Management of Acute Sinusitis via e-Visit

Introduction

Acute sinusitis is diagnosed in 12% of U.S. adults annually, ¹ and is the most common diagnosis in online health care delivery. ^{2 –4} Initial diagnosis of acute sinusitis is primarily based on the patient's history rather than a specific clinical examination finding, and complications are rare, ⁵ making it an ideal condition for management with nonface-to-face (non-F2F) care. Non-F2F care in the form of telephone-based nurse protocols has proven successful for management of acute sinusitis in terms of decreased antibiotic use and lower patient follow-up rates, ^6,7 but few studies have evaluated the effectiveness of managing sinusitis through e-visit.

More antibiotics are prescribed for acute sinusitis than any other condition in the ambulatory setting. ⁸ Yet antibiotic use for uncomplicated sinus infections has been found to yield only a marginal benefit, with only 5–11% of those treated with antibiotics having a shorter duration of illness compared with those not receiving antibiotics. ⁹ In addition, a significant amount of sinusitis is due to viral infections ¹⁰ which would not benefit from antibiotic treatment. With the current emphasis on antibiotic stewardship, concerns have been raised that non-F2F care may further contribute to unnecessary antibiotic prescribing. ¹¹ These concerns are not unwarranted; rates of antibiotic prescribing in excess of 90% for virtual sinusitis encounters have been reported. ^12,13 However, F2F encounters have similarly high antibiotic treatment rates ranging from 86% ¹⁴ to 94%. ^7,15,16 When comparing F2F with non-F2F sinusitis encounters, the current literature is conflicting with one study finding similar rates of antibiotic prescribing between F2F and non-F2F encounters, ¹⁷ and other studies finding lower rates for non-F2F care. ^6,18 Literature has also been mixed as to whether non-F2F visits result in better adherence to guideline-recommended antibiotics. ^6,17,19

Whether an e-visit is able to resolve the episode of care for a patient is crucial when considering cost savings and patient satisfaction. There is great variation in the rates of follow-up after e-visits for simple acute conditions, ranging from as low as 4% ²⁰ to as high as 34%. ²¹ Similar rates of follow-up for e-visits for acute sinusitis were seen when compared with F2F visits in convenience care ² and F2F visits in primary care, ¹² whereas others found higher follow-up rates for e-visits. ⁴ Phone call protocol treatment had similar follow-up rates when compared with F2F care in the primary care setting. ⁶

The purpose of this project was to investigate whether e-visits for the management of acute sinusitis have equivalent clinical outcomes for patients in terms of antibiotic treatment and follow-up rates when compared with the more established modalities of F2F visits and registered nurse-administered phone protocol.

Materials and Methods

A retrospective chart review was conducted of Mayo Clinic Rochester primary care empaneled patients, ages 18 to 75 years, who had a clinical encounter for acute sinus symptoms from May 1, 2016, to May 1, 2017, through an e-visit, at a retail health clinic, or through phone protocol. For the purpose of this study, an e-visit is an asynchronous text-based communication/interaction with a health care provider done over a secure online patient portal using branching, symptom-specific, diagnostic question sets. The patient submits answers to predetermined questions, which are reviewed by advanced practice providers (APPs) who respond to the patient through an online secure patient portal message, with treatment recommendations, patient education, and/or prescriptions. F2F encounters in this study occurred in Mayo Clinic Express Care (MCEC), a local retail clinic staffed by APPs, providing in-person care for minor acute illnesses. The same staff of APPs provided care for the e-visit and F2F visits at MCEC. The third encounter type was management and treatment through a phone-based nurse protocol for sinusitis. Primary care provider appointments were not included as an encounter type in this study because patients phoning in for a primary care appointment are run through algorithmic nurse triage before appointments are scheduled. During the nurse triage process, if symptoms were suggestive of a sinus infection, the nurses would administer the phone-based treatment protocol for sinusitis and the patient receives recommendations and antibiotic treatment (if indicated) without ever being evaluated in-person.

F2F visits and e-visits were retrieved by searching for diagnostic codes, whereas nurse phone protocol encounters were identified by using Mayo Clinic's Advanced Cohort Explorer (ACE) search tool. ACE uses text searching of electronic medical records (EMRs) to identify specific words and/or phrases. We used ACE to search EMRs for the specific protocol (by reference number) and triaging tools nurses use for documentation of phone protocol encounters for sinusitis.

Patient encounters were excluded from review and study inclusion if they had been evaluated by a health care provider for acute sinusitis within 30 days before the visit, as well as any patients treated with an antibiotic in the preceding 30 days for any upper respiratory infection. To capture patient follow-up, those living or traveling outside of Minnesota at the time of the encounter were also excluded.

A randomized selection of 150 encounters from each encounter type was created for manual review. In addition, demographic data, antibiotic prescribing rates, antibiotic type and duration, whether the patient was advised at the initial encounter to schedule follow-up, and 30-day outcomes including any follow-up for a related symptom/condition were noted.

This study was approved by the Mayo Clinic Institutional Review Board.

Results

In the 1-year time frame, there were 383 e-visits, 968 F2F visits at MCEC, and 2,084 phone protocol encounters noted for sinusitis, of which a random sample of 150 of each type was reviewed. Mean age of persons using e-visits was significantly younger than the other two encounter types (e-visit 40.3 years, F2F 44.7 years, phone protocol 46 years; p ≤ 0.001). Gender was predominately female for all encounter types (Table 1).

Discussion

Antibiotic treatment rates of all encounter types are lower than most reported in the literature. ^{7,12 –14,16} Within this study, e-visit antibiotic treatment rates are equivalent to or lower than the more traditional encounter types such as F2F and phone protocol. The actual number of patients starting a prescribed course of antibiotics may be lower than the number of prescriptions written because many providers at the F2F and e-visit encounters instructed their patients to use a guideline-recommended ²² watchful waiting approach. This differs from the phone protocol as there is no option to recommend watchful waiting. There are also differences in the home care recommendations given between the e-visit, F2F, and phone protocol encounters. At the time of this study, home care recommendations for phone protocol patients did not discuss nasal steroids, decongestants, or the avoidance of antihistamines, whereas e-visit patients typically received a prescripted set of instructions detailing use of these guideline-based recommendations ²² whether antibiotics were prescribed or not. Similar instructions were often provided in the form of preprinted patient education handouts at F2F visits.

The same group of nurse practitioners performed the e-visits and F2F encounters in the study, allowing for a more direct comparison of the two health care delivery modalities rather than differences in provider management. As already noted, F2F encounters were more likely to result in antibiotic treatment than e-visits. There are a number of factors that may contribute to the difference in prescribing rates. In this retrospective study, patients were able to self-select their encounter type. If a patient had been previously excluded from a phone protocol or e-visit encounter due to a comorbid condition or report of severe symptoms, the patient may be more likely to seek care through an F2F encounter for a subsequent illness. Another contributing factor may be the provider's perception of whether the patient desires antibiotics. Recent studies suggest that patient demand and desire for patient satisfaction drive many providers to prescribe unnecessary antibiotics, ^23,24 and the format of asynchronous non-F2F visits may allow providers to feel less pressure to prescribe unnecessarily.

Frequency of follow-up after e-visit was equivalent to that of F2F visits, and both e-visits and F2F encounters had significantly lower follow-up rates than phone protocol encounters. With phone encounters, obtaining follow-up is simple. If the patient does not meet protocol for treatment or if they do not agree with the care plan given, they can request to have an in-person visit scheduled before the phone encounter is over. e-Visit encounters in contrast are closed after the APP responds, so the patient is unable to reply to the APP without beginning another e-visit encounter. Length to first follow-up after e-visit and phone protocol was significantly shorter than for F2F encounters. Follow-up occurred more frequently in patients not receiving an antibiotic in the initial encounter, suggesting follow-up was for diagnostic clarification rather than treatment failure. For patients who did follow up, those who initially had an e-visit were more likely to complete another e-visit, or communicate with their primary care provider (PCP) through secure online message, that is, portal message, compared with other initial encounter types. Patients who initially seek e-visits may prefer the convenience of asynchronous text-based care for subsequent visits. When e-visit patients follow-up in-person, they more often choose the retail clinic setting. The APP's responding to e-visits typically recommend any needed follow-up be done at MCEC. Patients who initially had a phone encounter were more likely to follow-up with their PCP. This may be because, as mentioned, patients speaking to a nurse during a phone encounter are easily transferred to an appointment scheduler to schedule an appointment with their PCP.

There are limitations to consider. The health care providers performing the e-visit and F2F encounters were all APPs, so this may not be as generalizable to physician groups. This particular group of APPs is strongly committed to antibiotic stewardship. They have direct professional involvement on Mayo Clinic's institutional antibiotic stewardship committee, and have been involved with a number of research studies aimed at reducing antibiotic prescribing for conditions such as conjunctivitis, streptococcal pharyngitis, and pediatric acute otitis media. This may explain the lower antibiotic treatment rates than have been seen in other studies. To accurately assess the impact of the watchful waiting approach, pharmacy data would be needed to determine whether patients filled their antibiotic prescriptions. Lastly, the antibiotic treatment rate may not be fully representative of patients presenting with sinus complaints as providers may have used other diagnoses when not intending to treat with an antibiotic, such as sinus congestion or upper respiratory infection.