J Telemed Telecare 2008;14:404-409
doi:10.1258/jtt.2008.080105
© 2008 Royal Society of Medicine Press
EDUCATION & PRACTICETechnology |
Experience with the implementation of a web-based teledermatology system in a nursing home in Singapore
Lavanya Janardhanan *,
Yung H Leow 
,
Martin TW Chio ,
Yongmin Kim 
and
Cheong B Soh *
* Biomedical Engineering Research Centre, Nanyang Technological University, Singapore;
National Skin Centre, Singapore;
Department of Bioengineering, University of Washington, Seattle, Washington, USA
Correspondence: Associate Professor Soh Cheong Boon, 50 Nanyang Avenue, Nanyang Technological University, Singapore 639798 (Fax: +65 6793 3318; Email: ECBSOH{at}ntu.edu.sg)
 |
Summary
|
|---|
We introduced a web-based teledermatology system, the distributed
personal health information management system (DPHIMS), into
a nursing home in Singapore. The introduction was conducted
in two phases. Five staff nurses in Phase 1 and nine nurse aides
in Phase 2 performed the data entry and uploaded digital images
of the resident's skin condition. By the end of Phase 2, there
were 50 residents registered with DPHIMS. The average age of
the participants was 82 years and 84% were women. There were
31 first-time referral requests registered in the system during
Phase 2. The average time taken to complete a referral request
was 86 minutes. The average time taken by the dermatologist
to prepare and submit a diagnosis/treatment report was 11 minutes.
An online survey form was given to the nurses and the dermatologists
to gauge their level of satisfaction and their experience of
using DPHIMS. All the nurses said they would readily recommend
DPHIMS to other nurses. Overall, the dermatologists felt that
DPHIMS was helpful in obtaining specialist care for the residents.
However, some skin conditions required a face-to-face consultation.
Thus a mixture of face-to-face consultations and consultations
via teledermatology may be necessary to provide complete diagnosis
and treatment to patients. Our experience suggests that understanding
and addressing the organizational concerns is as important as
solving the technical problems.
|
Introduction
|
|---|
Store-and-forward teledermatology has been practised using the
web for some years.
1–8 We introduced a web-based teledermatology
system,
9,10 the distributed personal health information management
system (DPHIMS), into a nursing home in Singapore. The introduction
was conducted in two phases: Phase 1 (October 2005–July
2006) and Phase 2 (August 2006–July 2007). We evaluated
the difference in the usage and acceptance of DPHIMS between
Phase 1 and Phase 2.
Setting
The nursing home accommodated a total of 350 residents with an average age of 80 years. The majority of residents were bed-bound, paralyzed or suffered from dementia. Most of them neither spoke English nor had computer skills. The population of interest was the residents with non-emergency skin conditions.
There were 10 staff nurses, 44 nurse aides, one resident doctor and one matron at the nursing home. Five staff nurses (age range 30–50 years) in Phase 1 and nine nurse aides (age range 20–30 years) in Phase 2 performed the data entry and uploaded digital images of the resident's skin condition.11,12 The patients did not use DPHIMS themselves due to poor English and computer skills, and poor vision and health.
Two dermatologists from the National Skin Centre, Singapore, volunteered to give free diagnoses via DPHIMS during its introduction. They had excellent English and computer skills. During Phase 1, the dermatologists were located in Singapore whereas during Phase 2, one of the dermatologists moved to the UK but continued to volunteer his service.
Web-based teledermatology system
DPHIMS was developed as a secure and scalable personal electronic health record management system for the nursing home by implementing store and forward telemedicine to provide offline diagnosis for various non-emergency chronic diseases.9 Figure 1 depicts the process flow in DPHIMS. Email messages were sent automatically to dermatologists when a new patient was referred and to nurses when a diagnostic report was ready. A hard copy of the diagnosis/treatment report was sent to the matron and then to the resident doctor who reviewed it and prescribed the suggested medications. The diagnosis report page and image display page are shown in Figures 2 and 3.
The nurses were assigned the task of administering the medications.
In Phase 1, the staff nurses delegated this task to the nurse
aides. In Phase 2, the nurse aides took the responsibility for
administering medications and monitoring the progress of the
patients' skin conditions. In the event of relapse, recurrence
or no progress, the respective patient cases were uploaded into
DPHIMS again for follow-up diagnosis/treatment.
The dermatologists provided the single most likely diagnosis in the diagnostic report, and differential diagnosis if any. They also provided the treatment plan if available along with suggestions, comments or recommendations. An analysis of the types of diseases diagnosed and treated with DPHIMS was performed.
User training
The concept of teledermatology and DPHIMS were explained to the nursing staff, matron, resident doctor and managers of the nursing home. Individual training sessions lasting 60 minutes per nurse and 30 minutes per dermatologist were conducted on how to use DPHIMS. An additional 30-minute training session was given to the nurses on the techniques of photographing a patient's skin conditions.
|
User surveys
|
|---|
An online survey form was given to the nurses and the dermatologists
to gauge their level of satisfaction and their experience of
using DPHIMS. The survey for the nurses was conducted at the
end of Phase 1 and Phase 2. The survey for the dermatologists
was conducted at the beginning of Phase 1 and again at the end
of Phase 2.
The questionnaire consisted of multiple choice questions. Questions based on system satisfaction had four choices, and questions based on acceptance, experience and recommendation of the system had Yes/No choices. The survey also contained open-ended questions requesting the user to provide their comments and experience. Separate interviews were conducted for the nurses on matters regarding the acceptance of DPHIMS into their regular workflow. The system usage for both phases was studied using the database logs.
|
Patients
|
|---|
By the end of Phase 2, there were 50 residents registered with
DPHIMS. The average age of the participants was 82 years and
84% were women. During the introductory phases, 11 patients
died from causes not related to the skin disorders.
Sixteen percent (out of 49 patients) of the lesions had started less than a week before referral to DPHIMS, 16% had started a few weeks before, 46% had started a few months before and 22% had started a few years before. Eighty-three diagnostic reports were prepared by the two dermatologists, with 49 first-time reports via DPHIMS and 34 follow-ups.
|
Diagnoses
|
|---|
The diagnostic status of the residents before referral via DPHIMS
is summarized in Table
1. Five residents were not prediagnosed
for their skin conditions. The dermatologists were able to diagnose
and treat these patients over DPHIMS purely based on the patient's
medical records and skin images, without needing a face-to-face
consultation.
The diagnoses (primary diagnosis) made via DPHIMS for 49 patients
are summarized in Table
2. The category ‘Other’
showing 26 occurrences includes patients suffering from skin
disease conditions like erythematous papules, purpura simplex
and tinea cruris. The category ‘Other’ also includes
the 22 patients with no primary diagnosis and only differential
diagnoses. Their diagnoses were confirmed during subsequent
follow-up via DPHIMS.
Out of the 37 residents (11 died, 2 status unknown) diagnosed
via DPHIMS, 20 (54%) were recommended for a face-to-face consultation
with the dermatologists. However, for financial reasons these
patients were not sent to a dermatologist and continued to obtain
specialist care through DPHIMS. Fourteen residents out of the
20 showed positive progress while following the suggested treatment
plan. Their progress was updated when they were referred for
follow-up via the DPHIMS system. Thus, DPHIMS helped to avoid
unnecessary face-to-face consultations and enabled the elderly
residents to obtain specialist care without having to leave
their nursing home.
Some of the residents were destitute and without families. With virtually no income, they lived on the government subsidies provided to the nursing home. These subsidies provided for only one hospital visit every three months for a full check-up and for emergency visits. This limitation and the unavailability and/or refusal by a resident's family to pay for the dermatologist's visit were the two reasons that residents did not see the dermatologist.
|
Nursing staff attitudes
|
|---|
There was a perceptible difference in the nurses' attitude between
the two phases. The causes for this difference can be grouped
into three areas:
- Administrative problems, such as the non-availability of overtime pay to the staff nurses, were encountered in Phase 1. In Phase 2, the nurse aides were paid overtime based on the work done during each DPHIMS session. The active participation of the matron in Phase 2 helped to increase the acceptance of DPHIMS as compared to Phase 1;
- Computer skills. In Phase 1, the majority of the staff nurses were intimidated by computers and DPHIMS. Also because of their busy work schedule, they were not enthusiastic about learning to use DPHIMS. However, when Phase 2 began, most of the nurse aides had undergone formal computer training, enabling them to learn and use DPHIMS independently without much assistance;
- Responsibility. The job description of the staff nurses was more supervisory, like monitoring nurse aides. On the other hand, the job scope of nurse aides ranged from helping patients with everyday activities like brushing hair and bathing to administering medications. During Phase 2, the nurse aides gathered information from the patient's case-notes and when necessary referred to medical dictionaries to understand patient medications before each DPHIMS session. This not only made them more engaged in caring for patients' skin conditions but also gave them opportunities to enhance their knowledge of medications, diseases, symptoms of skin diseases and treatment.
Nurse surveys
Five staff nurses from Phase 1 and nine nurse aides from Phase 2 completed the online survey. All five nurses (100%) agreed that they found an improvement in the quality of health care after using DPHIMS compared to an ordinary clinic visit (Table 3). However, three of them (60%) mentioned that the medical care received via DPHIMS was not as good as that from an ordinary clinical visit. Two nurses (40%) found the system (entering data/uploading digital images) very convenient to use whereas three (60%) found it somewhat convenient to use. Four of them (80%) were satisfied with the overall concept and performance of DPHIMS.8
Seven (78%) nurse aides expressed good agreement and two nurse
aides (22%) expressed fair agreement about an overall improvement
in the quality of health care after using DPHIMS compared to
an ordinary clinic visit (Table
4). Three (33%) mentioned
that the medical care received after using DPHIMS was as good
as an ordinary clinical visit, one (11%) said that it was even
better, two (22%) disagreed and three (33%) were not sure. Eight
(89%) found the system convenient to use, and all nine (100%)
were satisfied with the overall concept and performance of DPHIMS.
The survey results for nurse satisfaction between Phase 1 and
Phase 2 were compared using the Mann-Whitney Rank Test. The
null hypothesis was that ‘a similar level of nurse system
satisfaction existed in both Phase 1 and Phase 2’. The
test result was not significant (
P = 0.84). This indicates that
the null hypothesis cannot be rejected and that respondents
in both phases had a similar level of satisfaction with DPHIMS.
In spite of the similar satisfaction level with DPHIMS, the
system's acceptance into the regular workflow was observed to
be greater in Phase 2. The possible reason for this discrepancy
is that the staff nurses in Phase 1 gauged only the efficiency
and ease of use of DPHIMS, but not its effect on their workflow.
During Phase 2, however, the survey responses by the nurse aides
encompassed their acceptance of the system into their workflow
in addition to their opinions on DPHIMS and teledermatology.
This reinforces the point that liking the concept of a telemedicine
system is very different from its acceptance into the normal
workflow for routine use.
The responses for the questions requiring a Yes/No answer are summarized below.
Ease of use of DPHIMS interface – The majority of the nurses (4/5) in Phase 1 and all of them (9/9) in Phase 2 agreed that the DPHIMS interface was easy to use. However, one nurse in Phase 1 without computer experience said that it was difficult to use.
Preference in using DPHIMS for storing and maintaining patient information – Most nurses (4/5 in Phase 1 and 8/9 in Phase 2) said that they would prefer DPHIMS for electronically storing, maintaining and retrieving patient information instead of filing the case-notes in paper form. This result is consistent with the findings of Getty et al.,13 who concluded that a majority of nurses have a favourable attitude towards computerization. The nurses were impressed with the ease with which patient details could be retrieved and updated using DPHIMS.
Recommending DPHIMS to other nurses – In both phases, all the nurses said they would readily recommend DPHIMS to other nurses.
|
Specialist surveys
|
|---|
A face-to-face consultation is commonly considered to be the
reference standard in health care. Hence for the Phase 1 survey,
the response of the dermatologist who felt that the medical
care received after DPHIMS was not as good as an ordinary clinic
visit was expected. Surprisingly, the other dermatologist answered
that the medical care via DPHIMS was even better than a normal
clinic visit. The reason was that he felt the system aided in
improving the logistics of making patient appointments and providing
timely consultations. He also felt that in some cases it saved
the elderly patients making a visit to the hospital for treatment.
From the survey responses, it was observed that the system satisfaction of the dermatologists in Phase 1 was greater than in Phase 2. The degree of satisfaction with the system was related to the patient's ability to attend a face-to-face consultation on the dermatologist's request. A few cases involved a request from the dermatologist for a face-to-face session to perform tests like fungal scraping or biopsy. Face-to-face consultations were also requested based on the severity of the illness. However, these patients were not able to make a face-to-face visit due to financial constraints. The decrease in dermatologist satisfaction could be attributed in part to the inability to perform laboratory tests when required and difficulty in building a rapport with the patients.
The dermatologists mentioned that the image quality was usually very good and satisfactory for diagnosis. This response is similar to that of the study by Philips et al.14 The dermatologists said that images involving scalp disease were difficult to view and diagnose, which is similar to the findings of Taylor et al.15 This was due to presence of hair, which prevented the camera from focusing automatically on the scalp. Obtaining a better view of the scalp requires manual focusing. One dermatologist added that a video camera could be used along with still images to improve diagnostic accuracy. The dermatologists agreed that the DPHIMS interface was easy to use and that they were able to use it without any assistance. The dermatologists also mentioned that the medical history was good enough for making a diagnosis.
The dermatologists expressed frustration over the inability to touch and feel the patient. This sometimes decreased their confidence in providing accurate diagnosis over DPHIMS, and they referred these patients for a face-to-face consultation. This response was again similar to the conclusions of Philips et al.14
The dermatologist who moved to the UK during Phase 2 was able to continue providing consultation, diagnosis and treatment for his existing patients and could even accept new patients. The ability to work at long distances impressed the dermatologists. They were excited about being able to retain their patients without having to transfer them to other dermatologists, irrespective of their physical location.
Overall, the dermatologists felt that DPHIMS could be used to help the residents at the nursing home obtain care from a specialist. According to them, DPHIMS could save unnecessary visits to a dermatologist and assist patients in obtaining early treatment and regular follow-up at low cost. They noted that ‘face-to-face consultation is the gold standard, but teledermatology is the next best thing’. A similar conclusion was reached in the review by Eedy and Wootton.16
|
System usage
|
|---|
At the nursing home, there were 19 first-time referral requests
registered in the system during Phase 1. The average time taken
to complete a referral request, i.e. registering the patient,
performing data entry, attaching digital images and uploading
the referral request to the main server, was 1 hour and 41 minutes.
Photographing a patient's skin conditions took an extra 20–40
minutes after completing the referral request. The average time
taken by the dermatologist to prepare and submit a diagnosis/treatment
report was 9 minutes.
There were 31 first-time referral requests registered in the system during Phase 2. The average time taken to complete a referral request was 1 hour and 26 minutes. The average time taken by the dermatologist to prepare and submit a diagnosis/treatment report was 11 minutes.
|
Lessons learned
|
|---|
The success or failure of a health system in a nursing home
is mainly determined by its acceptance by the various levels
of the administrative hierarchy. Users may initially respond
positively to surveys based on their impression of telemedicine
concepts. However, when faced with difficulties arising from
integrating the system into their workflow, for example, they
become less inclined to use it. Hence, when introducing such
a system, the existing workflow of the target users must be
studied to ensure minimum workflow disruption. It would also
be beneficial to provide additional compensation (e.g. certificate,
monetary allowance) to the target users in order to promote
the acceptance of new work processes.
Only certain types of skin conditions such as eczema and tinea corporis could be treated without requiring a face-to-face consultation. Treatment of skin diseases via a store-and-forward teledermatology system also depends on the confidence of the dermatologists in making a diagnosis. Thus a mixture of face-to-face consultations and consultations via teledermatology may be necessary to provide complete diagnosis and treatment to patients.
|
Conclusion
|
|---|
DPHIMS successfully demonstrated its potential for managing
skin conditions at the nursing home in Singapore. The system
was used regularly to perform diagnosis and follow-up for the
residents. Use of the system successfully avoided face-to-face
referrals in many cases and enabled residents to obtain complete
dermatologist care from the comfort of the nursing home.
We observed that for a system like DPHIMS to be successful, the workflow must be thoroughly understood. New telemedicine systems should introduce minimum changes to the existing workflow and produce the minimum inconvenience if they are to be adopted into everyday routine. Also, gaining acceptance from every level in the organization (management, matron, nurses) by understanding and resolving non-technical issues was important in the success of DPHIMS.
The dermatologists were satisfied with DPHIMS and could perform diagnosis effectively using the digital images and medical information provided. In a few instances they felt that not being able to meet the patient in a face-to-face session to make a definitive diagnosis was a limitation. Thus for DPHIMS to be used as a regular tool in nursing homes in Singapore, the residents may need to alternate between the traditional mode of health-care delivery and DPHIMS. For skin conditions that the dermatologist feels do not require a face-to-face visit, diagnosis can be made via DPHIMS in a timely fashion and with a reduced consultation fee. The nurse aides could manage DPHIMS and patient referrals. Benefits like overtime pay and recommendation from the matron for career advancement would motivate the nurse aides to use the system on a regular basis.
In conclusion, it appears that DPHIMS can make a positive impact on health care and can be expanded to other nursing homes, residential homes, community centres and general physician offices. DPHIMS could also be used for managing other conditions like dementia, post-stroke rehabilitation in addition to skin diseases.
Accepted June 29, 2008
|
References
|
|---|
- Zelickson BD. Teledermatology in the nursing home. Curr Probl Dermatol 2003;32:167–71[Medline]
- High WA, Houston MS, Calobrisi SD, Drage LA, McEvoy MT. Assessment of the accuracy of low-cost store-and-forward teledermatology consultation. J Am Acad Dermatol 2000;42:776–83[Medline]
- Whited JD, Hall RP, Simel DL, et al. Reliability and accuracy of dermatologists' clinic-based and digital image consultations. J Am Acad Dermatol 1999;41:693–702[Medline]
- Lesher JLJr, Davis LS, Gourdin FW, English D, Thompson WO. Telemedicine evaluation of cutaneous diseases: a blinded comparative study. J Am Acad Dermatol 1998;38:27–31[Medline]
- Lim AC, Egerton IB, See A, Shumack SP. Accuracy and reliability of store-and-forward teledermatology: preliminary results from the St George Teledermatology Project. Australas J Dermatol 2001;42:247–51[Medline]
- Harrison PV. Teledermatology in Morecambe Bay – 5 years experience in 1,441 patients. J Eur Acad Dermatol Venereol 1999;12 (Suppl. 2):S97 (Abstract)
- Emedicine Store-and-forward teledermatology. See http://www.emedicine.com/DERM/topic560.htm (last checked 26 June 2008)
- Barnard CM, Goldyne ME. Evaluation of an asynchronous teleconsultation system for diagnosis of skin cancer and other skin diseases. Telemed J E Health 2000;6:379–84[Medline]
- Prabhaharan K, Lavanya J, Goh KW, Kim Y, Soh CB. Distributed architecture towards tele-diagnosis. Proceedings of the First Distributed Diagnosis and Home Healthcare (D2H2) Conference. Arlington, VA, 2–4 April 2006. IEEE 2006:105–8
- Lavanya J, Goh KW, Leow YH, Distributed personal health information management system for dermatology at the homes for senior citizens. Proceedings of the 28th IEEE EMBS Annual International Conference, New York City, 30 August–3 September 2006. IEEE 2006:6312–15
- Pak HS. Basic guide to dermatologic photography. See http://www.atmeda.org/ICOT/telederm%20Forms/GuidetoDermatologicPhotography.pdf (last checked 26 June 2008)
- Wang M, Lau C, Matsen FA3rd, Kim Y. Personal health information management system and its application in referral management. IEEE Trans Inf Technol Biomed 2004;8:287–97[Medline]
- Getty M, Ryan AA, Ekins ML. A comparative study of the attitudes of users and non-users towards computerized care planning. J Clin Nurs 1999;8:431–9[Medline]
- Phillips CM, Balch D, Schanz S, Branigan A. Teledermatology: issues in remote diagnosis and management of cutaneous disease. Curr Probl Dermatol 2002;14:7–38
- Taylor P, Goldsmith P, Murray K, Harris D, Barkley A. Evaluating a telemedicine system to assist in the management of dermatology referrals. Br J Dermatol 2001;144:328–33[Medline]
- Eedy DJ, Wootton R. Teledermatology, a review. Br J Dermatol 2001;144:696–707[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Summary
Introduction
