Skip to main content

5. Pharmacological interventions

5.3 Monitoring treatments

See recommendations

Clinical Questions

1. Should ‘drug holidays’ from pharmacological treatment for ADHD be recommended and if so when?
2. What is the most clinically effective subsequent sequence of pharmacological/non-pharmacological treatment for people with ADHD when the initial treatment is ineffective, inadequate or treatment is not tolerated?

Clinical Questions

1. How should treatment effectiveness be monitored and supported?
2. How should adequacy of treatment response be assessed?
3. What are the indicators of remission and when should treatments be stopped?

Clinical practice gaps, uncertainties and need for guidance

There are currently inconsistencies in the timing and approach to monitoring treatment response and adverse effects of medications for ADHD, and in approaches to decision-making about stopping treatment, according to anecdotal reports. While an individual, person-centred approach is needed when prescribing and monitoring medication, consistent parameters are needed. A ‘drug holiday’ is an agreed cessation of medication for a period of time and is occasionally used to ‘catch-up’ on growth in children and adolescents. Guidance is needed on whether a drug holiday is helpful and safety issues to consider when starting and stopping medication.

Summary of evidence review

Subsequent sequence of pharmacological/non-pharmacological treatment when the initial treatment is ineffective, inadequate or treatment is not tolerated.

No new evidence was identified. NICE identified 6 RCTs in 9 publications to address this question in children and adolescents, and one RCT in adults; and none were identified for children under 5 years of age. Some comparison trials that were reported by NICE were deemed not clinically relevant (methylphenidate versus placebo to augment atomoxetine treatment). Factors to be considered when monitoring treatment, assessing treatment response, indications of remission and stopping treatment were addressed qualitatively by (NICE, 2018) (see section 5.1).

Lisdexamfetamine dimesylate versus placebo where previous methylphenidate treatment was stopped
No new evidence was found. NICE identified one very low-quality study which found a clinical benefit of lisdexamfetamine dimesylate, compared with placebo, for combined ADHD total, inattention and hyperactivity symptoms and Clinical Global Impression scale. No clinical difference was found for adverse events leading to hospitalisation/death/disability.

Lisdexamfetamine dimesylate versus atomoxetine where previous methylphenidate treatment was stopped
No new evidence was found. NICE identified one low-quality study which found a clinical benefit of lisdexamfetamine, compared with atomoxetine, for investigator-rated ADHD total, hyperactivity and inattention symptoms. No clinical difference for discontinuation of treatment due to adverse events or adverse events leading to hospitalisation/death/disability was found in one low-quality study, and other symptoms, and severity on the Clinical Global Impression scale.

Guanfacine in the morning or evening versus placebo augmented on top of previous stimulant treatment
No new evidence was found. NICE identified one low-quality study which found a clinical benefit of guanfacine, compared with placebo, for Clinical Global Impression scale. There was a clinical harm of methylphenidate in adverse events leading to hospitalisation/death/disability in one very low-quality study, and no clinical difference for discontinuation due to adverse events.

Clonidine versus placebo where previous stimulant treatment continued
No new evidence was found. NICE identified no clinical difference in investigator-rated ADHD total, inattention and hyperactivity symptoms and no clinical difference in discontinuing treatment due to adverse events in one very low-quality study.

Risperidone and parent training versus placebo where previous methylphenidate treatment was continued
No new evidence was found. NICE found in children and adolescents a clinical benefit of risperidone for parent-rated and teacher-rated ADHD total symptoms (one study of moderate to low quality), parent- and teacher-rated ADHD inattention symptoms (one study of moderate quality), oppositional defiant disorder (parent-rated, one study of low quality). In children and adolescents there was clinical harm of risperidone for teacher- and parent-rated ADHD hyperactivity symptoms (one study low to moderate quality). There was no clinical difference for ADHD inattention symptoms (one study of low quality) and teacher-rated and parent-rated other symptoms (2 studies of moderate to very low quality).

Guanfacine in the morning or evening versus placebo augmented on top of previous stimulant treatment
No new evidence was found. NICE identified one study of very low to low quality and found no clinical difference for ADHD total, inattention and hyperactivity symptoms, Clinical Global Impression scale, and adverse events leading to hospitalisation/death/disabilities.

Summary of evidence review – drug holidays

An updated evidence review was conducted with no evidence found. NICE identified one study (Martins et al., 2004), a blinded RCT conducted with children that compared the clinical effects of stopping pharmacological treatment at weekends over a 4-week period. The study was rated as high risk of bias and very low-certainty evidence. The study reported only parent-reported benefits for weekend breaks from methylphenidate use. No difference in ADHD symptoms was found between the treatment (drug holidays) and control group (continuous treatment) based on parent and teacher ratings. Reduced insomnia was found in the drug holiday group with a trend toward less interference on appetite.

Evidence to recommendation statement

Monitoring side effects and drug holidays
Evidence shows the clinically important differences in sleep disturbance, decreased appetite and weight changes in people taking ADHD medication (summarised in section 5.2). Due to concerns about decreased appetite and weight change, the GDG advised that weight should be checked every 3 months initially in children and 6 months thereafter and in children and adults. Young children should be monitored more frequently. There is a lack of research on the impacts of drug holidays. Evidence from the included study indicated no significant difference in symptoms and improvements in sleep and appetite. The NICE 2018 recommendations were therefore adapted to the Australian context, including the option of considering a planned break in treatment if growth concerns were indicated.

Sequencing of treatments
For sequencing of medication treatments, most outcomes were graded as low or very low quality and risk of bias was high to very high, and serious imprecision for 90% of the outcomes. The evidence for sequencing was of lower quality than the effectiveness trials and sequencing trails predominantly compared adding/substituting with a new medication and not adding/substituting with placebo. Therefore, the GDG broadly based their recommendations around the sequence of medication (see section 5.2) on the body of efficacy evidence in the general pharmacological efficacy review.

Recommendations

Clinical considerations for implementation of the recommendations

Availability of appointments for adequate follow-up with medical practitioners should ideally not be a barrier to monitoring if other clinicians are available to assist with providing relevant information (for example, a community nurse or primary care nurse), and if the person with ADHD or their caregivers are also engaged in structured monitoring. Additional medical appointments will need to be available for individuals who require medical monitoring.

There is currently a shortfall of clinicians to provide some of these services. Workforce development is required to increase the number of clinicians with expertise in ADHD and ensure that health inequity impacts are minimised by providing access through public services. The recommendations made here are generally well established in clinical practice and are therefore likely to be acceptable to stakeholders.

See Technical Report, sections 6.5, 6.6, 8.6 and 10.7 for further details

Next 5.4 Adherence to medication treatment

View Recommendations