Transcript
Acta Psychiatr Scand IY98: 98: 200-207 Printed n UK all ri hts reserved Copyrighf Munksgaard 1998 ACTA PSYCHlATRlCA SCANDINAVICA ISSN 0001 690X A 16-year follow-up study of schizophrenia and related disorders in Sofia, Bulgaria Ganev K, Onchev G, Ivanov I A 16-year follow-up study of schizophrenia and related disorders in Sofia, Bulgaria. Acta Psychiatr Scand 1998: 98: 200-207. Munksgaard 1998. A total of 60 patients with functional non-affective psychoses were assessed 16 years after their inclusion in the WHO co-ordinated study on reduction and assessment of psychiatric disability All patients at inclusion had a recent onset of a psychotic disorder. About one-third of the patients had a good outcome. The rest showed moderate to severe psychiatric symptoms and social disability. Comparison with other similar studies suggested that our results show a low mortality rate, high levels of clinical symptoms, high levels of social disability and a low percentage of institutionalized patients. These findings are discussed in the context of the high level of family involvement I in patients’ care, which could reflect a cultural factor. Introduction Ever since the formulation of the concept of ‘dementia praecox’, its course and outcome have attracted considerable attention. There are two main reasons for this. First, the course of schizo- phrenia has been used as a criterion for the valida- tion of its entity. Secondly, the often chronic and disabling course of the disorder determines its high ‘social costs’ and its tremendous impact on the quality of life both of affected subjects and of their relatives. There have been numerous follow-up studies of schizophrenia l), although the majority of these have been confined to its short-term course. The relatively few long-term studies generated results which are often difficult to compare and interpret due to the following important methodological limitations (2, 3): (i) large number of subjects lost to follow-up; (ii) sample selection bias, with samples often being hospital based rather than geographi- cally defined, and with different illness duration; (iii) idiosyncratic assessment, rather than standard- ized assessment instruments and procedures; (iv) retrospective assessment of samples that involves difficulties in recall and information bias. A further limitation of the long-term studies of schizophrenia is that almost all of the studies meeting sufficiently stringent criteria for research have been conducted in Western European coun- tries and in the USA One of the major findings of the WHO co-ordinated studies (4, 5) is that short- K. Ganev G. Onchev P. lvanov WHO Collaborating Centre for Research and Training in Mental Health, Sofia, Bulgaria Key words schizophrenia. longitudinal study. outcome assessment Kimon Ganev, WHO Collaborating Centre for Research and Training in Mental Health, 15, Dimitar Nestorov Street, Floor 11, Room 26, 1431 Sofia. Bulgaria Accepted for publication February 12, 1998 term outcome of schizophrenia is more favourable for patients in ‘developing’ compared to ‘developed’ countries. Although the division of countries into ‘developing’ and ‘developed’ has been subject to considerable criticism (6-8), this finding does raise the issue of the relative contributions of ‘illness’ vs. ‘culture’ to the life course of subjects after a first episode of severe psychotic illness. The Maine and Vermont three-decade studies (9) demonstrate that similar groups of patients with schizophrenia can differ very substantially in their illness course and outcome. The very high percent- age of good outcomes in Vermont (10) is probably a result of the model rehabilitation programme at that location. This finding could be viewed as evidence supporting the need to investigate environ- mental factors that contribute to good or poor outcome for patients with psychotic disorders. Another important issue in outcome studies of schizophrenia is the definition of the concept of ‘outcome’. There seems to be agreement that at least two dimensions (which appear to be inde- pendent, although related) should be included in the study of outcome, namely psychopathology and social disability. However, there is considerable diversity in the ways in which these (and other) measures of outcome are assessed and reported. This further complicates the issue of comparability of results between studies. As illustrated by the study by Marengo et al. ll), overall comparisons 200 Schizophrenia and related disorders in Bulgaria could be performed, although inferences should be treated very cautiously, and opportunities for detailed analyses of course and outcome are extremely limited. One solution is to design a number of studies with the same instruments rep- resenting more than one way of approaching each of the outcome dimensions. The WHO co-ordinated studies on schizophrenia provide opportunities for looking at similar patient samples in several centres assessed with identical sets of instruments. This paper reports the data obtained from a follow-up study of a sample which is part of the International Study of Schizophrenia (ISoS). IsoS, a transcultural investigation by the WHO in 18 centres, was designed to examine patterns of long-term course and outcome of severe mental disorders in different cultures. The aim of this paper is to describe the long-term outcome of schizophrenia and related disorders in a region (and culture) where few epidemiological studies have been carried out. The ambiguous posi- tion of Bulgaria on the ‘developing-developed’ dimension provides further opportunities to explore the influences of cultural variables on the course of severe mental disorder. This paper addresses the relationships between different dimensions of out- come, namely psychopathological symptoms, psycho- social adjustment and living arrangements. Material and methods Study sample The sample was initially assessed as part of the WHO co-ordinated study on reduction and assess- ment of psychiatric disability (RAPyD). The inclu- sion criteria for the disability study were as follows: (i) age 15-44 years; (ii) residence in the defined catchment area; (iii) absence of organic brain dis- ease, severe mental retardation, severe sensory deficits, and alcohol or drug dependence; (iv) pres- ence of (a) at least one of the following: halluci- nations, delusions of non-affective type, bizarre or grossly inappropriate behaviour, thought and speech disorder other than retardation or acceleration; or (b) at least two of the following: psychomotor disorder other than simple retardation or overac- tivity, overwhelming fear or anxiety, marked social withdrawal, marked self-neglect; (v) ICD-9 clinical diagnosis of either schizophrenia (code 295), para- noid state (code 297), or other non-organic psycho- sis, paranoid, other or unspecified (codes 298.3, 298.4, 298.8, and 298.9). The initial selection procedure followed a com- plex design in an attempt to include in the disability study patients at various stages after the onset of a severe mental illness. Subsequently, the sample was subdivided into two groups, namely the retro- spective sample (with an onset of illness more than 2 years before the initial assessment) and a pro- spective sample (with an onset of illness less than 2 years prior to the initial assessment). Only the latter sample was included in the present study. This sample describing the Bulgarian part of the study consists of 60 patients (21 men and 39 women) with a mean age of 27.2 years (range 16- 45 years) at the initial assessment. The mean age of subjects at follow-up was 43.2 years. The selec- tion procedure included both patients and out- patients from defined catchment areas in Sofia, one area for men and a larger area for women, explain- ing the higher number of women in the sample. Due to the administrative organization of the catchment regions, the psychiatric units that formed the basis of the study recruited women from a larger region than the catchment area for men. Ten of the women selected for the study came from the extended catchment area for women. Howeyer, the socio-economic variables and illness characteristics of these women did not show any significant differences from those of the other female patients. All patients were initially assessed during the period 1978-1980. The last follow-up was during the period 1994-1995. Study instruments Initial assessment. Patients were interviewed with the Present State Examination, 9th edition (12), and were rated on the Psychological Impairments Rating Scale (13). Other schedules were completed in an interview with a key informant. These covered psychiatric history, personal and family background (Past History and Sociodemographic Description Schedule), and social disability (Disa- bility Assessment Schedule) (14). Follow-up assessment. The assessment consisted of an interview with the subject and an interview with a key informant. All available clinical records were reviewed, and treating psychiatrists were contacted in order to minimize the underestimation of symp- toms in the periods between follow-ups (15). Every effort was made to ensure that data were obtained from all sources (16). The instruments used in the follow-up were the Present State Examination, 9th edition (PSE-9), the Psychological Impairments Rating Scale (PIRS), the Scale for Assessment of Negative Symptoms (SANS), the Disability Assess- ment Schedule (DAS), the Life Chart Schedule (LCS), the Broad Rating Schedule (BRS) and the Family Interview Schedule (FIS). Details of the study instruments have been reported in a previous publication (17). The data reported in this paper 20 1 Ganev et al. were collected using the LCS, DAS and Global Assessment of Functioning Scale (GAF) (18). Presentation of data Percentage values are reported with 95 con- fidence intervals (CI) according to the recommen- dation of Gardner Altman (19). Approximate 95 CI were calculated with SYSTAT version 7.0, where appropriate. Results Tracing and interviewing Of the 60 patients, 3 patients were lost to follow- up, although two of them were known to be alive a few months before follow-up. A total of 57 subjects (95 ) were traced to their place of resi- dence or death. Two subjects have died (both committed suicide during the first 2 years of the follow-up period). Eight subjects (13.3 ) refused a full psychiatric interview, although a brief inter- view was conducted with some of them. Five subjects (8.3 ) were not contacted because their relatives prevented the research team from doing so. Nine subjects (15 ) were not interviewed after extensive efforts most of them could be classi- fied as covert refusers. In all cases of subjects without full interviews, important data could be obtained from interviews with key informants, their treating psychiatrists, and by means of case records. Extensive clinical case-notes for most of the sub- jects provided extremely valuable information about the longitudinal course of symptoms and psychosocial adjustment. All of the instruments were completed for 33 subjects. In total, 38 subjects had a completed SANS. All of the 55 subjects who were traced and alive had completed the DAS, LCS and BRS. The overall quality of the information collected can be summarized as follows: data for all or almost all items, 33 subjects (55 ); data for most of the items, 15 subjects (25 ); data for some of the items (no interview with patient and no interview with informant living with subject), 9 subjects (15 ); very limited information nly for a few items, 2 subjects (3.3 ); completely lost to follow-up, 1 subject (1.7 YO). Analyses in the Results section refer to the 55 subjects traced and alive at follow-up. Data from subjects who had died n=2) or were lost to follow- up (n=3) are not presented. Long-term outcome of schizophrenia and related disorders One of the issues that arises in describing the outcome in follow-up studies is determination of the time frame for the ‘outcome’ period. The traditional 1-month period for rating the present mental state (12) and the current level of disability (14) would reflect more transient phenomena, rather than durable characteristics of the subjects’ functioning. We therefore adopted a twofold approach. We present data both for the last month and for the last 2 years. The second period was chosen to give a picture of the predominant pattern of clinical and psychosocial outcome approximately 15 years after the onset of a psychotic disorder. Living arrangements and employment during the last 2 years. Table 1 shows the marital status, main household composition and employment status of subjects during the last 2 years. A very large majority of the subjects had spent the last 2 years living independently. In total, 51 subjects (of a total of 55) had lived outside institu- tions for at least 21 months during the last 2 years. The importance of this finding can be further clarified by the extent of family involvement in caring for subjects. Table 2 shows the duration of the period of independent living, and whether the family pro- vided a substitute for custodial care. The table shows that a substantial proportion of patients who had been categorized as ‘independent community living’ were cared for by their families. Relatives provided a substitute for custodial care for more than a third of the patients who had not Table 1 Marital status, living arrangements and employment status in the last 2 years n ( ) Marital status Never married Married Divorced or separated Living arrangements Alone With spouse and/or children With parent(s) With extended family With sibling In institution Otherlunknown Employment status Full-time job >1 year Full-time lob <1 year Part-time job >1 year Part-time job <1 year Housewife >1 year Disability pension Disability pension >1 year Disability pension <1 year 27 (49 1 Yo) 16 (29 1 ) 12 (21 8 ) 13 (23 6 ) 12 (21.8 ) 14 (25 5 ) 9(164 ) 1 (1 8 ) 2 (3 6 ) 4 (7 3 ) 12 (21.8 ) 2 (3 6 ) 6 10 9 ) 6 (10 9 ) 3 (5 5 ) 27 (49 1 Yo) 3 (5 5 ) 202 Schizophrenia and related disorders in Bulgaria Table 2 Period of time spent living independently in the community in the last 2 years Family provided substitute for custodial care Living independently in the community Not at all Some of the time Most of the time Total Throughout the period 26 9 5 40 (74 1 ) For less than half the period 3 3 (5 6 ) For most of the period 4 4 3 11 (20 4 ) Total 33 (61 1 ) 13 (24.1 ) 8 (14 8 ) 54 (100 ia aOata for one subject are missing stayed in an institution during the last 2 years. The overall results show that 38.9 (95 CI, 25.9-51.9 ) of the families were involved to a substantial degree in providing for the living arrangements of the subjects at follow-up. More than 50 of the subjects (n=30) had received a disability pension during the last 2 years (see Table 1). The large majority of these subjects had received a disability pension for more than 1 year. The proportion of subjects who could be classified as ‘self-reliant’, e.g. who had held a job (or had been full-time housewives) for extended periods of time, was 43.6 (95 CI, 30.5-56.7 ). Symptoms n the last 2 years. The clinical state of the subjects during the last 2 years is summarized in Table 3. The presence or absence of at least one overt or two highly suggestive psychotic symptoms was rated for the 2-year outcome period. The largest proportion of subjects (45.5 , n=25) had had continuous psychotic symptoms during the last 2 years. Nearly the same percentages of subjects (38.2 , n=21) had had no psychotic symptoms during the last 2 years, and a relatively small proportion (12.7 , n = 7) had had an episodic course (psychotic episodes and remissions of at least 6 months’ duration). The presence and severity of negative symptoms were associated with the prominence of psychotic symptoms. However, over 25 of the subjects without psychotic symptoms had had negative symptoms during the outcome period. The overall finding was that nearly two-thirds of the patients (n=35; 63.6 ; 95 CI, 50.9-76.3 ) had had nega- tive symptoms during the last 2 years of the follow- up period. Another dimension of subjects’ psychopathological functioning is violent behaviour towards others or oneself. During the last 2 years, three subjects (5.9 of the 51 subjects for whom reliable data were available) had assaulted others, and four (7.7 of the 53 subjects for whom reliable data were available) had made suicide attempts (none of which resulted in death). Subjects who had had continuous psychotic symptoms during the last 2 years had committed all of the assaults. However, Table 3 Course of psychotic illnessa n the last 2 years n( ) 95 CI Never psychotic in last 2 years Remission Mainly complete Mainly incomplete Negative symptoms No Mild Prominent Episodic courseb n last 2 years Remission Mainly complete Mainly incomplete Negative symptoms No Mild Prominent Continuous courseC n last 2 years Negative symptomsd No Mild Prominent Neither episodic nor continuous Negative symptoms No Mild Prominent Total 21 (38 2 ) 21 8-55 2 18 3 15 5 1 7(127 ) 35-272 3 4 4 1 25 (45 5 1 28 0-62.2 1 8 15 2 (3 6 ) 0-14 7 0 1 55 (100 ) aPsychotic illness is defined as the presence of at least one of the following hallucinations. delusions. marked thought disorder, marked catatonic symptoms. bizarre or grossly inappropriate behaviour. or two of the following extreme reduction of interests and initiative. extreme social withdrawal severe excitement episodes of intense (psychotic) anxiety Continuous course, no remission longer than 6 months Data for one sublect are missing Completeness of remission is described as the predominant pattern of presence or absence of residual symptoms bEpisodic course, none of the episodes lasted longer than 6 months 203 anev t al uo - 70- 2 60- two of the four suicide attempts were made by subjects with an episodic course of the disorder. Treatment utilization in the last 2 years. The treat- ment programmes largely available in the study area include neuroleptic treatment in the commu- nity, and hospital treatment. Most of the subjects had been on continuous neuroleptic treatment during the last 2 years (35 (64.8 ) of the 54 subjects for whom data were available, 95 CI, 46.8-79.0 ); 7 subjects (13 ; 95 CI, 3.9-27.1 ) had used neuroleptics occasionally and 12 subjects (22.2 ; 95 CI, 9.8-37.9 ) had not taken neuro- leptics; data for one subject were missing. Ten of the 12 subjects who had not taken neuroleptics during the last 2 years had not been psychotic during this period. However, a substantial propor- tion of symptom-free patients at outcome (12 of 21 subjects) were taking neuroleptics. The large majority of patients (27 of 32 subjects) with con- tinuous or episodic psychotic symptoms in the last 2 years had been taking neuroleptics for most of the time. Most of the subjects (n=41; 74.5 ; 95 CI, 63.0-86.1 ) had not been admitted to a psychi- atric hospital during the last 2 years. Eleven sub- jects had spent 3 months or less in a psychiatric hospital, one subject had spent 16 months in a hospital, and two subjects had been in supervised accommodation for people with mental disorders all the time. Current disability. The GAF disability scores are shown in Fig. 1. Half of the patients (n=28; 50.9 ; 95 CI, 32.8- 67.3 ) had GAF disability scores between 31 and 50, corresponding to serious and major impairment of several areas (occupation, social and family relationships). Five subjects (9.1 ; 95 CI, 1.7- 22.5 ) showed an inability to function in almost all areas. However, a substantial proportion of patients (n=12; 21.8 ; 95 CI, 9.2-37.9 ) showed no or only minimal impairment during the last month. These results are very similar to the global evaluation of disability on the Disability Assessment Schedule. Nearly a third of the subjects (n=17; 30.9 ; 95 CI, 16.5-47.2 ) showed excel- lent or good adjustment at outcome, 25 subjects (45.5 ; 95 CT, 28.7-61.6 ) showed ‘fair’ or ‘poor’ adjustment, and 13 subjects (23.6 ; 95 CI, 11.0-39.4 ) showed very poor adjustment or severe maladjustment. Current psychopathology. The GAF symptom scores are shown in Fig. 1. Subjects showed a variety of outcomes on measures of current psycho- pathology. The patients were uniformly scattered P- 1 9 11 20 21 30 31 40 41 50 51 60 61 70 71 80 81 90 GAF score Fig. I Distribution of scores on Global Assessment of Functioning Scale (symptoms and disability) at outcomes. on the dimension of symptom severity: 12 subjects (21.8 ; 95 CI, 9.2-37.9 ) had no or minimal symptoms on outcome, 14 subjects (25.4 ; 95 C1, 11.8-42.0 ) had moderate symptoms, 16 sub- jects (29 ; 95 CI, 14.5-45.9 ) had severe symp- toms, and 13 subjects (23.6 ; 95 CI, 10.5- 40.0 ) had very severe symptoms. Relationship between symptom severity and level o disability. Figure 2 shows the scatter plot of subjects according to psychopathological symptoms and disability. There was considerable overlap in outcome as assessed by measures of symptom severity and 1 11 1 11 1 2 1 1 2 I 1 I1 1 1 1 1 I 1 1 1 30 42 5 4 66 1R GAF disability score Fig. 2. Scatter plot of Global Assessment of Functioning scores (symptoms and disability). Note that figures in the scatter plot denote number of subjects. 204
