Background

As the population ages, there are increasing numbers of older people presenting with gastrointestinal (GI) conditions amenable to major surgery. However, both under-investigation and under-treatment of older people are common, with rates of surgery declining with age [1-3]. There are large variations in outcomes in older people across different surgical units in the UK that are not wholly attributable to differences in socioeconomic circumstances or comorbidity. This implies that not all older patients receive the same level of care or access to resources [4-7]. In GI surgery, there is some concern that patients in centres with low elective surgery rates could be inappropriately denied the benefits of operative intervention including disease control and symptom improvement. This may lead to higher rates of emergency admission and intervention [3, 8]. Conversely, in centres with high rates of elective surgery, patients may be inappropriately subjected to the morbidity or mortality of surgery when there is limited or no benefit. This disparity is accentuated when older patients require emergency surgery [4, 9-11].

Adverse factors associated with ageing include comorbidity, polypharmacy, malnutrition, cognitive impairment, dependency and frailty, all of which are associated with increased all-cause mortality in the general population [12] and reduced quality of life following hospital admission [13]. There is also a natural decline in cardiorespiratory fitness with age. Major surgery in all patients leads to a short-term decrease in cardiorespiratory fitness and functional capacity [14]. However, in older adults this contributes towards long-term disability and loss of independence [15, 16]. Many older patients never regain their previous level of functioning after major surgery [16, 17].

Stratification and optimization of risk in the older population is a challenge. Guidelines for GI cancers focus on diagnosis and staging rather than assessment of suitability for treatment in this patient group [18-20]. Other guidance, such as that for diverticular disease, advises against the use of age in treatment decisions but does not offer alternative stratifiers [21]. Consequently, the process of assessment varies considerably between surgeons and surgical units [22]. Subjective assessment of functional status is a frequent aspect of surgical and anaesthetic assessments [23] but it has been suggested that this may result in misclassification of high-risk patients as low risk [24, 25]. In the elective setting, cardiopulmonary exercise testing (CPET) is gaining acceptance as an objective measure of cardiorespiratory fitness with the ability to predict complications and mortality after major abdominal surgery [24, 26, 27] however, it is not universally available [28]. Malnutrition is known to be a poor predictor of outcomes but many patients are not screened prior to treatment decisions [29]. Frailty, the state of enhanced vulnerability to minor stressors, has been shown to be predictive of poor outcomes and increased care needs post-discharge in both the elective [30] and emergency surgical settings [9, 10, 31] but still requires integration into surgical pathways in many hospitals [11]. Perioperative risk assessment, combining clinical assessment, objective tests and available risk calculators, must be individualized in both the elective and emergency settings to enable targeted optimization.

Optimization of outcomes in older patients with comorbidities and frailty requires multi-professional input which is often lacking [32, 33]. Many of the adverse factors associated with ageing may be modifiable if they are identified early in the patient pathway and by using evidence-based strategies and interventions. Strategies shown to be effective, particularly in the emergency setting, include senior clinician involvement, early cross-sectional imaging, timely access to theatres and risk-stratified decisions on postoperative destination (i.e., intensive care department vs. surgical ward) [11]. Interventions include multimodal exercise, nutrition and psychological interventions, termed ‘prehabilitation’ [34], comprehensive geriatric assessment and intervention, enhanced recovery after surgery protocols and delirium prevention pathways, for example. However, there is limited evidence of how these interventions are applied in current surgical practice and their effects on the functional recovery of older patients [35].

This study will objectively measure the functional recovery of older adults after a range of major GI operations using validated questionnaires completed by the patients with support from the research team. Evaluation of outcomes from the patient's perspective (patient reported outcomes, PROs) is acknowledged to be important in surgical research, particularly in older adults. Many studies in this area collect PROs as secondary outcomes [36, 37] and increasingly integrated patient interviews are being used to facilitate process evaluation [38].

The study will determine what optimization strategies are used in practice to mitigate against identified risk factors. It will also observe the functional trajectory of some patients who are deemed ‘unfit’ for major operative intervention. Barriers and facilitators to perioperative optimization and experiences of patients will be explored through semi-structured interviews. Mixed methods synthesis of results will enable qualitative data to help explain some of the differences observed in the observational study.

The research questions are as follows.

Objectives

The objectives are to determine:

Aims

The aims are to assess the functional recovery of older patients following elective and emergency GI surgery and the impact of variation in baseline health status and provision of perioperative support at different hospitals within a region.

Study design

This is a prospective, multi-centre, mixed methods observational pilot cohort with integrated qualitative patient evaluation.

Study schematic

An overview of the design is presented in Figure 1.

Eligibility criteria

Primary outcome

The primary outcome is functional recovery at 6 weeks measured using the World Health Organization Disability Assessment Schedule 2.0 (WHO DAS v2.0) [40] which has been validated as a qualitative measure of functional recovery in surgical populations [41, 42]

Secondary outcomes

Participant recruitment

Patients will be identified at multidisciplinary team meetings, surgical outpatient clinics, from elective operating lists and on call ‘take’ lists for elective and emergency presentations respectively and screened for eligibility. Patients will be approached by the local principal investigator (PI), delegated clinician or nursing study team members with the appropriate good clinical practice training. All potentially eligible patients will be recorded on the local screening log. Recruitment at Sheffield Teaching Hospitals commenced on 9 September 2020; the study is in set-up phase at the other sites. Study visits will be coordinated with usual clinical appointments or conducted by telephone or post to reduce the burden on patients. See Table 2 for a summary of the study timeline.

Data collection

At baseline, demographics, type of referral, preoperative assessment date and admission details will be collected for all patients. Comorbidities will be collected using the Charlson Comorbidity Index, a validated measure of prognostic impact of multiple chronic illnesses. Polypharmacy is defined as five or more regular medications. Preoperative blood tests relevant to the emergency and elective presentations will be collected. A detailed functional, nutritional and fitness assessment will be carried out using a number of validated questionnaires (Barthel's Activities of Daily Living [ADL], Lawton and Brody's Instrumental Activities of Daily Living, Clinical Frailty Scale [45], Mini Nutritional Assessment [46], International Physical Activity Questionnaire—Elderly [47], Mini-Cog© [48], WHO DAS and EQ-5D-5L) for the patient to complete themselves or with assistance from the research team and review of patient records (e.g., for CPET results). The order of administration of baseline questionnaires will be standardized as follows:

The following questionnaires are completed by the research team at baseline in the following order:

Members of the patient and public involvement (PPI) group have been consulted to ensure that the number of questionnaires is acceptable to patients and does not represent a significant burden. Validated questionnaires will be used in accordance with their respective user manuals. PRO measure instruments are summarized in Table 3.

At hospital discharge, patients will be asked to complete a questionnaire (see Appendix S1–S3) regarding preoperative, perioperative and postoperative optimization. Elective patients will be asked whether they participated in any form of prehabilitation (exercise, nutrition, psychological, geriatric), attended ‘surgery school’ or attended for transfusion, iron infusion, physiotherapy appointment, smoking cessation services or dietitian review and whether this was self-directed or arranged by the hospital. Elective and emergency patients will be asked about perioperative and postoperative optimization and specialty reviews (e.g., geriatrician, cardiology). The hospital records will be used to determine operative details, postoperative complications (using the Clavien–Dindo classification system), length of hospital stay and discharge arrangements.

Follow-up

At 6 weeks post-operation/procedure or decision not to operate the final pathology result, survival and readmission rate will be determined. Follow-up questionnaires (ADL, EQ-5D-5L and WHO DAS) will be completed to assess functional recovery at 6 weeks, 3 months and 6 months post-operation/procedure. These time-points have been chosen to look at postoperative recovery over time [17]. In addition, follow-up questionnaire results will be compared to baseline results. The order of administration of follow-up questionnaires will be standardized as follows:

Follow-up questionnaires will be completed by telephone with the patient. If a patient does not answer the telephone the researchers will attempt two more times before reporting the item as missing. Patients will then be contacted at their next follow-up time-point.

Integrated qualitative study

Qualitative patient interviews will be integrated with the quantitative study. Interviews will be semi-structured to enable an exploration of different themes with reference to a pre-prepared interview prompt sheet and the patients’ responses to the bespoke questionnaire. Interviews will focus on views on perioperative support measures, perceived barriers and facilitators to implementing these and mode of delivery. Perceptions of fitness and risk assessment and what this means to individuals will also be explored.

Maximal variation sampling will be used to select patients who have undergone different methods of assessment and optimization and across the full age spectrum (stratified recruitment to age groups 65–75, 75–85 and >85 years). From previous work in the field it is anticipated that approximately 20–30 interviews will be required, aiming for 10 in each age cohort across elective and emergency presentations. Participants will be identified from the cohort study if they have given consent for this aspect of the study. All interviews will be carried out by the chief investigator (CI) of the study, will be digitally recorded and transcribed verbatim. Interview transcripts will be anonymized prior to analysis and original recordings deleted.

Interview data analysis using the framework approach [49]will occur alongside recruitment, and recruitment will cease on achievement of data saturation. Three or four transcripts (10% total) will be double coded by an experienced qualitative researcher from the research team (MB) to ensure credibility and dependability of the qualitative findings.

Analysis of the qualitative and quantitative aspects will occur in tandem. This will enable the patient interviews to be used to explore the effects of different baseline health status and optimization strategies on patient attitudes and responses. They will also be used to explore whether patients with worse functional outcomes have different experiences to those with better outcomes. Triangulation of qualitative and quantitative data will enable us to develop a comprehensive understanding of the findings [50]. Findings from this study could help to inform prioritization of services.

Protocol amendments

Protocol amendments will be approved by the Health Research Authority and communicated to all local PIs and research and development teams by the study team.

Data management and monitoring

All data will be handled in accordance with the General Data Protection Regulation 2018 principles. All patients will be given a unique ID number which will be used in the database rather than their National Health Service (NHS) number (i.e., pseudo-anonymized). Data will be collected and recorded by hospital staff or members of the hospital research team on paper-based case report forms which will then be entered into a secure server running the Research Electronic Data Capture (REDCap) system based at the University of Sheffield [51]. REDCap allows collaborators to enter and store data in a secure system. Data will be monitored for quality and completeness by the study team. Missing data will be sought until they are received, confirmed as not available, or the study is at analysis. PRO data will not be monitored during the study to inform clinical care as this is an observational study only. Access to the entire study dataset will only be available to co-investigators of the study as detailed in the ethics application.

Statistical analysis and power calculation

Descriptive analyses will be performed to describe the recruited population of patients aged 65 years and over undergoing major GI surgery via emergency and elective pathways at different surgical units within a region. Descriptive analyses will also be used to detail the optimization pathways reported by patients and documented in their medical records. Univariate correlation analyses will be used to look for relationships between baseline variables and functional outcomes, but these will be exploratory only. As this is a pilot study, feasibility of recruiting from this patient population will be assessed by recording the number of people who decline participation. Adherence to the follow-up schedule will also be used to guide the design of future studies.

An opportunistic sample size of 120 has been estimated over the 6-month study recruitment period based on the number of patients undergoing major surgery at each of the units. Each surgical unit in South Yorkshire performs between 70 and 300 major elective GI resections per year and 114–300 emergency laparotomies per year, of which at least 50% will be over the age of 65 [11]. A high uptake rate of this simple, questionnaire-based study is anticipated based on a recent study of frailty in emergency laparotomy patients [10] and a postoperative study of quality of life after emergency laparotomy [52]. One aim of the study is to capture variation in practice; a sample of 120 should permit this, especially when spread across three or more centres. Methods to use pilot study data to estimate sample size for trials vary. A sample size of 120 exceeds the proposed size of 60–100 proposed by Teare et al. [53] to estimate effect size for continuous outcomes. It should also be compatible with other methods of effect size estimation [54].

Ethical approval

Ethical approval for this study was granted by an NHS Research Ethics Committee via the proportionate review system. Health Research Authority Approval has been obtained. This was granted by the Health and Care Research Wales (HCRW) Research Committee on 4 May 2020 (REC Reference 20/SC/0076).

All participating units must obtain approval from their local Research and Development Department consistent with the guidance from the Health Research Authority. It is the responsibility of the local PI to ensure that the relevant approvals are in place prior to commencing data collection.

Patient and public involvement

Patient and public involvement is integral to this study. The role of the PPI group is reported here according to the GRIPP2-SF guidelines [55]. The main aim of the PPI group so far has been to ensure that the research question is of relevance to the lay person and to review the study protocol. They also reviewed the patient-facing materials, which included checking that the number and time taken to complete the questionnaires was not much of a burden for patients. This resulted in a reduction in the number of baseline questionnaires and to changing one of the questionnaires to a shorter version. The CI met with the PPI group twice regarding this study and also communicated with them via email. One member is on the study steering group. This group will also be involved in disseminating the study findings. Future PPI involvement for a study of this nature might benefit from a speciality specific PPI group, such as those with experience of GI surgery, to facilitate dissemination of results.

Dissemination

Results from the study will be analysed after completion of data collection at all sites. Results from individual sites will be fed back to the local PIs for dissemination at hospital level. The PPI representatives will also be involved in disseminating to patient groups. Findings will be submitted to regional and national conferences in surgery, geriatrics or perioperative care. Manuscript(s) will also be prepared for publication in peer-reviewed journal(s).

Collaborators at participating hospitals with significant input, particularly regarding recruitment and follow-up of patients, will be eligible for collaborative authorship according to published guidelines [56].