Withdrawing off-label antipsychotics in people with intellectual disabilities: why does it fail?

In people with intellectual disability (ID), antipsychotics (AP) are often used
offlabel,
mostly for challenging behaviour, such as aggression or agitation. Studies in
both community (17-
27%) and inpatient (32-56%) settings show that the prevalence of antipsychotic
drug use in people with ID
is high [1-4]. AP are often used for a long period (> 10 years in 78 %) [1].
This is remarkable, since, in case
of challenging behaviour, there is no evidence for efficacy: a Cochrane review
concluded that there was no
evidence for the efficacy of AP for behavioural issues in ID [5] and Tyrer and
colleagues demonstrated that
risperidone, haloperidol and placebo all decreased aggression in patients with
ID after 4 weeks, with the
placebo group showing the greatest improvement [6]. Although evidence for any
effect of long-term use of
AP for challenging behaviour in ID is lacking, there is convincing evidence
that side effects, such as
diabetes, metabolic syndrome, extrapyramidal side effects, decreased threshold
for seizures, emotional
blunting and hyperprolactinemia, are - also in this population - common and
clinically relevant [7-10]. This
is particularly important because this is a vulnerable population with
preexisting other risk factors for
diabetes, metabolic syndrome, movement disorders, epilepsy and osteoporosis.
These side effects are - at
least partly - reversible after discontinuation of the AP [8, 11]. Despite
concerns about side effects and
dubious efficacy, successful discontinuation is not self-evident: controlled
studies focusing on long-term
complete discontinuation of off-label AP drugs for behavioural issues in ID
achieved full withdrawal in 33-
44% of patients [12, 13]. Strikingly, in these studies, no deterioration of
behaviour was found during and
after complete or incomplete withdrawal. In order to decrease the prevalence of
AP drug use in patients
with ID, it is important to understand why discontinuation is not successful in
the majority of patients. We
postulate 3 possible mechanisms:
1. The influence of the subjective interpretation of behavioural symptoms by
caregivers and family:
perceptions may be influenced by fear of worsening of behaviour after drug
reduction [14]. Subsequently,
attitude and apprehension may influence the behaviour of the patient with ID
due to interaction, which may
in turn contribute to drug reduction outcome. Successful withdrawal depends at
least in part on staff and
environmental characteristics [12]. In line with these findings, De Kuijper et
al demonstrated that, although
there was no worsening of behaviour during AP withdrawal as measured with a
standardized questionnaire
completed by caregivers, caregivers subjectively reported worsening of
behaviour during withdrawal,
possibly related to subjective beliefs [13]. However, this may also demonstrate
that caregivers detect more
subtle changes in behaviour, for which the questionnaire was not sufficiently
sensitive.
2. It cannot be excluded that some patients with ID and behavioural problems
might benefit from AP
treatment. Measures used to assess aggressive behaviour, which is an extremely
heterogeneous symptom,
may not always be sensitive enough to detect subtle treatment effects.
Furthermore AP may be effective for
psychiatric illnesses that remained previously undiagnosed, possibly due to a
lack of diagnostic procedures
and instruments. In addition, AP may positively influence (unrecognized)
disturbances of the circadian
rhythm (sleep-wake cycle). Circadian rhythm disturbances are relatively common
in people with ID [15] and,
since alterations in circadian rhythm are a side effect of AP, AP treatment as
well as AP discontinuation
may affect circadian rhythm [16, 17].
3. When AP are withdrawn after long-term treatment, brain Dopamine2 receptors
must downregulate. This
takes an unknown amount of time (probably depending on individual differences,
and on the dose and
duration of use) [18]. When the withdrawal rate exceeds the downregulating
process, withdrawal symptoms
may occur, such as agitation, mania, akathisia, and withdrawal-dyskinesia, due
to an excess of input from
dopamine. Also serotonergic, histaminergic, muscarinergic or adrenergic
(depending on the type of AP)
withdrawal or rebound symptoms may occur, such as decreased appetite, weight
loss, anxiety, sleep
problems, agitation, confusion, flu-like symptoms and psychosis [16]. These
symptoms may be
misinterpreted as recurrence of the original challenging behaviour, resulting
in a request to reinstitute the
AP.
There is general consensus that reluctance regarding off-label prescribing AP
for behavioural issues is
appropriate and that discontinuation of long-term off-label AP should be
considered in all patients. There is
an important need for evidence on the above mechanisms in order to develop
guidelines, not only for
diagnosis and treatment of behavioural problems in patients with ID, but also
for clear indications for
ongoing off-label use or discontinuation of AP, as stated by Tyrer and
colleagues in 2014 [19] :
*Drug treatment of challenging behaviour in people with intellectual disability
should no longer be on the
sidelines of evidence based medicine. If we are going to achieve parity of
esteem for people with mental
illness, we can no longer tolerate our ignorance on this subject. Quite apart
from the deficiencies in
evidence allowing dogma and opinion to rule, the cost of prescribing these
drugs is enormous. If they truly
are unnecessary, clinicians, pharmacists, service managers, and those who fund
services for people with
intellectual disability need to know, and soon*.

The study is designed to investigate why withdrawal of off-label antipsychotic
drugs for behavioral problems in people with intellectual disability often
fails by comparing two blinded groups (withdrawa group versus control group).
This has led to the following primary and secondary objectives.

Primary objective:
- Compare the number of dropouts in both groups (withdrawal versus control
group). If these groups have a similar dropout this will supports hypothesis 1.

Secondary objectives:
- Sleep problems, behavioral and psychiatric problems will be compared in both
groups. If the antipsychotic withdrawal group have more behavioral problems,
more not previously diagnosed psychiatric disorders or sleep problems,
antipsychotics may have effect and this will be support hypothesis 2.
- To investigate and compare withdrawal symptoms and side-effects in both
groups during withdrawal. If withdrawal symptoms more occur in the withdrawal
group (compared with the control group) hypothesis 3 will supported.

Other secondary objectives:
- To investigate patient characteristics that predict outcome of the withdrawal
study.
- To evaluate whether in measuring drug induced movement disorders the use of
electronic devices is superior to clinical rating scales.
- CYP polymorphism and serum levels of risperidone or pipamperone

Dubbelblind placebo controlled RCT

Withdrawal of antipsychotics (risperidone or pipamperone)

The study question can only be studied in this group, because people with ID
have specific characteristics. Participation requires effort from the patients
for the measurements. In this research there the risks for the participants are
similar as care-as-usual.
At the moment is withdrawing antipsychotics (AP) at people with Intellectual
Disability (ID) and challenging behavior (off-label use of AP) care- as-usal.
Therefore, there will be similar risks as the care-of-usual. The differences
are that we will observe them more and when there are problems probably we can
differ the therapy because of the addition of different examinations and
diagnostics for the underlying problem (eg. sleepproblems, withdrawal symptoms,
psychiatric disorders). Participants will have measurements at baseline, week
2, 4, 5, 6, 8, 10, 12, 13, 14, 16, and at follow-up (week 22 and 40). A
subgroup of the participants will have repeated measurements of movement
disorders at baseline.