Multimodal Machine Learning for Prediction of 30-day Readmission Risk in Elderly Population.

BACKGROUND: Readmission within 30 days is a prevalent issue among elderly patients, linked to unfavorable health outcomes. Our objective was to develop and validate multimodal machine learning models for predicting 30-day readmission risk in elderly patients discharged from internal medicine departments. METHODS: This was a retrospective cohort study which included elderly patients aged 75 or older, who were hospitalized at the Hadassah Medical Center internal medicine departments between 2014 and 2020. Three machine learning algorithms were developed and employed to predict 30-day readmission risk. The primary measures were predictive model performance scores, specifically area under the receiver operator curve (AUROC), and average precision. RESULTS: This study included 19,569 admissions. Of them, 3,258 (16.65%) resulted in 30-day readmission. Our three proposed models demonstrated high accuracy and precision on an unseen test set, with AUROC values of 0.87, 0.89, and 0.93, respectively, and average precision values of 0.76, 0.78, and 0.81. Feature importance analysis revealed that the number of admissions in the past year, history of 30-day readmission, Charlson score, and admission length were the most influential variables. Notably, the natural language processing score, representing the probability of readmission according to a textual-based model trained on social workers assessment letters during hospitalization, ranked among the top 10 contributing factors. CONCLUSIONS: Leveraging multimodal machine learning offers a promising strategy for identifying elderly patients who are at high risk for 30-day readmission. By identifying these patients, machine learning models may facilitate the effective execution of preventive actions to reduce avoidable readmission incidents.

Validation of a novel Artificial Pharmacological Intelligence (API) system for the management of patients with polypharmacy.

OBJECTIVE: Medication management of patients with polypharmacy is highly complex. We aimed to validate a novel Artificial Pharmacological Intelligence (API) algorithm to optimize the medication review process in a comprehensive, personalized, and scalable way. MATERIALS AND METHODS: The study was conducted on anonymized retrospective electronic health records (EHR) of 49 patients. Each patient's file was reviewed by the API system, a clinical pharmacist, and a judging committee. Validation was assessed by comparing the overall agreement of the judging committee (as the gold standard, blinded to the identity of the analyzer) to both the API system and clinical pharmacists' conclusions. Five medication-related problem (MRP) categories were assessed: duplication of therapy, age-related issues, incorrect dose, current side effects and future side effects' risk. For each category the overall validity parameters, agreement, positive predictive value (PPV), negative predictive value (NPV), sensitivity and specificity were analyzed. RESULTS: The agreement between the API system and the judging committee was 93.5 % (95 % CI 92.7-94.4), while the agreement between the clinical pharmacists and the judging committee was 73.9 % (95 % CI 72.5-75.3). The PPV was 92.2 % (90.9-93.5) and NPV was 94.2 % (93.1-95.2) for the API system and 76.3 % (69.8-82.8) and 73.5 % (72.3-74.8) respectively for the clinical pharmacists. DISCUSSION: AI systems can equip clinicians with sophisticated tools and scale manual processes such as comprehensive medication reviews, thus reducing MRPs and drug-related hospitalizations related to multidrug treatments. The API system validated in this study provided comprehensive, multidrug, multilayered analysis intended to bridge the innate complexity of personalized polypharmacy treatment. CONCLUSIONS: The API system was validated as a tool for providing actionable clinical insights non-inferior to a manual clinical review of a clinical pharmacist. The API system showed promising results in reducing MRPs.

Multi-Dimensional Validation of the Integration of Syntactic and Semantic Distance Measures for Clustering Fibromyalgia Patients in the Rheumatic Monitor Big Data Study.

This study primarily aimed at developing a novel multi-dimensional methodology to discover and validate the optimal number of clusters. The secondary objective was to deploy it for the task of clustering fibromyalgia patients. We present a comprehensive methodology that includes the use of several different clustering algorithms, quality assessment using several syntactic distance measures (the Silhouette Index (SI), Calinski-Harabasz index (CHI), and Davies-Bouldin index (DBI)), stability assessment using the adjusted Rand index (ARI), and the validation of the internal semantic consistency of each clustering option via the performance of multiple clustering iterations after the repeated bagging of the data to select multiple partial data sets. Then, we perform a statistical analysis of the (clinical) semantics of the most stable clustering options using the full data set. Finally, the results are validated through a supervised machine learning (ML) model that classifies the patients back into the discovered clusters and is interpreted by calculating the Shapley additive explanations (SHAP) values of the model. Thus, we refer to our methodology as the clustering, distance measures and iterative statistical and semantic validation (CDI-SSV) methodology. We applied our method to the analysis of a comprehensive data set acquired from 1370 fibromyalgia patients. The results demonstrate that the K-means was highly robust in the syntactic and the internal consistent semantics analysis phases and was therefore followed by a semantic assessment to determine the optimal number of clusters (k), which suggested k = 3 as a more clinically meaningful solution, representing three distinct severity levels. the random forest model validated the results by classification into the discovered clusters with high accuracy (AUC: 0.994; accuracy: 0.946). SHAP analysis emphasized the clinical relevance of "functional problems" in distinguishing the most severe condition. In conclusion, the CDI-SSV methodology offers significant potential for improving the classification of complex patients. Our findings suggest a classification system for different profiles of fibromyalgia patients, which has the potential to improve clinical care, by providing clinical markers for the evidence-based personalized diagnosis, management, and prognosis of fibromyalgia patients.

Pediatric overweight and obesity increased in Israel during the COVID-19 period.

Reports from many settings suggest that pediatric overweight and obesity increased in 2020 and 2021, presumably due to lifestyle changes associated with the COVID-19 pandemic. Many of these previous reports have relied on convenience samples or subsets of the population. Here, we present results of a longitudinal study of the entire population of Israel, a nation of approximately 9 million people, with the proportion with underweight, normal weight, overweight, and obesity at age 7 and at age 14-15, across the years 2017-2021. Our results show that the prevalence of overweight and obesity, which had been steady or improving through 2019, increased relatively quickly in 2020 and 2021. For example, among 7-year-olds, the percentage of children with obesity in 2019 was 6.8% (99% CI: 6.69-7.05), and by 2021, it had increased to 7.7% (99% CI: 7.53-7.93). There were important disparities in overweight and obesity based on SES; for example, the rate ratio for obesity comparing the poorest with the wealthiest 14-15-year-olds in 2019 was 1.63 (99% CI: 1.55-1.72). However, these disparities did not change meaningfully in 2020 and 2021, implying that while obesity did become more prevalent, this increase in prevalence was not differential across socioeconomic status. Like many other nations, Israel too experienced considerable increases in pediatric overweight and obesity in 2020-2021, erasing the improvements of the previous years among younger children.

Validation of a multi-frequency bioelectrical impedance analysis device for the assessment of body composition in older adults with type 2 diabetes.

BACKGROUND: Aging and type 2 diabetes (T2DM) are associated with an increased risk of sarcopenia. Diagnosis of sarcopenia is commonly done using dual-energy X-ray absorptiometry (DXA) in specialized settings. Another available method for assessing body composition is direct segmental multi-frequency bioelectrical impedance analysis (DSMF-BIA). Here, we examine the accuracy of a DSMF-BIA (InBody-770) for assessing body composition in older adults with T2DM when compared to DXA. METHODS: Eighty-four obese/overweight older adults (49 women, 71 ± 5 years) with T2DM who were recruited for the CEV-65 study and had both DSMF-BIA and DXA assessments at baseline were included. The analysis included Bland-Altman plots and intra class correlation coefficients. Sub-analyses were performed according to gender and following 10 weeks of interventions (diet, circuit training, and Empagliflozin). RESULTS: The leg lean mass results according to DSMF-BIA and DXA were 14.76 ± 3.62 kg and 15.19 ± 3.52 kg, respectively, with no difference between devices according to Bland-Altman analyses (p = 0.353). Assessment of appendicular skeletal mass index did not differ between DSMF-BIA and DXA (7.43 vs. 7.47 kg/m2; p = 0.84; ICC = 0.965, p < 0.0001; mean difference -0.068, p = 0.595). Gender and treatment interventions did not modify the accuracy of the DSMF-BIA when compared to DXA. CONCLUSIONS: In older adults with T2DM the degree of agreement between DSMF-BIA and DXA, was high, supporting the use of DSMF-BIA to measure muscle mass.

Longitudinal Adherence to Diabetes Quality Indicators and Cardiac Disease: A Nationwide Population-Based Historical Cohort Study of Patients With Pharmacologically Treated Diabetes.

Background Evidence of the cardiovascular benefits of adherence to quality indicators in diabetes care over a period of years is lacking. Methods and Results We conducted a population-based, historical cohort study of 105 656 people aged 45 to 80 with pharmacologically treated diabetes and who were free of cardiac disease in 2010. Data were retrieved from electronic medical records of the 4 Israeli health maintenance organizations. The association between level of adherence to national quality indicators (2006-2010: adherence assessment) and incidence of cardiac outcome; ischemic heart disease or heart failure (2011-2016: outcome assessment) was estimated using Cox proportional hazards models. During 529 551 person-years of follow-up, 19 246 patients experienced cardiac disease. An inverse dose-response association between the level of adherence and risk of cardiac morbidity was shown for most of the quality indicators. The associations were modified by age, with stronger associations among younger patients (<65 years). Low adherence to low-density lipoprotein cholesterol testing (≤2 years) during the first 5 years was associated with 41% increased risk of cardiac morbidity among younger patients. Patients who had uncontrolled low-density lipoprotein cholesterol in all first 5 years had hazard ratios of 1.60 (95% CI, 1.49-1.72) and 1.23 (95% CI, 1.14-1.32), among patients aged <65 and ≥65 years, respectively, compared with those who achieved target level. Patients who failed to achieve target levels of glycated hemoglobin or blood pressure had an increased risk (hazard ratios, 1.50-1.69) for cardiac outcomes. Conclusions Longitudinal adherence to quality indicators in diabetes care is associated with reduced risk of cardiac morbidity. Implementation of programs that measure and enhance quality of care may improve the health outcomes of people with diabetes.

Israeli COVID lockdowns mildly reduced overall use of preventive health services, but exacerbated some disparities.

BACKGROUND: During 2020, Israel experienced two COVID-19-related lockdowns that impacted the provision of primary and secondary preventive care. METHODS: We examined the month-by-month performance of selected preventive care services using data from Israel's national Quality Indicators in Community Healthcare program. Process of care measures included hemoglobin A1c (HbA1c) testing, cholesterol testing, colon cancer screening and mammography. Intermediate outcome measures included low-density lipoprotein control and HbA1c control. Measures were stratified by sex and by area-level socioeconomic position (SEP). Diabetes and mammography are presented in this abstract due to space limitations. RESULTS: Annual HbA1c testing among persons with diabetes decreased from 90.9% in 2019 to 88.0% in 2020. Performance of HbA1c tests during lockdown months was as low as half the usual amount. There were compensatory increases in testing during post-lockdown months that did not quite make up for the missed tests. In 2019, 9.0% of Israelis with diabetes had poor glycemic control (HbA1c ≥ 9.0); in 2020, it was 8.8%. In total, 4.5% fewer mammograms were performed in 2020 compared with 2019. Women in the lowest SEP level performed 10.4% fewer mammograms in 2020 than in 2019, while women in the highest SEP level performed 3.1% more mammograms. CONCLUSIONS: Prolonged COVID lockdowns in 2020 were associated with marked decreases in the performance of preventive health services during those months. Compensatory spikes following the end of lockdowns partly, but did not completely, make up for the missed care. COVID lockdowns may have exacerbated socioeconomic disparities in some preventive health services.

Advanced Hemodynamic Monitoring Allows Recognition of Early Response Patterns to Diuresis in Congestive Heart Failure Patients.

There are no clear guidelines for diuretic administration in heart failure (HF), and reliable markers are needed to tailor treatment. Continuous monitoring of multiple advanced physiological parameters during diuresis may allow better differentiation of patients into subgroups according to their responses. In this study, 29 HF patients were monitored during outpatient intravenous diuresis, using a noninvasive wearable multi-parameter monitor. Analysis of changes in these parameters during the course of diuresis aimed to recognize subgroups with different response patterns. Parameters did not change significantly, however, subgroup analysis of the last quartile of treatment showed significant differences in cardiac output, cardiac index, stroke volume, pulse rate, and systemic vascular resistance according to gender, and in systolic blood pressure according to habitus. Changes in the last quartile could be differentiated using k-means, a technique of unsupervised machine learning. Moreover, patients' responses could be best clustered into four groups. Analysis of baseline parameters showed that two of the clusters differed by baseline parameters, body mass index, and diabetes status. To conclude, we show that physiological changes during diuresis in HF patients can be categorized into subgroups sharing similar response trends, making noninvasive monitoring a potential key to personalized treatment in HF.

[ANTIBIOTIC USE IN COMMUNITY HEALTHCARE IN ISRAEL AS REFLECTED IN THE NATIONAL PROGRAM FOR QUALITY INDICATORS].

AIMS: We sought to evaluate the extent of overall and second-line systemic antibiotic use in the Israeli community, to compare them to international data and to monitor temporal trends. BACKGROUND: Antibiotic overuse is prevalent and has grave implications, primarily the emergence of resistant pathogens - an urgent public health concern worldwide. METHODS: The Israel National Program for Quality Indicators in Community Healthcare (QICH) obtains data from electronic medical records from the 4 health-plans, covering the entire civilian population. We assessed 2 quality indicators, compatible with those of the OECD: 1. Overall volume of antibiotics for systemic use dispensed. 2. Volume of second-line antibiotics as a proportion of all systemic antibiotics. Analysis was stratified by gender, age and socio-economic position (SEP). RESULTS: The volume of systemic antibiotics dispensed in 2016 was 20.76 DDD/1000 person/day, with second-line antibiotics comprising 22.0% of the total. These values have been stable since 2014, and are higher compared with the OECD averages (20.61 DDD/1000 person/day, and 17.02% in 2015). Both overall volume and the proportion of second-line antibiotics rose with age and were higher among women especially in the 20-40 years age bracket (overall volume of 23.98 DDD/1000 person/day, proportion of second-line antibiotics of 23.98% VS 17.41 and 19.17% in men). Higher overall use was observed among patients of low SEP. CONCLUSIONS: The observed volume of systemic antibiotics and the proportion of second-line antibiotics dispensed in the Israeli community were stable and high. Higher use was observed among older individuals, women and patients of low SEP. Our results call for the implementation of a national-level, community-based antibiotic stewardship program. QICH might serve to monitor such a program.

The quality of primary care provided to the elderly in Israel.

BACKGROUND: In view of increasing global and local trends in population ageing and the high healthcare utilization rates among the elderly, this study assesses the quality of primary care provided to the elderly population in Israel. It examines changes in quality over time, how quality varies across sub-groups of the elderly, and how quality in Israel compares with other countries. Data originate from the National Program for Quality Indicators in Community Healthcare (QICH), which operates in full collaboration with Israel's four HMOs. METHODS: The study population included all elderly Israeli residents aged 65 years or older during 2002-2015 (N = 879,671 residents in 2015). Seven elderly-specific quality indicators from within the QICH framework were included: influenza and pneumococcal vaccinations, benzodiazepine overuse, long-acting benzodiazepine use, body weight documentation, weight loss and underweight. In addition, two non-age specific quality indicators relating to diabetes mellitus were included: the rate of HbA1C documentation and uncontrolled diabetes. Data were collected from patient electronic medical records (EMR) in accordance with each HMO, and aggregated by three variables: gender, age, and socio-economic position (SEP). RESULTS: During the measurement period, vaccination rates significantly increased (Influenza: from 42.0% in 2002 to 63.2% in 2015; and pneumococcal vaccination: from 25.8% in 2005 to 77.0% in 2015). Body weight documentation (in 65-74 year old persons) increased from only 16.3% in 2003 to 80.9% in 2015. The rate of underweight (BMI < 23 kg/m2) and significant weight-loss (10% or more of their body weight) was only measured in 2015. The overall rate of benzodiazepine overuse remained steady from 2011 to 2015 at around 5%, while the rate of long-acting benzodiazepine use decreased from 3.8% in 2011 to 2.4% in 2015. The rate of HbA1c documentation for elderly diabetics was higher than for non-elderly diabetics in 2015 (92.2% vs 87.9%). The rate of uncontrolled diabetes was lower for the elderly than the non-elderly population in 2015 (6.9% vs. 15.7%). Gender disparities were observed across all measures, after age stratification, with worse indicator rates among females compared to males. SEP-disparities were not consistent across measures. In all indicators except benzodiazepine overuse, Israel showed a higher quality of care for the elderly in comparison with the international healthcare community. CONCLUSIONS: Overall, the quality of care received by elderly Israelis has improved substantially since measurements first began; yet, females receive lower quality care than males. Monitoring results of primary care quality indicators can contribute to population's successful aging; both chronic conditions at earlier ages (e.g. diabetes), and short-term hazardous conditions such as the use of potentially harmful medications and weight loss should be evaluated.

Trends in the performance of quality indicators for diabetes care in the community and in diabetes-related health status: an Israeli ecological study.

BACKGROUND: Israel is one of the few countries that have a national program for quality assessment of community healthcare. We aimed to evaluate whether improved performance in diabetes care was associated with improved health of diabetic patients on a national level. METHODS: We conducted a nationwide ecological study estimating improvements in diabetes-related quality indicators and health outcomes. We estimated both correlations between composite measures of diabetes-related quality indicators and selected outcomes, and assessed through a joinpoint analysis whether trends in selected outcomes changed 4 years after the inception of the national program. RESULTS: Between 2002 and 2010, the prevalence of diabetes in Israeli adults increased from 4.8% to 7.4%. During these years, an improvement was noticed in most quality indicators (from 53% to 75% for the composite score). Declines were noted in rates of blindness, diabetes-related end-stage kidney disease, lower limbs amputations and diabetes-related mortality. Significant accelerations in decline were noted for amputations in men and diabetes-related mortality in both Arab men and women 4 years after the inception of the national program. CONCLUSION: This study suggests that Israel's national program for quality indicators in diabetes care in the community has probably had a significant impact on the health status of the whole population and may have contributed to narrowing gaps in life expectancy between Israeli Jews and Arabs. Future studies based on individual-level data are needed to confirm these results.

Cancer and Aging - the Inflammatory Connection.

Aging and cancer are highly correlated biological phenomena. Various cellular processes such as DNA damage responses and cellular senescence that serve as tumor suppressing mechanisms throughout life result in degenerative changes and contribute to the aging phenotype. In turn, aging is considered a pro-tumorigenic state, and constitutes the single most important risk factor for cancer development. However, the causative relations between aging and cancer is not straight forward, as these processes carry contradictory hallmarks; While aging is characterized by tissue degeneration and organ loss of function, cancer is a state of sustained cellular proliferation and gain of new functions. Here, we review the molecular and cellular pathways that stand in the base of aging related cancer. Specifically, we deal with the inflammatory perspective that link these two processes, and suggest possible molecular targets that may be exploited to modify their courses.

Physician-facilitated designation of proxy decision-makers: family physician perceptions.

BACKGROUND: Among the challenges encountered during the care of patients at the end-of-life (EOL), eliciting preferences of patients with whom there is no ability to communicate is common and stressful for all those concerned and charged with patient care. Legal facilities available include patient delegation of proxy decision-makers (PDM) prior to communication incapacity. We sought to estimate family physician awareness and attitude with regard to these aspects of patient care. METHODS: A telephone survey of family physicians in the Jerusalem, Israel, district using a standard questionnaire. RESULTS: 74 family physicians responded to the survey. The response rate was 42 % and the cooperation rate was 66 %. Most of the respondents, (64 %), reported knowing that the PDM delegation facility exists, though only 24 % claimed to have suggested to their patients that they consider this option. Approximately three-quarters, (78 %), treat patients with whom they discussed other aspects of severe disease, disability or EOL. None of the physicians working predominantly with religiously observant groups reported suggesting PDM delegation. CONCLUSIONS: There is an apparent gap between family physician knowledge and their performance to empower the persistence of patient autonomy, should communication ability cease. System-wide interventions to increase EOL communication skills, starting at medical school and henceforth, are necessary in order to promote better EOL care and meaningful resource use.

The effect of medical students' gender, ethnicity and attitude towards poetry-reading on the evaluation of a required, clinically-integrated poetry-based educational intervention.

BACKGROUND: Art-based interventions are widely used in medical education. However, data on the potential effects of art-based interventions on medical students have been limited to small qualitative studies on students' evaluation of elective programs, and thus their findings may be difficult to generalize. The goal of this study is to examine, in an unselected students' population, the effect of students' gender, ethnicity and attitude towards poetry on their evaluation of a clinically-integrated poetry-based educational intervention. METHODS: A required Clinically- Oriented Poetry-reading Experience (COPE) is integrated into the 4th year internal medicine clerkship. We constructed a questionnaire regarding the program's effects on students. Students completed the questionnaire at the end of the clerkship. We performed a Confirmatory Factor Analysis, and examined the relationship between students' evaluation of the program and students' ethnicity, gender, attitude towards poetry-reading, and the timing of the program (early/late) during the fourth year. RESULTS: 144 students participated in the program, of which 112 completed the questionnaires. We identified two effect factors: "student-patient" and "self and colleagues". The average score for "student-patient" factor was significantly higher as compared to the "self and colleagues" factor.Evaluation the "student- patient" effect factor was higher among Arab and Druze as compared to Jewish students. Students' attitude towards poetry-reading did not correlate with the "student-patient" effect, but correlated with the "self and colleagues" effect. The evaluation of the "self and colleagues" effect was higher among students who participated in the program during their second as compared with the first clerkship. Students' gender was not associated with any of the effects identified. Students favored obligatory participation in COPE as compared with elective course format. CONCLUSIONS: According to students' evaluation, a format of integrated, obligatory poetry-based intervention may be suitable for enhancing "student-patient" aims in heterogeneous student populations. The higher evaluation of the "patient-student" effect among Arab and Druze as compared to Jewish students may be related to cultural differences in the perception of this component of medical professionalism. Further research can provide insight into the effect of cultural and ethnic differences on actual empathy of medical students in patient encounters.

Community healthcare in Israel: quality indicators 2007-2009.

BACKGROUND: The National Program for Quality Indicators in Community Healthcare in Israel (QICH) was developed to provide policy makers and consumers with information on the quality of community healthcare in Israel. In what follows we present the most recent results of the QICH indicator set for 2009 and an examination of changes that have occurred since 2007. METHODS: Data for 28 quality indicators were collected from all four health plans in Israel for the years 2007-2009. The QICH indicator set examined six areas of healthcare: asthma, cancer screening, cardiovascular health, child health, diabetes and immunizations for older adults. RESULTS: Dramatic increases in the documentation of anthropometric measures were observed over the measurement period. Documentation of BMI for adolescents and adults increased by 30 percentage points, reaching rates of 61% and 70%, respectively, in 2009. Modest increases (3%-7%) over time were observed for other primary prevention quality measures including immunizations for older adults, cancer screening, anemia screening for young children, and documentation of cardiovascular risks. Overall, rates of recommended care for chronic diseases (asthma, cardiovascular disease and diabetes) increased over time. Changes in rates of quality care for diabetes were varied over the measurement period. CONCLUSIONS: The overall quality of community healthcare in Israel has improved over the past three years. Future research should focus on the adherence to quality indicators in population subgroups and compare the QICH data with those in other countries. In addition, one of the next steps in assessing and further improving healthcare quality in Israel is to relate these process and performance indicators to health outcomes.

Severe malnutrition due to systemic lupus erythematosus associated protein losing enteropathy.

Systemic lupus erythematosus most often targets organs such as joints, serosa, skin, bone marrow, and the kidneys. Gastrointestinal complications are uncommon, and among these, protein losing enteropathy is particularly rare. We present a young woman who suffered from chronic abdominal pain and diarrhea, developed severe malnutrition, and was eventually diagnosed with systemic lupus erythematosus and associated protein losing enteropathy.

Janus Kinase V617F mutation in cigarette smokers.

The JAK2 V617F mutation is responsible for the constitutive activation of the erythropoietin receptor signaling pathway in most cases of polycythemia vera (PV). The mutation has also been described in healthy people. As smoking may result in secondary polycythemia, the goal of this trial was to examine the effect of smoking on the prevalence of the JAK2 mutation and its correlation to erythrocytosis. The study was case-control. Hospitalized smokers (n = 81) and nonsmokers (n = 61) were recruited. Serum was drawn for complete blood count, erythropoietin, ferritin and venous blood gases. JAK2 mutation was analyzed by highly sensitive allele-specific Quantitative Real Time PCR. The JAK2 mutation was found in 29/81 (35.8%) of smokers in comparison to only 9/61 (14.8%) of the control group (P = 0.007). The frequency of the mutation among smokers who were positive for the JAK2 mutation had a mean of 6.78 × 10(-4) ± 1.08 × 10(-3) vs. 1.51 × 10(-4) ± 2.04 × 10(-4) among nonsmokers (P = 0.027). Both frequencies are much lower than those found in PV. There was a medium correlation between older age and mutation frequency in nonsmokers (r= 0.67, P = 0.043). Hematocrit was higher in smokers (47.8 ± 6 vs. 41.7 ± 4.7, P < 0.0001), but no correlation was found to JAK2 mutation. In a cohort of hospitalized smokers and nonsmokers, JAK2 mutation was more prevalent and found in higher frequencies among smokers than nonsmokers. We suggest that accelerated erythropoiesis renders the cells susceptible to JAK2 mutation.