Of the 2062 posts observed (N= 2062), 1392 featured questions (n= 1392). Table 1 displays an overview of the user demographics, author categorisation and patterns of usage by group. Questions were asked by 787 unique users (n= 787). 686 were female (87%), 97 were male (12%) and four could not be identified due to lack of information (1%). The majority of users were categorised as presumed diagnosis (83%). 8% were classified as seeking diagnosis and 9% were classified as a family or friend. The one user classified as other described herself as a healthcare professional. With the exception of Group Four in which no Friends or Family were identified, the proportion of different user classifications were similar across all groups.
Based on the frequency of posts made by unique profile codes, the majority of the sample population were identified as one-time users (n = 531, 67%). 256 identified as repeat users (33%). Repeat users were not restricted to a single group: 37 (14%) of the 256 repeat users were observed participating across multiple groups. The modal number of posts featuring questions made by all repeat users was two across and within each of the four groups. 122 repeat users posted more than twice during the observation period (48%).
Location information was not available for a majority of the sample. Only 96 users (12%) provided location information within the text of their post such as the name of a country, city, or state and/or location specific information such as governmental programmes e.g. Medicare. Based on this information, 83 users were located in the United States of America (USA) (11%), eight in Europe (1%),three in Canada (<1%), one in India (<1%) and one in South Africa (<1%)
Table 1. Overview of User Demographic, Author Categorisation and Patterns of Usage.
|
Unique Users
(n= 787)
|
Group 1 Users (n = 607)
|
Group 2 Users (n = 121)
|
Group 3 Users (n = 86)
|
Group 4 Users (n = 22)
|
Multiple
Group Users
(n =37)
|
Gender, n (%)
|
Female
|
686 (87)
|
526 (87)
|
194 (86)
|
77 (90)
|
19 (86)
|
28 (76)
|
Male
|
97 (12)
|
77 (13)
|
17 (14)
|
9 (10)
|
3 (14)
|
9 (24)
|
Unknown
|
4 (0)
|
4 (0)
|
0 (0)
|
0 (0)
|
0 (0)
|
0 (0)
|
Author Category, n (%)
|
Presumed Diagnosed
|
656 (83)
|
504 (83)
|
102 (84)
|
69 (80)
|
19 (86)
|
28 (76)
|
Seeking Diagnosis
|
61 (8)
|
49 (8)
|
8 (7)
|
9 (10)
|
3 (4)
|
6 (16)
|
Friends/Family
|
69 (9)
|
53 (9)
|
11 (9)
|
8 (9)
|
0 (0)
|
3 (8)
|
Other
|
1 (0)
|
1 (0)
|
0 (0)
|
0 (0)
|
0 (0)
|
0 (0)
|
User Category, n (%)
|
One Time Users
|
531 (67)
|
418 (69)
|
83 (69)
|
68 (79)
|
17 (77)
|
n/a
|
Repeat Users
|
256 (33)
|
189 (31)
|
38 (31)
|
18 (21)
|
5 (23)
|
37 (100)
|
Mean
Mode
|
Mean - 3 Mode - 2
|
Mean - 3 Mode - 2
|
Mean - 3 Mode - 2
|
Mean – 2 Mode - 2
|
Mean – 2 Mode - 2
|
|
Note: 37 repeat users (14%) were observed participating across one or more of the observed groups. The mean depicts the average number of posts made by repeat users, and the mode refers to the most common number of posts made by these users.
Question Categories
Throughout the course of the observation period users of all four participating groups asked questions that could be classified into one of the following 8 categories: MG treatment (31%), Symptoms (19%), Living with MG (12%), Diagnosis (10%), Non-MG medication (11%), Tests (8%), Location (4%) and Other (4%). The proliferation of distinct users posting different question categories in each group (figure 1) indicates that the overall frequency of question categories was not attributable to a small number of prolific users asking about the same topic.
Figure 1. Frequency of users posting different question categories
The frequency of question categories asked by authors categorised as “Presumed Diagnosed” or “Friend/Family” followed a similar pattern to one another (table 2). The frequency of question categories by authors categorised as “Seeking Diagnosis” differed substantially. This was most notable in the proportion of the sample population asking questions in the Test and Diagnosis subcategories.
Table 2. Frequency of question categories asked by different author categories.
|
MG treatment
|
Symptom
|
Living with MG
|
Non-MG treatment
|
Diagnosis
|
Tests
|
Location
|
Other
|
Author Category, n (%)
|
Presumed Diagnosed
|
371 (33)
|
203 (18)
|
152 (14)
|
129 (12)
|
104 (9)
|
63 (6)
|
49 (4)
|
48 (4)
|
Seeking Diagnosis
|
16 (13)
|
46 (28)
|
8 (5)
|
16 (10)
|
28 (17)
|
38 (23)
|
5 (3)
|
3 (2)
|
Friends/Family
|
36 (34)
|
20 (19)
|
10 (9)
|
16 (15)
|
8 (7)
|
5 (5)
|
7 (7)
|
14 (4)
|
Other
|
0 (0)
|
0 (0)
|
0 (0)
|
0 (0)
|
0 (0)
|
1 (100)
|
0 (0)
|
0 (0)
|
Note: % refers to the rate of questions categorized into one of the eight category for each respective author category.
Question Subcategories
Questions concerning MG treatment were the most common type of question across all participating groups. In total, 305 users asked 439 questions related to MG treatment. All questions were further subcategorised to reflect eight distinct categories of MG treatment queried. These included: Symptomatic treatment (n= 97; 22%), Biologicals (n= 94; 21%), IVIG (immunoglobulins) (n= 65; 15%), General (non-specific) (n= 48; 11%), Corticosteroids (n= 47; 11%), Long-term immunosuppressants (non-corticosteroid) (n= 37; 8%), Thymectomy (n= 33; 16 8%), procedures (e.g. ports; n= 16; 4%), and rescue therapies (n= 2; 0%)
Six out of the eight MG treatment subcategories qualified for cross-platform analysis (table 3). The remaining two subcategories (procedures: 11, 3% and Rescue therapies: 2, 1%) were not featured in posts from two or more distinct users in three or more of the participating groups and therefor did not qualify for cross-platform analysis. Topics of interest included: Side Effects (n= 129; 29%), Experience (n= 114; 26%), Efficacy (n= 76; 17%), Dosage (n= 52; 12%), Onset of effect (n= 31; 7%), Other (n= 30;7%) and Duration of Action (n= 7;2%). No individual side effects were validated via cross-platform analysis. Complete frequency tables for all question categories are included in the supplementary material (supplements 3-9)
Table 3. Frequency of users posting MG treatment question subcategories validated using cross-platform analysis.
|
Group 1 Users
|
Group 2 Users
|
Group 3 Users
|
Group 4 Users
|
Unique Users
|
Total Users
|
Topic
|
MG Treatment
Subcategory, n (%)
|
Symptomatic treatment
|
59 (20)
|
12 (24)
|
7 (19)
|
2 (25)
|
76 (95)
|
80 (100)
|
Side Effects
|
Biologicals
|
59 (20)
|
13 (27)
|
9 (24)
|
2 (25)
|
76 (92)
|
83 (100)
|
Experience
|
IVIG (immunglobulins)
|
37 (13)
|
6 (12)
|
8 (22)
|
3 (38)
|
53 (98)
|
54 (100)
|
Side Effects
|
General (non-specific)
|
34 (12)
|
6 (12)
|
6 (16)
|
1 (13)
|
45 (96)
|
47 (100)
|
Experience
|
Long-term
oral-immunosuppressants
|
28 (10)
|
5 (10)
|
4 (11)
|
0 (0)
|
35 (96)
|
37 (100)
|
Side Effects
|
Thymectomy
|
26 (9)
|
2 (4)
|
2 (5)
|
0 (0)
|
29 (97)
|
30 (100)
|
Experience
|
Total Unique Users
|
289 (100)
|
49 (100)
|
37 (100)
|
8 (100)
|
305 (80)
|
383 (100)
|
Side Effects
|
Note: % refers to the rate of users relative to the total number of users across each of the four groups. Unique users is the total number of users minus repeat users who were observed posting the same question across multiple groups. Topic refers to the most common topic of interest. Long-term oral immunosuppressants are non-cortocosteroid.
Discussion of treatment across the participating groups was not limited to MG. 133 unique users asked 163 questions about non-MG treatment. The most common non-MG treatment questions concerned anaesthesia, which was asked about in 15 posts (9%), and antibiotics, which were discussed in nine posts (6%). No other non-MG medication, or group of non-MG medications, equalled ≥ 5% of the total questions in this category. Validation of these non-MG questions via cross-platform analysis was not possible as the majority of non-MG treatment discussion took place in Group One (80%).
Following MG treatment, the most popular category was symptom related queries. 216 unique users asked 271 questions related to symptoms. Only 59% of symptom questions matched the items included in the standardised symptom assessment tool Myasthenia Gravis Activities of Daily living (MG-ADL). MG-ADL symptoms discussed included: breathing (n= 28; 25%), eyelid droop (n= 21; 19%) swallowing (n= 18; 16%), double vision (n= 16; 14%), impairment of ability to rise from a chair (n= 11; 10%), talking (n= 11; 10%), impairment of ability to brush or comb hair (n= 3; 3%), and chewing (n= 3; 3%). Non-MG-ADL symptom related questions included a number of specific symptoms and signs that could not be classified under the MG-ADL (see figure 2). Full definitions of each Non-MG-ADL symptom is included in the accompanying material.
Figure 2. Frequency of MG questions observed across all groups.
Only three MG-ADL symptom subcategories and two non-MG-ADL symptom subcategories qualified for cross-platform analysis (table 4). Abnormal skin sensations were reported in 40 different posts and covered a wide variety of sensations including cramping, twitching, tingling and spasms. The 32 posts featuring questions categorised as pain were less diverse. Although reports of pain were not limited to one region of the body, neck, back and shoulder pain accounted for 40% of the non-MG-ADL symptom questions and joint and muscle pain (13%).
Table 4. Frequency of users posting MG-ADL and Non-MG-ADL symptom subcategories that were validated using cross-platform analysis.
|
Group 1 Users
|
Group 2 Users
|
Group 3 Users
|
Group 4 Users
|
Unique Users
|
Total Users
|
ADL Symptoms, n (%)
|
Breathing
|
20 (74)
|
5 (19)
|
2 (7)
|
0 (0)
|
25 (93)
|
27 (100)
|
Ptosis
|
14 (67)
|
2 (10)
|
5 (24)
|
0 (0)
|
21 (100)
|
21 (100)
|
Swallowing
|
11 (61)
|
2 (11)
|
3 (17)
|
2 (11)
|
17 (94)
|
18 (100)
|
Non-ADL Symptoms, n (%)
|
Abnormal sensations
|
28 (68)
|
6 (15)
|
3 (7)
|
4 (10)
|
39 (95)
|
41 (100)
|
Pain
|
23 (72)
|
6 (19)
|
3 (9)
|
0 (0)
|
32 (100)
|
32 (100)
|
Note: % refers to the rate of users relative to the total number of users across each of the four groups. Unique users is the total number of users minus repeat users who were observed posting the same question across multiple groups. Abbreviation: MG-ADL Myasthenia gravis activities of daily living (score).
Following the category “Symptoms”, “Living with MG” was the third most frequently category of question. 138 unique users asked 174 questions categorised as such. This was the widest of all question categories due to the scope of issues, which could be classified as aspects of routine or daily life, namely: dietary (n= 53; 30%), Health Insurance (35; 20%), non-medical living assistance (e.g. mobility aids, prisms glasses) (n= 31; 18%), personal relationships (n= 12; 7%), mental health and stress management (n= 11; 6%), weather (n= 10; 6%), exercise (n= 9; 5%), employment (n= 6; 3%), travel and/or holiday making (n= 4; 2%), and driving (n= 3; 2%). Only three subcategories were validated using cross-platform analysis (table 5).
Table 5. Frequency of users posting living with MG subcategories that were validated using cross-platform analysis.
|
Group 1 Users
|
Group 2 Users
|
Group 3 Users
|
Group 4 Users
|
Unique Users
|
Total Users
|
Living with MG
Subcategory, n (%)
|
Dietary
|
36 (73)
|
4 (8)
|
7 (14)
|
2 (4)
|
48 (98)
|
49 (100)
|
Non-medical Living Assistance
|
23 (77)
|
4 (13)
|
2 (7)
|
1 (3)
|
29 (97)
|
30 (100)
|
Health Insurance
|
20 (69)
|
5 (17)
|
2 (7)
|
2 (7)
|
28 (97)
|
29 (100)
|
Note: % refers to the rate of users relative to the total number of users across each of the four groups. Unique users is the total number of users minus repeat users who were observed posting the same question across multiple groups.
In total, 115 unique users asked 140 questions related to the diagnosis of MG (n= 76; 54%) or another disease (n= 64; 46%). Of the 57 unique users who asked about a non-MG diagnosis, the five most common non-MG diagnosis discussed were Covid-19 (n= 8; 14%), Thymoma (n= 5; 9%), Sleep Apnea (n= 3; 5%), Amyotrophic Lateral Sclerosis (ALS) (n= 3; 5%), Lambert-Eaton Syndrome (n= 3; 5%) and Amyotrophic Lateral Sclerosis (ALS) (n= 2; 4%). No non-MG diagnosis questions could be validated via cross-platform analysis.
90 unique users asked 110 questions related to tests. Test subcategories included, lab testing/ antibodies (44%), lung function (15%), nerve and muscle stimulation (15%), general unspecified (11%); imaging (6%), clinical examination (5%), and pathology (4%). With the exception of laboratory tests/antibodies, which asked about by 34 users in Group One (38%), six in Group Two (7%), three in Group Three (3%) and one in Group Four (1%), no other question subcategory could be validated using cross-platform analysis.
57 unique users asked 62 questions concerning contacts in their local area. Of those users 73% (n= 42) sought information concerning professional contacts with MG experience in their local area, 18% sought other diagnosed patients and 9% enquired about lay and professional contacts. Only a need for professional contacts, which was asked about by 32 users in group One (71%), eight in Group Two (18%), five in Group Three (11%) was validated using cross-platform analysis. No requests for professional contacts were made in Group Four.
55 questions asked by 44 unique users could not be categorised. As such, they could not be validated using cross-platform analysis and were excluded from further analysis. Due to data protection issues, it was not possible to include the exact wording of questions categorised as other.