QI principles were used by a multidisciplinary team to improve the quality of care in patients with glioma during the perioperative period. Ongoing dialogue across departments and reporting of system performance were important for sustaining process improvements.
Abstract
Purpose:
Although there is agreement on the oncologic management of patients with glioma, few guidelines exist to standardize other aspects of care, including supportive care.
Methods:
A quality improvement (QI) project was chartered to improve the care provided to patients with glioma. A multidisciplinary team was convened and identified 10 best-practice measures. Using a plan-do-study-act framework, the team brainstormed and implemented various improvement interventions between June 2011 and October 2012. Statistical process control charts were used to evaluate progress. A dashboard of quality measures was generated to allow for ongoing measurement and reporting.
Results:
The retrospective assessment phase consisted of 43 patients with diagnosis of glioma. A manual medical record review for these patients showed that compliance with 10 best-practice measures ranged from 23% to 100%. Several factors contributed to less-than-ideal process performance, including poor communication among disciplines and lack of familiarity with the larger system of care. After implementing improvement interventions, performance was measured in 96 consecutive patients with glioma. The proportion of patients who met criteria for 10 practice measures significantly improved (pre-QI work, 63%; post-QI work, 85%; P = .003). The largest improvement was observed in the measure assessing for preoperative notification of the neuro-oncology program (pre-QI work, 39%; post-QI work, 97%; P < .001).
Conclusion:
QI principles were used by a multidisciplinary team to improve the quality of care for patients with glioma during the perioperative period. Leadership involvement, ongoing dialogue across departments, and reporting of system performance were important for sustaining process improvements.
Introduction
Nearly 80% of primary malignant brain tumors are gliomas,1 and a majority of these are categorized as glioblastomas (GBMs), the most aggressive form.2 Although gliomas account for < 2% of all cancer diagnoses in the United States,3 their incidence has continued to increase as the population ages.4 Specifically, by 2050, the number of new patients with GBM is predicted to increase by 72%.5
Because gliomas are aggressive, they are often associated with poor survival1; the 5-year survival rate for patients with GBM who receive postoperative chemoradiotherapy is < 10%.6 The best predictors for survival include age at diagnosis, functional status, extent of resection, and O-methylguanine–DNA methyltransferase methylation status.6,7 Although there is no cure, scientific advances have led to the development of treatments that prolong life.7,8
There is evidence to inform treatment decisions with surgery, chemotherapy, and radiotherapy for patients with glioma; however, few guidelines exist that help standardize other phases of care, including supportive care.9 The Glioma Outcomes Project found a wide variation in patterns of care to manage patients with high-grade glioma.9 Despite consistency in following the evidence for imaging, surgery, and radiation treatment, the authors found great disparities in other aspects of care. For example, although deep venous thrombosis events occurred in 3% to 60% of patients with malignant glioma within 6 weeks of neurosurgery, only 7% of these patients received prophylactic low-dose heparin.9 In addition, practice patterns were sometimes in direct opposition to the evidence.9 Variations in patterns of care between academic and community practices were also associated with differences in survival outcome.9
These findings highlight the importance of developing cancer-specific clinical care pathways that can assist in selecting the appropriate care based on the reported literature.9,10 The hope is that these pathways may not only improve the quality of care patients receive but also reduce health care costs by minimizing treatment complications and eliminating the use of unnecessary treatment interventions.10 To our knowledge, there have been no prior reports on the use of quality improvement (QI) methodology to improve the care of the population diagnosed with glioma.
We report on the development of a clinical care pathway at Norris Cotton Cancer Center (NCCC) at the Dartmouth-Hitchcock Medical Center (Lebanon, NH) that addresses the acute health care needs of patients with glioma during the perioperative period. We relied on QI methodology in the development of this pathway. Our overall aim was to ensure that patients with glioma receive comprehensive, consistent, and timely care.
Methods
Organizing for Improvement
Patients with glioma represent 30% of the primary brain tumor population treated in the NCCC Neuro-Oncology Program. We chartered a multiphase QI project with the specific aims of improving the quality of care in patients with glioma, reducing process variation, and maximizing patient safety. The project began with the entry of patients into the neuro-oncology microsystem and ended with discharge of patients to survivorship or death.11 We defined the microsystem as a “small, naturally occurring frontline [unit] that provide[s] most clinical care to most people.”11(pp7-8)
We report on the first phase of this QI project, which focuses on time from initial diagnosis to entry into the surgical pathway and first visit with neuro-oncology to establish care. A multidisciplinary team was convened and included physicians, a physician's assistant, nursing staff, schedulers, and a social worker. Members represented disciplines including neuro-oncology, neurosurgery, radiation oncology, and care management as well as cancer center leadership. The team met twice per month from May to October 2011, then monthly from February 2012 to October 2012, and then quarterly to maintain project gains. The team was organized into three groups: entry into the microsystem, surgical care, and postoperative care. A coach trained in QI methodology was assigned to help the team evaluate current processes and identify improvement strategies.
Planning and Implementation
Before beginning this QI work, we reviewed the literature, and except for guidelines directing chemotherapy, surgery, and radiation treatment, we found limited information on which measures are associated with improved outcomes in the glioma population. On the basis of scant evidence and brainstorming, we proposed 10 objective measures that reflected timely and comprehensive care.
Using QI methodology,11–13 our team defined the nature and scope of the problem and used flowcharts to outline current processes. The team members generated three process maps, representing neurosurgery, medical neuro-oncology, and radiation oncology, and combined these into one system map. The team members then assigned the 10 objective best-practice measures to relevant steps in the process (Appendix Table A1, online only).
We abstracted performance data from available medical records and from a database created through prior QI work. Using a plan-do-study-act framework,12,13 the team members assessed current process performance. The members hypothesized that there were at least 15 contributors to poor performance and developed seven improvement interventions to address these concerns. Once we successfully implemented improvement efforts, our team members created a formal plan for sustaining the new process.
Proposed Improvement Strategies
The team members identified several variables that contributed to less-than-ideal process performance. There was no system to facilitate communication among services, and staff members often made decisions based on a narrow set of variables. Process participants also had limited awareness of the extent to which individual steps contributed to the larger system of care. There was no mechanism in place to facilitate ongoing process feedback.
Starting in June 2011, the team piloted several small tests of change to improve communication among the different services. Team members reviewed the current procedural terminology (CPT) codes for glioma surgery with the operating room (OR) scheduler. A checkbox for CPT code was added to the OR booking sheet to trigger the OR scheduler to notify the neuro-oncology team of any new admissions. A consult list was created in the electronic medical record and was routinely updated to reflect new patients requiring evaluation. Team members developed a standardized postoperative order set to guide antiepileptic drugs (AEDs) and steroid administration. Finally, discharge summaries were sent to the neuro-oncology team to ensure that follow-up occurred. Figure 1 shows the process flowchart, highlighting these various improvement strategies.
Figure 1.
Process flowchart for the acute care of patients with glioma (postintervention phase). NOTE. Codes refer to current procedural terminology codes for glioma surgery. Admit, hospital admission; appt, appointment; d/c, discharge; EMR, electronic medical record; ER, emergency room; N/O, neuro-oncology; N/S, neurosurgery; onc, oncology.
The team members generated a data dashboard to facilitate ongoing improvement work. The project leader reviewed the dashboard monthly with the statistician. A process owner was assigned to each QI initiative. The team continued to meet quarterly to review system performance and modify the process as indicated.
Outcome Measures
A score was created for each patient of maximum standards of care achieved. The numerator was the number of standards of care met, and the denominator was the number of standards of care needed for these patients. Because of symptoms and location of glioma, certain standards of care did not apply to a subset of patients. On the basis of the evidence, we defined appropriate use of AEDs as the initiation of an AED taper postoperatively in patients with no seizure history.14–16 We defined appropriate use of steroids as steroid use in patients with symptomatic edema.14,17
Statistical Analysis
To determine whether the maximum standard of care achieved improved over time, we analyzed the results for consecutive patients in an individual values and moving range chart. In this statistical process-control (SPC) chart, each data point represented a single observation.18 The SPC chart contains upper and lower control limits, which are set at three sigma, offering the best tradeoff between risks for type I and II errors.18 SPC charts differentiate between two types of variation: common and special causes. Processes are in statistical control when rates over time fall within the upper and lower control limits.18 A statistically significant change or a special-cause variation occurs when ≥ one point goes beyond the control limits or when there is a process shift in which ≥ eight successive values fall on the same side of the overall rate.18,19
SPC charts are robust statistical tools for detecting special-cause variation when the control limits are set at three sigma. Here, the probability of a type I error (ie, point falling outside control limit because of chance) is small (< 0.01).20 Similarly, the probability of ≥ eight successive values falling on the same side of the overall rate is low.20 Therefore, it is appropriate to infer that a statistically significant change occurred and recalculate the overall rate and limits.
The proportion of eligible patients who met each of the 10 individual measures pre- and post-QI work was reviewed using a histogram plot. Associations between categorical variables were determined using the χ2 statistic.
Results
Baseline Performance
The baseline assessment population included 43 patients with newly diagnosed glioma who received their surgical care at NCCC in 2010. We completed a manual review of their medical records to determine compliance with 10 best-practice measures. We found that before the initiation of the QI work, the overall adherence was 63% (range, 23% to 100%).
Patient Baseline Characteristics
Of the 43 patients included in the baseline assessment phase, 65% were men, with a mean age of 59.9 years (standard deviation [SD], 15.4 years). Tumor grade ranged from 7% for grade 1, 12% for grade 2, 19% for grade 3, and 63% for grade 4. Of the 96 patients included in the QI work, 66% were men, with a mean age of 59.4 years (SD, 15.6 years). Tumor grade ranged from 3% for grade 1, 8% for grade 2, 11% for grade 3, and 77% for grade 4.
Impact of Process Improvement
The statistical process control chart in Figure 2 shows consecutive patients and the proportion of best practice measures that were met pre- and post-QI work. Before the QI work, the overall mean was 63%, with wide variations, meaning patients could have these best practices met all the time or none of the time (lower and upper control limits were beyond 0% and 100%). After beginning the QI work in June 2011, compliance with the best-practice measures was significantly better (≥ eight consecutive patients having best-practice measure > historical overall mean of 63%), with an overall mean of 85% (P = .003). Besides an increase in the overall mean, there was also a reduction in variation, meaning that best practices would be met from 50% (lower control limit) to 100% (every time).
Figure 2.
Individual values and moving range chart of percent standards of care achieved by patients with glioma before and after quality improvement (QI) work. Control limits (CLs) are set at three sigma (pre-QI CL, 0% to 100%; post-QI CL, 50% to 100%). NOTE. Ovals indicate special-cause events (statistically significant change in underlying process whereby ≥ one point goes beyond CLs). Data not collected January through May 2011.
After starting the QI work, the team members identified two patients whose best-practice measures fell below the lower control limit of 50% (special-cause result). In one case, neuro-oncology was not consulted, because of an initial suspicion of brain metastasis. After discharge, the patient had a complicated fall, resulting in additional delays in care. In the other case, the patient was admitted for biopsy and discharged on the same day, and thus, neuro-oncology was not consulted before discharge. This event resulted in poor performance in the best-practice measures. The team reviewed the processes of care for these two patients, which facilitated discussion, highlighted the intricacies of the process, and stressed the importance of following the rules of the pathway. Figure 3 shows the degree to which individual measures rose with these QI efforts. The team members observed significant improvements in measures assessing for: one, use of standard postoperative order sets (43% v 76%; P = .001); two, preoperative notification of neuro-oncology (39% v 97%; P < .001); three, scheduling of a follow-up appointment before hospital discharge (27% v 71%; P < .001); four, evaluation with the social worker within 2 weeks (29% v 76%; P < .001); and five, the patient is presented at tumor board (83% v 97%; P = .01) within 2 weeks (71% v 99%; P < .001). We did not find any patients without symptomatic edema who were receiving steroids.
Figure 3.
Proportion of patients with glioma who met criteria for individual best-practice measures before and after quality improvement initiative, from January to December 2010 and June 2011 to September 2013. AED, antiepileptic drug; DC, discharge. (*) P = .01; (†) P < .001; (‡) P = .001.
Resources provided by the cancer center helped to support this effort in an ongoing manner. In addition, project sustainability was assured by continuous review of the data dashboard and quarterly updates to all team members at tumor board on the overall progress of the project.
Discussion
The care of patients with glioma has become increasingly complex and involves multiple disciplines that need to work in seamless synchrony to optimize patient safety and provide consistent treatment. A previous study found that patients with high-grade glioma who receive coordinated multidisciplinary care (informed by specific quality-of-care indicators) showed significant improvement in survival when compared with patients who were managed by standard referral pattern.21
Our multidisciplinary neuro-oncology group used QI methodology to improve the quality of care in patients with newly diagnosed gliomas at a National Cancer Institute–designated comprehensive cancer center. The proportion of patients who met best-practice measures, as defined by the group, rose significantly from 63% pre-QI work to 85% post-QI work. Concurrent review of individual results enabled team members to see how their work was connected to others, to identify process weaknesses, and to brainstorm for additional strategies to ensure sustainability. These collaborations also fostered camaraderie and motivated individuals to remain engaged in the work despite occasional challenges and setbacks. Developing data dashboards that were periodically reviewed with team members and assigning process owners were equally important in helping to sustain project gains.
Except for recommendations that guide surgery, chemotherapy, and radiation treatment, there is limited information on which clinical care process measures are associated with improved clinical outcomes in patients with glioma. The lack of clinical guidelines contributes to the variability in practice patterns observed in this population.9 At the same time, patients with a brain tumor undergoing surgery are at high risk for complications that can result from practice or system errors.22 We developed a clinical treatment pathway to help address these concerns and selected measures based on consensus after reviewing the scarce data available in the literature.
To our knowledge, there have been no prior reports on prospective QI work in patients with glioma. Recently, Rahman et al22 reported on the development of standard performance measures for adult patients with a brain tumor in the perioperative period after querying the Nationwide Inpatient Sample database for all hospitalizations involving a brain tumor. These measures included several Agency for Healthcare Research and Quality patient safety indicators and the Centers for Medicare and Medicaid hospital-acquired conditions.22 The most common patient safety indicators were postoperative respiratory failure, deep vein thrombosis, and sepsis, whereas the most common hospital-acquired conditions were falls, trauma, and pressure ulcers.22 In the second phase of this QI initiative, we have incorporated some of these outcomes, including 30-day postoperative complication rates (eg, falls and urinary tract infections) and hospital readmissions.
One of the limitations of our work was the inability to demonstrate that improvements in our selected quality measures would translate into improved outcomes, better use of resources, and enhanced patient and staff satisfaction. Many variables affect survival of patients with high-grade glioma, and the interventions we implemented were limited to patients entering our microsystem. A longitudinal QI project that includes all components of care and a large population of patients will be required to determine benefits in outcomes. Other sectors of health care, however, have already found that QI methodology can lead to significant improvements in clinical outcomes.23 In the future, it is also conceivable that clinical pathways may result in cost savings as clinics transition to a value-based payment structure.24,25
A small proportion of all patients were diagnosed with low-grade glioma. Except for timeliness of radiation oncology evaluation, we feel that our best-practice measures are equally relevant to this population. Because of the small numbers, however, we were unable to explore whether meaningful differences existed between these two groups.
There were several observations from our work showing that the streamlined process provided enhanced consistency of care. For example, the team was notified when a patient was admitted for surgery, information that resulted in more prompt presentation of patients at tumor board (pre- v post-QI work: 71% v 99% presented within 2 weeks), earlier assessment by social workers (pre- v post-QI work: 29% v 76% evaluated within 2 weeks), and scheduling of follow-up appointments before discharge. Through this QI initiative, the team also became aware of the priority of this work, enhancing multidisciplinary coordination, and the gains we achieved encouraged the group to remain engaged in the initiative. Furthermore, we identified process changes that were not dependent on the actions of individual staff members.
In conclusion, the timeliness and consistency of care provided to patients with glioma in the perioperative period can be improved through the use of QI principles. Involving clinical leaders, developing a system for ongoing measurement, and reporting of system performance are important for sustaining the gains of QI work. In the future, more of the burden for reimbursement will be placed on quality rather than quantity of care. It will be crucial for us to develop clinical pathways that accurately reflect quality care and are associated with improved clinical outcomes.
The next phase of our QI initiative is currently under way and focuses on the delivery of acute care to patients with glioma. Once this phase is complete, we plan to examine and postulate best-practice measurements for the chronic and palliative care of patients with glioma at our institution. We acknowledge that best practices can vary by institution and availability of services but hope that this work will provide an opportunity to start a dialogue and to standardize by consensus some minimal quality processes that can improve patients' quality of life and extend their survival.
Acknowledgment
Presented in part at the American Society of Clinical Oncology Quality Care Symposium, San Diego, CA, November 30-December 1, 2012. We thank Kati Fuller, Nancy Lapoint, Mary Robinson, Janet Stephenson, David Nalepinski, and Marylin Bedell for their assistance in carrying out this quality improvement initiative; Christopher Dant, PhD, for reviewing our manuscript and providing us with editing suggestions; and the original sponsors of this project, including Mark Israel, MD, Director of the Norris Cotton Cancer Center, and David Roberts, MD, Section Chief of Neurosurgery.
Appendix
Table A1.
Best Practice Measures for the Perioperative Treatment of Patients With New Diagnosis of Glioma*
| Measure No. | Best Practice Measure | Supporting Evidence |
|---|---|---|
| 1 | The Neuro-Oncology Program is notified preoperatively of new admissions by the neurosurgery service. | Patients with glioma benefit from multidisciplinary treatment.14; Mason WP: Curr Oncol 14:110-117, 2007. Well-coordinated and multidisciplinary care may be associated with improved outcomes.21 |
| 2 | The Neuro-Oncology Program evaluates a patient during their hospital admission. | Patients with glioma benefit from multidisciplinary treatment.14; Mason WP: Curr Oncol 14:110-117, 2007. Well-coordinated and multidisciplinary care may be associated with improved outcomes.21 |
| 3 | Standard post-operative orders are used. | Existing evidence suggests that clinical decision supports such as standardized order sets may lead to improved outcomes in health care settings. Amarasingham R: Arch Intern Med 169:108-114, 2009; Chan AJ: Int J Technol Assess Health Care 28:235-240, 2012; Ballard DJ, in Advances in Patient Safety: New Directions and Alternative Approaches. Rockville, MD, AHRQ, 2008 |
| 4 | Appropriate use of corticosteroids.† | Corticosteroid use in glioma patients reduces the risk for complications related to cerebral edema.14,17 |
| 5 | Appropriate use of AEDs.† | AEDs may be useful in patients with glioma who have a history of seizure. There is no benefit, however, to using AEDs in patients with glioma who have no history of seizure.14–16 |
| 6 | Patients are evaluated by the social worker within 2 weeks of admission. | Symptoms of psychological distress such as depression are common with brain cancer and can be associated with poorer quality of life. Fox SW, J Nurs Scholarsh 39:61-67, 2007 Addressing the psychosocial needs of patients is an important element of cancer treatment. AOSW Oncology Social Workers Standard of Practice, 2012, http://www.aosw.org; NCCN Clinical Practice Guidelines in Oncology, http://www.NCCN.org; American Society of Clinical Oncology: http://qopi.asco.org/Documents/QOPISpring2012MeasuresSummary_000.pdf; Spezeski J: http://www.braintumorcommunity.org/site/DocServer/Needs_Assessment_Report2009.pdf?docID=4441. |
| 7 | Follow-up appointment is scheduled with neuro-oncology prior to hospital discharge. | Patients with glioma benefit from multidisciplinary treatment.14 Mason WP: Curr Oncol 14:110-117, 2007. Well-coordinated and multidisciplinary care may be associated with improved outcomes.21 |
| 8 | Patients are evaluated by radiation-oncology within 2 weeks of discharge.‡ | Patients with glioma benefit from multidisciplinary treatment.14 Well-coordinated and multidisciplinary care may be associated with improved outcomes.21 |
| 9 | Patients are presented at tumor board. | Patients with glioma benefit from multidisciplinary treatment.14; Mason WP: Curr Oncol 14:110-117, 2007. Well-coordinated and multidisciplinary care may be associated with improved outcomes.21; Lutterbach J: Onkologie 28:22-26, 2005. |
| 10 | Within 2 weeks of surgery. |
Abbreviation: AEDs, anti-epileptic drugs.
Based on symptoms and location of glioma, certain standards of care did not apply to a subset of patients. For example, patients with minimal or no symptoms did not receive steroids.
AEDs and/or steroids are started in patients if clinically indicated.
Referral to radiation-oncology if clinically appropriate.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Author Contributions
Conception and design: Evelyn M. Schlosser, Karen Homa, Nathan E. Simmons, David H. Sargent, Linda P. Mason, Tobi J. Cooney, Nancy L. Kennedy, Camilo E. Fadul
Collection and assembly of data: Evelyn M. Schlosser, Jennifer A. Snide, Lesley A. Jarvis, Camilo E. Fadul
Data analysis and interpretation: Natalie B.V. Riblet, Evelyn M. Schlosser, Karen Homa, Jennifer A. Snide, Camilo E. Fadul
Manuscript writing: All authors
Final approval of manuscript: All authors
References
- 1.Schwartzbaum JA Fisher JL Aldape KD, etal: Epidemiology and molecular pathology of glioma Nat Clin Pract Neurol 2:494–503,2006 [DOI] [PubMed] [Google Scholar]
- 2.Ohgaki H, Kleihues P: Epidemiology and etiology of gliomas Acta Neuropathol 109:93–108,2005 [DOI] [PubMed] [Google Scholar]
- 3.Chandana SR Movva S Arora M, etal: Primary brain tumors in adults Am Fam Physician 77:1423–1430,2008 [PubMed] [Google Scholar]
- 4.Hess KR, Broglio KR, Bondy ML: Adult glioma incidence trends in the United States, 1977-2000 Cancer 101:2293–2299,2004 [DOI] [PubMed] [Google Scholar]
- 5.Johnson DR: Rising incidence of glioblastoma and meningioma in the United States: Projections through 2050 J Clin Oncol 30:131s,2012suppl abstr 2065 [Google Scholar]
- 6.Stupp R Hegi ME Mason WP, etal: Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial Lancet Oncol 10:459–466,2009 [DOI] [PubMed] [Google Scholar]
- 7.Reardon DA, Wen PY: Therapeutic advances in the treatment of glioblastoma: Rationale and potential role of targeted agents Oncologist 11:152–164,2006 [DOI] [PubMed] [Google Scholar]
- 8.Omuro A, DeAngelis LM: Glioblastoma and other malignant gliomas: A clinical review JAMA 310:1842–1850,2013 [DOI] [PubMed] [Google Scholar]
- 9.Chang SM Parney IF Huang W, etal: Patterns of care for adults with newly diagnosed malignant glioma JAMA 293:557–564,2005 [DOI] [PubMed] [Google Scholar]
- 10.Phillips C: Spotlight: Clinical pathways in cancer care catching on http://www.cancer.gov/ncicancerbulletin/090412/page6
- 11.Nelson EC Batalden PB Lazar JS, etal: Nelson EC, Batalden PB, Lazar JS.Understanding clinical improvement: Foundations of knowledge for change in health care systems Practice-Based Learning and Improvement: A Clinical Improvement Action Guide 1–12,2007ed 2 Oakbrook Terrace, IL: Joint Commission on Accreditation of Healthcare Organizations [Google Scholar]
- 12.Nelson EC: Nelson EC, Batalden PB, Lazar JS.Appendix A, improving care: Clinical improvement worksheets Practice-Based Learning and Improvement: A Clinical Improvement Action Guide 135–155,2007ed 2 Oakbrook Terrace, IL: Joint Commission on Accreditation of Healthcare Organizations [Google Scholar]
- 13.Caldwell C, Butler G, Posten N: Lean-Six Sigma for Healthcare: A Senior Leader Guide to Improving Cost and Throughput 2009ed 2 Milwaukee, WI: ASQ Quality Press [Google Scholar]
- 14.Stupp R Tonn JC Brada M, etal: High-grade malignant glioma: ESMO clinical practice guidelines for diagnosis, treatment and follow-up Ann Oncol 21:v190–v193,2010suppl 5 [DOI] [PubMed] [Google Scholar]
- 15.Glantz MJ Cole BF Forsyth PA, etal: Practice parameter: Anticonvulsant prophylaxis in patients with newly diagnosed brain tumors—Report of the Quality Standards Subcommittee of the American Academy of Neurology Neurology 54:1886–1893,2000 [DOI] [PubMed] [Google Scholar]
- 16.Sirven JI Wingerchuk DM Drazkowski JF, etal: Seizure prophylaxis in patients with brain tumors: A meta-analysis Mayo Clin Proc 79:1489–1494,2004 [DOI] [PubMed] [Google Scholar]
- 17.Roth P, Wick W, Weller M: Steroids in neurooncology: Actions, indications, side-effects Curr Opin Neurol 23:597–602,2010 [DOI] [PubMed] [Google Scholar]
- 18.Carey RG: Improving Healthcare With Control Charts: Basic and Advanced SPC Methods and Case Studies 7–8,2003 16–22.Milwaukee, WI: ASQ Quality Press [Google Scholar]
- 19.Hart MK, Hart RF: Statistical Process Control in Health Care 1–370,2002Pacific Grove, CA: Duxbury [Google Scholar]
- 20.Benneyan JC, Lloyd RC, Plsek PE: Statistical process control as a tool for research and healthcare improvement Qual Saf Health Care 12:458–464,2003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Back MF Ang EL Ng WH, etal: Improvements in quality of care resulting from a formal multidisciplinary tumour clinic in the management of high-grade glioma Ann Acad Med Singapore 36:347–351,2007 [PubMed] [Google Scholar]
- 22.Rahman M Neal D Fargen KM, etal: Establishing standard performance measures for adult brain tumor patients: A nationwide inpatient sample database study Neuro Oncol 15:1580–1588,2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.American Hospital Association. Trend Watch: Hospitals Demonstrate Commitment to Quality Improvement. www.aha.org/research/reports/tw/12oct-tw-quality.pdf.
- 24.Rauh SS Wadsworth EB Weeks WB, etal: The savings illusion: Why clinical quality improvement fails to deliver bottom-line results N Engl J Med 365:e48,2011 [DOI] [PubMed] [Google Scholar]
- 25.Society of Hospital Medicine. Physician Value-Based Payment Modifier. www.hospitalmedicine.org/Web/Advocacy/Key_Issues/Physician_Value_Based_Payment/Web/Advocacy/Value_Based_Payment.aspx?hkey=d37271dd-d4a8-459c-a354-f31d77bb5d8e.