Detection of Medication Taking Using a Wrist-Worn Commercially Available Wearable Device.

PURPOSE: Medication nonadherence is a persistent and costly problem across health care. Measures of medication adherence are ineffective. Methods such as self-report, prescription claims data, or smart pill bottles have been used to monitor medication adherence, but these are subject to recall bias, lack real-time feedback, and are often expensive. METHODS: We proposed a method for monitoring medication adherence using a commercially available wearable device. Passively collected motion data were analyzed on the basis of the Movelet algorithm, a dictionary learning framework that builds person-specific chapters of movements from short frames of elemental activities within the movements. We adapted and extended the Movelet method to construct a within-patient prediction model that identifies medication-taking behaviors. RESULTS: Using 15 activity features recorded from wrist-worn wearable devices of 10 patients with breast cancer on endocrine therapy, we demonstrated that medication-taking behavior can be predicted in a controlled clinical environment with a median accuracy of 85%. CONCLUSION: These results in a patient-specific population are exemplar of the potential to measure real-time medication adherence using a wrist-worn commercially available wearable device.

Long-term Effect of Machine Learning-Triggered Behavioral Nudges on Serious Illness Conversations and End-of-Life Outcomes Among Patients With Cancer: A Randomized Clinical Trial.

Importance: Serious illness conversations (SICs) between oncology clinicians and patients are associated with improved quality of life and may reduce aggressive end-of-life care. However, most patients with cancer die without a documented SIC. Objective: To test the impact of behavioral nudges to clinicians to prompt SICs on the SIC rate and end-of-life outcomes among patients at high risk of death within 180 days (high-risk patients) as identified by a machine learning algorithm. Design, Setting, and Participants: This prespecified 40-week analysis of a stepped-wedge randomized clinical trial conducted between June 17, 2019, and April 20, 2020 (including 16 weeks of intervention rollout and 24 weeks of follow-up), included 20 506 patients with cancer representing 41 021 encounters at 9 tertiary or community-based medical oncology clinics in a large academic health system. The current analyses were conducted from June 1, 2021, to May 31, 2022. Intervention: High-risk patients were identified using a validated electronic health record machine learning algorithm to predict 6-month mortality. The intervention consisted of (1) weekly emails to clinicians comparing their SIC rates for all patients against peers' rates, (2) weekly lists of high-risk patients, and (3) opt-out text messages to prompt SICs before encounters with high-risk patients. Main Outcomes and Measures: The primary outcome was SIC rates for all and high-risk patient encounters; secondary end-of-life outcomes among decedents included inpatient death, hospice enrollment and length of stay, and intensive care unit admission and systemic therapy close to death. Intention-to-treat analyses were adjusted for clinic and wedge fixed effects and clustered at the oncologist level. Results: The study included 20 506 patients (mean [SD] age, 60.0 [14.0] years) and 41 021 patient encounters: 22 259 (54%) encounters with female patients, 28 907 (70.5%) with non-Hispanic White patients, and 5520 (13.5%) with high-risk patients; 1417 patients (6.9%) died by the end of follow-up. There were no meaningful differences in demographic characteristics in the control and intervention periods. Among high-risk patient encounters, the unadjusted SIC rates were 3.4% (59 of 1754 encounters) in the control period and 13.5% (510 of 3765 encounters) in the intervention period. In adjusted analyses, the intervention was associated with increased SICs for all patients (adjusted odds ratio, 2.09 [95% CI, 1.53-2.87]; P < .001) and decreased end-of-life systemic therapy (7.5% [72 of 957 patients] vs 10.4% [24 of 231 patients]; adjusted odds ratio, 0.25 [95% CI, 0.11-0.57]; P = .001) relative to controls, but there was no effect on hospice enrollment or length of stay, inpatient death, or end-of-life ICU use. Conclusions and Relevance: In this randomized clinical trial, a machine learning-based behavioral intervention and behavioral nudges to clinicans led to an increase in SICs and reduction in end-of-life systemic therapy but no changes in other end-of-life outcomes among outpatients with cancer. These results suggest that machine learning and behavioral nudges can lead to long-lasting improvements in cancer care delivery. Trial Registration: ClinicalTrials.gov Identifier: NCT03984773.

European LeukemiaNet-defined primary refractory acute myeloid leukemia: the value of allogeneic hematopoietic stem cell transplant and overall response.

We sought to appraise the value of overall response and salvage chemotherapy, inclusive of allogeneic hematopoietic stem cell transplant (AHSCT), in primary refractory acute myeloid leukemia (prAML). For establishing consistency in clinical practice, the 2017 European LeukemiaNet (ELN) defines prAML as failure to attain CR after at least 2 courses of intensive induction chemotherapy. Among 60 consecutive patients (median age 63 years) correspondent with ELN-criteria for prAML, salvage was documented in 48 cases, 30/48 (63%) being administered intensive chemotherapy regimens and 2/48 consolidated with AHSCT as first line salvage. 13/48 (27%) attained response: CR, 7/13 (54%), CRi, 2/13 (15%), MLFS, 4/13 (31%). The CR/CRi rate was 9/48 (19%), with CR rate of 7/48 (15%). On univariate analysis, intermediate-risk karyotype was the only predictor of response (44% vs 17% in unfavorable karyotype; P = 0.04). Administration of any higher-dose (>1 g/m2) cytarabine intensive induction (P = 0.50), intensive salvage chemotherapy (P = 0.72), targeted salvage (FLT3 or IDH inhibitors) (P = 0.42), greater than 1 salvage regimen (P = 0.89), age < 60 years (P = 0.30), and de novo AML (P = 0.10) did not enhance response achievement, nor a survival advantage. AHSCT was performed in 12 patients with (n = 8) or without (n = 4) CR/CRi/MLFS. 1/2/5-year overall survival (OS) rates were 63%/38%/33% in patients who received AHSCT (n = 12) vs 27%/0%/0% in those who achieved CR/CRi/MLFS but were not transplanted (n = 5), vs 14%/0%/0% who were neither transplanted nor achieved CR/CRi/MLFS (n = 43; P < 0.001); the median OS was 18.6, 12.6 and 5.6 months, respectively. Although CR/CRi/MLFS bridged to AHSCT (n = 8), appeared to manifest a longer median OS (20 months), vs (13.4 months) for those with no response consolidated with AHSCT (n = 4), the difference was not significant P = 0.47. We conclude AHSCT as indispensable for securing long-term survival in prAML (p = 0.03 on multivariate analysis), irrespective of response achievement.

Where next for the design, delivery, and evaluation of community-based physical activity prescription? Emerging lessons from the United Kingdom.

Despite widespread use, community-based physical activity prescription is controversial. Data limitations have resulted in a lack of clarity about what works, under what circumstances, and for whom, reflected in conservative policy recommendations. In this commentary we challenge a predominantly negative discourse, using contemporary research to highlight promising findings and "lessons learnt" for design, delivery, and evaluation. In doing so, we argue for the importance of a more nuanced approach to future commissioning and evaluation. Novelty: Amalgamating learning from multiple research teams to create recommendations for advancing physical activity prescription.

Effect of Integrating Machine Learning Mortality Estimates With Behavioral Nudges to Clinicians on Serious Illness Conversations Among Patients With Cancer: A Stepped-Wedge Cluster Randomized Clinical Trial.

IMPORTANCE: Serious illness conversations (SICs) are structured conversations between clinicians and patients about prognosis, treatment goals, and end-of-life preferences. Interventions that increase the rate of SICs between oncology clinicians and patients may improve goal-concordant care and patient outcomes. OBJECTIVE: To determine the effect of a clinician-directed intervention integrating machine learning mortality predictions with behavioral nudges on motivating clinician-patient SICs. DESIGN, SETTING, AND PARTICIPANTS: This stepped-wedge cluster randomized clinical trial was conducted across 20 weeks (from June 17 to November 1, 2019) at 9 medical oncology clinics (8 subspecialty oncology and 1 general oncology clinics) within a large academic health system in Pennsylvania. Clinicians at the 2 smallest subspecialty clinics were grouped together, resulting in 8 clinic groups randomly assigned to the 4 intervention wedge periods. Included participants in the intention-to-treat analyses were 78 oncology clinicians who received SIC training and their patients (N = 14 607) who had an outpatient oncology encounter during the study period. INTERVENTIONS: (1) Weekly emails to oncology clinicians with SIC performance feedback and peer comparisons; (2) a list of up to 6 high-risk patients (≥10% predicted risk of 180-day mortality) scheduled for the next week, estimated using a validated machine learning algorithm; and (3) opt-out text message prompts to clinicians on the patient's appointment day to consider an SIC. Clinicians in the control group received usual care consisting of weekly emails with cumulative SIC performance. MAIN OUTCOMES AND MEASURES: Percentage of patient encounters with an SIC in the intervention group vs the usual care (control) group. RESULTS: The sample consisted of 78 clinicians and 14 607 patients. The mean (SD) age of patients was 61.9 (14.2) years, 53.7% were female, and 70.4% were White. For all encounters, SICs were conducted among 1.3% in the control group and 4.6% in the intervention group, a significant difference (adjusted difference in percentage points, 3.3; 95% CI, 2.3-4.5; P < .001). Among 4124 high-risk patient encounters, SICs were conducted among 3.6% in the control group and 15.2% in the intervention group, a significant difference (adjusted difference in percentage points, 11.6; 95% CI, 8.2-12.5; P < .001). CONCLUSIONS AND RELEVANCE: In this stepped-wedge cluster randomized clinical trial, an intervention that delivered machine learning mortality predictions with behavioral nudges to oncology clinicians significantly increased the rate of SICs among all patients and among patients with high mortality risk who were targeted by the intervention. Behavioral nudges combined with machine learning mortality predictions can positively influence clinician behavior and may be applied more broadly to improve care near the end of life. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT03984773.

Validation of a Machine Learning Algorithm to Predict 180-Day Mortality for Outpatients With Cancer.

IMPORTANCE: Machine learning (ML) algorithms can identify patients with cancer at risk of short-term mortality to inform treatment and advance care planning. However, no ML mortality risk prediction algorithm has been prospectively validated in oncology or compared with routinely used prognostic indices. OBJECTIVE: To validate an electronic health record-embedded ML algorithm that generated real-time predictions of 180-day mortality risk in a general oncology cohort. DESIGN, SETTING, AND PARTICIPANTS: This prognostic study comprised a prospective cohort of patients with outpatient oncology encounters between March 1, 2019, and April 30, 2019. An ML algorithm, trained on retrospective data from a subset of practices, predicted 180-day mortality risk between 4 and 8 days before a patient's encounter. Patient encounters took place in 18 medical or gynecologic oncology practices, including 1 tertiary practice and 17 general oncology practices, within a large US academic health care system. Patients aged 18 years or older with outpatient oncology or hematology and oncology encounters were included in the analysis. Patients were excluded if their appointment was scheduled after weekly predictions were generated and if they were only evaluated in benign hematology, palliative care, or rehabilitation practices. EXPOSURES: Gradient-boosting ML binary classifier. MAIN OUTCOMES AND MEASURES: The primary outcome was the patients' 180-day mortality from the index encounter. The primary performance metric was the area under the receiver operating characteristic curve (AUC). RESULTS: Among 24 582 patients, 1022 (4.2%) died within 180 days of their index encounter. Their median (interquartile range) age was 64.6 (53.6-73.2) years, 15 319 (62.3%) were women, 18 015 (76.0%) were White, and 10 658 (43.4%) were seen in the tertiary practice. The AUC was 0.89 (95% CI, 0.88-0.90) for the full cohort. The AUC varied across disease-specific groups within the tertiary practice (AUC ranging from 0.74 to 0.96) but was similar between the tertiary and general oncology practices. At a prespecified 40% mortality risk threshold used to differentiate high- vs low-risk patients, observed 180-day mortality was 45.2% (95% CI, 41.3%-49.1%) in the high-risk group vs 3.1% (95% CI, 2.9%-3.3%) in the low-risk group. Integrating the algorithm into the Eastern Cooperative Oncology Group and Elixhauser comorbidity index-based classifiers resulted in favorable reclassification (net reclassification index, 0.09 [95% CI, 0.04-0.14] and 0.23 [95% CI, 0.20-0.27], respectively). CONCLUSIONS AND RELEVANCE: In this prognostic study, an ML algorithm was feasibly integrated into the electronic health record to generate real-time, accurate predictions of short-term mortality for patients with cancer and outperformed routinely used prognostic indices. This algorithm may be used to inform behavioral interventions and prompt earlier conversations about goals of care and end-of-life preferences among patients with cancer.

Patient preference for time-saving telehealth postoperative visits after routine surgery in an urban setting.

BACKGROUND: Focusing on high-value delivery of health care, we describe our implementation of telephone postoperative visits as alternatives to in-person follow-up after routine, low-risk surgery in an urban setting. Our pilot program assessed telephone postoperative visit feasibility as well as patient satisfaction and clinical outcomes. METHODS: We offered telephone postoperative visits to all clinically eligible, in-state patients scheduled for appropriate low-risk operations. An advanced practitioner conducted the telephone postoperative visit within 2 weeks of the operation and discharged patients from routine follow-up if recovery was satisfactory. We reviewed the medical records to identify encounters and adverse events in the 30-day postoperative period. RESULTS: Telephone postoperative visits were opted for by 92/94 (98%) clinically eligible, in-state patients. Most patients cited convenience (55%), travel (34%), and time (22%) as their main motivations. The average patient opting in was 55 ± 16 years old (range 23-88, 8% > 65) and lived 22 ± 26 miles from our clinic (range 0.9-124). Of 50 patients completing telephone postoperative visits, 48 (96%, 2 were not asked) were satisfied with the telephone postoperative visit as their sole postoperative visit, 44 (88%) of whom required no additional follow-up. On average, telephone postoperative visits lasted 8.6 ± 3.9 minutes, compared with the 82.8 ± 33.4 minutes for preintervention, postoperative visit time. Adding travel times, we estimate each patient saved an average of 139-199 minutes or 94-96% of the time they would have spent coming to clinic. No instances of major morbidity or mortality were identified on chart review. CONCLUSION: Many patients find telephone postoperative visits more convenient than in-clinic visits. Moreover, estimates of time saved are compelling. Amid changing regulations and reimbursement, our findings support the growing use of telehealth for postoperative care of routine, low risk operations.

The provision of care to adults with an intellectual disability in the UK. A Special report from the intellectual disability UK chapter ILAE.

PURPOSE: This article reflects the report by the British Branch of the International League Against Epilepsy (ILAE) Working Group on services for adults with epilepsy and intellectual disability (ID). Its terms of reference was to explore the current status of aspects of the care of people with an ID and epilepsy. METHODS: Survey content was developed from key themes identified by consensus of the working group. An electronic survey was distributed via email. The sample population was the membership of the ILAE UK, Royal College of Psychiatrists (RCPsych) Faculty of ID, Epilepsy Nurses Association (ESNA), and the Association of British Neurologists (ABN). Following a six week response period the data was then collated, anonymised and distributed to the working group in order that opinion statements could be gathered. RESULTS: The time taken for individuals with both new-onset and established epilepsy to undergo routine investigation was commonly at least 1-3 months, far beyond recommendations made by NICE (CG20). A small minority of clinicians would not consider non-pharmacological interventions including epilepsy surgery, vagus nerve stimulation, and ketogenic diet for this population. Almost universally responders are actively involved in the assessment and management of key risk areas including risk of drowning, hospitalization, medication side effects, and sudden unexpected death in epilepsy (SUDEP). CONCLUSION: This investigation identifies key themes and recommendations relating to care delivery and meeting the complex needs of people with ID and epilepsy. Adults with ID and epilepsy appear to exist in a unique, but inadequate, segment of epilepsy care delivery.

Do dental undergraduates think that Thiel-embalmed cadavers are a more realistic model for teaching exodontia?

INTRODUCTION: Teaching exodontia to novice undergraduates requires a realistic model. Thiel-embalmed cadavers retain the flexibility of the soft tissues and could be used to teach exodontia. OBJECTIVE: The objective was to determine whether Thiel-embalmed cadavers were perceived to be a more realistic model by undergraduates in comparison with mannequins. MATERIALS AND METHODS: Over a period of 4 years (2011-2014), students were randomly assigned into two groups: those taught exodontia on mannequins only (NT) and those who also experienced cadaveric teaching (T). This was followed by an assessment. RESULTS: There were 174 students in the T group and 108 in the NT group. Sixty-five per cent of the T group and 69% of the NT group provided feedback. Ninety-eight per cent (98%) felt that they had been advantaged by being included in the group compared with 95% in the NT who felt disadvantaged. The majority (98%) thought that using the cadavers was advantageous and gave a realistic feel for soft tissue management (89%) and that it was similar to managing a patient (81%). Self-reported confidence in undertaking an extraction was not different between the two groups (P=.078), and performance in the extraction assessment was not significantly different between the two groups over the 4 years (P=.8). CONCLUSION: The Thiel-embalmed cadavers were well received by the students who found it a more realistic model for exodontia than a mannequin, even though this did not impact on their performance in a following assessment. Future work on these cadavers may be expanded to include surgical procedures.

Risk factors for arterial versus venous thrombosis in polycythemia vera: a single center experience in 587 patients.

In a recent International Working Group on Myeloproliferative Neoplasms Research and Treatment (IWG-MRT) study, prior arterial events and hypertension were predictors of subsequent arterial thrombosis whereas prior venous events and age ≥65 years predicted venous thrombosis in polycythemia vera (PV). In the current study, we sought to validate the above findings and identify additional predictors of arterial versus venous thrombosis. At a median follow up of 109 months, thrombosis after diagnosis occurred in 128 (22%) patients; 82 (14%) arterial and 57 (10%) venous events. On multivariate analysis, prior arterial events (<0.0001), hyperlipidemia (p = 0.03), and hypertension (p = 0.02) predicted subsequent arterial events. In comparison, prior venous events (p = 0.05), leukocytosis ≥11 × 109/L (p = 0.002), and major hemorrhage (p = 0.02) were predictors of subsequent venous events. Salient associations with arterial thrombosis included age ≥ 60 years, hypertension, diabetes, hyperlipidemia and normal karyotype whereas age ≤ 60 years, females, palpable splenomegaly and history of major hemorrhage were associated with venous thrombosis. TET2 or ASXL1 mutations did not impact arterial nor venous thrombosis. In conclusion, we identify distinct associations for arterial versus venous thrombosis in PV and confirm that a prior arterial or venous thrombotic event is the most reliable predictor of subsequent events.

Prognostic relevance of lymphocytopenia, monocytopenia and lymphocyte-to-monocyte ratio in primary myelodysplastic syndromes: a single center experience in 889 patients.

Current prognostic models for myelodysplastic syndromes (MDS), including the Revised International Prognostic Scoring System (IPSS-R), do not account for host immunity. We retrospectively examined the prognostic relevance of monocytopenia, lymphocytopenia and lymphocyte-to-monocyte ratio (LMR) in a cohort of 889 patients with primary MDS. After a median follow-up of 27 months, 712 (80%) deaths and 116 (13%) leukemic transformation were documented. In univariate analysis, subnormal absolute lymphocyte count (ALC) <0.9 × 109/l; P=0.001), ALC<1.2 × 109/l (P=0.0002), subnormal absolute monocyte count (AMC) <0.3 × 109/l (P=0.0003), LMR (P⩽0.0001) and LMR⩾5 (P=0.03) were all associated with inferior overall survival. In multivariable analysis that included other risk factors, significance was retained for LMR (P=0.02) and became borderline for ALC <1.2 × 109/l (P=0.06). Analysis in the context of IPSS-R resulted in P-values of 0.06 for ALC<1.2 × 109/l, 0.7 for monocytopenia and 0.2 for LMR. Leukemia-free survival was not affected by ALC, AMC or LMR. The observations from the current study suggest a possible detrimental role for altered host immunity in primary MDS, which might partly explain the therapeutic benefit of immune-directed therapy, including the use of immune modulators; however, IPSS-R-independent prognostic value for either ALC or AMC was limited.

Prognostic impact of bone marrow fibrosis in polycythemia vera: validation of the IWG-MRT study and additional observations.

In 2012, the International Working Group for Myeloproliferative Neoplasms (MPN) Research and Treatment (IWG-MRT) reported an associations between mild bone marrow (BM) fibrosis (⩾grade 1) in polycythemia vera (PV) and a lower incidence of thrombosis during the clinical course and a higher risk of fibrotic progression. The objective in the current study of 262 patients with PV was to validate these observations and also identify other risk factors for myelofibrosis-free survival (MFFS). About 127 (48%) patients displayed ⩾grade 1 reticulin fibrosis at the time of diagnosis; presenting clinical and laboratory features were not significantly different between patients with or without BM fibrosis. In univariate analysis, BM fibrosis had no significant impact on overall, leukemia-free or thrombosis-free survival, whereas a significant association was noted for MFFS (P=0.009, hazard ratio 2.9; 95% confidence interval 1.32-6.78); other risk factors for MFFS included leukocytosis ⩾15 × 109/l, presence of palpable splenomegaly and abnormal karyotype. During multivariable analysis, leukocytosis ⩾15 × 109/l, palpable splenomegaly and ⩾grade 1 BM reticulin fibrosis remained significant. The current study validates the previously observed association between ⩾grade 1 BM reticulin fibrosis in PV and subsequent fibrotic progression, and identifies leukocytosis and palpable splenomegaly as additional risk factors for fibrotic progression; additional studies are required to clarify the impact of BM fibrosis on thrombosis and that of abnormal karyotype on MFFS.