3 ms·

Cancer Progress: Much More Than You Wanted to Know

4 points by lainon 8y ago

dredmorbius 8y ago
As I'd commented recently on HN: a good friend died of cancer in the late 1980s. We were close enough that I ran interference on those trying to provide "helpful" advice, particularly of novel, or scam, cures. I came to know that therapeutic protocol pretty well, and have kept tabs on progress since. The cancer itself was, and is, rare and aggressive. Five-year survival rates remain about 20%. Cause is unknown -- no identified genetic or environmental factors -- though the relevant mutation was identified in the 1980s. Treatment is largely unchanged from the late 1950s / early 1970s. Heavy-dose, profound side-effect chemotherapy. Surgery. Radiation. Very poor quality of life, in exchange for a few extra months. Maybe. (And mind, in the midst of all of this, you simply do not know -- cause, best treatment, outcome ... and there is a powerful desire to do all that is possible. The uncertainty and helplessness, the lack of agency, are crippling.) Three decades on, and Francis Bacon is proved wrong: knowledge, without the capacity to act on it, is not in fact power. We've gotten better at imaging, gene sequencing, path lab. But we cannot differentially kill this particular malignancy faster than its host any more now than in 1988. Or 1970. It's a real bastard. I stumbled across a graph a month or three back showing auto deaths per passenger mile, in the US,from 1915 to present. It's common knowledge that there've been vast strides in auto safeety, especially in the 60s, 70s, and 80s. Common knowledge is wrong. The period of fastest improvement in auto safety was the decade 1915-25, where the mortality per passenger mile halved. Since then, progress has been remarkable consistent, with further halvings, though every two decades. Progress is incremental. Goddamned frustratingly so, with declining returns, but that's the inconvenient truth. In medicine, there's been a vast push to heroic measures, to 24/7/365 monitoring, to electronic patient recordss. There's much financial incentive there. (Internet pioneer Paul Baran warned of this and other dangers. In the 1960s. I've been on a bit of a PB kick of late. https://www.rand.org/pubs/papers/P3523.html https://www.rand.org/pubs/papers/P3523.html) But little health benefit. What's mattered most has been prevention, then access, then consistency. Remove contaminants. Get the lead out of Flint's water. (And elsewhere.) Immunise. Ensure mothers and children adequate nutrition. Remove lead, asbestos, particulates, tobacco. Reduce alcohol consumption. Diagnose, treat, or accomodate mental health. Get guns off the streets. Pay living wages. Stop evictions. Break the poverty cycle. Build social resilience. Reward providers based on community outcomes, not measures taken: value provided rather than cost. Recognise that these are the truly challenging problems, and tackle them.

4 points by lainon 8y ago

As I'd commented recently on HN: a good friend died of cancer in the late 1980s. We were close enough that I ran interference on those trying to provide "helpful" advice, particularly of novel, or scam, cures. I came to know that therapeutic protocol pretty well, and have kept tabs on progress since. The cancer itself was, and is, rare and aggressive. Five-year survival rates remain about 20%. Cause is unknown -- no identified genetic or environmental factors -- though the relevant mutation was identified in the 1980s. Treatment is largely unchanged from the late 1950s / early 1970s. Heavy-dose, profound side-effect chemotherapy. Surgery. Radiation. Very poor quality of life, in exchange for a few extra months. Maybe. (And mind, in the midst of all of this, you simply do not know -- cause, best treatment, outcome ... and there is a powerful desire to do all that is possible. The uncertainty and helplessness, the lack of agency, are crippling.) Three decades on, and Francis Bacon is proved wrong: knowledge, without the capacity to act on it, is not in fact power. We've gotten better at imaging, gene sequencing, path lab. But we cannot differentially kill this particular malignancy faster than its host any more now than in 1988. Or 1970. It's a real bastard. I stumbled across a graph a month or three back showing auto deaths per passenger mile, in the US,from 1915 to present. It's common knowledge that there've been vast strides in auto safeety, especially in the 60s, 70s, and 80s. Common knowledge is wrong. The period of fastest improvement in auto safety was the decade 1915-25, where the mortality per passenger mile halved. Since then, progress has been remarkable consistent, with further halvings, though every two decades. Progress is incremental. Goddamned frustratingly so, with declining returns, but that's the inconvenient truth. In medicine, there's been a vast push to heroic measures, to 24/7/365 monitoring, to electronic patient recordss. There's much financial incentive there. (Internet pioneer Paul Baran warned of this and other dangers. In the 1960s. I've been on a bit of a PB kick of late. https://www.rand.org/pubs/papers/P3523.html https://www.rand.org/pubs/papers/P3523.html) But little health benefit. What's mattered most has been prevention, then access, then consistency. Remove contaminants. Get the lead out of Flint's water. (And elsewhere.) Immunise. Ensure mothers and children adequate nutrition. Remove lead, asbestos, particulates, tobacco. Reduce alcohol consumption. Diagnose, treat, or accomodate mental health. Get guns off the streets. Pay living wages. Stop evictions. Break the poverty cycle. Build social resilience. Reward providers based on community outcomes, not measures taken: value provided rather than cost. Recognise that these are the truly challenging problems, and tackle them.