Predicting Persistent Opioid Use, Abuse, and Toxicity Among Cancer Survivors

Abstract Background Although opioids play a critical role in the management of cancer pain, the ongoing opioid epidemic has raised concerns regarding their persistent use and abuse. We lack data-driven tools in oncology to understand the risk of adverse opioid-related outcomes. This project seeks to identify clinical risk factors and create a risk score to help identify patients at risk of persistent opioid use and abuse. Methods Within a cohort of 106 732 military veteran cancer survivors diagnosed between 2000 and 2015, we determined rates of persistent posttreatment opioid use, diagnoses of opioid abuse or dependence, and admissions for opioid toxicity. A multivariable logistic regression model was used to identify patient, cancer, and treatment risk factors associated with adverse opioid-related outcomes. Predictive risk models were developed and validated using a least absolute shrinkage and selection operator regression technique. Results The rate of persistent opioid use in cancer survivors was 8.3% (95% CI = 8.1% to 8.4%); the rate of opioid abuse or dependence was 2.9% (95% CI = 2.8% to 3.0%); and the rate of opioid-related admissions was 2.1% (95% CI = 2.0% to 2.2%). On multivariable analysis, several patient, demographic, and cancer and treatment factors were associated with risk of persistent opioid use. Predictive models showed a high level of discrimination when identifying individuals at risk of adverse opioid-related outcomes including persistent opioid use (area under the curve [AUC] = 0.85), future diagnoses of opioid abuse or dependence (AUC = 0.87), and admission for opioid abuse or toxicity (AUC = 0.78). Conclusion This study demonstrates the potential to predict adverse opioid-related outcomes among cancer survivors. With further validation, personalized risk-stratification approaches could guide management when prescribing opioids in cancer patients.

Download Full-text

Educational intervention for physicians to address the risk of opioid abuse

Journal of Opioid Management ◽

10.5055/jom.2017.0399 ◽

2017 ◽

Vol 13 (5) ◽

pp. 303 ◽

Cited By ~ 3

Author(s):

Margaret K. Pasquale, PhD ◽

Richard L. Sheer, BA ◽

Jack Mardekian, PhD ◽

Elizabeth T. Masters, MS, MPH ◽

Nick C. Patel, PharmD, PhD ◽

...

Keyword(s):

Mental Health ◽

At Risk ◽

Patient Information ◽

Opioid Abuse ◽

High Dose ◽

Opioid Use ◽

Educational Materials ◽

Patients At Risk ◽

Ed Visits ◽

The Impact

Objective: To evaluate the impact of a pilot intervention for physicians to support their treatment of patients at risk for opioid abuse.Setting, design and patients, participants: Patients at risk for opioid abuse enrolled in Medicare plans were identified from July 1, 2012 to April 30, 2014 (N = 2,391), based on a published predictive model, and linked to 4,353 opioid-prescribing physicians. Patient-physician clusters were randomly assigned to one of four interventions using factorial design.Interventions: Physicians received one of the following: Arm 1, patient information; Arm 2, links to educational materials for diagnosis and management of pain; Arm 3, both patient information and links to educational materials; or Arm 4, no communication.Main outcome measures: Difference-in-difference analyses compared opioid and pain prescriptions, chronic high-dose opioid use, uncoordinated opioid use, and opioid-related emergency department (ED) visits. Logistic regression compared diagnosis of opioid abuse between cases and controls postindex.Results: Mailings had no significant impact on numbers of opioid or pain medications filled, chronic high-dose opioid use, uncoordinated opioid use, ED visits, or rate of diagnosed opioid abuse. Relative to Arm 4, odds ratios (95% CI) for diagnosed opioid abuse were Arm 1, 0.95(0.63-1.42); Arm 2, 0.83(0.55-1.27); Arm 3, 0.72(0.46-1.13). While 84.7 percent had ≥ 1 psychiatric diagnoses during preindex (p = 0.89 between arms), only 9.5 percent had ≥ 1 visit with mental health specialists (p = 0.53 between arms).Conclusions: Although this intervention did not affect pain-related outcomes, future interventions involving care coordination across primary care and mental health may impact opioid abuse and improve quality of life of patients with pain.

Download Full-text

Preoperative clinical and tumor genomic features associated with pathologic lymph node metastasis in clinical stage I and II lung adenocarcinoma

npj Precision Oncology ◽

10.1038/s41698-021-00210-2 ◽

2021 ◽

Vol 5 (1) ◽

Author(s):

Raul Caso ◽

James G. Connolly ◽

Jian Zhou ◽

Kay See Tan ◽

James J. Choi ◽

...

Keyword(s):

Lung Adenocarcinoma ◽

Clinical Stage ◽

Multivariable Analysis ◽

Multivariable Logistic Regression Model ◽

Genomic Features ◽

Genome Doubling ◽

Whole Genome Doubling ◽

Metastatic Lung ◽

Treatment Naïve ◽

Patients At Risk

AbstractWhile next-generation sequencing (NGS) is used to guide therapy in patients with metastatic lung adenocarcinoma (LUAD), use of NGS to determine pathologic LN metastasis prior to surgery has not been assessed. To bridge this knowledge gap, we performed NGS using MSK-IMPACT in 426 treatment-naive patients with clinical N2-negative LUAD. A multivariable logistic regression model that considered preoperative clinical and genomic variables was constructed. Most patients had cN0 disease (85%) with pN0, pN1, and pN2 rates of 80%, 11%, and 9%, respectively. Genes altered at higher rates in pN-positive than in pN-negative tumors were STK11 (p = 0.024), SMARCA4 (p = 0.006), and SMAD4 (p = 0.011). Fraction of genome altered (p = 0.037), copy number amplifications (p = 0.001), and whole-genome doubling (p = 0.028) were higher in pN-positive tumors. Multivariable analysis revealed solid tumor morphology, tumor SUVmax, clinical stage, SMARCA4 and SMAD4 alterations were independently associated with pathologic LN metastasis. Incorporation of clinical and tumor genomic features can identify patients at risk of pathologic LN metastasis; this may guide therapy decisions before surgical resection.

Download Full-text

Patients at Risk: Preoperative Opioid Use Affects Opioid Prescribing, Refills, and Outcomes After Total Knee Arthroplasty

The Journal of Arthroplasty ◽

10.1016/j.arth.2018.01.004 ◽

2018 ◽

Vol 33 (7) ◽

pp. S142-S146 ◽

Cited By ~ 25

Author(s):

Nicholas M. Hernandez ◽

Joshua A. Parry ◽

Tad M. Mabry ◽

Michael J. Taunton

Keyword(s):

At Risk ◽

Total Knee Arthroplasty ◽

Knee Arthroplasty ◽

Opioid Use ◽

Opioid Prescribing ◽

Patients At Risk ◽

Total Knee

Download Full-text

Abstract 038: A Novel Histology Based Classification System to Identify Patients at Risk for Postoperative Atrial Fibrillation

Circulation Research ◽

10.1161/res.113.suppl_1.a038 ◽

2013 ◽

Vol 113 (suppl_1) ◽

Author(s):

Mahek Mirza ◽

Anton Strunets ◽

Ekhson Holmuhamedov ◽

Jasbir Sra ◽

Paul H Werner ◽

...

Keyword(s):

Atrial Fibrillation ◽

At Risk ◽

Classification System ◽

Area Under The Curve ◽

Postoperative Atrial Fibrillation ◽

Right Atrial ◽

Class Iii ◽

Atrial Fibrosis ◽

Patients At Risk ◽

The Impact

Postoperative atrial fibrillation (PoAF) is a common complication in up to 40% of patients after cardiac surgery, increasing morbidity, hospital stay and costs. The myocardial substrate underlying PoAF is not fully characterized. The objective was to assess the impact of atrial fibrosis on incident AF and define the fibrosis threshold level predictive of PoAF. Methods: Right atrial appendages removed from patients undergoing elective CABG with no history of AF or class III/IV heart failure were used to characterize the ratio of collagen to myocardium (Masson’s trichrome; NIH ImageJ software; Fig A), which was correlated with incident AF. Percentage burden of fibrosis predictive of PoAF with high sensitivity and specificity was determined by ROC curve. Results: Of 28 patients (67±10 years, 64% males), 15 had PoAF. There were no age, gender or comorbidity differences between groups. Compared to the group that remained in sinus rhythm, patients with PoAF had a significantly higher ratio of extracellular collagen to myocardium (45±16% vs. 5±4%, p <0.001; Fig B). A threshold ratio of 12.7% collagen to myocardium (ROC area under the curve 0.997; z statistic 137; P<0.0001) with 96% sensitivity and 97% specificity identified those with PoAF (Fig C). A classification system based on histological extent of atrial fibrosis is proposed for identifying patients at risk for PoAF (Fig D). Conclusion: Ongoing studies will confirm the predictive value of this new classification system for identifying the atrial substrate predisposing PoAF and correlate with preoperative cardiac imaging and circulatory serum biomarkers to provide a novel noninvasive tool to stratify patients at risk for PoAF.

Download Full-text

Serum procalcitonin levels predict disseminated intravascular coagulation in patients with sepsis shock: A retrospective cohort study

10.21203/rs.3.rs-36566/v1 ◽

2020 ◽

Author(s):

Guang Fu ◽

Xi-si He ◽

Hao-li Li ◽

Hai-chao Zhan ◽

Jun-fu Lu ◽

...

Keyword(s):

Risk Factors ◽

Septic Shock ◽

Cohort Study ◽

Disseminated Intravascular Coagulation ◽

Area Under The Curve ◽

Multivariable Logistic Regression Model ◽

Intravascular Coagulation ◽

The Relationship ◽

Conventional Risk Factors

Abstract Background Complication of disseminated intravascular coagulation (DIC) is a determinant of the prognosis in patients with sepsis shock. Procalcitonin (PCT) has been advocated as a marker of bacterial sepsis. The purpose of this study was to evaluate the relationship between serum PCT levels and DIC with sepsis shock Methods A cohort study was designed which included patients that admitted in intensive care unit (ICU) between January 1, 2015 and December 31, 2018 and the follow-up to discharge. 164 septic shock patients were divided into DIC and non-DIC groups according to international society of thrombosis and homeostasis (ISTH). PCT was measured at the admission to ICU, and all the participants received routine biochemical coagulation test subsequently. Results PCT levels were considerably higher in septic shock patients who developed DIC than those who did not (54.6[13.6–200]vs12.6[2.4–53.3]ng/ml), respectively, P < 0.001). Multivariable logistic regression model revealed that PCT level was significantly associated with risk of DIC independent of conventional risk factors. In addition, curve fitting showed a linear relationship between PCT and DIC score. The Receiver Operating characteristic(ROC) curve suggested that the optimal cut-off point for PCT to predicting DIC induced by septic shock was 42.0 ng/ml, and the area under the curve (AUC) was 0.701(95% CI [0.619–0.784], P < 0.001). More importantly, incorporating PCT with other risk factors into the prediction model significantly increased the AUC for prediction of DIC induced by sepsis shock (0.801vs 0.706; P = 0.012). Conclusions Our study suggests that PCT levels on admission is significantly and independently associated with DIC development subsequently with septic shock, combining PCT levels with other risk factors could significantly improve the prediction of DIC induced by sepsis shock.

Download Full-text

MEWS++: Enhancing the Prediction of Clinical Deterioration in Admitted Patients through a Machine Learning Model

Journal of Clinical Medicine ◽

10.3390/jcm9020343 ◽

2020 ◽

Vol 9 (2) ◽

pp. 343 ◽

Cited By ~ 4

Author(s):

Arash Kia ◽

Prem Timsina ◽

Himanshu N. Joshi ◽

Eyal Klang ◽

Rohit R. Gupta ◽

...

Keyword(s):

Machine Learning ◽

At Risk ◽

Area Under The Curve ◽

Learning Model ◽

Clinical Deterioration ◽

Early Warning Score ◽

Support Vector ◽

Adult Age ◽

Machine Learning Model ◽

Patients At Risk

Early detection of patients at risk for clinical deterioration is crucial for timely intervention. Traditional detection systems rely on a limited set of variables and are unable to predict the time of decline. We describe a machine learning model called MEWS++ that enables the identification of patients at risk of escalation of care or death six hours prior to the event. A retrospective single-center cohort study was conducted from July 2011 to July 2017 of adult (age > 18) inpatients excluding psychiatric, parturient, and hospice patients. Three machine learning models were trained and tested: random forest (RF), linear support vector machine, and logistic regression. We compared the models’ performance to the traditional Modified Early Warning Score (MEWS) using sensitivity, specificity, and Area Under the Curve for Receiver Operating Characteristic (AUC-ROC) and Precision-Recall curves (AUC-PR). The primary outcome was escalation of care from a floor bed to an intensive care or step-down unit, or death, within 6 h. A total of 96,645 patients with 157,984 hospital encounters and 244,343 bed movements were included. Overall rate of escalation or death was 3.4%. The RF model had the best performance with sensitivity 81.6%, specificity 75.5%, AUC-ROC of 0.85, and AUC-PR of 0.37. Compared to traditional MEWS, sensitivity increased 37%, specificity increased 11%, and AUC-ROC increased 14%. This study found that using machine learning and readily available clinical data, clinical deterioration or death can be predicted 6 h prior to the event. The model we developed can warn of patient deterioration hours before the event, thus helping make timely clinical decisions.

Download Full-text

Integrating Symptoms Into the Diagnostic Criteria for Breast Cancer–Related Lymphedema: Applying Results From a Prospective Surveillance Program

Physical Therapy ◽

10.1093/ptj/pzaa162 ◽

2020 ◽

Vol 100 (12) ◽

pp. 2186-2197

Author(s):

Cheryl L Brunelle ◽

Sacha A Roberts ◽

Nora K Horick ◽

Tessa C Gillespie ◽

Jamie M Jacobs ◽

...

Keyword(s):

Breast Cancer ◽

At Risk ◽

Early Intervention ◽

Diagnostic Criteria ◽

Multivariable Analysis ◽

Primary Study ◽

Surveillance Program ◽

Patients At Risk ◽

Breast Cancer Related Lymphedema ◽

Prospective Longitudinal

Abstract Objective The objectives of this study were to determine whether patients reporting symptoms are more likely to develop lymphedema and to describe the temporal relationship between symptom onset and lymphedema. Methods This was a prospective longitudinal cohort study of 647 women treated for breast cancer and screened for lymphedema using arm volume measurements and subjective questionnaires (n = 647; 2284 questionnaires [median 3.5 per patient, range = 1–24]). Primary study outcome was lymphedema (relative volume change ≥10%). The Kaplan–Meier method was used to estimate cumulative lymphedema incidence. Cox proportional hazards models were used to assess the relationship between symptoms, other risk factors, and lymphedema. Results A total of 64 patients (9.9%) developed lymphedema. On multivariable analysis, patients reporting increased arm size (hazard ratio = 3.09, 95% CI = 1.62–5.89) were more likely to progress to lymphedema than those who did not report this symptom. Of those who developed lymphedema, 37 (58%) reported an increased arm size a median of 6.1 months before lymphedema onset (range = 68.6 months before to 50.2 months after lymphedema onset). Conclusion Patients at risk of lymphedema who report increased arm size might do so prior to lymphedema onset and are at 3 times the risk of lymphedema as patients not reporting this symptom. Even without objective or observable edema, these patients should be followed vigilantly and considered for early intervention. Symptoms should be incorporated into screening and diagnostic criteria for lymphedema. Impact This study shows that patients at risk for breast cancer–related lymphedema who report increased arm size should be considered at high risk for progression to lymphedema—even without edema on measurement or clinical examination—and should be followed vigilantly, with consideration of early intervention. Lay summary If you are at risk of lymphedema and you feel as though your arm size has increased, you might develop lymphedema, and you are at 3 times the risk of lymphedema as patients not reporting this symptom. Even without measurable or observable edema, you should be followed vigilantly and consider early intervention.

Download Full-text

Long-term opioid users with chronic noncancer pain: Assessment of opioid abuse risk and relationship with healthcare resource use

Journal of Opioid Management ◽

10.5055/jom.2018.0440 ◽

2018 ◽

Vol 14 (2) ◽

pp. 131 ◽

Cited By ~ 6

Author(s):

Anna D. Coutinho, BPharm, PhD ◽

Kavita Gandhi, BPharm, MS ◽

Rupali M. Fuldeore, BAMS, MS ◽

Pamela B. Landsman-Blumberg, MPH, DrPH ◽

Sanjay Gandhi, PhD

Keyword(s):

Risk Factors ◽

At Risk ◽

Opioid Abuse ◽

Low Risk ◽

Chronic Noncancer Pain ◽

Cart Analysis ◽

Risk Patients ◽

Noncancer Pain ◽

Abuse Risk

Objective: Identify opioid abuse risk factors among chronic noncancer pain (CNCP) patients receiving long-term opioid therapy and assess healthcare resource use (HRU) among patients at elevated abuse risk.Design: Data were obtained from an integrated administrative claims database. Classification and Regression Tree (CART) analysis identified risk factors potentially predictive of opioid abuse, which were used to classify the overall population into cohorts defined by levels of abuse risk. Multivariable logistic regression compared HRU across risk cohorts.Setting: Retrospective cohort study.Patients, participants: 21,072 patients aged ≥18 years diagnosed with ≥1 of 5 types of CNCP and a prescription for Schedule II or III/IV opioid medication used long-term (≥90 days).Main outcome measures: (1) Opioid abuse risk factors; (2) HRU differences between risk cohorts.Results: CART analysis identified four groups at elevated opioid abuse risk defined by three factors (age, daily opioid dose, and total days’ supply of opioids); sensitivity: 70.3 percent, specificity: 74.1 percent, and positive predictive value: 5.6 percent. The analysis results were used to classify patients into low-risk (72.5 percent), at-risk (25.4 percent), and opioid-abuser (2.2 percent) cohorts. In multivariable analysis, emergency department (ED) use was higher among at-risk vs low-risk patients (odds ratio [OR]: 1.14; p < 0.05); hospitalization and ED visits were higher for opioid-abusers vs low-risk patients (OR: 2.33 and 2.14, respectively; p < 0.05).Conclusions: This study identifies a subpopulation of CNCP patients at risk of opioid abuse. However, limited sensitivity and specificity of criteria defining this subpopulation reinforce the importance of physician discretion in patient-level treatment decisions.

Download Full-text

Health and cancer concerns among siblings of childhood cancer survivors: A report from the Childhood Cancer Survivor Study (CCSS).

Journal of Clinical Oncology ◽

10.1200/jco.2020.38.15_suppl.12117 ◽

2020 ◽

Vol 38 (15_suppl) ◽

pp. 12117-12117

Author(s):

Sonia Morales ◽

Sedigheh Mirzaei Salehabadi ◽

Deokumar Srivastava ◽

Todd M. Gibson ◽

Wendy M. Leisenring ◽

...

Keyword(s):

Risk Factors ◽

At Risk ◽

Health Status ◽

Cancer Risk ◽

Cancer Survivors ◽

Childhood Cancer ◽

Psychosocial Problems ◽

Matched Pair ◽

Future Health ◽

Grade 3

12117 Background: Siblings of long-term survivors of childhood cancer can be at risk for persistent concerns regarding their future health and risk for cancer. We examined self-perceived future health and cancer risk concerns among such siblings. Methods: 3,969 siblings (median age 29 [range 18–56] years) of 5+ year matched pair cancer survivors (n= 3,969; age 25 [6–48] years; time since diagnosis 19.6 [9.6-33.8] years) in the CCSS self-reported physical/psychosocial problems, including concerns regarding future health and cancer risk (dichotomized as concerned vs not concerned). Chronic health conditions (CHC) were graded using the Common Terminology Criteria for Adverse Events system: mild (grade 1), moderate (grade 2), severe/disabling (grade 3),or life-threatening (grade 4). Sibling demographics, their matched survivor’s diagnosis, era and treatment components, complications (death, relapse, disfigurement) as well as self-reported health status and CHCs for siblings and survivors were examined as potential risk factors for concern using multivariable logistic regression. Adjusted odds ratios (OR) and 95% confidence intervals (CI) are reported. Results: The prevalence of siblings reporting concerns regarding health and cancer risk decreased based on decades of matched survivor diagnosis: 1970-79 (73.3%; 63.9%), 1980-89 (67.2%; 62.6%), 1990-99 (45.7%; 52.3%). Risk factors for concerns included sibling poor/fair current health (future health OR 3.65, 95% CI 2.37-5.62; cancer risk OR 1.54, 1.12-2.13) compared to good/very good/excellent health. Sibling grade 2 (future health OR 1.46, 1.23-1.74; cancer risk OR 1.20, 1.01-1.42) or grade 3-4 CHCs (future health OR 1.37, 1.09-1.71; cancer risk OR 1.28, 1.03-1.58) were associated with greater concerns compared to those with less than grade 2 CHCs. Survivor treatment with chemotherapy/radiation was associated with elevated cancer risk concerns (OR 1.51, 1.13-2.02) compared to surgery/no therapy. Siblings of survivors with grade 3-4 CHCs (OR 1.35, 1.12-1.63) had greater future health concerns compared to those with less than grade 2 CHCs. Sibling bereavement was a risk factor for future health (OR 1.45, 1.04-2.03) and cancer risk (OR 1.44, 1.05-1.99) concerns. Conclusions: The prevalence of sibling concerns regarding future health and cancer have diminished in more recent decades. Subgroups of siblings are at-risk for concerns over future health and cancer risk, partially determined by medical characteristics of their survivor and their own health status.

Download Full-text

Characterising the severity of treatment resistance in unipolar and bipolar depression

BJPsych Open ◽

10.1192/bjo.2021.1004 ◽

2021 ◽

Vol 7 (6) ◽

Author(s):

Rachael W. Taylor ◽

Rebecca Strawbridge ◽

Allan H. Young ◽

Roland Zahn ◽

Anthony J. Cleare

Keyword(s):

Risk Factors ◽

At Risk ◽

Bipolar Depression ◽

Antidepressant Treatment ◽

Age At Onset ◽

Treatment Resistance ◽

Patient Characteristics ◽

Patients At Risk ◽

Resistant Depression ◽

Depressive Episodes

Background Treatment-resistant depression (TRD) is classically defined according to the number of suboptimal antidepressant responses experienced, but multidimensional assessments of TRD are emerging and may confer some advantages. Patient characteristics have been identified as risk factors for TRD but may also be associated with TRD severity. The identification of individuals at risk of severe TRD would support appropriate prioritisation of intensive and specialist treatments. Aims To determine whether TRD risk factors are associated with TRD severity when assessed multidimensionally using the Maudsley Staging Method (MSM), and univariately as the number of antidepressant non-responses, across three cohorts of individuals with depression. Method Three cohorts of individuals without significant TRD, with established TRD and with severe TRD, were assessed (n = 528). Preselected characteristics were included in linear regressions to determine their association with each outcome. Results Participants with more severe TRD according to the MSM had a lower age at onset, fewer depressive episodes and more physical comorbidities. These associations were not consistent across cohorts. The number of episodes was associated with the number of antidepressant treatment failures, but the direction of association varied across the cohorts studied. Conclusions Several risk factors for TRD were associated with the severity of resistance according to the MSM. Fewer were associated with the raw number of inadequate antidepressant responses. Multidimensional definitions may be more useful for identifying patients at risk of severe TRD. The inconsistency of associations across cohorts has potential implications for the characterisation of TRD.

Download Full-text