With the growing push toward used of electronic health records (EHRs), will the computer get in the way of the physician-patient interaction – literally or figuratively? One study, possibly the first of patient satisfaction with EHRs in psychiatric office practice, provides a unique opportunity to observe the effect of computers in the office on the physician-patient interaction and suggests that the disruption – at least for the psychiatrist-patient relationship – was not as was feared. Psychiatry is an area of special concern because communication is the interaction between physician and patient and confidentiality is a sensitive issue.
Dr. Randall F. Stewart and his colleagues at the University of New Mexico Health Sciences Center, Albuquerque (USA), surveyed psychiatric outpatients about their satisfaction with care before and after the introduction of EHRs. They found that these chronically mentally ill patients were equally satisfied with their psychiatrists’ care regardless of whether the doctor was using a paper-based record-keeping system or a computer. Furthermore, the patients who were expected to have the greatest difficulty with computers-those with a psychotic disorder-did not have any decrease in satisfaction when EHRs were introduced.
Because EHRs have not been widely adopted in the US, outcomes researchers now have a window, as they are introduced, to explore their effects using observational studies, Dr. Stewart said. Only about 20% of US hospitals use EHRs; the proportion is lower for outpatient medical practices, and in psychiatry the subject of office computers has received virtually no attention.
The study was conducted in a clinic that provided care for 20 to 40 patients a day. Most patients were seen for medication management, though psychotherapy was provided to a minority. Clinically stable, adult psychiatric outpatients were administered a modified form of the validated Patient Satisfaction Questionnaire 18 (PSQ-18), with some subscales dropped and some additional, locally-developed subscales added, having to do with anxiety, confidentiality, and concerns about computer use. The final composite questionnaire consisted of 23 questions: 1 measuring general satisfaction and 7 subscales addressing satisfaction with a variety of aspects of the doctor-patient interaction, most not directly concerned with computer use (Table 1).
Table 1: Subscales of the composite patient satisfaction questionnaire
Overall. General satisfaction with psychiatric care.Technical. Satisfaction with knowledge and abilities of the psychiatrist and how the office is equipped.
Interpersonal. Manner of psychiatrist, whether businesslike and impersonal or friendly and courteous.
Communication and Education. How well the psychiatrist explains the reasons for tests, answers questions, and understands and pays attention to what the patient says.
Time. Spending sufficient time with the patient.
Confidentiality of the patient record.
Anxiety about the future or about psychiatric care.
Computer use. Is the patient comfortable with it, or does it get in the way?
The questionnaires were administered to patients upon discharge from outpatient treatment during a period of paper record keeping and another period, beginning 4 months after electronic charting had been implemented so that psychiatrists had a chance to become comfortable with it.
The researchers found no difference in overall satisfaction or in any subscale of satisfaction between the 161 patients surveyed pre-EHR and the 141 surveyed post-implementation. This result was somewhat surprising, according to Philip J. Kroth, MD, a co-author. “We found the effect of the computer on patient satisfaction in psychiatry is really no different than previous studies have shown for traditional medical patients. There are apparently no special problems in the behavioural health care arena.” When the investigators stratified the results by primary diagnosis, they found another surprising result: patients with schizophrenia, despite difficulty processing reality and a tendency to view the recording of information with suspicion, were no more likely to dislike electronic than paper records, Dr Kroth said. The study was statistically powered to detect a 7% difference in satisfaction between encounters using paper records and the EHR; thus it is unlikely that the study failed to detect a large difference in satisfaction.
Anecdotally, Dr Stewart noted: “some physicians use computers very well. They greet the patient and give him their undivided attention, excusing themselves when they want to record or look something up on the computer.” Some even use the computer as a teaching tool-for example, showing the patient a correlation between blood drug levels and depression scores. “Patients tend to like it when the physician shares with them the information that’s on the computer screen,” he said. In contrast, in some offices the computer is placed against the wall and “the physician has his back to the patient and is asking questions over his shoulder while typing away.” Dr Stewart said that the study deliberately did not measure specific aspects of how psychiatrists implemented the EHR. Though this is a limitation of the study, the investigators thought it better to avoid intruding into the privacy of the office and inducing stress.
Another limitation, he said, is that the study did not measure outcomes directly, only patient satisfaction, a surrogate. “We were really interested in the dynamics of the psychiatrist-patient relationship, and patient satisfaction seemed the most readily available way to measure the quality of the interaction,” he said. He added that patient satisfaction is widely measured as a surrogate for quality of care, and the PSQ-18 provides a benchmark to compare this with other studies; also it is validated, freely available and easy to complete. However, it should be noted that their modified PSQ-18 was not validated.
Finally, the authors note that the participating patients may have been a limitation, in that the sampling strategy may have been biased toward patients who were more likely to participate, precisely because they were satisfied with the encounter. One can imagine that many patients are likely to say to their physician only, “I’m doing fine, thanks,” especially if they are there just to get a prescription refill.
The transition to EHRs is presenting outcomes researchers with ‘golden opportunities’ to study their effect using before-and-after designs, opportunities that will not always be available, he said.
In the UK, adoption of EHRs has been in fits and starts. While their impact in psychiatric practice has not been researched, it would not be surprising if they had little effect on the psychiatrist-patient relationship there too, according to Stefan Priebe, FRCP, Professor of Social and Community Psychiatry at Barts and the London School of Medicine and Dentistry, Queen Mary University of London, and an investigator of physician-patient communication.
Unlike the US, electronic records in the UK are being implemented from the top down, by the National Health Service via the country’s regional Trusts, he said. Implementation has been slow and uneven, delayed by controversy and unexpected costs, and has been a political issue since the effort began in 2002.
Most psychiatrists and other clinicians in the UK view EHRs as an additional burden, of uncertain benefit, Dr Priebe said. Where EHRs are used in the mental health setting, most psychiatrists prefer to keep a paper record and enter only the minimal required data on the computer when the patient is no longer present. The detailed notes, including sensitive information such as family history and suicide risk, typically don’t go into the electronic record, he said. Because patients’ EHRs are centrally accessible, there are real confidentiality issues; unlike the US, many NHS employees may have access to patients’ personal information. However, Dr. Priebe said that in his clinical experience, confidentiality of medical information is not a very important concern for UK patients with severe mental illness.