‘Single’ v. ‘panel’ appointed forensic mental observations: Is the referral process ethically justifiable?

T Schutte, MB ChB, FCPsych (SA); U Subramaney, MB BCh, BSc (Hons) (Psych), FCPsych (SA), MMed (Psych), PhD

Department of Psychiatry, School of Clinical Medicine, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

Corresponding author: T Schutte ( )

Objective. To compare the outcome and psychiatric morbidity of the forensic mental observation referrals, in the two legally created groups of detainees awaiting trial – the ‘singles’, representing the minor violent and non-violent offenders evaluated by a single-state appointed psychiatrist, v. the ‘panels’, representing the seriously violent offenders evaluated by two or more psychiatrists.

Methods. A retrospective record review covered 200 cases, comprising all individuals admitted to the forensic unit of Sterkfontein Hospital for 30 days psychiatric observation from January to August 2010. Pearson’s χ2 test for categorical data were used to determine statistical significance.

Results. Of 110 singles, 49 (44.55%) were found fit for trial and 40 (40.4%) were found criminally responsible. Of the 90 ‘panel’ cases 60 (66.67%) were found fit for trial and 57 (64.77%) were found criminally responsible (p=0.002 and p=0.001, respectively)

Conclusion. Those charged with seriously violent offences appear more likely to be found both fit and responsible, compared with those charged with less serious offences.

S Afr J BL 2013;6(2):64-68. DOI:10.7196/SAJBL.286

Section 79 of the Criminal Procedure Act1 deals, inter alia, with the court’s decision whether to appoint a single psychiatrist to perform a mental observation, or a panel of psychiatrists. According to the Act, in all cases where the accused is charged with ‘murder, culpable homicide, rape or another charge involving serious violence’ the court shall appoint a panel of at least two psychiatrists, one acting on behalf of the state and one appointed by the court for the accused. Should the court so decide, a third psychiatrist not in full time employment of the state, as well as a clinical psychologist, may be appointed to the panel. In all other cases, the court shall appoint a single psychiatrist to perform the observation on behalf of the state. These cases will include all non-violent or less violent/minor violent cases, including common assault, as well as property offences such as theft, common robbery, housebreaking, trespassing and malicious damage to property.

This legal procedure therefore divides the forensic patient population into two distinct groups, purely based on the seriousness of the alleged offence as determined by the Act. Furthermore, should the outcome of the observation indicate that the accused is either unfit to stand trial and/or was not criminally responsible for his/her actions at the time of committing the wrongful act, the court will also deal with these two groups of defendants in different ways. Those who committed offences involving serious violence (‘panels’) will be declared ‘state patients’ and referred back to the forensic hospital in terms of Section 42 of the Mental Healthcare Act,2 Act 17 of 2002, there to be indefinitely detained and treated pending a decision by a judge in chambers. Those who perpetrated minor offences (‘singles’) will be made involuntary psychiatric patients and referred to the general section of a psychiatric hospital, where they will be treated and dealt with as any other mental patient, and discharged based on clinical status in terms of Chapter V of the Mental Healthcare Act.

Despite numerous studies on the topic, the exact nature of the relationship between crime – violent or non-violent – and mental illness still remains elusive.

According to Taylor and Gunn,3 there are essentially three possible hypotheses:

• that the mentally ill are more prone to violence

• that mental illness has no tangible effect on violence/crime

• that mental illness reduces the risk of violence.

One can find support for all three proposals in the scientific literature.3

Though this is a very complex and confusing subject, there are some observations that appear to be generally agreed upon:

• A diagnosis of schizophrenia is strongly correlated with an increased risk of violence.4 , 5 One of the most commonly studied associations is the particular risk that the schizophrenic patient has of committing murder/homicide – several studies estimated that their overall risk of committing a murder is 10 - 20 times that expected in the general population.5 , 6

• The risk of committing murder is considered greater when the illness is associated with other important factors, such as a longstanding or extended duration of psychosis and the presence of active psychotic symptomatology (positive symptoms in particular).7 , 8

• The presence of an intellectual disability has also been strongly identified as a major risk factor for violence when compared with intellectually average individuals.9 , 10

Importantly, the above studies seem to be focused on serious violence and homicide in particular, and there seems to be only a small number of studies involving minor violent and non-violent offences committed by the mentally disordered. This may be due to the study design and selection criteria. Taylor and Gunn3 emphasise that most studies of this nature tend to look at either imprisoned or hospitalised offenders, who may represent the extremes of both violence and mental illness. They therefore decided to use remand prisoners – detainees awaiting trial – whom they felt would be most representative and least restricted in terms of violence and mental illness. This is comparable to the current study, in that all referrals for forensic observation in South Africa (SA), whether single or panel, are for detainees awaiting trial.

There are relatively few studies on this topic.11 The studies reviewed mostly focus on the outcome of the forensic observation process in terms of subjects’ triability and accountability, and the appropriateness of the referrals made by the courts.12-15 None compare outcomes in terms of the type of referral undertaken (single v. panel).


To highlight and compare the differences between single and panel forensic mental observations, in terms of both the final outcome of the observation process (as defined by the findings of fitness to stand trial and criminal responsibility) and of psychiatric morbidity. Fitness to stand trial entails the defendant’s capacity to follow the court proceedings so as to make a proper defence. Criminal responsibility reflects the defendant’s capacity to appreciate the wrongfulness of their actions and to be free to act in accordance with such appreciation, at the time of committing the alleged offence.


Study design, population and inclusion criteria

The study was a cross-sectional, retrospective forensic record and hospital file review. It included the first 200 admissions to the forensic unit of Sterkfontein Hospital of the year 2010, in terms of Section 79 of the Criminal Procedure Act.1 This included all admissions from 1 January to the end of August. The population included adult males and females, as well as adolescent males (forensic mental observations on adolescent females are currently not done at Sterkfontein Hospital).

Ethics approval was granted by the Human Research Ethics Committee of the University of the Witwatersrand, Johannesburg.

This report forms part of a larger study which included data on other socio-demographic, medical, psychiatric and forensic variables.

Data collection

Data were collected from the hospital clinical files, official prosecutor’s reports, charge sheets, social and collateral reports and the official psychiatric reports submitted to the court.

Data analysis

Data were analysed using Statistica version 10. Categorical data are presented as frequencies and percentages.

To compare associations between socio-demographic variables and outcomes in terms of fitness to stand trial and criminal responsibility in both groups, the study used bivariate and multivariate analyses with Pearson’s χ2 test and Fisher’s exact test for categorical data. Statistical significance was set at p<0.05.


The records of all 200 cases were available and there were no missing or grossly incomplete records. Forensic data are summarised in Table 1.

The forensic observations found a total of 109 (54.5%) subjects fit for trial and 97 (48.5%) criminally responsible (Figs 1 and 2).

Fig. 1. Subjects found fit to stand trial.

Fig.2. Subjects found criminally responsibile.

Type of referral and outcome

A total of 110 (55%) of the participants were referred as ‘single-psychiatrist’ cases or ‘singles’. The remaining 90 (45%) were referred as ‘panel-appointed’ cases or ‘panels’. In all of the latter, the panel of psychiatrists appointed came to unanimous conclusions regarding both the diagnosis and outcome in terms of fitness for trial and criminal responsibility.

Analysis showed that 49 (44.55%) of the singles were found fit for trial and 61 (55.45%) were regarded as unfit for trial. In the panel cases the situation was reversed: only 30 cases (33.33%) were found unfit for trial, whereas the majority, 60 (66.67%), were found fit to stand trial (χ2=9.77, degrees of freedom (df)=1 p=0.002). (Fig. 3).

Fig. 3. Comparison of fitness to stand trial for single- and panel-assessed subjects.

In 12 cases (6%), no comment could be made regarding criminal responsibility, due to a lack of sufficient information. In 1 case (0.5%) a finding of ‘diminished capacity’ was made. (Fig. 2). These 13 cases were excluded in the above and all other analyses of criminal responsibility, leaving 99 single cases and 88 panel cases (a total of 187 cases).

Regarding criminal responsibility, the difference between the two groups was more significant: 59 (59.6%) of the 99 singles were found not to have been criminally responsible for their actions, and 40 (40.40%) were found criminally responsible; in contrast only 31 (35.23%) of the 88 panel cases were regarded as not responsible, and 57 (64.77%) as criminally responsible (χ2=11.08, df=1, p=0.001) (Fig. 4).

Fig. 4. Comparison of criminal responsibility for single- and panel-assessed subjects.

Custodial status

Information on the detainees’ custodial status was available from the sources in 186 (93%) cases. Of the 86 (46.24%) who were found unfit for trial, 72 (83.72%) had been referred from prison and only 14 (28%) were granted bail. Of the 100 (53.76%) cases found fit to stand trial, 36 were granted bail and 64 were referred from prison (χ2=9.15, df=1, p=0.002). When comparing the accused’s custodial status with criminal responsibility and type of referral (single v. panel), no statistical significance was found (p=0.08 and p=0.54, respectively).

Reason for referral for observation

Odd or abnormal behaviour of the accused in court, during arrest or in custody seemed to be associated more strongly with referral of the single cases. In the panel cases, it seemed that the defence attorney was more likely to have difficulty in consulting with his/her client. Of the total of 60 (30%) cases referred due to their odd/abnormal behaviour in court, 44 (73.33%) were singles. Out of the 23 (11.5%) referred due to the attorney being unable to consult, 15 (65.22%) were ‘panel’ cases (χ2=10.56, df=1, p=0.0012).

There was a significantly higher association between cases in which an attorney was unable to consult, or which were referred as a result of the accused’s own statement/oral evidence in court, and the accused being found fit to stand trial: 18/23 cases (78.26%) referred due to the attorney being unable to consult were found fit for trial, as were all 15 accused (100%) who were referred due to their own statements in court. In contrast, of 60 cases referred due to odd or abnormal behaviour in court, 48 (80%) were found unfit to stand trial (χ2=43.59, df=2, p<0.0000).

Furthermore, the accused were more likely to have had no past psychiatric history revealed during observation in cases in which referrals were based on the attorney being unable to consult (60.87%) or on the accused’s own evidence, (66.67%). The majority of cases referred as a result of ‘documented proof of a psychiatric condition’ (35/43 (81.4%)) revealed a genuine past psychiatric history during observation (χ2=27.01, df=4, p<0.0000).

Past psychiatric history

Of the 83 cases (41.5%) where no past psychiatric history was present, 55 (66.27%) were found fit to stand trial. In contrast, of the 105 cases (52%) with a past psychiatric history, only 50 (47.62%) were found fit to stand trial (χ2=6.54, df=1, p=0.01). The same trend was seen regarding criminal responsibility, in which patients with a positive psychiatric history were more likely not to be found criminally responsible (χ2=12.65, df=1, p=0.0004). No significant association between past psychiatric history and the type of referral could be found (χ2=0.32, df=1, p=0.57).


In the current study, 48.5% of subjects were found both fit for trial and criminally responsible for their actions. This is consistent with figures of around 42 - 57% found in other SA studies of forensic mental observations.12-16 It appears that single cases are significantly more likely to be found both unfit for trial and not criminally responsible for their actions, compared with panel cases. This indicates that people with active and severe forms of mental illness are more likely to commit relatively minor or non-violent offences, while those with no significant active mental illness are more likely to commit offences involving serious violence. These findings contradict numerous studies which have found mental illness, specifically a diagnosis of schizophrenia, to be a significant risk factor for serious violence, especially murder.8-13

The current study is therefore in keeping with the work of Taylor and Gunn,3 who found that ‘serious personal and life-threatening violence was much more commonly committed by psychiatrically normal than by disturbed people’, and that only one third of murder suspects were considered mentally abnormal. The authors also speculate that their study population may have been ‘unnecessarily inflated’ by the authorities’ laying minor criminal charges against those who are clearly mentally ill, and should perhaps have been hospitalised instead.

This rings true from an SA perspective, where a resource-scarce mental healthcare system often lacks the capacity to deal with the enormous public demand for services.

It is possible that relatives of a difficult and aggressive mentally ill person may feel they have no option but to lay a relatively minor charge against their relative, such as common assault or malicious damage to property, not to seek justice or retribution but rather in a desperate attempt to seek help for their relative, as they perceive the mental healthcare system as being unable or unwilling to assist.

Furthermore, since the implementation of the current Mental Healthcare Act2 late in 2004, relatives battling to get assistance with an aggressive mentally ill person no longer have the option of accessing the mental healthcare system directly via the courts. Anecdotal reports and unpublished statistics from the forensic unit at Sterkfontein have shown an increase in the number of referrals for minor offences (singles) since the advent of the Act. Further research in this area is warranted.

The fact that those found unfit for trial were significantly more likely to be refused bail (i.e. sent for observation directly from prison) may indicate the courts’ reluctance to release potentially mentally unstable individuals out on bail. Surprisingly, the type of referral – which is essentially a function of the seriousness of the alleged offence – did not seem to play an important role in the decision whether to grant bail, as no significant association could be shown. It therefore appears that the courts regard the presence of possible mental illness as a greater factor in determing risk to the community than the seriousness of the crime of which the individual is accused.

Based on the findings of this study, and accepting the hypothesis that those without mental illness are more likely to commit serious crimes than the mentally ill, it is suggested that those with mental illness are being unfairly discriminated against during bail hearings and possibly inappropriately detained in prison, purely based on a stigma attached to being mentally ill.

‘Odd or strange or unusual behaviour in court’ seems to be associated most strongly with the accused ultimately being found unfit for trial (80% of cases). The majority of these cases (73%) were single observation cases. In contrast, an attorney being unable to consult had a significantly higher association with being found fit for trial, as well as with the panel referrals.

One possible explanation is that those charged with minor offences, who behave in a way that makes it obvious to the court that they are mentally ill, would be directly referred for mental observation (and kept in prison while on a waiting list, as discussed above). In these cases an attorney may not even be appointed before the case is referred.

It is difficult to explain why the panel cases would be more strongly associated with the attorney being unable to consult. Clearly, proper defence counsel is prudent when one is charged with a more serious offence. However, the fact that the majority of the panel cases were found both fit and responsible, and 78% of the cases in which an attorney was unable to consult were also found fit for trial, makes it hard to understand what the exact reasons were for the inability to consult. One could cautiously speculate that this claim may be a tactical move by the defence counsel, trying to defend a very difficult case. This is supported by the fact that the majority of these cases did not reveal any significant past psychiatric history.

The fact that all subjects whose cases were referred solely due to their own oral evidence in court were found fit for trial speaks for itself. One could speculate that an accused’s own account of their mental illness in court should not be regarded as sufficient evidence to warrant a referral for forensic observation.

Study limitations

The sample size may be too small to draw generalisable conclusions in all instances and correlates.

Throughout the study, fitness for trial was used as the major indicator of outcome of the observation cases: Categorical data on fitness were available in all 200 cases, whereas data on criminal responsibility were incomplete in 6% of cases. Criminal responsibility is also a retrospective analysis, often based extensively on collateral information rather than objective assessment. This is in contrast to fitness for trial, which is a more accurate ‘here and now’ assessment based more closely upon the actual 30 day observation findings.


The results of this study support the hypothesis that single observation cases, representing relatively minor and/or non-violent offences, are more likely to be found both unfit for trial and not criminally responsible for their actions, compared to the panel cases representing more serious acts of violence and aggression. This may imply that the mentally ill offender is more likely to commit a greater proportion of relatively minor offences than offenders with no current mental illness, who seem to be associated more frequently with more serious acts of violence.

These results may be useful to the referring courts and probation officers alike, specifically when reviewing their reasons for making the referral and in terms of developing a more rigorous screening procedure as part of their own mental inquiry. This may ultimately reduce the ever-growing waiting lists for forensic observation cases across the country.

This study may also aid in the destigmatisation of mental illness, where the mentally ill are all too often portrayed as the stereotypical ‘crazed killers’ and unfairly regarded as highly dangerous within their communities.


      1. South African Government. Criminal Procedure Act 51 of 1977. Pretoria: Government Printer, 1977.
      2. South African Government. Mental Healthcare Act 17 of 2002. Pretoria: Government Printer, 2002.
      3. Taylor PJ, Gunn J. Violence and psychosis. I. Risk of violence among psychotic men. Br Med J (Clin Res Ed) 1984;288:1945-1949.
      4. Grossman LS, Haywood TW, Cavanaugh JL, Davis JM, Lewis DA. State psychiatric hospital patients with past arrests for violent crimes. Psychiatr Serv 1995;46(8):790-795.
      5. Wootton L, Buchanan A, Leese M, et al. Violence in psychosis: Estimating the predictive validity of readily accessible clinical information in a community sample. Schizophr Res 2008;101(1-3):176-184. []
      6. Schanda H, Knecht G, Schreinzer D, Stompe T, Ortwein-Swoboda G, Waldhoer T. Homicide and major mental disorders: A 25 year study. Acta Psychiatr Scand 2004;110:98-107. []
      7. Fazel S, Gulati G, Linsell L, Geddes JR, Grann M. Schizophrenia and violence: Systematic review and meta-analysis. PLoS Med 2009;6(8):e1000120. []
      8. Humphreys MS, Johnstone EC, MacMillan JF, Taylor PJ. Dangerous behaviour preceding first admissions for schizophrenia. Br J Psych 1992;161:501-505.
      9. Link BG, Andrews H, Cullen FT. The violent and illegal behaviour of mental patients reconsidered. American Sociological Review 1992;57:275-292.
      10. Søndenaa E, Rasmussen K, Nøttestad JA. Forensic issues in intellectual disability. Curr Opin Psychiatry 2008;21(5):449-453. []
      11. Rojahn J, Matson JL, Naglieri JA, Mayville E. Relationships between psychiatric conditions and behaviour problems among adults with mental retardation. Am J Ment Retard 2004;109(1):21-33.
      12. Calitz FJW, van Rensburg PHJJ, et al. Psychiatric evaluation of offenders referred to the Free State psychiatric complex according to Sections 77 and/or 78 of the Criminal Procedure Act. South African Journal of Psychiatry 2006;12(3):47-50.
      13. Gagiano CA, van Rensburg PHJJ, Verschoor T. Unnecessary committals for forensic observation: Sections 77 and 78 of the Criminal Procedure Act 51 of 1977. S African L J 1991;108(1):714-718.
      14. Verster I, van Rensburg PHJJ. Mental disorders in patients referred for psychiatric observation after committing homicide. Journal of Juridical Science 1999;24(1):58-66.
      15. Pretorius PJ, van Rensburg PHJJ, Gagiano CA, Oosthuizen H. Daggagebruik deur wetsoortreders wat deur die hof vir 30 dae psigiatriese observasie verwys word. Geneeskunde 1996;March/April:43-50.
      16. Slabber M, van Rensburg PHJJ, Gagiano CA, Joubert G, Van Dyk A. The relationship between MRI brain abnormalities and psychiatric disorders. Geneeskunde 1996;38:6-12.

Article Views

Abstract views: 1741
Full text views: 3671

Comments on this article

*Read our policy for posting comments here