Questions on Diabetes from Patients and the Public

There are two Excel .csv files containing
questions on type 2 diabetes mellitus tagged by the specific topics each
question covers.
clinicquestions.csv has questions collected from 100 patients.
crowdsourcedquestions.csv has questions collected from 300 members of the
public.

Questions on type 1 diabetes were removed, as were questions on clinic
operations. Details on the coding can be found in the citation (below).

clinicquestions.csv
Number of instances: 152
Number of attributes: 25
Attribute characteristics: text; integer IDs,
one-hot encodings
Missing data: none

crowdsourcedquestions.csv

Number of instances: 284
Number of attributes: 26
Attribute characteristics: text; categorical
(Female/Male); integer IDs, one-hot encodings
Missing data: none

ATTRIBUTES:
gender – categorical (Male/Female)
person_ID – Integer identifier for the person asking the question
qx_ID – Integer identifier for the question
Qx – The text of the question itself, as asked, without
corrections or edits
22 topic categories one-hot encoded:
CAUSE ; RISK; PREVENTION; DIAGNOSIS; MANIFESTATIONS; TREATMENT; ANATOMY;
CURE; DIET; EXERCISE; WEIGHT; SELF-MANAGEMENT; DISEASE COMPLICATIONS; TREATMENT
COMPLICATIONS; PERSON or ORGANIZATION; PROGNOSIS; DISTRIBUTION of a DISEASE in
a POPULATION; INHERITANCE PATTERNS; RESEARCH; PSYCHOSOCIAL; Own HEALTH RECORD
RELATED; OTHER

CITATION:

Crangle CE, Bradley C, Carlin PF, Esterhay RJ, Harper R, Kearney
PM, Lorig K, McCarthy VJC, McTear M, Savage E, Tuttle MS, Wallace JG. (2018, to
appear) Exploring Patient Information
Needs: A Cross Sectional Study of Questions on Type 2 Diabetes. PLOS ONE