cultivating & crashing

an organic collection of notes, observations, and thoughts

Zuppa di cavolo riccio con funghi e fagioli

1/2 kg di cavolo riccio (o anche cavolo nero)
1 litro di brodo
1 fungo portobello grande (anche gli champignons vanno comunque bene)
250g di fagioli (o 1 scatoletta di fagioli precotti)
6 cucchiai di cipolla
2 spicchi d’ aglio
sale, pepe

Vari tipi di fagioli possono essere utilizzati, fagioli borlotti, neri, lenticchie. Se si usano fagioli secchi, farli ammollare per la notte precedente e poi cuocerli in acqua bollente finche’ non sono morbidi prima di utilizzarli in questa ricetta.

Soffriggere in una pentola grande meta’ della cipolla tagliata fine con l’ olio finche’ trasparente, poi aggiungere uno spicchio d’ aglio tagliato a pezzetti per altri 2 minuti, aggiungere i funghi tagliati a pezzi, sale e pepe. Quando i funghi sono morbidi, toglierli e metterli da parte, e nella stessa pentola soffriggere l’ altra meta’ della cipolla nell’ olio finche’ traslucida, poi aggiungere uno spicchio d’ aglio tagliato a pezzetti e cuocere per altri 2 minuti. A questo punto aggiungere il cavolo tagliato a pezzi e salato, cuocere per 5 minuti e poi aggiungere il brodo caldo e cuocere, coperto, per ulteriori 30 minuti. A questo punto aggiungere i funghi e i faglioli e cuocere ancora per 10 minuti.

Ricetta da Umbria

Fascinating stuff happening at the Douglas

Bayes and boots

It occurred to me today that a person’s shoe size is best understood in a Bayesian framework. My shoe size depends on the brand, the metric that’s being used (European, US, UK, Brazilian), and the kind of shoe. The value of my shoe size is not fixed, and we can assign probabilities to how likely it might be for it to be 37 or 38 for a sandal or boot, for instance. I think.

That said, I got new winter boots today and I’m in love with them. Part of me finds this really materialistic and obnoxious, but the other part of me wonders how I managed to make it through without boots for the range between 10 and -10C (as opposed to -10 to -40C).


I know it seems silly, but recently I’ve been really impressed by the mechanism of division in math. Maybe it’s because in my head I’ve always conceptualized it as breaking up a larger group into parts, and it’s only lately that I’ve started to think of it as standardization. When dividing a sum for a mean, you’re standardizing the numerator by the denominator. The denominator becomes 1, and for all the groups. Another example is when calculating out a conditional probability:


In this case, dividing by the probability of event E is standardizing — or conditioning on — the numerator by that probability. The probability of E and F is what it may be, but once divided by P(E) it is standardized to what P(EF) would be if P(E) were equal to 1; that is, it’s the P(EF) if the probability of E were the only universe that exists. So simple and so damn brilliant. If I were to go back in time and decide I wanted to live in the world of philosophy, I would have studied math.

Radio La Colifata

Just found out that the mental hospital where my aunt was institutionalized when I was little, El Borda, broadcasts a radio from inside the institution that aims to destigmatize mental illness. Listening to an episode where patients candidly discuss the plans, since revoked, to close the hospital.

Radio La Colifata

Masters of Love

Notes from this article in the Atlantic, because every relationship (romantic or otherwise) needs these things. Corroborates other research on supportive rather than authoritative parenting.

- turn toward partner at every bid
– scan social environment for things to appreciate + say thank you for
– scan partner for what s/he is doing right, respecting + showing appreciation
– think of kindness as a muscle — good relationships require sustained hard work
– inform expression of anger by kindness
– be generous about partner’s intentions (give benefit of doubt)
– appreciate the intent, even if poorly executed
– practice active constructive responding to positive news

Health in populations vs. health in individuals

Yesterday in my population health class we discussed how health is different (and dealt with differently) in populations than it is in individuals. These are two distinctions I found particularly interesting.

Diet and genes on total cholesterolThe above table shows how genetic makeup may be more important in determining a health outcome for an individual (an individual’s options are seen by reading the table horizontally), but that societal trends are more important in determining the same outcome in a population (a population’s outcomes are seen by reading the table vertically).

Another interesting distinction is the best way to deal with a given outcome. An individual works to prevent developing disease outcomes that s/he is at high risk for, whereas a population focuses on reducing the risk for people at medium risk, as this is where the bulk of the incident cases of the disease will occur.

Semenza JC. Strategies to intervene on social determinants of infectious diseases. Euro Surveill. 2010;15(27):pii=19611.

A strategy that targets individuals at high risk of developing a disease will see far less benefit overall than if a total population-based approach is implemented (compare change in means in graph D versus graph B). It’s the prevention paradox at work: more improvement overall is achieved by creating smaller benefits in more people than in huge benefits in the people who need it the most/who would experience the greatest relative benefit.

Imperfect data

The art of epidemiologic reasoning is to draw sensible conclusions from imperfect data.”

George W. Comstock

American Journal of Epidemiology 1990

Notes on Bayesian adaptive design, aka the future

Notes from a biostats seminar today by Kristine Broglio, M.S., statistical scientists at Berry Consultants. I was expecting another biostatistics talk where I would follow 10% and glaze over aggressive looking letter salads of the Greek alphabet the rest of the time. Instead, I heard about THE FUTURE. This is how controlled randomized trials will look like in the future, I am sure. It’s too spectacular not to.

fixed design is based on the math we were good at doing 200 years ago

– adaptive randomization (randomize to different groups)
– statistical modeling

can chase random highs by randomizing more patients to it, which will smooth random deviations, so can treat more patients in the best arm

Roger Perlmutter (Merck executive): we do 21st century biology in our laboratories and then do clinical trials that Hippocrates would have been comfortable with

would you rather be the last person enrolled in a trial or the first person to receive treatment?
Bayesian adaptive design shortens the gap between the two options.
would you rather be the first person enrolled in a trial or the last person in enrolled in the same trial?
clearly the latter.

are able to determine which arm of treatment is most likely to come out as the winner during interim analyses.
can stop more trials earlier because trial is more efficient

bayesian is a natural way to think about adaption. bayesian comparative design is how drs think anyway, but not limited to it. but frequentist can also use this design.

clinicians loved it, patients loved it, commercial statisticians loved it. only academic statisticians were not sure about it.

uses almost no priors, or really non-informative ones

Le monde est grand, je suis toute petite

I like epi because it makes you realize how small we are. What happens to us individually is on a tiny scale; events that we take so personally are best understood as waves of larger trends, and we’re never alone in them. When something is common we’re united in the commonality of it, but even when an occurrence is rare (an obscure cancer, a little-known disease), the persons experiencing it are experiencing it in the same marginal way (being the only one around who has it is how everyone who has it feels). In either case, we share that experience of it.

Makes me think also of how people in cities feel: everyone feels lonely and alienated from others, all crowded in the same densely-populated area.


Get every new post delivered to your Inbox.

Join 75 other followers