Early warning scoring systems versus standard observations charts for wards in South Africa: a cluster randomized controlled trial

BACKGROUND On South African public hospital wards, observation charts do not incorporate early warning scoring (EWS) systems to inform nurses when to summon assistance. The aim of this trial was to test the impact of a new chart incorporating a modified EWS (MEWS) system and a linked training program on nurses' responses to clinical deterioration (primary outcome). Secondary outcomes were: numbers of patients with vital signs recordings in the first eight postoperative hours; number of times each vital sign was recorded; and nurses' knowledge. METHODS/DESIGN A pragmatic, parallel-group, cluster randomized, controlled clinical trial of intervention versus standard care was conducted in three intervention and three control adult surgical wards in an 867-bed public hospital in Cape Town, between March and July 2010; thereafter the MEWS chart was withdrawn. A total of 50 out of 122 nurses in full-time employment participated. From 1,427 case notes, 114 were selected by randomization for assessment. The MEWS chart was implemented in intervention wards. Control wards delivered standard care, without training. Case notes were reviewed two weeks after the trial's completion. Knowledge was assessed in both trial arms by blinded independent marking of written tests before and after training of nurses in intervention wards. Analyses were undertaken with IBM SPSS software on an intention-to-treat basis. RESULTS Patients in trial arms were similar. Introduction of the MEWS was not associated with statistically significant changes in responses to clinical deterioration (50 of 57 received no assistance versus 55 of 57, odds ratio (OR): 0.26, 95% confidence interval (CI): 0.05 to 1.31), despite improvement in nurses' knowledge in intervention wards. More patients in intervention than control wards had recordings of respiratory rate (27 of 57 versus 2 of 57, OR: 24.75, 95% CI: 5.5 to 111.3) and recordings of all seven parameters (5 of 57 versus 0 of 57 patients, risk estimate: 1.10, 95% CI: 1.01 to 1.2). CONCLUSIONS A MEWS chart and training program enhanced recording of respiratory rate and of all parameters, and nurses' knowledge, but not nurses' responses to patients who triggered the MEWS reporting algorithm. TRIAL REGISTRATION This trial was registered with the Pan African Clinical Trials Registry (identifier: PACTR201309000626545 ) on 9 September 2013.