Introduction: The Antimicrobial Self-assessment Toolkit for Acute NHS Trusts (ASAT) was developed by a pharmacist reference group of an Advisory Non-Departmental Public Body on Antimicrobial Resistance and Healthcare Associated Infections (ARHAI). It was developed in conjunction with the Department of Health. The primary purpose of the ASAT is to identify and to measure the methods of implementation of antimicrobial stewardship programmes in acute NHS trusts. The face validity was previously tested by ARHAI. The overall aims of this programme of work were to investigate the validity of the ASAT and to make iterative changes to improve its validity. Ethical approval was not required for this PhD project because it was categorised as service evaluation by the LREC. Also, ethical approval from the University of Manchester Research Ethics Committee was deemed unnecessary at the time of the PhD project due to the nature of the data collected. Methods: A mixed methodology approach utilising a sequential exploratory strategy was used to investigate the validity of the ASAT. This PhD project was composed of four sequential studies which resulted in iterative changes to the ASAT, that is, from ASAT v15a to ASAT v18. In Study 1, cognitive interviews were conducted with eight antimicrobial pharmacists in order to investigate the content validity of ASAT v15a. In Study 2, both cognitive interviews and semi-structured interviews were conducted with 10 clinical microbiologists in order to investigate the content validity of ASAT v16. In Study 3, Rasch modelling and analyses using the Partial Credit Model (PCM) were conducted on the responses to ASAT v17 from 33 NHS trusts across England. In Study 4, simple OLS regression analyses were conducted using the NHS trust ‘ability’ estimates or calibrations and Clostridium difficile (CDI) rates of participating NHS trusts in order to investigate model fit and the predictive validity of the ASAT. Results: The cognitive interviews conducted in study 1 indicated that AMPs encountered cognitive difficulties along the cognitive processing pathway in response to ASAT v15a. These difficulties included comprehension in 27 (32.5%) questions and response generation/formatting in 13 (15.7%) questions. Also respondents indicated that the role of clinical microbiologists in ASPs was underrepresented in ASAT v15a. The interviews conducted in Study 2 were confirmatory in nature as they reflected the findings of Study 1. For example terms such as ‘formulary’ and ‘policy’ were misinterpreted by respondents. Rasch modelling and analysis showed that there were items within ASAT v17 which were underfitting and overfitting the Partial Credit Model. Item fit was investigated after removal of these items which resulted in improved fit for domains 2 and 5. ASAT v18 was developed after these analyses and was included items that were productive for measurement. On examination of the OLS regression analyses conducted in Study 4, it was seen that there was poor model fit and very limited predictive validity of the model. Conclusion: The iterative methodology utilised to investigate the validity and subsequently improve the ASAT was effective in establishing content and construct validity. However, the predictive validity of the ASAT was limited. This may be due to the outcome variable chosen for the OLS regression modelling. A more sensitive outcome measure such as compliance to treatment or prophylaxis guidelines may have been more effective at establishing predictive validity. The findings of this programme of work highlighted that there is further work required to validate the ASAT such as the determination of the appropriate weights and scores for ASAT domains and also the determination of the appropriate outcomes measures to determine the efficacy of ASPs. It is recommended that further validity testing should be conducted before a further iteration of the ASAT is used as a set of quality standards or as a hospital benchmarking tool
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:588132 |
| Date | January 2013 |
| Creators | Bailey, Chantelle |
| Contributors | Tully, Mary; Cooke, Jonathan |
| Publisher | University of Manchester |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | https://www.research.manchester.ac.uk/portal/en/theses/the-evaluation-of-the-antimicobial-selfassessment-toolkit-for-nhs-trusts(4d6f04ab-0011-48e3-818c-dd8c878e69ab).html |
Page generated in 0.0018 seconds