Advertisement

A Palliative Radiation Oncology Consult Service Reduces Total Costs During Hospitalization

Open ArchivePublished:March 08, 2018DOI:https://doi.org/10.1016/j.jpainsymman.2018.03.005

      Abstract

      Context

      Palliative radiation therapy (PRT) is a highly effective treatment in alleviating symptoms from bone metastases; however, currently used standard fractionation schedules can lead to costly care, especially when patients are treated in an inpatient setting. The Palliative Radiation Oncology Consult (PROC) service was developed in 2013 to improve appropriateness, timeliness, and care value from PRT.

      Objectives

      Our primary objective was to compare total costs among two cohorts of inpatients with bone metastases treated with PRT before, or after, PROC establishment. Secondarily, we evaluated drivers of cost savings including hospital length of stay, utilization of specialty-care palliative services, and PRT schedules.

      Methods

      Patients were included in our observational cohort study if they received PRT for bone metastases at a single tertiary care hospital from 2010 to 2016. We compared total costs and length of stay using propensity score-adjusted analyses. Palliative care utilization and PRT schedules were compared by χ2 and Mann-Whitney U tests.

      Results

      We identified 181 inpatients, 76 treated before and 105 treated after PROC. Median total hospitalization cost was $76,792 (range $6380–$346,296) for patients treated before PROC and $50,582 (range $7585–$620,943) for patients treated after PROC. This amounted to an average savings of $20,719 in total hospitalization costs (95% CI [$3687, $37,750]). In addition, PROC was associated with shorter PRT schedules, increased palliative care utilization, and an 8.5 days reduction in hospital stay (95% CI [3.2,14]).

      Conclusion

      The PROC service, a radiation oncology model integrating palliative care practice, was associated with cost-savings, shorter treatment courses and hospitalizations, and increased palliative care.

      Key Words

      Introduction

      Background

      Radiation treatment can be a highly effective, but also costly and burdensome palliative intervention for patients suffering from advanced cancer.
      • Bekelman J.E.
      • Epstein A.J.
      • Emanuel E.J.
      Single- vs multiple-fraction radiotherapy for bone metastases from prostate cancer.
      • Lutz S.
      • Spence C.
      • Chow E.
      • Janjan N.
      • Connor S.
      Survey on use of palliative radiotherapy in hospice care.
      • Hess G.
      • Barlev A.
      • Chung K.
      • Hill J.W.
      • Fonseca E.
      Cost of palliative radiation to the bone for patients with bone metastases secondary to breast or prostate cancer.
      • Fischberg D.
      • Bull J.
      • Casarett D.
      • et al.
      Five things physicians and patients should question in hospice and palliative medicine.
      One of the commonest reasons for referral to a radiation oncologist is for palliation of symptomatic bone metastases, a frequent site of metastatic disease with high prevalence (estimated 280,000 persons in the U.S.).
      • Li S.
      • Peng Y.
      • Weinhandl E.D.
      • et al.
      Estimated number of prevalent cases of metastatic bone disease in the US adult population.
      Pain relief rates of 50%–90% are commonly observed in patients undergoing PRT to the bone.
      • Chow E.
      • Harris K.
      • Fan G.
      • Tsao M.
      • Sze W.M.
      Palliative radiotherapy trials for bone metastases: a systematic review.
      • Jones J.A.
      • Simone 2nd, C.B.
      Palliative radiotherapy for advanced malignancies in a changing oncologic landscape: guiding principles and practice implementation.
      However, treatments can vary in duration and can span several weeks, resulting in debilitating adverse effects such as severe fatigue, esophagitis, and diarrhea.
      • Jones J.A.
      • Simone 2nd, C.B.
      Palliative radiotherapy for advanced malignancies in a changing oncologic landscape: guiding principles and practice implementation.
      • Chow E.
      • Zeng L.
      • Salvo N.
      • et al.
      Update on the systematic review of palliative radiotherapy trials for bone metastases.
      Patients overwhelmed by these burdens are at risk of being hospitalized during radiation treatment, or worse, of not completing it.
      • Jones J.A.
      • Simone 2nd, C.B.
      Palliative radiotherapy for advanced malignancies in a changing oncologic landscape: guiding principles and practice implementation.
      • Fesinmeyer M.D.
      • Mehta V.
      • Tock L.
      • et al.
      Completion of radiotherapy for local and regional head and neck cancer in Medicare.
      Shorter radiation courses, such as single-fraction radiation therapy (RT; one session) or hypofractionated RT (two to five sessions), are well-known to result in equally effective pain relief from bone metastases.
      • Bekelman J.E.
      • Epstein A.J.
      • Emanuel E.J.
      Single- vs multiple-fraction radiotherapy for bone metastases from prostate cancer.
      • Lutz S.
      • Berk L.
      • Chang E.
      • et al.
      Palliative radiotherapy for bone metastases: an ASTRO evidence-based guideline.
      • Hartsell W.F.
      • Scott C.B.
      • Bruner D.W.
      • et al.
      Randomized trial of short- versus long-course radiotherapy for palliation of painful bone metastases.
      Some prior studies have reported that single-fraction courses are more cost-effective, and less burdensome on patients and families than longer conventional fractionation schedules, yet they have been largely underutilized up to this point.
      • Bekelman J.E.
      • Epstein A.J.
      • Emanuel E.J.
      Single- vs multiple-fraction radiotherapy for bone metastases from prostate cancer.
      • Rutter C.E.
      • Yu J.B.
      • Wilson L.D.
      • Park H.S.
      Assessment of national practice for palliative radiation therapy for bone metastases suggests marked underutilization of single-fraction regimens in the United States.
      • Schreiber D.
      • Safdieh J.
      • Becker D.J.
      • Schwartz D.
      Patterns of care and survival outcomes of palliative radiation for prostate cancer with bone metastases: comparison of ≤5 fractions to ≥10 fractions.
      In an effort to improve on the existing pattern of prescribed palliative radiation therapy (PRT), we established the Palliative Radiation Oncology Consult (PROC) service in 2013 within the Mount Sinai Hospital's Department of Radiation Oncology. This service works closely with palliative care specialists and other supportive cancer care professionals to provide generalist-level palliative medicine to advanced cancer patients at the time of the initial radiation oncology consult. Most patients with advanced cancer necessitating palliative radiation were cared for under PROC. The following are the salient features of this novel service: 1) staffing by a radiation oncologist, nurse, administrative assistant, and rotating residents and fellows; 2) goals of care discussion occurring in the context of family meetings; 3) priority of communication among multiple specialists involved in patient's care; 4) early consultation by specialist-level palliative medicine; and 5) discussion on a regular basis during an associated specialty tumor board attended by PROC representatives. The impact of this service model on symptom outcomes of advanced cancer patients with bone metastases has been described elsewhere.
      • Chang S.
      • May P.
      • Goldstein N.E.
      • et al.
      A palliative radiation oncology consult service's impact on care of advanced cancer patients.
      The primary aim of this study was to assess the PROC service's impact on total hospitalization costs for advanced cancer patients receiving PRT for symptomatic bone metastases. Our secondary aim was to evaluate for potential drivers of cost savings including hospital length of stay (LOS), specialty-care palliative services, and changes in PRT regimens.

      Methods

      Intervention

      The PROC service was developed at Mount Sinai Hospital in October 2013 to care for both hospitalized and ambulatory care advanced cancer patients with any type of palliative radiation need. The service is staffed by a radiation oncologist with specific expertise in primary palliative care, a nurse, an administrative assistant, and rotating residents and fellows.

      Setting and Study Design

      We performed an observational cohort study to investigate total hospital costs for patients receiving PRT for symptomatic bone metastases under management of the PROC service during a hospitalization at Mount Sinai Hospital. Mount Sinai Hospital is a high-volume tertiary care center covering patients from diverse socioeconomic backgrounds.

      Participants

      Through electronic medical records and review of International Disease Classification code denoting secondary malignant neoplasm of bone (ICD-9 code 198.5 and ICD-10 code C79.51), we identified all patients between the ages of 18 and 95 years old treated at Mount Sinai Hospital with stage IV or otherwise incurable cancer and confirmed pathologic or radiographic evidence of osseous metastases. We included patients in the cohort analysis who began PRT at any time-point during a hospitalization occurred between February 2010 and December 2016, because electronic medical records were not readily available before 2010. To simplify analyses for patients with multiple inpatient radiation courses, we selected the first palliative radiation course that was administered during a hospitalization. Patients were excluded if they had incomplete medical records or were lost to follow-up. We grouped patients into two separate (pre and post) study cohorts in relation to the inception of the PROC service: before the PROC service was established (February 2010 to October 2013) and after its establishment (October 2013 to December 2016).

      Variables

      Outcomes of Interest

      Our primary outcome of interest was total cost of hospitalization, including both direct and indirect costs. Direct costs encompassed those that were specifically associated with the hospitalization: diagnostic tests, imaging, medications, procedures, room and board, and nurse staffing.
      • Taheri P.A.
      • Butz D.
      • Griffes L.C.
      • Morlock D.R.
      • Greenfield L.J.
      Physician impact on the total cost of care.
      Indirect, or overhead, costs were nonclinical costs to the hospital and included maintenance, operations, and administrative costs.
      • Taheri P.A.
      • Butz D.
      • Griffes L.C.
      • Morlock D.R.
      • Greenfield L.J.
      Physician impact on the total cost of care.
      • Weinstein M.C.
      • Siegel J.E.
      • Gold M.R.
      • Kamlet M.S.
      • Russell L.B.
      Recommendations of the panel on cost-effectiveness in health and medicine.
      Patients were excluded if they lacked corresponding cost data for the index hospitalization. We measured the average treatment effect on total cost between the two cohorts.
      To examine possible contributors to total costs, our secondary outcomes of interest include hospital LOS, utilization of inpatient palliative care consultation, and description of PRT characteristics. Hospital LOS was calculated by counting days from the date of admission to discharge. Stays in the emergency department that did not result in an eventual admission to the hospital were excluded. Patients were counted as having used palliative care services if their chart documented an inpatient consult with a palliative care provider sometime during their admission. Details of PRT delivery, such as treatment duration and dose prescription, were reviewed for each record. Dose fractionation was categorized into single-fraction, hypofractionation (two to five fractions), six to 10 fractions, and >10 fractions.

      Other Predictors

      We used categorical variables for gender, race, primary cancer type, and medical insurance (Table 1). In the context of our sample size, we used two strata for cancer type: hematologic cancer, the most frequent cancer type in our patient population, and non-hematologic. We used continuous variables for age and comorbidities summarized by a calculated Charlson comorbidity index.
      • Charlson M.E.
      • Pompei P.
      • Ales K.L.
      • MacKenzie C.R.
      A new method of classifying prognostic comorbidity in longitudinal studies: development and validation.
      Table 1Patient Demographic and Clinical Information
      Unadjusted SamplesPropensity Score Adjusted Samples
      Pre-PROC (n = 76)Post-PROC (n = 105)Absolute SD (%)Pre-PROC (n = 76)Post-PROC (n = 105)Absolute SD (%)
      Age
       Years (mean)61.3±13.863.6±12.4−17%63.6±1463.6±12.40%
       Range23–9531–89
      CCI
       Score (mean)5.7±2.95.9±2.6−7%5.9±2.75.9±2.62%
       Range2–151–15
      Gender, %
       Female35.545.7−21%44.845.72%
      Race, %
       Black30.339−19%37394%
       White32.929.57%30.329.52%
       Other36.831.411%32.731.43%
      Primary cancer, %
       Hematologic42.132.420%33.232.42%
      Insurance, %
       Medicare43.454.3−22%54.754.31%
       Medicaid18.428.6−24%27.328.63%
       Other38.217.148%1817.12%
      PROC = Palliative Radiation Oncology Consult; CCI = Charlson comorbidity index.

      Data Sources

      Information about demographics, hospitalizations, and radiation courses was collected from electronic medical records. Medical insurance status and cost data were extracted from the hospital's cost accounting system (Allscripts TSI). Cost data were acquired as an enumeration of specific costs for each department and products administered to each patient. These include radiology imaging, medications administered, radiation oncology costs, laboratory tests, specialty consultations, room and board, among others. All costs were standardized to U.S. dollars in 2016, the final year of data collection, using the Consumer Price Index.

      Bias

      Selection bias was not a primary concern in our study because membership of the cohorts was designated by timing (whether the hospitalization occurred before or after the PROC was implemented), and not according to patient-level factors. However, baseline clinical characteristics were likely correlated with treatment outcomes and health care utilization, and differing distributions of these characteristics between cohorts could lead to bias in analysis. In addition, smaller sample size increased the likelihood of baseline characteristics’ imbalance through natural variation across samples.
      To reduce the effects of potential confounders, propensity score kernel weights were calculated to balance all observed baseline factors in Table 1 between cohorts.
      • Garrido M.M.
      • Kelley A.S.
      • Paris J.
      • et al.
      Methods for constructing and assessing propensity scores.
      Balance across groups was evaluated with standardized differences both before and after weighting the samples. Samples differed across analyses because of missing data on relevant outcomes, for example, one patient lacked a variable RT cost because of archiving of cost details. Where samples were altered because of missing outcome data, a new sample-specific propensity score was created.
      • Green K.M.
      • Stuart E.A.
      Examining moderation analyses in propensity score methods: application to depression and substance use.

      Statistical Analysis

      Health care utilization data typically exhibit non-normal distributions—skewness, heteroscedasticity, and leptokurtosis—that pose challenges in analysis.
      • Jones A.M.
      Models for health care.
      Moreover, application of propensity score weights may reduce observed bias, but distort results by imbalancing unobserved factors.
      • Brooks J.M.
      • Ohsfeldt R.L.
      Squeezing the balloon: propensity scores and unmeasured covariate balance.
      To address these challenges, we analyzed associations between PROC and our two utilization outcomes (total hospital costs and hospital LOS) using two regression approaches that do not assume normal distribution of data, and with and without the propensity score weights.
      First, an unweighted comparison of each outcome of interest between groups was performed using the Mann-Whitney U test. Second, we fitted generalized linear models with a γ distribution and a log link, using a binary variable (pre-PROC | post-PROC) as our primary independent variable of interest and all predictor variables (age, Charlson comorbidity index, female gender, race, primary cancer, and medical insurance) as additional independent variables, and applying the propensity score weights. Estimated treatment effects were calculated as the average effect of moving a subject from the pre-PROC to the post-PROC group holding all other values constant. P-values < 0.05 were considered to be statistically significant in all analyses.
      Comparisons of categorical variables (PRT fractionation and inpatient palliative care consultation) were performed with χ2 test. Mann-Whitney U tests were used to compare continuous variables (PRT prescription dose and course duration).
      Propensity score matching was conducted by the Covariate Balancing Propensity Score package in R (version 3.4.0; The R Foundation for Statistical Computing, Vienna, Austria). All other statistical analyses were conducted using STATA (version 14.2; StataCorp, College Station, Texas).

      Results

      Participants

      We identified 181 patients who met our inclusion criteria, with 76 patients treated with inpatient palliative radiation before the establishment of PROC (pre-PROC) and 105 patients treated with inpatient palliative radiation after the establishment of PROC (post-PROC). Five patients were excluded for incomplete medical records or lost to follow-up. Seven patients were excluded for lack of corresponding cost data for their hospitalization.

      Patient Demographic and Clinical Information

      Demographic and clinical information for these patients are summarized in Table 1. Before matching, there were notable differences on the following baseline covariates between cohorts: age, female gender, black and other race, hematologic cancer type, and medical insurance. After matching, baseline imbalances were small with a mean absolute standardized difference of 2% and no absolute standardized difference >4% (Table 1).
      Twenty-one patients died during their hospitalization, nine (11.8%) in the pre-PROC cohort, and 12 (11.4%) in the post-PROC cohort (P = 0.93).

      Radiation Treatment Characteristics

      Details of PRT regimens are summarized in Table 2. PRT regimens in the post-PROC group were associated with greater proportion of single-fraction and hypofractionated treatments (P < 0.001), lower prescription dose (P < 0.001), and a seven day decrease in median course duration (<0.001).
      Table 2PRT Course Characteristics Before and After Establishment of PROC
      Pre-PROC (n = 76)Post-PROC (n = 105)P-value
      Fractional group, %
       Single-fraction9.220.0<0.001
       Hypofractionated
      (2–5 fractions)10.548.6
      6–10 fractions64.528.6
      >10 Fractions15.82.9
      Prescription dose, cGy
       Median30002000<0.001
       Range800–3750800–3000
      PRT course duration, days
       Median136<0.001
       Range0–450–38
      PRT = palliative radiation therapy; PROC = Palliative Radiation Oncology Consult; cGy = centiGray.

      Specialist-Level Palliative Care Utilization

      Eighty-six patients (81.9%) in the cohort treated after PROC establishment received specialist-level palliative care consults sometime in their hospitalization, compared with 36 patients (47.4%) treated before PROC establishment (P < 0.001).

      Total Hospitalization Costs and LOS

      Unweighted comparisons are presented in Table 3. Median total costs and LOS for the whole sample was $60,358 and 23 days, respectively. A comparison of utilization of pre-PROC and post-PROC finds that the establishment of the service was associated with a significant reduction (P < 0.001) in both total costs and LOSs.
      Table 3Hospital Utilization Data Before and After Establishment of PROC Service
      Whole Sample (n = 181)Pre-PROC (n = 76)Post-PROC (n = 105)P-valueGeneralized Linear Regression Model
      MedianRangeMedianRangeMedianRangeEstimated Treatment Effect95% CIP-value
      Total cost, $60,3586380–620,94376,7926380–346,29650,5827585–620,943<0.001−20,719−37,750 to −36870.017
      LOS, days232–139282–105192–139<0.001−8.5−14 to −3.20.002
      PROC = Palliative Radiation Oncology Consult; LOS = length of stay.
      Treatment effect estimates applying propensity score weights to balance the groups are presented in Table 3. The establishment of PROC is associated with an average reduction of $20,719 in total costs per patient (P = 0.017), and an average reduction of around nine days in LOS (P = 0.002).

      Discussion

      Our results suggest an association between implementation of a PROC service, reduced length of hospital stay, and concomitant cost-savings of hospitalization for advanced cancer patients. This works builds on previously published work in which we investigated the impact of a palliative radiation oncology service model on care of patients with bone metastases.
      • Chang S.
      • May P.
      • Goldstein N.E.
      • et al.
      A palliative radiation oncology consult service's impact on care of advanced cancer patients.
      The main distinction between our previous work and this report is that here we examined costs associated with hospitalizations for patients who started radiation treatment at some point during their hospitalization. The cohort of patients in our investigation represented only the hospitalized patients, whereas our previous study examined both outpatients and inpatients.
      Several studies support the financial benefits in terms of cost-savings of inpatient palliative care utilization, in the form of consultations or palliative care inpatient units.
      • Morrison R.S.
      • Penrod J.D.
      • Cassel J.B.
      • et al.
      Cost savings associated with US hospital palliative care consultation programs.
      • Nathaniel J.D.
      • Garrido M.M.
      • Chai E.J.
      • Goldberg G.
      • Goldstein N.E.
      Cost savings associated with an inpatient palliative care unit: results from the first two years.
      • May P.
      • Garrido M.M.
      • Del Fabbro E.
      • et al.
      Does modality matter? Palliative care unit associated with more cost-avoidance than consultations.
      Similarly, it is well established that shorter RT regimens are more cost-effective in general and have similar efficacy, especially in patients with limited prognoses as often is the case among advanced cancer patients with poor functional status.
      • Chow E.
      • Harris K.
      • Fan G.
      • Tsao M.
      • Sze W.M.
      Palliative radiotherapy trials for bone metastases: a systematic review.
      • Chow E.
      • Zeng L.
      • Salvo N.
      • et al.
      Update on the systematic review of palliative radiotherapy trials for bone metastases.
      • Pollicino C.A.
      • Turner S.L.
      • Roos D.E.
      • O'Brien P.C.
      Costing the components of pain management: analysis of Trans-Tasman Radiation Oncology Group trial (TROG 96.05): one versus five fractions for neuropathic bone pain.
      • Konski A.
      • James J.
      • Hartsell W.
      • et al.
      Economic analysis of radiation therapy oncology group 97-14: multiple versus single fraction radiation treatment of patients with bone metastases.
      • van den Hout W.B.
      • van der Linden Y.M.
      • Steenland E.
      • et al.
      Single- versus multiple-fraction radiotherapy in patients with painful bone metastases: cost–utility analysis based on a randomized trial.
      • Hartsell W.F.
      • Konski A.A.
      • Lo S.S.
      • Hayman J.A.
      Single fraction radiotherapy for bone metastases: clinically effective, time efficient, cost conscious and still underutilized in the United States?.
      To our knowledge, no previous study has examined the association between costs and a procedurally-oriented specialty service such as radiation oncology when key components of generalist-level palliative medicine are integrated within the scope of the palliative radiation oncology team's practice. The estimated saving of $20,719 per patient is equivalent to approximately 25% reduction of pre-PROC total costs, a proportion consistent with cost-savings found in the analyses of other hospital-based palliative care service interventions.
      • May P.
      • Normand C.
      • Morrison R.S.
      Economic impact of hospital inpatient palliative care consultation: review of current evidence and directions for future research.
      Specialist-level palliative care consults were significantly increased for patients in the post-PROC cohort. Patients were routinely referred by the PROC team for additional management of symptom control and goals of care planning. Patients also may had been referred through pathways external of the PROC service, such as the nascent creation of a criteria-based trigger system for patients on the inpatient solid tumor service at Mount Sinai Hospital.
      • Adelson K.
      • Paris J.
      • Horton J.R.
      • et al.
      Standardized criteria for palliative care consultation on a solid tumor oncology service reduces downstream health care use.
      Analyses of how palliative care consultation services reduce costs for inpatients have suggested that savings accrue through a combination of reduced intensity of stay and earlier discharge.
      • May P.
      • Garrido M.M.
      • Cassel J.B.
      • et al.
      Prospective cohort study of hospital palliative care teams for inpatients with advanced cancer: earlier consultation is associated with larger cost-saving effect.
      • May P.
      • Garrido M.M.
      • Cassel J.B.
      • et al.
      Cost analysis of a prospective multi-site cohort study of palliative care consultation teams for adults with advanced cancer: where do cost-savings come from?.
      Reduction of costs observed in our investigation was therefore likely multifactorial, resulting from a synergistic effect of increased generalist-level and specialty-level palliative care utilization that led to improved symptom management, more timely goals-of-care planning before initiation of RT, greater usage of shorter radiation regimens in appropriate patients, and consequently, reduced length of hospital stay.
      • Chang S.
      • May P.
      • Goldstein N.E.
      • et al.
      A palliative radiation oncology consult service's impact on care of advanced cancer patients.
      A previously published review of inpatient radiation treatments at a single center has shown that patients undergoing RT during a hospitalization often have prolonged LOSs because of clinical and non-clinical reasons that make discharge planning during a treatment course difficult.
      • Pintova S.
      • Holcombe R.F.
      • Blacksburg S.
      • Friedlander P.
      Impact of inpatient radiation on length of stay and health care costs.
      Ours is the first study to correlate the relationship between shorter radiation regimens for palliative radiation treatments and a reduction in length of hospital stay.
      This was an observational cohort study from a single institution. Although care has been taken to minimize observed confounding and membership of pre- and post-PROC groups (e.g., exposure assignation) should not have been associated with patient-level factors, it is nevertheless possible that our reported results derive as a consequence of unobserved factors or biases. Given the time-sensitive nature of our study, patients observed at later time points may be affected by predictors external to our PROC service, such as advancements in cancer therapy and radiotherapy techniques, general trend toward earlier palliative care consults, and systemic changes in health care to warrant decreased resource utilization. In addition, all utilization outcomes of interest are for inpatient hospital stay only and do not take into account costs incurred after discharge, for example, it is possible that earlier discharge results in higher formal and informal costs for families of patients.
      • Johnston B.M.
      • Normand C.
      • May P.
      Economics of palliative care: measuring the full value of an intervention.
      Although our results were still significant, the small sample size may have contributed to the wide confidence interval yielded by our general linear regression model for total cost. A larger sample would have allowed for control over more covariates (e.g., other primary cancer types, admission reason, etc.), which in turn might have improved estimate precision.
      There was a particular risk of unobserved heterogeneity in palliative care studies as concerns proximity to death, which may have impacted treatment choices and so utilization patterns.
      • Morrison R.S.
      • Penrod J.D.
      • Cassel J.B.
      • et al.
      Cost savings associated with US hospital palliative care consultation programs.
      However, stratifying by discharge status or otherwise controlling for mortality was also sub-optimal because this was not a baseline variable, but an outcome that treatment can theoretically impact.
      • May P.
      • Garrido M.M.
      • Cassel J.B.
      • Morrison R.S.
      • Normand C.
      Using length of stay to control for unobserved heterogeneity when estimating treatment effect on hospital costs with observational data: issues of reliability, robustness, and usefulness.
      In our study, in-hospital death was observed in 11.8% of pre-PROC patients and 11.4% of post-PROC patients (P = 0.93). In-hospital death was not significantly different between comparison and exposure groups, but it was possible that post-discharge survival differed across groups.
      Our primary outcome of interest was total cost of hospital admission, including both direct and indirect costs. In the short run, the PROC intervention will not impact indirect costs, which the hospital still has to pay for, and so our results potentially exaggerate the dollar impact of the service.
      • Smith T.J.
      • Cassel J.B.
      Cost and non-clinical outcomes of palliative care.
      However, economics best practice guidelines always recommend taking the widest possible perspective and in the long run the indirect costs associated with pre-PROC will indeed be defrayed in line with reduced direct costs.
      • Drummond M.F.
      • Jefferson T.O.
      Guidelines for authors and peer reviewers of economic submissions to the BMJ. The BMJ economic evaluation working party.
      Our radiation consult service was restricted to one institution and involved the multidisciplinary management of advanced cancer patients under a team of oncologists, palliative care doctors, pain specialists, and radiation oncologists. The practice model of our service is not standard of care at most other institutions hospitals. A large proportion of patients given PRT at our institution had a hematological malignancy (e.g., multiple myeloma). Over the last decade, inpatient admissions and all-cause health care costs have increased for the treatment of multiple myeloma.
      • Fonseca R.
      • Abouzaid S.
      • Bonafede M.
      • et al.
      Trends in overall survival and costs of multiple myeloma, 2000-2014.
      Other institutions treating for solid malignancies may see different trends in downstream health care utilization. However, given that our pilot program yielded promising results relating to improvements of quality of life and cost-effectiveness, we strongly believe other centers who provide radiotherapy services to inpatients may benefit from our experience and find ways to apply principles of palliative care that could lead to the similar improvements that we found.
      In conclusion, establishment of a PROC service integrated with palliative care was associated with reduced LOS and cost-savings of hospitalization for advanced cancer patients. Reduction of costs was likely multifactorial, resulting from increased palliative care utilization, shorter radiation regimens, and reduced length of hospital stay. We anticipate that other radiotherapy centers may benefit from adopting principles from our clinical service model.

      Disclosures and Acknowledgments

      This study was supported by grant no. 5P30AG028741 from the Claude D. Pepper Older Americans Independence Center at the National Institute of Aging/National Institutes of Health, a career development grant from the National Palliative Care Research Center, and a seed grant from the American Medical Association Foundation.
      Dr. Juan Wisnivesky is a member of the research board at EHE International, and has received consulting honoraria from Merck, AstraZeneca, and Quintiles, and research grants from Sanofi and Quorum Consulting. No other authors have competing financial interests to disclose.

      References

        • Bekelman J.E.
        • Epstein A.J.
        • Emanuel E.J.
        Single- vs multiple-fraction radiotherapy for bone metastases from prostate cancer.
        JAMA. 2013; 310: 1501-1502
        • Lutz S.
        • Spence C.
        • Chow E.
        • Janjan N.
        • Connor S.
        Survey on use of palliative radiotherapy in hospice care.
        J Clin Oncol. 2004; 22: 3581-3586
        • Hess G.
        • Barlev A.
        • Chung K.
        • Hill J.W.
        • Fonseca E.
        Cost of palliative radiation to the bone for patients with bone metastases secondary to breast or prostate cancer.
        Radiat Oncol. 2012; 7: 168
        • Fischberg D.
        • Bull J.
        • Casarett D.
        • et al.
        Five things physicians and patients should question in hospice and palliative medicine.
        J Pain Symptom Manage. 2013; 45: 595-605
        • Li S.
        • Peng Y.
        • Weinhandl E.D.
        • et al.
        Estimated number of prevalent cases of metastatic bone disease in the US adult population.
        Clin Epidemiol. 2012; 4: 87-93
        • Chow E.
        • Harris K.
        • Fan G.
        • Tsao M.
        • Sze W.M.
        Palliative radiotherapy trials for bone metastases: a systematic review.
        J Clin Oncol. 2007; 25: 1423-1436
        • Jones J.A.
        • Simone 2nd, C.B.
        Palliative radiotherapy for advanced malignancies in a changing oncologic landscape: guiding principles and practice implementation.
        Ann Palliat Med. 2014; 3: 192-202
        • Chow E.
        • Zeng L.
        • Salvo N.
        • et al.
        Update on the systematic review of palliative radiotherapy trials for bone metastases.
        Clin Oncol (R Coll Radiol). 2012; 24: 112-124
        • Fesinmeyer M.D.
        • Mehta V.
        • Tock L.
        • et al.
        Completion of radiotherapy for local and regional head and neck cancer in Medicare.
        Arch Otolaryngol Head Neck Surg. 2009; 135: 860-867
        • Lutz S.
        • Berk L.
        • Chang E.
        • et al.
        Palliative radiotherapy for bone metastases: an ASTRO evidence-based guideline.
        Int J Radiat Oncol Biol Phys. 2011; 79: 965-976
        • Hartsell W.F.
        • Scott C.B.
        • Bruner D.W.
        • et al.
        Randomized trial of short- versus long-course radiotherapy for palliation of painful bone metastases.
        J Natl Cancer Inst. 2005; 97: 798-804
        • Rutter C.E.
        • Yu J.B.
        • Wilson L.D.
        • Park H.S.
        Assessment of national practice for palliative radiation therapy for bone metastases suggests marked underutilization of single-fraction regimens in the United States.
        Int J Radiat Oncol Biol Phys. 2015; 91: 548-555
        • Schreiber D.
        • Safdieh J.
        • Becker D.J.
        • Schwartz D.
        Patterns of care and survival outcomes of palliative radiation for prostate cancer with bone metastases: comparison of ≤5 fractions to ≥10 fractions.
        Ann Palliat Med. 2017; 6: 55-65
        • Chang S.
        • May P.
        • Goldstein N.E.
        • et al.
        A palliative radiation oncology consult service's impact on care of advanced cancer patients.
        J Palliat Med. 2017; https://doi.org/10.1089/jpm.2017.0372
        • Taheri P.A.
        • Butz D.
        • Griffes L.C.
        • Morlock D.R.
        • Greenfield L.J.
        Physician impact on the total cost of care.
        Ann Surg. 2000; 231: 432-435
        • Weinstein M.C.
        • Siegel J.E.
        • Gold M.R.
        • Kamlet M.S.
        • Russell L.B.
        Recommendations of the panel on cost-effectiveness in health and medicine.
        JAMA. 1996; 276: 1253-1258
        • Charlson M.E.
        • Pompei P.
        • Ales K.L.
        • MacKenzie C.R.
        A new method of classifying prognostic comorbidity in longitudinal studies: development and validation.
        J Chronic Dis. 1987; 40: 373-383
      1. Bureau of Labor Statistics: CPI Database (All Urban Consumers). US Department of Labor, 2017 (Available from) (Accessed December 4, 2017)
        • Garrido M.M.
        • Kelley A.S.
        • Paris J.
        • et al.
        Methods for constructing and assessing propensity scores.
        Health Serv Res. 2014; 49: 1701-1720
        • Green K.M.
        • Stuart E.A.
        Examining moderation analyses in propensity score methods: application to depression and substance use.
        J Consult Clin Psychol. 2014; 82: 773-783
        • Jones A.M.
        Models for health care.
        in: Clements M.P. Hendry D.F. The Oxford Handbook of Economic Forecasting. Oxford University Press, New York2011
        • Brooks J.M.
        • Ohsfeldt R.L.
        Squeezing the balloon: propensity scores and unmeasured covariate balance.
        Health Serv Res. 2013; 48: 1487-1507
        • Morrison R.S.
        • Penrod J.D.
        • Cassel J.B.
        • et al.
        Cost savings associated with US hospital palliative care consultation programs.
        Arch Intern Med. 2008; 168: 1783-1790
        • Nathaniel J.D.
        • Garrido M.M.
        • Chai E.J.
        • Goldberg G.
        • Goldstein N.E.
        Cost savings associated with an inpatient palliative care unit: results from the first two years.
        J Pain Symptom Manage. 2015; 50: 147-154
        • May P.
        • Garrido M.M.
        • Del Fabbro E.
        • et al.
        Does modality matter? Palliative care unit associated with more cost-avoidance than consultations.
        J Pain Symptom Manage. 2018; 55: 766-774.e4
        • Pollicino C.A.
        • Turner S.L.
        • Roos D.E.
        • O'Brien P.C.
        Costing the components of pain management: analysis of Trans-Tasman Radiation Oncology Group trial (TROG 96.05): one versus five fractions for neuropathic bone pain.
        Radiother Oncol. 2005; 76: 264-269
        • Konski A.
        • James J.
        • Hartsell W.
        • et al.
        Economic analysis of radiation therapy oncology group 97-14: multiple versus single fraction radiation treatment of patients with bone metastases.
        Am J Clin Oncol. 2009; 32: 423-428
        • van den Hout W.B.
        • van der Linden Y.M.
        • Steenland E.
        • et al.
        Single- versus multiple-fraction radiotherapy in patients with painful bone metastases: cost–utility analysis based on a randomized trial.
        J Natl Cancer Inst. 2003; 95: 222-229
        • Hartsell W.F.
        • Konski A.A.
        • Lo S.S.
        • Hayman J.A.
        Single fraction radiotherapy for bone metastases: clinically effective, time efficient, cost conscious and still underutilized in the United States?.
        Clin Oncol (R Coll Radiol). 2009; 21: 652-654
        • May P.
        • Normand C.
        • Morrison R.S.
        Economic impact of hospital inpatient palliative care consultation: review of current evidence and directions for future research.
        J Palliat Med. 2014; 17: 1054-1063
        • Adelson K.
        • Paris J.
        • Horton J.R.
        • et al.
        Standardized criteria for palliative care consultation on a solid tumor oncology service reduces downstream health care use.
        J Oncol Pract. 2017; 13: e431-e440
        • May P.
        • Garrido M.M.
        • Cassel J.B.
        • et al.
        Prospective cohort study of hospital palliative care teams for inpatients with advanced cancer: earlier consultation is associated with larger cost-saving effect.
        J Clin Oncol. 2015; 33: 2745-2752
        • May P.
        • Garrido M.M.
        • Cassel J.B.
        • et al.
        Cost analysis of a prospective multi-site cohort study of palliative care consultation teams for adults with advanced cancer: where do cost-savings come from?.
        Palliat Med. 2017; 31: 378-386
        • Pintova S.
        • Holcombe R.F.
        • Blacksburg S.
        • Friedlander P.
        Impact of inpatient radiation on length of stay and health care costs.
        J Community Support Oncol. 2015; 13: 399-404
        • Johnston B.M.
        • Normand C.
        • May P.
        Economics of palliative care: measuring the full value of an intervention.
        J Palliat Med. 2017; 20: 222-224
        • May P.
        • Garrido M.M.
        • Cassel J.B.
        • Morrison R.S.
        • Normand C.
        Using length of stay to control for unobserved heterogeneity when estimating treatment effect on hospital costs with observational data: issues of reliability, robustness, and usefulness.
        Health Serv Res. 2016; 51: 2020-2043
        • Smith T.J.
        • Cassel J.B.
        Cost and non-clinical outcomes of palliative care.
        J Pain Symptom Manage. 2009; 38: 32-44
        • Drummond M.F.
        • Jefferson T.O.
        Guidelines for authors and peer reviewers of economic submissions to the BMJ. The BMJ economic evaluation working party.
        BMJ. 1996; 313: 275-283
        • Fonseca R.
        • Abouzaid S.
        • Bonafede M.
        • et al.
        Trends in overall survival and costs of multiple myeloma, 2000-2014.
        Leukemia. 2017; 31: 1915-1921