Transcript What are online communities?

Who benefits from using internet support groups for those affected by cancer?
Dr Henry WW Potts & Professor Amanda J Ramirez
{henry.potts, amanda-jane.ramirez}@kcl.ac.uk
Cancer Research UK London Psychosocial Group, Guy’s, King’s and St Thomas’ School of Medicine, King’s College, London
Introduction
Method
Internet usage is growing
· Over a third of UK homes are connected to the internet
and the figure is growing.1
· >100,000 medical websites exist.2
· 85% of UK doctors report some patients experienced
health benefits from using the internet, but 44% that
some patients experienced health problems.3
· 2001 ONS survey of UK public: 14% gave the internet
as somewhere they would go for information on cancer.4
· In two recent surveys of cancer patients, 10% in
Northern Ireland5 and 23% in London6 had used the
internet. Usage was more frequent among patients who
were younger5,6, better educated6 and from a higher
social class.6 There were no differences by gender or
diagnosis.5,6 The internet was rated as a high quality
information source.5
Information and support needs
· Most cancer patients want and need high quality
information tailored to their situation, yet a lack of
information remains a common complaint.7
· Social support reduces risk of psychiatric morbidity.8
· Face-to-face support groups can be useful for those
who participate in them.9
· The internet is not only a novel way of accessing static
information, but may also provide support for cancer
patients through contact with other patients and online
communities.
What are online communities?
· Technologies whereby groups of people can take part
in many-to-many communication in a virtual forum or
‘space’.
· Online communities defined as (1) people who
interact, (2) to a shared purpose, (3) under policies
that guide their interaction, (4) using computer
systems.10
· Technologies include mailing lists; Usenet & other
bulletin boards; chat rooms; Web fora.
· Online communities present several possible benefits
over traditional support structures.
· Cancer-related online communities date back to at
least 1992 (alt.support.cancer on Usenet founded).
· There is no way of tracking what or how many fora
exist.
Aims
·
·
·
To examine the profile of cancer-related internet
support groups.
To determine whether the extent of online activity
pertaining to different cancer types is related to
incidence figures or whether certain cancer types are
comparatively over- or underrepresented online.
To assess the potential for future research in this
area.
Yahoo! Groups is a major supplier of mailing lists with 100,000s in operation.
A search of Yahoo! Groups using cancer terms and Yahoo!’s own directory as
carried out (Mar/Apr 02). Publicly available information about each list at the
Yahoo! Groups website was collated. The number of mailing lists specific to
each type of cancer was recorded (see Table). This was compared with
incidence figures using χ2 and ordinal correlation (Kendall’s τb) as measures of
goodness of fit.
Results
We found 1147 English-language mailing lists that cover the experience or
treatment of cancer or cancer sequelae. Of these, 308 were active mailing
lists that provide support and/or information for those affected by cancer.
A screen shot of a Yahoo! Groups mailing list: colon_cancer_support
A comparison of the number of mailing lists to incidence figures by cancer type.
Cancer type
Number of mailing
lists specific to
cancer type
Incidence (US SEER
9 Registry, 1999)
Head & neck
16
5830
360
2380
150
Oesophagus
4
1240
310
200
50
Stomach
0
2110
Colo-rectal
9
14330
1590
1970
220
Liver
1
1360
1360
330
330
Pancreas
0
2650
∞
320
∞
Lung
2
15960
7980
1840
920
Sarcoma
5
910
180
470
90
Skin
5
Breast
36
23390
Gynaecological
13
Prostate
30note 2
Testis
7420note 1
Ratio: US cases per
no. of lists
∞
Incidence in those
under 55 (US SEER 9
Registry, 1999)
Ratio: US cases under
55 per no. of lists
320
∞
3098note 1
620
650
8320
230
8050
620
3090
240
18870
630
1550
50
1
730
730
700
700
Kidney
0
2820
∞
670
∞
Bladder
2
5230
2610
580
290
Brain/CNS
29
1720
60
820
30
Haematological
26
9530
370
2950
110
Other
3
2700
900
700
230
Childhood
11
800
70
800
70
1480
[1] Basal and squamous cancers not recorded.
[2] Most of the mailing lists for prostate cancer are part of an interrelated set in which people are expected to be on several different mailing lists, exaggerating the
apparent amount of activity.
The number of mailing lists specific to each cancer type was compared to the incidence of that cancer type. The internet is international
in usage, but English-speaking users are predominantly from the US, so US incidence figures are used (SEER Registry 9, 1999 results).
The table given is a summary of more detailed results.
The ratios illustrate how the number of mailing lists is only approximately related to the frequency with which that cancer type occurs. For
example, lung cancer is very common, but only two mailing lists specific to lung cancer were found, whereas 29 mailing lists pertaining
to brain/CNS tumours were found despite brain/CNS cancers being comparatively uncommon. The ratios with incidence in the under55s are somewhat more consistent, illustrating a better fit to the number of mailing lists.
Mailing lists cover a huge breadth and depth of patient types. There are lists
pertaining to:
• 58 different cancer types, including very rare cancers: e.g. tongue
carcinoma, Ewing’s sarcoma, optic glioma
• Various sequelae: e.g. pain, lymphoedema, cerebellar mutism
• Specific treatments: e.g. Temozolomide, Gleevec
• Religious groups: Christian, Muslim
• Other very specific groups: e.g. lesbians with breast cancer, war veterans
with prostate cancer
Profile by cancer type
·
We would expect more mailing lists to exist for cancers with greater
incidence, but there is only a weak relationship (correlation across cancer
types, τb = 0.43).
·
There are more mailing lists, and more subscribers to those lists,
pertaining to thyroid, brain/CNS and haematological cancers and fewer to lung
cancer than expected when compared to incidence (based on χ2 components
>10).
·
There is some suggestion that the number of mailing lists is greater for
cancer types with a younger age profile. The number of mailing lists shows a
stronger correlation with incidence in the under-55s (τb = 0.60). This
presumably reflects that internet users tend to be younger.
Conclusions and future research
·
A huge variety of mailing lists exist allowing individuals to
communicate
with others in similar circumstances.
·
Why do some comparatively rare cancers engender more online
activity?
Age may be an explanatory factor.
·
Next steps: questionnaire surveys of mailing list users to analyse the
benefits (and drawbacks) of online communities.
References
1.BBC News (2001). UK internet use growing. BBC News: Business. URL: <http://news.bbc.co.uk/1/hi/business/1292844.stm> Accessed 28 Oct 2002
2.Eysenbach G, Sa RE, Diepgen TL (1999). Shopping around the internet today and tomorrow: towards the millennium of cybermedicine. British Medical
Journal, 319, 1294
3.Potts HWW, Wyatt JC (2002). Online survey of doctors’ experience of patients using the internet. Journal of Medical Internet Research, 4, e5
4.K McCaffrey & J Wardle, pers. comm.
5.Mills ME, Davidson R (2002). Cancer patients’ sources of information: use and quality issues. Psycho-Oncology, 11, 371-8
6.D Wilkins, pers. comm.
7.Meredith C, Symonds P, Webster L et al. (1996). Information needs of cancer patients in west Scotland: cross sectional survey of patients’ views. British
Medical Journal, 313, 724-6
8.Pinder KL, Ramirez AJ, Richards MA, Gregory WM (1994). Cognitive responses and psychiatric disorder in women with operable breast cancer. PsychoOncology, 3, 129-37.
9.Fitch M (2000). Supportive care for cancer patients. Hospital Quarterly, 3, 39-46
10.Preece J (2000). Online Communities: Designing Usability, Supporting Sociability. Chichester: John Wiley.