Assessing suicide risk in children: guidelines for developmentally appropriate interviewing.

Although suicide is considered a leading cause of death for all age groups, resources and recommendations regarding methods of assessment of suicide risk in children appears to be scattered across related disciplines. Most risk assessment measures for "youth" are intended for use with adolescents, and the nature of children's developmental functioning presents particular challenges for accurate assessment. This article includes a brief review of risk factors and recommendations for preparing to conduct suicide risk assessments with children. Guidelines for mental health counselors who conduct developmentally appropriate risk assessments with children are detailed, and suggestions for consulting with caregivers are provided.

ASSESSING SUICIDE RISK IN CHILDREN: GUIDELINES FOR DEVELOPMENTALLY APPROPRIATE INTERVIEWING

Suicide is a leading cause of death for all age groups in the United States (National Institute of Mental Health [NIMH], 2003), accounts for approximately 5.8% of deaths in 10-14 year-olds (Gould, Shaffer, & Greenberg, 2003), and remains between the 3rd and 7th leading cause of death among 5-14 year olds (American Association of Suicidology [AAS], 2006; Wise & Spengler, 1997). Although the prevalence of completed suicide among children is low in comparison to other age groups (AAS, 2003), suicidal thoughts and behaviors in children are reportedly quite common (Brent & Kolko, 1990; Klimes-Dougan, 1998). Indeed, the AAS (2006) reported that youth make approximately 100-200 suicide attempts for each completed suicide.

Although clinicians are likely to address suicide at some point in their careers, many mental health counselors remain uncomfortable regarding the prospect of doing so (Shea, 1999); feelings are frequently amplified when attempting to address suicidal issues with children. According to Wise and Spengler, "low occurrence of childhood suicide, widespread myths and misconceptions, and a general lack of awareness about this phenomenon make assessment of childhood suicide one of the most difficult of diagnostic tasks" (1997, p. 318). These authors have argued that the prevalence of suicidal behavior in children is grossly underestimated due to statistical classification errors and adults' unwillingness to believe that children can and do plan and implement suicides. Similarly, Stefanowski-Harding (1990) identified lack of preparation for dealing with child suicide, discomfort, and denial as key elements to clinicians' difficulties in addressing child suicide. Nonetheless, mental health counselors must be prepared to assess for child suicide and intervene appropriately (Hendren, 1990; Juhnke, 1996; Stefanowski-Harding; Wise & Spengler).

Although mental health counselors are often on the front lines for assessing, referring, and treating children, information regarding suicide risk in children and recommendations for assessing suicide risk in this population tend to be scattered across disciplines. Evidence indicates some commonality of risk factors and course of suicidal behavior among children, adolescents, and adults (Gould et al., 2003; Pfeffer, 2003), but the unique nature of children's developmental functioning presents unique risk factors and particular challenges for accurate assessment (Hendren, 1990; Pfeffer, 2003; Stillion & McDowell, 1996; Wise & Spengler, 1997). The goal of this article is to review risk factors for childhood suicide and suicidal behavior, provide a commentary on current methods of assessing suicide risk in children, and discuss guidelines for conducting developmentally appropriate risk assessments with children.

RISK FACTORS FOR SUICIDE

A variety of theoretical and conceptual models related to suicidality are available to guide one's interpretation and understanding of suicide risk factors (see Berman, Jobes, & Silverman, 2006; Stillion & McDowell 1996; Westefeld et al., 2000). Although some models are grounded in specific biological, psychological, or sociological theories, Berman et al. reported "an increasing emphasis on theoretical integration, particularly among psychological theorizing" (p. 66). In this spirit, Stillion and McDowell provided a multidimensional Suicide Trajectory Model that aids in exploration of common risk factors and triggering events that may lead to suicidal ideation and behavior within and across developmental stages.

Stillion and McDowell (1996) proposed that individuals who experience certain biological, psychological, cognitive, and environmental risk factors will be more apt to respond to stressors with suicidal ideation or behavior. Although not all individuals who experience these risk factors become suicidal, the majority of those who are suicidal experience a number of such risk factors. Examination of risk factors, then, assists the clinician in calling attention to children who may need more comprehensive assessment (Wise & Spengler, 1997) and evaluating risk levels of individuals who present with suicidal ideation (Weller, Young, Rohrbaugh, & Weller, 2001). Specific risk factors for children ages 5-14 are reviewed in the following paragraphs.

Biological Risk Factors

Biologically, children are at risk for suicide due to higher degrees of impulsivity, and this greater degree of impulsivity is reflected in suicide attempts and gestures that are closely related to opportunity and require little planning (e.g., running in front of a car or jumping out a window) (Stillion & McDowell, 1996). Juhnke (1996) also identified biological risk factors including sex and age. Although very young boys and girls have similar rates of suicide attempts and completions, sex differences in rates of completed suicide emerge by age 10 (Wise & Spengler, 1997). The gap between the number of males and females who complete suicide continues to increase as age increases (Gould et al., 2003); by adolescence, boys complete and girls attempt suicide more often (AAS, 2006). In addition, the rates of suicide are very low among the very young and continue to increase as children mature (Juhnke); thus, age may be considered a risk factor even among 5 to 14 year-olds.

Psychological Risk Factors

Stillion and McDowell (1996) identified psychological risk factors among young children to include feelings of inferiority and the "expendable child syndrome" (p. 60). Children at risk for suicide frequently experience disturbance in psychological functioning and in one or more diagnosable mental health disorders (Brent & Kolko, 1990; Gould et al., 2003; Weller et al., 2001; Wise & Spengler, 1997). Common symptoms most often include depression (Juhnke, 1996; Wise & Spengler) but may also include anxiety, aggression, and impulsivity (Weller et al., 2001). Similarly, these children are more likely to internalize problems and stressors (Weller et al.) and present with poor coping skills (Wise & Spengler). As such, many children may present as hopeless about the future (Brent & Kolko; Juhnke; Stefanowski-Harding, 1990).

Cognitive Risk Factors

Cognitive risk factors for childhood suicide include immature views of death, concrete thinking styles, and a child's degree of attraction and repulsion to life and death (Stillion & McDowell, 1996). Certainly, a child who does not understand the nature of death and engages in rigid thinking will be less capable of understanding the implications of her or his behaviors. Juhnke (1996) further identified rational thinking loss as a risk factor for childhood suicide. Similarly, Pfeffer (2003) discussed the need for clinicians to be aware of potential auditory or visual hallucinations. A child who reports hearing voices commanding him or her to engage in specific behaviors, especially self-harm, should be considered as experiencing serious suicidal ideation because he or she is likely to experience decreased loss of control and ability to differentiate her or his own thoughts from the hallucinations. Similarly, rational thought loss and increased impulsivity resulting from recent drug and/or alcohol use are frequently associated with completed suicides in adolescents and adults (Juhnke) and may also be considered risk factors for children.

Environmental Risk Factors

Environmental risk factors include a constellation of early loss, parental conflict, chaotic or inflexible family structures, abuse, neglect, and parental suicidal behaviors (Stillion & McDowell, 1996). Mental health counselors should consider family factors key to identifying those at increased risk. The quality of family environment and the ability of caretakers to provide safe and nurturing relationships and environments can not be underestimated (Brent & Kolko, 1990; Gould et al., 2003; Stefanowski-Harding, 1990). Families experiencing increased distress, dysfunction, or violence frequently have difficulty providing for the needs of a distressed child (Wise & Spengler, 1997). Risk for the child increases dramatically when one or both parents are experiencing mental health problems and/or suicidal behavior (Brent & Kolko; Gould et al.; Stefanowski-Harding; Stillion & McDowell; Weller et al., 2001).

Similarly, children at risk for suicide are frequently isolated and alienated from their peers (Stefanowski-Harding, 1990), have poor social support (Weller et al., 2001), and display poor social skills or ability to function in social situations (Brent & Kolko, 1990). These children often feel lonely, disconnected, and do not enjoy the protective benefits of acceptance and peer group identification. In addition, suicidal children frequently present with problems at school including learning disabilities and/or academic failure (Gould et al., 2003; Stillion & McDowell, 1996; Wise & Spengler, 1997).

Precipitating Events

The presence of recent stressors and precipitating events must not be overlooked (Gould et al., 2003; Roberts, 2000; Stillion & McDowell, 1996; Weller et al., 2001; Wise & Spengler, 1997). Psychosocial stressors or changes may be particularly hard on a child and may include loss or threatened loss (Stefanowski-Harding, 1990) such as moving, crises in the family, health problems, and other events toward which the child feels little or no control. Similarly, environmental factors including recent exposure to suicide or "contagion" (Brent & Kolko, 1990; Gould et al.) may place a child at even higher risk for considering suicide as a solution to her or his stressors. Stillion and McDowell advised that triggering events for high-risk children often appear trivial to adults and may include events as simple as a denied privilege or seemingly ordinary punishment. Further, reviews of empirical literature indicate that prior suicidal ideation and attempts are the most predictive risk factors for future suicidal behavior and completion (Gould et al.; Juhnke, 1996; Weller et al.; Wise & Spengler). Thus, mental health counselors should consider children who have previously coped with stressors by thinking about suicide or attempting to take their lives particularly vulnerable to continued coping in such a manner. Finally, those with an organized plan for suicide should be considered as presenting with a much elevated risk.

PREPARING TO ASSESS CHILDREN FOR SUICIDE

Before engaging in crisis assessment or intervention, there is a particular need for the mental health counselor to prepare herself or himself by addressing her or his own thoughts, feelings, and fears regarding conducting this particular suicide assessment (Shea, 1999). Preparation may assist the counselor in recognizing the difficulties inherent in crisis work with children and minimize the likelihood of denying the seriousness of the situation (Stefanowski-Harding, 1990), acting on misinformation (Wise & Spengler, 1997), or nervously and indirectly rushing through an interview (Shea). In addition, the counselor may practice adapting questions typically included in a suicide assessment to cover a range of developmental levels.

Prior to the interview, the mental health counselor should consider any available information (e.g., age, developmental abilities, and the presence of particular risk factors) in an attempt to more closely approximate the child's likely linguistic and communicative abilities while anticipating possible needs. While far from perfect, this advance consideration may assist the counselor in shifting her or his mind to the developmental needs of the particular child; thus, the clinician may begin to consider how he or she may adapt language and communication to be appropriate for the child (Hendren, 1990).

Because a counselor must be comprehensive while working quickly, Hendren (1990) and Pfeffer (2003) recommended prioritizing needs to include identification of risk factors, assessment of coping abilities, and assessment of lethality. The mental health counselor may wish to construct a device or assessment tool to trigger her or his memory regarding the need to query specific risk factors or aspects of lethality. For example, the use of the adapted SAD-PERSONS (Juhnke, 1996) or another written reminder may help the counselor stay on track despite feeling overwhelmed by the task at hand.

GUIDELINES FOR DEVELOPMENTALLY APPROPRIATE INTERVIEWING

Suicide risk assessment frequently occurs within the context of crisis assessment and intervention; developmentally appropriate mental health counselors must simultaneously engage and involve the client while working toward an action plan in the child's best interest. Thus, the purpose of the interview is likely to alternate between problem-solving, environmental structuring, and assessment of the need for further services. Most often, the role of the clinician is to assess need, identify precipitating events, and mobilize resources in an expedient manner (Burgess & Roberts, 2000).

Roberts (2000) presented a 7-stage crisis intervention model that has been applied to suicide lethality assessment (Roberts & Yeager, 2005), counseling with suicidal adolescents (Jobes, Berman, & Martin, 2000), and children and adolescents experiencing psychiatric emergencies (Singer, 2005). The seven steps to crisis intervention overlap to some degree and include the following: "(1) plan and conduct a crisis assessment, (2) establish rapport and rapidly establish relationship, (3) identify major problems, (4) deal with feelings and emotions, (5) generate and explore alternatives, (6) develop and formulate an action plan, (7) establish follow-up plan" (Roberts, p. 1). This model may easily be adapted to assessing children for suicide risk and will be used to frame the following recommendations.

Plan and Conduct a Crisis Assessment

Initial moments. The counselor should begin the interview by first greeting the child and then any parents or caregivers who are available. It may be helpful to spend several moments attempting to engage the child in some sort of small talk, conversation, or activity that conveys interest and caring to the child. It is important to communicate to the child the specific reason for the interview and to structure the interview in such a way that is comforting rather than punitive (Pfeffer, 2003; Wilson & Powell, 2001). For example, the counselor may introduce the topic simply by saying, "Tell me why you are here today," or "Tell me why your teacher thought you should come here." These prompts will assist in ascertaining the child's mental status and orientation to the situation and provide an opportunity for the clinician to address any fears and misconceptions the child may have regarding the visit. Should the child reply with "I don't know," the clinician may wish to provide a short reason such as, "your mom is worried that you might be feeling very sad, and I am here to help."

When opening an interview, it is important to communicate to the child the specific reason for the interview, that he or she is not in trouble, that it is safe to tell the truth, and that there are no right or wrong answers (Aldridge & Wood, 1998; Bourg et al., 1999). Depending on the age of the child, the counselor may continue with further clarification such as, "It's okay to feel sad, and we want to make sure that you are safe. I would like to talk a little bit about what happened today. You aren't in any trouble, and I'm here to help. I won't be mad or sad about anything you say, so it's okay to tell the truth in here. If you don't want to talk about something, it's okay to tell me. Just tell me if you don't want to talk about it rather than telling a lie." The clinician should take breaks during this section to ascertain whether the child is able to understand these instructions; the child should also be provided an opportunity to ask any questions he or she may have regarding the interview.

The American Mental Health Counselors Association [AMHCA] (2000) Code of Ethics and American Counseling Association [ACA) (2005) Code of Ethics provide explicit attention to the complex nature of informed consent, a client's right to confidentiality, and a counselors' duty to protect the client from harm. These issues are particularly important when working with minors and their guardians, and AMHCA and ACA explicitly state that counselors must safeguard minor client's rights to confidentiality while involving guardians in the counseling process. Mental health counselors should discuss related expectations and limitations at the onset of counseling (Lawrence & Robinson Kurpius, 2001). The clinician, then, must find a balance between helping the client feel safe without violating a sense of trust. Careful consideration and advanced practice of wording, policies, and explanations around confidentiality will help mental health counselors to navigate this portion of the interview.

At this time, decisions must also be made regarding who is to be present for which portions of the interview. Although the presence of a caregiver may serve as a support for the child, caregiver presence also increases the possibility of the adult attempting to speak for or correct the child and the adult's emotional reaction to the content. The involvement of adults may result in the child withholding or changing information out of fear of repercussion or worry about upsetting the caregiver (Aldridge & Wood, 1998). Similarly, asking the child who he or she wishes to be present often puts the child in a difficult position. Thus, it is often helpful for the counselor to suggest a plan for the structure of the interview at the onset and adapt accordingly. For example, the counselor might say, "It's very important that I get to hear from to all of you today so that we might work together to come up with where to go from here. I would like to talk to you together for a few minutes, and then I would like to talk to--alone. After that, we'll come back together as a group." Should the child protest the departure of a caregiver, the counselor may adapt accordingly.

Assessment methods. The primary purpose of suicide assessment is to collect information regarding risk factors and suicidal ideation so that the clinician may make an educated decision in the client's best interest (Shea, 1999). Experts in the field have called for multi-method, multi-source, and multi-tiered approaches to assessment of suicidal risk and behavior in children (e.g., Goldston, 2003; Wise & Spengler, 1997). Although ideal, the need for this type of assessment often is tempered by the mental health counselor's need to obtain accurate assessment information in a short amount of time under a high degree of pressure. Thus, currently recommended assessment methods vary in the degree to which they are applicable for a given situation. These methods will be briefly reviewed in the following paragraphs.

Unstructured clinical interviews appear to be the most common form of assessing children for suicide risk. Wise and Spengler (1997) recommended screening all clients of all ages with a general question such as, "do things ever get so bad you think about hurting yourself" (p. 328). Responses to this question will indicate whether additional follow-up assessment for suicidal ideation or self-injurious behavior is needed. Similarly, the authors recommended that clients be queried regarding their experiences with death and psychological correlates of suicidal behavior including affective disorders and recent decreases in self-care or self-monitoring. Finally, unstructured parent interviews may be used to gather collateral information regarding client functioning, family environment, recent stressors, and nature of family communication and nurturance (Hendren, 1990; Weller et al., 2001; Wise & Spengler).

Observation of children's spontaneous activities such as play, fantasy, storytelling, and drawing also have been recommended as means of identifying suicidal themes; however, the clinician must be careful not to jump to hasty interpretations without considering context, themes over time, and additional content of play (Hendren, 1990; Wise & Spengler, 1997). While potentially rich in diagnostic information and conducted in a manner consistent with children's preferred modes of communication, the use of play for assessment of suicide risk may be limited by time constraints and a host of other external factors. Asking parents and teachers to describe their observations regarding emotional and behavioral themes to the child's play may, however, provide important clues regarding the child's current level of functioning. Indeed, observations of play, drawings, or stories that involve images of death and harm to self may spark parent or teacher concerns and prompt visits to the counselor for children who may have otherwise been unnoticed.

Goldston (2003) identified and reviewed a plethora of written assessment instruments, semi-structured assessment tools, and structured interviews that in some way address the issue of youth suicide. Most instruments are designed for use with adolescents, include only one or two questions specific to suicide, and are most useful for screening purposes. In many cases, instruments designed to assess for childhood depression or disruptive behavior disorders may contain one or two subquestions that directly access information regarding suicidal ideation, and empirical evidence indicates that children and adolescents are more likely to indicate the presence of suicidal ideation when the question is posed within a written instrument rather than directly from an interviewer (Klimes-Dougan, 1998). In addition, Goldston identified a number of assessment protocols regarding childhood suicidal ideation that have been proposed but not yet empirically studied. For example, Pfeffer (1986) proposed the Child Suicide Potential Scale (CSPS); this clinician-rated instrument provides a comprehensive listing of questions regarding potential precipitating events, stressors, psychological risk factors, family risk factors, and developmental understanding of death that assist the clinician in rating a child's suicide risk on a scale from one (nonsuicidal) to five (serious attempt). Conceptually, this tool is helpful for reminding clinicians of the various factors to be weighed in making a decision, but the length of the instrument may limit the clinical utility of the CSPS in most crisis settings.

Given that mental health disorders and coping styles are considered strong psychological correlates of suicidal behavior, structured interviews and/or psychological tests may be useful for systematically assessing these correlates (Hendren, 1990; Wise & Spengler, 1997). For example, the Children's Interview for Psychiatric Syndromes (CHIPS; Weller, Weller, Rooney, & Fristad, 1999) provides a listing of developmentally appropriate questions used to assess for specific DSM-IV disorders in children. In one of the few articles regarding assessment of suicide risk in children, Weller et al. (2001) recommended a complex series of interviews in which parents and children completed the parent-ChiPS and CHIPS and a number of other semi-structured interview tasks. While unlikely to be practical in a short-term setting, counselors may benefit from utilizing portions of the CHIPS interview schedule to assist in questioning children regarding symptoms of depression and other affective disorders in an age-appropriate manner.

Establish Rapport and Rapidly Establish Relationship

Interviewing and assessing children presents particular challenges, especially if the interviewer is unknown to the child and the interview content addresses threatening or painful information. Mental health counselors frequently conduct risk assessments during the early stages of a counseling relationship, and the need to assess may preempt steps a clinician would ordinarily take to build a relationship with a child. In addition, children may have difficulty answering questions posed by the interviewer because of the concrete nature of information processing, a lack of understanding regarding the concepts (e.g., "depression" or "suicide"), or difficulty understanding grammar or sentence structure (Wilson & Powell, 2001).

Similarly, questioning may be very stressful for young children (Aldridge & Wood, 1998), and children frequently look to the interviewer for friendly cues to approval. Mental health counselors often convey acceptance and caring to children via facial expressions and head-nodding; thus, counselors should be particularly aware of the messages being sent regarding the appropriateness of disclosures as subtle nonverbal expressions may influence children's ability and willingness to disclose (Bourg et al., 1999).

Bourg et al. (1999) recommended using primarily open-ended questions with closed-ended follow-up questions in combination with reflective responses. This mixture of approaches may assist the counselor in obtaining necessary information while avoiding overwhelming or leading the child. On the other hand, less mature children may have difficulty with open-ended questions and may be better able to respond to closed-ended questions. In either case, questioning should be slowly paced, and the interviewer should take particular care to avoid "why" or multiple-choice type questions (Aldridge & Wood, 1998). Mental health counselors are also advised to refrain from repeating yes-no questions as children may perceive that they answered the question incorrectly and thus change the answer in attempts to please the interviewer. Finally, Bourg et al. recommended using short sentences with easy words, eliminating unnecessary words or phrases, avoiding passive voice, and checking-in regarding the child's understanding of questions throughout the interview.

Identify Major Problems

When identifying the precipitating events and problems, mental health counselors may assist the child to first provide a "free narrative account" of the events leading up to the assessment or the appointment (Wilson & Powell, 2001). Wilson and Powell claimed that skilled yet nondirective responses during free narrative may elicit a great deal of information useful for assessing forensic situations; in my experience, children tend to be aware of triggering events or last straws. These open-ended prompts and reflections may assist the child to get the story out without overwhelming him or her with a barrage of questions. In addition, caregivers and teachers often have provided specific information regarding triggering events upon presenting for the appointment. For example, if the child has identified the reason for the visit to be an incident on the playground, the counselor may simply prompt, "tell me about what happened on the playground today." Other means of accessing the narrative may include asking the child to draw a picture of the activating event or asking him or her to tell the story through the use of hand puppets. Such a focus often matches children's highly concrete nature of relating stories.

When identifying the problem, it is helpful to start out generally and move into more specific questioning as the story evolves (Hendren, 1990; Jobes et al., 2000; Pfeffer, 2003; Shea, 1999). This may include asking closed-ended questions to build a more complete story. Vague comments or gestures suggesting possible suicidal ideation must be followed up during the interview. Follow-ups may be in the form of non-directive prompts such as "tell me more" or "what does--mean?"

At this point during the interview, mental health counselors need to address a child's thoughts and desires related to death and thoughts of "hurting" or killing oneself directly. Shea (1999) provided several techniques useful for eliciting suicidal ideation in adults that may be adapted for use with children. Specifically, keeping a focus on behavioral incidents or descriptions of events rather than thoughts about events is concrete and may assist in attenuating shame. A focus on obtaining a linear example of "what happened" may be particularly useful if a child has already engaged in a suicide attempt or gesture. In addition, the use of "gentle assumptions" rather than yes-no questions may be particularly helpful, especially if a behavior is known by others to have occurred. For example, a counselor may ask "How long have you felt like hurting yourself?" rather than "Have you felt like hurting yourself?" This will allow the clinician to introduce topics to the child in a way that suggests familiarity with the child's experience. Care must be taken, however, to avoid leading the child to accepting or discussing behaviors that are not actually present. Finally, Shea discussed the use of normalization to illicit suicidal ideation. For example, a counselor may reflect the child's account of the precipitating event and state, "You've been through a lot. Lots of kids I know would be really sad if something like that happened. Some might even think about wanting to die. Have you thought about ...?"

Pfeffer (2003) recommended the following developmentally appropriate probes for children:

   Did you ever think that you wanted to hurt yourself? Did you ever
   try to hurt yourself? Tell me about what you did. When did you do
   this? Did you ever think about killing yourself? Did you ever plan
   to kill yourself? Did you ever try to kill yourself? When did you
   try to do this?" (p. 215)

If a child has difficulty communicating this information verbally, it may be helpful to ask the child to draw a picture of what he or she thinks, or thinks about doing, when feeling very sad, angry, or scared. The process of drawing may help the client express suicidal ideation or provide the counselor with insight regarding the child's attempts at coping.

In all instances, the mental health counselor should assess for the existence of a suicide plan, intent, and means. In addition, the clinician will need to follow up with questions regarding perceived outcomes of taking such action. Questions regarding perceived outcomes will begin to tap children's understanding of death (Hendren, 1990; Stefanowski-Harding, 1990). This assessment may include use of questions such as, "if you were to stab your heart, what would happen?" and "what does that mean?" These questions will assist the counselor in assessing the child's understanding of death and potential motivators for suicidal behavior (e.g., death will reunite the child with a lost family member or pet, death is temporary, family will be happier). While assessing the child's understanding of death may be helpful in terms of developing interventions, Pfeffer (2003) advised clinicians to take all accounts seriously, as children are capable of planning and implementing suicide even without fully understanding the consequences of their actions. For a more comprehensive review of considerations regarding concepts of death, see Pfeffer's (1986) Child Suicide Potential Scale.

A major focus of the interview is likely to be on the precipitating event and current suicidal ideation, but it is important for the clinician to assess for the relative risk factors discussed previously. Hendren (1990) suggested the following questions as examples of developmentally appropriate means of tapping risk factors:

   Do your mommy and daddy have bad fights? Has anyone in your family
   ever died? Have you ever wanted to or tried to stop living? What do
   you think will happen to you if you do? Do you feel sad a lot of
   the time? Have you ever hurt yourself or someone else on purpose?
   Do your parents have any close friends or family who live nearby?
   Is there anyone who really cares about you? Has anything new or
   different happened to your family lately? (pp. 242-243)

Finally, while children are known to be the best sources of information regarding the presence or absence of suicidal ideation (Pfeffer, 2003), accurate information regarding the presence of risk factors and specific behaviors may be easily elicited from caregivers and may reduce the strain of continual questioning on an already distressed child.

Deal with Feelings and Emotions

Reflection of feeling and sustaining techniques often are important to clients of all ages during assessment (Murphy & Dillon, 2003); addressing and dealing with feelings are a crucial component of crisis intervention (Jobes et al., 2000; Roberts, 2000). In many cases, the simple availability of an empathic listener who demonstrates poise and a sense of control or grounding is comforting to those in crisis. Clinicians should pay special attention to cues given by children (e.g., panic response, covering face with hands, attempting to leave the room) that they are becoming overwhelmed. Taking breaks, ensuring the child's physical comfort, and assisting the child in taking deep breaths or utilizing relaxation strategies may prove to be invaluable to dealing with feelings during the session. At times, providing the child with a squish ball or some play-doh to hold during the interview may assist in releasing tension and anxiety while talking. In addition, Jobes et al. (2000) recommend specific questioning related to client's feelings around the events. Children who are less verbal may be asked to draw a picture of how they are feeling and/or to choose a "feeling face" from a poster that portrays a variety of emotional themes and intensities.

Generate and Explore Alternatives

As the mental health counselor works with client and family to build relationships and assess the current situation, focus of the interview will begin to shift from assessment to intervention. Because children are not independently capable of consenting to specific outcome plans, caregiver involvement and participation during this phase of the interview is critical. During this time, the clinician must assess the caregiver's perception of the problem and alternatives in order to co-construct an action plan that is suitable for all involved. This exploration should include a flank discussion of the clinician's concerns including indications of suicidal ideation and information regarding the presence or absence of a specific suicide plan. Depending on the child's maturity, quality of parent-child relationship, and nature of recommendations, the counselor may wish to meet separately with caregivers to secure support and address concerns prior to presenting the plan to the child. Either way, mental health counselors may work with caregivers to assess whether some symptom or stressor relief is possible for the child and whether the caregivers are willing and able to provide the support necessary to ensure the child's safety (Jobes et al., 2000). In my experience, most parents respond quite candidly to a simple question regarding their perceived ability to keep the child safe. Parental reports, however, should be carefully considered as parents often underestimate the suicidal content and behavior of their children, and underestimations are frequently amplified in the presence of other risk factors (e.g., when the parent is not well and in the midst of family stress) (Klimes-Dougan, 1998).

Fristad and Shaver (2001) suggested that parent education and consultation is enough to reduce risk and ensure adequate parent awareness and care for the child in most cases. In the event that caregiver and child reports are grossly mismatched or the parent is unable or unwilling to seek assistance for a child who presents with a high degree of risk, the clinician should consider initiating involuntary commitment procedures and/or reporting the guardian to social services for neglect. These actions should be considered carefully and used conservatively as they may be traumatic for the child and reduce the caregiver's likelihood of seeking and complying with future services.

Develop and Formulate Action and Follow-up Plans

As alternatives are generated, the mental health counselor should work closely with the child and family to activate necessary resources. Plans may be reviewed, and family members may find it helpful to write out specific action steps and recommendations. Similarly, the clinician may work with the family to coordinate skills and resources needed to keep the child safe. During these stages the counselor may assist by scheduling follow-up appointments, educating regarding access to 24-hour crisis services, and making arrangements for inpatient hospitalization if necessary. Upon completion of an action plan, the clinician should review the plan in its entirety and check with each individual--especially the child--to determine whether the plan is acceptable given current needs.

CONCLUSION

Suicidal behavior is not rare, and mental health counselors often are in key positions for identifying, assessing, and coordinating care for children who present with suicidal ideation and behaviors. Mental health counselors have an ethical responsibility to attend closely to clients' needs, and this responsibility includes conducting developmentally appropriate assessments and providing crisis intervention at a moment's notice. Such practices can be highly emotionally charged and procedurally uncomfortable for even experienced counselors. First steps to generating competence around this issue include educating oneself regarding risk factors and indicators, examining one's own beliefs and discomforts related to childhood suicide, and developing an action plan for learning about and conducting accurate assessments. Of course, consultation, supervision, and peer support are critical to ensuring the counselor's ability to make sound decisions, and debriefing after particularly difficult assessments is important. Suggestions for approaching and structuring suicide assessment interviews with children and working with caregivers have been provided; integration of these approaches into practice can prove useful for working with children at risk for suicide and suicidal behavior.

REFERENCES

Aldridge, M., & Wood, J. (1998). Interviewing children: A guide for child care and forensic practitioners. New York: John Wiley & Sons, Inc.

American Association of Suicidology. (2003). U.S.A. suicide: 2003 official final data. Retrieved September 15, 2006, from http://www.suicidology.org/ associations/1045/files/2003datapgb.pdf

American Association of Suicidology. (2006). Youth suicide fact sheet. Retrieved September 15, 2006, from http://www.suicidology.org/associations/1045/files/ Youth2003.pdf

American Counseling Association. (2005). Code of ethics. Alexandria, VA: Author.

American Mental Health Counselors Association. (2000). Code of ethics. Retrieved October 3, 2006, from http://http://www.amhca.org/code/#2

Berman, A. L., Jobes, D. A., & Silverman, M. M. (2006). Adolescent suicide: Assessment and intervention. Washington, D.C.: American Psychological Association.

Bourg, W., Borderick, R., Flagor, R., Kelly, D. M., Ervin, D. L., & Butler, J. (1999). A child interviewer's guidebook. Thousand Oaks, CA: SAGE Publications.

Brent, D. A., & Kolko, D. J. (1990). The assessment and treatment of children and adolescents at risk for suicide. In S. J. Blumenthal & D. J. Kupfer (Eds.), Suicide Over the Life Cycle: Risk Factors, Assessment, and Treatment of Suicidal Patients (pp. 253-302). Washington, D.C.: American Psychiatric Press, Inc.

Burgess, A. W., & Roberts, A. R. (2000). Crisis intervention for persons diagnosed with clinical disorders based on the stress-crisis continuum. In A. R. Roberts (Ed.), Crisis Intervention Handbook: Assessment, Treatment, and Research (2nd ed., pp. 56-76). Oxford: Oxford University Press.

Fristad, M. A., & Shaver, A. E. (2001). Psychosocial interventions for suicidal children and adolescents. Depression and Anxiety, 14, 192-197.

Goldston, D. B. (2003). Measuring suicidal behavior and risk in children and adolescents. Washington, D.C.: American Psychological Association.

Gould, M. S., Shaffer, D., & Greenberg, T. (2003). The epidemiology of youth suicide. In R. A. King & A. Apter (Eds.), Suicide in Children and Adolescents (pp. 1-40). New York: Cambridge University Press.

Hendren, R. L. (1990). Assessment and interviewing strategies for suicidal patients over the life cycle. In S. J. Blumenthal & D. J. Kupfer (Eds.), Suicide Over the Life Cycle: Risk Factors, Assessment, and Treatment of Suicidal Patients (pp. 235-252). Washington, DC: American Psychiatric Press, Inc.

Jobes, D. A., Berman, A. L., & Martin, C. E. (2000). Adolescent suicidality and crisis intervention. In A. R. Roberts (Ed.), Crisis Intervention Handbook: Assessment, Treatment, and Research (2nd ed., pp. 131-151). Oxford: Oxford University Press.

Juhnke, G. A. (1996). The adapted-SAD PERSONS: A suicide assessment scale designed for use with children. Elementary School Guidance & Counseling, 30, 252-258.

Klimes-Dougan, B. (1998). Screening for suicidal ideation in children and adolescents: Methodological considerations. Journal of Adolescence, 21,435-444.

Lawrence, G., & Robinson Kurpius, S. E. (2000). Legal and ethical issues involved when counseling minors in nonschool settings. Journal of Counseling & Development, 78, 130-136.

Murphy, B. C., & Dillon, C. (2003). Working with people in crisis. In Interviewing in action: Relationship, process, and change (2nd ed., pp. 195-220). Pacific Grove, CA: Brooks/Cole.

National Institute of Mental Health. (2003). Suicide facts [Brochure]. Author. Retrieved March 10, 2004, from http://www.nimh.nih.gov/research/ suifact.cfm

Pfeffer, C. R. (1986). Child suicide potential scales. In The suicidal child (pp. 173-203). New York: Guilford Press.

Pfeffer, C. R. (2003). Assessing suicidal behavior in children and adolescents. In R. A. King & A. Apter (Eds.), Suicide in Children and Adolescents (pp. 211-226). New York: Cambridge University Press.

Roberts, A. R. (2000). An overview of crisis theory and crisis intervention. In A. R. Roberts (Ed.), Crisis Intervention Handbook: Assessment, Treatment, and Research (2nd ed., pp. 2-30). Oxford: Oxford University Press.

Roberts, A. R., & Yeager, K. R. (2005). Lethality assessment and crisis intervention with persons presenting with suicidal ideation. In A. R. Roberts (Ed.), Crisis Intervention Handbook: Assessment, Treatment and Research (3rd ed., pp. 35-63). Oxford: Oxford University Press.

Shea, S. C. (1999). The practical art of suicide assessment: A guide for mental health professionals and substance abuse counselors. New York: John Wiley & Sons, Inc.

Singer, J. (2005). Child and adolescent psychiatric emergencies: Mobile crisis response. In A. R. Roberts (Ed.), Crisis Intervention Handbook: Assessment, Treatment and Research (3rd ed., pp. 319-361). Oxford: Oxford University Press.

Stallion, J. M., & McDowell, E. E. (1996). Suicide across the life span: Premature exits (2nd ed.). Washington, D.C.: Taylor & Francis.

Stefanowski-Harding, S. (1990). Child suicide: A review of the literature and implications for school counselors. School Counselor, 37, 328-339.

Weller, E. B., Weller, R. A., Rooney, M. T., & Fristad, M. (1999). Children's interview for psychiatric syndromes (CHIPS). Washington, D.C.: American Psychiatric Press, Inc.

Weller, E. B., Young, K. M., Rohrbaugh, A. H., & Weller, R. A. (2001). Overview and assessment of the suicidal child. Depression and Anxiety, 14, 157-163.

Westefeld, J. S., Range, L. M., Rogers, J. R., Maples, M. R., Bromley, J. L., & Alcorn, J. (2000). Suicide: An overview. The Counseling Psychologist, 28, 445-510.

Wilson, J. C., & Powell, M. (2001). A guide to interviewing children: Essential skills for counsellors, police, lawyers and social workers. New York: Routledge.

Wise, A. J., & Spengler, P. M. (1997). Suicide in children younger than age fourteen: Clinical judgment and assessment issues. Journal of Mental Health Counseling, 19, 318-335.

Casey A. Barrio is affiliated with the University of North Texas. E-mail: [e-mail].