Diagnostic Stability of Psychiatric Disorders in Baqiyatallah Hospital from 1997 to 2015

Seyed Abbas Tavalaei, Shervin Assari, Vahid Tavalaei, Roghieh Nooripour * 1 Behavioral Sciences Research Center, Baqiyatallah University of Medical Sciences, Tehran, IR Iran 2 Department of Psychiatry, School of Medicine, University of Michigan, Ann Arbor, USA 3 Counseling, Department of Counseling, Faculty of Psychology & Education, Ardekan University, Yazd, IR Iran 4 Counseling Department, Faculty of Psychology & Education, Alzahra University, Tehran, IR Iran Corresponding Author: Roghieh Nooripour, Counseling Department, Faculty of Psychology & Education, Alzahra University, Tehran, IR Iran. Tel: +98-9118187318, Fax: +98-2185692580, Email: nooripour.r@gmail.com


Background
Oversight or error in diagnosis leads to insufficient intervention and treatment in many patients, and this delays disease improvement [1].Diagnosis error and changes in diagnosis were attributed to diversity of psychiatric symptoms, disorder similarity, the different course of each disease, and comorbid disorders.In many instances, a diagnosis is not possible until extensive medical and psychological evaluation has taken place [2]; therefore, it is essential that psychiatrists use diagnostic sources other than symptoms [3].
In some cases, patients are hospitalized in a psychiatric ward not only for treatment, but also for diagnosis.Inpatient admission may be valuable in clarifying diagnoses.Improvement in diagnosis after hospitalization is attributed to the accurate observation of the patient, the use of diagnostic tests and imaging (because of access to a laboratory and radiology), and more neuro-psychological examinations [4].Therefore, it can be claimed that hospitalization may lead to a better psychiatric diagnosis.Over one-third of ICD-10 F20 schizophrenia cases at three years had non-schizophrenia diagnoses at onset [5].Patients with manic symptomatology at the beginning had a very unstable and changeable course [6].A survey of patients admitted four or more times to the same acute care psychiatric hospital over a period of 3 years revealed that only 34% of such patients were discharged with the same diagnosis on each admission [7].In another study, the diagnosis index of nearly half of patients changed over a 4year period [8].It was found that the percentage of patients with changed diagnoses is highest at first readmission; at each later readmission, the diagnosis of manic-depressive psychosis changed in about 10% of bipolar and 25% of unipolar cases, and a similar number of diagnoses was changed from other diagnoses to manic-depressive psychosis [9].During a two-year observation period, half of patients were readmitted, and the stability of diagnosis rate was 60% [10].
In approximately 56% of patients, the initial diagnosis of depressive disorder eventually changed during follow-up mainly to disorders in the schizophrenia spectrum (16%), but also to personality disorders (9%), neurotic, stress-related, and somatoform disorders (8%), and bipolar disorder (8%).Among the 18% of patients who were later diagnosed with depressive disorder, 23% were initially diagnosed with adjustment disorder [11], and 46.4% of hospitalized patients had at least one previous hospitalization with a psychiatric diagnosis other than bipolar disorder [12].A large proportion of inpatients had their diagnoses altered, especially during hospitalization [13].

Methods
In this retrospective study, 908 psychiatry ward inpatient records from Baqiyatallah hospital were randomly selected from all records throughout the years 1996-2015.Having primary and final diagnoses was the inclusion criterion.Demographic variables (age, sex, marital status, education, employment), mental health variables (primary and final psychiatric diagnoses, duration of hospitalization, psychiatric history, and medication history) were recorded.Ultimately, 429 cases were entered into the study.
Stability of diagnosis was studied in cases including those with unchanged diagnoses from admission to discharge.The groups diagnosed were mood disorders, anxiety disorders, psychotic disorders, and personality disorders.The axis I diagnosis was divided into mood disorders, anxiety disorders, psychotic disorders, and other disorders.Personality disorders were all considered to be axis II diagnoses.The psychiatric diagnoses and diagnostic axis were recorded separately.
Data input and statistical analysis were accomplished using SPSS software.To describe the qualitative variables, frequency and relative frequency tables were used; to describe quantitative variables, mean, median, standard deviation, etc. were used.
Descriptive Statistics (frequency, percentage) and diagnostic analysis, which is a kind of regression analysis, were used to analyze data (P<0.05).
The stability of diagnosis showed no significant relation with marital status, education, being a soldier or official employee, suicidality, smoking, malingering, economical problems, electro-convulsion therapy (ECT) (p>0.05)(Table 3).

Discussion
In this study, the diagnostic stability (DS) for mood disorders was 84%.In another study, the DS for mood disorders was 68% [14].The DS for mood disorders 5 years after the primary diagnosis was 40% [15], and the 30-40 year follow-up of mood disorders showed DS to be 78.3%[6].In another report, the DS for mood disorders was 67%.28.9% of subjects with an initial diagnosis of BPD had their diagnoses changed, whereas 16.1% of subjects with non BPD diagnoses had their diagnoses changed to BPD at a later episode [16].
The DS for anxiety disorders in this study was 63.8%.Results of another study showed that anxiety disorders had the greatest DS [17].In yet another study, anxiety disorders had the lowest diagnostic stability rate [18].
The current study determined the DS for psychotic disorders to be 46.3%.Two other studies determined the DS of schizophrenia to be 74% and 67% [19].In another study, patients with an initial episode of schizophrenic psychosis showed the greatest DS (93%) [4].Psychotic disorder was the most stable.Among schizophrenic patients, higher stability rates appeared for residual and disorganized types [20].
In the current study, disturbed personality showed 36.4% DS.Two other studies, however, reported a 36% and a 73% DS for disturbed personality [14].12.5% of evaluated patients with borderline personalities had the same diagnosis at their first consultation as 5 years afterwards [21].Dysthymic disorder was more frequently changed to major depressive episode than any other disorder [22].The 6-12 month follow-up of schizotypal, borderline, avoidant, and obsessive-compulsive patients and the group of major depressive disorder patients (as control group) showed that more subjects in each personality disorder group remained at a diagnostic threshold throughout the 12-month follow-up period than did those in the major depressive disorder group, although the number of criteria present decreased over time in all groups [23].Another study tracked the individual criteria of DSM-IV personality disorders -borderline, schizotypal, avoidant, and obsessive-compulsive personality disorders -and how they changed over 2 years.This study showed that the most prevalent and least changeable criteria over 2 years were paranoid ideation and unusual experiences for schizotypal personality disorder, affective instability and anger for borderline personality disorder, feeling inadequate and feeling socially inept for avoidant personality disorder, and rigidity and problems delegating for obsessivecompulsive personality disorder.The least prevalent and most changeable criteria were odd behavior and constricted affect for schizotypal personality disorder, self-injury and behaviors defending against abandonment for borderline personality disorder, avoiding jobs that are interpersonal and avoiding potentially embarrassing situations for avoidant personality disorder, and miserly behaviors and strict moral behaviors for obsessive-compulsive personality disorder [24].

Classification of Different Psychological Disorders Based on Diagnostic Stability
In the current study, mood disorders and personality disorders had the most and least diagnostic stability, respectively.Anxiety disorders and psychotic disorders were between these ranges of stability.Different studies in this field have shown different results.In one study, high levels of stability were found for schizophrenia, moderate levels for affective disorders, low levels for other non-organic psychotic conditions and atypical psychosis, and very low levels for schizoaffective disorder and other conditions [25].In another study, the highest diagnostic stability was found in patients with a diagnosis index of alcohol abuse, schizophrenia, and drug abuse, while the lowest stability was found in patients with neurotic, hysterical, and depressive disorders, acute psychoses, and bipolar disorders [26].The most temporally consistent 6-month categories were schizophrenia, bipolar disorder, and major depression; the least stable were psychosis not otherwise specified, schizoaffective disorder, and brief psychosis [27].In one study, schizophrenia and mania were the most stable diagnoses and organic disorders had the most variable diagnoses [28].In another study, mobility was most marked for the neurosis group and was approximately uniform for other groups [29].
In the current study, a significant relation was identified between stability of diagnosis and diagnosis axis, number of diagnoses, drug abuse, and somatic disease history.Stability of diagnosis showed no significant relation with marital status, education, being a soldier or official employee, suicidality, smoking, malingering, economical problems, or electro-convulsion therapy.Age, gender, ethnicity, substance abuse, and disease severity had prominent roles in diagnostic changes; however, one study found no variables associated with diagnostic instability apart from the diagnoses themselves [30].

Conclusion
According to the present study, the highest diagnostic stability rates were found in affective disorders and anxiety disorders, respectively.The least diagnostic stability rate was found for personality disorders; in 50% of these patients, the diagnosis was changed to affective disorder.The variables of substance abuse and organic disease history were related to the instability of diagnosis.Future prospective studies in this field will be of assistance.

Table 1 .
Descriptive statistics for early diagnoses of mood, anxiety, psychotic, and secondary character recognition

Table 2 .
Descriptive statistics for stability of diagnosis in bipolar and depression secondary diagnoses

Table 3 .
Diagnostic stability relationship and variables Data in table are presented as Mean SD or No. (%).