Table 2 Potential mechanisms and effects of clinical performance comparators and their theoretical and empirical support

Benchmarks

Increases feedback effectiveness by reducing complexity (enabling comparison with others enables recipients to better understand how well they are performing and which areas require improvement) and increasing social influence (by harnessing competition between recipients, changing recipients’ behaviour if they see others behaving differently, and trying to maintain their status in a group of high performing clinicians).

Theories (n = 4): Social Comparison Theory [31], Persuasion Theory [27], Social Norms Theory [33], Reference Group Theory [32].

Qualitative studies (n = 12): [34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52].

RCTs (n = 2): [53, 54]

Debilitates feedback effectiveness by directing attention away from the performance task at hand (e.g. prescribing appropriate medication) which allows recipients to explain away potentially bad performance if overall performance is low.

Theories (n = 1): Feedback Intervention Theory [7]

Induces both positive and negative emotions dependent on whether relative performance level is high or low respectively by increasing competition through social influence.

Theories (n = 1): Social Comparison Theory [31].

Qualitative studies (n = 7): [39, 49, 55,56,57,58,59].

Benchmarking against a reference group considered irrelevant or unfair by recipients (e.g. due to case-mix difference or inadequate statistical adjustment in outcome measures) inhibits feedback acceptance by decreasing credibility and perceived validity.

Theories (n = 1): Reference Group Theory [32].

Qualitative studies (n = 8): [36, 39, 40, 51, 52, 61,62,63].

Benchmarking against values that reflect mean or median performance inhibits action by limiting recipients’ perception of room for improvement (e.g. comparing against the mean only demonstrates discrepancies to half of recipients).

Theories (n = 2): Control Theory [9], Goal-setting Theory [13].

Qualitative studies (n = 3): [35, 59, 68].

RCTs (n = 3): [65,66,67].

Benchmarking against values (e.g. the 90th percentile) inhibit feedback acceptance by low performers if they consider the discrepancy too large and unachievable.

Theories (n = 1): Goal-setting Theory [13].

Qualitative studies (n = 2): [35, 62].

Benchmarking against identifiable individual peers may increase effectiveness because recipients can choose the most relevant peers for comparison and increases their sense of competition knowing that their own performance is reported to others.

Theories (n = 2): Social Comparison Theory [31], Reference Group Theory [32].

Benchmarking against identifiable individual peers inhibits feedback acceptance when recipients consider (semi)public reporting of their own performance inappropriate and a threat to their autonomy.

Qualitative studies (n = 5): [44, 48, 61, 71, 72].

Multiple benchmarks (multiple groups or values, or individual peer scores) facilitates feedback acceptance by increasing credibility because it helps recipients assess variation between professionals and judge whether potential discrepancies are clinically significant.

Theories (n = 2): Feedback Intervention Theory [7], Social Comparison Theory [31].

Qualitative studies (n = 6): [37, 40, 57, 59, 73, 74].

Multiple benchmarks allow recipients to make downward social comparisons (defensive response to feel better about themselves) instead of upward social comparisons which inhibit action.

Theory: Social Comparison Theory [31].

Trends

Facilitates action by decreasing the complexity in a way that helps recipients interpret and identify when clinical performance requires action, in particular, if the reference period includes sufficient time points at regular intervals dependent on the performance topic and number of observations each interval.

Theories (n = 1): Feedback Intervention Theory [7].

Qualitative studies (n = 11): [37,38,39, 44, 46, 50, 51, 55, 77,78,79,80,81,82,83].

Increases the observability of the feedback intervention which induces positive emotions by demonstrating how recipients’ clinical performance has improved over time as a consequence of their taken actions; higher improvement rates being associated with higher satisfaction.

Theories (n = 2): Feedback Intervention Theory [7], Johnson et al. [30].

Qualitative studies (n = 7): [44,45,46, 77,78,79,80].

Facilitates acceptance of feedback by increasing its credibility because performance is measured during a reference period that includes multiple time points (e.g. to eliminate the possibility of one-time coincidentally low performance).

Qualitative studies (n = 2): [39, 45].

Explicit targets

Facilitates action by reducing complexity of the feedback, making it easier for recipients to know what constitutes ‘good performance’ and therefore what requires a corrective response.

Theories (n = 3): Control Theory [9], Goal-setting Theory [13], Feedback Intervention Theory [7].

Qualitative studies (n = 2): [84, 85].

Targets from an external source that lacks power or credibility inhibit acceptance of negative feedback by inducing creates cognitive dissonance; recipients may respond by rejecting the target/feedback to resolve this dissonance and maintain the perception of self-integrity, rather than question their own competency as a clinician.

Theories (n = 4): Ilgen et al. [25], Cabana et al. [26], Theory of Cognitive Dissonance [28], Self-affirmation Theory [29].

Qualitative studies (n = 2): [68, 84].

Self-set targets (i.e. source is feedback recipients themselves) increase goal commitment and progress towards the target, but recipients may choose inappropriate targets (i.e. too low or unachievably high) to eliminate the discrepancy or because they do not know how to set targets.

Theories (n = 1): Goal-setting Theory [13].

Qualitative studies (n = 2): [85, 86].

Ambitious target values increase feedback effectiveness over simple targets as long as they are (considered) achievable.

Theories (n = 2): Goal-setting Theory [13], Feedback Intervention Theory [7].

Absolute target values are simple (decreasing complexity) than relative targets but can become outdated when achieved by most recipients which inhibits continuous quality improvement.

Theories (n = 1): Control Theory [9].

Relative targets based on benchmarking facilitate continuous quality improvement as can be automatically adjusted when the group performance changes, but also inhibits action because it creates uncertainty to recipients as to which performance levels should be targeted.

Qualitative studies (n = 1): [72].

Relative target values based on benchmarking inhibit feedback acceptance if recipients consider them unfair, in particular, if performance is just below target and variation between peers is small and clinically insignificant.

Qualitative studies (n = 2): [59, 84].