Comprehensive Geriatric Assessment in the perioperative setting; where next?

Abstract

Comprehensive Geriatric Assessment (CGA) is being employed in the perioperative setting to improve outcomes for older surgical patients. Traditionally CGA is delivered by a geriatrician led multidisciplinary team but with the acknowledged workforce challenges in geriatric medicine, it has been suggested that non-geriatricians may be able to deliver CGA. HOW-CGA developed a toolkit to facilitate the delivery of CGA by non-geriatricians in the perioperative setting. Across two hospital sites uptake and implementation of this toolkit was limited by a potential lack of face validity, behavioural and cultural barriers and an acknowledgement that geriatric medicine expertise is key to CGA and optimisation. In-keeping with this finding there has been an observed expansion in geriatrician led CGA services for older surgical patients in the UK. In order to demonstrate the effectiveness of perioperative CGA services, implementation science should be combined with health services research methodology and the use of big data through linked national audit.

Comprehensive Geriatric Assessment (CGA), perioperative, models of care, quality improvement, implementation science, older people

Key points

• Comprehensive Geriatric Assessment and optimisation (CGA) is traditionally delivered by geriatrician led multidisciplinary teams.

• Evidence supports the use of CGA methodology in the perioperative setting to improve clinical and patient reported outcomes for older surgical patients.

• The Hospital Wide Comprehensive Geriatric Assessment (HOW-CGA) team developed a toolkit to facilitate delivery of CGA by non-geriatricians in perioperative settings but when trialled this did not result in a change in clinical practice.

• Uptake and implementation of the HOW-CGA toolkit may have been impacted by limited face validity, behavioural and cultural barriers.

• HOW-CGA in the perioperative setting supports the need for geriatrician delivered CGA.

Commentary

Comprehensive Geriatric Assessment has increasing legitimacy as an evidence based approach to reduce institutionalisation, minimise time spent in hospital and improve life expectancy for older people [1]. These benefits have now been observed in several patient groups; community dwelling older people, medical inpatients and those undergoing surgery [1–3]. Implementing CGA based services across a variety of clinical settings has resource implications, particularly in terms of building an appropriately skilled workforce. The multidomain process of CGA is traditionally described as a geriatrician led, multidisciplinary delivered intervention. However, with the acknowledged workforce challenges in geriatric medicine, alternative ways of delivering CGA are necessary. The paper by Kocman et al. in this edition of the journal is timely as it aims to understand the implementation of CGA by non-geriatricians in the perioperative setting [4].

The authors of this paper should be commended for the use of a wide stakeholder group including clinicians, researchers and patients to co-design a comprehensive toolkit aiming to facilitate the delivery of CGA by non-geriatricians [4]. Furthermore, the evaluation was enhanced through the application of innovative methodology (Normalisation Process Theory or NPT). NPT provides a framework to develop, embed and evaluate the process of implementing complex interventions, with the aim of achieving sustainable delivery in clinical practice. It involves four pillars: coherence (can the intervention be clearly articulated?), cognitive participation (does the intervention make sense to those responsible for implementation?) collective action (will people rally round implementation and what are the barriers and enablers?) and reflexive monitoring (can and how will the intervention become so embedded that people forget what life was like previously?). Such methodology is critical to effective implementation and evaluation of complex healthcare interventions, especially at scale across multiple sites.

Kocman et al. co-designed a toolkit with a strong emphasis on theory and evidence using links to implementation techniques and evaluation tools such as run charts [4]. Toolkits have an established and clear role in knowledge translation, provided they are informed by quality evidence and theory and their impact is evaluated using rigorous mixed method studies to elucidate the factors underlying successful implementation [5]. Despite the robust approach to toolkit development in the accompanying study, the impact was limited by the failure of one site to engage with the toolkit, and the second site failing to routinely use the toolkit in the clinical pathway. The reluctance of site leads and participants to use the toolkit may reflect the volume of information it included, which may have made it difficult for clinicians to identify and use relevant sections. These findings have resonance with other studies. For example, a nested process evaluation of the EPOCH trial (which used a stepped wedge cluster randomised design in over 90 hospitals to evaluate the impact of a quality improvement intervention to support implementation of a complex pathway for patients undergoing emergency abdominal surgery) showed that despite delivery as planned at cluster level, intervention fidelity at hospital level was variable [6]. Self-assessment by local lead clinicians cited multiple reasons for this including difficulty in engaging colleagues and limited resources, both time and organisational. In contrast, the use of care bundles to engage clinicians and implement change has been effective in other settings [7]. This may be due to the face validity and brevity of care bundles which are limited to three to five components. As such, further refinement of the toolkit used in Kocman’s study, to include a separate section providing concise and practical resources to engage and support clinicians, may have facilitated better uptake. In addition, the presentation of the toolkit in this work may have implied a linear process to implementation resulting in a stalling of progress when challenges were encountered. The revised MRC framework instead advocates a non-linear approach to developing and evaluating complex interventions and may have provided a more appropriate strategy [8].

The work by Kocman et al. goes on to describe the key enablers and challenges in implementing CGA by non-geriatricians [4]. Acknowledging that the first enabler is to ‘win hearts and minds’, the authors cite coherence between participants in terms of first, recognition of the complexities in operating on older patients and second, the utility of CGA in addressing these issues. However despite the assertion that clinical leads and their colleagues needed ‘little convincing of the worth of CGA’, the results imply that clinicians did not believe in the power of CGA as a tool in the perioperative setting and could not reconcile delivery of preoperative CGA in a time-pressured surgical pathway. The lack of legitimisation regarding perioperative CGA, is clearly illustrated through the description of CGA as a ‘smouldering candle’ instead of a ‘burning platform’. This is in conflict with emerging evidence that timely perioperative CGA can positively improve outcomes [3, 9], and may have represented the fact that the case for change was not fully articulated or heard. The authors describe a further challenge; helping clinicians appreciate the role that CGA can play in the preoperative setting. Non-geriatricians may perceive the utility of preoperative CGA as describing a patient as ‘fit/not fit for surgery’ in a binary fashion, whereas evidence suggests that preoperative CGA can be used to holistically optimise patients thus modifying perioperative risk, informing shared decision making and improving longer term health outcomes [3, 9]. This disconnect between the perceived and potential role of CGA likely persists due to several factors. First, organisational issues such as treatment targets (‘just focussed on getting these patients through and hitting cancer targets’); second, consideration of the perioperative period as a discrete episode of care without appreciating it as a punctuation of longer term condition pathways; third, failing to recognise the opportunity offered by the perioperative period, to detect and optimise chronic conditions, and fourth, a lack of awareness of emerging evidence supporting the use of CGA in the perioperative setting.

The evidence supporting perioperative CGA now encompasses both emergency and elective settings. A Cochrane review of preoperative CGA, included seven papers in emergency repair of hip fracture and pooled analysis demonstrated a positive impact on mortality, length of hospital stay, institutionalisation and probable reduction in healthcare delivery costs [3]. In the elective setting, a randomised clinical trial examining preoperative CGA in vascular surgery demonstrated reduced postoperative complications and reduced length of stay [9]. In contrast, a randomised controlled study of preoperative CGA in elective colorectal surgery showed no benefit on the primary outcome of postoperative morbidity [10]. Whilst this study was under-powered, other major limitations included insufficient time for implementation of recommendations following preoperative CGA potentially diluting the impact on the primary outcome measure and possibly more importantly, a lack of fidelity to core components of CGA [11]. It is well accepted that ensuring fidelity (how well the evidence based intervention is replicated?) is critical to the translation of evidence into clinical services. Equally important, however, is adaption of the intervention to the local context, which can be achieved through use of quality improvement (QI) methodology. In keeping with this, other work in colorectal surgery, has shown how QI methodologies can be used to establish and sustain perioperative CGA services with improved clinical outcomes [12]. Certainly over the past five years there has been an expansion in geriatrician led perioperative CGA services in the UK, despite qualitative work describing human factors as potential barriers to the uptake of perioperative CGA services [13–15]. Alongside the supporting literature for perioperative CGA, another driver for this observed expansion may be national audit and its consequences. In the same way that evidence, guidelines, national audit and financial incentives supported the expansion of orthogeriatrics, we may now be seeing similar results related to the National Emergency Laparotomy Audit with recommendations soon to be supported by a best practice tariff [16].

The expansion of geriatrician led CGA services certainly presents a challenge in terms of workforce planning. In the purposively chosen study sites in Kocman’s paper [4], there appeared to be a clear recognition from non-geriatricians that the perioperative application of CGA was beyond their expertise and that co-management between surgeons, anaesthetists and geriatricians was desirable. In contrast, the geriatricians in this study expected the toolkit to facilitate implementation of perioperative CGA negating the need for ‘hands on’ geriatrician delivered care for frail surgical patients. While there was initial ‘buy in’ for the use of the toolkit from clinicians, collective action in implementing the toolkit was lacking as illustrated by the challenges in embedding a frailty screening tool as a preliminary step in the preoperative CGA process. The results of this paper therefore raise the perennial question of whether we can embed specialist geriatric medicine knowledge (for example using a toolkit as in this study, or through modular training or credentialing) or whether we need to embed the CGA specialist across hospital services. Whilst it remains clear that populations utilising healthcare services are predominantly older and therefore all staff require generic skills in geriatric medicine, specialist CGA and optimisation remains the preserve of geriatricians working within an interdisciplinary team. A similar conclusion was drawn over ten years ago in an editorial accompanying a study describing CGA interventions in community settings [17]. In the context of significant vacancies in consultant geriatrician posts across the UK, this conclusion presents a significant challenge. Enthusiasm for engagement with the concept of CGA in other specialties (particularly anaesthesia/perioperative medicine) is unfortunately challenged by a lack of training opportunities, and the general move in postgraduate healthcare training from generalism towards super-specialism. Options to address this shortfall need to be innovative. Initiatives such as Integrated Medical Training pathways for junior doctors, dual accreditation in single organ medicine and geriatric medicine or anaesthesia/perioperative medicine and geriatric medicine, and the development of the alternative workforce from pharmacy, nursing or therapy backgrounds will be necessary. Finally, the conclusions of this study illustrate that other specialities recognise the added value of geriatricians employing CGA methodology in novel clinical settings. Clearly this presents a challenge to the specialty of geriatric medicine but one that should be welcomed.

Declaration of Conflict of Interest: None.

Declaration of Sources of Funding: None.