Dr. Gillespie is a recipient of a Geriatric Academic Career Award administered by the Bureau of Health Professions, Health Resources and Services Administration, U.S. Department of Health and Human Services.
Original Research Contribution
Potential of Telemedicine to Provide Acute Medical Care for Adults in Senior Living Communities
Article first published online: 13 FEB 2013
© 2013 by the Society for Academic Emergency Medicine
Academic Emergency Medicine
Volume 20, Issue 2, pages 162–168, February 2013
How to Cite
ACADEMIC EMERGENCY MEDICINE 2013; 20, 162–168 © 2013 by the Society for Academic Emergency Medicine
The authors have no potential conflicts of interest to disclose.
- Issue published online: 13 FEB 2013
- Article first published online: 13 FEB 2013
- Manuscript Accepted: 12 SEP 2012
- Manuscript Revised: 4 AUG 2012
- Manuscript Received: 24 MAR 2012
Older adults dwelling in senior living communities (SLCs) often experience barriers to medical care when they experience acute illness. The potential of telemedicine as a substitute for standard routes of evaluating and caring for individuals with acute illness (e.g., in-person or telephone-based interactions with primary care providers and emergency department [ED] visits) was explored in this study.
In this cross-sectional, observational study, the authors conducted a 6-month retrospective review of the medical records of adults enrolled in a university-affiliated geriatrics practice that offers on-site primary medical care in SLCs. For each episode of acute care, patient demographics, medical history, and chief complaint were collected and presented to an expert panel of physicians, who determined whether telemedicine could have been used to provide acute evaluation and care. The care actually provided, including outcomes, was also noted. Descriptive statistics were used to characterize the population and potential for telemedicine care.
The medical records of 646 patients were reviewed, accounting for 1,535 unique episodes of acute care. The expert panel identified 576 visits (38%) as potentially appropriate for telemedicine-based acute care, with 38, 47, and 27% of phone, in-home, and ED visits being eligible, respectively. Chief complaints most likely to be deemed potentially appropriate were falls and dermatologic, respiratory, and gastrointestinal illnesses, representing 58% of visits identified for telemedicine-based acute care.
Telemedicine has a potentially significant role in the provision of acute care for older adults residing in SLCs. Studies are needed to evaluate the feasibility, acceptability, effectiveness, and efficiency of acute care telemedicine for this population.
El Potencial de la Telemedicina para Proporcionar Atención Médica Aguda a Adultos en Comunidades Residenciales para Mayores
Los ancianos en comunidades residenciales para mayores (CRM) a menudo experimentan barreras a la atención médica cuando ellos sufren una enfermedad aguda. Se exploró en este estudio el potencial de la telemedicina como un sustituto de los mecanismos estándar de evaluación y atención a los individuos con enfermedad aguda (ej: interacciones personales o telefónicas con los médicos de atención primaria (MAP) y visitas a los servicios de urgencias (SU)).
En este estudio observacional transversal, se llevó a cabo una revisión retrospectiva de 6 meses de las historias médicas de los adultos en un centro geriátrico afiliado a la universidad que ofrece atención primaria en CRM. Para cada episodio de atención aguda, se recogieron las características demográficas, la historia médica y la queja principal y se presentaron a un panel de médicos expertos, que determinaron si la telemedicina podía haberse usado para proporcionar la evaluación y la atención aguda. La asistencia proporcionada, incluyendo los resultados, también se anotó. La estadística descriptiva se utilizó para caracterizar la población y la atención potencial para la telemedicina.
Se revisaron las historias médicas de 646 pacientes, y se contabilizaron 1.535 episodios únicos de atención aguda. El panel de expertos identificó 576 visitas (38%) como potencialmente apropiadas para la atención aguda mediante telemedicina, y fueron elegibles un 38%, 47% y 27% de llamadas, y visitas al domicilio y a urgencias, respectivamente. Las quejas principales más probablemente consideradas apropiadas fueron las caídas y las enfermedades dermatológicas, respiratorias y gastrointestinales, que representaron el 58% de las visitas identificadas para atención aguda mediante telemedicina.
La telemedicina tiene un papel potencialmente significativo en la provisión de atención aguda a los ancianos que residen en una CRM. Se necesitan estudios para evaluar la viabilidad, la aceptabilidad, la efectividad y la eficiencia de la atención aguda mediante telemedicina para esta población.
The population of older adults (age > 65 years) in the United States is expected to double by 2030, thus increasing the demand for acute medical care by older adults. Senior living communities (SLCs), which include assisted living and independent living facilities, are home to a large number of older adults who need access to acute illness care. The exact number of residents in SLCs is unknown; however, in 2006 an estimated 1 million older adults lived in assisted living facilities and even more resided in independent living facilities.
Individuals living in SLCs seek medical care for acute illnesses frequently, and those living in assisted living facilities access medical care at rates greater than the general older adult population.[3-5] Unfortunately, older adult patients face numerous barriers to accessing acute care. Same-day evaluation by a patient's primary care provider (PCP) is often unavailable, likely because of the well-documented shortage of PCPs and geriatricians in the United States.[6-9] Even when urgent PCP appointments are available, transportation remains a major obstacle for older adults, many of whom do not drive, have difficulties accessing public transportation, and cannot take an ambulance to office visits. These barriers are further complicated by the high prevalence of functional and cognitive impairments within this population.
PCPs are then left with limited options, including evaluating a patient with an acute illness by phone, delaying care until an in-person visit is possible, or sending the patient to an emergency department (ED) via ambulance. Older adults seek medical care in EDs at a rate four times greater than younger populations, resulting in great cost.[11, 12] However, ED care may not be the optimal location to evaluate and treat patients from SLCs with nonemergent complaints.[13, 14] ED-based practitioners often lack sufficiently detailed patient information, such as health histories, medication lists, and baseline functional status. Institutional challenges such as the absence of social workers and patient boarding further limit the ability to deliver high-quality, patient-centered care.[13, 14]
New and innovative care programs are needed to deliver medical care to the growing population of older adults. Telemedicine, which refers to using health information technology for clinical care when distance and time separate the patient and provider, may offer such a program. This service varies in terms of intensity, ranging from telemonitoring systems to video and audio evaluation. The interactions between the patient and the provider can be asynchronous or synchronous. High-intensity telemedicine, which enables physicians to perform a full examination at a distance, has significantly increased access to medical care for acute illness in pediatric populations by enabling PCPs to care for patients without bringing them to the providers (thus eliminating transportation barriers). Studies of telemedicine for acute illnesses in children in schools and daycare facilities have found it to be highly feasible, acceptable, effective, and cost-effective.[15-18] Preliminary studies involving nursing home residents and other adult populations have also demonstrated the feasibility and acceptability of telemedicine, including its use for acute illness, and these studies have revealed enhancements in clinical care.[19-29] Thus, telemedicine may have significant value in the SLC setting; however, to our knowledge, no studies have evaluated this potential.
This aim of this study was to quantify the number, proportion, and characteristics of acute care episodes experienced by patients of an SLC-based geriatrics practice for which high-intensity telemedicine care could have been substituted. Through this study, we hoped to quantify and characterize the potential demand for these high-intensity telemedicine services to inform SLC-focused acute care telemedicine programs that can provide rapid, high-quality care for acute illnesses.
For this cross-sectional, observational study, we performed a structured, retrospective chart review of the active medical records in a geriatrics primary care practice. The university's Research Subjects Review Board approved this study as exempt from informed consent requirements. This study is compliant with the STROBE statement criteria for cross-sectional studies.
Study Setting and Population
The study was conducted in a university-affiliated primary care geriatrics practice. This practice, through its 11 physicians and 17 advanced practice providers, delivers primary care to adults residing in 19 SLC facilities. The providers hold regularly scheduled, on-site, primary care clinics at each facility. The frequency of the clinics is scheduled based on patient need. The practitioners provide care for the patients only while they reside in the SLC and only at the facility. Clinical coordination and administrative support are provided to the practice from a central office at a location separate from the SLCs. Paper-based medical records for the practice are housed in this office.
Patients were included in this study if they were being cared for by the practice's providers at any time between October 1, 2008, and March 31, 2009. Records for all requests for urgent or emergent care, as documented in the medical record during the study period, were abstracted.
To maximize the quality of the medical record abstraction, the recommendations of Gilbert et al. were used as much as possible. Standard abstraction forms were developed and revised to maximize usability and minimize ambiguity. Variables were defined, and the location of the information in the chart was identified. The study coordinator (RM), a medical student, performed the primary record review and abstraction, with the assistance of two research assistants who were trained by the study coordinator. During the study coordinator's training process, another investigator (MNS) re-reviewed the charts also to ensure accuracy. During the research assistants' training process, the study coordinator re-reviewed the charts to ensure accuracy. In both cases, this re-review process continued until the records were abstracted consistently and accurately. However, no formal test of interrater agreement was performed. Regular meetings occurred to resolve questions in abstraction. We were unable to blind the reviewers to the study objectives.
Data collected included patient demographics (patient age, sex, race, and facility of residence), medication list, and medical history. The following information for each episode of acute care during the study period was collected: mode of care provision (in an ED, by the PCP, or by phone), chief complaint(s), advance care directives (e.g., do not hospitalize, do not attempt resuscitation), length of symptoms, vital signs (if available), testing, diagnosis, treatment plan, and patient disposition. Records of acute care delivered outside the practice, such as in an ED, were also abstracted.
All episodes of urgent and emergent care were evaluated for telemedicine eligibility by a three-physician panel consisting of an emergency physician (MNS) and two geriatricians (DN, SMG). The panel members were selected based on their expertise in the care setting and patient population. The panel was provided an abstract of each acute care episode, including the information described above. The full medical record was also available, if needed, by the panel.
The panel first reviewed all episodes of care and determined whether each episode represented an initial contact for the illness or a follow-up contact. For instance, a patient might have called seeking relief from a sore throat (the initial contact) and then called back 2 days later to report continuing pain (follow-up contact). For purposes of this study, only the initial contact was eligible for analysis because the reasons for follow-up contact were highly variable.
The panel then reviewed each eligible episode of acute care to achieve a consensus decision as to whether the patient could have been triaged to a high-intensity telemedicine program for evaluation and care. The panel was told to consider high-intensity telemedicine (Table 1) as the available model. In this model, a technician (trained at approximately the emergency medical technician level) travels to the patient with the necessary equipment. The technician determines the history of present illness, confirms the patient's medication list, performs a clinical examination, and obtains laboratory testing as directed by protocols and the provider. Thus, from the telemedicine visit, the provider could obtain a history; vital signs; skin, ear, and throat images; pictures and video of the patient's symptoms; and laboratory results (e.g., blood work, urinalysis, and a 12-lead electrocardiogram). Additionally, the provider could video conference with the patient to obtain additional information.
|Communication with patient|
|• Telephone or videoconferencing|
|• Store-and-forward and real-time|
|• Protocol-driven data collection by trained technician|
|• Access all resources, including caregivers and facility staff|
|• Perform medication reconciliation|
|• Images (e.g., skin, tympanic membrane, throat, edema)|
|• Video (e.g., respiratory effort, gait, joint range of motion)|
|• Audio (e.g., lung, heart, bowel sounds)|
|• Pulse oximetry|
|• Limited point-of-care testing (e.g., urinalysis, rapid strep)|
|• Collaboration with mobile radiology|
|Age (yr), median (IQR)||85 (77–89)|
|Sex, female||456 (71)|
|Assisted living||525 (81)|
|Independent living||116 (18)|
|Black or African American||23 (4)|
|Dementia/cognitive impairment||369 (57)|
|Cardiovascular (includes CAD, MI, CHF)||278 (43)|
|Gastrointestinal diseases (nonliver diseases)||175 (27)|
|Pulmonary disease (includes asthma, COPD)||148 (23)|
|Renal disease||119 (18)|
|Liver diseases||8 (1)|
|Medications taken, median (IQR)||10 (8–13)|
|Activities of daily living deficiencies (n = 241, 37.3%)|
|Mini-Mental State Exam (n = 471, 72.9%)|
|Score, median (IQR)||25 (19–28)|
|Characteristic||Appropriate for Telemedicine (n = 576), n (%, 95% CI)||Not Appropriate for Telemedicine (n = 959), n (%, 95% CI)|
|How the acute illness was managed by the practice|
|Handled by telephone||388 (67, 64–71)||625 (65, 62–68)|
|Handled by PCP in person||107 (19, 15–22)||119 (12, 10–14)|
|Sent to an ED||81 (14, 11–17)||215 (22, 20–25)|
|Outcome after initial management|
|Cases handled by telephone|
|Managed by phone||222 (57, 52–62)||443 (71, 67–74)|
|Home visit set up (nonemergent)||13 (3, 2–5)||20 (3, 2–5)|
|Home visit set up (emergent)||117 (30, 26–35)||89 (14, 12–17)|
|Specialist visit set up||2 (0.5, 0–1)||12 (2, 1–3)|
|Direct admission to hospital||0 (0, N/A)||1 (0.2, 0–0.5)|
|Sent to an ED||24 (6, 4–9)||43 (7, 5–9)|
|Unknown||10 (3, 1–4)||19 (3, 2–4)|
|Handled by PCP in person|
|Managed during home visit||103 (96, 93–100)||111 (93, 89–98)|
|Specialist visit set up||1 (0.9, 0–3)||3 (3, 0–5)|
|Direct admission to hospital||0 (0, N/A)||1 (0.8, 0–2)|
|Sent to an ED||2 (2, 0–4)||1 (0.8, 0–2)|
|Unknown||1 (0.9, 0–3)||3 (3, 0–5)|
|Sent to an ED|
|Treated and released from an ED||39 (48, 37–59)||110 (51, 44–58)|
|Admitted to an inpatient unit||34 (42, 31–53)||90 (42, 35–48)|
|Died||0 (0, N/A)||2 (0.9, 0–2)|
|Other||0 (0, N/A)||1 (0.5, 0–1)|
|Unknown||8 (10, 3–16%)||12 (6, 3–9)|
|Fall||113 (20, 16–23)||212 (22, 19–25)|
|Dermatologic condition||91 (16, 13–19)||123 (13, 11–15)|
|Respiratory illness||67 (12, 9–14)||100 (10, 8–12)|
|Gastrointestinal illness||57 (10, 7–12)||93 (10, 8–12)|
|Musculoskeletal (not fall related)||45 (8, 6–10)||88 (10, 7–11)|
|Behavioral||49 (9, 6–11)||74 (8, 6–9)|
|Head/eyes/ears/nose/throat||34 (6, 4–8)||64 (7, 5–8)|
|Cardiovascular (includes presyncope/syncope)||34 (6, 4–8)||56 (6, 4–7)|
|Genitourinary||27 (6, 4–8)||37 (4, 3–5)|
|Other||28 (5, 3–7)||44 (5, 3–6)|
|Behavioral/possible infectious etiology||17 (3, 2–4)||26 (3, 2–4)|
|Lower extremity swelling||11 (2, 1–3)||27 (3, 2–4)|
|Neurologic||3 (1, 0–1)||15 (2, 1–2)|
|Advance Care Directive Statusa||Telemedicine Eligible (n = 474), n (%, 95% CI)||Not Telemedicine Eligible (n = 781), n (%, 95% CI)|
|Do not attempt resuscitation||376 (79, 76–83)||623 (80, 77–83)|
|Comfort care/do not hospitalize||38 (8, 6–10)||39 (5, 3–7)|
|Full code||98 (21, 17–24)||153 (20, 17–22)|
When deciding whether the patient could have been triaged to a high-intensity telemedicine program, the panel was told to consider the patient's advance directives (e.g., full code, comfort care, do not hospitalize), chief complaint, baseline health information, and other information (e.g., vital signs). Furthermore, the panel was told to consider the potential clinical conditions and risks to the patients, immediacy in which care was needed (e.g., care needed now vs. care can wait a few days), and the capabilities of the high-intensity telemedicine program in relation to the problem (e.g., ability to image the skin, inability to repair a laceration; Figure 1). If the panel members could not reach consensus, even after discussion, then the episode was considered ineligible for telemedicine.
The data were characterized using descriptive statistics, including 95% confidence intervals (CIs). Data were analyzed using SAS 9.2 (SAS Institute Inc., Cary, NC).
During the study period, 681 individuals were patients of the geriatrics group practice. Thirty-five medical records were unavailable for review, leaving 646 (95%) available for review and abstraction. SLC-based patients were usually female and disproportionately white, had many medical conditions, took many medications, and had few functional limitations. The majority of patients resided in assisted living facilities (Table 2).
These 646 patients accounted for 1,535 unique, initial episodes of acute care during the study period (Table 3). The PCPs handled 66% of the episodes by telephone, sent 19% to an ED for care, and arranged urgent PCP appointments for the remaining 15%.
The panel reached a consensus decision regarding telemedicine appropriateness for all acute episodes of care. Thirty-eight percent (n = 576) of the acute care episodes were identified as appropriate for high-intensity telemedicine. Table 3 shows how the practice handled the cases felt to be eligible for telemedicine, and the disposition of those cases. Of note, 388 cases (67% of the potential telemedicine cases) handled by telephone were deemed appropriate for telemedicine care. In contrast, 107 episodes (19%) that were managed by PCP evaluation at a SLC, and 81 (14%) acute care issues that were managed in an ED, were identified as eligible for telemedicine care.
Falls were the most common acute complaint, accounting for 325 episodes of care (21%) during the 6-month study period. Of these 325 encounters, 35% were suitable for telemedicine visits. Dermatologic (214, 14%) and respiratory (167, 11%) complaints were other major contributors to the total number of visits and were deemed telemedicine-eligible 43 and 40% of the time, respectively. Table 4 demonstrates the telemedicine eligibility, as stratified by advance care directive status. Care was provided in an ED 14 times for patients with advance care directives requesting no hospitalization. Half of these visits could have been supplanted by telemedicine.
To our knowledge, this is the first study to quantify the potential role of telemedicine in providing acute illness care for older adults residing in SLCs. Our high-intensity model of telemedicine has the advantage of using trained technicians at approximately the emergency medical technician level. Such a technician can obtain historical information from the patient and his/her caregivers; collect clinical examination findings such as vital signs, images, video, and audio; perform point-of-care testing; electronically document all findings; and package that information for providers to interpret and act upon (Table 1). The fact that 38% of acute illness care provided to older adults in a geriatrics SLC practice was felt to be appropriate for telemedicine care demonstrates significant potential for telemedicine as a mechanism to deliver care.
The expert panel in this study felt that 388 acute illness cases that were handled by telephone would be eligible for telemedicine. It is not surprising that this group constituted such a large proportion of the cases eligible for telemedicine (67%) and a significant proportion of cases handled by telephone (38%). Given the limited number of health care providers, the barriers to medical care experienced by the elderly, and the need to treat minor health issues in this population, many conditions are initially evaluated and treated by telephone.[9, 31] In these situations, the PCP must rely on information from the patient and his or her caregivers. Unfortunately, this information is often unreliable because of patients' limited understanding of medical issues and impaired cognitive ability, and these impediments to communication may affect formulation of the treatment plan. However, the involvement of trained telemedicine technicians for the collection of data for the provider can improve the quality of the information.
Our panel concluded that almost half of the acute illness cases handled in urgent PCP visits could also have been handled by telemedicine. Although we did not collect information regarding the delay between the call reporting the illness and the actual PCP visit, or the burden that the emergent visit placed on the PCPs, it is reasonable to hypothesize that primary care practices will realize increased efficiency in acute care delivery in a telemedicine-enhanced practice model. A high-intensity telemedicine program allows providers to evaluate patients with acute complaints without requiring deployment of medical providers to multiple SLCs, thus decreasing the time and effort needed for each encounter.
Of patients who went to an ED for care, our panel felt that a notable 27% would have been eligible for telemedicine care. Although we do not have specific information regarding the reasons patients went to an ED, we can hypothesize that they lacked sufficient access to care (e.g., their PCPs were unavailable and/or had insufficient data to comfortably treat the patients over the phone). Telemedicine addresses this access barrier by giving each patient an additional resource for accessing the health care system. Telemedicine has particular advantages over ED care. When going to an ED for care, patients must travel, which is, at minimum, an inconvenience. Patients are placed in an unfamiliar and noisy environment, which puts older adults at risk of developing delirium. An ED physician often has incomplete information about the patient's history, which can lead to complications, including medication errors, duplicated tests, and fragmented and confusing instructions for the patient.[33-35] Finally, ambulance transport and ED care are expensive.
Another important consideration, particularly in this population, is the goal of care. Some of these patients have stated that they wish to focus on comfort rather than curative therapies. They do not want to travel outside their homes for medical care, particularly to hospitals. For these patients, telemedicine is uniquely adapted to meeting their goals, as it has the potential to offer timely, patient-centered medical treatment and offers an alternative to a trip to an ED. Investigation of the benefits of high-intensity telemedicine deployment for the care of hospice patients is needed.
This study also demonstrates an important finding associated with new technologies: although technologic advances decrease the use of some health care options, they may increase the use of others; in other words, they replace inexpensive telephone management of disease with telemedicine visits. Further work is needed in the elderly population, as it has been performed in children, to demonstrate the comparative effectiveness of telemedicine for the care of those with acute illnesses.
As with any study, this preliminary work has a number of limitations. Because only one SLC-based primary care geriatrics practice was involved, our findings may not be generalizable to other practices or locations. Further, it is unlikely to be generalizable to the older adult population at large. However, given the absence of previous research in this field, and given the purposes of our study, our findings are important and validate the idea that telemedicine may be able to provide a significant volume of acute geriatrics care. It is also possible that some acute care visits were missed due to limitations in the documentation in the medical records reviewed. Thus, our estimate of the number of cases likely serves as a minimum that may be appropriate for telemedicine. However, we do not expect that these cases would be missed in a biased fashion; therefore, the percentages and rates of telemedicine care should be accurate. Additionally, the panelists did not review the episodes of acute care and come to a decision about telemedicine appropriateness prior to meeting as a group. It is possible that one panel member could have biased the opinions of the other members; however, the panel members monitored for this bias and did not feel that any member was particularly dominant. Finally, this study does not demonstrate the actual feasibility, acceptability, effectiveness, or efficiency of telemedicine for acute geriatrics care.
Older adults residing in senior living centers have a substantial volume of acute illness episodes that could be managed through a high-intensity telemedicine program. Further work is necessary to demonstrate the feasibility, acceptability, effectiveness, and efficiency of telemedicine for these patients.
The manuscript was copyedited by Linda J. Kesselring, MS.
- 1The growing burden of chronic disease in America. Public Health Rep. 2004; 119:263–70., .
- 2Residential Care and Assisted Living: State Oversight Practices and State Information Available to Consumers. AHRQ Publication No. 06-M051-EF. Rockville, MD: Agency for Healthcare Research and Quality, 2006.
- 3A National Study of Assisted Living for the Frail Elderly: Results of a National Survey of Facilities. Department of Health and Human Services. Available at: http://aspe.hhs.gov/daltcp/reports/facres.htm. Accessed Nov 19, 2012., ,
- 8Institute of Medicine. Retooling for an Aging America. Washington, DC: The National Academies Press, 2008.
- 9Economic and demographic trends signal an impending physician shortage. Health Affairs. 2002; 21:140–54., , , .
- 11The epidemiology of emergency medical services use by older adults: an analysis of the National Hospital Ambulatory Medical Care Survey. Acad Emerg Med. 2007; 14:441–7., , , et al.
- 12National Hospital Ambulatory Medical Care Survey: 2004 emergency department summary. Adv Data. 2006; 372:1–29., .
- 13Characteristics and acute care use patterns of patients in a senior living community medical practice. J Am Med Dir Assoc. 2012; 13:260–3., , , , .
- 18Potential to Reduce Healthcare Costs by Replacing Emergency Department With Telemedicine Visits. Platform Presentation, Pediatric Academic Societies' Annual Meeting, Honolulu, May 2008., , , ,
- 24Clinicians' and patients' experiences and satisfaction with unscheduled, nighttime, internet-based video conferencing for assessing acute medical problems in a nursing facility. AMIA Annu Symp Proc. 2003; 2003: 709–13., , , et al.
- 25Effect on telehealth on use of secondary care and mortality: findings from the Whole System Demonstrator cluster randomized trial. BMJ. 2012; 344:e3874., , , et al.
- 26Patient satisfaction with telemedicine: a comparison study of geriatric patients. J Telemed Telecare. 2001; 7(Suppl 2):85–6., .
- 28Therapist adherence and competence with manualized cognitive-behavioral therapy for PTSD delivered via videoconferencing technology. Behav Mod. 2007; 34:856–66., , , , , .
- 31Telephone Medicine: A Guide for the Practicing Physician. Philadelphia, PA: American College of Physicians – American Society of Internal Medicine; 2002., (eds).
- 36United States Government Accountability Office. Ambulance Providers: Costs and Expected Medicare Margins Vary Greatly. GAO Report 07-383. Washington, DC: GAO, May 23, 2007.