PA: Systems medicine the main representative
Systems medicine is an application in the medical field of systems biology [34], which developed at the beginning of this century [35]. Systems medicine is also termed P4 systems medicine (P4SM) because it claims to be Predictive, Preventive, Personalized, and Participatory. Predictive and Preventive are strictly interrelated items, as reported by [36]: “Moreover, P4 medicine will in the future be able to predict the potential future emergence of disease-perturbed networks in patients and then design ‘preventive drugs’ that will block the emergence of these disease-perturbed networks and their cognate diseases.” P4SM is Personalized: it considers each person as a unique individual and not as a statistical average. Therefore, each person is treated in a personalized way based on the big data collected during continuous monitoring. This intervention requires active participation and a positive contribution from the population; in this way, P4SM is able to provide actionable information, which it can use to improve health.
To achieve the P4 goals, P4SM has to integrate data from conventional sources (including paper patient records, clinical and pathological parameters, and molecular and genetic data) and new sources (originating from statistical, mathematical, and computational tools). The procedure requires teams that combine expertise from different disciplines as well as continuous spatial and temporal monitoring of every individual. That will allow the extrapolation, analysis, and relating of a huge amount of heterogeneous, structured, and unstructured data (big data) to determine the links between different phenomena and predict future ones.
To obtain a prediction regarding the status and behavior of a medically monitored person, it is necessary to proceed according to the following steps [2]. First, it is necessary to identify the system variables whose measurement and observation can be used to answer a particular question. Second, the interaction among those variables has to be characterized at the molecular, cellular, and whole-body level. Third, the consequences of such interaction have to be determined using a process whereby a complex system (the human body) may be assessed through reduced representation.
The consensus related to this new approach to medicine and its dissemination are highlighted by a number of international institutions working in the field. Examples include the Institute for Systems Biology in Seattle, United States [37], the European Institute for Systems Biology and Medicine [38], and the Suzhou Institute of Systems Medicine, China [39].
P4SM is the only international organized movement for the PA model. However, at the international level, PB is represented by numerous approaches, movements, and campaigns, often connected with one another. The PB approaches include the following: choosing wisely, watchful waiting, the Too Much Medicine campaign, slow medicine, complaints against overdiagnosis, and quaternary prevention. I hope that my list is fully comprehensive, and I present below a short summary of each one.
PB: Approaches and movements
Choosing wisely
Choosing wisely is probably the most widespread and structured movement featuring PB. Its mission is to promote dialogue between clinicians and patients by helping patients choose care that is as follows: supported by evidence; not duplicative of other tests or procedures already received; free from harm; truly necessary [40]. The core idea is to reduce overutilization of inappropriate and essentially harmful tests, treatment, and procedures. This movement is known for having launched in 2012 a campaign that invites all medical specialty societies to develop a list of five tests and procedures that physicians and patients should question [41].
Watchful waiting
Watchful waiting is not an organized movement: it is an approach to health problems. It is an alternative to more aggressive treatment, whereby time is permitted to pass before applying a medical intervention or therapy, and it makes use of patient involvement [42]. Prostate cancer has received considerable attention in this regard [43]. This approach is also promoted by the National Cancer Institute, which in its Dictionary of Cancer Terms describe the concept of watchful waiting as follows: “Closely watching a patient’s condition but not giving treatment unless symptoms appear or change. Watchful waiting is sometimes used in conditions that progress slowly. It is also used when the risks of treatment are greater than the possible benefits” [44].
Too much medicine campaign
At the end of the past century, a long debate began with The BMJ on the theme of “too much medicine?” [45]. The campaign later became formalized and has a dedicated Web site [46], which specifies the following: “The BMJ’s Too Much Medicine initiative aims to highlight the threat to human health posed by overdiagnosis and the waste of resources on unnecessary care. We are part of a movement of doctors, researchers, patients, and policymakers who want to describe, raise awareness of, and find solutions to the problem of too much medicine.”
Slow medicine
The organized movement of slow medicine began in Italy in May 2011. The background to the movement is respect for nature and the environment, a sense of justice, and an aversion to waste and consumerism [47]. These ideas are shared with those of another movement, Slow Food, with whom Slow medicine undertakes contact and collaboration. In the declaration of the slow medicine movement, seven “poisons” of fast medicine are described. Collaboration of slow medicine with choosing wisely came about with the creation of Choosing Wisely Italy [48].
Complaints against overdiagnosis
Overdiagnosis and overtreatment are the most insidious consequences of overuse of medical interventions [49]. Therefore, all the studies and authors that denounce it completely belong to PB. Here, the topics are not new; however, they have gained international resonance in recent years as mathematical models have been used to calculate the magnitude of overdiagnosis and overtreatment. Pathirana et al. [50] identified five drivers of overdiagnosis: culture; health systems; industry; professionals; and patients/public.
Quaternary prevention
This approach to medicine has been formalized in the concept of “quaternary prevention,” and it has particularly developed in the field of general practice [51]. Indeed the concept of quaternary prevention appears in the WONCA International Dictionary for General/Family Practice, where it is defined as “Action taken to identify patient at risk of overmedicalization, to protect him from new medical invasion, and to suggest to him interventions, which are ethically acceptable” [52].
I now present a comparison between PA and PB. First, I analyze their sensitivity and specificity; then, I assess their reduction to a harm-benefit ratio.
Sensitivity and specificity of the two paradigms
As systems medicine, PA pursues the goal of monitoring (in a continuous spatial and temporal modality) all individuals. Therefore, owing to this putative ability to predict and prevent each future disease, PA has a very high sensitivity: theoretically, almost 100% for any clinical condition. However, the specificity and positive predictive value are supposedly low, depending on such factors as the disease, its prevalence, and the threshold used. By contrast, PB acts with restraint and when there is certainty of a favorable balance in terms of harms and benefits. Thus, for each disease, PB has a high specificity and high positive predictive value: only a small part of the population will be diagnosed and treated, i.e., individuals who are more probably ill or for whom interventions are more appropriate. Obviously, this approach has low sensitivity, which entails the risk of losing sick people. A well-known ancient ethical dilemma, applied to single clinical problems (as I saw previously with breast cancer screening), here applies to the whole population: to include all sick people at the cost of treating many both overdiagnosed and healthy individuals (overdiagnosis, overtreatment), or not to include all sick people (underdiagnosis, undertreatment) with the advantage of not treating healthy individuals.
Reduction of comparison between PA and PB in harm-benefit assessment
Translating these considerations to the field of harm-benefit assessment, PA and PB have different approaches. In brief, PA entails high sensitivity, more benefits, and more harms; PB entails high specificity, fewer harms, and fewer benefits. Let us imagine ideal conditions under which both PA and PB offer the better performance. PA attempts to obtain maximum benefits for the population at the cost of increasing the damage somewhat (more benefits and a few more harms). PB tries to achieve minimum harms for the population at the cost of decreasing the benefits somewhat (fewer harms and slightly reduced benefits). This is the kind of incommensurability that I intend to address in comparing PA and PB: it is the impossibility of making a correct decision starting from these ideal conditions (undecidability). In my opinion, these are the core issues that arise from such a comparison. Therefore, the questions “Which is better?” and “Who is right?” are much too simplistic, or even nonsensical.