Degree of Variability in Performance on Breast Cancer Quality Indicators: Findings From the Florida Initiative for Quality Cancer Care

The Florida Initiative for Quality Cancer Care has identified how multiple aspects of breast cancer care can be improved.

Abstract

Purpose:

The Florida Initiative for Quality Cancer Care (FIQCC) comprises 11 Florida practice sites that participate in comprehensive reviews of quality of care specific to patients with cancer. Here, we examined site adherence to performance indicators to assess quality of care for patients with breast cancer (BC).

Methods:

Quality indicators were scripted on the basis of accepted guidelines from the Quality Oncology Practice Initiative, National Comprehensive Cancer Network, American College of Surgeons, and site-specific expert panel consensus. Comprehensive chart reviews, including both medical and surgical oncology quality measures, were conducted for patients with BC first seen in 2006 by a medical oncologist at one of the sites. Statistical comparisons were made by the Pearson χ² exact test, using Monte Carlo estimation.

Results:

Charts of 622 patients were reviewed. Of the 34 indicators, seven for medical oncology and four for surgical oncology fell below the 85% level of adherence. A statistically significant difference (P < .001) in variation of performance across the sites was found for the following medical and surgical oncology indicators: documentation of menopausal status, family history, informed consent, planned chemotherapy regimen and flow sheet, American Joint Committee on Cancer staging, HER2/neu status, reporting of margin orientation and inking of the margins, histological grade, having a sentinel lymph node biopsy for invasive BC, and obtaining a mammogram within 14 months of definitive surgery.

Conclusion:

The FIQCC has identified how multiple aspects of BC care can be improved. Findings are being used at the participating institutions to guide quality improvement efforts.

Introduction

In 1999, the Institute of Medicine's National Cancer Policy Board published “Ensuring Quality Cancer Care,” a landmark report on quality of cancer care.¹ The Board recommended developing systems to measure and monitor quality of care by means of a core set of indicators. As a result, many initiatives for quality cancer care were started. One such initiative, the National Initiative for Cancer Care Quality (NICCQ),² was sponsored by ASCO in 2000. The NICCQ used expert opinion and literature to formulate and assess breast cancer (BC) quality measures for patients presenting with stage I-III disease in five major US metropolitan areas. The 36 quality measures assessed eight components of clinical care, including pathology and documentation. Performance ranged from as low as 13% for referrals to as high as 97% for testing/treatment.³

Building on the NICCQ goals of assessing and reporting, ASCO sponsored the Quality Oncology Practice Initiative (QOPI)⁴ to provide feedback to medical oncology groups across the United States. QOPI is a practice-based quality self-assessment tool that relies on chart reviews focusing on pathology and chemotherapy documentation, pain assessment, symptom management, and BC-specific indicators. QOPI is now voluntarily used by > 500 practices nationwide. A study of 71 practices participating in QOPI demonstrated that, by providing feedback about deficiencies, use of QOPI resulted in significant positive changes in aspects of care for which performance was most deficient.⁵

The Florida Initiative for Quality Cancer Care (FIQCC), a physician- and practice-based quality improvement project, was conceived to study the barriers to delivering high-quality cancer care in Florida, the state with the second highest mortality rate from cancer in the United States.⁶ The FIQCC is a consortium of 11 Florida institutions (Appendix Figure A1, online only),⁷ with three academic and eight community sites, that are actively participating in a comprehensive practice-based system of quality self-assessment across a number of cancer types. FIQCC selected quality indicators on the basis of accepted QOPI, National Comprehensive Cancer Network, American College of Surgeons, and site-specific principal investigator (PI) panel consensus indicators. In this study, we assessed adherence to performance indicators among these sites. We hypothesized that variation existed among the sites on specific indicators and that practice volume and age of the patient cohorts, factors that can affect both delivery of care and treatment, also contributed to variation in adherence.

Methods

Representatives from the 11 oncology sites selected quality measures consistent with evidence-, consensus-, and safety-based guidelines that could be abstracted from medical records. The resulting indicators were organized by diagnosis (BC, colorectal cancer, non–small-cell lung cancer) and by domains of care (eg, symptom management). In this article, we focus only on BC results.

Selection of Indicators

A panel of breast medical and surgical oncology experts from the H. Lee Moffitt Cancer Center and Research Institute (Tampa, FL) met monthly for 6 months to formulate breast indicators based on ASCO, National Comprehensive Cancer Network, NICCQ, QOPI, and American College of Surgeons guidelines. Panel consensus of the selected quality indicators from PIs, coinvestigators, and collaborative site PIs was required before data collection.

Selection of Practices

The 11 FIQCC study sites, all affiliated with Moffitt (Appendix Figure A1), agreed to provide a qualified medical record abstractor and to follow FIQCC quality standards, data collection forms, and a data collection process. Comprehensive chart reviews were conducted for patients newly diagnosed with a primary invasive BC who had been first seen in 2006 by a medical oncologist at one of the 11 FIQCC sites.

Chart Reviews

When the consortium was formed, the first disease to be examined was colorectal cancer. The number of BC cases was predetermined to match the number of colorectal cancer cases, as there were fewer colorectal cases than BC cases. Charts of patients with BC were randomly selected from among those identified at each site until the predetermined number of charts had been abstracted. This amounted to a total of 19% (622 of 3,275) of the available charts.

Abstraction and Quality Control

Before chart abstraction at each site, an abstractor trainer traveled to each institution and, using sample charts, instructed the on-site abstractor on the specific information to be acquired. Quality control of the data retrieved was maintained through two audits, which were performed by the trainer when abstraction of the assigned total number of charts for review was one third and two thirds complete at each site. We included both medical oncology and surgical oncology quality measures. This study was approved by the institutional review boards at each participating site.

Statistical Analyses

We used descriptive statistics and graphic illustrations to summarize all indicator variables. The compliance proportion, with its 95% CI, was calculated on the basis of exact binomial distribution. Statistical comparisons across sites were made by the Pearson χ² exact test, using Monte Carlo estimation. Logistic regression models were used to evaluate effects of practice site and site volume (low v high, divided by a cutoff point of 300 patient charts to allow for a balanced distribution between the two groups) and age (< 50, 50-70, > 70 years) on each of indicator variables. Volume distribution was based on the estimated total volume of patients with BC per site for 2006. Odds ratio, with its 95% CI, was estimated from the models. All analyses used two-tailed tests of significance, with the significance level set at P < .001. On the basis of the NICCQ guidelines,³ we predetermined a threshold of 85% to be the goal for the level of concordance with the individual indicators.

Results

The charts of 622 patients (99% female) with invasive BC were reviewed (Table 1). The median age was 60 years (range: 22 to 95). Appendix Table A1 (online only) lists the parameters, quality indicators, and source for each medical and surgical oncology indicator reviewed.

Table 1.

Demographic Data

Characteristic	No.	%
Age, years
Median	60
Range	22-95
Race/ethnicity
White/Caucasian	501	80.5
Black/African American	57	9.2
Asian/Pacific Islander	4	0.6
Missing race	60	9.7
Hispanic/Spanish/Latino	37	6
Not Hispanic/Spanish/Latino	176	28
Missing ethnicity	409	66
Sex
Male	4	0.6
Female	618	99.4
Payer
Medicare	234	38
Medicaid	33	5
Private	306	49
Self-pay	15	2
Indigent/charity	27	4
Missing	7	2
Stage
I	246	43
II-III	297	51
IV	35	6

Medical Oncology Indicators

Results pertaining to medical oncology indicators are summarized in Appendix Table A2 (online only). Our results are shown as an overall score for all sites combined, followed by the range for the 11 participating sites.

All 622 charts were assessed for family history and menopausal status, with family history documented in 80% (495 of 622; range: 55% to 95%) and menopausal status documented in 49% (307 of 622; range: 15% to 78%) of charts. Of the 88 reported premenopausal women, 11% (10 of 88; range: 0% to 20%) indicated that fertility preservation had been discussed.

We found that 93% (578 of 622; range: 78% to 100%) of charts showed appropriate American Joint Committee on Cancer staging, and these were used for further abstractions. Documentation of the hormone receptor status (estrogen receptor/progesterone receptor) was evident in 100% (578 of 578; range: 98% to 100%) of charts, and 98% (566 of 578; range: 84% to 100%) showed documentation of HER2/neu status. Sites had excellent compliance for consideration or discussion of adjuvant hormone therapy for patients with both nonmetastatic and metastatic disease (Appendix Table A2).

Of the 330 patients who received chemotherapy, a discussion of chemotherapy was documented in 96% of patients at stage 1c-III: for stage I (T1c), 18 of 18 patients; for stage II-III, 285 of 297 patients (range: 81% to 100%). Consent was documented in 75% (247 of 330; range: 43% to 100%) and a chemotherapy flow sheet in 96% (316 of 330; range: 81% to 100%) of charts. Among all patients who received neoadjuvant or adjuvant chemotherapy, a planned chemotherapy regimen was documented in 75% (226 of 302; range: 43% to 98%). For charts of patients with nonmetastatic disease and chemotherapy documented, the planned dose of chemotherapy fell within the published regimen range in 75% (155 of 208; range: 40% to 89%). Initiation of adjuvant chemotherapy occurred within 8 weeks of definitive surgery for 84% of patients (233 of 279; range: 33% to 95%). Documentation of a surgical procedure after chemotherapy was shown for 89% (34 of 38; range: 33% to 100%) of patients receiving neoadjuvant therapy. A statistically significant difference in variation of performance across the sites was found for seven of 22 medical oncology quality indicators (Appendix Table A2).

Surgical Oncology Indicators

Surgical oncology results are summarized in Appendix Table A2. With respect to pathologic reporting, there was high (> 90%) compliance for presence of a pathology report and documentation of the American Joint Committee on Cancer or TNM staging, tumor size and grade, status of margins, and hormone receptor status. Pathology quality measure questions were applied only to patients who were staged nonmetastatic, had surgery, and had a pathology report in their medical records; that is, 537 charts of the initial 622 were eligible for analysis of pathology quality indicators. Of those 537 charts, 370 (69%; range: 44% to 90%) had documentation that the specimen was oriented and 474 (88%; range: 54% to 98%) stated that the margins were inked. The American Society of Breast Disease policy statement (June 2005) states that “specimen orientation should be performed on all partial breast excisions for a suspicious or malignant lesion.”⁸ Various methods of orientation can be used (clips or suture) providing two points of orientation (ie, superior, lateral) followed by inking. For purposes of this study, any inking was deemed compliant.

For inclusion in the remaining surgery quality measures, the patient had to have a nonmetastatic invasive cancer and documentation of surgical procedure. Sentinel lymph node biopsy is recommended by the American Society of Breast Surgeons in patients with clinically node-negative (T1N0, T2N0) disease based on palpation of the axilla, with or without axillary imaging.⁹ Extreme age, comorbid conditions, multicentric disease, and invasive tumors < 5 mm or > 5 cm were not considered as contraindications during chart abstraction. Sentinel lymph node biopsy was performed in 82% (447 of 542; range: 62% to 94%) of patients with invasive BC regardless of whether breast conservation or mastectomy was performed. Of the patients with a metastatic sentinel lymph node biopsy, 79% (123 of 156; range: 62% to 96%) went on to have a complete axillary node dissection. Isolated tumor cells in the sentinel lymph nodes were not considered node positive. Data reflecting micrometastatic, macrometastatic, or nodal burden were not available but may have confounded the compliance with performing a complete axillary lymph node dissection.

Sites showed moderate compliance regarding patients obtaining mammograms within 14 months of definitive surgery (77%; range: 26% to 98%). Although the four male patients in the series had mastectomies, they were excluded from this analysis. The rate of referral to radiation treatment within 1 year of definitive surgery was high (97%; range: 80% to 100%).

Statistically significant variations across practices were found for margin orientation, inking of the margins (Appendix Figure A2, online only), histological grading, performance of sentinel lymph node biopsy, and obtaining a mammogram within 14 months of definitive surgery.

Indicators by Volume and Age

Because patient volumes can affect delivery of health care, assessments of all indicators were made by dividing the sites into those with low volume (eight sites) and those with high volume (three sites). A marginally significant difference (P < .05) between all high- and low-volume sites was evident for two indicators. The high-volume sites outperformed the low-volume sites for documentation of inked margins. In contrast, the low-volume sites outperformed the high-volume sites on use of sentinel lymph node biopsy.

When stratified by age (< 50, 50-70, and > 70 years), statistically significant differences were evident for three of 34 indicators: documentation of family history, documentation of menopausal status, and obtaining a mammogram within 14 months of surgery. For family history, patients younger than 50 years had better adherence than those in 50 to 70 years age group (odds ratio [OR] = 1.87; 95% CI: 1.1 to 3.2) and the > 70 years group had poorer adherence than the < 50 group (OR = 0.49; 95% CI: 0.3 to 0.9). Likewise, for documentation of menopausal status, the < 50 group had better adherence than the 50 to 70 group (OR = 1.83; 95% CI: 1.2 to 2.7) and the > 70 group had worse adherence than the < 50 group (OR = 0.38; 95% CI: 0.2 to 0.6). There was no significant difference between the older groups for either of these two indicators compared with the 50 to 70 year group. Interestingly, patients > 70 years of age had a higher adherence to obtaining mammograms within 14 months of surgery than patients age 50 to 70 years (OR = 2.07; 95% CI: 1.1 to 3.0) or < 50 years (OR = 2.01; 95% CI: 1.0 to 3.9).

Discussion

This article represents a comprehensive examination, based on process indicators, of the quality of BC care provided at the 11 FIQCC practice sites. In addition to identifying numerous areas of excellence, our systematic review of medical oncology charts identified areas that needed improvement. We will discuss those indicators with less than optimal performance, as they represent important targets for quality improvement efforts. Variability among the sites was found for certain indicators, but in general, these differences did not correspond to the volume of patients seen at a site. Likewise, patient age (which is strongly related to menopause status) was also generally unrelated to performance.

In 2002, ASCO's QOPI included guidelines on chemotherapy consent and documentation. In our review, overall compliance of 75% of charts was reached for both documentation of consent and planned chemotherapy regimen. Exploring our data further, there was significant (P < .001) intersite variance for both parameters. Informed consent is a significant aspect of the medical oncologist's role,^1,10 and clinicians must be able to adequately inform patients regarding specifics of the recommended chemotherapy regimens.

Equally as important is an explanation of why a patient may not be appropriate for chemotherapy. We recognize that factors such as multiple comorbidities, poor Eastern Cooperative Oncology Group performance status, and postoperative complications play key roles in the chemotherapy decision-making process. Our analysis showed that this indicator (indicator 16 in Table 2) was documented in only 20% of charts, with high intersite variance. Unfortunately, the low sample size makes it difficult to draw firm conclusions.

Low rates of adequate chemotherapy dosing were noted across all sites reviewed, which is concerning, as others^11,12 have documented that suboptimal therapeutic chemotherapy dosing can lead to worse outcomes. Strong consideration should be given to the administration of adequate doses of chemotherapy to patients with BC as long as it is deemed safe. If inadequate dosing is necessary because of poor performance status or comorbid conditions, then clear documentation of the reasoning should be provided in the medical chart.

Fertility preservation was a prominent measurement, as compliance was low, whereas concordance between the sites was high. Although it was not a specific quality care measure according to QOPI, NICCQ, and FIQCC guidelines, ASCO¹³ and St Gallens¹⁴ have incorporated new guidelines regarding the importance of discussion of fertility preservation during the medical oncology consultation. The FIQCC panel included it as an exploratory indicator and found that documentation of discussions on fertility preservation was low across all sites. We acknowledge the difficulty in drawing specific conclusions on a small sample size but still recognize this as an area of deficiency.

Perhaps one theme that links these above indicators together is the concept that physicians might consider that the indicators are “implicit” in the medical chart and thus do not require specific documentation. For example, a 65-year-old woman diagnosed with BC is presumed to be postmenopausal, or an individual with multiple comorbidities would always undergo dose reduction for their first cycle of chemotherapy. Continuous measurements of quality indicators with associated feedback regarding the importance of documentation will lead to improvement in quality of cancer care. Therefore, unknown deficiencies such as those identified in this study can be identified, highlighted, and resolved.

With regard to surgical oncology indicators, we identified the need for improvement in margin orientation and inking, use of sentinel lymph node biopsy, and obtaining mammograms in postsurgical follow-up. One explanation for the lack of margin orientation is the continued use of primary surgical excision, as opposed to core needle biopsy, for diagnostic purposes of abnormal imaging. Most surgeons appeared not to have oriented the diagnostic biopsy specimen despite the policy statement from the American Society of Breast Diseases recommending that orientation (including inking) be performed on all surgical excision specimens if suspicious.⁸ If the diagnosis returns as malignant, then the specimen would be listed in our data abstraction as “no margin orientation.” Similarly, BC surgery can be performed in an ambulatory surgery center where on-site pathology is not available; thus the specimens would be placed in formalin without inking unless the surgeon performs the inking.

Use of sentinel node biopsy is dependent on several factors. Yet many experienced general surgeons still have not obtained the additional training required to perform this procedure. However, in our data, the high-volume centers showed significantly less frequent use of this procedure. One could rationalize that a selection gradient, or bias, might explain these results, such as the case of a frail elderly patient in whom the knowledge of the true sentinel node status would not change disease management or increase the use of preoperative axillary ultrasound with fine needle aspiration.

Failure to obtain a mammogram within 14 months of definitive surgery might be the result of missing documentation rather than not actually having the test performed. Traditionally, mammograms are under the purview of the surgeon, who might not forward the results to their medical oncology partners; thus information would not be included in a patient's medical chart (for abstraction).

The use of medical oncology charts as the sole source of data acquisition serves as a limitation of this study with regard to surgical/pathologic or radiation oncology indicators. This raises questions regarding whether problems with documentation can alter the results and limit data interpretation. As previously reported, supplementation of patient recorded data to physician records can result in quality scores increasing by 5% to 10%.^15,16 This effect is somewhat minimized within the FIQCC by the inclusion of patient history and new patient questionnaires. Another limitation is the small sample sizes for some of the indicators, which easily could skew the interpretation of compliance. In addition, depending on the level of evidence-based support and physician/center prioritization, some indicators will be more heavily weighted in the clinic than others, thus affecting the outcomes of some measurements. Although limitations do exist, the main strengths of the FIQCC are the diversity across sites, (ie, academic v private practice) and their broad geographic distribution, which make the results more broadly applicable.

All institutions were provided with the results shown in the figures and were un-blinded as to which letter (A-K) corresponded to their particular site. These findings are now being used at the participating institutions to guide quality improvement efforts. We anticipate that a comparison of the pre- and postimplementation data on those indicators targeted for improvement will show significant enhancements in quality of care over time. With methods developed and implemented as part of this future project and with refinement of the indicators, we plan to develop a model that could be used effectively by other local and regional networks throughout the United States to improve the quality of care for women with BC.

Appendix

Table A1.

Quality Indicators for Breast Cancer

Indicator Number	Parameter	Indicator	Source
1	Patient centric care: Documentation of family history	Was “the patient's family history of breast and ovarian cancer documented?”	Panel consensus
2	Patient centric care: Documentation of menopausal status	Was “the patient's menopausal status documented in the initial medical oncologist's note?”	Panel consensus
3	Patient safety: Diagnostic ascertainment	Was “there an explicit statement of the patient's staging according to the AJCC system?”	Panel consensus
4	Patient safety: Diagnostic ascertainment	For patients who do not have an explicit statement of staging in their chart, is “the tumor size, lymph node status, and metastatic status documented in the medical chart?”	Panel consensus
5	Patient safety: Diagnostic ascertainment	For nonmetastatic patients, was “there a copy of the surgical pathology report confirming malignancy in the medical oncology office chart?”	QOPI4 and panel consensus
6	Patient safety: Diagnostic ascertainment	For nonmetastatic patients who had a breast tumor surgically removed, does “the pathology report state the tumor size?”	NICCQ17 and panel consensus
7	Application of evidence and consensus-based treatment: Inked margins	For nonmetastatic patients who had a breast tumor surgically removed, does “the pathology report state that the margins were inked?”	NICCQ17 and panel consensus
8	Application of evidence and consensus-based treatment: Specimen oriented	For nonmetastatic patients who had a breast tumor surgically removed, does “the pathology report state that the specimen was oriented?”	Panel consensus
9	Application of evidence and consensus-based treatment: Margin status	For nonmetastatic patients who had a breast tumor surgically removed, does “the pathology report state the status of the margins?”	NICCQ17 and panel consensus
10	Patient safety: Diagnostic ascertainment	For nonmetastatic patients who had a breast tumor surgically removed, does “the pathology report state the histologic grade of the tumor?”	NICCQ17 and panel consensus
11	Patient safety: Diagnostic ascertainment	For nonmetastatic patients who had a breast tumor surgically removed, does “the pathology report state that the hormone receptor status of the tumor was checked and reported?”	NICCQ17 and panel consensus
12	Patient safety: Diagnostic ascertainment	Was “HER2 (determined by IHC or FISH) documented in the medical chart?”	Panel consensus
13	Application of evidence and consensus-based treatment: Use of herceptin	For patients who had positive HER2 status, did “the physician discuss, recommend, or refer treatment with Herceptin?”	NCCN guidelines and panel consensus
14	Application of evidence and consensus-based treatment: Sentinel lymph node biopsy	For nonmetastatic patients, was “sentinel lymph node biopsy (SLN) performed?”	NICCQ17 and panel consensus
15	Application of evidence and consensus-based treatment: Complete axillary lymph node dissection	For nonmetastatic patients who had a positive SLN, was “complete axillary lymph node dissection (CALND) performed?”	NICCQ17 and panel consensus
16	Patient centric care: Explanation for not administering treatment	For nonmetastatic patients who were assessed by neither SLN nor CALND, was “there any documentation in the medical record that the patient is NOT a candidate for adjuvant chemotherapy?”	NICCQ17 and panel consensus
17	Application of evidence and consensus-based treatment: Use of radiation	For nonmetastatic patients who had breast conserving surgery, did “the physician discuss, recommend, or refer for radiation therapy within one year of diagnosis?”	NICCQ,17 ACoS, and panel consensus
18	Patient safety: Diagnostic ascertainment	Was “hormone receptor status both ER and PR (i.e. positive, negative, or percentage) documented in the medical chart?”	Panel consensus
19	Application of evidence and consensus-based treatment: Use of hormone therapy	For nonmetastatic patients with ER+ or PR+ breast cancer, did “the physician discuss, recommend, or refer treatment with Tamoxifen or aromatase inhibitor within one year of diagnosis?”	NICCQ,17 ACoS, and panel consensus
20	Application of evidence and consensus-based treatment: Use of estrogen blockade	For stage IV patients with ER+ or PR+ breast cancer, did “the physician discuss, recommend, or refer treatment with Tamoxifen, Fulvestrant, or aromatase inhibitors within one year of diagnosis?”	ACoS and panel consensus
21	Application of evidence and consensus-based treatment: Use of chemotherapy	For patients with stage I, ER/PR negative disease and a tumor size greater than 1 cm, did “the physician discuss, recommend, or refer for chemotherapy?”	NCCN guidelines, NICCQ,17 and panel consensus
22	Application of evidence and consensus-based treatment: Use of chemotherapy	For patients with stage II or III disease, did “the physician discuss, recommend, or refer for chemotherapy?”	NCCN guidelines, NICCQ,17 and panel consensus
23	Patient centric care: Consent for chemotherapy treatment	For patients who received chemotherapy, was “there a signed consent for treatment in the chart or a practitioner's note that treatment was discussed and patient consented to treatment?”	QOPI4 and panel consensus
24	Patient safety: Use of flow sheets	For patients who received chemotherapy, was “there a flow sheet with chemotherapy notes and blood counts?”	QOPI4 and panel consensus
25	Patient safety: Chemotherapy dosage	For patients who received chemotherapy, was “the patient's planned dose of chemotherapy documented in the medical oncology note?”	NICCQ17 and panel consensus
26	Application of evidence and consensus-based treatment: Adherence to published chemotherapy regimens	For nonmetastatic patients whose planned dose of chemotherapy was documented, did “the patient's planned dose of chemotherapy, dose per cycle, and number of cycles fall within a range that is consistent with published regimens?”	NCCN guidelines, NICCQ17 and panel consensus
27	Patient safety: Documentation of body-surface area	For patients who received chemotherapy, was “the patient's body-surface area (BSA) documented?”	NICCQ17 and panel consensus
28	Application of evidence and consensus-based treatment: Use of chemotherapy	For nonmetastatic patients who received adjuvant chemotherapy, did “the patient receive an adjuvant chemotherapy regimen from the provided list?”	NCCN guidelines and panel consensus
29	Application of evidence and consensus-based treatment: Use of chemotherapy	For metastatic patients who received chemotherapy, did “the patient receive a chemotherapy regimen from the provided list?”	NCCN guidelines and panel consensus
30	Application of evidence and consensus-based treatment: Surgery following neoadjuvant therapy	For nonmetastatic patients who received neoadjuvant chemotherapy, did “the patient receive definitive surgery after neoadjuvant chemotherapy?”	Panel consensus
31	Application of evidence and consensus-based treatment: Adjuvant chemotherapy initiation	For nonmetastatic patients who received adjuvant chemotherapy, did “the patient start adjuvant chemotherapy within 8 wk of completion of surgical therapy?”	NICCQ17 and panel consensus
32	Application of evidence and consensus-based treatment: Adjuvant chemotherapy initiation	For nonmetastatic patients who received adjuvant chemotherapy, did “the patient start adjuvant chemotherapy within 4 mo of diagnosis?”	ACoS and panel consensus
33	Application of evidence-based surveillance: Mammogram	For nonmetastatic patients who did not have bilateral mastectomies, is “there evidence in the medical chart that the patient had a mammogram within 14 mo of completion of surgical therapy?”	Panel consensus
34	Patient-centric care	For patients who were documented to be premenopausal in chart, was “there evidence in the medical chart that fertility preservation was discussed with the patient?”	ASCO and panel consensus

Abbreviations: ACoS, American College of Surgeons; ER, estrogen receptor; FISH, fluorescence in situ hybridization; HER, human epidermal growth factor receptor; IHC, immunohistochemistry; NCCN, National Comprehensive Cancer Network; NICCQ, National Initiative for Cancer Care Quality; PR, progesterone receptor; QOPI, Quality Oncology Practice Initiative.

Table A2.

Compliance of Medical and Surgical Oncology Indicators

Indicator Type and No.	Indicator	Eligible Event (No.)	Rate (%)	Range (%)	P Value Across Sites
Medical oncology
29	Documentation: Metastatic chemotherapy from an approved list	22	100	100-100	NA
18	Documentation: Hormone receptor status	578	100	98-100	.345
27	Documentation: Body surface area	330	98	93-100	.056
19	Discussion/recommendation: Adjuvant AI or tamoxifen for ER/PR-positive breast cancer	441	98	90-100	.033
12	Documentation of HER2neu status	578	98	84-100	< .001
24	Documentation: Chemotherapy flow sheet	330	96	81-100	< .001
21 & 22	Discussion/Recommendation: adjuvant chemotherapy	330	96	81-100	.005
28	Documentation: Nonmetastatic chemotherapy from an approved list	270	94	86-100	.086
3	Documentation: AJCC staging in medical chart	622	93	78-100	< .001
20	Discussion/recommendation: Metastatic AI, tamoxifen, or fulvestrant	27	93	75-100	.896
32	Documentation: Initiation of adjuvant chemotherapy within 4 mo of diagnosis	279	92	78-100	.447
30	Documentation: Definitive surgery post neoadjuvant chemotherapy	38	89	33-100	.101
31	Documentation: Initiation of adjuvant chemotherapy within 8 wk of surgery	279	84	33-95	.005
13	Discussion/recommendation: Adjuvant trastuzumab for HER2neu-positive breast cancer	116	82	50-100	.221
1	Documentation: Family history	622	80	55-95	< .001
23	Documentation: Informed consent	330	75	43-100	< .001
25	Documentation: Planned chemotherapy regimen	302	75	43-98	< .001
26	Documentation: Planned chemotherapy dose fell within published regimen doses	208	75	40-89	.008
2	Documentation: Menopausal status	622	49	15-78	< .001
16	Documentation: Explanation for not administering treatment	20	20	0-100	.13
34	Discussion/recommendation: Fertility preservation*	88	11	0-20	.833
Surgical oncology indicators
5	Documentation: Copy of pathology report in chart	543	100	98-100	.337
11	Documentation: Hormone receptor status reported on pathology report	537	99	94-100	.033
6	Documentation: Tumor size on pathology report	537	98	91-100	.003
9	Documentation: Status of margins on pathology report	537	98	91-100	.182
17	Documentation: Referral to radiation oncology within 1 yr	294	97	80-100	.012
10	Documentation: Path states histologic grade of tumor	537	96	87-100	< .001
4	Documentation: Group stage by AJCC or complete TNM	622	93	78-100	< .001
7	Documentation: Inked margins on pathology report	537	88	54-98	< .001
14	Performance: SLN biopsy in clinical N0 patients	543	82	62-94	< .001
15	Performance: CALND for positive SLN biopsy	156	79	62-96	.166
33	Performance: Mammogram within 14 mo of surgery	468	77	26-98	< .001
8	Documentation: Specimen oriented on pathology report	537	69	44-90	< .001

Abbreviations: AI, aromatase inhibitor; AJCC, American Joint Committee on Cancer; CALND, complete axillary lymph node dissection; ER, estrogen receptor; HER, human epidermal growth factor receptor; N0, node negative; NA, not applicable; PR, progesterone receptor; SLN, sentinel lymph node biopsy.

^*Not a specific quality indicator.

Figure A1.

Locations of Florida Initiative for Quality Cancer Care Consortium institutions.⁷ Reprinted with permission from Cancer Control.

Figure A2.

Variance in performance across individual practice sites for surgical quality indicators. Percentages of practices indicating that (A) pathology report states that specimen was oriented, (B) pathology report states that specimen was inked.

Authors' Disclosures of Potential Conflicts of Interest

Although all authors completed the disclosure declaration, the following authors indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Employment or Leadership Position: None Consultant or Advisory Role: None Stock Ownership: None Honoraria: Christine Laronga, Genomic Health Research Funding: Paul B. Jacobsen, Pfizer Expert Testimony: None Other Remuneration: None

Author Contributions

Conception and design: Jhanelle E. Gray, Christine Laronga, Erin M. Siegel, David Shibata, Mokenge Malafa, Paul B. Jacobsen

Administrative support: Michelle Fletcher

Provision of study materials or patients: Jhanelle E. Gray, Christine Laronga, Fred Schreiber, Richard Brown, Richard Levine, Thomas Cartwright, Guillermo Abesada-Terk Jr, George P. Kim, Carlos Alemany, Douglas Faig, Philip Sharp, Merry-Jennifer Markham

Collection and assembly of data: Jhanelle E. Gray, Christine Laronga, Michelle Fletcher, Fred Schreiber, Richard Brown, Richard Levine, Thomas Cartwright, Guillermo Abesada-Terk Jr, George P. Kim, Carlos Alemany, Douglas Faig, Philip Sharp, Merry-Jennifer Markham

Data analysis and interpretation: Jhanelle E. Gray, Christine Laronga, Erin M. Siegel, Ji-Hyun Lee, William J. Fulp, Paul B. Jacobsen

Manuscript writing: Jhanelle E. Gray, Christine Laronga, Erin M. Siegel, Ji-Hyun Lee, Michelle Fletcher, William J. Fulp, Paul B. Jacobsen

Final approval of manuscript: Jhanelle E. Gray, Christine Laronga, Erin M. Siegel, Ji-Hyun Lee, William J. Fulp, Michelle Fletcher, Fred Schreiber, Richard Brown, Richard Levine, Thomas Cartwright, Guillermo Abesada-Terk Jr, George P. Kim, Carlos Alemany, Douglas Faig, Philip Sharp, Merry-Jennifer Markham, David Shibata, Mokenge Malafa, Paul B. Jacobsen