Zero-shot learning to extract assessment criteria and medical services from the preventive healthcare guidelines using large language models.

OBJECTIVES: The integration of these preventive guidelines with Electronic Health Records (EHRs) systems, coupled with the generation of personalized preventive care recommendations, holds significant potential for improving healthcare outcomes. Our study investigates the feasibility of using Large Language Models (LLMs) to automate the assessment criteria and risk factors from the guidelines for future analysis against medical records in EHR. MATERIALS AND METHODS: We annotated the criteria, risk factors, and preventive medical services described in the adult guidelines published by United States Preventive Services Taskforce and evaluated 3 state-of-the-art LLMs on extracting information in these categories from the guidelines automatically. RESULTS: We included 24 guidelines in this study. The LLMs can automate the extraction of all criteria, risk factors, and medical services from 9 guidelines. All 3 LLMs perform well on extracting information regarding the demographic criteria or risk factors. Some LLMs perform better on extracting the social determinants of health, family history, and preventive counseling services than the others. DISCUSSION: While LLMs demonstrate the capability to handle lengthy preventive care guidelines, several challenges persist, including constraints related to the maximum length of input tokens and the tendency to generate content rather than adhering strictly to the original input. Moreover, the utilization of LLMs in real-world clinical settings necessitates careful ethical consideration. It is imperative that healthcare professionals meticulously validate the extracted information to mitigate biases, ensure completeness, and maintain accuracy. CONCLUSION: We developed a data structure to store the annotated preventive guidelines and make it publicly available. Employing state-of-the-art LLMs to extract preventive care criteria, risk factors, and preventive care services paves the way for the future integration of these guidelines into the EHR.

Patient Navigation Plus Tailored Digital Video Disc Increases Colorectal Cancer Screening Among Low-Income and Minority Patients Who Did Not Attend a Scheduled Screening Colonoscopy: A Randomized Trial.

BACKGROUND: Up to 50% of people scheduled for screening colonoscopy do not complete this test and no studies have focused on minority and low-income populations. Interventions are needed to improve colorectal cancer (CRC) screening knowledge, reduce barriers, and provide alternative screening options. Patient navigation (PN) and tailored interventions increase CRC screening uptake, however there is limited information comparing their effectiveness or the effect of combining them. PURPOSE: Compare the effectiveness of two interventions to increase CRC screening among minority and low-income individuals who did not attend their screening colonoscopy appointment-a mailed tailored digital video disc (DVD) alone versus the mailed DVD plus telephone-based PN compared to usual care. METHODS: Patients (n = 371) aged 45-75 years at average risk for CRC who did not attend a screening colonoscopy appointment were enrolled and were randomized to: (i) a mailed tailored DVD; (ii) the mailed DVD plus phone-based PN; or (iii) usual care. CRC screening outcomes were from electronic medical records at 12 months. Multivariable logistic regression analyses were used to study intervention effects. RESULTS: Participants randomized to tailored DVD plus PN were four times more likely to complete CRC screening compared to usual care and almost two and a half times more likely than those who were sent the DVD alone. CONCLUSIONS: Combining telephone-based PN with a mailed, tailored DVD increased CRC screening among low-income and minority patients who did not attend their screening colonoscopy appointments and has potential for wide dissemination.

Up to half of people scheduled for a screening colonoscopy do not complete this test. There is a need for interventions to improve knowledge about colorectal cancer (CRC) screening, enhance access to screening by offering alternative test options, foster skills for completing screening, and mitigate barriers. The purpose of this study was to compare the effects of two interventions aimed at increasing CRC screening­a mailed tailored digital video disc (DVD) alone versus the mailed DVD plus telephone-based patient navigation (PN)­for patients who had not completed a scheduled screening colonoscopy. We enrolled 371 patients aged 45­75 years who had no CRC risk factors other than age, who were scheduled for a screening colonoscopy but did not attend their appointment. Participants were randomized to receive either: (i) a mailed tailored DVD; (ii) the mailed DVD plus phone-based PN; or (iii) usual care. Those who received the tailored DVD plus PN were four times more likely to complete CRC screening with stool test or colonoscopy compared to usual care. Combining telephone-based PN with a mailed, tailored DVD increased CRC screening among low-income and minority patients who did not attend a scheduled screening colonoscopy appointment.

Factors Associated with Cancer Prevention/Risk Reduction Behaviors among Latinos.

Improving understanding of behaviors that increase or reduce cancer risk for different Hispanic groups is a public health priority; such knowledge is sparse in new gateway immigration locations such as Indiana. The aims of this study were to: 1) describe cancer beliefs and cancer preventive/risk reduction behaviors (physical activity, tobacco, and alcohol use) among Hispanic adults; 2) examine differences in cancer beliefs and preventive behaviors by country/territory of birth, socioeconomic status, and area of residence (urban vs. rural); and 3) determine predictors of engagement in cancer prevention and risk reduction behaviors in this population. A cross-sectional online survey targeted adult Indiana residents who identified as Latino, Hispanic, or Spanish recruited using Facebook-targeted advertising. Complete survey data from 1520 respondents were analyzed using descriptive, unadjusted, and adjusted models. The majority of respondents believed they were unlikely to get cancer but held many other fatalistic beliefs about cancer. Only 35.6% of respondents had received the HPV vaccine, 37.6% reported they were currently smoking cigarettes, and 64% reported occasional or frequent drinking of alcohol. Respondents spent an average of 3.55 days per week engaged in moderate exercise. Differences were observed by country/territory of birth, income, and education but not by rural residence status. Predictors of cancer risk/risk reduction behaviors were identified. The Hispanic population in Indiana is diverse and effective interventions for cancer prevention should be culturally targeted based on country/territory of birth and individually tailored based on cancer-related beliefs.

The Experience of Peripheral Neuropathy Symptoms in Breast Cancer Survivors With Diabetes.

BACKGROUND: Diabetes (type 2) is a risk factor for developing peripheral neuropathy (PN) symptoms in breast cancer survivors (BCS). Because PN symptoms are associated with deficits in physical functioning and quality of life, more information is needed about the effects of PN symptoms on the lives of BCS with diabetes. OBJECTIVE: The aim of this study was to describe the experiences of PN among BCS with diabetes from their own perspectives. INTERVENTIONS/METHODS: This substudy is part of a larger investigation examining factors associated with cancer-related cognitive impairment in cancer survivors. Female early-stage (stage I-III) BCS with diabetes and PN symptoms were eligible to participate. A qualitative descriptive approach using purposive sampling and semistructured interviews was used. Participant narratives were summarized using standard content analytic techniques. RESULTS: Eleven BCS with diabetes and PN symptoms were interviewed. Participants described PN symptoms that were varied, were often persistent, and had troublesome effects on their physical functioning and quality of life. Participants used a variety of self-management strategies and prescription and over-the-counter medications to manage their PN symptoms. Some said that having both cancer and diabetes exacerbated the PN symptoms and complicated symptom management. CONCLUSION: Peripheral neuropathy symptoms can have a profound effect on the lives of BCS with diabetes and should be addressed by healthcare providers. IMPLICATIONS FOR PRACTICE: Clinical care for this population should include ongoing assessment of PN symptoms, conversations about the effects of these symptoms on everyday life, evidence-based treatment for the symptoms, and support for symptom self-management.

Cancer-related knowledge, beliefs, and behaviors among Hispanic/Latino residents of Indiana.

BACKGROUND: Cancer is the leading cause of death for Hispanics in the USA. Screening and prevention reduce cancer morbidity and mortality. METHODS: This study administered a cross-sectional web-based survey to self-identified Hispanic residents in the state of Indiana to assess their cancer-related knowledge, beliefs, and behaviors, as well as to identify what factors might be associated with cancer screening and prevention. Chi-square and Fisher's exact test were used to compare associations and logistic regression used to develop both univariate and multivariate regression models. RESULTS: A total of 1520 surveys were completed, median age of respondents was 53, 52% identified as men, 50.9% completed the survey in Spanish, and 60.4% identified the USA as their country of birth. Most were not able to accurately identify ages to begin screening for breast, colorectal, or lung cancer, and there were significant differences in cancer knowledge by education level. US-born individuals with higher income and education more often believed they were likely to develop cancer and worry about getting cancer. Sixty eight percent of respondents were up-to-date with colorectal, 44% with breast, and 61% with cervical cancer screening. Multivariate models showed that higher education, lack of fatalism, older age, lower household income, and unmarried status were associated with cervical cancer screening adherence. CONCLUSIONS: Among a Hispanic population in the state of Indiana, factors associated with cervical cancer screening adherence were similar to the general population, with the exceptions of income and marital status. Younger Hispanic individuals were more likely to be adherent with breast and colorectal cancer screening, and given the higher incidence of cancer among older individuals, these results should guide future research and targeted outreach.

Partnering to Increase Colorectal Cancer Screening: Perspectives of Community Advisory Board Members.

The Patient-Centered Outcomes Research Institute (PCORI) defines engagement in research as the meaningful involvement of patients, caregivers, clinicians, insurers, and others throughout the entire research process-from planning to conducting the study to disseminating study results. The purposes of this paper are to (a) describe methods used to engage community members across the various phases of a PCORI-funded comparative effectiveness trial to increase colorectal cancer screening; and (b) report results of qualitative and quantitative evaluations of community advisory board members' experiences on this project. Decisions to join and stay engaged with the study included feeling valued and appreciated, being compensated, the opportunity to contribute to research based on their skills and expertise, and being committed to colon cancer prevention efforts. Challenges identified by advisory board members included the significant time commitment, transportation, and meeting location. Lessons learned and guidance for researchers committed to patient and community engagement are described.