Better efficacy for the osmotic release oral system methylphenidate among poor adherents to immediate-release methylphenidate in the three ADHD subtypes


Susan Shur-Fen Gau, md, phd, Department of Psychiatry, National Taiwan University Hospital & College of Medicine, No. 7 Chung-Shan South Road, Taipei 10002, Taiwan. Email:


Aims:  To determine factors for switching to osmotic release oral system methylphenidate (OROS-MPH) among poor adherents to immediate-release methylphenidate (IR-MPH); and to compare the efficacy of OROS-MPH on the three attention-deficit/hyperactivity disorder (ADHD) subtypes in a multi-site prospective observational study in Taiwan.

Methods:  The sample included 240 children with ADHD, aged 6–16 years, who were poor adherents to IR-MPH, 137 of whom were switched to OROS-MPH. The child psychiatrists diagnosed the Diagnostic Statistical Manual of Mental Disorders (4th edition) ADHD subtypes and assessed the medical history, adherence, side-effects, global ADHD severity, and family/school effectiveness. Parents reported their child's behavioral symptoms.

Results:  The determinants for an OROS-MPH switch were higher dosage, shorter treatment and thrice-daily administration of IR-MPH, and more severe inattention symptoms. Hyperactivity and oppositional symptoms were greater in the ADHD combined and hyperactive-impulsive subtypes than the inattentive subtype. Switching to OROS-MPH significantly improved behavioral symptoms and family/school measures, and this was most evident in the ADHD-combined group, followed by the ADHD-inattentive group. Inattention influenced not only academic performance, but also overall classroom behaviors and the parent–child relationship, with the latter two also influenced by oppositional symptoms.

Conclusions:  This study suggests better efficacy for the OROS-MPH among poor adherents to IR-MPH; however, its effectiveness varied across the three ADHD subtypes ( number NCT00460720).

ATTENTION-DEFICIT/HYPERACTIVITY DISORDER (ADHD), a common childhood disorder with worldwide-pooled prevalence estimated at 5.29%,1 affects 7.5% of the school-aged children in Taiwan.2 Given the impact of ADHD on individuals, families and society, affected individuals usually need treatment3–5 and the published reports have clearly documented pharmacotherapy as an effective, fundamental and superior treatment strategy compared to psychosocial interventions alone regarding ADHD symptom reduction.6

Methylphenidate (MPH) is the most commonly used and extensively studied medication for ADHD,4 and its immediate-release (IR) formulation (IR-MPH) was the only stimulant medication used for ADHD in Taiwan until the osmotic release oral system (OROS) was launched in October 2003.5 Because IR-MPH has a relatively short half-life, maintenance of therapeutic efficacy typically necessitates administration two or three times a day,7 with this staggered dosing causing adherence problems,7 which, in turn, leads to suboptimal symptom management of ADHD and less-than-favorable outcomes in terms of psychosocial and academic functioning.6

Although Western studies7 and our previous studies5,8,9 have identified several predictors for poor adherence to IR-MPH such as multiple daily dosing,5,7 older age,5 and intolerable side-effects,5,7 little is known about determinants for OROS-MPH adherence.

A few investigations comparing MPH efficacy across three ADHD subtypes have reported that the efficacy of OROS-MPH treatment is particularly evident for the predominantly inattentive ADHD subtype (ADHD-I),10 and that MPH comparably reduced inattention in both ADHD-I and the combined subtype (ADHD-C), whereas reduction in hyperactivity was more evident in ADHD-C.11 However, Barbaresi et al.12 did not find different responses to MPH among the ADHD subtypes.

Therefore, this study aimed to: (i) identify potential predictors for the switch to OROS-MPH in poor IR-MPH adherents; (ii) determine the predictors for OROS-MPH adherence; and (iii) compare the changes in behavioral symptoms, parent–child interaction, classroom behavior, and academic performance among children with one of three ADHD subtypes.



This was a multi-site observational study involving 12 hospitals and 20 board-certificated child psychiatrists.9 The sample included 240 children with clinical diagnosis of DSM-IV ADHD (male, 81.7%) aged 6–15 years (10.4 ± 2.6), who were poor adherents after IR-MPH treatment for at least 3 months. The diagnoses of ADHD were made based on clinical interviews in accordance with the diagnostic criteria of DSM-IV ADHD by board-certificated child psychiatrists, who obtained information from parent and child interviews, observation of the child's behaviors, and rating scales reported by teachers. The ADHD children who had systemic disease or clinically significant gastrointestinal problems, and comorbid conditions with major psychiatric disorders except conduct disorder and oppositional defiant disorder were excluded.


The Joint Institute Review Board of the Department of Health, Taiwan (JIRB: 06-056-P) and the Institutional Review Boards of each study site approved this clinical study. Written informed consent was obtained from the participants and their parents prior to enrolment.

In Phase I, the investigators completed the Clinical Global Impression-ADHD-Symptom Severity (CGI-ADHD-S) rating, and determined whether there had been any side-effects during treatment. Meanwhile, the parents/caregivers completed the Chinese version of the Swanson, Nolan, and Pelham, version IV scale (SNAP-IV-C) at the clinic. Drug adherence was then assessed by investigators based on the reports of the patients and their parents/caregivers. Of the 607 IR-MPH-treated children with ADHD who underwent the baseline assessment,9 240 (39.5%) were poor adherents to IR-MPH.

Of these children, 137 were switched to OROS-MPH treatment (57%; 110 boys and 27 girls) and 103 remained on IR-MPH, based on the clinician's decision. After at least 3 weeks of OROS-MPH treatment, 124 patients (90.5%) completed the Phase-II assessment; however, 13 children missed their appointments. The assessments in Phase-II included clinical evaluation of drug adherence, CGI-ADHD-S, parent–child interaction, overall classroom behavior, overall academic performance, and side-effects. The parents also completed the SNAP-IV-C in this phase.


Adherence measurements

Poor adherence to IR-MPH at Phase I was defined as missing one or more doses on at least 2 school days weekly for 4 school weeks. The subjective evaluation of adherence was based on retrospective feedback from patients themselves, and their parents/caregivers. The objective evaluation of adherence was based on the frequency of missed doses on a daily basis. Poor adherence to OROS-MPH in Phase II of the investigation was defined where this medication was missed for 2 school days per school week during the treatment period (3 weeks or more) or if participants/parents reported OROS-MPH adherence was the same/worse than the IR-MPH analog.

Efficacy measures

Efficacy measures included the CGI-ADHD-S, which was evaluated by the investigators, and the Chinese SNAP-IV, as reported by the parents in both study phases. The children were also assessed by the investigators in Phase II (medication switch) to assess any changes in parent–child interaction, overall classroom behavior, and overall academic performance as compared to baseline based on a clinical interview with the participants and parents. The four possible ratings for the three efficacy measures were ‘much better’, ‘somewhat better’, ‘no change’ and ‘worse’. Before implementation of this study, all investigators had performed the interview training for 8 h to reach agreement with SS Gau.

Clinical Global Impression-ADHD Severity rating

The CGI-ADHD-S is a single-item rating of the global severity of ADHD symptoms based on the clinician's overall experience with other ADHD patients. Severity is rated on a 7-point scale (with 1 and 7 representing the symptom extremes of ‘normal/not at all ill’ and ‘among the most extreme cases’). The CGI-ADHD-S was evaluated in both study phases.

Chinese version of the Swanson, Nolan, and Pelham, version IV scale parent form

The 26 items of the SNAP-IV are rated on a 4-point Likert scale, with scores of 0–3 representing: ‘not at all’, ‘just a little’, ‘quite a bit’ and ‘very much’. The SNAP-IV consists of inattention, hyperactivity/impulsivity, and oppositionality corresponding to the core symptoms of DSM-IV ADHD and oppositional defiant disorder, respectively. The norm and psychometric properties of the Chinese version of the SNAP-IV (SNAP-IV-C) have been established in Taiwan.13

Safety measures

Safety measures assessed by the investigators were decreased appetite, dizziness/headache, gastrointestinal upset, poor sleep quality and other side-effects.

Data analysis

SAS 9.1 software (SAS Institute Inc, Cary, NC, USA) was used for data analysis, with the alpha level set at 0.05. For the comparisons of OROS-MPH switch and adherence status, the descriptive results were presented as frequencies and percentages for the categorical variables and as mean (standard deviation) for continuous variables. Univariate analysis using a logistic regression model and anova for categorical and continuous variables, respectively, were used for comparison of the switched (n = 137) and un-switched (n = 103) groups, and good (n = 99) and poor (n = 38) OROS-MPH adherents. A multiple logistical regression model was applied to identify the most significant correlates for OROS MPH switch among poor adherents and OROS-MPH adherence using backward model selection.

ancova, controlling for subjects' sex and age, was used to compare the t-scores for the SNAP-IV-C subscales and the mean CGI-ADHD-S scores assessed in Phase I between the three ADHD subtypes among the poor adherents; and to compare symptom reduction from Phase I (IR-MPH) to Phase II (OROS-MPH) in the SNAP-IV and CGI-ADHD-S between children with superior parent–child interaction, general classroom behaviors, and overall academic performance and their counterparts.

A linear mixed model with both fixed and random effects was employed to test intra-subject, inter-phase differences of repeated measures of the SNAP-IV-C and CGI-ADHD-S (IR-MPH vs OROS-MPH) for the three ADHD subtypes. We also tested the two-way interactions between the two assessments at Phase I and Phase II of symptoms and ADHD subtype, age and gender.


Sample description

Table 1 presents the comparison of demographics, variables related to diagnosis, treatment, and symptom severity between patients switched to OROS-MPH and patients maintained on IR-MPH, and between good and poor adherents to OROS-MPH. Of 240 poor adherents, 226 patients (95.0%) were of normal intelligence and 115 (47.9%) had a positive family history of ADHD, mostly involving siblings. Subtype diagnosis distribution was: ADHD-C (146, 60.9%); ADHD-I (74, 30.8%); and, ADHD-HI (20, 8.3%). The mean age at confirmed ADHD diagnosis was 7.4 ± 2.7 years (Table 1). The mean daily IR-MPH dosage was 20.2 ± 9.2 mg, and the mean duration of medication was 15.6 ± 18.5 weeks. More than half of the subjects (52.4%) were taking the IR form twice daily, with 26.6% using thrice daily dosing. The mean t-scores for the three SNAP-C subscales were: inattention, 66.1; hyperactivity, 65.5; and oppositional behavior, 62.0. The mean CGI-ADHD-S score was 3.3 ± 1.1 (i.e. mildly to moderately ill; Table 1).

Table 1.  Sample description and determinants for switching medication from IR-MPH to OROS-MPH
VariablesPoor adherence to IR-MPHIndividual analysis
No switch (n = 103)Switch to OROS–MPH
Good adherence (n = 99)Poor adherence (n = 38)Switch/no switchPoor/good adherence
OR95% CIOR95% CI
  • *

    P < 0.05,

  • **

    P < 0.01,

  • ***

    P < 0.0001.

  • ADHD, attention-deficit/hyperactivity disorder; CGI-ADHD-S, Clinical Global Impression-ADHD-Symptom Severity; CI, confidence interval; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders – 4th edition; IQ, intelligence quotient; IR-MPH, immediate-released methylphenidate; NTUH, National Taiwan University Hospital; OR, odds ratio; OROS-MPH, osmotic release oral system methylphenidate; SD, standard deviation; SNAP-IV, the Swanson, Nolan, and Pelham, version IV scale.

Sex, n (%)       
 Male86 (83.5)80 (80.8)30 (78.9)0.810.44–1.701.070.41–2.83
 Female17 (16.5)19 (19.2)8 (21.1)1.001.00
Age, mean (SD)10.3 (2.8)10.2 (2.5)11.2 (2.5)F(1,218) = 0.11P = 0.746F(1,122) = 3.60P = 0.060
Body mass index, mean (SD)17.8 (3.2)18.3 (4.1)19.4 (4.8)F(1,218) = 1.85P = 0.175F(1,122) = 1.38P = 0.243
ADHD history, n (%)       
 Yes77 (74.8)27 (27.3)11 (29.0)1.140.64–2.030.950.40–2.11
 No26 (25.2)72 (72.7)27 (71.0)1.001.00
Father education, n (%)       
 Senior high and below54 (52.4)50 (50.5)16 (42.1)1.190.71–1.980.710.34–1.52
 College and above49 (47.6)49 (49.5)22 (57.9)1.001.00
Mother education, n (%)       
 Senior high and below64 (62.1)56 (56.6)24 (63.2)1.170.69–1.971.320.61–2.84
 College and above39 (37.9)43 (43.4)14 (36.8)1.001.00
DSM-IV ADHD subtype, n (%)       
 Inattentive32 (31.1)29 (29.3)13 (34.2)1.020.58–1.800.870.39–1.96
 Hyperactive7 (6.8)11 (11.1)2 (5.3)1.450.55–3.842.140.44–10.45
 Combined64 (62.1)59 (59.6)23 (60.5)1.001.00
MPH dose/day, mean (SD)18.6 (9.4)21.5 (8.7)21.3 (9.1)F(1,237) = 5.86P = 0.016F(1,135) = 0.01P = 0.909
MPH dose/kg/day, mean (SD)0.50 (0.24)0.64 (0.29)0.53 (0.25)F(1,218) = 7.83P = 0.005F(1,122) = 3.36P = 0.069
Duration (week), mean (SD)18.0 (19.9)15.1 (18.7)9.6 (10.7)F(1,214) = 3.11P = 0.079F(1,116) = 2.43P = 0.122
Frequency of administration, n (%)       
 Twice daily47 (51.6)46 (54.1)17 (50.0)1.940.95–3.942.170.73–6.40
 Thrice daily18 (19.8)29 (34.1)9 (26.5)3.051.34–6.94*1.760.28–10.97
 Only morning dose26 (28.6)10 (11.8)8 (23.5)1.001.00
Age of onset, mean (SD)7.4 (3.0)7.1 (2.6)8.0 (2.5)F(1,221) = 0.02P = 0.879F(1,122) = 3.46P = 0.065
IQ, n (%)       
 Mental retarded (IQ < 70)7 (6.9)3 (3.1)2 (5.3)1.930.60–6.271.760.27–10.97
 Normal (IQ ≧ 70)95 (93.1)95 (96.9)36 (94.7)1.001.00
Site, n (%)       
 NTUH5 (4.9)23 (23.3)4 (10.5)2.971.05–8.44**2.090.63–6.99
 Other medical centers65 (63.1)32 (32.3)18 (47.4)0.420.24–0.74***0.650.29–1.46
 Non-medical centers33 (32.0)44 (44.4)16 (42.1)1.001.00
SNAP, mean of t-score (SD)       
 Inattention64.1 (12.3)67.9 (11.6)66.7 (12.2)F(1,238) = 4.89P = 0.028F(1,135) = 0.29P = 0.592
 Hyperactivity65.6 (15.1)66.1 (16.2)63.6 (14.9)F(1,238) = 0.01P = 0.935F(1,135) = 0.72P = 0.399
 Oppositional61.2 (11.8)63.2 (14.4)61.0 (14.1)F(1,238) = 0.64P = 0.425F(1,135) = 0.64P = 0.426
CGI-ADHD-S, mean (SD)3.2 (1.2)3.5 (1.1)3.2 (1.1)F(1,220) = 2.24P = 0.136F(1,125) = 1.53P = 0.218

Determinants for OROS-MPH switch

Univariate analysis revealed switching to OROS-MPH among 240 poor adherents was more likely with: (i) treatment at a national medical center or non-medical centers; (ii) higher mean daily dose; (iii) higher mean daily dose per kilogram of body weight; (iv) multi-dose IR-MPH administration; and (v) more severe inattention symptoms (Table 1). Multiple logistic regression analysis revealed that the most predictive variables for switching medication were greater inattention (P = 0.007) and shorter IR-MPH treatment (P = 0.028).

After switching from IR-MPH to OROS-MPH, 99 children (72.3%) were assessed as good adherents; however, adherence remained poor in 38 (27.7%). No predictive relationship was demonstrated for any variable in terms of OROS-MPH adherence. The average dose of OROS MPH was 24.9 ± 8.8 for 124 subjects who completed the Phase II assessments. The rates of side-effects were similar for the two variants, with decreased appetite as the most prevalent side-effect for both (21.1/21.8% for IR-MPH/OROS-MPH).

Symptom severity and ADHD subtype

The t-scores for hyperactivity and oppositional symptoms were significantly higher in children with ADHD-C and ADHD-HI relative to children with ADHD-I (Table 2); however, there was no such difference comparing the first two groups. Furthermore, no significant differences were demonstrated comparing inattention and CGI-ADHD-S score between the three subtypes.

Table 2.  Mean scores of the SNAP-IV subscales for three ADHD subtype groups
 ADHD subtypeStatistics
Combined (n = 146)Inattentive (n = 74)Hyperactive (n = 20)F valueP-valueComparison
Mean (SD)Mean (SD)Mean (SD)
  1. ADHD, attention-deficit/hyperactivity disorder; CGI-ADHD-S, Clinical Global Impression-ADHD-Symptom Severity; Com, combined; Hyp, hyperactive; Ina, inattentive; SD, standard deviation; SNAP-IV, the Swanson, Nolan, and Pelham, version IV scale.

SNAP subscales, t-score      
 Inattention67.0 (11.6)63.4 (12.5)68.9 (12.3)F(2,237) = 2.88P = 0.0582 
 Hyperactivity69.0 (13.8)56.0 (14.7)74.8 (14.0)F(2,237) = 25.87P < 0.0001Hyp > Ina, Com > Ina
 Oppositional64.6 (12.6)55.7 (12.5)66.0 (13.6)F(2,237) = 13.33P < 0.0001Hyp > Ina, Com > Ina
CGI-ADHD-S3.3 (1.2)3.3 (1.0)3.4 (1.3)F(2,219) = 0.16P = 0.853 

IR-MPH and OROS-MPH symptom-change comparison

At the study endpoint, 124 children had completed the Phase II assessment and were used to compare the treatment effect for the two MPH variants with respect to symptom change. Significant decreases in CGI score were demonstrated for all three ADHD subtypes. Moreover, children with ADHD-C had significantly lower scores on the three SNAP-IV subscales, with children with ADHD-I scoring significantly lower in inattention after switching to OROS-MPH relative to pre-switched treatment with IR-MPH (Table 3).

Table 3.  Change of symptoms between IR-MPH and OROS-MPH stratified by ADHD subtype
IR-MPH (n = 37)OROS-MPH (n = 37)StatisticsIR-MPH (n = 13)OROS-MPH (n = 13)StatisticsIR-MPH (n = 74)OROS-MPH (n = 74)Statistics
Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)
  1. ADHD, attention-deficit/hyperactivity disorder; CGI-ADHD-S, Clinical Global Impression-ADHD-Symptom Severity; IR-MPH, immediate-released methylphenidate; NS, not significant; OROS-MPH, osmotic release oral system methylphenidate; SD, standard deviation; SNAP-IV, the Swanson, Nolan, and Pelham, version IV scale.

SNAP-IV, t-score          
 Inattention64.6 (12.1)57.4 (11.0)F(1,33) = 13.78
P = 0.0008
68.9 (14.2)64.3 (11.1)F(1,9) = 1.20
P = 0.302
68.6 (11.1)60.8 (10.9)F(1,70) = 26.24
 Hyperactivity54.4 (14.4)54.6 (13.8)F(1,33) = 0.08
P = 0.780
75.2 (16.4)69.8 (14.3)F(1,9) = 0.96
P = 0.352
69.0 (14.1)61.8 (13.5)F(1,70) = 18.87
F(2,107) = 11.1
 Oppositional55.2 (12.9)52.7 (11.6)F(1,33) = 1.08
P = 0.307
66.8 (15.0)63.1 (15.5)F(1,9) = 2.26
P = 0.167
65.9 (13.8)58.1 (12.6)F(1,70) = 26.36
F(2,107) = 7.59
P = 0.0008
CGI-ADHD-S3.4 (1.0)2.9 (0.8)F(1,36) = 15.45
P = 0.0004
3.8 (1.3)2.8 (1.1)F(1,9) = 7.00
P = 0.027
3.3 (1.1)2.7 (0.9)F(1,70) = 33.61

Hyperactivity and oppositional symptoms varied across the three subtypes, with significant treatment effect observed only in ADHD-C (Table 3). There were no interactions between age, gender, ADHD subtype, and MPH variants on the symptom changes of inattention and CGI scores.

Degree of symptom reduction by efficacy-change group

The 124 children who completed Phase II assessments were further divided into two groups according to measures of functional changes in parent–child interaction, overall classroom behavior, and overall academic performance: (i) improved (much better and better); and (ii) not improved (no change and worse) (Table 4). There were 66 (53.2%) children having improvement in all, 21 (16.9%) in two, and 12 (9.7%) in only one of the three measures. Twenty-five (20.2%) were not improved in any of the three measures.

Table 4.  Symptom severity and changes by the presence and absence of improvement for the parent–child interaction, overall classroom behavior, and overall academic performance
Mean (SD)Parent–child interactionOverall classroom behaviorOverall academic performance
Improved (n = 86)Not improved (n = 38)P-valueImproved (n = 91)Not improved (n = 33)P-valueImproved (n = 75)Not improved (n = 49)P-value
  1. CGI-ADHD-S, Clinical Global Impression-attention-deficit/hyperactivity disorder-Symptom Severity; SD, standard deviation; SNAP-IV, the Swanson, Nolan, and Pelham, version IV scale.

SNAP-IV, t-score-phase I         
 Inattention67.7 (11.7)66.8 (12.1)0.68667.6 (11.2)67.0 (13.4)0.79367.1 (11.0)68.0 (12.9)0.670
 Hyperactivity65.6 (15.8)64.6 (16.8)0.74465.0 (15.8)66.0 (17.0)0.76264.5 (15.4)66.4 (17.1)0.516
 Oppositional63.3 (13.6)61.6 (16.3)0.54863.3 (14.1)61.3 (15.6)0.50761.9 (13.8)64.2 (15.5)0.376
CGI-ADHD-S3.4 (1.1)3.4 (1.0)0.8663.4 (1.1)3.4 (1.1)0.8313.4 (1.1)3.4 (1.1)0.864
SNAP-IV, t-score-phase II         
 Inattention58.3 (10.2)64.2 (12.0)0.01058.4 (10.4)64.5 (11.8)0.01056.9 (9.9)64.7 (11.2)0.000
 Hyperactivity59.1 (13.6)62.8 (15.8)0.22359.0 (13.5)63.4 (15.9)0.15358.4 (13.4)63.0 (15.3)0.103
 Oppositional55.2 (11.3)60.7 (15.5)0.04156.0 (12.1)59.3 (14.8)0.22855.7 (12.3)58.7 (13.8)0.240
CGI-ADHD-S2.7 (0.9)2.9 (0.8)0.1142.7 (0.8)2.9 (0.9)0.1742.6 (0.9)2.9 (0.8)0.065
Differences between phase I and II         
 Inattention−9.4 (11.7)−2.6 (11.8)0.010−9.2 (11.6)−2.5 (12.2)0.016−10.2 (11.4)−3.3 (12.0)0.007
 Hyperactivity−6.5 (13.4)−1.8 (11.6)0.251−6.0 (12.8)−2.6 (12.9)0.219−6.1 (12.4)−3.4 (13.7)0.780
 Oppositional−8.1 (10.5)−0.9 (11.0)0.000−7.3 (10.5)−2.0 (12.4)0.021−6.2 (9.9)−5.5 (13.2)0.745
CGI-ADHD-S−0.7 (0.9)−0.5 (0.8)0.015−0.7 (0.9)−0.5 (0.8)0.029−0.8 (0.8)−0.5 (1.0)0.016

There was no symptom severity difference in Phase I between those with improved efficacy and those without. However, the children with improved parent–child interaction, overall classroom behavior and overall academic performance scored significantly lower for inattention in Phase II compared to their counterparts. Children with improved parent–child interaction also scored significantly lower for oppositional symptoms in Phase II relative to those without improved efficacy.

Regarding the effect of degree of symptom reduction on efficacy change in Phase II, children with improvement in any of the three functional measures had significantly greater symptom reduction as demonstrated by the CGI-ADHD-S score and inattention symptoms than their counterparts (Table 4). Moreover, improved parent–child interaction and overall classroom behavior was significantly associated with reduction in oppositional symptoms.


As the first study on the effectiveness of methylphenidate across the three ADHD subtypes in Asian population, our major findings are that the ADHD symptom improvement between the two variants of MPH appeared to vary with the ADHD subtype, with favorable treatment response more evident in ADHD-C, followed by ADHD-I; and that reduction in inattention is associated with improved parent–child interaction, overall classroom behavior, and overall academic performance.

Determinants for OROS-MPH switch

Our finding implies a relationship between the switch to the extended-release formulation and symptom severity and/or between once-daily administration and adherence.5 Additionally, more severe inattention predicting OROS-MPH switch reflects that forgetfulness, one of the DSM-IV inattention symptoms, raised the concern regarding treatment adherence at school.5 Moreover, inattention, strongly related to academic performance, is generally considered the main parental concern in the context of Taiwanese families.3 Ethnic Chinese families in Taiwan are greatly influenced by Chinese culture and its Confucian tradition emphasizing children's academic achievement.3 Therefore, improving the child's attention in order to increase academic performance would be the main ADHD-treatment focus and drive the switch to OROS-MPH.8

Surprisingly, we found that poor adherents to IR-MPH were less likely to be switched to OROS-MPH in private medical centers. The Bureau of Taiwan National Health Insurance has contracted with almost all the medical institutions in Taiwan, covering 98% of the national population since 1995.14 Physicians at private medical centers hesitated to prescribe expensive medication, like OROS-MPH, under the limits of the global budget, which was executed in 2005 and 2006.

Our findings regarding the efficacy of OROS-MPH as compared to IR-MPH are similar to those of a randomized clinical trial in Taiwan,8 with comparable side-effects, less symptom severity and better parent–child interaction, classroom behavior and academic performance demonstrated in both studies.8 None of the variables assessed in Phase I predicted OROS-MPH adherence in Phase II. As this once-daily, extended-release form was designed to improve medication adherence, investigation of the reasons for the residual poor compliance will help subsequent development of pharmaceutical agents.

Symptom Severity and ADHD subtype

Our findings support more severe hyperactivity and oppositional symptoms but not inattention in ADHD-C and ADHD-HI compared to ADHD-I.11 However, our failure to demonstrate significant symptom differences between ADHD-C and ADHD-HI is not consistent with other studies15 that show increased inattention and hyperactivity in ADHD-C than ADHD-HI.

Previous findings of fewer inattention problems in the ADHD-HI subtype than the ADHD-I subtype15,16 are not supported by this study. In line with Baeyens et al.,17 we were unable to demonstrate significant differences in inattention symptoms comparing the three ADHD subtypes. Possible explanations for no greater oppositional symptoms18 and less inattention16 in ADHD-HI are that parents of children with ADHD-HI may over-report overall symptoms, or that the higher inattention ratings may reflect the focus on academic performance in the context of the Chinese family. This discrepancy may reflect some degree of disagreement between parent's reports on the SNAP-IV-C and the clinician's diagnosis based on information from multiple sources.

Treatment response and ADHD subtype

Treatment response is among the measures used to validate the ADHD subtype, mainly ADHD-C,18 despite relatively few studies on this topic.10,11 While significant reduction in inattention and hyperactivity-impulsivity symptoms has been observed in ADHD-C, it cannot be assumed that ADHD-I or ADHD-HI would obtain the same benefit.19 As one of first few to examine MPH treatment efficacy between ADHD subtypes,10,11 our findings lend evidence to support the proposition that MPH reduces inattention comparably for both ADHD-I and ADHD-C, whereas improvement in hyperactivity and oppositionality are more evident in ADHD-C.11 In contrast, we found that the effectiveness of OROS-MPH treatment after switching from IR-MPH is significant not only for the ADHD-I subtype10 but also for ADHD-C.

Overall, symptoms improved significantly after switching to OROS-MPH, and this was most evident in the ADHD-C group. No significant improvement for ADHD-HI may be explained by the lack of statistical power to detect the difference due to the small sample size. As ADHD-HI, which is a newly introduced DSM-IV ADHD subtype, comprises nearly one-fourth of ADHD diagnoses,18 and no study has specifically investigated treatment response for this subtype, conducting a treatment study of ADHD-HI appears warranted.

ADHD symptom changes and family/school measures

Pharmacological studies on ADHD generally have not further examined the relationship between ADHD symptom reduction and improvement in school/family measures.20 It appears reasonable to hypothesize that inattention is related to academic performance,20 which is supported by our finding of an association between greater reduction in inattention symptoms and better academic performance. However, our results do not support a relationship between hyperactivity-impulsivity and academic performance.21 Our study, similar to previous studies,22 demonstrated an association between greater reduction in inattention22 and in oppositional symptoms,22 but not hyperactivity,22 and a better parent–child relationship. Similarly, we reported that greater reductions in inattention21 and oppositional symptoms,23 but not hyperactivity, predicted better classroom behaviors. Unexpectedly and in contrast to other studies,21 we did not find that reduced hyperactivity has a significant effect on parent–child interaction,8 classroom behaviors,23 or academic performance.21 Taken together, we conclude that inattention influences not only academic performance but also other classroom behaviors and the parent–child relationship, and that the latter two are also influenced by oppositional symptoms. Lack of evidence supporting relationships between hyperactivity changes and any family/school efficacy measures needs further investigation.


The results of this study must be interpreted in the context of methodological limitations. First, the finding regarding determinants for the OROS-MPH switch might not be generalizable to other national populations because Taiwan's BNHI14 limits the right of clinicians to prescribe OROS-MPH for financial and policy reasons and MPH was the only approved medication for ADHD in Taiwan during the study period.8,9 Second, this is an open-label prospective observation study. Lack of blindness may decrease the persuasiveness of this study due to the potential bias by the investigators. Third, like most treatment studies, the measures were based on child and parent/caregiver reports and investigator assessments without teacher reports of classroom behaviors and academic performance. However, through interviewing the participants and parents, the investigators can obtain reliable information about classroom behaviors and judge the academic performance of ADHD patients according to reading the daily exam scores shown on the communication book.


Our findings imply that ADHD children with poor adherence to IR-MPH treatment will benefit from switching to OROS-MPH; identifying the reasons for non-adherence in those children who cannot be switched to OROS-MPH is needed to achieve optimal treatment outcome; and children with different ADHD subtypes appear to show varying response to MPH.


This work was supported by Jansessen-Cilag, Taiwan (Protocol ID: CCT-TWN-MA3) and the National Health Research Institute (NHRI-EX95-9407PC), Taiwan. We would like to express our thanks to all the child psychiatrists participating in this study.