Table 2.
Distribution of topics identified with raw example text among the symposium chat logs.
Topic identification number/labela | Thematic category | Raw example textb | Overall frequency, n (%) |
0: Documentation that adds value | Consensus building |
|
147 (8.3) |
1: Addressing poor usability | EHRc design |
|
122 (6.9) |
2: Sharing symposium resources | Symposium comments |
|
122 (6.9) |
3: Regulatory impact on clinician burden | Burden sources |
|
142 (8) |
4: Improved EHR user interface and design | EHR design |
|
128 (7.2) |
5: Role of quality measures and technology on burnout | Burden sources |
|
110 (6.2) |
6: Focusing documentation on patient narrative | Patient-centered care |
|
162 (9.1) |
7: Capturing data related to clinical practice | Burden sources |
|
113 (6.4) |
8: Determining data and documentation needs | Consensus building |
|
422 (23.8) |
9: Collectively reassessing documentation requirements in EHRs | Consensus building |
|
252 (14.2) |
N/Ad,e | N/A | N/A | 53 (3) |
aNumbering is based on the indices of an array to be consistent with programming code used across algorithms, which initiates with 0.
bRaw data are the actual chat messages of symposium attendees and have not been corrected for grammar.
cEHR: electronic health record.
dN/A: not applicable.
eExclusively comprised of person names, stop words, and other terms removed at the preprocessing stage.