Why do we need a new definition?

Since the 2000 publication of criteria for psychosis in Alzheimer’s disease (AD) (Jeste and Finkel, Reference Finkel and Burns2000), the awareness and understanding of cognitive impairment and dementia has advanced in several ways. AD, the most common form of dementia, has received considerable prominence, particularly with the updating of criteria by McKhann et al. (Reference McKhann2011), with the focus on clinical features that could be accessible to many practitioners. Recent scientific advances have helped initiate a more molecular approach to the diagnosis of AD (Jack et al., Reference Jack2018), which may require diagnostic evaluations that are not available to many patients across settings, particularly in low- and middle-income countries. Additionally, there is a growing awareness of multiple pathologies that co-occur in dementia and milder forms of cognitive impairment, making it difficult to define a single mechanism responsible for the observed symptoms.

Contrary to this molecular emphasis in diagnosis described by Jack et al. (Reference Jack2018), the approach of the International Classification of Diseases (ICD-10: World Health Organization, 2004) and DSM-5 is to use the characterization of major and mild NCDs. These classifications apply clinical criteria that are readily available and understandable to a diverse community of dementia practitioners across the world. As the IPA employed its broad survey methodology to reach clinicians and researchers across the world, it became apparent that the respondents wanted additional clarity to identify and define the symptoms of psychosis. For example, nearly 79% of survey respondents agreed that it would be helpful to provide examples of dementia-related delusions and hallucinations; 76% endorsed the separation of criteria for duration and severity; and 80% endorsed adding to the definition, the impact of psychosis on “patient or others’ functioning” (Cummings et al., Reference Cummings2020). The symptoms of hallucinations and delusions in the presence of the cognitive and functional impairment of dementia that occurs primarily in an aging population raise a number of questions. These include the nature of the symptoms, the temporal relationship to the onset of the dementia, and the impact of aging sensory and physical changes on behavioral manifestations. More than 70% of respondents acknowledged that the criteria must consider both cultural determinants and sensory impairment. The working premise in this effort established that an optimal definition must provide the opportunity to enhance the ability for all practitioners to more effectively diagnose and treat these problems.

The IPA process for revising definitional criteria

This new consensus emerged from a multipronged approach to inform the resultant criteria. A working group that included IPA leadership and association members created the survey to assess the existing criteria with a focus on identifying areas that could be improved upon. The IPA supported the distribution of the survey to members and affiliated member groups and later, provided them with a summary and analysis of the global survey results. A larger international, multi-stakeholder work group was then formed to review the relevant literature, share knowledge from their specific areas of expertise, and evaluate the survey responses. There was remarkable (but not total) consensus between the published literature and survey responses covering the areas of phenotypic definition, symptom overlap, shared biology, and therapeutic approaches. Where disagreements across sources existed, the multi-stakeholder group weighed the evidence and made final recommendations for criteria, which were then sent to the same broad IPA membership for review and comment.

In addition to the changes endorsed in the survey described above, several other revisions were made in the new criteria and they are listed in Table 1. In particular, the new criteria facilitate the diagnosis of psychosis in the mild NCD/mild cognitive impairment phase of neurodegenerative disorders. The new criteria also overtly broaden the criteria to include all non-AD NCDs too. The separation of duration and severity and the expansion of each provided additional descriptive information addressed some uncertainty from the initial survey about symptom duration. The follow-up survey found 72% endorsement of the 1-month interval for duration and 88% endorsement of definitions of severity (Cummings et al., Reference Cummings2020). In the initial survey, there was mixed enthusiasm with fewer than 50% endorsing the continued use of the term “negative symptoms” as an associated feature and 59% endorsing its replacement with “apathy,” which has recently been recognized as a separate syndrome (Robert et al., Reference Robert2018). There was high agreement for specifically excluding delirium from the definitional text and acknowledging the exclusion for shared cultural beliefs. Overall, 90% of respondents found the new definition provided to be an improvement over previous versions.

Table 1. Comparison of criteria (adapted with permission; The American Journal of Geriatric Psychiatry)

Strengths and limitations of the IPA process

The IPA process was designed to maximize input to the criteria from a broad group of stakeholders with an emphasis on the clinicians serving the patient population. While it includes many stakeholders, the framework remains in a traditional model of psychiatric symptom evaluation, with major changes reflected in how we categorize and define the cognitive impairment. The approach has served the IPA well and is consistent with how the field defines psychiatric conditions in general. A limitation of this approach is that there is minimal information available on the validation of instruments that could independently identify symptom clusters and syndromes within dementia populations. Additionally, the condition is not readily associated with a single biology that could be used to target interventions. However, as the science and biology of psychosis grows, these criteria can be re-evaluated just as they have been now, to integrate this information. The criteria require prospective assessment and research in this area will be guided by them while identifying their strengths and weaknesses.

Work for the future

The process of building this consensus highlighted the need for better tools to identify and measure psychotic symptoms. Some have observed the breadth of tools that exist with overlapping items, including some with apparent convergent validity, and have suggested that item analysis and more standardized administration techniques may be a valuable future contribution. Of particular note is the observation that antecedent events and analysis of causes and consequences of specific behaviors and symptoms are critical steps to identifying appropriate psychosocial and other nonpharmacological interventions. The work of the consensus team acknowledges the need to demonstrate clinical meaningfulness, the value of the definition to enhance clinical decision-making, and to create evidenced-based clinical care pathways. Prospective validation of the criteria is a critical next step in demonstrating the validity, reliability, and utility of the criteria. IPA will promote these activities through its unyielding commitment to serving its multidisciplinary members and to advancing knowledge in our field. All of these efforts are devoted to improving the lives of older adults at risk for, or living with mental and cognitive disorders.

The IPA is proud to have had the opportunity to support this effort using its established methodologies and engaging its global multidisciplinary community. This consensus definition for psychosis in major and mild NCD will help advance the goals of detection and recognition of these conditions, providing a common ground for creating pathways of care that include both nonpharmacological and pharmacological interventions. Regulatory issues exist across multiple agencies, including those that oversee care delivery in different settings as well as those that regulate pharmaceutical development. Criteria for training of care providers, establishing optimal staff-to-patient ratios, and identifying environmental requirements to serve persons with specific neuropsychiatric symptoms to optimize their care are as critical as criteria for demonstrating pharmaceutical efficacy in clinical trials. The proposed definition provides a common ground for better characterizing populations to engage in research, clinical protocols for the evaluation of interventions, and development of standards of care for these vexing problems. Always ready for future discovery, the IPA supports this work as the “next step” in addressing a serious public health problem. We are hopeful that this effort will promote future novel and even more valuable perspectives, into these vitally important features of the dementia syndrome. We are excited that new research efforts may now be catalyzed leading to innovations in dementia care practices and service delivery that demonstrably improve the lives of older adults across the globe.