Identifying characteristics of women cancer survivors with a recent history of falls.

2017 ◽  
Vol 35 (5_suppl) ◽  
pp. 113-113
Author(s):  
Kerri M. Winters-Stone ◽  
Thais Reis ◽  
Sydnee Stoyles ◽  
Nathan Dieckmann
Keyword(s):  
Cancer Survivors ◽  
Stepwise Regression ◽  
Stance Phase ◽  
Secondary Data ◽  
Self Report ◽  
Cancer Type ◽  
Chair Stand ◽  
Leg Strength ◽  
History Of ◽  
Base Of Support

113 Background: Epidemiologic data suggest that fall rates are increased among women with a history of cancer compared to women never diagnosed with cancer. However, the unique characteristics of women cancer survivors who fall are not completely understood. Our study aimed to identify the demographic, clinical and physical attributes of women treated for cancer and who experienced falling in the past year. Methods: Secondary data analysis of baseline data from aging women cancer survivors (greater than 50 years of age) about to participate in clinical exercise trials (n = 611). Based on self-report of having fallen in the last year (yes/no), women were compared on the following: age, cancer type (breast vs. other), cancer stage, time since diagnosis, presence of neuropathy or pain, comorbidities, BMI, physical activity, maximal leg press strength, chair stand time, walk speed, gait patterns, and the short physical performance battery (sPPB). Stepwise regression was run to determine attributes significantly associated with fall history. Results: 28% of women reported falling in the last year (n = 173) and 79% of fallers experienced a related injury. Women cancer survivors who fell were significantly more likely than women who did not fall to have: not received chemotherapy (25% vs. 13%), higher morbidity scores (2.2 vs. 1.8), higher BMI (30.6 vs. 29.2 kg/m2), more neuropathy (49 vs. 39%), wider base of support (10.0 vs. 8.8 cm), more of the gait cycle spent in the stance phase (64 vs. 63%), longer chair stand times (12.8 vs. 11.9 sec.), and lower PPB scores (10.3 vs. 10.8). In stepwise regression models, receipt of chemotherapy, comorbidities, maximal leg strength, neuropathy, base of support, and % time in the stance phase of gait were significantly associated with the odds of having fallen in the last year. Conclusions: Women cancer survivors over 50 years old have a prevalence of falls approaching the 33% reported by the general population of women over 65; however, the rate of injurious falls is much higher in our sample of women cancer survivors. Fall prevention should be considered in women cancer survivors at an earlier age than usual for older women, particularly for survivors at higher risk for falls, and focus on exercise to improve gait and leg strength.

scholarly journals Falls, Functioning, and Disability Among Women With Persistent Symptoms of Chemotherapy-Induced Peripheral Neuropathy

2017 ◽  
Vol 35 (23) ◽  
pp. 2604-2612 ◽  
Author(s):  
Kerri M. Winters-Stone ◽  
Fay Horak ◽  
Peter G. Jacobs ◽  
Phoebe Trubowitz ◽  
Nathan F. Dieckmann ◽  
...  
Keyword(s):  
Peripheral Neuropathy ◽  
Cancer Survivors ◽  
Physical Function ◽  
Functional Decline ◽  
Self Report ◽  
Functional Rehabilitation ◽  
Persistent Symptoms ◽  
Care Plans ◽  
Leg Strength ◽  
Base Of Support

Purpose Chemotherapy-induced peripheral neuropathy (CIPN) may persist after treatment ends and may lead to functional decline and falls. This study compared objective and self-report measures of physical function, gait patterns, and falls between women cancer survivors with and without symptoms of CIPN to identify targets for functional rehabilitation. Methods A secondary data analysis of 512 women cancer survivors (age, 62 ± 6 years; time since diagnosis, 5.8 ± 4.1 years) categorized and compared women self-reporting symptoms of CIPN (CIPN+) with asymptomatic women (CIPN−) on the following: maximal leg strength, timed chair stand, physical function battery, gait characteristics (speed; step number, rate, and length; base of support), self-report physical function and disability, and falls in the past year. Results After an average of 6 years after treatment, 47% of women still reported symptoms of CIPN. CIPN+ had significantly worse self-report and objectively measured function than did CIPN−, with the exception of maximal leg strength and base of support during a usual walk. Gait was slower among CIPN+, with those women taking significantly more, but slower and shorter, steps than did CIPN− (all P < .05). CIPN+ reported significantly more disability and 1.8 times the risk of falls compared with CIPN− ( P < .0001). Increasing symptom severity was linearly associated with worsening function, increasing disability, and higher fall risk (all P < .05). Conclusion This work makes a significant contribution toward understanding the functional impact of CIPN symptoms on cancer survivors. Remarkably, 47% of women in our sample had CIPN symptoms many years after treatment, together with worse function, greater disability, and more falls. CIPN must be assessed earlier in the clinical pathway, and strategies to limit symptom progression and to improve function must be included in clinical and survivorship care plans.

scholarly journals 1053 Sleep Duration And Timing Associated With History Of Breast Prostate And Skin Cancer: Data From A Nationally-representative Sample

SLEEP ◽  
2020 ◽  
Vol 43 (Supplement_1) ◽  
pp. A400-A400
Author(s):  
B Piro ◽  
S Garland ◽  
P Jean-Pierre ◽  
B Gonzalez ◽  
A Seixas ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Prostate Cancer ◽  
Skin Cancer ◽  
Cancer Survivors ◽  
Sleep Duration ◽  
Sleep Disturbances ◽  
Sleep Problems ◽  
Cancer Type ◽  
Cancer Data ◽  
History Of

Abstract Introduction Sleep disturbances are a common problem among cancer survivors. Also, cancer patients can have altered circadian rhythms and these changes can continue to affect the patient long after the conclusion of their treatment. This analysis aims to investigate how the sleep and wake times of cancer survivors differ from the rest of the population, depending on the type of cancer. Methods Data from the 2015-2016 National Health and Nutrition Examination Survey were used. Population-weighted data on N=5,581 individuals provided complete data. History of breast, prostate, and skin cancer (melanoma or other) was self-reported. Sleep duration was self-reported in half-hour increments, and typical bedtime and waketime was self-reported. Covariates included age, sex, and race/ethnicity. Weighted linear regressions with sleep duration, bedtime and waketime were examined, with each cancer type as predictor. Results Prevalence was 1.7% for prostate cancer, 1.5% for breast cancer, 2.3% for non-melanoma skin cancer, and 0.8% for melanoma. In adjusted analyses, prostate cancer was associated with an additional 26.5 minutes of average total sleep (95%CI 2.2,50.9, p=0.03), a 23.1 bedtime minutes earlier (95%CI -40.4,-5.8, p=0.009), and no difference in waketime. Breast cancer was associated with a bedtime that was 41.1 minutes later (95%CI 10.3,72.0, p=0.009) and a waketime that was 48.7 minutes later (95%CI 12.5,84.9, p=0.008), but no difference in sleep duration. No statistically significant effects were seen for either type of skin cancer, melanoma or non-melanoma. Conclusion Prostate cancer was associated with an earlier bedtime and associated increased sleep time. Breast cancer, on the other hand, was associated with a phase delay of the sleep period but no change in sleep duration. Skin cancer was not associated with differences in sleep duration or timing. These findings may have implications for not only treatment of sleep problems in different types of cancer, but also possible circadian mechanisms. Support Dr. Grandner is supported by R01MD011600

scholarly journals Alcohol Use Among Patients With Cancer and Survivors in the United States, 2000–2017

2020 ◽  
Vol 18 (1) ◽  
pp. 69-79 ◽  
Author(s):  
Nina N. Sanford ◽  
David J. Sher ◽  
Xiaohan Xu ◽  
Chul Ahn ◽  
Anthony V. D’Amico ◽  
...  
Keyword(s):  
Risk Factor ◽  
Alcohol Use ◽  
Cancer Survivors ◽  
Binge Drinking ◽  
Cancer Diagnosis ◽  
The United States ◽  
Self Report ◽  
Smoking History ◽  
Cancer Type ◽  
Moderate Drinking

Background: Alcohol use is an established risk factor for several malignancies and is associated with adverse oncologic outcomes among individuals diagnosed with cancer. The prevalence and patterns of alcohol use among cancer survivors are poorly described. Methods: We used the National Health Interview Survey from 2000 to 2017 to examine alcohol drinking prevalence and patterns among adults reporting a cancer diagnosis. Multivariable logistic regression was used to define the association between demographic and socioeconomic variables and odds of self-reporting as a current drinker, exceeding moderate drinking limits, and engaging in binge drinking. The association between specific cancer type and odds of drinking were assessed. Results: Among 34,080 survey participants with a known cancer diagnosis, 56.5% self-reported as current drinkers, including 34.9% who exceeded moderate drinking limits and 21.0% who engaged in binge drinking. Younger age, smoking history, and more recent survey period were associated with higher odds of current, exceeding moderate, and binge drinking (P<.001 for all, except P=.008 for excess drinking). Similar associations persisted when the cohort was limited to 20,828 cancer survivors diagnosed ≥5 years before survey administration. Diagnoses of melanoma and cervical, head and neck, and testicular cancers were associated with higher odds of binge drinking (P<.05 for all) compared with other cancer diagnoses. Conclusions: Most cancer survivors self-report as current alcohol drinkers, including a subset who seem to engage in excessive drinking behaviors. Given that alcohol intake has implications for cancer prevention and is a potentially modifiable risk factor for cancer-specific outcomes, the high prevalence of alcohol use among cancer survivors highlights the need for public health strategies aimed at the reduction of alcohol consumption.

Comparison of physical function and falls among women with persistent symptoms of chemotherapy-induced peripheral neuropathy.

2016 ◽  
Vol 34 (3_suppl) ◽  
pp. 130-130 ◽  
Author(s):  
Kerri M Winters-Stone ◽  
Coleman Hilton ◽  
Shiuh-Wen Luoh ◽  
Peter Jacobs ◽  
Sarah Faithfull ◽  
...  
Keyword(s):  
Peripheral Neuropathy ◽  
Cancer Survivors ◽  
Physical Function ◽  
Post Treatment ◽  
Self Report ◽  
Cancer Site ◽  
Gait Patterns ◽  
Base Of Support ◽  
Asymptomatic Women ◽  
Time Since Diagnosis

130 Background: Chemotherapy-induced peripheral neuropathy (CIPN) is a side effect of cancer treatment that may persist and impact physical function. Studies that quantify functional consequences associated with CIPN in post-treatment cancer survivors are rare, based on self-report, or use small samples. The purpose of this study was to compare objective and self-report measures of physical function, gait patterns, and falls between women cancer survivors with or without symptoms of CIPN. Methods: Baseline assessments from 678 women cancer survivors enrolled in exercise trials were available for analysis. Women who self-reported symptoms of CIPN (CIPN+) were compared to asymptomatic women (CIPN-) on the following: maximal leg press strength (LPmax); timed chair stand (CS), physical performance battery (PPB), gait patterns (speed (SP), step number (SN), stride length (SL), base of support (BOS), % time in double support (%DS)), self-report physical function (PF) and disability (DIS), and falls in the past year (% fallers). Group comparisons were made using analysis of covariance, adjusting for time since diagnosis and cancer site (breast or other). Results: After excluding women who had diabetes, were premenopausal at cancer diagnosis or had not received chemotherapy, 462 cases were analyzed (age: 62±6yrs; time since diagnosis: 5.8±4.1yrs). CIPN+ (N = 210) and CIPN- (N = 252) groups significantly differed on all measures (*p < 0.05; **p < 0.01), except maximal leg strength and base of support during a usual walk, with worse performance for CIPN+. Conclusions: In our sample of women cancer survivors, 45% had symptoms of CIPN an average of 6 years post treatment and significantly worse physical function, altered gait patterns and more falls than asymptomatic women. CIPN should be addressed early in treatment and strategies to improve function and mobility and prevent falls need to be tested. [Table: see text]

Patient preferences for cancer survivorship care: Results of an online survey.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 12064-12064
Author(s):  
Deanna J. Attai ◽  
Matthew S. Katz ◽  
Elani Streja ◽  
Jui-Ting Hsiung ◽  
Beverly A Zavaleta ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Primary Care ◽  
Cancer Survivors ◽  
Patient Preferences ◽  
Early Stage ◽  
Cancer Type ◽  
Survivorship Care ◽  
Follow Up Care ◽  
History Of

12064 Background: Nearly 17 million cancer survivors live in the United States. Workforce shortages are projected to diminish the number of available medical oncologists (MOs) to care for newly diagnosed patients with cancer and for the growing number of cancer survivors. Models of survivorship care include oncologist-led, primary care-led, and shared care approaches. Recent proposals recommend a risk-stratified approach to care, guided by individual and cancer-specific factors, but there is little evidence regarding patient preferences for non-oncologist survivorship care. Methods: We developed a survey in partnership with patient advocates. The primary endpoints were patient-reported comfort with survivorship care by a primary care provider (PCP) or in a dedicated survivorship clinic. We distributed the survey to online, cancer-specific patient communities from June to August 2020. Logistic regression analyses were adjusted for patient age, race and ethnicity, insurance, and cancer type and stage. Results: Of 1166 responses, 975 surveys were complete and available for analysis. Respondents were primarily women (91%), white (92%), and US residents (73%); 78% had a college or graduate degree. Two-thirds had private insurance. Thirty-six different cancer types were reported; 61% of respondents had breast cancer, and 25% were treated for more than one type of cancer. Most respondents (83%) had nonmetastatic disease, 74% reported experiencing late effects of cancer therapy, and almost all (93%) had a PCP. Only 21% of respondents were comfortable seeing a PCP (versus MO) for survivorship care, including cancer follow-up, side effect management, and monitoring for recurrence or progression. About half (55%) were comfortable with follow-up in a survivorship clinic instead of with their MO. Multivariable analyses showed no significant associations between age, race or ethnicity, insurance, cancer type, or stage at diagnosis and comfort with follow-up care from a PCP or in a survivorship clinic. In analyses restricted to the 439 respondents with a history of early-stage breast cancer, the 239 (54%) who were within 1 to 5 years of diagnosis were less comfortable with PCP versus MO follow-up, compared with the 52 (12%) who were > 15 years from diagnosis (OR 0.40; 95% CI 0.20–0.75; p= 0.004). In this sub-analysis, time from diagnosis was not associated with comfort being seen in a survivorship clinic. Conclusions: In our study, most patients with a history of cancer were not comfortable receiving follow-up care from their PCP. It is often recommended that survivors of early-stage breast cancer transition to primary care for follow-up and surveillance, but our study revealed comfort with this approach only many years after diagnosis. While both PCP survivorship training and patient confidence in PCP follow up is needed, preferences of cancer survivors should be considered in designing new models of survivorship care.

PERKEMBANGANKAWASAN PECINAN SEMARANG

2017 ◽  
Vol 10 (6) ◽  
pp. 361
Author(s):  
Monica Latu Melati ◽  
Ariadne Kristia Nataya ◽  
Alfonsus Arianto Wibowo
Keyword(s):  
Urban Form ◽  
Secondary Data ◽  
Primary Data ◽  
Physical Factors ◽  
Review Analysis ◽  
Urban Heritage ◽  
Observation Area ◽  
Cultural Artefact ◽  
History Of ◽  
Special District

Abstract:Semarang Chinatown  is a special  district in Semarang City  known with its chineese culture, where chineese citizen of Semarang have been living  for centuries. The sustained chineese culture in this area makes Semarang Chinatwon as an urban heritage and cultural artefact in Semarang City. The aims for this paper are to investigate the factors shaping Chinatown Semarang, the development of Chinatown Semarang from time  to time, the urban form elements in Semarang Chinatown, and the correlation between morphological components of Semarang Chinatown. This writing use some review methods, first theoritical overview to get secondary data about physical or non-physical factors forming city, second observation area such as collecting photos and interviewing to get primary data. Data review analysis use qualitative data analysis which is configure with the problems and aims that have been appointed.Keywords:elements of urban form, morphological components, history of Semarang ChinatownAbstrak: Kawasan Pecinan Semarang adalah sebuah kawasan di kota Semarang yang sangat kental dengan budaya Tionghoa. Di sinilah warga keturunan Tionghoa sejak berabad-abad silam menetap di Semarang. Adanya budaya Tionghoa yang masih sangat terjaga menjadikan Kawasan Pecinan Semarang ini sebagai kawasan urban heritage dan artefact budaya di kota Semarang. Tujuan penulisan adalah untuk menemukan faktor pembentuk Kawasan Pecinan Semarang, mengetahui perkembangan Kawasan Pecinan Semarang dari masa ke masa, mengetahui pola bentuk dan elemen kawasan pada Kawasan Pecinan Semarang, serta mengetahui kaitan antara faktor pembentuk kawasan terhadap perkembangan Kawasan Pecinan Semarang. Penulisan ini menggunakan metode kajian berupa tinjauan teori untuk memperoleh data sekunder mengenai faktor-faktor pembentuk kota baik secara fisik maupun non fisik, serta observasi lapangan berupa pengumpulan foto yang dilengkapi dengan wawancara untuk memperoleh data primer. Analisis data kajian dilakukan dengan menggunakan analisis data kualitatif yang disesuaikan dengan permasalahan dan tujuan yang telah ditetapkan.Kata kunci:Elemen Kawasan, Faktor Pembentuk Kawasan, PerkembanganSejarah Kawasan Pecinan Semarang

Profiles of Psychosocial and Clinical Functioning in Adolescence and Risk for Later Depression

2019 ◽  
Author(s):  
Thomas M Olino ◽  
Daniel Klein ◽  
John Seeley
Keyword(s):  
Psychosocial Functioning ◽  
Adolescent Depression ◽  
Depressive Disorders ◽  
Self Report ◽  
Initial Assessment ◽  
Predictive Utility ◽  
Regression Methods ◽  
History Of ◽  
History Of Depression

Background: Most studies examining predictors of onset of depression focus on variable centered regression methods that focus on effects of multiple predictors. In contrast, person-centered approaches develop profiles of factors and these profiles can be examined as predictors of onset. Here, we developed profiles of adolescent psychosocial and clinical functioning among adolescents without a history of major depression. Methods: Data come from a subsample of participants from the Oregon Adolescent Depression Project who completed self-report measures of functioning in adolescence and completed diagnostic and self-report measures at follow-up assessments up to approximately 15 years after baseline. Results: We identified four profiles of psychosocial and clinical functioning: Thriving; Average Functioning; Externalizing Vulnerability and Family Stress; and Internalizing Vulnerability at the baseline assessment of participants without a history of depression at the initial assessment in mid- adolescence. Classes differed in the likelihood of onset and course of depressive disorders, experience of later anxiety and substance use disorders, and psychosocial functioning in adulthood. Moreover, the predictive utility of these classes was maintained when controlling for multiple other established risk factors for depressive disorders. Conclusions: This work highlights the utility of examining multiple factors simultaneously to understand risk for depression.

scholarly journals Incidence and multidimensional predictors of occasional and recurrent falls among Malaysian community‐dwelling older persons

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Theng Choon Ooi ◽  
Devinder Kaur Ajit Singh ◽  
Suzana Shahar ◽  
Nor Fadilah Rajab ◽  
Divya Vanoh ◽  
...  
Keyword(s):  
Incidence Rate ◽  
Older Persons ◽  
Test Score ◽  
Community Dwelling ◽  
Chair Stand ◽  
Stand Test ◽  
Chair Stand Test ◽  
Recurrent Falls ◽  
History Of

Abstract Background Falls incidence rate and comprehensive data on factors that predict occasional and repeated falls from large population-based studies are scarce. In this study, we aimed to determine the incidence of falls and identify predictors of occasional and recurrent falls. This was done in the social, medical, physical, nutritional, biochemical, cognitive dimensions among community-dwelling older Malaysians. Methods Data from 1,763 Malaysian community-dwelling older persons aged ≥ 60 years were obtained from the LRGS-TUA longitudinal study. Participants were categorized into three groups according to the presence of a single fall (occasional fallers), ≥two falls (recurrent fallers), or absence of falls (non-fallers) at an 18-month follow-up. Results Three hundred and nine (17.53 %) participants reported fall occurrences at an 18-month follow-up, of whom 85 (27.51 %) had two or more falls. The incidence rate for occasional and recurrent falls was 8.47 and 3.21 per 100 person-years, respectively. Following multifactorial adjustments, being female (OR: 1.57; 95 % CI: 1.04–2.36), being single (OR: 5.31; 95 % CI: 3.36–37.48), having history of fall (OR: 1.86; 95 % CI: 1.19–2.92) higher depression scale score (OR: 1.10; 95 % CI: 1.02–1.20), lower hemoglobin levels (OR: 0.90; 95 % CI: 0.81-1.00) and lower chair stand test score (OR: 0.93; 95 % CI: 0.87-1.00) remained independent predictors of occasional falls. While, having history of falls (OR: 2.74; 95 % CI: 1.45–5.19), being a stroke survivor (OR: 8.57; 95 % CI: 2.12–34.65), higher percentage of body fat (OR: 1.04; 95 % CI: 1.01–1.08) and lower chair stand test score (OR: 0.87; 95 % CI: 0.77–0.97) appeared as recurrent falls predictors. Conclusions Having history of falls and lower muscle strength were predictors for both occasional and recurrent falls among Malaysian community-dwelling older persons. Modifying these predictors may be beneficial in falls prevention and management strategies among older persons.

Do Migrants Receive Tuberculosis Education in China? Evidence From the China Migrants Dynamic Survey

2021 ◽  
pp. 109019812110003
Author(s):  
Zheng Zhu ◽  
Mengdi Guo ◽  
Tingyue Dong ◽  
Beibei Gong ◽  
Xia Zhao ◽  
...  
Keyword(s):  
Low Socioeconomic Status ◽  
Age Groups ◽  
Secondary Data ◽  
Total Sample ◽  
Self Report ◽  
Primary Methods ◽  
Logistic Regression Models ◽  
Key Population ◽  
Bulletin Boards ◽  
Nationally Representative

Background Migrants are the key population for tuberculosis (TB) transmission in China. However, it remains unknown how many migrants have received TB education and through what means. Objectives To identify the rate and methods of TB education among migrants in China by using nationally representative data. Method This study used secondary data analysis. The data were derived from the China Migrants Dynamic Survey 2014–2017. A total sample of 745,926 migrants was included in the following analysis. Information on TB education was collected through a self-report questionnaire. We used hierarchical logistic regression models to explore the relationship between the independent variables and the receipt of TB education. Results Only 30.4% ( n = 226,458) received TB education. Among all age-groups, participants between 65 and 69 years old had the highest TB education rate (33.4%). Bulletin boards (86.5%–91%), media (73% to 86.7%), and books/magazines (59.2%–67.4%) were the most common ways for migrants to receive TB education. Conclusions Our study showed the rates of TB education in each region of China and indicated the significant disparity among the seven regions. Traditional media, off-line medical consultation, community advocacy, and bulletin boards should be the primary methods of delivering TB education. TB education campaigns targeting migrants with a low socioeconomic status should be actively promoted.

419 STOP-Bang Score and History of Radiation Predicts Risk of Obstructive Sleep Apnea in Cancer Patients: A Machine Learning Study

SLEEP ◽  
2021 ◽  
Vol 44 (Supplement_2) ◽  
pp. A166-A166
Author(s):  
Ankita Paul ◽  
Karen Wong ◽  
Anup Das ◽  
Diane Lim ◽  
Miranda Tan
Keyword(s):  
Machine Learning ◽  
Obstructive Sleep Apnea ◽  
Sleep Apnea ◽  
Cancer Patients ◽  
Prediction Accuracy ◽  
Screening Tools ◽  
Cancer Type ◽  
Sleep Study ◽  
Obstructive Sleep ◽  
History Of

Abstract Introduction Cancer patients are at an increased risk of moderate-to-severe obstructive sleep apnea (OSA). The STOP-Bang score is a commonly used screening questionnaire to assess risk of OSA in the general population. We hypothesize that cancer-relevant features, like radiation therapy (RT), may be used to determine the risk of OSA in cancer patients. Machine learning (ML) with non-parametric regression is applied to increase the prediction accuracy of OSA risk. Methods Ten features namely STOP-Bang score, history of RT to the head/neck/thorax, cancer type, cancer stage, metastasis, hypertension, diabetes, asthma, COPD, and chronic kidney disease were extracted from a database of cancer patients with a sleep study. The ML technique, K-Nearest-Neighbor (KNN), with a range of k values (5 to 20), was chosen because, unlike Logistic Regression (LR), KNN is not presumptive of data distribution and mapping function, and supports non-linear relationships among features. A correlation heatmap was computed to identify features having high correlation with OSA. Principal Component Analysis (PCA) was performed on the correlated features and then KNN was applied on the components to predict the risk of OSA. Receiver Operating Characteristic (ROC) - Area Under Curve (AUC) and Precision-Recall curves were computed to compare and validate performance for different test sets and majority class scenarios. Results In our cohort of 174 cancer patients, the accuracy in determining OSA among cancer patients using STOP-Bang score was 82.3% (LR) and 90.69% (KNN) but reduced to 89.9% in KNN using all 10 features mentioned above. PCA + KNN application using STOP-Bang score and RT as features, increased prediction accuracy to 94.1%. We validated our ML approach using a separate cohort of 20 cancer patients; the accuracies in OSA prediction were 85.57% (LR), 91.1% (KNN), and 92.8% (PCA + KNN). Conclusion STOP-Bang score and history of RT can be useful to predict risk of OSA in cancer patients with the PCA + KNN approach. This ML technique can refine screening tools to improve prediction accuracy of OSA in cancer patients. Larger studies investigating additional features using ML may improve OSA screening accuracy in various populations Support (if any):

Sign in / Sign up

Export Citation Format

Share Document