Western medicine is undergoing a transition toward transparency of quality and costs, and healthcare systems are striving to achieve the Triple Aim, a framework for improving the patient experience of care, improving the health of populations, and reducing the per capita cost of healthcare. Meanwhile, there is growing recognition of the impact of social determinants of health and a new federal

requirement for nonprofit hospitals to implement prevention strategies. A specialized meal delivery program called Simply Delivered for ME (SDM) was formed in an effort to improve care and reduce 30-day hospital readmission rates. 

The Maine Medical Center (MMC) partnered with the Southern Maine Agency on Aging to offer SDM on a voluntary basis to high-risk Medicare patients already enrolled in the Community-based Care Transition Program (CCTP) at MMC. We report the results of the 2-year intervention in terms of 30-day hospital readmission rates and cost measures (ie, return on investment and cost savings). 

Of the 622 MMC patients who received SDM during the 24 months, the 30-day readmission rate was 10.3% (compared with the 16.6% 30-day rate of hospital readmission at baseline [ie, before the adoption of CCTP]) for all-cause readmissions. The cost savings for reduced readmissions were $212,160. The return on investment was 387%, or a benefit-cost ratio of $3.87 for every $1.00 spent on meals. Programs such as SDM may reduce the rate of hospital readmission among high-risk older adults and, thereby, yield lower healthcare costs. 

Social determinants of health matter across the age spectrum, from birth to old age, and across disease progression, from prevention to tertiary treatment.

We examined the impact of a care transition intervention with or without the addition of a meal delivery program titled Simply Delivered for ME (SDM) upon hospital discharge and its effect on 30-day readmission rates. More than 600 high-risk elderly patients received SDM over a 24-month period, which was associated with a 38% decreased rate of 30-day readmissions (10.3% vs 16.6% at baseline).

The associated cost savings for these lower readmissions were estimated to be $212,160, or a benefit-cost ratio of $3.87 for every $1.00 spent on meals.

Western medicine is beginning to recognize what social scientists have known for years: There are social determinants of health. For too long, we have focused on treatment of disease, instead of prevention, and not comprehensively enough to reduce the disease burden. Social determinants are consistently and prominently noted by the World Health Organization and, more recently, by our nation’s health protection agency, the CDC, along with several other agencies, as summarized in a recent scan of large-scale initiatives.

 is hardly unknown, yet today, in much of the world and in pockets of this country, people’s physiological and safety needs are not being met. The United States dedicates far fewer dollars per person on social services than does almost every other nation within the Organisation for Economic Co-operation and Development.

 The Economist Intelligence Unit released a white paper on the healthcare systems of 166 nations that ranked the United States highest in healthcare spending and 33rd in quality outcomes, making it a “poor-value” healthcare system.

 As our healthcare system is striving toward the Triple Aim of improving population health, improving the patient experience of care, and reducing per capita cost,

 it is time to embrace a new paradigm to achieve these aims.

A recent study found that states with a higher ratio of social spending to health spending had significantly better health outcomes for 7 measures: adult obesity, asthma, mentally unhealthy days, days with activity limitations, and mortality rates for lung cancer, acute myocardial infarction, and type 2 diabetes.

 In countries where collaborations happen across sectors, successful programs exist, such as the Integrated Care Pathways for Airway Diseases.

 The Affordable Care Act has a new requirement for nonprofit hospitals: conducting a community health needs assessment every 3 years and using the results to launch an implementation strategy.

 This affords the opportunity to expand collaborations with social services agencies. Evidence already exists to support these collaborations in terms of the positive health outcomes and reduced healthcare spending they generate.

 Cost savings are often realized by decreased rates of readmission, especially in the geriatric population.

Simply Delivered for ME (SDM) was a collaboration between a hospital and a social service agency: The Maine Medical Center (MMC) partnered with the Southern Maine Agency on Aging (SMAA) to offer a Community-based Care Transition Program (CCTP) with an optional add-on program, SDM. SDM offered specialized meals (eg, vegetarian, pureed) to patients after they were discharged from the hospital; caregivers were also allowed to participate. Up to a 7-day meal supply was delivered weekly to participants’ doorsteps. The purpose of this report is to examine the potential impact of the CCTP on the readmission rates of high-risk Medicare patients and the potentially additive effect of delivering meals to these individuals post transition from the hospital to the home or an alternative setting, such as long-term care.

The study population of patients from MMC were on Medicare and designated as at high risk of readmission according to CMS (ie, CMS Hierarchical Condition Category [HCC] score >1.6). We used a time-series design with rolling enrollment over a 2-year period (July 2013-July 2015).

In 2012, SMAA was selected by CMS for participation in the CCTP. SMAA used as the basis for their program the Coleman Model of Care Transition Intervention (CTI), a 4-week process that focused on patient engagement post discharge by coaching individuals to take more active roles in their healthcare. Patients enrolled in CTI received specific tools and skills training reinforced by a transition coach (ie, a nurse or social worker) who received extensive training in motivational interviewing and teach-back methods. The transition coaches were responsible for reinforcing the following components: medication reconciliation and self-management, use of a patient-centered health record to help guide patients through the care process, primary care provider and specialist follow-up, and patient understanding of red-flag indicators of worsening condition and appropriate next steps. Within the post—hospital discharge setting, the goals of the CCTP were to improve quality of care, reduce readmissions for high-risk beneficiaries, and document measurable savings to the Medicare program. During the 4 weeks of the transition process, referrals were made to other social service agencies as deemed appropriate by the visiting coach. To enhance the care transition process, SMAA introduced meals post discharge, the SDM program, which was offered on a voluntary basis. SDM is similar to Meals on Wheels, but distinguishable by the following: (1) it was marketed to patients as part of the CCTP at no cost, (2) it was titled Simply Delivered for ME and delivered in a labeled shopping bag, (3) it consisted of a weekly delivery of frozen meals, (4) it included a 7-meal supply delivered within 4 days of discharge, and (5) no milk or bread was provided.

The patients were in either the CCTP-only group or the CCTP-plus-SDM group; the 30-day hospital readmission rate was calculated by group status. A hospital staff member calculated that rate, by group, and compared those rates with the baseline rate (the year prior to adoption of CCTP for patients with HCC >1.6). Statistical significance was determined by nonoverlapping 95% CIs. Cost savings and return on investment were calculated based on the cost of providing the meals and on the 30-day readmission rate, assuming an average cost per readmission

of $16,320 per high-risk patient, as established at baseline. Although the baseline costs may seem high, they are the true costs experienced by MMC for the year preceding the adoption of CTPP and were the impetus for adoption of the new project.

The project served 1745 people, 1177 patients and 568 caregivers, in southern Maine between July 2013 and June 2015. Data from MMC included 622 patients (data from settings other than MMC are not available) whose characteristics are displayed in the 

. The mean age was 71.7 years, and 56.6% were female. Among the 622 MMC discharges who received SDM during the 24 months, the 30-day readmission rate was 10.3% (95% CI, 8.1%-12.9%; n = 64 readmissions). This is 16.3% lower than MMC’s CCTP 30-day readmission rate of 12.3% (95% CI, 10.0%-15.2%; n = 77 readmissions) since the beginning of the CCTP in 2012, and a 38.0% improvement over MMC’s baseline (pre-CCTP) 30-day all-cause readmission rate of 16.6% (95% CI, 13.8%-19.7%; n = 103 readmissions). The cost savings for the 13 fewer readmissions with the addition of SDM to CCTP were $212,160. The cost of providing 7 days of meals to the 622 discharges was $43,540. Thus, the return on investment was 387%, or a benefit-cost ratio of $3.87 for every $1.00 spent.

Based on a literature review (2004-2014), Taylor and colleagues

 summarized the impact of social service investments on health outcomes and healthcare costs. More than 80% of these studies reported positive effects on either health outcomes (62.5%) or costs (15.6%) or both (22.0%). The authors stress that additional research addressing both health and cost outcomes and evidence of cost savings will be essential for more widespread endorsement. We were able to examine a proxy for health status (30-day hospital readmissions) and to calculate costs, providing further evidence for both.

In our analyses, the 30-day readmission rate for those receiving coordinated care transition and SDM was 38% lower than baseline, although the difference was not statistically significant. Being enrolled only in CCTP (vs CCTP plus SDM) was also associated with a reduction in the readmission rate. Indeed, we need comprehensive approaches for mitigating “posthospital syndrome,” a condition described by Krumholz,

 which suggests that these approaches need to target stressors that contribute to the high baseline rate of readmissions. Having a resource, such as a CCTP transition coach or delivered meals, can alleviate many of the likely stressors that patients face once they return home.

A synthesis of 9 studies of coordinated care found that 4 of the coordinated care interventions showed decreased healthcare costs and 2 others had positive health outcomes. The 2 that included outreach (eg, home-delivered meals) showed both lower costs and improved outcomes.

 The authors also synthesized 11 nutritional support studies and found that 7 showed improved health outcomes, but none showed decreased healthcare costs. In contrast, our analyses showed an association with both cost savings and positive health outcomes.

SDM is similar to Meals on Wheels, which has demonstrated a wide array of beneficial effects.

 reviewed 8 studies on home delivery of meals to older adults and found that the majority reported significant improvement in diet quality and nutrient intake, reduced food insecurity and nutritional risk, and other benefits, such as increased socialization and higher quality of life. The authors concluded that home-delivered meals help older adults maintain independence and remain in their homes.

The coordinated care transition intervention in our report can be compared with the Bridge Model,

 which combines care coordination, case management, and patient engagement. The model, like the CCTP, is designed to provide a seamless transition and improve the overall quality of transitional care for older adults. The authors of the Bridge Model study believe that the emphasis on value and quality support further development and expansion of transitional care strategies, which offer promising avenues to fulfill the Triple Aim while also impacting population health and controlling per capita costs.

 noted barriers to widespread adoption of these collaborations across disciplines (eg, there are no medical codes to bill for social services and collaborations are challenged by various care delivery models, organizational structures, and financial contracting systems). Indeed, in the present report, external grant funding was an essential element, without which the SDM program ceased.

Our recommendations are to: (1) establish a plan for patients discharged to skilled nursing facilities before returning home; (2) ensure that dietary restrictions are identified by the transition coach and confirmed by site staff before the meals are delivered (eg, food allergies or need for pureed, vegetarian, or renal-sensitive meals); (3) brand SDM, to distinguish the program from Meals on Wheels, which some perceived negatively as a program for a frail and low-income population; and (4) consider the effects of home-delivered meals on caregivers.

One limitation of our report was that the analyses were limited to just 1 hospital with 622 patients, although there were more than 1000 participants. Although we cannot reliably generalize our findings to other hospitals, we believe that the sample size and 2-year duration are sufficient to suggest potentially beneficial effects. We do not have access to the data for further analyses to explore the potential bias of our sample that volunteered for SDM. Our cost estimates are valid for this hospital but, again, perhaps not generalizable, as baseline costs for the high-risk patients were quite high ($16,320 per patient).

Despite these limitations, we observed an association with positive outcomes for adopting coordinated care at transition, and even more so when accompanied by home-delivered meals. Our findings also suggest that it can be cost-saving to the healthcare system involved. We hope that, moving forward, barriers to collaboration will be overcome and that social determinants of health can be addressed as common practice.&ensp;

The authors would like to acknowledge Rocket Wong, formerly of MMC (Portland, Maine), for the original analyses of these data and for the abstraction of the deidentified MMC data. They would also like to acknowledge Dan Knox of SMAA for his reanalysis of the data for this manuscript.

 Husson University School of Pharmacy (SLM, CP), Bangor, ME; Southern Maine Agency on Aging (NC, KB), Scarborough, ME.

 The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (NC, KB); acquisition of data (NC); analysis and interpretation of data (SLM); drafting of the manuscript (SLM, CP); critical revision of the manuscript for important intellectual content (SLM, NC, CP, KB); administrative, technical, or logistic support (NC, CP); and supervision (KB).

 Nancy Connelly, MBA, Southern Maine Agency on Aging, 136 US Rte 1, Scarborough, ME 04074. Email: nconnelly@smaaa.org.

1. Koo D, O’Carroll PW, Harris A, DeSalvo KB. An environmental scan of recent initiatives incorporating social determinants in public health. 

. 2016;13:E86. doi: 10.5888/pcd13.160248.

3. Bradley EH, Elkins BR, Herrin J, Elbel B. Health and social services expenditures: associations with health outcomes. 

. 2011;20(10):826-831. doi: 10.1136/bmjqs.2010.048363.

4. Health outcomes and cost: a 166-country comparison. The Economist Intelligence Unit website. eiu.com/public/topical_report.aspx?campaignid=Healthoutcome2014. Published 2014. Accessed September 21, 2016.

5. Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, and cost. 

. 2008;27(3):759-769. doi: 10.1377/hlthaff.27.3.759.

A Guide to Measuring the Triple Aim: Population Health, Experience of Care, and Per Capita Cost

. Cambridge, MA: Institute for Health Care Improvement; 2012.

7. Taylor LA, Tan AX, Coyle CE, et al. Leveraging the social determinants of health: what works? 

. 2016;11(8):e0160217. doi: 10.1371/journal.pone.0160217.

8. Bousquest J, Addis A, Adcock I, et al; European Innovation Partnership on Active and Healthy Ageing, Action Plan B3; Mechanisms of the Development of Allergy, WP 10; Global Alliance against Chronic Respiratory Diseases. Integrated care pathways for airway diseases (AIRWAYS-ICPs). 

. 2014;44(2):304-323. doi: 10.1183/09031936.00014614.

9. New requirement for 501(c)(3) hospitals under the Affordable Care Act. Internal Revenue Service website. irs.gov/charities-non-profits/charitable-organizations/new-requirements-for-501c3-hospitals-under-the-affordable-care-act. Updated August 27, 2017. Accessed September 16, 2016.

10. Deschodt M, Devriendt E, Sabbe M, et al. Characteristics of older adults admitted to the emergency department (ED) and their risk factors for ED readmission based on comprehensive geriatric assessment: a prospective cohort study. 

. 2015;15:54. doi: 10.1186/s12877-015-0055-7.

11. Leppin AL, Gionfriddo MR, Kessler M, et al. Preventing 30-day hospital readmissions: a systematic review and meta-analysis of randomized trials. 

. 2014;174(7):1095-1107. doi: 10.1001/jamainternmed.2014.1608.

12. Krumholz HM. Post-hospital syndrome—an acquired, transient condition of generalized risk. 

. 2013;368(2):100-102. doi: 10.1056/NEJMp1212324.

13. Campbell AD, Godfryd A, Buys DR, Locher JL. Does participation in home-delivered meals programs improve outcomes for older adults? results of a systematic review. 

. 2015;34(2):124-167. doi: 10.1080/21551197.2015.1038463.

14. Zhu H, An R. Impact of home-delivered meal programs on diet and nutrition among older adults: a review. 

. 2013;22(2):89-103. doi: 10.1177/0260106014537146.

15. Alvarez R, Ginsburg J, Grabowski J, Post S, Rosenberg W. The social work role in reducing 30-day readmissions: the effectiveness of the Bridge Model of transitional care. 

. 2016;59(3):222-227. doi: 10.1080/01634372.2016.1195781.

16. Gottlieb LM, Garcia K, Wing H, Manchanda R. Clinical interventions addressing nonmedical health determinants in Medicaid managed care. 

. 2016;22(5):370-376. doi: 10.1007/s40471-014-0031-3.

Articles

TRENDS FROM THE FIELD

Offering a meal delivery service to patients in a community-based care transition program was associated with cost savings and a 38% lower 30-day hospital readmission rate.

Simply Delivered Meals: A Tale of Collaboration

 To describe participation rates, results, and lessons learned from a lung cancer screening (LCS) demonstration project.

 Prospective observational study at 1 of 8 centers participating in a national Veterans Health Administration LCS demonstration project.

 An electronic health record (EHR) algorithm and tobacco pack-year (TPY) information prompt identified patients potentially eligible for LCS. LCS invitation was planned to consist of shared decision-making materials, an invitation letter to call the LCS manager, a reminder letter, and an outreach phone call for nonresponders. The outreach call was subsequently dropped due to time constraints on the LCS manager. Lung nodules and incidental findings on LCS low-dose computed tomography (LDCT) were recorded in templated radiology reports and tracked with EHR notes.

 Of 6133 potentially eligible patients, we identified 1388 patients with eligible TPY information: 918 were invited for LCS and 178 (19%) completed LCS. LCS completion was more likely in patients in the mailing-plus-call outreach group (phase I) compared with the mail-only group (phase II) (22% vs 9%; 

<.001). Among those completing an LDCT, 61% had lung nodules requiring follow-up: 43% of the nodules were less than 4 mm in diameter, 12 patients required further diagnostic evaluation, and 2 had lung malignancies. There were 179 incidental LDCT findings in 116 patients, and 20% were clinically significant.

 Important considerations in LCS are accurate identification of eligible patients, balancing invitation approaches with resource constraints, and establishing standardized methods for tracking numerous small lung nodules and incidental findings detected by LDCT.

Lung cancer screening (LCS) is a complex process that is best supported by a managed system approach in order to accurately identify eligible patients, provide consistent shared decision making (SDM), ensure standardized low-dose computed tomography interpretation, and track results over time.

Optimal approaches to patient invitation for LCS and SDM are unclear.

A better understanding of the clinical significance of small (<4 mm diameter) lung nodules and incidental findings is needed.

Lung cancer is the leading cause of cancer morbidity and mortality in the United States.

 Given the high disease burden and aggressive nature of lung cancer, considerable effort has been directed at early detection and treatment through lung cancer screening (LCS) trials. The National Lung Screening Trial (NLST) reported a 20% relative reduction in lung cancer mortality with low-dose computed tomography (LDCT) compared with chest x-rays.

 The primary principle of LCS is detection and surveillance of small lung nodules over time for changes that are suspicious for malignancy.

The NLST served as the primary basis for the recent United States Preventive Services Task Force recommendation for annual LCS with LDCT for high-risk individuals.

 Several professional societies have also endorsed annual LDCT for high-risk individuals,

 and in 2015, CMS added LCS as a reimbursable preventive service.

 Despite the benefits of LDCT for LCS on both lung cancer and all-cause mortality, there are concerns about high costs and potential associated harms, including false-positive results,

 and psychological distress, particularly for patients who receive an indeterminate result.

To gain information about the feasibility of implementing these recommendations, the Veterans Health Administration (VHA) completed a National Demonstration Project. The Minneapolis Veterans Affairs Health Care System (MVAHCS) was 1 of 8 demonstration sites. We report the initial results of LCS at the MVAHCS to provide more detailed information than was collected in the National VHA Demonstration Project regarding tobacco pack-year (TPY) information, patient uptake rates of LCS in response to different invitation approaches, characteristics of lung nodules detected on LDCT, and the clinical significance of incidental findings on LDCTs that are unrelated to LCS.

Initial LCS results at the MVAHCS between January 1, 2014, and May 22, 2015, were analyzed. We employed a national VHA electronic health record (EHR) algorithmic program to identify potential eligible patients who met the preliminary LCS criteria at the time of an appointment with their primary care provider (PCP). The criteria were: being aged 55 to 80 years; having no diagnosis codes in the EHR for hospice care or lung, esophageal, pancreatic, or liver cancer; having no chest CT in the previous year; and not being previously coded in the EHR as not expected to live more than 6 months. If a patient met all of these criteria, the algorithm activated an EHR prompt for the appointment check-in nurse to collect TPY information (ie, current cigarette smoking status, years smoked, and average lifetime packs per day).

To provide equitable access and prevent exceeding the initial screening capacity for LCS, the National VHA Demonstration Project recommended gradual local implementation by a random rolling activation of EHR prompts per individual PCPs. At our site, providers opted out of this approach because of an existing program of a manager-driven lung nodule tracking system. We therefore elected to randomly choose patients with eligible TPY information (≥30 pack-years and either currently smoking or quit <15 years ago) for invitation to LCS using a 2:1 ratio (2 patients selected for LCS invitation for every 1 usual care patient), by the LCS program manager, an MPH with training in health education and extensive clinical experience. The MVAHCS Internal Review Board determined that patient-level randomization for invitation to

the Demonstration Project was not considered research, but rather a quality improvement and feasibility evaluation method. Randomization was implemented in blocks, with groups assigned using random number generator—based software algorithms.

All MVAHCS patients had access to a comprehensive Tobacco Cessation Program, which includes assessment of smoking status via an annual EHR prompt, patient education materials, individual and group behavioral therapy, and pharmacotherapy.

Eligible patients were mailed shared decision-making (SDM) materials developed by the VHA National Center for Health Promotion and Disease Prevention

]) and a letter inviting them to call the LCS program to discuss LCS and schedule an LDCT. A second letter was mailed a week later. At the outset of the local demonstration project, the program manager attempted to complete phone contact with all invitation nonrespondents. During an LCS call, the patient and program manager would review the SDM material and discuss any LCS-related questions. Patients were encouraged to quit smoking, and the program manager offered to place a consult to the tobacco cessation team (phase I). Patients were encouraged to contact their PCP with any additional personal questions regarding LCS. After 9 months, in response to resource constraints, we converted to LCS invitation by mailing only (SDM material, plus 2 letters; phase II).

LCS LDCT appointments were scheduled for eligible patients requesting screening after the SDM mailing. LDCTs were performed with multidetector helical technique in a single breath-hold during a suspended state of full inspiration. Axial images were

obtained from the lung apices to the costophrenic sulci, with a 0.8-mm slice thickness. Multiplanar reconstructions were obtained at a 1.0-mm interval for interpretation. Maximum intensity projection reconstructions were also performed.

 No veterans had an LDCT outside the VHA system.

Radiologists utilized a National Demonstration Project LCS LDCT report template that included a standardized description of the lung nodules of greatest concern for malignancy. Reports contained specific lung nodule codes captured by an electronic tool provided by the National VHA Demonstration Project. LCS program managers accessed the tracking tool, documented the nodule of greatest concern, and entered the guideline-based follow-up plan

 in the EHR template that interfaced with the tracking tool. Any incidental findings noted in the impression were recorded in the EHR template, which the PCP was asked to digitally sign as notification of findings. If there were significant incidental findings, defined as findings in the impression for which the radiologist had recommended follow-up action, the manager completed an additional significant incidental finding EHR template that the PCP was asked to sign. All LCS patients were mailed standardized LCS LDCT result letters and educational leaflets, including information about lung nodules and LCS, along with reiteration of smoking cessation recommendations.

We obtained baseline information for age, gender, race/ethnicity, smoking status, TPY information, and LCS LDCT completion dates from the EHR. When analyzing the uptake rate, we compared uptake at 219 days (minimum number of days of follow-up post randomization and mailing of the invitation for all participants) to allow for equal opportunity for follow-up. Additional information collected from the EHR and tracking tool included LDCT results. The National VHA Demonstration Project classified any lung nodule with a diameter of 2 mm or greater as requiring follow-up LDCT. LDCT results were categorized as: (1) no nodule to be tracked; (2) nodule not known to be benign, requiring tracking to establish stability over time (solid nodules without suspicious features ≤8 mm, ground-glass nodules >5 mm, or mixed-density nodules with a solid component <5 mm); (3) diagnostic evaluation needed for possible malignancy (any nodule with suspicious features, known to be new, or growing based on prior imaging); or (4) lung cancer diagnosis as a result of the initial LDCT followed by pathologic confirmation or by multidisciplinary conference consensus when tissue could not be obtained. We defined an incidental LDCT finding as any finding other than pulmonary nodules that was recorded in the report impression. We defined a significant incidental finding as any finding in the impression that required further follow-up in the opinion of an independent internist and the LCS staff pulmonologist who reviewed the EHRs.

We report demographic and smoking characteristics of eligible patients invited for LCS in the demonstration project. To identify factors potentially associated with LCS uptake among invitation participants, we compared demographic and smoking characteristics by LDCT screening completion status within the follow-up period using Pearson’s χ

 or Fisher’s exact tests for categorical measures and Wilcoxon 2-sample rank-sum tests for continuous measures. We compared LCS uptake between phases I and II using a Pearson’s χ

 test. We summarize initial screening results for patients who completed any LDCT screening through May 22, 2015.

The preliminary EHR eligibility algorithm identified 6133 unique patients with primary care visits from January 2, 2014, through August 15, 2014. There were 4745 patients (77%) excluded from invitation to LCS for the following reasons: 3248 with incomplete TPY information (53%), 89 out of age range (1%), 2 smokers who quit more than 15 years ago or had fewer than 30 pack-years (<1%), 1394 never-smokers (23%), and 12 patients who declined to provide TPY information (<1%).

A total of 1388 patients were eligible for LCS invitation. Of these, 926 were invited for LCS and 462 were not invited but had opportunity for referral outside of the demonstration project. Eight of the 926 patients were noted to be ineligible upon chart review by the program manager prior to invitation and were excluded from further analysis, leaving 918 patients invited for LCS. Twenty-seven patients died before being able to complete screening. Baseline characteristics of patients invited for LCS are described in 

The LCS program manager completed calls to 280 patients invited for LCS; the remaining 638 patients received SDM materials and invitation letters only. Patients received LCS within 1 month of accepting screening or per patient preference.

Of 918 patients invited for LCS, 178 (19%) completed an LDCT. Of the 766 invited participants in phase I, 165 (22%) had an LDCT. Of the 152 phase II patients who did not receive an outreach phone call, 13 (9%) had an LDCT. The difference between the 9% and 22% uptake rates was statistically significant (

<.001). Due to resource constraints, planned calls could not be made for 486 of the 766. Of the remaining 280 with complete phone status, 165 (59%) had an LDCT. The rate of LCS uptake, with follow-up time fixed to 219 days, was highest for the first cohort of eligible veterans but then remained steady during the rest of phase I. The rate of LCS declined during phase II. Increased awareness by PCPs was not a source of bias given their low rate of referral. We found no significant differences in demographic or smoking characteristics between those who completed an LDCT and those who did not (

Of the 178 patients who completed an initial LDCT during the follow-up interval, 39% either had clinically benign nodules that did not require further follow-up or had no nodules. Sixty-one percent had a nodule requiring follow-up; of these, 43% had a nodule <4 mm, 20% had a nodule ≥4 mm to <5 mm, 7% had a nodule ≥5 mm to 6 mm, 18% had a nodule >6 mm to 8 mm, and 12% had a nodule >8 mm in diameter. Twelve patients required diagnostic evaluation, and 2 had malignant nodules. The time from LDCT to lung cancer diagnosis ranged from 14 to 76 days. A total of 179 incidental findings were reported in 116 patients; 65% of patients had at least 1 incidental finding. Twenty percent of the incidental findings were considered clinically significant, most commonly abdominal abnormalities such as renal masses.

Despite EHR prompts to check-in nurses, complete TPY information was not captured in the national data tracking tool for most patients during the demonstration project. We retrospectively determined that this was mainly due to inaccurate initial TPY information entry by check-in nurses into restrictive EHR data fields. We retrospectively reanalyzed the TPY data to obtain more accurate estimates of the number of potentially LCS-eligible patients at our institution. Among the 6099 patients with complete TPY reminders after reanalysis, 76% did not meet LDCT eligibility criteria. Of those, 30% never smoked, 16% had fewer than 30 pack-years, and 54% quit more than 15 years ago. Of the 24% who met criteria for further LCS consideration, 46% were former smokers and 54% were current smokers.

Our initial local experience demonstrates the importance of a well-designed, systematic approach to the complex process of LCS. EHRs must be designed to capture accurate, complete TPY data to determine LCS eligibility. Patient participation may vary by invitation method, with potentially higher uptake rates following more direct patient outreach. Reliable LCS tracking tools that interface with the EHR are useful to efficiently manage large numbers of patients with lung nodules, many of which have a very low probability of malignancy yet require complex, prolonged surveillance. LCS systems should be designed to address numerous incidental findings of varying clinical significance.

LCS uptake by patients seemed dependent on the invitation approach. Uptake was higher when a phone call was added to SDM and letters. This uptake was comparable with the National VHA Demonstration Program results.

 Person-to-person SDM, although desirable, is resource intensive and time consuming. Given the large number of potentially eligible patients, initially adopting a less aggressive invitation method, such as provision of written SDM materials followed by mailings only, may allow healthcare institutions to offer equitable screening access for all interested patients. However, LCS uptake at our facility was lower with this approach. Although PCPs at our institution opted out of the demonstration project, their participation in LCS SDM is of potential value to patients, given the unique insights they may have regarding their patients. Conversely, restricting SDM to PCPs might subject patients to possible provider bias for or against LCS. In addition to phone counseling, new LCS programs might consider alternative strategies, such as online interactions on MyHealtheVet, screening “drop-in” clinics, or lung health fairs or other community events, to reach patients.

 We found a higher rate of patients with lung nodules than was reported in the NLST (59% vs 27%), although our results were similar to the overall rate detected in the National VHA Demonstration Project.

 One possible explanation for the higher rate in our program is that LDCT technology and protocols have advanced since the NLST, including thinner reconstruction intervals and use of maximum intensity projection reconstruction.

 Interreader variability among radiologists can also account for differences in nodule detection sensitivity and false-positive rates.

 However, the most likely explanation is a difference in the definition of a lung nodule: Based on existing guidelines at the time,

 the National VHA Demonstration Project defined a nodule 2 mm or greater in diameter as requiring follow-up compared with 4 mm or greater in NLST.

 By the NLST definition, 33% of our patients had a nodule requiring follow-up, similar to the frequency reported in the NLST. More recently, the American College of Radiology provided guidelines for tracking lung nodules detected during LCS.

 These guidelines consider solid lung nodules less than 6 mm or new solid lung nodules less than 4 mm to have a very low likelihood of malignancy, thus their recommendation to continue annual LCS.

This assumes that all such patients will remain eligible for LCS, although some will not be, due to age or other changing eligibility criteria. It also does not address issues of liability if no follow-up LDCT is recommended or performed or how to engage in SDM with patients who might make different decisions about future LDCT dependent on its purpose. Would decisions differ if the LDCT was a screening scan (ie, no known disease) versus a scan to follow up on a small, albeit very low-risk, lung nodule? Until further evidence and guidance is provided regarding tracking sub-4 mm nodules, tracking nodules 2 mm or greater in diameter for at least 1 year continues to be common practice.

Increased detection of small lung nodules with a very low likelihood of malignancy, as well as other incidental findings, may have potential adverse consequences. Patients with nodules requiring follow-up may experience anxiety while awaiting the next LDCT.

 The extent to which indeterminate nodules and incidental findings affect patient psychological well-being remains uncertain; however, an analysis of the NLST results found no increase in anxiety in patients with a false-positive nodule (requiring only serial LDCT surveillance) or a significant incidental finding on screening compared with patients with a negative LDCT.

 Increased detection of small lung nodules increases resource utilization with uncertain clinical benefit

 and may result in additional diagnostic evaluations with associated costs and complications.

We detected more clinically significant incidental findings than the NLST.

 It is unclear whether our definitions and methods of determining clinical significance were comparable with those of the NLST, and the consequences of detecting clinically significant incidental findings have not been systematically characterized.

Our study had several limitations. This was a single-site evaluation, utilizing a VHA EHR with potentially more complete data capture and patient tracking than may be available in other healthcare systems. However, efficient and accurate data capture is also a strength of our study, as it is an essential requirement for successful LCS implementation. The high rate of TPY information capture error during the demonstration project limited the selection sample to patients with complete initial TPY information. Because of the change from calling nonrespondents in phase I to mailing outreach only in phase II of this demonstration project, we were given an opportunity to estimate the screening rates between 2 direct invite methods, albeit with limitations. First, there was no random invitation assignment. Second, there could be temporal or other reasons for the observed decrease in screening rate that we are unable to account for in this study. Third, due to this unplanned change, systems were not in place to completely capture which invite method the patient received. Due to that limitation, we are unsure if the completed calls from phase I were perhaps from patients very motivated to be screened who called in early&#8203; or if they were initially nonrespondents who received an outreach call. We estimated that if one does not call initial nonrespondents (phase II), the screening rate would be 9%, which estimates the proportion of motivated patients willing to call in to seek early screening. The significant difference in screening rates between phase I and phase II warrants further investigation to better assess the difference an outreach call to nonresponders can make to LCS uptake rates. The number of lung malignancies we detected on initial LCS LDCT was lower than that reported in the NLST,

 possibly due to our relatively small patient sample size. Finally, our homogenous sample in a healthcare system limits generalizability. Although many of the lessons learned from our demonstration site apply to screening programs at other institutions, the VHA is a unique setting.

We implemented LCS as part of a National VHA Demonstration Project. LCS is unprecedented in its complexity compared with other cancer screenings, due in part to requirements for structured SDM and detailed guidelines for follow-up of large numbers of lung nodules. Patient participation rates were dependent on accurate TPY information capture and method of LCS invitation. Efficiencies included integration of the EHR with the LCS tracking system and standardized data capture templates for radiology reports and lung nodule tracking, all directed by a full-time LCS program manager. Methods of LCS invitation must optimize equity and access within the confines of institutional capacity and resources. Smoking cessation is a critical component of LCS, and more information is needed regarding the effects of LCS on smoking behavior when offered in conjunction with proactive smoking cessation. Further evaluation is needed to determine optimal methods of patient risk stratification, LCS invitation approaches, SDM, and the risks and benefits associated with the detection of small lung nodules and incidental findings.

Pulmonary Section, Department of Medicine (AEF, KLR), Center for Chronic Disease and Outcomes Research (SEL, MRP, SSF, BAC, AKB, DBN, EAD), and Department of Radiology (BJB), Minneapolis Veterans Affairs Health Care System, Minneapolis, MN; Department of Medicine, University of Minnesota Twin Cities Medical School (MRP, SSF, DBN, KLR), Minneapolis, MN.

 This material is the result of work supported in part by the Minneapolis Veterans Affairs Health Care System, the Veterans Health Administration National Center for Health Promotion and Disease Prevention, and the Minneapolis Veterans Affairs Health Care System Center for Chronic Disease and Outcomes Research. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

 Concept and design (AEF, SEL, MRP, DBN, KLR); acquisition of data (AEF, SSF, BAC, AKB, EAD, BJB, KLR); analysis and interpretation of data (AEF, SEL, SSF, BAC, AKB, DBN, KLR); drafting of the manuscript (AEF, SEL, DBN, BJB, KLR); critical revision of the manuscript for important intellectual content (SEL, MRP, SSF, BAC, DBN, BJB, KLR); statistical analysis (BAC, DBN, KLR); provision of patients or study materials (EAD); administrative, technical, or logistic support (SSF, AKB, EAD); and supervision (MRP).

 Angela E. Fabbrini, MPH, Minneapolis Veterans Administration Health Care System (111N), 1 Veterans Dr, Minneapolis, MN 55417. Email: Angela.Fabbrini@va.gov.

. Atlanta, GA: American Cancer Society; 2015. cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annual-cancer-facts-and-figures/2015/cancer-facts-and-figures-2015.pdf. Accessed December 20, 2015.

2. Aberle DR, Adams AM, Berg CD, et al; National Lung Screening Trial Research Team. Reduced lung-cancer mortality with low-dose computed tomographic screening. 

. 2011;365(5):395-409. doi: 10.1056/NEJMoa1102873.

3. Moyer VA; U.S. Preventive Services Task Force. Screening for lung cancer: U.S. Preventive Services Task Force recommendation statement. 

. 2014;160(5):330-338. doi: 10.7326/M13-2771.

4. Providing guidance on lung cancer screening to patients and physicians. American Lung Association website. lung.org/assets/documents/lung-cancer/lung-cancer-screening-report.pdf. Published April 30, 2015. Accessed February 2, 2016.

5. Detterbeck FC, Mazzone PJ, Naidich DP, Bach PB. Screening for lung cancer: diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. 

. 2013;143(suppl 5):e78S-e92S. doi: 10.1378/chest.12-2350.

6. Jaklitsch MT, Jacobson FL, Austin JH, et al. The American Association for Thoracic Surgery guidelines for lung cancer screening using low-dose computed tomography scans for lung cancer survivors and other high-risk groups. 

. 2012;144(1):33-38. doi: 10.1016/j.jtcvs.2012.05.060.

CLINICAL

Results, lessons, and challenges of a local lung cancer screening program within a national demonstration project.

Initial Results of a Lung Cancer Screening Demonstration Project: A Local Program Evaluation

 To describe the extent of and longitudinal changes in physician practice variation with respect to implant costs, institutional postacute care (PAC) provider utilization, and total episode payments, as well as to evaluate the association between physician volume and quality and these outcomes.

We combined claims and internal hospital cost data for

34 physicians responsible for 3614 joint replacement episodes under bundled payment at Baptist Health System (BHS). Multilevel multivariable generalized linear models were employed and the intraclass correlation (ICC) was used to quantify between-physician variation.

 There was significant between-physician variation in implant costs, institutional PAC provider utilization, and total episode payments not explained by observable variables (

<.001 for all). Over 5 years, the ICC decreased from 0.26 to 0.06, 0.15 to 0.13, and 0.12 to 0.10 for implant costs, institutional PAC provider utilization, and total episode payments, respectively, but differences were not statistically significant. Both higher physician case volume and quality were associated with decreased total episode payments and institutional PAC provider utilization, but not with changes in implant costs.

practice variation was observed under bundled payment at BHS

and decreased to a greater degree for implant costs than institutional PAC provider utilization or total episode payments. Institutional PAC provider utilization and total episode payments were associated with physician volume and quality. Although some organizational strategies achieve gains by reducing physician practice variation, variation reduction is not an absolute requisite for success under bundled payment.

Amid the proliferation of orthopedic bundles, organizations can benefit from understanding whether strategies must reduce physician practice variation in order to succeed. Analyzing the experience of a long-standing participant in Medicare orthopedic joint replacement bundles, we found that although some strategies achieve gains by reducing physician practice variation, variation reduction is not an absolute requisite for success under bundled payment.

There was significant physician practice variation in implant costs, institutional postacute care (PAC) provider utilization, and total episode payments.

Over time, physician practice variation decreased to a greater degree for implant costs than for institutional PAC provider utilization or total episode payments.

To increase value and address substantial cost and quality variation in care, CMS has rapidly expanded orthopedic bundled payment initiatives. By receiving a single “bundled” amount for an entire episode of care, hospitals and physicians are accountable for quality and costs under bundled payment.

After early success bundling acute care among a small group of providers in the voluntary Acute Care Episode (ACE) demonstration,

 CMS extended its focus across the care continuum. In 2012, it launched the Bundled Payments for Care Improvement (BPCI) initiative,

 a large national program through which many hospitals voluntarily accept bundled payment for joint replacement episodes encompassing acute and postacute care (PAC). More recently, in April 2016, CMS used the BPCI framework to initiate the Comprehensive Care for Joint Replacement model,

 a mandatory program that initially held approximately 800 hospitals in 67 metropolitan areas financially accountable for the costs and quality of episodes spanning hospitalization through 90 days of PAC.

Early evidence from BPCI suggests that bundled payment can reduce episode costs and produce savings for both CMS and hospitals while possibly increasing quality.

 Because physician practice style contributes to large variation in hospital and postacute spending for surgeries such as hip replacement,

 variation reduction is frequently championed in bundle payment strategy.

 However, there is little empirical evidence about whether organizational strategies must reduce physician practice variation in order to succeed.

In this study, we describe physician practice variation at Baptist Health System (BHS), a continuous participant in Medicare joint replacement bundles since 2009. BHS achieved high performance—demonstrating notable reductions in total episode payments and hospital savings during a period when joint replacement expenditures rose nationwide

—by engaging physicians to reduce surgical implant costs and postdischarge utilization of institutional PAC providers.

 We evaluate the extent of and longitudinal changes in physician practice variation with respect to implant costs, institutional PAC provider utilization, and total episode payments, as well as the associations among physician volume and quality and these 3 outcomes.

The study was divided into 4 periods. The first was the baseline period (July 2008-December 2008), when BHS received nonbundled fee-for-service payments prior to ACE participation. The second was the ACE period (July 2009-June 2012), during which BHS implemented joint replacement bundles for acute hospitalization. The ACE period consisted of 3 years: ACE year 1 (July 2009-June 2010), ACE year 2 (July 2010-June 2011), and ACE year 3 (July 2011-June 2012). The third was the transition period (July 2012-September 2013), when BHS prepared to transition to BPCI Model 2 and did not receive bundled payment. The fourth was the BPCI period (October 2013-June 2015), in which BHS implemented joint replacement bundles for episodes spanning acute hospitalization through 30 days of PAC. The BPCI period consisted of 2 years: BPCI year 1 (October 2013-June 2014) and BPCI year 2 (July 2014-June 2015).

We obtained both Medicare claims and internal hospital cost data directly from BHS.

 Using Medicare data, we constructed care episodes that were consistent with the health system’s BPCI arrangement with CMS and encompassed acute hospitalization plus 30 days of PAC.

 Data were not available during the transition period between ACE and BPCI when BHS was not paid by CMS under bundled payment.

Our analytic sample consisted of 3614 patient episodes from a panel of 34 physicians who performed at least 1 joint replacement surgery in both the ACE and BPCI periods. Overall, 51 orthopedic surgeons affiliated with BHS performed a combined total of 3725 joint replacement surgeries on Medicare beneficiaries under Medicare Severity Diagnosis Related Group 470, Major joint replacement or reattachment of lower extremity without major complications or comorbidities. We excluded physicians who performed surgeries only in the ACE period (n = 9) or BPCI period (n = 8).

Patient demographics and clinical data were calculated from Medicare claims, and patient illness severity was estimated using the Elixhauser Comorbidity Score.

 Medicare data were also used to calculate 2 quality of care measures, 30-day readmission and emergency department (ED) visit rates, based on CMS demonstration project specifications.

 Characteristics of operating physicians (gender, years of experience, allopathic vs osteopathic degree, graduation from US or non-US medical school, and board certification status) were obtained from publicly available information on consumer websites.

We evaluated variation in 3 outcomes of interest: implant costs, total episode payments, and institutional PAC provider utilization. Internal hospital cost data were used to quantify the costs of orthopedic implants (implant costs), while BHS used a time-driven activity-based costing approach

 to quantify other nonsupply hospital costs. Medicare data were used to calculate the sum of all Medicare payments for an episode of care (total episode payments) by combining claims for all acute hospital facility payments, physician fees during hospitalization, and total PAC payments through 30 days post discharge. Total PAC payments consisted of those corresponding to outpatient visits, ED visits, readmissions, physician fees, durable medical equipment, and utilization of PAC providers, such as skilled nursing facilities (SNFs), inpatient rehabilitation facilities (IRFs), and home health agencies (HHAs).

We used Medicare data to define a dichotomous variable (institutional PAC provider utilization) as discharge from hospitalization to either SNFs or IRFs as institutional PAC providers, compared with discharge home or home with HHA services. Patients were discharged to 1 of 4 destinations in nearly all (99.4%) episodes.

Finally, we defined several variables to evaluate associations between physician practice characteristics and our outcomes. We assessed practice quality by calculating the proportion of episodes for each physician with prolonged length of stay (PLOS), a validated measure of complications of orthopedic lower extremity joint replacement.

 PLOS evaluates the pattern of patient discharges as a function of length of stay (ie, number of days in the hospital) and identifies a point at which discharge is less likely than continued hospitalization, representing a complication. We assess practice volume using total case volume over the study period. Due to data skewness, we analyzed volume as a continuous variable, as well as a dichotomous variable with the top 10 physicians by volume (each with more than 100 cases and representing more than 70% of total cases) defined as high-volume physicians.

Descriptive statistics were reported using means and SDs for continuous variables and percentages for categorical variables. Multilevel generalized linear models, adjusted for study year and patient and physician characteristics, were employed to evaluate the extent of physician variation in the 3 outcomes of interest. Models were clustered at the physician level and included a physician random effect. To account for multiple patients per operating physician and heteroscedasticity, all models utilized clustered standard errors with the Huber-White correction.

We utilized linear random effects models with a log link and gamma distribution for the outcomes of implant cost and total episode spending.

 To evaluate institutional PAC provider utilization, we used a random effects model with a logit link and binominal distribution to estimate the logit of the probability of discharge to institutional PAC providers as a linear function of covariates and operating physician random effect.

The significance of between-physician variation was determined by testing whether the variance of the physician random effect differed significantly from 0 and was reported using the intraclass correlation (ICC). The ICC is a measure that reflects the proportion of overall variance explained by variation between, rather than within, clusters of individual observations. In this study, the calculated ICC corresponds to the proportion of variation explained by differences in practice patterns among physicians (ie, between-physician variation), as opposed to among cases for individual physicians (ie, within-physician variation). A linear threshold method was used to calculate ICC for institutional PAC provider utilization.

 To compare changes in the ICC across the study period, we used bootstrapping to calculate standard errors for ICCs in ACE year 1 and BPCI year 2 before using the 

 statistic and pairwise tests for each outcome to evaluate whether there were significant differences between the 2 years.

Random effects models were also used to calculate and compare between-physician variation and ICC for implant costs and total episode payments across our study period (ACE year 1 vs BPCI year 2). Because bundle definitions did not include PAC until the BPCI program period, we compared between-physician variation and ICC for PAC provider utilization among ACE year 1, ACE year 3, and BPCI year 2.

Separately, we used random effects models to test the associations among physician quality and case volume and our 3 outcomes of interest. All cost and spending figures were adjusted for inflation and reported in 2015 US$ equivalents. Implant cost data were aggregated at the level of individual physician per quarter and weighted by episode volume. Analyses were conducted using SAS version 9.4 (SAS Institute; Cary, North Carolina). All tests of significance were 2-tailed and considered statistically significant at an α of .05. The University of Pennsylvania Institutional Review Board approved the study.

The overall patient and program characteristics across the BHS bundled payment program have been described previously.

 From that sample, we included 34 physicians in this analysis, 97% of whom were male, 100% whom had obtained allopathic medical degrees, 91% of whom had graduated from US medical schools, and 82% of whom were board certified (

). On average, physicians in this cohort had 26 years of experience and a total volume of 106 cases across our study period, ACE year 1 to BPCI year 2.

The mean age of patients cared for by these physicians was 72 years. They were predominantly female, and the mean Elixhauser Comorbidity Score was 1.1, with the 5 most common coexisting clinical conditions being hypertension (76%), diabetes (21%), hypothyroidism (21%), obesity (19%), chronic obstructive pulmonary disease (16%), and depression (10%). Readmissions and ED visits occurred in 5% and 6% of cases, respectively.

Overall, mean implant costs decreased 5.8% across the study period, from $5026 in the first year of bundled payment (ACE year 1) to $4732 through 5 years (BPCI year 2). There was also an overall downward shift in distribution of implant costs among physicians. Compared with ACE year 1, for example, more physicians in BPCI year 2 had average implant costs between $4500 and $5000 (22 vs 9) and fewer physicians had average implant costs of $5550 or greater (1 vs 10) (

After accounting for patient and physician characteristics, higher implant cost was associated with younger patient age and earlier study year. There was a decline in between-physician variation across study year and program period (

This study demonstrates that variation reduction is an important, but not requisite, component of organizational success under orthopedic bundled payment.

June 2018

Simply Delivered Meals: A Tale of Collaboration

Initial Results of a Lung Cancer Screening Demonstration Project: A Local Program Evaluation

Physician Practice Variation Under Orthopedic Bundled Payment

Prevalence and Predictors of Hypoglycemia in South Korea

A Longitudinal Examination of the Asthma Medication Ratio in Children

Cost Sharing for Antiepileptic Drugs: Medication Utilization and Health Plan Costs

Satisfaction With Care After Reducing Opioids for Chronic Pain

Assessing Markers From Ambulatory Laboratory Tests for Predicting High-Risk Patients

Identifying Children at Risk of Asthma Exacerbations: Beyond HEDIS

Placement of Selected New FDA-Approved Drugs in Medicare Part D Formularies, 2009-2013