Studies are documenting reliability of diagnoses
Early indications are positiveThe jury is still out on how much wound care can be provided through telemedicine, but preliminary indications are optimistic.
For instance, in a British investigation, 79 patients with pigmented lesions were referred from a general medical clinic to a dermatologist. After reading the referral letters, the dermatologist judged that 35 of the patients required appointments "soon" or "urgently." When the dermatologist reviewed the letters and still video images, he concluded that only 12 patients needed to be seen soon or urgently. And when the dermatologist examined the patients in person, he deemed that 10 patients required an appointment soon or urgently.1
In a preliminary study at the Medical College of Georgia School of Medicine in Augusta, 22 patients were evaluated both in person and by telemedicine consultation. Researchers found 100% agreement between the corresponding diagnoses. Similar success was reported by researchers in Norway.1
In yet another study, conducted at the University of East Carolina in Greenville, NC, 60 patients with 78 skin lesions were referred to the dermatology clinic. All were evaluated via a live interactive telemedicine system and an in-person evaluation. The consistency rate between diagnoses made by the two methods of examination was 80%. An additional 5% of the diagnoses were interpreted as being the same, but the dermatologists called the diseases by slightly different names.
If, in general, diagnoses ascertained via telemedicine are found to be comparable with those obtained during direct observation, telemedicine could spare patients a tremendous inconvenience while simultaneously reducing costs.
A reasonable expectation would be decreased numbers of face-to-face physician-patient consultations and the prevention of unnecessary trips to the emergency room and subsequent hospital admissions, says researcher Rashid Bashshur, PhD, professor of health management and policy at the University of Michigan in Ann Arbor. "We’re not at the point of advocating a massive adoption of telemedicine. We’re taking things one step at a time — first determining clinical reliability, then costs and benefits," he says.
Putting telemedicine to the test
Bashshur is coordinating a pilot study comparing the efficacy and accuracy of telemedicine evaluations with face-to-face meetings for patients with pressure ulcers. Study subjects undergo both personal and telemedicine examinations by different dermatologists, each of whom issues a diagnosis. The corresponding results are then compared for do see if the diagnoses are in agreement.
Telemedicine for dermatology is also under study at the University of Miami in Florida. In one protocol, dermatologists are using both still images and in-person examinations to make diagnoses, much the same as in the University of Michigan study, says Anne Burdick, MD, PhD, associate professor of dermatology at the University of Miami School of Medicine.
"We’re looking at the concordance of telemedicine diagnoses with in-person diagnoses," Burdick explains. "If we can show that [telemedicine] is just as accurate, it might push more insurance companies to consider this a valid method."
She notes that the Health Care Financing Administration pays for telepathology and teleradiol-ogy, but not telemedicine exchanges requiring physician-patient interaction.
Still images work well for some specialties
In the University of Miami study, dermatologists examine skin lesions on patients at an AIDS clinic on the campus. Each diagnosis is entered into the patient’s log. Using the Picasso Still Image Phone technology developed by AT&T, the dermatologist sends images of the patient’s skin lesions to Burdick or another dermatologist participating in the study, who also can interview patients as the images are received. Burdick or a colleague will make a second diagnosis based on the telemedicine examination, and the two diagnoses will be compared.
Though some telemedicine systems are capable of recording and transmitting real-time, full-motion video, Burdick notes that still images are probably adequate for specialties such as dermatology and ophthalmology, where movement is not an essential factor for making a diagnosis. "But if a neurologist is looking at a Parkinson’s tremor or during psychiatric interactions, you certainly need movement and eye contact." Smell and touch cannot yet be incorporated into telemedicine, though engineers are experimenting with methods to fill in those gaps.
1. Burdick AE, Berman B. Teledermatology. Adv in Dermatol 1997; 12:19-46.