Individuals with intellectual disabilities (ID) exhibit neurodevelopmental deficits that impair both intellectual functioning and adaptive behavior. ID affects approximately 1% of the overall population (1), making it a relatively common neurodevelopmental disorder. The principal etiologies of ID include genetic conditions (such as Down syndrome and Fragile X Syndrome, FXS), prenatal factors (e.g. fetal alcohol syndrome or infections during pregnancy), perinatal complications (such as oxygen deprivation during birth), and postnatal causes (including traumatic brain injury, severe infections, and exposure to toxins). Additionally, environmental factors such as malnutrition, lack of stimulation, and social deprivation can contribute to the development of ID (2).

The impact of ID reaches beyond those directly affected, as it presents significant challenges for their families and communities as well. This results in considerable socioeconomic consequences, such as higher healthcare costs, special education needs, and additional support services (3).

Behaviors that challenge (BC), including irritability, aggression, self-injury, and other disruptive manifestations, pose significant difficulties for individuals with ID (4). These behaviors impact the quality of life of the individuals and their caregivers, necessitating effective intervention strategies. Pharmacological interventions, particularly atypical antipsychotics (5, 6), are routinely used in clinical practice to manage these behaviors (7, 8).

This practice is often in contrast with international (9) and national recommendations, such as National Institute for Health and Care Excellence (NICE) guidelines (10, 11) and 2018 Canadian consensus guidelines (12). These guidelines recommend that antipsychotics should not be used to treat BC in people with ID unless a comorbid psychotic disorder is present or unless other non-pharmacological approaches have been attempted and proven ineffective (13). As well as not being very effective on average in preventing CB, the use of those medications in ID population often involves potentially inappropriate prescriptions (14, 15), leading to polymedication (16) drug-drug interaction issues and adverse drug reactions (17). Despite these recommendations, antipsychotics continue to be frequently prescribed for BC in this population, often outside their officially approved indications. In fact, a substantial proportion of these prescriptions are considered off label. Importantly, “off-label” use in this context can refer to several distinct scenarios — including prescriptions outside approved age ranges, for diagnoses not specifically indicated, or for behavioral symptoms in the absence of a formal psychiatric disorder. Each of these carries different safety and efficacy profiles, which may significantly influence clinical outcomes and raise ethical concerns. Recognizing this heterogeneity is crucial to fully grasp the complexities of antipsychotic use in individuals with ID.

In clinical practice, antipsychotics seem to be partially effective for specific BC such as irritability and aggression (18), but they do not appear as effective for other BC such as self-injurious behaviors (SIBs), repetitive behaviors, and hyperactivity (19, 20).

Due to the significant discrepancy between guideline recommendations and common clinical practices despite the low level of effectiveness, we conducted a literature review to seek robust evidence on the appropriate use of antipsychotics for managing BC in individuals with ID. This review focuses specifically on CBs in individuals with ID, rather than those occurring in the context of autism spectrum disorder (ASD). Although ASD and ID frequently co-occur and may share certain behavioral features, they represent distinct clinical populations with differing neurodevelopmental profiles, diagnostic frameworks, and treatment responses. Including studies primarily focused on ASD would have introduced significant heterogeneity, potentially limiting the applicability of findings to individuals with ID. This distinction is clinically meaningful, as highlighted by Thurm et al. (21), who emphasize the diagnostic and phenotypic divergence between ID and ASD despite overlapping symptoms.

The aim of this comprehensive review was to synthesize existing literature on the efficacy and safety of antipsychotic drugs interventions in this specific patient population, while highlighting potential limitations. By critically examining these findings within the context of clinical practice, we seek to inform clinical decision-making and improve therapeutic approaches.

A comprehensive search was conducted on PubMed using Medical Subject Headings (MeSH) terms and keywords to identify relevant studies on antipsychotic pharmacological interventions for individuals with ID and CBs. The search strategy included terms like “Intellectual Disability,” “Drug Therapy,” “Psychopharmacology,” “Antipsychotic Agents,” “Aggression,” “Disruptive Behavior,” “Conduct Disorder,” “Self-Injurious Behavior,” “Hostility,” and “Impulsive Behavior.” An example of the search string used is: (“intellectual disability” AND “antipsychotic agents” AND “aggression”). The search covered studies published from January 1990 to June 2024, focusing on RCTs or cross-sectional or longitudinal observational studies. Review papers, abstracts, and case reports were excluded. Articles were screened based on title and abstract screening, and retained articles were assessed for full text eligibility, leading to the final included articles. This review was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines (22), providing a standardized framework for identifying and selecting studies. Eligibility criteria for article selection were defined using the PICOS (Population, Intervention, Comparison, Outcomes, Study design) approach (23), as detailed in Supplementary Table 1 .

As anticipated, significant gaps remain in the literature regarding the efficacy and safety of antipsychotics drugs to treat BC in people with ID. By applying our criteria, we were only able to select a small number of studies, all of which included a small number of subjects (346 subjects all together, maximum 86 for the largest). Furthermore, the formal risk of bias assessment revealed that, although most RCTs were of moderate to high methodological quality, several observational studies presented moderate to serious concerns—particularly regarding confounding and participant selection. These limitations, detailed in Supplementary Table 3 , underscore the fragility of the current evidence base. For this reason, a meta-analysis was not conducted. The small number of eligible studies, their heterogeneous designs, and the variability in outcome measures, interventions, and follow-up durations precluded meaningful quantitative synthesis. It is also worth noting that we deliberately excluded the term “autism spectrum disorder” from our search strategy, as previously explained in the introduction. While this decision likely limited the number of retrieved studies—as suggested by findings from a recent systematic review and meta-analysis by Deb et al. (40)—it was necessary to maintain population specificity and avoid the heterogeneity introduced by overlapping but distinct neurodevelopmental profiles. Nevertheless, there are notable similarities between our findings and those reported by Deb et al., particularly the lack of strong evidence supporting the use of antipsychotics to treat BC in both ASD and ID populations.

In their study, Deb et al. found some evidence for risperidone (14 RCTs) and preliminary evidence for aripiprazole (5 RCTs) to significantly reduces BC in children with ASD. They found no evidence for adults ASD population. However, our findings do not support a similar level of evidence and show mixed results regarding the efficacy of risperidone in ID population and no conclusive results regarding aripiprazole. Specifically, only one study (28) demonstrated some efficacy of long-term, low-dose risperidone in reducing irritability in children, adolescents, and adults with ID. Conversely, another study (26) reported no significant difference between risperidone and placebo in reducing BC.

Regarding adverse effects of antipsychotic treatment, Deb et al. reported a higher incidence of side effects such as weight gain and sedation in ASD populations. These findings are consistent with our observations in individuals with ID.

Overall, the studies presented showed that atypical antipsychotics, which are currently still preferred, can be just as effective as typical antipsychotics in this off-label use. Risperidone was the compound for which we identified the most data. In our selection, olanzapine was next, while aripiprazole was only studied in one small open-label trial in people with FXS. It should be noted that in the study of Deb et al., on subjects suffering from ASD, aripiprazole was involved in 5 RCTs, 4 of which were positive versus placebo for treating BC. The presence of an ID was only specified in one of the studies and was 63% of the included people. Regarding typical antipsychotic drug we only found data on low dose zuclopenthixol in 40 people (30, 31) as well as negative results with haloperidol in 30 people (26).

Although commonly accepted as effective, it is interesting to note that the largest study, a randomized controlled trial, showed no effect of risperidone compared to placebo for treating aggressive BC, after one month of treatment. In fact, a clear reduction in aggressive behaviors was reported in the 3 groups, risperidone, haloperidol and placebo, during the 1st month of treatment, underscoring the magnitude of the placebo effect (26). This observation challenges the assumption that antipsychotic medication is always necessary to achieve behavioral improvement in this population. It suggests that non-specific factors, such as increased attention, structured environments, or caregiver expectations, may contribute meaningfully to short-term outcomes. These insights reinforce the importance of reassessing treatment need regularly and considering non-pharmacological approaches as first-line interventions.

Another interesting point was that the efficacy of low doses of risperidone appeared to be equal to that of higher doses. High doses, which are associated with a higher risk of adverse effects, should therefore be avoided (28).

Regarding the types of BC, although there were some evidences in favor of atypical antipsychotics diminishing aggression and agitation, their impact on SIBs remains controversial. This might be explained by the sedative properties of some antipsychotics, for instance olanzapine, which may help mitigate aggressive behavior (41), but these sedative properties alone may not be enough to limit self-aggressive behaviors, as Ruedrich et al. (20) demonstrated in their study on risperidone and olanzapine. On the counterpart, Janowsky et al. and Erickson et al. (6, 29) reported positive outcomes on SIBs using olanzapine and aripiprazole respectively. These divergent findings underscore the variability in how atypical antipsychotics affect SIBs. Such inconsistency may be rooted in the multifactorial nature of SIBs, which involves dysregulation across several neurotransmitter systems—including dopaminergic, serotonergic, glutamatergic, and GABAergic pathways—as well as structural and functional abnormalities within fronto-limbic-striatal circuits (42). These neurobiological alterations can impair emotional regulation, behavioral inhibition, and reward processing, thereby contributing to the onset and persistence of SIB. This neurobiological and behavioral heterogeneity may help explain the mixed treatment outcomes observed across studies. It also highlights the need for improved phenotyping and the adoption of multimodal treatment strategies combining pharmacological and behavioral components.

Finally, the need to continue treatment in the medium to long term (>1 year) is being questioned. While some studies support long-term low-dose antipsychotic prescriptions (30, 31), others suggest the possibility of deprescribing without exacerbating BC. For example, Ramerman et al. (5) indicated that long-term treatment with risperidone can be withdrawn without increasing irritability. These findings advocate for considering deprescribing strategies, especially given the limited long-term efficacy and significant side effects associated with these medications, also opening windows for implementing non-pharmacological strategies which appear to be moderately effective (43).

Ramerman’s findings were further supported by various experiences, case reports and expert views on deprescribing psychotropic medication in ID population presenting challenging behaviors. A qualitative study by de Kuijper et al. (44) suggested that withdrawing antipsychotics is feasible with careful planning and support, involving healthcare professionals, patients, and their families to ensure a successful process. Similarly, a systematic review on deprescribing psychotropic medication for BC in people with ID (45), highlighted the complexities and potential benefits of deprescription by analyzing 54 studies and identifying both positive outcomes and substantial barriers. They emphasized that deprescribing can lead to behavioral and health improvements, particularly when implemented within an interdisciplinary care model, but also warned of risks such as relapse, dyskinesias, and the need for re-prescribing. The authors stressed the importance of pre-planned deprescribing strategies, stakeholder involvement, shared decision-making, and regular monitoring to ensure successful and safe implementation. Key factors for success include the role of primary caregivers, staff knowledge and skills (46), and realistic expectations about psychotropic medications, which are often overestimated.

An example of antipsychotics deprescription’s strategies is the structured withdrawal programme implemented by Shankar et al. (13). This program was enforced in Cornwall, UK and involved multiple steps and stakeholder engagement. Meetings were held with patients, carers, and professionals to discuss the withdrawal process, including potential risks and benefits. An oversight committee coordinated the program, integrating input from primary and secondary care, and facilitating regular Multi-Disciplinary Team discussions for case reviews. Specific tools were developed to assess and visualize risks and patient outcomes. Gradual dose reductions of 10-25% every 6–8 weeks were performed. A follow-up plan was established for each patient, with contingency strategies to manage any adverse effects during and after the withdrawal process. Results of their study show that with this structured approach, it was possible to completely withdraw antipsychotics in 46.5% of adults with ID and reduce the dose by over 50% in an additional 11.3% with no significant changes in the challenging behaviors.

These findings support the potential for wider implementation of structured withdrawal programmes to reduce inappropriate prescriptions on antipsychotics and long-term side effects, thereby improving the overall quality of life for individuals with ID.

One useful tool to guide clinicians in deprescription might be the Tool for Optimizing Prescription in Intellectual Disability (TOP-ID) (47). TOP-ID is a structured prescription and deprescription guide developed specifically for adults with ID. It aims to address common clinical situations such as pain management, gastrointestinal disorders, sleep disorders, and BC, providing a systematic approach to optimize medication use and minimize adverse effects. TOP-ID was developed using a four-step consensus-based process that included a review of the literature, semi-structured interviews, and a two-round Delphi process with 18 experts from various medical fields. This rigorous process ensured that the tool is based on both clinical expertise and the best available evidence. TOP-ID holds promise for clinical practice by providing a valuable tool for clinicians to make informed decisions about prescribing and deprescribing for this vulnerable population.

Furthermore, given the scarce evidence on the efficacy of antipsychotics, especially for long-term use, the importance of incorporating non-pharmacological methods cannot be overstated. These include behavioral therapies (e.g., positive behavioral support (PBS) or adapted dialectical behavior therapy (DBT)), environmental modifications, and caregiver training. For instance, a longitudinal study by Brown et al. demonstrated that adapted DBT can effectively reduce challenging behaviors in adults with intellectual and developmental disabilities (48) Similarly, recent evidence from a multicenter trial by Bruinsma et al. (49) suggests that PBS delivered by trained staff can improve quality of life among individuals with more severe CB or lower levels of intellectual functioning. Integrating both pharmacological and non-pharmacological interventions in a holistic approach may prove more effective in managing BC in individuals with ID (50, 51). Such comprehensive strategies, which include environmental modifications and behavioral therapies, could potentially provide more sustainable outcomes for individuals with ID, thereby ensuring a more balanced and effective management of their needs.

This review has several limitations, including the small sample sizes of the included studies and significant heterogeneity across various factors, such as the type of antipsychotic used, dosage, duration of treatment, patient comorbidities, concomitant medications, underlying causes of intellectual disability, severity of ID, types of BC, age of patients, assessment scales, and study settings. These variations complicate the ability to generalize findings and make direct comparisons, thus limiting the overall applicability of the results. Additionally, given the considerable timespan covered by the included studies (1990–2024), it is important to acknowledge that both the definitions of ID and the pharmacological and non-pharmacological interventions have evolved substantially over time. This temporal variability represents an additional limitation, as it may affect the interpretation and comparability of findings across decades.

This review underscores the lack of consistent results on the use of antipsychotics for managing BC in ID population in contrast to the observed clinical practices, relying often on medication.

The place of antipsychotic drugs in the care of BC has therefore to be specified.

There is no doubt that medication can be useful in behavioral crisis, but it is essential to think systematically, and repeatedly, about the medical and environmental causes of BC and to use non-pharmacological measures to prevent BC before using medication. Another important aspect is to assess carefully concomitant somatic and psychiatric disorders and to treat them.

Tools like TOP-ID can guide clinicians in this process by systematizing the approach and offering a step-by-step path that leads to more informed and patient-centered decisions, thereby minimizing the long-term use of antipsychotics and their associated side effects. Antipsychotics should, in fact, only be prescribed for a limited duration, unless justified by clinical response and ongoing need. This recommendation aligns with current international guidelines and reflects the principle that antipsychotics in individuals with ID should always be used within a framework of regular, structured reassessment of both indication and effectiveness.

Efforts should also be focused on implementation of deprescribing measures, widely described in the literature but sometimes difficult to set-up in everyday practice.

Future research should prioritize well-designed naturalistic studies and longitudinal observational designs to strengthen the evidence base for the use of antipsychotics in individuals with ID. These studies should focus on optimizing dosing strategies, exploring the long-term impact of these medications, and investigating additional pharmacological and non-pharmacological options. Despite the inherent challenges in conducting such studies within the ID population, the frequent use of antipsychotics in clinical practice makes it vital to equip clinicians with appropriate strategies and tools to optimize both the prescription and deprescription of psychotropic medications for challenging behavior. Such an approach can help tailor treatments to individual needs, minimize adverse effects, and ensure consistent monitoring, thus standardizing care across various clinical settings.