(Chest. 2004;125:1753-1760.)
© 2004
American College of Chest Physicians
Use of a Clinical Pathway To Manage Unsuspected Radiographic Findings*
William E. Holden, MD;
David M. Lewinsohn, MD, PhD;
Molly L. Osborne, MD, PhD, FCCP;
Chris Griffin, MD, MBA;
Ann Spencer, RN;
Carol Duncan, RN and
Mark E. Deffebach, MD
* From the Medical Service (Drs. Holden, Lewinsohn, Osborne, and Deffebach, and Mss. Spencer and Duncan), Pulmonary and Critical Care Section, and the Diagnostic Radiology Service (Dr. Griffin), Portland Veterans Administration Medical Center, Portland OR.
Correspondence to: William E. Holden, MD, P3-PULM, Portland VA Medical Center, 3710 SW US Veterans Hospital Rd, Portland, OR 97201; e-mail: william.holden{at}med.va.gov
 |
Abstract
|
|---|
Study objectives: To describe our 5-year experience with a clinical pathway used to ensure the timely communication and evaluation of unsuspected radiologic findings (URFs) noted on clinically requested chest imaging.
Design: Prospective data collection on clinical practice.
Setting: Academically affiliated Veterans Affairs medical center.
Participants: Pulmonary physicians, nurses, and radiologists.
Results: Over a period of 5 years, 1,629 URFs were referred to the pathway (from chest radiographs, 1,359 [83.4%]; from CT scans, 270 [16.6%]). Most URFs (78%) were nodules, with a specific diagnosis made in one third of URFs, and with a specific diagnosis thought to be clinically significant in another one third of URFs. The most common diagnosis was neoplasm, with over two thirds of these diagnoses being lung cancer. One third of lung cancers detected were either stage 1 or 2, with 1 in 17 of all URFs being stage IA lung cancer. The cost of the pathway was estimated at $28,600 per year.
Conclusions: URFs noted on chest imaging are frequently clinically significant, and a systematic approach to managing URFs, such as a clinical pathway, can significantly improve care in a large teaching hospital.
Key Words: chest radiograph • clinical pathway • lung cancer • lung nodule • radiology • radiographic findings • radiology information systems
|
Introduction
|
|---|
Radiographic studies of the chest are generally requested for specific clinical indications that are described in a short phrase drafted by the clinician and appended to the request. The radiologist notes this "clinical information" as the radiograph or CT scan is examined. Frequently, however, findings are detected that are not related to the clinical problem or to the reason for the study, as outlined in the clinical information provided. These unsuspected radiographic findings (URFs) may have important clinical information (eg, a new, solitary pulmonary nodule suspicious for lung cancer), or they may have little or no clinical importance. Often, the radiologist cannot establish the clinical significance, and communication between the radiologist and the clinician is required for clarification of the finding’s significance. Additionally, URFs often may require follow-up actions (eg, a search for older comparison studies), additional validation studies (eg, further radiographic views of the chest or the addition of an IV contrast agent to a CT scan), or simply follow-up over time. While URFs are common, their spectrum and clinical significance are unknown. Furthermore, URFs have the potential for delay in diagnosis and treatment, or to be "lost to follow-up" because of the myriad communication and systems issues inherent to the process of dealing with them, especially in a large hospital environment. These include an increasing number and variety of providers involved in contemporary medical practice. The lack of timely and appropriate evaluation of URFs can pose significant medical and legal risks.
Our institution devised and implemented a clinical pathway to manage URFs in 1996. The purposes of the pathway are to ensure the timely communication of URFs to clinicians, and to facilitate the further evaluation and diagnosis of URFs. In this report, we review our 5-year experience using the pathway with the following objectives. First, we have characterized the type of radiographic findings that presented as URFs. Second, we have analyzed the outcomes and clinical significance of URFs. Finally, we have analyzed the monetary costs of the URFs pathway in time and personnel. Our conclusion is that the majority of URFs have clinically important information, including a sizable number of cases of early-stage lung cancer. The dollar costs, while significant, seem reasonable to us in the context of the beneficial impacts of the pathway. Our impression is that a systematic approach to the management of URFs using a clinical pathway can significantly improve the efficiency of patient care in a large teaching medical center.
|
Materials and Methods
|
|---|
Setting
The pathway was instituted in an urban US Veterans Affairs medical center that was affiliated with a university teaching center and a series of smaller, regional hospitals and clinics that were interconnected by an electronic, paperless chart.
Organization of the URFs Clinical Pathway
The pathway was implemented in early 1996 and has continued to the present with minimal modifications, including changes in personnel. All patients referred to the pathway from April 1, 1996, through March 31, 2001, are included in this report. The examining radiologist, based on the clinical information provided, identified URFs and initiated the pathway (Fig 1
). URFs found on the radiographic study were defined as findings that were not related to or explained by the clinical information provided by the clinician. The URFs often were the only detected findings, or were identified in conjunction with other, clinically suspected findings. Any imaging study (eg, shoulder radiographs or abdominal CT studies) of all or parts of the chest demonstrating URFs was included in the pathway. Routine chest radiographs at our institution are performed using a computed radiography system (Fuji; Tokyo, Japan) with digital images examined on 4-megapixel diagnostic quality monitors (Megascan Corporation; North Billerica, MA) using software developed by the Department of Veterans Affairs (Vista-Rad). When possible, all chest radiographs in ambulatory patients were two-view studies (ie, posteroanterior and lateral). Chest CT studies of the thorax were performed with IV contrast, unless there were contraindications, using a four-detector scanner (Light-Speed; GE Medical Systems; Waukesha, WI) or a one-detector scanner (Cti; GE Medical Systems).
View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1.. Flow diagram for the URF clinical pathway. The radiologist initiated the pathway, directing the finding to a pulmonary consultation team through a pulmonary nurse specialist, who collated data and notified the primary care provider. The pulmonary consultation team evaluated the URF, occasionally with the help of a thoracic radiologist, and a plan for evaluation or follow-up was drafted. The URF then was observed over time until a diagnosis was made, stability was assured, or the URF resolved.
|
|
A pulmonary nurse specialist received notification of the URFs by an electronic message and collated relevant clinical information from the electronic chart along with available past radiologic studies for review by the pulmonary service consultation team. The consultation team reviewed the URFs and relevant clinical information, occasionally with the help of a thoracic radiologist, and formulated a preliminary plan for further evaluation. At this point, the patient’s primary care provider was notified of the URFs along with the suggested plan for evaluation, and was given the opportunity to provide further information and to participate in the decision-making process. If no primary care provider was identified, the pulmonary consultation team assumed responsibility for the evaluation. The plan for further evaluation and follow-up of the URFs, as well as the patient’s electronic chart, was entered into an electronic database (see below), and the consultation team either directed or monitored the evaluation and follow-up until the URF was resolved. Resolution of the URF could occur in one of several ways. First, the URF might resolve or might not be confirmed by subsequent studies. Second, a specific diagnosis of the URF might be made. Third, the URF might demonstrate stability over time. Finally, a minority of URFs was unresolved because the patient was lost to follow-up or died before the evaluation was complete. The pathway remained active until terminated by the pulmonary consultation team.
Database Organization and Analysis: Outcomes and Clinical Significance
We entered each URF into an electronic database (Access; Microsoft Corporation; Redmond, WA) along with the relevant clinical information, radiologic interpretations, and the formulated plan for evaluation. As follow-up studies were obtained, prior studies reviewed, or other information became available, the additional information was entered in the database. For the purposes of our evaluation of the URF pathway, we grouped the findings by radiologic characteristics, as described by the referring radiologist, into one of the following six types: (1) nodular densities (ie, densities with reasonably sharp outlines1); (2) infiltrates (ie, densities with ill-defined borders1); (3) pleural shadows; (4) bony or chest wall shadows; (5) mediastinal or hilar shadows; or (6) other types of shadows. The outcome of the pathway evaluation and follow-up of each URF was determined and was placed into one of the following five outcome categories: (1) URF resolved or not subsequently seen; (2) URF stable over time; (3) definitive diagnosis made of URF; (4) URF unresolved (ie, patient died or was lost to follow-up); or (5) continued follow-up of URF was necessary. During the evaluation or follow-up of primary URFs, we occasionally uncovered additional URFs. These secondary URFs were analyzed separately in the same way as outlined above for the primary URFs.
We also assessed whether each URF was clinically significant using the following criteria. URFs were considered to be clinically significant if a definitive diagnosis was made. URFs were considered to be of indeterminate significance if they were persistent and shown to be stable over time, or if the finding was considered to be significant enough to require further follow-up. Nodular abnormalities demonstrated to be stable over 2 years, whether on prospective evaluation or by comparison to older studies, were considered likely to be benign.23 However, despite 2 years of follow-up, some of the latter URFs were still under observation or were lost to follow-up at the time of this analysis. It is possible that in time these URFs will be further classified. Reasons for continued and prolonged follow-up included the very slow evolution of the findings or medical conditions that precluded invasive evaluation. Stability over time was demonstrated either by obtaining older, comparison studies that demonstrated the finding, or by following the URFs prospectively with comparison studies showing no change. URFs were considered to be clinically insignificant if they resolved or were not subsequently seen during the evaluation.
Determination of Costs of the URF Pathway
The primary costs of the URF clinical pathway were in personnel time and salary. Because URFs were reported by radiologists reading clinically indicated studies, our assumption was that the URF pathway did not add additional or unnecessary costs to studies or procedures for the evaluation of the reported findings. Prior to the implementation of the pathway, URFs either were communicated directly to the clinicians by the radiologist, or the clinicians were alerted to the URFs by the recorded interpretation of the radiologic study. In either case, the clinicians then proceeded with further evaluation or follow-up of the URFs, although the responsibility for clinical follow-up was occasionally diffuse and poorly defined. The only new and additional features added to this process by the URF pathway were: (1) a consistent, timely communication of the URFs to the clinician, (2) the activities of the pulmonary nurse specialist in collating information and prior studies, and communicating with the primary care provider, and (3) the activities of the pulmonary consultation team in reviewing the URFs, and formulating a suggested plan of evaluation and follow-up. It could be argued that the pulmonary consultation team would have been involved in the evaluation of many URFs even in the absence of the clinical pathway (ie, a pulmonary consultation would have been requested). For this reason, we did not include the time of the consultation team in directing or performing further studies, but included only those activities that related to the formulation of the plan of evaluation or follow-up, and collating/recording the results. The time (in hours) needed for pathway activities was first estimated. The dollar costs then were calculated based on the hourly salaries of the personnel involved in each activity.
|
Results
|
|---|
From April 1, 1996, through March 31, 2001, a total of 1,629 URFs was referred to the clinical pathway (Fig 2
). The average age of patients referred to the pathway was 66.3 ± 11.9 years (mean ± SD) and, consistent with the Veterans Affairs patient population, 96% were men. The number of URFs increased sharply in the third year and remained relatively constant in the following 2 years. To place these numbers in context, the number of patient encounters in our hospital also increased over the 5 years, from 291,776 per year in 1996 to 379,247 in 2001, which is an increase of 29.9%. Interestingly, while the number of chest radiographs at our institution remained relatively constant (20,000 to 21,000 studies per year, respectively), there was a dramatic increase in the number of CT studies performed (600 to 1,750 studies per year, respectively [a 192% increase]) over the time frame of the study.
View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2.. Growth of the URF pathway over a 5-year period. The vertical axis shows the number of URFs per year. The number of URFs increased sharply after the second year and remained steady thereafter. A total of 1,629 cases were analyzed over the 5-year period.
|
|
Types of Radiographic Findings and Diseases Presenting as URFs
The vast majority of URFs (78%) were nodules or other densities that were too small to characterize specifically (Fig 3
). The second most common type of URF (10%) consisted of abnormal shadows in the mediastinum or hila. Infiltrates, pleural densities, and bony or chest wall shadows were less common. Only 3% of URFs could not be readily placed into one of the above categories and were classified as "other shadows."
View larger version (49K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3.. Types of radiographic findings presenting as URFs. The great majority of URFs (78%) were nodules, or similar densities/opacities that were too small to be characterized further. Densities or shadows in the hila or mediastinum, infiltrates, other densities, pleural densities, and bone or chest wall shadows accounted for the remainder of URFs.
|
|
Predictably, during the evaluation of the primary URFs we often uncovered other unsuspected findings. For example, a CT study performed to investigate a primary URF would reveal yet another nodular density that was clinically unsuspected. We found 75 of these secondary URFs, or one secondary URF, for every 22 primary URFs evaluated. The types of secondary URFs were quite similar to the primary URFs in that the majority (66%) were nodules of similar densities.
Clinical Outcomes and Significance of URFs
In our analysis of clinical outcomes, we assigned each primary or secondary URF to one of five categories (Table 1
). A specific diagnosis was made in just under one third of primary URFs (32.8%) and secondary URFs (32.9%). Most other primary and secondary URFs either resolved or were stable with the passage of time, and a smaller number required continued observation or were unresolved at the time of this report.
To assess the clinical significance of URFs, we reasoned that a clinically significant finding was one that led to an eventual specific diagnosis. URFs that demonstrated stability over time (either retrospectively or prospectively) or that were suspicious enough to require further observation were considered to be of indeterminate significance. In contrast, URFs that resolved or were not subsequently seen on follow-up examinations were considered to be insignificant. Using these criteria, 32.8% of primary URFs and 32.9% of secondary URFs were clinically significant, 21.7% of primary URFs and 32.9% of secondary URFs were of indeterminate significance, and the remainder of primary and secondary URFs were clinically insignificant. Of the clinically significant URFs, two thirds of primary URFs (66.7%) and over half of secondary URFs (56.2%) required immediate treatment.
Diagnoses of URFs
The diseases presenting as either primary or secondary URFs were organized into one of the following three diagnostic categories: neoplastic; inflammatory; or benign (Table 2
). It is noteworthy that half of the specific diagnoses made in primary URFs (50.5%) were neoplastic, and that over two thirds of these neoplasms (197 of 268; 73.5%) were primary lung cancer. Similar findings were seen in secondary URFs, among which 41.7% of secondary URFs were neoplasms and 80% were primary lung cancer. Primary lung cancer was by far the most common single diagnosis in both primary and secondary URFs, accounting for 36.3% and 33%, respectively, of the total number of diagnoses made. As might be expected in an analysis of unexpected findings, the pathway non-small cell lung cancers tended to be in an early stage, with a majority (57%) of the total number of cancers (ie, primary and secondary URFs) being at either stage 1 or 2 (Fig 4
). One third of the non-small cell lung cancers found by the URF pathway (33.7%) were at stage IA, the most favorable stage for responsiveness to treatment. Stated another way, 1 of every 17 URFs was a stage IA primary lung cancer. In contrast, the proportion of early-stage lung cancers (ie, stage I or II) diagnosed hospital-wide (ie, cancers diagnosed by the URF pathway and in all other venues) in our institution over the 5-year period was 34.2%, and 16.5% of them were stage IA. The most common cell type reported by the pathologist was squamous cell carcinoma (32%), followed by adenocarcinoma (23%) and small cell carcinoma (15%) [Fig 4].
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4.. Stage of non-small cell lung cancers presenting as URFs with the total numbers shown as bars and the percentage of each stage shown within each bar. The inset shows the distribution of cell types for all cancers reported by the pathologist. Squamous cell carcinoma accounted for 32%, adenocarcinoma accounted for 23%, and small cell carcinoma accounted for 15% of all cancer diagnoses through the URF pathway.
|
|
Personnel Costs of the URF Pathway
The dollar costs of the URF pathway were overwhelmingly due to personnel (ie, nurse and physician) time. The pathway required 10 h per week of nurse specialist time, and 2.5 h per week of pulmonary physician time. The total weekly costs for these personnel were estimated to be $600, with an annual cost estimated at $28,600. Costs in time were insignificant for our radiology department, and consisted mainly of the time needed to record the finding and to notify the pulmonary nurse specialist. We reasoned that the main function of the pathway was to ensure and facilitate the evaluation of URFs and that the costs of the evaluation of the URFs as part of a pathway would not differ from the costs if the pathway did not exist. Most of the pathway costs were due to pulmonary nurse specialist time to draw data together for review, to record the findings, and to communicate with primary care providers and other members of the pathway team. However, there was a significant amount of pulmonary physician time and cost in analyzing the URFs and organizing the evaluation or follow-up.
|
Discussion
|
|---|
In this report, we describe our 5-year experience with a clinical pathway approach to the analysis of URFs. We instituted the clinical pathway in 1996 in response to a sentinel event in which an unsuspected finding on a chest radiograph was noted, but was not adequately communicated or evaluated. URFs are a frequent occurrence, and, prior to this report, the radiologic profile and clinical significance of these findings were unknown. The primary goals of the pathway were to ensure the timely communication of URFs to the responsible clinicians and to facilitate the evaluation of URFs. In the process, we have learned additional information about the nature and significance of URFs. The most important findings were as follows. First, the vast majority of URFs were nodules or similar densities. Second, a specific diagnosis was made in roughly one third of URFs, whereas most of the remaining URFs were either stable over time or were not subsequently seen, and a minority remained unresolved or are still under observation. Third, in cases in which a specific diagnosis was made, over half were neoplasms, and two thirds of these neoplasms were primary lung cancer. Fourth, over half of the primary lung cancers were at an early stage (ie, stage I or II), and over one third were at stage IA, the most favorable prognostic stage for long-term survival. Finally, the dollar costs of the URF pathway were modest, consisting largely of nurse and physician time. It is our conclusion that these costs are favorably offset by the benefits of the pathway in guaranteeing the timely communication of URFs to the responsible clinicians and in facilitating a URF evaluation.
Our Veterans Affairs hospital is a tertiary care, teaching hospital affiliated with a university medical center. We are also affiliated with a series of smaller, regional hospitals and clinics that are interconnected by an electronic, paperless chart. The electronic chart was implemented during the time of this study, and facilitated communication between providers and the collation of data. Over a period of several years prior to 1996, we noted a small, but significant, number of URFs that either were delayed in evaluation or were lost to follow-up, occasionally with adverse effects on patient care. A multitude of system and communication problems contributed to this problem, many related to the increasing fragmentation and complexity of contemporary medical care. The clinical pathway was designed to be simple, yet to address the myriad issues involved. Although we have not systematically studied the impact of the URF pathway on a delay in diagnosis since the pathway has been in place, no further instances of litigation in cases of URFs have occurred.
In the United States, the responsibility for resolving URFs has largely been the domain of radiologists. The standard for the communication of urgent findings outlined by the American College of Radiology states, "If there are urgent... findings, radiologists should communicate directly with the referring physician... ."4 In general, this standard would hold for conditions that require treatment before the printed report would reasonably be expected to be delivered to the referring physician.5 Expectations for the communication of unexpected, but not urgent, findings are less clearly defined, but Berlin6 has recommended that radiologists should verbally communicate URFs to the referring physician, especially in the case of URFs on routine (eg, preoperative) radiographs, on which referring physicians are less likely to expect abnormalities, and therefore might not seek out the results as carefully. The significant feature of our URF clinical pathway is that we consider the responsibility for communication to be institutional, and have broadened the lines of communication to include pulmonologists and the primary care physician, along with the requesting provider, in cases in which URFs occurred in radiographs of the thorax. In addition, we have incorporated a process to both evaluate and follow URFs to diagnosis, resolution, or determined stability. The key and novel features of this process are that it is multidisciplinary (including a thoracic radiologist, a pulmonologist, a pulmonary nurse specialist, and the responsible primary care provider) and that it eliminates delays in evaluation, as well as preventing the URFs from being lost to follow-up.
The majority of URFs were nodules, or were of similar densities that were too small to specifically characterize as nodules (Fig 3). Radiologists often use the terms density or opacity to describe lesions that are too small to specifically characterize as nodules. The location of these lesions was in the peripheral lung fields on either the radiograph or CT study. That a majority of URFs occurred in the lung fields is logical, since it is easier to detect smaller lesions in the lung fields compared with the more complex hilar, mediastinal, or chest wall images on both chest radiographs and CT scans. The smallest lesions detected on the pathway were in the range of 5 to 6 mm on chest radiographs, and 3 mm on CT scans. These measurements are in the range of the minimal detection sizes for lung carcinoma by chest radiographs and CT studies cited in the literature.78 Since > 65% of patients visiting our emergency care unit are present or past smokers who are at risk for the development of primary lung carcinoma, the detection and prompt treatment of occult carcinoma is a high priority. It is clear that survival following therapy for patients with early-stage non-small cell lung carcinoma is improved compared with that for patients with advanced-stage disease.9 Neoplastic disease was the most common specific diagnosis made for both primary and secondary URFs, and the vast majority of these neoplasms were primary lung carcinoma. Moreover, over two thirds of these lung carcinomas were early stage (ie, stage I or II) disease. The URF pathway was designed as a method to ensure that the follow-up of suspicious lesions occurs, and we believe that the efficiency of evaluation is improved in a large, tertiary care center such as our Veterans Affairs medical center. The pathway was easily interpolated into the practice of our hospital, and feedback from primary care physicians and subspecialists has been positive. This might not be the case in community hospitals, where variable referral lines among physicians define clinical practice. On the other hand, if the advantages (ie, a guarantee of evaluation of URFs and facilitation of follow-up) of the pathway are deemed to be advantageous, medical staffs in community hospitals could consider the use of a similar pathway or other models to facilitate the communication and tracking of URFs.
Our pathway led to the diagnosis of a number of early-stage lung cancers. However, while we were always alert to the possibility that a reported URF might represent an early-stage lung cancer, the pathway was not designed to detect cancers that would otherwise not have been detected. It was designed to ensure that a lung cancer detected early received a prompt evaluation and was not lost to follow-up, only to later reappear at an advanced stage. This is in clear contrast to lung cancer screening programs that are specifically designed to detect early-stage lung cancer that would otherwise not be detected. In our experience, 17% of subjects with URFs were eventually shown to have lung cancer, and 5.7% had stage I lung cancer. Interestingly, these numbers are comparable to lung cancer rates in subjects with abnormal CT scan findings in CT scan-based lung cancer screening programs. In the two largest North American trials, the Early Lung Cancer Action Project study10 and the Mayo Clinic study,11 lung cancer was detected in 2.8% to 11.6% of subjects with abnormal CT scan findings.
The distribution of lung cancer cell types that we detected as URFs in the course of routine medical care was clearly different than that detected by prior studies using either chest radiograph screening or CT scan-based screening. Squamous cell carcinoma was the single most common pathology in our patients, noted in 32% of all cancers (Fig 4). Using chest radiographs for lung cancer screening, the Mayo Lung Project12 found that adenocarcinomas were more frequent than squamous cell carcinoma and that most of the "excess" cancers detected through radiograph screening were adenocarcinoma. In the Mayo Clinic11 and Early Lung Cancer Action Project10 CT scan-based screening studies, adenocarcinoma was detected three times more frequently than was squamous cell carcinoma.
In our analysis of clinical significance, we defined URFs as clinically significant if the pathway made a definitive diagnosis of the finding. URFs that were shown to be stable over time or that required further follow-up were considered to be of indeterminate significance. We reasoned that once a URF was identified an explanation of the finding was required. Using this reasoning, even benign findings (eg, a nodule that turns out to be a nipple shadow on follow-up studies) have significance, albeit sometimes indeterminate. Clearly, a definitive diagnosis of a benign shadow is helpful for the interpretation of future studies. A similar argument can be made for the significance of a finding found to be stable over time (ie, this is helpful in the interpretation of future studies). If URFs resolved during evaluation or were not subsequently confirmed by further studies, we considered them to be clinically insignificant. Using these definitions, over half of both primary and secondary findings were either clinically significant or of indeterminate significance. It is our impression that the pathway also has improved the efficiency in the evaluation of URFs and has improved patient care, although this is very difficult to quantify. The efficiency is due to a rapid communication of the URF to the physician who ordered the radiologic study and the primary care physician, and to a determination of the optimal follow-up or evaluation needed by our pulmonary consultation team in consultation with these physicians. These benefits must be considered in the context of the costs of the pathway in time and money.
Our analysis of costs was based on the assumption that all URFs require evaluation, whether or not a clinical pathway is used in the evaluation. In this context, the only additional features added by our pathway were: (1) the personnel time needed for enlisting patients and for communication to clinicians, (2) the personnel time needed for the collation of data (and data entry) by the pulmonary nurse specialist, and (3) the personnel time of the pulmonary consultation team in formulating a plan for follow-up. Personnel time was the major expense for these activities, so we calculated the dollar costs of the pathway using the salary costs of the nurse specialist and physicians to perform these functions. The costs of patient enlistment, database management, and systems management were included in this calculation. We acknowledge that there are possible hidden costs that were not recognized by our analysis. For example, it could be argued that the pathway led to procedures, or complications of procedures, that might not otherwise have occurred. The evaluation of URFs that eventually were determined to be stable or of no clinical significance also might have led to hidden costs. However, our assumption is that these URFs eventually would have been recognized and would have required evaluation, even in the absence of our URF pathway. The major advantages of our pathway are that URFs do not go unrecognized and the efficiency of evaluation is, hopefully, improved. Our analysis of the costs suggests that they are a reasonable trade-off for the benefits gained.
|
Footnotes
|
|---|
Abbreviation: URF = unsuspected radiographic finding
Received for publication October 22, 2002.
Accepted for publication December 4, 2003.
|
References
|
|---|
- Felson, B (1973) Chest roentgenology. WB Saunders. Philadelphia, PA:
- Ost, D, Fein, A Evaluation and management of the solitary pulmonary nodule. Am J Respir Crit Care Med 2000;162,782-787[Free Full Text]
- Tan, BB, Flaherty, KR, Kazerooni, EA, et al The solitary pulmonary nodule. Chest 2003;123(suppl),89S-96S
- American College of Radiology. ACR standard for communication: diagnostic radiology; standards. 2001,5-6 American College of Radiology. Reston, VA:
- Berlin, L Communication of the urgent finding. AJR Am J Roentgenol 1996;166,513-515[Free Full Text]
- Berlin, L Communication of the significant but not urgent finding. AJR Am J Roentgenol 1997;168,329-331[Free Full Text]
- Kubik, A, Polak, J Lung cancer detection: results of a randomized prospective study in Czechoslovakia. Cancer 1986;57,2427-2437[CrossRef][ISI][Medline]
- White, CS, Romney, BM, Mason, AC, et al Primary carcinoma of the lung overlooked at CT: analysis of findings in 14 patients. Radiology 1996;199,109-115[Abstract/Free Full Text]
- Mountain, CF Revisions in the International System for Staging Lung Cancer. Chest 1997;111,1710-1717[Abstract/Free Full Text]
- Henschke, CI, McCauley, DI, Yankelevitz, DF, et al Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet 1999;354,99-105[CrossRef][ISI][Medline]
- Swensen, SJ, Jett, JR, Hartman, TE, et al Lung cancer screening with CT: Mayo Clinic experience. Radiology 2003;226,756-761[Abstract/Free Full Text]
- Colby, TV, Tazelaar, HD, Travis, WD, et al Pathologic review of the Mayo Lung Project cancers: is there a case for misdiagnosis or overdiagnosis of lung carcinoma in the screened group? Cancer 2002;95,2361-2365[CrossRef][ISI][Medline]
TOP
Abstract
Introduction
