K. S. (2001). Pharmacokinetic and pharmacodynamic drug interactions in the treatment of attention-deficit hyperactivity disorder. Clinical Pharmacokinetics, 40(10), 753–772. Meszaros, G., Horvath, L. O., & Balazs, J. (2017). Self-injury and externalizing pathology: A systematic literature review. BMC Psychiatry, 17(1), 160. National Health Insurance Administration (2013). National health insurance administration insurance regulations. Formely Bureau of National Health Insurance. Olesen, J. B., Hansen, P. R., Erdal, J., Abildstrøm, S. Z., Weeke, P., Fosbøl, E. L., ... Gislason, G. H. (2010). Antiepileptic drugs and risk of suicide: A nationwide study. Pharmacoepidemiology and Drug Safety, 19(5), 518–524. Paternite, C. E., Loney, J., Salisbury, H., & Whaley, M. A. (1999). Childhood inattention-overactivity, aggression, and stimulant medication history as predictors of young adult outcomes. Journal of Child and Adolescent Psychopharmacology, 9(3), 169–184. Patorno, E., Bohn, R. L., Wahl, P. M., Avorn, J., Patrick, A. R., Liu, J., & Schneeweiss, S. (2010). Anticonvulsant medications and the risk of suicide, attempted suicide, or violent death. JAMA: The Journal of the American Medical Association, 303(14), 1401–1409. Søndergård, L., Lopez, A. G., Andersen, P. K., & Kessing, L. V. (2008). Mood-stabilizing pharmacological treatment in bipolar disorders and risk of suicide. Bipolar Disorders, 10(1), 87–94. Savill, N., & Bushe, C. J. (2012). A systematic review of the safety information contained within the Summaries of Product Characteristics of medications licensed in the United Kingdom for Attention Deficit Hyperactivity Disorder. how does the safety prescribing advice compare with national guidance? Child and Adolescent Psychiatry and Mental Health, 6(2). Subcommittee on Attention-Deficit/Hyperactivity Disorder (2011). ADHD: Clinical practice guideline for the diagnosis, evaluation, and treatment of attention-deficit/ hyperactivity disorder in children and adolescents. Pediatrics, 128(5), 1007–1022. U.S. Food and Drug Administarion (2008). Statistical review and evaluation: Antiepileptic drugs and suicidality. [23 May 2008] http://www.fda.gov/ohrms/dockets/ac/ 08/briefing/2008-4372b1-01-FDA.pdf. Vijayakumar, L. (2005). Suicide and mental disorders in Asia. International Review of Psychiatry, 17(2), 109–114. Wilens, T. E., Biederman, J., & Spencer, T. J. (2002). Attention deficit/hyperactivity disorder across the lifespan. Annual Review of Medicine, 53(1), 113–131. World Health Organization (1975). International statistical classification of diseases, ninth revision (ICD-9). Geneva: World Health Organization.
105
Contents lists available at ScienceDirect
Research in Developmental Disabilities journal homepage: www.elsevier.com/locate/redevdis
Research Paper
Suicide risk reduction in youths with attention-deficit/ hyperactivity disorder prescribed methylphenidate: A Taiwan nationwide population-based cohort study
MARK
Sophie Hsin-Yi Lianga,b, Yao-Hsu Yangc,d,e, Ting-Yu Kuod, Yin-To Liaof,g, Tzu-Chin Linf,g, Yena Leeh, Roger S. McIntyreh,i, Brent A. Kelsenj, Tsu-Nai Wangk, ⁎ Vincent Chin-Hung Chena,l, a
Department of Medicine, Chang Gung University, Taoyuan, Taiwan, ROC Department of Child Psychiatry, Chang Gung Memorial Hospital at Taoyuan, Taoyuan, Taiwan, ROC c Department of Traditional Chinese Medicine, Chiayi Chang Gung Memorial Hospital, Chiayi, Taiwan, ROC d Health Information and Epidemiology Laboratory of Chang Gung Memorial Hospital, Chiayi, Taiwan, ROC e School of Traditional Chinese Medicine, College of Medicine, Chang Gung University, Taoyuan, Taiwan, ROC f Department of Psychiatry, Chung Shan Medical University Hospital, Taichung, Taiwan, ROC g Department of Psychiatry, School of Medicine, Chung Shan Medical University, Taichung, Taiwan, ROC h Mood Disorders Psychopharmacology Unit, University Health Network, Toronto, Canada i Department of Psychiatry and Pharmacology, University of Toronto, Toronto, Canada j Language Center, National Taipei University, New Taipei City, Taiwan, ROC k Department of Public Health, College of Health Science, Kaohsiung Medical University, Kaohsiung, Taiwan, ROC l Chang Gung Medical Foundation, Chiayi Chang Gung Memorial Hospital, Chiayi, Taiwan, ROC b
AR TI CLE I NF O
AB S T R A CT
Number of reviews completed is 2.
Background: Attention-deficit/hyperactivity disorder (ADHD) youths have increased suicide risk. Nevertheless, the beneficial effects of methylphenidate (MPH) on suicide attempt have received relatively little attention. Aims: To investigate the MPH usage and the risk of suicide attempt among ADHD youths. Methods: We identified 84,898 youths less than 18 years old with ADHD diagnosis between 1997 and 2013 from National Health Insurance, and examined whether MPH use affected suicide attempt risk using Cox proportional-hazards models. Outcome and results: Among ADHD youths, reduction of suicide risk was found in patients prescribed 90–180 days of MPH after adjusting for confounding factors (hazard ratio (HR): 0.41, 95% confidence interval (CI): 0.19–0.90) and a greater reduction in those prescribed more than 180 days of MPH (HR: 0.28, 95% CI: 0.17–0.48). Conclusions and implications: We observed a 59% suicide attempt risk reduction among ADHD youths prescribed between 90 and 180 days and a 72% risk reduction in those prescribed more than 180 days of MPH. The protective benefit observed by the group prescribed MPH for longer duration underscores the importance of psychoeducation and compliance enhancement as part of ADHD management. Indication bias is identified as a limitation of this study, and future self-case control study to investigate the association between suicide attempt and ADHD medication is
Keywords: Attention-deficit/hyperactivity disorder Suicide attempt Methylphenidate
Abbreviations: ADHD, attention-deficit/hyperactivity disorder; ASD, autism spectrum disorder; MPH, methylphenidate; NHIRD, National Health Insurance Research Database; HR, hazard ratio; CI, confidence interval ⁎ Corresponding author. Current address: Chang Gung Medical Foundation, Chiayi Chang Gung Memorial Hospital, 6. West Sec. Chiapu Road, Putzu City, Chiayi Hsien, 61363, Taiwan, ROC. E-mail address: [email protected] (V.C.-H. Chen). http://dx.doi.org/10.1016/j.ridd.2017.10.023 Received 1 June 2017; Received in revised form 9 October 2017; Accepted 25 October 2017 0891-4222/ © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
suggested. What this paper adds: This nationwide population-based cohort study showed that among ADHD youths, reduction of suicide risk was observed in patients prescribed MPH for duration 90 days and longer, underscoring the importance of appropriate ADHD pharmacotherapy and enhancing drug compliance.
1. Introduction Attention-deficit/hyperactivity disorder (ADHD) is the most common childhood neurodevelopmental disorder that can affect people across their lifespan and is characterized by persistent hyperactivity, impulsivity, impaired attention and distractibility (Wilens, Biederman, & Spencer, 2002). It is well established that ADHD is significantly associated with impairments in day-to-day life, daily functioning at school or at work, as well as on social and emotional development (Buitelaar & Medori, 2010), family stress and healthcare consumption (Banaschewski et al., 2014), increased risk of accident and injury (Chang, Lichtenstein, D’Onofrio, Sjolander, & Larsson, 2014) and increased suicidality (Impey & Heun, 2012). Pharmacotherapy including MPH and atomoxetine can effectively reduce ADHD symptoms at all ages (Asherson, Chen, Craddock, & Taylor, 2007). Additional benefits of stimulant therapy for core symptoms of ADHD are decreased risk for substance use disorder (Faraone & Wilens, 2007), injury (Dalsgaard, Leckman, Mortensen, Nielsen, & Simonsen, 2015), fracture (Chou, Lin, Sung, & Kao, 2014a) and improved functional outcomes (Subcommittee on Attention-Deficit/Hyperactivity Disorder, 2011). It is estimated that suicide claims approximately 1 million lives worldwide every year, and as many as 60% of incidences occur in Asia (Vijayakumar, 2005). In most of the developed world, suicide is among the top 10 leading causes of death for individuals of all ages, and is the third leading cause of death among adolescents, after motor vehicle accidents and homicide (Bridge, Goldstein, & Brent, 2006). Numerous studies have found that ADHD is associated with suicide-related events (Chronis-Tuscano et al., 2010; Impey & Heun, 2012; James, Lai, & Dahl, 2004) and self-harm (Allely, 2014; Meszaros, Horvath, & Balazs, 2017). In a review of six prospective studies measuring annual suicide rates, ADHD subjects were reported to have 1.3–13.6 times greater annual suicide rates in comparison to the general population (Impey & Heun, 2012). A Finnish study of a treatment-naive 1986 birth cohort found that, when compared with adolescents without ADHD, those with ADHD had more suicidal ideation (57% vs. 28%, p < 0.01) and self-harm (69% vs. 32%, p < 0.01), with a clear conclusion that ADHD is a risk factor for both suicidal ideation and self-harm (Hurtig, Taanila, Moilanen, Nordström, & Ebeling, 2012). In Taiwan, Chou et al. conducted a national population-based study from the National Health Insurance Research Database (NHIRD) and found that ADHD patients were at a 4.65-fold greater risk of developing self-poisoning than the control patients were (0.9% vs 0.1%; P < .001, Odds ratio: 4.65, 95% Confidence interval (CI): 2.4–8.9) (Chou, Lin, Sung, & Kao, 2014b). Evidence has confirmed that the rates of suicide ideation, self-injury and suicide attempts are significantly increased in untreated ADHD populations (James et al., 2004). However, the roles of ADHD treatment across disparate measures of suicidality are mixed (Bangs, Tauscher-Wisniewski et al., 2008; Bangs, Wietecha, Wang, Buchanan, & Kelsey, 2014; Chen et al., 2014; Paternite, Loney, Salisbury, & Whaley, 1999). A longitudinal population-based study using a Swedish national registered database exploring the incidence rate of suicide-related events amongst individuals with ADHD did not identify an increase in risk for suicide-related events with ADHD treatment (hazard ratio (HR) = 0.89, 95% CI 0.79–1.00) (Chen et al., 2014). A meta-analysis of 14 clinical trials of adverse-event data associated with atomoxetine treatment reported a statistically significant elevated risk for suicidal thoughts amongst atomoxetine-treated children and adolescents when compared with patients receiving a placebo (Mantel-Haenszel incidence difference of 0.5, 95% CI 0.1–0.8) (Bangs, Tauscher-Wisniewski et al., 2008). Recently, in a separate meta-analysis, Bangs, Wietecha et al. (2014) reported there was no evidence of increased risk for suicidal behavior in atomoxetine-treated pediatric or adult patients (Bangs, Wietecha et al., 2014). Relatively few studies using data for suicide-related events associated with MPH have been reported (Paternite et al., 1999). Summaries of Product Characteristics (SPCs) for MPH list suicidality as a contraindication for usage, contain a warning relating to the emergence of suicidal behaviours, and list suicidal ideation as a rare and serious adverse event (Savill & Bushe, 2012). The mixed findings regarding the effects of MPH across measures of suicidality provides the impetus for our study herein. Thus, as our starting point, we hypothesize that MPH treatment reduces the risk of suicide attempts and that the beneficial effect is influenced by duration of exposure to treatment amongst younger populations newly diagnosed with ADHD whose data is captured within the nation-wide population database in Taiwan.
2. Methods 2.1. Data source Taiwan’s National Health Insurance (NHI) was implemented in March 1995 and established a single-payer insurance system covering the delivery of health care for 92% of the national population at that time. This government-supported insurer centralized the disbursement of all healthcare funds including outpatient, ambulatory, dental service and hospital inpatient care, and, as of 2009, NHI coverage had grown to 99.5% of medical claims (National Health Insurance Administration, 2013). 97
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
2.2. Identification of ADHD cohort In this study, we used a dataset containing participants with at least one or more diagnoses using the following International Classification Diagnostic [ICD]–9-CM codes (World Health Organization, 1975) 290–319 or A-code (a simplified coding system used mainly for ambulatory care before 2000) from the total population cohort of Taiwan NHIRD, which included all patients who received a diagnosis of ADHD (ICD-9 codes: 314.*). The date of the initial ADHD diagnosis was defined as the index date of ADHD. The definition of this ADHD cohort is based on the patients who received diagnosis of ADHD (ICD-9 codes: 314.*) with at least two records of psychiatric outpatient visits within one year or one admission diagnosis between January 1, 1997, and December 31, 2013. A total of 84,898 patients were included in the study cohort. These patients accrued follow-up time beginning on January 1, 1997, and ended on the date the event happened, death, or withdrawal from the registry or on December 31, 2013. Enrolees having a diagnosis of suicide attempt before a diagnosis of ADHD (n = 524), birth years before January 1, 1996, or later than December 31, 2005, (n = 74,860) and the ADHD diagnosis in 2013 (n = 4662) were excluded (Fig. 1). 2.3. Covariates Covariates considered in this analysis included gender, age, the level of urbanization (most urban, urban, rural, and most rural), physical illness ((seizure (ICD-9 codes 345), asthma (ICD-9 codes 493)), psychiatric comorbidities ((mental retardation (ICD-9 code 317–319), autism (ICD-9 code: 299), depression (ICD-9 codes: 296.2, 296.3, 300.4, 311), anxiety disorder (ICD-9: 300)), usage of
Fig. 1. Flow diagram of criteria for inclusion and exclusion in the present study, 1996–2013.
98
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
medication ((MPH, sedative/hypnotic/anxiolytics (excluding ATC code: N05BB01-hydroxyzine), antidepressant drugs, anticonvulsant drugs)). 2.3.1. MPH usage MPH is the only stimulant approved for the treatment of ADHD in Taiwan. Atomoxetine, a non-stimulant pharmacological agent, was approved and launched in Taiwan in 2007. According to the guideline of the Bureau of NHI in Taiwan, MPH is the first-line treatment for ADHD. Atomoxetine is only approved for usage in individuals with ADHD who have had insufficient outcomes with MPH or who have comorbid tic or anxiety disorders. As only 4.3% of individuals with ADHD in the database were prescribed atomoxetine (Lee et al., 2016), we confined our analysis to MPH in this study. The duration of MPH use was calculated and divided into four groups (0, 1–90, 91–180, > 180 days) with days of exposure defined as cumulative duration of MPH prescription within the follow-up time until suicide attempt or the end of study period. 2.4. Outcome measures Suicide attempt was operationalized based on the following ICD-9 codes: E950–E959 including self-inflicted poisoning Table 1 Characteristics of ADHD youths with and without MPH usage. Variables
Frequency (N = 84,898)
MPH = 0 day (N = 19,181)
MPH > 0 day (N = 65,717)
count
%
count
%
count
%
79.6 20.4
74.8 25.2
< .0001
31.0 63.2 5.8
53,247 12,470 7.6 ± 2.4 10,398 50,637 4682
81.0 19.0
19.3 73.9 6.8
14,339 4842 6.9 ± 2.5 5947 12117 1117
15.8 77.1 7.1
< .0001
95.4 4.7
18,102 1079
94.4 5.6
62,847 2870
95.6 4.4
< .0001
59.6 40.4
10,922 8259
56.9 43.1
39,648 26,069
60.3 39.7
< .0001
86.9 13.1
16,862 2319
87.9 12.1
56,881 8836
86.6 13.5
< .0001
90.6 9.4
16,766 2415
87.4 12.6
60,134 5583
91.5 8.5
< .0001
97.4 2.6
18,827 354
98.2 1.9
63,892 1825
97.2 2.8
< .0001
78.2 21.8
17,068 2113
89.0 11.0
49,332 16,385
75.1 24.9
< .0001
78.7 21.3
15,712 3469
81.9 18.1
51,074 14,643
77.7 22.3
< .0001
86.7 13.3
17,269 1912
90.0 10.0
56,327 9390
85.7 14.3
< .0001
75.3 24.7
14,918 4263
77.8 22.2
49,007 16,710
74.6 25.4
< .0001
35.3 47.9 11.5 5.3
7726 8795 1892 768
40.3 45.9 9.9 4.0
22,278 31,844 7880 3715
33.9 48.5 12.0 5.7
< .0001
99.9 0.1
19,144 37
99.8 0.2
65,654 63
99.9 0.1
0.0006
Gender Male 67,586 Female 17,312 Age 7.4 ± 2.4 0–5 16,345 6–11 62,754 12–18 5799 Covariates Seizure No 80,949 Yes 3949 Asthma No 50,570 Yes 34,328 Mental retardation No 73,743 Yes 11,155 ASD No 76,900 Yes 7998 Depression No 82,719 Yes 2179 Anxiety No 66,400 Yes 18,498 Sedative/hypnotics/anxiolyticsa No 66,786 Yes 18,112 Antidepressant drug No 73,596 Yes 11,302 Antiepileptic drug No 63,925 Yes 20,973 Residenceb 1(City) 30,004 2 40,639 3 9772 4(Villages) 4483 Suicide attempt No 84,798 Yes 100
ASD: autism spectrum disorder. a Sedative/hypnotics/Anxiolytics: excluded ATC code: N05BB01. b Residential areas are subgrouped by urbanized level.
99
P value
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
(E950–E952), self-inflicted injury (E953–E959) and adding E980–E989 (undetermined cause) to broaden the definition of suicide attempts (World Health Organization, 1975).
2.5. Data analysis To describe the distribution of the study population, the first step examined the differences between the without MPH use and with MPH use groups through Chi-square tests for categorical variables. To compare the risk for suicide attempts between the two groups, a multivariate Cox proportional hazards regression was employed to determine the risk of suicide attempt. The results are presented as hazard ratios (HRs) with 95% confidence intervals (CIs). Sensitivity analysis of risk for suicide attempts was performed according to the duration of MPH usage stratified into four groups: 0, 1–90, 91–180, above 180 days. The results of comparisons with a P-value of less than or equal to 0.05 were considered to represent statistically significant differences. All data management and statistical analyses were performed using SAS, Version 9.3 software (SAS Institute Inc., Cary, NC, USA).
Table 2 Cox’s proportional hazards model for suicide attempt in ADHD youths. Variables
Univariate Hazard ratio
Gender Male 0.80 Female(ref.) 1.00 Age 0–5(ref.) 1.00 6–11 1.27 12–18 1.03 MPH usage 0 day(ref.) 1.00 > 0 day 0.51 Covariates Seizure No(ref.) 1.00 Yes 1.14 Asthma No(ref.) 1.00 Yes 1.14 Mental retardation No(ref.) 1.00 Yes 1.11 ASD No(ref.) 1.00 Yes 0.43 Depression No(ref.) 1.00 Yes 1.89 Anxiety No(ref.) 1.00 Yes 0.78 Sedative/hypnotics/anxiolyticsa No(ref.) 1.00 Yes 1.48 Antidepressant drug No(ref.) 1.00 Yes 0.92 Antiepileptic drug No(ref.) 1.00 Yes 2.55 Urbanized levelb 1(City) (ref.) 1.00 2 1.84 3 1.53 4(Villages) 3.11
Multivariate 95%CI
P value
Hazard ratio
95%CI
P value
0.51–1.26
0.337
0.88 1.00
0.55–1.40
0.5877
0.78–2.06 0.30–3.49
0.3358 0.9688
1.00 1.30 0.85
0.79–2.14 0.25–2.95
0.3061 0.8017
0.34–0.76
0.001
1.00 0.45
0.29–0.69
0.0002
0.50–2.61
0.7539
1.00 0.65
0.27–1.55
0.3289
0.77–1.69
0.5127
1.00 1.12
0.76–1.67
0.566
0.65–1.90
0.7041
1.00 1.09
0.62–1.90
0.7652
0.18–1.06
0.0664
1.00 0.45
0.18–1.12
0.0866
0.77–4.65
0.1643
1.00 2.19
0.79–6.04
0.1323
0.47–1.29
0.3282
1.00 0.77
0.45–1.30
0.3233
0.97–2.26
0.0691
1.00 1.32
0.85–2.04
0.2147
0.52–1.63
0.7822
1.00 0.78
0.42–1.47
0.4444
1.72–3.78
< 0.0001
1.00 2.61
1.74–3.93
< 0.0001
1.13–2.99 0.75–3.14 1.48–6.53
0.014 0.2468 0.0028
1.00 1.84 1.49 3.17
1.13–2.99 0.73–3.07 1.50–6.69
0.0141 0.2772 0.0025
ASD: autism spectrum disorder. a Sedative/hypnotics/anxiolytic: excluded ATC code: N05BB01. b Residential areas are subgrouped by urbanized level.
100
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
2.6. Ethics The study data did not indicate any identifying information. Institutional Review Board approval was approved at Chang Gung Memorial Hospital. 3. Results 3.1. Sample characteristics We identified 84,898 youths with an incident ADHD diagnosis from this dataset between 1997 and 2013. Of the total subjects, 100 (0.1%) youths met the criteria for suicide attempts during the surveillance period: 37 (0.2%) in the group without any MPH use and 63 (0.1%) in the group with MPH use (p = 0.0006). Statistically significant differences in demographic data and comorbidities and medication use are reported in Table 1. In the group with MPH use, subjects tended to be more male, older age, less urbanized, have a higher percentage of comorbidity with mental retardation and anxiety disorder, have a lower percentage of seizure, asthma, ASD, and more usage of sedative/hypnotic/anxiolytic medication, antidepressant and antiepileptic medication compared to the group without MPH use (Table 1). 3.2. Adjusted hazard ratio analysis of suicide attempt among ADHD youths In the fully adjusted Cox regression model for competing risk analysis, which included adjustment for sex, age, level of urbanization, psychiatric illness, physical illness and medication use, we found a risk reduction of suicide attempt amongst patients with ADHD prescribed MPH (HR: 0.45, 95% CI: 0.29–0.69, p = 0.0002) (Table 2). Additional factors associated with the incidence of suicide attempt were the usage of antiepileptic drugs (HR: 2.61, 95% CI: 1.74–3.93) and less urbanization (HR: 3.17; 95% CI: 1.50–6.69). 3.3. Sensitivity analysis for the risk of suicide attempt stratified by duration of MPH use Furthermore, in the sensitivity analysis of the Cox’s proportional hazards model for suicide attempt stratified by days of MPH usage, the result showed that this protective factor for suicide attempt in ADHD youths was only found in the group who took MPH 90 days to 180 days (HR: 0.41, 95% CI: 0.19–0.90, p = 0.0252), or above 180 days (HR: 0.28, 95% CI: 0.17–0.48, p < .0001), but not in the group with usage between 1 and 90 days (HR: 0.86, 95% CI: 0.53–1.41) (Table 3). 4. Discussion To our knowledge, this is the first study using a national population-based and longitudinal dataset to investigate the duration of MPH usage in individuals with ADHD on measures of suicide attempt. We found a 59% suicide attempt risk reduction among ADHD youths prescribed between 90 days to 180 days and a 72% risk reduction for those with more than 180 days of MPH compared to those without MPH usage, after adjusting for confounding factors. However, this protective effect was not apparent if the subjects received MPH less than 90 days. 4.1. Reduction of suicide attempt risk and prescribed ADHD medication The suicide attempt in children and adolescents diagnosed with ADHD was found to be higher than the normal population from analysis using the NHIRD in Taiwan. The incidence rate of self-poisoning in sex-matched control children and adolescents was 1.36 per 1000 person-years, while in ADHD youths was 6.13 per 1000 person-years (Chou et al., 2014a, 2014b). However, information was not provided regarding the effect of ADHD medication and their analysis only focused on one type of suicide attempt – selfpoisoning. Our study extends current knowledge on the role of ADHD medication in suicide attempt in ADHD youths. We observed a Table 3 Cox’s proportional hazards model for suicide attempt stratified by duration of MPH usage in ADHD youths. Variables
MPH usage 0 day(ref.) 1–90 days 91–180 days > 180 days
Frequency
Univariate
Multivariate
count
%
Hazard ratio
95%CI
P value
Hazard ratio
95%CI
P value
19181 16778 9305 39634
22.6 19.8 11.0 46.7
1.00 0.99 0.49 0.32
0.61–1.61 0.23–1.05 0.19–0.53
0.9744 0.067 < 0.0001
1.00 0.86 0.41 0.28
0.53–1.41 0.19–0.90 0.17–0.48
0.5476 0.0252 < 0.0001
Adjusted variables: gender, age, medication, seizure, asthma, mental retardation, autism-spectrum disorder, depression, anxiety, sedative/hypnotics/anxiolytic (excluded ATC code: N05BB01), antidepressant drug, antiepileptic drug, urbanized level.
101
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
reduction of risk for suicide attempt in ADHD patients with longer duration of medication treatment. Our finding is similar to that reported by Paternite et al. (1999) who conducted a retrospective study examining the contributions of childhood ADHD symptoms and aspects of MPH treatment (including dosage, response and duration) to the prediction of young adult outcomes in boys in a clinical setting in the United States (Paternite et al., 1999). In the foregoing study, higher dosage of MPH was associated with fewer suicide attempts (r = −0.23, p < 0. 05) (Paternite et al., 1999). A longitudinal population-based study using a Swedish national registered database explored the incidence rate of suicide-related events during ADHD drug treatment periods compared to drug nontreatment periods (Chen et al., 2014). The study reported that the ADHD treatment-emergent risk for suicide-related events among ADHD patients was 0.89 (95% CI 0.79–1.00). In psychostimulant users, a reduced rate of suicide-related events was found during treatment periods (HR = 0.81, 95% CI 0.70–0.94), whereas among non-stimulant/mixed users, no significant increase during nonstimulant treatment periods was seen (HR = 0.96, 95% CI 0.77–1.20). 4.2. Longer duration of prescribed medication reduced the risk of suicide attempt In this current study, the protective effect for suicide attempt was observed only in youths with ADHD prescribed more than 90 days. Our results contrast with previous studies: for example, Paternite et al.’s (1999) study found that there was no correlation between treatment duration of MPH in childhood and suicide attempts amongst adult men with ADHD (Paternite et al., 1999). Chen et al.’s (2014) study conducted sensitivity analysis regarding the duration of ADHD treatment and showed no risk reduction either in the treatment status ended in 30 days (HR = 0.96; 95% CI 0.86–1.08) or 3 months (92 days) within patient groups (HR = 0.90, 95% CI 0.80–1.01) (Chen et al., 2014). However, Chen’s study didn’t conduct further analysis of ADHD treatment into stimulant group or non-stimulant group by the duration of treatment. Studies have revealed that the full therapeutic effect of stimulant treatment on ADHD core symptoms might not be observed until 24 weeks after the initiation of treatment, which provided an adequate duration and dosage of stimulant usage (Clemow & Bushe, 2015; Gazer-Snitovsky, Brand-Gothelf, Dubnov-Raz, Weizman, & Gothelf, 2015). We conducted further sensitivity analysis according to the cumulative defined daily dosage (DDD) to analyze the dosage effect of MPH (Appendix A). A Cox proportional-hazards model estimated the risks for suicide attempt between the group prescribed MPH subgrouped by four cumulative dosages: 0, 0–28 DDDs, 28–84 DDDs, and > 84 DDDs. The result was similar to the present study which showed a significant trend toward decreasing risk of suicide with increasing cumulative DDDs of MPH (p for trend < 0.0001). This protective effect was only observed in patients prescribed more than 84 DDDs of MPH after adjusting for confounding factors (HR: 0.29, 95% CI: 0.17–0.48, p < .0001). 4.3. Possible explanation of the risk reduction of longer duration of treatment on suicide attempt In our study, the protective effect likely stemmed from effective treatment for ADHD, including decreasing the extent of ADHD core symptoms along with improvements in interpersonal, psychosocial, and education/vocational functioning (Bangs, TauscherWisniewski et al., 2008). A neurobiological correlate of the foregoing hypothesis is the result from a separate functional MRI study that reported increased activation in the inferior frontal gyrus after 12 weeks MPH therapy, which was associated with decreased severity of impulsivity (Chou, Chia, Shang, & Gau, 2015). Moreover, a meta-analysis of brain image studies revealed there is increased striatal dopamine transporter density in subjects with ADHD in previously medicated patients and lower density in drugnaïve subjects (Fusar-Poli, Rubia, Rossi, Sartori, & Balottin, 2012). A separate neurobiological mechanism could involve molecular systems implicated in synaptoplasticity. For example, plasma brain-derived neurotrophic factor (BDNF) levels have been reported to be increased in youths with ADHD after 6-weeks MPH treatment (Amiri et al., 2013). Evidence suggests that BDNF is implicated in the pathogenesis, as well as the phenomenology, of mood disorders (e.g. suicidality) (Dwivedi, 2009). It could be hypothesized that the beneficial effects that we have observed on measures of suicide attempt after 180 days of MPH treatment were mediated by increased expression in neurobiological systems involved in synaptoplasticity, learning, and memory. 4.4. Association between antiepileptic drugs and suicide risk among ADHD youths In this study, we found usage of antiepileptic drugs among ADHD youths was associated with a 2.6-fold increased suicide risk (HR: 2.61, 95% CI: 1.74–3.93, p < .0001) after adjusting for confounding factors. Antiepileptic drugs are a heterogeneous pharmacological class, which are widespread and commonly used to treat epilepsy and other conditions in children and adolescents, including aggressive behaviours, pervasive developmental disorder, movement disorder, bipolar disorder, neuralgia and migraine (Golden, Haut, & Moshé, 2006; Landmark, Larsson, Rytter, & Johannessen, 2009). In 2008, the Food and Drug Administration (FDA) in the USA issued a safety warning to antiepileptic drugs prescribers of a 2-fold increase in the risk of suicidal behavior or ideation (odds ratio: 1.80, 95% CI: 1.24–2.66) (U.S. Food and Drug Administarion, 2008). Since the publication of the FDA alert on the risk of suicide associated with antiepileptic drugs, several studies have emerged which appear to validate the previously reported result of a 2- to 4-fold overall increase in suicide risk (Olesen et al., 2010; Patorno et al., 2010). Among the studies related to antiepileptics and suicide risk associated with psychiatric comorbidities, Arana, Wentworth, Ayuso-Mateos, and Arellano (2010) focus on with or without psychiatric comorbid and depression in epilepsy patients (Arana et al., 2010). Studies that were conducted with patients with bipolar disorder showed mixed results (Gibbons, Hur, Brown, & Mann, 2009; Søndergård, Lopez, Andersen, & Kessing, 2008; Leon et al., 2012). As far as we know, there are few studies exploring the association of suicide risk between those subjects comorbid ADHD and prescribed antiepileptic drugs. However, evidence of significant pharmacokinetic or pharmacodynamic interactions between 102
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
MPH and antiepileptic drugs were not found in a systematic review (Markowitz & Patrick, 2001). Future study related to antiepileptic drug and suicide risks among ADHD youths is needed. 4.5. Strengths and limitations 4.5.1. Strengths Key strengths of the study were the nationally representative sample, physician-based diagnosis to identify ADHD cases, longitudinal dimension of the analysis, clear information of medication prescription, and range of covariates considered including sociodemographic, general health and mental health factors. The clinical diagnoses of ADHD in the registration database has been reported to have high validity (Dalsgaard, Hansen, Mortensen, Damm, & Thomsen, 2001). Both recall bias and selection bias are avoided with registration medical claim data from the NHIRD. The reverse causal relationship between ADHD and suicide attempt is eliminated by excluding suicide attempt prior to ADHD diagnosis. 4.5.2. Limitations There are also important limitations encompassing lack of data of subjects’ body weight, medication compliance, effect of behavior therapy or psychoeducation, as well as stressful life event, smoking and family psychopathology. Moreover, we included ADHD patients only using MPH and excluded those receiving atomoxetine or combined atomoxetine and MPH. Atomoxetine use in ADHD patients is rare in Taiwan (4.3%) (Lee et al., 2016), and including atomoxetine use in our study resulted in the similar protective effect. Furthermore, the possible shortcoming relating to indication bias, the severity of ADHD, is overlooked in this study. To avoid indication bias, future study conducted employing a self-case control design to investigate the association between suicide attempt and ADHD medication is suggested. The group under longer medication treatment may reflect higher severity of ADHD, which itself is a risk factor of suicide attempt. Nevertheless, we found a suicide attempt reduction in the group with a longer duration of medication treatment. One possible interpretation of this is that longer duration of medication usage reflects a phenomenon of better compliance and better insight and motivation for treatment in both youths and their parents (Gajria et al., 2014). In addition, studies from other countries have shown that children from more highly educated families tend to have higher medical or health-related service utilization (Fusar-Poli et al., 2012). Moreover, some studies have shown higher youth suicide attempt rates for those from socio-economically disadvantaged and poor parenting and schooling backgrounds (Agerbo, Nordentoft, & Mortensen, 2002). However, Gau et al. (2008) found that higher paternal education level was one of the risk factors of poor drug adherence of MPH usage in a national survey study of adherence in ADHD youths in Taiwan (Gau et al., 2008). 5. Conclusion The current study employing data from the NHIRD found a decreased risk of suicide attempt in youths with ADHD prescribed MPH treatment for greater than 90 days compared to those without any medication use and greater suicide reduction after more than 180 days use. However, this protective effect of MPH treatment was not observed among those prescribed with medication for less than 90 days compared to those without any medication prescription. Our results indicate that providing appropriate ADHD pharmacotherapy and enhancing drug compliance not only decreases ADHD core symptoms but also may decrease suicide attempt. In clinical practice, it is important for healthcare providers to emphasize the importance of duration and adherence to medication in all individuals prescribed stimulants for ADHD. The generalization of the protective effects of MPH on different types of suicide attempts needs to be verified through further studies. Funding source The present study is supported in part by the Ministry of Science and Technology, Taiwan, ROC (MOST 102-2314-B-040-004MY3) and Chang Gung Medical Foundation, Chiayi Chang Gung Memorial Hospital, Taiwan, ROC (CMRPG6E0261). Conflict of interest Conflict of interest Author Vincent Chin-Hung Chen has received speaking honoraria from Pfizer, Eli Lilly, Janssen, Astellas, GlaxoS-mithKline,and AstraZeneca, and has been an investigator in two clini-cal trials from Eli Lilly and Janssen. All other authors declare that they have no conflicts of interest. Acknowledgements The authors would also like to thank the Health Information and Epidemiology Laboratory (CLRPG6G0041) for their comments and assistance in data analysis. This study was supported by a grant from Chang Gung Memorial Hospital, Chia-Yi Branch, and based on the National Health Insurance Research Database provided by the Central Bureau of National Health Insurance, the Department of Health, and managed by the National Health Research Institutes. The interpretation and conclusions contained herein do not represent those of Bureau of National Health Insurance, Department of Health, or National Health Research Institutes.
103
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
Appendix A See Table A1.
Table A1 Cox’s proportional hazards model for suicide attempt stratified by cumulative defined daily dose (DDD) of MPH exposure in ADHD youths. Variables
MPH usage 0 DDD(ref.) 0 –28 DDDs 28–84 DDDs > 84 DDDs
Frequency
Univariate
Multivariate
count
%
Hazard ratio
95%CI
P value
Hazard ratio
95%CI
P value
19,064 13,839 13,145 38,850
22.5 16.3 15.5 45.8
1.00 0.90 0.65 0.33
0.53–1.51 0.36–1.18 0.20–0.54
0.6818 0.1592 < 0.0001
1.00 0.77 0.56 0.29
0.46–1.31 0.30–1.03 0.17–0.48
0.3389 0.0614 < 0.0001
Adjusted variables: gender, age, medication, seizure, asthma, mental retardation, autism-spectrum disorder, depression, anxiety, sedative/hypnotics/anxiolytic (excluded ATC code: N05BB01), antidepressant drug, antiepileptic drug, urbanized level.
References Agerbo, E., Nordentoft, M., & Mortensen, P. B. (2002). Familial, psychiatric, and socioeconomic risk factors for suicide in young people: Nested case-control study. British Medical Journal, 325(7355), 74–77. Allely, C. S. (2014). The association of ADHD symptoms to self-harm behaviours: A systematic PRISMA review. BMC Psychiatry, 14(1), 133. Amiri, A., Torabi Parizi, G., Kousha, M., Saadat, F., Modabbernia, M. J., Najafi, K., & Atrkar Roushan, Z. (2013). Changes in plasma brain-derived neurotrophic factor (BDNF) levels induced by methylphenidate in children with attention deficit-hyperactivity disorder (ADHD). Progress in Neuropsychopharmacology & Biologiical Psychiatry, 47, 20–24. Arana, A., Wentworth, C. E., Ayuso-Mateos, J. L., & Arellano, F. M. (2010). Suicide-related events in patients treated with antiepileptic drugs. New England Journal of Medicine, 363(6), 542–551. Asherson, P., Chen, W., Craddock, B., & Taylor, E. (2007). Adult attention-deficit hyperactivity disorder: Recognition and treatment in general adult psychiatry. The British Journal of Psychiatry, 190(1), 4–5. Banaschewski, T., Johnson, M., Lecendreux, M., Zuddas, A., Adeyi, B., Hodgkins, P., ... Coghill, D. R. (2014). Health-related quality of life and functional outcomes from a randomized-withdrawal study of long-term lisdexamfetamine dimesylate treatment in children and adolescents with attention-deficit/hyperactivity disorder. CNS Drugs, 28(12), 1191–1203. Bangs, M. E., Tauscher-Wisniewski, S., Polzer, J., Zhang, S., Acharya, N., Desaiah, D., ... Allen, A. J. (2008). Meta-analysis of suicide-related behavior events in patients treated with atomoxetine. Journal of the American Academy of Child and Adolescescent Psychiatry, 47(2), 209–218. Bangs, M. E., Wietecha, L. A., Wang, S., Buchanan, A. S., & Kelsey, D. K. (2014). Meta-analysis of suicide-related behavior or ideation in child, adolescent, and adult patients treated with atomoxetine. Journal of Child and Adolescent Psychopharmacology, 24(8), 426–434. Bridge, J. A., Goldstein, T. R., & Brent, D. A. (2006). Adolescent suicide and suicidal behavior. Journal of Child Psychology and Psychiatry, 47(3–4), 372–394. Buitelaar, J., & Medori, R. (2010). Treating attention-deficit/hyperactivity disorder beyond symptom control alone in children and adolescents: A review of the potential benefits of long-acting stimulants. European Child and Adolescent Psychiatry, 19(4), 325–340. Chang, Z., Lichtenstein, P., D'Onofrio, B. M., Sjolander, A., & Larsson, H. (2014). Serious transport accidents in adults with attention-deficit/hyperactivity disorder and the effect of medication: A population-based study. JAMA Psychiatry, 71(3), 319–325. Chen, Q., Sjölander, A., Runeson, B., D’Onofrio, B. M., Lichtenstein, P., & Larsson, H. (2014). Drug treatment for attention-deficit/hyperactivity disorder and suicidal behaviour: Register based study. British Medical Journal, 348. Chou, I. C., Lin, C. C., Sung, F. C., & Kao, C. H. (2014a). Attention-deficit hyperactivity disorder increases the risk of deliberate self-poisoning: A population-based cohort. European Psychiatry, 29(8), 523–527. Chou, I., Lin, C. C., Sung, F. C., & Kao, C. H. (2014b). Attention-deficit-hyperactivity disorder increases risk of bone fracture: A population-based cohort study. Developmental Medicine & Child Neurology, 56(11), 1111–1116. Chou, T.-L., Chia, S., Shang, C.-Y., & Gau, S. S.-F. (2015). Differential therapeutic effects of 12-week treatment of atomoxetine and methylphenidate on drug-naïve children with attention deficit/hyperactivity disorder: A counting stroop functional MRI study. European Neuropsychopharmacology, 25(12), 2300–2310. Chronis-Tuscano, A., Molina, B. S., Pelham, W. E., Applegate, B., Dahlke, A., Overmyer, M., & Lahey, B. B. (2010). Very early predictors of adolescent depression and suicide attempts in children with attention-deficit/hyperactivity disorder. Archives of General Psychiatry, 67(10), 1044–1051. Clemow, D. B., & Bushe, C. J. (2015). Atomoxetine in patients with ADHD: A clinical and pharmacological review of the onset, trajectory, duration of response and implications for patients. Journal of Psychopharmacology, 29(12), 1221–1230. Dalsgaard, S., Hansen, N., Mortensen, P. B., Damm, D., & Thomsen, P. (2001). Reassessment of ADHD in a historical cohort of children treated with stimulants in the period 1969–1989. European Child & Adolescent Psychiatry, 10(4), 230–239. Dalsgaard, S., Leckman, J. F., Mortensen, P. B., Nielsen, H. S., & Simonsen, M. (2015). Effect of drugs on the risk of injuries in children with attention deficit hyperactivity disorder: A prospective cohort study. The Lancet Psychiatry, 2(8), 702–709. Dwivedi, Y. (2009). Brain-derived neurotrophic factor: Role in depression and suicide. Neuropsychiatric Disease and Treatment, 5, 433–449. Faraone, S. V., & Wilens, T. E. (2007). Effect of stimulant medications for attention-deficit/hyperactivity disorder on later substance use and the potential for stimulant misuse, abuse, and diversion. Journal of Clinical Psychiatry, 68(Suppl. 11), 15–22. Fusar-Poli, P., Rubia, K., Rossi, G., Sartori, G., & Balottin, U. (2012). Striatal dopamine transporter alterations in ADHD: pathophysiology or adaptation to psychostimulants? A meta-analysis. American Journal of Psychiatry, 169(3), 264–272. Gajria, K., Lu, M., Sikirica, V., Greven, P., Zhong, Y., Qin, P., & Xie, J. (2014). Adherence, persistence, and medication discontinuation in patients with attentiondeficit/hyperactivity disorder—A systematic literature review. Neuropsychiatric Disease and Treatment, 10, 1543. Gau, S. S.-F., Chen, S.-J., Chou, W.-J., Cheng, H., Tang, C.-S., Chang, H.-L., ... Chou, M.-C. (2008). National survey of adherence, efficacy, and side effects of methylphenidate in children with attention-deficit/hyperactivity disorder in Taiwan. Journal of Clininical Psychiatry, 69(1), 131–140. Gazer-Snitovsky, M., Brand-Gothelf, A., Dubnov-Raz, G., Weizman, A., & Gothelf, D. (2015). High familial correlation in methylphenidate response and side effect profile. Journal of Attention Disorders. http://dx.doi.org/10.1177/1087054715580844. Gibbons, R. D., Hur, K., Brown, C. H., & Mann, J. J. (2009). Relationship between antiepileptic drugs and suicide attempts in patients with bipolar disorder. Archives of General Psychiatry, 66(12), 1354–1360.
104
Research in Developmental Disabilities 72 (2018) 96–105
S.H.-Y. Liang et al.
Golden, A. S., Haut, S. R., & Moshé, S. L. (2006). Nonepileptic uses of antiepileptic drugs in children and adolescents. Pediatric Neurology, 34(6), 421–432. Hurtig, T., Taanila, A., Moilanen, I., Nordström, T., & Ebeling, H. (2012). Suicidal and self-harm behaviour associated with adolescent attention deficit hyperactivity disorder—A study in the Northern Finland Birth Cohort 1986. Nordic Journal of Psychiatry, 66(5), 320–328. Impey, M., & Heun, R. (2012). Completed suicide, ideation and attempt in attention deficit hyperactivity disorder. Acta Psychiatrica Scandinavica, 125(2), 93–102. James, A., Lai, F. H., & Dahl, C. (2004). Attention deficit hyperactivity disorder and suicide: A review of possible associations. Acta Psychiatrica Scandinavica, 110(6), 408–415. Landmark, C. J., Larsson, P. G., Rytter, E., & Johannessen, S. I. (2009). Antiepileptic drugs in epilepsy and other disorders—A population-based study of prescriptions. Epilepsy Research, 87(1), 31–39. Lee, M.-J., Yang, K.-C., Shyu, Y.-C., Yuan, S.-S., Yang, C.-J., Lee, S.-Y., ... Wang, L.-J. (2016). Attention-deficit hyperactivity disorder, its treatment with medication and the probability of developing a depressive disorder: A nationwide population-based study in Taiwan. Journal of Affective Disorders, 189, 110–117. Leon, A. C., Solomon, D. A., Li, C., Fiedorowicz, J. G., Coryell, W. H., Endicott, J., & Keller, M. B. (2012). Antiepileptic drugs for bipolar disorder and the risk of suicidal behavior: A 30-year observational study. The American Journal of Psychiatry, 169(3), 285. Markowitz, J. S., & Patrick, K. S. (2001). Pharmacokinetic and pharmacodynamic drug interactions in the treatment of attention-deficit hyperactivity disorder. Clinical Pharmacokinetics, 40(10), 753–772. Meszaros, G., Horvath, L. O., & Balazs, J. (2017). Self-injury and externalizing pathology: A systematic literature review. BMC Psychiatry, 17(1), 160. National Health Insurance Administration (2013). National health insurance administration insurance regulations. Formely Bureau of National Health Insurance. Olesen, J. B., Hansen, P. R., Erdal, J., Abildstrøm, S. Z., Weeke, P., Fosbøl, E. L., ... Gislason, G. H. (2010). Antiepileptic drugs and risk of suicide: A nationwide study. Pharmacoepidemiology and Drug Safety, 19(5), 518–524. Paternite, C. E., Loney, J., Salisbury, H., & Whaley, M. A. (1999). Childhood inattention-overactivity, aggression, and stimulant medication history as predictors of young adult outcomes. Journal of Child and Adolescent Psychopharmacology, 9(3), 169–184. Patorno, E., Bohn, R. L., Wahl, P. M., Avorn, J., Patrick, A. R., Liu, J., & Schneeweiss, S. (2010). Anticonvulsant medications and the risk of suicide, attempted suicide, or violent death. JAMA: The Journal of the American Medical Association, 303(14), 1401–1409. Søndergård, L., Lopez, A. G., Andersen, P. K., & Kessing, L. V. (2008). Mood-stabilizing pharmacological treatment in bipolar disorders and risk of suicide. Bipolar Disorders, 10(1), 87–94. Savill, N., & Bushe, C. J. (2012). A systematic review of the safety information contained within the Summaries of Product Characteristics of medications licensed in the United Kingdom for Attention Deficit Hyperactivity Disorder. how does the safety prescribing advice compare with national guidance? Child and Adolescent Psychiatry and Mental Health, 6(2). Subcommittee on Attention-Deficit/Hyperactivity Disorder (2011). ADHD: Clinical practice guideline for the diagnosis, evaluation, and treatment of attention-deficit/ hyperactivity disorder in children and adolescents. Pediatrics, 128(5), 1007–1022. U.S. Food and Drug Administarion (2008). Statistical review and evaluation: Antiepileptic drugs and suicidality. [23 May 2008] http://www.fda.gov/ohrms/dockets/ac/ 08/briefing/2008-4372b1-01-FDA.pdf. Vijayakumar, L. (2005). Suicide and mental disorders in Asia. International Review of Psychiatry, 17(2), 109–114. Wilens, T. E., Biederman, J., & Spencer, T. J. (2002). Attention deficit/hyperactivity disorder across the lifespan. Annual Review of Medicine, 53(1), 113–131. World Health Organization (1975). International statistical classification of diseases, ninth revision (ICD-9). Geneva: World Health Organization.
105