RESUMEN

Background: Family health history (FHH) is central to human genomic profiling construction; however, there is no protocol for documenting FHH in a pedigree format in Vietnam.Aim: A "Gia Su Suc Khoe" (GSSK) tool was developed to create a user-friendly interface for collecting FHH and offering diseases' risk assessment.Results: A tool was described (https://giasusuckhoe.vn/) with good feedback from genetic counselors and family-medicine doctors. Among 20 surveys, 100% of respondents noted that the report accurately reflected their FHH and were satisfied with the tool's display. About 74% of familial conditions were covered. Overall, all constructive feedback has been adapted into the updated version.Conclusion: Gia Su Suc Khoe has the potential to significantly improve healthcare delivery and outcomes in Vietnam.

[Box: see text].

RESUMEN

Background: Physical activity is an essential aspect of living a healthy lifestyle, which offers a myriad of benefits for both the body and the mind. Nonetheless, a family history of obesity can pose significant challenges to achieving and maintaining a healthy weight, which is further exacerbated by the academic stress. Individuals who have a role model that prioritizes physical activity are more likely to develop positive and healthy habits in their lives. Objectives: The study aimed to examine how role models affect physical activity and its impact on addressing challenges of family obesity history and academic stress in school children. Methods: This cross-sectional study was conducted from October 2019 to April 2020 in three schools in Amritsar district selected by lottery method of simple random sampling. Out of 4226 students, 355 overweight and obese students were interviewed regarding their physical activity. An informed written assent and consent was taken from the children and guardians respectively. Mothers of 6 to 11-year-old students and students aged 12 to 16 were interviewed using a pre-designed, validated, semi-structured questionnaire. Results: 51.3% had a family history of obesity while 14.9% and 14.6% of parents of overweight students had hypertension and diabetes, respectively. Over half (52.7%) lacked interest in physical activity and almost 33.5% of males were inactive. Students who chose movie actors as role models (71.1%) and those who found their role models' information reliable (83.2%) were more likely to participate in physical activity. Of the 247 students who were knowledgeable about healthy living, a significant majority of 70.4% engaged in physical activity while 76% of overweight/obese students reported academic stress as a barrier. Conclusions: Study shows majority engage in physical activity regardless of gender. Over 70% have family history of obesity and healthy lifestyle knowledge. Stress reduces physical activity and students with movie/sports/fitness role models have better activity levels. Reasons for not engaging included dislike for sports and lack of time.

RESUMEN

Comprehensive analyses of the association between a family history of lung cancer and lung cancer risk are limited, especially in the Korean population. We used baseline data from the Korean Genome and Epidemiology Study, conducted between 2001 and 2013. This study enrolled 198,980 individuals. Lung cancer diagnoses and family histories were determined using questionnaires. Multivariable logistic regression analysis was performed to evaluate the effect of family history on the risk of lung cancer. Of 198,980 individuals, 6296 (3.2%) and 140 (0.1%) had a family history of lung cancer and lung cancer, respectively. Individuals with a family history of lung cancer in first-degree relatives (FDRs) had a higher risk of lung cancer development than those without (adjusted odds ratio [aOR] = 2.28, 95% confidence interval [CI] = 1.11-4.66). This was more pronounced in young individuals (<60 years) who had affected relatives diagnosed with lung cancer before the age of 60 years (aOR = 3.77, 95% CI = 1.19-11.88). In subgroup analyses, this association was more evident in women, never smokers, and young individuals. A family history of lung cancer, especially in FDRs, is a significant risk factor for lung cancer development in Korea.

RESUMEN

OBJECTIVE: Family health history (FHx) is an important tool in assessing one's risk towards specific health conditions. However, user experience of FHx collection tools is rarely studied. ItRunsInMyFamily.com (ItRuns) was developed to assess FHx and hereditary cancer risk. This study reports a quantitative user experience analysis of ItRuns. METHODS: We conducted a public health campaign in November 2019 to promote FHx collection using ItRuns. We used software telemetry to quantify abandonment and time spent on ItRuns to identify user behaviors and potential areas of improvement. RESULTS: Of 11,065 users who started the ItRuns assessment, 4305 (38.91%) reached the final step to receive recommendations about hereditary cancer risk. Highest abandonment rates were during Introduction (32.82%), Invite Friends (29.03%), and Family Cancer History (12.03%) subflows. Median time to complete the assessment was 636 s. Users spent the highest median time on Proband Cancer History (124.00 s) and Family Cancer History (119.00 s) subflows. Search list questions took the longest to complete (median 19.50 s), followed by free text email input (15.00 s). CONCLUSION: Knowledge of objective user behaviors at a large scale and factors impacting optimal user experience will help enhance the ItRuns workflow and improve future FHx collection.

Asunto(s)

RESUMEN

Background: Health disparities are costly and preventable differences in disease progression that disproportionately affect minority communities such as African Americans. Practices to reduce health disparities can be rooted in prevention, particularly through screening tools. Family Health History tools are preventative screening mechanisms meant to explore family history to better understand how an individual's health can potentially be predicted or impacted. These tools are underutilized in the African American community. Contributions to this underutilization include a lack of cultural tailoring in the tools, a lack of health literacy in community members, and a lack of effective health communication. The Family Health History Study will create a culturally appropriate Family Health History toolkit to increase family health history utilization and ultimately decrease health disparities. Methods: The proposed sample will be composed of 195 African American adults ages 18 + who live in Genesee County, Michigan. The study consists of two phases: the development phase and the randomized pilot study phase. The goal of the development phase (n = 95) is to explore how Family Health History toolkits can be modified to better serve the African American community using a community based participatory research approach and to create a culturally tailored family health history toolkit. In the pilot study phase, 100 participants will be randomized to the culturally tailored toolkit or the current standard Family Health History toolkit. Outcomes will include feasibility and acceptability of the intervention. Discussion: This study will result in a culturally appropriate Family Health History tool that is co-developed with community members that can be utilized by African American adults to better understand their family health histories. Trial registration: Clinicaltrials.gov: NCT05358964 Date: May 5, 2022.

RESUMEN

Genetic testing is key in modern healthcare, particularly for monogenic disorders such as familial hypercholesterolemia. This Tohoku Medical Megabank Project study explored the impact of first-degree relatives' dyslipidemia history on individual responses to familial hypercholesterolemia genomic results. Involving 214 participants and using Japan's 3.5KJPN genome reference panel, the study assessed preferences and intentions regarding familial hypercholesterolemia genetic testing results. The data revealed a significant inclination among participants with a family history of dyslipidemia to share their genetic test results, with more than 80% of participants intending to share positive results with their partners and children and 98.1% acknowledging the usefulness of positive results for personal health management. The study underscores the importance of family health history in genetic-testing perceptions, highlighting the need for family-centered approaches in genetic counseling and healthcare. Notable study limitations include the regional scope and reliance on questionnaire data. The study results emphasize the association between family health history and genetic-testing attitudes and decisions.

Asunto(s)

RESUMEN

Family health history (FHH) is an evidence-based genomics tool used in cancer prevention and education. Chinese Americans, the largest Asian American group, face unique barriers in FHH collection and communication. This study aimed to evaluate the efficacy of culturally and linguistically appropriate community health worker (CHW)-delivered FHH-based breast cancer (BC) education and services to Chinese Americans. A total of 1129 Chinese Americans received FHH-based BC education and service delivered by our trained Chinese American CHWs. Participants responded to evaluation surveys before, immediately after, and 3 months after the education and service. Participating Chinese Americans showed significant increases in rates of collecting FHH of BC, discussing FHH of BC with family members, informing their primary care physicians of their FHH of BC, and discussing their FHH of BC with their primary care physicians at 3 months post-education and service compared to the baseline data (all Ps < 0.01). Attitudes, intention, and self-efficacy related to FHH of BC communication and collection and FHH of BC knowledge were improved both immediately after and 3 months after the delivery of the education and services (all Ps < 0.01). Within 3 months, ~ 14.3% of participants who had a high risk of BC based on FHH reported visiting geneticists for genetic evaluation. Our Chinese American CHW-delivered FHH-based BC education and services showed initial success in increasing knowledge, collection and communication of BC-related FHH, and genetic service utilization among Chinese American participants. This study can serve as a starting point for conducting more robust studies, such as randomized controlled trials, in the future.

Asunto(s)

RESUMEN

BACKGROUND: Primary care clinicians screen for breast cancer risk factors and assess the risk level of their patients. Women at high risk for breast cancer (eg, 5-year risk of at least 3% or lifetime risk of ≥20%) are eligible for enhanced screening and/or chemoprophylaxis. However, many clinicians do not identify women at high risk and offer appropriate referrals, screening, or chemoprophylaxis. METHODS: We reviewed a sample of 200 charts of women ages 35 to 50 years old with a family history of breast cancer. We identified factors that contribute to their risk for breast cancer and used the Tyrer-Cuzick Risk Assessment Calculator to determine their personal lifetime risk. We then assessed whether these patients received counseling for chemoprophylaxis, referrals, or screening. We also looked for correlations between combinations of risk factors and increased lifetime risk. RESULTS: Out of 200 charts reviewed, 71 women were identified as high risk for breast cancer (lifetime risk of ≥20%). Of those 71 women, just 17 were referred to a high-risk clinic for enhanced screening and/or chemoprophylaxis. Three risk factors, mammographic breast density of category C or D, first degree relatives with breast cancer, and age first given birth if after 30 years old had a significant impact on lifetime risk for breast cancer. DISCUSSION: Primary care clinicians can use these independent risk factors as cues to pursue a more formal calculation of a woman's lifetime risk for breast cancer and make appropriate referrals for enhanced screening and chemoprophylaxis counseling if indicated.

Asunto(s)

RESUMEN

BACKGROUND: Identifying children at high risk of developing asthma can facilitate prevention and early management strategies. We developed a prediction model of children's asthma risk using objectively collected population-based children and parental histories of comorbidities. METHODS: We conducted a retrospective population-based cohort study using administrative data from Manitoba, Canada, and included children born from 1974 to 2000 with linkages to ≥1 parent. We identified asthma and prior comorbid condition diagnoses from hospital and outpatient records. We used two machine-learning models: least absolute shrinkage and selection operator (LASSO) logistic regression (LR) and random forest (RF) to identify important predictors. The predictors in the base model included children's demographics, allergic conditions, respiratory infections, and parental asthma. Subsequent models included additional multiple comorbidities for children and parents. RESULTS: The cohort included 195,666 children: 51.3% were males and 17.7% had asthma diagnosis. The base LR model achieved a low predictive performance with sensitivity of 0.47, 95% confidence interval (0.45-0.48), and specificity of 0.67 (0.66-0.67) using a predicted probability threshold of 0.20. Sensitivity significantly improved when children's comorbidities were included using LASSO LR: 0.71 (0.69-0.72). Predictive performance further improved by including parental comorbidities (sensitivity = 0.72 [0.70-0.73], specificity = 0.69 [0.69-0.70]). We observed similar results for the RF models. Children's menstrual disorders and mood and anxiety disorders, parental lipid metabolism disorders and asthma were among the most important variables that predicted asthma risk. CONCLUSION: Including children and parental comorbidities to children's asthma prediction models improves their accuracy.

Asunto(s)

RESUMEN

Objective: Family health history (FHx) is an important tool in assessing one's risk towards specific health conditions. However, user experience of FHx collection tools is rarely studied. ItRunsInMyFamily.com (ItRuns) was developed to assess FHx and hereditary cancer risk. This study reports a quantitative user experience analysis of ItRuns. Methods: We conducted a public health campaign in November 2019 to promote FHx collection using ItRuns. We used software telemetry to quantify abandonment and time spent on ItRuns to identify user behaviors and potential areas of improvement. Results: Of 11065 users who started the ItRuns assessment, 4305 (38.91%) reached the final step to receive recommendations about hereditary cancer risk. Highest abandonment rates were during Introduction (32.82%), Invite Friends (29.03%), and Family Cancer History (12.03%) subflows. Median time to complete the assessment was 636 seconds. Users spent the highest median time on Proband Cancer History (124.00 seconds) and Family Cancer History (119.00 seconds) subflows. Search list questions took the longest to complete (median 19.50 seconds), followed by free text email input (15.00 seconds). Conclusion: Knowledge of objective user behaviors at a large scale and factors impacting optimal user experience will help enhance the ItRuns workflow and improve future FHx collection.

RESUMEN

The collection of family health history (FHH) is an essential component of clinical practice and an important piece of data for patient risk assessment. However, family history data have generally been limited to diseases and have not included medication history. Family history was a key component of early pharmacogenetic research, confirming the role of genes in drug response. With the substantial number of known pharmacogenes, many affecting response to commonly prescribed medications, and the availability of clinical pharmacogenetic (PGx) tests and guidelines for interpretation, the collection of family medication history can inform testing decisions. This paper explores the roots of family-based pharmacogenetic studies to confirm the role of genes in these complex phenotypes and the benefits and challenges of collecting family medication history as part of family health history intake.

RESUMEN

CONTENT: This study examines the potential utility of genetic testing as a supplement to family health history to screen for increased risk of inherited disease. Medical conditions are often misreported or misunderstood, especially those related to different forms of cardiac disease (arrhythmias vs. structural heart disease vs. coronary artery disease), female organ cancers (uterine vs. ovarian vs. cervical), and type of cancer (differentiating primary cancer from metastases to other organs). While these nuances appear subtle, they can dramatically alter medical management. For example, different types of cardiac failure (structural, arrhythmia, and coronary artery disease) have inherited forms that are managed with vastly different approaches. METHODS: Using a dataset of over 6,200 individuals who underwent genetic screening, we compared the ability of genetic testing and traditional family health history to identify increased risk of inherited disease. A further, in-depth qualitative study of individuals for whom risk identified through each method was discordant, explored whether this discordance could be addressed through changes in family health history intake. FINDINGS: Of 90 individuals for whom genetic testing indicated significant increased risk for inherited disease, two-thirds (66%) had no corroborating family health history. Specifically, we identify cardiomyopathy, arrhythmia, and malignant hyperthermia as conditions for which discordance between genetic testing and traditional family health history was greatest, and familial hypercholesterolaemia, Lynch syndrome, and hereditary breast and ovarian cancer as conditions for which greater concordance existed. CONCLUSION: We conclude that genetic testing offers utility as a supplement to traditional family health history intake over certain conditions.

Asunto(s)

RESUMEN

Family health history (FHH) is an essential foundation for personalized disease prevention. As the incidence of early-onset chronic diseases is increasing among college students, it is important to provide them with the education required to learn about their FHH. This study aimed to assess college students' interest in receiving FHH education, preferred topics, and desired learning methods. We invited college students to complete an online survey from a large research-intensive university. A total of 2276 college students completed the survey. Nearly half of the participants self-identified as non-Hispanic white (45.5%). Slightly more than half of the sample (53.9%) were not interested in receiving FHH education mainly due to low prioritization. Among those who expressed interest in obtaining FHH education, the three most desired learning topics were the ability to interpret FHH information (76.1%), the application of FHH in disease prevention (72.0%), and FHH information collection strategies (63.6%). Computer-based learning (51.1%) was the most preferred educational method. Moreover, females, older individuals, those who have FHH in first-degree relatives, and participants who were members of racial and ethnic groups showed greater interests in receiving FHH education (ps < 0.05). Strategies to promote college students' awareness, collection, and use of FHH are needed.

Asunto(s)

RESUMEN

PROBLEM ADDRESSED: To better understand the factors associated with family cancer history (FCH) information and cancer information seeking, we model the process an individual undergoes when assessing whether to gather FCH and seek cancer information and compare models by sociodemographics and family history of cancer. We used cross-sectional data from the Health Information National Trends Survey (HINTS 5, Cycle 2) and variables (e.g., emotion and self-efficacy) associated with the Theory of Motivated Information Management to assess the process of FCH gathering and information seeking. We completed path analysis to assess the process of FCH gathering and stratified path models. RESULTS: Those who felt they could lower their chances of getting cancer (emotion) were more confident in their ability to complete FCH on a medical form (self-efficacy; B = 0.11, p < .0001) and more likely to have discussed FCH with family members (B = 0.07, p < .0001). Those who were more confident in their ability to complete a summary of their family history on a medical form were more likely to have discussed FCH with family members (B = 0.34, p < .0001) and seek other health information (B = 0.24, p < .0001). Stratified models showed differences in this process by age, race/ethnicity, and family history of cancer. IMPLICATIONS FOR PUBLIC HEALTH RESEARCH AND PRACTICE: Tailoring outreach and education strategies to address differences in perceived ability to lower chances of getting cancer (emotion) and confidence in the ability to complete FCH (self-efficacy) could help encourage less engaged individuals to learn about their FCH and gather cancer information.

Asunto(s)

RESUMEN

BACKGROUND: Obtaining comprehensive family health history (FHH) to inform colorectal cancer (CRC) risk management in primary care settings is challenging. OBJECTIVE: To examine the effectiveness of a patient-facing FHH platform to identify and manage patients at increased CRC risk. DESIGN: Two-site, two-arm, cluster-randomized, implementation-effectiveness trial with primary care providers (PCPs) randomized to immediate intervention versus wait-list control. PARTICIPANTS: PCPs treating patients at least one half-day per week; patients aged 40-64 with no medical conditions that increased CRC risk. INTERVENTIONS: Immediate-arm patients entered their FHH into a web-based platform that provided risk assessment and guideline-driven decision support; wait-list control patients did so 12 months later. MAIN MEASURES: McNemar's test examined differences between the platform and electronic medical record (EMR) in rates of increased risk documentation. General estimating equations using logistic regression models compared arms in risk-concordant provider actions and patient screening test completion. Referral for genetic consultation was analyzed descriptively. KEY RESULTS: Seventeen PCPs were randomized to each arm. Patients (n = 252 immediate, n = 253 control) averaged 51.4 (SD = 7.2) years, with 83% assigned male at birth, 58% White persons, and 33% Black persons. The percentage of patients identified as increased risk for CRC was greater with the platform (9.9%) versus EMR (5.2%), difference = 4.8% (95% CI: 2.6%, 6.9%), p < .0001. There was no difference in PCP risk-concordant action [odds ratio (OR) = 0.7, 95% CI (0.4, 1.2; p = 0.16)]. Among 177 patients with a risk-concordant screening test ordered, there was no difference in test completion, OR = 0.8 [0.5,1.3]; p = 0.36. Of 50 patients identified by the platform as increased risk, 78.6% immediate and 68.2% control patients received a recommendation for genetic consultation, of which only one in each arm had a referral placed. CONCLUSIONS: FHH tools could accurately assess and document the clinical needs of patients at increased risk for CRC. Barriers to acting on those recommendations warrant further exploration. TRIAL REGISTRATION NUMBER: ClinicalTrials.gov NCT02247336 https://clinicaltrials.gov/ct2/show/NCT02247336.

Asunto(s)

RESUMEN

BACKGROUND: Systematically assessing disease risk can improve population health by identifying those eligible for enhanced prevention/screening strategies. This study aims to determine the clinical impact of a systematic risk assessment in diverse primary care populations. METHODS: Hybrid implementation-effectiveness trial of a family health history-based health risk assessment (HRA) tied to risk-based guideline recommendations enrolling from 2014-2017 with 12 months of post-intervention survey data and 24 months of electronic medical record (EMR) data capture. SETTING: 19 primary care clinics at four geographically and culturally diverse U.S. healthcare systems. PARTICIPANTS: any English or Spanish-speaking adult with an upcoming appointment at an enrolling clinic. METHODS: A personal and family health history based HRA with integrated guideline-based clinical decision support (CDS) was completed by each participant prior to their appointment. Risk reports were provided to patients and providers to discuss at their clinical encounter. OUTCOMES: provider and patient discussion and provider uptake (i.e. ordering) and patient uptake (i.e. recommendation completion) of CDS recommendations. MEASURES: patient and provider surveys and EMR data. RESULTS: One thousand eight hundred twenty nine participants (mean age 56.2 [SD13.9], 69.6% female) completed the HRA and had EMR data available for analysis. 762 (41.6%) received a recommendation (29.7% for genetic counseling (GC); 15.2% for enhanced breast/colon cancer screening). Those with recommendations frequently discussed disease risk with their provider (8.7%-38.2% varied by recommendation, p-values ≤ 0.004). In the GC subgroup, provider discussions increased referrals to counseling (44.4% with vs. 5.9% without, P < 0.001). Recommendation uptake was highest for colon cancer screening (provider = 67.9%; patient = 86.8%) and lowest for breast cancer chemoprevention (0%). CONCLUSIONS: Systematic health risk assessment revealed that almost half the population were at increased disease risk based on guidelines. Risk identification resulted in shared discussions between participants and providers but variable clinical action uptake depending upon the recommendation. Understanding the barriers and facilitators to uptake by both patients and providers will be essential for optimizing HRA tools and achieving their promise of improving population health. TRIAL REGISTRATION: Clinicaltrials.gov number NCT01956773 , registered 10/8/2013.

Asunto(s)

RESUMEN

OBJECTIVE: To measure the effect of maternal family history of hypertension on preterm birth (PTB) and to identify factors that modified this association. METHODS: A case-control study was nested in a prospective cohort of the entire pregnant population in Wuhan, China, from 2011 to 2013. Home-visit interviews were scheduled for all PTBs and their controls, to collect extensive information on maternal exposures to behavioral, environmental, and intergenerational risk factors of PTB. The effects of maternal family history of hypertension on PTB were measured by logistic regression analyses, controlling for potential confounders. Potential effect modifiers were examined using stratified analyses. RESULTS: There were 2393 PTBs and 4263 full-term births out of all eligible births. A positive association was observed between maternal family history of hypertension and PTB, after adjusted for potential confounders (adjusted odds ratio: 1.17 [1.03, 1.33]). A higher effect was observed when mothers were exposed to certain noise during pregnancy (adjusted odds ratio: 1.37 [1.14, 1.65]) and/or when they did not take multivitamins during pregnancy (adjusted odds ratio: 1.46 [1.20, 1.78]), whereas, this association was weaker and no longer significant when mothers took multivitamins during pregnancy (adjusted odds ratio: 1.00 [0.84, 1.19]) and/or when they were not exposed to certain noise during pregnancy (adjusted odds ratio: 1.01 [0.85, 1.12]). The modification effect from maternal multivitamin intake was significant on both spontaneous and medically indicated PTBs, and the modification effect from maternal exposure to certain noise was only significant on spontaneous PTB. CONCLUSIONS: Increased PTB risk was observed for pregnant women with a family history of hypertension in Wuhan, China. This effect was stronger when pregnant women did not take multivitamin and/or exposed to certain noise during pregnancy, than those who took multivitamin and/or unexposed to certain noise.

Asunto(s)

RESUMEN

Family health history (FHH) is a data type serving risk assessment, diagnosis, research, and preventive health. Despite technological leaps in genomic variant detection, FHH remains the most accessible, least expensive, and most practical assessment tool for assessing risks attributable to genetic inheritance. The purpose of this manuscript is to outline a process to assist primary care professionals in choosing FHH digital tools for patient care based on the new ISO/TS 82304-2 Technical Specification (TS), which is a recently developed method to determine eHealth app quality. With a focus on eHealth in primary care, we applied the quality label concept to FHH, and how a primary care physician can quickly review the quality and reliability of an FHH app. Based on our review of the ISO TS's 81 questions, we compiled a list of 25 questions that are recommended to be more succinct as an initial review. We call this process the FHH Quick App Review. Our 'informative-only' 25 questions do not produce a quality score, but a guide to complete an initial review of FHH apps. Most of the questions are straight from the ISO TS, some are modified or de novo. We believe the 25 questions are not only relevant to FHH app reviews but could also serve to aid app development and clinical implementation.

Asunto(s)

RESUMEN

Background: An alternative to population-based genetic testing, automated cascade genetic testing facilitated by sharing of family health history, has been conceptualized as a more efficient and cost-effective approach to identify hereditary genetic conditions. However, existing software and applications programming interfaces (API) for the practical implementation of this approach in health care settings have not been described. Methods: We reviewed API available for facilitating cascade genetic testing in electronic health records (EHRs). We emphasize any information regarding informed consent as provided for each tool. Using semi-structured key informant interviews, we investigated uptake of and barriers to integrating automated family cascade genetic testing into the EHR. Results: We summarized the functionalities of six tools related to utilizing family health history to facilitate cascade genetic testing. No tools were explicitly capable of facilitating family cascade genetic testing, but few enterprise EHRs supported family health history linkage. We conducted five key informant interviews with four main considerations that emerged including: 1) incentives for interoperability, 2) HIPAA and regulations, 3) mobile-app and alternatives to EHR deployment, 4) fundamental changes to conceptualizing EHRs. Discussion: Despite the capabilities of existing technology, limited bioinformatic support has been developed to automate processes needed for family cascade genetic testing and the main barriers for implementation are nontechnical, including an understanding of regulations, consent, and workflow. As the trade-off between cost and efficiency for population-based and family cascade genetic testing shifts, the additional tools necessary for their implementation should be considered.

RESUMEN

BACKGROUND: Family health history has been recognized as an essential factor for cancer risk assessment and is an integral part of many cancer screening guidelines, including genetic testing for personalized clinical management strategies. However, manually identifying eligible candidates for genetic testing is labor intensive. OBJECTIVE: The aim of this study was to develop a natural language processing (NLP) pipeline and assess its contribution to identifying patients who meet genetic testing criteria for hereditary cancers based on family health history data in the electronic health record (EHR). We compared an algorithm that uses structured data alone with structured data augmented using NLP. METHODS: Algorithms were developed based on the National Comprehensive Cancer Network (NCCN) guidelines for genetic testing for hereditary breast, ovarian, pancreatic, and colorectal cancers. The NLP-augmented algorithm uses both structured family health history data and the associated unstructured free-text comments. The algorithms were compared with a reference standard of 100 patients with a family health history in the EHR. RESULTS: Regarding identifying the reference standard patients meeting the NCCN criteria, the NLP-augmented algorithm compared with the structured data algorithm yielded a significantly higher recall of 0.95 (95% CI 0.9-0.99) versus 0.29 (95% CI 0.19-0.40) and a precision of 0.99 (95% CI 0.96-1.00) versus 0.81 (95% CI 0.65-0.95). On the whole data set, the NLP-augmented algorithm extracted 33.6% more entities, resulting in 53.8% more patients meeting the NCCN criteria. CONCLUSIONS: Compared with the structured data algorithm, the NLP-augmented algorithm based on both structured and unstructured family health history data in the EHR increased the number of patients identified as meeting the NCCN criteria for genetic testing for hereditary breast or ovarian and colorectal cancers.