Can telemedicine be used to improve communication between primary and secondary care?

TECHNICAL FEASIBILITY

Four of the early consultations were subject to serious technical failure, such as loss of sound or vision, or both. The images were not adequate for the dermatologist. Apart from those referred to dermatology, the patients had conditions that could be dealt with on the basis of their history, test results, and, occasionally, a visual examination. This usually required patients to show the distribution of their symptoms or the range of movement of joints. In one case the endocrinologist asked a patient to drink a glass of water in order to examine a goitre. The quality of the images were found to be satisfactory for these purposes.

PARTICIPANTS' SATISFACTION

The self administered questionnaires were completed by 43 (80%) of the patients, 43 (80%) of the general practitioners, and 48 (89%) of the consultants.

Doctors' satisfaction—Table 1 shows the levels of satisfaction reported by the general practitioners and consultants. Both groups of doctors generally recorded positive responses for all of the items used in the evaluation. The consultants were somewhat more critical of the technical performance, rating both sound and vision positively in only 54% of evaluated consultations. However, they expressed higher levels of satisfaction with the level of communication and information received. The consultants classified 44 (92%) of the referrals as most certainly or certainly appropriate. No referral was classified as inappropriate.

Patients' satisfaction was measured in both general statements and through specific parameters such as rapport, shyness, and confidentiality. The two overall measures were the patient's general satisfaction rating and their willingness to teleconsult again (tables 2 and 3).

FEASIBILITY OF TELECONSULTATIONS

The use of teleconferencing did not pose substantial technical problems. While the areas that received the least positive appraisal were technical performance and administrative arrangements, it must be appreciated that all the participants were novices with the technique. It is likely that increased practice and training would result in greater skill. Clear reasons were identified for the negative technical scores, such as too poor an image. The dermatologist's dissatisfaction with the quality of the images was in part due to the financial constraint placed on the technical investment in the practices, the level of skill of the users, and the limitations of current “off the shelf” software.

Other negative scores were due to administrative problems such as missing medical records. The relatively poorer ratings on technical performance by consultants is probably because they placed greater demands on the technology, which had to provide them with adequate sound and vision. Also, the general practitioners and patients usually used the equipment in “hands free” mode, which reduced the quality of the sound received by the consultant.

While there was scope for technical improvement, the basic quality of the audio-visual link did not seem to inhibit satisfactory consultations, providing a link was successfully made. Moreover, the collaboration needed to overcome any practical problems during the consultation seemed to increase the rapport between the participants. Most of the technical difficulties would be readily resolvable with improved software and a relatively inexpensive upgrade in telecommunications.

DRAWBACKS OF TELECONSULTATIONS

The inability of consultants to perform a physical examination may be perceived as a drawback for teleconsulting. In our study, however, some physical examinations did occur. These were visual examinations of patients, usually involving their demonstrating the problem or site of complaint. In some cases an examination by proxy was performed, with the general practitioner being guided by the consultant. In one case the ear, nose, and throat surgeon requested an appointment with the patient for direct visual inspection of their nasopharynx. With an appropriate level of investment—for example, in fibreoptic instruments—this could be achieved from the general practitioner's surgery.14

Concern has been expressed about legal liability and teleconsultations. Essentially, there is no difference between a teleconsultation and a conventional outpatient referral.15 The onus is on the general practitioner and the consultant to make clear who is taking responsibility. Contemporaneous written records are required as for the usual type of referral.

Teleconsulting does create logistical problems in that all parties have to be present at the same time. This was manageable in our small scale trial, but it would undoubtedly cause greater difficulties if a service was introduced on a larger scale. Preliminary experience indicates that there may be a feasibility ceiling of two to three teleconsultations a week for a general practitioner. This would, however, represent a large proportion of suitable referrals, as teleconsulting would not be appropriate for all referrals. This level of teleconsulting could be of considerable educational value, and the possibility of teleconsulting being made eligible for the post graduate educational allowance are being explored.

Joint consultations may involve more general practitioner time. Research indicates, however, that there are fewer return visits to the general practitioner after a joint consult.9 Our own study was too short to register any possible changes in referral behaviour. For patients, the reliability of appointment times is likely to be a considerable improvement on conventional outpatient clinics.

It is encouraging that the teleconsultations in our study could be accomplished effectively with relatively low level equipment based on desktop PCs in the doctors' usual workplace. The expected development of multimedia electronic record systems and the NHS-wide electronic network make it likely that the quality of, and access to, teleconsultations will improve.

LIMITATIONS OF STUDY

While our preliminary study showed teleconsulting to be feasible and acceptable, the viability of this particular form of telemedicine is not yet proved and there are several factors which limit the conclusions that can be drawn from the study. The sample of general practitioners and hospital consultants is unlikely to be representative of the profession as a whole. Selection bias was strongest among the general practitioners, who were self selected enthusiasts. The initial evaluation of technical innovation will almost always be with enthusiasts, but this is likely to influence the doctors' satisfaction rather than that of patients. The patients who took part were largely unselected, with only one refusal to participate being reported by the general practitioners.

A larger, more rigorous study will be needed to evaluate this use of innovative technology. This will require a properly designed randomised control trial of adequate size and incorporating appropriate outcome measures. A pilot study of such a trial is currently in progress.

We acknowledge the support and guidance of the steering committee's members: Professor A Haines, Dr J Roberts, Ms D Holmes, and Dr D Harris. In particular, we acknowledge the support and encouragement of Mr P Garner, care project manager, BT Laboratories. We also gratefully acknowledge the help of the Royal Free Hospital NHS Trust and the following hospital specialists and general practitioners: Dr O Epstein, Mr R Morgan, Dr P Bouloux, Mr N Goddard, Mr R Quiney, Professor A MacLean, Professor B Taylor, Dr V van Someron, Dr D Flynn, Dr P McLaren, Dr A Jones, Mr D Economides, Dr H Chester, Dr J Sandford, Dr K Hoffman, Dr K Fraser, Dr A Parker, Dr P Wiseman, Dr S Corcoran, Dr C Mitchell, and Dr A Selwyn.