- Email: [email protected]
Trends in schizophrenia
851
LETTERS to the EDITOR
Trends in
schizophrenia
3.
SiR,—Mr Der and his colleagues (March 3, p 513) review data on hospital admissions in the UK with a diagnosis of schizophrenia in the past two decades and ask "Is schizophrenia disappearing?". In suggesting a positive answer they consider alternative explanations and write that "to exclude these conclusively, and to exclude other possible artefacts, it would have been necessary to study schizophrenia diagnosed with consistent criteria in all individuals in a defined population". In his review of time trends in schizophrenia Hafnerl concluded that only three" of the nine studies considered did not have considerable shortcomings. These three studies (two in Norway and one in Australia) covered time periods of 40, 63, and 120 years, and all three concluded that age-corrected admission rates for schizophrenia had been fairly stable. 0degaard2 wrote that "the astonishing stability of first admission rates over half a century indicates that in the countries investigated the true incidence of schizophrenia remained essentially unchanged". Given that schizophrenia occurs worldwide, and that when the clinical syndrome is restrictively defined incidence rates are very sirnilar5 across wide cultural, climatic, and industrial variations, it would be surprising if there were real and substantial variations over short periods within one country. One problem is the definition of schizophrenia. Over the time span considered by Der et al there has been an increase in interest in the reliability of diagnosis and an emphasis on operational and restrictive criteria. This is very likely to have led to greater caution in applying the diagnosis. A feature of Der’s table 11 is that numbers in the category "other psychoses" have remained high in the face of a decrease in admissions (presumably attributable to increases in community care) in almost every other category except senile and presenile dementia. At the end of the period the first admission rate in this category (13 per 100 000) is higher than rates for affective psychoses (9) and schizophrenia (8 to 9). There seems to be plenty of room here (or, indeed, in the surprisingly high rate of 33 for "all other conditions") to accommodate the more equivocal cases of schizophrenia, especially if there is a true continuumb that includes schizophrenic, affective, and intermediate psychoses. Another change affecting the past three decades or so has been neuroleptic medication. These drugs have improved the outcome of psychosis, and there is a plausible case’ for claiming that they have permitted the steady decline in the mental hospital population over this period. It would be surprising if these drugs had not also ameliorated some of the most severe outcomes, the infrequency of which Der et al adduce as additional evidence for a decrease in incidence. That neuroleptic drugs or any simple environmental influence have drastically altered the incidence of psychosis seems improbable. Is it not more likely that psychosis is a fairly constant accompaniment of the human genetic condition, but that our ways of subdividing its varied manifestations are subject to fashion? Division of Psychiatry, Clinical Research Centre, Northwick Park Hospital, Harrow HA1 3UJ, UK
T.
4.
Astrup C. The increase of mental disorders. National Case Register of Mental Disorders, Gaustad Hospital, Oslo, 1982. Krupinski J, Alexander L. Patterns of psychiatric morbidity in Victoria, Australia in relation to changes in diagnostic criteria 1848-1978. Soc Psychiatry 1983; 18: 61-67.
N, Jablensky A, Korten A, et al. Early manifestations and first-contact incidence of schizophrenia in different cultures. Psychol Med 1986; 16: 909-28. 6. Editorial. A continuum of psychosis. Lancet 1987; ii: 889-90. 7. Office of Health Economics. Mental health in the 1990’s: from custody to care? London: Office of Health Economics, 1989. 5. Sartorius
SiR,—Mr Der and his colleagues report a substantial reduction in first-admission rates for schizophrenia and suggest that this trend indicates a true decline in the incidence of schizophrenia. They argue that if the reduction in admissions is to be accounted for by a tendency for psychiatric patients to be treated in the community a parallel fall in admissions for other conditions would be expected. Reference to Der and colleagues’ statistical sources suggests that this is just what happened. Data for 1970-86 show that first admission rates for schizophrenia fell by 40% for men and 50% for women, accompanied by comparable falls for affective psychosis (36%), neurotic disorders (54%), and personality disorders (25%). Der et al state that 21 % of all psychiatric first admissions in 1952 were due to "schizophrenia and related paranoid conditions", compared with 9% in 1986. However, pre-1970 statistics are not directly comparable with later ones, and the decline in first admissions did not begin until the late 1960s. For these reasons it is more instructive to compare the first-admission rates for schizophrenia with those for other diagnoses for all the years 1970 to 1985 (table). The table
suggests
that the decline in admissions for
schizophrenia needs to be seen in the context of a similar fall for other diagnoses. This relation is even more obvious if admissions for alcohol-related conditions are discounted, an exclusion that seems justified because the category of alcohol-related conditions is the only one to show an increase in admissions, presumably related to an increase in alcohol consumption. In our view much of the general reduction in first-admission rates may have resulted from changes in attitudes of the mentally ill, their families, and their doctors as to what constitutes suitable grounds for admission to a psychiatric unit. A similar controversy exercised the minds of physicians a century ago. The question then was whether the steady increase in admissions to asylums reflected a true increase in lunacy. The "nosocomialists" argued that factors other than the disease itself determined whether a person with that disease was included on a FIRST ADMISSIONS DUE TO ALL PSYCHIATRIC ILLNESS AND TO SCHIZOPHRENIA, ENGLAND AND WALES, BOTH SEXES
J. CROW
1. Hafner H. Epidemiology of
2.
schizophrenia. In: Hafner H, Gattaz WF, Janzarik W, eds. Search for the causes of schizophrenia. Berlin: Springer-Verlag 1987; 47-74. Ødegaard Ø. Hospitalised psychoses in Norway: time trends 1926-1965. Soc Psychiatry 1971; 6: 53-58.
Percentages in italics refer to admissions due to alcohol-related admissions
schizophrenia
but
discounting
852
register of cases. Prominent in that school of thought was D. Hack Tuke, who identified the following factors which tended to exaggerate the incidence of insanity:2 "an ever increasing encroachment on the mass of unregistered lunacy", "a large exodus of patients from workhouses and the care of relatives to country asylums", and "the value and comfort of asylums was increasingly appreciated". As nosocomialists of the 1990s we suggest that these three tendencies, operating in reverse, may have accounted for the apparent decline in the incidence of schizophrenia over the past two decades. Maudsley Hospital, London SE5, UK
MARTIN J. PRINCE MICHAEL C. PHELAN
1. Hare EH. Was insanity on the increase? Br J Psychiatry 1983; 142: 439-55. 2. Tuke DH. Alleged increase of insanity. Ment Sci 1894; 40: 219-31. J
SIR,-Mr Der and colleagues reject better outpatient care as an explanation for their findings and do not mention neuroleptic drugs. These agents, especially those in depot form, are generally accepted as having been a major factor contributing to the success of outpatient treatment of psychosis. If in earlier years general practitioners and outpatient psychiatrists had used ineffective lower dosage regimens and more effective doses recently, first-admission rates would be expected to diminish. In Australia, where a decline in
SIR,-Mr Der and colleagues have used data from the Mental Enquiry to record a decline in first-admission rates for schizophrenia. They conclude that this is a real fall in incidence, though alternative explanations are considered, including the possibility that the decline might reflect a shift towards treatment of schizophrenic patients as outpatients or that there has been a shift in the time during the illness when diagnosis is made, from the first to subsequent admissions. In the Mental Health Enquiry "first admission" refers to inpatient care regardless of diagnosis. A diagnosis of schizophrenia made during a first admission would be counted as a "first admission" with schizophrenia but someone given a diagnosis of schizophrenia for the first time at a second or later admission would be identified as a "readmission" with schizophrenia. Such Health
deferments would reduce the apparent number of new cases. A shift towards outpatient care or a postponement of diagnosis merits special consideration because of the profound changes in psychiatric care during the period studied by Der et al. The changes included the use of long-acting phenothiazines to maintain schizophrenic patients in the community, open hospital and therapeutic community policies, shorter admissions, and increased precision in making the diagnosis of schizophrenia. 1,2
first-admission rates has also been observed,! the usage of flupenazine decanoate (’Modecate’) offers support to these ideas. Until 1987 modecate was the only depot neuroleptic readily available. The manufacturers (E. R. Squibb & Sons Pty) tell me that sales rose from 560 kg in 1975 to 1552 kg in 1986. Of sales in 1975, 10% were to dispensing chemists (primarily outpatients) and 90% to hospitals, but by 1986 40% of sales were to dispensing chemists and only 60% to hospitals. Der et al found no change in the incidence of mania and refute the suggestion that mania has been diagnosed in cases where formerly schizophrenia would have been the diagnosis. However, better community care might have masked the increase in mania expected with a change in diagnosis. Cookson et al2 have observed that, with adequate doses of neuroleptics, manic patients usually show a much earlier resolution of symptoms than do schizophrenics. Knowledge of this and increasing familiarity with higher and more effective doses of neuroleptics may have encouraged clinicians to attempt outpatient treatment of mania more frequently than schizophrenia, so allowing first-admission rates of mania to remain unchanged. The validity of using first-admission rates as representing total incidence of schizophrenia requires comment. Der and colleagues cite Cooper et al, who have produced good evidence to support this contention, using figures from one area and over a 2-year period (1978-80). Der et al, however, used figures from many different areas and reviewed a 34-year period. Admissions policies have probably varied from time to time and place to place-eg, at Napsbury Hospital in Hertfordshire, R. D. Scott and his team in the late 1960s and early 1970s pursued a policy of not admitting except in extreme cases.3 They had observed the negative effects of admission and noted their ability to offer alternative domiciliarybased approaches. Similar approaches were adopted by other area teams. The availability of beds has changed in some areas, smaller numbers usually being available. First-admission rates may represent different proportions of the total incidence at different times and in different places, with figures from recent years probably representing a lesser proportion of cases than in earlier years. Bentley Clinic, Perth, Western Australia 6107
PETER M. GRAHAM
1. Parker
G, O’Donnell M, Walter S. Changes in the diagnosis of the functional psychoses associated with the introduction of lithium. Br J Psychiatry 1985; 146: 377-82.
2 Cookson JC, Silverstone T, Rees L. Plasma prolactin and growth hormone levels in manic patients treated with pimozide. Br J Psychiatry 1982; 140: 274-79. 3. Scott RD. Cultural Frontiers in the Mental Health Service. Schizophrenia Bulletin
1974; (fall) No. 10: 58-73.
Age standardised rates for first-ever recorded diagnosis of schizophrenia (-)and schizophrenia at first-ever psychiatric contact (- - - -).
Xortrendmma!es.————=12(p<0 01),- - - -=16(p<001) -X2=13(p005),- - -
853
Linkage Study data include all contacts with specialist psychiatric services, and because data are linked sequences of episodes of the care of individuals, schizophrenia at first-ever contact can be distinguished from schizophrenia first diagnosed at a later stage. Truncated age-standardardised rates for people aged 15 and over, standardised by the direct method with the 1981 population of Oxfordshire as the standard, are shown in the figure. The total number of people with a first-recorded diagnosis of schizophrenia was much higher than the number diagnosed as schizophrenia at first-ever psychiatric contact alone. In males both these measures, and first-ever inpatient admission rates alone, showed a significant decline in the mid-to-late 1970s similar to that reported by Der et al. The pattern for females was much less clear-cut. Further explanations for the decline in first-contact rates for males include changes in diagnostic fashion, any increased tendency for general practitioners to treat schizophrenics without referral to a specialist, a decrease in severity, and (in local studies) any tendency for people predisposed to schizophrenia to migrate out of the population studied or any increase in inward migration by people not predisposed. Our data add some support to the suggestion that schizophrenia has declined in males, though not in females, in the period 1975-86. Oxford Record
In our view, however, the evidence is not conclusive. We do have comparable data for years before 1975. Unit of Clinical Epidemiology, Oxford University, Oxford Regional Health Authority, Oxford OX3 7LF, UK
not
JACQUELINE ALARCON VALERIE SEAGROATT MICHAEL GOLDACRE DE
1. World Health
Organisation The international pilot study of schizophrenia. Geneva. WHO, 1973. 2. Cooper JE, Kendell RE, Garland BJ, Sharpe L, Copeland JRM, Simon R. Psychiatric diagnosis in New York and London: a comprehensive study of mental hospital admissions (Maudsley Monogr no 20). London: Oxford Univesity Press, 1972.
SiR,—Elsewhere I argue that Kraepelin’s and Bleuler’s concepts of dementia praecox and schizophrenia were mistakenly derived from a population suffering mainly from physical disorders, notably encephalitis lethargica and its sequel, post-encephalitic parkinsonism. The years following von Economo’s descriptions of this infection and its consequences (after Kraepelin and Bleuler had completed their major writings) were characterised by diagnostic confusion; clinicians diagnosed schizophrenia but found it difficult to distinguish the condition from post-encephalitic parkinsonism. Efforts at differentiation were marked by contradiction and non-specificity. More important, they were based on the unfounded assumption that Kraepelin and Bleuler had provided evidence in support of their concepts. In other words, the task of differential diagnosis begged the question of the validity of "dementia praecox" and "schizophrenia".2 The suggestion that Kraepelin and Bleuler were dealing, at least in part, with the neurological sequelae of infectious diseases is very different from the hypothesis that schizophrenia is an infectious disease.3,4That claim, too, is based on an unjustified confidence in the validity of Kraepelin’s and Bleuler’s concepts. My suggestion is that the original concepts of dementia praecox and schizophrenia were mistakenly inferred from at least one now well-described infectious disease and its sequelae. It is difficult to know what implications this claim would have for subsequent rates of schizophrenia diagnosis. Reliable figures for the prevalence of the virus assumed to be implicated in encephalitis lethargica are not available, although it is generally believed to have declined sharply after the epidemic of 1916-27.5 von Economo, however, thought that the virus was passed from mother to fetus, and Sacks has cautioned against the naive assumption that such viruses "disappear" after highly visible epidemics.6 Also the concept of schizophrenia has gradually been transformed from the strongly physical and neurological, as well as behavioural, concept of Kraepelin and Bleuler, to the more behavioural/experiential one of DSM-IIIR. It is quite possible that many patients today who are said to have schizophrenia only superficially resemble Kraepelin’s
and Bleuler’s cases. Nonetheless I suspect that the apparent decline in incidence of schizophrenia is not unrelated to the decline in post-encephalitic parkinsonism. Questions such as "Where have all the catatonic (and severely deteriorated) schizophrenics gone?" may have answers that are at least similar to those to the question "Where have all the cases of post-encephalitic parkinsonism gone?" Department of Psychology, Polytechnic of East London,
MARY BOYLE
London E15 4LZ, UK
1. Boyle M. Schizophrenia: a scientific delusion? London: Routledge, 1990. 2. Boyle M. Is schizophrenia what it was? A re-analysis of Kraepelm’s and Bleuler’s populations. J Hist Behav Sci (in press). 3. Crow TJ. A re-evaluation of the viral hypothesis: is psychosis the result of retroviral integration at a site close to the cerebral dominance gene? Br J Psychiatry 1984; 145: 243-53. 4. Hare EH. Schizophrenia as an infectious disease In: Kerr A, Snaith P, eds. Contemporary issues in schizophrenia. London Royal College of Psychiatrists/ Gaskell, 1986. 5. Pallis CA. Parkinsonism: natural history and clinical features Br Med J 1981; iii. 683-90. 6. Sacks O. Parkinsonism: a so-called new disease? Br Med J 1971; iii: 111.
Coexistence of Raynaud’s syndrome and
erythromelalgia SiR,—Erythromelalgia is known as the inverse Raynaud’s syndrome; the former disease can be induced by treatment of the latter. We report a woman with Raynaud’s syndrome and erythromelalgia. A 40-year-old woman had had, for 5 years, episodes of pallor and blue discolouration of the fingers and toes after cooling and emotional exertion. During these 5 years she also had intolerance to warming of the feet, which manifested attacks of pulsating pain and burning sensations. She had about ten cold and five to seven hot induced paroxysms daily. There were no trophic changes or oedema of the fingers and toes. Distal pulsation remained. Adson’s and Allen’s tests were negative. A cold provocation test induced pallor and blue discolouration of the fingers and toes. A hot provocation test induced pulsating pain and red discolouration of the feet. Biochemical investigation revealed accelerated ADPinduced platelet aggregation. The patient was otherwise healthy. She was given nifedipine (40 mg daily), dihydroergocristine (1-5mg daily), aspirin (750 mg daily), and hyperbaric oxygen sessions (1 ’7 atmospheres absolute, 45 min, seven sessions). After 3 weeks she was discharged and no longer had either hot or cold induced paroxysms.
I cannot explain with certainty how or why the two conditions started together. Their coexistence may have resulted from the dysfunction of arteriovenous anastomoses which are implicated in both diseases. Pathological expansion of the anastomoses while small arteries and arterioles are in cold spasm leads to an attack in Raynaud’s disease, whereas in a warm environment such expansion
might cause erythromelalgia. Kuibyshev Medical Institute, Post Box 703, Post Office 100, Kuibyshev 443100, USSR
G. E. SLUTSKER
Short-term continuous infusion of mitozantrone for advanced breast cancer SIR,-Professor Harris and colleagues (Jan 27, p 186) compare four cycles of mitozantrone with continued treatment in responding breast cancer patients, and record similar disease-free survival in the two groups. Their findings are corroborated by our results in patients with non-pretreated breast cancer, who received continuous infusion of mitozantrone (11mg/m2 daily) over 2 weeks, repeated every 4 weeks for four cycles-a schedule chosen, on the basis of an earlier phase-I study,’ to keep side-effects to a minimum while maintaining optimum efficacy. Ten patients were investigated (mean age 53 years, range 36-68), and two had received cyclophosphamide/methotrexate/5-fluorouracil adjuvant therapy. Responses were complete in two (9 and over 19 months), partial in
LETTERS to the EDITOR
Trends in
schizophrenia
3.
SiR,—Mr Der and his colleagues (March 3, p 513) review data on hospital admissions in the UK with a diagnosis of schizophrenia in the past two decades and ask "Is schizophrenia disappearing?". In suggesting a positive answer they consider alternative explanations and write that "to exclude these conclusively, and to exclude other possible artefacts, it would have been necessary to study schizophrenia diagnosed with consistent criteria in all individuals in a defined population". In his review of time trends in schizophrenia Hafnerl concluded that only three" of the nine studies considered did not have considerable shortcomings. These three studies (two in Norway and one in Australia) covered time periods of 40, 63, and 120 years, and all three concluded that age-corrected admission rates for schizophrenia had been fairly stable. 0degaard2 wrote that "the astonishing stability of first admission rates over half a century indicates that in the countries investigated the true incidence of schizophrenia remained essentially unchanged". Given that schizophrenia occurs worldwide, and that when the clinical syndrome is restrictively defined incidence rates are very sirnilar5 across wide cultural, climatic, and industrial variations, it would be surprising if there were real and substantial variations over short periods within one country. One problem is the definition of schizophrenia. Over the time span considered by Der et al there has been an increase in interest in the reliability of diagnosis and an emphasis on operational and restrictive criteria. This is very likely to have led to greater caution in applying the diagnosis. A feature of Der’s table 11 is that numbers in the category "other psychoses" have remained high in the face of a decrease in admissions (presumably attributable to increases in community care) in almost every other category except senile and presenile dementia. At the end of the period the first admission rate in this category (13 per 100 000) is higher than rates for affective psychoses (9) and schizophrenia (8 to 9). There seems to be plenty of room here (or, indeed, in the surprisingly high rate of 33 for "all other conditions") to accommodate the more equivocal cases of schizophrenia, especially if there is a true continuumb that includes schizophrenic, affective, and intermediate psychoses. Another change affecting the past three decades or so has been neuroleptic medication. These drugs have improved the outcome of psychosis, and there is a plausible case’ for claiming that they have permitted the steady decline in the mental hospital population over this period. It would be surprising if these drugs had not also ameliorated some of the most severe outcomes, the infrequency of which Der et al adduce as additional evidence for a decrease in incidence. That neuroleptic drugs or any simple environmental influence have drastically altered the incidence of psychosis seems improbable. Is it not more likely that psychosis is a fairly constant accompaniment of the human genetic condition, but that our ways of subdividing its varied manifestations are subject to fashion? Division of Psychiatry, Clinical Research Centre, Northwick Park Hospital, Harrow HA1 3UJ, UK
T.
4.
Astrup C. The increase of mental disorders. National Case Register of Mental Disorders, Gaustad Hospital, Oslo, 1982. Krupinski J, Alexander L. Patterns of psychiatric morbidity in Victoria, Australia in relation to changes in diagnostic criteria 1848-1978. Soc Psychiatry 1983; 18: 61-67.
N, Jablensky A, Korten A, et al. Early manifestations and first-contact incidence of schizophrenia in different cultures. Psychol Med 1986; 16: 909-28. 6. Editorial. A continuum of psychosis. Lancet 1987; ii: 889-90. 7. Office of Health Economics. Mental health in the 1990’s: from custody to care? London: Office of Health Economics, 1989. 5. Sartorius
SiR,—Mr Der and his colleagues report a substantial reduction in first-admission rates for schizophrenia and suggest that this trend indicates a true decline in the incidence of schizophrenia. They argue that if the reduction in admissions is to be accounted for by a tendency for psychiatric patients to be treated in the community a parallel fall in admissions for other conditions would be expected. Reference to Der and colleagues’ statistical sources suggests that this is just what happened. Data for 1970-86 show that first admission rates for schizophrenia fell by 40% for men and 50% for women, accompanied by comparable falls for affective psychosis (36%), neurotic disorders (54%), and personality disorders (25%). Der et al state that 21 % of all psychiatric first admissions in 1952 were due to "schizophrenia and related paranoid conditions", compared with 9% in 1986. However, pre-1970 statistics are not directly comparable with later ones, and the decline in first admissions did not begin until the late 1960s. For these reasons it is more instructive to compare the first-admission rates for schizophrenia with those for other diagnoses for all the years 1970 to 1985 (table). The table
suggests
that the decline in admissions for
schizophrenia needs to be seen in the context of a similar fall for other diagnoses. This relation is even more obvious if admissions for alcohol-related conditions are discounted, an exclusion that seems justified because the category of alcohol-related conditions is the only one to show an increase in admissions, presumably related to an increase in alcohol consumption. In our view much of the general reduction in first-admission rates may have resulted from changes in attitudes of the mentally ill, their families, and their doctors as to what constitutes suitable grounds for admission to a psychiatric unit. A similar controversy exercised the minds of physicians a century ago. The question then was whether the steady increase in admissions to asylums reflected a true increase in lunacy. The "nosocomialists" argued that factors other than the disease itself determined whether a person with that disease was included on a FIRST ADMISSIONS DUE TO ALL PSYCHIATRIC ILLNESS AND TO SCHIZOPHRENIA, ENGLAND AND WALES, BOTH SEXES
J. CROW
1. Hafner H. Epidemiology of
2.
schizophrenia. In: Hafner H, Gattaz WF, Janzarik W, eds. Search for the causes of schizophrenia. Berlin: Springer-Verlag 1987; 47-74. Ødegaard Ø. Hospitalised psychoses in Norway: time trends 1926-1965. Soc Psychiatry 1971; 6: 53-58.
Percentages in italics refer to admissions due to alcohol-related admissions
schizophrenia
but
discounting
852
register of cases. Prominent in that school of thought was D. Hack Tuke, who identified the following factors which tended to exaggerate the incidence of insanity:2 "an ever increasing encroachment on the mass of unregistered lunacy", "a large exodus of patients from workhouses and the care of relatives to country asylums", and "the value and comfort of asylums was increasingly appreciated". As nosocomialists of the 1990s we suggest that these three tendencies, operating in reverse, may have accounted for the apparent decline in the incidence of schizophrenia over the past two decades. Maudsley Hospital, London SE5, UK
MARTIN J. PRINCE MICHAEL C. PHELAN
1. Hare EH. Was insanity on the increase? Br J Psychiatry 1983; 142: 439-55. 2. Tuke DH. Alleged increase of insanity. Ment Sci 1894; 40: 219-31. J
SIR,-Mr Der and colleagues reject better outpatient care as an explanation for their findings and do not mention neuroleptic drugs. These agents, especially those in depot form, are generally accepted as having been a major factor contributing to the success of outpatient treatment of psychosis. If in earlier years general practitioners and outpatient psychiatrists had used ineffective lower dosage regimens and more effective doses recently, first-admission rates would be expected to diminish. In Australia, where a decline in
SIR,-Mr Der and colleagues have used data from the Mental Enquiry to record a decline in first-admission rates for schizophrenia. They conclude that this is a real fall in incidence, though alternative explanations are considered, including the possibility that the decline might reflect a shift towards treatment of schizophrenic patients as outpatients or that there has been a shift in the time during the illness when diagnosis is made, from the first to subsequent admissions. In the Mental Health Enquiry "first admission" refers to inpatient care regardless of diagnosis. A diagnosis of schizophrenia made during a first admission would be counted as a "first admission" with schizophrenia but someone given a diagnosis of schizophrenia for the first time at a second or later admission would be identified as a "readmission" with schizophrenia. Such Health
deferments would reduce the apparent number of new cases. A shift towards outpatient care or a postponement of diagnosis merits special consideration because of the profound changes in psychiatric care during the period studied by Der et al. The changes included the use of long-acting phenothiazines to maintain schizophrenic patients in the community, open hospital and therapeutic community policies, shorter admissions, and increased precision in making the diagnosis of schizophrenia. 1,2
first-admission rates has also been observed,! the usage of flupenazine decanoate (’Modecate’) offers support to these ideas. Until 1987 modecate was the only depot neuroleptic readily available. The manufacturers (E. R. Squibb & Sons Pty) tell me that sales rose from 560 kg in 1975 to 1552 kg in 1986. Of sales in 1975, 10% were to dispensing chemists (primarily outpatients) and 90% to hospitals, but by 1986 40% of sales were to dispensing chemists and only 60% to hospitals. Der et al found no change in the incidence of mania and refute the suggestion that mania has been diagnosed in cases where formerly schizophrenia would have been the diagnosis. However, better community care might have masked the increase in mania expected with a change in diagnosis. Cookson et al2 have observed that, with adequate doses of neuroleptics, manic patients usually show a much earlier resolution of symptoms than do schizophrenics. Knowledge of this and increasing familiarity with higher and more effective doses of neuroleptics may have encouraged clinicians to attempt outpatient treatment of mania more frequently than schizophrenia, so allowing first-admission rates of mania to remain unchanged. The validity of using first-admission rates as representing total incidence of schizophrenia requires comment. Der and colleagues cite Cooper et al, who have produced good evidence to support this contention, using figures from one area and over a 2-year period (1978-80). Der et al, however, used figures from many different areas and reviewed a 34-year period. Admissions policies have probably varied from time to time and place to place-eg, at Napsbury Hospital in Hertfordshire, R. D. Scott and his team in the late 1960s and early 1970s pursued a policy of not admitting except in extreme cases.3 They had observed the negative effects of admission and noted their ability to offer alternative domiciliarybased approaches. Similar approaches were adopted by other area teams. The availability of beds has changed in some areas, smaller numbers usually being available. First-admission rates may represent different proportions of the total incidence at different times and in different places, with figures from recent years probably representing a lesser proportion of cases than in earlier years. Bentley Clinic, Perth, Western Australia 6107
PETER M. GRAHAM
1. Parker
G, O’Donnell M, Walter S. Changes in the diagnosis of the functional psychoses associated with the introduction of lithium. Br J Psychiatry 1985; 146: 377-82.
2 Cookson JC, Silverstone T, Rees L. Plasma prolactin and growth hormone levels in manic patients treated with pimozide. Br J Psychiatry 1982; 140: 274-79. 3. Scott RD. Cultural Frontiers in the Mental Health Service. Schizophrenia Bulletin
1974; (fall) No. 10: 58-73.
Age standardised rates for first-ever recorded diagnosis of schizophrenia (-)and schizophrenia at first-ever psychiatric contact (- - - -).
Xortrendmma!es.————=12(p<0 01),- - - -=16(p<001) -X2=13(p
853
Linkage Study data include all contacts with specialist psychiatric services, and because data are linked sequences of episodes of the care of individuals, schizophrenia at first-ever contact can be distinguished from schizophrenia first diagnosed at a later stage. Truncated age-standardardised rates for people aged 15 and over, standardised by the direct method with the 1981 population of Oxfordshire as the standard, are shown in the figure. The total number of people with a first-recorded diagnosis of schizophrenia was much higher than the number diagnosed as schizophrenia at first-ever psychiatric contact alone. In males both these measures, and first-ever inpatient admission rates alone, showed a significant decline in the mid-to-late 1970s similar to that reported by Der et al. The pattern for females was much less clear-cut. Further explanations for the decline in first-contact rates for males include changes in diagnostic fashion, any increased tendency for general practitioners to treat schizophrenics without referral to a specialist, a decrease in severity, and (in local studies) any tendency for people predisposed to schizophrenia to migrate out of the population studied or any increase in inward migration by people not predisposed. Our data add some support to the suggestion that schizophrenia has declined in males, though not in females, in the period 1975-86. Oxford Record
In our view, however, the evidence is not conclusive. We do have comparable data for years before 1975. Unit of Clinical Epidemiology, Oxford University, Oxford Regional Health Authority, Oxford OX3 7LF, UK
not
JACQUELINE ALARCON VALERIE SEAGROATT MICHAEL GOLDACRE DE
1. World Health
Organisation The international pilot study of schizophrenia. Geneva. WHO, 1973. 2. Cooper JE, Kendell RE, Garland BJ, Sharpe L, Copeland JRM, Simon R. Psychiatric diagnosis in New York and London: a comprehensive study of mental hospital admissions (Maudsley Monogr no 20). London: Oxford Univesity Press, 1972.
SiR,—Elsewhere I argue that Kraepelin’s and Bleuler’s concepts of dementia praecox and schizophrenia were mistakenly derived from a population suffering mainly from physical disorders, notably encephalitis lethargica and its sequel, post-encephalitic parkinsonism. The years following von Economo’s descriptions of this infection and its consequences (after Kraepelin and Bleuler had completed their major writings) were characterised by diagnostic confusion; clinicians diagnosed schizophrenia but found it difficult to distinguish the condition from post-encephalitic parkinsonism. Efforts at differentiation were marked by contradiction and non-specificity. More important, they were based on the unfounded assumption that Kraepelin and Bleuler had provided evidence in support of their concepts. In other words, the task of differential diagnosis begged the question of the validity of "dementia praecox" and "schizophrenia".2 The suggestion that Kraepelin and Bleuler were dealing, at least in part, with the neurological sequelae of infectious diseases is very different from the hypothesis that schizophrenia is an infectious disease.3,4That claim, too, is based on an unjustified confidence in the validity of Kraepelin’s and Bleuler’s concepts. My suggestion is that the original concepts of dementia praecox and schizophrenia were mistakenly inferred from at least one now well-described infectious disease and its sequelae. It is difficult to know what implications this claim would have for subsequent rates of schizophrenia diagnosis. Reliable figures for the prevalence of the virus assumed to be implicated in encephalitis lethargica are not available, although it is generally believed to have declined sharply after the epidemic of 1916-27.5 von Economo, however, thought that the virus was passed from mother to fetus, and Sacks has cautioned against the naive assumption that such viruses "disappear" after highly visible epidemics.6 Also the concept of schizophrenia has gradually been transformed from the strongly physical and neurological, as well as behavioural, concept of Kraepelin and Bleuler, to the more behavioural/experiential one of DSM-IIIR. It is quite possible that many patients today who are said to have schizophrenia only superficially resemble Kraepelin’s
and Bleuler’s cases. Nonetheless I suspect that the apparent decline in incidence of schizophrenia is not unrelated to the decline in post-encephalitic parkinsonism. Questions such as "Where have all the catatonic (and severely deteriorated) schizophrenics gone?" may have answers that are at least similar to those to the question "Where have all the cases of post-encephalitic parkinsonism gone?" Department of Psychology, Polytechnic of East London,
MARY BOYLE
London E15 4LZ, UK
1. Boyle M. Schizophrenia: a scientific delusion? London: Routledge, 1990. 2. Boyle M. Is schizophrenia what it was? A re-analysis of Kraepelm’s and Bleuler’s populations. J Hist Behav Sci (in press). 3. Crow TJ. A re-evaluation of the viral hypothesis: is psychosis the result of retroviral integration at a site close to the cerebral dominance gene? Br J Psychiatry 1984; 145: 243-53. 4. Hare EH. Schizophrenia as an infectious disease In: Kerr A, Snaith P, eds. Contemporary issues in schizophrenia. London Royal College of Psychiatrists/ Gaskell, 1986. 5. Pallis CA. Parkinsonism: natural history and clinical features Br Med J 1981; iii. 683-90. 6. Sacks O. Parkinsonism: a so-called new disease? Br Med J 1971; iii: 111.
Coexistence of Raynaud’s syndrome and
erythromelalgia SiR,—Erythromelalgia is known as the inverse Raynaud’s syndrome; the former disease can be induced by treatment of the latter. We report a woman with Raynaud’s syndrome and erythromelalgia. A 40-year-old woman had had, for 5 years, episodes of pallor and blue discolouration of the fingers and toes after cooling and emotional exertion. During these 5 years she also had intolerance to warming of the feet, which manifested attacks of pulsating pain and burning sensations. She had about ten cold and five to seven hot induced paroxysms daily. There were no trophic changes or oedema of the fingers and toes. Distal pulsation remained. Adson’s and Allen’s tests were negative. A cold provocation test induced pallor and blue discolouration of the fingers and toes. A hot provocation test induced pulsating pain and red discolouration of the feet. Biochemical investigation revealed accelerated ADPinduced platelet aggregation. The patient was otherwise healthy. She was given nifedipine (40 mg daily), dihydroergocristine (1-5mg daily), aspirin (750 mg daily), and hyperbaric oxygen sessions (1 ’7 atmospheres absolute, 45 min, seven sessions). After 3 weeks she was discharged and no longer had either hot or cold induced paroxysms.
I cannot explain with certainty how or why the two conditions started together. Their coexistence may have resulted from the dysfunction of arteriovenous anastomoses which are implicated in both diseases. Pathological expansion of the anastomoses while small arteries and arterioles are in cold spasm leads to an attack in Raynaud’s disease, whereas in a warm environment such expansion
might cause erythromelalgia. Kuibyshev Medical Institute, Post Box 703, Post Office 100, Kuibyshev 443100, USSR
G. E. SLUTSKER
Short-term continuous infusion of mitozantrone for advanced breast cancer SIR,-Professor Harris and colleagues (Jan 27, p 186) compare four cycles of mitozantrone with continued treatment in responding breast cancer patients, and record similar disease-free survival in the two groups. Their findings are corroborated by our results in patients with non-pretreated breast cancer, who received continuous infusion of mitozantrone (11mg/m2 daily) over 2 weeks, repeated every 4 weeks for four cycles-a schedule chosen, on the basis of an earlier phase-I study,’ to keep side-effects to a minimum while maintaining optimum efficacy. Ten patients were investigated (mean age 53 years, range 36-68), and two had received cyclophosphamide/methotrexate/5-fluorouracil adjuvant therapy. Responses were complete in two (9 and over 19 months), partial in