Bastien Chabe-Ferret (he/his/him)
@bastiencf.bsky.social
Senior Lecturer - Economics / Demography / Health - BA Film student - Middlesex Uni London
Feminist / Antiracist / 🏳️🌈🏳️⚧️ ally
https://sites.google.com/site/bastiencf/home
Feminist / Antiracist / 🏳️🌈🏳️⚧️ ally
https://sites.google.com/site/bastiencf/home
TFW your favourite Nobel prize winner cites your paper :-)
May 30, 2025 at 10:19 AM
TFW your favourite Nobel prize winner cites your paper :-)
Good riddance
January 21, 2025 at 6:39 PM
Good riddance
Given the recent inflow of new comers here, I thought I would re-post this thread about my recent BMC public health paper. I believe it will be of interest to everyone working on the determinants and consequences of #breastfeeding, esp. in low & middle income countries #EpiSky #EconSky
December 9, 2024 at 5:06 PM
Given the recent inflow of new comers here, I thought I would re-post this thread about my recent BMC public health paper. I believe it will be of interest to everyone working on the determinants and consequences of #breastfeeding, esp. in low & middle income countries #EpiSky #EconSky
And a happy Monday to you too 😊
November 18, 2024 at 9:20 AM
And a happy Monday to you too 😊
Couldn’t choose 🙈
November 16, 2024 at 8:15 AM
Couldn’t choose 🙈
For exclusive breastfeeding (EBF), the study found that in many cases, non-exclusivity is due to ingestion of plain water. Including water (quasi-EBF) leads to significant differences (20pp on avg), especially large in West and Middle Africa & slightly decreasing over time. 5/N
October 10, 2024 at 2:14 PM
For exclusive breastfeeding (EBF), the study found that in many cases, non-exclusivity is due to ingestion of plain water. Including water (quasi-EBF) leads to significant differences (20pp on avg), especially large in West and Middle Africa & slightly decreasing over time. 5/N
These discrepancies are also NOT classical measurement error: the underestimation of EIB is worse for South Asia, & has increased over time, leading to downward-biased trends. 4/N
October 10, 2024 at 2:14 PM
These discrepancies are also NOT classical measurement error: the underestimation of EIB is worse for South Asia, & has increased over time, leading to downward-biased trends. 4/N
🚨 Key finding: some @WHO/@UNICEF indicators are not robust to small changes in their definition, in particular early initiation of breastfeeding (EIB) & exclusive breastfeeding under 6 months (EBF). Continued breastfeeding at 1-2 years (CBF) is much more reliable. 2/N
October 10, 2024 at 2:13 PM
🚨 Key finding: some @WHO/@UNICEF indicators are not robust to small changes in their definition, in particular early initiation of breastfeeding (EIB) & exclusive breastfeeding under 6 months (EBF). Continued breastfeeding at 1-2 years (CBF) is much more reliable. 2/N
New article just out in BMC Public Health analyzing how breastfeeding indicators are calculated using @DHSprogram (DHS) data, revealing issues in measurement and reporting of indicators compiled by WHO & UNICEF: a🧵 rdcu.be/dSA5n #BreastfeedingResearch #EpiTwitter #EconTwitter #EpiSky #EconSky 1/N
October 10, 2024 at 2:12 PM
New article just out in BMC Public Health analyzing how breastfeeding indicators are calculated using @DHSprogram (DHS) data, revealing issues in measurement and reporting of indicators compiled by WHO & UNICEF: a🧵 rdcu.be/dSA5n #BreastfeedingResearch #EpiTwitter #EconTwitter #EpiSky #EconSky 1/N
Bluesky now has over 10 million users, and I was #1,792,353!
Fun fact of the day 🙃
Fun fact of the day 🙃
September 18, 2024 at 7:42 AM
Bluesky now has over 10 million users, and I was #1,792,353!
Fun fact of the day 🙃
Fun fact of the day 🙃
First, we checked if other boroughs, not targeted by the campaign, experienced similar movements away from their long-run trend during the campaign. We found that 2 experienced similar drops in magnitude, but statistically not significant, & 3 others saw significant hikes 15/N
September 10, 2024 at 5:16 PM
First, we checked if other boroughs, not targeted by the campaign, experienced similar movements away from their long-run trend during the campaign. We found that 2 experienced similar drops in magnitude, but statistically not significant, & 3 others saw significant hikes 15/N
Our preferred estimates indicate that 7 suicides were possibly averted thanks to the campaign, with a cost-effectiveness of about £5,000 per averted death. This makes the intervention in Barnet as cost-effective as others more directly targeted to at-risk populations like: 12/N
September 10, 2024 at 5:15 PM
Our preferred estimates indicate that 7 suicides were possibly averted thanks to the campaign, with a cost-effectiveness of about £5,000 per averted death. This makes the intervention in Barnet as cost-effective as others more directly targeted to at-risk populations like: 12/N
These four exercises produce a range of estimates of suicides averted and cost-effectiveness of the campaign which is summarised in this table: 11/N
September 10, 2024 at 5:14 PM
These four exercises produce a range of estimates of suicides averted and cost-effectiveness of the campaign which is summarised in this table: 11/N
Using this measure of exposure, we estimated the treatment effect of the campaign, using the entire data first, and then removing the period from January to September 2021 to use a more conservative baseline. 10/N
September 10, 2024 at 5:14 PM
Using this measure of exposure, we estimated the treatment effect of the campaign, using the entire data first, and then removing the period from January to September 2021 to use a more conservative baseline. 10/N
We then made use of data on the number of clicks per months the different services advertised by the campaign received, to determine a plausible way to measure the exposure to the campaign. We took it to be the average number of months of exposure over the past 6 months. 9/N
September 10, 2024 at 5:13 PM
We then made use of data on the number of clicks per months the different services advertised by the campaign received, to determine a plausible way to measure the exposure to the campaign. We took it to be the average number of months of exposure over the past 6 months. 9/N
We used those to perform first some “naïve” on/off and pre/on/post campaign comparisons, controlling for sex and a third-degree polynomial in month to even out movements due to seasonality: 8/N
September 10, 2024 at 5:13 PM
We used those to perform first some “naïve” on/off and pre/on/post campaign comparisons, controlling for sex and a third-degree polynomial in month to even out movements due to seasonality: 8/N
We obtained info on 1,408 suspected suicides for March 2021 to November 2023. Merging this with data on population by age, sex & borough of residence in the 2021 census, we illustrate the pace at which suicides occur for different geographic areas, aka Kaplan-Meier curves: 3/N
September 10, 2024 at 5:11 PM
We obtained info on 1,408 suspected suicides for March 2021 to November 2023. Merging this with data on population by age, sex & borough of residence in the 2021 census, we illustrate the pace at which suicides occur for different geographic areas, aka Kaplan-Meier curves: 3/N
A little self-love on a Monday morning. Conditional acceptance and praise from student evaluations... I think I needed that! Hope your week starts well too :-)
November 20, 2023 at 9:32 AM
A little self-love on a Monday morning. Conditional acceptance and praise from student evaluations... I think I needed that! Hope your week starts well too :-)