- Email: [email protected]
Child abuse and neglect
Comment
Program, participants were asked to fill out the Quality of Well-Being Scale each year.4 The questionnaire revealed that patients with impaired tolerance to glucose, who received an intensive lifestyle intervention, rated their quality of life more highly than those who received a pharmacological intervention (metformin).4 Fourth, there is the issue of time horizon. A long-term perspective should be taken if a health intervention has an impact beyond the period for which data from clinical trials are available. In this case, modelling techniques, such as decision, epidemiological, and Markov models, can be used to project the costs and consequences of the intervention.5 In the Diabetes Prevention Program, the cost-effectiveness of metformin improved drastically when the time horizon was extended from within the trial to lifetime.6 Besides the time horizon, when there are multiple stakeholders, it may be helpful to conduct economic evaluations from different perspectives. The US Panel on Cost Effectiveness in Health and Medicine recommends that, for comparability, the societal perspective be provided as a reference case along with other perspectives.7 Public-health agencies might also find it useful to have evaluations done from the budgetary perspective. Conducting economic evaluations alongside clinical trials is a viable means to evaluate health interventions, especially when such evaluations are supplemented by modelling techniques.8–10 However, if such studies are to contribute to policy debates, it is important to generate
information in a way that allows us to do what economic evaluations do best—to inform decision-making about resource allocation that maximises population health.11 *Kumiko Imai, Ping Zhang Division of Diabetes Translation, Centers for Disease Control and Prevention, Atlanta, GA 30341, USA [email protected] We declare that we have no conflict of interest. 1
Taylor RS, Drummond MF, Salkeld G, Sullivan SD. Inclusion of cost effectiveness in licensing requirements of new drugs: the fourth hurdle. BMJ 2004; 329: 972–75. 2 UCLA Stroke/Vascular Neurology Program. Frequently asked questions. 2002: http://strokeprotect.mednet.ucla.edu/faq.htm (accessed April 13, 2005). 3 Diabetes Prevention Program Research Group. Costs associated with the primary prevention of type 2 diabetes mellitus in the Diabetes Prevention Program. Diabetes Care 2003; 26: 36–47. 4 The Diabetes Prevention Program Research Group. Within-trial cost-effectiveness of lifestyle intervention or metformin for the primary prevention of type 2 diabetes. Diabetes Care 2003; 26: 2518–23. 5 Drummond MF, O’Brien B, Stoddart GL, Torrance GW. Methods for the economic evaluation of health care programmes, 2nd edn. Oxford: Oxford University Press, 1997. 6 Herman WH, Hoerger TJ, Brandle M, et al. The cost-effectiveness of lifestyle modification or metformin in preventing type 2 diabetes in adults with impaired glucose tolerance. Ann Intern Med 2005; 142: 323–32. 7 Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-effectiveness in health and medicine. New York: Oxford University Press, 1996. 8 O’Sullivan AK, Thompson D, Drummond MF. Collection of health economic data alongside clinical trials: is there a future for piggy-back evaluations? Value Health 2005; 8: 67–79. 9 Ramsey SD, McIntosh M, Sullivan SD. Design issues for conducting cost-effectiveness analyses alongside clinical trials. Annu Rev Public Health 2001; 22: 129–41. 10 O’Brien B. Economic evaluation of pharmaceuticals: Frankenstein’s monster or vampire of trials? Medical Care 1996; 34 (suppl): DS99–108. 11 Murray CJL, Evans DB, Acharya A, Baltussen RMPM. Development of WHO guidelines on generalized cost-effectiveness analysis. Health Econ 2000; 9: 235–51.
Child abuse and neglect Published online May 5, 2005 DOI:10.1016/S0140-6736(05) 66389-1 See Articles page 1786
1750
The paper in today’s Lancet by Harriet MacMillan and colleagues focuses attention once again on the important topic of child abuse and neglect, which continues to be a significant health issue in many countries. A recent population-based survey in the UK, for example, showed that the prevalence of all forms of abuse is considerably higher than that indicated by objective measures.1 Furthermore, the negative sequelae have been documented across various developmental domains, including cognition, language, learning, and socioemotional development.2 Whilst there is some evidence emerging about the potential effectiveness of prevention initiatives,3,4 there is considerably less consensus about what works when
abuse has already occurred. Many different approaches have been used to date (eg, family therapy, multisystem therapy, parenting programmes, educational and skilldevelopment programmes, problem solving, stress resolution, and anger management) but MacMillan and colleagues’ study is one of the first rigorous evaluations of the effectiveness of home visits by nurses in stopping child abuse in families already in the child-protection system (ie, preventing abuse recurrence). MacMillan and colleagues’ results show that not only was there no difference in the recurrence of child physical abuse and neglect in the intervention group compared with standard services, but also that, on the basis of data from hospital records, there was a higher www.thelancet.com Vol 365 May 21, 2005
Comment
www.thelancet.com Vol 365 May 21, 2005
Rights were not granted to include this image in electronic media. Please refer to the printed journal
Photolibrary.com
recurrence of abuse in the intervention group. The authors quite rightly consider whether this finding represents a harmful effect of the intervention or the type of surveillance bias that occurs in abuse prevention studies. It seems likely that it was the latter, and such bias has led some to question the appropriateness of measuring abuse as an outcome in studies of this nature.5 It is important that interventions to reduce abusive parenting have a sound theoretical basis, linking the specific content or mechanisms of the intervention with the problem being addressed. The intervention being assessed in MacMillan and colleagues’ study was underpinned by an ecological model6 that aimed to target problems at several levels (eg, from the individual, through the family, to the environment). The use of an ecological model is a valuable theoretical basis from which to proceed, and one that has proved effective in the primary prevention of abuse.4 In many cases, journal restrictions on article length, however, preclude the possibility of knowing exactly what such complex psychosocial interventions actually comprise, making it difficult for future programme developers to build on existing knowledge or to assess why an intervention did not have a bigger impact. There is increasing recognition that the effectiveness of intervention programmes, such as home visiting, might depend on the success with which the programme provider establishes a trusting relationship with the recipient.7 If such relationships are central to the programmes’ effectiveness, as the literature on both counselling and psychotherapy suggests, there is an urgent need for studies to begin measuring the success with which this is achieved. It is not clear whether trusting relationships were a core part of MacMillan and colleagues’ intervention, or indeed whether the intervention was lacking other key components, such as sufficient intensity. Whilst greater intensity would have increased the cost, it seems likely that higher costs would have been justified, given the massive expenditure required for child abuse and neglect at both an individual and societal level. Trials that involve participants as active agents in initiating psychological and behavioural changes are fundamentally different from trials of treatments that involve the participants as passive recipients (as is the
case for many medical interventions). Such trials are delivered by people with differing levels of skills, to families with different histories, preoccupations, circumstances, different levels of readiness for change, and different levels of commitment to the process, all of which determine whether an intervention of this nature proves to be effective. The implication of these differences is that the impact of such interventions is likely to be small, and there is, as such, a need for much larger (and more expensive) studies. Indeed, the survival curve in MacMillan and colleagues’ study shows a 10% reduction in physical abuse, which would seem to be an important improvement. However, the sample size was based on a 25% reduction, which might have been overly optimistic. MacMillan and colleagues’ results make for depressing reading. More than 50% of abused children remaining in the home continued to be abused, and there is little evidence currently available about how to prevent this abuse from happening without removing the child from the home. The benefits of standard services are for the better part not known, but as MacMillan points out, most fall seriously short of the sort of service evaluated in their study. Whilst these findings might highlight the importance of the primary prevention of child abuse and neglect, primary prevention will not be effective in all 1751
Comment
cases, and there is an urgent need to identify an effective secondary preventive intervention to reduce the recurrence of abuse, and to limit the impact that such abuse has on children’s health. The secondary prevention of abuse might require that the balance of investment is now in favour of establishing from existing evidence the core components of potentially successful interventions.
1
2
3
4
5
*Jane Barlow, Sarah Stewart-Brown Warwick Medical School, Medical School Building, University of Warwick, Coventry CV4 7AL, UK [email protected] We declare that we have no conflict of interest.
6
7
Cawson P, Wattam C, Brooker S, Kelly G. Child maltreatment in the United Kingdom: a study of the prevalence of child abuse and neglect. London: National Society for the Prevention of Cruelty to Children, 2001. Cicchetti D. Perspectives from developmental psychopathology. In: Cicchetti D, Carlson V, eds. Child maltreatment: theory and research on the causes and consequences of child abuse and neglect. New York: Cambridge University Press, 1989. Olds DL, Eckenrode J, Henderson CR, et al. Long-term effects of home visitation on maternal life course and child abuse and neglect: fifteen year follow-up of a randomised trial. JAMA 1997; 278: 637–43. Olds DL, Henderson C, Tatelbaum R, Chamberline R. Preventing child abuse and neglect: a randomised trial of nurse home visitation. Pediatrics 1986; 78: 65–78. Roberts I, Kramer MS, Suissa S. Does home-visiting prevent childhood injury? A systematic review of randomised controlled trials. BMJ 1996; 312: 29–33. Bronfenbrenner U. The ecology of human development: experiments by nature and design. Cambridge, Massachusetts: Harvard University Press, 1979. Davis H, Spurr P, Cox A, Lynch MA, Roenne A, Hahn K. A description and evaluation of a community child mental health service. Clin Child Psychol Psychiatry 1997; 2: 221–38.
Nod2 and Crohn’s disease: many connected highways See Research Letters page 1794
1752
In this issue of The Lancet, David van Heel and co-workers analyse the innate immunity driven by the mononuclear cells of patients with Crohn’s disease. Using nanomolar concentrations of muramyldipeptide in an ex-vivo system, they report that pathways act synergistically for the nucleotide oligomerisation domain 2 (Nod2, also known as caspase-recruitment domain 15 or Card15) and toll-like receptor (TLR). Co-activation of these two pathways dramatically increased the production of proinflammatory chemokines. This observation argues for an integrated response of innate immunity. Innate immunity is driven by a few pathogenassociated molecular patterns that are present in most of the non-eucaryote cells. The host molecules involved in recognition of pathogen-associated molecular patterns are known as pattern-recognition receptors. Two main families of pattern-recognition receptors have been discovered in human beings. TLRs are transmembranous molecules able to recognise lipopeptides (TLR1 and 6), lipoteichoic acid (TLR2 and 6), DNA from viruses (TLR3, 7, and 8), lipopolysaccharide (TLR4), flagellin (TLR5), bacterial DNA (TLR9), and still unknown other molecules.1 Nods are seen as the intracellular counterpart of the TLRs. Nod1 (also known as Card4) and Nod2 are activated by some fractions of peptidoglycan, a major component of the bacterial wall. The poorly purified pathogen-associated molecular patterns used in initial experiments might have led to
erroneous conclusions. For example, Nod2 was initially considered as a lipopolysaccharide receptor, but it is now known that it recognises muramyldipeptide, a peptidoglycan fraction. More recently, Travassos et al showed that TLR2 is a receptor for lipoteichoic acid but not for peptidoglycan.2 As a result, Nod1 and Nod2 now appear with the peptidoglycan-recognition proteins as the only known proteins involved in the host response induced by peptidoglycan. Peptidoglycan-recognition proteins are considered as lytic enzymes (secreted or stored in vesicles) while the Nods are seen as intracellular sensors of peptidoglycan even if we do not yet know whether peptidoglycan products interact physically with the Nods. The observation that pattern-recognition receptors act in synergy is not a surprise: living systems are highly integrated. Also, previous studies converged toward this conclusion. The serine/threonine kinase Rip2 (receptorinteracting protein 2, also known as RICK and CARDIAK) is activated by Nod1 and Nod2. However, in Rip2 knock-out mice, the response to lipopolysaccharide, peptidoglycan, and double-stranded RNA is impaired.3,4 These observations might be partly related to contaminated preparations of pathogen-associated molecular patterns as discussed above. However, the demonstration that Rip2 can be recruited to TLR2signalling complexes after peptidoglycan stimulation and that the effect of the reduced cytokine production www.thelancet.com Vol 365 May 21, 2005
Program, participants were asked to fill out the Quality of Well-Being Scale each year.4 The questionnaire revealed that patients with impaired tolerance to glucose, who received an intensive lifestyle intervention, rated their quality of life more highly than those who received a pharmacological intervention (metformin).4 Fourth, there is the issue of time horizon. A long-term perspective should be taken if a health intervention has an impact beyond the period for which data from clinical trials are available. In this case, modelling techniques, such as decision, epidemiological, and Markov models, can be used to project the costs and consequences of the intervention.5 In the Diabetes Prevention Program, the cost-effectiveness of metformin improved drastically when the time horizon was extended from within the trial to lifetime.6 Besides the time horizon, when there are multiple stakeholders, it may be helpful to conduct economic evaluations from different perspectives. The US Panel on Cost Effectiveness in Health and Medicine recommends that, for comparability, the societal perspective be provided as a reference case along with other perspectives.7 Public-health agencies might also find it useful to have evaluations done from the budgetary perspective. Conducting economic evaluations alongside clinical trials is a viable means to evaluate health interventions, especially when such evaluations are supplemented by modelling techniques.8–10 However, if such studies are to contribute to policy debates, it is important to generate
information in a way that allows us to do what economic evaluations do best—to inform decision-making about resource allocation that maximises population health.11 *Kumiko Imai, Ping Zhang Division of Diabetes Translation, Centers for Disease Control and Prevention, Atlanta, GA 30341, USA [email protected] We declare that we have no conflict of interest. 1
Taylor RS, Drummond MF, Salkeld G, Sullivan SD. Inclusion of cost effectiveness in licensing requirements of new drugs: the fourth hurdle. BMJ 2004; 329: 972–75. 2 UCLA Stroke/Vascular Neurology Program. Frequently asked questions. 2002: http://strokeprotect.mednet.ucla.edu/faq.htm (accessed April 13, 2005). 3 Diabetes Prevention Program Research Group. Costs associated with the primary prevention of type 2 diabetes mellitus in the Diabetes Prevention Program. Diabetes Care 2003; 26: 36–47. 4 The Diabetes Prevention Program Research Group. Within-trial cost-effectiveness of lifestyle intervention or metformin for the primary prevention of type 2 diabetes. Diabetes Care 2003; 26: 2518–23. 5 Drummond MF, O’Brien B, Stoddart GL, Torrance GW. Methods for the economic evaluation of health care programmes, 2nd edn. Oxford: Oxford University Press, 1997. 6 Herman WH, Hoerger TJ, Brandle M, et al. The cost-effectiveness of lifestyle modification or metformin in preventing type 2 diabetes in adults with impaired glucose tolerance. Ann Intern Med 2005; 142: 323–32. 7 Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-effectiveness in health and medicine. New York: Oxford University Press, 1996. 8 O’Sullivan AK, Thompson D, Drummond MF. Collection of health economic data alongside clinical trials: is there a future for piggy-back evaluations? Value Health 2005; 8: 67–79. 9 Ramsey SD, McIntosh M, Sullivan SD. Design issues for conducting cost-effectiveness analyses alongside clinical trials. Annu Rev Public Health 2001; 22: 129–41. 10 O’Brien B. Economic evaluation of pharmaceuticals: Frankenstein’s monster or vampire of trials? Medical Care 1996; 34 (suppl): DS99–108. 11 Murray CJL, Evans DB, Acharya A, Baltussen RMPM. Development of WHO guidelines on generalized cost-effectiveness analysis. Health Econ 2000; 9: 235–51.
Child abuse and neglect Published online May 5, 2005 DOI:10.1016/S0140-6736(05) 66389-1 See Articles page 1786
1750
The paper in today’s Lancet by Harriet MacMillan and colleagues focuses attention once again on the important topic of child abuse and neglect, which continues to be a significant health issue in many countries. A recent population-based survey in the UK, for example, showed that the prevalence of all forms of abuse is considerably higher than that indicated by objective measures.1 Furthermore, the negative sequelae have been documented across various developmental domains, including cognition, language, learning, and socioemotional development.2 Whilst there is some evidence emerging about the potential effectiveness of prevention initiatives,3,4 there is considerably less consensus about what works when
abuse has already occurred. Many different approaches have been used to date (eg, family therapy, multisystem therapy, parenting programmes, educational and skilldevelopment programmes, problem solving, stress resolution, and anger management) but MacMillan and colleagues’ study is one of the first rigorous evaluations of the effectiveness of home visits by nurses in stopping child abuse in families already in the child-protection system (ie, preventing abuse recurrence). MacMillan and colleagues’ results show that not only was there no difference in the recurrence of child physical abuse and neglect in the intervention group compared with standard services, but also that, on the basis of data from hospital records, there was a higher www.thelancet.com Vol 365 May 21, 2005
Comment
www.thelancet.com Vol 365 May 21, 2005
Rights were not granted to include this image in electronic media. Please refer to the printed journal
Photolibrary.com
recurrence of abuse in the intervention group. The authors quite rightly consider whether this finding represents a harmful effect of the intervention or the type of surveillance bias that occurs in abuse prevention studies. It seems likely that it was the latter, and such bias has led some to question the appropriateness of measuring abuse as an outcome in studies of this nature.5 It is important that interventions to reduce abusive parenting have a sound theoretical basis, linking the specific content or mechanisms of the intervention with the problem being addressed. The intervention being assessed in MacMillan and colleagues’ study was underpinned by an ecological model6 that aimed to target problems at several levels (eg, from the individual, through the family, to the environment). The use of an ecological model is a valuable theoretical basis from which to proceed, and one that has proved effective in the primary prevention of abuse.4 In many cases, journal restrictions on article length, however, preclude the possibility of knowing exactly what such complex psychosocial interventions actually comprise, making it difficult for future programme developers to build on existing knowledge or to assess why an intervention did not have a bigger impact. There is increasing recognition that the effectiveness of intervention programmes, such as home visiting, might depend on the success with which the programme provider establishes a trusting relationship with the recipient.7 If such relationships are central to the programmes’ effectiveness, as the literature on both counselling and psychotherapy suggests, there is an urgent need for studies to begin measuring the success with which this is achieved. It is not clear whether trusting relationships were a core part of MacMillan and colleagues’ intervention, or indeed whether the intervention was lacking other key components, such as sufficient intensity. Whilst greater intensity would have increased the cost, it seems likely that higher costs would have been justified, given the massive expenditure required for child abuse and neglect at both an individual and societal level. Trials that involve participants as active agents in initiating psychological and behavioural changes are fundamentally different from trials of treatments that involve the participants as passive recipients (as is the
case for many medical interventions). Such trials are delivered by people with differing levels of skills, to families with different histories, preoccupations, circumstances, different levels of readiness for change, and different levels of commitment to the process, all of which determine whether an intervention of this nature proves to be effective. The implication of these differences is that the impact of such interventions is likely to be small, and there is, as such, a need for much larger (and more expensive) studies. Indeed, the survival curve in MacMillan and colleagues’ study shows a 10% reduction in physical abuse, which would seem to be an important improvement. However, the sample size was based on a 25% reduction, which might have been overly optimistic. MacMillan and colleagues’ results make for depressing reading. More than 50% of abused children remaining in the home continued to be abused, and there is little evidence currently available about how to prevent this abuse from happening without removing the child from the home. The benefits of standard services are for the better part not known, but as MacMillan points out, most fall seriously short of the sort of service evaluated in their study. Whilst these findings might highlight the importance of the primary prevention of child abuse and neglect, primary prevention will not be effective in all 1751
Comment
cases, and there is an urgent need to identify an effective secondary preventive intervention to reduce the recurrence of abuse, and to limit the impact that such abuse has on children’s health. The secondary prevention of abuse might require that the balance of investment is now in favour of establishing from existing evidence the core components of potentially successful interventions.
1
2
3
4
5
*Jane Barlow, Sarah Stewart-Brown Warwick Medical School, Medical School Building, University of Warwick, Coventry CV4 7AL, UK [email protected] We declare that we have no conflict of interest.
6
7
Cawson P, Wattam C, Brooker S, Kelly G. Child maltreatment in the United Kingdom: a study of the prevalence of child abuse and neglect. London: National Society for the Prevention of Cruelty to Children, 2001. Cicchetti D. Perspectives from developmental psychopathology. In: Cicchetti D, Carlson V, eds. Child maltreatment: theory and research on the causes and consequences of child abuse and neglect. New York: Cambridge University Press, 1989. Olds DL, Eckenrode J, Henderson CR, et al. Long-term effects of home visitation on maternal life course and child abuse and neglect: fifteen year follow-up of a randomised trial. JAMA 1997; 278: 637–43. Olds DL, Henderson C, Tatelbaum R, Chamberline R. Preventing child abuse and neglect: a randomised trial of nurse home visitation. Pediatrics 1986; 78: 65–78. Roberts I, Kramer MS, Suissa S. Does home-visiting prevent childhood injury? A systematic review of randomised controlled trials. BMJ 1996; 312: 29–33. Bronfenbrenner U. The ecology of human development: experiments by nature and design. Cambridge, Massachusetts: Harvard University Press, 1979. Davis H, Spurr P, Cox A, Lynch MA, Roenne A, Hahn K. A description and evaluation of a community child mental health service. Clin Child Psychol Psychiatry 1997; 2: 221–38.
Nod2 and Crohn’s disease: many connected highways See Research Letters page 1794
1752
In this issue of The Lancet, David van Heel and co-workers analyse the innate immunity driven by the mononuclear cells of patients with Crohn’s disease. Using nanomolar concentrations of muramyldipeptide in an ex-vivo system, they report that pathways act synergistically for the nucleotide oligomerisation domain 2 (Nod2, also known as caspase-recruitment domain 15 or Card15) and toll-like receptor (TLR). Co-activation of these two pathways dramatically increased the production of proinflammatory chemokines. This observation argues for an integrated response of innate immunity. Innate immunity is driven by a few pathogenassociated molecular patterns that are present in most of the non-eucaryote cells. The host molecules involved in recognition of pathogen-associated molecular patterns are known as pattern-recognition receptors. Two main families of pattern-recognition receptors have been discovered in human beings. TLRs are transmembranous molecules able to recognise lipopeptides (TLR1 and 6), lipoteichoic acid (TLR2 and 6), DNA from viruses (TLR3, 7, and 8), lipopolysaccharide (TLR4), flagellin (TLR5), bacterial DNA (TLR9), and still unknown other molecules.1 Nods are seen as the intracellular counterpart of the TLRs. Nod1 (also known as Card4) and Nod2 are activated by some fractions of peptidoglycan, a major component of the bacterial wall. The poorly purified pathogen-associated molecular patterns used in initial experiments might have led to
erroneous conclusions. For example, Nod2 was initially considered as a lipopolysaccharide receptor, but it is now known that it recognises muramyldipeptide, a peptidoglycan fraction. More recently, Travassos et al showed that TLR2 is a receptor for lipoteichoic acid but not for peptidoglycan.2 As a result, Nod1 and Nod2 now appear with the peptidoglycan-recognition proteins as the only known proteins involved in the host response induced by peptidoglycan. Peptidoglycan-recognition proteins are considered as lytic enzymes (secreted or stored in vesicles) while the Nods are seen as intracellular sensors of peptidoglycan even if we do not yet know whether peptidoglycan products interact physically with the Nods. The observation that pattern-recognition receptors act in synergy is not a surprise: living systems are highly integrated. Also, previous studies converged toward this conclusion. The serine/threonine kinase Rip2 (receptorinteracting protein 2, also known as RICK and CARDIAK) is activated by Nod1 and Nod2. However, in Rip2 knock-out mice, the response to lipopolysaccharide, peptidoglycan, and double-stranded RNA is impaired.3,4 These observations might be partly related to contaminated preparations of pathogen-associated molecular patterns as discussed above. However, the demonstration that Rip2 can be recruited to TLR2signalling complexes after peptidoglycan stimulation and that the effect of the reduced cytokine production www.thelancet.com Vol 365 May 21, 2005