Toward a Terminology for Functional Recovery in Schizophrenia: Is Functional Remission a Viable Concept?

Abstract

Recovery in schizophrenia is receiving increasing attention. Part of the increased focus is based on the recent working criteria for clinical remission in schizophrenia and the realization that many people with schizophrenia meet these criteria for remission. In this article, we consider whether functional disability can also be evaluated in a “remission” model. In so doing, we evaluate the concept of clinical remission, evaluate the possibility of remission of other generally stable features of schizophrenia such as negative symptoms, and make some heuristic terminological recommendations. We also propose a “level and breadth” model for the definition of functional remission and examine some of the alternate influences that could produce suboptimal everyday functioning, including effort, motivation, and societal barriers toward functional achievement.

Introduction

There are multiple features of schizophrenia, including disability in everyday functioning (social functioning, everyday living skills, productive activities, and independence in living), cognitive impairments, various comorbidities (substance abuse, medical illness, and medication side effects), and other symptoms such as depression and anxiety that are not part of the formal diagnostic criteria for this illness. For years, the primary focus on schizophrenia was on the reduction of clinical symptoms (positive, negative, and disorganized) and their direct consequences, including discharge from long-term care and on relapse prevention. This focus clearly does not address the majority of the problems experienced by people with schizophrenia, and symptomatic reduction is now viewed as only one of several meaningful treatment goals.¹ Now it is more widely accepted that recovery is possible, even to the extent of full absence of all symptoms and disabilities and that optimal recovery should be a goal for people with schizophrenia, although not all individuals with the illness will make significant progress toward this goal.²

Recovery is often viewed as a process and not an outcome, and the multiple definitions of recovery that have been advanced are reviewed by Freese and Saks in this issue. This process may involve improvements in different elements of the illness at different time periods. For instance, psychotic symptoms are often responsive early in the course of illness, while improvements in everyday functioning may take longer to achieve. Comorbidities are an important consideration, in terms of both prevention as well as treatment, and some do not develop until later in the illness course. Finally, subjective well-being is a domain that needs to be continuously considered over the course of the illness, in that subjective well-being does not have a clear correlational relationship with all the other symptomatic and functional features of the illness.

There are multiple elements of recovery. These include freedom from troubling psychotic symptoms and relapses, satisfaction with life and daily activities, and suitable functioning in everyday life. In this article, we will focus on the issues associated with “functioning,” that is the performance of daily activities that are required for self-maintenance (earning an income and maintaining a residence), as well as social activities. This special issue is aimed at functional recovery, and we will discuss the issues of freedom from psychosis and life satisfaction because they inform our thinking about reductions in impairments in functioning in everyday life. Our primary goal in this article is to facilitate the process of definition of functional recovery and its component parts, in order to promote additional research and treatment efforts aimed at these topics.

Definitions and Terminology for Functional Recovery

One of the major issues associated with functional recovery as a goal is an almost complete lack of consensus on the appropriate terminology and standards used to index levels of accomplishments in the process of functional recovery. This is particularly challenging because, to a large extent, there are no clear standards for suitable levels of accomplishment in these same functional domains in the healthy population. It is easier to define the absence of substantial psychotic symptoms than it is to determine how an individual’s condition should be evaluated if they are showing mixed functional outcomes: such as living independently, generally symptom free, but have no productive activities or social contact, and are satisfied with that situation. The need to apply value judgments and to consider an individual’s history and motivations is clearly embedded in the task of defining functional recovery. We acknowledge that we are making several such value judgments and will identify and discuss each below.

The complexity of this issue is highlighted by using the domain of impairments in cognitive performance as a point of reference. In some ways, cognitive impairment should be an easier area to index impairment or lack thereof, as well as treatment-related or spontaneous reductions in impairment, because there are clear normative standards for a wide range of cognitive tests in the healthy population. Scores on various cognitive performance measures can be converted into percentile scores and adjusted on the basis of demographic factors that contribute to performance. There are no such detailed norms for outcomes such as residential independence, vocational outcomes, or social functions.

Despite the availability of these normative standards and substantial understanding of the demographic determinants of performance in healthy individuals, there is still a major controversy as to the prevalence of definite cognitive impairment and, conversely, “neuropsychological normality” in schizophrenia.³ This controversy originates from the argument that normal range performance on the part of people with schizophrenia is actually impaired relative to premorbid functioning.⁴ Clearly, consideration of such historical factors is important in the definition of “normal” or “average” cognitive functioning. Such historical factors are likely to also be important in the evaluation of functional recovery in people with schizophrenia.

Remission of Clinical Symptoms

To demonstrate the substantially greater challenges in defining levels of unimpaired everyday functioning compared with the absence of psychosis, an evaluation of recent advances in the definition and measurement of clinical remission will be instructive. A recent consensus definition of “clinical remission” has been proposed.⁵ Following from previous efforts in the domain of affective disorders,⁶ these criteria were based on a process of consensus development employing a committee of international experts on schizophrenia. The proposed clinical remission criteria include definitions for remission of clinical symptoms that are referenced into severity scores obtainable from several different clinical symptom rating scales. The symptoms of central focus in this conceptualization are those that are part of the active phase of illness criteria (ie, the “A” criteria from the multiaxial classification system) according to the Diagnostic and Statistical Manual of Mental Disorders. Remission, according to these criteria, is defined as the absence of any of the central clinical symptoms of schizophrenia. These central symptoms include all the positive, negative, and disorganized symptoms of the illness. “Absence” is defined as a level of severity for every one of these symptoms at a level no greater than “mild.” Thus, the presence of clinical remission is predicated on the presence of no symptoms that exceed the level of mild. The temporal duration of remission was also defined as a part of these criteria and set to a 6-month period. Thus, the presence of clinical remission is operationally defined in terms of symptom severity, and a meaningful duration of the absence of symptoms is required in order to substantiate this state of remission.

This definition of remission has spurred considerable interest. The article presenting these criteria has been cited over 125 times and has spurred multiple diverse research efforts. For instance, several different studies have examined the prevalence of remission, defined with these criteria, following pharmacological treatments⁷; examined other elements of functional outcome (eg, employment and independent residential status) in patients who did and did not meet criteria for remission⁸; examined the convergence between the development of remission and improvements in cognitive functioning⁹; and examined the differential importance of remission as compared with cognitive impairments for achieving other functional milestones.¹⁰

This clinical remission concept has, therefore, served as a valuable heuristic. Interestingly, application of this definition has also made it clear that, even with fairly stringent criteria for the presence of clinical remission, a substantial proportion of people with schizophrenia at various stages of the illness meet remission criteria and do not have substantial current psychotic symptoms, both cross-sectionally and over longitudinal follow-up periods. This encouraging rate of remission stands in marked contrast to the estimated rates of functional disability^11,12 in schizophrenia.

Defining clinical remission is not without some limitations. For instance, being in current remission does not provide information about which symptoms have improved and which were never present. Further, the attention paid to clinical remission is in some sense a return to a previous exclusive focus on clinical symptoms to the exclusion of other relevant features of the illness. Remission is a very conservative criterion and a clinical worsening that leads to loss of remission may be such a small change that it does not come close to a relapse.

Functional Remission

While the concept of remission has an intuitive appeal when thinking about symptoms, because of their episodic nature even in untreated patients, the development of a remission criterion for functional disability is somewhat more challenging. There are several levels where this is a challenge, including whether the term “remission” is the best term to apply to traits that historically do not manifest much variability, which domains of everyday functioning should be considered, what duration requirements make sense, and what standards should be used to define remission. When meeting criteria for clinical remission, symptoms can be present but must be mild in nature. Following this logic, complete achievement of wide-ranging normality in everyday functioning or “full recovery” should not be required to consider a person to be in functional remission, but the standards for what defines “minimal impairment” requires some thought and some research.

Is Functional Remission a Tenable Concept?

In terms of the applicability of the term remission to functional deficits, the enduring nature of disability in most patients seems contrary to the cyclical characteristics of clinical symptoms in most patients, particularly positive symptoms. It has been argued that the course of cognitive and functional disability is stable over the lifetime in schizophrenia.¹³ Many people with schizophrenia have never achieved certain social, educational, or vocational milestones, even during premorbid periods, and lifelong disability could simply be a continuation of poor premorbid functioning.¹⁴ Some patients with schizophrenia show a deteriorating course during the premorbid periods, while others are stably impaired and others are unimpaired.¹⁵ These differences in premorbid courses are associated with treatment response immediately after the time of illness onset.¹⁶ On the other hand, it is clear that functional deterioration does occur after the onset of illness in some patients with schizophrenia, with many patients manifesting levels of disability after the onset of their illness that are substantially worse than their premorbid functioning.¹⁷ Some patients appear to show functional decline over the entire course of illness, with increasingly lower functioning in residential and vocational roles and increasingly more severe social dysfunctions.¹⁸ Finding worsening in functional status does not imply that reversal of decline is necessarily possible, but the fact that disability is apparently dynamic suggests that remission is a tenable concept. This argument is also supported by the fact that some people who receive a diagnosis of schizophrenia appear to have long-term outcomes without any impairments or symptoms (see Freese and Saks, this issue for a detailed discussion).

Negative symptoms may give us some guidance in terms of the viability of a remission concept for functional disability. It has long been argued that negative symptoms manifest substantial temporal stability, and this has been demonstrated with sophisticated statistical techniques,¹⁹ but many patients with schizophrenia have minimal enough negative symptoms such that they meet criteria for remission. It is not possible to determine from the finding that a patient is currently in remission whether their negative symptom severity improved to this level or was never impaired in the first place. Recent studies using mixed model regression approaches have produced evidence for change in negative symptoms with treatment in both acute and chronic samples.²⁰ The findings suggest that ostensibly stable features of the illness can manifest remission and that similarly stable functional deficits may themselves be dynamic in nature.

What Domains of Functioning Should Be Considered?

There are multiple domains of everyday functional activities where adequate performance on a day-to-day basis is required for normal functioning in Western cultures. These include productive activities, residential and self-maintenance activities, and social relationships. For people with schizophrenia, adherence to medication treatments is also associated with reduced risk for relapse and readmission or emergency care. While different social and familial roles may impact on the levels of achievement in these areas (eg, homemaking vs outside employment), these 3 broad domains are likely to be relevant to everyone in Western cultures.

Medication adherence poses a different challenge. There is little doubt that nonadherence in patients who are medication responsive is associated with increased risk for worsening. Many patients show evidence of nonadherence based on poor organizational ability, and medication adherence interventions reduce risk of symptomatic worsening.²¹ We have chosen not to include adherence in our functional remission domains because it has been extensively defined and studied in other areas.

What Is the Role of Effort?

Success in all these functional domains requires effort in order to perform the relevant skilled acts. Further, individuals who lack the necessary skills to perform the acts must acquire them in some way as well. Attending rehabilitation clinics, receiving individual or group treatment, and then attempting to apply the skills learned require both an interest in learning these skills and a willingness to make an attempt to perform them. Individuals must make some first steps toward both using the skills that they have and toward learning new skills if they lack critical competencies. Is it better to try and fail than never to try at all? Clearly this is a value judgment, and in our first attempts to define levels of functional remission, we have weighted individuals making efforts as having made more progress than individuals who have made no efforts.

Our solution as a first step in this heuristic process is to present a 2-component conception of functional remission. We have separated the level of success, within functional domains, as well as the breadth of success across functional domains. As noted above, we have chosen to believe that exerting effort is better than not and have rated outcomes where attempts are not yet successful as a better outcome than failure to exert any effort. Naturally, there are certain circumstances where this is likely to be untrue, such as someone who has made repeated failed attempts without incremental change in basic abilities and where continued failed attempts could be “toxic” for the person. Although somewhat arbitrary, we have elected to use “progress” as an index of the various levels of functional remission. For instance, attempting to achieve functional milestones without success is rated as a step closer to remission than making no attempts. As a result, criterion 1 for development of functional remission is the level of success that is obtained. Criterion 2 is the breadth of success across domains, with the domains defined as social, productive activities, and independent living. Thus, higher levels of success across multiple domains would bring an individual closer to functional remission.

What Level of Functioning Is Consistent With Remission?

When evaluating these domains of everyday functioning, it must be considered that within the realm of healthy functioning, there is considerable variability in the extent to which individual functional goals are met. For instance, it is possible to work at a job that is below the individual’s potential and to live in a substandard residence. Further, preference for levels of social contact varies considerably within the healthy population, and there is clearly marked variation in social activities across individuals. Thus, specification of what constitutes “healthy range” functioning requires a variety of value judgments.

As a result, defining what level of functioning would be consistent with remission is a part of the definitional challenge. Return to premorbid functioning is an intuitively appealing and easy-to-reference benchmark. However, “premorbid” functioning may itself be affected by the prodrome of the illness,²² and the time during which the functioning is assessed (ie, early prodromal and premorbid) may actually lead to an underestimate of lifetime maximum potential. Further, healthy real-world functioning is also intrinsically developmental, with marked differences in expectations for individuals in their teenage years (when the illness often develops) and early to middle adulthood. Thus, the topography of everyday living skills is dynamic and independent real-world functioning (other than academic performance) in adolescence is intrinsically more limited than in adulthood.

Breadth of Success

Our conceptual reason for focusing on breadth of success is the fact that achievement in productive, residential, and social domains occurs to a greater or lesser degree in the healthy population. Further, clinical remission criteria focus on minimal symptomatology across the whole array of the A criteria for schizophrenia. Thus, functional remission will eventually be defined by achievement of some level of success across some combination of functional domains. Exactly which breadth of achievement and at what level of achievement is the question to be answered in terms of defining functional remission.

Table 1 presents some of these issues and levels of success across benchmarks that could define possible definitions of functional remission. Return to premorbid baseline functioning is not presented independently because the level of premorbid functioning can be indexed by these levels of achievement. As can be seen in this table, within each functional domain, we propose multiple levels of achievement. We have attempted to use quite concrete real-world outcomes that could be assessed with reliability. Different cultural groups within the United States and different worldwide cultures may have clearly different reference points.

Table 1.

A Level and Breadth Approach to Defining Functional Remission

Criterion 1

Level of accomplishment

Not actively working toward change/functional improvement

Making attempts—no real world success: attending vocational rehabilitation without employment; functions in locked board and care; attends social activities without participating

Making progress—modest real-world change: supported residence; supported part-time employment; participates in group social activities

Partial success—nonsupported staffed residence; part-time work or school; partially functional as a homemaker; some self-initiated individual social activities

Full success—independent living; full-time competitive employment or education; active social life with friends and dating

Criterion 2

Breadth of accomplishment across functional domainsa

Making progress in 1 domain

Making progress in 2 domains

Making progress in 3 domains

Partial success in 1 domain and making progress in 1 other

Partial success in 1 domain and making progress in 2 others

Partial success in 2 domains and making progress in the other

Partial success in 3 domains

Full success in 1 domain, making progress in the 2 others

Full success in 1 domain and partial success in 1 other making progress in the third

Full success in 1 domain and partial success in 2 others

Full success in 2 domains and making progress in the other

Full success in 2 domains, partial success in another

Full success in 3 domains

^aFunctional domains are productive activities, independent living, and social.

Some examples may help to clarify these heuristic classifications. For instance, an individual who manages to live in an unsupported residence with no attempts at socialization and/or productive activities would be seen as partially successful in one domain. If this individual was also working part-time, this would be partial success in 2 domains. If the same person was also attending a social networking group and making contacts there that did not extend outside the group, this would be partial success across 3 domains.

As can be seen in table 1, there are multiple combinations of the breadth of outcomes. What will require consensus in the field is the combination that reflects functional remission. Partial success in all 3 functional domains of interest is an indicator of broader progress then full success in 1 area combined with no attempts to function in the other 2, but the benchmarks for partial functional success are not consistent with remission. Full success in 2 areas combined with partial success in 1 seems a reasonable possibility for defining functional remission, but it clearly could be argued that the specific functional domains where full success is required may be critical. For example, it could be argued that independence in residential status is a requirement for functional remission, while partial achievement in other areas could be accepted. This is clearly an issue for future debate and research.

What Is the Time Frame Used to Index Remission?

The time frame for clinical remission is indexed to 6 months. Given the rapid nature of the response of clinical symptoms to treatment, with much clinical response seen in 2 weeks,²³ a 6-month duration for remission is likely capturing maintenance of clinical symptomatic improvement. In contrast, many functional goals in the real world would require a substantially longer time period to occur and evaluation of maintenance might require an even longer period of observation. Even with basic competencies in place, finding a place to live, a job, or romantic partner can take time. As we have suggested elsewhere, development of basic competence is likely to be a prerequisite for real-world functional improvements.²⁴ As a result, development of functional remission is likely to take a longer period of time in most cases than development of clinical remission and a long-term period, such as 6 months after initial development of remission, for evaluation of the maintenance of functional remission seems to be a minimum requirement.

Environmental Issues

Another major issue in the functional remission domain is intrinsic and extrinsic motivations for real-world functional achievement. One of the most substantial predictors of unemployment in schizophrenia in the United States is the receipt of disability compensation.²⁵ Regardless of the willingness of an individual to work (ie, exert effort), there is minimal motivation or even disincentives to do so. In many circumstances in the United States, because of the rules associated with insurance and disability payments, seeking employment at levels above the minimal level can compromise both financial compensation and health insurance. Thus, an individual without developed work skills or a meaningful work history, who receives compensation for a disability, may find that seeking entry-level employment would be an irrational choice. In a related vein, real-world functional performance is also affected by societal factors, including stigma. Thus, a patient who is motivated and able to work may find himself shut out of the job market by negative attitudes among employers or a weak economy that favors those with good work histories.

Environmental factors that preclude performance of real-world activities should be considered in the determination of the presence of remission. If an individual is capable of full-time work, but receives more compensation than they would earn, then their compensation status needs to be considered as a factor in their current employment. Similarly, if an individual could live independently but prefers to spend the money that it would cost to live along on other productive activities, such as education, then this extraneous factor needs to be considered as well. Directly relevant to this issue, we recently examined the correlation in a sample of older people with schizophrenia between normal performance on neuropsychological tests and functioning in 3 different real-world outcomes areas: independence in residential status, martial status, and employment status.¹⁰ The relationship between neuropsychological normality and residential outcomes was clear and statistically significant: normal-performing patients were more likely to be living independently and to be financially responsible for their dwelling than patients who were performing in the impaired range. The impaired patients were significantly more likely to be living in a fully supported residence than normal performers as well. Interestingly, marital status and employment were both unassociated with cognitive status. Societal influences may impact on employment, while the lack of opportunities during critical developmental periods may have an adverse impact on social and marital outcomes.

These issues clearly indicate that the development of meaningful criteria for functional remission will require understanding premorbid abilities and likely outcomes if schizophrenia had never developed, as well as cultural and familial expectations for achievement and social outcomes, and opportunities and disincentives present in the immediate environment. Defining functional remission with too high a bar may mean that functional remission in schizophrenia requires “supernormal” functioning. Requiring that all individuals have equivalent performance across functional domains may disadvantage those from cultures where certain activities are not performed by members of one sex or the other. Excessive reliance on personal/subjective definitions of functional remission could define as remitted a seriously ill group who lack awareness of limitations and who report satisfaction with levels of functioning that are truly impoverished and have long-term adverse effects.

Another major issue to be considered is that of whether elements of functional disability can be more reasonably accounted for by other symptoms, such as negative symptoms. If an individual has both the cognitive abilities and the skills required to perform functional acts but does not do so because they experience anhedonia then their functional disability may be due entirely to symptoms. The inability to anticipate a pleasurable experience may remove motivation to perform the acts.²⁶ Similarly, high levels of anergic symptoms may lead an individual to simply not engage in enough behavior to achieve functional outcomes.²⁷ Most directly, and as discussed before, symptoms of amotivation are a potent interfering factor for functional outcomes. What will need to be determined in the end result is if remission is a purely behavioral criteria, ie, what does the person do in the real world regardless of the reasons for it, or if other contributors need to be considered as well. This element of the disability picture strongly suggests that functional abilities (ie, functional capacity) need to be examined when attempting to determine if a person with schizophrenia is in functional remission or not.²⁸ Measuring the ability to behave in a manner consistent with remission is likely to be a critical step in the assessment process and to determine the causes for lack of functional remission when it is seen.

Subjective Response

Subjective reaction to illness and current subjective illness burden are important factors. First, the willingness of a patient to participate in, contribute to, and sustain involvement in their functionally relevant aspects of treatment is likely linked to their subjective impression of current illness burden and likely future treatment benefit. Patients who are “objectively” doing well (symptomatic remission and partial functional remission) but do not share this positive view are likely to be at high risk for a variety of adverse outcomes. Conversely, patients who see themselves as unimpaired or not ill, despite disability and symptoms, should clearly not be considered to be in functional remission simply because they do not consider themselves disabled. While it has been known for years that unawareness of illness in the realm of lack of insight into psychotic symptoms is common in schizophrenia,²⁹ it has become quite apparent in recent years that many patients manifest discrepancies compared with observer reports in evaluating their cognitive, social, and functional abilities and their level of success in the achievement of real-world outcomes.³⁰ While subjective reports of illness burden and perceived disability have been found to have a substantial correlation in schizophrenia, these self-reports are often found not to correlate with observer ratings or direct measurements of functional skills performance.³¹ This applies as well to self-estimated cognitive impairment, where subjective appraisals of cognitive functioning are often correlated close to 0 with performance on NP tests, while observer and clinician ratings of impairment were found to be more highly associated with patient performance.³²

The implication of these issues is that the global concept of recovery needs to consider patient impression, but functional remission cannot be entirely dependent on subjective response. We argue that it is critical to consider both real-world functioning, objectively measured, and subjective response to life situations (including treatment) as separate components of the recovery process. Satisfaction with one’s life can arise from many sources, and there is no clear definition of some threshold level of everyday functioning above which satisfaction from achievement, in social, financial, or residential domains, would intrinsically accrue in either patient or healthy populations.

Conclusions

Functional remission appears to be a tenable concept. It is possible to identify the domains of life functioning where improved functioning on the part of people with schizophrenia would lead to greater autonomy. We see functional remission as a separate domain of functioning from clinical remission and subjective response and argue that the process of recovery includes all these domains. It is more challenging in the domain of functional remission than clinical remission to identify the levels of performance that would be clearly remitted, but we have proposed that both level and breadth of accomplishment should be considered and that the time frame for sustained functional remission be similar to that for clinical remission. Both external and internal factors other than ability can affect real-world outcomes and are likely to reduce the changes for functional remission even in people who have the necessary skills to achieve these milestones. We suggest assessments of patients that separate ability from real-world outcome, in order that other factors can be considered, but also believe that functional remission should be indexed to real-world outcomes and evaluated accordingly. If other factors, such as negative symptoms, are found to be the primary determinant of poor role performance, this should clearly be designated. Such a separation of objective and subjective elements of functional remission and the clear identification of situations wherein adequate ability is not being expressed by good functional outcomes will lead to improved conceptualizations of symptoms and disability in schizophrenia. Improved assessment and classification will allow for treatment interventions to be optimally delivered after they have been developed.

Funding

National Institute of Mental Health (63116 and 78775 to P.D.H.); Department of Veterans Affairs VISN-5 Mental Illness Research, Education, and Clinical Center; A.S.B., Principal Investigator.