Introduction

Kendall and Jablensky [1] stated, “Psychiatric disorders might merge into one another with no natural boundary in between”. Explicit diagnostic criteria like those contained in DSM and ICD enable us to achieve a high degree of diagnostic reliability. By improving reliability, they have raised the lower bound for validity. The objective of DSM-V and ICD-11 is to transcend the limitations of the current DSM paradigm and to encourage a research agenda that goes beyond our current ways of thinking about psychiatric classification and diagnosis. Hitherto, subthreshold disorders are defined by the absence of one of the symptoms required to meet full diagnostic criteria, or by failing to meet the criterion for duration. The concept of subthreshold disorders includes the atypical ones, not elsewhere classified and non-specific.

Psychiatric classifications have gone through a progress of limping and splitting for many decades. The number of overall mental disorders in ICD-7 was 78, in ICD-8 210, in ICD-9 569, and in ICD-10 444. While DSM-I contained 106 mental disorders, DSM-II had 140, DSM-III 200, and DSM-IV-TR 284.

Longitudinal designs of psychiatric classifications should be advocated since little is known about the natural histories and courses of subthreshold conditions [2]. Are there self-limiting risk factors for the more severe conditions, or do these have a stable course of their own? Virtually nothing is known about the effects of treatment, both psychosocial and psychopharmacological, of subthreshold disorders (STD). The morbidity and prevalence of STD can be exemplified by a few examples: In post-Traumatic Stress Disorder (PTSD), no significant differences were found between patients with full PTSD and those with subthreshold PTSD regarding the degree of impairment. It is rather difficult to identify clinical criteria that delineate between full-blown PTSD and subthreshold PTSD. The prevalence of PTSD depends on the resilience, religious, and coping behavior of the population under study. PTSD symptoms occur in 62% of patients without any history of trauma, which raises some doubts about the validity of this diagnosis [3].

We do not need to medicalize social ills or to increase the number of diagnoses, but the degree of functional impairment produced by subthreshold disorders requires their inclusion in our current nosology. The addition of the dimensional approach to the categorical ones can accommodate subthreshold disorders in our classificatory systems.

Subthreshold disorders

An example of problems associated with subthreshold disorders comes from the affective disorders. Subthreshold symptoms in bipolar disorder impair functioning and diminish quality of life. A reassessment of bipolar-spectrum disorders including patients with subthreshold symptoms revealed a prevalence of at least fivefold greater than who found with traditionally defined syndromal diagnostic criteria. Appropriate therapeutic interventions in subthreshold manifestations should be considered even when threshold level symptoms are absent [4]. In a 20-year follow-up study, patients with type I and type II bipolar disorder were found to spend about half the time in subthreshold affective conditions. Many healthy people report depressive and hypomanic symptoms and many are identifiable as manifesting depressive, hypomanic, and cyclothymic temperaments, which appear to predispose to the respective affective disorders and personality disorders. Only about 15% of the population reports no such symptoms over their lifetime and are ‘super normal’, with very low scores for vegetative lability and neuroticism [5].

The majority of work in this area has focused on subthreshold major depressive disorder (MDD). Studies have demonstrated that depressive symptoms in various combinations, including subthreshold conditions, are present in as much as 24% of the population. This suggests that subthreshold and Full Syndromic (FS) disorders can be considered as falling along a spectrum, with subthreshold disorders being viewed as quantitatively milder than, but qualitatively similar to FS disorders [6]. A number of studies have demonstrated that subthreshold MDD is associated with significant psychosocial impairment, exhibits familial cosegregation, and predicts the later development of FS MDD. Similar findings have been reported for subthreshold bipolar disorder and schizophrenia [7]. An important caveat of our classifications is the high prevalence of comorbidity because of the overlap of psychiatric symptoms and the difficulties encountered in delineating syndromes. As long as we depend on cluster of symptoms in our diagnosis and not on specific etiology or pathology, co-occurrence will appear as a sort of pseudocomorbidity rather than comorbidity.

How does psychosis exist in nature and how does this relate to schizophrenia?

The various definitions of schizophrenia combine a mix of positive, negative, cognitive, depressive, and manic symptoms in the context of need for care. The symptoms that characterize patients with psychosis seen in mental health clinics are also prevalent in the community [8]. A recent meta-analysis reported prevalence rates (around 8%) and incidence rates (around 3%) of positive psychotic experiences that are around 10 (prevalence) and 100 (incidence) times greater than reported rates for psychotic disorder. Subclinical psychotic experiences are typically expressed in adolescence and young adulthood and are usually transitory [8].

Psychotic experiences

If psychotic experiences are so frequent in the healthy population, their diagnostic value will be low in relation to the rare disorder of schizophrenia. Research has shown that even if the follow-up period of healthy probands with psychotic experiences was extended to more than 15 years, the probability of developing a psychotic disorder did not exceed 25% [9]. A comparison has been made in a growing number of studies between dimensional (positive, negative, disorganized, manic, and cognitive) and categorical representations (schizophrenia, bipolar disorder, schizoaffective disorder, somatoform disorder, etc.). The sample of patients had a range of psychotic disorders. In forms of the utility regarding etiology, treatment, and outcome, the combinations of categorical and dimensional approaches do better than either representation alone [10].

The only subthreshold diagnosis in both classifications is found in ICD-10 as “Mixed Anxiety Depression Disorder” (MADD) presented in ICD-10 F41.2 under “Other Anxiety Disorders”. Data from the National Psychiatric Morbidity survey suggest that MADD may account for half of all cases of common mental disorders in Great Britain. The impact of MADD upon health-related quality of life is similar to that of pure anxiety and depression, but somewhat less than that of other comorbid anxiety and depressive disorders [11]. 12% of those with MADD reported a lifetime suicide attempt. 20% of all disability days in Great Britain occurred in people with MADD, accounting for around half of all the disability days occurring in people with common mental disorders [11]. Critics have rightly queried the tendency to extend the boundaries of what is considered a mental disorder, arguing that this involves the medicalization of normal human distress. Mixed anxiety depressive disorders are pathological producing an accountable impairment in the quality of life and not a normal human stress. Data suggest that many cases of MADD have merely slipped through the gaps in the current classificatory system. Inclusion of MADD would seem to be amply justified as a necessary correction to omissions in the current classification, rather than an attempt to lower the threshold of criteria to include minor cases of dubious psychopathological significance [11].

Other subthreshold psychiatric disorders

There is a growing literature on subthreshold anxiety disorders, substance use disorders, conduct disorder, antisocial personality disorder, and eating disorders [12]. A family study of a variety of subthreshold disorders (MDD, bipolar disorder, anxiety disorders, alcohol use disorders, drug use disorders, and conduct disorder/antisocial personality disorder) in a large community sample of young adults was conducted. There is a possibility that a subthreshold disorder may be associated with multiple full syndromic (FS) disorders, and vice versa. Just as there is high comorbidity between FS disorders, there is also high comorbidity between subthreshold conditions [13].

A study on social anxiety disorder (SAD) above and below the diagnostic threshold found that the 12-month prevalence rate for above-threshold SAD was 2%, while it was 3% for subthreshold and symptomatic social anxiety. There was higher comorbidity and greater impairment compared to the control group. Social anxiety below the diagnostic threshold is clearly associated with adverse outcomes [14].

Subthreshold Eating Disorders were frequently observed in transcultural studies. In a study with 371 Egyptian adolescent girls (age range 13–18 years; 4.4% of the target population), it was found that according to the operational criteria of the ICD-10 research criteria (1993), only two atypical cases of anorexia nervosa were revealed (0.54% of the sample), 26 atypical cases of bulimia nervosa (8.6% of the sample), and 32 cases of unspecified eating disorders (9.7% of the sample). A need for culture-sensitive nosology for eating disorders was recommended [15].

Perhaps the most challenging field of research on cultural issues beyond the questions of boundaries between normality and abnormality is in personality functions. Cultural factors can contribute to the debate on whether personality disorders should or should not be considered autonomous mental disorders, or whether some of them are best considered as variants of Axis I disorders (e.g. schizotypal personality disorder as a variant of schizophrenia) [16]. The reliability and validity of the current diagnostic criteria of personality disorders in both ICD-10 and DSM-IV is low when applied in the Arab and North African region. The schizoid, obsessive, and avoidant personality disorders have other religious connotations, which even may be desirable traits. It may be particularly problematic cross-culturally to apply categorical criteria developed in one culture to members of other cultures where relevant categorical thresholds may differ [17]. Religious interpretation of personality disorders in traditional societies may differ in its perception: Schizotypal features connote closeness to God, schizoid features to kind person—schizoid patients are usually described as “cold”, and “kind”, paranoid features to carefulness, avoidant has a religious connotation as such a behavior is praised by some conservative religious sectors who believe that mixing both sexes is an unacceptable and that shyness and avoidance are cherished and anankastic traits are perceived as meticulousness in following religious rituals [17]. The cultural impact on societal perception of mental symptoms in traditional and religious societies has an influence on what is a mental disorder and what is an accepted social trait, e.g. negative symptoms may connote deeper contemplation about God, while positive symptoms may be attributed as a gift from God for extraordinary perception [17].

Dimensional approaches

A dimensional concept (with a continuum from normal to pathological) was proposed for schizophrenia (schizothymic–schizoid–schizophrenia); and for affective disorders (cyclothymic temperament–cycloid ‘psychopathy’–manic-depressive disorder) [18]. The term ‘spectrum’ was first used in psychiatry in 1968 for the schizophrenia spectrum, which integrated schizoid personalities [19]. In 1977, Akiskal proposed a cyclothymic-bipolar spectrum, and in 1981 Klerman suggested a mania spectrum. A clear distinction should be made between ‘disorder’ and ‘diagnosis’. ‘Disorder’ means the clinical condition of a patient. ‘Diagnosis’ is a label used to represent information about that clinical condition. The reliability, validity, sensitivity, and specificity of a diagnosis all relate to the correspondence between the diagnosis and the disorder itself. Psychiatry has yet no “gold standard’ for identifying mental disorders [20]. Examples of dimensional approaches commonly used in contemporary psychiatry are the Hamilton Scale for Depression [21], the Positive and Negative Syndrome Scale (PANSS) for schizophrenia [22], and the seven-point Clinical Global Impressions Scale [23]. The top–down is the historical standard for psychiatry. In the top–down approach, experts deliberate consult; review their clinical experience, survey existing literature, and, in some cases, perform secondary data analyses of existing data to ultimately decide upon the criteria [20]. The bottom–up approach tends to be more objective as it depends on the reference population(s), in which the measures are developed, and the statistical assumptions made by those doing the analyses. A widely used example is the Childhood Behavior Checklist [17, 24]. It is due time to be more objective using the bottom up approach in our diagnostic system as currently many diagnoses depend on the top–down approach. The top–down approach depends on the experience and assumptions of the senior psychiatrists while the bottom–up approach depends on data gathered and analyzed from the patient’s population.

Advantages and disadvantages of categorical and dimensional approaches

Clinicians and researchers like to diagnose psychiatric conditions in a manner that is etiologically and therapeutically meaningful. Readiness of the field for dimensional diagnostic approaches is not consistent across all the psychiatric disorders represented in DSM, e.g. in the case of substance use disorders, where harmful use, dependence, addiction whether psychological or physical reflect a dimensional approach of severity. The measures of frequency of substance use provide a relatively straightforward way to incorporate a severity dimension, whereas psychoses do not have a convenient analogous proxy for severity. This does not demonstrate that psychoses cannot be measured in a dimensional fashion, but it illustrates a difficulty to be addressed in considering a shift from exclusively categorical definitions of psychiatric disorders to more continuous measures of psychopathology. The focus and concern of ICD and DSM has always been on “diagnoses”, that is, a clinical expert’s opinion as to whether some disorder is present in a particular patient. The purpose of any diagnostic system, such as DSM and ICD, is not to say what is “normal or abnormal” or what is or is not “acceptable” in any society. Nor is it an effort to “medicalize” society’s problems. In addition, the purpose of these classification systems is also not to channel clients to psychiatrists rather than to clinical psychologists, sociologists, or other mental health providers. DSM and ICD do not mention “insanity”, which is a legal rather than a medical term. Kendell and Jablensky [1] note that carefully defined categorical diagnostic criteria have resulted in significant improvements in diagnostic agreement (reliability) and communication, more precise criteria and research instruments based on definite criteria. The categorical diagnosis is an international reference providing a worldwide common language and public access to diagnostic definitions, thus improving communication with patients [17].

The disadvantages of a categorical classification system are many, e.g. the limitations, both clinical and statistical, imposed by labeling patients only based on whether their signs and symptoms collectively rise above a defined threshold. One example is that a diagnosis cannot be made, e.g. if the diagnostic criteria require a duration of at least 28 days, in a patient whose illness manifestations have so far lasted 27 days. This patient will not receive the diagnosis in question and is thus considered equivalent to someone who has never had that illness. But this patient receives a completely different diagnostic result compared to someone whose disease duration has reached 29 days [20]. Categorical criteria are important for determining which patients are sufficiently ill to justify treatment, as they specify the degree of severity and impairment that requires therapeutic intervention but dimensions are much better suited to understanding relationships between social and biological variables which reflect the psycho psychobiological approach of psychiatry [25].

Practicing clinicians are accustomed to adopt a dimensional perspective in the severity of illness, in order to develop a treatment plan and assess clinical progress [26], e.g. the severity of depression regarding intervention, ECT for psychotic depression, serotonin noradrenergic reuptake inhibitors (SNRI) for Melancholic depression while serotonin reuptake inhibitors (SSRI) for nonmelancholic depression are preferred. There are some benefits of dimensional diagnostic categories, namely providing a diagnosis-specific quantitative score, when desired, using a consistent methodology across studies or individual patients. A basic level of quantification increases statistical power without diminishing the utility of the categorical definitions [20]. There is still some controversy over dimensional classification systems considering the increased complexity in clinical communication associated with them. Dimensional classifications strive to rate severity both within and across areas of psychopathology. There may be simple severity scores for each syndrome, whether that syndrome rises to the level of categorical diagnosis or not. This may ensure that treatment efforts address the full range of current psychopathology. Dimensional classification may be potentially advantageous for a better understanding of public health and epidemiological data [27] as it will include all stages of illness whether with impairment or without and it will include so many disabling disorders which may change the prevalence in epidemiological studies like subthreshold disorders.

Conclusions

The addition of continuous, “dimensional” measures into the various diagnostic domains might help to resolve some of the critical taxonomic issues currently facing the field of mental health. It was overtly recognized that both categorical and dimensional approaches to diagnosis are important both for clinical work and for research, and that the ideal taxonomy would offer both. To avoid diagnostic chaos, the dimensional scale must reflect the categorical definition and the two must have a clear and obvious relationship to one another. The idea that psychiatrists should extend their attempts to produce wellness in their patients to those without diagnosed mental illness may be attractive at first glance but does not bear up to scrutiny. It is clear that the inclusion of subthreshold or subsyndrome psychiatric disorder should have a place in our current nosology. The degree of suffering, morbidity, and functional impairment is almost equal to that of full syndromal disorders. We do not need to widen the horizon of psychiatric diagnosis by having more entities and categories. The inclusion of those neglected disorders, which could not be accommodated in the current nosology whether ICD-10 or DSM-IV-RT should be attempted in the proposed dimensional system in ICD-11 and DSM-V so that we can be able to ameliorate the suffering of that group of neglected mental disorders. Significant ethical problems may arise in trying to medicalize common problems that are culturally mediated. We thus still face an ethical dilemma. Should we treat people who do not have a disorder by the current criteria of international classifications? Should we take “quality of life” and “disability” more into consideration in our diagnostic system? Should subthreshold disorders be included in the dimensional spectrum instead of adding more categorical disorders?