nature.com

Evaluation of the digital genetic assistant in technology assisted genetic counseling for genetic carrier screening

Abstract

Rapid growth in genetic testing usage resulted in declining availability of genetic counselors (GCs) per ordered tests, prolonging the waiting times for face-to-face (F2F) counseling. We evaluated the digital genetic assistant (DGA) for reproductive genetic carrier screening (RGCS) in a real-life clinical setting using a “couple-based” paradigm. The platform provides digital patient intake and automated counseling for low-risk individuals, as well as GC-facing tools that reduce administrative burden in patient-related activities. Among 225 couples undergoing RGCS during the study period, 4% had high-risk results requiring F2F counseling and an additional 4% of low-risk couples requested F2F counseling, suggesting that DGA use reduced GC-participant F2F interactions by 69.3%. GC evaluations revealed that the DGA triaging algorithm was accurate and surveys demonstrated high degrees of user comprehension and satisfaction. These results highlight the utility of digital platforms for patient intake and delivery of low-risk results in settings with limited genetic counseling resources.

Introduction

Genetic testing is becoming more prevalent in mainstream medical practice, providing the promise of early diagnosis and disease prevention in many medical areas, such as prenatal diagnosis, pediatrics, neurology, and oncology1,2,3,4. This trend is fueled by the increasing availability of genetic tests, a decrease in cost of DNA sequencing methods, as well as public awareness[5](https://www.nature.com/articles/s41746-025-01573-7#ref-CR5 "Wetterstrand, K. A. DNA sequencing costs: data. National Human Genome Research Institute

https://www.genome.gov/about-genomics/fact-sheets/DNA-Sequencing-Costs-Data

(2023)."). However, genetic and genomic testing is complex and requires extensive training to appropriately provide guidance and recommendations. Genetic counselors (GCs) are domain experts in this area of medicine, but workforce shortages have resulted in extreme bottlenecks and delayed care, warranting exploration of supplemental solutions that can bridge this gap in patient care^{[6](https://www.nature.com/articles/s41746-025-01573-7#ref-CR6 "Abacan, M. A. et al. The global state of the genetic counseling profession. Eur. J. Hum. Genet. 27, 183–197 (2019).")}.

The purpose of carrier screening for genetic conditions before or early during pregnancy is to identify individuals and/or couples who carry variants associated with genetic diseases, and thus are at risk of transmitting a heritable disorder to their offspring. Detection allows appropriate genetic counseling and reproductive choices, including prenatal diagnosis and preimplantation genetic testing. Reproductive genetic carrier screening (RGCS) in Israel has been widespread since the 1990s, and patients are encouraged to undergo screening by their obstetricians. Carrier screening is routinely performed for all pregnant women through their health maintenance organizations, and funding is provided for a basic set of conditions by the Israeli Ministry of Health (MOH). Testing for an extended set of conditions is commercially available by several providers, including genetics institutes within medical centers.

The routine carrier screening workflow in Israel (Fig. 1), where screening has been practiced since the early 1990s, was developed when there were only a few tested conditions. Testing was then performed in a stepwise manner. First, the female partner was tested for a limited number of conditions; if she was found to be a carrier of an autosomal recessive disease, she received genetic counseling regarding the disease, its inheritance and the couples’ reproductive risk, either by a face-to-face counseling session (F2F) or by telephone. Her reproductive partner was then referred for testing of the same gene. If both partners are found to be carriers of the same AR disease, or the woman was found to be a carrier of an X-linked disease, they receive F2F genetic counseling to discuss diagnosis and reproductive options. This process is long and time-consuming, especially when facing the growing availability of genetic tests and the shortage of GC services.

Fig. 1: Genetic counseling workflows for reproductive genetic carrier screening (RGCS).

figure 1

The top workflow details the steps in traditional, or routine, “woman-first” counseling and testing paradigm. In Israel, it is recommended to refer the female partner for carrier screening prior to pregnancy due to the test timelines (to allow high-risk couples to make informed decisions regarding the pregnancy). The woman receives a basic counseling session, including a short explanation about the test, focused medical intake and relevant family history, signs an informed consent, and provides a blood sample for testing. In case of a negative test result, a lab report is sent to the couple by the genetic counselor (GC). If the woman is found to carry an autosomal recessive (AR) disease-causing variant, the couple receives a second short genetic counseling session where the male partner is referred for testing. If the couple is found to be “high-risk” (i.e., both partners are variant carriers in the same AR gene, or the woman is a carrier of an X-linked variant), they are referred for a third genetic counseling session to receive a full explanation regarding the implications and their reproductive options. The bottom workflow shows the “couple-based” testing paradigm utilizing the digital genetic assistant (DGA). This workflow allows the DGA to provide information regarding several steps of the “traditional” process, which spares GC time (e.g. initial medical intake, test explanation and consenting, and stepwise testing when a female is found to be an AR variant carrier), thus streamlining the RGCS flow. This allows to forfeit all pretest patient-GC encounters for all couples, and to reduce the personal encounters for the low-risk couples. Steps requiring participant and/or GC interactions are highlighted in different colors (blue- patient; pink- GC; purple- patient and GC).

Full size image

The digital genetic assistant (DGA) is a medically supervised platform that was developed by Igentify Ltd. to streamline the genetic counseling process by providing digital patient intake and automated genetic counseling for low-risk individuals, as well as clinician-facing tools that reduce administrative burden in patient-related activities (PRAs) (Fig. 1). In essence, the DGA is meant to enable clinicians to increase efficiency while still providing an accessible and relevant patient experience. Additionally, the DGA provides all patient-facing materials at an 8th grade comprehension level and in several languages, which further increases accessibility for patients with different health literacy levels, as well as non-native speakers. Several other tools exist, aiming to bridge similar gaps in genetic counseling workflows, such as genetics advisor (https://www.geneticsadviser.com/), Gia (Invitae, https://www.invitae.com/us/providers/gia-chatbot) and My Gene Counsel (https://www.mygenecounsel.com/). These tools lower the barriers to patient engagement and aid in streamlining patient intake and return of results. However, to the best of our knowledge, the DGA is the only tool that provides a multi-media, uniquely personalized experience for patients, simulating some aspects of a genetic counseling session.

In this study, we evaluated the clinical utility of the DGA as well as its ability to streamline the genetic counseling workflow and gauge patients and GCs satisfaction levels. This study evaluated the DGA specifically in the context of carrier screening; if proven successful, its use could then be extended to other areas of genomic medicine.

Results

Study participants and demographics

Throughout the study period, 450 individuals (225 couples) underwent ECS testing in conjunction with the DGA platform. All couples who were interested in the ECS testing and fulfilled inclusion and exclusion criteria, were offered to participate in the study and consented to do so.

The most common participant ethnicities were multi-ethnic (n = 213/450, 47.3%) and Ashkenazi Jewish (n = 152/450, 33.8%), see Supplementary Fig. 2. Participant age and pregnancy status were available for the TASMC cohort only (n = 278/450, 61.8%). Data regarding participant age were available for 141 participants (31.3%), ranging from 27–53 years with a median age of 35 years. Pregnancy-related data were available for 129 female participants (57%, n = 129/225). Of these, 93 (72%, n = 93/129) were not pregnant at the time of testing, while 36 (28%, n = 36/129) were pregnant, with pregnancy week ranging from 3–34, and a median of 12 weeks. Data regarding medical history was reported by 49.6% (n = 138/278) individuals. As both recruiting centers are situated in the center of Israel, the participating individuals represent a diverse demographic population, characteristic of the Israeli society, with the exception of underrepresentation of the Arab population, which is mainly populated in the North and South of the country (only two Muslim Arab couples took part in the study, Supplementary Fig. 2).

Consent video comprehension

Questionnaire data regarding participant understanding about the test was retrieved and analyzed for the 278 TASMC cohort participants in order to study the effectiveness of the consent video in adequately educating patients about the test. Out of 278 participants, 229 (82.3%, n = 229/278) answered all three comprehension test questions correctly on the first attempt. Thirty-three participants (11.8%, n = 33/278) answered incorrectly at least once, but successfully corrected their answers without being directed to re-watch the video. Eight participants (2.9%, n = 8/278) were required to re-watch the video due to more than two failed attempts to answer a question correctly, and only one participant (0.4%, n = 1/278) had to re-watch the video twice due to multiple failed attempts at answering the questions.

We analyzed the distribution of wrong answers among the three questions in order to understand whether specific questions were an obstacle for the participants (Supplementary Fig. 3). We found that 88.1% (n = 245/278) answered the first question correctly on the first attempt, 92.1% (n = 256/278) answered the second question correctly on the first attempt, and 95.7% (n = 266/278) answered the third question correctly on the first attempt.

Overall, the data show that most of the participants understood the explanatory video, and although the first question was slightly more challenging, the participants were able to answer all three questions correctly on their second attempt, and only nine participants had to re-watch the video. After re-watching the video, participants were given the comprehension test again and ultimately, all participants completed it successfully. Based on our data, there were no participants who dropped out at the comprehension test stage, indicating that the comprehension evaluation process is effective, and participants are able to progress successfully.

Testing results and risk assessment

Nine couples (4%, n = 9/225) were found to be at high-risk due to a pathogenic variant in the same AR gene (n = 8), or because of a female carrier of an X-linked condition (n = 1) (Fig. 2). These couples were invited to F2F genetic counseling. The remaining 216 couples (96%, n = 216/225) were determined to be at low risk. The low-risk couples consisted of 51 couples (23.6%, n = 51/216) where neither partner was found to be a carrier of a tested pathogenic variant, 113 couples (52.3%, n = 113/216) where only one partner was a carrier, and 52 couples (24.1%, n = 52/216) where both were carriers of a pathogenic variant but in different AR genes (Fig. 2). All low-risk couples received an automated personalized digital video and written consult note via the patient-app (link to the app sent via email). Only nine low-risk couples (4.2%, n = 9/216) requested a F2F or telephone counseling session after having received the DGA generated video and consult note. When queried, these couples explained that they wanted to learn more about the ECS test they took and to explore options for further genetic testing.

Fig. 2: Reproductive genetic carrier screening (RGCS) results.

figure 2

The distribution of RGCS testing results among 225 couples tested in the study, revealed 4% high-risk couples.

Full size image

Impact on GC workflow

When comparing the routine workflow with the DGA-assisted one (Fig. 1), a noteworthy reduction in the number of live low-risk genetic counseling encounters can be derived. Traditionally, in Israel, stepwise testing of couples for genetic screening is applied, where the female partner is tested first, and only if she is found to be a carrier for an AR disease, her reproductive partner is also tested (Fig. 1). In this routine process, all the couples with at least one carrier of a pathogenic variant receive a F2F or telephone genetic counseling to explain the interim results and refer the partner for subsequent testing. Had the routine workflow been applied to the cohort that participated in this study, 77.3% of the couples (n = 174/225) would have received a live counseling session. By testing both reproductive partners simultaneously and generating personalized videos via the DGA for pretest education and post-test results delivery instead of a live counseling session, GCs did not perform any live pretest education or consenting, and only provided F2F encounters for 8.0% of couples (n = 18/225), of whom half were high-risk couples (n = 9/18), and half were low-risk couples that requested additional information (n = 9/18). Therefore, this work scheme allowed us to forfeit all the pretest encounters for all couples, and to reduce the personal encounters for low-risk couples by 69.3% (n = 156/225, including: 113 couples where only one partner is a carrier, 52 couples where both are carriers of AR variants in different genes, minus the nine couples that requested F2F counseling despite being low-risk).

DGA performance

Our analyses revealed that the DGA algorithm performed according to GC expectations, i.e., accurate triaging of low and high-risk couples, in all but two cases where participants were found to carry two variants in the same gene (CFTR and CPT2), while their partners did not carry any variants in these genes. Although, it was not clear at the time whether both variants were observed in cis or in trans, these cases were initially flagged as “low-risk.” The algorithm was consequently modified to flag such cases as “high-risk” in the future and refer them to F2F genetic counseling and workup for the double-carrier’s parents to decipher the phasing. In both cases, the GCs identified the discrepancy during their evaluation, before sending the results, and the patients were invited for F2F counseling. Both these cases are part of the 4% of high-risk patients who require additional counseling.

Participant satisfaction

Participant satisfaction was assessed via a Likert scale of 1–7 (7 = highest satisfaction) and data was collected via an online survey and completed by 317 participants from both sites (response rate of 70%). Both partners were requested to complete the survey but since survey responses were voluntary, some participants did not complete the survey in its entirety, resulting in different group sizes for each question. Participants who gave a Likert score of 5 or higher were considered to have positive attitudes with regard to the parameter being measured. Results revealed that 91.9% of participants (n = 136/148) felt the pre-test educational video was clear, 88.0% (n = 265/301) had no follow-up questions about the genetic testing process after viewing the pre-test educational video, 87.7% (n = 271/309) felt the post-test video explained their results clearly, 95.4% (n = 62/65) felt comfortable receiving results in a video format, and 83.1% (n = 182/219) would recommend this service to a friend (Fig. 3). Additional results of the participant satisfaction survey can be found in Supplementary Material (Supplementary Figs. 4, 5).

Fig. 3: Participant satisfaction with the DGA platform and process.

figure 3

Representative results of the participant satisfaction survey, answered by 317 individual participants, showing an overall satisfaction with the digital counseling process.

Full size image

We found no statistically significant differences in participant satisfaction scores. Potential correlation between the RGCS result (carrier vs. non-carrier status) and participant satisfaction was examined for the post-test survey using a two-tailed student t-test, and revealed p-values of 0.86 and 1, respectively, indicating no clear difference in participant satisfaction levels between both groups. We also did not find statistically significant differences in participant satisfaction scores between those who reported a family medical history and those who did not (two-tailed t-test, p-value 0.98).

GCs satisfaction

Four GCs from both institutions who used the system answered a GC user experience survey to reflect on their satisfaction with using the clinician-facing platform. Survey responses were measured using a Likert scale ranging from 1–5 where 1 = very difficult and 5 = very easy. Satisfaction scores were calculated based on the sum of weighted averages, where each score was multiplied by the percentage of answers for that score (e.g., if one GC gave a score of 2 and three GCs gave a score of 4; 0.25*2 + 0.75*3 = 3.5). The average GCs scores ranged between 3.33–5 (Fig. 4). The genetic counseling related items received a high score of 4–4.25, suggesting that the DGA effectively streamlined participant handling. Three aspects of the DGA received a rating of 5; time saved when using the platform compared to a routine workflow, comprehensiveness of the explanatory videos, and comprehensiveness of the result videos. The GCs commented that the information collected during the patient interview via the DGA was not as comprehensive as collected in standard clinical practice. However, most of these were not pertinent to the carrier screening process, such as family history of cancer and other late-onset conditions. The GCs also felt that the technical aspects of the DGA could be improved. This is reflected by an average score of 3.33–4 for many of the features surveyed (Fig. 4). These issues were improved in subsequent DGA versions.

Fig. 4: Genetic counselor (GC) satisfaction survey.

figure 4

Results of the DGA satisfaction survey, answered by four GCs who used the system; results are presented as weighted averages.

Full size image

Discussion

While the demand for genetic testing is ever-increasing, it is not reciprocated by an increase in the number of available GCs. A 2019 study estimated that there are no more than 7000 GCs in 28 developed countries6,7. Furthermore, genetic services are time- and labor-intensive. In one study, GCs reported seeing an average of 10 patients per week with a mean F2F session of 47 min. Moreover, each session requires additional time for other PRAs. The most time-consuming task was genetic report writing, followed by case preparation, follow-up, and other administrative tasks8. A recent study showed that higher levels of burnout among GCs were associated with insufficient administrative support and lack of autonomy, among others9. The authors also found that higher levels of GC burnout were associated with lower levels of empathy and a greater desire to reduce the amount of time spent on clinical care9. In Israel, the demand for genetic services is very high and waiting times for a GC appointment may take several months according to a survey made by Deloitte in collaboration with the Israeli Society of Medical Genetics. Identifying methods to reduce GC workload is important for providing genetic counseling in a timely manner while reducing GC burnout.

In recent years, efforts are being put into developing genetic-specific technological platforms to facilitate the genetic counseling process, both for patients and genetic staff10. While there are a growing number of these digital platforms, most are limited to providing educational materials10,11 via tools that enable uniform pretest education (such as a chatbot)12,13 or post-test results delivery systems14, or both15. Some of these platforms include in-house development and internal use in commercial laboratories, e.g., Color (https://www.color.com/product/overview), Invitae (https://www.invitae.com/en/individuals), and Pathway Genomics (https://www.pathway.com), as well as more interactive platforms, which allow patients to make choices and manage their genetic results16,17. Only a few platforms provide support for the entire process, from pretest counseling and education to the delivery of post-test results; one such platform is the ‘Genetics Adviser’’18.

To the best of our knowledge, the GC-supervised DGA by Igentify Ltd. is the only platform which is personalized based on the participant’s medical information and genetic test results, and one of the few available solutions that support the entire process, including a pretest personalized explanatory video, a digital consent process following a comprehension test, and personalized delivery of post-test results. It should be emphasized that the DGA is a medically supervised tool and does not make autonomous decisions. It employs medical algorithms suggesting actions and recommendations to the GC, who in turn needs to approve, modify, or reject each suggestion. All GCs that have used the DGA reported that the entire digital process saved time compared to common practice, that the pretest educational video was clear, and the results video contained sufficient pertinent information (Supplementary Fig. 4).

As expected, the use of the DGA was associated with a significant reduction in the number of F2F genetic counseling sessions. To date, the routine workflow does not provide post-test F2F counseling for couples where neither partner is a variant carrier (n = 51/225, 22.7% in this series; Fig. 2). In these cases, providing an additional personalized video to explain test results and their implications elevates the current standard of care. The bulk of cases that previously required time and effort included couples where only one partner is a variant carrier (n = 113/225, 50.2% in this series; Fig. 2), or where both are carriers of a pathogenic variant, but in different AR genes (n = 52/225, 23.1% in this series; Fig. 2). For these couples, the couple-based testing paradigm together with the DGA results delivery model, has the potential to save 72.9% of F2F sessions. Moreover, our results show that the DGA was well received by participants (Fig. 3). Nonetheless, a small number of low-risk participants (4%, n = 9/225) still requested a F2F or telephone conversation. These conversations were generally brief and focused because the couple had already viewed the video prior to counseling, allowing them to attend the counseling session in a more prepared manner. Therefore, the overall reduction in genetic counseling sessions in this study was 69.3%. Similar results were presented at the 2023 annual meeting of the International Society for Prenatal Diagnosis by Labcorp, using the same DGA tool19. In a cohort with 96 cases, they noted a high degree of patient satisfaction, positive user experience, as well as high patient-knowledge assessment scores, further demonstrating the effectiveness of this platform for ECS.

Our study demonstrates that the DGA platform can significantly compensate for the shortage of GCs, thereby reducing burnout, particularly in the area of RGCS. Follow-up studies could reinforce our findings and further assess the extent to which this or other tools are beneficial to the GCs.

It should be noted that the technology-based solutions used in this study are designed to an eighth-grade level education, thus they do not address the needs of populations with low education and literacy, or those who struggle with technology. Additionally, individuals with an Arab ethnicity were underrepresented in the study; further studies are needed to examine the use of a digital counseling system in this population. This study also did not examine satisfaction for couples who had negative past pregnancy-related experiences, e.g., after a miscarriage, a fetus with congenital defects, or abnormal pregnancy testing, who might find the DGA impersonal. Such couples, as well as those uncomfortable with a digital tool, may still request F2F genetic counseling sessions; however, with the widened use of such digitalized solutions, the GCs would have additional time to counsel couples with the need for F2F interaction. Furthermore, only four GCs worked with the system and answered the GC satisfaction questionnaire. This is a limited group size that should be expanded in future studies, as well as tested in other populations. In addition, we tested the DGA only in the setting of couple testing. It would be valuable to test the workflow in case where the male partner is not available. Moreover, the study did not compare participant satisfaction between those who performed the flow using the DGA system and those who underwent F2F counseling. Such a comparison could provide valuable insights into the effectiveness and patient satisfaction of ‘digital’’ versus ‘traditional’’ genetic counseling methods. Additionally, demographic data collection methods varied between both study sites due to differences in institutional protocols, which may have influenced the data collected.

To summarize, this study presents an evaluation of a DGA tool developed by Igentify Ltd. to facilitate ECS counseling and testing procedures. Our results show that the DGA saved 69.3% of F2F and telephone interactions between GCs and patients and demonstrated a high degree of participant understanding and satisfaction. The use of such platforms can be further developed for other areas amenable to streamlining, such as counseling and consenting for non-invasive prenatal testing, prenatal invasive testing, next generation sequencing testing, oncogenetics, and more. These digital counseling platforms can provide accessible and personalized genetic counseling to individuals in remote areas, increase genetic counseling availability, and reduce costs. We anticipate that, in the future, technological solutions could be useful in reducing the time spent by GCs doing routine tasks, thus allowing GCs to focus on complex cases requiring intense counseling.

Methods

Study participants and ethics

Couples undergoing RGCS testing at the Genetics Institutes at Tel Aviv Sourasky Medical Center (TASMC) and Sheba Medical Center (SMC) from December 2020 through August 2021 were enrolled in the study. RGCS was performed using ThermoFisher CarrierScan®, which tests for 1643 curated pathogenic variants in 365 genes prevalent in the diverse Israeli population20. Participation in this study was limited to reproductive partners undergoing testing simultaneously. The availability of both partners for testing was the main inclusion criteria for the study. Exclusion criteria consisted of cases where only one of the reproductive partners was available for testing, or if an individual had a rare ethnicity that is not covered by the RGCS panel (e.g., temporary migrant workers from South-East Asia); these individuals were not offered participation or testing. Demographic information, including ethnicity, age, and pregnancy status, was collected.

The study was approved by the TASMC (#0569-19-TLV) and the SMC (#6346-19-SMC) Helsinki ethics committees, and participants were informed that by filling out the questionnaires, they are providing consent to the study.

Digital counseling and genetic testing procedure: the participant’s aspects

To initiate the counseling and testing process, both partners were provided an individual link or QR code to self-enroll through the DGA web application (Fig. 1). Following enrollment, each participant completed a medical intake, which collects relevant personal and family medical history (see ‘Enrollment questionnaire’’ in Supplementary Material). Next, participants were shown a personalized educational video about the RGCS test, based on the information provided in the medical intake. The medical intake contains questions regarding personal or family history of conditions that may have a genetic background, including multiple birth defects or early death; intellectual disability, developmental delay, or autism; muscular or neurological disease; cancer; hereditary diseases or genetic syndromes; deafness or blindness, as instructed by the Israeli MOH. The DGA uses a complex algorithm that generates participant-specific videos within a matter of seconds, and incorporates individual patient information, such as name, week of gestation and medical history, to create a personalized clinically-relevant experience. For example, if a patient reports a family history of blindness, the algorithm identifies this information and includes a specific scene in the video and in the consult note that explains the importance of genetic counseling in such cases. This analysis occurs through a series of predefined rules that ensure the accuracy and relevance of the content. Following the explanatory video, participants were required to undergo a short comprehension evaluation by answering three multiple-choice questions regarding information presented in the video (see ‘Patient Comprehension Test’’ in Supplementary Material). Participants who failed to correctly answer a question were referred back to the educational video. This step is repeated until patients are able to answer all three questions correctly. While there is no specific mechanism to refer patients to F2F counseling if they struggle with the questions, it is important to note that this scenario is very rare and that the participant can ask for F2F counseling at any stage. Following successful completion of this step, participants were asked to sign an electronic consent form and to undergo testing.

After completion of digital patient intake, participants had their DNA sampled at the Genetics Institute via blood samples for RGCS testing. Residual risk was calculated automatically in the DGA system based on recommendations by Dungan et al.21.

It should be noted that GCs were available at the Genetics Institutes to discuss any queries that may arise by the participants throughout the entire process.

Digital counseling and genetic testing procedure: the Genetics Institute aspects

The Igentify DGA platform includes a dedicated Patient Management Dashboard (Supplementary Fig. 1), which enables GCs to monitor the progression of a participant throughout the process. Once a patient sample was collected and sent to the testing laboratory, all relevant data appeared in a Laboratory Dashboard. When test results were released by the laboratory, GCs received a notification, which presented all the relevant patient and test information, including alerts regarding personal or family history. At this stage, couples were triaged by the system and the GCs were alerted on potential high-risk scenarios. High-risk couples were defined as those where both partners carry a pathogenic variant in the same AR disease-causing gene, or when the female partner was found to carry an X-linked condition, while low-risk couples were defined as those who do not carry a pathogenic variant, or when one or both are carriers of a pathogenic variant in different AR genes.

Each of the DGA recommendations regarding medical history and RGCS test results was medically supervised, carefully reviewed, and approved or modified by an authorized GC, prior to finalizing the report. This step provided the GC with all the required information for test results sign-out and release via the DGA to the participants. While reviewing all the cases, the GC could establish the need for F2F genetic counseling for reasons other than abnormal RGCS test results, e.g. a high-risk designation could be given by the GC based on family or personal medical history, such as a first-degree relative with cognitive impairment, offspring with birth defects, etc. This information provided the GC with the opportunity to calculate recurrence risks and recommend further genetic evaluation, as needed. The DGA algorithm does not compare the responses by both partners; therefore, the system cannot highlight incongruent information. As such, it is the GCs responsibility to review the data before releasing the final report.

It should be noted that the purpose of the GC supervision is to make sure that nothing is missed by the DGA system, and not to decide on whether the genetic screening test should be carried out. Therefore, the RGCS test could be pursued by the patients in parallel to individualized genetic counseling related to medically relevant information disclosed by the patients during the enrollment (e.g., positive family history).

Digital counseling and genetic testing procedure: post-test results delivery and counseling

Low-risk couples received RGCS results via an automatically generated personalized video (see https://vimeo.com/508041112 for an example), as well as an automatically generated consult note. The consult note includes demographic information, details on the test type, a genetic result summary with risk indications, and recommendations including instructions on follow-up testing or genetic counseling. Participants were also given the option to request a F2F or a telephone session to clarify information they were unsure of. Importantly, couples with low-risk results and without a family history of disease were not required to see a GC during the process. All participants who reported a personal or family history were referred for F2F genetic counseling focusing on the reported conditions; however, this did not exclude them from the use of DGA for carrier screening or from participation in the study.

High-risk couples were invited to an F2F genetic counseling session. In these cases, the GCs had already received the patient’s initial information and genetic test results, and a consult note was already drafted by the system. These capabilities significantly streamlined the counseling process.

DGA performance evaluation

We aimed to assess the accuracy of the DGA algorithm in interpreting the RGCS results, combined with the clinical information, into accurate and tailored explanations and recommendations for the patients. This was achieved by GC supervision on the DGA output. Once the test results were available in the DGA system, the GCs reviewed the participants’ questionnaires and ensured that the DGA’s consult notes and videos included all relevant recommendations. These recommendations included genetic counseling based on family history, consanguinity, or special disclaimers dependent on pregnancy week. The GCs assessed the consult notes and videos for any errors, and requested adjustments to the algorithm accordingly.

User satisfaction

Participants and the GCs that used the DGA were asked to complete questionnaires (see ‘Patient satisfaction questionnaires’’ in Supplementary Material) to assess their satisfaction, using a Likert scale of 1–7 (7 = highest satisfaction). The participants were asked to complete two questionnaires, one after the enrollment process, and the other upon receiving the test results. Data were collected via SurveyMonkey (https://www.surveymonkey.com/).

We examined whether the ECS result (carrier vs. non-carrier status) had any effect on patient satisfaction, using a two-tailed student t-test. Participants who answered the surveys anonymously (n = 118/317, 37.2%) were excluded from this analysis since, in those cases, we could not correlate result type to satisfaction scores. The analysis was performed on two questions from the post-test survey (questions 8 and 16 in ‘Participant satisfaction questionnaire: Second survey—Following results’’ in Supplementary Material). Data were divided into two groups and compared; group one consisted of participants who answered the survey questions and had carrier status, and group two consisted of participants who answered the survey questions and had no findings.

Data availability

Data not presented in the manuscript will be provided upon reasonable request to the corresponding authors. The data are not publicly available due to privacy or ethical restrictions.

Code availability

The data evaluating the use and performance of the DGA system was not analyzed by code. The code to the DGA system is not publicly available.

References

Ayme, A. et al. Determinants of genetic counseling uptake and its impact on breast cancer outcome: a population-based study. Breast Cancer Res. Treat. 144, 379–389 (2014).

CASPubMedGoogle Scholar

Gabai-Kapara, E. et al. Population-based screening for breast and ovarian cancer risk due to BRCA1 and BRCA2. Proc. Natl. Acad. Sci. USA 111, 14205–14210 (2014).

CASPubMedPubMed CentralGoogle Scholar

Mattick, J. S. et al. The impact of genomics on the future of medicine and health. Med. J. Aust. 201, 17–20 (2014).

PubMedGoogle Scholar

Metcalfe, S. A. Genetic counselling, patient education, and informed decision-making in the genomic era. Semin. Fetal Neonatal. Med. 23, 142–149 (2018).

PubMedGoogle Scholar

Wetterstrand, K. A. DNA sequencing costs: data. National Human Genome Research Institutehttps://www.genome.gov/about-genomics/fact-sheets/DNA-Sequencing-Costs-Data (2023).

Abacan, M. A. et al. The global state of the genetic counseling profession. Eur. J. Hum. Genet. 27, 183–197 (2019).

PubMedGoogle Scholar

Ormond, K. E. et al. Genetic counseling globally: where are we now? Am. J. Med. Genet. Part C. Semin. Med. Genet. 178, 98–107 (2018).

PubMedGoogle Scholar

Attard, C. A., Carmany, E. P. & Trepanier, A. M. Genetic counselor workflow study: the times are they a-changin’? J. Genet. Couns. 28, 130–140 (2019).

PubMedGoogle Scholar

Caleshu, C. et al. Contributors to and consequences of burnout among clinical genetic counselors in the United States. J. Genet. Couns. 31, 269–278 (2022).

PubMedGoogle Scholar

Gordon, E. S., Babu, D. & Laney, D. A. The future is now: technology’s impact on the practice of genetic counseling. Am. J. Med. Genet. Part C Semin. Med. Genet. 178, 15–23 (2018).

PubMedGoogle Scholar

Gasteiger, N. et al. Digital interventions for genomics and genetics education, empowerment, and service engagement: a systematic review. J. Community Genet. 14, 227–240 (2023).

PubMedPubMed CentralGoogle Scholar

Al-Hilli, Z. et al. A randomized trial comparing the effectiveness of pre-test genetic counseling using an artificial intelligence automated chatbot and traditional in-person genetic counseling in women newly diagnosed with breast cancer. Ann. Surg. Oncol. 30, 5990–5996 (2023).

PubMedGoogle Scholar

Schmidlen, T, Schwartz, M, DiLoreto, K, Kirchner, H. L. & Sturm, A. C. Patient assessment of chatbots for the scalable delivery of genetic counseling. J. Genet. Couns. 28, 1166–1177 (2019).

PubMedGoogle Scholar

Suckiel, S. A. et al. GUÍA: a digital platform to facilitate result disclosure in genetic counseling. Genet. Med. 23, 942–949 (2021).

PubMedPubMed CentralGoogle Scholar

Biesecker, B. B. et al. Web platform vs in-person genetic counselor for return of carrier results from exome sequencing a randomized clinical trial. JAMA Intern. Med. 178, 338–346 (2018).

PubMedPubMed CentralGoogle Scholar

Bombard, Y. et al. The genomics ADvISER: development and usability testing of a decision aid for the selection of incidental sequencing results. Eur. J. Hum. Genet. 26, 984–995 (2018).

CASPubMedPubMed CentralGoogle Scholar

Tabor, H. K. et al. My46: a web-based tool for self-guided management of genomic test results in research and clinical settings. Genet. Med. 19, 467–475 (2017).

PubMedGoogle Scholar

Clausen, M., et al. Genetics adviser: the development and usability testing of a new patient digital health application to support clinical genomic testing. Genet. Med. Open 2, 101814 (2024).

PubMedPubMed CentralGoogle Scholar

Cutillo, D., et al. P381: The digital genetic assistant: effectiveness of a novel patient approach to carrier screening. Genet. Med. Open 1, 100417 (2023).

Google Scholar

Reches, A. et al. Expanded targeted preconception screening panel in Israel: findings and insights. J. Med. Genet. 61, 783–787 (2024). https://doi.org/10.1136/jmg-2023-109629

Dungan, J. Expanded carrier screening: what the reproductive endocrinologist needs to know. Fertil. Steril. 109, 183–189 (2018).

PubMedGoogle Scholar

Download references

Acknowledgements

The authors wish to thank the couples participating in the study and answering the online survey. The study supported by a grant from The Israel Innovation Authority (#68143).

Author information

Author notes

These authors contributed equally: Yuval Yaron, Vered Ofen Glassner, Michal Berkenstadt, Adi Reches, Alina Kurolap, Hagit Baris Feldman.

Authors and Affiliations

The Genetics Institute and Genomics Center, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel

Yuval Yaron, Vered Ofen Glassner, Adi Reches, Alina Kurolap & Hagit Baris Feldman

Department of Obstetrics and Gynecology, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel

Yuval Yaron & Adi Reches

School of Medicine, Faculty of Medical and Health Sciences, Tel Aviv University, Tel Aviv, Israel

Yuval Yaron, Michal Berkenstadt, Elon Pras & Hagit Baris Feldman

The Danek Gertner Institute of Human Genetics, Sheba Medical Center, Tel-Hashomer, Israel

Michal Berkenstadt, Nurit Goldstein, Haike Reznik Wolf, Liat Ries Levavi, Liat Abo Gutstein & Elon Pras

Igentify Ltd, Caesarea, Israel

Yael Furman, Mori Anouchi, Galit Delmar & Doron M. Behar

Authors

Yuval Yaron

View author publications