Epidemiology is the study of patterns of disease in populations -- who's at risk, what's the cause, how to prevent it. For more than 100 years after John Snow epidemiologists concentrated on proximate single causes or specific exposures -- cigarettes and lung cancer, Legionnaire's disease and Legionella bacteria, HIV and AIDS -- but recently the field of "social epidemiology" has gained some traction. This is the study of the social determinants of health and disease. The proximate cause of AIDS is HIV, but often, HIV is contracted through drug use and shared needles, or patterns of multiple sex partners. And, in most places, those at highest risk are poor, so that it's perfectly legitimate to say that poverty causes AIDS. So, what's the actual 'cause' of the disease? And, where does Public Health intervene to control it?
There has long been debate over whether epidemiology has a theoretical framework, or whether it is just a set of established statistical methods. Social epidemiologist Nancy Krieger, in her seminal paper in 1994 called "Epidemiology and the web of causation: has anyone seen the spider?" discussed just this, writing that epidemiology at the time, yes, was interested in the 'web of disease causation' -- what causes disease in populations? -- but was neglecting the search for the spider, the maker of the web. This paper kick-started the field of social epidemiology.
Krieger wrote:
[This paper] emphasizes why epidemiologists must look first and foremost to the link between social divisions and disease to understand etiology and to improve the public’s health, and in doing so exposes the incomplete and biased slant of epidemiologic theories reliant upon a biomedical and individualistic world-view.She's just published a new paper in the American Journal of Public Health, "History, Biology and Health Inequities: Emergent Embodied Phenotypes and the Illustrative Case of the Breast Cancer Estrogen Receptor" in which she argues that health inequities can only be reduced if diseases are considered not just as static biological entities, but within their social and evolutionary contexts, and as individual histories (she calls this broad view of, in this case the estrogen receptor, the "emergent embodied phenotype" -- we can ignore the jargonizing of what is a rather obvious idea). The paper was brought to our attention by Susan Oyama, who has done some very good work on developmental systems, broadening the understanding of genes in context, among other things.
There's a new field in biology, too; 'systems biology'. This is the study of interconnected biological systems -- metabolic pathways, gene interactions, cell signaling networks. This is intended to be a more holistic approach to biology rather than a reductionist one. The idea is that describing these interactions will help us to understand the 'emergent' traits that are complex diseases. It is hoped that there will be practical applications -- understanding networks will, e.g., elucidate 'druggable' pathways that pharmaceutical companies can then intervene on, to prevent or cure disease.
Ken and I have often criticized the idea that the new field of 'systems biology' is going to be all it promises. Like social epidemiology, systems biology is a somewhat self-congratulatory jargonized term whose proponents suggest the very reasonable idea of taking a broader view of causation, in an attempt to take into account as many factors as possible that might explain how genes function or how traits are made or the causes of a disease. It's basically a way of enumerating identifiable interactions among components.
So, rather than being satisfied with knowing that mutations in a specific gene on their own cause a particular disease, the idea is to take a larger view: Enumerate the regulatory pathways or gene interactions (that is, between proteins) so that there are more targets for pharmaceutical intervention.
Seoul from space |
Metabolism is an example in which chains or cycles or hierarchies of interactions pass molecules from one stage to the next. Even if the interaction networks are the mechanism of action, there is no reason to expect that the effects of variation in the components will lead to simple or additive variation in the result.
However, the temptation (whether explicit or implicit) is to try to rescue complexity by treating networks as self-contained causal units. Take sets of tens or hundreds of genes and treat each as a single causal unit, and magically you've reduced the causal dimensions you need to consider by an order of magnitude! But this is wishful thinking, and doesn't really simplify causation or the hunt to understanding, because we know that genes contribute to multiple networks, that vary, overlap, and have alternative pathways that are used under different circumstances.
Social epidemiology faces similar complexity.
Broad causal networks may seem to smooth out causal relationships until they look generalizable, but they aren't necessarily any smoother in social than molecular life. There's a common issue in population sciences called the "ecological fallacy." This is when an association that is true on the population level is inferred to be true on the individual level. The county may always vote Republican, or this neighborhood may be a wealthy one, but it can't be assumed that everyone in the county votes Republican, or that everyone in the neighborhood has an income above the average. Or that the neighborhood causes political preference or wealth in a given individual.
Social epidemiology by its nature and objectives searches for population level associations and attempts to infer from those associations causal relationships that apply at the level of the individual. Poverty causes AIDS. Racism causes stress which causes high blood pressure. In a sense, as reasonable and plausible as these ideas, and as much as they must reflect causal processes in some way, the ecological fallacy is always lurking, because not everyone with AIDS is poor, and not everyone who is poor gets AIDS, not everyone who has experienced racism has high blood pressure, and so on. And, even eliminating poverty won't prevent further cases of AIDS.
Unlike systems biology, where the idea is to intervene in the networks with targeted pharmacological agents, it's hard to know what to do with the information that social epidemiology is producing. Public Health is, ultimately, an applied science. It has very little that could be called real theory, beyond the use of statistical methods to design and evaluate studies -- that is, assuming the kinds of repeatabliity that any statistical sampling requires, similar to saying that every roll of dice has the same probability of coming up 6. Dice rolls may be repeatable events, but to a great extent, humans aren't. And as with systems biology, the idea is not just to enumerate interactions but to identify targets of intervention. But, when the field is identifying "poverty" or "racism" as causal factors, what can be done to intervene?
Indeed, one might say that at least simple biological systems like the krebs energy cycle or photosynthesis require networks of interactions among a step-wise hierarchy of truly enumerable molecular components in quantitative relationships, and a very high degree of repeatability so they can be studied experimentally and evaluated with standard statistical methods. But are 'poverty' and 'racism' even serious concepts of similar type? If not, then current statistical or other study-design methods may simply be inappropriate, too vague, or unable to provide the type of answers we would like to get, especially if we desire to understand causation at the individual level.
Public Health, and epidemiology in particular, has a legacy of successful research and intervention in infectious diseases -- these are 'point source' diseases, with largely replicable causality, and if you cut the problem off at the source, you prevent the disease. This has its parallels with human genetics, where clearly Mendelian, single-gene disorders are relatively easy to explain. It's when causation gets complex that genetics -- and Public Health -- get into difficulties.
But, ok, granted, Public Health is a population-level field, and eliminating all cases of a disease has never been asked of it. But it's curious to see the field seeming to homogenize causation at a time when complexity is a buzzword in other fields. Though, again, this is quite in line with much of genetics, which has its own legacy of successes with point causation (Mendelian disease), just not so much with complex diseases. And which also mixes population-level effects of particular variants with the ability of those variants to 'personalize' medical care.
Have you seen the photos that Col. Chris Hadfield is posting on his Facebook page? He's in the Space Station, taking pictures of Earth in his free time and sharing them with the world. They are stunning. He photographs cities, regions, large chunks of continents; they are lit up at night or green, or snow-covered, or cloud-covered, or dry-as-a-bone desert, or stretches of ocean during the day. When he wonders what something is in one of his pictures, he crowdsources the answer. The other day it was a spot in Tehran that looked curious to him. It turned out it had once been an airport, and is now a playground, which Iranians told him.
If a region is experiencing drought, that's easy to see from these photos. But, if there are cracks in the macadam on those long stretches of highway, or a bridge is unstable, or the rivers are polluted, you can't see it from space. At least, not from these pictures. You do get the big picture, but you have to zoom in to discover where to intervene to prevent a catastrophe due to crumbling infrastructure.
Social epidemiology may be a lot like Col. Hadfield's photographs -- beautiful descriptions, but so far removed from the individual that it's impossible to actually predict who's going to get sick and why, never mind useful for figuring out where to intervene. Even identifying the risk factors themselves is notoriously difficult. And one need not give it a special name, which is often a way we academics have of making their ideas seem new or particularly insightful. Just 'epidemiology' will do very well--and the point is to include social as well as other potentially causal factors (and, as we hope is clear, we are not just picking on public health rhetoric, because genetics is just as much affected by unnecessary jargonizing).
Public Health is by goals and design a population-based field, and it works well when the message is that cholera is waterborne, so we need to keep our water sources clean, or that vaccination can protect entire populations against infectious disease. Its methods have saved countless lives. But, like trying to make risk predictions from genetic data from whole populations, it's hard to know what to do with these descriptions of the causes of complex diseases from such a distance.
And ultimately, of course, if we go out enough concentric circles of causation, it's life that is the cause of disease, and the only prevention is death.
I have long enjoyed reading the entries in this blog, admired the often bold and refreshing ideas that depart from current “reductionist” thinking with which I have always been so much in agreement that I saw no need to post comments. But in this interesting analysis of “social epidemiology” and the challenge of complex causation, the field ‘systems biology’ is mentioned, a bit out of the blue, only to be used in the authors argumentation in a way that is neither academically appropriate nor pedagogically useful. It seems that the term ‘systems biology’ and the attention it attracts is exploited to erect a straw man, by distorting its meaning such that it conveniently serves as a screen to project all the insufficiency of modern biology that the authors wish to rant.
ReplyDeleteHere at the Institute for Systems Biology (ISB) – the first research institute that carries this label and which has help define the field –many of my colleague and I beg to differ: Systems biology is precisely not about “enumerate[ing] the regulatory pathways or gene interactions … so that there are more targets for pharmaceutical intervention”. This is the old-fashioned brute-force approach that is the domain of genomics and other –omics, bioinformatics, GWAS, etc. that systems biology proudly leaves behind and that the authors leave out in their piece. Of course attacking these outdated approaches already vilified ad nauseam elsewhere is less glamorous than criticizing adolescent systems biology
Only those who have borrowed the label of ‘systems biology’ to give their work more appeal, but who remain reductionist in their heart and mind, will work on the “purely additive model of what genes do”. So let me correct: Systems biology is precisely about what makes the whole more than the sum of its part. Systems biology goes beyond entirety of analysis: it embraces analysis of entirety. It unites holistic with molecular approaches. Yes, it espouses systems thinking but does not eschew systematic approach. It embraces big biology but knows that population averages discard valuable information about individuality. It seeks proximate causation but appreciates ultimate causation. The authors correctly criticize that there is a limit to identifying “targets of intervention”. But what if there ARE still unidentified simple targets of intervention or “point source diseases” with “replicable causality” –modern versions of cholera scenario - don’t we owe those afflicted by such disease our effort to find a cure – since it is targetable? Hence systems biology, even if we feel the pressure from the rise of post-genomic holism must continue to emphasize the ‘systematic’ approach when adopting ‘systems thinking’ and continue to honor the stamp collecting mentality of old fashioned molecular pharmacology given that we have tools to do so efficiently.
The full spirit behind systems biology, with all its nuances, cannot be described here. It certainly cannot be packaged and redefined in the simple “jargonized terms” as is conveniently done in this blog. To see how the spirit and culture of thought behind systems biology is actually aligned with the authors admirable philosophy of biology one needs only take a closer look at the vision of “P4 medicine” – a new personal health and wellness initiative spearheaded by ISB: P4 Medicine stands for Predictive, Preventive, Personalized, and Participatory Medicine. It is clear that the authors’ vision individuality in epidemiology is covered by these 4 P’s
Granted, systems biology, perhaps the most misunderstood concept in modern biology, lacks a consensus definition such that its malleable meaning can be adjusted to the intention of the user. Then why project an imagined discipline with all its insufficiency onto it instead of helping to shape the concept of ‘systems biology’ precisely such that it embraces the very reasonable vision of a form of biology that best serves mankind that the authors of this blog have long presented?
Your response highlights the issues we were noting, and is quite useful. Our point was not to deny much less denigrate the idea that genomic effects work via interactions. We've written on that extensively, indeed, as an antidote to the reductionist and highly deterministic kinds of competition-obsessed way that so much of the life sciences rest on (implicitly or explicitly).
DeleteOur book, Mermaid's Tale, after which we named this blog, is all about that. We didn't use the term 'systems biology' nor did we sling around others like 'GRN' for gene regulatory networks, but our whole point was about what we refer to as cooperative interactions as the basis of life.
Our point here, and I think you've also agreed, is that terms like 'systems biology' are thrown around very often as if they explained something by implying that systems are causal units that could replace genes, thereby making complexity simple.
I also personally think our enterprise, and indeed our culture more broadly, rests too much on labeling and relabeling. We don't need in my personal view, to coin separate names for things, which too often are marketing and promotional tools. And I think that has become too prominent in how we advance, and fund, science. One could hardly argue with the 4 P's, though are they anything new to medicine, that can't be found even in Hippocrates?
I mean, Sewall Wright before any of us were even gleams in our parents' eyes was clearly aware of, and studying, interactions, developed methods (like path analysis) for evaluating them, etc.
So we might argue about how useful, or in what ways, such integrative thinking will be, or what the best strategies are. But that wasn't our point.
If our wording triggered a vigorous response, well, that's good and could lead to a discussion of both the scientific and the strategic issues.
At least, that's how I think of it.
If we've misunderstood systems biology, at least we are in broad company, it being the most misunderstood concept in modern biology! Thanks for your comments, and your explanation. As we apparently share some aspects of our worldview, I am sorry to have crossed you.
DeleteI hope that the fundamental issue we were trying to wrestle with in this post is not based on a misrepresentation of systems biology but is in fact a real issue. And it's something that epidemiology and genetics share, as I still think SB does, and that is how to translate or infer the broader pictures that are of increasing interest to these fields to the individual. This is a problem for clinicians hoping to translate research in these fields into practice, and for Big Pharma hoping to translate the understanding of gene networks and so forth into medicine. And, as we hoped to point out in the post, for the field of Public Health, which by nature is population-based, how to intervene when causation gets more and more generic is a real question.
Looking at these issues by zooming out gives you a picture, yes, but it can be a picture that's so broad that it's hard to know what to do with it. The conundrum of course is that the reductionist, close-up picture is also limited in its own way.
We don't like to annoy readers that generally resonate with what we have to say! And we could have tamped down our (usual) rant, yes, but I think there are serious questions here, that we might still agree are worth considering.
I would submit that the reason there even appears to be a "fundamental issue" that deserves to be wrestled with is because the metaphysical notion of causality is invoked throughout. I would challenge the authors to imagine writing the posting while making the same points without using the word " cause " once. Otherwise, one risks stepping outside the boundaries of science. ilya shmulevich (also at ISB).
ReplyDeleteWell, you can use whatever word you want, and I agree that it is a bit metaphysical or, at least, 'axiomatic'. But science is about more than just temporarily-ordered co-occurrence (A->B). That's involved each of the 4P's.
DeleteThe problem with the idea of Whatever-You-Call-It (WYCI) is that one needs to be able to observe A and predict B, or to understand B one needs to identify A.
If all we are doing is noting some ordered co-occurrences that has no predictive power unless we make some assumptions about WYCI.
The problem of confounding, which bedevils epidemiology, illustrates some of the issues. It becomes relevant if the purported predictor really isn't a WYCI, because then predictions become unreliable to an unknowable extent.
This has of course all been a challenge going back at least to Aristotle's famous attempts to discuss these subjects, and has woven its way through philosophy and epistemology for all the following centuries.
As to your characterization of what constitutes 'science', that, too, becomes a matter of definition. To some, unless enables us to predict and manipulate the world, it's not 'science'. Here (I would say) without WYCI, for reasons I just said, it's not so useful. To others, it's about understanding the world, and that basically involves WYCI.
Of course, at any given time our knowledge is limited, so we are forced to substitute some aspects of empirical co-occurrence for WYCI, but when or where one decides that the empirical goal is met, is an individual judgment.
So, unless I totally misunderstand your point, I don't see what other way of discussing the issues you suggest.
Further, if one's purpose is to predict, prevent, treat, somehow intervene in a disease cycle, one needs to choose where to intervene. Whether that's science or not, or whether you call it causation or not seems to me to be irrelevant, but I may be missing your point.
DeleteAs Ken says, the problem of causation is a deep one, but important for applied WYCIs. It's easy when an effect is large. We didn't need something like systems biology when infectious diseases and Mendelian diseases were the major concern.
When effects are additive (or whatever) and small, and that's where we are with complex diseases, every field that is trying to understand where and how to intervene -- genetics, epidemiology, public health in general, psychology, sociology, pharmacology, and, yes, systems biology -- seems to want to expand the circle of causation in an attempt to somehow capture the co-occurrence of relevant events. Or, the relevant co-occurrence of events.
When the circle is expansive enough to capture every possible interaction, is that helpful for applied WYCIs?
Or, do we completely misunderstand your point?
I will let Bertrand Russell clarify the matter:
ReplyDelete“In short, every advance in a science takes us farther away from the crude uniformities which are first observed, into greater differentiation of antecedent and consequent, and into a continually wider circle of antecedents recognized as relevant. The principle ‘same cause, same effect,’ which philosophers imagine to be vital to science, is therefore utterly otiose. As soon as the antecedents have been given sufficiently fully to enable the consequent to be calculated with some exactitude, the antecedents have become so complicated that it is very unlikely they will ever recur. Hence, if this were the principle involved, science would remain utterly sterile….The law of causality, I believe, like much that passes muster among philosophers, is a relic of a bygone age, surviving, like the monarchy, only because it is erroneously supposed to do no harm.”
(from On the notion of cause)
Well, Ilya, it's a nice quote from a deep thinker, but that doesn't make it right (or wrong). To me, the issues are serious, and from what reading I've done recently, Russell was essentially saying what had been debated extensively and gradually realized in the 19th century.
DeleteI think this raises all the issues about whether probabilistic--um, 'patterning', if I can't say 'cause'? -- is real or is just a matter of our ignorance and imperfect measurement.
Unless quantum physicists have some fundamentally different answer, and I don't understand enough of that to have an opinion, the notion of probability as an inherent kind of cause is as elusive as that of determinism.
Relevant, I think, to this discussion is whether a probabilistic pattern has some underlying distribution (binomial, Gaussian, etc.). If so, then really I think the distribution just takes the place of deterministic 'cause', and the word probability becomes a synonym for cause, by reflecting a real, underlying, regular pattern generator. Then probabilities become a new kind of deterministic 'laws' of nature, whatever their rather mysterious mechanism.
This would essentially be a frequentist view. On the other hand, if this is not the case, and we view probability as just a description of the unique events we have observed, then those probabilities don't reflect underlying 'laws' and they may be adequate data-fitting methods but without sound epistemological predictive value. Even a Bayesian view that probabilistic description just tends us to prefer one belief over another, that in itself implies a belief that there is some regular, if elusive, underlying truth.
In the absence of some more law-like theory, be it probabilistic or classically deterministic, prediction becomes a matter of extrapolation, and how can we do that without assuming regularity?
If we take a collective-effect view, which I think is the way that probabilistic science developed in the 19th and into the 20th century, then prediction and orderliness rest on some form of law of large numbers or central tendency (can I use the world 'law' here?). That may well be the way of the world, and seems to be what Russell's quote hints at, but it makes individual prediction rather tenuous even if group prediction is somewhat less so.
I personally tend to think the world is in fact law-like, because my imagine is too feeble to understand how something can meander around 'randomly' without each movement involving some sort of deterministic 'push'. But of course I could be completely wrong.
So, in the absence of deterministic causality, promises like 'personalized medicine', and much that we say about evolution, would become more hollow than is already the case, in what most assume (wrongly, perhaps) is just temporary and due to our current, but correctable, ignorance. Because even collective principles of large numbers are retrospective data fitting. We have plenty of examples where major contributing (causal?) factors of disease or evolution, change in a way that can't be predicted.
So, while I think I see the point you and Russell are making, I don't see how that salvages the goals or promises of science, in terms of trying to understand how our cosmos works--much less to apply in practice, especially where the variance in the collective process is great relative to the measured contributing variables and their retrospectively assessed effects.
So I can't tell if we disagree or agree, especially, of course, if I've misunderstood your point.
The absence of causality does nothing to dim the promise of personalized medicine. Let me quote from a recent paper: “In sum, biological modeling, and the translational medicine consequent to it, must handle the parallelism and redundancy required for system efficiency and survivability. This problem has been faced before by engineers and scientists in complex system analysis, albeit, not as complex as biological systems. Only via interwoven regulation can a system be sufficiently fault tolerant to survive in a rapidly changing interactive environment. We confront the modeling and control of systems capable of autonomous reconfiguration. This problem has been faced by engineers and scientists in complex system analysis since the 1930s.” [Bittner, M. L., and E. R. Dougherty, “Newton, Laplace, and the Epistemology of Systems Biology,” Cancer Informatics, 2012:5, 185-190]. For a full discussion of these issues, see Dougherty, E. R., and M. L. Bittner, Epistemology of the Cell: A Systems Perspective on Biological Knowledge, John Wiley, New York, 2011.
ReplyDeleteThanks for your comment. I'm not quite sure I understand your approach, though. So, you've got a biological model. Now what?
DeleteBut, I should note that we are not arguing the absence of causality. Only that it's difficult, when faced with an ever broadening understanding (model, description, whatever you want to call it) of causality (poverty 'causes' AIDS, gene networks 'cause' disease) to know where to intervene. Our point was a practical one, Public Health and medicine in general being applied disciplines, as much as a philosophical one.
The word 'cause' and its equivalents have created mountains of discussion since the classic philosophers. Inevitably, except for the rhetoric or semantics, one needs prediction to be a part of personalized (or public) medicine or health.
DeleteCorrelates are not necessarily good predictors. Good predictors--even if they are in the form of aggregates of factors somehow digested, as in various systems approaches--imply causation, whether it's probabilistic or deterministic (words with problems of their own). This is certainly true of specific risk factors, because if they are not 'causal' in the common-sense connotation, eliminating them won't eliminate the consequent.
So one needs to disentangle rhetoric and semantics from the actual empirical objectives. We ourselves do modeling all the time, and we try to address complex 'causation'.
The issues we have written about have to do with or to what extent current approaches will bear fruit. People have various views, and the 'systems' view is having its day; we'll see what it reaps, and hopefully the harvest will be bountiful.