Conceptualization and Study of Antipsychotic Medication Use: From Adherence to Patterns of Use

Treatment guidelines supported by evidence recommend antipsychotic medication (APM) as the frontline treatment for schizophrenia (1, 2). Yet, appropriate APM use remains a contentious and controversial topic. At the heart of the debate lies an often strong divergence between prescriber and user perspectives, manifested in the different ways that tradeoffs between side effects and benefits are weighted (3). Researchers and practitioners also diverge in their views of APM, with a growing group of critical clinicians emphasizing minimization and structured discontinuation or deprescribing (4, 5). Those who endorse APM use tend to emphasize the medication’s positive effect on traditional objective outcomes, such as reductions in medical service use, symptoms, emergency department visits, and relapse and mortality, while others focus on the impact of side effects and potential long-term neurotoxicity (1). Academic debate aside, it is well established that a large proportion of people who experience psychosis do not use APM as prescribed, and nonadherence rates range between 40% and 60% (6). This Open Forum offers a flexible user-centered alternative to the more traditional, binary adherence framework. We propose a reconceptualization of what would traditionally be framed as APM “adherence research” that focuses on patterns of use and articulate a set of pragmatic recommendations.

Emerging Evidence for Diverse Patterns of APM Use

In recent years, controversy has increased about APM use, with responses ranging from moderate reservations (e.g., questions about whether a subgroup of youths diagnosed as having first-episode psychosis [FEP] might do well without APM) (7) to strong opposition, particularly regarding the unknown neurological consequences of long-term use (8). Along this spectrum, critics have emphasized the risk in prescribing APM too fast, sometimes because of misdiagnosis or overdiagnosis (9). Emerging evidence has also challenged common assumptions that the use of APM is universally beneficial over the long term and that discontinuation necessarily leads to adverse outcomes (9, 10). For example, several naturalistic follow-up studies have found that, over time, individuals discontinue APMs at relatively high rates and many of those who discontinue nevertheless report high quality of life (10). More recent evidence from a randomized controlled trial (RCT) on APM discontinuation suggests that a subgroup of individuals with FEP may function well without medications and supports the implementation of structured discontinuation strategies (11). Additionally, two recent feasibility trials, both of which were designed to inform fully powered RCTs, have examined the use of medication-free FEP treatment (12, 13). Other service users may follow nonstandard APM dosing regimens or augment or substitute treatment with nontraditional drugs or medications, such as cannabis. At the individual level, these heterogeneities are reflected in the different ways people understand and experience psychosis and APM use, in their motivation to alleviate symptoms, and in their preference for managing tradeoffs. For example, some people prefer symptoms over side effects, and others have reported missing certain symptoms once they have been eliminated or suppressed (14).

These developments in the spheres of both research and advocacy arguably push for a shift in focus of adherence research, from a narrow, dichotomous perspective (taking medications as prescribed versus not) to exploration of the range of ways people with psychosis choose to use medication. With this context in the foreground, we argue that it is incumbent upon the field, particularly among researchers actively focused on the study of medication use, adherence, and effectiveness, to fundamentally rethink the concept of adherence and its measurement.

Reconceptualization of Adherence Measurement in Research

Most existing adherence measures and metrics have been developed from the perspective of clinical researchers. Within this paradigm, adherence is typically considered a categorical concept, with little attention to the many alternative ways in which service users may utilize medications to support recovery goals without “adhering” as prescribed. Core adherence measures in psychiatry do not aim to capture personal medicine behaviors that support alternative approaches to mental health and recovery, such as tapering or discontinuation of APM, extended or sporadic dosing, person-developed augmentation strategies, or other forms of “mindful” nonadherence. Even where implemented, extant shared decision making (SDM) tools and approaches often fail to include the option to taper or discontinue APM or to neutrally present both the short- and long-term risks, as well as benefits, of APM. Furthermore, many of the samples included in adherence research are recruited from inpatient and outpatient settings and are unlikely to represent the broader population of individuals with a current or prior diagnosis of psychosis who use community services or choose not to use services at all. Sample selection bias, especially in clinical trials, is often compounded by narrow eligibility criteria that exclude the profiles of many clients.

Pragmatic Recommendations for Future APM Research

To better capture different patterns of APM use in research, we propose a set of specific, actionable changes. These recommendations are described below.

Shift to patterns of APM use.

A shift is needed from dichotomous operationalizations of adherence (adherent/nonadherent to medication as prescribed) to the measurement of longitudinal patterns of medication use, including mindful or intentional alterations of prescribed pharmacotherapy, such as patient-initiated extended dosing.

Support a bottom-up research approach.

Where possible, longitudinal psychosis research should be augmented with qualitative follow-up protocols that help the field understand the circumstances under which service users successfully employ alternative APM use patterns and/or discontinuation. Additionally, the field must ensure that future measures and tools are meaningfully coproduced and involve researchers and/or community members with personal experience of long-term antipsychotic use and, where applicable, discontinuation. Researchers should consistently explore the use of person-centered statistical techniques, such as latent class and latent trajectory analysis, which may help to identify the characteristics of “hidden” groups that may benefit from different use patterns. Last, in keeping with Deegan’s personal medicine framework (15), we must consistently include measures of alternative strategies service users may employ in addition to or in place of medications, with the explicit goal of APM augmentation or substitution.

Address racial disparities in APM research.

The field should ensure representation of diverse APM users in clinical trials and other research, specifically including those not served by more traditional academic medical centers or outpatient clinics. In light of the disproportionately high rates of schizophrenia among Black and Latinx individuals and the tendency to prescribe APM differently to Black and Latinx versus White service users, more research is especially needed on patterns of use among those who are Black, Indigenous, and people of color (BIPOC).

Use SDM interventions and tools.

SDM in APM management should include an open discussion between at least two equal experts, one with clinical training and one with lived experience; at least two medication-related options to discuss, including tapering or discontinuation; and a shared decision that aligns with the patient’s goals, preferences, and values (17). SDM in psychopharmacology can bridge the gap between the individual’s values and preferences regarding APM use (value-based medicine) that often diverge from those held by the psychiatrist (evidence-based medicine) (18). Value-based care is especially significant when caring for BIPOC clients, because many BIPOC APM users express distrust of prescribers, lack confidence in treatment recommendations, and do not follow through with care. Because most APM users ultimately make their own decisions about how to move forward with their treatment—a “hidden issue” in clinical care, with providers often unaware of clients’ discontinuation or nonadherence—there is a need to lay the cards on the table and openly discuss the shared risks and benefits of taking or tapering APM (3).

Conclusions

Individuals with long-term psychosis often do not share whether or how they are using APM nor the strategies they may utilize as alternatives to APM. In combination with overly simplistic, dichotomized measures of adherence, this tendency has arguably led to the impoverishment of research on APM use. Simultaneously, in spite of current tensions and controversies, it seems likely that stakeholders with various viewpoints would agree that a deeper understanding of the use and discontinuation of APM and associated consequences is of great importance to the field.

In this Open Forum, we have described a series of challenges pertaining to the conceptualization and operationalization of adherence in APM research, offering concrete ways to advance the field. We hope that these ideas can help energize and diversify current research efforts. Ultimately, we hope that these efforts will contribute to a transformed culture of care in which service users have meaningful opportunities to explore their options and to choose whether to take medication as well as the kinds of medications, doses, and patterns of use that best support their recovery processes. A reorientation is needed away from the current dichotomous viewpoints and toward a diversity of views and experiences. We believe a shift from studying adherence to a prescribed regimen to studying user-directed patterns of use will help move research past often-polarized positions and insufficient attention to service user perspectives and concerns. This shift may help empower those with the most at stake—those prescribed APM—to make truly informed personalized choices among a range of evidence-based options emerging from APM users’ real-world patterns of use.