Does 'Uncertainty Tolerance' Influence Clinical Decisions Regarding Antibiotics? Informing Stewardship for Healthcare Staff in Aged-Care Settings

Antibiotic overprescribing continues to be a problem in residential aged-care settings, with at least 50% of prescribing considered to be inappropriate. High levels of diagnostic uncertainty, lack of diagnostic testing, time constraints and pressure from families shape risk perceptions of staff involved in providing care and have been linked to prescribing antibiotics ‘just in case’. The primary aim of this thesis was to investigate the role of uncertainty and its tolerance in influencing the inappropriate initiation of antibiotics in residential aged-care settings.

Study 1 involved a systematic review of the literature regarding aged-care staff attitudes relevant to antimicrobial prescribing and stewardship. Multiple domains of attitudes, lack of theory defining these domains and large variations in measurement of these domains complicated the synthesis of findings. Attitudes were predominantly captured through beliefs and endorsement of behaviours and measures pertaining to affective responses were lacking. Additionally, perceptions of professional risk (e.g., deregistration) regarding reducing antibiotics have not been measured, despite fear regarding malpractice being associated with prescribing antibiotics ‘just in case’.

Research points to the key role of the registered nurse in driving clinical decisions in the context of providing clinical assessments, monitoring residents and coordinating care between multiple stakeholders in aged-care facilities. As such, Study 2 aimed to develop a measure that quantified nursing behaviours under conditions of uncertainty related to common infections in the aged-care context. A new vignette-based measure was developed and administered to 157 nurses. Factor analysis identified three scales, including a 4-item factor labelled as Immediate Assessment and Antibiotics (IAA). The IAA scale demonstrated significant positive correlations with the domain-specific or clinical tolerance of uncertainty but not general tolerance of uncertainty. Using the same sample, Study 3 then explored the relationship between situational anxiety, clinical tolerance of uncertainty and the clinical decisions favouring antibiotics. Nurses’ anxiety fully mediated the relationship between tolerance of uncertainty and perceived need for antibiotics and assessment, however only 6% of variance was predicted by the model. Taken together, findings from Studies 2 and 3 provide evidence supporting the influence of affective states, particularly situational anxiety arising from uncertainty in influencing some of the antibiotic initiation in RACFs.

Lastly, Study 4 involved interviews conducted with 16 aged-care nurses regarding the management of their uncertainty, within the context of antibiotic prescribing and stewardship. The theoretical framework of medical uncertainty proposed by Han et al. (2021) was used to map strategies engaged by nurses in the aged-care setting. Nurses described relying on ignorance-focused strategies (e.g., initiating diagnostic evaluation, consulting colleagues) to manage issues of scientific uncertainty, while uncertainty-focused strategies (e.g., maximizing attention, decision deferral) were used to navigate issues of both, scientific and practical uncertainty. Response (e.g., cultivating virtues) and relationship-focused strategies (e.g., sharing with families) emerged as important in managing issues of practical and personal uncertainty, particularly in sharing burdens associated with uncertainty and decision-making. A smaller number of nurses also described using response-focused strategies of ‘reflection’ and ‘avoidance’ to manage competing priorities in their roles that were previously not described in the existing taxonomy. The ongoing and iterative nature of managing risk was evident in the wide variety of strategies that nurses described using under conditions of uncertainty.

Together these studies demonstrate the importance of clinicians’ affective states, particularly anxiety, in influencing clinical decisions associated with the inappropriate initiation of antibiotics in aged-care facilities. The role of affect has tended to be neglected in designing stewardship interventions to date, with a focus on educational stewardship that reduces scientific uncertainty. Aged-care staff describe aversive affective states, such as anxiety, frustration and guilt and they fear consequences (e.g., worsening of symptoms, complaints from families) when facing scientific, practical and ethical issues of uncertainty in making clinical decisions. It is therefore unsurprising that they are required to draw on several strategies to manage the different types of uncertainty experienced when providing care for residents. Findings from this thesis indicate the need for acknowledging the impact of domain-specific tolerance of uncertainty on situational anxiety that nurses experience and its’ potential for influencing clinical decisions related to antibiotics. As such, it may be helpful to support clinicians in their decision-making through the inclusion of interventions that explicitly target the negative states arising from uncertainty and increase clinicians’ tolerance of uncertainty in the clinical context.