- Email: [email protected]
Aversive therapy
919
Aversive therapy of aversive treatments for severe selfinjurious behaviour (SSIB) in developmentally disabled children and adults has been hotly debated. Some people believe this therapy is inhuman and demeaning to patients and may brutalise those who administer it. However, condemnation is unhelpful unless one can suggest a realistic alternative. S SIB is in itself very distressing to all concerned; this condition may be associated with serious tissue damage and even death and is often very difficult to treat by rewardbased behavioural interventions. SSIB is almost always associated with profound learning difficulties. Milder stereotyped behaviours are found in 10-15% of children and adults with severe learning difficulties and usually respond to non-aversive treatments. Thus, use of aversive methods, when other techniques have failed, would be entertained in only 1 in 10 000 of the mentally handicapped population.1,2 The aetiology of SSIB is poorly understood. Four main possibilities have been proposed.3 An organic component is well recognised in some syndromes (eg, Lesch-Nyhan or de Lange syndromes) and the close association between SSIB and extreme organic dysfunction in patients with severe learning difficulties likewise indicates causal organic factors (in some specific syndromes SSIB responds to pharmacological treatments1). A second hypothesis is that SSIB may be an extension of well-recognised developmental stereotypes such as rocking, handflapping, and head-banging; these behaviours are self-stimulating and tension-reducing for the individual and are also used as a means of protective communication with care givers and to seek their attention.3Neither hypothesis can explain all cases or the degree of persistence of SSIB in handicapped individuals. The third possibility is that SSIB is a sensory self-stimulation that the individual finds highly rewarding, albeit damaging and painful; the higher frequency of SSIB in individuals with sensory losses and in institutionalised patients with few external stimuli accords with this view. The last main factor may be behavioural learning—eg, when caregivers unwittingly reinforce the stereotyping by their responses. One or more of these factors may contribute to the aetiology of SSIB in an individual case and the behaviour may occur initially in response to a seemingly unimportant event. CarrA offered a series of steps for screening individuals with SSIB to determine the most likely cause and so guide choice of interventions. Behavioural treatments such as positive and differential reinforcement of non-self-injurious behaviours are most widely used. Aversive techniques are not synonymous with electric shock--eg, in overcorrection, the individual may be required to make arm or leg movements that are incompatible with the self-injurious movements.s Since SSIB is usually controllable by use of restraints and protective The
use
by non-aversive behavioural methods, aversive techniques incorporating unpleasant tastes and smells and especially electric shock, which is understandably regarded with extreme antipathy, need be considered only in exceptional cases. The value of aversive methods is that they can provide a last-ditch chance of stopping protracted, serious, and even life-threatening self-injury; aversive stimulation, consequent upon the injurious behaviour, can act speedily and have lasting effects.3 If, after a detailed functional analysis of the behaviour, clothing
or
all reasonable attempts with non-aversive methods have failed, it would be wrong not to consider an
approach which, although outwardly distasteful, can rapidly end a very distressing episode for the patient. Individual responses to aversive and non-aversive of SSIB are highly variable and it is often predict the likelihood of success. Skilled staff should try non-aversive methods first and, if these fail, be experienced enough to turn to aversive techniques. Until lately it has been difficult to administer and regulate mild electric shock therapy; the advent of microtechnology now allows smaller electric shocks to be administered remotely via a device worn by the patient (a sensor applied to the head to detect head banging is linked by radio to an electric shock stimulator on an arm or leg). Those opposed to aversive methods feel that ease of application will increase the number of cases in which such therapy is used, with the attendant risk of abuse, and may diminish interest in developing less punitive techniques. Linschied and colleagues2 reported the clinical evaluation of the self-injurious behaviour inhibiting system, in which a stimulator supplies a carefully regulated shock contingent upon head movements in severe head banging. They acknowledge that the system is appropriate in only a few cases but argue that it permits precise and automatic delivery of shocks, that it does not interfere with the patients’ normal activities, that the electrical stimulus is a highly discrete event, and that it is possible to administer the shocks without physical risk. In all these respects contingent electrical stimulation permits far more acceptable aversive stimulation than methods such as squirting of lemon juice in the patient’s mouth, water mist, or facial covering. Brief, mild electrical stimulation produced rapid and large decreases in longstanding and previously untreatable SSIB. Thus, the disadvantages of aversive methods are that they carry the risk of causing harm or suffering if wrongly used; that they are open to misapplication when other methods could perhaps have been introduced; and that staff who are overstretched or inexperienced may be tempted to turn to them as short-cuts. Nevertheless, Linschied and co-workers have shown that, with careful control, aversive treatment can bring substantial benefits with few side-effects. Whenever aversive therapies are to be tried both patients and staff must have safeguards: the treatments difficult to
920
programme should be reviewed by an independent ethical committee, which could examine and agree the need for such treatment in individual cases. This approach would go a long way to meeting treatment
the objections of those who are fearful of misuse of the techniques and yet ensure that treatment can be made available to the very few people with highly distressing and damaging behaviour. Campbell HJ. Causes of severe self-injurious behaviour. In: Mittler P, deJong JM, eds. Frontiers of knowledge in mental retardation. Vol II. Biomedical aspects. Baltimore: University Park Press, 1980: 285-92. Linschied TR, Iwata BA, Ricketts RW, Williams DE, Griffin TC. Clinical evaluation of the self-injurious behaviour system. J Appl Behav Anal 1990; 23: 53-78. Murphy G, Wilson B. Self injurious behaviour. Kidderminster: British Institute of Mental Handicap, 1985. Carr EG. The motivation of self-injurious behaviour: a review of some hypotheses. Psychol Bull 1977; 84: 800-16. Conley OS, Wolery MR. Treatment by over-correction of self-injurious eye-gouging in pre-school blind children. J Behav Ther Exp Psychiatry 1980; 11: 121-25.
1. Corbett JA,
2.
3. 4. 5.
Safety of high-titre measles vaccine Measles vaccine has been in use in most developed countries for more than 20 years; its safety record is good and its protective efficacy is very satisfactory. Where high vaccine uptake rates are achieved the disease is generally well controlled, although sporadic cases, mainly from importations, and occasional outbreaks in poorly protected population groups, are still encountered. The usual recommendation is that the vaccine should be given at 12-18 months when maternally acquired antibodies have waned and maximum antibody responses can be assured. However, in many developing countries, where infants tend to lose maternal antibody earlier, where general resistance to infection is often poor, and where crowded living conditions lead to early exposure, measles commonly occurs in the second 6 months of life and is an important cause of death. 1,2 Consequently, since 1976 the World Health Organisation (WHO) has recommended that the vaccine should be given at 9 months-a compromise between the need to provide early protection and the inferior antibody response achieved by vaccination at an early age.3 A high-titre vaccine (defined as more than 5-0 loglo infectious units) first produced in the Institute of Immunology, Zagreb, Yugoslavia, by further attenuation of the Edmonston strain (EZ vaccine) was found to induce a good antibody 4 response when given as early as 4 months of age.4 Since 1989, therefore, WHO has recommended the phased introduction of high-titre measles vaccine to be given at 6 months of age in countries with a high measles mortality under the age of 9 months.5 Evidence from several trials and from limited surveillance of the use of high-titre vaccines indicates that they are both safe and effectived However, in this issue (p 903) investigators in Senegal report an excess mortality in infants who were followed for 24-39 months after being given one of two high-titre
(EZ or Schwarz) at 5 months of age, by comparison with children given a placebo at 5 months followed by standard low-titre Schwarz vaccine at 10 months and with non-participants. This finding emerged from a study that was designed as a randomised controlled trial of two high-titre vaccines in a community with an established routine mortality surveillance system based on weekly household visits. If the observed relation was causal, it will have profound implications for the continued use of high-titre measles, but there are good reasons for exercising caution before reaching such a conclusion: (a) the study was not designed to test a hypothesis about late mortality in recipients of the different vaccines, so methodological difficulties were encountered; (b) there was no significant difference in mortality during the first 6 months of follow-up; and (c), most important, there were various causes of death with no suggestion of a common pathogenesis. A review of the Senegal evidence by an independent expert panel convened by WHO concluded that the results could not be used for decision-making and advised no change in the existing policy on use of high-titre vaccines.6Nevertheless, further careful monitoring of use of the vaccine and possible associated adverse events, including morbidity as well as mortality, is essential to confirm or refute the Senegal findings. WHO have asked all investigators conducting trials of high-titre vaccines to include evaluation of late events in their protocols. Further studies should include other high-titre vaccines, in particular the AIK-C strain. vaccines
Leoning WEK, Coovadia HM. Age-specific occurrence rates of measles in urban, peri-urban, and rural environments: implications for time of vaccination. Lancet 1983; ii: 324-26. 2. Taylor WR, Kalisa R, Ma-Disu M, Weiman JM. Measles control efforts in urban Africa complicated by high incidence in the first year of life. Am J Epidemiol 1988; 127: 788-94. 3. Ministry of Health of Kenya and the World Health Organisation. Measles immunity in the first year after birth and the optimum age for vaccination in Kenyan children. Bull WHO 1977; 55: 21-30. 1.
4. Whittle HC, Rowland MGM, Mann GF, Lamb WH, Lewis RA. Immunisation of 4-6-month old Gambian children with EdmonstonZagreb measles vaccine. Lancet 1984; ii: 834-37. 5. Expanded Programme on Immunisation. Global Advisory Group. Weekly Epidemiol Rec 1990; 65: 5-12. 6. Expanded Programme on Immunisation. Safety and efficacy of high titre measles vaccine at 6 months of age. Weekly Epidemiol Rec 1991; 66: 249-51. 7. Whittle H, Hanlon P, O’Neill K, et al. Trial of high-dose EdmonstonZagreb measles vaccine in the Gambia: antibody response and side-effects. Lancet 1988; ii: 811-14. 8. Markowitz LE, Sepulveda J, Diaz-Ortega JL, et al. Immunisation of six month old infants with different doses of Edmonston-Zagreb and Schwarz measles vaccine. N Engl J Med 1990; 332: 580-87. 9. Aaby P, Jensen TG, Hansen HL, et al. Trial of high-dose EdmonstonZagreb measles vaccine in Guinea-Bassau: protective efficacy. Lancet 1988; ii: 809-11. 10. Job JS, Halsey NA, Boulos R, et al. Successful immunisation of infants at 6 months of age with high-dose Edmonston-Zagreb measles vaccine. Pediatr Infect Dis J 1991; 10: 303-11. 11. Khanum S, Uddin N, Garelick H, Mann G, Tomkins A. Comparison of Edmonston-Zagreb and Schwarz strains of measles vaccine given by aerosol or subcutaneous injection. Lancet 1987; i: 150-53. 12. Tidjani O, Grunitsky B, Guérin N, Lévy-Brahl D, Lecam N, Zukereff C. Serological effects of Edmonston-Zagreb, Schwarz, and AIK-C measles vaccine strains given at ages 4-5 or 8-10 months. Lancet 1989; ii: 1357-60.
Aversive therapy of aversive treatments for severe selfinjurious behaviour (SSIB) in developmentally disabled children and adults has been hotly debated. Some people believe this therapy is inhuman and demeaning to patients and may brutalise those who administer it. However, condemnation is unhelpful unless one can suggest a realistic alternative. S SIB is in itself very distressing to all concerned; this condition may be associated with serious tissue damage and even death and is often very difficult to treat by rewardbased behavioural interventions. SSIB is almost always associated with profound learning difficulties. Milder stereotyped behaviours are found in 10-15% of children and adults with severe learning difficulties and usually respond to non-aversive treatments. Thus, use of aversive methods, when other techniques have failed, would be entertained in only 1 in 10 000 of the mentally handicapped population.1,2 The aetiology of SSIB is poorly understood. Four main possibilities have been proposed.3 An organic component is well recognised in some syndromes (eg, Lesch-Nyhan or de Lange syndromes) and the close association between SSIB and extreme organic dysfunction in patients with severe learning difficulties likewise indicates causal organic factors (in some specific syndromes SSIB responds to pharmacological treatments1). A second hypothesis is that SSIB may be an extension of well-recognised developmental stereotypes such as rocking, handflapping, and head-banging; these behaviours are self-stimulating and tension-reducing for the individual and are also used as a means of protective communication with care givers and to seek their attention.3Neither hypothesis can explain all cases or the degree of persistence of SSIB in handicapped individuals. The third possibility is that SSIB is a sensory self-stimulation that the individual finds highly rewarding, albeit damaging and painful; the higher frequency of SSIB in individuals with sensory losses and in institutionalised patients with few external stimuli accords with this view. The last main factor may be behavioural learning—eg, when caregivers unwittingly reinforce the stereotyping by their responses. One or more of these factors may contribute to the aetiology of SSIB in an individual case and the behaviour may occur initially in response to a seemingly unimportant event. CarrA offered a series of steps for screening individuals with SSIB to determine the most likely cause and so guide choice of interventions. Behavioural treatments such as positive and differential reinforcement of non-self-injurious behaviours are most widely used. Aversive techniques are not synonymous with electric shock--eg, in overcorrection, the individual may be required to make arm or leg movements that are incompatible with the self-injurious movements.s Since SSIB is usually controllable by use of restraints and protective The
use
by non-aversive behavioural methods, aversive techniques incorporating unpleasant tastes and smells and especially electric shock, which is understandably regarded with extreme antipathy, need be considered only in exceptional cases. The value of aversive methods is that they can provide a last-ditch chance of stopping protracted, serious, and even life-threatening self-injury; aversive stimulation, consequent upon the injurious behaviour, can act speedily and have lasting effects.3 If, after a detailed functional analysis of the behaviour, clothing
or
all reasonable attempts with non-aversive methods have failed, it would be wrong not to consider an
approach which, although outwardly distasteful, can rapidly end a very distressing episode for the patient. Individual responses to aversive and non-aversive of SSIB are highly variable and it is often predict the likelihood of success. Skilled staff should try non-aversive methods first and, if these fail, be experienced enough to turn to aversive techniques. Until lately it has been difficult to administer and regulate mild electric shock therapy; the advent of microtechnology now allows smaller electric shocks to be administered remotely via a device worn by the patient (a sensor applied to the head to detect head banging is linked by radio to an electric shock stimulator on an arm or leg). Those opposed to aversive methods feel that ease of application will increase the number of cases in which such therapy is used, with the attendant risk of abuse, and may diminish interest in developing less punitive techniques. Linschied and colleagues2 reported the clinical evaluation of the self-injurious behaviour inhibiting system, in which a stimulator supplies a carefully regulated shock contingent upon head movements in severe head banging. They acknowledge that the system is appropriate in only a few cases but argue that it permits precise and automatic delivery of shocks, that it does not interfere with the patients’ normal activities, that the electrical stimulus is a highly discrete event, and that it is possible to administer the shocks without physical risk. In all these respects contingent electrical stimulation permits far more acceptable aversive stimulation than methods such as squirting of lemon juice in the patient’s mouth, water mist, or facial covering. Brief, mild electrical stimulation produced rapid and large decreases in longstanding and previously untreatable SSIB. Thus, the disadvantages of aversive methods are that they carry the risk of causing harm or suffering if wrongly used; that they are open to misapplication when other methods could perhaps have been introduced; and that staff who are overstretched or inexperienced may be tempted to turn to them as short-cuts. Nevertheless, Linschied and co-workers have shown that, with careful control, aversive treatment can bring substantial benefits with few side-effects. Whenever aversive therapies are to be tried both patients and staff must have safeguards: the treatments difficult to
920
programme should be reviewed by an independent ethical committee, which could examine and agree the need for such treatment in individual cases. This approach would go a long way to meeting treatment
the objections of those who are fearful of misuse of the techniques and yet ensure that treatment can be made available to the very few people with highly distressing and damaging behaviour. Campbell HJ. Causes of severe self-injurious behaviour. In: Mittler P, deJong JM, eds. Frontiers of knowledge in mental retardation. Vol II. Biomedical aspects. Baltimore: University Park Press, 1980: 285-92. Linschied TR, Iwata BA, Ricketts RW, Williams DE, Griffin TC. Clinical evaluation of the self-injurious behaviour system. J Appl Behav Anal 1990; 23: 53-78. Murphy G, Wilson B. Self injurious behaviour. Kidderminster: British Institute of Mental Handicap, 1985. Carr EG. The motivation of self-injurious behaviour: a review of some hypotheses. Psychol Bull 1977; 84: 800-16. Conley OS, Wolery MR. Treatment by over-correction of self-injurious eye-gouging in pre-school blind children. J Behav Ther Exp Psychiatry 1980; 11: 121-25.
1. Corbett JA,
2.
3. 4. 5.
Safety of high-titre measles vaccine Measles vaccine has been in use in most developed countries for more than 20 years; its safety record is good and its protective efficacy is very satisfactory. Where high vaccine uptake rates are achieved the disease is generally well controlled, although sporadic cases, mainly from importations, and occasional outbreaks in poorly protected population groups, are still encountered. The usual recommendation is that the vaccine should be given at 12-18 months when maternally acquired antibodies have waned and maximum antibody responses can be assured. However, in many developing countries, where infants tend to lose maternal antibody earlier, where general resistance to infection is often poor, and where crowded living conditions lead to early exposure, measles commonly occurs in the second 6 months of life and is an important cause of death. 1,2 Consequently, since 1976 the World Health Organisation (WHO) has recommended that the vaccine should be given at 9 months-a compromise between the need to provide early protection and the inferior antibody response achieved by vaccination at an early age.3 A high-titre vaccine (defined as more than 5-0 loglo infectious units) first produced in the Institute of Immunology, Zagreb, Yugoslavia, by further attenuation of the Edmonston strain (EZ vaccine) was found to induce a good antibody 4 response when given as early as 4 months of age.4 Since 1989, therefore, WHO has recommended the phased introduction of high-titre measles vaccine to be given at 6 months of age in countries with a high measles mortality under the age of 9 months.5 Evidence from several trials and from limited surveillance of the use of high-titre vaccines indicates that they are both safe and effectived However, in this issue (p 903) investigators in Senegal report an excess mortality in infants who were followed for 24-39 months after being given one of two high-titre
(EZ or Schwarz) at 5 months of age, by comparison with children given a placebo at 5 months followed by standard low-titre Schwarz vaccine at 10 months and with non-participants. This finding emerged from a study that was designed as a randomised controlled trial of two high-titre vaccines in a community with an established routine mortality surveillance system based on weekly household visits. If the observed relation was causal, it will have profound implications for the continued use of high-titre measles, but there are good reasons for exercising caution before reaching such a conclusion: (a) the study was not designed to test a hypothesis about late mortality in recipients of the different vaccines, so methodological difficulties were encountered; (b) there was no significant difference in mortality during the first 6 months of follow-up; and (c), most important, there were various causes of death with no suggestion of a common pathogenesis. A review of the Senegal evidence by an independent expert panel convened by WHO concluded that the results could not be used for decision-making and advised no change in the existing policy on use of high-titre vaccines.6Nevertheless, further careful monitoring of use of the vaccine and possible associated adverse events, including morbidity as well as mortality, is essential to confirm or refute the Senegal findings. WHO have asked all investigators conducting trials of high-titre vaccines to include evaluation of late events in their protocols. Further studies should include other high-titre vaccines, in particular the AIK-C strain. vaccines
Leoning WEK, Coovadia HM. Age-specific occurrence rates of measles in urban, peri-urban, and rural environments: implications for time of vaccination. Lancet 1983; ii: 324-26. 2. Taylor WR, Kalisa R, Ma-Disu M, Weiman JM. Measles control efforts in urban Africa complicated by high incidence in the first year of life. Am J Epidemiol 1988; 127: 788-94. 3. Ministry of Health of Kenya and the World Health Organisation. Measles immunity in the first year after birth and the optimum age for vaccination in Kenyan children. Bull WHO 1977; 55: 21-30. 1.
4. Whittle HC, Rowland MGM, Mann GF, Lamb WH, Lewis RA. Immunisation of 4-6-month old Gambian children with EdmonstonZagreb measles vaccine. Lancet 1984; ii: 834-37. 5. Expanded Programme on Immunisation. Global Advisory Group. Weekly Epidemiol Rec 1990; 65: 5-12. 6. Expanded Programme on Immunisation. Safety and efficacy of high titre measles vaccine at 6 months of age. Weekly Epidemiol Rec 1991; 66: 249-51. 7. Whittle H, Hanlon P, O’Neill K, et al. Trial of high-dose EdmonstonZagreb measles vaccine in the Gambia: antibody response and side-effects. Lancet 1988; ii: 811-14. 8. Markowitz LE, Sepulveda J, Diaz-Ortega JL, et al. Immunisation of six month old infants with different doses of Edmonston-Zagreb and Schwarz measles vaccine. N Engl J Med 1990; 332: 580-87. 9. Aaby P, Jensen TG, Hansen HL, et al. Trial of high-dose EdmonstonZagreb measles vaccine in Guinea-Bassau: protective efficacy. Lancet 1988; ii: 809-11. 10. Job JS, Halsey NA, Boulos R, et al. Successful immunisation of infants at 6 months of age with high-dose Edmonston-Zagreb measles vaccine. Pediatr Infect Dis J 1991; 10: 303-11. 11. Khanum S, Uddin N, Garelick H, Mann G, Tomkins A. Comparison of Edmonston-Zagreb and Schwarz strains of measles vaccine given by aerosol or subcutaneous injection. Lancet 1987; i: 150-53. 12. Tidjani O, Grunitsky B, Guérin N, Lévy-Brahl D, Lecam N, Zukereff C. Serological effects of Edmonston-Zagreb, Schwarz, and AIK-C measles vaccine strains given at ages 4-5 or 8-10 months. Lancet 1989; ii: 1357-60.