A.M. Alhamad,1 M.H. Al-Sawaf,2 A.A. Osman3 and I.S. Ibrahim4
ABSTRACT Consultation–liaison psychiatry has emerged as an important sub-specialty in the general hospital setting during recent years as a result of psychiatric acute wards moving into these hospitals. This has inspired the need for better structured research to establish its relevance and effectiveness. We, therefore, carried out a prospective cohort study at King Fahad General Hospital. We report the interaction of sociodemographic, clinical and diagnostic factors, time lag of referral and diagnostic ability of referring physicians. A total of 206 patients were referred over a period of 6 months. Sensitivity and specificity of the diagnostic skills of the referring doctors were found to be generally poor, particularly for anxiety.
Aspects différentiels de la psychiatrie de liaison dans un hôpital saoudien : I. Orientation et indices cliniques
RÉSUMÉ La psychiatrie de liaison est devenue une importante sous-spécialité dans les hôpitaux généraux ces dernières années du fait que ces hôpitaux se sont dotés de moyens d’accueil et de prise en charge psychiatrique en urgence. Ceci a soulevé la nécessité de mieux structurer la recherche afin d’établir sa pertinence et son efficacité. Nous avons donc réalisé une étude de cohorte prospective à l’Hôpital général Roi Fahd. Nous rapportons l’interaction de facteurs sociodémographiques, cliniques, diagnostiques, ainsi que du délai de transfert du patient et de la capacité diagnostique des médecins référents. Au total, 206 patients ont été adressés à une structure spécialisée sur une période de six mois. On a trouvé que la sensibilité et la spécificité des compétences des médecins référents en matière de diagnostic étaient généralement faibles, notamment pour l’anxiété.
1Department of Psychiatry, College of Medicine, King Saud University, Riyadh, Saudi Arabia (Correspondence to A.M. Alhamad:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
).
2Department of Psychiatry; 4 Department of Biostatistics, King Fahad Hospital, Jeddah, Saudi Arabia.
3Division of Psychiatry, King Faisal Specialist Hospital and Research Centre, Jeddah, Saudi Arabia.
Received 29/07/03; accepted: 03/03/04
EMHJ, 2006, 12(3-4): 316-323
Introduction
A fundamental aim of consultation–liaison psychiatry is the integration of psychiatry into the medical mainstream [1,2]: in essence, to revive the orthodox, holistic medical approach that has prevailed since the time of Hippocrates, who asserted that “in order to cure the human body, it is necessary to have knowledge of the whole” [3]. To effectively achieve this, consultation–liaison psychiatry must first establish its pertinence to other branches of medicine in the general hospital setting, and second, display adequately and reliably the magnitude and the impact of psychiatric morbidity and mortality in physically ill inpatients [1,3]. Both objectives are only achievable through reliable and adequate research in the field, which should not only investigate the pattern of referral and the compliance with psychiatrists’ recommendations, but also assess the knowledge and attitude of both the hospital doctors and the patients, which are crucial determinants in referrals for psychiatric consultation [4]. To the best of our knowledge, however, none of the literature has addressed these issues simultaneously in the same setting. Moreover, the many difficulties inherent in research into consultation–liaison psychiatry [5], combined with a lack of standard research measures [6,7], has rendered results from different studies widely variable and inconsistent.
Three closely related but distinct reasons might be responsible for this wide variability in results, the nature of the patient population under investigation, the knowledge of hospital doctors and their attitude towards psychiatry, and the attitude of the patients themselves towards psychiatric consultation in the context of their current physical illness [3,4,8,9].
We carried out this study to demonstrate the corresponding influence of each of these 3 dimensions on the pattern of referral for psychiatric consultation in a general hospital. In this paper we report the sociodemographic data, the referral indices, the clinical characteristics, the diagnostic categories and concordance, and the consent of patients. Knowledge and attitude of referring doctors and patients are reported in part II of the study.
Methods
We carried out a prospective cohort study of all patients referred consecutively for psychiatric consultation at King Fahad Hospital in Jeddah, Saudi Arabia, over a period of 6 months between July and December 1994. These patients were referred either from inpatient departments or the extended emergency room, where patients stay up to 48 hours only.
King Fahad Hospital in Jeddah is a large, 930-bed general hospital which serves mainly as a secondary care centre, but having 3 units which function basically as tertiary care units, the open-heart surgery unit, the organ transplant unit and the haemodialysis unit. The psychiatric consultation service is provided by an outpatient unit with 2 full-time consultants, 2 residents, 2 psychologists, 2 social workers and 1 nurse.
We designed 2 forms to collect the data needed for our study. Form A was to be filled in personally by the referring consultant or specialist (consultee). The questions covered basic demographic data, the reason for referral, the degree of urgency, the medical diagnosis, active psychiatric symptoms he was able to elicit, and provisional psychiatric diagnosis.
Form B was to be filled in personally by the responding psychiatrist (consultant). This covered clinical assessment, degree of urgency and final psychiatric diagnosis according to the World Health Organization ICD-10 classification system [10]. The degree of urgency of the referral was divided in both forms into 4 categories to determine referral time lag. These were immediate (within 48 hours of admission), early (3–7 days after admission), delayed (8–15 days after admission) and late referral (> 15 days after admission).
Results
A total of 206 patients were referred over the period of the study, of which 188 came from inpatient departments and 18 from the extended emergency room. This represented an overall referral rate of 3.9% of the total admissions to the hospital over the study period (5263 patients), 3.8% for inpatient departments and 6.3% for the extended emergency room.
There were 120 referrals of male patients (58.5%), giving a male to female ratio of 1.4:1; this compares to a ratio of 1.8:1 in the general hospital population. Mean age was 37.14 years (standard deviation 15.2), range 11–80 years. There were only 10 patients (4.9%) older than 60 years. The highest rate of referral was from the nephrology and haemodialysis unit, 8.1%; the others are summarized in Table 1.
The medical department tended to refer patients earlier than other departments (Table 2). Analysis of the time lag in relation to diagnosis, reason for referral, age and consent of the patient is shown in Table 3. There was a statistically significant difference between time lag for patients under 40 years of age (shorter time lag) and those over 40 years of age (χ2 = 5.88; P < 0.05). Substance abuse patients diagnosed by the consultee were referred earlier (χ2 = 7.52; P < 0.01), while organic brain syndrome patients diagnosed by the consultant were significantly delayed in referral (χ2 = 7.52; P < 0.01). An interesting finding was that the earlier the patient was referred, the greater the likelihood he had consented to referral and this was statistically significant also (χ2 = 9.92; P < 0.02).
Only 7 patients (3.7%) were referred for evaluation of competence for surgical treatment and the diagnoses were 3 with organic brain syndrome, 2 phobic states of surgery and 2 personality disorders.
Table 4 shows the concordance of diagnoses made by the consultees and the consultants.
Table 5 shows sensitivity and specificity for each diagnosis. Only 190 patients out of the total 206 referred were given diagnoses by the consultees, hence the difference between total numbers of patients for consultees and consultants. Depression and drug abuse were rated relatively better in both specificity and sensitivity, while anxiety disorders showed the lowest sensitivity and specificity rates. Specificity for personality disorder was zero.
Discussion
In view of the large array of variables which may prove difficult to control, and the limitations in generalizability of the results obtained from any one study, besides the small size and the short duration of the study, we must initially acknowledge the limitations of our findings. However, we believe that our findings may highlight a few, but nevertheless significant, issues in this field, being probably the first study of its kind.
Demographic characteristics
In contrast to the majority of previous reports [4,8–13], in this study males exceeded females in both the hospital population and referral rates. This reversed sex ratio is in accord with the general trend in the Saudi hospital population [14]. Some researchers consider this reversal of ratio may be because females tend to seek help from traditional local support systems for longer periods before they seek help from medical centres or because females have higher acceptability and tolerance of psychiatric symptoms and the tendency to conceal them in view of their local, more stigmatizing, effects on females, which may be true even among hospital populations [8,14].
The age range and mean age did not differ significantly from other reports in the literature, though patients older than 60 years were relatively under-represented, a finding which invites some investigation as psychiatric disorders tend to be more common among this group. The rate, 4.9%, was significantly lower than that reported among inpatients (13.3%) and also much lower than the usually reported ranges of 16%–23% in the literature [3,4,8].
Referral indices
The mean referral rate of 3.8% for the inpatients is consistent with the usually reported rates of 3%–4% [3,4,8]. An important, though not surprising, finding was the high referral rate (8.1%) from the nephrology and haemodialysis unit. There is an acknowledged high psychiatric morbidity among patients on regular haemodialysis [15,16].
The finding that a good number of patients (18.4%) were referred for the mere presence of past history of psychiatric disorder or for taking some kind of a psychotropic drug, even though they were free of symptoms at the time, may reflect the poverty of psychiatric knowledge among hospital doctors. Medico-legal factors can also influence psychiatric referral as it has been shown that 25% of referrals are for evaluation of competence in the absence of gross evidence of mental illness and 75% of such referrals were found to be mentally competent. This was explained by the medico-legal consequences, physicians’ anxiety and lack of appropriate understanding of patients’ rights during treatment [9].
The timing of consultation is influenced by many factors [17,18], and younger patients were more likely to be referred early (3–7 days), which is closely related to the previously mentioned finding of under-representation of the elderly, and could be attributed to similar reasons.
Intrapersonal factors, i.e. the individual characteristics of the patient and his/her response to the referral, and the stigmatizing effect of the consultation in the young patients compared to the elderly may explain the time-lag difference. It is also possible that the consultees were less concerned about the psychiatric symptoms in older patients, perhaps owing to their wrong assumptions that such symptoms are normally expected in this age group, and do not warrant psychiatric consultation, or perhaps they found it more difficult to identify such symptoms in the elderly [19].
The shorter time lag in referrals from the medical units in comparison to the surgical ones may reflect a better knowledge of psychiatry in the internists, or that symptoms appear later in surgical patients, possibly related to the stress of surgery.
Substance abuse was the only diagnosis which precipitated immediate referral by many of the consultees, most likely in anticipation of difficulties once the withdrawal symptoms appeared. However, organic brain syndrome, which was expected to be referred early, showed a statistically significant delay [20]. This could be a reflection of the hospital doctors’ reluctance to refer such cases because of their wrong assumption that organic brain syndrome belongs solely to their own domain, or because of their negative expectation of the help a psychiatrist might contribute in the management of such cases [20].
Consent of patients
There are 3 possible explanations for the interesting finding of the strong association between consent of the patient and immediate referral. First, it might reflect the severity of symptoms, which may dissipate the ambivalence and fear of the consultee of informing the patient about the possibility of his symptoms being psychological. Second, earlier referral might reflect the better knowledge, and hence more positive attitude towards psychiatry, on the part of the consultee, who therefore feels comfortable in informing his patient. Third, the longer the patient stays in the hospital under the care of the same doctor, the stronger the dominance of the doctor over the patient, and the more likely he or she will be to take decisions on the patient’s behalf without consent [21].
Diagnostic ability of consultees
Deficiencies in the diagnostic abilities of hospital doctors may affect their ability to correctly recognize psychiatric symptoms. This could seriously distort the pattern of referrals and obscure the prevalence rates of these symptoms among their patients [22]. Our findings generally indicated that such abilities among our hospital doctors were disappointingly poor. Even in a disorder such as organic brain syndrome, which is strongly perceived by the majority of doctors to be solely within their own domain, the sensitivity of diagnosis was second poorest. The difficulty in correctly diagnosing these cases was also manifested in the unduly prolonged time lag in referral. Diagnosis of depression and anxiety disorders, which are by far the commonest disorders among hospital patients [2,3,8], also showed inadequate sensitivity and specificity. The poor specificity in diagnosing psychosis may be due to the consultees’ difficulty in differentiating it from other conditions [22]. Sixteen out of the 26 patients who were diagnosed as having psychosis by the consultees were found to be suffering from either personality disorder, severe depression, or organic brain syndrome (5, 6, and 5 patients respectively).
Conclusion
Our findings, for referral rates, diagnostic abilities of consultees or the relationship of time lag to other significant variables, reflect results which are generally comparable to reports in the literature. The poor knowledge of psychiatry of our hospital doctors necessitates the simultaneous investigation of the correlation of these factors to the knowledge and attitude of the consultees and the referred patients. This will be reported in the second part of our study. This supports the liaison model [11,13] which gives priority to the teaching and education of hospital staff rather than to the provision of clinical services.
References
- Lipowski ZJ. Introduction to consultation–liaison psychiatry. Psychiatry update, 1983, 3(3):177–87.
- Gomez J. Liaison psychiatry–mental health problems in the general hospital. Beckenham, Kent, Croom-Helm Ltd, 1987:103–28.
- Lipowski ZJ. Consultation–liaison psychiatry: an overview. American journal of psychiatry, 1974, 131(6):623–30.
- Lipowski ZJ, Wolston EJ. Liaison psychiatry-referral patterns and their stability over time. American journal of psychiatry, 1981, 138(12):1608–11.
- Lipowski ZJ. Physical illness and psychiatric disorder—a neglected relationship. Psychiatria fennica, 1979, 10:32–57.
- Goldberg D, Huxley P. Mental illness in the community: the pathway to psychiatric care. London, Tavistock Publications, 1980:37–9.
- Williams P, Tarnopolsky A, Hand D. Case definition and case identification in psychiatric epidemiology: review and assessment. Psychological medicine, 1980, 10(1):101–14.
- Abu-Hijleh NS. Psychiatric consultation in Jordan University Hospital. Jordan medical journal, 1987, 21(2):149–57.
- Craig TJ. An epidemiologic study of a psychiatric liaison service. General hospital psychiatry, 1992, 4:131–7.
- ICD-10. International classification of diseases: classification of mental and behavioural disorders. Geneva, World Health Organization, 1993.
- Huyse FJ et al. European consultation–liaison psychiatric services: the ECLW collaborative study. Acta psychiatrica scandinavica, 2000, 101(5):360–6.
- Mayou R, Hawton K. Psychiatric disorder in the general hospital. British journal of psychiatry, 1986, 149:172–90.
- Smith GC, Clarke DM, Herrman HE. Consultation liaison psychiatry in Australia. General hospital psychiatry, 1993, 15(2):121–4.
- Osman AA, Al-Khateeb SO, Ali AS. The pattern of admission to Jeddah psychiatric hospital. Saudi medical journal, 1993, 14(4):334–9.
- Sensky T. Psychiatric morbidity in renal transplantation. Psychotherapy and psychosomatics, 1989, 52(1–3):41–6.
- Barrett BJ et al. Clinical and psychological correlates of somatic symptoms in patients on dialysis. Nephron, 1990, 55(1):10–5.
- De-Jonge P et al. Timing of psychiatric consultations: the impact of social vulnerability and level of psychiatric dysfunction. Psychosomatics, 2000, 41(6):505–11.
- Handrinos D, McKenzie D, Smith GC. Timing of referral to consultation–liaison psychiatry unit. Psychosomatics, 1998, 39(4):311–7.
- Wilkinson P, Bolton J, Bass C. Older patients referred to a consultation–liaison psychiatry clinic. International journal of geriatric psychiatry, 2001, 16(1):100–5.
- Stevens LE, de-Moore GM, Simpson JM. Delirium in hospital: does it increase length of stay? Australian and New Zealand journal of psychiatry, 1998, 32(6):805–8.
- Schnyder U et al. Reference to psychiatric consultation in the discharge letter of general hospital inpatients. International journal of psychiatry in medicine, 1997, 27(4):391–402.
- McGorry P, Copolov DL, Singh BS. The validity of the assessment of psychopathology in psychosis. Australian and New Zealand journal of psychiatry, 1989, 23:469–82.