Diabetes Education and Support Tele-Visit Needs Differ in Duration, Content, and Satisfaction in Older Versus Younger Adults

Background: Diabetes education and support are critical components of diabetes care. During the COVID-19 pandemic, when telemedicine took the place of in-person visits, remote Certified Diabetes Care and Education Specialist (CDCES) services were offered to address diabetes education and support. Specific needs for older adults, including the time required to provide education and support remotely, have not been previously reported. Methods: Adults with diabetes (primarily insulin-requiring) were referred to remote CDCESs. Utilization was individualized based on patient needs and preferences. Topics discussed, patient satisfaction, and time spent in each tele-visit were evaluated by diabetes type, age, sex, insurance type, glycosylated hemoglobin (HbA1c), pump, and continuous glucose monitor (CGM) usage. t-Tests, one-way analysis of variance, and Pearson correlations were employed as appropriate. Results: Adults (n = 982; mean age 48.4 years, 41.0% age ≥55 years) with type 1 diabetes (n = 846) and type 2 diabetes mellitus (n = 136, 86.0% insulin-treated), 50.8% female; 19.0% Medicaid, 29.1% Medicare, 48.9% private insurance; mean HbA1c 8.4% (standard deviation 1.9); and 46.6% pump and 64.5% CGM users had 2203 tele-visits with remote CDCESs over 5 months. Of those referred, 272 (21.7%) could not be reached or did not receive education/support. Older age (≥55 years), compared with 36–54 year olds and 18–35 year olds, respectively, was associated with more tele-visits (mean 2.6 vs. 2.2 and 1.8) and more time/tele-visits (mean 20.4 min vs. 16.5 min and 14.8 min; p < 0.001) as was coverage with Medicare (mean 2.8 visits) versus private insurance (mean 2.0 visits; p < 0.001) and lower participant satisfaction. The total mean time spent with remote CDCESs was 53.1, 37.4, and 26.2 min for participants aged ≥55, 36–54, and 18–35 years, respectively. During remote tele-visits, the most frequently discussed topics per participant were CGM and insulin pump use (73.4% and 49.7%). After adjustment for sex and diabetes type, older age was associated with lack of access to a computer, tablet, smartphone, or internet (p < 0.001), and need for more education related to CGM (p < 0.001), medications (p = 0.015), hypoglycemia (p = 0.044), and hyperglycemia (p = 0.048). Discussion: Most remote CDCES tele-visits were successfully completed. Older adults/those with Medicare required more time to fulfill educational needs. Although 85.7% of individual sessions lasted <30 min, which does not meet current Medicare requirements for reimbursement, multiple visits were common with a total time of >50 min for most older participants. This suggests that new reimbursement models are needed. Education/support needs of insulin-treated older adults should be a focus of future studies.


Introduction
Diabetes self-management education and support (DSMES) provided by qualified health care professionals, including Certified Diabetes Care and Education Specialists (CDCESs), has been shown to improve diabetes self-management and outcomes, 1-14 but health care coverage is variable and many insurers will only reimburse for education sessions that last for at least 30 min. [13][14][15] In addition to lack of access, barriers to participation include distance, cost, and convenience (i.e., taking time off from work). [13][14][15][16] Some barriers to receiving professional care for diabetes can be ameliorated by telehealth. 17 Telehealth has the potential to reach many more people with diabetes (PWD), and numerous studies have demonstrated efficacy. 7 Internet access as well as access to computers, tablets, and/or smartphones and technological acumen correlate with age and education level. 18,19 The specific needs for education and support of older adults with insulin-requiring diabetes using telehealth have not been well studied. The time required to meet diabetes education and support needs of older adults using telemedicine, including needs related to the use of new and remote monitoring devices and insulin delivery systems, is important to understand but has not been previously reported.
With the onset of the COVID-19 pandemic, in-person adult diabetes care visits ceased, and remote provider and CDCES visits were started. The CDCESs provided remote education and support, including device education (insulin pump, continuous glucose monitor [CGM], and meter support), lifestyle, and psychosocial support, and addressed the management of hypoglycemia and hyperglycemia, financial concerns, sick day management, and the insulin injection technique.
Previous studies have shown patient satisfaction with provider telemedicine visits to be high. [20][21][22][23][24][25] Patients preferred telemedicine for reasons such as convenience, illness, or wanting to avoid waiting rooms for risk of becoming infected with COVID-19 during the pandemic. 26 However, there is a relative lack of data specifically pertaining to patient satisfaction with remote diabetes education and support across the lifespan.
In this study, real-world experience working with remote CDCESs to facilitate care for PWD on complex insulin regimens is reported, and differences in the duration and content of these tele-visits, and patient satisfaction are examined in older versus younger adults.

Research design
A cross-sectional design was used to assess the utilization, impact, and satisfaction related to remote diabetes education and support tele-visits.

Participants
One diabetes center in Syracuse, New York referred adults (ages ‡18 years) to a team of remote CDCESs to provide diabetes education, training, and support during the early months of the COVID-19 pandemic (from April 2020 to September 2020). Referrals were made for any English-speaking adult with type 1 diabetes (T1D) mellitus with an upcoming medical visit. Based on needs, providers could refer individuals with type 2 diabetes (T2D) mellitus, primarily using multiple daily insulin injections. Adults who were non-English speaking were provided services by the diabetes center using translation services that were previously in place. This study was reviewed by The Institutional Review Board for the Protection of Human Subjects at SUNY Upstate Medical University.
Diabetes care, education, and support tele-visits A remote team of CDCESs provided diabetes education, training, and support. This service was free of charge to participants, with the cost being covered by COVID relief funds given by the Leona M. and Harry B. Helmsley Charitable Trust. These funds permitted the provision of unrestricted comprehensive DSMES based on participants' needs and level of desire. Medical care provided by physicians and advanced practice practitioners was billed as usual.
The following information was made available to the remote CDCESs: participants' age, sex, type of diabetes, medications, diabetes devices, most recent glycosylated hemoglobin (HbA1c; all HbA1c values ‡14.0% were recorded as 14.0%), and contact information. Additional information was provided as needed, including any known potential barriers to education, the presence of diabetes complications, and urgency of the referral. This information was uploaded to a secure cloud content management and file sharing service.
The initial contact attempt was made by phone, and follow-up communications between the remote CDCESs and the participant were performed by either phone or video. If the CDCES determined a video tele-visit would be helpful for the participant, this would be offered. The team of remote CDCESs were detailed and used talk-back methods to help the individual use their software and devices as needed. Video visits were used for the majority of individuals who were started on CGM. The CDCESs were also able to instruct individuals using the phone while also referring the participant to additional starter guides and videos as indicated. Up to three attempts were made to reach each participant.
Participants were considered to have a tele-visit with a CDCES when any diabetes education or medical information was reviewed or discussed. When televisits were performed by video, the platform Zoom for Healthcare was used. The remote CDCES team's call center platform was used to obtain the duration of each tele-visit (minutes).
Information collected by the CDCESs included the frequency of visits, topics discussed, device data, any dosing changes made within approved parameters, and/or urgent recommendations made by the remote team. This information was uploaded into the EMR daily, and it was forwarded to the participant's provider for review. A CDCES at the diabetes center was always available to receive calls from the remote CDCESs for urgent issues.
After the first 1-2 months, titration of insulin doses by the remote CDCES was initiated within pre-defined parameters. If the remote CDCES had high priority recommendations or concerns after contacting a participant, such as dose change recommendations outside of the set parameters, this information was communicated to the diabetes center CDCES to address. A virtual huddle between the primary site and remote CDCES teams was initially conducted daily, as new workflows and procedures were initiated, and then 1-3 times per week for the subsequent 2-3 months.
Participant satisfaction with remote CDCES tele-visits Surveys were completed between May 2020 and August 2020 after at least one visit with a remote diabetes educator. Participants were asked to complete the survey electronically by using Research Electronic Data Capture (REDCap). If there was no response within 1 week or if there was no e-mail address available, the survey was administered by phone. Each participant was called a maximum of two times.
A phone script was used when calling participants to standardize survey administration. Special attention was given to ensure that little feedback was given by research assistants over the phone. Survey questions used a 5-point Likert scale: strongly disagree (1) through strongly agree (5). Participants could also provide additional comments. Questions were developed with guidance from previously used telehealth surveys such as the Telemedicine Satisfaction Questionnaire 27 and Telehealth Usability Questionnaire. 28

Data analyses
Differences between age (in years), number of people in each insurance category (commercial, Medicare, or Medicaid), sex (male or female), and type of diabetes (T1D vs. T2D) were analyzed by using the t-test. The mean number of tele-visits by sex and differences in survey item mean scores by age was assessed by t-test or one-way analysis of variance, as appropriate. Pearson correlation coefficients were calculated to assess the associations between the number of tele-visits, the duration of all tele-visits for each participant, and average call time duration and age.
To assess the relationship between Likert scaled items and age, Pearson correlation coefficients were used. Bivariate correlation coefficients were calculated for all educational topics (number of times addressed) by age, and partial correlations were also calculated to adjust for sex and diabetes type. Statistical analyses were performed by using SPSS version 27.

Participant characteristics
A total of 1301 referrals were made for 1269 adults, 1089 with T1D and 165 with T2D. Excluded from these analyses were those with pre-diabetes (n = 7), cystic fibrosis-related diabetes (n = 4), and obesity without diabetes (n = 4). Of the 1254 participants with T1D and T2D, 982 (78.3%) had at least one visit with a remote CDCES. The others (21.7%) were unable to be reached or declined education services.
Participant characteristics are shown in Table 1. The mean age was 48.4 years (standard deviation [SD] 17.3). Approximately half of the participants were females (50.8%), and commercial insurance was the primary coverage for 48.9% of the participants. The mean HbA1c was 8.4% (SD 1.9; range 4.2-14.0%). There were 64.5% using CGM, 46.6% using insulin pumps, and 18.4% had no access to a computer, tablet, or smartphone.

Tele-visit characteristics
A total of 2203 tele-visits were conducted with 982 participants. Participants attended a mean of 2.2 televisits (SD 1.8; range 1-16 tele-visits; Table 2), and 2.5% of participants had seven or more tele-visits ( Fig. 1). The mean (SD) length of time of an individual tele-visit was 17.6 min (SD 13.8; range 1.0-158.3 min;    Table 2). Adults with T2D comprised 13.8% of the participants. Those with T2D had a greater number of televisits compared with those with T1D (mean number of tele-visits: T2D = 3.4 vs. T1D = 2.1; p < 0.001) and a longer total duration of tele-visits (T2D = 65.3 min vs. T1D = 36.6 min; p < 0.001), respectively ( Table 2). The discussions during the visits were categorized into top-ics shown in Table 4. A CDCES could discuss multiple topics with a participant within a single visit. The CGM education was discussed with 73.4% participants in a total of 1413 remote tele-visits. Preparation for provider telemedicine visits was the next most frequently discussed topic, with 58.7% of participants receiving instruction a total of 707 times.
Device download instructions were provided to 557 participants (Table 4). Other topics included prevention and treatment of hypoglycemia, sick day management, medication management, insulin injection techniques, and lifestyle support (for the full list see Table 5).

Differences by age
The mean total number of tele-visits was greater among participants with Medicare (2.8) compared with Medicaid and commercial insurance (2.2, p = 0.001 and 2.0, p < 0.001, respectively; Table 2). Those aged ‡55 years had a greater mean total number of tele-visits (2.6) compared with those aged 18-35 years (mean 1.8; p < 0.001) and those aged 36-54 years (mean 2.2; p = 0.041). The average duration of each visit in those aged ‡55 years was 20.4 min compared with those 18-35 years old (14.8 min), and those aged 36-54 years old (16.5 min), p < 0.001.
Older age was correlated with a greater number of tele-visits (r = 0.193; p < 0.001), higher average time  Older adults compared with younger adults were less likely to have access to a computer, tablet, smartphone, or internet (Table 6). Older age also correlated with a greater need for education and support related to CGM use, medication management, hyperglycemia, hypoglycemia, lifestyle support, and psychosocial support (Table 6).

Satisfaction survey results
In total, 211 surveys were collected and 198 were included for data analysis (response rate 35.5%). The mean age of responders was 52.7 years (SD 17.1); 78.0% had T1D, 56.1% were female, and 44.9% had commercial insurance. Participants had a mean HbA1c of 8.0% (SD 1.6), 64.1% used CGM, and 49.5% used an insulin pump. Participants who completed the survey had a mean of 2.7 (SD 2.2) tele-visits; the average visit was 19.5 min (SD 13.5 min), and the total average time was 54.3 min (SD 71.9 min). Surveys were excluded if: (1) the participant did not recall having a CDCES visit, (2) the survey was not fully completed, (3) the survey was a duplicate, or (4) the participant had a diagnosis other than T1D or T2D.
Although the majority of participants stated they were satisfied with the remote education visit (93.4%), those who were younger had greater satisfaction (r = 0.175; p = 0.014). Younger participants felt that the remote education visits were convenient (r = À0.167, p = 0.019) and that the remote visits saved them time (r = À0.244, p = 0.001; Table 7). Those with Medicaid were more likely to agree or strongly agree with the statement ''In the past, cost was a barrier to receiving

Discussion
Early in the pandemic, over 5 months, 78.3% of referred patients completed remote CDCES visits and most had two or more visits based on participant need for education and support. Older adults had more remote CDCES visits, the tele-visits on average lasted longer, and the total time with a CDCES was lon-ger than in younger adults. Overall, 85.7% of the individual remote visits lasted <30 min. Importantly, almost none of these CDCES tele-visits would have been billable to insurance since most insurers will only reimburse for education sessions that last for at least 30 min. However, most participants had multiple tele-visits, which in total lasted >30 min. Deficits in cognitive function are common in aging, especially in the presence of diabetes, and include impaired sustained and selective attention and working memory. [29][30][31][32] Clinically significant cognitive impairment was found in 48% of a sample of individuals with diabetes ‡60 years old. 30 More frequent, shorter sessions may be more productive and provide more  opportunities for reinforcement than longer sessions in older adults since attention tends to fade and information overload is more common with longer remote education visits. Frequent but shorter tele-visits are easier to schedule and attend than frequent in-person visits, given transportation, staffing, and other operational and social considerations. This experience suggests that Medicare and other insurers should re-evaluate their policies that preclude reimbursement for individual DSMES visits with a duration of <30 min. We believe that shorter, more focused remote discussions can improve comprehension and self-management, but this was not tested. Studies evaluating the efficacy of shorter and longer DSMES visits are needed in older adults, but they are difficult to perform in real-world settings.
Remote CDCES visits have been used successfully to initiate and train adults in the use of CGM. 33 The CGM has been shown to reduce hypoglycemia in older adults with both T1D and insulin-treated T2D and increases the sense of safety for the PWD. [34][35][36][37] Since initiating and sustaining CGM use in people with T1D is a priority, it was requested that the remote CDCESs discuss CGM with all individuals with T1D not using CGM. It is, therefore, not surprising that CGM use was a frequent topic. We have previously shown that older adults with T1D have more barriers to starting CGM, 38 many of which can be addressed with more education and support.
We were not able to assess whether individuals started and sustained CGM use as a result of a CDCES remote visit since providers were also recommending CGM and, simultaneously, there were other policy changes that made CGM more available and affordable for many older adults. However, anecdotally, participants told providers that their sessions with the CDCESs were helpful in better understanding and using CGM.
The needs of PWD vary over time due to changes in life circumstances, the development of complications, and the availability of new technologies and medications that require education and training for proper use and optimal benefit. The Consensus report of the American Diabetes Association, the Association of Diabetes Care and Education Specialists, the Academy of Nutrition and Dietetics, the American Academy of Family Physicians, the American Academy of Physician Assistants, the American Association of Nurse Practitioners, and the American Pharmacists Association recommend DSMES services at initial diagnosis, annually, when new complicating factors affect selfmanagement and during transitions. 39 Medicare covers 10 h of DSMES in the 1st year that beneficiaries receive this service and 2 h/year for each year thereafter. 40 Although individuals usually require more education initially, there are times in subsequent years, such as when new diabetes technologies are prescribed or when complications develop, that >2 h of education per/year are needed. This is especially true for older adults who are experiencing cognitive decline, when more education is needed for both the PWD and their caregivers to ensure safety.
In the short time that many older adults interacted with the remote CDCESs, it was found that they needed more instruction related to technology when compared with those in younger age groups. Data are not available spanning a full year, but it is likely that 2 h of education time is insufficient for many older adults, especially if starting CGM or a new insulin delivery system.
All staff have the necessary technology to complete remote DSMES visits (computer with secure remote video capability). Although some PWD only have telephone access, increasingly they are obtaining access to a smartphone, tablet, or computer technology. Since our results suggest that PWD benefit from remote telephone and/or video DSMES visits as well as in-person visits, both should be equally reimbursed.
Adults with T2D had more DSMES tele-visits and longer total time was spent with the CDCESs. This is likely due to the fact that individuals with T2D were referred by their medical providers, because they identified specific education needs. In contrast, all adults with T1D and upcoming provider visits were referred without consultation with their providers. It is hypothesized that, if a selection process was used for referral of those with T1D, the total duration of tele-visits would also be longer.
Lower participant satisfaction with remote diabetes education in older adults was observed. Reasons for this may include less knowledge and comfort with using devices than those who are younger. It is possible that, over time and with more experience and use of technology, this will change. Other contributing factors were lesser availability of computers and video equipment in the home, concern for privacy, and visual and/or hearing impairments.
This report has several limitations. Remote CDCESs were only available for a limited period of time (5 months). These data are from a single center, glycemia data and quality of life data were not collected, and potential long-term benefits could not be measured.
A ''control'' (in-person) group was not possible since the use of the remote CDCESs was triggered by the start of the COVID-19 pandemic and the inability to conduct routine in-person visits.
These data are mostly from adults with T1D or T2D on complex insulin regimens, so these results cannot be generalized to others with T2D. The remote DSMES was provided without charge so that all adults could participate without incurring a cost, acknowledging that some would not have participated if there had been a charge. The relatively small number of participants who completed the satisfaction survey was also a limitation.
There are also strengths of this article. Importantly, it is the first real-world study, to our knowledge, involving a large number of PWD (n = 1254), to present data related to the length of time needed for remote CDCES visits, and included PWD over the adult lifespan.

Conclusions
Older adults with T1D or T2D on complex insulin regimens needed a greater number and duration of remote visits. This population may do better with more frequent, shorter tele-visits. Since the majority of the remote visits lasted <30 min, these services would not have been reimbursed under current insurance policies. Changes in reimbursement for needed remote education and support services should be considered. Reimbursement should be based on cumulative time rather than by time spent in an individual visit.
The level of satisfaction was high with diabetes education tele-visits, supporting this method of intervention. However, although there are many benefits associated with the use of remote visits, there are disparities in the ability to access these services. Further studies are needed to examine the adaptations and approaches that could improve the use of telehealth for older adults.