Initial Experience With a Free, High-Volume, Low-Dose CT Lung Cancer Screening Program

The National Lung Screening Trial demonstrated a significant mortality benefit for patients at high risk for lung cancer undergoing serial low-dose CT. Currently, the National Comprehensive Cancer Network and several United States–based professional associations recommend CT Lung screening for high-risk patients. In the absence of established reimbursement, the authors modeled and implemented a free low-dose CT lung cancer screening program to provide equitable access to all eligible patients. Elements of the program reported in this article include a decentralized referral network, centralized program coordination, structured reporting, and a patient data management system. The experience and initial results observed in this clinical setting closely match the performance metrics of the National Lung Screening Trial with regard to cancer detection and incidental findings rates. To eliminate health care disparities a vigorous lobbying effort will be needed to expedite reimbursement and make CT lung screening equally available to all patients at high-risk. The National Lung Screening Trial demonstrated a significant mortality benefit for patients at high risk for lung cancer undergoing serial low-dose CT. Currently, the National Comprehensive Cancer Network and several United States–based professional associations recommend CT Lung screening for high-risk patients. In the absence of established reimbursement, the authors modeled and implemented a free low-dose CT lung cancer screening program to provide equitable access to all eligible patients. Elements of the program reported in this article include a decentralized referral network, centralized program coordination, structured reporting, and a patient data management system. The experience and initial results observed in this clinical setting closely match the performance metrics of the National Lung Screening Trial with regard to cancer detection and incidental findings rates. To eliminate health care disparities a vigorous lobbying effort will be needed to expedite reimbursement and make CT lung screening equally available to all patients at high-risk. Lung cancer causes more deaths among men and women in the United States than breast, colorectal, and prostate cancers combined, with approximately 450 people dying from lung cancer every day [1National Cancer InstituteGeneral information about non-small cell lung cancer.http://www.cancer.gov/cancertopics/pdq/treatment/non-small-cell-lung/healthprofessionalGoogle Scholar]. Despite continuing advancements in surgery, radiation, and chemotherapy, lung cancer remains a highly lethal disease, with 16% overall 5-year survival only marginally improved from 12% in the 1970s [2National Cancer InstituteSurveillance Epidemiology and End Results.http://seer.cancer.gov/faststats/selections.php? - OutputGoogle Scholar]. Although primary prevention (smoking cessation) has saved countless lives by decreasing the rate of smoking from >40% in 1965 to <20% today [3Gallup OrganizationOne in five U.S. adults smoke, tied for all-time low.http://www.gallup.com/poll/156833/one-five-adults-smoke-tied-time-low.aspxGoogle Scholar], many former heavy smokers remain at high risk and now represent the largest group of patients diagnosed with lung cancer [4Ebbert J.O. Yang P. Vachon C.M. et al.Lung cancer risk reduction after smoking cessation: observations from a prospective cohort of women.J Clin Oncol. 2003; 21: 921-926Crossref PubMed Scopus (150) Google Scholar]. In fact, given the large number of former heavy smokers in the aging baby boom population, lung cancer mortality seems poised to rise in the absence of effective secondary prevention (screening) [5United States age distributionhttp://www.nationmaster.com/country/us/Age_distributionGoogle Scholar]. In 2011, the National Lung Screening Trial (NLST) reported a 20% lung cancer–specific mortality benefit in high-risk current and former heavy smokers who underwent 3 rounds of annual low-dose CT (LDCT) lung screening compared with annual chest radiography [6NLST Trial TeamReduced lung-cancer mortality with low-dose computed tomographic screening.N Engl J Med. 2011; 365: 395-409Crossref PubMed Scopus (6467) Google Scholar]. Shortly after publication of the results of this large, National Cancer Institute–sponsored, randomized controlled trial, the National Comprehensive Cancer Network (NCCN) released guidelines recommending annual LDCT lung screening for two specific groups of high-risk individuals meeting stringent criteria [7National Comprehensive Cancer NetworkNCCN guidelines for lung cancer screening (V.1.2012).http://www.nccn.org/professionals/physician_gls/pdf/lung_screening.pdfGoogle Scholar]. Group 1 (NLST population, NCCN category 1 recommendation—uniform consensus based on high-level evidence): •55-74 years old•At least 30 pack-year smoking history•Current or former smokers (quit within past 15 years) Group 2 (NCCN category 2B recommendation—consensus based on lower level evidence): •>50 years old•>20 pack year smoking history•Current or former smokers (quit for any length of time)•One additional lung cancer risk factor Lahey Hospital & Medical Center subsequently tasked a multidisciplinary steering committee to design and implement an NCCN Guidelines®-based CT lung cancer screening program to commence January 9, 2012. The committee included representatives from radiology, pulmonology, radiation oncology, medical oncology, internal medicine, administration, finance, philanthropy, business development, and marketing. In this article, we report the critical elements of our lung cancer screening program and our initial results, and we conclude by highlighting points of discussion about lung screening in general. The most important and vexing decision we made during the conception of our screening program was whether to charge for the screening examinations. Most public and private payers, including CMS, currently do not reimburse for CT lung screening, with a few notable exceptions [8AnthemComputed tomography scans with or without computer aided detection (CAD) for lung cancer screening.http://www.anthem.com/ca/medicalpolicies/policies/mp_pw_a053266.htmGoogle Scholar, 9BlueCross BlueShield of MontanaScreening for lung cancer using CT scanning or chest radiographs.https://www.bcbsmt.com/MedReview/Policies/ScreeningLungCancer/v101.aspxGoogle Scholar, 10Empire BlueCross BlueShieldComputed tomography scans with or without computer assisted detection (CAD) for lung cancer screening.http://www.empireblue.com/medicalpolicies/policies/mp_pw_a053266.htmGoogle Scholar, 11Matthews A. Wellpoint to cover lung CT scans for heavy smokers. The Wall Street Journal.http://blogs.wsj.com/health/2011/12/01/wellpoint-to-cover-lung-ct-scans-for-heavy-smokers/Google Scholar]. Self-pay rates for CT lung screening (ranging from $99 to $1,000) create access disparities among high-risk individuals of varying financial means [12The Advisory Board CompanyStarting a CT lung cancer screening program.http://www.advisory.com/Research/Oncology-Roundtable/Oncology-Rounds/2012/07/Starting-a-CT-lung-cancer-screening-programGoogle Scholar]. Out-of-pocket costs also discourage asymptomatic high-risk patients from undergoing recommended screening examinations. These potent economic and psychological barriers may in large part explain the commonly reported low rates of enrollment in existing self-pay LDCT programs. Low-volume screening is potentially ineffective, as 320 individuals need to be screened to save 1 life according to the NLST [6NLST Trial TeamReduced lung-cancer mortality with low-dose computed tomographic screening.N Engl J Med. 2011; 365: 395-409Crossref PubMed Scopus (6467) Google Scholar]. To fulfill what we felt was an ethical responsibility to provide equal screening access to all persons at high risk regardless of socioeconomic status and to encourage persons at high risk to present for screening, we decided to offer CT lung screening at no cost to patients until CMS and commercial insurers establish reimbursement on a broad scale. We feel this approach is consistent with the philosophy of the Patient Protection and Affordable Care Act, which seeks to eliminate health care disparities and barriers to preventative services [13US Department of Health and Human ServicesHealth disparities and the Affordable Care Act.http://www.healthcare.gov/news/factsheets/2010/07/health-disparities.htmlGoogle Scholar]. These considerations are at the core of our Rescue Lung, Rescue Life movement [14Rescue Lung, Rescue Life.https://www.facebook.com/pages/Rescue-Lung-Rescue-Life/213549102061234Google Scholar]. Our institutional compliance department required that the program fulfill several conditions to be permitted to offer free CT lung screening. Objective patient qualification criteria needed to be established that would be followed without exception. No participant could be billed, not even those with insurance providing coverage for CT lung screening. Finally, at termination of the free offering, it must be stopped for all participants indiscriminately. Detailed business modeling of the program was performed before its inception and presented to senior management for ultimate program approval. The business model requires the availability of downtime on installed base CT scanners. Our PET/CT scanner typically is idle early in the morning, between the injection of radiotracer and scanning of our first patient. During this downtime, the PET/CT scanner can accommodate 5 CT lung screening examinations (25 appointments/week). An additional 10 appointments/week are available late in the day, when outpatient activity at the institution decreases and both technologist staff members and CT scanner capacity become available. To fully serve our patient population, we predicted that we would eventually need approximately 100 to 120 lung screening appointments/week and therefore estimated the cost of adding 1 dedicated 40-hour overnight and weekend shift (1 technologist and 2 technologist aides), which could accommodate as many as 200 additional screening slots per week. We assumed that overnight and weekend scan times would be acceptable to patients, considering the potential benefit of this examination performed at no cost. Although we do not charge for the initial or annual follow-up screening examinations, workup of any positive findings requiring downstream diagnostic CT examinations, clinical assessment, or intervention is billed to the patient's insurance in the customary fashion and provides the revenue stream that supports the program. It should be noted that a universal health insurance mandate is in effect in the state of Massachusetts [15Attorney General of MassachusettsMandatory health insurance.http://www.mass.gov/ago/doing-business-in-massachusetts/health-care/health-insurance-mandate.htmlGoogle Scholar]. We used inputs from both the NLST data and NCCN Guidelines® to assess the potential financial impact of offering free lung screening. The resulting interactive, spreadsheet-based tool set created for this purpose allows the operator to perform a sensitivity analysis on the financial impact of multiple variables, including the percentage of examinations with positive results, the percentage of examinations with nodules 4 to 8 mm, the expected rate of uninsured individuals, radiologist and technologist cost, patient retention rate, annual patient enrollment, and cancer treatment revenue. Of note, at very high scanning volumes (>6,000 screenings/year) >1 full-time equivalent radiologist may be required to interpret the resulting examinations. By modeling our own situation, we estimated that in the first 2 years of our screening program, 60% to 80% of the revenue available to offset the cost of free screening would be derived from treating lung cancer. In years 3 to 10, the revenue derived from interval diagnostic LDCT follow-up of small pulmonary nodules and lung cancer treatment become equally important revenue sources. These financial models are included with many other documents on our lung screening program information CD-ROM, which is available at no cost to interested centers upon request. Lahey Hospital & Medical Center did not participate in NLST, the International Early Lung Cancer Action Program, or any other CT lung screening trial [16Henschke C.I. Yankelevitz D.F. et al.International Early Lung Cancer Action Program InvestigatorsSurvival of patients with stage I lung cancer detected on CT screening.N Engl J Med. 2006; 355: 1763-1771Crossref PubMed Scopus (1435) Google Scholar]. Therefore, we needed to build the comprehensive lung cancer screening program infrastructure from the ground up. Accordingly, a lung cancer screening program coordinator and a patient navigator position were established. A radiology department working group consisting of a radiologist, an administrative director, CT, PET, and scheduling team leaders, and a PACS software engineer created the program elements listed in Table 1.Table 1CT lung screening program elementsItemPurposeToll-free number (855-CT-CHEST)Central acceptance and routing of program-directed patient inquiriesIntake formsUsed by general radiology schedulers to qualify patients and stratify them into one of the two NCCN high-risk groupsCall center scriptExplains to callers the importance of being asymptomatic at the time of screening and directs inquiries of those not meeting criteria for screeningFAQ documentExplains what to expect before, during, and after screening; gives the benefits and risks of screening; and provides information on smoking cessation at patients' levels of understandingCustom database applicationPulls patient-specific data from the RIS to facilitate and manage patient intake, scheduling, and follow-upPatient letter libraryResults-specific, standardized patient notification lettersProgram literaturePhysician-directed program information literatureScanning protocolsLow-dose lung cancer screening scanning protocolsNote: FAQ = frequently asked questions; NCCN = National Comprehensive Cancer Network; RIS = radiology information system. Open table in a new tab Note: FAQ = frequently asked questions; NCCN = National Comprehensive Cancer Network; RIS = radiology information system. We created a standardized CT lung screening reporting system (LungRADS) modeled on BI-RADS® [17American College of RadiologyBI-RADS Atlas.http://www.acr.org/Quality-Safety/Resources/BIRADSGoogle Scholar]. In addition to diagnostic categories for lung nodules, the system includes a binary modifier (category S) to address the occurrence of clinically significant incidental findings observed on LDCT lung screening examinations (Fig. 1). LungRADS also incorporates an NCCN guidelines–based nodule lexicon, a structured reporting format, and a nodule size range methodology that reports mean size in 1-mm ranges (ie, 4-5 mm, 7-8 mm) rather than a discrete number (ie, 4, 4.3, or 5.6 mm). This approach respects the spatial resolution limitation of LDCT, decreases interobserver and intraobserver variability, and avoids guideline “trigger” points (4, 6, and 8 mm for solid nodules), which could result in confusion with regard to the appropriate follow-up interval (should a 6-mm nodule undergo 3-month or 6-month follow-up?). We also anticipate that reporting size ranges will reduce the need to explain to understandably concerned patients and physicians why a hypothetical 4.7-mm nodule which previously measured 4.3 mm has not definitively grown. LungRADS addresses multiple needs that should be common to all LDCT lung screening programs: •Facilitate adherence of radiologist recommendations to screening guidelines (in our specific case, NCCN Guidelines).•Facilitate structured, focused training of interpreting radiologists before credentialing.•Facilitate the communication of examination results among the various involved health care providers.•Allow periodic quality review to assess guideline adherence and ongoing evaluation of performance metrics such as positive screening rates.•Facilitate structured database storage and tracking of findings.•Automatically generate results-specific patient notification letters.•Facilitate appropriate follow-up examination coding.•Triage risk categories within the screened population and mechanize referral of the small number of patients (3%-4%) with suspicious findings (LungRADS category 4) to multidisciplinary team management. On the basis of our initial experience, we suggest that there is a significant opportunity for the ACR to convene experts in the field to develop a practice guideline for LDCT lung screening. In that context, we suggest that LungRADS or a similar structure reporting system would be a key component of such a guideline, providing common language to communicate lung screening examination results across providers and institutions. Primary care physicians (PCPs) are best positioned to help patients decide on the appropriateness of preventative care interventions such as lung cancer screening. They have knowledge of their patients' overall health and share the downstream responsibility of managing examination findings. For these reasons, we require PCP orders on all patients before screening. The PCP provider base represents an established and experienced decentralized preventive care network essential to operating a low-cost, high-volume screening program. We advocate that program patient navigator time is best leveraged by organizing community information seminars and advising individual patients and physicians on an ad hoc basis throughout the screening We an continuing medical with the elements and to effective with screening the common that the current absence of established reimbursement means that CT lung screening is not recommended or of benefit Lung on CT lung cancer Scholar, of from the of and the College of on the review and clinical practice guideline on the of CT screening for lung cancer by the Scholar, et and of CT screening for lung a Google Scholar, et for guidelines for lung cancer screening using low-dose computed tomography scans for lung cancer and other high-risk PubMed Scopus Google Scholar, et Cancer lung cancer screening Cancer J Scopus Google and all and findings in a similar fashion to screening multidisciplinary team of physicians available at Lahey Hospital & Medical Center to provide management recommendations and treatment for all patients with LungRADS category 4 suspicious only about of a provider patient will qualify for screening. Our business team at and PCP to these rounds to about the workup and treatment of lung cancer. a result of this continuing medical patients are now to the program our PCP network, and we have no costs for To our in the we patient a of frequently asked the risks and benefits of lung cancer screening and providing for smoking patient is by 1 or 2 before screening and within 3 after screening by a patient navigator to the screening or follow-up and a patient An of the is in Our program on January 9, and we a of screened patients on on the reported data are In the first few of our program, of patients about screening did not group of NCCN high-risk and of patients only the NCCN group 2 the rate of patients not to an in of the high-risk criteria by both patients and are presented in Table Although the age of patients in our program is similar to that in the NLST, our overall age is more of the general population than the trial population in NLST, in which patients to years were [6NLST Trial TeamReduced lung-cancer mortality with low-dose computed tomographic screening.N Engl J Med. 2011; 365: 395-409Crossref PubMed Scopus (6467) Google age NLST = National Lung Screening Open table in a new tab Note: NLST = National Lung Screening of screening examinations positive results including those suspicious for similar to findings reported in the NLST NLST = National Lung Screening = not NLST = National Lung Screening = not = NLST = National Lung Screening = not Open table in a new tab Note: The rate of clinically significant incidental findings in our program was compared with during the first year of the This may be by the that approximately of patients screened of the chest or available for which these findings as not clinically patients in our program were diagnosed with lung and patients are currently being followed for pulmonary nodules Cancer in Open table in a new tab No were analysis of these initial patients is LDCT screening program must be to the benefit and the required elements of such a program include a screening a standardized reporting system and dedicated to manage appropriate patient as as communication of positive findings. a program were implemented in the United CT lung screening has the potential to save lives every et number of lung cancer deaths potentially by screening in the United PubMed Scopus Google Scholar]. to this across the of health care providers in this most of have no experience in lung screening, of the of screening is Our own institutional experience that findings are and that of the lung cancer screening into clinical practice is The ACR is positioned to develop a comprehensive practice guideline on lung cancer screening to of such programs. The current of established reimbursement a to lung cancer screening. access to this screening, free of economic is by of a screening program free of charge to The to this approach of a free screening program will according to situation, including available CT scanner and business and patient screening requires a decentralized patient referral network and to the small number of suspicious findings for evaluation with PCP is critical to build a program, given the for and effective lung cancer screening a dedicated and is A vigorous lobbying effort is needed to expedite reimbursement to eliminate the health care disparities this and make CT lung screening equally available to all patients at high Lung cancer screening with LDCT is an of the radiologists can to patient and effective CT lung screening in the absence of established reimbursement requires several key for equitable patient decentralized and referral dedicated patient management between multiple medical eligible patient