Differences in the process of care before and after the extended LPS
| January–March 2014 (n=373) | January–March 2015 (n=381) | Difference (abs) | p Value | |
| Psychosocial assessment n (%)* | 213 (57) | 258 (68) | +45 (+11%) | 0.003 |
| Assessor for those who had a, n (%)* | (n=213) | (n=258) | ||
| Psychiatrist | 109 (51) | 77 (30) | −32 (−21%) | <0.001 |
| Liaison nurse | 71 (33) | 166 (64) | +95 (+31%) | |
| Other | 24 (5.9) | 13 (5.0) | −1 (−0.9%) | |
| Unknown | 9 (5.9) | 2 (0.8) | −10 (−5.1%) | |
| Psychosocial assessment by hour of day, n (%)* | ||||
| Monday to Friday 09:00–17:00 hours | 133 (62) | 117 (46) | −16 (−16%) | <0.001 |
| All other times | 61 (29) | 121 (47) | +60 (+18%) | |
| Unknown | 19 (8.9) | 20 (7.8) | +1 (-1.1%) | |
| Median time from ED arrival to psychosocial assessment† | 11 hours 44 min | 8 hours 30 min | −3 hour 14 min | <0.017 |
| Median time from medical assessment to psychological assessment† | ||||
| Overall | 9 hours 30 min (n=185) | 6 hours 53 min (n=230) | −2 hour 37 min | <0.001 |
| Attendances between 09:00 and 17:00 hours | 2 hours 51 min (n=44) | 2 hours 59 min (n=60) | +8 min | 0.078 |
| Attendances during other hours | 10 hours 20 min (n=141) | 8 hours 28 min (n=170) | −1 hour 52 min | 0.003 |
| Admission to ward, n (%)* | ||||
| No | 121 (32) | 117 (31) | −4 (−1%) | 0.110 |
| Yes—observation ward | 212 (58) | 238 (63) | +26 (+5%) | |
| Yes—ITU | 9 (2.5) | 2 (0.5) | −7 (−2%) | |
| Yes—other | 24 (6.6) | 21 (5.6) | −3 (−1%) | |
| Median (p25–75) duration of hospital stay if not admitted (hours) | No data‡ | 12 (7–21) | ||
| Mean duration of hospital stay if admitted (days)§ | 1.7 (4.1) | 1.4 (2.8) | −0.37 | 0.26§ |
| Median (p25–75) duration of hospital stay if admitted to ward/ITU (days)¶ | 1 (1–1) | 1 (1–1) | = | 0.004 |
| Total admission days (including ITU) | 480 | 393 | −87 | |
| Outcome of ED attendance, n (%)* | ||||
| Psychiatric inpatient admission | 11 (2.9) | 11 (2.9) | -/- | 0.96 |
| Crisis team | 46 (12) | 58 (15) | +12 (+3%) | 0.25 |
| Other community team | 27 (7.2) | 58 (15) | +31 (+8%) | 0.001 |
| Self-harm clinic | 57 (15) | 18 (4.7) | −39 (−10%) | <0.001 |
| Alcohol nurse service | 19 (5.1) | 20 (5.2) | +1 (+0.1%) | 0.92 |
| Home/GP care only | 94 (25) | 89 (23) | −5 (−2%) | 0.56 |
| Social services | 0 (0) | 3 (0.8) | +3 (+0.8%) | 0.086 |
| Voluntary agency | 0 (0) | 13 (3.4) | +13 (+3.4%) | <0.001 |
| Custody | 25 (6.7) | 14 (3.7) | −11 (−3%) | 0.061 |
| Died | 0 (0) | 1 (0.3) | +1 (+0.3%) | 0.51 |
| Patients self-discharging before a psychosocial assessment is carried out, n (%)* | 73 (20) | 51(13) | −22 (−7%) | 0.022 |
| Episodes with repeat self-harm<90 days, n (%) | ||||
| Total repeat episodes (min–max)** | 105 (0–19) | 97 (0–9) | −8 (−8%) | 0.79 |
| First repeat after index, n (%) | 48 (18) | 54 (17) | +6 (-1%) | |
| First episode of self-harm, n (%) | 3 (5.6) | 2 (4.3) | −1 (−1.3%) | |
| Previous self-harm, n (%) | 45 (21) | 52 (20) | +7 (−1%) |
*p Value based on Χ2 test.
†p Value based on log-rank test.
‡This element was not documented from the start of the self-harm register.
§p Value based on bootstrap corrected t-test statistic.
¶p Value based on Wilcoxon-Mann-Witney test.
**p Value based on Wilcoxon-Mann-Witney test for # attendances per patient.
††p Value based on Cox regression, adjusted for previous self-harm, sex and suicide risk.
ED, emergency department; ITU, intensive treatment unit; LPS, liaison psychiatric service.