|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Ensuring Quality Cancer Care by the Use of Clinical Practice Guidelines and Critical PathwaysFrom the Massey Cancer Center and Department of Internal Medicine, Medical College of Virginia at Virginia Commonwealth University, Richmond, VA. Address reprint requests to T.J. Smith, MD, Division of Hematology/Oncology, Medical College of Virginia Box 980230, Virginia Commonwealth University, Richmond, VA 23298-0230; email: tsmith@ hsc.vcu.edu.
PURPOSE: We describe the impact of clinical practice guidelines (CPGs) on improvement in oncology treatment processes or outcomes. METHODS: We performed a comprehensive search of the literature from 1966 to the present and a directed review of the literature. RESULTS: Improvements have been demonstrated in compliance with evidence-based guidelines or evidence-based medicine, and in short-term length of stay, complication rates, and financial outcomes. The data suggest that patient satisfaction can be maintained despite a shorter length of stay. There has been one example of province-wide improvement in disease-free and overall survival of breast cancer patients coincident with the adoption of CPGs. The components of successful guidelines can be summarized as follows: (1) development is based on evidence, with the guideline formulated by key physicians in the group; (2) the guidelines are disseminated to all affected health care professionals for critique; (3) implementation includes direct feedback on performance to physicians or general feedback on system performance; and (4) there is accountability for performance according to the guidelines. This accountability can consist of voluntary peer pressure to conform to evidence-based medicine, and it does not require a financial reward or penalty. CONCLUSION: Some attempts to improve practice have been moderately successful in achievement of reduced health care costs, reduced hospital length of stay, and possibly improved outcomes. Other methods that are still in use have been demonstrated to have little effect. Programs that have not succeeded have relied on voluntary change in practice behavior without incentives to change or have had no accountability component. Further research is needed to assess how guidelines are enacted in organizations other than those demonstrably committed to improvement, ways to improve compliance of health care providers who are not committed to change, and methods to improve accountability.
CLINICAL PRACTICE guidelines (CPGs) are one approach to improved quality of care and cost control. CPGs are defined as "systemically developed statements to assist both practitioner and patient decisions about appropriate health care for specific clinical circumstances."1 We sought information on the impact of CPGs as part of a project on quality of care,2 and we addressed the following questions: (1) the state of guidelines for cancer care; (2) evidence for use; and (3) the effect on cancer care quality. The development of CPGs (to include clinical pathways, care maps, and so on) has been driven by concerns regarding the rapid escalation of cancer care costs and the quality of cancer care. Concerns regarding practice variations, quality of care, and suboptimal patient outcomes seem to be well founded, with a clear-cut relationship between high volume and/or specialization and better quality of care.3 The data from noncancer branches of medicine demonstrate some evidence for improved care and better outcomes.4-7 In the most rigorous review, explicit guidelines improved the process of care in 55 of 59 published evaluations, and in nine of 11 reports that assessed outcomes.5 However, the influence of guidelines has not been evaluated rigorously in cancer treatment. The purpose of this article is to review the current data on creation, dissemination, evaluation, and use of CPGs in oncology.
We searched MEDLINE from 1966 to the present using the terms CPGs, clinical pathways, care maps, outcomes management, and oncology. A directed search of primary papers was undertaken. Only studies that included an evaluation component are included here, and there are no other exclusions. We have included both European and American studies because the methods are similar.
Background of CPGs Guidelines and evidence. One of the conundrums of guideline preparation is that if good scientific evidence exists for a procedure, then practice should be uniform and no guideline is needed. However, the least uniform practices are most likely to lack uniform evidence on which to base a guideline. Cook et al11 demonstrated that systematic reviews of the evidence can assist in the process of guideline creation and implementation in several ways: (1) to provide the basis for the guideline; (2) to point out areas of needed research; and (3) to educate practitioners and help keep them up to date with the literature. They point out that if the data to make a guideline are inconclusive, then the guideline should be targeted for early assessment to ensure that its outcomes are acceptable. Barriers to physician compliance. Grilli and Lomas12 attempted to define the characteristics of guidelines and practitioners that lead to compliance and improved care. Change in practice was most likely to occur when the guideline was based on solid data, provided discrete and concrete steps for change, and did not require knowledge or skills outside of the practitioner’s current realm.12 This evaluation was performed only for guidelines with voluntary implementation and no accountability, so the study may not be applicable to health care systems that incorporate accountability and specific education. Katterhagen13 outlined the obstacles to overcome in his attempts to implement guidelines in a California health care system: "a lack of market awareness and need to decrease costs and produce superior outcomes"; "anger over declining autonomy and power"; and "falling incomes." He noted that physicians "basically respond to three forces: financial reward, a desire for a good reputation, and peer pressure." He also advocated 10% of time spent on guideline development and 90% on implementation, and direction of 80% of change efforts toward the 20% of medical staff members who are the opinion leaders;13 however, these hypotheses have not been tested. Using guidelines to monitor practice. To assess its impact, a guideline must be converted into an evaluation tool. This process has been well described in Agency for Health Care Policy and Research (AHCPR) literature,14,15 but the design of a guideline explicitly for this process is often overlooked. Each guideline should have readily discernible benchmarks or targets for good practice.16 A framework for analysis of guideline success. Grimshaw and Russell5 reported a practical plan to improve effectiveness of CPGs by a strategy of development, dissemination, and implementation. The guidelines with highest probability of success include those with internal development, specific educational intervention, and a patient-specific reminder at the time of consultation. The authors suggested that national external, published-only, general reminder programs have a low probability of success, and they reviewed the evidence that guidelines by themselves do not change process or outcomes substantially. We have modified their original framework to include accountability, ie, the effort to hold physicians responsible for their clinical practice either by direct audit, direct practice monitoring, or incentives to perform within set limits (Table 1).
Quality of Published Guidelines There have been several comparisons of the conclusions of alternative guidelines, as listed in Table 2,17-27 but no studies have been published on cancer outcomes under alternative guidelines. Follow-up care of colorectal cancer patients could be simplified substantially if the recommendations from the largest randomized clinical trial were followed. Breast cancer care at Roswell Park Cancer Institute fit the National Comprehensive Cancer Network guidelines, with acceptable variations, and these guidelines were similar to others. All groups who have developed guidelines for non–small-cell lung cancer have concluded that treatment is warranted to improve survival. Table 3 summarizes the probability of success of each program on the basis of the model of development, dissemination, implementation, and accountability16,28-39; further explanation is provided below. Specific trials are described in detail because many of the failed methods are still in common use.
Specific Trials United States, 1987: The community hospital oncology program (CHOP). Ford et al28 concluded that CHOP, designed to improve cancer care in United States hospitals, provided "no evidence of diffusion of guideline principles to the majority of practicing physicians." Although the study was performed many years ago, the methods are still in common use and shed light on current results. In summary, the health care professionals who saw the most cancer patients in each community developed "site-specific patient management guidelines for staging, medical management, nursing, and rehabilitation."28 For women with positive lymph nodes, only 73% received a medical oncology consultation despite 10 years of experience to demonstrate the benefit of adjuvant treatment. Some of the most important variance was associated with physician characteristics, not patient or disease characteristics. Physicians who were principal investigators or on the executive committee conformed slightly better, but there was no evidence of diffusion to other doctors, as listed in Table 4. Younger physicians tended to give more therapy, and physicians with more recent and better cancer training tended to perform better when measured against the accepted guidelines.
There was no demonstrable impact of the program, and the data fail to support the theory that participation in the development process enhances adoption and diffusion. Data on actual patient outcomes were not studied. The CHOP participants that were able to ensure staging did so only by requiring completion of forms before submission of specimens to the pathology department, or loss of privileges if forms were not completed. As the authors note, these measures were successful but did not require guidelines. The authors conclude that for CPGs, "leadership and organizational commitment seem to be the necessary ingredients."28 On the CPG analysis chart (Table 1), this intervention would be predicted to have a low chance of success. The guideline was developed locally, but there was no plan for implementation and no accountability to change practice. United States, 1994: Lung, breast, and colorectal cancer management. Patton and Katterhagen29 described a successful program to standardize care for cancer patients in Northern California, as was necessary to adapt to managed care influences. The authors used a pre/post method to compare patients on the guideline with those treated the previous year. Length of stay (LOS) and costs declined substantially, as listed in Table 5. The exact contribution of their critical pathways to the change in outcomes cannot be assessed, as there is no possible control group (a similar health system immune to the same cost pressures) in Northern California.
The authors noted a number of standardized patient procedures that were perceived to provide better medical and nursing care after the guideline, which included the following: (1) mandatory preoperative teaching for respiratory care and bowel cleansing; (2) use of patient-controlled analgesia with morphine rather than pro re nata meperidine or morphine (which resulted in a decrease in postoperative confusion from 11% to 2%); and (3) discharges distributed throughout the week rather than concentrated in the preweekend period. The likelihood that this program would succeed was high because of internal development, specific educational intervention, patient-specific reminders at the time of encounter, and monitored practice with feedback given. The authors note that the success of their system depends on understanding the process of medical care and the power of group process to change practice once consensus is reached. United States and Canada, 1994: Management of prostatectomy and other urologic procedures. Radical prostatectomy is the most common urologic cancer operation. After implementation of a clinical pathway, LOS decreased from 5 days to 3.6 days, and costs decreased from $7916 to $6934 (-12%).30 The team concluded that standardized preoperative and postoperative management led to significant cost savings. The hospital LOS each year had already decreased from 7 days to 6.1 days, to 5 days, to 3.6 days, so a trend was evident even before the guideline was adopted. After implementation, formal quality of life scores were high, as was patient satisfaction despite the decreased LOS, but there were no data before the clinical pathway was adopted.40 The likelihood of success with this program was high because of internal development, specific educational intervention, patient-specific reminders at the time of encounter, and monitored practice with feedback given. United States, 1995: Outcomes management for gynecologic oncology. In 1994, Morris et al16 at M.D. Anderson Cancer Center established practice guidelines and collaborative care paths for total abdominal hysterectomy with oophorectomy and staging biopsies, tumor reduction surgery, radical hysterectomy with node dissection, and vulvar resection with groin dissection. The program was initiated by a group of four gynecologic oncologists and one nurse practitioner, with a strong administrative mandate to control costs and maximize patient outcomes. The draft care paths were circulated to all involved groups for comment. Accountability was maximized by documentation for each care path, including standard forms, standard data collection sets, patient education forms, and so on. The results from the first 30 patients who underwent surgery in accordance with the care path were compared with 29 patients matched for age, indications for surgery, stage, and attending surgeon. These planning patients were chosen from the time the care paths were under discussion to ascertain whether there was a Hawthorne effect of better behavior under observation. Another control group of 73 patients was chosen from the time before the care paths were under discussion. All groups were comparable, and differences among the groups could not account for the findings. Results are listed in Table 6.
The implementation of the care path and accountability significantly improved outcomes, decreased lengths of stay, decreased costs, and kept patient satisfaction high. The authors could not identify any adverse outcomes, even as they further reduced LOS to 3 days. The authors made some specific points on the basis of their experience but not otherwise verified: (1) to work, care paths must be physician driven; (2) the paths must define desired outcomes, eg, discharge when pain is controlled instead of on a specific day; (3) there must be ongoing assessment of achievement of the desired outcome; and (4) care paths should be defined using formal methods based on evidence and consensus. The likelihood of success of this program was high because of internal development, education of all providers, and high accountability by monitoring patient-specific outcomes at the time of encounter. This program seems to have improved the process, immediate surgical outcomes, and cost of surgical care. This study is immediately applicable to most surgical cancer admissions in the United States. United States, 1994 to present: The American Society of Clinical Oncology (ASCO). Before the publication of the ASCO guidelines on hematopoietic growth factors, the majority of projected uses fit the ASCO guidelines; however, there were significant variances from evidence-based use. For example, there was no evidence to support use in febrile neutropenia, afebrile neutropenia, acute leukemia, and dose intensification outside a clinical trial. In 1994, the survey demonstrated that a significant number of practitioners used colony-stimulating factors (CSFs) in these ways. Of note, the use of CSFs was strongly related to type of practice; health maintenance organizations and academic physicians used less CSFs and more often chose a dose-reduction strategy.31 The actual impact of the ASCO guidelines has been difficult to prove. Surveys were conducted in 1997 after the membership had been exposed to the 1994 guideline and 1996 update, both published in the Journal of Clinical Oncology. The data can be interpreted in two ways: either they demonstrate substantial improvement compared with usage before the guideline,41 or they demonstrate continued overuse that never has been supported by the evidence. There was less overuse of CSFs in one scenario (treatment of febrile neutropenia); use decreased from 39% to 29% (P < .05.) However, the evidence-based use should be approximately 0%. Treatment of febrile neutropenia with right lower lobe pneumonia diminished slightly, from 54% to 46% (P < .05). Again, the evidence-based use should be approximately 0%. Significant routine overuse of CSFs persisted in primary prophylaxis in a range from 6% to 39% dependent on scenario; the use should be approximately 0%, as there is no evidence to support efficacy. Significant overuse (up to 30% of respondents stated they always, usually, or sometimes use CSFs) as treatment for afebrile neutropenia persisted despite published evidence of no benefit.42 Physician characteristics were again likely to predict use; those in academic medical centers or health maintenance organizations were more likely to use dose-reduction strategies and less likely to use CSFs than physicians in private practice. In the 1997 survey, oncologists were questioned on use of CSFs for acute leukemia, a new indication since the 1994 guidelines.43,44 Physician or practice characteristics were strongly correlated with propensity to use growth factors: those in a fee-for-service practice or those who were hematologists or hematologists/oncologists were more likely to use CSFs. The majority of physicians use a CSF in ways proven to be effective; 79% of respondents said they would start CSF treatment after chemotherapy or when bone marrow aplasia was documented, both of which are proven strategies. However, 21% would initiate CSF after the patient became neutropenic or neutropenic and febrile, and both of these are unproven strategies.45 A low likelihood of success of the ASCO CSF guidelines would be predicted, as demonstrated, given the lack of implementation and accountability components. In addition, there currently are no disincentives to administration of CSFs, CSFs rarely cause harm, and they can be a significant source of income to a practice. United States, 1996: Cancer pain relief in Utah. Several studies have documented inadequate pain relief among cancer patients46-49 and terminally ill patients.50 In 1994, the AHCPR established a clinical practice guideline for management of cancer pain. Baseline data collected on 10 patients at seven acute care hospitals in Utah demonstrated low compliance with the guidelines. Follow-up data in 1996, after a statewide education program, demonstrated improved compliance with six core guidelines, and data were reported on seven audited components (Table 7).32
There are not enough data to ascertain whether the program is successful, because pain scores were not measured. The high baseline use of opiates may make it difficult to demonstrate improvement in pain scores. On the basis of the need for better pain management, the program has been initiated statewide, but no results are available. United States, 1998: Implementation of cancer pain guidelines. DuPen et al33 transformed the AHCPR cancer pain guidelines into algorithms, then randomized 81 cancer patients to standard practice or intervention practice with the algorithm. The setting was the outpatient clinic of 26 Washington State oncologists in 13 practices. Pain intensity was significantly reduced in the pain algorithm group compared with the standard community practice group, with no other changes in quality of life or symptom control. The authors conclude that the pain algorithm successfully reduced pain scores. An integral part of the solution was a nurse practitioner who ensured that prescribers received the pain assessment data, because the optimization of pain medicine relies heavily on pain assessment scores. The adherence to guidelines was no different between the two groups, which suggests that the added nursing assistance provided assessment but not better compliance. The likelihood of success of this program was high because of internal development, education of all providers, and high accountability by monitoring patient-specific outcomes at the time of encounter. United States, 1996: The American Urologic Association early prostate cancer guidelines. In December 1995, the American Urologic Association recommended that surgery, radiation, or surveillance be offered as treatment options for patients with localized prostate cancer. The guideline was highly publicized in the literature, at scientific meetings, and in the prostate cancer patient community. After the guidelines were released, 1,000 men were surveyed. These patients comprised 500 who had undergone radical prostatectomy before implementation of the guideline, and 500 patients who received treatment 6 months after the guideline was established; 674 patients responded. The average number of treatments offered before the guideline was 1.98, which increased to 2.30 after release of the guideline (P = .003). The total number of specialists consulted before and after release of the guidelines was 1.35 and 1.47, respectively (P = .03). A copy of the patient guide to treatment for localized prostate cancer was reported received by 30% of the patients.34 The authors concluded that the purpose of the guidelines was served: more patients had the opportunity to make informed choices. These guidelines would be expected to have an above-average probability of implementation on the basis of the specific educational intervention and because the guidelines were mailed to each practitioner. In addition, the required change in practice is minimal and fits with societal changes in prostate cancer treatment. Alternative explanations for success could include the intense media exposure and increased consumer movement in the prostate cancer community.
Relevant Experience in the World The processes of care were relatively uniform. For patients with indications for treatment, compliance was 95% for the radiation, 77% for the chemotherapy, and 68% for the tamoxifen arms. No patients received radiation without a radiation oncology consultation or chemotherapy without a medical oncology consultation. Referral to and indications for adjuvant radiation therapy were high at 95.4% for adjuvant treatment after breast conservation. Only seven patients received radiation without a demonstrable indication. Chemotherapy compliance also was high at 95.9%. Of 83 women with an indication for chemotherapy, 64 (77.1%) received chemotherapy treatment. All patients who consulted a medical oncologist received chemotherapy if it was indicated. Nineteen patients did not receive indicated chemotherapy: 13 of these patients had not consulted a medical oncologist, and six patients consulted a community medical oncologist. Referral, and hence treatment, were far more likely for patients younger than age 50 years. Compliance for tamoxifen when indicated (including use when indicated and nonuse when not indicated) was 89.2%, but only 67.8% of patients with an indication to use tamoxifen actually received it. The impact of the guidelines seems to be impressive, but there may be other explanations for the good care. There was an influential multidisciplinary clinic in 1991 at BCCA that had some impact on practice patterns as well as the structure of the guidelines. The care in British Columbia has tended to be centralized, with much emphasis on the expert opinions of the BCCA leaders (G. Browman, personal communication, 1998).
This study represents the only guideline that may be associated with improved survival. Olivotto et al51 demonstrated that the improvements in disease-free survival (DFS) and overall survival (OS) noted during the time period of guideline implementation were similar to those observed in clinical trials. For patients younger than 50 years, DFS increased from 65% to 76% and OS increased from 65% to 75%. For women age 50 to 89 years, DFS at 7 years increased from 63% to 70%, and OS increased from 54% to 58% (P The guidelines would be predicted to have a high probability of success, and they did, because of province-wide development, dissemination, and implementation plans. There were more variations from the guidelines among patients of the community oncologists and those who were not referred to other specialists by their surgeon. Ontario, 1995: Experience with paclitaxel for metastatic breast cancer—Changing practice with accountability. Levine et al36 described the Ontario experience with paclitaxel, wherein reimbursement for the drug was contingent on practitioner evaluation of cancer response; 412 of 418 forms were completed, and most patients received the specified dose. The disease response rate was only 17%. The likelihood of success of this program was high because of internal development, education of all providers, and high accountability by monitoring patient-specific outcomes at the time of encounter. Form completion and compliance with guidelines can be improved by a patient-specific reminder at the time of the encounter, accountability (if the forms are not completed, no drug can be dispensed), and reimbursement. An alternative explanation is that physicians who bother to treat with new drugs can fill out forms. Italy, 1987: The Italian National Research Council evaluation. Grilli et al37 evaluated a program similar to CHOP that was initiated in Italy after repeated demonstrations of a gap "between best care and that which is actually delivered in the community." Acceptance was voluntary, and there was no feedback on performance or accountability. For the audit, 742 breast, 641 colorectal, and 100 ovarian cancer patient records were reviewed. A gold standard was defined, which consisted of 36 of the 53 recommendations that were most important and that met with the highest level of agreement; this standard was clearly and definitively reported in the text of the guidelines. Dissemination and acceptance of the survey guidelines were unsatisfactory, as listed in Table 8. Physician practice volumes and personal characteristics were significantly associated with treatment choices. High-volume physicians more often favored breast-conserving surgery and adjuvant treatment, appropriate colorectal cancer surgery, and adjuvant therapy and chemotherapy for advanced ovarian cancer (P < .01 or .05.) Physicians who were aware of the guidelines made better therapy choices, even when adjusted for volume and other confounding variables.
The likelihood of success of this program was expected to be low given its national development, lack of education and implementation, and lack of accountability for change. The authors note that results were disappointing and that efforts to improve cancer care with a guideline diffusion approach seem to have had a negligible effect on cancer treatment. The authors state that for practice to change, the scientifically sound message must reach the right professional audience, who must be willing and able to act on it. France, 1995: Improving care with breast and colon cancer guidelines in a French cancer center. Implementation of CPGs is a regional quality assurance project initiated by the comprehensive cancer center in Lyon, Centre Leon Berard.38 The center has 235 beds, 80 oncologists, and treats 3,000 new cases each year. The goals were to assist oncologists in their decision making, reduce variation, optimize health benefit, ensure rational chemotherapy prescription within the center, and provide local leadership. A task force of medical, radiation, and surgical oncologists developed the guidelines in 1993 based on evidence with consensus, and all practitioners then reviewed the guidelines. The guidelines were disseminated in 1994 and made available in paper, computer disk, and online at the center. Practitioners did not receive feedback but were aware of the ongoing audit and accountability. A before/after study was conducted, and 1993 and 1995 were used as comparison years. Patients included 100 randomly selected breast cancer patients, 77 preintervention colon cancer patients, and 81 postintervention colon cancer patients. The initial treatment sequence was audited and compared with the guidelines. The main outcome was the number of medical decisions that conformed to the guidelines. The secondary outcome was the number of medical decisions judged to be based on either the guidelines or on published evidence (as determined beforehand). The observed compliance rate in 1995 was significantly better than that in 1993. In 1993, only 42% of initial treatments conformed to the CPGs or were based on scientific evidence, compared with 68% in 1995. Specific results for compliance with guidelines are listed in Table 9.
Overall, there was a striking increase in the number of breast cancer treatment decisions compliant with the guideline or based on scientific evidence: from 19% to 54% (P < .001). For colorectal cancer overall, the compliance for the whole treatment sequence increased from 50% to 70% (P = .009). In 1993, 71% of the colon cancer treatment decisions were compliant or judged as in accordance with the scientific evidence, which increased to 81% in 1995 (not statistically significant). The compliance rates did not change with the exception of adjuvant chemotherapy administration. The likelihood of success of this program was high because of internal development, education of all providers, and high accountability through monitoring patient-specific outcomes at the time of encounter. The success of the guidelines program in improvement of quality was attributed in part to the local development process. The guidelines were made widely available to the 80 physicians at medical meetings, and specific reminders were mailed to each physician. Finally, the guidelines were readily available throughout the center. Of note, no specific penalty or reward system except peer pressure was used. France, 1996: Improving care With guidelines in a French regional cancer program. At the 1998 ASCO meeting, this process was reported to improve compliance rates in a French Cancer Network.39 The Centre Leon Bernard and 28 hospitals established a regional cancer network. The guidelines were reviewed by all the practicing physicians, then used in a continuing education program of 12 meetings, and specific reminders were mailed to all physicians in 1995. The compliance rate for breast cancer improved from 12% in 1994 to 37% in 1996; for colon cancer, the compliance rate improved from 14% to 45% (P < .001). The range of improvements was similar to those reported in the original study, with improvements in diagnosis, surgery, chemotherapy, and radiation (data not shown). Again, these modest changes in practice were accomplished without formal penalty or reward other than education, awareness, and peer pressure. The likelihood of success of the program was high because of internal development, specific education intervention, patient-specific reminders at the time of encounter, and practice monitoring. It seems to have accomplished a major improvement in the process of care, and information on patient outcomes will follow.
Where the Data Are Absent
There have been a number of successful clinical practice guideline/critical pathway efforts in the United States, Canada, and France. Improvements have been demonstrated in compliance with evidence-based guidelines or evidence-based medicine, short-term LOS, complication rates, and financial outcomes. There has been one example of documented improvement in province-wide DFS and OS in breast cancer patients. The published data suggest that a high level of patient satisfaction can be maintained even as patients are discharged sooner. A specific process can be recommended for CPGs on the basis of experience to date with successful guidelines. Development of the guidelines should include all local users, with opportunity to provide critique. The guidelines should be disseminated through specific mailing and educational conferences, not solely by publication in a journal. Implementation of the guidelines should be accomplished through patient-specific reminders; ie, the guideline must be readily available for reference and the physician should be prompted to use it. Accountability for process and outcomes is a key component and probably the most important aspect of guideline program success. Such accountability can take the form of peer pressure, financial incentives, or administrative reward or sanction. Conversely, programs that have not succeeded have relied on voluntary change in practice behavior or have had no accountability component. The simple provision of information available through meetings, mailings, or publication has not lead to significant or timely change in prescribing patterns or clinical practice. Areas in which research is needed include at least the following: (1) health care systems without an explicit goal of improvement; (2) limits on care that may be judged to be of benefit by the patient but not by the guideline committee; and (3) guideline attempts beyond the most common types of first-line care. In conclusion, some attempts to improve practice have been moderately successful, and other methods still in current use have been demonstrated to have little effect. Costs and LOS can be reduced, practice variations minimized, and patient quality of care and satisfaction maintained. Unsuccessful programs have relied on national dissemination, without a patient-specific implementation plan, and not accountability. Further research is needed on the performance of guidelines outside of organizations demonstrably committed to improvement, ways to improve compliance of health care providers who are not committed to change, methods to improve accountability, and methods to improve incentive programs.
Supported in part by a Faculty Scholar Award, Project on Death in America, Open Society, New York (T.J.S.); grant no. RFP CO 94388-63 from the Office of Cancer Communications, National Cancer Institute (T.J.S.); a Faculty Research Award from the American Cancer Society (B.E.H.); a grant from the National Cancer Policy Board of the Institute of Medicine/National Academy of Sciences; and a grant from the Virginia Commonwealth University Department of Internal Medicine Center for Outcomes Research.
1. Field MJ, Lohr KN: Clinical Practice Guidelines. Washington DC: Institute of Medicine, 1990 2. Hewitt M, Simons P: Ensuring quality cancer care. Washington DC: Institute of Medicine, 1999
3.
Hillner BE, Smith TJ, Desch CE: Hospital and physician volume or specialization and outcomes in cancer treatment: Importance in quality of cancer care. J Clin Oncol 18: 2327-2340, 2000
4.
Grol R: Beliefs and evidence in changing clinical practice. BMJ 315: 418-421, 1997 5. Grimshaw JM, Russell TI: Effect of clinical guidelines on medical practice. Lancet 342: 1317-1322, 1993[Medline] 6. Hayward RS, Laupacis A: Initiating, conducting and maintaining guidelines development programs. CMAJ 148: 507-512, 1993[Medline]
7.
Woolf SH: Practice guidelines: A new reality in medicine—III: Impact of patient care. Arch Intern Med 153: 2646-2655, 1993
8.
Lee S, Earle C, Weeks J: Outcomes research in oncology: History, conceptual framework, and trends in the literature. J Natl Cancer Inst 92: 195-204, 2000 9. Winn RJ, Smith TJ: The American Society of Clinical Oncology guidelines program. Pharm Pract Manage Q 16: 31-38, 1996 10. Winn RJ: The role of oncology clinical practice guidelines in the managed care era. Oncology (Huntingt) 9: 177-183, 1995 (suppl)[Medline]
11.
Cook DJ, Greengold NL, Ellrodt AG, et al: The relation between systematic reviews and practice guidelines. Ann Intern Med 127: 210-216, 1997 12. Grilli R, Lomas J: Evaluating the message: The relationship between compliance rate and the subject of a practice guideline. Med Care 32: 202-213, 1994[Medline] 13. Katterhagen G: Physician compliance with outcome-based guidelines and clinical pathways in oncology. Oncology (Huntingt) 10: 113-121, 1996 (suppl)[Medline] 14. Schoenbaum SC, Sundwall DN, Bergman D: Using Clinical Practice Guidelines to Evaluate Quality of Care: Issues (vol 1). Rockville, MD: Agency for Health Care Policy and Research Publication 95: 0045 1995 15. Schoenbaum SC, Sundwall DN, Bergman D: Using Clinical Practice Guidelines to Evaluate Quality of Care: Methods ( (vol 2). Rockville, MD: Agency for Health Care Policy and Research Publication 95: 0046 1995 16. Morris M, Levenback C, Burke TW, et al: An outcomes management program in gynecologic oncology. Obstet Gynecol 89: 485-492, 1997[Medline] 17. Schoemaker D, Toouli J, Black R, et al: Yearly colonoscopy, CT liver and chest radiography do not influence 5-year survival of colorectal cancer patients. Gastroenterology 114: 7-14, 1998[Medline] 18. Engstrom PF, Benson AB, Cohen A, et al: NCCN colorectal cancer practice guidelines: The National Comprehensive Cancer Network. Oncology (Huntingt) 10: 140-175, 1996 (suppl)[Medline] 19. Smith TJ, Bear HD: Standard follow-up of colorectal cancer patients: Finally, we can make practice guidelines based on evidence. Gastroenterology 114: 211-213, 1998[Medline] 20. Winn RJ, Brown NH, Botnick WZ: Reproducibility of guidelines: A comparison of the NCCN and ASCO lung cancer guidelines. Oncology (Huntingt) 13: 35-39, 1999 21. American Society of Clinical Oncology: Clinical practice guidelines for the treatment of unresectable non–small-cell lung cancer. J Clin Oncol 15: 2996-3018, 1997[Abstract] 22. Evans WK, Newman T, Graham I, et al: Lung cancer practice guidelines: Lessons learned and issues addressed by the Ontario Lung Cancer Disease Site Group. J Clin Oncol 15: 3049-3059, 1997[Abstract]
23.
Smith TJ, Hillner BE, Mitchell RB: Decision analysis in non–small-cell lung cancer: Not back to the drawing modeling board—Back to the bedside. J Clin Oncol 15: 870-872, 1997 24. Edge SB: Breast cancer practice guidelines: Evaluation and quality improvement. Oncology (Huntingt) 11: 151-154, 1997[Medline] 25. Morrow M, Bland KI, Foster R: Breast cancer surgical practice guidelines: Scope and format of guidelines. Oncology (Huntingt) 11: 877-886, 1997[Medline] 26. Carlson RW, Goldstein LJ, Gradishar WJ, et al: NCCN breast cancer practice guidelines: The National Comprehensive Cancer Network. Oncology (Huntingt) 10: 47-75, 1996 (suppl)[Medline]
27.
Hamel MB, Phillips RS, Davis BR, et al: Outcomes and cost-effectiveness of initiating dialysis and continuing aggressive care in seriously ill hospitalized adults: SUPPORT Investigators—Study to understand prognoses and preferences for outcomes and risks of treatments. Ann Intern Med 127: 195-202, 1997 28. Ford LG, Hunter CP, Diehr P, et al: Effects of patients management guidelines on physician practice patterns: The Community Hospital Oncology Program experience. J Clin Oncol 5: 504-511, 1987[Abstract] 29. Patton MD, Katterhagen JG: Critical pathways in oncology: Aligning resource expenditures with clinical outcomes. J Oncol Manage 6: 16-21, 1994 30. Litwin MS, Smith RB, Thind A, et al: Cost-efficient radical prostatectomy with a clinical care path. J Urol 155: 989-993, 1996[Medline] 31. American Society of Clinical Oncology: 1997 update of recommendations for the use of hematopoietic colony-stimulating factors: Evidence-based, clinical practice guidelines. J Clin Oncol 15: 3288, 1997[Medline] 32. Rischer JB, Childress SB: Cancer pain management: Pilot implementation of the AHCPR guideline in Utah. Jt Comm J Qual Improv 22: 683-700, 1996[Medline]
33.
Du Pen SL, Du Pen AR, Polissar N, et al: Implementing guidelines for cancer pain management: Results of a randomized controlled clinical trial. J Clin Oncol 17: 361-370, 1999 34. Thompson I, Optenberg S, Segura J, et al: AUA prostate cancer clinical guidelines have a significant impact upon patient care in a national study population. J Urol 59: 257a, 1995 (abstr)
35.
Olivotto A, Coldman AJ, Hislop TG, et al: Compliance with practice guidelines for node-negative breast cancer. J Clin Oncol 15: 216-222, 1997 36. Levine M, Browman G, Newman T, et al: The Ontario Cancer Treatment Practice Guidelines Initiative. Oncology (Huntingt) 10: 19-22, 1996 (suppl)[Medline] 37. Grilli R, Apolone G, Marsoni S, et al: The impact of patient management guidelines on the care of breast, colorectal, and ovarian cancer patients in Italy. Med Care 29: 50-63, 1991[Medline]
38.
Ray-Coquard I, Phillip T, Lehmann M, et al: Impact of a clinical guidelines program for breast and colon cancer in a French cancer center. JAMA 278: 1591-1595, 1997 39. Ray-Coquard I, Philip T, De Laroche G: Impact of a clinical guidelines program on medical practice in a French cancer network. Proc Am Soc Clin Oncol 17: 421a, 1998 (abstr 1624) 40. Litwin MS, Shpall AI, Dorey F: Patient satisfaction with short stays for radical prostatectomy. Urology 49: 898-906, 1997[Medline] 41. Somerfield M, Bennett CL, Feinglass J, et al: Decrease in misuse/overuse of hematopoietic colony stimulating factors (CSFs) for solid tumors and lymphomas: Resulting from ASCO surveys related to the 1994 and 1996 ASCO CSF Guidelines. Proc Am Soc Clin Oncol 17: 421a, 1998 (abstr 1625)
42.
Hartmann LC, Tschetter LK, Habermann TM, et al: Granulocyte colony-stimulating factor in severe chemotherapy-induced afebrile neutropenia. N Engl J Med 336: 1776-1780, 1997
43.
American Society of Clinical Oncology: Recommendations for the use of hematopoietic colony-stimulating factors: Evidence-based, clinical practice guidelines. J Clin Oncol 12: 2471-2508, 1994
44.
American Society of Clinical Oncology: Outcomes of cancer treatment for technology assessment and cancer treatment guidelines. J Clin Oncol 14: 671-679, 1996
45.
Bennett CL, Bishop MR, Tallman MS, et al: The association between physician reimbursement in the US and use of hematopoietic colony stimulating factors as adjunct therapy for older patients with acute myeloid leukemia: Results from the 1997 American Society of Clinical Oncology survey—Health Services Research Committee of the American Society of Clinical Oncology. Ann Oncol 10: 1355-1359, 1999
46.
Von Roenn JH, Cleeland CS, Gonin R, et al: Physician attitudes and practice in cancer pain management: A survey from the Eastern Cooperative Oncology Group. Ann Intern Med 119: 121-126, 1993
47.
Cleeland CS, Gonin R, Hatfield AK, et al: Pain and its treatment in outpatients with metastatic cancer. N Engl J Med 330: 592-596, 1994
48.
Cleeland CS, Gonin R, Baez L, et al: Pain and treatment of pain in minority patients with cancer. Ann Intern Med 127: 813-816, 1997 49. Zenz M, Zenz T, Tryba M, et al: Severe undertreatment of cancer pain: A 3-year survey of the German situation. J Pain Symptom Manage 10: 187-191, 1995[Medline]
50.
SUPPORT Principal Investigators: A controlled trial to improve care for seriously ill hospitalized patients: The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT). JAMA 274: 1591-1598, 1995
51.
Olivotto IA, Bajdik CD, Plenderleith IH, et al: Adjuvant systemic therapy and survival after breast cancer. N Engl J Med 330: 805-810, 1994 Submitted July 16, 1999; accepted February 22, 2001.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
.05.) The timing correlated with implementation of the BCCA guidelines, but causality cannot be demonstrated. For example, the influence of strong expert opinion in the centralized cancer treatment system, education interventions and conferences on the evidence rather than the guidelines, or increased accountability and pressure to conform all could be partly responsible. In addition, it is not possible to ascertain whether patients of cancer centers and community oncologists experienced the same survival. 
