Table 3. Key findings.

Initial programme theory in the form of ‘If-Then’ statements in relation to the CMO configurations (observed associations), key quotations and reflections of the focus group meeting.

Initial programme theory	CMO configuration	Key quotations from the included studies	Reflections of the focus group meeting
Theory 1. If the organisation has appropriate systems and skills in place to monitor, track and evaluate the results of an audit (both anticipated and unanticipated) then the improvements are more likely to be incorporated and sustained [120].	CMOc1. Externally initiated audits create QI awareness although their impact on improvement diminishes over time	“Once the on-site survey concluded, improvements continued, but the rate levelled off. Essentially, the positive trend prior to accreditation increased during accreditation and continued post-accreditation, but it began to plateau.” [51] (p.5)	The focus group confirmed this pattern. The focus group noted that external audits are often experienced as an imposed way to comply with external requirements. The group confirmed that the number of years that an organisation participated in external audits does negatively affect the extent of changes that take place; they had the experience that, after about three years from initiating the audit, the rate of improvements tailed off: “The audit has, especially at the start, considerable effect. The conversations are in fact new, and they [healthcare professionals and organisational members] talked for the first time with someone [the auditor] who could help them to have a closer look at how they are performing and this person could help them further.” [Participant 1]
Theory 2. If an audit focuses on continuous improvement rather than on assurance and compliance then the audit is more proactive and facilitates change [13, 121].	In a context where the audit is mandated by an external party (C), organisational members are instructed to participate in the audit process and collect data for the audit to be able to comply with the requirements (Mresource). Although these audit activities are imposed, the interest and awareness of healthcare professionals and organisational members in QI are raised (Mreasoning), even after the external pressure has diminished (C). After a couple of years, the audit no longer provides a challenge to healthcare professionals and organisational members to improve quality (Mreasoning) and the rate of improvement levels off (PO).
Theory 3. If healthcare professionals are actively involved in audit processes and recognise the need for improvement then they identify themselves with the findings of an audit which makes them more likely to accept the findings and, where appropriate, implement change [5, 122, 123].	CMOc2. A sense of urgency felt by healthcare professionals triggers engagement with an audit	“The present inquiry began when the senior author reviewed a day in the OR on which cases started late, families expressed frustration with long preoperative stays at the hospital for short procedures, the OR finished later than scheduled, and little operation had occurred because the room was often not occupied by a patient.” [82] (p.783)	CMO configuration not discussed during the focus group meeting
	In a context where an urgency to improve quality is present (C), healthcare professionals undertake actions and initiate local audits as an improvement instrument (Mresource), are willing to participate in the audit as they feel engaged (Mreasoning), and recognise the need for improvement (Mreasoning), thereby increasing the likelihood of actual improvements in patient care (PO).	“The organizational focus remained to be a source of frustration. Twenty interviewees expressed their concerns on having to repeat all the organizational items in the self-review and site visit in a fourth participation round. As a result, only 12 persons would still find a fourth participation worthwhile without major changes in the program.” [106] (p.483)
Theory 4. If the driving forces behind an audit are champions who advocate the audit then this increases the willingness of healthcare professionals to adapt and accept suggested improvements, which results in greater adoption of the improvements by individual healthcare professionals in an organisation [7, 120].	CMOc3. Champions are vital for an audit to be perceived by healthcare professionals as worth the effort	“A group of internal medicine residents, led by a resident champion, considered several potential quality goals (…). A retrospective chart review of 100 randomly selected patients discharged from the medical service in June 2009 revealed that communication with primary care physicians was documented in only 55% of cases. Therefore, the residents challenged their colleagues to increase the rate of documented primary care physicians contact to 80% or higher.” [69] (p.473)	CMO configuration not discussed during the focus group meeting
Theory 5. If an organisation has leadership committed to quality of care then this positively influences the organisational culture and thus the implementation of quality improvement measures [7].	A healthcare professional who is widely recognised as a champion (C) and who is adequately supported by the organisation (C) acts as a role model for the espoused change (Mresource) and challenges colleagues to implement quality improvements in practice (Mresource). Healthcare professionals consider the audit relevant since one of their peers is the driving force, and this fosters trust (Mreasoning). This results in increased attention to the quality of delivered care (PO) and a stronger commitment to QI by healthcare professionals (PO).
Theory 6. If an audit is initiated in a bottom-up way by healthcare professionals then this allows ownership and autonomy, which fosters trust, goodwill and confidence such that implementation and sustained surveillance of the suggested improvements are more likely to occur [124].	CMOc4. Bottom-up initiated audits are more likely to bring about sustained change	“A clearer job description, better communication between professionals and more unambiguous agreements on how to arrange postoperative care were mentioned as being important subjective outcome measures. These logistic alterations and specific arrangements on how a professional would provide care or information to the patient, are difficult aspects to evaluate by objective (hard) end-points, but improvement in this area is of great importance when improving quality of care is the main goal.” [80] (p.275)	The focus group acknowledged the importance of respecting each other’s competences and being sensitive to each other’s work and needs by healthcare professionals. They also recognised improved reflection, feedback and collaboration as an outcome: “Regarding clinical audits that are performed within professional groups, I noted that these reflections strengthened the understanding among professionals because they can discuss with each other the outcomes of the audit and, as a result, this boosts solidarity” [Participant 2]
	In a context where healthcare professionals work closely together, respect each other’s competences and are sensitive to each other’s work and needs (C), they are able to define quality improvements (Mresource) that are recognised as valuable by all healthcare professionals involved (Mreasoning), ensuring appropriateness and fostering acceptance of sustainable improvements in practice (Mreasoning). Increased and improved communication between healthcare professionals (PO) favours reflection and feedback (Mreasoning) that will result in improved multidisciplinary care collaboration (PO).
No initial programme theory	CMOc5. Knowledge-sharing within externally mandated audits triggers participation by healthcare professionals	“At the beginning of the self-assessment phase, staff seated around the table had divided into three groups, each of which spoke to the moderator but not to the other groups. By the end of the self-assessment phase, staff from different sites [of the organisation] sat in mixed groups around the table. They also exchanged protocols, discussed means of implementing common working procedures, and collaborated on better integrating the patient pathway within the organization.” [83] (p.8)	The focus group acknowledged the importance of knowledge sharing among healthcare professionals. The clinical relevance of the audit topic was an important factor for healthcare professionals to decide to participate: “What do we need in order to deliver high-quality care, good education and to conduct good research? If that’s your starting point, then everyone is willing to contribute.” [Participant 3]
	In a context in which the audit is mandated by an external party (C), healthcare professionals are brought together to exchange ideas and knowledge (Mresource), which makes them to see the audit as a learning opportunity (Mreasoning), thus increasing their willingness to participate (Mreasoning). This will result in better understanding of the challenges facing the organisation (PO), changes being more quickly implemented and spread throughout the organisation (PO) and improved communication within and between teams (PO).
No initial programme theory	CMOc6. Audit data support healthcare professionals in raising issues in their dialogues with those in leadership positions	“Not only did the accreditation recommendations cause management to adjust and modify many practices, staff also used them to convince management and the Board of Directors to adopt particular measures.”[83] (p.9)	The focus group recognised that, within a safe organisational culture, healthcare professionals feel free to raise their concerns: “I think that if there’s a safe culture, that they [healthcare professionals] actually feel empowered, and that they dare to share issues and to discuss data of audits with leaders.” [Participant 4]
	In a context of a safe organisational culture (C), healthcare professionals are able to substantiate their requests for changes using audit data (Mresource) and feel strengthened in their dialogue with organisational leaders to convince them of the need for change in practice (Mreasoning). These requests for changes by healthcare professionals will result in organisational leaders adopting a more active approach to QI (PO).
No initial programme theory	CMOc7. Audits legitimise the provision of feedback to colleagues, which flattens the perceived hierarchy and encourages constructive collaboration	“I would never have told a MD [Medical Doctor] to wash his or her hands without this [hand hygiene] tool” and “It [the audit] gave me justification to tell others to follow [hand hygiene] policy.” [72] (p.52)	The focus group agreed that a safe learning climate is a prerequisite if healthcare professionals are to address each other’s performance. In their view, the audit can be an instrument that facilitates an open discussion of quality of care and creates greater awareness of a certain policy: “If you normally feel you are not in a position to say something about a certain subject, then the audit has the effect of empowering you to do something about it. (…) The audit facilitates discussing issues that otherwise would not have occurred.” [Participant 3]
	In a context of a safe learning culture (C), healthcare professionals are encouraged to give feedback to each other during the audit (Mresource). Professionals feel that a specific audit finding justifies (Mreasoning) and empowers them in giving feedback to other staff for non-adherence (Mreasoning). Providing feedback flattens the existing perceived hierarchy (PO) and results in better team collaboration (PO).