Background
Enhanced recovery programs have been associated with improved outcomes after gynecologic surgery. There are limited data on the effect of enhanced recovery programs on healthcare costs or healthcare service use.
Objective
The purpose of this study was to evaluate differences in hospital charges for women who undergo surgery for a suspected gynecologic cancer that is managed in an enhanced recovery program as compared with conventional perioperative care.
Study Design
We performed a retrospective cohort study of women who underwent open abdominal surgery for a suspected gynecologic cancer before and after the implementation of an enhanced recovery after surgery program. Consecutive patients from May to October 2014 and from November 2014 to November 2015 comprised the conventional perioperative care (before enhanced recovery after surgery) and enhanced recovery after surgery cohorts, respectively. Patients were excluded if they underwent surgery with a multidisciplinary surgical team or minimally invasive surgery. All technical and professional charges were ascertained for all healthcare services from the day of surgery until postoperative day 30. Charges for adjuvant treatment were excluded. Charges were classified according to the type of clinical service provided. The primary outcome was the difference in total hospital charges between the pre–enhanced recovery after surgery and the enhanced recovery after surgery groups. Secondary outcomes were between group differences in hospital charges within clinical service categories.
Results
A total of 271 patients were included in the analysis (58 patients in the pre–enhanced recovery after surgery and 213 patients in the enhanced recovery after surgery cohort). A total of 70,177 technical charges and 6775 professional charges were identified and classified. The median hospital charge for a patient decreased 15.6% in the enhanced recovery after surgery group compared with the pre–enhanced recovery after surgery group (95% confidence interval, 5–24.5%; P =.008). Patients in the enhanced recovery after surgery group also had lower charges for laboratory services (20% lower; 95% confidence interval, 0–39%; P =.04), pharmacy services (30% lower; 95% confidence interval, 14–41%; P <.001), room and board (25% lower; 95% confidence interval, 20–47%; P =.005), and material goods (64% lower; 95% confidence interval, 44–81%; P <.001). No differences in charges were observed for perioperative services, diagnostic procedures, emergency department care, transfusion-related services, interventional radiology procedures, physical/occupational therapy, outpatient care, or other services.
Conclusion
Hospital charges and healthcare service use were lower for enhanced recovery patients compared with patients who received conventional perioperative care after open surgery for a suspected gynecologic cancer. Enhanced recovery programs may be considered to be high value in healthcare because they provide improved outcomes while lowering resource use.
The benefit of enhanced recovery after surgery (ERAS) programs after gynecologic surgery has been documented. Such programs apply evidence-based perioperative management strategies with the goal to standardize patient care and mitigate the physiologic dysfunction that can occur as a result of surgery and postoperative recovery. Implementation of ERAS programs have resulted in shorter postsurgical admissions, fewer postoperative complications, decreased use of opioid pain medication, and improvements in functional recovery. There has been broad interest in the creation of ERAS programs nationally and globally.
Why was this study conducted?
This investigation was performed to further characterize the value of enhanced recovery after surgery (ERAS) programs in gynecologic cancer surgery. Although the clinical benefits of ERAS programs have been described, less focus has been placed on the resources that are used to achieve improved outcomes. This study evaluates whether implementation of an ERAS program impacts healthcare costs in women who undergo open surgery for a suspected gynecologic cancer.
Key findings
Compared with historic control subjects, women who undergo open gynecologic surgery had lower hospital charges after implementation of an ERAS program. Most of the decrease in healthcare spending seemed to be related to shorter postoperative hospital stay in the ERAS cohort, which suggests that ERAS may be cost-saving by reducing healthcare resource use.
What does this add to what is known?
This investigation offers evidence that ERAS programs in gynecologic surgery provide value through both improved outcomes, such as shorter postoperative hospital stays and reduced opioid consumption, and reduced healthcare costs and healthcare resource use. This analysis also characterizes how ERAS programs provide cost-savings principally by reduction in hospital services use through shorter postoperative length of stay.
Value in healthcare is described frequently as an equation of “health outcomes achieved per dollar spent.” In the context of ERAS programs, considerable interest has focused on the improvement in perioperative outcomes. Less attention has been directed towards the cost that is required to achieve those outcomes. Ascertaining “cost” in the American healthcare system is difficult, given the substantial variability that is observed in the expense to a provider to offer a service, the amount billed to a patient or payer, and the reimbursement actually provided for that service.
Cost reductions have been observed or projected from the implementation of ERAS programs in other surgical literature. A recent systematic review and metaanalysis of patients who underwent major abdominal surgery across a variety of surgical disciplines estimated a mean cost reduction of $5100 for ERAS patients. Reports from several ERAS programs in gynecologic surgery have also reported cost-savings as secondary outcomes. Cost-savings that were observed with ERAS programs may be attributed to reductions in the length of postoperative hospital admission. Our goal with this investigation was to evaluate differences in hospital costs, both in aggregate and within discrete categories of perioperative clinical services for patients who undergo gynecologic surgery in an ERAS program compared with conventional postoperative care.
Materials and Methods
A retrospective cohort study was performed as a secondary analysis of our institutional quality improvement project that implemented an ERAS program in gynecologic surgery that was established in November 2014 ( Table 1 ). In brief, the program consists of management strategies that can be categorized into the preoperative (clear fluids until 2 hours before surgery, avoidance of bowel preparation, preoperative analgesia), intraoperative (opioid-sparing multimodal anesthesia, goal-directed fluid therapy, minimization of the use of surgical drains), and postoperative phases (limited intravenous fluids, multimodal opioid-sparing analgesia, quick resumption of a general diet, encouragement of ambulation, discouragement of bedrest). Before implementation of the ERAS program, perioperative treatment of patients who undergo open surgery with a gynecologic oncologist was not standardized with regards to the use of bowel preparations, postoperative intravenous fluids, resumption of diet after surgery, pain management, blood product transfusion, bedrest, and postoperative mobilization.
| Phase of care | Intervention | Enhanced recovery approach |
|---|---|---|
| Preoperative | Counseling | Preoperative teaching and optimization |
| Diet | Nutritional counseling during preoperative visit | |
| No solid food after midnight | ||
| Clear liquids until 2 hours before surgery | ||
| Carbohydrate loading | ||
| Bowel preparation | None | |
| Premedication | Tramadol | |
| Pregabalin | ||
| Celecoxib | ||
| Acetaminophen | ||
| Prophylactic heparin | ||
| Intravenous fluids | Saline lock intravenously | |
| Intraoperative | Antibiotics | Prophylaxis per American College of Obstetricians and Gynecologists guidelines |
| Anesthesia | Emphasis on total intravenous anesthesia | |
| No epidurals | ||
| Wound infiltration with local anesthetic | ||
| Intravenous fluids | Goal-directed (noninvasive cardiac monitoring) | |
| Nasogastric tube/drain placement | Not used on a routine basis | |
| Foley catheter | Removal morning of postoperative day 1 | |
| Postoperative | Intravenous fluids | Intravenous fluids 40 mL/hr until morning of postoperative day 1 |
| Saline lock intravenously when tolerating 500 mL by mouth | ||
| Multimodal analgesia | Acetaminophen | |
| Ibuprofen | ||
| Pregabalin | ||
| Oxycodone as needed | ||
| Hydromorphone intravenously as needed | ||
| Diet | Dietitian counseling | |
| Regular diet on arrival to hospital floor | ||
| Oral hydration | ||
| Ambulation | Ambulate ≥8 time per day | |
| All meals in chairs | ||
| Out of bed ≥8 hours daily | ||
| Blood transfusion | Transfuse for hemoglobin <7 g/dL |
Hospital charge data, defined as the amount billed by a healthcare provider for a service, were collected for all consecutive patients for 1 year (November 2014 to November 2015) after the implementation of the ERAS program (ERAS cohort). Patients who underwent surgery within 6 months before ERAS implementation (May 2014 to October 2014) served as historic control subjects (pre-ERAS cohort). Patients were eligible for inclusion in the analysis if they underwent open gynecologic surgery. Patients who had concurrent procedures with any additional surgical services or who underwent minimally invasive surgery were excluded. Demographic and clinical data were ascertained: age, race, body mass index, American Society of Anesthesiologists score, Charlson comorbidity index, procedure length, tumor site, tumor type, perioperative complications, surgical complexity score, reoperation, and readmission. Perioperative complications were categorized with the use of the Clavien-Dindo classification system. Ovarian, fallopian tube, and primary peritoneal cancer cases were characterized as low, medium, or high according to the surgical complexity score developed by Aletti et al. In patients who underwent reoperation, their categorization into either the pre-ERAS or ERAS groups was by their initial surgery. Itemized data for all technical and professional services were collected from the day of surgery until 30 days after surgery. The technical component of a charge reflects the cost of equipment, supplies, non–physician personnel, facilities, and other elements that are related to the provision of service. The professional component reflects the fee for the physician’s work or expertise in the delivery of a diagnostic or therapeutic service. For patients who received adjuvant therapy related to their cancer diagnosis, charges for these services were excluded from the analysis. Hospital charges were classified according to clinical service group, which included laboratory services, pharmacy services, perioperative care (eg, post anesthetic care unit services), diagnostic procedures, interventional radiology procedures, Emergency Department care, transfusion-related services, pathology-related services, physical and occupational care, room and board–related services, charges for material goods, outpatient care, and other. Patients who experienced postoperative complications or hospital readmission that extended their hospital stay beyond postoperative day 30 were not excluded from the analysis.
The data collection and management were approved as a component of an institutional quality improvement project (QI-6033). The retrospective analysis of these data was approved by the institutional review board (PA18-1136). Numeric values represent percent differences in hospital charges both in aggregate and within categorized clinical services. As is the case among many healthcare organizations, numeric cost data are protected and considered proprietary information at our institution. Absolute numeric values regarding hospital costs are unable to be disclosed in this report to protect institutional cost-to-charge ratios.
The primary objective was to determine whether there were any differences in the total median hospital charges for patients who were treated in an ERAS program compared with those who received conventional postoperative care. The secondary objectives were to determine whether there were differences in median hospital charges in categorized clinical services between the pre-ERAS and ERAS groups. Descriptive statistics were used to summarize the demographic and clinical characteristics of the pre-ERAS and ERAS cohorts. A Wilcoxon rank-sum test was used to compare distributions among continuous variables, and a Fisher’s exact test was used to test associations among categoric variables. Wilcoxon rank-sum tests were used to evaluate for differences in the distributions of charges between the pre-ERAS and ERAS cohorts for both the total hospital charges and for the service-related charge categories. The absolute percent change from pre-ERAS to ERAS was calculated between the median charges overall and for each charge category. Bootstrap methods were used to calculate the 95% confidence interval (CI) for the absolute percent change. The bootstrap sample included 1000 unrestricted random samples of k size, where k is the number of observations selected from each group (58 pre-ERAS and 213 ERAS; bootstrap sample size, 271,000). The lower and upper limits of the 95% CI was derived from the 2.5 and 97.5 percentiles of the calculated median percent changes from the bootstrap samples, respectively. A probability value of <.05 was considered statistically significant. Analyses were performed with SAS software (version 9.4; SAS Institute, Cary, NC).
Results
There were 271 patients included in the analysis, 58 patients and 213 patients in the pre-ERAS and ERAS cohorts, respectively. A total of 70,177 technical charges and 6775 professional charges were attributed to the entire study population. The pre-ERAS and ERAS groups had similar demographic and clinical characteristics ( Table 2 ). The median ages for the pre-ERAS and ERAS groups were 58 (range, 32–85) and 57 (range, 18–79) years, respectively ( P =.808). Median operating times were 221.5 (range, 98–484) and 210 (range, 33–497) minutes for the pre-ERAS and ERAS groups, respectively ( P =.271). Complications of any grade were observed in 43.1% (25/58) and 38.5% (82/213) of patients in the pre-ERAS and ERAS groups, respectively ( P =.547). The case mix of the 2 groups was similar in terms of surgical complexity ( P =.153). The majority of procedures were performed by gynecologic oncologists (99.6%; 270/271). Three patients (5.2%; 3/58) in the pre-ERAS group and 4 patients (1.9%; 4/213) in the ERAS group underwent reoperation. Eight patients (13.8%; 8/58) in the pre-ERAS group and 27 patients (12.7%; 27/213) in the ERAS group were readmitted after surgery. Gastrointestinal complications (eg, ileus, nausea/emesis, obstruction, perforation) were the most common reason for readmission in both the pre-ERAS (37.8%; 3/8) and ERAS groups (44.4%; 12/27). Median length of stay was 1 day shorter for the ERAS group compared with the pre-ERAS group (4 vs 3 days; P <.001). Patients were compliant with at least 70% of the interventions in the institutional ERAS program.
