Frequent but fragmented: use of note templates to document outpatient visits at an academic health center - PMC Skip to main content
PMC home page
Journal of the American Medical Informatics Association : JAMIA logoLink to Journal of the American Medical Informatics Association : JAMIA
. 2021 Oct 19;29(1):137–141. doi: 10.1093/jamia/ocab230

Frequent but fragmented: use of note templates to document outpatient visits at an academic health center

Adam Rule 1,✉, Michelle R Hribar 2
PMCID: PMC8714279  PMID: 34664655

Abstract

Recent changes to billing policy have reduced documentation requirements for outpatient notes, providing an opportunity to rethink documentation workflows. While many providers use templates to write notes—whether to insert short phrases or draft entire notes—we know surprisingly little about how these templates are used in practice. In this retrospective cross-sectional study, we observed the templates that primary providers and other members of the care team used to write the provider progress note for 2.5 million outpatient visits across 52 specialties at an academic health center between 2018 and 2020. Templates were used to document 89% of visits, with a median of 2 used per visit. Only 17% of the 100 230 unique templates were ever used by more than one person and most providers had their own full-note templates. These findings suggest template use is frequent but fragmented, complicating template revision and maintenance. Reframing template use as a form of computer programming suggests ways to maintain the benefits of personalization while leveraging standardization to reduce documentation burden.

Keywords: electronic health records, clinical documentation, note templates

INTRODUCTION

In most electronic health records (EHRs), users can write notes by copying text from prior notes or inserting text and data from templates (eg, Epic SmartPhrases/SmartTexts, Cerner Autotext/Smart Templates).1 These methods take less time and produce more consistent notes than manual typing, but can propagate outdated or extraneous information, threatening note quality and patient safety.2,3 Many studies have examined how physicians copy note text and provided guidance for how to do so safely.4–10 However, recent evidence suggests most text in clinical notes comes from templates.11

EHRs have included note templates—documentation aids that insert predefined text or data from the patient’s chart into the note—for decades,12–14 but we know surprisingly little about how they are used in practice.13–15 One study from 2002 examined how physicians edited templated exam findings,13 while another from 2008 observed template use at 2 academic medical centers.14 Template use has changed dramatically since then. In the 2002 and 2008 studies, clinicians worked with administrators to create templates that represented an entire note and used one per note.14 Now there is evidence to suggest clinicians routinely create their own templates (some inserting just short phrases), use several in the same note, and use them in collaboration with other members of the care team.15 For example, a clinician might use a full-note template to insert section headers and patient data to structure their note and aid chart review, their scribe might use a short-phrase template during the visit to insert a diagnosis they have difficulty spelling, and the clinician later use an attestation template to add text certifying the scribe’s edits.

Personalization lets clinicians tailor templates to their preferences and workflow and is associated with increased EHR satisfaction.16 However, it may also encourage the inclusion of extraneous information (eg, information perceived to be, but not actually required for billing) and duplication of effort. Several studies have found implementing standard note templates for residents and training them on documentation best practices (eg, limiting lists, encouraging novel input) decreased documentation time while increasing note quality.17–19

Recent policy changes have reduced documentation requirements for outpatient notes,20,21 providing an opportunity to rethink documentation practices. Navigating the tradeoffs between personalization and standardization will require a better understanding of current template use.

Objective

Describe how primary providers (eg, physicians, nurse practitioners, physician assistants) and other members of the care team (eg, trainees, nurses, medical assistants, technicians, scribes) together used note templates to write the provider progress note for outpatient visits at an academic health center.

MATERIALS AND METHODS

Setting

We conducted this retrospective cross-sectional study at Oregon Health & Science University (OHSU), an academic health center in Portland, OR. OHSU implemented an institution-wide EHR in 2006 (EpicCare, Epic Systems, Verona, WI). OHSU’s Institutional Review Board approved this study.

Data collection

Prior to data collection, we hypothesized that template use was frequent but fragmented; that most office visits were documented using at least one template, but most templates were only ever used by one person. To test these hypotheses, we collected EHR metadata from our institution’s data warehouse on all outpatient office visits—including telehealth visits—conducted at OHSU in 2018–2020. For each visit, we identified the primary provider, specialty, and progress note. We also collected metadata on all SmartPhrases—a type of template in EpicCare—used to document each visit. These metadata included the template’s contents, who invoked the template, who “owned” the template, and context of use (eg, charting, letter writing). While EpicCare has other templating tools, such as SmartTexts, SmartPhrases are the easiest to personalize so we focused on their use. In the remainder of this paper, we use “SmartPhrase” and “template” interchangeably. To examine template use while writing notes, we only analyzed templates used while charting visits that produced a progress note. We excluded data from therapeutic specialties (ie, Occupational Therapy) and specialties with <1000 outpatient office visits across 2018–2020, leaving metadata from 52 specialties for analysis.

SmartPhrase templates at OHSU are invoked by typing the template’s name (eg, .GENPEDCLINICNOTE) into an interactive note editor, which inserts the template’s contents at that location. These templates can contain static text, lists of text alternatives to be selected from a dropdown menu (eg, the patient was {agitated, calm}), links that import data (eg, @AGE@), and placeholders for manually typed text (ie, ***). The template owner is the user who created the template, either de novo or by modifying another user’s template. Users cannot invoke others’ templates by default but can use system templates created by the EHR vendor or local administrators and templates shared with them.

Measures and data analysis

We measured the frequency of template use by counting the number of visits where a template was used and the median number of templates used each visit. We also examined who used templates, including how often the person who invoked the template was the primary provider or the template’s owner, and the number of unique people who invoked a template each visit. We measured how fragmented template use was by counting the number of unique people who invoked each template. These analyses were conducted in Python (version 3.7).22

To examine template contents, we used regular expression matching to count templates that included text-selection lists, data links, or manual entry placeholders. We identified common types of templates by manually categorizing the 1000 most used templates. Each template was iteratively categorized by the first author until a consistent set of categories emerged. The second author used this taxonomy to categorize the 100 most used templates, achieving high inter-rater reliability (ie, Cohen’s kappa of 0.89). Automatically labeling templates with 5+ data links as full-note templates—a type of template identified in the manual categorization—achieved a precision of 0.924 and recall of 0.947 on the set of 1000 manually labeled templates. We used this threshold to identify full-note templates at scale in a secondary analysis of the frequency and fragmentation of full-note template use.

RESULTS

Frequency of template use

We included 2 490 899 (85.1%) of the nearly 3 million outpatient office visits conducted at OHSU between 2018 and 2020 in this study. Of these, 89.3% were documented using at least one template. This rate was stable across time (88.3% of visits in 2018 and 90.4% in 2020) and visit type (89.1% of in-person vs. 91.2% of telehealth visits). See Figure  1 for rates of template use by specialty, and Supplementary Appendix  Table S1 for the number of encounters by specialty.

Figure 1.

Figure 1.

Open in a new tab

Frequency of template use as measured by percent of outpatient encounters in each specialty where a template was used to write the progress note.

A median of 2 templates was used per visit. Fully 65.7% of visits involved the use of 2 or more templates and 39.5% involved 2 or more people invoking separate templates. Of the 5549 EHR users who invoked a template between 2018 and 2020, 2443 (44.0%) were the primary provider for at least one visit and 4321 (77.9%) used at least one template they created themselves. Templates were most often invoked by the primary provider (63.9% of uses) and template owner (62.4% of uses).

Template contents

Of the 100 230 unique templates used, 38.0% included a data link, 21.5% a list for text selection, 46.6% a placeholder for manual text entry, and 42.7% contained just static text. We identified 5 template types by manually categorizing the 1000 most used templates: full-note templates, attestations/signatures, short phrases, datapoints/panels, and screenings/procedures (Figure  2). The most common type was the full-note template that imported a near complete progress note, often using multiple data links. Such full-note templates—identified at scale by labeling all templates with 5+ data links as a full-note template—were used at 64.8% of visits.

Figure 2.

Figure 2.

Open in a new tab

Examples of 5 common types of template used to write outpatient notes.

Fragmentation of template use

We observed 100 230 unique templates being used, only 17.3% of which were ever used by more than one person. Similarly, just 19.7% of the 23 382 unique full-note templates (ie, templates with 5+ data links) were ever used by more than one person. Looking at the full-note templates that providers used most frequently, there were 1350 unique templates across the 1806 providers who conducted at least 100 encounters during the study period. The ratio of preferred full-note templates to providers (0.75 overall) ranged from 1.0 in 18 specialties (ie, each provider had their own preferred full-note template) to 0.42 in Pediatrics, where many providers—especially residents—used a standard template provided by the department.18 Many templates owned by different providers were similar or identical to one another. For example, 103 different templates documented responses to the same set of contraindication questions for a flu vaccine.

DISCUSSION

We analyzed EHR metadata from outpatient visits conducted at OHSU between 2018 and 2020, finding template use was both pervasive and fragmented. Templates were used to document most visits (89.3%), likely because they can be remarkably useful. Templates can reduce documentation time, enforce standards, aid chart review, and provide decision support. But like copy-pasting, templates can import inaccurate information, including unreconciled medication lists or default exam findings that were not actually observed. Several groups have developed best practices for copy-paste,9,10,23 but there is little evidence to guide template design.18,19 One recent study found 56% of text in outpatient notes came from templates, but just 14% from copy-paste.11 Templates may now be the dominant method of writing notes. More evidence is needed on their risks, benefits, and best use.

We also found template use was remarkably fragmented, with more than 100 000 unique templates used across 2018–2020. Most (83%) were only ever used by one person. Individual template development can align documentation workflows with clinical workflows,16 but it may also duplicate effort and increase EHR burden, such as when policy changes require thousands of individual templates be updated simultaneously. Fragmentation may also make it easier for clinicians to continue using templates with outdated clinical guidance. Prior work has found implementing standard resident note templates can increase note quality and reduce documentation time.18,19 Even for more experienced clinicians, standardization may reduce the EHR burden, both regulatory and self-imposed.24

How should individuals and organizations navigate the tradeoffs between personalized and standardized templates? Consider that modern note templates are custom functions that interpolate text and data, are invoked by typing, and can modify the user interface (eg, render dropdown lists). This is sophisticated computer programming, and might benefit from being viewed as such.15 For example, EHR designers might consider how template development environments can better support users with development tools like tab completion, syntax highlighting, and contextual documentation (eg, tooltips describing what each data link does). Health systems might treat their collection of templates like an application programming interface by developing standards for naming, documentation, and scope of use. Version control systems—which let individual developers “fork” and edit documents while preserving a link to the original so changes can be propagated to their copy without overwriting their edits—might let individual clinicians tailor default templates to their unique workflows while still making it possible to quickly propagate changes across an organization when policies change.

Limitations

This study has several limitations which future work could address. First, it was conducted in the outpatient setting of one academic health center using one EHR. Future work might generalize these findings to other contexts. Second, it examined one templating tool (ie, Epic SmartPhrases), but not other documentation aids such as dictation, copy-paste, or form-driven note writers which may be used alongside these templates. These results then present a lower bound on the frequency of using documentation aids to write outpatient progress notes. Finally, we did not link specific patterns of template use to clinical or workflow outcomes such as patient safety or documentation time. For example, while some fragmentation may contribute to harmful variation in patient care (eg, templates with outdated clinical guidelines) other fragmentation may be benign (eg, of signature templates).

Future directions

With so little current evidence to guide template use, there are many promising directions for future research. One is characterizing documentation workflows; who invokes templates, when they use them, how templates scaffold team-based care, and how templates are used alongside dictation and copy-paste. A second promising direction is understanding template lifecycles; what prompts people to create templates, how others edit them, and what causes some templates to go “viral” while others go unused. A third direction is collecting data to inform template governance; investigating what training users receive on template design, how organizations monitor template use, and how many templates contain deprecated features or outdated guidance. A fourth promising direction is template design; both testing how existing templates might be rewritten, and how new templating features (eg, version control, contextual documentation) might enhance note quality, reduce EHR burden, and advance patient care. A final direction is assessing usability; whether notes produced by various methods of note entry are more accurate, trustworthy, or efficient to use.

CONCLUSION

This study demonstrates that template use in outpatient care is frequent but fragmented. Given the widespread use of templates, and increasing flexibility of documentation requirements, there is great potential to reduce EHR burden by rethinking how note templates are designed, managed, and used.

FUNDING

Supported by grants R00LM12238, P30EY10572, and T15LM007088 from the National Institutes of Health (Bethesda, MD), and by unrestricted departmental funding from Research to Prevent Blindness (New York, NY). The funding organizations had no role in the design or conduct of this research.

AUTHOR CONTRIBUTIONS

AR and MRH contributed to the research design, data analysis, and manuscript preparation.

SUPPLEMENTARY MATERIAL

Supplementary material is available at Journal of the American Medical Informatics Association online.

Supplementary Material

ocab230_Supplementary_Data
Click here for additional data file. (15.2KB, docx)

ACKNOWLEDGMENTS

Thank you to Ryan Bales for assistance with data collection, and Dr. Ben Orwoll for comments on earlier drafts of this manuscript.

CONFLICT OF INTEREST STATEMENT

None declared.

DATA AVAILABILITY

The data underlying this article cannot be shared publicly for the privacy of individuals that participated in the study.

REFERENCES

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ocab230_Supplementary_Data
Click here for additional data file. (15.2KB, docx)

Data Availability Statement

The data underlying this article cannot be shared publicly for the privacy of individuals that participated in the study.

Articles from Journal of the American Medical Informatics Association : JAMIA are provided here courtesy of Oxford University Press

RESOURCES