The U.S. population is aging, and although care for chronic medical conditions has improved, optimization for surgery is a complicated process. Medical and surgical complications continue to threaten patients who are elderly and who have complicating comorbidities.1,2
Using a variety of methods and risk assessments, frailty has been demonstrated in 4.1 percent to 50.3 percent of surgical patients and is predictive of mortality, postoperative complications, and disposition at discharge.3,4 Complications after surgery are costly—they can increase case mix-adjusted costs by $9,419 to $13,832 per case.5 More notably, complications worsen patient experience, are of concern to regulatory bodies, and can contribute to higher use of health care resources.
Enhanced recovery programs (ERPs) are being used nationwide to reduce complications, lengths of stay (LOS), and cost of care per patient.6-8 It is widely accepted that no single implemented change will improve outcomes of surgery across all patient populations and that the approach to perioperative and postoperative care must incorporate multiple disciplines, modalities, and components to optimize patient care.
Identification of local problem
Connecticut has some of the highest hospital complication and serious safety event rates in the nation. Multiple efforts to improve these outcomes have been condensed into a constellation of enhanced recovery processes to optimize patients and improve outcomes after surgery. These programs vary location to location but overall have had a significant impact on patient readiness for surgery, expectation management for patients and families, and improvement in the overall patient experience. Innovative surgical and anesthetic techniques have contributed to improvements, but low-cost, high-impact efforts like an ERP can make all the difference.
Complications documented by local data assessments, the Centers for Disease Control and Prevention’s National Healthcare Safety Network (NHSN), and the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP®) data indicated that our hospital was an outlier with respect to overall morbidity, LOS, postoperative surgical site infection (SSI), sepsis, transfer to extended care facilities in the postacute setting, hospital-acquired infections (HAIs), and postoperative opiate use. All listed complications were important to patient experience. Events like ileus, reinsertion of a nasogastric tube, and other data were not publicly reported.
Saint Francis Hospital and Medical Center, Trinity Health Of New England, Hartford, CT, sought to improve patient care throughout the regional health system and to disseminate lessons learned locally and from peer hospitals in the state to reduce variation and hospital-acquired conditions statewide. The Connecticut Surgical Quality Collaborative (CSQC) was instrumental in obtaining grant funding for technology and education statewide, allowing for uniform data collection and for techniques to evolve toward being more unified with evidence-based best practices. This collaboration has been fruitful, with creation of a resource website, development of the Connecticut Geriatric Program in Surgery, and the sharing of data to improve care regionally. Lastly, we have identified several opportunities for systemic improvements throughout this program and are on the journey toward consistent practice, high reliability, and the provision of safe, effective surgical care.
Saint Francis Hospital has evolved a constellation of practices to reduce HAIs and conditions over the last several years. As early adopters of large-volume, risk-adjusted surgical databases like ACS NSQIP, we have improved multiple metrics across the continuum of care. We have reduced ventilator-associated events, reduced SSIs, and adopted ERPs in caring for our colorectal surgery patients to improve every aspect of their experience. These processes are being used for patients undergoing hysterectomy and other procedures, and other programs in Connecticut have had similar success.
This project began with pilot data on the implementation of team-based training, best practice use, and the idea that state programs would benefit from transparent data sharing. A grant was sought and awarded, and we began our leadership and collaboration with the CSQC, a partner of the Connecticut Chapter of the ACS. With a goal of enhancing the patient experience, lowering costs of care, and minimizing complications, we were set to lead the journey toward zero harm.
Putting the QI activity in place
Saint Francis Hospital is a not-for-profit, urban, tertiary care, Level I trauma center. It is the largest Catholic hospital in the northeast and a primary partner in the development of the CSQC, a surgical safety collective including every acute care hospital in Connecticut.
We used the following data to measure quality improvement:
- NHSN published infection and event data, which provides our hospital event rates and actionable items in addition to benchmarks
- ACS NSQIP risk-adjusted data, which allowed us to benchmark surgical patients at Saint Francis against surgical patients at other CSQC institutions and nationally reported benchmarked data
- Locally collected outcomes, including identified surgical and medical complications and adverse or unexpected events, rates of transition to nursing facilities postacute care, readmission data, LOS, and postoperative milligram morphine equivalents, to monitor opiate use and prescribing
Other clinical data, such as tolerance of enteral nutrition, use of urinary catheters, and more, are collected in real time.
Our own performance with patient experience has been subpar in comparison with the remainder of the nation. Our complications, HAIs, and other parameters are higher than most hospitals in the U.S.
In 2016, the noted performance reflected in the ACS NSQIP data suggested the need for improvement, and the ERP was instituted. Monthly development and implementation meetings were convened to educate surgery staff, and the colon surgery specialty group committed to the process with early adoption of outpatient and postacute measures. Inpatient measures were implemented simultaneously.
Description of the quality improvement activity
We targeted improvement in these events in an effort to enhance the patient experience and sought support from the private foundation of a medical professional liability provider that had shown interest in our preliminary work. With that support and help from the CSQC, we were able to implement the ERP for colon and rectal surgery. More recently, further efforts to focus ERP on frail populations (patients with diabetes, patients older than 70 years old, and patients with American Society of Anesthesiology [ASA] Classification >2) have been explored. Perioperative optimization in these patients has resulted in similar improvements in outcomes.
Briefly, an ERP was instituted over the last three years. Consecutive patients undergoing colorectal resections performed after January 1, 2016, have been followed. Outcome measures including overall complications, postacute disposition to skilled nursing facility, readmissions, LOS, duration to flatus, and morphine milligram equivalents (MME) were collected prospectively. Statistical analysis and overall assessment on patient experience was performed for all interventions.
Overall complications were assessed in the patients before institution of ERP and subsequent to the start of the program. Subgroups of patients included frail patients, such as highly comorbid patients (those with ASA >2), patients of advanced age (defined as age >70), and patients with diabetes (defined as preoperative glycosylated hemoglobin >6.5 percent).
Initially, 183 consecutive patients underwent colorectal resection before the ERP was initiated. Subsequent to the program starting, 509 consecutive patients were included in the ERP. For the sake of data analysis, seven patients were excluded because of the absence of important data. Demographics collected included age, gender, body mass index, and tobacco use, which did not differ between groups.
Complications were identified by individual chart abstraction and included surgical complications (return to operating room [OR], leak, bleeding, wound drainage, and others), and medical complications (acute kidney injury, hypotension, oliguria, myocardial infarction, thromboembolic event, and others, including events that are infrequently publicly reported, such as ileus, wound erythema, dislodgement and replacement of devices, dysuria and urinary retention, changes in level of care, and more). Data from the state hospital association, locally collected hospital data, and annual and semiannual reports from ACS NSQIP were used to identify events and outliers.
This process began in 2016 and continues today.
Resources used and skills needed
Existing staff, including nursing providers in the perioperative center, a single physician assistant supported by office staff members, and clinical health care professionals, all were engaged. Leadership included a dedicated quality nurse provider, the service line executive director, and chair of surgery, as well as a dedicated colon and rectal surgery group. Surgeons in a private specialty colon and rectal surgery practice were the primary surgical providers and supplied the push needed to engage the remainder of the hospital staff.
Getting results
Overall, complications of any kind occurred in 47.5 percent of the pre-ERP patients. Though this number seems high overall, it includes even transient increases in creatinine, adynamic ileus, vomiting events requiring interventions, and other patient dissatisfiers. In ERP patients, this number dropped to 23.2 percent (see Tables 1 and 2). Separate analysis revealed improvements in frail patient groups as well. Comorbid patients (n = 237) and patients with diabetes (n = 139) demonstrated identical improvements (p <0.001) (see Tables 3 and 4). Patients of advanced age (n = 255) improved significantly with respect to surgical and medical complications, days to flatus, and hospital LOS (p <0.001) (see Table 5).
Table 1. Saint Francis Hospital ERP patient data
