The development and validation of a modified Situation-Background-Assessment-recommendation (SBAR) communication tool for reporting early signs of deterioration in patients

Master Thesis

2015

Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher

University of Cape Town

License
Series
Abstract
Background: Errors in communication are prevalent in healthcare and affect patient safety and cause unnecessary patient deaths. Reporting early signs of physiological or clinical deterioration could improve patient safety and prevent 'failure to rescue' or unexpected intensive care admissions, cardiac arrest or death. The structured Situation-Background-Assessment-Recommendation (SBAR) communication tool enables nurses to provide doctors with pertinent information about a deteriorating patient in a logical order, based on a complete assessment. In addition, nurses have increased confidence in their findings and are better able to initiate a call and to convince a doctor to provide orders promptly or see a patient. Aim: The aim of this sub-study of a randomized controlled trial was to develop and validate a modified SBAR communication tool incorporating components of a local MEWS vital signs observations chart. Methods: The modified SBAR communication tool was developed following a review of available published examples and validated by employing a mixed methods approach: 1) cognitive interviews (n=3 nurses, 2 doctors), 2) determining the index of content validity with nurses (n=5), physicians (n=5) and surgeons (n=8) and 3) inter-rater reliability testing, with calculation of kappa values (n=2 nurses). Results: Cognitive interviews prompted more changes to the modified SBAR communication tool than determined by the content validity index. For cognitive interviews, there were 15/42 (35.71 %) modifications: 11 items were added (26.19 %) and three removed, (7.14 %) resulting in 49 items whereas for content validity index there were 4/49 (8.16%) modifications, 5/49 (10.20%) items removed and one item added (2.04%). Four of 49 items (8.16%) rated as relevant by <70% of nurses and doctors were revised or deleted. No additional modifications were needed following review by surgeons, as all items were rated as relevant by the pre-determined ≥70% of experts. Inter-rater reliability of the SBAR tool was established by two nurses who were mostly in substantial to full agreement on 37/45 items on the modified tool. The exceptions were: 'Calling from' (Cohen's Kappa-0.05) and 'this is a change from' (Cohen's Kappa-0.07), representing agreement below the level of chance. However, the high percentage agreement and nature of the questions suggest that the questions are sound. Percentage agreement amongst participants for these items was 91 % (95% confidence interval (CI): 71 to 99 ) and 86% (95% CI: 65 to 97 ) respectively. Deciding whether a doctor should see the patient now (Cohen's Kappa 0.09) or in the next 30 minutes, achieved fair agreement (Cohen's Kappa 0.20). This reflects a difference in clinical judgement as the decision when to call for assistance depended on the individual nurse's clinical judgement. IRR was not possible to test on 4/45 items, as those items required a response by the person being summoned. Overall, nine of 42 items were removed, 12 were added and 19 substantially modified, leaving 45 items. Conclusion: The modified SBAR communication tool was valid and reliable for use in a local context in conjunction with the Cape Town Modified Early Warning Score (MEWS) vital EWS) vital signs chart.
Description

Includes bibliographical references

Keywords

Reference:

Collections