Patient and Physician Perceptions of Virtual Visits for Parkinson's Disease: A Qualitative Study

Abstract

Background and Introduction:

Delivering care through telemedicine directly into the patient's home is increasingly feasible, valuable, and beneficial. However, qualitative data on how patients' and physicians' perceive these virtual house calls are lacking. We conducted a qualitative analysis of perceptions of these visits for Parkinson's disease to (1) determine how patients and physicians perceive virtual visits and (2) identify components contributing to positive and negative perceptions.

Materials and Methods:

Qualitative survey data were collected from patients and physicians during a 12-month randomized controlled trial of virtual house calls for Parkinson's disease. Data from 149 cases were analyzed using case-based qualitative content analysis and quantitative sentiment analysis techniques.

Results:

Positive and negative perceptions of virtual visits were driven by three themes: (1) personal benefits of the virtual visit, (2) perceived quality of care, and (3) perceived quality of interpersonal engagement. In general, participants who identified greater personal benefit, high quality of care, and good interpersonal engagement perceived visits positively. Technical problems with the software were commonly mentioned. The sentiment analysis for patients was strongly favorable (+2.5) and moderately favorable for physicians (+0.8). Physician scores were lowest (−0.3) for the ability to perform a detailed motor examination remotely.

Discussion:

Patients and providers generally view telemedicine favorably, but individual experiences are dependent on technical issues.

Conclusions:

Satisfaction with and effectiveness of remote care will likely increase as common technical problems are resolved.

Introduction

Chronic conditions benefit from specialist care and regular follow-up.¹ However, access to specialist care is often limited by distance, distribution of physicians, and patient disability.^1

–4 In recent years, virtual house calls have emerged as a potentially feasible alternative to office visits, with evidence of similar efficacy and high levels of satisfaction expressed by both patients and physicians.^5

–8

While it is known that patients and providers generally like telemedicine,⁵ qualitative data on what motivates perceptions of virtual visits are lacking.⁸ Implementation theory suggests that perceptions can serve as facilitators or barriers to the uptake of new technology and affect adoption of new approaches to care.^9,10 Thus, understanding how patients and providers perceive virtual visits is critical to wide-spread implementation.¹⁰ To date, most research in telemedicine has focused on feasibility and efficacy of novel devices, remote monitoring, or site-to-site telemedicine consultation.¹¹ Only seven studies have explored videoconferencing delivered by physicians to patients at home.^{12

–20} Furthermore, qualitative data on the patients' or providers' perceptions of these home virtual visits are almost entirely lacking.²¹ Therefore, we conducted a qualitative study of patients and physicians during a randomized controlled trial to (1) determine how patients and physicians perceive virtual visits, and (2) identify components contributing to positive and negative perceptions.

Materials and Methods

Trial Design

As described previously, we conducted a 12-month randomized controlled trial to compare usual care with usual care augmented by four virtual visits with a Parkinson's disease specialist delivered directly into the home.^5,22,23 Internal review board approval from all participating sites and subject consent was obtained. A total of 20 physicians (Parkinson's Specialists) participated in the study. A total of 195 patient participants were randomized across 18 research sites. Individuals were randomized 1:1 to (i) usual care in the community (control) or (ii) usual care plus four virtual visits with one of the specialists. The control group optionally received a one-time virtual visit after study completion. Virtual visits consisted of a remote consultation with a Parkinson's disease specialist through secure videoconferencing. Most patients and physicians had no prior telemedicine experience. Visits consisted of patient history, Parkinson's-specific examination, and treatment recommendations. After each visit, patients and specialists completed an online survey with five open-ended questions (Table 4).

Analyses

Quantitative study findings of satisfaction with virtual house calls are published elsewhere.⁵ For the qualitative analyses, two techniques were used: (1) case-based qualitative content analysis and (2) quantitative sentiment analysis.

Case-based qualitative content analysis

Case-based qualitative content analysis techniques allow for a more contextualized examination of responses.^24,25 Cases were defined as the total feedback from a patient–physician pair. Feedback was received for a total of 149 cases, with between 1 and 4 visits per case. Individual responses were placed into matrix form with rows representing a single visit (n = 399) and columns representing patients' or physicians' responses to the five open-ended survey questions (Table 4).

Data analysis was performed by a researcher previously uninvolved with the study (J.R.M.) and by a study researcher (M.J.E.), using Atlas.ti.²⁶ Coding was conducted jointly, and differences in coding decisions were discussed to achieve consensus. For first cycle analysis, open coding techniques were used in conjunction with a preliminary coding protocol²⁷: (1) visit type; (2) patient versus physician feedback; (3) positive, negative, mixed, or neutral responses; and (4) concepts associated with each sentiment (such as “connection trouble” or “convenience”). Concept mapping techniques were used to derive a hierarchical thematic structure and define associations between developing concepts.²⁸ For second cycle analysis, pattern coding was used to further develop categories and themes.^27,29

Sentiment analysis

Sentiment analysis of free-text comments that accompany Likert-scaled item responses can provide more nuance to survey results. We performed automatic sentiment analysis on patients' and the physicians' comments. For each comment, scores representing the magnitude and polarity of the comment's opinion were calculated: the more negative the score, the more unfavorable the opinion conveyed; likewise, the more positive the score, the more favorable the opinion.

Several word-level sentiment dictionaries were combined, and sentiment scores made uniformly binary: −1 for words conveying negative sentiment and +1 for positive words. For each free-text survey response, the sentiment scores of the words were summed to give a sense of the overall sentiment of the response—negative, neutral (a score of 0), or positive. Summary statistics were then compiled (median and interquartile range, mean ± standard deviation) for each survey, overall and by visit, and to identify the highest frequency words used in each survey.

Validity

Measures to enhance validity included use of diverse coders, peer debriefing, and triangulation of analytic techniques, combining both qualitative and quantitative analysis of free-text data.³⁰ The coding schema is shown in Figure 2, and participant quotes are presented with numeric (study identification number) and categorical identifiers (P = Patient; D = Doctor).

Results

Study Population

The recruiting diagram is shown in Figure 1. Table 1 summarizes baseline characteristics of the study population. Participants were largely white (96%), college educated (73%), older (mean age 66 years), and had Parkinson's disease for 8 years. Familiarity with the Internet was high (96% used the Internet or e-mail at home), and half had previously conducted a video call (54%). Baseline satisfaction with usual care was high (83% satisfied or very satisfied).⁵

Fig. 1.

Flowchart of study participants.

Table 1.

Baseline Characteristics of the Study Population

CHARACTERISTIC	ALL RANDOMIZED PARTICIPANTS (N = 195)	VIRTUAL HOUSE CALLS (N = 97)	USUAL CARE (N = 98)
Demographics
Age as of screening, [mean (SD)]	66.4 (8.1)	65.9 (7.8)	66.9 (8.5)
Women [n (%)]	91 (46.7)	49 (50.5)	42 (42.8)
Race [n (%)]
White	187 (95.9)	92 (94.9)	95 (96.9)
Other	3 (1.5)	3 (3.1)	0 (0)
Prefer not to answer	5 (2.6)	2 (2.1)	3 (3.1)
Ethnicity [n (%)]
Hispanic/Latino	3 (1.5)	1 (1.0)	2 (2.0)
Not Hispanic/Latino	183 (93.9)	92 (94.9)	91 (92.9)
Prefer not to answer	9 (4.6)	4 (4.1)	5 (5.1)
Bachelor's degree or higher education [n (%)]	143 (73.3)	71 (73.2)	72 (73.5)
Presently married	151 (77.4)	75 (77.3)	76 (77.6)
Participants with care partners, relationship to participant [n (%)]
Spouse/partner	132 (67.7)	63 (65.0)	69 (70.4)
No response/no care partner	54 (27.7)	29 (29.9)	35 (35.5)
Other	2 (1.0)	2 (2.1)	0 (0)
Internet use and familiarity
Participants who use the Internet or e-mail at home [n (%)]	187 (95.9)	95 (97.9)	92 (93.9)
Participants who have ever used the Internet to look for health or medical information online [n (%)]	189 (96.9)	96 (99.0)	93 (94.9)
Participants who have ever used their desktop or laptop computer to participate in a video call or video chat [n (%)]	105 (53.8)	43 (44.3)	62 (63.3)
Clinical characteristics
Parkinson's disease duration—year (n = 185) [mean (SD)]	8.0 (5.6)	8.3 (6.2)	7.6 (4.9)
Montreal Cognitive Assessment Score [0–30]^a [mean (SD)]	26.2 (2.8)	26.1 (2.7)	26.4 (2.9)
Parkinson's disease care
Participants who have seen a Parkinson's disease specialist in the last 12 months [n (%)]	143 (73.3)	64 (66.0)	79 (80.6)
Participants who are satisfied or very satisfied with their current Parkinson's disease care [n (%)]	143 (73.3)	67 (69.1)	76 (77.5)

All values are mean (SD) unless otherwise noted.

The number of responses is 195 unless otherwise noted.

Lower scores indicate greater disability.

SD, standard deviation.

Qualitative Content Analysis

Patients' perceptions of virtual house calls

Patients' perceptions of the virtual visits were based on three major themes: (1) personal benefits of the virtual visit, (2) perceived quality of care, and (3) perceived quality of interpersonal engagement. The coding schema used to develop themes is shown in Figure 2, and substantiating quotes are presented in Tables 2 and 3.

Fig. 2.

Themes underlying patients' and physicians' perceptions of virtual house calls, with substantiating coding schema.

Table 2.

Themes Underlying Patients' and Physicians' Positive Perceptions of Virtual House Calls, with Substantiating Quotes

THEMES	SUBSTANTIATING QUOTES
Personal benefits of the virtual house call: Less travel Less wait time Less stress Less expense More comfort More convenient Access to specialist	P210: Very convenient compared to 3.5 hour drive each way! Less expensive. P297: On a cold rainy day it was so nice not to have to worry about getting a ride … I could concentrate on what I wanted to ask and the info the doctor provided. P596: Convenience - no traffic or parking issues therefore less stressful. P158: It was so good to not have to ride 45 minutes in a handicapped van.” P761: “THIS IS SO MUCH BETTER THAN GOING TO THE DOCTOR'S OFFICE 100 MILES AWAY.” (emphasis in original) P424: I was especially happy that this visit spared my wife and I the necessity of having to cope with today's inclement weather (snow). It would have taken us four hours or more to drive to and from JHMC in Baltimore. I also felt I was able to concentrate more on what [the Doctor] said, as well as less affected by the frequent distractions that too often plague my visit. P424: I feel like the virtual visit allows the same positive benefits as a face-to-face benefits, only it is much more convenient—saved travel time 3–5 hours. P587: It's very nice to be able to be home and comfortable rather than driving and stressed. I feel like the visit format gives much more accurate view of how I'm doing. And I don't feel at all like I'm missing out because we're communicating via computer rather than in person. P703: I didn't have to drive 350 miles round trip and survive city traffic for a hurried 20 to 30 minute visit with the specialist! P795: We travel a total of 6 hrs at the present time to see my dr (and are usually in there for 15 minutes at the most); [Virtual visit] evaluations have been very through and comfortable. P458: The virtual visit required very little energy on my part, compared with having to be properly dressed for a doctor's visit, drive to the office, check in and wait, finally see the doctor, and then have to drive home. This is an excellent way to see a neurologist in a regular and convenient form. P494: It was so easy to talk with the specialist. Much less stressful than fighting traffic to go see my specialist. If I were to take the taxi it would cost $65. I am sure there are some downsides involved (i certainly had technical difficulties for one). Once these were solved I was satisfied P714: Telemedicine works well, at least in my case, to reduce the expense, time & energy needed to travel to a PD specialist…Interestingly, [National Public Radio] recently had a report that insurance companies were finding that telemedicine didn't actually save them very much money. The report never mentioned the time, expense and energy telemedicine saves the patient.
	D769: The patient is isolated in rural Oregon. His wife has just been diagnosed with Alzheimer's disease… wondering if he could continue with his care long term via telemedicine.
Perceived quality of care More time with doctor More frequent visits More thorough assessment Benefitting from care	P795: The virtual visits are great—The [doctor] does not rush and takes time to talk with me about any concerns I may have. I find that the visits are equal or more extensive than my regular neurology app. …I hope this can be implemented for many patients. P862: I received much more help than I expected. What to look into, how to schedule meds, importance of PT, and so much more! P587: I felt like I had almost identical treatment to an actual office visit, but so nice to do from home rather than driving an hour each way…and [being] stressed. I feel like the visit format gives much more accurate view of how I'm doing. And I don't feel at all like I'm missing out because we're communicating via computer rather than in person. P941: I could easily see and understand the doctor. He [had] me perform the same tests as he would have in his office. He spent quality time with me and I feel that I had his undivided attention.
	D655: It was very rewarding to see this patient via this study. She was very poorly managed before the study… seeing her transformation into an active, happy woman on appropriate medication has been great. D765: Very satisfying to see patient playing ping pong with her grandson while waiting for me to come to computer for visit. D427: The virtual visit went well. I was able to evaluate all items satisfactorily. D598: The ability to see the patient, get a good history and do a limited exam is good. There is still significant difficulty in trying to assess for tremor vs. dyskinesia in a patient given pixelated video. D750: I enjoyed doing this—most of their needs were non-motor and telemedicine was very appropriate for this. Some of their complaints (painful muscle cramps and weakness while on a statin) would have benefited from hands-on examination
Quality of interpersonal engagement Good communication Felt seen and understood Liked doctor/patient	P424: I personally have been very pleased thus far. For some reason I find it easier to be more comfortable expressing my PD via a remote device than I do during a face-to-face visit. P870: THE PHYSICIAN I CONNECT WITH IS CARING AND THOROUGH. POSITIVE EXPERIENCE!… I HOPE TO CONTINUE P600: The specialist and I felt the overall quality of the video feed was excellent today. We could see and hear each other … For a couple of minutes it seemed like I was just sitting in her office. P810: The doctor was very compassionate and friendly and answered all of my questions. I could see myself using telemedicine between my doctor appointments to deal with problems about my health that may occur. P131: Excellent face-to-face time with specialist, without the distractions that sometimes are present when at the clinic. P513: Very good visit. [The doctor] is excellent. Caring, listens to concerns, and is knowledgeable.
	D462: This was visit 4—evidence of the patient-physician bond that can be established using telemed is that we both had great difficulty saying ‘goodbye’. He asked for another visit, and it was so hard to say there would be no more. I felt as if I were abandoning him D998: This visit highlighted positives of doing video visits from people's homes to discuss sensitive topics—the participants' spouse died recently and this medium felt good/private to discuss his death openly, not sure if the same would have come from talking in a cold/sterile doctor office. D609: Visit interaction was great, but it was very difficult to determine actual ratings for rapidly alternating movements D713: Great visit. Terrific patient.

D, doctor; P, patient; PD, Parkinson's disease.

Table 3.

Themes Underlying Patient and Physician Negative Perceptions of Virtual House Calls, with Substantiating Quotes

THEMES	SUBSTANTIATING QUOTES
Lack of personal benefits of the virtual house call Inconvenient (technical problems) Frustrating No benefit versus usual care Lack of continuity of care	P698: Computer problems delayed the start. I had to remove an existing version of the software, empty the trash and then download and install a new one. I was assisted by staff but the situation makes me anxious P810: I am very happy with my present neurologist but for people who don't have someone they are comfortable with locally telemedicine would be a definite plus P18: It is frustrating to feel like I need to change neurologists to get the care I should be getting… I'm a little confused whether I am to follow up with PCP or neuro. P625: There were tech issues with the microphone, which meant we had to be on the phone and computer at the same time. Very frustrating. Otherwise things were fine.
	D181: Technical issues make evaluation limited and experience frustrating D675: Need to improve technical quality for it to be useful D875: For unexplainable reasons [we] had 40 minutes of difficulty connecting. … had to reset my account…had to restart my computer, the patient had to schedule a new appointment, and it all took a ton of time. D772: Patient used iPAD for visit. During first 1/2 of visit the AV was lost 3 times. … Did not negatively affect quality of visit but it was bothersome and distracting. D468: The first 10 mins on the call patient was unaware that I had dialed and was busy with other things. I tried calling her cell phone and she did not answer, so I left her message. She eventually got her husband to help but she was unable to hear me (I could hear her fine). We conducted the phone using video for visual input and the telephone for audio. D667: My screen kept freezing, we tried connecting multiple times and using multiple locations without improvement of the video. We were able to continue the appointment with me being able to view the patient but she could only hear my voice. Also, much of the patient's speech cut out (maybe 20%), and I had to ask her to repeat herself.
Decreased quality care Technical troubles substantially affected care Trouble conducting exam Less control over situation	P772: Early on technical [we had] issues with the software and connecting. By the third session, these issues had pretty much disappeared. P986: It took too long to connect … The connection over the cable modem was interrupted half way through …The crispness (lack thereof) of the images likely hampered the specialist's evaluation of my movements.
	D181: Video quality remained poor and, combined with physical space limitations of room, camera angle and lightening (as well as the fact that she was in a wheeled desk chair and normally ambulates with a cane) made the exam limited and difficult. D413: 1. Difficulty bringing examined body part into webcam view. 2. A window in the background washed out part of viewing area. 3. Could not provide a view in which the entire body may be seen. 4. Frame rate was marginal, difficult to ascertain bradykinesia and the degree of it. D462: There were limitations in the assessment because: 1. patient had hearing impairment and did not always understand, 2. could not see his feet and outstretched arms well due to camera being too close; 3. subject could not walk alone, so gait could not be assessed. There were limitations to visibility given the camera position, so not all movements were in good focus. This was complicated by the fact that the patient does not ambulate unassisted. He was in the room alone. I was not comfortable asking him to stand or walk, so the evaluation was incomplete. D424: Talking about virtual, the assessment of fine motor deficits was virtually impossible. …Another area of concern again was the space - the patient's wife assisted in moving the camera around to capture the feet and other body parts … It certainly made the assessment more difficult. There was no way I could visualize his entire body (from head to toe) for walking - there was not enough distance to walk within the room. D229: This was a very difficult, cumbersome, and ineffective visit…When the quality of the connection is iffy or plain bad, it interferes with [the exam]—more so when the patient's motor score is in the moderately abnormal range and less so when they have minimal findings. The connection had to be restarted as at one point it got frozen. When reestablished, it still remained very choppy, freezing frequently. On the patient's walking test, she would jump from one end of the room to another, depending on the connection/screen freezing. Another problem was the low lighting in the room, yet another that during most of the evaluation the camera was out of focus (perhaps it was too dim for the autofocus to even find the target), …Many readjustments and awkward maneuvering were necessary to more or less cover leg exam, posture, etc.
Poorer quality of interpersonal engagement Difficult communication Less comfortable Lack of intimacy Lack of confidence in physician	P424: Virtual visits tend to be slightly more impersonal and matter-of-fact. But that's not a factor of great concern, just an impression. P229: Not as personal. P698: I don't feel that I'm building a relationship with the doctor. That fact makes it harder to accept her suggestions plus the ‘exam’ seemed rushed/routine today so some issues didn't come up. P785: As I progress with Parkinson's disease I feel the human 3D experience is more comforting. What can I say? I am a wimp! P745: I felt limited in asking questions and expressing myself. It just wasn't as personal as being in the same room. P782: Had problems connecting… Quality of communication was poor—cutting out. Dr was excellent. P698: It felt rushed and the doctor never really got to know me. P430: Try to remember you are dealing with a person with Parkinson's, try to be patient and compassionate. remember you are on a computer and your body language tells your interest level to the patient
	D573: Connected right away but very poor quality video with frequent arrests and lags and poor quality sound with varying volume and intermittent cutting out–[it was] no better than a phone call (or perhaps worse because we might have at least had good audio) D459: Connection was poor, lots of breaking up, having to repeat myself; patient couldn't hear me despite my mic being all the way up and his speakers being all the way up.

Of 149 cases, 124 (83.2%) expressed positive perceptions of the virtual visits based on substantial personal benefit, high-quality care, and high-quality patient–physician interactions. Nineteen expressed neutral perceptions (12.6%) and reported fewer personal benefits or reported lower quality of care or interpersonal engagement. Only six (4%) expressed negative perceptions.

Personal benefits of the virtual house calls

Personal benefits cited included increased convenience, decreased stress and effort of travel, decreased costs, increased personal comfort, and convenient access to a specialist. The following quote illustrates these benefits:

P624: “I love everything about it! It is difficult for me to get to town…[I] take half a day to go to visit a doctor who would spend less than 5 minutes … I have had greater attention given to me on these virtual visits than I have had from in person appointments. … Not having to drive and sit for long periods of time in waiting room is a BIG bonus.”

Patients who reported fewer personal benefits indicated they lived closer to their physician or were already satisfactorily engaged with a Parkinson's disease specialist. For example:

P810: “The convenience was a big plus and would be even more so if my neurologist was 2 hours or more away …. I am very happy with my present neurologist but for people who don't have someone they are comfortable with locally, telemedicine would be a definite plus.”

Perceived quality of care

Quality of care was mentioned in 68 cases (45.6%). Most of these (65/68) indicated that they were pleased with the quality of care and felt it to be comparable to or better than traditional care. Few (3/68) felt that quality of care was lower. Concepts cited included more thorough assessments, more time with the doctor, and increased frequency of visits.

P795: “The virtual visits are great. [The doctor] does not rush and takes time to talk with me …I find that the visits are equal or more extensive than my regular neurology appointment.”

P18: “With the virtual visit, I am given a more thorough, longer lasting visit. I like the doctor and have more faith in his philosophy. He listens and asks more questions. I feel like I am getting better care from my virtual visits.”

Few (4/149) expressed concerns that inability to perform a hands-on assessment decreased the quality of care or limited the exam.

Quality of interpersonal engagement

Quality of the interpersonal engagement was mentioned in 88 cases (59.1%). Patients who described positive interactions with the physician (71/88) also perceived virtual visits positively. Some reported trouble engaging with the physician (17/88), but still felt positive about the virtual visits (9/17), while others expressed neutral to negative perceptions of the visits (8/17). Patient-identified components of quality interpersonal engagement included good communication, feeling seen and understood, feeling that the doctor listened, and liking the doctor. Comments from patients included the following:

P226: “[There was a] Remarkable feeling of closeness: I felt seen and understood.”

P601: “The excellent expertise of my doctor and the relaxed atmosphere encourage an in depth conversation that is sometimes more difficult in a hospital/office setting. It encourages note taking that I don't consider doing elsewhere. I like this approach very much.”

Few patients (6/149) expressed negative perceptions of the visits. Those who did also reported technical problems, feelings that the virtual visit was less personal, or confusion as to whom they should follow up with:

P789: “I found the virtual visit too impersonal.”

P387: “I do not develop a relationship with the doctor virtually. She can't touch me, it seems artificial, and I don't feel connected the same way I would if we were in the same room.”

In summary, most patients liked the virtual visits and expressed wishes that the virtual visits could be used as part of routine Parkinson's care. When asked if there was anything they disliked about the virtual visit, many responded “NOTHING.” One patient expressed it this way:

P582: “The doctor is extremely knowledgeable, listens well and is very effective in addressing my problems. If you had a higher category that was ‘As good as I can imagine’, that is what I would have checked. …THANK YOU!!!”

Physicians' perceptions of virtual house calls

In contrast to patients, who heavily emphasized the personal benefits of care, physicians' free-text responses were predominantly focused on technical problems (present in 187/399 visits; 46.9%). This included connection trouble (noted in 51/339 visits; 12.8%), audio (31/399 visits; 7.8%) and visual (129/399 visits; 32.3%) difficulties, which limited the physician's ability to see, hear, and communicate with the patient, and which thus interfered with ability to conduct the examination. These difficulties were generally greatest in initial visits and diminished over successive visits. Typical comments from physicians included ones such as the following:

D181: “Audio was fine but video quality remained poor and, combined with physical space limitations of room, camera angle and lightening (as well as the fact that she was in a wheeled desk chair and normally ambulates with a cane) made the exam limited and difficult, and the experience frustrating.”

D339: “It was another arduous technically flawed experience with trouble connecting, poor video and audio quality and dropped calls. There was absolutely nothing convenient about it from my perspective and I could have done the same thing in less time with a telephone call.”

Inability to touch the patient, although less commonly cited as a problem by physicians (16/399 visits; 4%), occasionally contributed to difficulty conducting the examination.

D187: “This was a case where feeling for rigidity or seeing the person in person with better visual clarity could have been helpful. However, the patient was doing well and it would probably not have changed management.”

Despite technological difficulties, physicians made many comments like these:

D665: “It was very rewarding to see this patient … She was very poorly managed before the study… seeing her transformation into an active, happy woman on appropriate medication has been great.”

D462: “This was visit 4—evidence of the patient-physician bond that can be established using telemedicine is that we both had great difficulty saying ‘goodbye’. He asked for another visit, and it was so hard to say there would be no more. I felt as if I were abandoning him.”

Sentiment Analysis

The overall mean sentiment score for each survey (i.e., combined for all visits) is shown in Table 4, and permits the following observations:

Table 4.

Summary of Survey Sentiment

RESPONDER	SURVEY ITEM	RESPONSES INCLUDED IN ANALYSIS (N)	PERCENT NONRESPONSE (%)	MEAN ± STANDARD DEVIATION
Patient	Please let us know what you liked about your virtual house call.	100	49.7	2.0 ± 2.9
Patient	Please let us know what you did NOT like about your virtual house call.	93	53.3	0.5 ± 1.7
Patient	Please provide any feedback about the visit.	248	41.6	2.5 ± 3.0
Physician	Please provide any feedback about the visit.	284	33.2	0.8 ± 3.0
Physician	Please provide any feedback about the experience of performing the motor assessment remotely.	220	48.2	−0.3 ± 1.9

1. The mean score for general patient feedback (“Participant virtual house call”) suggests a relatively favorable opinion.

2. The overall mean score for patients' dislike of virtual house calls is close to zero, which reflects neutrality in their comments rather than outright expression of dislike.

3. The mean score for general physician feedback (“Doctor virtual house call”) is lower compared with patients, perhaps reflecting a more tempered consideration of telemedicine.

4. Physicians' comments on performing remote motor assessment are scored near zero, suggesting balanced or neutral opinion.

Highest frequency words found in items that direct commenters toward a particular valence can give insight into patients' common opinions. Tables 5 and 6 shows the most frequently found words in comments from “Like about virtual house call” and “Dislike about virtual house call.” The patients liked, for example, the convenience of virtual house calls, with concomitant reductions in their time and their drive to the doctor's office. Although patients disliked technical problems (e.g., computer and connection), their most common dislike was “nothing.”

Table 5.

Highest Frequency Words from Patient Surveys, “Like About Virtual House Call” and “Dislike About Virtual House Call”

	LIKE	N	DISLIKE	N
1	Visit(s)	45	“Nothing”	20
2	Doctor	22	Visit	15
3	Virtual	22	Computer	11
4	Convenience	21	Virtual	10
5	Time	21	Connection	9
6	Neurologist	17	Doctor	9
7	Drive	16	Personal	7
8	See	12	See	7
9	Person	11	Time	7

Table 6.

Highest Frequency Words from Physician Survey, “Feedback About the Experience of Performing the Motor Assessment Remotely”

	WORD	N
1	Difficult	61
2	Assess	46
3	Patient	45
4	See	42
5	Video	33
6	Tremor	33
7	Camera	31
8	Good	25
9	Quality	24
10	Bradykinesia	23

Discussion

In this study, patients' and physicians' perceptions of virtual visits were positive and driven by perceived benefits of telemedicine, quality of care, and quality of interpersonal interactions. As frequently reiterated by patients in our study, reduced burden of healthcare access (convenience, comfort, and not having to leave home/travel) is one of the major benefits of virtual over office-based care. These benefits maybe greatest for patients with multiple comorbidities and longer duration of chronic illness.³¹ While there are patently disadvantages to virtual care (e.g., technological hurdles and touchless examinations), the potential benefits to chronically ill patients warrant ongoing efforts to improve the utility and feasibility of this care modality.

It is important to note that pursuit of convenient care does not imply quality should be compromised. Responses from this and other studies suggest that quality of care remains important to both patients and providers, and affects how telemedicine is perceived.³² However, definitions of “quality care” may differ substantially between groups.³³ Physicians in this study, as in others, indicated that the ability to conduct the physical exam was limited by audio/visual problems, which translated into a sense of diminished quality of care.^34,35 Conversely, patients rarely commented on limitations of the examination, instead prioritizing communication and relational closeness. With increasing emphasis on providing patient-centered care, finding a middle ground that meets the needs of both parties will be essential to preparing telemedicine for mainstream use. Encouragingly, the majority of concerns expressed by providers were related to technical issues, which will likely be resolved with time and increased technological familiarity.^36,37

As with office-based care, patients' perceptions of quality of care in telemedicine appears to be strongly connected to the quality of the patient–provider interpersonal engagement.^31,32,38,39 Whether virtual or otherwise, the patient–provider relationship will likely retain a central role in how care is perceived. Our data further suggest that interpersonal engagement during telemedicine may in some ways be considered superior by patients, who value a comfortable environment with fewer external distractions and greater opportunity to ask questions. Similar findings have been noted elsewhere.⁴⁰

Overall, sentiment in patient responses was strongly positive, while provider sentiment was only moderately positive, most likely due to persistent technical issues. The strong positive valence of patients' perceptions is well supported by prior research,¹⁸ as well as concurrent survey data showing 96% were satisfied with the virtual visits.⁵ In contrast, providers' surveys revealed a stronger quantitative satisfaction ranking (83% positive) than was evident in the qualitative data.⁵ This maybe due to the fact that providers used the free-text responses to discuss technical issues pertaining to the visit. Thus, our qualitative data may not fully reflect providers' overall feelings toward telemedicine, and should be interpreted cautiously. Despite this, the provider data give valuable insight into issues that are central to telemedicine, and highlight the connection between technical quality and overall impact on quality of care in the telemedicine context.

The study had several limitations. First, participation in this study, was not equitable and likely reflected the digital divide.^3,41 Study participants were overwhelmingly white, well educated, and more familiar with the Internet than the general population.^3,41,42 Second, not all participants and physicians responded to every survey, and responses may have been biased toward those who felt most strongly about the visits. Third, we specifically asked what patients did or did not like, which may have elicited more negative or positive responses than patients might have volunteered otherwise. Because data were collected in survey rather than interview format, we were unable to explore or further contextualize responses, and we could not assess if nonresponders' or partial responders' perceptions qualitatively differed from those who responded.

Future efforts may address these limitations. Inclusion of underserved populations and those with less familiarity with the Internet may help determine the generalizability of these results. Use of interviews with patients and providers, as opposed to online surveys, maybe helpful in reducing response bias due to impaired motor function.

Despite the noted limitations, this study provides valuable insight into factors underlying patients' and providers' perceptions of in-home virtual visits. These findings may be applicable to other conditions associated with aging (e.g., Alzheimer's disease),^2,43 reduced mobility (e.g., amyotrophic lateral sclerosis), or limited specialty access (e.g., rare conditions).⁴⁴ Although a new technology, the age-old issues of quality care and interpersonal engagement remain central to healthcare. Thus, prioritizing good interpersonal communication, quality of care, and benefits to patients will help ensure that telemedicine achieves its potential to deliver quality patient-centered care to anyone, anywhere.

Footnotes

Acknowledgments

Statistical analysis conducted by Dr. J.R.M., PhD, NP-C, University of Rochester School of Nursing, University of Rochester, Rochester, NY, and by Dr. J.J.J., PhD, Department of Biostatistics and Computational Biology, University of Rochester, Rochester, NY. The Patient-Centered Outcomes Research Institute had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the article. Software used in the study was provided and supported by SBR Health (Cambridge, MA), Vidyo (Hackensack, NJ), and ID solutions (Indianapolis, IN).

Author Contributions

J.R.M.: Analysis or interpretation of data, design or conceptualization of the study, and drafting or revising the article for intellectual content. M.J.E.: Analysis or interpretation of the data and drafting or revising the article for intellectual content. J.J.J.: Analysis or interpretation of data, design or conceptualization of the study, and drafting or revising the article for intellectual content. C.A.B.: Analysis or interpretation of data and drafting or revising the article for intellectual content. D.B.B.: Analysis or interpretation of data and drafting or revising the article for intellectual content. K.M.B.: Analysis or interpretation of data and drafting or revising the article for intellectual content. C.M.B.: Analysis or interpretation of data and drafting or revising the article for intellectual content. P.N.S.: Analysis or interpretation of data and drafting or revising the article for intellectual content. R.S.: Analysis or interpretation of data and drafting or revising the article for intellectual content. A.W.W.: Analysis or interpretation of data and drafting or revising the article for intellectual content. E.R.D.: Analysis or interpretation of data, design or conceptualization of the study, and drafting or revising the article for intellectual content.

Industry Sponsor

This study was sponsored by the Patient-Centered Outcomes Research Institute (AD-12-11-4701).

Disclosure Statement

Dr. E.R.D. serves on the medical advisory board of and has stock options in Grand Rounds. All other authors report no disclosures.

References

Cheng

, Swarztrauber

, Siderowf

, et al. Association of specialist involvement and quality of care for Parkinson's disease. Mov Disord, 2007; 22:515–522.

Dorsey

, George

, Leff

, Willis

. The coming crisis: Obtaining care for the growing burden of neurodegenerative conditions. Neurology, 2013; 80:1989–1996.

Willis

, Schootman

, Evanoff

, Perlmutter

, Racette

. Neurologist care in Parkinson disease: A utilization, outcomes, and survival study. Neurology, 2011; 77:851–857.

Willis

, Schootman

, Tran

, et al. Neurologist-associated reduction in PD-related hospitalizations and health care expenditures. Neurology, 2012; 79:1774–1780.

Beck

, Beran

, Biglan

, Boyd

, Dorsey

, Schmidt

, Simone

, Willis

, Galifianakis

, Katz

, Tanner

, Dodenhoff

, Aldred

, Carter

, Fraser

, Jimenez-Shahed

, Hunter

, Spindler

, Reichwein

, Mari

, Dunlop

, Morgan

, McLane

, Hickey

, Gauger

, Richard

, Mejia

, Bwala

, Nance

, Shih

, Singer

, Vargas-Parra

, Zadikoff

, Okon

, Feigin

, Ayan

, Vaughan

, Pahwa

, Dhall

, Hassan

, DeMello

, Riggare

, Wicks

, Achey

, Elson

, Goldenthal

, Keenan

, Korn

, Schwarz

, Sharma

, Stevenson

, Zhu

National randomized controlled trial of virtual house calls for Parkinson disease. Neurology [Epub ahead of print]; DOI: 10.1212/WNL.0000000000004357.

Sood

, Watts

, Johnson

, Hirth

, Aron

. Telemedicine consultation for patients with diabetes mellitus: A cluster randomised controlled trial. J Telemed Telecare, 2017; [Epub ahead of print]; DOI: 10.1177/1357633X17704346

Irving

, Stewart

, Spallek

, Blinkhorn

. Using teledentistry in clinical practice, an enabler to improve access to oral health care: A qualitative systematic review. J Telemed Telecare, 2017; [Epub ahead of print]; DOI: 10.1177/1357633X16686776.

Flodgren

, Rachas

, Farmer

, Inzitari

, Shepperd

Interactive telemedicine: Effects on professional practice and health care outcomes. Cochrane Database Syst Rev, 2015:CD002098.

May

Towards a general theory of implementation. Implement Sci, 2013; 8:18.

10.

Caffery

, Martin-Khan

, Wade

. Mixed methods for telehealth research. J Telemed Telecare, 2016; pii: 1357633X16665684.

11.

Wootton

Twenty years of telemedicine in chronic disease management—an evidence synthesis. J Telemed Telecare, 2012; 18:211–220.

12.

Rasmussen

, Lauszus

, Loekke

. Telemedicine compared with standard care in type 2 diabetes mellitus: A randomized trial in an outpatient clinic. J Telemed Telecare, 2016; 22:363–368.

13.

McCrossan

, Morgan

, Grant

, et al. A randomised trial of a remote home support programme for infants with major congenital heart disease. Heart, 2012; 98:1523–1528.

14.

Morgan

, Craig

, Grant

, Sands

, Doherty

, Casey

. Home videoconferencing for patients with severe congential heart disease following discharge. Congenit Heart Dis, 2008; 3:317–324.

15.

Leon

, Caceres

, Fernandez

, et al. A new multidisciplinary home care telemedicine system to monitor stable chronic human immunodeficiency virus-infected patients: A randomized study. PLoS One, 2011; 6:e14515.

16.

Whitlock

, Brown

, Moore

, et al. Telemedicine improved diabetic management. Mil Med, 2000; 165:579–584.

17.

Dallolio

, Menarini

, China

, et al. Functional and clinical outcomes of telemedicine in patients with spinal cord injury. Arch Phys Med Rehabil, 2008; 89:2332–2341.

18.

Gardner

, Jenkins

, O'Neil

, Wood

, Spurrier

, Pruthi

. Perceptions of video-based appointments from the patient's home: A patient survey. Telemed J E Health, 2015; 21:281–285.

19.

Dorsey

, Venkataraman

, Grana

, et al. Randomized controlled clinical trial of “virtual house calls” for Parkinson disease. JAMA Neurol, 2013; 70:565–570.

20.

Achey

, Beck

, Beran

, et al. Erratum to: Virtual house calls for Parkinson disease (Connect. Parkinson): Study protocol for a randomized, controlled trial. Trials, 2016; 17:7.

21.

Tan

, Teo

, Srivastava

, et al. Improving access to care for women veterans suffering from chronic pain and depression associated with trauma. Pain Med, 2013; 14:1010–1020.

22.

Achey

, Beck

, Beran

, et al. Virtual house calls for Parkinson disease (Connect. Parkinson): Study protocol for a randomized, controlled trial. Trials, 2014; 15:465.

23.

Dorsey

, Achey

, Beck

, et al. National randomized controlled trial of virtual house calls for people with Parkinson's disease: Interest and barriers. Telemed J E Health, 2016; 22:590–598.

24.

Hsieh

, Shannon

. Three approaches to qualitative content analysis. Qual Health Res, 2005; 15:1277–1288.

25.

Ayres

, Kavanaugh

, Knafl

. Within-case and across-case approaches to qualitative data analysis. Qual Health Res, 2003; 13:871–883.

26.

Atlas.ti 7 [computer program]. Berlin: Scientific Software Development GmbH, 2013.

27.

Saldaña

The coding manual for qualitative researchers, 2nd ed. Washington, DC: Sage, 2013.

28.

Mammen

JR.

Computer-assisted concept mapping: Visual aids for knowledge construction. J Nurs Educ, 2016; 55:403–406.

29.

Ryan

, Bernard

. Techniques to identify themes. Field Methods, 2003; 15:85–109.

30.

Marshall

, Rossman

. Designing qualitative research, 5 ed. London: SAGE, 2011.

31.

Valikodath

, Leveque

, Wang

, et al. Patient attitudes toward telemedicine for diabetic retinopathy. Telemed J E Health, 2017; 23:205–212.

32.

Polinski

, Barker

, Gagliano

, Sussman

, Brennan

, Shrank

. Patients' satisfaction with and preference for telehealth visits. J Gen Intern Med, 2016; 31:269–275.

33.

Abuosi

AA.

Patients versus healthcare providers' perceptions of quality of care: Establishing the gaps for policy action. Clin Govern Int J, 2015; 20:170–182.

34.

Moloczij

, Mosley

, Moss

, Bagot

, Bladin

, Cadilhac

. Is telemedicine helping or hindering the delivery of stroke thrombolysis in rural areas? A qualitative analysis. Intern Med J, 2015; 45:957–964.

35.

Shah

, Morris

, Jones

, et al. A qualitative evaluation of a telemedicine-enhanced emergency care program for older adults. J Am Geriatr Soc, 2013; 61:571–576.

36.

Kolltveit

, Gjengedal

, Graue

, Iversen

, Thorne

, Kirkevold

. Conditions for success in introducing telemedicine in diabetes foot care: A qualitative inquiry. BMC Nurs, 2017; 16:2.

37.

Zayapragassarazan

, Kumar

. Awareness, knowledge, attitude and skills of telemedicine among health professional faculty working in teaching hospitals. J Clin Diagn Res, 2016; 10:JC01–JC04.

38.

Sandberg

, Trief

, Izquierdo

, et al. A qualitative study of the experiences and satisfaction of direct telemedicine providers in diabetes case management. Telemed J E Health, 2009; 15:742–750.

39.

Agha

, Schapira

, Laud

, McNutt

, Roter

. Patient satisfaction with physician-patient communication during telemedicine. Telemed J E Health, 2009; 15:830–839.

40.

Toh

, Pawlovich

, Grzybowski

. Telehealth and patient-doctor relationships in rural and remote communities. Can Fam Physician, 2016; 62:961–963.

41.

Dahodwala

, Xie

, Noll

, Siderowf

, Mandell

. Treatment disparities in Parkinson's disease. Ann Neurol, 2009; 66:142–145.

42.

Norris

Digital divide: Civic engagement, information poverty, and the Internet worldwide. New York: Cambridge University Press, 2001.

43.

Cialone

, Augustine

, Newhouse

, Vierhile

, Marshall

, Mink

. Quantitative telemedicine ratings in Batten disease: Implications for rare disease research. Neurology, 2011; 77:1808–1811.

44.

Wilkinson

, Spindler

, Wood

, et al. High patient satisfaction with telehealth in Parkinson disease: A randomized controlled study. Neurol Clin Pract, 2016; 6:241–251.