Validation of the SIMPLE STEPS model of suicide assessment.
According to the most recent statistics, suicide is the tenth leading cause of death in the United States and ranked second among youth (American Association of Suicidology, 2015). According to scholarly literature, working with client suicidality is one of the most difficult and stress provoking issues that counselors face (Bonner, 1990; Kleespies, Penk, & Forsyth; McAdams & Foster, 2000). Though it is difficult, given the high prevalence of suicide, counselors must assess, treat, and attempt to prevent suicide on a daily basis.
In an effort of prevention, counselors frequently recommend suicide prevention hotlines as a resource to provide suicidal clients with 24-hour support. For several decades, suicide hotlines have represented the nation's most advertised and used method of suicide prevention (Budinger, Cwik, & Riddle, 2015). In fact, when The Los Angeles Suicide Prevention Center opened its first hotline in 1958, experts claimed it to be the most significant event in the history of the suicide prevention movement (Spencer-Thomas & Jahn, 2012). Currently, hotlines are widely known and highly accessible and provide a valuable resource to individuals who may not seek other means of professional help (Witte, Gould, Munfakh & Kleinman, 2010). Suicide prevention hotlines have also been seen as providing the truest look at one's suicidality because people typically call these hotlines in their highest state of suicidal ideation and lethality (McGlothlin, 2008).
Over the years, suicide prevention hotlines have become more scrutinized and standardized, as evidenced by accreditation standards set forth by the American Association of Suicidology (AAS; 2011). In an effort to provide exemplary crisis programs, AAS developed standards to assist agencies in refining and enhancing their services. A crucial element to these programs is lethality assessment. The current AAS (2011) accreditation standards manual recommends that the following factors be included when determining lethality: estimation of immediate risk, ideology, and intent to complete, assessment of plans and means, previous attempts, state of depression, history of mental illness, alcohol and drug use, survivor status, history of stress and loss, degree of hopelessness and helplessness, and availability of support. Ultimately, it is up to the crisis worker to determine if, when, and how to include all components of the assessment. There is no standardized method. The decisions to potentially break confidentiality in order to save the caller's life are based on this risk assessment (Joiner et al., 2007).
While the above standards are intentionally general in nature, specific standardized models have been developed to assess suicide risk factors. Most notably, mnemonic devices are typically used to help hotline workers and counselors assess suicide risk factors and in turn assess suicide lethality. According to Granello and Granello (2007) some of the most basic mnemonic devices for assessing suicidality include:
* NO HOPE by Shea, 2002 (No meaning, Overt change in clinical presentation, Hostile environment, Out of hospital recently, Predisposing personality factors, and Excuses for dying)
* PIMP (Plan, Intent, Means, and Prior attempts)
* PLAID (Prior attempts, Lethality, Access to means, Intent, and Drugs / alcohol)
* SLAP (Specificity, Lethality, Availability, and Proximity of help)
In addition, Patterson, Dohn, Bird, and Patterson (1983) developed the SAD PERSONS mnemonic (Sex, Age, Depression and hopelessness, Prior attempts and psychiatric illness, Excessive alcohol and drug use, Rational thinking loss, Separated / widowed / divorced, Organized serious attempt, No social supports, and Stated future intent) which has been used for years and also revised to be used specifically with children (Juhnke, 1996).
In 2003 The AAS established a more comprehensive mnemonic device for educating the public on suicide risk factors called the IS PATH WARM model (Lester, McSwain & Gunn, 2011). The mnemonic stands for Suicide Ideation, Substance abuse, Purposelessness, Anger, Trapped, Hopelessness, Withdrawing, Anxiety, Recklessness, and Mood change. While analyzing the IS PATH WARM model, Lester, McSwain, and Gunn found little evidence of these 10 warning signs differentiating between suicidal individuals and non-suicidal individuals. However, they did pose that the IS PATH WARM model provided crisis counselors with a tool to triage clients or to help make determinations of suicide risk.
In 2008, McGlothlin developed the SIMPLE STEPS model and has since revised this model slightly to include the importance of protective factors. Currently, the SIMPLE STEPS model represents the following criteria for assessing suicide lethality:
* Suicidal: does the person verbalize intension to kill himself / herself?
* Ideation: pervasiveness of suicidal thoughts; "On a scale from 1 to 10 (1 being not likely at all and 10 being definite), how likely are you to commit suicide within the next 72 hours?" (p. 37)
* Method: nature of the person's plans on committing suicide; the lethality and availability of the means; the comprehensiveness of the plan
* Perturbation: degree of emotional distress or degree of emotional pain
* Loss: actual or perceived losses of people or objects
* Earlier Attempts: existence of previous attempts, and the details of the event (i.e., what was going on in his life, what happened so that he did not complete, etc.)
* Substance Use: use of alcohol, prescribed and non-prescribed drugs, and medication compliance
* (Lack of) Troubleshooting Skills: Openness to alternatives other than suicide; current ability to solve problems
* Emotions / Diagnosis:--Assessment of emotional attributes (hopelessness, helplessness, worthlessness, loneliness, agitation, depression, and impulsivity) and diagnoses commonly associated with completed suicide (e.g., substance abuse, mood disorders, personality disorders, etc.)
* (Lack of) Protective Factors: individual strengths, resources, supports or skills within the person or in their family or community that could mitigate the risk of attempting/completing suicide
* Stressors and Life Events: evaluation of current and past life stressors; "What is going on in your life that has leads you to thinking of suicide?" (p. 37)
The SIMPLE STEPS (McGlothlin, 2008) model of suicide assessment provides clinicians with a comprehensive framework for appraising suicide lethality. It is grounded in (a) the creator's experience as a hotline volunteer and crisis counselor, (b) empirically determined risk factors (Rudd et al., 2006), and (c) various theoretical assertions in the science of suicidology (SpencerThomas & Jahn, 2012). It may be utilized for individuals across the lifespan (McGlothlin, 2008). The SIMPLE STEPS model infuses solution-focused perspectives to empower individuals by focusing on strengths, highlighting the possibility of change, focusing on the present and future, and encouraging a collaborative relationship between helper and person considering suicide (Berg & Miller, 1992; McGlothlin, 2008). To date, the SIMPLE STEPS model has been used extensively in practice, though empirical evidence has not been produced on its validity until this article. The purpose of this study was to examine how the variables in the SIMPLE STEPS model of suicide assessment related to suicide lethality, thereby establishing some basic validity evidence.
Data Collection and Preparation
Data were collected from a suicide prevention hotline in a major mid-western city. For each caller to the hotline, a "contact sheet" was completed by as hotline worker who had been through a minimum of 70 hours of training on suicidality and lethality assessment. Contact sheets are essentially the case notes from each call to the hotline and include a minimum of the following standardized areas of information: age, sex, marital status, mental health concerns, medications, lethality assessment, reason for calling, current problems, suicide plan, means, history of suicide, alcohol and drugs, and outcome of call. Hotline workers are trained to complete the contact sheets by inserting verbatim statements from the callers into the majority of the above categories. Therefore, the majority of the data that were collected were originally in a qualitative format (i.e., string data).
Between January 2008 and December 2013, 18,218 contact sheets were collected. At this point, four procedures were taken to prepare the qualitative contact sheets for quantitative analysis and ultimately the regression analysis: (1) culling out cases that were not usable, (2) identifying qualitative statements that could be converted into dichotomous variables, (3) grouping responses of similar meaning i.e., creating grouped variables (GVs), and (4) creating Simple Steps variables (SSV).
First, contact sheets were culled for unusable or irrelevant cases. For example, prank or obscene calls were excluded. Calls that lasted less than five minutes and calls that did not include an assessment of overall suicide lethality were excluded because they provided little information. Duplicate calls and contact sheets that were difficult to read were excluded. Lastly, callers who contacted the hotline multiple times and were seen as "chronic callers" were excluded as they possess a different context to their suicidality. As a result of this procedure, 4,795 (26.3%) of the overall sample were excluded from the overall sample. Therefore, a working sample of 13,423 callers to a suicide prevention hotline over a six-year period was obtained.
Second, some variables were easily turned into quantitative measures. For example, each caller's suicidal lethality was determined and placed on a five-point scale (low = 1, low-moderate = 2, moderate = 3, moderate-high = 4, and high = 5). Also, some items were easily converted into dichotomous "yes" (1) or "no" (2) variables; for example, whether or not a client had a history of suicide attempts and whether or not they currently consumed alcohol or drugs.
Third, in order to make meaning out of many of the qualitative strings, they needed to be converted into GVs. Since verbatim caller statements were used, a structure of the statements needed to be created. For example, under the section of "plan" statements such as "jumping from a tower", "jumping from a bridge", and "jumping off something really high" were all labeled "jumping". Even more complex was the conversion of qualitative strings under the category of "reason for call". Many callers presented several reasons for calling and for wanting to commit suicide. Originally, the reasons for calling the hotline were put into 101 GVs. However, after further analysis by the three researchers, these 101 GVs were refined even further into 34 GVs. For example, panic attacks, agoraphobia, social anxiety, anxiety, worry, being scared, and being afraid created the GV of "anxiety and fear".
Lastly, the three researchers discussed the concepts of each of the Simple Steps variables (SSV). The hotline data were examined for conceptually matching variables that could be used to operationalize the Simple Steps variables. These conceptually matching variables were entered into equations used to compute the Simple Steps Variables. These equations are listed below. It should be noted that SSVs were created from GVs derived from mostly from "reasons for calling" and "current problems" listed. A single concept, for example loss, might have been given as a reason for calling, a problem, both a reason for calling and a problem, or might have been neither a reason for calling nor a problem. This why in some of the equations below a concept might be listed twice in the same equation. For clarity, in these equations, variables that were listed as reasons for calls are simply named. Variables that were listed as problems are listed follow by an asterisk (for example loss*).
Equations for the Simple Steps Variables (SSV) were:
(S1) Suicidal = yes or no;
(I) Ideation = ideations + suicide emotions;
(M) Means = plan + means + cutting now;
(P) Perturbation = extreme negativity + emotional;
(L) Loss = divorce/custody/cheating + grief + loss + loss* + grief*;
(E1) Earlier attempts = yes or no;
(S2) Substance Use = medication issues + drugs/alcohol + alcohol/drugs* + alcohol/drug history* + medication*;
(T) [Lack of] Trouble Shooting Skills = coping with life events, coping, resources, + distance means;
(E2) Emotion = lonely + anger + hopelessness + helplessness + mental health diagnosis + depression + anxiety/fear;
(P2) [Lack of] Protective Factors = reaching out + protective factors+ protective*; and
(S3) Stress = violence/abuse + relationship + financial + medical + legal + stress + caregiver stress + life events + housing + job + GLBT stress + school + sleep.
The quantification of the "reason for call" and "problem" variables occurred in the following manner. If a participant listed a specific reason for call or problem, it was given a score of 1. If it was not listed, it was scored as zero. The scores on the group variables and problem variables were summed to obtain the participants score on a SSV. Values for the SSVs were computed using the SSV equations reported above.
Of the 13,423 callers, 7,873 (58.7%) were female and 5,550 (41.3%) were male. There was an age range of 8 years old to 89 years old with a mean age of 25. A specific age breakdown of the participants is 8-14 (n = 356, 2.6%), 15-24 (n = 2,908, 23.8%), 25-44 (n = 3,562, 26.6%), 45-64 (n = 2,737, 20.4%), 65+ (n = 314, 2.2%), and unknown age (n = 3,546, 24.4%).
In terms of the suicide lethality of this sample it was determined that 5,226 (38.9%) were assessed as having low lethality, 4,485 (33.4%) had low-moderate lethality, 2,867 (21.4%) had moderate lethality, 550 (4.1%) had moderate-high lethality, and 295 (2.2%) had high lethality.
The relationship of the SSVs to lethality assessment (LA) was examined by standard multiple regression analysis. The SSVs were the predictors in the regression equation. Values for the SSVs were computed using the SSV equations reported above. LA, which used a 1-5 rating scale, was used as the criterion in the regression equation. The SSVs were weighted equally when entering them into the analysis since the literature provided no justification for weighting the variables any other way.
The results of the regression predicting LA from SSVs can be found in Table 1. Assumptions of normally distributed errors, non-colinearity, and uncorrelated errors were checked and met. The combination of SSVs predicted Lethality Assessment with [R.sup.2] =.623.
The purpose of this study was to examine how the variables in the SIMPLE STEPS model of suicide assessment related to suicide lethality in a large-scale sample of callers (n = 13,423) to a suicide prevention hotline over a six-year period. Overall, it was found that this model provides a valid measure of assessing suicide lethality. While each of the eleven SIMPLE STEPS variables were predictive of higher levels of suicide lethality, some variables contributed more to lethality than other variables. For example, Troubleshooting, Emotion and diagnosis, Verbalizing suicidality, and Protective factors are associated with suicide lethality. Most unique to these data is the notion that troubleshooting and protective factors are strong predictors of suicide lethality. Many authors within the last 20 years have placed importance on protective factors and troubleshooting and coping skills (Simon & Hales, 2012) in relation to suicide lethality; yet not to the degree that these data have indicated. In addition, the ability to troubleshoot crises, ability to cope with difficulties, and protective factors are not in many other mnemonic suicide assessments.
Implications for Counseling Practice
The results of this study have implications for the practice of assessing suicidal clients with mnemonic devices and the general treatment of suicidal clients. First, when assessing clients for suicidality, counselors cannot solely rely on basic mnemonic devices such as PIMP, PLAID, and SLAP. The data from this study suggest that there are more factors predicting suicide lethality than the basic mnemonic devices include. Table 2 provides a detailed comparison of the common mnemonic suicide assessments. Clearly the SIMPLE STEPS model is not only more comprehensive in nature, but the data from this study suggest that many of the other mnemonic assessments are missing critical variables that equate to higher levels of suicide lethality (i.e., troubleshooting, protective factors).
In most clinical settings, clinicians treat suicidal clients with the initial strategy of reducing depressive symptoms and reducing suicidal behaviors (Simon & Hales, 2012). While the data from this study do not dispute this, the data do suggest that there may be more that needs to be done. The data suggest that all of the components of the SIMPLE STEPS model contribute to perceived suicide lethality. Therefore, progress may be made toward reducing suicide lethality by working on other factors contributing to suicide lethality in addition to depression and suicidal behaviors.
For example, working with suicidal clients is difficult, stress provoking and complex. Clinicians may have difficulty seeing changes in depressive states, especially if the depression is severe and medication has not been effective. The data suggest that improving troubleshooting skills, decreasing stressors, working on issues of loss, increasing protective factors, etc. can be highly effective as well. The clinical task is not just depression reduction and suicidal behavior reduction. The clinical task (as suggested by the data in this study) is to work on all of the SIMPLE STEPS variables. Helping improve a client's state on just one of the SIMPLE STEPS variables could (to some degree) reduce overall suicide lethality.
This study is not without limitations. First, the researchers were suspicious about the accuracy of the data entry. For example, hotline workers complete a handwritten contact sheet on each call and in turn, the information on that contact sheet is entered into a database. In the initial look at the data entered into the database, some cases did not make sense (i.e., a call lasted 3 minutes but all fields were entered and a thorough assessment took place). The researchers took measures to delete such cases that were difficult to understand however, we realize that some of the data may have been entered incorrectly. It is the thoughts of the authors that given the size of the final sample (N = 13,423) the small amount of unknown yet possible data entry mistakes would not materially affect the significance of this study.
Also related to using data from another source, another limitation of this study the fact that many of the initial data were string data and needed to be converted into quantitative values. Therefore, the researchers were required to make judgments on how to convert qualitative data into quantitative values. Consultation amongst the three authors along with making judgments based on the literature was used to aid in the validity of these judgments.
This is a correlational study; as such, one cannot derive cause-effect conclusions from the analysis. Caution must also be exercised in concluding that the SSVs predict lethality per se; the criterion measure was perceived lethality as judged by the same individual who rated the SSVs, rather than an independent measure of lethality.
While this study provides support for the SIMPLE STEPS model of suicide assessment, more research needs to be done on the utility of this model. For example, though the participants in this study were callers from a suicide prevention hotline and provides an accurate portrayal of suicidality, the use of the SIMPLE STEPS model in face-to-face settings would provide further evidence into the utility of this model in different settings. Furthermore, the 11 different variables of the model could be analyzed more with different populations to see if a client's sociocultural and multicultural attributes contribute to how each variable contributes to one's overall suicide lethality. The researchers realize that this large-scale study provided an overall glimpse into the validity of this model, more research needs to be done generally in the field of suicide assessment, prevention, and treatment, given the large societal impact suicide has and how suicide affects all clinicians.
American Association of Suicidology (2011). Organization accreditation standards manual (10th ed.), Washington, DC: Author.
American Association of Suicidology. (2015). U.S.A. Suicide: 2013 official final data. Retrieved from http://www.suicidology.org/Portals/14/docs/Resources/FactSheets/2013datapgsv3.pdf
Berg, I. K., & Miller, S. D. (1992). Working with the problem drinker: A solution-focused approach. New York, NY: Norton.
Bonner, R. L. (1990). A "MAP." to the clinical assessment of suicide risk, journal of Mental Health Counseling, 12, 232-236.
Budinger, M. C., Cwik, M. F., & Riddle, M. A. (2015). Awareness, attitudes, and use of crisis hotlines among youth at-risk for suicide. Suicide and Life-Threatening Behavior, 45, 192-198.
Granello, D. H., & Granello, P. F. (2007). Suicide: An essential guide for helping professionals and educators. Boston, MA: Pearson.
Joiner, T., Kalafat, J., Draper, J., Stokes, H., Knudson, M., Berman, A., & McKeon, R. (2007). Establishing standards for the assessment of suicide risk among callers to the National Suicide Prevention Lifeline. Suicide and Life-Threatening Behavior, 37, 353-365. doi: 10.1521/ suli.2007.37.3.353
Juhnke, G. A. (1996). The Adapted SAD PERSONS: A suicide assessment scale designed for use with children. Elementary School Guidance (5 Counseling, 30, 252-258.
Kleespies, P. M., Penk, W. E., & Forsyth, J. P. (1993). The stress of patient suicidal behavior during clinical training: Incidence, impact, and recovery. Professional Psychology: Research and Practice, 24, 293-303. doi: 10.1037/0735-7028.24.3.293
Lester, D., McSwain, S., & Gunn, J. F. (2011). A test of the validity of the IS PATH WARM warning signs for suicide. Psychological Reports, 108, 402-404.
McAdams, C. R., & Foster, V. A. (2000). Client suicide: Its frequency and impact on counselors. journal of Mental Health Counseling, 22, 107-121.
McGlothlin, J. (2008). Developing clinical skills in suicide assessment, prevention and treatment. Alexandria, VA: American Counseling Association.
Patterson, W. M., Dolrn, H. H., Bird, J., & Patterson, G. A. (1983). Evaluation of suicidal patients: The SAD PERSONS Scale. Psychosomatics, 24, 343-249. doi: 10.1016/S0033-3182(83)73213-5
Rudd, M. D., Bernan, A. L., Joiner, T. E., Nock, M. K., Silverman, M. M., Mandrusiak, M., ... Witte, T. (2006). Warning signs for suicide: Theory, research, and clinical applications. Suicide and Life-Threatening Behavior, 36, 255-262.
Shea S. (2002). The practical art of suicide assessment. A guide for mental health professionals and substance abuse counselors. Hoboken, NJ: John Wiley & Sons.
Simon, R. I., & Hales, R. E. (2012). The American Psychiatric Publishing textbook of suicide assessment and management. (2n^ ed.). Washington, DC: American Psychiatric Publishing.
Spencer-Thomas, S., & Jahn, D. R. (2012). Tracking a movement: U.S. milestones in suicide prevention. Suicide and Life-Threatening Behavior, 42, 78-85. doi: 10.1111/j. 1943-278X.2011.00072.x
Witte, T. K., Gould, M. S., Munfakh, J. L. H., Kleinman, M., & Joiner, T.E. (2010). Asessing suicide risk among callers to crisis hotlines: A confirmatory factor analysis, journal of Clinical Psychology, 66, 941-964. cloi: 10.1002/jclp.20717
Kent State University
Jason McGlothlin, Counselor Education and Supervision, Kent State University; Betsy Page, Counselor Education and Supervision, Kent State University; Kelsey Jager, Counselor Education and Supervision, Kent State University. Correspondence concerning this article should be addressed to Jason McGlothlin, Counselor Education and Supervision, 310 White Hall, PO Box 5190, Kent State University, Kent, Ohio 44242-0001. E-mail: email@example.com
Table 1 Simple Linear Regression Analysis Summary of Scores for Simple Steps Variables Predicting Scores on Lethality Assessment Predictor B SE B [beta] (Constant) -2.653 .072 (S1) Suicidal .374 .013 .184 (I) Ideation .250 .017 .082 (M) Means -.064 .016 -.031 (P) Perturbation .093 .022 .022 (L) Loss .141 .013 .057 (E 1) Earlier Attempts .164 .011 .083 (S 2) Substance Use .047 .004 .067 (T) [Lack of] Troubleshooting .467 .007 .498 (E2) Emotion and Diagnosis .140 .004 .211 (P 2) [Lack of] Protective Factors .238 .009 .153 (S 3) Stress .041 .004 .060 Note. [R.sup.2] = .623; F = 901.129; all ps < .001 Table 2 Comparison of Mnemonic Suicide Assessments SIMPLE IS PATH SAD STEPS WARM PERSONS NO HOPE Alcohol/Drug Use S2 S E Anger/Agitation E2 A1 Anxiety E2 A2 Depression E2 D Earlier Attempts E1 P Emotional Volatile Individuals E2 P Helplessness E2 Hopelessness/Meaninglessness E2 H D N Ideation/Intent I I S2 E Impulsivity E2 Lack of Support P2 N Lethality of method M 0 Loss (Actual and Perceived L Means M Medication compliance S2 Mental Health Diagnosis E2 P Mood Chance E1 M 0 Perturbation P1 Plan M 0 Poor Relationships S3 H Protective Factors P2 Proximity of Help Purposelessness/Worthlessness E2 P Rational thinking loss E2 R Recent Hospital Release S3 0 Recklessness R Separated/Widowed/Divorced L S Sex and Age SA Stressors S3 Trapped T Troubleshooting T Verbalized Suicide Intent S1 S2 Withdrawing/Loneliness E2 W PLAID SLAP PIMP Alcohol/Drug Use D Anger/Agitation Anxiety Depression Earlier Attempts A P Emotional Volatile Individuals Helplessness Hopelessness/Meaninglessness Ideation/Intent I Impulsivity Lack of Support Lethality of method L Loss (Actual and Perceived Means L A M Medication compliance Mental Health Diagnosis Mood Chance Perturbation Plan P S P Poor Relationships Protective Factors Proximity of Help P Purposelessness/Worthlessness Rational thinking loss Recent Hospital Release Recklessness Separated/Widowed/Divorced Sex and Age Stressors Trapped Troubleshooting Verbalized Suicide Intent Withdrawing/Loneliness P Note: SIMPLE STEPS: S1 = Suicidal, I = Ideation, M = Method, P1 = Perturbation, L = Loss, E1 = Earlier Attempts, S2 = Substance use, T = (Lack of) Troubleshooting, E2 = Emotion/Diagnosis, P2 = Protective factors, and S3 = Stressor and life events. IS PATH WARM: I = Suicide ideation, S = Substance abuse, P = Purposelessness, A1 = Anger, T = Trapped, H = Hopelessness, W = Withdrawing, A2 = Anxiety, R = Recklessness, M = Mood change. SAD PERSONS: S1 = Sex, A = Age, D = Depression and hopelessness, P = Prior attempts and psychiatric illness, E = Excessive alcohol and drug use, R = Rational thinking loss, S2 = separated/widowed/divorced, O = Organized serious attempt, N = No social supports, and S3 = Stated future intent. NO HOPE: N = No meaning, O1 = Overt change in clinical presentation, H = hostile environment, O2 = Out of hospital recently, P = Predisposing personality factors, and E = Excuses for dying. PLAID: P = Prior attempts, L = Lethality, A = Access to means, I = Intent, D = Drugs/alcohol. SLAP: S = Specificity, L = Lethality, A = Availability, and P = Proximity of help. PIMP: P1 = Plan, I = Intent, M = Means, and P2 = Prior attempts.
|Printer friendly Cite/link Email Feedback|
|Author:||McGlothlin, Jason; Page, Betsy; Jager, Kelsey|
|Publication:||Journal of Mental Health Counseling|
|Date:||Oct 1, 2016|
|Previous Article:||Excoriation: what counselors need to know about skin picking disorder.|
|Next Article:||Mental health practitioners' knowledge of colleagues' problems of professional competency.|