Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Apr 17.
Published in final edited form as: Community Ment Health J. 2023 Jun 20;59(8):1639–1646. doi: 10.1007/s10597-023-01155-5

Suicide Safety Plan Self-knowledge in Serious Mental Illness: Psychiatric Symptom Correlates and Effects of Brief Intervention

Emma M Parrish 1, Avery Quynh 2, Vanessa Scott 2, Samantha A Chalker 2,3, Cindy Chang 2,3, Snigdha Kamarsu 4, Elizabeth W Twamley 2,3, Colin A Depp 2,3
PMCID: PMC13085597  NIHMSID: NIHMS2159511  PMID: 37340170

Abstract

Psychiatric symptoms experienced by people with serious mental illness (SMI; schizophrenia, bipolar and other psychotic disorders) may impact suicide safety planning. This study examined safety plan self-knowledge, or an individual’s own knowledge and awareness of their safety plan, in a sample of people with SMI. Participants (N = 53) with SMI at elevated risk of suicide completed a 4-session intervention that included safety plans, with one intervention group that included mobile augmentation. Self-knowledge was assessed from previous safety plans at 4-, 12- and 24-weeks. Fewer warning signs generated was correlated with greater psychiatric symptoms (r = −.306, p = .026) and suicidal ideation (r = −.298, p = .030). Fewer coping strategies generated was correlated with greater suicidal ideation (r = .−323, p = .018). Preliminarily, there was greater self-knowledge of warning signs over time among participants in the mobile intervention. These preliminary findings highlight the relationship of safety plan self-knowledge to symptoms and suggest that mobile augmentation of safety planning may be beneficial. Trial Registration: NCT03198364.

Keywords: Serious mental illness, Safety planning, Schizophrenia, Bipolar disorder, Suicide prevention

Introduction

People with serious mental illness (SMI; i.e., schizophrenia spectrum, bipolar disorder with or without psychotic features, and other psychotic disorders) have a higher rate of suicidal ideation and behavior compared to the general population (Aydın et al., 2019; Chen & Dilsaver, 1996; Fu et al., 2021). Indeed, suicidal behaviors are approximately 20–50 times higher in people with schizophrenia as compared to the general population (Cassidy et al., 2018). It is estimated that 1 in 20 people with schizophrenia (Hor & Taylor, 2010) and 1 in 5 people with bipolar disorder die by suicide (Miller & Black, 2020). Though suicide-focused interventions developed for people with psychosis show promise in an emerging evidence base (Bornheimer et al., 2020), people with SMI are excluded from most trials of suicide prevention interventions (Villa et al., 2020). Moreover, those with SMI are frequently disengaged from care (Dixon et al., 2016), which may further limit access to suicide specific interventions. Thus, there is limited knowledge of how gold-standard suicide prevention interventions, including safety planning interventions, may or may not be effective for people with psychosis.

Safety planning and lethal means restriction, standards of care in suicide prevention, directly focus on intervening on the suicidal crisis, which has been shown to be more effective than indirect interventions (e.g., treating depression; (Calati & Courtet, 2016; Meerwijk et al., 2016). Safety planning interventions vary in their components, but typically consist of identifying personal warning signs and coping strategies, people, and places to decrease isolation and help in a crisis, and lethal means counseling (i.e., means reduction). These safety plans are most effective when the plans are complete, contain appropriate recommendations for warning signs and coping strategies, and are personalized by including specific individualized recommendations (Gamarra et al., 2015). A recent meta-analysis found safety planning-type interventions are effective in reducing suicidal behaviors (Nuij et al., 2021). However, no research has specifically examined safety planning among those with serious mental illness (SMI).

An important step toward understanding how safety planning intervention should or could be adapted to people with SMI is how symptoms (e.g., psychotic symptoms; suicidal ideation) may intersect with aspects of safety planning. There is some evidence of uptake of some suicide prevention components may be less evident in people with SMI, as people with SMI who have high utilization of crisis services do not utilize crisis lifelines at a level commensurate with emergency services or hospitalization (Chalker et al., 2022). Additionally, there is evidence that people with schizophrenia have greater difficulty regulating their emotions while they are concurrently experiencing psychotic symptoms (Strauss et al., 2019). Thus, it is possible that people with SMI with higher levels of psychotic symptoms may have difficulty with engaging in the safety plan creation process, as well as engaging in effective coping strategies in a moment of crisis. It is also possible that greater distress related to suicidal ideation itself may interfere with effective safety plan implementation. For instance, there is evidence that negative affect and distress influences emotion regulation strategy choices among adolescents in healthy samples (Lennarz et al., 2019). Thus, distress related to suicidal ideation severity may interfere with the implementation of momentary emotion regulation strategies, though it is unclear how this relationship would apply to people with SMI. Furthermore, it is possible that psychotic symptoms and/or suicidal ideation may interfere with knowledge of safety plan contents, but this is yet to be investigated. Recall of therapeutic content has previously been evaluated in bipolar disorder(Lee & Harvey, 2015) but not to our knowledge in reference to safety planning.

It is also possible that reminders of safety plan content through mobile augmentation may improve an individual’s own knowledge and awareness of their safety plan (i.e., safety plan self-knowledge), but this has not yet been studied. The current study is a secondary analysis of a clinical trial that included two intervention arms of a brief suicide-specific intervention, with one arm that included a mobile augmentation of the intervention (described elsewhere, Depp et al., 2021, 2022, and below). Participants in this trial were assessed at 4 weeks following the intervention, as well as a 12-week and 24-week follow up.

The goal of the present study was to examine safety plan self-knowledge and related outcomes in this population in the context of a pilot suicide prevention intervention for people with SMI. The specific aims of the present study were to (1) evaluate the association of safety plan self-knowledge with clinical symptoms and suicide ideation severity and (2) understand change in self-knowledge of safety plan elements after different stages of the completion of in-person intervention at 4 weeks, completion of telephone coaching after 12 weeks, and 24-week follow up and across intervention conditions.

Methods

Participants

The Safety and Recovery Therapy (START) study was a pilot trial of a cognitive behavioral therapy intervention, with one arm including a mobile device, to reduce suicide risk among people with SMI (see elsewhere for full description; Depp et al., 2021, 2022). Participants were recruited between 2018 and 2020 from multiple psychiatric walk-in clinics and community mental health centers in San Diego County. Inclusion criteria were as follows: (1) had a primary diagnosis of a schizophrenia spectrum disorder, bipolar disorder (with or without psychotic features), or other psychotic disorder confirmed by the Mini International Neuropsychiatric Interview (Sheehan et al., 1998), (2) had current suicidal ideation (identified by scoring a “2” or above on the Columbia—Suicide Severity Rating Scale in the past month; (Posner et al., 2008) or had a suicide attempt in the three months prior to the initial assessment, and (3) were treated in an outpatient mental health center. Participants were excluded if they: (1) were non-English speaking, (2) were unable to navigate a touch screen due to dexterity or vision problems, (3) had a current diagnosis of a substance use disorder or were intoxicated during the research visit, and (4) had a private or public conservatorship. For this secondary analysis, only participants with safety plan self-knowledge data were included (N = 53) out of 77 participants included in the full trial (smaller sample due to missing data and attrition).

Procedures

Participants were randomly assigned to the therapy and mobile condition (mSTART) or to the therapy condition (START alone) prior to the study inception (full study protocol is described in (Depp et al., 2021). The participants and therapist were not blind to the condition, but study staff who collected data were blinded. After the completion of baseline assessments, participants were scheduled to meet with a study therapist for four sessions, once a week (see (Depp et al., 2021) for full protocol). The therapy was designed to improve self-efficacy in coping with suicidal symptoms by having participants implement crisis-management strategies. During both versions of the START intervention, participants reviewed early warning signs and triggers for a suicidal crisis, as well as coping strategies for these triggers, in an open-ended safety planning approach. The number of strategies each participant discussed with their therapist was not standardized between participants. These strategies were incorporated into a safety plan that also included a list of people participants could call for distraction and/or help in a crisis, crisis lines, and lethal means counseling. Participants in both conditions also received follow up phone contacts from a clinician at weeks 6, 8, and 10 following treatment, or until they were connected with outpatient care. During the mSTART, participants received mobile augmentation of therapy content for 12 weeks following the intervention, including general CBT techniques, current mood, and queries about safety plan elements. Relevant to safety planning, participants were queried once per day via mobile phone if they noticed any of their early warning signs, and if so, whether they engaged in a skill or coping strategy (e.g., thought challenge, behavioral coping strategy) derived in person meetings. Mobile augmentation was personalized to the participant’s unique safety plan. The assessments were repeated after the therapy period at 4 weeks after the baseline, 12 weeks after, and 24 weeks after.

Measures

Sociodemographic Characteristics

All participants were assessed at baseline for basic sociodemographic information (which included age, sex, race/ethnicity, education, and marital status), and employment history.

Repeated Assessments

Participants completed assessments of the following constructs during baseline and the three follow-up assessment periods: severity of positive, negative, and general psychopathology using the Brief Psychiatric Rating Scale (BPRS; (Overall & Gorham, 1962), current suicide severity via the Scale of Suicide Ideation (SSI; (Beck et al., 1979) total score, and use of coping behaviors with the Coping Self-Efficacy Scale (CSE; (Chesney et al., 2006). Higher scores on all measures reflect greater levels of the construct. Lifetime history of suicide attempts were assessed using the Columbia-Suicide Severity Rating Scale (C-SSRS; (Posner et al., 2008).

Safety and Recovery Plan Self-Knowledge

When participants came in for their 4-, 12-, and 24-week assessments, they were asked to freely list up to seven warning signs and triggers and up to seven coping strategies that they had discussed in therapy. Original safety plan information was not available, so only safety plan self-knowledge data is utilized in the current study. See Supplementary Material for full measure text.

Qualitative Data Coding

First, the number of warning signs and coping strategies listed were counted. Next, word count was calculated as a secondary measure of safety plan self-knowledge using Microsoft Word’s word count calculator. Both steps were performed by two independent raters (A.Q. and V.S.) to ensure accuracy. Qualitative coding was based on an existing measure, the Safety Planning Intervention Scoring Algorithm for VA (SPISA-VA), a measure utilized by the U.S. Department of Veterans Affairs to evaluate safety plans (Brown et al., 2019). The brief version of this scale (SPISA-Brief) has been utilized in other studies of safety plan quality and completeness (Goodman et al., 2020). We modified this scale to fit the safety plan self-knowledge measure, which only included warning signs and coping strategies. Generated safety plan components were coded for two characteristics: personalization and appropriateness. Personalization was defined as how specific a component was (e.g., “reading” would not be considered a personalized coping strategy and would be coded as “0”, whereas “reading a book in my favorite fantasy series” would be considered personalized and coded as “1”). Appropriateness reflected whether a warning sign or coping strategy was deemed to be relevant, helpful, and suitable for a safety plan (e.g., “work harder” would not be considered an appropriate coping strategy and would be coded as “0”, whereas “go for a walk around my neighborhood” would be considered appropriate and would be coded as “1”).

Overall, the dataset yielded 141 unique safety plan self-knowledge assessments from 53 participants. Of these 141 unique safety plan self-knowledge measures, 6 participants did not understand the task of the safety plan self-knowledge assignment on one of their safety plan self-knowledge forms (i.e., answers that were clearly irrelevant) and these individual forms were excluded from analysis (n = 135 for final analysis).

Statistical Analyses

Two independent raters (A.Q. and V.S.) coded all generated warning signs and coping strategies for a total of 2676 individual data points across all timepoints (IRR = 84.9%).

Safety plan self-knowledge data were aggregated by participant so that each participant received a mean number of warning signs and coping strategies generated, as well as a mean word count, across all three timepoints (4-week, 12-week, and 24-week), in order to account for restricted range at each timepoint. Then, descriptive statistics were calculated to characterize (1) the amount of safety plan self-knowledge (i.e., mean number and word count of generated warning signs and coping strategies) and (2) appropriateness and personalization of safety plan self-knowledge (i.e., number of appropriate warning signs generated, number of personalized warning signs generated, number of appropriate coping strategies generated, and number of personalized coping strategies generated). An independent samples t-test was conducted to determine if amount, appropriateness, and personalization of safety plan self-knowledge differed between participants with a schizophrenia spectrum disorder and those with a bipolar disorder or other psychotic disorder. To determine if safety plan self-knowledge correlated with clinical symptoms, Pearson correlations were performed with these items and the three outcome measures: BPRS, SSI, and CSE. Spearman correlations were performed with the word count of warning signs and coping strategies, and the BPRS, SSI, and CSE, due to a non-normal distribution of the word count data. Following this, a linear mixed model was performed to understand how safety plan self-knowledge changed over time (4-week, 12-week, and 24-week). Finally, an interaction term of intervention group (i.e., START vs. mSTART) x time was included in these linear mixed models to investigate if the intervention group had an impact on self-knowledge, and an effect size of this interaction was calculated (Morris, 2008). Of note, original safety plan data was not available for analysis, so this study only reports findings related to safety plan self-knowledge data.

Results

Demographic Information

See Table 1 for full demographic information. Participants had a mean age of 45.9 (SD = 12.4) and were 52.8% female. The participants identified their races as 62.3% White, 28.3% African American, 5.7% American Indian/Alaska Native, and 3.8% Asian. Additionally, 35.8% of participants identified as Hispanic or Latino. The mean years of education for this sample was 11.8 (SD = 1.2; range 8–18). Most participants (83.0%) were not currently employed. Participants were also experiencing high levels of current suicidal ideation, with a mean SSI of 14.6 (SD = 7.2). Furthermore, participants had a mean of 4.6 lifetime suicide attempts (SD = 10.2), ranging from 0 to 57 suicide attempts.

Table 1.

Demographics and clinical characteristics (N = 53)

M(SD), Range or N, %
Age 45.9 (12.4), 21–65
Gender 28, 52.8% Female
25, 47.2% Male
Race 33, 62.3% White
15, 28.3% African American
3, 5.7% American Indian/Alaska Native
2, 3.8% Asian
Ethnicity 19, 35.8% Hispanic or Latino
Years of education 11.8 (1.2), 8–18
Employment status 44, 83.0% Not Employed
9, 17.0% Employed
Diagnosis 22, 41.5% Schizoaffective Disorder
18, 34.0% Bipolar Disorder
10, 18.9% Schizophrenia
3, 5.7% Other Psychotic Disorder
BPRS mean 61.2 (12.4), 36–95
SSI mean 14.6 (7.2), 1–33
CSE mean 118.5 (48.9), 9–260
Number of lifetime suicide attempts 4.6 (10.2), 0–57
Open in a new tab

BPRS Brief Psychiatric Rating Scale, SSI scale of suicidal ideation, CSE coping self efficacy

Descriptive Information

See Table 2 for full safety plan self-knowledge descriptive information. 16 participants noted that they did not remember any aspect of their safety plan on 26 safety plan self-knowledge measures. Across all participants and timepoints, the mean number of Warning Signs generated was 2.5 (SD = 1.6) with an average word count of 7.5 (SD = 7.8). Regarding appropriateness and personalization of warning signs, on average, participants generated 2.0 (SD = 1.5) appropriate warning signs, and 1.3 (SD = 1.4) personalized warning signs across all timepoints. Across all participants and timepoints, the mean number of Coping Strategies generated was 2.6 (SD = 1.7) with an average word count of 13.0 (SD = 13.3). On average, participants generated 2.2 (SD = 1.7) appropriate coping strategies and 1.9 (SD = 1.7) personalized coping strategies across all timepoints. An independent samples t-test revealed that Warning Sign and Coping Strategy number, word count, appropriateness, and personalization did not significantly differ between participants with a schizophrenia spectrum disorder and other diagnoses (all p’s > 0.119).

Table 2.

Descriptive statistics: self-knowledge of warning signs and coping strategies

M (SD) Range
Warning signs
 Number generated (max = 7) 2.5 (1.6) 0–7
 Word count 7.5 (7.8) 1–52
 Appropriateness 2.0 (1.5) 0–5.7
 Personalization 1.3 (1.4) 0–4.3
Coping strategies
 Number generated (max = 7) 2.6 (1.7) 0–7
 Word count 13.0 (13.3) 0–69
 Appropriateness 2.2 (1.7) 0–6.0
 Personalization 1.9 (1.7) 0–6.0
Open in a new tab

Correlations of Safety Plan Self-Knowledge with Clinical Measures

See Table 3 for correlational outcomes.

Table 3.

Correlations of symptom measures, warning sign, and coping strategy (N = 53)

r or ρ (p)
BPRS SSI CSE
Warning sign generated counta − 0.31 (0.026)* − 0.30 (0.030)* 0.05 (0.720)
Coping strategy generated counta − 0.27 (0.053) − 0.32 (0.018)* 0.05 (0.705)
Warning sign generated average word countb − 0.25 (0.076) − 0.30 (0.030)* 0.07 (0.626)
Coping strategy generated average word countb − 0.27 (0.040)* − 0.31 (0.025)* − 0.04 (0.757)
Open in a new tab

BPRS Brief Psychiatric Rating Scale, SSI scale of suicidal ideation, CSE coping self efficacy.

*

Indicates significant at p < .05

**

Indicates significant at p < .01.

a

Indicates Pearson correlation.

b

Indicates Spearman correlation

Number of Safety Plan Components Generated

Regardless of appropriateness or personalization, generating fewer warning signs was correlated with higher psychiatric symptoms on the BPRS (r = −.31, p = .026) and higher suicide symptom severity on the SSI (r = −.30, p = .030). Similarly, regardless of appropriateness or personalization, fewer generated coping strategies was correlated with higher scores on the SSI (r = .−32, p = .018). There was no significant correlation between coping skills on the CSE and number of warning signs (r = .05, p = .720) or coping strategies (r = .05, p = .705) recalled.

Word Count of Safety Plan Component Self-knowledge

Spearman correlations examining correlations of symptom measures with word count revealed a similar picture. A lower word count for coping strategies generated was related to higher scores on the BPRS (rho = −0.29, p = .040). A lower word count for warning signs generated related to higher severity of suicidal ideation on the SSI (rho = −0.30, p = .030), and a lower word count for coping strategies related to higher severity of suicidal ideation on the SSI (rho = −0.31, p = .025). There were no significant correlations with the CSE with word count of warning signs (rho = 0.07, p = .626) or coping strategies (rho = −0.04, p = .757) generated.

Change in Self-knowledge Over Time and Intervention Effects

The number of warning signs generated did not significantly change over the three timepoints, F(88.8) = 0.10, p = .906. The number of coping strategies generated also did not significantly change over the three timepoints, F(135) = 0.40, p = .674. However, there was a moderate effect of an interaction between timepoint and group, with participants in the mSTART condition generating more warning signs over time than participants in the START alone condition, F(88.8) = 2.52, p = .086, with a moderate effect size from 4 to 24 weeks (effect size = 0.47). This interaction was not significant when examining self-knowledge of coping strategies over time, F(135) = 0.25, p = .781, with a small effect size from 4 to 24 weeks (effect size = 0.13).

Discussion

The current study examined whether aspects of safety plan self-knowledge were associated with psychiatric symptoms, self-efficacy, and suicidal ideation among individuals with SMI. This study contributes applied knowledge for how to interpret the degree to which an individual can generate elements of their safety plan and how much they elaborate on warning signs and coping strategies. In general, participants’ generated warning signs and coping strategies were appropriate or personalized, though some warning signs or coping strategies listed were inappropriate, ineffective, or not specific to the individual. We also found that individuals who generated fewer warning signs and coping strategies, as well as fewer words for each, tended to have more severe psychiatric symptoms and suicidal ideation. Additionally, there was a moderate effect of intervention condition, with the intervention group that included a mobile component demonstrating lower self-knowledge declines.

First, greater psychiatric symptoms were associated with fewer generated warning signs, lower word count for warning signs, and lower word count for coping strategies. It is possible that greater levels of psychiatric symptoms make it more difficult for individuals to list their warning signs and coping strategies, which is especially concerning considering that individuals with SMI experiencing high levels of symptoms are likely to face structural barriers that contribute to treatment dropout (Benjet et al., 2022). Because we did not have participants’ original safety plans to compare to, it is difficult to identify whether fewer warning signs generated represents recall or simply that the individual had generated fewer on their original safety plan. Thus, it is also possible that individuals with more severe psychiatric symptoms generated fewer warning signs and coping strategies on their original safety plans and in less detail, suggesting lower engagement in initial safety planning process. Interestingly, number of coping strategies generated was not associated with psychiatric symptoms, suggesting that those with more severe symptoms generate the same number of strategies but may list them in less detail. Nevertheless, these findings suggests that individuals with more severe psychiatric symptoms, who are already at higher risk for suicide (Cassidy et al., 2018; Hawton et al., 2005; Skodlar et al., 2008), may experience barriers to engaging in and/or generating their safety plan in detail.

With respect to suicidal ideation, fewer warning signs, lower word count for warning signs, fewer coping strategies, and lower word count for coping strategies were all significantly associated with greater suicidal ideation. This finding aligns with past research demonstrating the association between safety planning intervention and decreased suicide risk (Ferguson et al., 2022), extending this finding to safety plan self-knowledge. Lower self-knowledge of one’s safety plan may prevent individuals from using the safety plan to remember their warning signs and use coping skills. It is also possible that individuals generating fewer components of their safety plan and having lower word count may have simply had less content on the warning signs section of their original safety plan. Conversely, individuals with more severe suicidal ideation may struggle more with generating warning signs and coping strategies. This possibility is concerning as it suggests that individuals with more severe suicidal ideation may not be generating or elaborating on aspects of their safety plan, thus making it difficult for them to identify when they need to use coping skills and subsequently implement them.

Notably, self-efficacy was not significantly associated with number of warning signs, number of coping strategies, word count for warning signs, or word count for coping strategies. It may be that because the scale assessed coping with problems in general, it is possible that it does not capture self-efficacy related to coping with suicidal thoughts. Another possibility relates to introspective accuracy, or one’s ability to accurately appraise their ability to complete cognitive and functional tasks (Mervis et al., 2022). Introspective accuracy is impaired in individuals with SMI such that people tend to overestimate their own functioning in several domains (Mervis et al., 2022; Silberstein & Harvey, 2019). It may be possible that people with SMI overestimate their ability to cope relative to their own safety plan self-knowledge. For example, a client may report that their self-efficacy is high in terms of calling a suicide crisis line, but they do not follow through with this behavior in the moment. Future research is needed to better understand whether self-efficacy for coping with suicidal thoughts is related to self-knowledge and elaboration of warning signs and coping strategies, and whether inaccurate appraisals of one’s own abilities are implicated in this self-knowledge process.

Self-knowledge of safety plan components did not significantly change over time. Although we did not find significant differences between the intervention conditions in safety plan self-knowledge, there was a signal for lower self-knowledge declines in the mSTART group. It is possible that mobile devices may be useful for increasing exposure to safety plan elements and interventions, however, the current study is underpowered to detect a strong intervention effect and it is unclear why this effect is not present for coping strategies.

Our study has several key limitations. A major limitation is that our study did not assess for number of warning signs and coping strategies when the safety plan was created. Because this information is not available, lower self-knowledge may be a function of less content on the original safety plan; therefore, it is not possible to determine whether these variables reflect problems with self-knowledge or with initial safety plan creation. Additionally, we did not include measures additional variables that may be related to safety plan self-knowledge, such as co-occurring mental health issues, cognitive impairment, and objective indicators of capacity to manage crises. In addition to these variables, study therapist style in delivering the safety plan intervention may have influenced safety plan self-knowledge. While fidelity measures assessed whether a safety plan was completed, these fidelity measures did not assess safety plan details or degree of personalization. Furthermore, our study’s definition of SMI may be consistent with some definitions of SMI, but not all, including some broader definitions that include personality disorders or other serious mood disorders (Martínez-Martínez et al., 2020). Our study is also limited in that it does not have a non-SMI comparison group, but it is possible that safety plan self-knowledge may differ between people with SMI and other high risk psychiatric groups.

Future research should control for number of warning signs and coping strategies, and their respective word counts, at initial safety plan creation. Additionally, future studies should also examine these questions while individuals are in crisis to accurately measure their ability to generate their safety plan while it is needed. Furthermore, given the presence of cognitive impairments in SMI, future research should directly assess the impact of cognitive impairment on safety plan self-knowledge. Given the complex skills needed to free recall and verbally communicate aspects of the safety plan, future work could use different methods of assessing self-knowledge (e.g., behavioral observation, written response, comparing recalled safety plans to original safety plans) to disentangle memory versus verbal communication difficulties. Additionally, memory for therapeutic content has been previously studied in bipolar disorder (Lee & Harvey, 2015), and similar methods may be expanded to suicide research and approaches to measure memory for safety plans. Moreover, the potential impact of demographic factors on safety plan self-knowledge should be investigated. Finally, future research should investigate if clinician style and delivery of the safety planning intervention impacts safety plan self-knowledge, personalization, and appropriateness.

Although preliminary, these findings may have several clinical implications. First, they provide support for the clinical utility of assessing how many warning signs and coping strategies an individual generates, as well as the degree to which they elaborate on each warning sign or coping strategy. Second, clinicians may consider the degree to which psychiatric symptoms are affecting an individual’s ability to fully engage in safety planning as well as implement the safety plan effectively, and they may wish to use cognitive interventions or learning strategies to increase likelihood of remembering the safety plan. Given evidence that elaborative rehearsal improves episodic long-term memory (Oberauer, 2019), strategies that facilitate implementation of the safety plan (e.g., taking a picture of it on a cellphone, leaving a copy where an individual frequently looks) may have positive effects (Oberauer, 2019). Clinicians may want to pay particular attention to ensuring clients remember their warning signs. The creators of the safety plan describe recognition of warning signs as the first step to developing the safety plan, considering that addressing a problem before it fully emerges is one of the most effective ways of averting a suicidal crisis (Stanley & Brown, 2012). Additionally, clinicians may also be able to help normalize how common it is to forget elements of a safety plan, as well as promote the use of compensatory strategies in relationship to safety planning. Third, it is possible that the use of mobile devices to provide reminders of safety plan elements may aide in safety plan self-knowledge, and these reminders may contribute to over-learning of these strategies.

Supplementary Material

Supplementary material

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s10597-023-01155-5.

Acknowledgements

Dr. Twamley gratefully acknowledges the support of a VA Rehabilitation Research and Development Research Career Scientist Award. Dr. Chalker acknowledges support by CDA Award Number IK2RX004239–01 from the United States (U.S.) Department of Veterans Affairs; Rehabilitation Research and Development Service.

References

  1. Aydın M, İlhan B, Tekdemir R, Çokünlü Y, Erbasan V, & Altınbaş K (2019). Suicide attempts and related factors in schizophrenia patients. Saudi Medical Journal, 40(5), 475. 10.15537/SMJ.2019.5.24153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Benjet C, Borges G, Orozco R, Aguilar-Gaxiola S, Andrade LH, Cia A, Hwang I, Kessler RC, Piazza M, Posada-Villa J, Sampson N, Stagnaro JC, Torres Y, Viana MC, Vigo D, & Medina-Mora ME (2022). Dropout from treatment for mental disorders in six countries of the Americas: A regional report from the World Mental Health Surveys. Journal of Affective Disorders, 303, 168–179. [DOI] [PubMed] [Google Scholar]
  3. Beck AT, Kovacs M, & Weissman A (1979). Assessment of suicidal intention: The scale for suicide ideation. Journal of Consulting, 47(2), 343–352. [Google Scholar]
  4. Brown G, Stanley B, & Batdorf W (2019). Safety planning intervention: Consultant training workshop. Advanced safety planning intervention training program. Office of Mental Health and Suicide Prevention Department of Veterans Affairs. [Google Scholar]
  5. Bornheimer LA, Zhang A, Li J, Hiller M, & Tarrier N (2020). Effectiveness of suicide-focused psychosocial interventions in psychosis: A systematic review and meta-analysis. Psychiatric Services, 71(8), 829–838. 10.1176/appi.ps.201900487. [DOI] [PubMed] [Google Scholar]
  6. Calati R, & Courtet P (2016). Is psychotherapy effective for reducing suicide attempt and non-suicidal self-injury rates? Meta-analysis and meta-regression of literature data. Journal of Psychiatric Research. 10.1016/j.jpsychires.2016.04.003 [DOI] [Google Scholar]
  7. Cassidy RM, Yang F, Kapczinski F, & Passos IC (2018). Risk factors for suicidality in patients with schizophrenia: A systematic review, meta-analysis, and meta-regression of 96 studies. Schizophrenia Bulletin, 44(4), 787–797. 10.1093/SCHBUL/SBX131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chalker SA, Parrish EM, Ceren M, Depp CS, Ilgen CA, Goodman MA, Twamley M, E. W., & Doran N (2022). Crisis service utilization following completion of a suicide safety plan for veterans with and without affective and nonaffective psychosis. Journal of Psychiatric Research, 154, 219–223. 10.1016/j.jpsychires.2022.07.034. [DOI] [PubMed] [Google Scholar]
  9. Chen YW, & Dilsaver SC (1996). Lifetime rates of suicide attempts among subjects with bipolar and unipolar disorders relative to subjects with other axis I disorders. Biological Psychiatry, 39(10), 896–899. 10.1016/0006-3223(95)00295-2. [DOI] [PubMed] [Google Scholar]
  10. Chesney MA, Neilands TB, Chambers DB, Taylor JM, & Folkman S (2006). A validity and reliability study of the coping self-efficacy scale. British Journal of Health Psychology, 11(3), 421–437. 10.1348/135910705X53155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Depp C, Ehret B, Villa J, Perivoliotis D, & Granholm E (2021). A brief mobile-augmented suicide prevention intervention for people with psychotic disorders in transition from acute to ongoing care: Protocol for a pilot trial. JMIR Research Protocols, 10(2), e14378. 10.2196/14378 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Depp C, Parrish E, Chalke S, Ehret B, Kamarsu S, Perivoliotis D, & Granholm E (2022). Pilot feasibility trial of a brief mobile augmented suicide prevention intervention for serious mental illness. Psychiatric Rehabilitation Journal, 46, 74. [Google Scholar]
  13. Dixon LB, Holoshitz Y, & Nossel I (2016). Treatment engagement of individuals experiencing mental illness: Review and update. World Pscyhiatry, 15(1), 13–20. 10.1002/wps.20306. [DOI] [Google Scholar]
  14. Ferguson M, Rhodes K, Loughhead M, McIntyre H, & Procter N (2022). The effectiveness of the safety planning intervention for adults experiencing suicide-related distress: A systematic review. Archives of Suicide Research, 26(3), 1022–1045. 10.1080/13811118.2021.1915217. [DOI] [PubMed] [Google Scholar]
  15. Fu XL, Qian Y, Jin XH, Yu HR, Wu H, Du L, Chen HL, & Shi YQ (2021). Suicide rates among people with serious mental illness: A systematic review and meta-analysis. Psychological Medicine. 10.1017/S0033291721001549 [DOI] [Google Scholar]
  16. Gamarra JM, Luciano MT, Gradus JL, & Stirman SW (2015). Assessing variability and implementation fidelity of suicide prevention safety planning in a regional VA Healthcare System. Crisis, 36(6), 433–439. 10.1027/0227-5910/a000345. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Goodman M, Brown GK, Galfalvy HC, Spears AP, Sullivan SR, Kapil-Pair KN, Jager-Hyman S, Dixon L, Thase ME, & Stanley B (2020). Group (“Project Life Force”) versus individual suicide safety planning: A randomized clinical trial. Contemporary Clinical Trials Communications, 17, 100520. 10.1016/j.conctc.2020.100520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hawton K, Sutton L, Haw C, Sinclair J, & Deeks JJ (2005). Schizophrenia and suicide: Systematic review of risk factors. British Journal of Psychiatry, 187(01), 9–20. 10.1192/bjp.187.1.9. [DOI] [Google Scholar]
  19. Hor K, & Taylor M (2010). Suicide and schizophrenia: A systematic review of rates and risk factors. Journal of Psychopharmacology, 24(4_suppl), 81–90. http://journals.sagepub.com/doi/abs/10.1177/1359786810385490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lee JY, & Harvey AG (2015). Memory for therapy in bipolar disorder and comorbid insomnia. Journal of Consulting and Clinical Psychology, 83(1), 92–102. 10.1037/a0037911. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lennarz HK, Hollenstein T, Lichtwarck-Aschoff A, Kuntsche E, & Granic I (2019). Emotion regulation in action: Use, selection, and success of emotion regulation in adolescents’ daily lives. International Journal of Behavioral Development, 43(1), 1–11. 10.1177/0165025418755540. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Martínez-Martínez C, Richart-Martínez M, & Ramos-Pichardo JD (2020). Operational definition of serious mental illness: Heterogeneity in a review of the research on quality-of-life interventions. Journal of the American Psychiatric Nurses Association, 26(3), 229–244. 10.1177/1078390320902823. [DOI] [PubMed] [Google Scholar]
  23. Meerwijk EL, Parekh A, Oquendo MA, Allen IE, Franck LS, & Lee KA (2016). Direct versus indirect psychosocial and behavioural interventions to prevent suicide and suicide attempts: A systematic review and meta-analysis. The Lancet Psychiatry. 10.1016/S2215-0366(16)00064-X [DOI] [Google Scholar]
  24. Mervis JE, Vohs JL, & Lysaker PH (2022). An update on clinical insight, cognitive insight, and introspective accuracy in schizophrenia-spectrum disorders: Symptoms, cognition, and treatment. Expert Review of Neurotherapeutics, 22(3), 245–255. 10.1080/14737175.2022.2049757. [DOI] [PubMed] [Google Scholar]
  25. Miller JN, & Black DW (2020). Bipolar disorder and suicide: A review. Current Psychiatry Reports, 22(2), 6. 10.1007/s11920-020-1130-0. [DOI] [PubMed] [Google Scholar]
  26. Morris SB (2008). Estimating Effect Sizes from Pretest-Posttest-Control Group designs. Organizational Research Methods, 11(2), 364–386. 10.1177/1094428106291059. [DOI] [Google Scholar]
  27. Nuij C, van Ballegooijen W, de Beurs D, Juniar D, Erlangsen A, Portzky G, O’Connor RC, Smit JH, Kerkhof A, & Riper H (2021). Safety planning-type interventions for suicide prevention: Meta-analysis. The British Journal of Psychiatry. 10.1192/bjp.2021.50 [DOI] [Google Scholar]
  28. Oberauer K (2019). Is rehearsal an effective maintenance strategy for working memory? Trends in Cognitive Sciences, 23(9), 798–809. 10.1016/j.tics.2019.06.002. [DOI] [PubMed] [Google Scholar]
  29. Overall J, & Gorham D (1962). The brief psychiatric rating scale. Pscyhological Reports, 10(3), 799–812. [Google Scholar]
  30. Posner K, Brent D, Lucas C, Gould M, Stanley B, Brown G, Fisher P, Zelazny J, Burke A, Oquendo MJ, & Mann J (2008). Columbia-suicide severity rating scale (C-SSRS). Columbia University Medical Center. [Google Scholar]
  31. Sheehan D, Lecrubier Y, Sheehan KH, Amorim P, Janvas J, Weiller E, Hergueta T, Baker R, & Dunbar GC (1998). The mini-international neuropsychiatric interview (MINI): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry, 59(20), 22–33. [Google Scholar]
  32. Silberstein J, & Harvey PD (2019). Impaired introspective accuracy in schizophrenia: An independent predictor of functional outcomes. Cognitive Neuropsychiatry, 24(1), 28–39. 10.1080/13546805.2018.1549985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Skodlar B, Tomori M, & Parnas J (2008). Subjective experience and suicidal ideation in schizophrenia. Comprehensive Psychiatry, 49(5), 482–488. 10.1016/j.comppsych.2008.02.008. [DOI] [PubMed] [Google Scholar]
  34. Stanley B, & Brown GK (2012). Safety planning intervention: A brief intervention to mitigate suicide risk. Cognitive and Behavioral Practice, 19(2), 256–264. 10.1016/j.cbpra.2011.01.001. [DOI] [Google Scholar]
  35. Strauss GP, Zamani Esfahlani F, Visser KF, Dickinson EK, Gruber J, & Sayama H (2019). Mathematically modeling emotion regulation abnormalities during psychotic experiences in schizophrenia. Clinical Psychological Science, 7(2), 216–233. 10.1177/2167702618810233. [DOI] [Google Scholar]
  36. Villa J, Ehret BC, & Depp CA (2020). Systematic review of the inclusion of people with psychosis in suicide-specific clinical trials. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 41(3), 233–236. 10.1027/0227-5910/a000628. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material
NIHMS2159511-supplement-Supplementary_material.docx (13.6KB, docx)

RESOURCES