Monday, July 17, 2017

Inexact Sciences

Thoughts on the place of science in an era of false conviction

Some recent articles, noted by a few of us in journals regularly monitored by HCR bloggers, provide real food for thought in our New World Order of alternate facts, fake news and truthiness.

In a recent number of the still intrepidly pay-wall-free Guardian,  development economist John Rapley summarizes his new book Twilight of the Money Gods. This summary is the best we in the colonies can do until this month's UK publication of the full volume makes it to our shores. (Rapley, a true globalist, both an academic and a public intellectual experienced at teaching in several countries, is also a not infrequent Guardian contributor.)

Comes Rapley now to hammer home the point that research domains--his focus is economics--can ossify to the point that they become more like religions or belief systems, pace Mary Douglas's classic work Purity and Danger, than anything resembling exact sciences. By abandoning humility to project rigor and group-think, they undermine their own credibility. Think about the following statements, then, and mentally substitute "academic medicine" for economics.
"... [academic] departments were increasingly hiring and promoting young economists who wanted to build pure models with little empirical relevance."
One thinks, for example, about all the spurious metrics currently being applied to the assessment of "meaningful use" in electronic health records. Rapley goes on to elaborate and pose a partial solution. He notes that...
"...[w]hen the church retains its distance from power, and a modest expectation about what it can achieve, it can stir our minds to envision new possibilities and even new worlds. Once economists apply this kind of sceptical scientific method to a human realm in which ultimate reality may never be fully discernible, they will probably find themselves retreating from dogmatism in their claims.
By "power" one can read the unholy alliances with Big Money and managerialism oft-decried in HCR. By "retreating from dogmatism" one can read a renewal of alliances instead--well, let's go with Rapley's words: with ", demographic and anthropological work, and [the need] to work more closely with other disciplines."

His solution, then, through such alliances, is the restoration of narrative to a central role for poo-bahs of some of the more overly sterile reaches of EBM, medical informatics, and decision analysis. He quotes the authors of the recent Phishing for Phools, for whom "storytelling is a 'new variable' for economics, since 'the mental frames that underlie people’s decisions' are shaped by the stories they tell themselves."

Harvard cardiologist Lisa Rosenbaum, who's produced some admirable and at times controversial writing for years in NEJM and elsewhere, has a new piece that appeared simultaneously with Rapley's. She offers a way of refracting some of these same ideas through the lens of scientific medicine--or, more particularly, the crisis of trust by a "weary public" in the scientific process. Dr. Poses, in these e-pages of HCR, has frequently addressed this same issue of relativism where, to use Rosenbaum's example, people (this includes people with major political power in today's US of A) will be damagingly dubious about issues such as climate change. Yet at the same time we seem them devouring (so to speak) the claims of nutritional charlatans. Here I suppose she's speaking mainly of the views of those who both fund and popularize medical science, and the way the traditional establishment of science has been dethroned or at least, by many, seriously questioned.

It's a conundrum. The mandarins of science build their ivory tower and labor mightily in the effort of boundary maintenance (says Rapley); and yet they've lost ground (says Rosenbaum). "The fear of venturing into the fray," she observes, "means that the public hears far more from science’s critics than its champions. This imbalance contributes to 'science is broken' narratives ranging from claims about the pervasiveness of medical error to the insistence that benefits of our treatments are always overhyped and their risks underplayed."

Yet the prescriptions Rapley and Rosenbaum both offer seem rather similar: go back to telling the story. And here one can only agree most heartily. Telling the story works at two levels. At the level of domain knowledge, whether we're talking about economic behavior, clinical decision making, or the interpretation of Big Data, we need more interdisciplinary efforts to connect the dots between physiological, psychological, social and economic processes. And, then, at the level of telling the story about science itself--something about which Rosenbaum herself is greatly concerned--individuals in all the domains of science and medicine need to "learn to tell stories that emphasize that what makes science right is the enduring capacity to admit we are wrong. Such is the slow, imperfect march of science."

1 comment:

Anonymous said...

Good piece that works across many industries and endeavors. We have all become so focused on the details, and computers allow us to look at the details, we lose sight of the larger picture.

Steve Lucas