The SUAS-S is a 20-item self-report rating scale measuring the patient’s attitude towards suicide, suicide-related behaviour and suicidal ideation on the day of reporting and during the previous seven days. Each item is scored in the range of 0–4 on a Likert-type scale and resulting in a scale sum score with a range of 0–80. The scale is designed to measure levels of suicidality and to be sensitive to temporal changes in suicide-related symptoms . The SUAS-S items cover five thematic areas [6, 8]: “affect” (items 1, 2, 9, 12 and 13), “bodily states” (items 3, 8 and 10), “control and coping” (items 6, 7, 11 and 15), “emotional reactivity” (items 4, 5 and 14) and “suicidal thoughts and behaviour” (items 16–20). The original SUAS-S scale has been found to possess good concurrent validity compared with the interview version .
Fifty-five patients consecutively referred to a psychiatric outpatient clinic from January 2008 until January 2009, were asked to participate in the project. Fifty-three (96%) of the patients provided informed consent and were included in the study. Of these, one patient dropped out before all scales were completed and was not included in the further analysis. Patients with an on-going substance abuse and/or psychotic disorder were not eligible to participate in the study. Patients were recruited from a non-emergency setting.
Deliberate self-harm (DSH) was defined as “an act with non-fatal outcome in which an individual deliberately initiated behaviour with the intention to cause self-harm, for example, self-cutting or jumping from a height, by ingestion of an illicit drug, a non-ingestible substance or an excess of a prescribed substance”[10, 11]. Suicide attempt (SA) was defined as “a potentially self-injurious behaviour with a nonfatal outcome, for which there is evidence that the person intended, at some level, to kill himself/herself”. The Clinical Global Impression of Severity of Suicidality Scale (CGI-SS)  was used in the baseline assessment to provide an overall clinician rated measure of the clinical risk of suicidality in each participant. The CGI-SS has five levels of severity of suicidality; 1 = not at all suicidal, 2 = mildly suicidal, 3 = moderately suicidal, 4 = severely suicidal, and 5 = attempted suicide. Interviews yielding these scores were made independent of and blind to the participants' SUAS-S scores. Study subjects were categorically classified as “non-suicidal” if they received scores 1 or “suicidal” if they were scored 2–5 on the CGI-SS. The baseline assessment also included measuring depressive symptoms with the Beck Depression Inventory (BDI) , feelings of hopelessness with Beck’s Hopelessness Scale (BHS) , general psychiatric symptoms with the Symptom Check List-90 R (SCL-90 R) , suicidal ideation with the five items screening section of Beck’s Scale for Suicidal Ideation (BSS-5)  and general functioning level with the Global Assessment of Functioning (GAF) . The SUAS-S was administered a second time one week after the first administration. Psychiatric diagnoses were made at baseline using the Mini International Neuropsychiatric Interview (M.I.N.I.)  for DSM-IV axis I disorders and the Structured Clinical Interview for DSM-IV Personality Disorders (SCID-II)  for axis II disorders. Interviews and diagnoses were all made by the first author, an experienced psychiatrist. In cases where it was unclear whether criteria for a specific diagnosis were met, the case was discussed with the second author (a professor of psychiatry) and settled by consensus. In cases where the patients had more than one DSM IV axis I or axis II diagnosis, only the main diagnosis on each axis was considered in the analyses. Diagnostic data were collected to provide a good description of this clinical sample. The study was, however, not designed to study the SUAS-S properties in diagnostic subgroups.
Means and standard deviations, or medians and quartiles, were given for sum scores. The chi-squared test was used to assess differences in distribution between the groups for categorical variables. Differences in sum scores of two independent groups were tested by two-sample t-test or Mann–Whitney U test. Internal consistency was computed as the Cronbach’s alpha coefficient, and test–retest reliability was measured with Intra Class Correlation (ICC) with 95% Confidence Intervals (CIs). Associations between the sum scores on the SUAS-S and BSS-5, and their precision, were estimated by Pearson’s or Spearman’s correlation coefficient and bootstrapped 95% CI with 10,000 replications. Discriminatory ability of suicidality for the SUAS-S and BSS-5 sum scores was examined and compared through the use of nonparametric receiver operating characteristic (ROC) curve analyses. The tests were two-sided, and the significance level was set at 0.05. The analyses were performed with STATA 11  and the graphs were made in R 2.11.1 .
The study was approved by the Regional Committee for Medical Research Ethics, South-East Norway. All patients gave written informed consent.