Advances in communications technology now make it possible for an expert in one location to assess a diabetic foot ulcer in a patient who is miles away. But telehealth efforts face a number of hurdles, the most daunting of which may be patient privacy.
By Emily Delzell
After debriding a potentially devastating diabetic foot ulcer, surgeons from the VA Medical Center Pittsburgh can check on their patients across town at a long-term care facility to see how healing is progressing—and neither the doctor nor the patent has to travel. This postoperative vascular telehealth clinic, conducted using video, is just one example of how telemedicine is delivering integrated, specialized medical care without limitations of physical distance.
This evolving technology also comes with concerns and challenges, including issues of widespread implementation, costs and reimbursement, and patient privacy, as well as questions regarding the quality of data transmitted.
“The ultimate role telehealth plays in the future of US medicine will be determined in large part by what kind of a healthcare system is designed,” said endocrinologist Frederick R. DeRubertis, MD, chief of medicine at the VA Medical Center Pittsburgh.
In the US, the Department of Veterans Affairs (VA) has been at the forefront of telemedical care, and DeRubertis noted that the VA’s goal for fiscal year 2012 is to have 15% of its patients involved in telemedical care, a target that is set to rise to 50% by 2014.
“I suspect that with more—and more rapid—patient-provider interaction we will see both short- and long-term payoffs in human and economic costs,” DeRubertis said. “If, for example, ulcers can be detected and treated early, you can prevent amputation. Better control of blood lipids, glucose, and pressure pay off in the long-term with reduced cardiovascular mortality.”
Telemedicine, he added, may be a way to advance the improved outcomes and efficiencies promised by “patient-centered care” and “medical homes,” concepts heard frequently in healthcare today as antidotes to the fractured communication and coordination common in current medical systems.
“At the VA we call this concept a ‘patient aligned care team,’” he said. “All these terms are euphemisms for developing more integrated coordinated care systems—models in which a primary care provider is the main coordinator of care with support—meaning real-time support—from a number of specialists. Implementing such a system economically and efficiently means we almost have to use telehealth modalities to integrate care and information.”
Broadly defined, telemedicine encompasses videoconferencing, transmission of still and video images, e-health (including patient and physician portals), remote monitoring of vital signs, online continuing medical education, nursing call centers, and telephone consults.1
In diabetes care, as in most telemedicine focused on clinical care and monitoring of patients, telemedicine falls into three primary categories: store-and-forward, which involves acquisition of clinical images that are sent from one clinician to another for consultation; remote monitoring of patients for clinical indicators; and interactive approaches involving real-time communication between patients and healthcare providers.1,2-6
Telemedical care is being implemented in different ways in different nations, said Mark S. Granick, MD, chief of plastic surgery at the University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark.
Granick, who last month gave a presentation comparing telemedicine approaches in the US, Israel, and France at the 2012 Diabetic Foot Global Conference in Los Angeles, noted reimbursement is a significant issue in the US.
“In most of western Europe the physicians are salaried. In the US, this is not the case. While there are billing codes for telemedicine, they are not paid,” he said.
Virtual wound care
In 2003, Wilkins and colleagues evaluated the accuracy of a store-and-forward telemedicine system for evaluating ulcers in 70 patients (118 wounds; 99 chronic ulcers, 19 surgically repaired ulcers) seen in two different VA medical centers. Patients included inpatients and outpatients with stage II, III, and IV pressure ulcers and outpatients with diabetic foot or venous stasis ulcers.4
Six physicians worked in pairs, alternately evaluating wounds in person (up to six visits) and remotely using a telemedicine system that involved collection and database storage of digital images of the ulcer, measures of ulcer area and volume, and other ulcer and patient data.
Investigators compared agreement between in-person and remote physicians. Although agreement ranged from “fair” to “high” for diagnosing cellulitis, necrosis, and healing (percentage agreement between telemedical assessments and all physician visits ranged from 57.3% to 95.7%), it was not significantly less than interphysician agreement on in-person assessments, and the authors recommended using such a system to increase access to specialized wound care.
Widespread use of telewound care is still not a reality, however, said Julie C. Lowery, PhD, associate director of Health Services Research & Development at the VA HSR&D Center of Excellence, Ann Arbor, MI.
“Some sites, such as ours, are doing wound care teleconsultations on an informal basis—i.e., primary care providers at sites without wound care expertise are e-mailing digital images of wounds to a wound care expert at another site seeking treatment advice,” said Lowery, coauthor of the 2003 study2 and other studies on the use of telemedicine for diabetes care.3,4
Granick and colleagues recently completed a study7 comparing accuracy of on-site wound evaluation with that of remotely viewed images. They found what he called a “surprisingly good correlation” between the two modalities.
He noted the biggest challenge in the US to implementation of telemedical wound care is patient privacy.
“It is quite easy, for example, to use your iPhone to see a patient’s wound and talk directly to them. That information is not secure and would be in violation of HIPAA [the Health Insurance Portability and Accountability Act],” he said.
Besides HIPAA concerns, Granick said, the next largest hurdle in telemedical wound care is acquisition and use of interpretable images.
Granick and his surgical colleagues evaluated 43 inpatients and 100 consecutive outpatients on-site and remotely with store-and-forward technology.7 For the inpatient arm they used a 4.0-megapixel camera to photograph wounds and calculated agreement among on-site and remote surgeons regarding diagnosis and management.
In the outpatient arm patients were seen in busy urban clinic, where a medical student, untrained in medical photography, took images with a 2.0-megapixel camera; a surgeon performed concurrent in-person assessment. The remote surgeon received no other patient information besides the image.
Concordance between in-person and remote evaluation with images taken with the 4.0-megapixel camera ranged from 46.6% (edema) to 86.1% (necrosis) for 15 indicators. Overall, concordance for wound description was 46% to 86% and 65% to 81% for wound management.
In the outpatient arm, on-site examination and remote evaluation through store-and-forward technology showed 46% to 86% concordance for wound description and 65% to 81% concordance for wound management.
Granick noted, “We pushed the limit in our study using an untrained photographer and a low-resolution camera. We still were able to make the right assessments. However, the better the image, the easier the evaluation.”
Lowery’s research4 suggests that looking at a digital image of a wound is not the same as looking at a wound in person.
“People must be careful when using digital images to diagnose a wound and make treatment recommendations,” she said.
In a study by Kim et al (Lowery is a coauthor) on the accuracy of a web-based system for monitoring chronic wounds, investigators found the telemedicine system produced generally fair accuracy for the diagnosis of cellulitis, necrosis, or healing, and high accuracy for diagnosis of osteomyelitis.4
“We couldn’t say whether the differences between telemedicine and in-person assessments were due to technical issues or simply to variability in physician judgment,” Lowery said.
She noted, however, that remote experts can recognize these limitations and decide if: the image is of sufficient quality for accurate assessment; if better images or other patient data are needed; or if the patient should be seen in person.
“I believe digital images can reduce the need for many patients to travel to a referral facility, and we have some unpublished pilot data suggesting this is true,” she said. “To optimally use digital images for wound care, both the referring provider and consultant need to recognize technological limitations and request additional information or an in-person visit when there is uncertainty.”
In the UK and Europe investigators are developing wound assessment technology with consistent, secure images that can be used for quantitative measurements.
A 2010 study by Bowling et al5 utilized a bespoke camera-like device, computer software, and single-use disposable optical targets to produce 3D images of diabetic foot wounds. The software allows remote clinicians to pan, tilt, and zoom in on images and measure wound length, depth, volume, and surface area by marking critical wound features with a computer mouse (see figure).
The investigators had two aims: to quantify variability among physicians using the system and to examine the system’s ability to reliably reproduce visual cues from wounds and involved tissue to allow for key clinical observations.
To assess operator variability, five clinicians repeatedly took images of three wounds from different patients. One clinician measured the wound area in selected 3D images 10 times, allowing computation of the intraobserver variation due to measurement. The remaining four clinicians then measured the wounds in each of the selected 3D images.
Variation of measurement was less than 4.2%, with the exception of interobserver variation due to measurement (11.9%). Maximum intraobserver variation due to measurement (2.4%) was much lower than the corresponding maximum interobserver variation (11.9%).
“Clinicians in our study often did not agree on the exact location of the wound boundary, but could use the computer tools to mark the wounds in the images to indicate their opinion with a high degree of repeatability,” said James A. Paterson, DPhil, study coauthor and chief technical officer of Eykona Technologies, which developed the system.
To measure the system’s ability to identify clinically relevant features, clinicians evaluated 20 different wounds at two centers. A study clinician physically examined each patient and wound, and then answered 11 questions about clinical observations (investigators termed this assessment the “criterion standard”); the clinician then imaged the wound. Three other clinicians recorded their observations using only the corresponding 3D images.
In the majority of patients, concordance of remote assessment with the criterion standard was greater than 50% (accuracy ranged from 50% to 100%) for all questions except two related to whether there was evidence that debridement of the wound or wound area would improve healing. Investigators attributed lower agreement for these measures to the subjective nature of the questions.
A third question, which asked whether the wound was moist or exuding, also had relatively poor concordance (accuracy of remote assessment, 50%-80%). Researchers noted the 3D image produces a “dry” appearance and acknowledged this is as a current technical limitation.
“Much of the value in our system is being able to measure wound size across time to determine whether healing is occurring,” Paterson said. “We believe our device is easier to use and has a lower training burden than, for example, a cell phone or digital camera for the specific task of collecting quantitative data from wounds. Independent research suggests that use of a such device to quantitatively monitor the healing process ultimately saves time and money.”
Home monitoring and telemanagement
Preventing DFUs and other complications of diabetes requires significant management by physicians as well as patients, and research has evaluated the use of various home-monitoring systems to track blood glucose, blood pressure, foot health, and other measures in patients with diabetes.
In 2010 DeRubertis and colleagues reported on DiaTel, a randomized controlled trial evaluating telemedicine interventions of two different intensities in veterans with type 2 diabetes whose blood glucose levels were 8% or greater after at least one year of standard drug therapy.6 One group (n=77) received a monthly care coordination phone call from a diabetes nurse educator who asked about clinical measures and medication compliance; when the nurse felt issues warranted intervention, patients were referred to their primary care provider.
Patients in the active management plus telemonitoring group (n=73) had telemonitoring units installed in their homes. They submitted daily information through a secure network on measures such as blood glucose and blood pressure. A nurse practitioner reviewed data each day and adjusted medications. The system also delivered daily educational messages and recommendations for checking weight and foot health.
Investigators measured clinical indicators in both groups at baseline and three and six months. Compared with individuals receiving a monthly phone call, the telemonitoring plus active care group demonstrated significantly larger decreases in hemoglobin A1C levels at three months (1.7% vs 0.7%) and six months (1.7% vs 0.8%).
“The instrumentation matters less than how clinicians respond to information,” DeRubertis said. “The real difference between the two groups was that one had a dedicated midlevel provider who looked at data daily and made frequent adjustments in patients’ management.”
In the Netherlands, investigators are using a home monitoring tool to look specifically at foot health in patients with diabetes.
Bus et al developed a foot-imaging device that produces high quality images of the plantar foot that can be sent through the Internet to a central database.
In a 2010 study utilizing the system, the same four clinicians (a surgeon, two wound care specialists, and human movement scientist) made both live and repeated remote assessments for presence of ulceration, abundant callus, or absence of signs in 10 patients with diabetes type 1 or 2.8
Agreement scores between live and remote images produced by the definitive device (investigators also tested a prototype) were 100% for ulcer, 98% for abundant callus, and 97% for absence of signs.
Bus et al noted that though all clinicians reported the images allowed adequate assessment of clinical signs, they said they had some difficulty deciding whether the level of callus warranted referral.
A 2012 feasibility study by the Dutch investigators assessed the device’s use in 22 patients recruited from the outpatient clinics of two specialized diabetic foot centers.9 The patients had either type 1 or 2 diabetes and investigators classified them as being at high risk for ulceration based on the presence of peripheral neuropathy and at least one foot deformity and/or prior foot ulceration.
Patients used the device to image their feet three times a week for four months. A remote diabetic foot specialist assessed the images; if the clinician diagnosed ulcer or redness patients were referred for treatment in 24 hours; a diagnosis of abundant callus, blisters, or other signs of prelesions resulted in referral for treatment within three days. Some patients received multiple referrals during the study period.
Although the foot specialist diagnosed only three patients with foot ulcer, all referrals were justified. For the diagnosis of abundant callus, 31 of 32 referrals received treatment. Both patients and clinicians reported using the system was quick and perceived usability, based on a questionnaire, was high. Health-related quality of life among patients improved slightly, but not significantly.
Investigators concluded the device is feasible for home use and noted the next step was examining the cost-effectiveness of such a system.
VA research suggests telemedical care for conditions such as diabetes may lower overall healthcare costs.
A 2008 study evaluated the use of telemedicine to coordinate the care of veterans with chronic conditions, including diabetes. Investigators enrolled patients in a telemedical care program that included use of health informatics, home telehealth, and disease management technologies. Enrollment began in 2003 and by 2007, 31,570 veterans were taking part in the program.10
After enrollment participants demonstrated a 25% reduction in numbers of bed days of care and a 19% reduction in numbers of hospital admissions compared with these measures before the program’s implementation.10
“Telehealth can be used as a strategic tool to enhance patients’ access to specialized care, implement preventive care more widely, and, hopefully, improve quality of care and outcomes,” DeRubertis said.
1. American Telemedicine Association. Telemedicine defined. http://www.americantelemed.org/i4a/pages/index.cfm?pageid=3333. Accessed March 20, 2012.
2. Hopp FP, Hogan MM, Woodbridge PA, Lowery JC. The use of telehealth for diabetes management: a qualitative study of telehealth provider perceptions. Implement Sci 2007;2:14.
3. Kim HM, Lowery JC, Hamill JB, Wilkins EG. Patient attitudes toward a web-based system for monitoring chronic wounds. Telemed J E Health 2004;10(Suppl 2):S26-S34.
4. Kim HM, Lowery JC, Hamill JB, Wilkins EG. Accuracy of a web-based system for monitoring chronic wounds. Telemed J E Health 2003;9(2):129-140.
5. Bowling FL, King L, Paterson JA, et al. Remote assessment of diabetic foot ulcers using a novel wound imaging system. Wound Repair Regen 2011;19(1):25-30.
6. Stone RA, Rao RH, Sevick MA, et al. Active care management supported by home telemonitoring in veterans with type 2 diabetes: the DiaTel randomized controlled trial. Diabetes Care 2010;33(3):478-484.
7. Trovato MJ, Scholer AJ, Vallejo E, et al. eConsultation in plastic and reconstructive surgery. Eplasty. 2011;11:e48.
8. Bus SA, Hazenberg CE, Klein M, Van Baal JG. Assessment of foot disease in the home environment of diabetic patients using a new photographic foot imaging device. J Med Eng Technol 2010;34(1):43-50.
9. Hazenberg CE, Bus SA, Kottink AI, et al. Telemedical home-monitoring of diabetic foot disease using photographic foot imaging−a feasibility study. J Telemed Telecare 2012;18(1):32-36.
10. Darkins A, Ryan P, Kobb R, et al. Care coordination/home telehealth: the systematic implementation of health informatics, home telehealth, and disease management to support the care of veteran patients with chronic conditions. Telemed J E Health 2008;14(10):1118-1126.