Fax: (919) 668-7345
Quality measures for the use of adjuvant chemotherapy and radiation therapy in patients with colorectal cancer†‡§
A systematic review
Article first published online: 12 OCT 2006
Copyright © 2006 American Cancer Society
Volume 107, Issue 10, pages 2352–2360, 15 November 2006
How to Cite
Prosnitz, R. G., Patwardhan, M. B., Samsa, G. P., Mantyh, C. R., Fisher, D. A., McCrory, D. C., Cline, K. E., Gray, R. N. and Morse, M. A. (2006), Quality measures for the use of adjuvant chemotherapy and radiation therapy in patients with colorectal cancer. Cancer, 107: 2352–2360. doi: 10.1002/cncr.22278
Presented at the Annual Meeting of the American Radium Society, Maui, Hawaii, May 6–10, 2006.
This article was based on research conducted by the Duke Evidence-Based Practice Center under contract to the Agency for Health Care Research and Quality (Contract 290-02-0025), Rockville, MD, in collaboration with the National Cancer Institute, the Centers for Disease Control and Prevention, and the Centers for Medicare and Medicaid Services.
The authors of this article are responsible for its contents. No statement in this article should be construed as an official position of the Agency for Healthcare Research and Quality, the National Cancer Institute, the Centers for Disease Control and Prevention, the Centers for Medicare and Medicaid Services, or the U.S. Department of Health and Human Services.
- Issue published online: 8 NOV 2006
- Article first published online: 12 OCT 2006
- Manuscript Accepted: 25 AUG 2006
- Manuscript Revised: 24 AUG 2006
- Manuscript Received: 29 JUN 2006
- Veterans Affairs Health Services Research and Development Research Career Award
- health care quality;
- colorectal cancer;
- adjuvant therapy;
- radiation therapy
Chemotherapy (CT) and radiation therapy (RT) are essential components of adjuvant (preoperative or postoperative) therapy for many patients with colorectal cancer (CRC); however, quality measures (QMs) of these critical aspects of CRC treatment have not been characterized well. Therefore, the authors conducted a systematic review of the literature to determine the available QMs for adjuvant CT and RT in patients with CRC and rated their usefulness for assessing the delivery of quality care.
The MEDLINE and Cochrane data bases were searched for all publications that contained potential/actual QMs pertaining to adjuvant therapy for CRC. Identified QMs were rated by using criteria developed by the National Quality Forum.
Thirty-two articles met the established inclusion/exclusion criteria. Those 32 articles contained 12 potential or actual QMs, 6 of which had major flaws that limited their applicability. The most useful QMs identified were 1) the percentage of patients with AJCC Stage III colon cancer who received postoperative CT and 2) the percentage of patients with Stage II or III rectal cancer who received chemoradiotherapy.
To the authors' knowledge, very few QMs pertaining to adjuvant CT or RT for CRC have been published to date, and only half of those measures were rated as useful, acceptable, and valid in the current literature review. Future research should focus on refining existing QMs and on developing new QMs that target important leverage points with respect to the provision of adjuvant therapy for patients with CRC. Cancer 2006. © 2006 American Cancer Society.
Chemotherapy (CT) and radiotherapy (RT) play important roles in the treatment of patients with colorectal cancer (CRC). In patients with AJCC Stage II or III rectal cancer, randomized studies have indicated that concurrent, postoperative chemoradiotherapy (CRT) reduces the risk of locoregional recurrence and improves overall survival.1, 2 In patients with Stage III colon cancer, it has been demonstrated repeatedly that postoperative CT also improves survival.3 In 1990, the National Institutes of Health (NIH) issued a Consensus Statement declaring that all patients with Stage III colon cancer should receive postoperative CT and that all patients with Stage II or III rectal cancer should receive postoperative CRT.4 After additional studies demonstrated a reduced risk of local recurrence with preoperative CRT compared with adjuvant CRT in patients with rectal cancer, this guideline was refined by the North Central Cancer Network (NCCN) to recommend preoperative treatment.
Despite the obvious importance of adjuvant CT and RT given preoperatively or postoperatively in the management of patients with CRC, little is known regarding the quality of adjuvant CT and RT for CRC or the optimal means of measuring that quality. To explore this issue, we conducted a systematic review of the literature to identify the available quality measures (QMs) for adjuvant CT and RT for CRC and then rated them in terms of importance/usability, scientific acceptability, and validity.
MATERIALS AND METHODS
A structured literature review was conducted to identify potential or actual QMs pertaining to the adjuvant treatment of CRC. A computerized search of the MEDLINE® data base (from 1966 to January 2005), the Cochrane Database of Systematic Reviews, and the National Guideline Clearinghouse was performed. A study was included for review if the study population consisted of adults with CRC and if it did 1 of the following: contained an explicit QM, contained a measure that assessed the adequacy or completeness of documentation, tested the validity or reliability of a QM, or described a data source that was used to construct a QM. Studies were excluded if they used data from outside the U.S., if they were published before 1990, or if they were not published in English. The internal validity of each QM initially was assessed by each reviewer during the data abstraction stage using criteria developed by the National Quality Forum. These criteria consisted of “importance,” “usability,” and “scientific acceptability.” (Although feasibility was important, it was considered beyond the scope of this project and was not assessed.) A measure was considered “important” if 1 or more audience would find that the information produced from a measure was useful for some purpose. A measure was considered “usable” if the intended audience could understand the results of the measure and find them useful for decision making. A measure was considered “scientifically acceptable” if it produced consistent and credible results when implemented. Five subcriteria were used to rate scientific acceptability: precise specification, reliability, validity, adaptability, and adequacy of risk adjustment. For all measures, we scored each criterion on a scale from 1 to 5 scale, with 1 denoting “poor” and 5 denoting “ideal.” From the complete set of ratings criteria, we devised a summary rating scale for each QM using the criteria of importance and usability (I), scientific acceptability (S), and extent of testing (T). The I-S-T rating was determined by examining all studies that used the particular measure to identify the highest rated example. The 3 criteria were rated subjectively on a scale from 1 to 5 by 2 reviewers with 1 denoting “poor” and 5 denoting “ideal.” However, because several QMs were derived from secondary sources that contained insufficient information for a comprehensive I-S-T rating, we permitted reviewers to use their expert judgment to rate such measures. The interrater agreement for these subjective ratings was calculated. A detailed description of our methodology has been published elsewhere.5
In total, 3429 abstracts were reviewed, and 843 of those articles were selected for full text evaluation. Ultimately, 73 articles met the established inclusion/exclusion criteria, and 32 of those articles pertained to adjuvant CT or RT. The interrater reliability for inclusion/exclusion was from 0.23 to 0.46.
Our literature search yielded 10 general-process QMs pertaining to adjuvant CT, RT, or concurrent CRT; 6 were considered unacceptable because of substantial flaws in their design or application, and 4 QMs were deemed acceptable. These measures are listed in Table 1. The unacceptable measures were not rated on the I-S-T scale. Both the acceptable and the unacceptable QMs are described qualitatively below.
|Quality measure||Reference(s)||Quality measure level assignment*|
|Is the measure important and usable? (I)||Is it scientifically acceptable? (S)||Is it well tested? (T)|
|Percentage of patients with Stage III colon cancer who received postoperative CT||6–17||5||5||5|
|Percentage of patients with Stage III colon cancer who were offered postoperative CT||8,23||5||4||2|
|Percentage of patients with Stage III colon cancer who were referred to a medical oncologist for consideration of postoperative CT||24||5||3||1|
|Percentage of patients with Stage II or III rectal cancer who received concurrent CRT||15,16,18,20||5||4||1|
|Percentage of patients with Stage II and III rectal cancer who received postoperative CT or RT||25–27||—||—||—|
|Percentage of patients with colon cancer, stage not specified, who received postoperative CT||28–31||—||—||—|
|Percentage of patients with colon or rectal cancer receiving postoperative CT||27,32,36||—||—||—|
|Percentage of patients with Stage II or III rectal cancer who were referred for consideration of postoperative CT||24||—||—||—|
|Percentage of patients who received postoperative RT who also received postoperative CT for cancer of the sigmoid colon or rectum||29||—||—||—|
|The percentage of patients with late-stage rectal cancer (Stage ≥III) who received ≥1 course(s) of postoperative CT within 1 y of initial cancer surgery||34||—||—||—|
Percentage of patients with Stage III colon cancer who received postoperative CT
Using a variety of data sources and methods of analysis, several different investigators have attempted to measure the percentage of patients with Stage III colon cancer receiving postoperative CT (personal communication).6–16 The strengths and weaknesses of each of these efforts to measure this important process of care are discussed below.
Using data from the California Cancer Registry and information obtained from physician surveys and office records, Ayanian et al.17 measured the percentage of patients diagnosed with Stage III colon cancer from 1996 to 1997 who received postoperative CT. That study did not validate this QM but did validate the data sources. The main limitation of the analysis was the use of a single state registry, which may not reflect national patterns of care.
Schrag et al.9, 10, 18 measured the percentage of patients diagnosed with Stage III colon cancer from 1991 to 1996 who received postoperative CT within 3 months of surgery based on data from the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked registry. There was no validation of the receipt of CT beyond the usual policies of the SEER registry. The strengths and weaknesses of the study stem from the use of the linked SEER-Medicare data base. Although this data set contains a large number of patients from 13 different geographic areas of the U.S., it does not include patients from the entire U.S., and it is limited to patients age 65 years and older who are insured by Medicare.
Sundararajan et al.13 and Keating et al.19 also measured the percentage of patients with Stage III colon cancer who received CT based on data from the SEER-Medicare linked registry. Sundararajan et al. measured the percentage of patients who received 5-fluorouracil within 4 months of surgery from 1992 to 1996, whereas Keating et al. measured the percentage of patients who received any type of CT from 1993 to 1999 but did not specify the interval from surgery to the initiation of CT. Keating et al. examined county-specific utilization rates as a function of market share of fee-for-service (FFS) plans versus Health Maintenance Organizations (HMOs). The limitations of these studies are those inherent to the use of the SEER data set.
Tropman et al.20 used the tumor registries of North and South Carolina to measure the percentage of patients with Stage III colon cancer who received CT according to the consensus-based guidelines that were in effect at the time of diagnosis. Tumor registry data were supplemented by chart review. A strength of this study is the use of chart review, which allowed validation of the tumor registry data. The major limitation of this study is the geographic restriction to 2 states.
The Institute of Medicine (IOM) (“Assessing the Quality of Cancer Care”6) and the Colon Cancer Workgroup (personal communication) proposed a similar QM. However, instead of measuring the percentage of patients with Stage III colon cancer who received postoperative CT, those investigators proposed measuring the percentage of patients with “lymph node-positive colon cancer” who receive “a full course of postoperative CT after surgery.” This QM conforms to the 2005 guidelines of the NCCN.21 The main shortcoming of this QM is that it may not be feasible; in general, using current administrative data bases, it cannot be verified that a patient received a full course of CT, although the length of time from start to completion of CT often is used a surrogate for this endpoint.
The National Cancer Database (NCDB) contains information regarding the patterns of care for patients with colon cancer and rectal cancer based on disease stage. Several publications have been based on this data base. Steel16 used data from surveys that were conducted by the NCDB during the years 1985, 1988, and 1990 to report the percentages of patients with Stage 0 through Stage IV CRC who received surgery only, surgery and CT, surgery/radiation/CT, other, or supportive care. Additional publications by Steele et al. detailed the patterns of care for patients with CRC who were treated in 199116 and 1992.12 Jessup et al.7 presented a similar analysis of this data base for the years 1986, 1987, 1988, 1990, 1991, 1992, and 1993. The main limitation of the NCDB is that participation in it is voluntary, a feature that introduces potential bias (i.e., hospitals that have poor rates of compliance with the guidelines may opt not to participate).
Percentage of patients with Stage III colon cancer who are offered postoperative CT
The main limitation of most measures that assess rates of delivery of postoperative therapy is that they fail to take into account legitimate reasons why patients may not receive postoperative therapy, such as comorbid conditions or other contraindications and patient preferences. To avoid this pitfall, some investigators have chosen to measure the rate at which patients are offered CT rather than the rate of CT delivery.8, 22 Potosky et al.22 took this approach by using SEER data. Because the SEER data base is based only on hospital records, Potosky et al. also contacted treating physician to verify that postoperative CT was offered. The study did not explicitly validate this QM.
One limitation of administrative data bases is that they seldom contain the information needed to ascertain why a therapy was omitted. To overcome this limitation, Mahoney et al.8 used the New Jersey Tumor Registry to identify a group of patients diagnosed with Stage III colon cancer from 1989 to 1996 who did not receive postoperative CT and then reviewed their medical records to determine the reasons why. The downside of this approach is that medical records reviews are costly and time consuming.
Percentage of patients with Stage III colon cancer who were referred to a medical oncologist for consideration of adjuvant CT
Using the administrative data base of a Massachusetts HMO, Oliveria et al.23 measured the percentage of patients with colon cancer who were referred to a medical oncologist for consideration of postoperative CT on a stage-by-stage basis. Referral to a medical oncologist presumably is a prerequisite for the receipt of CT in accordance with accepted guidelines and, thus, may be thought of as a surrogate endpoint for this QM. The study did not specifically validate the QM, but reliability of the data sources was ascertained by reviewing medical records to confirm data in the computerized data base. The main drawback of this QM is that referral to a medical oncologist does not guarantee that postoperative CT will be recommended and administered according to accepted guidelines.
Percentage of patients with Stage II or III rectal cancer who received concurrent postoperative CRT
Using data from 3 regional registries of the California Cancer Registry, Ayanian et al.17 determined the percentage of patients age 18 years and older who were diagnosed with Stage III adenocarcinoma of the colon and survived for at least 30 days after surgery who received postoperative CRT. Information regarding postoperative treatment was obtained from physician surveys and office records. The study did not explicitly validate this QM, but it did validate the data sources.
Using similar methods, several authors have employed the linked SEER-Medicare data base to determine the percentage of patients with Stage II or III rectal cancer who received concurrent postoperative CRT. Keating et al.19 determined county-specific utilization rates and compared them according to the market share of FFS health plans versus HMOs in those counties. Schrag et al.15 looked at 3 different numerators: 1) the number of patients who received postoperative CT (with 5-fluorouracil) and RT within 4 months of surgery, 2) the number of patients who received RT alone, and 3) the number of patients who received postoperative CT only. Their results also were presented separately by stage. Finally, Neugut et al.14 measured the percentage of patients who received postoperative CT, RT, or both. The impact of age, gender, and race on the rate of postoperative therapy was evaluated. Their results also were presented separately by stage. The strengths and weaknesses of the SEER-Medicare data base discussed above apply to all of 3 of these studies.
Percentage of patients with Stage II and III rectal cancer who received postoperative CT or RT
Because the NIH and NCCN guidelines specify that concurrent CRT should be administered to patients with Stage II or III rectal cancer, measures that report the rates of adjuvant CT (without considering RT) or adjuvant RT (without considering CT) are of limited utility.24–26
Percentage of patients with colon cancer, stage not specified, who received postoperative CT
A few QMs described rates of postoperative CT for CRC but did not specify the stage of the patients' disease.27–30 Because the guidelines are stage-specific, these QMs are of limited utility.
Percentage of patients with colon or rectal cancer who received postoperative CT
Percentage of patients with Stage II or III rectal cancer who were referred for consideration of postoperative CT
Some investigators have examined the rate of referral to medical oncologists without attempting to determine the rate of referral to radiation oncologists.23 Because CRT is the standard of care for patients with Stage II and III rectal cancer, these QMs are inadequate.
Percentage of patients who received postoperative RT who also received postoperative CT for cancer of the sigmoid colon or rectum
Because this QM employs an incorrect denominator, its usefulness is limited. The correct denominator should be based on the cohort of patients diagnosed with Stage II or III rectal cancer, not just those patients who received postoperative RT.28
Percentage of patients with late-stage rectal cancer (Stage ≥III) who received ≥1 course of postoperative CT within 1 year of initial cancer surgery
QMs pertaining to adjuvant CT are not particularly useful if the denominator includes patients with both Stage III and IV disease, because there are separate guidelines for these 2 groups of patients (personal communication).
Our literature search yielded 2 potential technical-process QMs pertaining to adjuvant CT, RT, or concurrent CRT (Table 2). Both of these measures pertain to RT, and both were deemed acceptable. No technical-process measures for CT or CRT were identified. These 2 measures and their I-S-T ratings are shown in Table 2 and also are described qualitatively below.
|Quality measure||Quality measure level assignment*|
|Is the measure important and usable (I)?||Is it scientifically acceptable (S)?||Is it well tested (T)?|
|Adherence to RT management treatment guidelines for patients with adenocarcinoma of the rectum or sigmoid colon (Kline et al., 199737)||5||4||2|
|Rate of use of modern RT techniques and adherence to recommendations of the NCI-sponsored randomized control trials in rectal cancer patients (Minsky et al., 199838)||5||4||2|
Adherence to RT management treatment guidelines for patients with adenocarcinoma of the rectum or sigmoid colon
Kline et al.35 used data from the Patterns of Care Study (PCS) (1989–1990) to evaluate adherence to consensus guidelines developed by the PCS Treatment Planning Committee pertaining to the delivery of RT for rectal cancer. Academic centers, hospital-based practices, and free-standing facilities were evaluated. Multiple quality indicators were assessed. Examples include 1) the use of custom blocks to shape the fields, 2) the use of port films to verify accurate positioning of the patient during treatment, and 3) the use of small bowel contrast to identify the position of the small bowel and shield it as needed. The use of a random sampling of health care facilities in the U.S. suggests that the results are a valid representation of RT practice in the U.S. Although some refinements have been made in the techniques used to plan RT for rectal cancer (e.g., computed tomography-based planning vs. fluoroscopic simulation), the basic treatment techniques (3-field or 4-field) essentially have remained the same.
Rate of use of modern RT techniques and adherence to recommendations of NCI-sponsored, randomized, controlled trials in patients with rectal cancer patients
Minsky et al.36 used data from the PCS in Rectal Cancer (1992–1994)37 to evaluate the rate of use of modern RT techniques and adherence to recommendations of the NCI-sponsored, randomized controlled trials in rectal cancer. “Modern” RT techniques were defined by a panel of experts (the PCS Rectal Cancer Committee). Many factors that were considered indicative of high-quality RT were measured in the survey. Examples include treatment of all fields each day, an attempt to exclude the small bowel in patients who are treated postoperatively, and inclusion of the scar in all treatment fields in patients who undergo abdominoperineal resection. The data source was not validated. Similar to the report by Kline et al., the use of a random sampling of health care facilities in the U.S. suggests that the results are a valid representation of RT practice in the U.S.
The IOM has defined “quality” as it pertains to health care as the extent to which health services for individuals and populations increase the possibility of desired health outcomes and are consistent with current professional knowledge.38 For CRC, it is agreed generally that CT for Stage III colon cancer and CRT for Stage II and III rectal cancer constitute quality care, because these interventions improve health outcomes. Nonetheless, there are few important, scientifically acceptable, well tested measures of the quality of adjuvant therapy for patients with CRC. In the current review, we identified only 6 acceptable process measures, 4 of which were general and 2 of which were technical. Despite years of academic research, few formal quality measurement projects have been conducted. The QMs we identified nearly all were potential QMs derived from academic publications that described patterns of care and compared these actual patterns with established guidelines. Furthermore, all of the QMs we identified pertained to postoperative CT or CRT, although preoperative therapy is gaining acceptance as the standard of care.
The 4 acceptable general-process measures identified by our review have both strengths and weaknesses. QM1 (percentage of patients with Stage III colon cancer receiving postoperative CT: I5S5T5) is accepted widely as an important indicator of quality care for patients with colon cancer but has significant limitations. Most significant among these is that the rate of adjuvant CT may not be the ideal measure of quality: Some patients are either unable to tolerate CT or refuse it.
One way to improve QM1 would be to define the denominator more precisely to include patients with Stage III colon cancer who undergo a complete surgical resection and have no contraindications to CT and no severe comorbidities. Comorbid conditions should be included in administrative data bases. Alternatively, Karnofsky performance status could be coded in the data base.
QM2 (percentage of patients with Stage III colon cancer who are offered postoperative CT: I5S5T5) improves on QM1 by including in the numerator patients who are offered CT but ultimately refuse to undergo treatment because of their personal preferences. In the study by Ayanian et al., 30% of patients did not receive CT because they refused it.17 However, like QM1, this measure fails to take into account patients who have comorbid conditions (or other contraindications) that preclude the use of CT. Mahoney et al. observed that a significant proportion of patients who did not receive CT were considered “too old” or had significant comorbid illnesses.8 Although this measure may seem to be a more accurate assessment of adherence to guidelines that call for the administration of adjuvant CT, no study has elucidated clearly the reasons for patient refusal of CT. A patient who understands the risks and benefits of CT and makes an informed choice to forego treatment simply is exercising their individual autonomy. However, a patient who refuses CT because they do not fully understand the risks and benefits of treatment simply may not have been educated properly by his or her health care providers.
QM3 (percentage of patients with Stage III colon cancer who are referred to an oncologist for consideration of postoperative CT: I5S3T1) pertains to an important leverage point in practice. Because CT typically is prescribed by medical oncologists, it is logical to assume that referral to a medical oncologist is a necessary prerequisite to the receipt of CT by patients with Stage III colon cancer. Although referral to a medical oncologist probably is necessary for the provision of high-quality care, it does not guarantee that high-quality care will be delivered. There is no guarantee that the patient will accept guideline-based recommendations for adjuvant therapy, nor is there any guarantee that the medical oncologist will follow the established guidelines. Thus, the relation between QM3 and outcomes, although it is plausible, is less direct than the relation for QM1 and QM2. Further research is needed to determine how this measure correlates with a measure of whether patients were offered CT.
QM4 (the percentage of patients with Stage II and III rectal cancer who received both CT and RT: I5S4T5) has limitations that are similar to those of QM1; namely, it does not account for patients who fail to receive adjuvant treatment because of comorbid conditions or contraindications or because they refuse it. Depending on the situation, it may be reasonable to withhold CT, RT, or both from a patient who has Stage II or III rectal cancer.
Although it is important to know whether a patient received CRT, it is equally important to assess the technical quality of its administration. Obtaining information on the details of CT and RT administration, however, presents a significantly greater challenge than determining simply whether CT or RT was used at all. The 2 potential QMs for the technical administration of RT identified in the current report demonstrate that this type of measure is feasible, but they also illustrate some of the hazards of attempting to measure the technical quality of rapidly evolving treatments.
The most significant challenge in measuring the quality of RT may be obtaining data in a timely fashion. If excessive delay occurs between the collection of data needed to measure the quality of RT and the analysis/reporting of these data, then the resulting QMs will not be applicable to modern practice. Because of this time lag, neither of the 2 technical-process QMs that we identified is ready for immediate use. However, with modern communications technology, it is possible to rapidly ascertain, analyze, and report the data needed for quality measurement.
The principle of “real-time” RT quality assurance has been demonstrated by several ongoing trials sponsored by the Radiation Therapy Oncology Group (RTOG). The RTOG has established mechanisms for 1) credentialing of facilities that participate in RTOG trials and 2) rapid review and approval of RT plans for specific patients enrolled on these trials. The costs of reviewing every CRC patient's RT treatment plan likely would be prohibitively great. However, facilities may be encouraged or required to undergo random sampling of cases as part of a quality-assurance mechanism.
There are a number of avenues for improving the technical measures of the quality of RT for patients with rectal cancer. Because technology within RT is evolving rapidly, the technical measures need to be updated by a panel of radiation oncologists with expertise in treating patients with rectal cancer. The main shortcoming of the technical measures reviewed in this report, however, is that none have been linked to outcomes such as reduced locoregional recurrence rates, improved disease-free or overall survival, or reduced complication rates. An implicit assumption has been made that the technical quality of RT is linked to at least some of these outcomes, but to our knowledge there are very little if any data demonstrating that such a link exists. Establishing links between specific technical measures of quality RT and these outcomes should be a priority of future research on this type of QM.
Technical-process QMs for CT administration were conspicuously absent from the literature. The development of such measures is an important avenue of future investigation. High-quality CT requires that patients receive the correct agents, at the right dose, on the proper schedule. The toxicities of CT can be substantial, and mortality is reported. Measures of the quality of CT administration should document efforts to avoid toxicity, the ultimate rate of toxicity, and an assessment of whether the toxic effects of CT were managed appropriately. The facility in which CT is administered also is an important determinant of quality. Such issues as ensuring sterility, correct dose and concentration, correct diluent, correct vessel and tubing for holding the CT, and care in how the drug actually is administered to the patient are crucial not only for patients with CRC but also for CT in general.
After we completed our study, Malin et al.39 reported on their analysis of the percentage of Stage I to III breast cancer survivors and Stage II to III CRC survivors in 5 metropolitan statistical areas across the U.S. who received recommended care that was specified by a comprehensive set of explicit QMs. Data for their analysis came from patient surveys and review of medical records. To develop their QMs, a National Institute for Cancer Care Quality project team reviewed existing quality indicators, guidelines, and review articles along with randomized trials and other controlled studies of breast cancer and CRC treatment that were identified through MEDLINE searches. Some of the QMs reported in their study were similar to the QMs that we identified in the literature, as described above. Examples include 1) the percentage of patients with Stage III colon cancer (and high-risk Stage II colon cancer) receiving CT and 2) the percentage of patients with Stage II and III rectal cancer receiving CRT. However, Malin et al. also reported some QMs that were not addressed previously by other authors, including QMs that addressed the timing of CT, the documentation of the CT plan, and the administration of doses of CT consistent with those reported in the literature.
In summary, there are few measures of the quality of adjuvant CT or RT for patients with CRC, and the existing measures all have significant limitations. The greatest impediment to the widespread implementation of most general-process measures may be that current data bases are inadequate. Even the largest administrative data base used in the QMs we identified, the linked SEER-Medicare data base, contains data on only a small fraction of all patients with CRC who are diagnosed and treated in the U.S. Furthermore, this large data base provides information only on patients age 65 years and older who are insured by Medicare. Improving the quality of care for patients with CRC will require the development of a national data base that encompasses the entire U.S. and includes at least a representative sample of all patients with CRC, regardless of age or insurance status.
The development and implementation of optimal technical-process measures is an equally daunting task. One of the biggest challenges will be identifying the points in the process of care that have the greatest impact on patient outcome. Another difficult undertaking will be the timely collection of data on complex processes for which the standard of care may be evolving rapidly.
Considerable progress had been made in evaluating the quality of adjuvant therapy for patients with CRC over the last 10 years. However, the optimization of these measures will require a much greater effort and a considerable expansion of the resources dedicated to that effort.
- 6Institute of Medicine. Assessing the Quality of Cancer Care: An Approach to Measurement in Georgia. Committee on Assessing Improvements in Cancer Care in Georgia. National Cancer Policy Board. Washington, DC: The National Academies Press; 2005.
- 11Colorectal Cancer. In: SteeleGDJr., WinchesterDP, MenckHR, MurphyGP, editors. National Cancer Data Base Annual Review of Patient Care, 1994. Atlanta, GA: American Cancer Society. 1994: 24–42.
- 12Colorectal Cancer. In: SteeleGDJr, WinchesterDP, MenckHR, MurphyGP, editors. National Cancer Data Base Annual Review of Patient Care, 1995. Atlanta, GA: American Cancer Society. 1995: 66–83., .
- 21North Central Cancer Network. Clinical practice guidelines in oncology: colon cancer. Version 3, 2005. Available at: http://www.nccn.org/professionals/physician_gls/PDF/colon.pdf Accessed July 7, 2005.
- 38Institute of Medicine. Measuring the quality of health care. A statement by the National Roundtable on Health Care Quality. Washington, DC: National Academy Press; 1999..