Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jul 26.
Published in final edited form as: Int J Eat Disord. 2025 Dec 29;59(4):744–752. doi: 10.1002/eat.70022

Suicidal Ideation in Adult Women: The Unique Roles of Bingeing, Purging, and Restricting

Holly K Spinner 1, Sara Prostko 2, Thomas E Joiner 3, Pamela K Keel 4,*
PMCID: PMC13401432  NIHMSID: NIHMS2197162  PMID: 41466532

Abstract

Objective:

Suicidal ideation represents a serious experience common in individuals with eating disorders. To inform screening and clinical assessment and advance theoretical work, the current study compared clinical characteristics between adult women with and without current suicidal ideation and examined whether bingeing, fasting, and purging behaviors are uniquely associated with suicidal ideation above and beyond existing psychiatric diagnoses.

Method:

Data come from four studies conducted between 2000 and 2023 (N=732) that evaluated current suicidal ideation in all participants. Participants completed diagnostic interviews assessing lifetime psychiatric diagnoses and current eating disorder diagnoses and behavioral symptom frequencies. 553 participants had DSM-5 eating disorders (54.6% bulimia nervosa, 15.0% purging disorder (PD), 28.2% other specified (excluding PD), 1.4% anorexia nervosa, and 0.7% binge-eating disorder).

Results:

Individuals with current suicidal ideation had a greater prevalence of most psychiatric diagnoses and greater frequency of bingeing, fasting, and purging. Additionally, purging frequency was associated with current suicidal ideation independent of bingeing, fasting, and lifetime psychiatric comorbidities.

Discussion:

Results support prior work suggesting that purging may fall on the self-harm continuum with non-suicidal self-injury. Further, findings support screening for purging and suicide risk assessment in individuals with eating disorders.

Keywords: suicidal ideation, eating disorders, purging, bingeing, fasting, comorbidities

Summary:

• Individuals with current suicidal ideation have a greater prevalence of lifetime mood disorders, substance use disorders, obsessive-compulsive disorder, and post-traumatic stress disorder, and current bulimia nervosa and purging disorder.

• More frequent purging is associated with current suicidal ideation above and beyond the frequency of other eating disorder behaviors and lifetime history of other psychiatric disorders.

• Findings underscore the importance of screening for purging and suicide risk assessments in individuals with eating disorders.

Introduction

Suicidal ideation is a major health concern (Jobes & Joiner, 2019), and impacts 25–33% of individuals with an eating disorder (Smith et al., 2018). Suicidal ideation predicts future suicide attempt (Franklin et al., 2017), likely explaining elevated suicide attempts and deaths in individuals with eating disorders compared to the general population (Udo et al., 2019; Smith et al., 2018). Despite this, relatively little empirical attention has been paid to suicidal ideation within eating disorder populations, with prior work focused primarily on suicide attempts or suicide mortality (Smith et al., 2019). Suicide attempts and deaths are a critical public health issue; however, leading suicide researchers declared suicidal ideation to be an essential research target to reduce human suffering (Jobes & Joiner, 2019). Identifying unique associations between specific eating disorder features and suicidal ideation can inform screening and assessments in clinical settings and advance theoretical work explaining elevated suicide risk in eating disorders (Smith et al., 2018; Smith et al., 2019).

Death by suicide occurs in up to 20% of individuals with anorexia nervosa (AN), representing a nearly two-fold increase in cause of premature death (Arcelus et al., 2011). Recent studies support elevated suicide risk in bulimia nervosa (BN), binge eating disorder (BED), and other specified feeding or eating disorder (OSFED) (Duffy et al., 2021; Schmidt et al., 2024; Sohn et al., 2023). Efforts to identify how specific eating disorder symptoms (e.g., bingeing, fasting, and purging) are associated with suicidal ideation have produced mixed findings. Some studies have reported both fasting and purging to be associated with suicidal ideation (Hunt & Levinson, 2021; Shahnaz & Klonsky, 2021). Other studies found only purging to be significantly associated with suicidal ideation (Joiner et al., 2022; Obeid et al., 2020), and Wang and colleagues (2019) found only fasting to be associated with increased likelihood of suicidal ideation. Additionally, one study found that suicidal ideation prevalence increased as eating disorder symptoms increased, regardless of symptom type (Lipson & Sonneville, 2019).

The Interpersonal Theory of Suicide offers insight for the potential association between suicidal ideation and eating disorder behaviors. This theory posits that suicidal ideation arises from the combination of thwarted belongingness (i.e., an unmet need of social connectedness) and perceived burdensomeness (i.e., that one’s death is worth more than one’s life to one’s friends, family, and society) (Joiner, 2005; Van Orden et al., 2010). Research has shown that fasting is associated with thwarted belongingness and perceived burdensomeness potentially due to secrecy and social isolation caused by fasting (Forrest et al., 2016; Kwan et al., 2016). Further, bingeing is associated with elevated perceived burdensomeness potentially due to the associated guilt, shame, low self-esteem, and self-hatred (Forrest et al., 2016; Kwan et al., 2016). Similarly, purging is associated with perceived burdensomeness conceivably through guilt and shame from purging and potentially in-part by increased caregiver burden (Forrest et al., 2016; Sepulveda et al., 2014).

Much of existing research on suicidal ideation and eating disorders has not accounted for influence of frequently comorbid diagnoses, such as mood, anxiety, and substance use disorders, which also significantly predict suicidal ideation (Franklin et al., 2017). Bodell et al. (2013) found that BN but neither AN nor DSM-IV Eating Disorder Not Otherwise Specified was uniquely associated with elevated suicidality, controlling for comorbid disorders. Although findings hint at aspects of clinical presentation that might be most linked to suicidal ideation in eating disorders, the authors did not examine symptom-level associations. Lipson and Sonneville (2019) addressed this gap and found that associations between bingeing and purging and suicidal ideation were no longer significant after controlling for demographic features, anxiety, and depression. A summary of the sample breakdown from prior work can be found in Table 1.

Table 1.

Summary of sample makeup from previous work on eating disorder behaviors and suicidality

Sample Size (N) Participants with EDs (N) ED breakdown (N) Age (M [SD]) Sex (% Female) Clinical vs. Community
Bodell et al. (2013) 364 220 AN = 30
BN = 50
EDNOS = 148		 35 (9) 100 Community
Duffy et al. (2021) 257 257 AN = 146
BN = 45
BED = 16
OSFED = 50		 28.6 (9.2) 91.1 Clinical
Forrest et al. (2016) 98 98 AN = 34
BN = 27
BED = 1
OSFED = 30
UFED = 6		 26.9 (7.9) 100 Clinical
Hunt & Levinson (2021) 181 181 AN = 59
BN = 40
OSFED = 82		 29.8 (10.0) 85.8 Clinical
Joiner et al. (2022) 936 936 AN = 560
BN = 137
BED = 49
OSFED = 111
ARFID = 79		 26.2 (9.6) 87.8 Clinical
Kwan et al. (2016) a 574;
1,791		 -- -- 19.7 (2.6) 60.6 Community
Lipson & Sonneville (2019) b 71,712 -- -- -- 55 Community
Obeid et al. (2020) c 3,575 315 -- 14.4 (1.6) 61 Both
Schmidt et al. (2024) 302 257 AN = 11
BN = 12
BED = 6
OSFED = 114		 20.37 (2.8) 100 Community
Shahnaz & Klonsky (2021) d 387 -- -- 35.6 (10.6) 62.3 Community
Wang et al. (2019) 82 82 AN = 40
OSFED = 26
ARFID = 16		 18.3 (3.2) 100 Clinical
Open in a new tab

Note. Sohn et al. (2022) was a systematic review and meta-analysis and thus study characteristics were not included in the table. Sepulveda et al. (2014) assessed caregivers of individuals with EDs and thus study characteristics were not included in the table. ED = eating disorder, AN = anorexia nervosa, BN = bulimia nervosa, EDNOS = eating disorder not otherwise specified, BED = binge eating disorder, OSFED = other specified feeding or eating disorder, UFED = unspecified feeding or eating disorder, ARFID = avoidant-restrictive food intake disorder.

a

Kwan et al. (2016) included two studies with differing sample sizes. Moreover, ED diagnoses were not reported

b

Lipson & Sonneville (2019) did not report an average age. Rather, they categorized participant by age group (18–22, 23–25, 26–30, 31+). Of participants, 66.7% were 18–22 years old, 12.2% 23–25 years old, 10.3% 26–30 years old, and 10.85% 31+ years old. Moreover, they did not report specific ED diagnoses.

c

Obeid et al. (2020) did not report specific ED diagnoses. Moreover, N = 315 captures clinical participants and some community participants also had an ED, but this N was not reported.

d

Shahnaz & Klonsky (2021) did not report specific ED diagnoses.

The current study aimed to clarify conflicting literature on the influence of eating disorder symptoms on suicidal ideation by (1) assessing differences in clinical characteristics between individuals with and without current suicidal ideation, and (2) examining how bingeing, fasting, and purging behaviors are uniquely associated with suicidal ideation above and beyond existing psychiatric diagnoses in a community sample. For our first aim, we hypothesized that prevalence of eating disorder symptoms and lifetime presence of psychiatric diagnoses will be greater in individuals with current suicidal ideation. For our second aim, we hypothesized that eating disorder behaviors will be associated with current suicidal ideation above and beyond lifetime mental health diagnoses and that purging will explain variance in current suicidal ideation above and beyond bingeing and fasting. Findings have the potential to provide clinical insight as to who is the most at-risk for suicidal ideation, enhancing screening and assessments, and to inform theoretical models of suicidal ideation.

Method

Participants

Participants come from studies focused on binge/purge type eating disorders conducted between 2000 and 2023 that included assessment of current suicidal ideation (Keel et al., 2005, 2008, 2018, 2025). The first three studies had a minimum body mass index (BMI) threshold that precluded AN diagnosis. Across all studies, participants were adult women with a mean (SD) age of 21.30 (3.98) recruited from the surrounding communities using a combination of outreach strategies, including paid advertisements on public transportation, newspaper, radio, billboards, and television, brochures, and posters, and, for later studies, we also used paid advertisements on social media and mass e-mails to students at enrolled at local universities (N=732). Participant demographics are reported in Table 2. Studies 1 and 2 used DSM-IV criteria for eating and other disorders. Proposed and adopted DSM-5 eating disorder criteria were available and utilized beginning with the third study, which used DSM-IV criteria for other disorders. Study 4 used DSM-5 criteria for eating and other psychiatric disorder diagnoses. Importantly, changes that expanded diagnostic criteria along with minimum BMI thresholds for study eligibility, ensured that all participants who met criteria for a DSM-IV eating disorder also met full criteria for the same DSM-5 eating disorder in all four studies. Participants in Study 1 (n=111) were women with BN (n= 39), purging disorder (PD; n=37), and non-eating disorder controls (n=35). In Study 2 (n = 104), participants were women with BN (n=53), PD (n=16), and non-eating disorder controls (n=35) who had not participated in Study 1. In Study 3 (n = 118), participants were women with BN (n=57), PD (n=30), and non-eating disorder controls (n=31). Participants in Study 4 (n = 399) were women with BN (n=153), AN (n=8), BED (n=4), OSFED (n=156) that included atypical AN (n=9), BN of low frequency/limited duration (n=118), BED of low frequency/limited duration (n=3), unspecified feeding or eating disorders (n=26), and non-eating disorder controls (n=78). There were too few cases of AN and BED to include in diagnosis specific analyses, but these cases were included in all transdiagnostic analyses, and all OSFED diagnostic groups in Study 4 were combined into one group labeled OSFED (excluding PD). Study 4 required all participants with eating disorders to endorse DSM-5 binge-eating episodes; therefore, Study 4 included no individuals with PD. Across studies, there were 553 women with current eating disorders including BN (n=302), PD (n=83), and OSFED (excluding PD; n=156). In all studies, participants were excluded if they had a medical condition or other treatment that may influence their appetite, weight, or ability to participate in study procedures. For Study 3, participants were excluded if they had current major depressive disorder or current substance dependence disorder to minimize confounds for that study’s aims. However, there was no restriction on lifetime major depressive disorder or substance use disorder in that study. In all studies, non-eating disorder controls could not report a history of eating disorder symptoms. Studies 2, 3, and 4 required non-eating disorder controls to score below 10 on the Three-Factor Eating Questionnaire (Stunkard & Messick, 1985) cognitive restraint subscale. Studies 3 and 4 further required non-eating disorder controls to deny dietary restriction for the purpose of weight loss on the Eating Disorders Examination (EDE) (Fairburn & Cooper, 1993; Fairburn et al., 2014).

Table 2.

Participant demographics

M (SD) N (%)
Age (years) 21.30 (3.98)
BMI (kg/m2) 23.48 (3.52)
Race
 American Indian 3 (0.4)
 Asian/Pacific Islander 40 (5.5)
 Black/African American 68 (9.3)
 White/Caucasian 581 (79.4)
 More than one race 23 (3.1)
Ethnicity
 Hispanic or Latina 122 (16.7)
Open in a new tab

Note. BMI = body mass index.

Measures

Suicidal Ideation

In Studies 1, 2, and 3 (n=333) current suicidal ideation was assessed using the Beck Depression Inventory (Beck et al., 1961) suicide item, which measures suicidal ideation on a Likert scale in the past week (including today): (0 = I don’t have thoughts of killing myself, 1 = I have thoughts of killing myself, but I would not carry them out, 2 = I would like to kill myself, 3 = I would kill myself if I had the chance). This one-item measure of suicidality has been shown to have both high reliability and construct validity (Joiner et al., 2025). In Study 4 (n=399), the Structured Clinical Interview for DSM-5 Disorders (SCID-5; First et al., 2015) suicidal ideation screener item, “Have you ever wished you were dead or wished you could go to sleep and not wake up? (Tell me about that.)” with the follow-up confirming presence of thoughts of killing the self. If present, this was followed with “Did you have any of these thoughts in the past week (including today)?” which was used to assess the presence of current suicidal ideation. Current suicidal ideation was coded as 1 = present or 0 = not present, with 1 indicating thoughts of killing oneself regardless of current intent.

Eating Disorder Symptoms and Diagnoses

The EDE editions 14.0D (Studies 1–3; Fairburn & Cooper, 1993) and 17.0D (Study 4; Fairburn et al., 2014) were used to assess frequency of eating disorder symptoms and to make eating disorder diagnoses. To assess for frequency of current bingeing, participants were asked to describe any episodes in which they felt they had eaten too much, how much they consumed, over what period of time, in what context, and if they experienced loss of control during these episodes during the past 28 days. Episodes that involved a large amount of food, more than most people would eat under similar contexts, and accompanied by a loss of control were coded as binge eating, and the total number of binge-eating episodes was assessed for the past 28 days. To assess for current fasting, participants were asked about the number of days they fasted for 8 or more waking hours in the past 28 days. This was rated on a scale anchored from 0 = 0 days to 6 = 28 days. Prior research has measured fasting using this item in place of the EDE Restraint subscale to ensure behavior is being assessed as opposed to cognitive restraint, which includes intentions to restrict regardless of actual reduced food intake (Wang et al., 2019). Finally, to assess for frequency of current purging, participants were asked how many times they had made themselves vomit, taken laxatives, and/or taken diuretics as a means of controlling their shape or weight over the past 28 days. Number of episodes over the past 28 days were divided by 4 to report bingeing and purging frequency as episodes per week. Given that excessive exercise has been associated with capability for suicide (Rogers et al., 2019; Smith et al., 2013), suicide attempts (Shahnaz & Klonsky, 2021), and suicidal ideation among individuals with a previous suicide attempt (Laasko et al., 2013), exploratory analyses were conducted to examine compensatory exercise. Good interrater reliability was shown for the EDE across studies (kappas ≥ .72 or intraclass correlations ≥ 0.71; Keel et al., 2005, 2008, 2018, 2025).

Other Psychiatric Diagnoses

In Studies 1, 2, and 3, psychiatric diagnoses were made using the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I; First et al., 1995). In Study 4, the SCID-5 (First et al., 2015) was administered. Diagnoses assessed included mood disorders, anxiety disorders, obsessive-compulsive disorder (OCD), post-traumatic stress disorder (PTSD), and substance use disorders (SUD). The lifetime presence of an anxiety disorder across all studies was coded based upon DSM-5 criteria (American Psychiatric Association, 2013) and therefore diagnoses of OCD and PTSD were included as their own category and did not count towards diagnosis of lifetime anxiety disorder. Conversely, lifetime SUD included individuals meeting criteria for abuse and dependence based on SCID-I interviews and those meeting criteria for SUD based on SCID-5 interviews. Across studies, interrater reliability for lifetime psychiatric diagnoses was good (kappas ≥ .81; Keel et al., 2005, 2008, 2018, 2025).

Data Analyses

Data were evaluated for normality prior to running analyses. Bingeing and purging frequency variables were log transformed before being used in parametric analyses. Missing data analysis revealed that 0.1–1.4% of data were missing and were missing completely at random (p=.665; Little, 1988) and therefore listwise deletion was employed (Schafer & Graham, 2002). We used t-tests to compare frequency of eating disorder behaviors and chi-square analyses to assess differences in prevalence of eating disorder and other psychiatric diagnoses between participants with and without current suicidal ideation. To determine how eating disorder diagnoses differed between groups, we first compared groups on presence vs. absence of any eating disorder. Following this, we compared groups on presence of each eating disorder diagnostic category (BN, PD, OSFED excluding PD) vs. absence of any eating disorder. Hierarchical binary logistic regression was used to identify contributions of eating disorder behaviors to current suicidal ideation above and beyond diagnostic comorbidities. Diagnostic comorbidities were entered in step 1, bingeing and fasting were entered in step 2, and purging was entered in step 3. Age and BMI were also included as covariates in the model. As an exploratory analysis, after mean centering eating disorder behavior frequency variables and BMI, we examined if BMI moderated the associations between eating disorder behaviors and suicidal ideation. An interaction term (BMI X eating disorder behavior frequency) was entered in step 4 for each of the three behaviors in three separate models. Sensitivity analyses controlled for study source as a covariate, and these results are included in an online supplement. All analyses were conducted using IBM SPSS 29.0.1.

Results

Clinical Characteristics

Table 3 displays clinical characteristics for the full sample, for groups with and without current suicidal ideation and comparisons between groups. Current suicidal ideation was endorsed by 8.9% of participants. There were significant differences in the lifetime prevalence of mood disorders, SUDs, OCD, and PTSD, with higher prevalence in individuals with current suicidal ideation for all disorders. Effect sizes were small (φ ≤ .17). We did not find significant differences in the prevalence of lifetime anxiety disorders across the suicidal ideation vs. no-suicidal ideation group. The prevalence of eating disorders was significantly higher in individuals with current suicidal ideation than without, with a small effect size (φ = .16). Specifically, prevalence of current BN and PD were higher in individuals with current suicidal ideation, with small effect sizes (φ ≤ .29). There was no significant difference in the prevalence of OSFED (excluding PD) in individuals with and without current suicidal ideation. Finally, there were significant differences in frequency of eating disorder symptoms such that individuals with current suicidal ideation reported more frequent binge-eating episodes, fasting, and purging, with medium to large effect sizes (Cohen’s d ≥ .72).

Table 3.

Clinical characteristics by current suicidal ideation

Full Sample (N= 732) Current Suicidal Ideation
 χ2(1) / t(df)
Yes (n = 65) No (n = 667)
ED behaviors
 Binge [M (SD)] a, b 1.76 (2.87) 3.82 (5.08) 1.56 (2.47) −4.23(71.55)**
 Fast [M (SD)] 1.05 (1.41) 1.42 (1.51) 1.02 (1.40) −2.19(728)*
 Purge [M (SD)] a, b 2.60 (4.69) 7.07 (7.11) 2.16 (4.13) −6.83(72.67)**
ED diagnosis c
Any ED [N (%)] 553 (75.5) 63 (96.9) 490 (73.5) 17.65**
BN [N (%)] 302 (41.3) 39 (60.0) 263 (39.4) 20.06**
PD [N (%)] 83 (11.3) 13 (20.0) 70 (10.5) 22.23**
OSFED (excluding PD) [N (%)] 156 (21.3) 9 (13.8) 147 (22.0) 2.37
Lifetime diagnosis
 Mood disorder [N (%)] 441 (60.2) 56 (86.2) 385 (57.7) 21.64**
 Anxiety disorder [N (%)] 290 (39.6) 30 (46.2) 260 (39.0) 1.74
 OCD [N (%)] 75 (10.2) 13 (20.0) 62 (9.3) 7.67*
 PTSD [N (%)] 154 (21.0) 23 (35.4) 121 (19.6) 11.25**
 SUD [N (%)] 281 (38.4) 33 (50.8) 248 (37.2) 5.39*
Open in a new tab
*

p < .05,

**

p ≤ .001

Note. ED = eating disorder. BN = bulimia nervosa. PD = Purging Disorder. OSFED = other specified feeding and eating disorders. OCD = obsessive-compulsive disorder. PTSD = post-traumatic stress disorder. SUD = substance use disorder. OSFED (excluding PD) includes participants with atypical AN, BN of low frequency/limited duration, BED of low frequency/limited duration, and unspecified feeding or eating disorders.

a

Levene’s test for equality of variances was significant (p < .05) and t-statistic and degrees of freedom come from t-test with equal variances not assumed.

b

Bingeing and purging frequency variables were log transformed.

c

Chi-square tests compared participants with each eating disorder diagnosis to participants with no eating disorder.

Unique Associations between Eating Disorder Symptoms and Suicidal Ideation

Table 4 displays results for the hierarchical binary logistic regression examining the unique contributions of lifetime psychiatric diagnoses and eating disorder symptoms to likelihood of current suicidal ideation. Lifetime psychiatric disorders contributed significantly to likelihood of current suicidal ideation (Nagelkerke R2 = .11, χ2 = 34.76, p < .001), due to the association between lifetime mood disorder and increased likelihood of suicidal ideation (B = 1.56, p < .001). Adding binge-eating frequency and fasting frequency improved model fit (ΔNagelkerke R2 = .03, Δχ2 = 10.96, p = .004), reflecting unique significant contributions of lifetime mood disorder (B = 1.28, p = .004) and greater frequency of bingeing (B = .61, p < .001) to current suicidal ideation. Adding purging frequency further improved model fit (ΔNagelkerke R2 = .08, Δχ2 = 27.15, p < .001). In the final model, lifetime mood disorder (B = 1.17, p = .009) and greater purging frequency (B = .83, p < .001) contributed significantly to likelihood of current suicidal ideation. After accounting for purging frequency, bingeing frequency was no longer significantly associated with current suicidal ideation (p = .30). In exploratory analyses, the interactions between BMI and binge frequency (ΔNagelkerke R2 < .01, Δχ2 = 1.78, p = .18), fast frequency (ΔNagelkerke R2 < .01, Δχ2 = .58, p = .45), and purge frequency (ΔNagelkerke R2 < .01, Δχ2 = 1.03, p = .31) did not add a significant contribution to the variance explained by the model. Moreover, there was no significant interaction between BMI and binge frequency (B = −.09, p = .19), fast frequency (B = −.02, p = .46), or purge frequency (B = −.05, p = .31), suggesting that the association between eating disorder behavior frequency and suicidal ideation do not differ across the weight spectrum. Results including compensatory exercise can be found in the on-line supplement. There were no differences in compensatory exercise frequency among individuals with and without current suicidal ideation (p = .90) and compensatory exercise was not associated with current suicidal ideation above and beyond comorbid diagnoses (p = .53).

Table 4.

Hierarchical binary logistic regression model predicting suicidal ideation based on bingeing, fasting, and purging

B SE Wald OR 95% CI Pseudo-R2 χ2(df)
Step 1 .11 34.76(7)**
 Age .04 .03 1.90 1.04 (.99, 1.10)
 BMI −.05 .04 1.34 .96 (.88, 1.03)
 Mood disorder 1.56** .43 12.93 4.75 (2.03, 11.12)
 Anxiety disorder −.16 .32 .25 .87 (.46, 1.60)
 OCD .47 .39 1.47 1.60 (.75, 3.44)
 PTSD .58 .32 3.29 1.79 (.95, 3.36)
 SUD .23 .29 .63 1.26 (.71, 2.22)
Step 2 .14 45.72(9)**
 Age .03 .03 1.40 1.04 (.98, 1.10)
 BMI −.03 .04 .70 .97 (.89, 1.05)
 Mood disorder 1.28* .44 8.46 3.60 (1.52, 8.54)
 Anxiety disorder −.11 .32 .12 .90 (.48, 1.68)
 OCD .36 .40 .82 1.43 (.66, 3.12)
 PTSD .48 .33 2.07 1.62 (.84, 3.11)
 SUD .27 .29 .87 1.31 (.74, 2.33)
 Binge frequency a .61** .18 11.08 1.84 (1.28, 2.63)
 Fasting frequency −.03 .10 .11 .97 (.79, 1.18)
Step 3 .22 72.86(10)**
 Age .01 .03 .02 1.01 (.95, 1.07)
 BMI .02 .05 .14 1.02 (.93, 1.11)
 Mood disorder 1.17* .45 6.91 3.22 (1.35, 7.72)
 Anxiety disorder <.01 .34 <.01 1.00 (.52, 1.94)
 OCD .58 .42 1.86 1.77 (.78, 4.06)
 PTSD .41 .35 1.35 1.50 (.76, 2.99)
 SUD .16 .30 .28 1.17 (.65, 2.12)
 Binge frequency a .20 .19 1.08 1.22 (.84, 1.78)
 Fasting frequency −.12 .11 1.37 .88 (.72, 1.09)
 Purge frequency a .83** .16 26.08 2.30 (1.67, 3.16)
Open in a new tab
*

p < .01,

**

p < .001.

Note. Pseudo-R2 is Nagelkerke R2. OCD = obsessive-compulsive disorder. PTSD = post-traumatic stress disorder. SUD = substance use disorder.

a

Bingeing and purging frequency variables were log transformed.

Discussion

This study aimed to assess differences in clinical characteristics, including frequency of eating disorder behaviors (i.e., bingeing, fasting, and purging), among adult women with and without current suicidal ideation and to examine how frequency of eating disorder behaviors are associated with current suicidal ideation above and beyond comorbid psychiatric diagnoses. Findings were partially consistent with our first hypothesis, demonstrating that women with current suicidal ideation were more likely to have most lifetime psychiatric diagnoses but did not have elevated lifetime histories of anxiety disorders or current OSFED (excluding PD). Separating PD from other OSFED diagnoses may have diminished associations between OSFED and suicidal ideation found by others (Duffy et al., 2021; Izquierdo et al., 2023). This may indicate that PD is more strongly associated with suicidal ideation than other OSFED diagnoses. In terms of eating disorder behaviors, women with current suicidal ideation reported more frequent bingeing, fasting, and purging than women without current suicidal ideation.

Additionally, findings supported our second hypothesis that purge frequency explained additional variance in current suicidal ideation above and beyond both comorbid psychiatric disorders and binge and fast frequencies. These findings support prior associations between purging and suicidal ideation (Joiner et al., 2022; Obeid et al., 2020) and extend this by showing associations after controlling for comorbid diagnoses. Prior works have commented that greater frequency of purging may only serve as an indicator of illness severity, which may explain its role in predicting suicidal ideation (Dieffenbacher et al., 2024; Edler et al., 2007; Forney et al., 2016; Haedt et al., 2006). However, high comorbidity indicates greater clinical severity (Kessler et al., 2005; Kingsbury et al., 2019). Therefore, our results suggest that purging frequency is associated with current suicidal ideation above and beyond clinical severity.

Conversely, results contradict a prior report that restrictive eating predicted suicidal ideation, and binge eating and purging did not (Wang et al., 2019). This contradiction may be due to differences in study samples, as the current study included adult women across the weight spectrum, and Wang and colleagues (2019) included adolescents and young adults with low-weight eating disorders. This may suggest that the association between eating disorder behaviors and suicidal ideation may differ based on age, as suggested by Joiner and colleagues (2022). At lower BMIs and at younger ages, fasting and restrictive eating may represent more severe, life-threatening behaviors than captured in our sample. However, exploratory findings indicated that BMI does not moderate the association between fasting and current suicidal ideation. The current study may have needed more participants with a low BMI to detect any moderation effects. Our findings support the notion that purging may act as a form of self-harm similar to non-suicidal self-injury due to its active nature, potentially explaining its unique association with suicidal ideation (Jacobson & Luik, 2014).

The present study had several strengths. Notably, our large sample was community-based and consisted of women with multiple types of eating disorders. Structured clinical interviews with high interrater reliability were used to determine eating disorder and other psychiatric diagnoses, and use of the EDE was particularly valuable in teasing apart frequencies of binge eating, fasting, and purging. Prior studies on suicidal ideation and eating disorder behaviors have largely been limited to individuals with a smaller range of diagnoses (Hunt & Levinson, 2021; Wang et al., 2019) or to clinical samples (Joiner et al., 2022). Additionally, many studies examined only lifetime suicidal ideation (Arnold et al., 2023; Bodell et al., 2013; Schmidt et al., 2024), and the assessment of current suicidal ideation allows us to draw the conclusion that the suicidal ideation did not simply come and go before eating disorder symptoms developed. Moreover, the inclusion of a large number of DSM-5 psychiatric comorbidities allowed us to examine unique contributions of specific eating disorder features to suicidal ideation.

Importantly, findings should be examined in the context of study limitations. First, the cross-sectional nature of our data does not permit temporal or causal inferences between the frequency of eating disorder behaviors and suicidal ideation. Our sample was restricted to adult women, which may limit the generalizability of our findings to other populations affected by eating disorders and suicidality such as men and adolescents. Findings for our community-based sample may differ from findings in clinical samples, due to associations between treatment seeking and increased comorbidity, impairment, and distress (Feinstein et al., 1986), which has been shown to be associated with suicidal ideation (Groh et al., 2024). Drawing both cases and controls from the community may have reduced the impact of Berkson’s Bias on associations between symptom levels and suicide risk (Feinstein et al., 1986). As the current sample included a very small proportion of individuals with AN and participants were primarily young adults, associations between eating disorder behaviors and suicidal ideation may differ from those found for samples that differ in age or diagnostic composition. Our assessment of suicidal ideation relied on single-item assessments, potentially limiting our ability to detect important variation in suicidality, a complex and nuanced construct. Moreover, although comorbidity can serve as a proxy for clinical severity, the inclusion of a clinical severity scale such as the Clinical Global Impression may be useful in future work to confirm if purging frequency is associated with suicidal ideation above and beyond clinical severity. Finally, our focus on fasting did not allow us to examine other forms of restriction. Moreover, only a small proportion of our sample endorsed fasting and AN was underrepresented, which may account for non-significant findings. If we were to examine multiple forms of restricting, our results may have provided additional insight into the relationship between eating disorder behavior and suicidal ideation.

In summary, purging is associated with current suicidal ideation above and beyond bingeing, fasting, and other psychiatric comorbidities among adult women primarily without low weight eating disorders. These findings provide important clinical implications for screening, detection, and ultimate reduction of suicide risk in adult women. Specifically, purging behavior should be included in screens as a clinically significant symptom in its own right and as an indicator of increased suicide risk. Clinicians should remain vigilant in assessing for suicide risk across eating disorders. Future research should examine associations longitudinally, to evaluate temporal precedence as well as possible bidirectional associations between purging and suicide risk.

Supplementary Material

OnlineSupplement

Funding statement:

This work was supported by grants from the National Institute of Mental Health (R03MH061320; R01MH061836; R01MH111263, PI: Pamela K. Keel). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Contributor Information

Holly K. Spinner, Department of Psychology, Florida State University

Sara Prostko, Department of Psychology, Florida State University.

Thomas E. Joiner, Department of Psychology, Florida State University

Pamela K. Keel, Department of Psychology, Florida State University.

Data availability statement:

Data are available upon request.

References

  1. Arnold S, Correll CU, & Jaite C (2023). Frequency and correlates of lifetime suicidal ideation and suicide attempts among consecutively hospitalized youth with anorexia nervosa and bulimia nervosa: results from a retrospective chart review. Borderline Personality Disorder and Emotion Dysregulation, 10(1). 10.1186/s40479-023-00216-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arcelus J, Mitchell AJ, Wales J, & Nielsen S (2011). Mortality rates in patients with anorexia nervosa and other eating disorders: a meta-analysis of 36 studies. Archives of General Psychiatry, 68(7), 724–731. 10.1001/archgenpsychiatry.2011.74 [DOI] [PubMed] [Google Scholar]
  3. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.) 10.1176/appi.books.9780890425596 [DOI] [Google Scholar]
  4. Beck AT, Ward CH, Mendelson M, Mock J, & Erbaugh J (1961). An inventory for measuring depression. Archives of General Psychiatry, 4(6), 561–571. 10.1001/archpsyc.1961.01710120031004 [DOI] [PubMed] [Google Scholar]
  5. Bodell LP, Joiner TE, & Keel PK (2013). Comorbidity-independent risk for suicidality increases with bulimia nervosa but not with anorexia nervosa. Journal of psychiatric research, 47(5), 617–621. 10.1016/j.jpsychires.2013.01.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dieffenbacher AL, Meule A, & Voderholzer U (2024). Purging frequency and number of purging methods as indicators of severity in bulimia nervosa: Interactive effects with binge eating episodes. European Eating Disorders Review, 33(2), 343–359. 10.1002/erv.3147 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Duffy ME, Siegfried N, Bass Garrett., & Joiner TE. (2021). Presence and severity of suicidal thoughts and behaviors across the eating disorder diagnostic spectrum: A pilot study. Journal of Clinical Psychology, 77(4), 1045–1053. 10.1002/jclp.23095 [DOI] [PubMed] [Google Scholar]
  8. Edler C, Haedt AA, & Keel PK (2007) The use of multiple purging methods as an indicator of eating disorder severity. International Journal of Eating Disorders, 40(6), 515–520. 10.1002/eat.20416 [DOI] [PubMed] [Google Scholar]
  9. Fairburn CG & Cooper Z (1993). The eating disorder examination. In Fairburn C & Wilson GT (Eds.), Binge eating: Nature, assessment and treatment. (12th ed., pp. 317–331). New York, NY: Guilford Press. [Google Scholar]
  10. Fairburn CG, Cooper Z, & O’Connor M (2014). Eating Disorder Examination (Edition 17.0D). https://www.credo-oxford.com/pdfs/EDE_17.0D.pdf
  11. Feinstein AR, Walter SD, & Horwitz RI (1986). An analysis of Berkson's bias in case-control studies. Journal of chronic diseases, 39(7), 495–504. 10.1016/0021-9681(86)90194-3 [DOI] [PubMed] [Google Scholar]
  12. First MB, Spitzer RL, Gibbon M, & Williams JBW (1995). Structured clinical interview for DSM-IV Axis I disorders—Patient edition (SCID/P): Biometrics Research Department: New York State Psychiatric Institute, New York, 1995. [Google Scholar]
  13. First MB, Williams JB, Karg RS, & Spitzer RL (2015). Structured clinical interview for DSM-5 disorders-research version (SCID-5 for DSM-5, research version; SCID-5-RV). American Psychiatric Association. [Google Scholar]
  14. Forney KJ, Buchman-Schmitt JM, Keel PK, & Frank GKW (2016). The medical complications associated with purging. International Journal of Eating Disorders, 49(3), 249–259. 10.1002/eat.22504 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Forrest LN, Bodell LP, Witte TK, Goodwin N, Bartlett ML, Siegfried N, Eddy KT, Thomas JT, Franko DL, & Smith AR (2016). Associations between eating disorder symptoms and suicidal ideation through thwarted belongingness and perceived burdensomeness among eating disorder patients. Journal of Affective Disorders, 195, 127–135. 10.1016/j.jad.2016.02.017 [DOI] [PubMed] [Google Scholar]
  16. Franklin JC, Ribeiro JD, Fox KR, Bentley KH, Kleiman EM, Huang X, Musacchio KM, Jaroszewski AC, Chang BP, & Nock MK (2017). Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychological Bulletin, 143(2), 187. 10.1037/bul0000084 [DOI] [PubMed] [Google Scholar]
  17. Groh A, Bahlmann l., Colic L, Schulz A, Kastner UW, Polzer U, Walter M, Sobanski T, & Wagner G. Psychological distress among individuals with a suicide attempt or suicidal ideation and suicide attempts patterns: First two years of the pandemic. Frontiers in Psychiatry, 15. 10.3389/fpsyt.2024.1366191 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Haedt AA, Edler C, Heatherton TF, & Keel PK (2006). Importance of multiple purging methods in the classification of eating disorder subtypes. International Journal of Eating Disorders, 39(8), 648–654. 10.1002/eat.20335 [DOI] [PubMed] [Google Scholar]
  19. Hunt RA, & Levinson CA (2021). Self-reported interoceptive sensibility does not moderate the relationship between eating disorder symptoms and suicidal thoughts and behaviors. Behavior Therapy, 52(5), 1137–1144. 10.1016/j.beth.2021.03.002 [DOI] [PubMed] [Google Scholar]
  20. Izquierdo AM, Nelson JD, Daza A, Gasbarro A, Hardin R, Marino J, & Fischer S (2023). Predictors of current suicidal ideation in a multi-diagnostic sample of individuals with eating disorders. Journal of Eating Disorders, 11, 94. 10.1186/s40337-023-00789-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jacobson CM, & Luik CC (2014). Epidemiology and sociocultural aspects of non-suicidal self-injury and eating disorders. In Claes L & Muehlenkamp JJ (Eds.), Non-Suicidal Self-Injury in Eating Disorders (pp. 19–34). Springer-Verlag Berlin Heidelberg. 10.1007/978-3-642-40107-7_2 [DOI] [Google Scholar]
  22. Jobes DA, & Joiner TE (2019). Reflections on suicidal ideation. Crisis, 40(4), 227–230. 10.1027/0227-5910/a000615 [DOI] [PubMed] [Google Scholar]
  23. Joiner T (2005). Why people die by suicide. Harvard University Press. [Google Scholar]
  24. Joiner TE, Robison M, Manwaring J, Rienecke RD, Le Grange D, Duffy A, Mehler PS, & Blalock DV (2025). The reliability and construct validity of a single-item assessment of suicidal ideation among patients with mood and anxiety disorders. Journal of Clinical Psychology, 81(5), 290–297. 10.1002/jclp.23768 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Joiner TE, Robison M, McClanahan S, Riddle M, Manwaring J, Rienecke RD, Le Grange D, Duffy A, Mehler PS, & Blalock DV (2022). Eating disorder behavior as predictors of suicidal ideation among people with an eating disorder. International Journal of Eating Disorders, 55(10), 1352–1360. 10.1002/eat.23770 [DOI] [PubMed] [Google Scholar]
  26. Keel PK, Bodell LP, Ali SI, Starkey A, Trotta J, Luxama JW, Halfhide C, Hill NG, Appelbaum J, & Williams DL (2025). Examining weight suppression, leptin levels, glucagon-like peptide 1 response, and reward-related constructs in severity and maintenance of bulimic syndromes: Protocol and sample characteristics for a cross-sectional and longitudinal study. JMIR Research Protocols, 14, e66554. 10.2196/66554 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Keel PK, Haedt A, & Edler C (2005). Purging disorder: An ominous variant of bulimia nervosa? International Journal of Eating Disorders, 38(3), 191–9. 10.1002/eat.20179 [DOI] [PubMed] [Google Scholar]
  28. Keel PK, Haedt-Matt AA, Hildebrandt B, Bodell LP, Wolfe BE, & Jimerson DC (2018). Satiation deficits and binge eating: Probing differences between bulimia nervosa and purging disorder using an ad lib test meal. Appetite, 127, 119–125. 10.1016/j.appet.2018.04.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Keel PK, Wolfe BE, Gravener JA, & Jimerson DC (2008). Co-morbidity and disorder-related distress and impairment in purging disorder. Psychology Medicine, 38(10), 1435–1442. 10.1017/S0033291707001390 [DOI] [PubMed] [Google Scholar]
  30. Kessler RC, Chiu WT, Demler O, & Walters EE (2005). Prevalence, severity, and comorbidity of twelve-month DSM-IV disorders in the national comorbidity survey replication (NCS-R). Archive of General Psychiatry, 62(6), 617–627. 10.1001/archpsyc.62.6.617 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kingsbury M, Sucha E, Horton NJ, Sampasa-Kanyinga H, Murphy JM, Gilman SE, & Colman I (2019). Lifetime experience of multiple common mental disorders and 19-year mortality: Results from a Canadian population-based cohort. Epidemiology and Psychiatric Sciences, 4(29), e18. 10.1017/S2045796018000859 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kwan YM, Gordon KH, Carter DL, Munich AM, & Grossman SD (2016). An examination of the connections between eating disorder symptoms, perceived burdensomeness, thwarted belongingness, and suicide risk among undergraduate students. Suicide and Life-Threatening Behavior, 47(4), 493–508. 10.1111/sltb.12304 [DOI] [PubMed] [Google Scholar]
  33. Laakso E, Hakko H, Räsänen P, Riala K, & STUDY T (2013). Suicidality and unhealthy weight control behaviors among female underaged psychiatric inpatients. Comprehensive Psychiatry, 54(2), 117–122. 10.1016/j.comppsych.2012.06.012 [DOI] [PubMed] [Google Scholar]
  34. Lipson SK, & Sonneville KR (2020). Understanding suicide risk and eating disorders in college student populations: Results from a national study. International Journal of Eating Disorders, 53(2), 229–238. 10.1002/eat.23188 [DOI] [PubMed] [Google Scholar]
  35. Little RJA (1988). A Test of Missing Completely at Random for Multivariate Data with Missing Values. Journal of the American Statistical Association, 83(404), 1198–1202. 10.2307/2290157 [DOI] [Google Scholar]
  36. Obeid N, Norris ML, Valois DD, Buchholz A, Goldfield GS, Hadjiyannakis S, Henderson KA, Flament M, Hammond NG, Dunn J, & Spettigue W (2020). Bingeing, purging, and suicidal ideation in clinical and non-clinical samples of youth. Eating Disorders, 28(3), 289–307. 10.1080/10640266.2019.1642033 [DOI] [PubMed] [Google Scholar]
  37. Rogers ML, Duffy ME, Buchman-Schmitt JM, Datoc AE, & Joiner TE (2019). Exercise dependence: Associations with capability for suicide and past suicidal behavior. Journal of Clinical Psychology, 75(1), 165–177. 10.1002/jclp.22696 [DOI] [PubMed] [Google Scholar]
  38. Schafer JL & Graham JW (2002). Missing data: Our view of the state of the art. Psychological Methods, 7(2), 147–177. 10.1037/1082-989X.7.2.147 [DOI] [PubMed] [Google Scholar]
  39. Schmidt K, Fitzgerald E, & Keel PK (2024). Suicidal thoughts and attempts in a transdiagnostic eating disorder sample: Do diagnostic severity criteria predict risk?. European Eating Disorders Review, 32(5), 952–962. 10.1002/erv.3104 [DOI] [PubMed] [Google Scholar]
  40. Sepulveda AR, Anastasiadou D, Pellegrin Y, Andrés P, Graell M, Carrobles JA, & Morandé G (2014). Impact of caregiving experience on mental health among caregivers: A comparison of eating disorder patients with purging and non-purging behaviors. Eating and Weight Disorders-Studies on Anorexia, Bulimia and Obesity, 19(1), 31–39. 10.1007/s40519-013-0064-5 [DOI] [PubMed] [Google Scholar]
  41. Shahnaz A, & Klonsky ED (2021). Clarifying the association of eating disorder features to suicide ideation and attempts. Journal of Clinical Psychology, 77(12), 2965–2977. 10.1002/jclp.23266 [DOI] [PubMed] [Google Scholar]
  42. Smith AR, Fink EL, Anestis MD, Ribeiro JD, Gordon KH, Davis H, Keel PK, Bardone-Cone AM, Peterson CB, Klein MH, Crow S, Mitchell JE, Crosby RD, Wonderlich SA, le Grange D, & Joiner TE Jr (2013). Exercise caution: over-exercise is associated with suicidality among individuals with disordered eating. Psychiatry research, 206(2–3), 246–255. 10.1016/j.psychres.2012.11.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Smith AR, Velkoff EA, Ribeiro JD, & Franklin J (2019). Are eating disorders and related symptoms risk factors for suicidal thoughts and behaviors? A meta-analysis. Suicide and Life-Threatening Behavior, 49(1), 221–239. 10.1111/sltb.12427 [DOI] [PubMed] [Google Scholar]
  44. Smith AR, Zuromski KL, & Dodd DR (2018). Eating disorders and suicidality: What we know, what we don’t know, and suggestions for future research. Current Opinion in Psychology, 22, 63–67. 10.1016/j.copsyc.2017.08.023 [DOI] [PubMed] [Google Scholar]
  45. Sohn MN, Dimitropoulos G, Ramirez A, McPherson C, Anderson A, Munir A, … & Devoe DJ. (2023). Non-suicidal self-injury, suicidal thoughts and behaviors in individuals with an eating disorder relative to healthy and psychiatric controls: A systematic review and meta-analysis. International Journal of Eating Disorders, 56(3), 501–515. 10.1002/eat.23880 [DOI] [PubMed] [Google Scholar]
  46. Stunkard AJ, & Messick S (1985). The three-factor eating questionnaire to measure dietary restraint, disinhibition and hunger. Journal of Psychosomatic Research, 29(1), 71–83. 10.1016/0022-3999(85)90010-8 [DOI] [PubMed] [Google Scholar]
  47. Udo T, Bitley S, & Grilo CM (2019). Suicide attempts in US adults with lifetime DSM-5 eating disorders. BMC medicine, 17(1), 120. 10.1186/s12916-019-1352-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Van Orden KA, Witte TK, Cukrowicz KC, Braithwaite SR, Selby EA, & Joiner TE Jr. (2010). The interpersonal theory of suicide. Psychological Review, 117(2), 575–600. 10.1037/a0018697 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Wang SB, Mancuso CJ, Jo J, Keshishian AC, Becker KR, Plessow F, Izquierdo AM, Slattery M, Franko DL, Misra M, Lawson EA, Thomas JJ, Eddy KT (2019). Restrictive eating, but not binge eating or purging, predicts suicidal ideation in adolescents and young adults with low-weight eating disorders. International Journal of Eating Disorders, 53(3), 472–477. 10.1002/eat.23210 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

OnlineSupplement
NIHMS2197162-supplement-OnlineSupplement.pdf (169.7KB, pdf)

Data Availability Statement

Data are available upon request.

RESOURCES