Table 2.
Measurement of study variables.
Variable | Description | Measurement | |||
Dependent variable |
|
|
|||
|
Patient satisfaction | Whether the patient expresses appreciation toward the physician | 0: The patient does not thank the physician 1: The patient expresses thanks to the physician during the interaction |
||
Independent variable |
|
|
|||
|
Response load | The physician’s average patient receipts per week | ln (the number of patients the physician receives per week divided by the number of days the physician is on duty per week) | ||
|
More detailed style | Average word count of the physician’s messages | ln (the total number of words the physician sends in one interaction divided by the total number of messages the physician sends during the interaction) | ||
|
Emotional comfort | Whether the physician comforts the patient | 0: The physician does not comfort the patient during the interaction | ||
Moderator variable |
|
|
|||
|
Free service | Whether the service is free | 0: The patient chooses a paid service 1: The patient chooses a free service |
||
Control variable |
|
|
|||
|
Gender | Patient’s gender | 0: Female 1: Male |
||
|
Age | Patient’s age | Number of years | ||
|
Department | Physician’s department | Category variables for 6 major departments | ||
|
Patient residence | Whether the patient lives in Guangdong | 0: The patient lives outside Guangdong 1: The patient lives in Guangdong |
||
|
Symptom description | Length of the patient’s symptom description | Shown in the patient’s order before the interaction | ||
|
Interaction frequency | Total number of exchanges during the chat | Shown in the text data of the physician-patient interaction | ||
|
Illness severity | Whether the patient’s illness warrants further treatment and diagnosis offline | 0: The physician does not suggest an in-person visit 1: The physician suggests an in-person visit |