Table 3.

Overview of study characteristics

Study: authors, year, location		Sample description	Study design	Assessment of PMDD	Assessment of suicidality	Data analysis	Overview of findings	Quality Rating
Suicide cognitions: Studies assessing suicidal thoughts, ideation, plans or risk (in reverse chronological order)
1	de Carvalho et al. (2018)Brazil	727 women, aged between 21 and 32 years	Cohort: Analysis of data collected as part of a larger cross-sectional population-based study (psychosocial and biological factors in bipolar disorder: a population-based cohort of young adults)	Mini International Neuropsychiatry Interview (MINI Plus) translated into Brazilian Portuguese (Amorim 2000) The MINI assesses PMDD based upon DSM-IV criteria (American Psychiatric Association 2000)	Mini International Neuropsychiatry Interview (MINI Plus) translated into Brazilian Portuguese (Amorim 2000)	Chi-square tests were used to assess factors associated with PMDD. Poisson regression was used to obtain estimates of prevalence ratios and to adjust the analysis for potential confounders	128 women (17.6%) were identified to have PMDD Significant co-morbidity of current suicide risk and PMDD p = 0.002. Women with PMDD were two to three times more likely to report suicide risk than women without PMDD	59.0% Good
2	Ogebe et al. (2011) USA and Nigeria	537 adolescents, aged between 13 and 21 years	Cohort: Female adolescents attending four outpatient clinics in Lagos and Maiduguri (Nigeria) and Akron (USA) were recruited	A modified version of the Premenstrual Symptoms Screening Tool (PSST; Steiner et al. 2003) The PSST assesses PMDD based upon DSM-IV criteria (American Psychiatric Association 2000)	The Modified Mini International Neuropsychiatry Interview (Sheehan et al. 1998)	Pearson Chi-square tests were to compare the distribution of categorical variables	22 adolescents (4.1%) were identified to have PMDD 18% of adolescents with PMDD experienced co-morbid suicidal ideation, compared with 24% and 10.9% of adolescents with moderate/severe PMS and no/mild PMS, respectively	66.7% Good
3	Yonkers et al. (2003) USA	907 women, with an average age of 31 years	Cohort: Women attending six obstetric-gynaecological practices in Connecticut were invited to participate	The Premenstrual Attitudes and Perception Scale (PAPS) which was developed for this study and the Last Menstrual Period Module (LMPM) based upon the Daily Rating of the Severity of Problems (Endicott et al. 1993)	The Brief Patient Health Questionnaire (BHQ) from the Primary Care Evaluation of Mental Disorders (Spitzer et al. 2000)	Data not reported	24% of women with PMDD endorsed suicidal thoughts at any level (several days, more than half of the days, every day). 20% endorsed these thoughts for several days	38.5% Moderate
Suicide attempts: Studies assessing suicide attempts (in reverse chronological order)
4	Shams-Alizadeh et al. (2018) Iran	120 women aged between 13 and 40 years and 120 matched controls Controls were matched with respect to age and occupational status	Control: Women admitted to the hospital following a suicide attempt and a control group of women accompanying patients on other wards were recruited	Psychiatric interview based on DSM-5 criteria for PMDD (American Psychiatric Association 2013)	Women admitted to the hospital following a suicide attempt	Pearson’s Chi-square test was used to assess the relationship between the independent variables and attempting suicide	Significantly higher frequency of PMDD in women who had attempted suicide (37/120 = 30.8%) compared with control group (6/120 = 5%); p = 0.001. No relationship between suicide attempts in PMDD and menstrual phase	64.1% Good
5	Ducasse et al. (2016) France	232 women aged between 18 and 53 years, with a median age of 33.8 years	Cohort: Women admitted to the hospital following a suicide attempt were recruited	Premenstrual Assessment Form Shortened version (Allen et al. 1991) using questions that met criteria from the DSM-5 (American Psychiatric Association 2013)	Women admitted to the hospital following a suicide attempt	Associations were calculated with odds ratios and 95 confidence intervals and logistic regression models	High prevalence of PMDD in women who had attempted suicide (23%: 51/232) Severity of suicide attempt was not significantly associated with PMDD. No significant difference between presence of PMDD and menstrual cycle phase of attempt	61.5% Good
6	Soydas et al (2014) Turkey	70 women with PMDD, aged between 20 and 41 years and 78 healthy controls	Control: Women with PMDD who were referred to an outpatient psychiatry service and a control group comprised of hospital staff and staff relatives were recruited	Premenstrual Syndrome Scale (PMSS; Gencdogan 2006)	Structured Clinical Interview for Axis I disorders (SCID; First et al. 1997), translated into Turkish (Özkürkçügil et al. 1999)	Chi-square tests were used to assess the relationships between variables	Significantly higher frequency of historical suicide attempts in women with PMDD (7.1%) compared with control group (0%); p = 0.024	56.4% Good
7	Baca-Garcia et al. (2004) Spain	125 women and 83 controls. Average ages were 30.6 years and 32.7 years, respectively	Control: Women admitted to the hospital following a suicide attempt and a control group of blood donors were recruited	Clinical interviews based upon DSM-IV criteria for PMDD (American Psychiatric Association 2000)	Women admitted to the hospital following a suicide attempt Impulsivity was assessed using the Beck Suicidal Intent Scale (SIS-I, Beck et al. 1974). Lethality was assessed using the Lethality Rating Scale (Beck et al. 1975)	Fisher’s exact tests were used to compare diagnosis and suicide attempt features between suicide attempters with and without PMDD. Chi-square tests were used to assess menstrual cycle phase	Significantly higher prevalence of PMDD in women attending the emergency room following a suicide attempt (54%: 67/125) compared with controls (6%: 5/83); p < 0.001. No significant difference regarding lethality and impulsivity of attempts No increase in suicide risk during luteal phase	43.6% Moderate
8	Wittchen et al. (2002)	1488 women, aged 14–24 years	Cohort: Analysis of data collected as part of a larger cross-sectional population-based study	PMDD module from the Munich Composite International Diagnostic Interview (M-CIDI; Wittchen and Pfister 1997) based upon DSM-IV criteria (American Psychiatric Association 2000)	Munich Composite International Diagnostic Interview (M-CIDI; Wittchen and Pfister 1997)	Odds ratio and 95% confidence intervals were calculated to compare PMDD and non-PMDD cases	112 (7.4%) women were identified to have PMDD High risk of suicide attempts among PMDD cases but only moderately increased risks of suicidal ideation. Higher prevalence of women reporting at least one suicide attempt during their lifetime (15.8%) compared with non-PMDD women (3.2%)	59.0% Good
Suicide cognitions + Suicide attempts: Studies assessing suicidal thoughts, ideation, plans or risk and suicide attempts
9	Pilver et al. (2013) USA	3965 women, aged 18–40 years with an average age of 28.8 years	Cohort: Analysis of data collected as part of a larger cross-sectional population-based study (collaborative psychiatric epidemiology surveys CPES data set)	Premenstrual Syndrome module of the World Mental Health Composite International Diagnostic Interview (WMH-CIDI), based upon DSM-IV criteria	Statements about suicidal behaviours where positive responses were scored “1” and negative responses “0”	Three sets of logistic regression models were constructed to assess the magnitude and direction of the association between PMDD status and suicidal behaviours and to control for possible sources of confounding	PMDD was strongly and independently associated with suicidal behaviours. Women with PMDD significantly more likely than women with PMS and no premenstrual symptoms to report suicide attempts (16.2%, 7.4%, 4.9%), suicidal ideation (37.4%, 22% and 13.3%) and suicide plans (19.1%, 7.6% and 4.6%); p = < 0.05	74.4% Good
10	Hong et al. (2012) South Korea	2499 Korean women, aged between 18 and 49 years	Cohort: Analysis of data collected as part of a larger cross-sectional population-based study (Korean Epidemiologic Catchment Area (KECA) study)	A translated version of the PMDD module from the Composite International Diagnostic Interview (WHO-CIDI; Kessler and Üstün 2004) based upon DSM-IV PMDD criteria (American Psychiatric Association 2000)	Korean version of the Composite International Diagnostic Interview module on suicide (K-CIDI; Cho et al. 2002)	Odds ratios and 95% confidence intervals and logistic regression analyses were used to compare PMDD and non-PMDD cases	59 women (2.4%) were identified to have PMDD Both 12-month and lifetime prevalence of suicide ideations, plans and attempts were significantly associated with PMDD Lifetime PMDD vs non-PMDD: suicidal ideation (45.8% vs.17.3%), suicide plans (16.9% vs. 4.2%) and suicide attempts (13.6% vs. 3.9%)	79.5% Excellent