eMJA     The Medical Journal of Australia

Home | Issues | eMJA shop | My account | Classifieds | Contact | More... | Topics | Search   

Abstract

Introduction

With a few exceptions, the management of DSP in Australian hospitals appears to be on an ad hoc basis. There are several relatively unstructured models of management for presentations of DSP to hospital. Commonly the patient is managed in the Emergency Department and, if required, admission occurs under the general physician of the day or under the specialty which best corresponds to the patient's toxicological problem. Other models include medical management for most patients occurring entirely within the Emergency Department/Intensive Care axis with or without formal admission. In at least one centre, 30% of DSP presentations were not formally admitted with all assessment and management occurring in the Emergency Department.⁵ In many centres, expertise in toxicology comes from outside the service providing the patient care (for example from Poisons Information Centres) rather than expertise being developed within the unit.

In these models of management, not every DSP patient receives formal psychiatric assessment. In spite of evidence that psychiatric intervention after parasuicide is worthwhile,⁶ concentrating psychiatric resources on only those patients who require admission and those who have the most medically severe poisoning is common.⁷ However, significant suicidal risk is present for many patients with toxicologically "trivial" poisonings.  

Service structure

All DSP patients are formally admitted under the HATS clinical toxicologist. That clinician retains primary responsibility for care during the whole admission. A preformatted toxicology data collection form was developed which is also the formal admission record. The psychiatry team assess all DSPs and, on request, other poisonings. Referral to the Drug and Alcohol service occurs as necessary. The patient is determined as medically fit for discharge by the toxicologist and the decision on appropriate destination is made by the psychiatrist. Medical follow up, when required, is the responsibility of the toxicologist.

A relational database for collecting data on poisoned patients⁴ was written (by IMW) in late 1986. The complexity of this database has progressively increased and an extensive psychiatric component (written by GLC) became operational in January 1996.

HATS was transferred to the Newcastle Mater Misericordiae Hospital (NMMH) in 1991 and took on an Area role (population 385,000). All poisoning cases are transferred, either directly or after assessment at a closer hospital, to the NMMH unless they are too ill. If they are admitted to another hospital, they are admitted under the care of HATS.

HATS provides a 24 hour a day telephone consulting service for the Upper and Lower Hunter Areas (population 100,000) and a tertiary referral service when required. The clinical pathway for this managed care is shown in Figure 1.

HATS medical management is provided by a full time equivalent clinical toxicologist (currently two clinical pharmacologists) and a registrar in clinical toxicology. The after hours service is provided by the two clinical pharmacologists with the addition of another clinical pharmacologist and a drug and alcohol specialist with an interest in toxicology. Every toxicology admission is seen by the medical team (at least once a day), seven days a week.

HATS psychiatric care is provided by a psychiatrist, a psychiatry registrar and a clinical nurse consultant in psychiatry. An after hours service is provided by part of an area roster of psychiatry registrars and four to six "second on call" psychiatrists. Every deliberate self-poisoning admission is seen by the psychiatry team (at least once a day), seven days a week. Even when the patient is consciously or cognitively impaired, our practice is to begin the psychiatric assessment by obtaining a collateral history from family/friends and health care workers before proceeding with individual psychiatric assessment. In the event the patient presents with a toxicologically trivial poisoning after 5 pm or on weekends (which would allow discharge before the routine morning psychiatric review) then the psychiatric registrar on call will come in to do the assessment and decide on appropriate discharge destination.  

Philosophy and strategies

Nursing perspective

Psychiatric perspective

Medical perspective

Outcomes and resource use

In 1995 there were 736 admissions (see Table 1). The median (range) hospital stay is calculated because the data are not normally distributed. The lower quartile is at 10.5 hours and the upper quartile at 27.5 hours. The distribution of hospital stay for DSP patients is shown in Figure 2. In 1994/95, for all hospitals with an Emergency Department in the Greater Newcastle area, DSP comprised 1.2% of medical admissions the majority of which presented to NMMH where they comprised 7.3% of medical admissions. Of the 520 DSP admissions who received formal psychiatric assessment 492 (94.6%) had one or more formal diagnoses¹¹ made of psychiatric disorder, personality disorder or other condition (V codes).

ALOS data for HATS compared with national data and all NSW public hospitals are shown in Tables 2 and 3 respectively. Table 2 compares data derived from the HATS database for 1991-1994 with national data for 1992. Table 3 compares data from the NSW Health Department ISC (1994/95) with the mean for all public hospitals in NSW. Both comparisons show a substantial reduction in bed stay for HATS. To reduce the effect of presentations that may not be admitted at other hospitals we further analysed HATS data for 1993-1995. For all admissions ALOS was 1.50 days; for all admissions with a hospital stay > > 12 hours (72.6% of admissions) ALOS was 1.69 days; for all admissions that required intensive care admission (16.8%), ALOS was 2.59 days.

There has been no evidence of compromised patient care with the reduction in bed stay as mortality from DSP during this period (1987-1995) has been 0.6% (24 deaths in 3856 DSP admissions; 95%CI 0.4 - 0.9). Most of these patients had an out of hospital arrest and death was inevitable on presentation.⁴ NSW Health Department data for 1992 (the most recent year for which death data were available) show 13 inpatient deaths in 2876 admissions in NSW (0.5%; 95%CI 0.2 - 0.8). HATS data for 1992 show one inpatient death in 512 DSP admissions (0.2%; 95%CI 0.0 - 1.1). These proportions are not significantly different (Chi-square = 2.04, P= 0.36). Standardised mortality ratios for suicide in NSW show the Hunter Area has no greater all cause suicide mortality.¹²

HATS's prospective data collection on all presentations in a defined population is a very powerful tool for observational research. Thus, we have been able to identify a number of significant public health issues related to patterns of drug use,^13,14 relative toxicity of drugs within classes^15,16,17 and the impact of safety packaging of medications.¹⁸

The NSW Health Department, as part of its health outcomes strategy, has provided funds to HATS to develop the management model (and the database) so that it can be trialed at other centres in NSW.  

Discussion

It could be argued that shorter length of stay in the Hunter is due to our policy of admitting all patients with deliberate self poisoning regardless of severity. However, the greatest difference in ALOS occurs in those patients with complications or comorbidities (Table 2 & 3) who require admission under any policy. Figure 2 shows that 90% of all our patients stay in hospital for less than 50 hours, which is less than the ALOS for uncomplicated poisoning in NSW public hospitals (Table 2). In addition, HATS bed stay for all admissions (complicated and uncomplicated) after excluding those admitted for less than 12 hours is still shorter than the ALOS for uncomplicated admissions to all NSW hospitals. The ALOS for HATS admissions requiring intensive care is more than 2 days shorter than the ALOS for all complicated admissions to NSW hospitals.

The model of management of self poisoning described in this article is, we believe, unique in Australia. The differences we have identified in this model are:

• all toxicology presentations in one Area Health Service are diverted (by Ambulance services and Emergency Departments) to one hospital

• all DSP presentations are admitted

• all admissions are to one team

• the team is multi-disciplinary with medical, psychiatric, drug and alcohol, and nursing participation

• all DSP admissions receive psychiatric assessment

• a 24-hour service for management and advice is provided

  The critical elements are that the service is centralised to one hospital and within that hospital clinical responsibility is centralised to one clinical unit where all DSP is admitted.

  We argue that all DSP presentations should be admitted for several reasons:

• DSP is a presenting symptom for another problem that requires assessment and intervention

• most DSP admissions (94.6%) have a diagnosable psychiatric disorder, personality disorder or other psychiatric condition

• formal admission facilitates more efficient and effective assessment and management of both medical and psychiatric issues

• as the bulk (>50%) of presentations occur after hours, overnight admission is required to ensure psychiatric assessment

This model has resulted in a substantial and significant reduction in bed stay which provides an increase in available beds for other purposes (opportunity cost) in the Area. There are two ways of calculating monetary cost if funding were on a Diagnosis Related Groups (DRG) basis. The first is to multiply the bed days saved by the DRG cost of a bed day as in Table 2 and Table 3. The second is to assume the saved beds will be occupied by patients attracting further DRG funding. Assigning a monetary value to this is not possible. It is clear, however, that based on DRG funding these saved bed days are worth more to the Area than the cost of running the service.

The reduction in bed stay has not been accompanied by a worsening in outcome as defined by in hospital mortality from DSP or standardised mortality ratios for all cause suicide. While the number of admissions to HATS in 1992 appears disproportionate it is consistent with the proportion of admissions to other major hospitals^1,2 and reflects our policy of admitting all DSP presentations. It appears likely the Department of Health figures for admissions significantly underestimate the number of presentations for DSP to NSW hospitals. If so, while the magnitude of the saving per admission may be uncertain, on a state wide basis the potential savings from implementing our model are even greater.

Without further data, determining the reasons for the shorter ALOS is not possible but anecdotal comparisons with other hospitals suggest the following possibilities:

• centralized, evidenced-based management of specific poisonings resulting in

• earlier recognition of nontoxic or minimally toxic exposure

• more efficient gastrointestinal decontamination

• better management of significant toxic exposure

• more efficient use of psychiatric assessment, aftercare and discharge planning

• increased involvement of nursing staff in a multidisciplinary approach

Our current model of management has evolved utilizing the skills and experience of those interested in poisoning in Newcastle. The only new position created was the registrar position in Clinical Pharmacology. We do not believe, however, that replicating the model or its outcomes is dependent on replicating our subspeciality mix. Nevertheless, the identification of a team to manage poisoning is critical. In many health areas the Emergency physicians may be the logical choice for such a team. This would require an extension of admitting rights into the general hospital or a collaborative venture with an identifiable medical team. A specific group of psychiatrists is also critical.

We believe that Area Health Services should consolidate acute toxicology services on one site. A potential disadvantage of consolidation is loss of skills in the management of toxicological problems in other hospitals in the Area. This could be offset by making the service part of registrar and nursing training rotations. The advantages of consolidation include:

• individualising patient care

• continuity of care

• a better learning curve via greater experience

• training, education and research

Overall the efficiencies evident in this model are a product of the reorganisation of largely existing resources to provide a multidisciplinary team approach to the management of poisoned patients. The provision of care is based on a philosophy that these patients are entitled to a legitimate sick role. The major stumbling block to establishing a similar dedicated service is in making the decision to reorganise existing services. As House et al state, after reviewing services in the UK, "there is much to recommend in clinical diversity, but nothing to recommend [in] unplanned and incoordinated service provision".⁶  

Acknowledgements

References

1. Pond SM. Prescription for poisoning. Med J Aust 1995;162:174-5
2. McGrath J. A survey of deliberate self-poisoning. Med J Aust 1989;150:317-22.
3. Buckley NA, Whyte IM, Dawson AH. There are days . . . and moons. Self-poisoning is not lunacy. Med J Aust 1993;159:786-9.
4. Buckley NA, Whyte IM, Dawson AH, McManus PR, Ferguson NW. Self-poisoning in Newcastle, 1987-1992. Med J Aust 1995;162:190-3.
5. Davis AT, Kosky RJ. Attempted suicide in Adelaide and Perth: changing rates for males and females, 1971-1987. Med J Aust 1991;154:666-85.
6. House A, Owens D, Storer D. Psycho-social intervention following attempted suicide: is there a case for better services? Int Rev Psychiat 1992;4:15-22.
7. Tengel E, Cook NG, Kreeger IS. Attempted suicide. Chapman & Hall, London, 1958.
8. Parsons T. The social system. London, Routledge and Kegan Paul, 1951.
9. Buckley NA, Dawson AH, Whyte IM. HyperTox - a hypertext teaching program in toxicology. http://www.newcastle.edu.au/department/md/htas/toxi0002.htm
10. World Health Organisation: International Classification of Disease - Clinical Modification, 1978. WHO, 1992.
11. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. Washington, DC, American Psychiatric Association, 1994.
12. Stewart G, Chipps JA, Sayer G. Suicide mortality in NSW local government areas. NSW Public Health Bulletin 1995;7:1-10.
13. Smith AJ, Whyte IM. New drugs for old: an issue for debate? Med J Aust 1988;149:581-2.
14. Dawson AH, Whyte IM. Compound analgesics. Med J Aust 1990;152:334.
15. Buckley NA, Dawson AH, Whyte IM, Henry DA. Greater toxicity in overdose of dothiepin than of other tricyclic antidepressants. Lancet 1994;343:159-162.
16. Buckley NA, Dawson AH, Whyte IM, O'Connell DL. Relative toxicity of benzodiazepines in overdose. BMJ 1995;310:219-21.
17. Buckley NA, Whyte IM, Dawson AH. Cardiotoxicity more common in thioridazine overdose than with other neuroleptics. J Toxicol-Clin Toxicol 1995;33:199-204.
18. Buckley NA, Newby DA, Dawson AH, Whyte IM. The effect of the introduction of safety packaging for carbamazepine on toxicity in overdose in adults. Pharmacoepid Drug Safety 1995;4:351-4.