aschwalbc.bsky.social
PhD candidate + MSc Epidemiology @LSHTM | 🇵🇪 in 🇬🇧 | MD @CayetanoHeredia | Researcher @IMTAVH_UPCH | Learning R and Greek | Imagine pleasant nonsense.
45 posts
2,026 followers
280 following
Regular Contributor
Active Commenter
comment in response to
post
Tagging @cugartegil.bsky.social now that he joined Bluesky. #TBSky
comment in response to
post
Massive thanks to my supervisors: Rein Houben (@reinhouben.bsky.social), Katherine Horton (@kchorton.bsky.social), and César Ugarte-Gil. Your knowledge and passion for TB research
have been invaluable to me. Huge thanks to Pete Dodd (@petedodd24.bsky.social) for being a great advisor throughout.
comment in response to
post
Agree. Symptom-screening would miss a high proportion of individuals with asymptomatic TB, even if less strict definitions of symptomatology are applied.
comment in response to
post
As promised: tinyurl.com/2pmudzrk
comment in response to
post
14/ Huge thanks to my co-authors: @kchorton.bsky.social @laragosce.bsky.social @reinhouben.bsky.social and many many others not yet in Bluesky. @lshtm.bsky.social @lshtm-tbmod.bsky.social @tbmac.bsky.social
comment in response to
post
13/ This modelling study - tinyurl.com/2pmudzrk - shows how population-wide screening can rapidly reduce TB prevalence in Viet Nam. Front-loaded costs promise reduced morbidity, mortality, and progress towards the End TB Strategy, as BAU is unlikely to meet targets.
NOT YET PEER-REVIEWED
comment in response to
post
12/ Estimated ICERs against the BAU for all algorithms were below the estimated cost-effectiveness threshold range for Viet Nam. The two-step CXR+NAAT algorithm averted 4.29m DALYs (95%UI:2.86-6.14) at US$225 (95%UI:85-520) per DALY averted compared with BAU.
NOT YET PEER-REVIEWED
comment in response to
post
11/ Total intervention costs were substantial. However, they yield persistent cost savings, likely continuing beyond the 25-year time horizon. Most costs were for screening in NAAT-based algorithms and treatment in the CXR-only algorithm.
NOT YET PEER-REVIEWED
comment in response to
post
10/ All screening algorithms resulted in lower cumulative TB incidence, TB deaths, and DALYs compared with BAU between 2025 and 2050. Reductions were similar for NAAT-based algorithms but greater with the CXR-only algorithm.
NOT YET PEER-REVIEWED
comment in response to
post
9/ NAAT-based algorithms required at least six rounds to reach the prevalence threshold, while CXR-only required three. The threshold was not reached within the time horizon under BAU. NAAT-only achieved a prevalence reduction consistent with the ACT3 trial after three rounds.
NOT YET PEER-REVIEWED
comment in response to
post
8/ We determined the annual screening rounds needed to reduce pulmonary TB prevalence below 50 per 100,000 people. We evaluated algorithms using incremental cost-effectiveness ratios, showing additional costs (US$) per DALY averted compared to business-as-usual (BAU) by 2050.
comment in response to
post
7/ We calibrated a deterministic transmission model to TB epidemiology in Viet Nam. We designed three population-wide screening algorithms from 2025: sputum nucleic acid amplification tests (NAAT, Xpert MTB/RIF Ultra) only; CXR followed by NAAT; and CXR-only without microbiological confirmation.
comment in response to
post
6/ Despite the evident promise of population-wide screening interventions, their implementation as a central component of the TB elimination strategy remains under debate. Key challenges include determining the optimal duration, frequency of repeated screening, and the best screening algorithm.
comment in response to
post
5/ Contemporary evidence is mixed, but the ACT3 trial (Viet Nam, 2014–2018) - tinyurl.com/3xfvzru5 - showed a significant reduction in TB prevalence. This trial consisted of annual, community-wide screening over three years, using a symptom-agnostic approach similar to historic screening efforts.
comment in response to
post
4/ Mass screening interventions for TB have a history of success! Notable examples include chest radiography (CXR) 🩻 campaigns in Glasgow (tinyurl.com/yc8xmvp6) and the KolÃn district (tinyurl.com/yaa29rdj).
comment in response to
post
3/ Through this approach: i) individuals are diagnosed and treated after an extended period of infectiousness, contributing to ongoing transmission; and ii) a large gap of undiagnosed individuals is left, as not everyone with TB experiences symptoms or is able to access care.
comment in response to
post
2/ The conventional approach to TB prevention and care is passive detection, where diagnosis and treatment are only provided to individuals with symptoms who seek and receive healthcare. This approach is insufficient in order to achieve ambitious End TB Strategy targets.
comment in response to
post
Nice to see it out!
comment in response to
post
Ditto
comment in response to
post
On my wish list for next year!
comment in response to
post
Hi Olivier, can I join?
comment in response to
post
Thanks, Gabriele. Appreciate it! I'm in the process of dragging the whole TB Modelling Group from LSHTM to Bluesky!
comment in response to
post
12/ Finally, if you are in Bali for the #UnionConf, I will be presenting this work on Friday (Session OA42) and Saturday (Session SP37). The work is also available in the e-poster stand for TBScience (TBS-EP01).
comment in response to
post
11/ Huge thanks to my co-authors: Pete Dodd (@petedodd24.bsky.social), Hannah Rickman, César Ugarte-Gil, Katherine Horton (@kchorton.bsky.social), and Rein Houben (I will convince them to join 🦋). Additional thanks to Marcel Behr (@mbehr-mcgill.bsky.social), Dave Moore, and Tomos Prys-Jones.
comment in response to
post
10/ In summary, we use new scientific insights that account for the dynamic nature of TB to present the first medically actionable target estimate of the global population harbouring viable Mtb infection.
tinyurl.com/yc6kbay9
NOT YET PEER-REVIEWED
comment in response to
post
9/ Of all recent infections, over 87% were found in the three regions of Southeast Asia, Western Pacific, and Africa. At the country level, India, China, and Indonesia together account for about 50% of recent global Mtb infections.
NOT YET PEER-REVIEWED
comment in response to
post
8/ We found that in 2022, 156 million people (95%UI:127-199) were recently infected with viable Mtb, equating to 2.0% (95%UI:1.6-2.5) of the global population, with substantial regional variability.
NOT YET PEER-REVIEWED
comment in response to
post
7/ By integrating these insights, our model generated estimates for 171 countries (covering 99.6% of the world population), all six WHO regions, and globally.
NOT YET PEER-REVIEWED
comment in response to
post
6/ Thirdly, we focused on recent infections, acknowledging that the highest risk of progression to disease occurs within the first two years since infection.
comment in response to
post
5/ Secondly, we also account for the evidence that many individuals may self-clear Mtb infection, calibrating these rates using data from @kchorton.bsky.social et al. - tinyurl.com/ykap23kx - on infection and disease pathways in a simulated cohort.
comment in response to
post
4/ Firstly, we adjusted annual infection risk (ARI) trends for immunoreactivity reversion and age-based social mixing to address ARI underestimation, offering a more accurate infection force estimate.
NOT YET PEER-REVIEWED.
comment in response to
post
3/ Recent insights into the natural history of TB warrant a more accurate estimate of the global burden of viable Mtb infection. Our approach extends the methods of the 2016 paper by including key new insights around reversion, age-specific mixing patterns and self-clearance.
comment in response to
post
2/ A previous study estimated that a quarter of the global population had a 'latent' TB infection - tinyurl.com/2ap23d6k. However, this estimate essentially reflects individuals exposed to Mtb and still exhibiting immunoreactivity, rather than viable Mtb infection.