Secondary Logo

Journal Logo

Original Articles: Gastroenterology

Impact of a Medical Scribe Program on Outpatient Pediatric Gastroenterology Clinic

Rahhal, Riad; Goad, Lindsay; Bishop, Warren

Author Information
Journal of Pediatric Gastroenterology and Nutrition: February 2021 - Volume 72 - Issue 2 - p 220-225
doi: 10.1097/MPG.0000000000002954
  • Free

Abstract

See “Medical Scribes: A First Step” by Huang on page 181.

What Is Known

  • Medical scribe experiences have been described in primary care, emergency room, and some adult subspecialty settings.
  • Medical scribes have been proposed to improve efficiency of clinical operations and address concerns about rising physician burnout.

What Is New

  • Medical scribes in an academic outpatient setting can reduce pediatric gastroenterologist time needed per encounter with shorter lag time in note completion.
  • The presence of medical scribes did not demonstrate any measurable negative impact on patient satisfaction or billing complexity.

The expanding use of electronic medical record (EMR) systems and gradual elimination of traditional transcription services in favor of typing and voice recognition software has shifted the burden of clinical documentation to clinicians. Research on physician time utilization in emergency medicine and internal medicine has shown that providers spend significantly more time on medical documentation than on direct patient contact (1,2). Studies from primary care specialties have also demonstrated that clinicians are spending more time on documentation, which compromises personal time raising concerns about career satisfaction and burnout (3,4).

To improve efficiency of clinical operations and address concerns about rising physician burnout, utilization of medical scribes or medical documentation assistants has been proposed. Tasks assigned to a medical scribe often focus on medical documentation but can possibly extend to further provider assistance, resulting in different training requirements (5,6). The Joint Commission, a national US organization for program accreditation for hospitals and healthcare organizations, defines a scribe as an “unlicensed, certified, or licensed person who provides documentation assistance to a physician or other licensed independent practitioner consistent with the roles and responsibilities defined in the job description, and within the scope of his or her certification or licensure.” The Joint Commission describes other potential scribe roles, which may include aiding in navigating the EMR, locating information (such as test results) and supporting documentation for medical record coding but it does not support using scribes to enter orders for providers (7).

Medical scribe services have been introduced in some specialties especially in emergency departments (EDs) to improve efficiency (8,9). A systematic review and meta-analysis, including 17 studies, mostly from academic EDs, showed overall positive effects on provider productivity and satisfaction (10). In gastroenterology, the literature on medical scribes is extremely scarce and includes an inpatient experience showing enhanced documentation of medical complexity, resulting in improved reimbursement (11). For non-ED pediatric-related experiences, the published literature is limited to a single study in a general pediatric inpatient setting (12). That study did not achieve its primary goal of earlier discharges but was associated with increased provider and parent satisfaction, more complete and succinct notes, and less perceived provider documentation time. There is limited literature in adults suggesting that high patient satisfaction can still be maintained with a scribe program even with shorter provider-patient contact time. Overall, there remains a gap in knowledge about the impact of scribes on patient satisfaction especially in the pediatric population.

To our knowledge, there is no prior published research on the impact of medical scribes on outpatient pediatric subspecialty clinic operations. The primary aim of our study was to address this knowledge gap. We assessed the impact of a medical scribe program in a pediatric gastroenterology outpatient clinic on patient clinic flow and provider revenue (by medical documentation billing complexity). Secondary aims included assessing the effect of the scribe program on patient satisfaction and medical documentation efficiency.

METHODS

Medical Scribe Program and Clinic Setting

Medical students were recruited to scribe in the Pediatric Gastroenterology Clinic during the summer between their first and second years of medical school training. Fourteen medical scribes were trained over the course of approximately 10 hours. This included training on the EMR system, shadowing an experienced scribe, and performing a supervised scribe shift (half-day in clinic). Scribes attended standard half-day clinics (8.00 a.m. to noon, or 1.00 p.m. to 5.00 p.m.). At the beginning of each encounter, scribes were introduced to families as medical documentation assistants who were part of the healthcare team. Their role included only documentation of history and examination findings (the latter dictated by the physician during the examination), using a computer inside the examination room. During each shift, a medical scribe was assigned to work with 1 provider. The program extended from April to September during the summer of 2014.

The scribe program provided documentation support to 7 pediatric gastroenterology faculty physicians, 2 fellows, and 1 physician assistant (PA). All faculty physicians had full time positions and had been in clinical practice for a median of 7 years (range 1–25 years). The scribe program was limited to outpatient visits conducted in the Pediatric Subspecialty Clinic at the main university campus. The clinic setting included an attached venipuncture/laboratory station, procedure/endoscopy suite, and infusion area allowing patients to see 1 or more subspecialty providers, have laboratory testing performed, undergo endoscopy procedures, and receive prescribed infusions (such as infliximab or iron infusions) on the same day whenever possible. Typical clinic encounter templates allowed 20 minutes for established (return) patient visits and 40 minutes for new patient visits. Provider schedules, including duration of clinic encounters, were not altered in anticipation of having a scribe. assignment of scribes to provider clinics was based on provider clinic schedules and the availability of scribes.

Study Design and Patient Cohort

This was a retrospective cohort study of outpatient pediatric gastroenterology encounters in the location specified above. Patients were identified through clinic charts and billing records. The institutional review board approved the study. Inclusion criteria included evaluation and management (E&M) outpatient encounters with faculty physicians from the Pediatric Gastroenterology division during the scribe program duration. Exclusion criteria included:

  • 1. Encounters with nonfaculty physicians (fellows or PA)
  • 2. Encounters for laboratory testing, nursing services (including infusions), procedures and imaging studies
  • 3. Encounters with total clinic time lasting >90 minutes (considered outliers, likely because of multiple subspecialty visits on same day, postvisit tests and/or nursing services, or improper use of dashboard tracking system)
  • 4. Encounters with total clinic time lasting <2 minutes (considered outliers, likely because of improper use of dashboard tracking system)
  • 5. Encounters without outpatient billing codes (considered inappropriately assigned and/or billed)

Patient encounters were categorized as new or established E&M services depending on prior encounters with the provider group, as defined by the Centers for Medicare & Medicaid Services.

Outcome Measures of Interest Included

  • 1. Documenting medical note billing complexity
  • 2. Number of delinquent dictations
  • 3. Efficiency of patient flow through clinic (defined as the time from physician entering patient room to checkout from clinic)
  • 4. Patient satisfaction
  • 5. Faculty physician satisfaction with the scribe program

Medical Note Complexity and Billing

In the United States, outpatient E&M services are classified as new (includes consult and self-referral) or established encounters. Each service is assigned a CPT (Current Procedural Terminology) code based on documentation of key components (history, physical examination, complexity of medical decision-making), which categorizes the service level into 1 of 5 levels (1–5). High service levels correspond to higher care complexity, allowing assignment of higher work relative value units (wRVUs) to providers as well as increased billing. For this study, note complexity was assessed to determine impact on revenue, specifically whether medical billing was different when scribes were involved. High note complexity was defined as a level 4 or 5, and both new consult and new self-referral encounters were treated as 1 category. We used the 2016 Medicare RVU reimbursement rate for wRVU calculation. We abstracted CPT codes for each outpatient encounter from electronic billing records. Comparisons were made in note complexity and associated wRVUs per encounter type (new vs established) between groups (with or without medical scribe).

Lag Time in Note Completion

The effect of medical scribe presence on charge lag time was assessed by tracking time to note completion and percent of delinquent notes. A note was considered delinquent if documentation was not finalized in the EMR system by the responsible faculty physician after 7 days from the encounter date.

Time Tracking System

Data on patient flow through the clinic were extracted using an electronic dashboard integrated in the EMR system to track patient flow. This dashboard is used by check-in staff, nursing staff and physicians, and tracks the following time points: time at patient check-in to clinic (by front desk staff), time patient is retrieved from waiting room (by nursing staff), time care provider enters the room (by physician or nursing staff) and time at discharge from clinic (by front desk staff). For this study, total clinic time was defined as time from physician entering the patient room to checkout from the clinic at the front desk. This time frame was chosen as it: represents more accurate assessment of physician time utilization, avoiding variability related to delays in placing patients inside examination rooms, and had the highest reliability of documentation on the dashboard (least missing data, as described later).

Satisfaction

Patient satisfaction surveys were provided at all encounters, starting 2 weeks after the scribe program implementation, to assess family satisfaction with the visit and acceptability of a medical scribe in the examination room. Surveys were distributed over a period of 8 weeks, collected by nursing or front desk staff and included questions on the following:

  • 1. Concern about presence of scribe in the examination room using a 5-point Likert type scale (1 = very concerned to 5 = no concerns at all, or nonapplicable if no scribe present)
  • 2. Perceived delays for more than a few minutes at different stages of clinic encounter (including check-in, waiting room, and inside examination room)
  • 3. Perceived overall quality of service using a 5-point Likert type scale (1 = very poor to 5 = excellent)

Provider satisfaction surveys were also given to faculty physicians at the end of the scribe program to assess perceived program benefits and included questions on the following (high score defined as score of 5, on a scale of 1–5):

  • 1. Interest in continuing the scribe program (1 = Strongly disagree to 5 = Strongly agree)
  • 2. Did scribe program make documentation easier (1 = Strongly disagree to 5 = Strongly agree)

Statistical Analysis

Continuous variables were described as means and standard deviations for normally distributed data and medians and interquartile ranges for data with skewed distribution and were compared by a Student t-test or the Wilcoxon signed-ranks test, respectively. Categorical variables were described as percentages and proportions and were compared by a chi-square test. We conducted a linear regression analysis, with total clinic time as a continuous outcome variable with specific provider (1 through 7) and presence of scribe as categorical predictor variables. For the analysis of quality-of-service survey data, Likert-scale scores 4 and 5 were grouped together (categorized as Excellent/Good service) whereas scores 1, 2, and 3 were grouped together (categorized as Fair/Poor service). A probability value <0.05 was considered significant. Statistical analysis was performed with SAS 9 statistical software (SAS Institute, Cary, NC).

RESULTS

Study Cohort

After excluding patients seen by nonfaculty physicians and excluding 10 encounters without outpatient billing codes, a total of 1970 clinic encounters were included in the study. This cohort included 529 new and 1441 established encounters of which 164 new and 529 established encounters (35% of total) were conducted in the presence of a scribe. Patient characteristics included a median age of 9.8 years with 53% girls without statistically significant differences in basic demographics (age or sex) between scribe and nonscribe patient cohorts.

Medical Note Complexity and Billing

A total of 1970 encounters were included in the analysis on complexity level documentation and associated wRVUs. No statistically significant differences were noted in assigned note complexity when comparing encounters assigned levels 4 or 5 between encounters conducted with and without a scribe present (Table 1). This was the case for new encounters and established encounters. Assigned wRVUs per encounter type (new and established), which is based on medical note complexity, were not different between groups.

TABLE 1 - Medical note complexity and clinic efficiency
All encounters Encounters with scribe Encounters without scribe P value
Established and new encounters, N 1970 693 1277
Established encounters only, N 1441 529 912
New encounters only, N 529 164 365
Revenue
 % High billing note complexity
  Established encounters only 85% 83% 86% 0.14
  New encounters only 94% 94% 93% 0.67
 Assigned RVUs per encounter, median (IQR)
  Established encounters only 1.5 (0) 1.5 (0) 1.5 (0) 0.14
  New encounters only 2.4 (0.6) 2.4 (0.6) 2.4 (0.6) 0.33
Patient flow
 Total clinic time, median (IQR), minutes
  Established and new 22 (17) 21 (14.5) 23 (17) 0.002
  Established encounters only 20 (14) 18 (13) 21 (16) 0.003
  New encounters only 29 (17) 29 (17) 29.5 (17.5) 0.81
IQR = interquartile range; N = number of encounters, high billing complexity for note documentation refers to level 4 or 5 based on assigned current procedural terminology (CPT) code; RVUs = relative value units.

Lag Time in Note Completion

Time to note completion was statistically shorter for new patient encounters in the presence of a medical scribe with a median time to note completion of 2 days (IQR 4) versus 3 days (IQR 6) with and without a medical scribe, respectively (P = 0.044). For established encounters, time to note completion was similar in both groups with a median of 1 day (IQR 5–6) (P = 0.50). Significantly, more notes were finalized by the third day after the clinic encounter when a scribe was present (63% with scribe vs 57% without scribe, P = 0.02). Note delinquency (defined as unfinalized notes 7 days after encounter) was not statistically different between groups at 16% and 18% with and without scribes, respectively (P = 0.4). A wide range of note delinquency rates was seen among individual providers (Fig. 1). One provider had significantly lower note delinquency with scribes (3% vs 8%, P = 0.03) whereas another had significantly more note delinquency with scribes (59% vs 41%, P = 0.002). When comparing encounters conducted with and without scribes, note delinquency did not differ based on encounter type (among new and among established encounters).

F1
FIGURE 1:
Comparison of percent note delinquency for all providers and for individual providers with and without the presence of a scribe. P ≤ 0.05, ∗∗ P ≤ 0.01.

Clinic Efficiency

Dashboard use to track patients required direct input by a healthcare team member and was most reliable at the time points of the provider entering the examination room and when the patient was discharged from clinic, both of which were documented in 60% of encounters. As stated previously, this time frame was chosen to represent total clinic time. After excluding outliers (>90 minute and <2 minute encounters) and those with missing dashboard data, 1175 encounters were analyzed, including 432 encounters (37%) conducted in the presence of medical scribes. Percent missing tracking data was not statistically different between encounters conducted with and without a scribe (P = 0.6). Total clinic time for established encounters was statistically shorter in the presence of a medical scribe (median 18 vs 21 minutes, P < 0.01), corresponding to a 14% reduction. For new encounters, total clinic time was not statistically different with and without a scribe (median 29 vs 29.5 minutes, respectively, P = 0.81). When all encounters were combined, total clinic time remained statistically shorter when a medical scribe was present (median 21 vs 23 minutes, P < 0.001). When assessing total clinic time by individual provider (Fig. 2), a wide range of differences was noted between encounters conducted with and without scribes with 2 providers showing significantly shorter total clinic time when a scribe was present.

F2
FIGURE 2:
Comparison of total patient clinic time for all providers and for individual providers with and without the presence of a scribe. P ≤ 0.05, ∗∗∗ P ≤ 0.001.

Providers conducted a wide range of encounter numbers in the presence of scribes (22--153 encounters per provider). Each provider also worked with multiple scribes (range 4–9 scribes) during the study time period. No statistical difference was noted in total clinic time when comparing among the different scribes. As scribe assignment to provider clinic was not randomized, we conducted linear regression analysis and adjusted for specific provider (1 through 7), which showed that the presence of a scribe significantly decreased encounter time by 2 minutes (P = 0.02), irrespective of provider.

Satisfaction

Satisfaction surveys were completed by patients for 298 encounters (15% of total), which included 118 encounters (40%) in the presence of a medical scribe. This corresponded to a response rate of 17% among encounters conducted in the presence of a scribe. Survey responses showed that the majority had no concerns about the presence of a medical scribe in the examination room with 94% reporting no concerns at all (score of 1) with mean score of 1.1 (range 1–5). Two survey responders noted dissatisfaction (scores of 4 and 5) with the presence of a medical scribe but no specific comments were provided. The survey included a question to rate the experience with the provider (on a scale of 1--5, Poor = 1, Excellent = 5). The mean score was similar at 4.8 with and without scribes (P = 0.51). The analysis of overall quality of service showed uniformly high satisfaction scores noted by 100% and 99% of respondents in the presence or absence of a medical scribe, respectively (P = 1.0). Survey questions about perceived delays at 1 or more stages of the clinic visit (including check-in, waiting room, examination room, or lab) noted reports of delays at different clinic locations, ranging between 2% and 13% of locations but without statistically significant differences between encounters with or without a medical scribe.

All faculty completed the satisfaction surveys. these indicated high satisfaction with the scribe program, and all reported being in favor of continuing the scribe program (high score of 5 out of 5 reported by 71% of providers, mean score 4.7, range 4--5). Similarly, providers perceived that the presence of scribes made timely documentation completion easier (mean score 4.7, range 4--5). Objective time assessment of time spent on documentation was, however, not part of this study.

DISCUSSION

This study is the first to describe small but measurable real-life impacts of a medical scribe program in an outpatient academic pediatric subspecialty setting. It was also unique in using local students as scribes. We demonstrated that the scribe program had benefits on patient clinic flow with reduced clinic time needed especially for established encounters, highlighting an opportunity to increase revenue. In addition, our scribe program was associated with modest reduction in time to note completion, an attractive advantage for healthcare providers, without negative effects on medical note complexity documentation and associated RVUs. On the basis of survey responses, the scribe program was well received by patients and participating physicians. It is important to recognize that the patient survey response rate was only 15% among all encounters and 17% among those conducted in the presence of a scribe.

The role of medical scribes is to accurately document encounters between medical providers and patients, utilizing note templates to record histories and examination findings, with provider supervision, in real time. Scribes can also perform other tasks, such as searching for test results and providing patient summaries (13). There has been a documented increase in scribe utilization in primary care services, as well as in some adult subspecialty clinics (outside EDs), including adult urology, rheumatology, endocrinology, and dermatology. All showed a positive impact of scribes (14–17). Our findings are consistent with some of the published literature in other specialties. Arya et al (9) showed increased patient number seen and RVUs generated per hour in ED setting associated with the use of scribes. Another study in an ED showed a 13% overall increase in patient encounters per hour with a range of 6% to 15% between different providers (18). A prospective controlled study in outpatient adult cardiology compared standard visits (20 and 40-minute slots for established and new encounters, respectively) to a scribe system (15 and 30-minute slots for established and new encounters, respectively) showing increased number of patients seen and wRVUs generated per hour while maintaining high patient satisfaction (14). In this study, the authors noted shorter overall provider-patient contact time although provider-patient interaction was greater with scribes, possibly explaining the high satisfaction among patients. In our study, we also found that shorter clinic time did not have any negative impact on patient satisfaction. It is important to note that not all studies showed a positive effect of scribes on efficiency of clinical operations (10,19). Therefore, it is relevant to consider specific local environments and cultures within each clinical setting to maximize success and buy-in. Our results demonstrated variability in scribe impact on clinic time based on individual provider with some providers experiencing more efficiency but not others. Based purely on clinic efficiency, a clinical practice may have to be selective in whom to offer scribe services among its providers to optimize the benefits of this service.

Having a scribe service comes at a cost, and some would question the sustainability of such a program without a cost-benefit analysis, which was outside the scope of this study. We hypothesize that the costs of a scribe can be offset by additional patient encounters to each half-day. It is important to consider multiple factors when conducting a cost-benefit analysis. This includes the expense of the scribe services, which can vary widely depending on whether this is based on local clinic/hospital scribe recruitment and training, adding to roles and responsibilities of existing clinic staff (such as medical assistants), or contracting with an independent company. In addition, as not all providers demonstrated improved clinic efficiency when adding a scribe, selecting specific providers to work with scribes may help maximize the financial benefits. When weighing the costs and benefits, it is important to consider measurable and unmeasurable metrics. In our study, we determined potential gains in time efficiency, which can be translated to adding more patients per clinic half day to increase revenue. There was no loss in revenue associated with the scribe service (based on note complexity/wRVUs). It is, however, important to consider other potential benefits, including lower physician documentation burden and better patient-physician interaction. The prospect of scribes enabling more direct provider-patient interaction is particularly appealing and deserves more detailed assessments in future studies. Participating physicians were interested in continuing the scribe program and perceived that scribes reduced the time they spent documenting services. Although we were not able to directly track physician time spent in documentation, we believe that shorter time for documentation is likely. We believe that use of scribes can positively influence provider career satisfaction, reduce burnout and improve retention, which also provides an opportunity to reduce direct costs of provider recruitment and replacement.

Our scribe trial recruited medical students, which also provided them with an informal educational experience during a summer vacation. This model could potentially be generalizable to students in other healthcare disciplines with proper training. Advantages to our approach included having scribes already familiar with our hospital EMR system, with prior exposure to medical terminology and concepts, and to elements of medical documentation. We recognize that using students for a scribe system has limitations including availability, turnover, and training requirements. For medical students, care must also be taken to separate scribing activities from formal learning activities. More traditional scribes are often existing medical staff, such as nurses or medical assistants, or commercial medical scribes. Our study shows sufficient benefits to encourage additional trials, with attention not only to economics and time-saving but also of educational value to students and of provider job satisfaction and burnout.

CONCLUSION

With the increasing adoption of EMR systems, physician concerns have grown about time needed for clinical documentation. Medical scribes may offer a positive route to improve physician productivity and workflow efficiency in clinic with reduced time spent on documentation. Our study showed no negative impact on patient satisfaction or medical note complexity documentation and demonstrated provider acceptability as well as some improvement in clinic workflow and efficiency. Further perspective studies are needed to better assess this emerging service and determine economic feasibility in the setting of pediatric subspecialty healthcare services.

REFERENCES

1. Hill RG Jr, Sears LM, Melanson SW. 4000 clicks: a productivity analysis of electronic medical records in a community hospital ED. Am J Emerg Med 2013; 31:1591–1594.
2. Weigl M, Muller A, Zupanc A, et al. Participant observation of time allocation, direct patient contact and simultaneous activities in hospital physicians. BMC Health Serv Res 2009; 9:110.
3. Kuhn T, Basch P, Barr M, et al. Clinical documentation in the 21st century: executive summary of a policy position paper from the American College of Physicians. Ann Intern Med 2015; 162:301–303.
4. Shanafelt TD, Hasan O, Dyrbye LN, et al. Changes in burnout and satisfaction with work-life balance in physicians and the general US working population between 2011 and 2014. Mayo Clin Proc 2015; 90:1600–1613.
5. Woodcock DV, Pranaat R, McGrath K, et al. The evolving role of medical scribe: variation and implications for organizational effectiveness and safety. Stud Health Technol Inform 2017; 234:382–388.
6. Reick-Mitrisin V, MacDonald M, Lin S, et al. Scribe impacts on US health care: benefits may go beyond cost efficiency. J Allergy Clin Immunol 2020; 145:479–480.
7. The Joint Commission. Documentation Assistance Provided By Scribes. https://www.jointcommission.org/en/standards/standard-faqs/ambulatory/record-of-care-treatment-and-services-rc/000002210/. Accessed August 26, 2017.
8. Allred RJ. Ewer S Improved emergency department patient flow: five years of experience with a scribe system. Ann Emerg Med 1983; 12:162–163.
9. Arya R, Salovich DM, Ohman-Strickland P, et al. Impact of scribes on performance indicators in the emergency department. Acad Emerg Med 2010; 17:490–494.
10. Heaton HA, Castaneda-Guarderas A, Trotter ER, et al. Effect of scribes on patient throughput, revenue, and patient and provider satisfaction: a systematic review and meta-analysis. Am J Emerg Med 2016; 34:2018–2028.
11. Tischendorf JJ, Crede S, Herrmann P, et al. Encoding of diagnosis by medical documentation assistant or ward physician. Influence on the mapping of Diagnostic Related Group (DRG) performance. Dtsch Med Wochenschr 2004; 129:1731–1735.
12. Tegen A, O’Connell J. Rounding with scribes: employing scribes in a pediatric inpatient setting. J AHIMA 2012; 83:34–38.
13. Yan C, Rose S, Rothberg MB, et al. Physician, scribe, and patient perspectives on clinical scribes in primary care. J Gen Intern Med 2016; 31:990–995.
14. Bank AJ, Obetz C, Konrardy A, et al. Impact of scribes on patient interaction, productivity, and revenue in a cardiology clinic: a prospective study. Clinicoecon Outcomes Res 2013; 5:399–406.
15. Koshy S, Feustel PJ, Hong M, et al. Scribes in an ambulatory urology practice: patient and physician satisfaction. J Urol 2010; 184:258–262.
16. Nambudiri VE, Watson AJ, Rubenstein MH, et al. Association of patient satisfaction with medical scribe use in an academic dermatology practice. JAMA Dermatol 2018; 154:480–482.
17. Danila MI, Melnick JA, Curtis JR, et al. Use of scribes for documentation assistance in rheumatology and endocrinology clinics: impact on clinic workflow and patient and physician satisfaction. J Clin Rheumatol 2018; 24:116–121.
18. Walker KJ, Ben-Meir M, Phillips D, et al. Medical scribes in emergency medicine produce financially significant productivity gains for some, but not all emergency physicians. Emerg Med Australas 2016; 28:262–267.
19. Heaton HA, Nestler DM, Jones DD, et al. Impact of scribes on patient throughput in adult and pediatric academic EDs. Am J Emerg Med 2016; 34:1982–1985.
Keywords:

clinic flow; documentation; gastroenterology; outpatient; pediatric; scribe; subspecialty

Copyright © 2020 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition