The COVID-19 pandemic led to significant changes in healthcare delivery, including the increased use of telemedicine, and understanding the impact of these changes on cancer risk assessment is critical to ensuring the effective delivery of these important services. There have been several studies examining the feasibility and non-inferiority of telemedicine services over time, showing that the use of telemedicine services in many healthcare specialties does not sacrifice quality of care provided [16,17,18]. However, prior studies before the COVID-19 pandemic have less relevance to our current healthcare landscape post-COVID, as telemedicine has now become a more common and familiar fixture in our healthcare framework. Throughout the pandemic the GI-CREP at our institution utilized both in-person and telemedicine appointments, and we utilized this unique clinic set-up to perform side-by-side comparison of these two different modalities. This study illustrates that although telemedicine was favored by most patients, it did not substantially impact the likelihood of GI-CREP appointment completion, of genetic testing recommendation by providers, or of consent for genetic testing by patients. However, telemedicine use was associated with lower rates of genetic testing completion and a longer turnaround time for receipt of results.

The use of telemedicine can reduce barriers to care for certain patient populations [19]. Patients can avoid traveling long distances, reduce time off work, and importantly, avoid potential COVID-19 exposures. The latter is especially important in the cancer-risk evaluation setting as patients with active cancers may be immunocompromised. In this study’s survey cohort, most patients indicated that they preferred the telemedicine modality with a common reason being that it reduces both the time and costs associated with traveling to the clinic. While telemedicine helps to lower these barriers, it can also introduce new barriers. In order to complete a virtual appointment, the patient must have reliable phone or internet service, and be familiar with the technology used to access their visit. Although it is relatively common for US adults to have access to an internet connection, there are some who do not and therefore may have difficulty accessing a telemedicine appointment. Pew reported in 2021 that approximately 7% of US adults do not have access to the internet, and the National Telecommunications and Information Administration (NTIA) reports that 1 in 5 US households does not have internet access in the home [20, 21]. Telemedicine appointments are also subject to technical glitches such as frozen video or audio cutting out, despite proficiency with technology. In a systematic review article published in 2022, technical issues were found to be the most common issues reported with the use of telehealth [22].

Although telemedicine may lower some barriers, our study found that being scheduled for a telemedicine appointment was not associated with higher rates of appointment completion compared to traditional in-person visits. Possible explanations for why appointment completion rates remain similar for in-person and telemedicine visits are that individuals may be less invested in a telemedicine visit. For example, it may be easier to forget about the appointment if patients do not need to make special arrangements to attend. It may be more difficult for individuals with cancer to attend an extra appointment, regardless of modality, when their schedule can already be full of healthcare visits. Furthermore, technology issues related to telemedicine appointments may prevent patients from logging in to attend their appointments. However, as patients in this cohort were not randomly assigned to a telemedicine or in-person visit type, self-selection bias is certainly possible, which could only be overcome by a future clinical study where individuals are randomized to appointment type.

Overall, we found that 81% of patients completed a GI-CREP appointment, which was higher than completion rates pre-pandemic (75%).11 Pre-pandemic we showed that certain factors including Medicaid insurance, self-identified Black race, and a personal history of cancer were associated with lower rates of GI-CREP appointment completion [11]. Fortunately in this study there were no differences in appointment completion rate based on race/ethnicity during the COVID-19 pandemic (Table 1). However, disparities in appointment completion remained based on insurance status and history of cancer, with individuals with Medicaid insurance and a personal history of cancer having lower rates of GI-CREP appointment completion during the COVID-19 pandemic. Given the presence of disparities based on insurance status and personal history of cancer that are observed both pre- and post-pandemic, it is unlikely that pandemic-specific factors were the primary drivers of these disparities.

Our findings demonstrated that the completion rate for genetic testing was significantly different between telemedicine and in-person appointments. In fact, patients seen via telemedicine had a more than three-fold higher chance of not completing genetic testing after consenting compared to patients seen in-person (18.3% of telemedicine patients compared to 5.2% of in-person patients, Table 1). The high rate of genetic testing completion for in-person appointments is likely because patients typically have their blood or saliva sample collected during the in-person appointment. For patients seen via telemedicine, sample collection happens after visit completion, with a saliva kit typically mailed to the patient’s personal address, which could take days or weeks depending on mail delays. During this time, individuals may have second thoughts about following through with their testing, lose interest in the possible results, or become busy and not prioritize the sample collection [23]. Additionally, there could also be increased rates of inadequate or improper sample collection by patients when they are collecting their sample at home without the direct guidance of a genetics professional [24]. Based on the results from this study, new processes were implemented within GI-CREP to improve testing completion rates. Briefly, this process involves requesting monthly reports from commercial laboratories listing patients with outstanding genetic testing samples. Using these lists, a genetic counseling assistant (GCA) will first remind patients about sample collection through the electronic medical record, and if the sample remains outstanding the GCA will attempt to contact the patient by phone. Future studies will be able to assess whether these newly implemented processes improve testing completion rates.

This study also observed significant differences in turnaround time for the return of genetic testing results. Return of results took more than twice as long for telemedicine visits compared to their in-person visit counterparts (32 days for telemedicine compared to 13 days for in-person). It is likely that these delays in return of results are at least partially related to extra time required to mail a sample collection kit to patients and the delay of independent sample collection on the patient’s own accord [24]. The discrepancy in turnaround time between telemedicine and in-person visits is an important issue to consider in the continued use of telemedicine services, especially in the cancer risk assessment setting when genetic testing results could potentially impact treatment options or surgical decision-making.

Limitations of this study include data collection from a single institution GI cancer risk evaluation program. Another limitation is that patients were offered the first available GI-CREP appointment but could choose to wait longer for a different appointment type, and therefore they were not randomly assigned telemedicine or in-person appointments. However, in contrast to cancer genetics programs that transitioned entirely to telemedicine during the COVID-19 pandemic, our clinic structure allowed a head-to-head comparison of these two different appointment modalities within the same time period. Another potential limitation is how translatable our data may be moving forward from the COVID-19 pandemic, as this research was conducted towards the beginning of the pandemic where patients may have had limited experience with telehealth and may have been uncomfortable visiting healthcare facilities in-person.

In conclusion, telemedicine has been a vital service delivery option for GI cancer risk evaluation during the COVID-19 pandemic. When comparing in-person and telemedicine visits, there were no differences in providers recommending genetic testing nor in rate of patient consent for genetic testing. However, more than threefold more patients seen by telemedicine did not complete genetic testing, and return of results took more than twice as long compared to in-person visits. Therefore, it is important to continue to address logistical challenges related to telemedicine including ensuring reliable and expedient sample return through use of increased GCA-driven patient reminders, while simultaneously maintaining the option for in-person visits for patients who are not comfortable using telehealth as well as those where expedited return of results is important for immediate clinical decision-making.