A pre-visit questionnaire that patients complete from home, paired with automatic upload to the electronic health record and a same-day prompt for the family physician, increased new family history documentation roughly 94-fold compared with usual care, according to a matched-pair effectiveness-implementation trial published in the September/October 2025 issue of the Annals of Family Medicine. The intervention also triggered tangible downstream actions, including screening recommendations and specialist referrals.
Family history has long occupied an awkward space in primary care. Clinicians know it matters – it is, after all, what the authors, citing Ginsburg and colleagues, describe as the “most useful tool for risk assessment for common chronic diseases” – yet it remains stubbornly under-collected, inconsistently updated, and often buried somewhere unsearchable in the electronic health record (EHR). A Canadian team led by Dr June C. Carroll at Mount Sinai Hospital’s Granovsky Gluskin Family Medicine Centre has now shown that a relatively low-tech, well-engineered workflow can change that picture rather dramatically.
What the intervention actually did
The study, conducted across three University of Toronto-affiliated practices between September 2021 and June 2022, tested a multifaceted bundle rather than a single tool. Patients aged 30-69 received an emailed invitation one to two weeks before a scheduled appointment, along with a consent form and the Family History Screening Questionnaire (FHSQ) – an instrument originally validated in Australian primary care and adapted here to screen for breast, ovarian, colorectal, and prostate cancer, melanoma, ischaemic heart disease, and type 2 diabetes. The team added items on prior genetic testing and self-reported genetic disorders.
Completed questionnaires uploaded automatically into the PS Suite EHR via the Ocean e-mailing platform, and on the day of the visit the family physician received an EHR prompt flagging that new family history information was available for review. Physicians had also attended a brief webinar on the clinical utility of family history, and patient-facing education was delivered via waiting room televisions, clinic websites, and email links to a YouTube explainer. Control patients and physicians, drawn from the UTOPIAN practice-based research network and matched on age, sex, and EHR type, received no intervention.
In essence, this was an exercise in implementation science as much as anything else, drawing explicitly on the Consolidated Framework for Implementation Research and the literature on what an ideal family history strategy looks like – patient-administered, EHR-integrated, update-able, low-friction.
A 94-fold difference, and what it means in the clinic
The headline number is striking. Within 30 days of the clinic visit, new family history was documented in the EHR for 93 of 576 (16.1%) intervention patients, compared with just 5 of 2,203 (0.2%) control patients – an adjusted odds ratio of 94.2 (95% CI, 36.8–240.8; P<.001). For cancer family history specifically, 7.8% of intervention patients had new documentation, against 0.1% of controls. Heart disease family history showed a similar pattern (3.3% vs <0.01%).
For clinicians, the implications run deeper than tidier charts. Of patients who reported discussing their family history at the visit (n=296), 24.5% said a screening test had been recommended, 7.8% reported lifestyle advice, 7.5% were referred to a non-genetics specialist, and 2.4% were referred for genetics assessment. The authors note that this genetics referral rate, although superficially modest, aligns reasonably well with the estimated 1.33% population prevalence of pathogenic or likely pathogenic variants in the four conditions queried – hereditary breast and ovarian cancer syndromes, Lynch syndrome, and familial hypercholesterolaemia.
In other words, the questionnaire is not just generating documentation for its own sake – it appears to be identifying patients who genuinely warrant further preventive action, and prompting that action within the visit itself.
Why this worked when previous attempts have stumbled
The literature on family history collection tools is, to put it generously, mixed.
Murray and colleagues, whose 2013 study the authors discuss at length, found that around half of patient-reported family history information collected through portals was never reviewed by clinicians – often because the data sat in a parallel system rather than flowing into the EHR.
Carroll and colleagues frame their result as a product of three converging design choices. As they write in the discussion, the effects were “likely due to time efficiency (patient completion at home, EHR integration); addressing knowledge gaps (easily accessible education, evidence-based clinical decision support); and system integration (incorporation into usual care flow with physician reminders).”
The same-day EHR prompt is probably doing significant work here. Family history that arrives at the moment of clinical decision-making, in the system the physician is already using, is family history that gets acted on. Patient uptake also outstripped comparable studies: 36.4% of invited patients consented, versus 9.8% to 14.6% in Murray’s intervention arms.Famil
Discordant reporting between patients and physicians
One of the more interesting wrinkles in the data concerns who reported what. Patients were considerably more likely than their family physicians to say that new actions had been taken on the basis of family history. Patients reported screening recommendations at 24.5% of visits; physicians reported changes in screening or management at just 3.4%. Referrals to non-genetics specialists were noted by 7.5% of patients but only 0.8% of physicians.
The authors offer a sensible interpretation: physicians may have been recommending screening based on family history they already knew about, while patients experienced the recommendation as flowing from the new conversation. Lifestyle counselling, similarly, may not register to physicians as a “change in management.” Either way, the gap is a useful reminder that patient-reported and clinician-reported outcomes capture genuinely different things.
Limitations and equity considerations
The authors are upfront about the study’s constraints. Baseline family history documentation was already high – 81.6% of intervention patients and 86.4% of controls had some family history on file before the trial began – which may have produced a ceiling effect. Only about a third of invited patients consented, and those who did were probably enriched for positive family history. Conducting the trial during the COVID-19 pandemic, when physicians had limited bandwidth for new workflows, may have suppressed the true effect size.
Equity is the larger concern. The intervention cohort was 94% urban, 73% White, and 84% university-educated, with 59% reporting household incomes above $100,000. Participation required English literacy, internet access, computer skills, and comfort sharing health information online. Without thoughtful adaptation, a tool like this risks widening rather than narrowing the gap between patients who benefit from preventive genomics and those who do not.
What comes next
The team is planning to revise the FHSQ in response to patient feedback – clearer definitions of “family,” expanded condition lists to include stroke, dementia, and mental health, and broader cultural and linguistic adaptation – and to roll it out as a quality improvement project. Future iterations will assess whether ordered tests actually match patients’ risk levels and whether patients follow through on recommendations.
The authors conclude: “This strategy showed significant improvement in collection and documentation of FH. Patients welcomed the opportunity to provide FH information before appointments. Factors contributing to the intervention’s success included being completed by the patient and seamless EHR integration with a reminder.”
Reference:
Carroll, J. C., Greiver, M., Kukan, S., et al. (2025). An innovative strategy for collecting family health history: An effectiveness-implementation trial in primary care clinics. Annals of Family Medicine, 23(5), 399–406. https://doi.org/10.1370/afm.240472
