An electronic family health history tool to identify and manage patients at increased risk for colorectal cancer: protocol for a randomized controlled trial

A Cluster Randomized Trial of a Family Health History Platform to Identify and Manage Patients at Increased Risk for Colorectal Cancer

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.

Expanding Family Health History to Include Family Medication History

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.

Family history tools for primary care: A systematic review

Background

Many medical family history (FH) tools are available for various settings. Although FH tools can be a powerful health screening tool in primary care (PC), they are currently underused.

Objectives

This review explores the FH tools currently available for PC and evaluates their clinical performance.

Methods

Five databases were systematically searched until May 2021. Identified tools were evaluated on the following criteria: time-to-complete, integration with electronic health record (EMR) systems, patient administration, risk-assessment ability, evidence-based management recommendations, analytical and clinical validity and clinical utility.

Results

We identified 26 PC FH tools. Analytical and clinical validity was poorly reported and agreement between FH and gold standard was commonly inadequately reported and assessed. Sensitivity was acceptable; specificity was found in half of the reviewed tools to be poor. Most reviewed tools showed a capacity to successfully identify individuals with increased risk of disease (6.2–84.6% of high and/or moderate or increased risk individuals).

Conclusion

Despite the potential of FH tools to improve risk stratification of patients in PC, clinical performance of current tools remains limited as well as their integration in EMR systems. Twenty-one FH tools are designed to be self-administered by patients.

Identification and Impact Analysis of Family History of Psychiatric Disorder in Mood Disorder Patients With Pretrained Language Model

Mood disorders are ubiquitous mental disorders with familial aggregation. Extracting family history of psychiatric disorders from large electronic hospitalization records is helpful for further study of onset characteristics among patients with a mood disorder. This study uses an observational clinical data set of in-patients of Nanjing Brain Hospital, affiliated with Nanjing Medical University, from the past 10 years. This paper proposes a pretrained language model: Bidirectional Encoder Representations from Transformers (BERT)-Convolutional Neural Network (CNN). We first project the electronic hospitalization records into a low-dimensional dense matrix via the pretrained Chinese BERT model, then feed the dense matrix into the stacked CNN layer to capture high-level features of texts; finally, we use the fully connected layer to extract family history based on high-level features. The accuracy of our BERT-CNN model was 97.12 ± 0.37% in the real-world data set from Nanjing Brain Hospital. We further studied the correlation between mood disorders and family history of psychiatric disorder.

Digital health interventions for the management of mental health in people with chronic diseases: a rapid review

OBJECTIVE

Determine the effectiveness of digital mental health interventions for individuals with a concomitant chronic disease.

DESIGN

We conducted a rapid review of systematic reviews. Two reviewers independently conducted study selection and risk of bias evaluation. A standardised extraction form was used. Data are reported narratively.

INTERVENTIONS

We included systematic reviews of digital health interventions aiming to prevent, detect or manage mental health problems in individuals with a pre-existing chronic disease, including chronic mental health illnesses, published in 2010 or after.

MAIN OUTCOME MEASURE

Reports on mental health outcomes (eg, anxiety symptoms and depression symptoms).

RESULTS

We included 35 reviews, totalling 702 primary studies with a total sample of 50 692 participants. We structured the results in four population clusters: (1) chronic diseases, (2) cancer, (3) mental health and (4) children and youth. For populations presenting a chronic disease or cancer, health provider directed digital interventions (eg, web-based consultation, internet cognitive-behavioural therapy) are effective and safe. Further analyses are required in order to provide stronger recommendations regarding relevance for specific population (such as children and youth). Web-based interventions and email were the modes of administration that had the most reports of improvement. Virtual reality, smartphone applications and patient portal had limited reports of improvement.

CONCLUSIONS

Digital technologies could be used to prevent and manage mental health problems in people living with chronic conditions, with consideration for the age group and type of technology used.