An axis for risk management in classificatory sys-tems as a contribution to efficient clinical practice

Abstract

Comprehensive clinical assessment and patient management plans have been enhanced by the development of multiaxial classificatory systems. Assessment of risk is an essential clinical task for which the conclusions are not currently reflected in the multiaxial diagnostic schemata. Developments in the understanding of risk and its management make possible consideration of its place in multiaxial systems. The structure and principles of a potentially workable axis, summarizing current knowledge of risk in the domains of suicide, self-neglect and violence to others, are described. Clinicians are more likely to use this axis than the multiple, emerging, risk assessment guidelines. Incorporating risk management would be a practical addition to presently available axes and be very widely clinically applicable.

Consideration has often been given to placing greater emphasis on the utility of classificatory systems, particularly because of the lack of progress in developing an etiologically based system and the recognition that a “naturalistic” approach to classification might be unrealistic 1-3.

The classificatory system and clinical formulations are central to the clinical logic of connecting the assessment information to the patient recovery plan 4. Multiaxial classificatory systems can be thought of as attempts to standardize regularly informative components of formulation into a classificatory framework.

In recent years there has been increasing emphasis on the concept of risk assessment. For example, there have been several publications in the area of risk of suicide (5,6), risk of harm/violence to others7 and more extended risks to the patient themselves, such as self-neglect 8. The understanding of these risk factors has been gradually increased by more precise epidemiologically guided research. Public and health service concerns about the consequences of inadequate risk management have led to the gradual emergence of a number of guidelines 9-11. Almost inevitably these guidelines, which connect risk assessment and risk management, concentrate on only one of the three major risk areas referred to above, despite the recognition that a single, comprehensive clinical management or recovery plan best serves patient/consumer needs.

We would argue that incorporating the clinical management consequences of risk assessment as one dimension of a multiaxial classificatory system would increase both clinical effectiveness and efficiency. This paper sets out a possible structure for such an axis, with its rationale.

RISK ASSESSMENT VS. RISK MANAGEMENT

It has been noted that, when predicting risk of violence, psychiatrists are likely to be very often wrong 12-15. We also know that by developing the skills of risk formulation 12and risk management 16they are likely to achieve better results. The distinction between the tasks of risk assessment for clinical management and event prediction is subtle but significant. A classic study in this regard was conducted by Lidz et al 17, who reported that clinicians were reasonably accurate in assessing dangerousness, since the patients who did prove to be violent on follow-up over six months were detected with reasonable sensitivity. On the other hand, many patients who were rated as dangerous by clinicians did not prove to be more violent than the other patients (low specificity).

A clinical determination that a patient presents sufficient risk to justify intervention is one goal of assessment of risk. Risk assessment must identify clinical or situational factors which can be modified to reduce risk. It is noteworthy that inquiries into homi-cides by persons with mental illness have consistently found that only a minority of incidents are predictable, whilst the majority are preventable with good quality clinical assessment, communication and intervention 18,19. We can use our psychiatric training to introduce interventions according to the needs of an individual and master the art of risk management by constantly considering the dynamic nature of risk and paying attention to the needs and deficits of an individual.

The issue of shifting focus from risk prediction to risk management becomes more relevant when one considers the ethical implications of the two 14. Often the outcome of risk assessment is that a patient with a history of violence is identified as “potentially violent”, which easily gets distorted as “violent”. These adjectives accumulate in the file and are of little utility unless ways are identified to manage risk. Our responsibility as psychiatrists does not end with stating that a given patient is potentially dangerous. The ethical justification for risk assessment by a treating psychiatrist is risk reduction through risk management. Risk changes with time and circumstance and therefore the risk of violence needs to be assessed and reviewed regularly. While these factors are described in the context of assessment of risk of violence to others, the same principles apply to the other two main types of risk that clinicians routinely assess in general adult psychiatric settings.

AXIS DESIGN ISSUES

The major organizing principle for our proposed axis is that it should inform and assist the development of patient recovery plans. It will do that best by incorporating both positive and negative risk factors which need to be addressed or harnessed to facilitate patient recovery.

Clinicians most commonly undertake three types of risk assessment – violence, suicide and self-neglect – which are embedded in the legislations on compulsory treatment in many places 14,20. In order to be accepted and widely used, a risk axis will need to be simple yet comprehensive. It should be sufficiently comprehensive not only to capture all the types of risk assessed, but also to be able to address the unique aspects of each risk. It needs to be able to capture all three types of risk in one format, rather than the tri-partite guidelines which are beginning to appear in a number of nations – for example, in the UK 9and in New Zealand 10. Hav-ing a separate system for each type of risk is confusing and burdensome for clinicians, and therefore more likely to be observed in the breach than in the action. It also means there are often several different management plans in different parts of the clinical file.

A history of violence is known to evoke strong emotions and aversion in the people conducting such risk assessment 14. It is likely that in patients who have committed previous violent acts, clinicians may either miss or underestimate other types of risks such as of suicide or self-neglect. Incorporating the three types of risk in one axis will encourage their assessment in a manner similar to how detection of personality disorder and physical illnesses have improved with the introduction of multiaxial diagnostic systems 21-23.

A retrospective study 24, based on a case note review that looked at the practicality of extracting risk-related information, found that on average it took 5 hours to conduct a thorough review, rendering retrospective case note reviews an impractical, incom-plete and misleading way of conducting the three types of risk assessment. The authors recommended prospective recording as a more practical method if used selectively, but cautioned that it required a standardized approach to clinical recording and case note mainte-nance. It may be worth noting that taking a (multidisciplinary) team approach to risk assessment may not only reduce biases in clinical decision making 25, but also speed the process due to cumulative knowledge about the risk issues.

We note that each type of risk has both dynamic or clinical factors and static or historical factors, which are assessed by clini-cal or actuarial methods respectively. It has been argued that for better outcomes the two methods should be combined 7,26. A risk axis could enable clinicians to attend to both tasks and serve as an “aide memoire”, yet have sufficient in-built flexibility to allow in-dividual or unique aspects of the patient's presentation to be taken into account in the clinical recovery plan.

We believe, as stated above, that risk assessment should be carried out primarily with a view to managing the risk, otherwise the task becomes unethical and disadvantageous to the patient. Therefore the risk axis should be able to inform the development of the indi-vidual care plan. For each of the three types of risk (self-neglect, suicide and violence to others), static, dynamic and management factors (targeting on the latter may well reduce the risk) will need to be described in a manner that informs the patient recovery plan. Some risk factors and their managements are common to all three.

Static factors for risk of self-neglect include male gender, older age, poverty, living alone and physical problems (e.g., history of hip fracture/stroke) 8; dynamic factors include clinically significant depressive symptoms, cognitive impairment, a deteriorating physical condition, non-compliance with treatment and/or support consistent with self-neglect, hoarding of rubbish and persistent ne-glect of rotting food, denial of danger from malfunctioning appliances, disconnection of essential services and leaving home with doors unlocked and open (27). To the best of our knowledge, no studies have looked at factors that may have a specific protective effect against the risk of self-neglect.

Static risk factors for suicide have been identified in a recent systematic review 10: they include sex (while more male die by suicide, many more females attempt suicide), age (aged 15-24 years and those over 60 years), history of previous attempts, ethanol and drug abuse, sexual abuse, comorbid anxiety disorders (particularly panic disorder), personality disorders (antisocial and border-line), conduct disorder and oppositional defiance disorder, and identifiable stressful events. Identified dynamic factors include de-pression, impaired rational thinking, presence of organized plan, loneliness or debilitating medical illness, and experiences of adver-sity. Management or protective factors are presence of support networks, relief about not completing suicide, people relying on them for ongoing care, a sense of unfinished business, framework for meaning (e.g., religious belief), beliefs about the need to care for children, good self-esteem, self-confidence and awareness of significant others about their suicidal thoughts.

Finally, static factors for risk of violence to others include previous violence, young age at first violence, psychopathy, early maladjustment, personality disorder, prior supervision failure; dynamic factors include relationship instability, employment problems, substance use problems, lack of insight, negative attitudes, active symptoms of major mental illness, impulsivity and unresponsive-ness to treatment. Management or protective factors include level and type of personal support, dealing with stressors, working on medication adherence.

All the above could be combined in a qualitative or quantitative format which could be completed as a part of a multiaxial summary of the clinical assessment process.

CONCLUSIONS

The assessment of risk of self-neglect, suicide and violence to others is a task that clinicians routinely undertake. However, current classificatory systems do not make any provision for it. A dedicated risk management axis would help clinicians by integrating the findings of the assessment into the clinical recovery plan and may improve the utility of the classificatory systems by aligning them better to routine clinical work. Such an axis will need to combine actuarial and clinical factors. Our understanding of actuarial factors associated with the three types of risks has improved greatly in the recent years, making the development of such an axis now possible.