When people fall through the cracks of despair - it is time to make a net of hope: The effectiveness and feasibility of a suicide prevention action network (supranet) in Dutch mental health care

In this thesis, chapter 2 presents the protocol of the SUPRANET Care implementation study. The SUPRANET Care study (N=13) uses an uncontrolled longitudinal prospective design, collecting data on on the following three study outcomes by using a 2-level structure (institutional level and team level): 1] reduced suicide rates, 2] improved registration of suicide attempts, and 3] improved professionals’ knowledge and adherence towards the suicide prevention guideline. In chapter 3, a Delphi study is described. An expert panel of 90 participants scored 11 quality indicators (QIs) on two criteria (relevance and action orientation). Data analysts (n=6) from the participating institutions rated the QIs on a third criterion: feasibility. We found that 5 QIs were scored as relevant and action oriented: 1] screening for suicidal thoughts and suicidal behavior, 2] safety plans, 3] early follow-up on discharge, 4] continuity of care, and 5] family- or caretaker involvement. Only one indicator reached consensus on feasibility: early follow-up on discharge. In chapter 4 of this thesis, a total of 400 mental health care professionals were included, working in 22 teams (crisis teams and ambulatory care teams) across 12 institutions. Although the majority of mental health professionals (N=400) scored relatively high on professionals’ knowledge, and attitude during baseline, most of them were not aware to what extent the guideline was used by their colleagues (i.e. adherence towards the guideline). After the data was analyzed on team level, we found that crisis teams scored significantly better on professionals’ knowledge (B(95%-CI)=2.00(1.32-2,67); p<0.001), attitude (B(95%-CI)=0.48(0.19-0.77); p< 0.001), and guideline adherence compared to ambulatory care teams (B(95%-CI)=2.56(0.95-4.17); p=0.002). Finally, our findings also showed that adherence towards the guideline significantly differed between institutions. In chapter 5, we found no significant differences in suicide rates between both groups relative to baseline. However, we did find that institutions labeled as above average registered significantly more suicide attempts during the first six months of the intervention (78.8 registered suicide attempts per 100.000 patients; p=0.001,95%CI= 51.3 per 100.0000, 106.4 per 100.000) and throughout the rest of the study period, they continued to perform significantly better on this outcome compared to institutions assigned as below average (8,7 registered attempts per 100.000 patients per half year; p=0.004,95%CI= 3.3 per 100.000, 14.1 per 100.000). As for the third study outcome, institutions classified as above average showed a significantly stronger improvement in professionals’ adherence over time (B(95%-CI)=1.39(0.12,2.65);p=0.032) compared to institutions that peformed less well. In Part II, chapter 6 we described a meta-analysis including 18 studies (N=5380), and found that guideline-based treatments significantly reduced the severity of psychopathological symptoms compared to treatment-as-usual (TAU) (d=0.29,95%-CI=(0.119-0.40); p< 0.001). Patients also recovered more quickly if they were treated according to the guideline recommendations (HR=1.54, 95%-CI=(1.29-1.84); p=0.001). These results imply that patients did not only benefit more of guideline-adherent treatments vs TAU, but that this improvement was also reached more quickly. Furthermore, we found that providers in the guideline condition were significantly more adherent to the guideline recommendations compared to providers in the TAU-condition (secondary outcome) (d=0.18,95%-CI=(0.02-0.34); p=0.025). Finally, in chapter 7, two network meta-analyses (NMAs) were performed, including 43 studies (N=3273). It appeared that ST, MBT and DBT were more efficacious in the treatment of borderline personality disorder (BPD) compared to GT, and TAU. We also found significant differences between several active comparisons, however these findings were based on very few trials and these results should be interpreted with caution. Several treatments (more than) halved the risk of (fatal) suicide attempts, reporting RRs around 0.5, however, these results were not significantly better compared to TAU or to other psychotherapies.