In recent years, ATPDs have received increased attention in the psychiatric research setting. This attention may be prompted by an opportunity to improve classification during the revision of the Eleventh International Classification of Disease (ICD-11). To contribute to this effort, the aims of our study were to describe the clinical features of the index episode of ATPD in patients in Latvia, to analyse changes in the diagnosis longitudinally and to explore potential correlations between the sociodemographic characteristics of patients and their disease characteristics. We also tried to identify the longitudinal diagnostic stability of ATPD diagnosis in Latvia and correlate stressful life events before the first episode to the occurrence of the characteristics of the progression of the disease. We hope that our study results can help predict the development of disease and inform potential changes to ICD-11.
The sample at RCPAD can be regarded as a representative sample of the clinical inpatient population with ATPD in Latvia, as the hospital is the largest hospital in the country (with approximately 440 beds) and provides care for approximately 40% of the Latvian population. In a study carried out by this team in 2010, an average of 73 new ATPD cases a year were found, giving an incidence of 8.1 cases per 100,000 inhabitants . This incidence is high in comparison to the ‘Nottingham’ study, which had only 3.9 cases per 100,000 population , but closer to a Danish cohort study which identified an incidence of 9.6 per 100,000 population .
In agreement with previous studies, we found a higher prevalence of ATPD in females [2, 3, 21, 29, 30]. Women composed 60.7% of the population diagnosed with ATPD in this study. This is similar to the proportion observed by Aadamsoo et al. in Estonia (60.0%) . The average age at first psychotic episode in our study was higher in women than in men; this is also in agreement with previous studies, but this can be due to ‘gender X age’ effect .
There are only a few studies investigating the familial psychiatric morbidity of individuals with ATPD. Marneros and Pillmann reported a higher rate of mental disorders among family members of patients with ATPD than among the relatives of healthy controls, but no significantly increased frequency of psychotic disorders among these family members was found . In our study, only 15.6% (n = 16) the patients were found to have a family history of psychiatric disorder.
We also surveyed the level of formal education among this patient cohort. Similar to results reported by Marneros and Pillman, about one quarter of the patients in this study had completed higher education .
Almost a third of the patients were married. Interestingly, patients with a ‘pure’ ATPD diagnosis were significantly (p = 0.0001) more likely to be married than the other patients and composed 80% of the married patient group. Aadamsoo et al. found that 42.3% of the ATPD patients were married compared to only 11.4% of patients with a schizophrenia diagnosis . This may indicate that before the first episode of psychosis, the level of social functioning is higher among the group of patients with a ‘pure’ ATPD diagnosis than among those with a diagnosis which changes to schizophrenia.
During the first psychotic episode, the patients were treated in the hospital an average of 21.6 hospital bed days. This is similar to other ATPD studies. Interestingly, the patients whose ATPD diagnosis was later changed to schizophrenia were treated longer upon their first psychotic episode than patients from the ‘pure’ ATPD group (p = 0.004).
Just over 40% of patients were readmitted during the follow-up period. This could indicate a better prognosis for ATPD than for schizophrenia. If the ATPD diagnosis was changed, in 70.7% of cases, it was changed to a diagnosis of schizophrenia. This is similar to the data of Aadamsoo et al., in which 64.0% of changed ATPD diagnoses were changed to schizophrenia .
Although the follow-up period here averaged only 26.5 months, our previous retrospective study with a follow-up period of 6 years showed that diagnosis conversion took place within the first 2 years after initial hospitalisation for 81.0% of patients . Thus, we believe that our current study data captures a significant portion of the trends in diagnosis conversion for ATPD patients.
ATPD has lower rates of relapse in developing countries as compared to industrialised countries and has a relatively high diagnostic stability in Europe . The overall stability rate in our study was high (67.4%; 2.2-year follow-up period), and this is similar to those reported by Marneros and Pillman (54.0%; 4.7-year follow-up period), Jørgensen et al. (52.0%; 3-year follow-up period) and Castagnini et al. (48.4%; 5-year follow-up period) [2, 14, 29].
Aadamsoo et al. reported a lower stability rate at 34.0% (2-year follow-up period) . This may be due to the methodological differences between our studies. We used the methods similar to those in Castagnini and Berrios, which include patients who were not readmitted after the first episode, but Aadamsoo et al. did not include those patients in their calculations [4, 9].
The percentage of F23.0 patients in our study population (21.5%) was smaller than that described by Aadamsoo et al. (25.0%) in Estonia. However, the F23.1 diagnosis was much more frequent at 62.7% in the Latvian study compared to 29.0% in Estonia . This finding may demonstrate differences in diagnostic interpretations by the psychiatrists in the two neighbouring countries.
Studying the changes in diagnosis longitudinally for patients with F23.0 diagnoses vs. those with F23.1 diagnoses, we determined that patients with an F23.0 diagnosis were significantly more likely to have their diagnoses remain ‘pure’ ATPD. Similar results were found in both Japan and Estonia [4, 30]. This could play a significant role in the development of amendments to the new classification (ICD-11).
Some differences between the clinical features of the first episode of psychosis associated with a ‘pure’ ATPD diagnosis and those associated with ATPD which was later converted to schizophrenia were observed. The Latvian study shows that thought disorder was found in 58.6% of patients with ATPD that later developed schizophrenia (p = 0.002) with a statistically significant relevance (Spearman correlation value 0.326). This could be a strong statistically significant ‘predictor’ of ATPD diagnosis conversation to schizophrenia (in binary logistic regression, Wald's criterion was high 9.435). Thus, in the first-episode ATPD patients with thought disorder, the odds ratio of getting a diagnosis change to schizophrenia is 4.3 times higher than those in patients without thought disorder.
Studies of reactive psychosis (which is included in ICD-10 under ATPD diagnosis and DSM IV under brief psychotic disorder) were very popular in Scandinavia during the 1970s and 1980s [1, 31–36]. All of these studies identified a correlation between the onset of ATPD and stressful life events preceding the first episode of psychosis [2, 18, 21, 23]. We also found that a large proportion of patients in our study had experienced stressful life events during the 6 months prior to their first psychotic episode. Thus, our research supports the argument that stressful life events can be an important factor facilitating the development of this disease.
In SRRS, each one of the stressful life events is awarded a life change unit (LCU), depending on how traumatic it was felt to be by a large sample of participants. In this study, to compare SLE as a risk factor, we divided the SLE into two groups: (1) high (100–45) LCU level: death of significant other, separation/divorce, serious illness/operation and serious problems at work and (2) low (45–11) LCU level: change of job/school, major journey, relocation of residence and serious problems in family. We did not find any statistically significant difference between the groups . But, perhaps, the relatively small sample size precluded finding statistically significant differences between the groups of patients.
In the literature, there is a considerable number of reports on the correlation of premorbid personality with schizophrenia; however, there is a lack of research about correlation of ATPD with premorbid personality profile . Empirical data presented by Jørgensen et al. who, in a sample of 51 ATPD patients assessed with the International Personality Disorder Examination (IPDE) and found relatively high prevalence of personality disorders and ATPD .
Pillmann et al. in their study used the Neuroticism-Extroversion-Openness Five Factor Inventory (NEO-FFI) self-rating scale. They found no differences between healthy controls and ATPD patients (n = 42) . We used the ‘Mini-Mult’ scale by Kincannon, which assesses the personality profile more objectively than the NEO-FFI used in other studies . In our study, 17.6% had a personality profile within the norm and were from the ‘pure’ ATPD diagnosis group. This may indicate that before the first episode of psychosis, the level of social functioning is higher among the group of patients with a ‘pure’ ATPD diagnosis than among those with a diagnosis which changes to schizophrenia.
A large portion of ATPD patients showed deviations from the norm in personality profiles, similar to that found by Jørgensen et al., where 63% of ATPD patients also had a prevalence of personality disorders . The Latvian results are similar those of other studies, but when we tried to compare ‘Mini-Mult’ scores of the ‘pure’ ATPD group and the group whose diagnosis later converted to schizophrenia, we could not find any statistically significant difference .
The limitations of this study are similar to those of other studies assessing. No structured life event scale was used, and life event data may have been influenced by recall bias, as was demonstrated in the ‘Nottingham’ study . One of the most important limitation is possible bias due to differences in interview methods, sampling or clinical assessment which will affect diagnostic stability and outcomes for ATPD subtype results . The limitations of the personal profile results are intrinsic to the ‘Mini-Mult’ method. Some authors argue that the ‘Mini-Mult’ shows success in screening, but the full MMPI is preferable for screening descriptive features, and personality assessment of patients occurs after the first psychotic episode . In the Latvian study, the sample size was rather small, and the follow-up period was shorter than those in other studies. Due to this, the study may be underpowered for finding statistically significant differences between the groups of patients.