Background:Implementation clinicians/researchers often use pre-implementation Likert scale questionnaires aligned to the Theoretical Domains Framework (TDF) to identify clinician-perceived barriers. Results are often reported as means which do not typically assist with diagnostics. The questionnaires are rarely re-administered post-implementation. We present our novel approach to categorising clinicians’ perceptions of care as barriers/enablers, and the changes observed post-implementation.
Methods:Our questionnaire comprised 57 Likert scale questions (Q) regarding continence care, mapped to 13 of 14 TDF domains. Responses were categorised into two groups: positive (agree/strongly agree) and neutral/negative (unsure/disagree/strongly disagree). Percentages for each question was calculated, labelled and colour-coded based on response rates: barrier (<60% positive responses/red), neutral (60-69% positive/orange), or enabler (>69% positive/green). Pre- to post-implementation changes were evaluated.
Clinicians from fifteen adult wards that adopted the SCAMP practice-change package (12 Australian hospitals: metropolitan=4, regional=8) were invited to complete the online questionnaire both before and immediately after the 6-month implementation phase. The practice-change package focused on inpatient clinicians, employing strategies to address barriers identified pre-implementation.
Results:Clinician participation pre-implementation: hospitals=12/12, n=291, female=82%, nurses=88%. onset. Post-implementation: hospitals=10/12 (no implementation phase at 2/12 hospitals as coincided with onset COVID-19), n=195, female=86%, nurses=94%.
Post-implementation seven TDF domains (25/27 questions) had improved by 11-30% (p-value<0.05): skills (4/4), knowledge (6/6), memory (1/1), beliefs about capabilities (2/2), emotion (3/3), environmental context/resources (6/7), and social/professional role (3/4). Three domains were high-scoring enablers at both timepoints (≥87% positive, p-value>0.05): goals (10/10), beliefs about consequences (6/6) and reinforcement (1/1). Three domains were mainly unchanged (p-value>0.05): social influences (4/4), intentions (5/6) and behaviour regulation (3/4).
Conclusion:Categorising and colour-coding responses offered a practical, visual method for stakeholders to interpret Likert data, prioritise barriers to address, identify changes in perceptions post-implementation, help explain which implementation strategies appeared successful in changing practice and which ones need refining.
Implications:Our novel methods are useful for clinicians/researchers driving practice-change.