Ben OutdoorsOnly GoldacreVerified account

@bengoldacre

Doctor, nerd cheerleader, Bad Science person, stats geek, procrastinator. I run at making tools+papers from data ben@badsci

Oxford, England
badscience.net
Joined June 2007
1,938 Photos and videos Photos and videos

Tweets

You blocked @bengoldacre

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @bengoldacre

  1. Pinned Tweet
    May 14

    NEW OpenSAFELY PAPER Prevalence of "Long Covid" in 58 MILLION patients' primary care records. You read that correctly, this study was conducted in 58 MILLION patients full GP records, analysed in situ. Now, a puzzle…

    44 replies 130 retweets 253 likes
    Show this thread
    Show this thread
    Undo
  2. Retweeted
    May 25

    The challenges and opportunities of mental health data sharing in the UK A cast of thousands have written this important new paper in Inc. 👏🏽

    1 reply 11 retweets 16 likes
    Show this thread
    Show this thread
    Undo
  3. Retweeted
    May 24

    New from me and in : We expanded our earlier work in examining results reporting requirements of the FDA Amendments Act 2007 to additional areas of compliance across the entire dataset.

    1 reply 16 retweets 27 likes
    Show this thread
    Show this thread
    Undo
  4. Retweeted
    May 24

    How well are clinical trial sponsors reporting their results? Another excellent analysis by and via part of

    1 reply 4 retweets 3 likes
    Undo
  5. May 24

    My group in Oxford has published LOTS on who does, and doesn't, report their clinical trial results. But what about the rest: late registrations, data verification, certificates of delay, document sharing? Our lashes detail in

    New from me and in : We expanded our earlier work in examining results reporting requirements of the FDA Amendments Act 2007 to additional areas of compliance across the entire dataset.
    Show this thread
    1 reply 9 retweets 20 likes
    Undo
  6. May 21

    3 likes
    Undo
  7. May 21

    How we built a full end-to-end federated analytics pipeline to analyse 58 million NHS patients' GP records securely, with all data kept in situ, as a small band of dedicated geeks, during a horrible pandemic.

    1 reply 12 retweets 50 likes
    Undo
  8. Retweeted
    May 16

    10 WAYS TO TO "BOOST" YOUR IMMUNE SYSTEM DURING THE COVID-19 PANDEMIC: 1. VACCINES 2. VACCINES 3. VACCINES 4. VACCINES 5. VACCINES 6. VACCINES 7. VACCINES 8. VACCINES 9. VACCINES 10. VACCINES

    94 replies 658 retweets 3,664 likes
    Undo
  9. May 18

    The mighty DataLab posse. Couldn't be more proud. Look upon our works and collaborate!

    13 replies 8 retweets 36 likes
    Undo
  10. May 17

    Worth saying - as per paper - it might be any number of things causing this low rate of Long COVID: whether patients present, whether doctors know and believe in the diagnosis, whether they diagnose, whether they use the codes, whether the codes are designed optimally, and more.

    Huge National study of Long covid finds most Cases being not recorded by GPs with a prevalence of only around 4/10,000 of the population.
    33 replies 5 retweets 25 likes
    Undo
  11. Retweeted
    May 17

    This thread is an encouraging masterclass about tackling an important pandemic issue with Science and team work. Lovely reminder that grown ups are helping us out of this, behind the scenes.

    The rate of Long Covid, looking at diagnostic codes in primary care, is really low: 40 cases per 100,000 people. This seems much lower than other estimates. So what is happening? Patients not presenting? Missed diagnoses? A high diagnostic threshold? Low awareness of the codes?
    Show this thread
    7 replies 5 retweets 21 likes
    Undo
  12. Retweeted
    May 17

    See this excellent thread on "Long Covid". There is a BIG mismatch between assumed prevalence and GP reporting overall - and between practices and the two main GP reporting systems.... ⬇️⬇️⬇️

    The rate of Long Covid, looking at diagnostic codes in primary care, is really low: 40 cases per 100,000 people. This seems much lower than other estimates. So what is happening? Patients not presenting? Missed diagnoses? A high diagnostic threshold? Low awareness of the codes?
    Show this thread
    7 replies 5 retweets 14 likes
    Undo
  13. May 17

    OpenSAFELY PAPER Prevalence of "LONG COVID" in 58 million patients' primary care records is oddly low. Why? Patients not being seen? GP beliefs? GP coding behaviour? Software design? Diagnostic code design? We dive deep into the data with open science and a crack team. THREAD:

    NEW OpenSAFELY PAPER Prevalence of "Long Covid" in 58 MILLION patients' primary care records. You read that correctly, this study was conducted in 58 MILLION patients full GP records, analysed in situ. Now, a puzzle…
    Show this thread
    29 replies 8 retweets 26 likes
    Undo
  14. Retweeted
    May 16

    Indeed, Friday night I walked round Oxford disheartened by what it could be and how much better it would be for businesses if George, Broad, Magdalen Streets, the High & St Aldates were open to humans!

    What generates more revenue for businesses? These people or the 3 or 4 parked cars they replaced?
    Show this thread
    8 replies 11 retweets 104 likes
    Undo
  15. Retweeted
    May 15

    Seems entirely appropriate that one of the lead cheerleaders of the anti-lockdown marches today (aka the Awful Poo Lady - ) has a past making extremely dubious claims to academic credentials

    7 replies 12 retweets 40 likes
    Undo
  16. Retweeted
    May 11

    This was one of my paper ideas and some guy on Reddit just went and did it

    54 replies 498 retweets 2,422 likes
    Undo
  17. Retweeted
    May 15

    Fabulous, and fabulously important, thread

    NEW OpenSAFELY PAPER Prevalence of "Long Covid" in 58 MILLION patients' primary care records. You read that correctly, this study was conducted in 58 MILLION patients full GP records, analysed in situ. Now, a puzzle…
    Show this thread
    2 replies 2 retweets 8 likes
    Undo
  18. Retweeted
    May 14

    Such a useful thread here . Need good data to build focused interventions. Over half of OTs using COG-OT told us they are working with covid patients. Worry has been people not sharing Long covid with GPs so lots of opportunities with 👍👍

    NEW OpenSAFELY PAPER Prevalence of "Long Covid" in 58 MILLION patients' primary care records. You read that correctly, this study was conducted in 58 MILLION patients full GP records, analysed in situ. Now, a puzzle…
    Show this thread
    1 reply 2 retweets 3 likes
    Undo
  19. Retweeted
    May 14

    Long COVID is much rarer in health records than self reported. But for some not so obvious reasons. The term ‘Long COVID’ is not used in electronic medical record systems. So may be that some doctors are trying to add, can’t find it, so it’s systematically underreported.

    NEW OpenSAFELY PAPER Prevalence of "Long Covid" in 58 MILLION patients' primary care records. You read that correctly, this study was conducted in 58 MILLION patients full GP records, analysed in situ. Now, a puzzle…
    Show this thread
    6 replies 11 retweets 19 likes
    Show this thread
    Show this thread
    Undo
  20. Retweeted
    May 14

    This is an excellent and comprehensive thread - well done on impressive and fast collaborative work

    NEW OpenSAFELY PAPER Prevalence of "Long Covid" in 58 MILLION patients' primary care records. You read that correctly, this study was conducted in 58 MILLION patients full GP records, analysed in situ. Now, a puzzle…
    Show this thread
    6 retweets 4 likes
    Undo
  21. Retweeted
    May 14

    Interesting thread on surprisingly low prevalence of long Covid in primary care records

    NEW OpenSAFELY PAPER Prevalence of "Long Covid" in 58 MILLION patients' primary care records. You read that correctly, this study was conducted in 58 MILLION patients full GP records, analysed in situ. Now, a puzzle…
    Show this thread
    2 retweets 5 likes
    Undo

@bengoldacre hasn't Tweeted yet.

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·