F1000Research F1000Research 2046-1402 F1000 Research Limited London, UK 10.12688/f1000research.181906.1 Research Article Articles Documentation of Electronic Nicotine Delivery System, Nicotine Pouch and Combustible Tobacco Product Use in Electronic Health Record Clinical Notes

[version 1; peer review: awaiting peer review]

 Zhang Mingda Conceptualization Resources Supervision Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0001-8138-2722 a 1 Pope Derek Methodology Project Administration Writing – Original Draft Preparation Writing – Review & Editing 1 Altria Client Services LLC, Richmond, Virginia, 23219, USA mingda.zhang@altria.com

Competing interests: Both authors are employed by Altria Client Services LLC. The authors have no additional conflict of interest.

 2 6 2026 2026 15 862 20 5 2026 Copyright: © 2026 Zhang M and Pope D 2026 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background

Due to the lack of standardized codes, documentation of electronic nicotine delivery system (ENDS) use is non-existent in healthcare claims data and severely limited in structured electronic health record (EHR) data. We conducted a comprehensive assessment on the documentation of terms related to ENDS, nicotine pouch, and combustible tobacco use in unstructured EHR clinical notes.

 Methods

Utilizing retrospective data from Optum’s EHR and Clinical Notes Databases (2007–2023), 13.6 million patients had ≥1 diagnosis code for combustible tobacco use history, and 5.2 million also had ≥1 clinical note. Searches were conducted for terms for ENDS, nicotine pouches, or combustible tobacco in the clinical notes.

 Results

There were 4.4 million patients with ≥1 clinical note containing a term for ENDS, nicotine pouches, or combustible tobacco, with 66.1 million notes. Seventy-seven ENDS-related terms were identified, in addition to 27 ENDS brand terms, with the terms used evolved over time. There were 8 terms and 10 brand names for nicotine pouches, and 118 terms for combustible tobacco.

 Conclusions

Extensive ENDS use data are available in unstructured EHR clinical notes that can be extracted for research. When combined with additional data sources (structured EHR, claims, and survey linkage), these data may enable robust studies on the health impacts associated with the use of ENDS. Standardization and consistent documentation practices will further enhance the utility of these data for research.

 electronic health records; electronic nicotine delivery systems; tobacco use; oral tobacco; real-world evidence Altria Client Services LLC. Altria Client Services LLC. Abbreviations EHR

Electronic health records

 ENDS

Electronic nicotine delivery systems

 HPHC

harmful and potentially harmful constituents

 ICD

International Classification of Diseases

 NLP

Natural language processing

 SFP

Smoke-free products

 1. Introduction

There is a general consensus that using smoke-free products (SFP) like electronic nicotine delivery systems (ENDS) and nicotine pouches is associated with lower health risks than cigarette smoking. 1, 2 Compared to combustible cigarettes, SFPs have much lower levels of harmful and potentially harmful constituents (HPHC). 3– 8, 9– 11 Adults who switched from cigarettes to SFP not only substantially reduced their exposure to the corresponding HPHCs 12– 15 but also show either significant reductions or positive trends in the biomarkers of potential harm (BoPH). 16, 17 Epidemiological evidence on the health impact of switching from cigarette smoking to SFP is emerging but still limited, reporting mixed results. 18– 22 Interpretation of results from these early epidemiological studies often faces challenges in the cross-sectional design, limited number of participants in the studies, as well as varying degrees of control for important confounding factors, including history of cigarette smoking. Real-world data from healthcare claims and electronic health records (EHR) may enable a new approach to assess the health impact associated with the use of ENDS as they capture data on product usage patterns and health outcomes in diverse, real-world settings. 19, 23– 25 However, due to the lack of standardized codes, the documentation of ENDS use is non-existent in healthcare claims data 26 and severely limited in the structured fields of EHR data, 27, 28 in contrast to documentation for combustible tobacco products.

Previous studies using data from the EHR system of the Department of Veterans Affairs (VA) reported that documentation of ENDS use has increased substantially in recent years in the clinical notes and the terms used to describe ENDS are diverse and changed over time. 29, 30 The ENDS product landscape in the US has continued to evolve rapidly. 31 More recently, new SFPs like nicotine pouches have also emerged, for which standardized codes are not available either. We conducted a survey on the documentation of ENDS, nicotine pouches as well as combustible tobacco products in clinical notes using recent data from much larger commercial EHR datasets that more closely reflect the general US population. This study aimed to describe the documentation of these tobacco products’ use in the unstructured clinical notes between 2007 and 2023 among a population of patients with a history of combustible tobacco use, since most adults who use ENDS and nicotine pouches had current or prior smoking history. 19, 32

2. Materials and methods 2.1. Study design and data source

This was a retrospective study using data from January 2007 through August 2023 (study period) from Optum® de-identified Electronic Health Record data set (Optum EHR Database) 33 which is derived from the electronic health records of a network of healthcare provider organizations across the United States. Clinical encounter data are aggregated from more than 2,000 hospitals and 7,000 clinics. The Optum EHR Database currently has more than 107 million unique patients across the United States, with an average of 56 months of observed data per patient. The Optum EHR Database is robust, longitudinally linked, HIPAA compliant, and statistically de-identified. The EHR clinical notes database comprises a subset of more than 45 million patients from the EHR Database 34 that includes free-text clinical notes sourced from more than 50 large healthcare systems throughout the US. This study did involve any human subjects, identifiable personal information, other HIPPA identifiers, or any other information that could lead to the identification of a patient. No patient’s or provider’s identity or medical records, or electronic notes, or facility identifiers were disclosed for the purposes of this study. The study protocol was reviewed by the WCG Institutional Review Board (WCG IRB ID: IRB00000533) and determined to be exempt from IRB review; therefore, informed consent was not required.

 2.2. Study population

The study population ( Table 1) consisted of all patients who met the following criteria.

Patients with at least one EHR record with an International Classification of Diseases (ICD) diagnosis code ( Table 2) for the history of combustible tobacco use during the study period, and

Presence of at least one clinical note during the study period, and

Presence of at least one term of interest related to ENDS, nicotine pouches, or combustible tobacco in the clinical notes during the study period.

Terms of interest used to define the study population were classified into three product categories: ENDS, nicotine pouch, and combustible tobacco. ENDS terms fall into two groups: variants of vape and variants of e-cigarette. Combustible tobacco terms included terms related to current and former cigarette smoking, cigar smoking, and the use of water pipes. An initial list of terms for each category was constructed by subject matter experts incorporating data from existing literature. Additional terms were identified during the manual annotation of clinical notes in the process of developing a novel natural language processing (NLP) algorithm for extracting ENDS use data from EHRs (to be reported separately). The final lists of terms searched in this study for ENDS are listed in Table 3.

When searching for the terms of interest in clinical notes, capital and lower-case letters were treated as equivalent, i.e., the following terms were all interchangeable: e-cig, E-cig, and E-CIG. Additionally, the query looked for any sequence containing the term of interest, i.e., notes containing “vapes” or “vaped” would all be flagged as containing “vape”. Therefore, patient and note counts for longer terms (e.g., vapes, vaped) represent subsets of the overlapping shorter terms (e.g., vape).

 2.3. Analysis

Patients with at least one clinical note containing a term (e.g., “vape”) or a set of terms (e.g., all vaping-related terms) ( Table 3) were flagged, and the count of notes containing terms of interest was retained. Analyses were conducted for each year in the study period (2007–2023) to better understand how ENDS use documentation has evolved in recent years (excluding January to August 2023, which is a partial year). No statistical testing was performed. A patient with multiple notes containing term(s) of interest within each product category (i.e., ENDS, nicotine pouch, or combustible,) was counted only once. Similarly, a note containing multiple instances of term(s) of interest within each product category was only counted once.

 3. Results 3.1. Study population

Among the 107 million unique patients in the Optum’s EHR Database, 13.6 million patients had at least one diagnosis code for history of combustible tobacco use (EHR Population, Table 1). Of these, 5.2 million met the inclusion criterion of having at least one clinical note (Notes Population, Table 1). A total of 4.4 million patients had at least one term of interest related to ENDS, nicotine pouches, or combustible tobacco products in 66.1 million clinical notes (Study Population, Table 1), which served as the study population. Detailed breakdown of patient and note counts in the full EHR Database by ICD diagnosis code is provided in Table 2.

Patient and note attrition from the full EHR/Notes data asset (1/1/2007–8/31/2023).
Criteria Patients Notes
Remaining Remaining
n % n %
EHR Population * - Patients with at least 1 EHR record with an ICD dx code for combustible tobacco use 13,583,236
Notes Population- At least 1 clinical note in the observation period 5,194,000 38.2% 862,491,151
Study Population- At least 1 note containing a term for ENDS, nicotine pouches or combustible tobacco 4,371,006 84.2% 66,101,820 7.7%
≥1 combustible tobacco term ** 3,443,072 78.8% 48,432,066 73.3%
≥1 ENDS term OR ≥ 1 ENDS brand term ** 1,207,884 27.6% 7,738,198 11.7%
≥1 nicotine pouch term OR ≥ 1 nicotine pouch brand term ** 1,217 0.03% 2,653 0.004%

Initial population (first row) is among the full set of patients/notes in Optum® EHR data asset. Each of the two subsequent row is a subset of the prior row

Subset of Study Population; patient and note counts are unique within each row; a patient/note may appear in more than one row

 Patient and Note Counts in the Optum’s EHR Dataset with ICD Diagnosis Codes (2007–2023).
Code Code Type Description Patient Count Note Count
3051 ICD-9 Dx Tobacco use disorder 391,536 2,943,225
98984 ICD-9 Dx Toxic effect of Tobacco 510 3,334
V1582 ICD-9 Dx Personal history of tobacco use 197,364 1,721,680
F17210 ICD-10 Dx Nicotine dependence, cigarettes, uncomplicated 591,683 3,999,763
F17211 ICD-10 Dx Nicotine dependence, cigarettes, in remission 14,258 137,125
F17213 ICD-10 Dx Nicotine dependence, cigarettes, with withdrawal 3,535 29,067
F17218 ICD-10 Dx Nicotine dependence, cigarettes, with other nicotine-induced disorders 9,489 101,900
F17219 ICD-10 Dx Nicotine dependence, cigarettes, with unspecified nicotine-induced disorders 9,004 80,870
T65221A ICD-10 Dx Toxic effect of tobacco cigarettes, accidental (unintentional), initial encounter 135 1,056
T65221D ICD-10 Dx Toxic effect of tobacco cigarettes, accidental (unintentional), subsequent encounter 6 65
T65221S ICD-10 Dx Toxic effect of tobacco cigarettes, accidental (unintentional), sequela 5 37
T65222A ICD-10 Dx Toxic effect of tobacco cigarettes, intentional self-harm, initial encounter 121 962
T65222D ICD-10 Dx Toxic effect of tobacco cigarettes, intentional self-harm, subsequent encounter 4 5
T65222S ICD-10 Dx Toxic effect of tobacco cigarettes, intentional self-harm, sequela 70 713
T65223A ICD-10 Dx Toxic effect of tobacco cigarettes, assault, initial encounter 1 17
T65223D ICD-10 Dx Toxic effect of tobacco cigarettes, assault, subsequent encounter 1 2
T65223S ICD-10 Dx Toxic effect of tobacco cigarettes, assault, sequela - -
T65224A ICD-10 Dx Toxic effect of tobacco cigarettes, undetermined, initial encounter 65 592
T65224D ICD-10 Dx Toxic effect of tobacco cigarettes, undetermined, subsequent encounter 11 96
T65224S ICD-10 Dx Toxic effect of tobacco cigarettes, undetermined, sequela 5 68
Z720 ICD-10 Dx Tobacco use 408,168 2,996,951
Z87891 ICD-10 Dx Personal history of nicotine dependence 844,203 6,469,782

Patient and Note counts are unique within each row; a patient/note may appear in more than one row.

 3.2. Documentation of terms of interest in EHR notes

A total of 222 combustible tobacco terms, 78 ENDS terms plus 52 ENDS brand terms, and 10 nicotine pouch terms plus 12 nicotine pouch brand terms were pre-identified. Searching for these terms in the clinical notes found 118 combustible tobacco product terms, 77 ENDS terms plus 27 ENDS brand terms (104 total terms), and 8 nicotine pouch terms plus 10 nicotine pouch brand terms (18 total terms). The remaining pre-identified search terms were not found in the notes. A summary of patient and note counts for each of the three product categories among the Study Population is provided in Table 1.

3.2.1. ENDS-Related Terms

Of the 77 ENDS-related terms found in notes, 31 were variations of “e-cigarette” (e-cig variant terms) and 46 were variations of “vape” (vape variant terms). An additional 27 ENDS brand terms were identified. The full set of ENDS-related and ENDS brand terms searched in this study, and the corresponding number of patients and notes identified, can be found in Table 3.

Patient and Note Counts for Terms Related to Electronic Nicotine Delivery System (ENDS) Use (2007–2023).
Vaping Variants E-cigarette Variants ENDS Brand Names
Term Patients Notes Term Patients Notes Term Patients Notes
Vaping 2,951,664 14,341,471 E-Cigarette or E Cigarette 2,220,214 10,248,902 Hqd 77,586 87,154
Vape 1,343,839 10,200,711 electronic cigarettes 1,280,997 9,914,734 Leap 36,686 81,403
Vapes 113,899 289,839 e-cigarettes or e cigarettes 1,390,672 6,153,289 Glas 9,539 27,516
Vape pen or Vape-pen 15,392 32,676 e-cig or e cig 218,652 1,040,704 Juul 9,185 21,936
Vaped 11,907 28,402 electronic cigarette 155,683 397,016 Juul Vape 304 672
Using Vape 3,940 8,753 ecig 51,832 108,991 Phix 340 358
Vape pens or Vape-pens 3,923 6,019 e-cigs or e cigs 16,697 47,779 Mr Fog 90 145
Smoking Vape 2,399 3,149 ecigarette 2,905 7,354 My Blu 90 95
Vaping device 1,252 2,095 electric-cigarette or electric cigarette 2,335 5,948 Lost Mary 46 70
Smokes Vape 778 1,682 E smoking or E-smoking 4,235 5,768 Lava 2 25 65
Electronic Vapor 307 931 e-cigg or e cigg 787 3,459 Bidi Stick 1 54
E-vape 310 912 electric-cigarettes or electric cigarettes 1,331 3,295 Breeze Pro 19 41
E vape or E-Vapes 310 912 ecigarettes 1,403 3,125 Elfbar 18 27
Electronic vape 347 655 electric-cig or electric cig 359 1,742 Breeze Plus 14 20
Disposable Vape or Disposable-Vape 281 453 e-ciggs or e ciggs 370 1,484 Blu Vape 9 12
Pod system or Pod-System 206 281 electric-cigs or electric cigs 133 569 Vuse Alto 6 10
Electronic Vapor cigarette 90 216 e-ciggarettes or e-ciggarettes 95 354 Esco Bars 5 8
Vape pods 80 148 e-ciggarette or e ciggarette 131 353 Logic Power 8 8
Electronic Vapor cigarettes 75 148 e cigaretes 75 212 Logic Pro 6 7
Electric Vapor 33 124 e-cigaretes 75 212 Blu Plus 5 5
Vape pod or Vape-pod 66 112 electronic ciggarette 57 186 Lava Plus 4 5
EVape 45 106 electric-cigg or electric cigg 32 161 Vuse Solo 1 4
Electronic vapes 31 68 eciggarettes 5 145 Vuse Vibe 3 4
Electric Vape 26 50 electronic ciggarettes 39 127 Suorin Air 2 2
Disposable vape pen 33 45 e-cigarete or e cigarete 35 126 Blu Classic 1 1
Electric Vaping 11 37 electronic cigarete 14 87 Elfbar Bc5000 1 1
Vape mod or Vape-mod 11 20 e-cigarette device or e cigarette device 62 68 Hyppe Max Flow 1 1
Vape mods or Vape-mods 3 18 electronic cigaretes 17 49 21St Century Smoke 0 0
Pod vape or Pod-Vape 8 10 Electronic Nicotine cigarettes 18 34 Blu vape bar 0 0
Electric vapes 3 3 eciggarette 18 26 Eb Design Bc5000 0 0
Vape pen kit 1 1 electric-cigarete or electric cigarete 3 26 Epic Series By Jak 0 0
Electronic Nicotine cigarette 17 22 Flair Infinity 0 0
electric-ciggarette 11 15 Fume Extra 0 0
electric ciggarette 11 15 Fume Infinity 0 0
electric-cigaretes or electric cigaretes 3 15 Fume Ultra 0 0
electric-ciggs or electric ciggs 7 11 Greensmartliving 0 0
ecigarete 6 10 Hqd Cuvie Plus 0 0
e-ciggaretes or e ciggaretes 6 7 Juicy Bar Jb5000 0 0
electriccigarette 6 6 Kado Bar Br5000 0 0
ecigaretes 3 5 Kangvape Onee Stick 0 0
electric-ciggarettes or electric ciggarettes 3 5 Kangvape Tc8000 0 0
Electric Nicotine cigarette 2 5 Loon Air Plus 0 0
ecigarette device 4 4 Loon Maxx 0 0
e-ciggarete or e ciggarete 2 2 Loon Typhoon 0 0
electriccigarettes 1 1 Lost Mary Mo5000 0 0
Electric Nicotine cigarettes 1 1 Lost Vape Orion 0 0
eciggarete 0 0 Mngo Stick 0 0
Mr Fog Switch 0 0
My Blu Intense 0 0
Njoy Ace 0 0
Njoy Daily 0 0
Sugar Extrax 0 0

Counts are unique within each row; a patient/note may appear in more than one row.

Counts for longer terms (e.g., vapes, vaped) represent subsets of overlapping shorter terms (e.g., vape).

The unique patient and unique note count distribution associated with key ENDS-related terms (vape, e-cig, and ENDS brand terms) are enumerated in Table 4. Because a patient could have multiple terms (e.g., both e-cig and vape variant terms) in their notes, a patient/note may appear in more than one row (e.g., both “≥1 Vape term” and “≥1 E-cig term”) in Table 4. Overall, during the 2007–2023 study period, vape variant terms were more frequently used than e-cigarette terms by both patient count (1,114,448 vs. 718,117) and note counts (7,270,473 vs. 3,265,350), respectively, whereas ENDS brand names were infrequently captured in the notes (5,458 patients, 16,544 notes). Combining 718,117 patients with “≥1 E-cig term” with 1,114,448 patients for “≥1 Vape term” yielded a total of 1,205,838 patients for “≥1 Vape term OR ≥1 E-cig term”, an increase of only 91,390, indicating that 87% [1 – (91,390/718,117)] patients with e-cig variant terms also have vape variant terms in their clinical notes. Similarly, 86% [1 – (7,727,143-7,270,473)/3,265,350] notes with e-cig variant terms also included vape variant terms. With its broadest definition, the bottom row of Table 4 represents the total numbers of unique patients (1,207,884, 27.6%) and notes (7,738,198, 11.7%) with documentation of ENDS use in the Study Population.

Patient and Note Counts for ENDS-related Terms among the Study Population (2007–2023).
Description Patients Notes
≥1 Vape term 1,114,448 7,270,473
≥1 E-cig term 718,117 3,265,350
≥1 ENDS brand term 5,458 16,544
≥1 Vape term OR ≥ 1 E-cig term 1,205,838 7,727,143
≥1 Vape term OR ≥ 1 ENDS brand term 1,116,643 7,281,820
≥1 E-cig term OR ≥ 1 ENDS brand term 721,137 3,279,077
≥1 Vape term OR ≥ 1 E-cig term OR ≥ 1 ENDS brand term 1,207,884 7,738,198

Patient and Note counts are unique within each row; a patient/note may appear in more than one row.

From 2007–2011, only e-cig variants were used in clinical notes. Documentation of e-cig variant terms continued to increase rapidly until 2021 and then decreased thereafter. Vape variants first occurred in clinical notes in 2012, remained sparse until 2018 when their documentation increased dramatically, and surpassing e-cig variants in 2022 [ Figure 1]. Overall, vape variant terms were more frequently used than e-cig variants during the 2007–2023 study period by patient count. Note count evolved in a pattern closely mirroring that for patient count shown in Figure 1, with much higher counts as expected (data not shown).

Patient counts with documentation of ENDS-use by leading terms found in EHR clinical notes in the study population from January 2007 to August 2023.

The same patient may be counted under multiple terms if these terms are found in the patient’s clinical notes. Patient counts for longer terms (e.g., vapes) represent a subset of overlapping shorter terms (e.g., vape). Patient counts are cumulative over time, i.e., counts in earlier years are subsumed in those for subsequent years.

In addition to the shift from e-cig to vape variants over time, specific terms used for both e-cig and vape variants have also evolved over time. For example, “e smoking” was the most popular e-cig variant up to 2018 when its use peaked and was taken over by “e-cig” in 2019, which has remained the most popular e-cig variant since then. “Vape” was the most popular vape variant between 2012 and 2015, surpassed by “vaping” in 2016, and regained the lead between 2019 and 2020. Vaping became the dominant term used since 2021.

Twenty-seven ENDS brand terms were identified ( Table 3, with corresponding patient and note counts). There were no ENDS brand terms in the clinical notes until 2011. The highest patient counts for the most popular brand remain below 2,500 per year throughout 2011–2023. Compared to the e-cig and vape variant terms discussed above, ENDS brand names were rarely documented in notes, limiting the feasibility of conducting brand-specific ENDS studies using EHR.

3.2.2. Nicotine Pouch Terms

Documentation of nicotine pouch use was first recorded for one patient in 2011 and remained almost non-existent between 2011 and 2018, with fewer than 10 patients per year. While increasing, documentation of nicotine pouch use in clinical notes is still sparse. There were only a total of 1,217 patients with such documentation in clinical notes among the study population, compared to 1,207,884 for ENDS ( Table 1).

3.2.3. Combustible Tobacco Terms

The 118 combustible tobacco terms documented in clinical notes can be divided into 4 general groups: 49 for current smoking, 34 for former smoking, 21 for cigar, and 14 for water pipe. The terms with the two highest patient count for each of these 4 groups are, “current smoker” (238,277) and “continues to smoke” (173,614); “former smoker” (1,152,965) and “ex-smoker” (109,708); “cigar” (44,446) and “smokes cigar” (8,083); and “hookah” (6,015) and “smokes hookah” (929), respectively.

While all 13,583,236 patients in the EHR Population ( Table 1) had at least one ICD code indicating combustible product use history, combustible tobacco use was documented for 3,443,072 patients (25.3%), in 48,432,066 of the 862,491,151 clinical notes (5.6%) ( Table 1).

 4. Discussion

In this retrospective survey, we investigated terms used to document ENDS, nicotine pouch, and combustible tobacco use in EHR clinical notes among patients with a history of combustible tobacco use, and changes in the terms used over time. The number of terms used for each product category shows large variations in how tobacco product use is documented in clinical notes which likely resulted from inconsistencies in documentation between EHR systems as well as among individual practitioners, and present major challenges in utilizing such data for studies on the health impact of SFP use.

Documentation on ENDS use in clinical notes has increased substantially in recent years to become a valuable data source for assessing the real-world impact of transitioning from combustible tobacco to ENDS, as well as on the potential health impact of ENDS use among adults who did not have a history of using combustible products. However, while there was at least one documentation of ENDS use for 27.6% of the study population, ENDS use was documented in only 11.7% of the notes, indicating ENDS use is still under-documented in clinical notes, which is consistent with earlier reports. 28, 29 Clinicians used a large variation of terms to document ENDS use, and the terms used evolved substantially over time, which poses challenges for extracting consistent ENDS use information from clinical notes. Initially, e-cigarette variant terms were exclusively used before 2012, but by 2022, the latest full-year data available, vape variants were more commonly used (2.3 million notes and 599.6 thousand patients vs. 2.1 million notes and 567.3 thousand patients for e-cig variants). In contrast, ENDS brand names were captured infrequently in clinical notes (e.g., 4,400 notes, 2,600 patients in 2022). This evolution reflects changes in product popularity and terminology used by both consumers and healthcare providers. Understanding these changes is crucial for accurately capturing ENDS use history in future research that utilizes clinical notes. Results on ENDS use documentation from this study using data from a general US population are in line with earlier report based on data from the Department of Veteran Affairs 29 and can inform the creation of an ontology for ENDS, supporting further research into ENDS-related outcomes.

Consistent with earlier reports, 28– 30, 35 combustible tobacco product use is also substantially under documented in clinical notes, for about a quarter of the patients with ICD codes for combustible product use history in the Optum EHR dataset. Nevertheless, clinical notes contain details on combustible tobacco product use not captured by ICD codes and therefore still have value for real-world evidence studies. For example, since ICD diagnosis codes do not differentiate combustible product types, information from clinical notes will be particularly useful for studies on products other than cigarettes (e.g., cigars and water pipes). Combining both structured and unstructured data from EHR would likely yield more accurate data on combustible tobacco use than from either source alone.

Documentation of nicotine pouch use, while increasing, is still too limited, with the highest annual patient count in the hundreds, for meaningful real-world evidence studies. It will likely take several years for patient counts with nicotine pouch documentation in the EHR to reach a meaningful level, making them valuable for health outcomes research. More diligent and consistent documentation of tobacco product use in EHR by clinicians will enhance their utility for real-world evidence studies. Standardization of SPF use documentation in EHRs (e.g., through the adoption of ICD codes, addition of structured fields to EHR systems, or the use of consistent terms) can substantially increase the efficiency as well as the quality of such real-world evidence studies.

A strength of this study is the use of a large EHR database comprising 107 million unique patients across the United States, with more than 45 million patients having clinical notes as the data source. This data source enables the results reported here to closely reflect the current practices of tobacco product use documentation in clinical notes in the US. Another strength is the study period; trends in documentation were captured from 2007 through 2023. The study has several limitations. The search strategy used in this study did not consider the context in which the terms appeared within the clinical notes, which would overestimate patients and notes documenting actual ENDS use, as cases were counted even when the notes indicated no ENDS use. However, data on positive confirmation of no ENDS use would also be of value for identifying such cohorts for comparison in future studies. The study population identification approach used in this study does not capture patients with documentation of tobacco use in their clinical notes without the ICD diagnosis codes for combustible tobacco in their EHR. However, the missed population is likely to be relatively small since combustible tobacco use has been more broadly documented with ICD codes than documentation of tobacco use in clinical notes. 28– 30, 35 Since the terms identified in the study were based on an initial list from subject matter experts plus additional terms identified during the annotation of a subset of randomly sampled clinical notes, there might be other terms in the clinical notes not captured by this approach. The number of such terms are likely to be relatively small due to the large number of notes manually annotated during the study.

 5. Conclusion

A large variation of terms has been used to document ENDS use in the unstructured clinical notes. Such documentation has increased substantially in recent years, generating a large body of ENDS use data. Importantly, these data can be extracted with tools like NLP algorithms for real-world evidence studies on the impact of transitioning from combustible tobacco to ENDS use. Clinical notes also contain additional information on combustible tobacco use not captured by ICD codes. In contrast, documentation of nicotine pouch use in clinical notes is still very limited. Findings from this study on the variation, distribution, and evolution over time, of terms used to document combustible tobacco, ENDS, and nicotine pouches in clinical notes can help researchers in planning and designing real-world evidence studies on the health impact of transitioning from combustible tobacco to modern smoke-free tobacco products, including ENDS and nicotine pouches. Combining data from clinical notes with data from other sources (e.g., structured EHR, claims, consumer surveys) will enable more robust real-world evidence studies. Standardization in the documentation of modern SFP use in EHR will further enhance the utility of these data for real-world evidence research.

 Institutional review board statement

Ethical review and approval were waived for this study due because this study did not involve any human subjects, identifiable personal information, other HIPPA identifiers, or any other information that could lead to the identification of a participant. All database records used in the study are statistically de-identified and certified to be fully compliant with US patient confidentiality requirements set forth in the Health Insurance Portability and Accountability Act of 1996. The study protocol was reviewed by WCG Institutional Review Board (WCG IRB ID: IRB00000533) which determined it to be exempt from IRB review and informed consent was therefore not required.

 Informed consent statement

Not applicable.

 Data availability statement

This study was reviewed by an Institutional Review Board and was determined to be exempt from human subjects research oversight. The underlying electronic health record (EHR) data used are not publicly available due to restrictions on access to proprietary EHR data and the need to protect patient privacy. The study used de-identified data from Optum’s EHR and Clinical Notes databases, which are subject to data use agreements that prohibit public sharing. These datasets contain sensitive health information and are licensed for research use under specific contractual and ethical restrictions. Researchers interested in the data may request a license from Optum, subject to license agreements and approval by Optum. Requests should be directed to Pope, Derek at Derek.A.Pope@altria.com. Aggregate results generated in this study are provided within the manuscript. Additional aggregate results can be made available upon reasonable request.

 Acknowledgments

The authors gratefully acknowledge Abhinav Nayyar, Shashi Khan, Shikha Anand, Tamanna Tara, and Pankaj Kang of Optum for their contributions to data curation and model development and Noelle Gronroos and Ami Buikema of Optum for their contributions to study design and interpretation. The authors would also like to thank Dr. Yan Wang for her comments on the draft manuscript.

Third party trademarks are the property of their respective owners, are used for reference only, and are not intended to suggest any affiliation.

 References Abrams DB Glasser AM Pearson JL : Harm Minimization and Tobacco Control: Reframing Societal Views of Nicotine Use to Rapidly Save Lives: Erratum. Annu. Rev. Public Health. 2024. NASEM: National Academies of Sciences, E., and Medicine: Public HealthConsequences of E-Cigarettes. 2018. 10.17226/24952 U.S. Department of Health and Human Services: How tobacco smoke causes disease: the biology and behavioral basis for smoking-attributable disease: a report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention;2010;0160840783. Food, U.; Administration, D: Harmful and potentially harmful constituents in tobacco products and tobacco smoke; established list. Fed. Regist. 2012;77:20034–20037. Koh HK Fiore MC : The tobacco industry and harm reduction. JAMA. 2022;328:2009–2010. 10.1001/jama.2022.20679 Oldham MJ DeSoi DJ Rimmer LT : Insights from analysis for harmful and potentially harmful constituents (HPHCs) in tobacco products. Regul. Toxicol. Pharmacol. 2014;70:138–148. 24973503 10.1016/j.yrtph.2014.06.017 Polosa R : Examining the evidence for the health impact of combustion-free products: progress and prospects for tobacco harm reversal and reduction. Intern. Emerg. Med. 2021;16:2043–2046. 34524625 10.1007/s11739-021-02837-2 PMC8563566 Perfetti TA Rodgman A : The complexity of tobacco and tobacco smoke. Beiträge zur Tabakforschung International/Contributions to Tobacco Research. 2011;24:215–232. 10.2478/cttr-2013-0902 Goniewicz ML Knysak J Gawron M : Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob. Control. 2014;23:133–139. 23467656 10.1136/tobaccocontrol-2012-050859 PMC4154473 Reilly SM Cheng T Feng C : Harmful and Potentially Harmful Constituents in E-Liquids and Aerosols from Electronic Nicotine Delivery Systems (ENDS). Chem. Res. Toxicol. 2024;37:1155–1170. 38924487 10.1021/acs.chemrestox.4c00093 PMC11256903 Wagner KA Flora JW Melvin MS : An evaluation of electronic cigarette formulations and aerosols for harmful and potentially harmful constituents (HPHCs) typically derived from combustion. Regul. Toxicol. Pharmacol. 2018;95:153–160. 29567331 10.1016/j.yrtph.2018.03.012 Anic GM Rostron BL Hammad HT : Changes in Biomarkers of Tobacco Exposure among Cigarette Smokers Transitioning to ENDS Use: The Population Assessment of Tobacco and Health Study, 2013-2015. Int. J. Environ. Res. Public Health. 2022;19. 10.3390/ijerph19031462 Dai H Khan AS : A Longitudinal Study of Exposure to Tobacco-Related Toxicants and Subsequent Respiratory Symptoms Among U.S. Adults with Varying E-cigarette Use Status. Nicotine Tob. Res. 2020;22:S61–s69. 33320254 10.1093/ntr/ntaa180 PMC8628872 Holt NM Shiffman S Black RA : Comparison of biomarkers of exposure among US adult smokers, users of electronic nicotine delivery systems, dual users and nonusers, 2018-2019. Sci. Rep. 2023;13:7297. 37147399 10.1038/s41598-023-34427-x PMC10163269 Lizhnyak PN Noggle B Wei L : Understanding heterogeneity among individuals who smoke cigarettes and vape: assessment of biomarkers of exposure and potential harm among subpopulations from the PATH Wave 1 Data. Harm Reduct. J. 2022;19:90. 35978343 10.1186/s12954-022-00673-x PMC9387076 Christensen CH Chang JT Rostron BL : Biomarkers of Inflammation and Oxidative Stress among Adult Former Smoker, Current E-Cigarette Users-Results from Wave 1 PATH Study. Cancer Epidemiol. Biomarkers Prev. 2021;30:1947–1955. 10.1158/1055-9965.Epi-21-0140 Shiffman S Oliveri DR Goldenson NI : Comparing Adult Smokers Who Switched to JUUL versus Continuing Smokers: Biomarkers of Exposure and of Potential Harm and Respiratory Symptoms. Nicotine Tob. Res. 2024;26:494–502. 37837438 10.1093/ntr/ntad197 Brouwer AF Jeon J Jimenez-Mendoza E : Changing patterns of cigarette and ENDS transitions in the USA: a multistate transition analysis of adults in the PATH Study in 2017–2019 vs 2019–2021. Tob. Control. 2024;33:570–579. 36977570 10.1136/tc-2022-057905 PMC10533746 Cohen G Cook S : Observational Studies of Exposure to Tobacco and Nicotine Products: Best Practices for Maximizing Statistical Precision and Accuracy. iScience. 2024. Selya A : The “Gateway” hypothesis: evaluation of evidence and alternative explanations. Harm Reduct. J. 2024;21:113. 38849956 10.1186/s12954-024-01034-6 PMC11157890 Rodu B Plurphanswat N : Cross-sectional e-cigarette studies are unreliable without timing of exposure and disease diagnosis. Intern. Emerg. Med. 2023;18:319–323. 36434423 10.1007/s11739-022-03141-3 Rodu B Plurphanswat N Rodu J : Inaccurate and misleading meta-analysis of E-cigarettes and population-based diseases. Intern. Emerg. Med. 2025;20:1381–1388. 40350507 10.1007/s11739-025-03956-w PMC12331790 Asfar T Jebai R Li W : Risk and safety profile of electronic nicotine delivery systems (ENDS): an umbrella review to inform ENDS health communication strategies. Tob. Control. 2024;33:373–382. 36252567 10.1136/tc-2022-057495 PMC10043882 Jackson SE Brose L Buss V : Moderation of the real-world effectiveness of smoking cessation aids by mental health conditions: a population study. PLoS Mental Health. 2024;1:e0000007. 41661824 10.1371/journal.pmen.0000007 PMC12798440 Zhang M : Economic Outcomes Associated with Transitions of Tobacco Use Among a Cohort of Adult Male Smokers using U.S. Real-World Data. Proceedings of the ISPOR 2023, Boston, May 7-10, 2023. 2023. Zhang M Cheng HG Noggle B : Assessing healthcare cost changes associated with transitioning away from cigarette smoking using healthcare claims data: an exploratory study among adult male patients with COPD. Harm Reduct. J. 2024;21:227. 39716190 10.1186/s12954-024-01141-4 PMC11665026 Jose T Hays JT Warner DO : Improved Documentation of Electronic Cigarette Use in an Electronic Health Record. Int. J. Environ. Res. Public Health. 2020;17. 32823985 10.3390/ijerph17165908 PMC7460036 Winden TJ Chen ES Wang Y : Towards the Standardized Documentation of E-Cigarette Use in the Electronic Health Record for Population Health Surveillance and Research. AMIA Jt Summits Transl Sci Proc. 2015;2015:199–203. 26306269 Conway M Alba PR Zhu SH : Vaping at the VA: Developing an Annotated Corpus of Electronic Cigarette Mentions in Clinical Notes at the Department of Veterans Affairs. AMIA Annu. Symp. Proc. 2021;2021:343–351. Conway M Mowery DL South BR : Documentation of ENDS Use in the Veterans Affairs Electronic Health Record (2008-2014). Am. J. Prev. Med. 2019;56:474–475. 30777165 10.1016/j.amepre.2018.10.019 Lindson N Butler AR McRobbie H : Electronic cigarettes for smoking cessation. Cochrane Database Syst. Rev. 2025;2025(1):Cd010216. 10.1002/14651858.CD010216.pub9 Tattan-Birch H Jackson SE Shahab L : Oral nicotine pouch use in Great Britain: a repeat cross-sectional study, 2020-25. Lancet Public Health. 2026;11:e26–e34. 41418803 10.1016/s2468-2667(25)00296-8 Optum: Gain visibility into patient care and outcomes with EHR data. 2024;2024. Reference Source Guerra Currie A-MLB Shintani K Balankura T : Natural language processing methods. Natural Language Processing Methods. 2023. Young-Wolff KC Klebaner D Folck B : Documentation of e-cigarette use and associations with smoking from 2012 to 2015 in an integrated healthcare delivery system. Prev. Med. 2018;109:113–118. 29360481 10.1016/j.ypmed.2018.01.012 PMC7004208