by Jennifer M. Giddens and David V. Sheehan, MD, MBA
J. Giddens is Co-founder of the Tampa Center for Research on Suicidality, Tampa, Florida; and Dr. Sheehan is Distinguished University Health Professor Emeritus, University of South Florida College of Medicine, Tampa, Florida.
Innov Clin Neurosci. 2014;11(9–10):191–193
Funding: There was no funding for the development and writing of this article.
Financial Disclosures: J. Giddens is the author and copyright holder of the Suicide Plan Tracking Scale (SPTS) and is a named consultant on the Sheehan-Suicidality Tracking Scale (S-STS), the Sheehan-Suicidality Tracking Scale Clinically Meaningful Change Measure Version (S-STS CMCM), the Pediatric versions of the S-STS, and the Suicidality Modifiers Scale; Dr. D. Sheehan is the author and copyright holder of the S-STS, the S-STS CMCM, the Pediatric versions of the S-STS, the Sheehan Disability Scale (SDS), and the Suicidality Modifiers Scale, is a co-author of the SPTS, and owns stock in Medical Outcomes Systems, which has computerized the S-STS.
Key words: Suicide, suicide assessment, suicidality, patient-rated suicide assessment, S-STS, Suicidality Modifiers Scale
Abstract: Objective: This case study explores to what extent, if any, a subject’s reporting varies depending upon whom the subject believes will view the data and the relationship the subject has with the reviewer. It also explores the variance in reporting if several days pass between the timeframe in question and the time of data collection. Method: The subject answered three suicidality-related scales (the Sheehan-Suicidality Tracking Scale, the Suicidality Modifiers, and the Alphs Dichotomous Impulsivity and Hopelessness Two Questions) for 24 distinct timeframes. The scales were rated three different ways for each timeframe. The first was for only the patient. Immediately afterward, the scales were answered for the subject’s therapist. A few days later, the scales were again answered by the patient, but only for the patient. The three different interviews for the same timeframe were compared to investigate any deviations. Results: This case study found clinically relevant deviations between the three ratings completed for the same timeframe. Conclusion: This case study illustrates that a patient’s reporting of his or her symptoms of suicidality using a patient-rated scale can vary depending upon the context, distance from timeframe in question, and the patient’s relationship with the reviewer of the data.
Introduction
When patients complete a patient-rated suicidality scale regarding a particular timeframe (e.g., a prior week or month or since their last visit with their clinician), confidence that the data are accurate and will not change over time is desired in order for clinicians to effectively treat their patients. As with most scales, the reported scores progressively decrease with the passage of time when patients reflect back on the same timeframe.[1]
Some patients have reported to the authors of this article that they had, at times, been less than truthful with their clinicians (sometimes on purpose and other times because their perspective of symptoms had changed). A few reported being dishonest even with themselves.
To explore this phenomenon further, we designed this case study to document to what extent, if any, a subject’s reporting of suicidality may vary depending upon who will view the data, the subject’s relationship to the reviewer of the data, and whether the reporting varies if several days have passed between the timeframe in question and the time of data collection.
Methods
A 29-year-old female subject, who was diagnosed with Asperger syndrome and had chronic suicidality, collected data for 24 timeframes over three months. The data included the 11/11/11 versions of the Sheehan-Suicidality Tracking Scale (S-STS), the Suicidality Modifiers (SM) scale (used in University of Alabama at Birmingham S-STS validation study), and the Alphs Dichotomous Impulsivity and Hopelessness Two Questions (IH Questions) (used in the University of Alabama at Birmingham InterSePT Scale for Suicidal Thinking-Plus [ISST-Plus] validation study).[2–4] Data were collected using the computerized versions of all three scales.[5] In addition, the subject documented any other notes she thought relevant to her scoring choices. The S-STS is an 11-question scale about suicidal phenomena and has one question about non-suicidal self-injury. The SM has five questions on each of two topics: impulsivity and hopelessness. Each of these two scales (S-STS and SM) uses a 0 to 4 (5-point Likert) scoring system with the following descriptive anchors: 0=not at all, 1=a little, 2=moderately, 3=very, 4=extremely.
The two IH Questions have seven descriptive anchors each that were converted into 1 to 7 numeric values: For the question on usual caution/impulsivity, 1=extremely cautious, 2=very cautious, 3=moderately cautious, 4=in the middle, 5=moderately impulsive, 6=very impulsive, and 7=extremely impulsive; and for the question on usual hopefulness/hopelessness, 1=extremely hopeful, 2=very hopeful, 3=moderately hopeful, 4=in the middle, 5=moderately hopeless, 6=very hopeless, and 7=extremely hopeless. During data collection, the subject was under the care of both a therapist and a psychiatrist.
The subject completed the computerized versions of each of the three scales three different times for each timeframe. The first dataset was for the subject alone and was not to be shared (self-version [immediate]). The second dataset, completed just after the first, was for her therapist (self-version [for therapist]). The third dataset was for the subject, but was completed 1 to 5 days following the first. This is referred to as self-version (days later). The subject answered the scales for these three data collection points for each of the 24 timeframes (timeframes varied between 3 to 5 days throughout the study [mean of 3.96 days]).
The results of the reliability study data of these three scales is being prepared in another report.
The three different versions (self-version [immediate], self-version [for therapist], and self-version [days later]) of each of the three scales were compared for each of the same timeframes to determine what deviations occurred in the question answers and notes.
Results
Comparison of self-version (immediate) to self-version (for therapist). For the purpose of this comparison, the self-version (immediate) is used as the standard against which the self-version (for therapist) was judged. On the S-STS, there were six interviews on which the subject failed to report to her therapist any preparatory behaviors at all. Some of these timeframes contained as many as three preparatory behaviors. There were deviations in the counts of both passive and active suicidal ideation. The deviation was as high as 150 and 230 events, respectively, during one timeframe. Deviations occurred in the severity and type of suicidal planning during in as many as 20 of the 24 timeframes.
On the SM, the hopelessness questions deviated in 22 of the24 timeframes. There were 17 aggregate point deviations in the seriousness of the loss of desire to resist impulsivity across all timeframes.
The IH Questions only deviated on the level of usual hopefulness/ hopelessness question, which occurred in seven timeframes. Overall, there was deviation on 41 percent of all scale questions across all timeframes (302 out of a total of 744 questions).
The deviation in the note topics was minimal compared to the individual scale questions. However, the subject did not report some of the note topics to the therapist that related to the severity of her symptoms, to details about planning an attempt (including possible dates), and notes that directly related to the therapist that were noted in the self-version (immediate). These deviations occurred in five, 10, and three timeframes, respectively.
Comparison of self-version (immediate) to self-version (days later). For the purpose of this comparison, the self-version (immediate) is used as the standard against which the self-version (days later) was judged. On the S-STS, there were deviations in the counts of active suicidal ideation (320 events across all 24 timeframes). The mean deviation, when there was deviation on this question, was higher here (107 events) than it was comparing the self-version (immediate) to the self-version (for therapist) (89 events). When deviations occurred, almost all questions only deviated by a mean of 1 point of seriousness.
On the SM, the mean deviation of any question, when any deviation occurred on the question, was only 1 point of seriousness. On the IH Questions, there was no deviation at all.
Overall, there was deviation on five percent of all scale questions across all timeframes (35 out of a total of 744 questions). The deviation in note topics mainly related to the subject’s current severity of suicidality symptoms (4 out of 24 timeframes).
Comparison of self-version (days later) to self-version (for therapist). For the purpose of this comparison, the self-version (days later) is used the standard against which the self-version (for therapist) was judged. On the S-STS, there were deviations in the counts of passive and active suicidal ideation and the count of non-suicidal self-injury (occurring in 18, 21, and 12 timeframes, respectively). There were also deviations in the severity and type of suicidal planning, which occurred in at least 18 of the 24 timeframes. The seriousness of preparatory behaviors deviated an aggregate total of 25 points during only six timeframes.
On the SM, the deviations occurred in all of the hopelessness questions during all 24 interviews.
On the IH Questions, there was no deviation at all in the usual caution/impulsivity question and only minimal deviation in the usual hopefulness/hopelessness question.
Overall, there was deviation on 45 percent of all scale questions across all 24 timeframes (337 out of a total of 744 questions). The subject did not report some of the note topics that were included in self-version (days later) to the therapist that related to the severity of her symptoms, to details about planning an attempt (including possible dates), and that were directly related to the therapist. These deviations occurred in eight, 10, and three timeframes, respectively.
Discussion
This case study illustrates the potential for important information to be missed if a patient is only asked to rate a suicidality scale once for a specific timeframe. It is possible that some suicidal phenomena occur too close to the timeframe in question for a patient to immediately acknowledge the presence, the gravity, or the details associated with those phenomena. Coping often involves acutely minimizing the gravity of the symptoms. In settings where suicidality rating scales are regularly used, it may be helpful for a clinician to routinely ask the patient if there were any details about the patient’s suicidality during earlier timeframes that were not previously shared with the clinician. Doing so may serve as a way for the clinician to gather information that is still clinically relevant and that occurred during a prior timeframe but was not previously reported. Our findings do not support the common assumption that the near-term suicidality scores about the same timeframe decrease as time passes.[1] Indeed, in this individual case, we found that opposite occurred.
There were deviations in five percent of all questions between the self-version (immediate) to the self-version (days later). The total deviations on all questions between the self-version (immediate) and the self-version (for therapist) were much higher at 41 percent of all questions. This suggests it was easier for the subject to be honest with herself than it was for her to be honest with her therapist.
Limitations. The use of only one subject for this case study means these findings may not be generalizable to other cases of suicidality. Knowing that the collected data would be given to someone else may have biased the subject’s reporting of events.
Conclusion
This case study illustrates that a subject’s reporting of symptoms of suicidality using a patient-rated scale can vary depending upon the context, distance from timeframe in question, and the subject’s relationship with the clinician, rater, or reviewer of the data. It is possible similar issues relate to clinician-rated suicidality scales.
References
1. Sundin EC, Horowitz MJ. Horowitz’s Impact of Event Scale evaluation of 20 years of use. Psychosomat Med. 2003;65(5):870–876.
2. Sheehan DV, Alphs L, Mao L, et al Comparative validation of the S-STS, the ISST-Plus, and the C-SSRS for assessing the suicidal thinking and behavior FDA 2012 suicidality categories. Innov Clin Neurosci. 2014;11(9–10):32–46.
3. Alphs L (personal communication). Suicidality Modifiers (SM) were developed by Sheehan DV, Alphs L, and Giddens JM for the study “Comparative validation of the ISST-Plus, the S-STS, and the C-SSRS for assessing suicidal thinking and behavior,” which was presented as a poster presentation at the 14th International Congress on Schizophrenia Research (ICOSR). Orlando, FL: April 21–25, 2013.
4. Alphs L (personal communication). Two dichotomized spectrum test questions (one assessing impulsivity/caution dichotomous spectrum, the second using a hopefulness/hopelessness dichotomous spectrum) were developed for “Comparative validation of the ISST-Plus, the S-STS, and the C-SSRS for assessing suicidal thinking and behavior,” which was presented as a poster presentation at the 14th International Congress on Schizophrenia Research (ICOSR). Orlando, FL: April 21–25, 2013.
5. Dolphin Electronic Data Capture (eMINI Professional Version 2.1.1 / R131112.1 Database Version 2.26) [Software]. (1994–2012). http://medical-outcomes.com.
