TABLE 1. Studies Investigating Consistency in Self-Harm Reports: Across Different Measures or Over Time.

Publication	Country	Sample	Measure	Comparison of different measures	Consistency in reports over time	Other design	Results
Adolescent samples
Ougrin and Boege (2013)	United Kingdom	Adolescent inpatients and outpatients n = 100, 12–17 years	Self-harm	Yes Self-report questionnaire and clinical record (reported during the clinical assessment)	No	No	3 (3%) indicated in the clinical record they had self-harmed but did not report self-harm on the questionnaire. 20 (20%) reported at least one episode of self-harm on the questionnaire that was not recorded in the clinical record
Hart et al. (2013)	United States	Longitudinal community sample of adolescents n = 678, Assessed annually between age 12 and 22 years	Suicide attempts	No	Yes Also examined characteristics associated with discrepant reporting	No	88.5% inconsistently reported a suicide attempt at some point during the study; 65.3% were inconsistent the year following the self-harm event Consistent and inconsistent reporters did not differ on clinical or demographic variables, but consistent reporters had higher lifetime suicidal ideation
Bjärehed et al. (2013)	Sweden	Adolescent community sample n = 1,052, from grade 7 (mean age 13.7 years) and 8 (mean age 14.7 years)	Non-suicidal self-injury	Yes Self-report questionnaire and follow-up interview	No	No	97 adolescents were selected for interview. 32/66 (48%) participants who reported self-harm on the questionnaire did not disclose self-harm during the follow-up interview.
Kidger et al. (2012)	United Kingdom	Adolescent community sample n = 4810, 16 years	Suicide attempts	No	No	Yes Examined suicidal thoughts among those with suicidal self-harm	Approximately 10% of those who reported wanting to die during the most recent episode of self-harm said they had never had thoughts of killing themselves
Christl et al. (2006)	Germany	Longitudinal community sample of adolescents/young adults n = 3021, 14–24 years at baseline	Suicide attempts	No	Yes Also examined characteristics associated with discrepant reporting	Yes Compared drop-out rates among those with and without suicidal behaviour	One third of baseline suicide attempters (n = 15/45), did not report a suicide attempt at follow-up 4 years later 81% of discrepant reporters were female and 59% were aged 14–17 at baseline. Greater consistency in reporting was associated with a higher number of psychiatric disorders Those with a suicide attempt at baseline were at least 1.6 times more likely to drop out of the study then those without suicide attempts or ideas
Ross and Heath (2002)	Not specified	Adolescent community sample n = 440 (from 2 schools), average age 14–15 years	Self-mutilation (defined as deliberate alteration or destruction of body tissue without suicidal intent)	Yes Self-report screening questionnaire and follow-up interview	No	No	School sample 1: 38.8% who reported self-harm on the questionnaire were not classified as having self-harmed following the interview (19/49) School sample 2: 24.4% who reported self-harm on the questionnaire were not classified as having self-harmed following the interview (10/41)
Velting et al. (1998)	United States	Adolescent outpatients n = 48, 12–20 years, mean 15.3 years	Suicide attempts	Yes Self-report questionnaire and interview Also investigated explanations and characteristics associated with discrepant reporting	No	No	Discrepancies in reporting were found amongst 50% of the sample (24/48). Discrepancies primarily due to confusion with the operational definition of suicidal behaviour (i.e., confused attempt and ideation or confused attempt and gesture). The discrepant and non-discrepant groups were comparable on measures of suicidal intent, ideation, and hopelessness and on their diagnostic profiles
O'Sullivan and Fitzgerald (1998)	Ireland	Adolescent community sample n = 88 age 13–14 years	Suicide attempt	Yes Self-report screening questionnaire and follow-up interview	No	No	45 adolescents completed a follow-up interview. 5/7 (71%) participants who reported a suicide attempt on the questionnaire did not disclose self-harm during the follow-up interview.
Adult samples
Eikelenboom et al. (2014)	The Netherlands	Longitudinal cohort of adults with depressive or anxiety disorders n = 1973, aged 18–65 years at baseline, (mean age 42.4 years)	Suicide attempts	No	Yes Also examined characteristics associated with discrepant reporting	No	23% of baseline suicide attempters, did not report their attempt at follow-up 2 years later (63/274) Consistent reporting was associated with a greater number of suicide attempts, and more severe current psychopathology. No differences were found for recency of the event, age, sex, or education
Morthorst et al. (2011)	Denmark	Patients admitted to hospital following a suicide attempt n = 243, age 12+, mean age 31 years	Suicide attempts, assessed 1 year after baseline	Yes Self-report (telephone interview) and hospital records	No	No	Seven suicide attempts listed in the hospital records were not reported by participants. Nine patients reported a suicide attempt that was not listed in the hospital records
Plöderl et al. (2011)	Austria	Adult community sample n = 1385, age 18–84 years, Mean 37.8 years	Suicide attempts	No	No	Yes Examined intent to die among those reporting suicidal self-harm	One quarter (15/60) of individuals reporting a suicide attempt were false positives (lacked intent or attempt aborted) 0.8% (n = 11) were identified as false negatives (reported no suicide attempt on the screen question but reported a self-harm event with intent to die in follow-up questions). 2/11 (18%) false negatives resulted in injuries requiring hospital treatment There were no differences between true positives and false positives regarding age or education or lethality of method
Linehan et al. (2006)	United States	Adult clinical sample: Five cohorts, three with borderline personality disorder	Self-harm	Yes Self-report interview and 1) therapist notes 2) participant diary cards 3) medical records	No	No	Agreement with therapist notes (presence/absence of self-harm) was 83% Good agreement with diary cards (mean 4.5 acts at interview vs. mean 4.3 acts on diary cards) 82% of episodes reported by participants as being medically treated had a corresponding medical record. There were no false negatives —all medically treated episodes were reported by participants.
Nock and Kessler (2006)	United States	Predominately Adult community sample n = 5,877, aged 15–54 years	Suicide attempts	No	No	Yes Examined intent to die among those reporting suicidal self-harm	112/268 (42%) of those reporting a lifetime history of suicide attempt reported no intent to die