Recent Research into Risk Assessment of Children

Ludwig.F. Lowenstein Ph.D

Southern England Psychological Services



Expert witnesses and others have always been concerned with how much risk can be taken, if any, in the concern for the safety of children. Safety of children should ensure that they are in no danger of physical, sexual or emotional abuse or neglect. The prediction of danger to children, though vital, can be problematic.

It is vital for those who assess risk to use any type of assessment, measurement, interview etc. likely to lead to the most accurate prediction. There is a tendency for many who assess dangerousness to over-estimate it, as it is better “to err on the side of caution” then the reverse. This despite the fact that few individuals are likely to be a danger compared to those who are not likely to act as such (Menzies et al, 1994).

There has been an increase in the use of the actuarial method for predicting that risk is minimal. Hence false positives may be prevented, i.e. people who are predicted to do something wrong but do not do so. False negatives are people who are not expected to do anything dangerous but do so; they must also be identified (Gardner et al, 1996)

Ackerman, (1999) clearly indicates the factors likely to increase the risk of dangerousness or violence. These include:

  1. History of violence (Steadman & Robbins, 1998; Wrightsman et al, 1998)
  2. The use of substances such as alcohol and drugs (Steadman et al, 1998; Steadman & Robbins, 1998)
  3. Psychotic illness including hallucinations and delusions such as fear of being threatened (Melton et al, 1997).
  4. Being immature or under the age of 30 (Tardiff et al, 1997; Melton et al, 1997).
  5. Anti-social personality disorder i.e. psychopathy (Tardiff et al, 1997, a, b; Harris&Rice, 1997).
  6. Failure to take psychotropic medication (Monohan & Steadman, 1994).

The most recent research into risk assessment of children has been divided into four parts:

  1. Risk factors associated with children’s welfare.
  2. Problems associated with risk assessments.
  3. Assessment methods for ascertaining risk to children.
  4. Therapeutic approaches to reduce risk towards children

A. Risk Factors Associated with Children’s Welfare

Risk factors for physical child abuse were studied by Christmas et al (1996). The review of the current literature in relation to this considers foremost a history of abuse in the past, the depression of one or both parents, single parenting, the socio-economic status of the family, social isolation, the maternal age with younger women being more vulnerable and partaking in substance abuse.

Mentally ill mothers who have killed in the past were studied by Jacobsen & Miller(1998), One proposal is to “fast-track” cases involving parents with long-standing mental disorders by automatically terminating parental rights. This approach assumes that a severe and chronic mental disorder is incompatible with safe parenting. Three cases were studied where children were killed. The conclusion was evaluation of parenting competency of mentally ill parents is important. An American study carried out in the state of Massachusetts by Whitney & Davis (1999) considered the importance of an internal domestic violence programme better to identify and serve families where partners abuse and child abuse overlap.

Child custody disputes there are particular dangers to children as well as to adults due to the acrimony which exists between the former partners. Austin (2000,a) studied the relocation ui child custody and its impact on children. Courts usually allow the child to move away with the residential parent, unless there is potential harm to the child. The forensic violence risk assessment literature provides an analogous conceptual framework for understanding the prediction of harm. Harm is likely to be exacerbated when comments are made by the resident parent towards a child to develop a condition commonly termed parental alienation syndrome (Lowenstein 1998a; 1998b; 1999 a,b,c; 2001; Gardner 1992; 1998; 2001).

Finally Gambrill & Shlonsky (2001) considered that risk assessment studies in child welfare have largely focused on identifying individual or family risk factors. This is often associated with future harm. These risks include, lack of proper assessment of service needs, inadequate linkage of available services to desired outcomes, and an agency culture that is reactive rather man pro-active in pursuit of risk reduction.

B. Major Problems Associated with Risk Assessment

There are major problems in carrying out risk assessments. To carry out no risk assessment is obviously wrong, but to consider that risk assessments are always reliable is equally fallacious. A review of the literature up to 1996 by Lyons et al (1996) of risk assessments included the examining of psychometric properties including reliability and validity and outcomes of the implementation as a response to the crisis of growing intakes of child protective services agencies. Risk assessment models and their evaluation were search based on two criteria: 1) the evaluation published or presented at a conference; 2) the evaluation conducted by an independent evaluator. The result highlighted the need for development and research into risk assessment procedures.

Kelly & Milner (1996) suggested that case conference decisions e.g. by social worker, were inherently more risky than those taken by professionals with “individual responsibility” because of the way in which cases are framed in terms of losses. A different view was expressed by Milner et al (1998). They stressed that despite the number of assessment techniques currently available, researchers and practitioners had few methods of clearly identifying risk and mitigating factors and that direct assessment techniques currently offer the best assessment strategies. Unfortunately, a child care worker was the poorest predictor of further risk.

Bell (1999) studied 22 Local Authorities in Great Britain and their social workers undertaking child protection investigation. He noted that while social workers were committed to being, and believed their practice to be, participative, the duel tasks of making risk assessments for conferences and working in partnership with problem families created conflicts of interest and rights. The impact of this on social workers engagement with family members, on their assessment, and on the decision-making process was explored. It was concluded that the difficulties identified, were endemic and pointed the need for a more broadly-based child care service to appropriately meet the welfare needs of the family. Also the contradictory nature of the conference task needed to be addressed by clarifying the legal base of the intervention and developing other models of decision making.

Drury-Hudson (1999) examined the process of decision making in child protection, particularly as it related to the decision whether or not to remove a child from home. A group of novice social workers was compared with expert practitioners, with particular focus on the types of knowledge that novices draw upon when making such decisions. A three stage qualitative methodology was employed. All parts of the study utilised a case vignette of a neglect scenario. Findings were reported in respect of the use of theoretical, empirical and procedural knowledge and suggested novices tended to kek a clear understanding of the factors associated with child maltreatment. While they have a superficial awareness of the concept of risk assessment, they failed to weigh factors appropriately and apply them to their practice. This supports the findings of the previous study.

An Australian study by Goddard et al (1999) considered the practice of risk assessment in children via the protection services. It highlighted the complexity of the concept of risk as the basis for a future oriented assessment activity. The authors suggested that this change of time frame (from what has happened to what might happen) was likely to be detrimental to children. Through a critical review of the literature, the authors questioned whether risk predictions were possible and discussed limitations of risk assessment instruments which omitted some risk factors and may have totally ignored the perspective of the child. The authors therefore challenged the validity of risk assessment instruments in statutory settings and suggested that the protection of the organisation was likely to be the major objective in their implementation. This might be considered a kind of self-protection rather than risk assessment of children!

Houston & Griffiths (2000) argued that objectivist paradigms failed to provide valid and reliable measures of risk. Risk should be explored from an alternative subjectivist paradigm. They therefore advocated the reinstatement of the individual to the family and the relationship as the guiding rationale for social work intervention. This view is likely to be highly criticised by many social services, especially when there is a threat of child sex or physical abuse.

Risk reduction interventions in the case of child custody relocation cases were considered by Austin (2000,b). When a custodial parent chooses to relocate to a new community, the child of divorce faces a life transition that was potentially even more traumatic than the parental break-up. The courts generally allow the custodial parent to move away with the child. Divorce affects research into child risk assessments and a recent model of risk assessment for relocation suggested factors that predicted potential harm or protection for the child.. Family mediators and psychotherapists had important roles to play in reducing the risk. (This will be discussed in the last section dealing with the therapeutic approaches to reduce risks for children.) It must be added that there are risks caused by the involvement of social services as noted in one study by DePanfilis & Zuravin (2001) which found that families with a previous substantiated report of child abuse were 22% less likely to be open for services than. families without prior substantiated reports. In cases substantiated for neglect they were 20% less likely to be open for services than physical abuse cases.

C. Assessment Methods for Ascertaining Risk to Children

Notwithstanding the problems associated with risk assessment risk must still be assessed albeit using imperfect tools. Jagannathan & Camasso used the Washington Risk Assessment Matrix and found three distinct risk profiles which made risks more likely. These were:

  1. Children with behaviour problems were more likely to be at risk; as were
  2. children from disadvantaged households; and,
  3. children with an unemployed parent.

Psychological tests were particularly useful in the assessment of risk of children, who had been, and might yet again be, sexually abused. Babiker & Herbert (1998) found these tools to be less intrusive and therefore less damaging to children and their families. This is in contrast to clinical interviews most often used by social services. It was however suggested that risk assessment should ideally use several procedures in a multi-dimensional approach to assessing child sexual abuse. This method was also favoured by Lowenstein (1998).

Baird et al (1999) also found there were no 100% reliable methods of risk assessment. It was only possible to reduce the risk in the end by using a number of methods, and monitoring the situation carefully and intensively (Lowenstein, 1998).

Assessing the credibility in allegations of marital violence in the high conflict child custody type situation was studied by Austin (2000,c). The author states that forensic psychology had not systematically examined the problem of evaluating the credibility of allegations of marital violence in the context of a child custody case. A risk assessment approach to marital violence in the custody evaluation context must be viewed as a serious matter involving the children. The actuarial approach was considered to classify more accurately cases of risk (Baird & Wagner, 2000). Again the Washington State Risk Assessment Matrix was favoured by Camasso & Jagannathan (2000). Fuller et al (2001) gave preference to the Illinois Child Endangerment Risk Assessment. Here results indicated that the age of the youngest child, single parent households and the number of child problems as well as type of maltreatment could be more effectively diagnosed as to risk.

D. Therapeutic and Other Approaches to Reducing the Risk Toward Children

Christmas et al (1996) considered the most effective therapeutic method to be cognitive behaviour therapy especially when the parents suffer from depression. When dealing with parents who cause risks to their children also important were sex education, and instruction skills in anger management. On the whole a multi-faceted approach is to be preferred.

Fisher & Beech (1998) emphasised the use of a comprehensive assessment of sex offenders using psychological methods and risk factors for offending, families after sexual abuse has taken place. The emphasis needs to be on reducing social inadequacies and dealing with pro-offending attitudes in the case of sexual abuse and the denial which frequently accompanies this and relapse prevention.

A question to be answered as far as Turnell & Edwards (1999) were concerned was: “How can child protection professionals build partnerships with parents where there is suspected or substantiated child abuse or neglect. The authors felt there was a need for practical hands-on strategies for building a partnership with parents, which may, in the long run, prevent abuse and family dissolution. The emphasis was on family reunification practices supposed to be carried out following the identification of specific risks of re-entry of family members (Terling,1999).

At present protection service agencies tend to be bureaucratic, technocratic, regulatory mechanisms for detecting and managing abuse and neglect. Krane & Davies (2000) suggested a “mothering narrative”, and thereby prevent risks to children emerging. Initially however it is important to filter out “high risk” cases from the rest. A view, currently growing, is that there is limited empirical support for the wholesale adoption of managed care of children and their families (Embry et al, 2000). The focus it is felt, should be on the reduction of maltreatment risks between case opening and closing. A large study by Lyle & Graham (2000) consisting of 245 families with a total of 592 children resulted in significant decreases in risk scores from intake to case closing. Dealing with the whole family in the form of family therapy was a necessary approach according Hohmann-Marriot (2001) rather than just an individual or individuals without the whole family.

Summary& Abstract

Much must be done to improve the accuracy of risk assessment of children. Agencies and social services tend to err on the side of extreme caution rather than truly weigh the risks. This is undoubtedly due to the anxieties associated with the consequences, should tragic mistakes occur. Hence children are often unnecessarily prevented from being with their families. This could be considered an abuse in itseh0.

The best risk assessment techniques require a number of approaches including psychological testing, especially personality assessments, intensive monitoring of children at risk and interviewing members of families. More accurate specific methods including psychological tests are needed to assess and weigh risks more accurately.


  1. Ackerman, MJ. Book: Essentials of Forensic Psychological Assessment. 1999. John Wiley & Sons Inc. New York.
  2. Austin, W G. A forensic psychology model of risk assessment for child custody relocation law. Family and Conciliation Courts Review. 2000 (a), Apr; Vol 38(2): 192-207.
  3. Austin, W G. Risk reduction interventions in the child custody relocation case. Journal of Divorce and Remarriage. 2000 (b); Vol 33(1-2): 65-73.
  4. Austin, W G. Assessing credibility in allegations of marital violence in the high-conflict child custody case. Family and Conciliation Courts Review. 2000 Oct; Vol 38(4): 462-477.
  5. Babiker, G; Herbert, M. Critical issues in the assessment of child sexual abuse. Clinical Child and Family Psychology Review. 1998 Dec; Vol 1(4): 231-252.
  6. Baird, C; Wagner, D; Healy, T; Johnson, K. Risk assessment in child protective services: Consensus and actuarial model reliability. Child Welfare. 1999 Nov-Dec; Vol 78(6): 723-748.
  7. Baird, C; Wagner, D. The relative validity of actuarial-and consensus-based risk assessment systems. Children and Youth Services Review. 2000 Nov-Dec; Vol 22(11-12): 839-871.
  8. Bell, M. Working in partnership in child protection: The conflicts. British Journal of Social Work. 1999 Jun; Vol 29(3): 437-455.
  9. Camasso, M J; Jagannathan, R. Modeling the reliability and predictive validity of risk assessment in child protective services. Children and Youth Services Review. 2000 Nov-Dec; Vol 22(11-12): 873-896.
  10. Christmas, A L; Wodarski, J S; Smokowski, P R. Risk factors for physical child abuse: A practice theoretical paradigm. Family Therapy. 1996; Vol 23(3) 233-248.
  11. DePanfilis, D; Zuravin, S J. Assessing risk to determine the need for services. Children and Youth Services Review. 2001 Jan; Vol 23(1): 3-20.
  12. Drury-Hudson, J. Decision making in child protection: The use of theoretical, empirical and procedural knowledge by novices and experts and implications for fieldwork placement. British Journal ofSocial Work. 1999 Feb; Vol 29(1): 147-169.
  13. Embry, R A; Buddenhagen, P; Bolles, S. Managed care and child welfare: Challenges to implementation. Children and Youth Services Review. 2000 Feb; Vol 22(2): 93-116.
  14. Fisher, D; Beech, A. Reconstituting families after sex abuse: The offender’s perspective. Child Abuse review. 1998 Nov-Dec; Vol 7(6) (Spec Issue): 420-434.
  15. Fuüer, T L; Wells, S J; Cotton, E E. Predictors of maltreatment recurrence at two milestones in the life of a case. Children and Youth Services Review. 2001 Jan; Vol 23(1): 49-78.
  16. Gambrill, E; Shlonsky, A. The need for comprehensive risk management systems in child welfare. Children and Youth Services Review. 2001 Jan; Vol 23(1): 79-107.
  17. Gardner, W; Lidz, C W; Mulvey, E P; Shaw E C. Clinical versus actuarial predictions of violence in patients with mental illness. Journal of Consulting and Clinical Psychohgy. 1996. Vol 64: 602-609.
  18. Gardner, R A. Book: True and False Accusations of Child Sex Abuse. (1992). Creative Therapeutic s, New Jersey, USA.
  19. Gardner, R A. Book: The Parent Alienation Syndrome – A Guide for Mental Health and Legal Professionals (second edition, 1998) Creative Therapeutics, Inc., New Jersey, USA.
  20. Gardner, R A. Book: Therapeutic Interventions for Children with Parental Alienation Syndrome. (2001). Creative Therapeutics, Inc., New Jersey, USA.
  21. Goddard, C R; Saunders, B J; Stanley, J R; Tucci, J. Structured risk assessment procedures: Instruments of abuse? Child Abuse Review. 1999 Jul-Aug; Vol 8(4): 252-263.
  22. Harris, G T; Rice, M E. Risk appraisal and management of violent behaviour. Psychiatrie Services. 1997. Vol 48:1168-1176.
  23. Hohman-Marriot, B E. Marriage and family therapy research: Ethical issues and guidelines. American Journal of Family Therapy. 2001 Jan-Feb; Vol 29(1): 1-11.
  24. Houston, S; Griffiths, H. Reflections on risk in child protection: Is it time for a shift In paradigms? Child and Family Social Work. 2000 Feb; Vol 5(1): 1-10.
  25. Jabobson, T; Miller, L J. Mentally ill mothers who have killed: Three cases addressing the issue of future parenting capability. Psychiatric Services. 1998 May; Vol 49(5): 650-657.
  26. Jagannathan, R; Camasso, M J. Risk assessment in child protective services: A canonical analysis of the case management function Child Abuse and Neglect. 1996 M; Vol 20(7): 599-612.
  27. Kelly, N; Milner, J. Child protection decision-making. Child Abuse Review. 1996 May; Vol 5(2): 91-102.
  28. Krane, J; Davies, L. Mothering and child protection practice: Rethinking risk assessment. Child and Family Social Work. 2000 Feb; Vol 5(1): 35-45.
  29. Lyle, C G; Graham, E. Looks can be deceiving: Using a risk assessment instrument to evaluate the outcomes of child protection services. Children and Youth Services Review. 2000 Nov-Dec; Vol; 22(11-12): 935-949.
  30. Lyons, P; Doueck, H J; Wodarski, J S. Risk assessment for child protective services: A review of the empirical literature on instrument performance. Social Work Research. 1996 Sep; Vol 20(3): 143-155.
  31. Lowenstein, L F. Parent alienation syndrome: A two step approach toward a solution. Contemporary Family Therapy. 1998,a Dec; Vol 20(4): 505-520.
  32. Lowenstein L F. Book: Paedophilia (1998,b) Able Publishers, 13 Station Road, Nebworth, Herfordshire, SG3 6AP, UK.
  33. Lowenstein, L F. Parent alienation syndrome (PAS). Justice of the Peace. 1999,a Jan; Vol 163(3): 47-50.
  34. Lowenstein, L F. Parent alienation syndrome: What the legal profession should know. Medico-LegalJournal. 1999, b; Vol 66(4): 151-161.
  35. Lowenstein, L F. Parent alienation and the judiciary. Medico-Legal Journal. 1999,c; Vol 67 part 3: 121-123.
  36. Lowenstein, L F. Tackling parental alienation. Justice of the Peace. 2001 Feb; Vol 165(6): 102
  37. Melton, G B; Petrila, J; Poythress, N G; Slobogin, C. Book: Psychological Evaluations for the Courts.(2nd edition). 1997. New York: Guilford Press.
  38. Menzies, R; Webster, C; McMain, S; Staley, S; Scaglione, R. The dimensions of dangerousness revisited. Law and Human Behaviour. 1994. 18:1-28
  39. Milner, J S; Murphy, W D; Valle, L A; Tolliver, R M Assessment issues in child abuse evaluations. Chapter in a book: Lutzker, J R (Ed). (1998). Handbook of child abuse research and treatment. Issues in clinical child psychology. (pp 75-115;. New York, NY, US: Plenum Press xvi, 595p.
  40. Monohan, J; Steadman, H. Book: Violence and mental disorder: Developments in risk assessment. 1994. Chicago: University of Chicago Press.
  41. Steadman, H; Mulvey, E; Monahan, J; Robbins, P Appelbaum, P; Grisso, R; Roth, L; Silver, E. Violence by people discharged from acute psychiatric inpatient facilities and by others in the same neighbourhoods. Archives of General Psychiatry. (1998); No 55: 393-401.
  42. Steadman, H; Robbins, P. The MacArthur violence risk assessment study. (1998 Oct). Paper presented at the Competency, Coercion and Risk of Violence conference, Marquette University Law School, Milwaukee, WI.
  43. Tardiff, K; Marzuk, P M; Leon, A C; Portera, L. A prospective study of violence by psychiatrie patients after hospital discharge. Psychiatric Services. 1997, a. No 48: 678-681.
  44. Tardiff, K; Marzuk, P M; Leon, A C; Portera, L; Weiner, C. Violence by patients admitted to a private psychiatric hospital. American Journal of Psychiatry. 1997, b. No 154: 88-93.
  45. Terling, T. The efficacy of family reunification practices: Re-entry rates and correlates or re-entry for abused and neglected children reunited with their families. Child Abuse and Neglect. 1999 Dec; Vol 23(12): 1359-1370.
  46. Turnell, A; Edwards, S. Book: Signs of Safety: A solution and safety oriented approach to child protection casework. New York, NY, US: W W Norton and Co, Inc (1999). xi, 216pp.
  47. Whitney, P; Davis, L. Child abuse and domestic violence in Massachusetts: Can practice be integrated in public child welfare setting? Journal of American Professional Society on the Abuse of Children. 1999 May; Vol4(2): 158-166.
  48. Wrightsman, L S; Nietzel, M T; Fortune, W H. Book: Psychology and the legal system. 1998. Pacific Grove, CA: Brooks/Cole.