Content and Style Comparison of Physician Communication in Teledermatology and In-Person Visits

Abstract

Background: The body of research is rapidly growing regarding the use of telemedicine in patient care, including cost-effectiveness, patient access, patient outcomes, etc. Less has been done describing physician communication during different aspects of the clinical visit (i.e., education, assessment, treatment, etc.) during actual versus virtual patient visits. The purpose of this study was to evaluate dermatology healthcare providers' communication via both modalities with regard to content and style. Subjects and Methods: In-person and teledermatology patient visits were observed, audio-recorded, and transcribed over an 8-month period. A content analysis was performed. Results: The Wilcoxon rank sum test was used to compare the content differences between visit modalities for each category. A p value of 0.05 was considered as significant for all tests. There were no statistically significant differences between modalities in the average number of physician words in seven of eight communication categories: small talk, clinical assessment, psychosocial issues, patient education, patient compliance, patient treatment, and administrative issues (p value range, 0.16–0.91). As well, the same communication themes occurred in each modality to essentially the same degree. For instance, assessment and discussion of treatment occurred in 100% of in-person and teledermatology visits, as did small talk. Conclusions: This research indicates that physician providers communicate with similar style and content whether using teledermatology or in-person.

Introduction

In the past decade, dramatic technological improvements have increased the quality of telehealth while decreasing the costs of utilizing the technology. Telehealth capabilities have increased access to healthcare for the elderly and for rural and urban underserved populations.^1
–3 Specifically, evidence has demonstrated that the same diagnoses are made and the same treatments are recommended using telehealth compared with in-person visits for dermatology patients.⁴ Furthermore, both patients and doctors have a positive response to videoconferencing,⁵ and, in general, patients are very satisfied with telehealth.³ However, few studies have investigated how videoconference-based healthcare visits affect the style and quantity of communication between physicians and patients. Verbal communication, a vital building block of the doctor–patient relationship, is key to successful healthcare provision whether visits are conducted in-person or via telehealth. Thus, it is important to learn whether the modality of patient encounters changes the content or manner by which doctors communicate with their patients.

Several studies have focused on doctor–patient communication in the field of telehealth. Liu et al.⁶ found that time spent in consultation during telehealth visits was significantly shorter than the time spent in face-to-face visits. However, telehealth and in-person visits were similar in the number of closed and open-ended questions and the number of times the patient and physician spoke at the same time (simultaneous utterances). Liu et al.⁶ observed that telehealth had fewer utterances in categories of empathy, praise, and facilitation and more utterances in requests for repetition. However, the doctors involved in this study had little prior experience with telehealth. Thus, Liu et al.⁶ recommended the need for telehealth training programs for doctors addressing new forms of non-verbal communication and different tactics for information exchange. Fleming et al.⁷ also recommended the integration of telehealth into primary care resident outpatient training and reinforced the need for more research on communication and telehealth. According to Onor and Misan,⁸ a well-managed clinical interview lays the groundwork for the doctor–patient relationship and suggest that the best environment for telehealth visits is one with minimal distractions.

The present study is a follow-up to a previous study¹ investigating the impact of virtual teledermatology on communication by comparing the content of in-person visits with that of virtual visits. Demiris et al.¹ found that the communication patterns of teledermatology visits were comparable to those of in-person visits. There were no statistically significant differences between the two modes of visits in the number of visits in which clinical assessment, education, treatment, compliance, psychosocial issues, and administrative issues were addressed and in the average duration of the visit. However, a statistically significant difference was found between the two modes of visits in the number that contained small talk. More telehealth visits (29.6%) included small talk than in-person visits (24%). Another finding was that the majority of teledermatology visits occurred without technical or equipment difficulty or failure.¹ The present study differed from the original study in several ways: it included 10 residents and 4 faculty physicians instead of 1 physician; for practical reasons, word counts were used for each utterance category instead of timing utterance categories; and this study also included demographic and diagnostic information about the patients. Total time of visit, total physician word counts, and “other” word counts were recorded.

The primary aim of this study was to compare the content and style of physician communication between in-person dermatology visits and teledermatology visits. The hypothesis was that there is no difference in content and style of physician communication between these two modalities. Secondary objectives were to examine the associations between patient demographics and diagnoses to mode of visit and how the presence of dermatology residents impacted physician communication.

Subjects and Methods

This study was conducted at the University of Missouri (Columbia, MO) Dermatology and Teledermatology Clinics. The Teledermatology Clinic is a long-standing, established clinic that operates in conjunction with the Missouri Telehealth Network. This cooperative network provides specialized care to underserved, rural areas in Missouri. The Missouri Telehealth Network serves over 175 remote sites in 51 Missouri counties. To date more than 29,900 interactive video encounters have been conducted by current network participants. In fiscal year 2010 over 60 medical professionals in 31 specialties conducted over 6,900 encounters via video on the Missouri Telehealth Network.⁹ Teledermatology visits were conducted during two half-days per week between the University of Missouri Dermatology Clinic and remote telehealth sites in Missouri.

One hundred one in-person and live-interactive teledermatology visits were audio-recorded over an 8-month period. A total of 10 dermatology residents and 4 dermatology faculty members participated. Days chosen for recording visits were based on the availability of study personnel, on the number of in-person or teledermatology visits, and, ultimately, on the willingness of arrived patients to participate. Because of the length of the study the sample is considered to be representative of the patient population in both in-person and live-interactive teledermatology visits. All subjects were English-speaking, were 18 years of age or over, and had decision-making capacity as determined at the time of enrollment. Prior to each visit, the patient gave his or her written informed consent to participate. In addition to the audio recording, demographic information was documented, including date, patient age, gender, new versus established (return) patient, diagnosis, and length of visit (time) by study personnel present at the time of the visit.

Four faculty dermatologists and 10 dermatology residents participated in this study. In a procedure remaining consistent with the ambulatory clinical training process of the Department, a dermatology resident initiated and conducted the first part of the visit, followed by the attending physician and resident visiting the patient together. These interactions were recorded, and physician conversation was analyzed based on predetermined content categories (Table 1). The content categories selected for coding were based on a hybrid design previously tested in a study of virtual visits in the field of home care.¹⁰ The hybrid coding design took features of the Davis Observational Code, a 20-item direct observation scale for physician–patient interactions tested extensively for reliability and validity.¹¹ This hybrid analysis protocol was also used in the previous study by Demiris et al.¹

Table 1.

Codes and Categories Used for Physician Communication

CODE	CATEGORY	DEFINITION
C1	Informal talk	Small talk including session initiation and closure greeting, discussion about the weather, current events, etc.
C2	Clinical assessment	Collection of clinical data such as past diagnosis, temperature, blood pressure, pain, skin condition, and other symptoms or patient questions
C3	Technical issues	Information not dealing with the patient directly, asking to adjust the camera or volume, asking for supplies, equipment, etc.
C4	Psychosocial issues	Discussing patient's personal problems or concerns, matters unrelated to the patient's medical condition, such as discussion about the spouse, family members, friends, etc.
C5	Education	Providing in-depth information about a specific medical condition or disease
C6	Discussion of treatment	Explanation of procedures, anticipated progress, choice of medication, dosage, or education about the medication and how to use it
C7	Promoting compliance	7.1: Reminders (for example, “Don't forget to take your medication twice a day!”)
		7.2: Identifying barriers to compliance (for example, “Tell me why you're having trouble taking your medication.”)
		7.3: Positive reinforcement (for example, “You're doing great; keep up the good work!”)
		7.4: Negative reinforcement (for example, “If you don't change the dressing, the wound could become infected.”)
C8	Administrative issues	Information dealing with the patient directly, such as the specifics of patient's access to care—scheduling a new appointment, calling prescriptions to pharmacy, scheduling or confirming upcoming in-person or virtual visit or visit to another healthcare provider, and reminders for numbers to call or persons to contact in case of a question or emergency

Each recording was transcribed into an electronic word document including patient and physician conversation. The total word count was also documented. The counted words from the transcribed documents were divided into “physician words” and “other words.” Physician words included both resident and faculty conversation. Other words included conversation by the patient, the patient's family, and the nurse or telehealth personnel, if present. Subsequently the physician conversation in each transcribed document was coded separately by two individuals. Each coder was trained and instructed on the coding procedure by the same study coordinator. Coding training consisted of familiarizing the coder with the eight codes, definitions of each content category, descriptions of the qualifications or exclusions for that category, and examples of conversation that met the category criteria. After individual coding was completed, the two coders jointly reviewed each document (patient visit) to compare their codes for each “utterance.” An utterance refers to any simple sentence, clause, or phrase. Discrepancies were resolved by a third reviewer. Once coding for a visit was finalized, the number of words occurring under each of the eight code categories was tallied. Also, visits for each modality were examined to determine what communication categories were present.

Results

Patient Demographics

Few demographic differences existed between the teledermatology group and the in-person group. Patients seen in the clinic had an average age of 50 years (range, 19–81 years; standard deviation, 17 years); teledermatology patients had an average age of 47 years (range, 18–92 years; standard deviation, 18 years). Thirty-one males and 61 females participated. In each group, female patients outnumbered males about 2:1. No significant difference was found in age (t value=0.82, p=0.41) or gender (chi-squared=0.26, p=0.61) between groups. We evaluated whether subjects were new or returning dermatology patients. There was a significant difference (chi-squared=5.06, p=0.025) in patient visit status between visit types. Approximately 73% of the in-person subjects were return patients, whereas only 50% of the teledermatology subjects were return patients.

Length of Visits

Teledermatology visits were slightly shorter than in-person visits. In-person visits averaged 25 min in length (range, 9–51 min; median, 24 min); teledermatology visits averaged 21 min in length (range, 11–51 min; median, 21 min). This mean difference in length of visit was not statistically significant (Wilcoxon rank sum 2,168.5, p=0.09).

Physician and Resident Participation

Ten residents and four faculty members took part in the study. One faculty member saw 75% of the in-person study patients; she also saw 96% of the teledermatology study patients. As one of the two faculty members who regularly do teledermatology, this faculty member sees primarily adults in the Telemedicine Clinic, whereas the other teledermatologist sees primarily pediatric patients (who were ineligible for this study). The other 25% of study visits were attended by two other faculty members who do not regularly do teledermatology. Residents participated in all in-person and live-interactive telemedicine visits. Resident and physician conversations for each visit were combined as “physician communication.”

Physician Communication

Seven of eight different communication categories (described in Subjects and Methods) were evaluated between groups. The results of Box's M test (mean=74, F=3.3, significance <0.0001) did not support the homogeneity of covariance matrices for mutivariable analysis of variance. The Wilcoxon rank sum test was used to compare the content differences between visit modalities for each category. A p value of 0.05 was considered as significant for all tests (Table 2). There were no significant differences between the two groups in any of seven components (p value range, 0.16–0.91). Even though there were “technical issues” (C3) in both modalities, this component area was not statistically comparable because the technical issues during the teledermatology visits entailed dealings with camera or lighting. Discussion of technical issues (C3) occurred in 95% of the in-person visits and 100% of the teledermatology visits. An average of 3% of physician communication (words) in in-person visits dealt with technical aspects, primarily in requesting nursing assistance; in teledermatology visits this increased to 8% because of discussion of camera and lighting issues as well.

Table 2.

Wilcoxon Rank Sums and p Values

CODE	DESCRIPTION	WILCOXON RANK SUM	P VALUE
C1	Informal talk	2,028.5	0.55655
C2	Clinical assessment	1,938.5	0.91261
C4	Psychosocial issues	1,914.5	0.39924
C5	Education	2,048.5	0.45447
C6	Discussion of treatment	2,008.0	0.66923
C7	Promoting compliance	2,050.0	0.42810
C7.1	Reminders	2,054.0	0.22066
C7.2	Identifying barriers	1,855.0	0.13818
C7.3	Positive reinforcement	2,189.5	0.03966
C7.4	Negative reinforcement	1,959.0	0.90289
C8	Administrative issues	1,774.0	0.16172

The largest proportions of physician communication concentrated on treatment (C6) and assessment (C2) for both visit modalities (Table 3). Assessment and treatment conversation combined made up 69% and 67% of average physician words during in-person visits and teledermatology visits, respectively. Informal talk (C1) and education (C5) consisted of the next largest conversation components; however, informal talk had more words (was “longer”) than patient education, on average, during in-person visits, whereas education was “longer” than informal conversation during teledermatology visits. Overall, during in-person visits, the physician communication (faculty and residents) made up 59% of the entire visit; in teledermatology visits it made up 65% of the visit communication (i.e., all conversation including patient, family, nurses, etc.).

Table 3.

Words by Category for In-Person and Teledermatology Visits

VARIABLE	MEAN	PERCENTAGE OF MD WORDS	SD	MEDIAN
Informal talk	149/97	11/7	154/49	95/88
Clinical assessment	353/373	26/28	177/174	349/342
Technical issues	40/104	3/8	28/65	33/89
Psychosocial issues	1/4	<1/<1	7/17	0/0
Education	121/110	9/8	108/127	107/60
Discussion of treatment	592/521	43.5/39	425/306	473/499
Promoting compliance (sum)	14/13	1/1	21/25	6/1
Reminders	3/2	<1/<1	7/6	0/0
Identifying barriers	2/6	<1/<1	10/22	0/0
Positive reinforcement	8/3	<1/<1	12/8	2/0
Negative reinforcement	1/2	<1/<1	7/10	0/0
Administrative issues	89/104	6.5/8	74/71	73/96

Data are in-person/teledermatology results.

MD, physician; SD, standard deviation.

Physicians ensured compliance by (a) giving reminders (C7.1), (b) discussing barriers to compliance (C7.2), (c) giving positive reinforcement (C7.3), and (d) giving negative reinforcement (C7.4). The Wilcoxon rank sum test was also used to compare the differences in the C7 subcategories. The results indicated that there were no statistically significant differences except for “positive reinforcement” (C7.3) (p=0.0366). Significantly more positive reinforcement was given during in-person visits than in teledermatology visits. During in-person visits, ensuring compliance was predominantly in the form of positive reinforcement, but during teledermatology visits, it was predominantly in the form of identifying barriers. Ensuring compliance was the second smallest portion of communication (average of 13 and 14 words per visit for teledermatology and in-person visits, respectively) after psychosocial conversation. The code category that held the fifth place in amount of physician conversation for both visits was “administrative” (C8). “Other words” were the words spoken by anyone other than a physician; this included the patient, nurses, family members accompanying the patient, telehealth personnel, and/or study personnel. These words in addition to the physician words make up the total number of words spoken during a visit (Table 4). Significantly fewer “other words” were spoken, on average, during teledermatology as opposed to in-person visits (p=0.01).

Table 4.

Physician Versus Other Communication During Visits

	MEAN		SD		PERCENTAGE OF TOTAL WORDS
VARIABLE	IP	TD	IP	TD	IP	TD
MD words	1,360	1,325	521	465	59%	65%
Other words	942	723	484	397	41%	35%
Total words	2,302	2,048	879	730

IP, in-person; MD, physician; SD, standard deviation; TD, teledermatology.

Content analysis also included where the categories of communication occurred in the two visit modalities (Table 5). Clinical assessment, treatment discussion, and informal talk took place in all visits. Administrative issues and technical issues were discussed in almost all visits of both modalities. Education, ensuring compliance, and psychosocial issues occurred in 75%, 50%, and less than 10% of visits, respectively. A particular communication theme occurred a similar percentage of time in each visit modality. For instance, patient education (C5) was discussed in 83% of in-person visits and 76% of teledermatology visits. In no category (C1–C8) were these differences significant.

Table 5.

Communication Categories Found in In-Person and Teledermatology Visits

CATEGORY	IN-PERSON VISITS (N=42)	TELEDERMATOLOGY VISITS (N=50)	P VALUE
Informal talk	42 (100%)	50 (100%)	NA
Clinical assessment	42 (100%)	50 (100%)	NA
Technical issues	40 (95%)	50 (100%)	0.206
Psychosocial issues	1 (2.5%)	3 (6%)	0.623
Education	35 (83.5%)	38 (76%)	0.446
Discussion of treatment	42 (100%)	50 (100%)	NA
Promoting compliance	25 (59.5%)	26 (52%)	0.47
Administrative issues	41 (97.5%)	48 (96%)	1.000

Measurements are rounded to the nearest 0.5%.

NA, not applicable.

Diagnoses

During the study visit, diagnosis was documented as the one primarily discussed during the visit. Each of the diagnoses was categorized into a broader diagnostic category, if one existed. Consequently all diagnoses fell into the following seven categories: acne, neoplasm, dermatitis/eczema, inflammatory ailments, psoriasis, infestations, and hair loss. Overall, the most common diagnosis in this study population was dermatitis/eczema, which accounts for 34% of the diagnoses (Table 6). This parallels the predominance of patients with this diagnosis in our clinics overall. Half of the in-person diagnoses were neoplasm, and nearly half of the teledermatology diagnoses were dermatitis. The same three diagnoses (dermatitis/eczema, inflammatory disease, and neoplasm) made up the top three diagnoses in both in-person and teledermatology visits: 80% and 82%, respectively. Just over 75% of patients diagnosed with neoplasm were seen face-to-face, and over 75% of patients diagnosed with dermatitis/eczema were seen via teledermatology.

Table 6.

Diagnoses for In-Person and Teledermatology Visits (n=91)

TYPE OF VISIT, DIAGNOSIS	NUMBER
In-person
Neoplasm	21
Dermatitis/eczema	7
Inflammatory disease	5
Psoriasis	4
Acne	3
Hair loss	1
Infestations	0
Total	41
Teledermatology
Dermatitis/eczema	24
Inflammatory disease	11
Neoplasm	6
Psoriasis	6
Acne	1
Hair loss	1
Infestations	1
Total	50

Discussion

The major finding from this study was that in-person and teledermatology visits are comparable in style and content of physician communication. The findings generally corroborate those of a previous study.¹

Although faculty dermatologists were well experienced with teledermatology (years of experience at the time ranged from 4 to 11 years), a limitation of the study may be that 96% of teledermatology visits (p<0.001) were conducted by the same attending physician. However, 10 residents were also involved in the study visits. Residents conducted half or more of each teledermatology visit. As was customary during in-person visits, the residents conducted the first portion of the examination, primarily the assessment. Case management (diagnosis, treatment, and education) was conducted by the resident and faculty provider, with each faculty provider's proportion dependent on the resident's experience, knowledge, and confidence. The resident physicians had varying levels of telehealth experience. Their presence potentially “diluted” the influence of having predominantly one faculty physician participate in this study, suggesting that telehealth may be an equivalent form of communication for most practitioners, not just a few, highly qualified individuals.

In a previous study, the length of teledermatology visits was significantly shorter than in-person visits because teledermatology visits were scheduled for 10-min intervals, and there was only one videoconferencing suite.¹ In the current study there was no statistically significant difference between the lengths of the two visit modalities as both modalities provided for 20-min appointments, and a second teledermatology suite was added. Over half of the visits in each modality group were longer than the scheduled 20 min. Additionally, although 36% of in-person visits were 30 min or over, only 7% of teledermatology visits were this long. Telehealth visits may more often fall within the scheduled appointment time because of no procedures being performed during videoconferencing or because of fewer interruptions by ancillary staff, phone calls, etc. Patients and family may be less inclined to converse during these visits because of unfamiliarity with telehealth. Indeed, there were significantly fewer “other words” spoken during teledermatology visits (p=0.01).

Although physicians dominated the conversation in both visit modalities, others (predominantly patients, but also patients' family members, nurses, and clinic and study personnel) spoke less during teledermatology visits. This could be a result of less familiarity with teledermatology. In general, patients tend to be more familiar with the routine of in-person doctor visits than with telehealth visits. Additionally, significantly more in-person patients were return patients than were the teledermatology patients. It has been shown that as patients become more comfortable with their physician, surroundings, and mode of health delivery, they have increased participation.⁸ During teledermatology visits, the patient is usually the only person on the screen, making it difficult for “others” to speak because they do not feel as if they are a part of the visit.

There were no statistically significant differences between modalities in the average number of words in the following content categories: small talk, clinical assessment, psychosocial issues, patient education, patient compliance, patient treatment, and administrative issues. As well, the same communication themes occurred in each modality to the essentially same degree. As expected, assessment and discussion of treatment occurred in all in-person and teledermatology visits, but so did informal/small talk. This result differs from the original study's findings regarding small talk: 29.6% of teledermatology visits included small talk, significantly outnumbering the 20% of in-person visits in which small talk occurred.¹ Liu et al.⁶ found that physicians felt that “too much time was spent on small talk” and that telehealth visits were significantly shorter. Perhaps more enlightening is that promoting compliance, psychosocial discussion, and disease education occurred in similar percentages of in-person and teledermatology visits, making both modalities comparable in the “finer points” of physician communication, indicating that the two visit types are also comparable in the quality of care. Physicians were equally well able to focus on the patients' personal issues and to express empathy and encouragement. The amount of communication in one category does not affect the amount of communication in another.

For both modalities, the majority of physician discussion was spent on patient treatment. Previous studies have shown that patients forget substantial amounts of information provided during their visit; independent of age, the more information that was provided about prognosis, the less information the patients recalled.¹² Therefore, when patients were given only pertinent and necessary information, they were able to remember that information better. This has led experts to believe that pathophysiology of the patient's disease should not be the focus of the interview, but, rather, what the patient should do to improve his or her condition. Thus, having the focus of the interview on treatment is in the best interest of the patient.

This study also analyzed patient demographics and diagnoses. There were no significant differences found between the age and gender of patients in in-person and teledermatology visits. However, the in-person group consisted of significantly more return patients. The difference in diagnoses found in each group in this study may be attributed to different patient populations. More patients with neoplasms were seen in-person than in teledermatology visits. More dermatitis cases were seen in teledermatology than in-person visits. It is likely that these differences were the result of the diagnosis listed when visits were scheduled or as a result of procedural necessity due to diagnosis. A retrospective descriptive analysis of patients coming to the university clinic for follow-up after a telemedical visit was conducted by Edison et al.¹³ The need for a procedure or clinic-based intervention was a reason for the visit in 79% of these encounters. Thus the findings of this study regarding scheduling in-person versus teledermatology clinic visits in reference diagnosis are to be expected. Also unexpected is that procedural treatment recommendations occurred more often in patients physically returning to the university dermatology clinic, compared with having more medical treatment recommendations observed in the telemedical group.¹³

Conclusions

Greater confidence in the use of telehealth is warranted. This research suggests that physicians communicate with patients using similar style and content doing “virtual visits” through videoconferencing or in person. Investigation of this nature should be expanded to determine if there is consistency and similar findings across specialties.

Footnotes

Acknowledgments

Kristen Twenter, a student at the University of Missouri, contributed many hours conducting the audio recordings of clinic and telemedicine visits for this study.

Disclosure Statement

No competing financial interests exist.

References

Demiris

, Edison

, Vijaykumar

. A comparison of communication models of traditional and video-mediated healthcare delivery. Int J Med Inform, 2005; 74:851–856.

Fleming

. Ethical implications in the use of telehealth and teledermatology. Pak

, Edison

, Whited

. Teledermatology: A user's guide. Cambridge, MA: Cambridge University Press, 2008; 97–108.

Whited

. Summary of the status of teledermatology research. Teledermatology Special Interest Group, American Telemedicine Association June 7, 2010. www.americantelemed.org/files/public/MemberGroups/Teledermatology/Teledermatology_Research_2010.pdf. 2011 April 22.

Edison

, Ward

, Dyer

, Lane

, Chance

, Hicks

. Diagnosis, diagnostic confidence, and management concordance in live-interactive and store-and-forward teledermatology compared to in-person examination. Telemed J E Health, 2008; 14:889–895.

Ball

, McLaren

, Summerfield

, Lipsedge

, Watson

. A comparison of communication modes in adult psychiatry. J Telemed Telecare, 1995; 1:22–26.

Liu

, Sawada

, Takizawa

et al. Doctor-patient communication: A comparison between telemedicine consultation and face-to-face consultation. Intern Med, 2007; 46:227–232.

Fleming

, Riley

, Boren

, Hoffman

, Edison

, Brooks

. Incorporating telehealth into primary care resident outpatient training. Telemed J E Health, 2009; 15:277–282.

Onor

, Misan

. The clinical interview and the doctor-patient relationship in telemedicine. Telemed J E Health, 2005; 11:102–105.

Missouri Telehealth Network. Missouri Telehealth Network overview. 2010. http://telehealth.muhealth.org/about%20mtn/about_projects.html. 2011 April 26.

10.

Demiris

, Speedie

, Finkelstein

. The nature of communication in virtual home care visits. Proc AMIA Annu Symp, 2001; 135–138.

11.

Callahan

, Bertakis

. Development and validation of the Davis Observation Code. Fam Med, 1991; 23:19–24.

12.

Jansen

, Butow

, van Weert

et al. Does age really matter? Recall of information presented to newly referred patients with cancer. J Clin Oncol, 2008; 26:5450–5457.

13.

Edison

, Chance

, Martin

, Braudis

, Whited

. Users and nonusers of university-based dermatology services following a teledermatology encounter: A retrospective descriptive analysis. Telemed J E Health, 2011; 17:14–18.