The impact of pre-treatment muscle strength and physical performance on treatment modification in older adults with cancer following comprehensive geriatric assessment.

BACKGROUND: Grip strength (GS) and the short physical performance battery (SPPB) have been shown to predict clinical outcomes in older adults with cancer. However, whether pre-treatment GS and SPPB impact treatment decisions following comprehensive geriatric assessment (CGA) is poorly understood. Our objective was to assess the impact of low GS and/or SPPB on treatment modification to initially proposed treatment plans in older adults with cancer following CGA. METHODS: This was a retrospective cohort study of older adults who had undergone CGA before receiving cancer treatment. Data were retrieved from a prospective database in an academic cancer centre and medical records. Treatment modification following CGA was defined as reduced treatment intensity or transition from active treatment to supportive care. Multivariable logistic regression assessed the impact of pre-treatment GS and SPPB on treatment modification following CGA. RESULTS: In total, 515 older adults (mean age: 80.7y) who had undergone CGA prior to cancer treatment were included. Low muscle strength and/or physical performance was observed in 66.4% of participants. Treatment was modified in 49.5% of the cohort following CGA. Low GS and/or SPPB combined was predictive of treatment modification (OR = 1.77, 95%CI = 1.07-2.90, P = 0.025) in multivariable analysis. Additional predictors of treatment modification included palliative treatment intent, comorbidities and malnutrition. CONCLUSIONS: Low GS and/or SPPB combined prior to cancer treatment predicts treatment modification in older adults with cancer and may be useful in treatment decision-making. Management of poor muscle strength and physical performance should be offered to optimize patient care and potentially improve treatment outcomes.

Impact of patients' sex in assessments and treatment recommendations in an older adult cancer clinic.

INTRODUCTION: In multiple settings, sex disparities have been seen in diagnosis, treatment, and outcomes. This study sought to determine whether there are sex differences in a geriatric oncology clinic concerning results of the comprehensive geriatric assessment (CGA) and treatment recommendations. MATERIALS AND METHODS: This is a retrospective cohort study including patients ≥65 years old referred for consultation on cancer treatment decision-making who underwent a CGA between July 2015 and December 2020, in a single Canadian academic geriatric oncology (GO) clinic. We examined differences by sex, stratified by disease site, stage, treatment intent, CGA results by domain, final treatment plan, and referrals for abnormal CGA findings. Differences were assessed using chi-square, Fisher's exact, or t-test as appropriate. Multivariate logistic regression was performed to examine whether sex impacted recommendations to reduce treatment intensity. RESULTS: In the study period, 328 patients were assessed in the GO clinic (mean age 81 years). The most common cancer types were gastrointestinal (42.1%), hematologic (18.3%), and head and neck (17.3%). More males than females were assessed in the GO clinic (62.2% versus 37.8%, respectively). This proportion did not change over time (p = 0.58). The GO clinic recommended to reduce treatment intensity in 140 cases (42.7%), with no difference between sexes in adjusted models (43.6% of females and 42.2% of males, p = 0.80). There were no differences in any CGA domain by sex. There were also no differences in referrals made by the GO clinic to optimize abnormal CGA domains by sex. DISCUSSION: Sex itself did not impact treatment decision-making, nor referrals to optimize abnormal CGA domains in our GO clinic using CGA-based care.

Implementation of clinical recommendations from the geriatric oncology clinic.

INTRODUCTION: While evidence on the value of routine geriatric assessment (GA) in cancer care for older patients is growing, there is limited data on the geriatric oncology (GO) clinic's specific recommendations and how they are implemented. In this study, we aimed to assess and evaluate the implementation of recommendations from the GO clinic at Princess Margaret Cancer Center, Toronto, Canada, within six months of the initial visit. MATERIALS AND METHODS: A retrospective chart review was conducted on 100 consecutive adults age 65+ visiting the GO clinic from 2018 to 2019. For each patient, we evaluated the number and type of recommendations from the GO clinic. Recommendations were grouped based on clinical judgement. Of the recorded recommendations, we measured the rate of implementation within six months of the initial visit including who implemented the recommendations and why recommendations were not implemented. Data were analyzed using descriptive statistics. RESULTS: One hundred patients visiting the GO clinic (mean age of 80.5 years, 62% male, 52% with planned curative intent, with the genitourinary site being most common) received a median of six recommendations (range of 2-12), regardless of sex, cancer stage, cancer site, and treatment intent. Medication optimization (27%), patient education (26%), and referral to allied health (14%) were the top recommendations from the GO clinic. At six-month follow-up, 83% of all recommendations were implemented, of which 94% were performed by the GO clinic team. Patient education was implemented at a 100% rate by the GO clinic at the time of initial assessment. GO follow-up visit and other diagnostic tests (hearing test, vision test) were the recommendations with the lowest implementation rates, at 51% and 31%, respectively. The most common reasons for recommendations not being implemented were patient transfer to palliative care/death and patient declining recommendations due to busy appointment schedules. DISCUSSION: A median of six recommendations were made per patient. The vast majority of recommendations were implemented, predominantly by the GO team. Overall, the study helps evaluate recommendations provided to patients visiting GO clinics, identify potential gaps, and assist with resource planning for optimal cancer care for older adults.

Do clinicians address impairments in muscle strength and physical performance for older adults with cancer?

INTRODUCTION: Low physical function is associated with adverse outcomes in older adults with cancer, but evidence on real-world, clinical management of low physical function in oncology is lacking. We explored whether impairments in muscle strength and/or physical performance triggered downstream management by clinicians, the types of recommended strategies, and the reasons for not providing a referral/strategy for addressing such impairments in older cancer survivors prior to treatment. MATERIALS AND METHODS: We conducted a retrospective, cross-sectional study of older adults who completed a comprehensive geriatric assessment (CGA) prior to cancer treatment in a tertiary cancer centre. Muscle strength and physical performance were assessed through grip strength and the Short Physical Performance Battery (SPPB), respectively. Patients who exhibited an impairment in grip strength and/or SPPB were classified as having abnormal objective physical function. Downstream management strategies and clinicians' reasons for not providing referrals were retrieved from clinical notes and an institutional database. RESULTS: In total, 515 older adults (mean age: 80.7 years) were included. Low grip strength and/or SPPB combined was observed in 66.4% (n = 342) of participants, of whom 54.1% (n = 185) received an acceptable intervention. However, 41.2% (n = 141) were not provided with a referral/strategy by clinicians to address such impairments following CGA. No reasons were provided in clinical notes for not addressing impairments in physical function for 100 participants (70.9%). DISCUSSION: Many older adults with cancer have impaired physical function prior to treatment. However, we found that such impairments are not systematically addressed by clinicians, and documentation was often suboptimal, identifying gaps in patient care that need to be addressed.

Implementing geriatric assessment and management for older Canadians with cancer: Adherence to and satisfaction with the intervention, results of the 5C study.

INTRODUCTION: Geriatric assessment and management (GAM) is recommended by professional organizations and recently several randomized controlled trials (RCTs) demonstrated benefits in multiple health outcomes. GAM typically leads to one or more recommendations for the older adult on how to optimize their health. However, little is known about how well recommendations are adhered to. Understanding these issues is vital to designing GAM trials and clinical programs. Therefore, the aim of this study was to examine the number of GAM recommendations made and adherence to and satisfaction with the intervention in a multicentre RCT of GAM for older adults with cancer. MATERIALS AND METHODS: The 5C study was a two-group parallel RCT conducted in eight hospitals across Canada. Each centre kept a detailed recruitment and retention log. The intervention teams documented adherence to their recommendations. Medical records were also reviewed to assess which recommendations were adhered to. Twenty-three semi-structured interviews were conducted with 12 members of the intervention teams and 11 oncology team members to assess implementation of the study and the intervention. RESULTS: Of the 350 participants who were enrolled, 173 were randomized to the intervention arm. Median number of recommendations was seven. Mean adherence to recommendations based on the GAM was 69%, but it varied by type of recommendation, ranging from 98% for laboratory tests to 28% for psychosocial/psychiatry oncology referrals. There was no difference in the number of recommendations or non-adherence to recommendations by sex, level of frailty, or functional status. Oncologists and intervention team members were satisfied with the study implementation and intervention delivery. DISCUSSION: Adherence to recommendations was variable. Adherence to laboratory investigations and further imaging were generally high but much lower for recommendations regarding psychosocial support. Further collaborative work with older adults with cancer is needed to understand how to optimize the intervention to be consistent with patient goals, priorities, and values to ensure maximal impact on health outcomes.

Impact of Geriatric Assessment and Management on Quality of Life, Unplanned Hospitalizations, Toxicity, and Survival for Older Adults With Cancer: The Randomized 5C Trial.

PURPOSE: American Society of Clinical Oncology recommends that older adults with cancer being considered for chemotherapy receive geriatric assessment (GA) and management (GAM), but few randomized controlled trials have examined its impact on quality of life (QOL). PATIENTS AND METHODS: The 5C study was a two-group parallel 1:1 single-blind multicenter randomized controlled trial of GAM for 6 months versus usual oncologic care. Eligible patients were age 70+ years, diagnosed with a solid tumor, lymphoma, or myeloma, referred for first-/second-line chemotherapy or immunotherapy or targeted therapy, and had an Eastern Cooperative Oncology Group performance status of 0-2. The primary outcome QOL was measured with the global health scale of the European Organisation for the Research and Treatment of Cancer QOL questionnaire and analyzed with a pattern mixture model using an intent-to-treat approach (at 6 and 12 months). Secondary outcomes included functional status, grade 3-5 treatment toxicity; health care use; satisfaction; cancer treatment plan modification; and overall survival. RESULTS: From March 2018 to March 2020, 350 participants were enrolled. Mean age was 76 years and 40.3% were female. Fifty-four percent started treatment with palliative intent. Eighty-one (23.1%) patients died. GAM did not improve QOL (global QOL of 4.4 points [95% CI, 0.9 to 8.0] favoring the control arm). There was also no difference in survival, change in treatment plan, unplanned hospitalization/emergency department visits, and treatment toxicity between groups. CONCLUSION: GAM did not improve QOL. Most intervention group participants received GA on or after treatment initiation per patient request. Considering recent completed trials, GA may have benefit if completed before treatment selection. The COVID-19 pandemic may have affected our QOL outcome and intervention delivery for some participants.

Objective measures of physical function and their association with cognitive impairment in older adults with cancer prior to treatment.

INTRODUCTION: Objective measures of physical function are associated with cognitive function in community-dwelling older adults. Many older adults experience cognitive declines prior to cancer treatment initiation. Thus, it is unclear whether the association between low physical function and cognitive impairment is generalizable to older adults with cancer prior to treatment. Our objective was to examine whether objective measures of physical function were associated with cognitive impairment in geriatric oncology patients prior to treatment. MATERIALS AND METHODS: We used prospectively collected data from an institutional database within a cancer centre and electronic medical records of older adults who had undergone a geriatric assessment before cancer treatment. Objective measures of physical function included grip strength and the Short Physical Performance Battery (SPPB). Cognitive impairment was assessed via the Mini-Cog. Multivariable logistic regression was used to determine whether grip strength and SPPB were associated with cognitive impairment prior to cancer treatment in all patients, as well as in patients with moderate-to-high comorbidity as part of a sensitivity analysis. RESULTS: A total of 386 older adults (mean age 80.9 years) were included in the analysis. Most participants (65.3%) had low grip strength and/or low SPPB, whereas 42.2% were cognitively impaired. Neither low grip strength (odds ratio [OR] = 1.55, 95% confidence interval [CI] = 0.92-2.63, p = 0.097) nor low SPPB (OR = 1.29, 95%CI = 0.69-2.42, p = 0.41) alone or combined (OR = 1.05, 95%CI = 0.59-1.88, p = 0.85) were significantly associated with cognitive impairment in multivariable analyses of all patients. However, low SPPB was significantly associated with cognitive impairment in the sensitivity analysis restricted to patients with moderate-to-high comorbidity (OR = 4.05, 95%CI = 1.50-10.95, p = 0.006). Dependence in one or more instrumental activities of daily living [IADLs] was consistently associated with cognitive impairment in the main and sensitivity analyses. DISCUSSION: Low physical performance and IADL dependence are associated with cognitive impairment in patients with moderate-to-high comorbidity prior to cancer treatment. Scrutiny is advised for these patients to assess for possible cognitive impairment. Larger studies are warranted to confirm our findings.

Economic Evaluation of a Geriatric Oncology Clinic.

Geriatric assessment (GA) is supported by recent trials and guidelines yet rarely implemented due to a lack of resources. We performed an economic evaluation of a geriatric oncology clinic. Pre-GA proposed treatments and post-GA actual treatments were obtained from a detailed chart review of patients seen at a single academic centre. GA-based costs for investigations and referrals were calculated. Unit costs were obtained for surgical, radiation, systemic therapy, laboratory, imaging, physician, nursing, and allied health care (all in 2019 Canadian dollars). A six-month time horizon and government payer perspective were used. Consecutive patients aged 65 years or older (n = 152, mean age 82 y) and referred in the pre-treatment setting between July 2016 and June 2018 were included. Treatment plans were modified for 51% of patients. Costs associated with planned treatment were CAD 3,655,015. Costs associated with GA and related interventions were CAD 95,798. Final treatment costs were CAD 2,436,379. Net savings associated with the clinic were CAD 1,122,837, or CAD 7387 per patient seen. Findings were robust in multiple sensitivity analyses. Combined with mounting trial data demonstrating the clinical benefits of GA, our data can inform a strong business case for geriatric oncology clinics in health care environments similar to ours, but additional studies in diverse health care settings are warranted.

Quality of care of consultations from the geriatric oncology clinic: "Are we addressing the needs of patients?"

INTRODUCTION: A comprehensive geriatric assessment (CGA) is recommended for older adults with cancer in the pre-treatment setting to optimize care. A CGA systematically evaluates multiple domains to develop a holistic view of the patient's health and facilitate timely interventions to ameliorate patient outcomes. For a CGA to be most effective, optimization of each abnormal domain should occur. However, there is limited literature exploring this issue. MATERIALS AND METHODS: Consultations of patients seen in a Geriatric Oncology clinic from June 2015 to June 2018 were reviewed. The percentage of "no recommendations made" in the consultation letter following the identification of impairment in each of eight geriatric domains was calculated. Trends over time were examined by stratifying the data into three periods ("Year 1", "Year 2", and "Year 3") and conducting a logistic regression analysis. RESULTS: A total of 365 consultation notes were reviewed. The patients were predominately older (mean age 79.9 years), male (66.9%), with genitourinary (38.6%) or gastrointestinal (23.3%) cancers. The most common stage was metastatic (40.6%). The most common treatment intent and modality were palliative (50.4%) and hormonal (50.9%), respectively. The geriatric domains that had the greatest frequency of impairments were medication optimization (76.2%), functional status (68.8%), and falls risk (64.9%). The domains that had the highest frequency of "no recommendations made" following identification of impairment were nutrition (39.8%), social support (39.5%), and mood (26.4%). The prevalence of "no recommendations made" decreased over time in social support (54.6% in Year 1 to 27.8% in Year 3, p = 0.043) and possibly nutrition (53.1% in Year 1 to 34.3% in Year 3, p = 0.088) but not for mood (p = 0.64). CONCLUSIONS: Nutrition, social supports and mood were the CGA domains with the highest proportion of "no recommendations made" following an identification of impairment. This is the first quality assurance study to identify social supports, mood, and nutrition domains as less frequently addressed following an identification of an impairment amongst older patients with cancer. Subsequent prospective research is required to understand reasons for these observations and identification of barriers to address these geriatric domains amongst older adults with cancer.

An International Expert Delphi Consensus to Develop Dedicated Geriatric Radiation Oncology Curriculum Learning Outcomes.

PURPOSE: The management of older adults with cancer is rapidly becoming a significant challenge in radiation oncology (RO) practice. The education of future radiation oncologists in geriatric oncology is fundamental to ensuring that older adults receive high-quality care. Currently RO trainees receive little training and education in geriatric oncology. The objective of this study was to define core geriatric RO curriculum learning outcomes relevant to RO trainees worldwide. METHODS AND MATERIALS: A 2-stage modified Delphi consensus was conducted. Stage 1 involved the formation of an expert reference panel (ERP) of multiprofessional experts in geriatric oncology and/or RO and the compilation of a potential geriatric RO learning outcomes set. Stage 2 involved 3 iterative rounds: round 1 and round 2 (both online surveys), and an intervening ERP round. These aimed at identifying and refining ideal geriatric RO learning outcomes. Invited participants for round 1 and 2 included oncology health care professionals with expertise across RO, geriatric oncology, and/or education and consumers. Predefined Delphi consensus definitions were applied to the results of rounds 1 and 2. RESULTS: An ERP of 11 experts in geriatric oncology and/or RO was formed. Seventy potential knowledge- and skill-based learning outcomes were identified. In round 1, 103 of 179 invited eligible Delphi participants completed the survey (58% response rate). The ERP round was conducted, resulting in the exclusion of 28 learning outcomes. In round 2, 54 of 103 completed the survey (52% response rate). This identified a final total of 33 geriatric RO learning outcomes. CONCLUSIONS: The geriatric RO learning outcomes described in this study form an international consensus that can inform RO training bodies worldwide. This represents the first fundamental step in developing a global educational framework aimed at improving RO trainee knowledge and skills in geriatric oncology.

Role of the vulnerable elders survey-13 screening tool in predicting treatment plan modification for older adults with cancer.

BACKGROUND: The Vulnerable Elders Survey (VES-13) is commonly used to identify older patients who may benefit from Comprehensive Geriatric Assessment (CGA) prior to cancer treatment. The optimal cut point of the VES-13 to identify those whose final oncologic treatment plan would change after CGA is unclear. We hypothesized that patients with high positive VES-13 scores (7-10)have a higher likelihood of a change in treatment compared to low positive scores (3-6). METHODS: Retrospective review of a customized database of all patients seen for pre-treatment assessment in an academic geriatric oncology clinic from June 2015 to June 2019. Various VES-13 cut points were analyzed to identify those individuals whose treatment was modified after CGA. Area under the curve (AUC) was calculated and subgroups of patients treated locally or systemically were also examined to determine if performance varied by treatment modality. RESULTS: We included 386 patients with mean age 81, 58% males. Gastrointestinal cancer was the most common site (31%) and 60% were planned to receive curative treatment. The final treatment plan was modified in 59% overall, with 52.7% modified with VES-13 scores 7-10, 50.8% with scores 3-6 and 28.1% with scores <3 (P = 0.002). VES-13 performance in predicting treatment modification was similar for cut points 3 (AUC 0.58), 4 (0.59), 5 (0.59), and 6 (0.59) and in those considering local treatment vs. chemotherapy. CONCLUSIONS: A positive VES-13 score was associated with final oncologic treatment plan modification. A high positive score was not superior to the conventional cut point of ≥3.

Teaching Professionalism in Postgraduate Medical Education: A Systematic Review.

PURPOSE: This systematic review sought to summarize published professionalism curricula in postgraduate medical education (PGME) and identify best practices for teaching professionalism. METHOD: Three databases (MEDLINE, Embase, ERIC) were searched for articles published from 1980 through September 7, 2017. English-language articles were included if they (1) described an educational intervention addressing professionalism, (2) included postgraduate medical trainees, and (3) evaluated professionalism outcomes. RESULTS: Of 3,383 articles identified, 50 were included in the review. The majority evaluated pre- and posttests for a single group (24, 48%). Three (6%) were randomized controlled trials. The most common teaching modality was small-group discussions (28, 56%); other methods included didactics, reflection, and simulations. Half (25, 50%) used multiple modalities. The professionalism topics most commonly addressed were professional values/behavior (42, 84%) and physician well-being (23, 46%). Most studies measured self-reported outcomes (attitude and behavior change) (27, 54%). Eight (16%) evaluated observed behavior and 3 (6%) evaluated patient outcomes. Of 35 studies that evaluated statistical significance, 20 (57%) reported statistically significant positive effects. Interventions targeting improvements in knowledge were most often effective (8/12, 67%). Curriculum duration was not associated with effectiveness. The 45 quantitative studies were of moderate quality (Medical Education Research Study Quality Instrument mean score = 10.3). CONCLUSIONS: Many published curricula addressing professionalism in PGME are effective. Significant heterogeneity in curricular design and outcomes assessed made it difficult to synthesize results to identify best practices. Future work should build upon these curricula to improve the quality and validity of professionalism teaching tools.

Downstream consequences of abnormal cognitive screening in older adults seen pretreatment in a geriatric oncology clinic.

INTRODUCTION: Although screening for cognitive impairment (CI) is an important part of a comprehensive geriatric assessment (CGA), little is known about the downstream work-up of abnormal screening or its impact on cancer treatment. We characterized the downstream workup in diagnosing CI and its impact on cancer treatment decision-making. METHODS: Patients who underwent a pre-treatment CGA at an academic Geriatric Oncology (GO) clinic between July 2015 and June 2018 and had a positive Mini-Cog (≤ 3 out of 5) screen were included. Data were collected from medical charts and database review. Analyses were primarily descriptive. RESULTS: Of 82 patients seen in the pre-treatment setting, 46 (56.1%) had a positive Mini-Cog screen. Of those, 12 (26.1%) were diagnosed with dementia, 8 (17.4%) were diagnosed with mild cognitive impairment and 10 (21.7%) had CI not otherwise specified. Although 46 patients had a positive screen, only 30 patients (65.2%) were classified as cognitively "abnormal" in the GO team final assessment. Change to oncologic treatment due to CI was seen in 12 (40.0%) cases. Increased delirium risk was identified in 9 (75.0%) of 12 surgical cases; however, delirium prevention was only recommended in 5 cases (55.6%). Strategies to optimize patients with CI included targeting falls prevention (n = 13), home/personal safety (n = 7), medication safety (n = 7), and nutrition (n = 6). Pharmacotherapy for cognition was not recommended in any case. CONCLUSION: Undiagnosed CI is prevalent in the GO setting and influenced treatment in 40.0% of cases. Gaps were identified in clinician and patient/caregiver education around delirium risk. Addressing these issues may improve patient care.

Performance of the Vulnerable Elders Survey 13 screening tool in identifying cancer treatment modification after geriatric assessment in pre-treatment patients: A retrospective analysis.

PURPOSE: Geriatric assessment (GA) is recommended for older adults ≥ 70 years with cancer to guide treatment selection. Screening tools such as the Vulnerable Elders Survey (VES-13) and G6 have been used to identify patients at highest need of GA. Whether either tool predicts a change in oncologic treatment following GA is unclear. METHODS: Patients attending a geriatric oncology clinic between July 2015 and June 2017 who completed a VES-13 and underwent subsequent GA were included. Clinical information was extracted from a prospectively maintained database. G6 scores were assigned retrospectively. Patients were stratified into those who were "VES-13 positive" (score ≥ 3) and "VES-13 negative" (score < 3). Logistic regression was used to explore the relationship between VES-13 score, G6 score, and treatment modification. RESULTS: Ninety-nine patients were seen prior to initiating cancer treatment. The median VES-13 score was 7; with 81.8% of patients scoring ≥3. The treatment plan was modified in 47.5% of patients after GA. VES-13 score was predictive of treatment plan modification (63.0% among VES-13 positive versus 16.7% among VES-13 negative patients; p = 0.001). G6 performed similarly to the VES-13. The only statistically significant predictor of treatment change in multivariable analysis was performance status. CONCLUSION: VES-13 positive patients are more likely to undergo treatment modification to reduce treatment intensity or supportive care only. The VES-13 may provide oncologists with a rapid, reliable way of identifying vulnerability in older adults with cancer who may need further GA prior to commencing cancer treatment.

Beyond the black box of geriatric assessment: Understanding enhancements to care by the geriatric oncology clinic.

OBJECTIVE: Comprehensive geriatric assessment (CGA) of older adults with cancer aids treatment decision-making and prognostication. Much less is known about the supportive care elements or enhancements to care afforded by the CGA. We characterized the enhancements to care provided by a geriatric oncology clinic and determined how these vary by indication for referral. MATERIALS AND METHODS: All patients age 65 or older referred to a single academic geriatric oncology clinic between July 2015 (clinic opening) and June 2017 were included. Treatment enhancements were prospectively recorded in 5 categories: educational support, comorbidity management, symptom management, oncologic treatment delivery, and peri-operative management recommendations. Indications for referral were categorized into 3 groups: pre-treatment (n = 97, 44%), on active treatment (n = 89, 41%), and survivorship phase (n = 33, 15%). Data were analyzed using descriptive statistics and multivariable logistic regression. RESULTS: 219 patients were seen during the study period (mean age 79.7 years, 69% male). Overall, educational support (96%) and comorbidity management (95%) were the most common enhancements, whereas peri-operative management (10%) was the least common and provided only to pre-treatment patients. Enhancements to cancer treatment delivery were offered more often to patients pre-treatment than on active treatment (61% versus 41%, p < 0.001). Other enhancements to care did not vary by indication for referral. CONCLUSION: Educational support and comorbidity management are nearly universally offered. Most enhancements to care do not vary by indication for referral. Understanding the enhancements to care provided by geriatric oncology clinics can help with resource planning and program design.

A standardized bleeding risk score aligns anticoagulation choices with current evidence.

OBJECTIVES: Atrial fibrillation (AF), the most common arrhythmia in elderly patients, accounts for 15% of strokes. Oral anticoagulation (OAC) can reduce the risk of stroke by 60% but is underprescribed. The HAS-BLED score (Hypertension, Abnormal renal or liver function, Stroke, Bleeding, Labile INR, Elderly, Drugs) can predict OAC bleeding complications. The authors hypothesized that use of HAS-BLED can help align decision making with current evidence. METHODS: The authors developed a survey with four clinical vignettes designed to highlight the complexity in deciding whether to anticoagulate elderly patients with AF. Physicians were randomly assigned to receive the survey either including the HAS-BLED score and the estimated annual risk of bleeding (intervention) or without (control). Following each vignette, participants were asked: (1) whether they would recommend OAC and (2) to estimate the risk of bleeding and stroke. The "appropriate" anticoagulation decision was defined as the choice that minimized the risk of stroke and major bleeding. RESULTS: A total of 203 physicians were recruited for the survey, with 55 responses obtained (27%). Physicians who were given the HAS-BLED score were 18% more likely to choose appropriate anticoagulation (74% vs. 56%, P < .05). The HAS-BLED score assisted physicians in both choosing to anticoagulate when appropriate and not to anticoagulate when the risk of bleeding outweighed the benefit. Overall, physicians were poor at estimating the risk of stroke (42% correct) and major bleeding (31% correct). CONCLUSIONS: Presentation of the HAS-BLED score led to an 18% improvement in appropriate OAC choices. Future study should evaluate incorporation of HAS-BLED use in real-time clinical situations.