Conference Coverage

E-visits less likely to generate antibiotic prescriptions for common ailments

 

Key clinical point: When patients were seen in person and by e-visit for the same complaint, antibiotics were given more frequently in person.

Major finding: Antibiotics were given in the office but not the e-visit in 73% of cases.

Data source: Prospective study of 113 office visits that were paired with independently assessed e-visits for the same patient and complaint.

Disclosures: Dr. Player reported no conflicts of interest.


 

AT NAPCRG 2017

MONTREAL– When the same patient was assessed in person and via an electronic visit (e-visit) for several common complaints, a prescription for antibiotics was more likely to be generated from the face-to-face encounter.

In a recent study, if antibiotics were prescribed in one setting, but not the other, the office visit rather than the e-visit was where the antibiotic prescription was written in 73% of cases. Visits for sinus problems and vaginal symptoms made up over 80% of these cases of nonconcordant prescribing.

Doctor, patient, medication Thinkstock.com
“This lessens concerns about antibiotic overprescribing in e-visits,” Marty Player, MD, said at the annual meeting of the North American Primary Care Research Group. He described the results of a recent study that evaluated 113 office visits paired with a mock e-visit for the same date, patient, and complaint.

The study compared the diagnosis and treatment of five common acute conditions in an outpatient and e-visit setting, examining the concordance of both diagnosis and treatment between the two settings for complaints of vaginal irritation or discharge, urinary symptoms, sinus problems, rash, and diarrhea.

Outcomes tracked included concordance between the office visits and mock e-visits for the diagnosis, whether antibiotics were prescribed, and the general choice of antibiotics. Determinations about concordance were made by a third provider who was not involved with either the in-person visit or the mock e-visit, said Dr. Player, of the department of family medicine at the Medical University of South Carolina, Charleston.

Nonconcordance in treatment could occur either because an antibiotic was prescribed in one setting, but not the other, or because the broad choice of antibiotic class differed between the two settings.

Adult patients who came to the outpatient clinic and agreed to be enrolled in the study also completed the e-visit questionnaires appropriate to their condition before they saw the provider in an office visit. Thus, mock e-visits were created that mirrored the office visit with the e-visit format used in practice.

At a later point in time, the blinded e-visit questionnaires were given to e-visit providers who treated the patients as they would if the questionnaires had been generated in an actual e-visit.

The study generated a total of 142 office visits with accompanying mock e-visits, but 29 were excluded for lack of completeness or inappropriateness for e-visit care. In all, 113 paired visits were evaluated. All but seven patients (94%) were female; slightly more than half (53%) of patients were aged 45 years or older.

About one-third of visits (34%; n = 38) were for vaginal discharge or irritation. Sinus problems were reported by 36 patients (32%). Twenty-five patients (22%) reported urinary problems, while eight patients (7%) reported diarrhea. Six patients (5%) complained of a rash.

In total, 78 visit pairs (69%) were assessed as being concordant. Of the 35 nonconcordant visits, over half (54%) were for sinus problems, 40% were for vaginal discharge or irritation, and 6% were for rash. None of the visits involving urinary problems or diarrhea were assessed as nonconcordant.

Examining the data another way, Dr. Player and his coinvestigators also looked at how many visits involved antibiotic prescribing, and how many of those visits were assessed as nonconcordant. Of the 96 patients (85%) who were prescribed antibiotics, 37 had office and mock e-visits that were assessed as discordant in antibiotic prescribing.

Of these visit pairs, about half (51%) were for sinus problems, and a third (32%) were for vaginal complaints. Urinary complaints made up 11% of the nonconcordant visit pairs where antibiotics were prescribed, and rashes made up the remaining 5%.

Diagnostic concordance was seen in about two-thirds of rash (67%) and vaginal discharge (63%) visit pairs. Concordance of diagnosis for sinus problems occurred in fewer than half (47%) of visit pairs.

Dr. Player said that the investigators excluded visits involving urinary or vaginal complaints that did not have an accompanying urinalysis or vaginal wet mount. This decision was made because the standard of care for both office visits and e-visits requires these laboratory tests for diagnosis, he said.

The study design came with some limitations, said Dr. Player. “Patients self-select for e-visits, and the patients in this study might be different from those in true e-visit encounters,” he said. Also, the diagnosis and treatment of sinus problems, rash, and diarrhea relied on clinical judgment alone in each visit setting. Still, he said, the study supports what many clinicians report anecdotally: Patients want to leave the office knowing that the clinician has “done something” for them, and often, that means walking out with a prescription in hand.

Dr. Player reported no conflicts of interest.

On Twitter @karioakes

   Comments ()

Recommended for You

News & Commentary

Quizzes from MD-IQ

Research Summaries from ClinicalEdge