Development and validation of a questionnaire on nurses' knowledge and recognition of early signs of clinical deterioration

Master Thesis


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

University of Cape Town

Introduction: There is evidence-based concern that nurses on general wards do not recognise signs of physiological and clinical deterioration and delay calling for more skilled assistance for review of a patient showing signs of deterioration. Aim: The development and validation of a questionnaire to assess factors influencing general ward nurses’ ability to recognise and respond to patient deterioration; nurses’ knowledge of physiological and clinical parameters associated with patient deterioration; and nurses’ self-reported clinical reasoning ability. Methodology: A mixed methods sequential 4-phase study design was employed: 1) an indepth literature review to identify and develop content domains and item statements for a prototype questionnaire; 2) determining the index of content validity (CVI) (n=5 expert registered professional nurses) of all item statements; 3) conducting cognitive interviews (n=3 expert registered professional nurses) to explore face validity and the quality of the revised prototype questionnaire; and 4) assessing stability of the final validated questionnaire through test-retest reliability testing (n=30 nurses: Registered Professional Nurses with four years of training, Enrolled Nurses with two years of training, Enrolled Nursing Auxiliaries with one year of training) two weeks apart. Results: The CVI exceeded the pre-set proportion of ≥70% agreement for 56/65 (86.2%) item statements scoring 3 (relevant only needing minor editing) or 4 (extremely relevant); removal of 3/65 (4.6%) items from the prototype questionnaire. Cognitive interviews then resulted in amendment of 30/78 (38.5%) item statements; removal of 2/78 (2.6%) from the revised prototype questionnaire. The weighted kappa statistic for level of agreement beyond chance for nurse respondents’ test-retest data was fair (0.21-0.4) for 18/47 (38.3%) items, moderate (0.41-0.6) for 12/47 (25.5%) items and substantial (0.61-0.8) for 13/47 (27.7%) items. Registered Professional Nurses’ responses between time 1 and time 2 were more consistent than for Enrolled Nurses and Nursing Auxiliaries. Conclusion and recommendations: The researcher-developed questionnaire was validated by registered professional nurses, but there is concern about its stability, tested on three categories of nurses. The questionnaire should be reassessed for content and face validity using a sample inclusive of all categories for nurses who take and interpret patients’ vital signs in an attempt to improve the reliability of the questionnaire.