Institution Name: Prince of Wales Hospital, Sydney, Australia
Primary Author and Title: Christina M. Norris, MBBS, FRACP
Co-Authors and Titles: Jugdeep Dhesi, MBChB, PhD; Gregory Keogh, MBBS, FRACS; Philip Crowe, MBBS, DPhil, FRCSC, FRACS; Robert Gandy, MBBS, FRACS; Barbara Toson; and Jacqueline CT Close, MBBS, MD
Name of Case Study: The Care of Older People in Surgery (COPS) Service
As the population ages, older people are increasingly presenting to hospital with surgical problems requiring assessment and management. Older people face different challenges than their younger counterparts, including an increased risk of complications, death and functional decline.1,2 The presence of frailty, cognitive impairment and multimorbidity also contribute to poorer postoperative outcomes and necessitate holistic and comprehensive care.1,3,4 Collaborative care between surgeons and geriatricians is known to improve outcomes for older patients with hip fracture.5 Evidence is emerging for similar models of care in other surgical populations which utilize comprehensive geriatric assessment including patients undergoing vascular and gastrointestinal surgery.6-8 Furthermore, a recent study has demonstrated the benefit of geriatrician review in reducing mortality of older people undergoing emergency laparotomy.9 However, despite this evidence translation into clinical practice has been slow to occur.10-12 Whether a similar model of care may improve outcomes for those in the acute general surgical setting is less clear.
The Prince of Wales Hospital in Sydney, Australia, has a long history of collaborative care for older patients admitted to hospital with orthopaedic and vascular surgical problems. Patients admitted with general surgical problems were noted to have a high rate of hospital-acquired complications with significant levels of functional dependency and unplanned readmission. This is particularly true for older patients undergoing emergency laparotomy.10 Given the success of the established models of collaborative care, the departments of general surgery and geriatric medicine sought to determine whether similar benefits could be achieved in the emergency general surgical population, with a view to improving their outcomes.
Prince of Wales Hospital is a 450-bed metropolitan teaching hospital and tertiary referral center in Sydney, Australia. It has a well-established department of geriatric medicine that operates a shared model of care for older orthopaedic trauma patients and vascular surgery patients. Input into other surgical specialties was otherwise provided on an individual consult basis.
Over the course of a year, approximately 500 older (75+ years) patients are admitted to the general surgery service on an emergent basis. Less was known about these patients until an observational cohort study (n=303) was undertaken in 2016 demonstrated that a large proportion (41 percent) of patients experienced complications during admission, with the most common complication being delirium (18 percent). A significant proportion (26 percent) of patients also experienced decline in mobility and function during their stay.
Armed with the information from the observational cohort study, literature demonstrating the capacity for collaborative care to improve outcomes and a shared desire to improve care for the older general surgical patient, senior members of both departments agreed to pilot a new model of care. A decision was made to focus on older emergency general surgical patients given the apparent vulnerability of this population and the high rate of complications they experienced. Before proceeding, funding had to be secured with an understanding that the work would be time-limited and that if it failed to add value, the service would cease. Likewise, success would then necessitate funding to support the service on an on-going basis.
Discussions were held with key stakeholders and grant funding was sought through innovation and translational research grants from the Local Health District and state Department of Health respectively.
This study also follows several international initiatives and guidelines for older surgical patients. In the United Kingdom, geriatrician review is considered a standard of care for older people undergoing emergency laparotomy and is known to improve outcomes for this population.9,13 Similarly, the Proactive Care of Older Patient Undergoing Surgery (POPS) model of care has been well described in the literature with evidence of increasing uptake of similar collaborative models of care between surgeons and geriatricians across the UK.12,14 In the United States, the American Geriatrics Society, American Society of Anesthesiology, American College of Surgeons National Surgical Quality Improvement Program and the Society for Perioperative Assessment and Quality Improvement (SPAQI) have all produced guidelines for older surgical patients, including recommending that high risk older surgical patients receive geriatrician review with comprehensive geriatric assessment considered the gold standard for management of frailty.15-18
In order to address identified opportunities to improve care, a decision was made to utilize a collaborative model of care with each patient receiving shared care from a general surgeon and a geriatrician. The model was introduced on 19th September 2016 and was delivered to patients aged 75+ years admitted to a general surgical specialty on an emergent basis with a planned period for the pilot to run until 31st January 2018. In addition to shared care, all patients received physiotherapy from extra physiotherapy hours resourced by the grant. Patients underwent comprehensive geriatric assessment (CGA) on admission and were reviewed daily (Monday to Friday) by an aged care fellow who worked closely with nursing and allied health staff on the ward to ensure the delivery of coordinated care and facilitate early discharge planning.
The Care of Older People in Surgery (COPS) service includes a geriatrician (0.2 full time equivalent [FTE]), geriatric registrar (0.6 FTE) and physiotherapist (0.5 FTE). This service operates collaboratively and in tandem with the usual surgical services which include an Acute Surgical Unit for emergency general surgical admissions consisting of a general surgeon, a surgical registrar, senior resident and clinical nurse consultant. Funding was sought through two grant initiatives including an innovation grant locally via the South Eastern Sydney Local Health District as well as at a state-level with a translational research grant through New South Wales Health. The cost of this initiative totaled $AU284,000 over 18 months.
After initial discussions, the model of care was introduced as described above. Throughout implementation there were multiple problems encountered. These ranged from a high volume of allied health referrals—beyond that generated directly by the COPS service; lack of experience by nursing staff in caring for patients with delirium; and access to food immediately upon upgrade of diet. For each problem encountered Plan Do Study Act cycles were utilized to identify the root cause and appropriate solutions. Daily rapid-rounding with allied health staff for all patients on the general surgical ward was implemented to promote greater efficiency and prioritization of referrals which were found to have been generated by junior medical staff recognizing the importance of allied health for other patients not receiving care from the COPS service. A range of education strategies were utilized to allow nursing staff to gain greater knowledge and feel more confident in managing older delirious patients. An additional refrigerator was purchased for the ward and stocked with a range of options for older patients to consume upon diet upgrade, without having to wait for kitchen staff to bring their meal.
More broadly, although there was a general wish for the general surgical and geriatric departments to work more collaboratively, not all staff ‘bought in’ to the model from the first day. Concerted efforts to meet with consultants and senior staff on a semi-regular basis, continued support at an executive level and promotion of communication at all levels within the team with their surgical counterparts allowed for the development of professional relationships. By the end of the implementation period, communication occurred freely allowing for better decision making—at both a medical and a surgical level—as well as more efficient discharge planning.
This initiative was evaluated using a mixed methods approach. Firstly, the impact of COPS on clinical outcomes was evaluated through a retrospective case-matched cohort study with patients matched from before and after the implementation of the model of care. Patients were matched using age (+/- 3 years), sex and Australian Refined Diagnosis Related Group (AR-DRG). The primary outcome measure was the rate of hospital acquired complications with secondary outcomes measures including specific complications such as delirium, 30 day mortality, length of stay and unplanned readmission. Data were extracted directly from patient files using a range of pre-defined criteria for specific hospital acquired complications. Ethics approval was granted by the local Human Research Ethics Committee.
A total of 352 patients or 176 pairs were included in the analysis. Only a quarter of patients underwent surgical intervention with similar rates of surgical intervention between groups. Patients in the intervention group were significantly less likely to experience a complication compared with the control group OR 0.69 (95 percent CI 0.53-0.89, p=0.004). Patients were also significantly less likely to experience specific complications such as delirium OR 0.64 (95 percent CI 0.44- 0.92, p=0.017) and hospital acquired infection OR 0.58 (95 percent CI 0.34-0.99, p 0.045). Patients experienced lower rates of functional decline and unplanned readmission. There was no significant difference in length of stay between groups (median total length of stay 4.1 vs 3.9 days).
Secondly, a qualitative study was undertaken to explore patient and staff experience with the model of care. Whilst this feedback was overwhelmingly positive there were clear opportunities still remaining for further improvement. One of the main opportunities offered by staff was a desire for the service to be expanded to other patients younger than 75 years, those admitted electively and patients admitted to other surgical specialties. Staff were also asked to give three words that reflected their experience with working with the COPS service.
The word cloud representing this feedback is displayed in Figure 1.
Thirdly, system-level data was used to determine whether the model of care provided any benefit in terms of coding and cost. Patients with Gastrointestinal or Hepatobiliary AR-DRG diagnoses were compared from two 16 month periods before and after the intervention. A total of 894 patients were analyzed, with 415 admitted prior to the intervention and 479 after. Patients in the intervention group were found to have significantly higher rates of major or intermediate coded complexity, resulting in a higher mean total National Weight Average Unit (2.31 vs 1.88, Independent t test p=0.036). For the 2017–2018 financial year, this equates to a difference of 0.43 total NWAU per patient, or $AU2111.30 based on the national efficient price at the time of $AU4910 per unit. For this cohort of 479 patients this reflects an additional reimbursement of $AU1,011,312.70 during the intervention period compared with the total outlay for staffing of $AU284,000 for the same time period.
Despite the success of the service and clear evidence demonstrating benefit, securing long term funding for the model was not forthcoming. At the end of the implementation period there was a strong desire for continuation of the service from surgeons, nursing staff and allied health alike. Furthermore, there were requests for the service to be expanded. However, only after many discussions and continued advocacy from the department of surgery with the executive was funding able to be continued. These discussions have resulted in the COPS service being sustained with a part-time geriatrician, full-time geriatric registrar and part time physiotherapist as well as allowing the service to be expanded to additional surgical specialties.
Cohort |
|||||
Control N = 176 |
Intervention N = 176 |
p-value |
|||
Count |
% |
Count |
% |
||
Death at 30 days
|
7
|
4.0%
|
2 |
1.1%
|
0.091
|
Any Complication*
|
69
|
39.2%
|
40
|
22.7%
|
0.001
|
Delirium
|
39
|
22.2%
|
20
|
11.4%
|
0.007
|
Hospital acquired infection
|
23
|
13.1%
|
10
|
5.7%
|
0.017
|
Exacerbation of congestive cardiac failure
|
13
|
7.4%
|
4
|
2.3%
|
0.024
|
Acute Kidney Injury
|
24
|
13.6%
|
15
|
8.5%
|
0.126
|
Bowels not opened for >3 days
|
31
|
17.7%
|
15
|
8.6%
|
0.011
|
New dependence for pADLs
|
29
|
22.3%
|
13
|
8.7%
|
0.002
|
Unplanned Readmission
|
31
|
18.1%
|
18
|
10.3%
|
0.038
|
*Complications include: death during admission, delirium, hospital acquired infection, acute coronary syndrome, arrhythmia, exacerbation of heart failure, venous thromboembolism, acute kidney injury, inpatient fall and hospital acquired pressure injury.
Although there was a clear desire for a collaborative model of care between the departments of general surgery and geriatric medicine, the results of the observational cohort study provided much needed data to demonstrate both an opportunity for improvement and the basis for change. Similarly, this data also allowed for a means of comparison to demonstrate whether success had been achieved after intervention. The decision to pilot the model over a defined period of time provided a clear signal that if success was not achieved, the model would stop. Similarly, when success was achieved the knowledge that the model of care would stop indefinitely without funding provided an impetus for securing permanent funding in order to prevent the cessation of the service.
The success of this intervention—like any novel model of care—can also be attributed to buy-in from key stakeholders and leadership from the outset. Continued commitment at senior levels and ongoing support from clinical champions to ensure that a collegiate and collaborative environment was maintained, along with communication with key stakeholders such as nursing staff and allied health assisted with successful implementation. The COPS team similarly made an effort to ensure staff felt comfortable approaching them to discuss patients or other aspects of care. These measures facilitated implementation and created a model of care which was responsive to the needs of patients within the environment it was being delivered in, whilst being flexible and adaptable to challenges as they were faced.