Call Girls Coimbatore Just Call 8250077686 Top Class Call Girl Service Available
An Investigation Into The Early Detection Of Psychosis And Depression In Individuals Via Social Media Amanda COLLINS-EADE
1. An investigation into the early detection
of psychosis and depression in
individuals via Social Media (Twitter)
Amanda Collins-Eade, Ed de Quincey*, JohnAmanda Collins-Eade, Ed de Quincey*, John
Foster, Lynne JumpFoster, Lynne Jump
University of Greenwich, UK;University of Greenwich, UK;
*Keele University, UK*Keele University, UK
2. IntroductionIntroduction
• A spate in suicides has been linked by the media to social networking.
There is also a rise in admissions from young people self harming in the UK.
Luxton et al, (2012)
• Whilst Twitter is used to vent feelings and gain emotional support there are
fears that people who frequently post ‘vague’ but worrying messages may
alienate online friends who may not take them seriously. (Forest and Wood
2012; Seidman 2014)
• Vague posts on social media should not be hastily dismissed because they
can serve as an early warning system for timely mental health intervention.
(Hoffman 2012) However, identifying worrying behaviour in online posts is
not straight-forward. Families are not trained to differentiate between
teenage angst or recognise signs of early mental health problems.
• Twitter currently allows third party access to user’s posts and profile
information, therefore a possible area to explore in early detection of mental
health problems. (Park and Calamaro 2013)
3. MethodMethod
• Pilot study undertaken to analyse Tweets
that identify possible online signs of
distress
• Key words/phrases were used from known
mental health assessment tools and
collected via the Twitter search API
• Of 20,800 UK tweets collected, a subset of
2,500 were analysed and rated
4. Key words used to collect tweets from Twitter
Key words Key words Key words
I have nothing to look
forward to
Seeing things
Wish I was dead
I look ugly
Someone is controlling my
thoughts
Too scared to go out
People hate me Nervous and apprehensive
Something terrible is
going to happen
Suicidal thoughts People are watching me
Hearing voices
Suicidal thoughts People are telling me to do
things
Feeling irritated
I’m being watched I hate myself Better off dead
5. Tweet examplesTweet examples
• ‘I always have a feeling people are talking about me.’
• ‘I feel like I’m nothing, I’m worthless, I hate myself’
• ‘I hate having anxiety, I can barely leave the house
without feeling sick, overthinking and trying to talk
myself out of it’
• ‘This life is super hard My life isnt worth living its all
going down the drain anyways so y not just finish it
instead of wating for my time’
6. ResultsResults
• Not all tweets containing these phrases indicated mental health
concerns.
• Tweets such as ‘Can’t wait to see you tomorrow,’ mental health
charity adverts and lyrics from popular bands.
• Some tweets had a strong depressive tone but finished with
‘LOL’ or a smiley face emoji. #sarcasm hashtag indicates that a
tweet should not be taken literally but was not used in these
tweets
7. ResultsResults
Out of 2,500 tweets;
•783 ‘seeing things’, 1% (n=26) identified as a concern.
•462 ‘I am ugly’ (n= 236) 9.4% of these tweets were a concern.
•625 ‘I hate myself’ (N= 326) 13% were a concern.
•115 ‘suicidal thoughts’ (N=46)1.8% were a concern
•115 ‘wished they were dead’ (N=59) 2.4% were a concern
•74 ‘hearing voices’ (N=10) 0.4% were a concern
•19 ‘too scared to go out’ (N=6) 0.2% were a concern
8. Inter rater reliabilityInter rater reliability
• 2 independent judges (nurses) were asked to rate 100 of the
2,500 rated tweets
• This was to explore whether the tweets could be construed as a
concern
• Asked to rate whether they were ‘concerned’, ‘not concerned’ or
‘unsure’
• Children’s nurse (Rater 1) agreed with the author re concern
16.4% of time. Agreement was higher when unsure, (52.9%)
With no mental health concern, agreement reached 53.6%.
Kappa = 0.060 (p<0.307)
9. Inter rater reliabilityInter rater reliability
• Mental health nurse (Rater 2) agreed with the author as to what
was concerning 56.4% of the time. “unsure” ratings reached
agreement of 47.1%. Kappa = 0.151 (p <0.018).
• Rater 1 and rater 2 agreed with one another as to what was
concerning 22.2%. Unsure ratings reached 40% agreement.
Kappa = 0.003 (p<0.947)
• No strong agreement between the judges as to what
constituted a worrying tweet
• One potential explanation relates to some judges tending to
choose “unsure" more frequently than others.
10. Last thoughtsLast thoughts
• Whilst the percentages are small, the results suggest that
distress is expressed on Twitter.
• Low self-esteem, self-image issues and mental health
difficulties posted to online friends, family and other Twitter
users they may not be familiar with.
• Actual or vague thoughts of self-harm or suicide seen on social
media may cause concern or other responses from those
reading them. More work is needed to identify concerns as
health professionals struggled to agree on an expression of
distress.
• Future work includes considering removal of the “unsure” option
and repeating with different independent Judges.
11. Thank you for your interest!Thank you for your interest!
12. ReferencesReferences
• Beck, A. T., Steer, R. A., and Carbin, M., (1988) Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation, Clin psych
review, 8(1) 77-100
• Brown, J., (2014) ‘Why don’t you kill yourself?’ Hannah Smith’s grieving father calls for halt to sinister world of internet abuse [online available at]
http://www.independent.co.uk/news/uk/home-news/why-dont-you-kill-yourself-hannah-smiths-grieving-father-calls-for-halt-to-sinister-world-of-internet-abuse-
(Accessed 29th December 2014)
• Department of Health (2001) Mental Health policy implementation guide, London,HMSO Department of Health
(2012) Preventing suicide in England, A cross Government outcomes strategy to save lives, London, DH
• Flux, E,.(2012 ) Is it ok to vaguebook? [online available at] http://lipmag.com/opinion/is-it-okay-to-vaguebook/ (Accessed 29th
December 2014)
• Forest, A. L., and Wood, J. V., (2012) When social networking is not working: Individuals with low self esteem recognise but do not reap the benefits of
self disclosure on Facebook, Psychological Science, 23(3) 295-302.
• Health and Social Care Information Centre (2013) Hospital statistics on teenagers: girls predominate in self-harm cases, boys in assaults, HSCIC
• Hoffman, J, (2012) Online cries for help can signal depression, specialists say, [online available at] http://health.heraldtribune.com/2012/02/24/online-
cries-for-help-can-signal-depression-specialists-say/ (Accessed 29th
December 2014)
• Hsu, C,. (2012) Facebook May Detect Adolescent Depression Before Doctors or Parents: [online available at]
http://www.medicaldaily.com/articles/9172/20120224/facebook-twitter-depression-psychology-social-networking-risk.htm (Accessed 29th December 2014)
• Krawiecka, M., Goldberg, D., and Vaughn, M. A., (1977) Standardised psychiatric assessment scale for rating chronic psychotic patients, Acta
Psychiatrica Scandinavica, 55, 299-308
• Lester, H., Tait, L., Khera, A., Birchwood, M., Freemantle, N,. and Patterson, P., (2005) Primary care teaching – The development and implementation of
an educational intervention on first episode psychosis for primary care: Medical education, 39(10) 1006-1014
• Luxton D D, June J D, Fairall J D, (2012) Social media and suicide: A public health perspective, American Journal of public health, 102, s195-s200
• Moreno, M. A., VanderStoep, A., Parks, M. R., Zimmerman, F. J., Kurth, A., Christakis, D. A., (2009) Reducing At-Risk Adolescents' Display of Risk
Behavior on a Social Networking Web Site, A Randomized Controlled Pilot Intervention Trial, Arch Pediatr Adolesc Med, 163(1):35-41
• Papyrus (2014) Suicide remains highest cause of death in young people, [online] available at; http://hopelineuk.org.uk/more/news/685 (Accessed 23rd
June 2015)
• Park, B. K., and Kalamaro, C., (2013) :A systematic review of social networking sites: Innovative platforms for health research targeting adolescents and
young adults, J. Nursing Schol, 45 (3) 256-264
• Seidman, G., (2014) Expressing the ‘true self’ on Facebook, Computers in human behaviour, 31, 367-372
• de Quincey, E., and Kostkova, P., (2010) Early warning and outbreak detection using social networking websites: The potential of Twitter, Electronic
Healthcare, Berlin Heidelberg: Springer p21-24
• de Quincey, E., Kyriacou, T., Williams, N. and Pantin, T., (2014) Potential of Social Media to determine hay fever seasons and drug efficacy.
Planet@Risk, 2(4), Special Issue on One Health: 293-297, Davos: Global Risk Forum, GRF Davos [Online available at] http://planet-
risk.org/index.php/pr/article/view/96/236 (Accessed 13 December 2014)