Table 1.

Evidence to recommendations table for deprescribing APs: Does deprescribing (dose reduction or frank discontinuation) APs compared with continuous AP use result in benefits or harms for adults > 18 y (excluding those prescribed APs for treatment of psychosis) in primary care and LTC settings?

DECISION DOMAIN	SUMMARY OF REASON FOR DECISION	SUBDOMAINS INFLUENCING DECISION
QoE: Is there high- or moderate-quality evidence? Yes ☑ No □ (See references 1–10 in the evidence reviews at CFPlus*)	The QoE for the success of deprescribing is high High-quality evidence suggests that chronic AP medication can be withdrawn in many older people with Alzheimer dementia and NPS without detrimental effects on their behaviour and without substantial withdrawal symptoms In terms of relapse (measured by a change in NPI score), there was no significant difference between people withdrawn from and those continuing APs at 3 mo (MD = −1.49, 95% CI −5.39 to 2.40) The QoE for effectiveness of atypical APs for insomnia is very low One RCT (N = 13) demonstrated no statistical difference in total sleep time, onset of sleep latency, or sleep satisfaction for quetiapine vs placebo over 2 wk for primary insomnia. The trial was very low quality owing to imprecision and risk of bias	The baseline symptom level might have an influence on the success of deprescribing APs. Patients with more severe baseline scores were more likely to experience relapses (defined as a 30% increase in the NPI score) in 2 studies. Withdrawal in patients with severe behavioural baseline scores might not be successful or should not be attempted
Balance of benefits and harms: Is there certainty that the benefits outweigh the harms? Yes ☑ No □ (See references 1–9 in the evidence reviews at CFPlus*)	Overall, benefits of AP deprescribing appear to outweigh harms Available evidence suggests that “many older people with Alzheimer’s dementia and NPS can be withdrawn from chronic antipsychotic medication without detrimental effects on their behaviour”¹⁸ Effectiveness of atypical APs for insomnia There is very low certainty surrounding a lack of evidence that atypical APs are effective for managing insomnia (1 small RCT showing non-significant improvements in sleep parameters, and small uncontrolled trials). There is minimal information surrounding harms of atypical APs for insomnia; however, their use for other indications suggests potential for harm (eg, EPS, somnolence, metabolic disturbances, anticholinergic adverse effects) The magnitude of benefits of deprescribing in terms of cognition, psychomotor status, reductions in adverse effects of AP, or mortality are unclear. Declercq et al report that “Individual studies did not report significant differences between groups on any other outcome except one trial that found a significant difference in a measure of verbal fluency, favouring discontinuation. Most trials lacked power to detect clinically important differences between groups”¹⁸	Is the baseline risk for benefit similar across subgroups? Yes ☑ No □ Should there be separate recommendations for subgroups based on risk levels? Yes □ No ☑ Is the baseline risk for harm similar across subgroups? Yes □ No ☑ There is insufficient evidence to assess any differences in risk of harm between groups. In patients with severe baseline BPSD, the likelihood of successfully deprescribing is probably lower; careful consideration should be given to plans for close monitoring and intervention if deprescribing is considered in these patients Should there be separate recommendations for subgroups based on harms? Yes ☑ No□ The main difference in the likelihood of benefiting from AP deprescribing relates to the baseline severity of BPSD. Declercq et al state that “Caution is required in older nursing home residents with more severe NPS, as two studies suggest these peoples’ symptoms might be worse if their [AP] medication is withdrawn”¹⁸
Values and preferences: Is there confidence in the estimate of relative importance of outcomes and patient preferences? Yes ☑ No □	Reasons for prescribing APs include aggressive behaviour (physical and verbal), easier management of patients during daily care, as a sleep aid, or to help caregivers cope. Other viable options, such as nonpharmacologic alternatives, are less widely used owing to limited access, being highly resource dependent, and requiring additional staff training. APs can have a small effect in decreasing caregiver burden. Thus, there might be resistance from home-care staff when decreasing AP use or pressure from nursing home staff to prescribe APs. Inadequate staffing, additional workload, and increased demands are barriers to decreasing APs. Caregivers find the use of APs for controlling behaviour harmful. In addition, caregivers observe better patient QoL when APs are not used. Families would like more information on the side effects of APs	Perspective taken: the perspectives of the patient and caregivers are central to the decision to deprescribe APs, but so is the availability of professional health care support to monitor and accompany the process Source of values and preferences: literature review, pilot study of guidelines in both LTC and outpatient settings Source of variability, if any: variability difficult to estimate Method for determining values satisfactory for this recommendation? Yes ☑ No□ All critical outcomes measured? Yes ☑ No □ Although critical outcomes were assessed with a broad approach, evidence about cost implications of potential increases in caregiver burden could not be accurately quantified
Resource implications: Are the resources worth the expected net benefit? Yes ☑ No □	As there is little evidence about cost implications of deprescribing APs, and none about cost-effectiveness, it is difficult to precisely estimate this trade-off. It is likely that in some cases deprescribing APs might lead to increased caregiver resource requirements; on the other hand, patients will no longer be exposed to numerous potential side effects of APs (increased risk of falls, stroke, death, somnolence, confusion, dizziness, EPS, metabolic disturbances, weight gain, anticholinergic side effects, tardive dyskinesia, orthostatic hypotension, cardiac conduction disturbances, sedation, cognitive slowing); medication costs will also decrease	Feasibility: Is this intervention generally available? Yes ☑ No □ Opportunity cost: Is this intervention and its effects worth withdrawing or not allocating resources from other interventions? Yes ☑ No □ Is there a lot of variability in resource requirements across settings? Yes ☑ No □ Resource requirements depend in part on severity of baseline symptoms of patients for whom APs are deprescribed and on the success of deprescribing
Strength of main recommendation: strong	The strong recommendation is based on the lack of evidence of substantial harms of deprescribing APs for BPSD, the evidence for benefits of avoiding unnecessary exposure to APs, the societal costs of inappropriate AP use, and the feasibility of this intervention in primary care and LTC; for insomnia, there is a lack of evidence for efficacy of APs and there is potential for harm
Remarks and values and preference statement	These recommendations place a high value on the minimal clinical risk of deprescribing, reducing the inappropriate use of APs and their side effects, and the associated resource use given the high cost, both monetary and nonmonetary, associated with long-term AP use. They place some value on the potential for harms from attempted deprescribing and on potentially increased caregiver resource use as a result of deprescribing APs

AP‑antipsychotic, BPSD‑behavioural and psychological symptoms of dementia, EPS‑extrapyramidal symptoms, LTC‑long-term care, MD‑mean difference, NPI‑Neuropsychiatric Inventory, NPS‑neuropsychiatric symptoms, QoE‑quality of evidence, QoL‑quality of life, RCT‑randomized controlled trial.