Qualitative Research

ELECTRONIC VISIT PROGRAM DESCRIPTIVE RESEARCH REPORTS

AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018 901

Implementation and evaluation of a pharmacist-led electronic visit program for diabetes and anticoagulation care in a patient-centered medical home

Emily M. Hawes, Pharm.D., BCPS, CPP, Department of Family Medicine, UNC School of Medicine, Chapel Hill, NC, and UNC Eshelman School of Pharmacy, Chapel Hill, NC.

Erika Lambert, Pharm.D., CPP, BCPS, UNC Medical Center, Chapel Hill, NC, and UNC Eshelman School of Pharmacy, Chapel Hill, NC.

Alfred Reid, M.A., Department of Family Medicine, UNC School of Medicine, Chapel Hill, NC.

Gretchen Tong, Pharm.D., CPP, UNC Family Medicine Center, Chapel Hill, NC, and Department of Family Medicine, UNC School of Medicine, Chapel Hill, NC.

Mark Gwynne, D.O., UNC Health Alliance, Chapel Hill, NC, and Department of Family Medicine, UNC School of Medicine, Chapel Hill, NC.

Address correspondence to Dr. Hawes ([email protected]).

Copyright © 2018, American Society of Health-System Pharmacists, Inc. All rights reserved. 1079-2082/18/0602-0901.

DOI 10.2146/ajhp170174

Purpose. Results of a study evaluating quality-of-care, financial, and pa- tient satisfaction outcomes of pharmacist-conducted telehealth visits for diabetes management and warfarin monitoring are reported.

Methods. A retrospective pre–post study was conducted to determine the impact of an electronic visit (e-visit) program targeting 2 groups of outpatients: adults with uncontrolled diabetes and warfarin-treated adults performing patient self-testing (PST) for monitoring of International Normalized Ratio (INR) values.

Results. A total of 36 patients participated in the e-visit program during the 2-year study period. Among warfarin-treated patients, the percentage of INR values in the desired range increased relative to preenrollment val- ues (from 62.5% to 72.7%, p = 0.07), and the frequency of extreme INR values (values of <1.5 or >5.0) decreased (from 4.8% to 0.01%, p = 0.01); the margin per patient was $300 during the first year and $191 annually thereafter. In the diabetes group, a decrease from baseline in glycosylated hemoglobin values of 3.4 percentage points was observed at 5.7 months after enrollment (p < 0.001), with significant improvements in frequencies of statin use, aspirin use, and blood pressure control; the margin was $100 per patient. The overall median patient satisfaction survey score was 39 of 40.

Conclusion. An online e-visit model for warfarin monitoring was an effi- cient, safe, and cost-effective method for implementing PST. Pharmacist- led management of diabetes through e-visits, often in combination with in-person visits, generated revenue while significantly improving clinical outcomes.

Keywords: anticoagulation, diabetes, electronic visits, patient self-testing, pharmacist, telemedicine, warfarin

Am J Health-Syst Pharm. 2018; 75:901-10

Telemedicine, defined as “use of electronic information and com- munication technologies to provide and support health care when dis- tance separates participants,” has introduced novel ways for patients and providers to interact.1 It encom- passes exchange of medical infor- mation via electronic communica- tion (e.g., live videoconferencing, store-and-forward imaging, Internet communication).

Given its potential to yield time and cost savings, telemedicine technology has the capability to grow beyond ini- tial implementations providing access to remote locations.2,3 Patient portals within the electronic medical record (EMR) are an example of telemedicine technology that has increased patient access to medical records through se- cure, online portals. Patients can view laboratory results, request medica- tion refills or appointments, and seek

DESCRIPTIVE RESEARCH REPORTS ELECTRONIC VISIT PROGRAM

902 AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018

medical advice on a platform that is efficient and convenient.

Electronic visits (e-visits), or com- munication between patients and providers through a secured electron- ic channel, are gaining popularity for nonurgent medical services, including chronic disease management. E-visits do not require that a patient travel to a site to interact with a medical pro- vider in person or through clinical video technology (CVT) and can occur anywhere with a computer and In- ternet access. In current e-visit mod- els, patients complete a standardized questionnaire, and providers respond within 12–24 hours with instructions that may include prescriptions or re- ferrals.4,5 E-visits offer an alternative to telephone consultations, which are frequently nonreimbursable, and in-office visits. E-visits often cost less than office visits and entail minimal to no indirect cost to patients.6,7 Data on e-visits are limited, and most perti- nent studies we identified focused on e-visits conducted by physicians and advanced provider practitioners not including pharmacists.4-8

Pharmacist-managed anticoagula- tion services have been demonstrated to yield positive outcomes with re- spect to safety, effectiveness, and cost savings. Several studies have indicated that pharmacist-led anticoagulation services in a variety of settings, in- cluding services provided via CVT, can achieve significantly better control of International Normalized Ratio (INR) values than can be achieved with usu- al care.9-16 In recent years, increased attention has surrounded the concept of patient self-testing (PST) for anti- coagulation monitoring. Compared with standard laboratory monitoring, home INR monitoring has been shown to enhance quality of life and patient satisfaction, improve INR control, and reduce mortality and thromboembo- lism occurrence, with no increased risk of major bleeding events.17-32

Several studies have shown that pharmacist care improves glycosylated hemoglobin (HbA

1c ) control, blood

pressure control, and adherence to

KEY POINTS • Results of a pre–post study

indicated that a pharmacist-led electronic visit (e-visit) pro- gram for patients with diabetes and warfarin-treated patients conducting patient self-testing (PST) resulted in significant improvements in International Normalized Ratio and glycosyl- ated hemoglobin values, a posi- tive financial margin, and highly satisfied patients.

• An e-visit model can be a safe, efficient, and cost-effective method for implementing PST and an effective follow-up alter- native to in-person visits or tele- phone consultations in diabetes management.

• The study was the first to evaluate chronic disease man- agement through pharmacist- conducted e-visits.

primary prevention therapies and has cost benefits.33-38 Additionally, diabetes management with various telemedi- cine interventions has been shown to yield significant improvements in HbA

1c control relative to usual care, but

the evaluated analyses did not include services conducted by pharmacists.39,40

In North Carolina, clinical phar- macist practitioners (CPPs) have a long history of providing direct pa- tient care for chronic disease state management by initiating, changing, or discontinuing pharmacotherapy, ordering laboratory tests, and billing for services. CPPs have the skill set to offer anticoagulation and diabetes management services as an e-visit.41 Many anticoagulation clinics do not routinely provide home INR manage- ment services to patients due to the high time burden and low reimburse- ment rates.42-44 However, a systematic online INR monitoring program in-

volving e-visits by a CPP may be a so- lution for providing quality warfarin management in a more cost-effective manner. Likewise, offering e-visits as a follow-up option for interested patients with uncontrolled diabetes may be efficient, convenient, and ef- fective in enhancing diabetes control. Thus, a CPP-led e-visit program was implemented for warfarin-treated pa- tients conducting PST and for patients with uncontrolled diabetes at a family medicine center (FMC). To our knowl- edge, there is no literature evaluating e-visits conducted by pharmacists for chronic disease state management. The objective of the study described here was to evaluate the impact of a pharmacist-led e-visit program for an- ticoagulation and diabetes manage- ment on clinical outcomes, financial outcomes, and patient satisfaction.

Methods

Practice setting. A pre–post retrospective and observational study was conducted at the FMC, a physician-owned academic medical center that is part of a large health system in North Carolina. The FMC annually serves approximately 19,000 patients through 56,000 visits. The FMC is recognized as a National Com- mittee for Quality Assurance level 3 patient-centered medical home (PCMH) that provides full-scope pri- mary care. A CPP is a licensed phar- macist who has met specific criteria specified by the North Carolina Medi- cal Board and North Carolina Board of Pharmacy and is authorized to pro- vide drug therapy management under the supervision of a licensed physi- cian. A CPP’s scope of care is specific to each practice site and guided by a protocol outlining disease states for which the CPP can prescribe and disease-specific drug therapy.41 Within the FMC, 2 CPPs provide direct pa- tient care through in-person visits in the pharmacotherapy clinic. Patients are referred to the pharmacotherapy clinic by FMC practitioners for en- hanced management of a variety of chronic diseases. The FMC uses an

ELECTRONIC VISIT PROGRAM DESCRIPTIVE RESEARCH REPORTS

AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018 903

Epic EMR and the MyChart patient portal (both from Epic Systems Corpo- ration, Verona, WI). MyChart provides patients access to their medical rec- ords and enables secure 2-way mes- saging between patients and provid- ers. The MyChart messaging feature is the mode of secure communication used for e-visits between patients and CPPs. An Epic smartphone applica- tion is also available for download.

Program development and de- scription. Anticoagulation. Prior to 2014, individual primary care provid- ers (PCPs) at the FMC had provided home INR management on a case-by- case basis, but this practice was dis- couraged by clinic administrators due to low clinic revenue, a high time bur- den, and logistic issues. For example, there were instances when patients with home INR monitoring devices had out-of-range INRs but their PCP was unavailable, requiring that on- call physicians address those situa- tions. Furthermore, PCPs attempted to use Current Procedural Terminol- ogy (CPT) code G0250 for billing, but this yielded approximately $9 in reimbursement per encounter and required evaluation and documenta- tion of 4 INR test results.45 In response to frequent patient requests for home INR warfarin management, the phar- macotherapy clinic run by CPPs began exploring options to make the process more efficient, safe, and cost-effective. Hence, the decision was made to pi- lot test an online e-visit monitoring model.

In 2014, the pharmacotherapy clinic began offering the current CPP- managed online home INR monitor- ing program on a rolling basis to pa- tients who were eligible for PST and agreed to the program’s expectations. Patients whose condition was stable, who were adherent to warfarin ther- apy for at least 3 months, who were emotionally, mentally, physically, and financially able to perform PST, and who had MyChart access were eligible. Once approved, the home monitoring device and supplies were shipped to the FMC, and a CPP con-

ducted initial in-person training on use of the device, the CoaguChek XS system (Roche Diagnostics, Indianap- olis, IN). The training included dem- onstration of use and care of the INR monitor, collection of a blood sample, education regarding the process of reporting home INR test results, and documentation of the patient’s abil- ity to perform testing. Once training was completed, the CPP billed for ser- vices using CPT code G0248. Ongoing testing supplies were provided to the patient by the manufacturer of the CoaguChek XS system. Subsequent visits were conducted through Epic’s MyChart messaging portal. Every 2 weeks on a predetermined day, the patient checked his or her INR, sub- mitted the result online or by phone to Roche, and electronically sent a com- pleted questionnaire (modified from Bussey and Bussey18) and INR data in a MyChart message to the CPP by 1 p.m. The CPP scheduled an afternoon pharmacotherapy clinic appointment (a 30-minute session) weekly to com- plete the PST e-visits. Via MyChart messaging, the CPP then responded to the patient with a management plan, including warfarin dosing and lifestyle adjustments, and provided education, ordered laboratory tests, and/or sent prescriptions to the pharmacy. The CPP also reviewed the Roche Coagu- Chek Link Portal website to confirm that the patient had reported the INR data. Any patient with a home INR of >4.5 was required to visit the FMC for venipuncture INR verification and evaluation by a CPP or another pro- vider. CPPs submitted reimbursement claims via “incident-to” billing (i.e., billing for services provided incident to, or in conjunction with, those pro- vided by the supervising physician) using CPT code 99444 for 1 electronic encounter each month, as agreed upon during enrollment.

Diabetes. In January 2015, the e-visit program was expanded to in- clude CPP-conducted diabetes man- agement for interested patients. Pa- tients were recruited during CPP visits in the pharmacotherapy clinic, by re-

ferral from FMC providers, or through EMR reporting to identify patients with an HbA

1c concentration of >9%.

Patients were eligible for participation if they needed pharmacist-enhanced diabetes care, received primary care at the FMC, and agreed to online evalu- ation visits (billed using CPT code 99444) and use of MyChart. Eligible patients were initially enrolled by the CPP during a telephone encounter or ideally during an in-person visit in the pharmacotherapy clinic to establish a relationship, to allow assessment of health literacy and understanding of diabetes, and to provide baseline disease state education. Patients who needed additional follow-up were given the option of in-person visits, e-visits, or a combination based on patient preference. Insurance cover- age and expected copayment were evaluated prior to patient enrollment and communicated to the patient. On a predetermined date or based on pa- tient inquiry, the CPP sent the patient an individualized questionnaire to answer via MyChart. The CPP often responded with additional questions, resulting in back-and-forth messages between the patient and the CPP be- fore the conclusion of the encounter. Some patients receiving insulin used the “glucose flowsheet” available in Epic to report blood glucose read- ings and insulin dosing in a chart for- mat. At the end of the encounter, the CPP provided a management plan including medication adjustments and lifestyle modifications, provided education, ordered laboratory tests, sent prescriptions, and developed a follow-up strategy. After the patient indicated understanding and agree- ment, the CPP wrote a note, complet- ed the encounter, and billed for the e-visit. One CPP conducted e-visits for both anticoagulation and diabetes management.

Study design. The study, ap- proved by the university’s institutional review board, was a single-center, retrospective, observational pre–post study including a patient satisfaction survey. Patients were included if they

DESCRIPTIVE RESEARCH REPORTS ELECTRONIC VISIT PROGRAM

904 AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018

were at least 18 years of age, received primary care at the FMC, and par- ticipated in at least 1 pharmacist-led e-visit via MyChart from 2014 through 2016. All patients were asked to partic- ipate in a survey evaluating satisfac- tion with the e-visit program over the phone or through a MyChart message.

Statistical analysis. Categorical variables were represented as frequen- cies and percentages, and continuous variables were expressed as mean ± S.D. or median with interquartile range (IQR) values. For calculating the statistical significance of postimple- mentation versus preimplementation data, nominal variables were analyzed via chi-square testing, and continu-

ous values were analyzed using ei- ther a paired Student’s t test for nor- mally distributed data or Wilcoxon’s matched-pairs signed rank test for non-normally distributed data. The a priori level of significance was <0.05. Assumptions used in the financial analysis appear in the appendix.

Outcomes. The primary objective of the study was to assess the effect of a pharmacist-led e-visit program on clinical outcomes, financial indica- tors, and patient satisfaction. For the anticoagulation group, the primary outcome was the difference in the proportions of visits associated with an INR value in the target range of 2.0–3.0 before and after enrollment.

For the diabetes group, the primary outcome was the difference in mean HbA

1c values before and after enroll-

ment. The secondary endpoint for the anticoagulation group was the difference in the occurrence of ex- treme INRs (values of <1.5 and >5.0) before and after enrollment and the frequency of adverse events, defined as recurrent embolization or bleed- ing requiring medical intervention, in the postenrollment phase. Second- ary endpoints for the diabetes group included changes in the proportions of patients who had HbA

1c values of

<8% and <7%, had controlled blood pressure, and were taking aspirin or statin therapy (if indicated), as well as antidiabetic medication utilization and frequency of hypoglycemia, in the postenrollment phase. Additional outcomes for both groups included numbers and types of visits, time spent by the CPP on e-visits, reimbursement outcomes, and patient-reported sat- isfaction with the e-visit program. As- sumptions for our financial analysis included overhead costs, with calcula- tions based on average values in our practice and Medicare rates for ap- propriate billing codes at the time of the study. A modified version of the Satisfaction with Pharmacist (SWiP) scale was used to assess patient sat- isfaction with the e-visit program and was administered by a pharmacy resident.46 The survey consisted of 7 questions that were modified from the SWiP scale for use in the study along with 3 additional questions specific to the e-visit program. The maximum score of 40 indicated the highest level of satisfaction.

Results

A total of 36 patients participated in e-visits. Nine patients were en- rolled for warfarin management, and 29 were enrolled for diabetes man- agement. Two patients were enrolled for both anticoagulation and diabe- tes management. Demographic and outcomes data for enrolled patients are summarized in Tables 1 and 2. We found no relationship between any

Table 1. Baseline Patient Demographicsa

Variable Value

Diabetes group (n = 29)

Mean ± S.D. age (yr) 52.1 ± 7.8

No. (%) women 19 (66)

Race, no. (%)

Black or African-American 24 (83)

White or Caucasian 5 (17)

Mean ± S.D. HbA 1c

concentration (%) 10.9 ± 1.9

No. (%) with controlled blood pressureb 15 (52)

No. (%) receiving aspirinc 18 (86)

No. (%) receiving statind 23 (82)

Anticoagulation group (n = 9)

Mean ± S.D. age (yr) 54.3 ± 12.1

No. (%) women 5 (56)

Race, no. (%)

Black or African-American 0

White or Caucasian 9 (100)

Indication for anticoagulation with warfarin, no. (%)

Atrial fibrillation 2 (22)

Aortic valve replacement 2 (22)

Venous thromboembolism 5 (56)

Fraction (%) encounters associated with target INR 65/104 (63) aHbA

1c = glycosylated hemoglobin, INR = International Normalized Ratio.

bDefined as blood pressure values of <140/90 mm Hg in accordance with 2017 American Diabetes Association (ADA) guidelines.

cOnly 21 patients met 2017 ADA guideline criteria for aspirin therapy for primary or second- ary prevention or did not have a contraindication.

dOut of 28 patients, 1 patient had a contraindication to statin therapy.

ELECTRONIC VISIT PROGRAM DESCRIPTIVE RESEARCH REPORTS

AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018 905

evaluated outcome and patient age or sex, overall number of visits, or pro- portion of encounters conducted via e-visits.

Anticoagulation program. The mean duration of e-visit participation for anticoagulation patients was 10.4 months (median, 12 months; range, 4–16 months). Seven of the 9 patients were still receiving INR management via e-visits with a CPP at the time of data collection for the study. Two pa- tients discontinued due to relocating out of state. Prior to enrollment, pa- tients were receiving warfarin man- agement via in-person clinic visits with a CPP. INR control improved after e-visit participation (Table 2). Fur- thermore, no patients experienced an adverse event related to recurrent em-

bolism or bleeding requiring medical intervention while managed through e-visits.

Diabetes program. The median length of patient participation in the diabetes e-visit program was 4 months (range, 1–12 months). The majority of patients (n = 27) were enrolled during in-person visits, while 2 patients were enrolled by telephone. After enroll- ment, 19 patients (66%) participated in only e-visits, while the others had a combination of visit types. HbA

1c

was measured at a mean ± S.D. of 5.7 ± 4.5 months after enrollment (me- dian, 4 months; range, 1–19 months; IQR, 2.5–8.5 months). The mean HbA

1c

value significantly decreased after participation (Table 2). After enroll- ment, 20 patients (69%) achieved an

HbA 1c

value of <8%; of those patients, 9 (45%) achieved an HbA

1c value of

<7%. E-visit participation increased the percentage of patients with con- trolled blood pressure, as well as the percentage receiving statin and aspir- in therapy when indicated. The inten- sity of statin therapy also improved, with an increase in use of moderate- to high-intensity doses and a decrease in use of low-intensity doses.

There was no significant change in the mean number of antidiabetic agents prescribed (2.0 before enroll- ment versus 2.1 after enrollment); however, there were differences in the types of antidiabetic agents used (Table 2). Most patients who were treated with metformin or basal insu- lin were maintained on those thera-

Table 2. Primary and Secondary Outcomes Before and After Program Initiationa

Variable Before Program After Program p

Anticoagulation group (n = 9)

Fraction (%) INR values of 2.0–3.0 65/104 (63) 144/198 (73) 0.07

Fraction (%) INR values of <1.5 or >5.0 5/104 (5) 1/198 (0.5) 0.01

Diabetes group (n = 29)

Mean ± S.D. HbA 1c

concentration (%) 10.9 ± 1.9 7.5 ± 1.0 <0.0001

No. (%) patients with HbA 1c

concentration of <8% 1 (3) 20 (69) <0.0001

No. (%) patients with HbA 1c

concentration of <7% 0 9 (31) 0.0002

No. (%) patients with controlled blood pressure 15 (52) 27 (93) <0.001

No. (%) patients receiving aspirin 18 (86) 21 (100) 0.07

No. (%) patients receiving statin therapyb 23 (82) 28 (100) 0.06

Low-intensity 3 (13) 0 0.23

Moderate-intensity 10 (44) 16 (57) 0.19

High-intensity 10 (44) 12 (44) 0.79

Mean ± S.D. no. antidiabetic agents per patient 2.0 ± 0.8 2.1 ± 0.6 0.59

No. (%) patients receiving antidiabetic agent

Metformin 24 (83) 25 (86) >0.99

Sulfonylurea 13 (45) 4 (14) 0.2

DPP-IV inhibitor 2 (7) 2 (7) >0.99

Basal insulin 13 (45) 12 (41) >0.99

Bolus insulin 2 (7) 3 (10) >0.99

Mixed insulin 0 3 (10) 0.24

GLP-1 agonist 1 (3) 12 (41) <0.0001 aINR = International Normalized Ratio, HbA

1c = glycosylated hemoglobin, DPP-IV = dipeptidyl peptidase 4, GLP-1 = glucagon-like peptide 1.

bAll enrolled patients were prescribed or switched to a moderate- or high-intensity statin therapy, as recommended by the 2013 American College of Cardiology–American Heart Association guidelines.

DESCRIPTIVE RESEARCH REPORTS ELECTRONIC VISIT PROGRAM

906 AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018

pies. Fewer patients remained on a sulfonylurea, and more patients were initiated on a glucagon-like peptide-1 (GLP-1) receptor agonist during enrollment. Five patients (17%) reported at least 1 episode of hypoglycemia during e-visit partici- pation, but none required emergent medical attention.

Two patients receiving diabetes care through e-visits had uncon- trolled gout prior to enrollment. By initiating and adjusting allopurinol therapy, the CPP reduced gout flares by gradually decreasing uric acid con- centrations to a target range (from 9.3 to 5.4 mg/dL and from 9.9 to 4.9 mg/ dL in the 2 patients, respectively).

Encounters, clinical time, and billing. The CPP reported that pa- tient care responsibilities (including MyChart messaging with the patient, billing, sending prescriptions, and note documentation) for each e-visit were completed in 5–10 minutes for warfarin e-visits and 10–20 minutes for diabetes e-visits. Traditional in-person pharmacotherapy clinic appointments are allotted a 30-minute time frame. Each week the CPP blocked out 30 minutes to conduct a single pharma- cotherapy session covering an aver- age of 4.5 home INR monitoring pa- tients, which equated to 6.7 minutes per e-visit.

Services were provided and incident-to billing was conducted in conjunction with a supervising phy-

Table 3. Financial Outcomes

Anticoagulation Program

Variable Year 1 After Year

1 Diabetes Program

Clinical pharmacist practitioner time (hr/mo)

2.1 plus 13.5 for initial training sessions

2.1 38

No. patients/mo 9 9 29

Median revenue per patient ($) 44/mo 28/mo 43/mo (172 total)

Cost of care per patient ($) 19/mo 12/mo 18/mo (72 total)

Margin per patient ($)a 25/mo 16/mo 25/mo (100 total) aBased on a total of 2.4 e-visits and 1.4 in-person visits over a 4-month duration.

sician, who cosigned the encounter documents. Reimbursement rates were similar to the published North Caro- lina Medicare Fee Schedule rates.45 Table 3 summarizes the findings of the financial analysis.

At the time of data collection, the 9 patients in the anticoagulation group had participated in a total of 198 elec- tronic encounters from 2014 through 2016. The patients were enrolled on a rolling basis. CPT code G0248 was billed for each patient once after train- ing and documentation. E-visits were billed once monthly, representing 48% of all electronic encounters during the study period and resulting in billing of 12 e-visits per patient using CPT code 99444 annually.

In the diabetes group, at the time of data collection patients had par- ticipated in a total of 69 e-visits and 41 in-person visits after enrolling, for a mean of 2.4 e-visits (range, 1–13 vis- its) and a mean of 1.4 in-person vis- its (range, 0–4 visits) with a CPP. All e-visits were billed using CPT code 99444, and the in-person visits were billed according to the level of ser- vice (codes 99211–99214, with code 99213 typically used). With the use of code 99213 reimbursement rates for in-person visits and code 99444 rates for e-visits, the total revenue for managing the 29 patients was close to $5,000.

All study patients were insured. Al- though coverage for e-visits varied by

insurance carrier, patients were more commonly covered by private insur- ance companies versus government- funded insurance (e.g., Tricare, North Carolina Medicaid, Medicare Part A or Part B, federal employee plans). All study patients had insurance plans that covered CPT code 99444, which included the following plans: BCBS Blue Options/PPO/ADV, BCBS Blue Care HMO, BCBS Blue Value, Cov- entry Wellpath–Coventry Healthcare, Humana Medicare Adv, North Caro- lina State Health Plan, UMR, Unicare Commercial–Unicare Medicare Supp, United Healthcare, and United Health- care Medicare Adv.

Patient satisfaction. Nineteen patients (53%) participated in the vol- untary survey (Table 4). The median score was 39 out of 40 points, indicat- ing a high level of satisfaction. Patients reported the highest level of satisfac- tion in response to the questions relat- ed to the e-visit practice model (ques- tions 8–10).

Discussion

The goal of the study was to evalu- ate the impact of a pharmacist-led e-visit program on clinical outcomes, financial indicators, and patient satis- faction. CPP use of e-visit telemedicine technologies to deliver care resulted in significantly improved INR and HbA

1c

control for patients who participated in anticoagulation and diabetes man- agement, respectively. The program improved clinical care quality and financially resulted in a net positive margin, and patients reported high levels of satisfaction. To our knowl- edge, ours was the first study to evalu- ate chronic disease management by a pharmacist through e-visits.

Most studies addressing a phar- macist’s role in direct chronic disease management within a PCMH using telemedicine were conducted within the Department of Veterans Affairs healthcare system and involved the use of live CVT.16,47-49 Our study ex- panded on previous findings regard- ing real-time telehealth technology by including a novel e-visit model,

ELECTRONIC VISIT PROGRAM DESCRIPTIVE RESEARCH REPORTS

AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018 907

with nonurgent secure messaging, in a PCMH at a physician-owned aca- demic FMC.16,47

No patients in the warfarin PST group experienced an embolic or bleeding event, and the percentage of extreme INR values significantly decreased after program enrollment. Prior to e-visit participation, warfarin therapy was managed in the CPP-led pharmacotherapy clinic. After pro- gram implementation, the percent- age of INRs within the target range increased, albeit nonsignificantly, by about 10%. This finding was consis- tent with those of other studies com- paring standard in-person antico- agulation management via PST with Internet-based feedback.17,23-29 In our study the prescribed INR monitoring frequency was every 2 weeks, which varied from the intervals in most PST studies, which involved weekly moni- toring. This frequency was set to re- duce providers’ time burden and limit out-of-pocket costs that patients in- cur with increased testing frequency. Of note, the quality of warfarin man- agement in our study was like that in studies in which PST occurred more frequently.17-19,24,26,28,30-32 Despite dem- onstrated effectiveness, PST is often not implemented due to high pro- vider time demands and limited rev- enue.42-44 By using e-visit billing and systematic online secure messaging, the provider time was decreased to a mean of 14 minutes monthly, yielding a return of almost $30 for that time in addition to improved quality.

In the diabetes management group, our study demonstrated a sig- nificant mean HbA

1c reduction of 3.4%

through use of e-visits conducted by a CPP. This finding was consistent with findings in previously published analyses evaluating diabetes man- agement with various telemedicine technologies; however, in our study a CPP directed care.39,40 Only 9 patients (31%) had achieved a target HbA

1c

value of <7% at the time of data col- lection. Of note, HbA

1c was measured

less than 3 months after enrollment in 7 patients. Our findings are con-

sistent with results reported with pharmacist-conducted diabetes man- agement via CVT.46 E-visits are argu- ably even more convenient than CVT given that they can occur wherever Internet access or smartphone con- nectivity is available, at a time that is convenient for the patient, and with- out travel to a clinic. Most e-visits oc- curred after an initial in-person visit; therefore, these e-visits were an effec- tive follow-up alternative to in-person visits and nonreimbursable telephone consultations.50

In the diabetes group, significantly more patients achieved preventive care targets, including blood pres- sure control, statin use, and aspirin therapy, specified in accordance with guideline recommendations.51,52 Pa- tients were maintained on a mean of 2 antidiabetic medications, a figure essentially unchanged from baseline despite improving glycemic control. This finding highlighted the observa- tion that medication burden did not increase and thus did not contribute

to a reduction in patient adherence. After enrollment, fewer patients were maintained on a sulfonylurea, and more patients were initiated on GLP-1 receptor agonists, which pose a low- er risk of hypoglycemia and weight gain.53 Although weight loss has been reported by patients receiving GLP-1 agonists,54-56 it was not measured or documented in our study. The low frequency of patients experiencing hypoglycemia at any point after en- rollment (n = 5) and the lack of reports of severe hypoglycemia reinforce our stance that e-visits may be a safe al- ternative to in-person visits. Of note, prestudy reports of hypoglycemia were not collected.

Within a PCMH, clinical and finan- cial outcomes and patient satisfac- tion are important indicators of the quality of services. Patient satisfaction is positively correlated with health outcomes, safety, and adherence to medications and preventive care mea- sures.57 Of the patients who responded to the survey, all indicated high levels

Table 4. Patient Satisfaction Survey Results (n = 19)a

Survey Item Median

Score (IQR)

1. My pharmacist advises me on the proper use of my medicines.

4 (4–4)

2. My pharmacist advises me on the adverse (side) effects of my medicines. 4 (4–4)

3. I have confidence in my pharmacist. 4 (4–4)

4. My pharmacist is available to answer my questions. 4 (3–4)

5. My pharmacist helps me with the arrangements necessary to obtain my medicines and supplies. 4 (3–4)

6. My pharmacist is aware of my treatment-related needs. 4 (3–4)

7. My pharmacist responds to my treatment-related needs. 4 (3–4)

8. How comfortable did you feel in communicating with someone by MyChart messaging? 4 (4–4)

9. How convenient was the MyChart messaging? 4 (4–4)

10. Was the lack of physical presence of the provider acceptable? 4 (4–4)

Total score (maximum = 40) 39 (34–40) aQuestions 1–7 were modified from the Satisfaction with Pharmacist scale.46 Questions

8–10 were specific to the e-visit program. Scores indicated response to survey statements or questions: 0 = not at all, 1 = a little bit, 2 = somewhat, 3 = quite a bit, 4 = very much. IQR = interquartile range.

DESCRIPTIVE RESEARCH REPORTS ELECTRONIC VISIT PROGRAM

908 AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018

of satisfaction (Table 4). Only part of the SWiP survey (questions 1–7) has been validated46; however, it was im- portant to receive feedback about the e-visit service. A similar modified SWiP survey was used by Maxwell and colleagues.47 In our study, the highest levels of satisfaction were related to questions 8–10, which were specific to the e-visit model. Patients felt com- fortable communicating with a CPP through MyChart messaging, found the service convenient, and accepted the lack of a physical CPP presence. Patient satisfaction with pharmacist- led services in the study was consis- tent with satisfaction levels reported in studies of other pharmacist-led services for chronic disease manage- ment.47,58-60 More than half of the pa- tients responded to the patient satis- faction survey, which was higher than the FMC’s average rate of response to patient surveys (e.g., less than 10% for Press Ganey surveys [Press Ganey Associates, Inc., South Bend, IN]).

The study was not without limi- tations. The small patient sample limited the external validity of the findings. However, it is notable that a statistically significant difference in outcomes was found despite the small sample size. Given require- ments regarding patient access to and use of MyChart, competency in us- ing a home INR device or glucometer, and insurance coverage for e-visits, participating patients may have had higher-than-typical levels of health literacy and access to care. The strict inclusion criteria likely contributed to a small sample size. In addition, our patient population was young relative to populations in similar stud- ies.47 Based on the above points, the study results cannot be generalized to a broad chronic disease population. Since the study, insurance coverage for e-visits may have expanded to in- clude more carriers. Patients were in- volved in varying levels of care prior to enrollment. Anticoagulation patients were managed by a CPP for at least 3 months before enrollment. Given that many patients with diabetes par-

ticipated in e-visits and in-person visits, it is difficult to strictly assess the impact of e-visits alone. All INR values required manual reporting to the monitoring device manufacturer. Using a wirelessly connected moni- tor could reduce reporting errors. The same home testing device was used by all enrolled anticoagulation patients to improve monitoring efficiency, ac- curacy, and consistency for the CPP. While direct revenue from CPP bill- ing was collected, further analysis to evaluate the economic impact of the e-visit program (including indi- rect and direct revenue) on a larger scale is warranted. Although data regarding screening of patients were not collected, many patients who declined the service reported a pref- erence for in-person visits. The study was retrospective in design, and its findings could be strengthened by evaluating outcomes in a prospec- tive randomized clinical trial enroll- ing a larger sample of patients from a more diverse population. A larger sample size would also facilitate multivariate assessment of the influ- ence of factors.

Conclusion

An online e-visit model for warfa- rin monitoring was an efficient, safe, and cost-effective method for imple- menting PST. Pharmacist-led man- agement of diabetes through e-visits, often in combination with in-person visits, generated revenue while signifi- cantly improving clinical outcomes.

Acknowledgments The authors thank, Pharm.D. student, Sarah Chou for assisting with insurance coverage inquiries and conducting patient satisfaction surveys.

Disclosures The authors have declared no potential conflicts of interest.

References 1. Field MJ, ed., for Institute of Medicine

Committee on Evaluating Clinical Ap- plications of Telemedicine. Telemedi- cine: a guide to assessing telecommu- nications in health care (1996). www.

ncbi.nlm.nih.gov/books/NBK45440/ (accessed 2016 Dec 8).

2. Department of Veterans Affairs. VA telehealth services. www.telehealth. va.gov/ (accessed 2016 Dec 8).

3. Russo JE, McCool RR, Davies L. VA telemedicine: an analysis of cost and times savings. Telemed J E Health. 2016; 22:209-15.

4. Mehrotra A, Paone S, Martich GD et al. A comparison of care at e-visits and physician office visits for sinus- itis and urinary tract infections. JAMA Intern Med. 2013; 173:72-4.

5. Adamson SC, Bachman JW. Pilot study of providing online care in a primary care setting. Mayo Clin Proc. 2010; 85:704-10.

6. Rohrer JE, Angstman KB, Adamson SC et al. Impact of online primary care visits on standard costs: a pilot study. Popul Health Manag. 2010; 13(2):59-63.

7. Courneya PT, Palattao KJ, Gallagher JM. HealthPartners’ online clinic for simple conditions delivers sav- ings of $88 per episode and a high patient approval. Health Aff. 2013; 32:385-92.

8. Bellon JE, Stevans JM, Cohen SM et al. Comparing advanced practice providers and physicians as providers of e-visits. Telemed J E Health. 2015; 21:1019-26.

9. Bishop MA, Streiff MB, Ensor CR et al. Pharmacist-managed International Normalized Ratio patient self-testing is associated with increased time in therapeutic range in patients with left ventricular assist devices at an academic medical center. ASAIO J. 2014; 60:193-8.

10. Bungard TJ, Gardner L, Archer SL et al. Evaluation of a pharmacist- managed anticoagulation clinic: improving patient care. Open Med. 2009; 3:e16-21.

11. Challen L, Agbahiwe S, Cantieri T et al. Impact of point-of-care im- plementation in pharmacist-run anticoagulation clinics within a community-owned health system: a two-year retrospective analysis. Hosp Pharm. 2015; 50:783-8.

12. Chilipko AA, Norwood DK. Evalu- ating warfarin management by pharmacists in a community teach- ing hospital. Consult Pharm. 2014; 29:95-103.

13. Garton L, Crosby JF. A retrospective assessment comparing pharmacist- managed anticoagulation clinic with physician management using International Normalized Ratio sta- bility. J Thromb Thrombolysis. 2011; 32:426-30.

ELECTRONIC VISIT PROGRAM DESCRIPTIVE RESEARCH REPORTS

AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018 909

14. Holden J, Holden K. Comparative effectiveness of general practitioner versus pharmacist dosing of patients requiring anticoagulation in the community. J Clin Pharm Ther. 2000; 25:49-54.

15. Young S, Bishop L, Twells L et al. Comparison of pharmacist managed anticoagulation with usual medical care in a family medicine clinic. BMC Fam Pract. 2011; 12:88.

16. Singh LG, Accursi M, Black KK. Im- plementation and outcomes of a pharmacist-managed clinical video telehealth anticoagulation clinic. Am J Health-Syst Pharm. 2015; 72:70-3.

17. Bereznicki LR, Jackson SL, Peterson GM. Supervised patient self-testing of warfarin therapy using an online system. J Med Internet Res. 2013; 15:e138.

18. Bussey HI, Bussey M. Warfarin man- agement: International Normalized Ratio self-testing and warfarin self- dosing. Circulation. 2012; 126:e52-4.

19. Bussey HI. Transforming oral antico- agulation by combining International Normalized Ratio (INR) self testing and online automated manage- ment. J Thromb Thrombolysis. 2011; 31:265-74.

20. Gardiner C, Williams K, Mackie IJ et al. Patient self-testing is a reliable and acceptable alternative to laboratory INR monitoring. Br J Haematol. 2005; 128:242-7.

21. Matchar DB, Jacobson A, Dolor R et al. Effect of home testing of Interna- tional Normalized Ratio on clinical events. N Engl J Med. 2010; 363:1608- 20.

22. Meyer S, Frei CR, Daniels KR et al. Impact of a new method of warfarin management on patient satisfaction, time, and cost. Pharmacotherapy. 2013; 33:1147-55.

23. Christensen H, Lauterlein JJ, Sørensen PD et al. Home manage- ment of oral anticoagulation via telemedicine versus conventional hospital-based treatment. Telemed J E Health. 2011; 17:169-76.

24. Desantis G, Hogan-Schlientz J, Liska G et al. STABLE results: warfarin home monitoring achieves excellent INR control. Am J Manag Care. 2014; 20:202-9.

25. Ferrando F, Mira Y, Contreras MT et al. Implementation of SintromacWeb, a new Internet-based tool for oral anticoagulation therapy telecontrol: study on system consistency and pa- tient satisfaction. Thromb Haemost. 2010; 103:1091-101.

26. O’Shea SI, Arcasoy MO, Samsa G et al. Direct-to-patient expert system and

home INR monitoring improves con- trol of oral anticoagulation. J Thromb Thrombolysis. 2008; 26:14-21.

27. Ryan F, Byrne S, O’Shea S. Random- ized controlled trial of supervised patient self-testing of warfarin therapy using an Internet-based ex- pert system. J Thromb Haemost. 2009; 7:1284-90.

28. Watzke HH, Forberg E, Svolba G et al. A prospective controlled trial comparing weekly self-testing and self-dosing with the standard man- agement of patients on stable oral anticoagulation. Thromb Haemost. 2000; 83:661-5.

29. Harper P, Pollock D. Improved anti- coagulant control in patients using home International Normalized Ratio testing and decision support provided through the Internet. Intern Med J. 2011; 41:332-7.

30. Menéndez-Jándula B, Souto JC, Oliver A et al. Comparing self- management of oral anticoagulant therapy with clinic management: a randomized trial. Ann Intern Med. 2005; 142:1-10.

31. Stafford L, Peterson GM, Bereznicki LR et al. Clinical outcomes of a collaborative, home-based postdis- charge warfarin management service. Ann Pharmacother. 2011; 45:325-34.

32. Bloomfield HE, Krause A, Greer N et al. Meta-analysis: effect of patient self-testing and self-management of long-term anticoagulation on major clinical outcomes. Ann Intern Med. 2011; 154:472-82.

33. Collier IA, Baker DM. Implementa- tion of a pharmacist-supervised out- patient diabetes treatment clinic. Am J Health-Syst Pharm. 2014; 71:27-36.

34. Fera T, Bluml BM, Ellis WM. Diabetes Ten City Challenge: final economic and clinical results. J Am Pharm Assoc. 2009; 49:383-91.

35. Bunting BA, Cranor CW. The Asheville Project: long-term clinical, human- istic, and economic outcomes of a community based medication thera- py management program for asthma. J Am Pharm Assoc. 2006; 46:133-47.

36. Wubben DP, Vivian EM. Effects of pharmacist outpatient interventions on adults with diabetes mellitus: a systematic review. Pharmacotherapy. 2008; 28:421-36.

37. Sullivan J, Jett BP, Cradick M, Zuber J. Effect of clinical pharmacist interven- tion on hemoglobin A1C reduction in veteran patients with type 2 diabetes in a rural setting. Ann Pharmacother. 2016; 50:1023-7.

38. Wang Y, Weo QQ, Ko Y. Economic evaluations of pharmacist-managed

services in people with diabetes mel- litus: a systematic review. Diabet Med. 2016; 33:421-7.

39. Flodgren G, Rachas A, Farmer AJ et al. Interactive telemedicine: effects on professional practice and health care outcomes. Cochrane Database Syst Rev. 2015; 9:CD002098.

40. Faruque LI, Wiebe N, Ehteshami- Afshar A et al. Effect of telemedicine on glycated hemoglobin in diabetes: a systematic review and meta-analy- sis of randomized trials. CMAJ. 2016; 189:E341-64.

41. North Carolina Board of Pharmacy. Clinical pharmacist practitioners. www .ncbop.org/pharmacists_cpp. htm (accessed 2017 Feb 5).

42. Phibbs CS, Love SR, Jacobson AK et al., for the THINRS Executive Committee and site investigators. At-home versus in-clinic INR moni- toring: a cost-utility analysis from the Home INR Study (THINRS). J Gen Intern Med. 2016; 31:1061-7.

43. Jennings I, Kitchen D, Keeling D et al., for the BCSH Committee. Patient self- testing and self-management of oral anticoagulation with vitamin K antagonists: guidance from the British Committee for Standards in Haema- tology. Br J Haematol. 2014; 167:600-7.

44. Ansell J, Jacobson A, Levy J et al., for the International Self-Monitoring Association for Oral Anticoagula- tion. Guidelines for implementa- tion of patient self-testing and patient self-management of oral anticoagulation. International consensus guidelines prepared by International Self-Monitoring Asso- ciation for Oral Anticoagulation. Int J Cardiol. 2005; 99(1):37-45.

45. Centers for Medicare and Medicaid Services. 2017 Medicare physician fee schedule. www.cms.gov/apps/ physician-fee-schedule/search/ search-results.aspx?Y=0&T=0&HT=1 &CT=3&H1=99444&H2=G0250&H3= G0248&H4=99211&H5=99213&M=5 (accessed 2017 Mar 3).

46. Hernandez L, Hang SH, Cella D et al. Development and validation of the Satisfaction with Pharmacist scale. Pharmacotherapy. 2000; 20:837-43.

47. Maxwell LG, McFarland MS, Bakcer JW, Cassidy RF. Evaluation of the impact of pharmacist-led telehealth clinic on diabetes-related goals of therapy in a veteran population. Pharmacotherapy. 2016; 36:348-56.

48. Aberger E, Migliozzi D, Follick MJ et al. Enhancing patient engagement and blood pressure management for renal transplant recipients via home

DESCRIPTIVE RESEARCH REPORTS ELECTRONIC VISIT PROGRAM

910 AM J HEALTH-SYST PHARM | VOLUME 75 | NUMBER 12 | JUNE 15, 2018

electronic monitoring and Web- enabled collaborative care. Telemed J E Health. 2014; 20:850-4.

49. Ishani A, Christopher J, Palmer D et al. Telehealth by interprofessional team in patients with CKD: a ran- domized controlled trial. Am J Kidney Dis. 2016; 68:41-9.

50. Weidman-Evans E, Evans J, Eastwood R, Fort A. Implementation of a pharmacist-run telephonic insulin titration service. J Am Pharm Assoc. 2012; 52:e266-72.

51. American Diabetes Association. Standards of medical care in diabetes—2017. http://professional. diabetes.org/sites/professional. diabetes.org/files/media/dc_40_s1_ final.pdf (accessed 2017 Mar 3).

52. Stone NJ, Robinson J, Lichtenstein AH et al. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the Ameri- can College of Cardiology/American Heart Association Task Force on Prac- tice Guidelines. Circulation. 2014; 129(25, suppl 2):S1-45.

53. Inzucchi SE, Bergenstal RM, Buse JB et al. Management of hyperglycemia in type 2 diabetes: a patient-centered ap- proach. Diabetes Care. 2012; 35:1364- 79.

54. UKPDS Group. UK Prospective Diabetes Study 7: response of fast- ing plasma glucose to diet therapy in newly presenting type II diabetic patients. Metabolism. 1990; 39:905- 12.

55. Goldstein DJ. Beneficial health effects of modest weight loss. Int J Obes Relat Metab Disord. 1992; 16:397-415.

56. Pastors JG, Warshaw H, Daly A et al. The evidence for the effectiveness of medical nutrition therapy in diabetes management. Diabetes Care. 2002; 25:608-13.

57. Doyle C, Lennox L, Bell D. A system- atic review of evidence on the links between patient experience and clinical safety and effectiveness. BMJ Open. 2013; 3:e001570.

58. Schuessler TJ, Ruisinger JF, Hare SE et al. Patients satisfaction with pharmacist-led chronic disease state management programs. J Pharm Prac. 2016; 29:484-9.

59. McFarland MS, Wallace JP, Parra J, Baker J. Evaluation of patient satisfac- tion with diabetes management provided by clinical pharmacists in the patient-centered medical home. Patient. 2014; 7:115-21.

60. Bishop L, Young S, Twells L et al. Patients’ and physicians’ satisfaction with a pharmacist managed anticoag- ulation program in a family medicine clinic. BMC Res Notes. 2015; 8:1-7.

Appendix—Assumptions and codes used in financial analysisa

Weeks per month: 4

Pharmacist salary: $55

CPP clinical time for PST e-visits: 6.7 min

CPP clinical time for diabetes e-visit: 15 min

CPP clinical time for initial home INR training session (performed as a group visit when possible): 90 min

CPT code for in-person CPP visits: 99213

Healthcare Common Procedure Coding System code and 2017 North Carolina Medicare rates45

99211: office/outpatient visit established, $21

99212: office/outpatient visit established, $44

99213: office/outpatient visit established, $74

G0248: demonstrate use of home INR monitor, $111. Average reimbursement rate in study was $80 more than Medicare rates, and higher rate was used for financial analysis

G0250: physician INR test review, interpretation, and management: $9

99444: online evaluation and management, established patient: $28. (Not covered by Medicare Part A and Part B. Reimbursement rate based on internal data.)

aCPP = clinical pharmacist practitioner, PST = patient self-testing, INR = International Normalized Ratio.

Copyright of American Journal of Health-System Pharmacy is the property of American Society of Health System Pharmacists and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.