1990 - Expressed Emotion And Lithium Prophylaxis

S.Priebe C.Wildgrube B.MullerOerlinghausen

... 'I

Expressed emotion and hospital admission in lithium prophylaxis Department FRG

of Psychiatry,

Free University of Berlin,

INTRODUCTION: The key-relatives' expressed emotion (EE) as assessed by the CamberwellFamily Interview (CFI) have repeatedly been shown to predict the relapse rate for schizophrenic and depressive patients. Only the cut-off point for high-EE in critical remarks is different for the two diagnostic groups. In schizophrenic patients, this higher relapse rate also depends on the amountof the face-to-face contact with the high-EE relative and can be lowered by a maintenance neuroleptic treatment, so that patients living with high-EE relatives would particularly benefit from such a prophylactic long-term medication. In some studies, high-EE relatives changed to low-EEin line with a positive clinical course of the patient (2). This study was to investigate in which way the key-relative's EE is related to hospital admissions within the last 3 years in a sample of patients being on long-term prophylactic lithium treatment. METHOD AND SAMPLE: We selected

all 28 patients

(21 female,

7 male; mean

age = 49 years) being treated in our outpatient clinic for lithium prophylaxis for at least 3 years (x=10ys) and living closely with a key-relative. 24 patients were diagnosed as bipolar affective, and 4 patients as schizoaffective disorders according to ICD-9 classification. The serum levels throughout the last 3 years were regarded as appropriate in all patients (O.5-1.1mmol/I). In 7 cases, the key-relatives were unavailable or reluctant to take part in the study. So, 21 relatives were interviewed. 16 "relatives" were spouses, 2 were parents, 2 were children, and 1 was a fellow-nun of the patient. All interviews were done in our department and by the same interviewer. The rater was blind to the clinical features of the patients at the time of the rating. RESULTS: Whenthe cut-off point for high-EEis 5 or 6 critical remarks, there were only 5 high-EE relatives. When the threshold for defining high-EE i~ 3 critical remarks, then 7 relatives must be considered as being high-H. In both cases, there is no clear difference between patients living with high- or low-EErelatives as far as the the hospital admissions during the last 3 years are concerned. If all 11 relatives with 2 or more critical commentsare allocated to high-EE status, hospitalized recurrencies occur more often in patients living with those relatives. Yet, this tendency fails to reach statistical significance. Tab.1 shows the distribution of patients' hospital admissions and their relatives' critical remarks, also indicating that the amountof the direct face-to-face contact per week of patients and potentially high-EE relatives is not clearly linked with relapses either. In spite of this general result, there is an association between hospital admissions and one CFl-scale. Table 2 showsthat a high score in emotional overinvolvement (EOI) is more often associated with a hospital admission of the patient.

Lithium: Inorganic Pharmacology and Psychiatric Use Edited by Nicholas J.Birch @ IR L Press Limited,

Oxford,

England.

29

S.Priebe et al. Table 1: Number of clinical remarks made by the relatives and hospital admission of the patients within the last 3 years; in parenthesis the approximate face-to-face contact per week between patients and high-EE relatives. NO of critical

remarks

o 1 2 4 7 9 11 12

Table 2: EOI-score

o

admission

(20h, 6Oh) (40h, 50h) ( 35h) (42h) (35h )

One hospital (N=6) "2

admission

o 2 (30h, 44h)

o

1 (35h)

o o

1 (50h)

Emotional overinvolvement-score of the relatives and hospitalized recurrencies of the patients within the last 3 years. No hospital (N=15)

0 1 2 3 4

Table 3:

No hospital (N=15) 7 1 2 2 1 1 1

admission 7 7 1 0 0

One hospital (N=6) 3 0 1 1 1

admission

Emotional overinvolvement-score of the relatives and the patients' Morbidity Indices over the whole period of lithium treatment and over the last 3 years. EO!-score

Morbidity Index over the whole period of lithium treatment over the last 3 years

o or 1 (N=17)

EOI-score 2 or 3 or 4(N=4)

.087 (SD=.06)

.108 (SD=.04)

.028 (SD=.06)

.160 (SD=.10)

Whenrelatives are subdivided into those with a score of 0 and 1 in the EOI-scale on the one hand, and those with a score of 2 and more on the other (1 point lower than usual), we are left with 4 patients living with the second group of particularly overinvolved relatives. Tab.3 shows the Morbidity Indices (based upon hospitalized recurrencies only) over the whole period of lithium treatment being not muchhigher in that small group comparedwith the remaining 17 patients (1). This indicates that those 4 patients have not had more or longer inpatient treatments during the whole lithium prophylaxis. But the Morbidity Index concerning only the last 3 years is 5 times as high as in the patients not living with an in that way overinvolved relative. 30

Lithium prophylaxis

DISCUSSION: Frequencies of critical remarks are comparatively low in this study. This may be due to special conditions in affective psychosis to a selection process within an outpatient clinic for very long-term treatment or to changes during a mostly successful treatment (3). In case a predictive validity of EE is assumed for these disorders as well, there are at least 2 possible explanations for the lack of a significant correlation between the relatives' EE and the patients' hospital admissions within the last 3 years: Firstly, lithium is particularly protective for patients living with high-EE relatives. Secondly, the treatment was associated with a change of the relatives' EE status. Of course, only hospital admissions are considered as outcome criterion in this study; the analysis of recurrencies not leading to a hospita I admission gives a different picture indicating a stronger association between EE and morbidity index (unpublished data). Finally, patients living with emotionally highly involved relatives seem to have a higher risk for hospital ized recurrencies, even after several years of prophylactic lithium treatment. Yet it has to be taken into account that this result is based upon very few patients. REFERENCES:

1.

2. 3.

Coppen, A., Peet, M., Bailey, J., Noguera, B., Burns, B.H., Swani, M.S., Maggs, R., Gardner, R. (1973) Double-blind and open prospective studies of lithium prophylaxis in affective disorders. Psych. Neurol.Neurochir.76, 501-510. Konigsberg, H.W.&Handley, R. (1986) Expressed emotion: From predictive index to clinical construct. Amer.J.Psychiat.143, 13611373. Yolk, J. & Muller-Oerlinghausen, B. (1987) Quality of interepisodic periods in manic-depressive patients under lithium long-term treatment. This volume.

31