The new biology

Engel supposed that the BMM supposed biology was or was reducible to physics and chemistry (Engel, Reference Engel1977). This reductionist supposition was pretty much orthodoxy at that time, but is now history. Biology (as physiology) has transformed itself in the last few decades, in a revolution largely driven in fact by biomedicine. Biology (as physiology), with biomedicine, is now a closely integrated combination of two kinds of science: one is the physics and chemistry of energy exchanges and transformations (quantified in enthalpy change equations, for example), but there is in addition something new: models of mechanisms for the regulation and control of the physics and chemistry, and of other systems, maintaining functions, typically using feedback by information-transfer. The new ‘regulatory’ paradigm, also known as the ‘information-processing’ revolution, appeared clearly in molecular biology in the 1960s and 1970s, such as Jacob and Monod's ground-breaking work on genetic regulatory mechanisms in the synthesis of proteins (Jacob & Monod, Reference Jacob and Monod1961; Lewis, Reference Lewis2013).

The new paradigm in biology can be dated to the physicist Ernst Schrödinger (Reference Schrödinger1944) ground-breaking definition of life as local areas in which the overall direction of the 2nd law of thermodynamics is reversed – life decreases entropy, temporarily (Morange, Reference Morange2020; Schrödinger, Reference Schrödinger1944). Schrödinger saw that this would have to involve control of energy production, and he further hypothesized that this was done by genes. An intellectual pathway can be traced from Schrödinger's new conceptualization of biological systems to von Bertalanffy's general system theory (Von Bertalanffy, Reference Von Bertalanffy1968) to Engel's Reference Engel1997 paper. Von Bertalanffy proposed that negative entropy was achieved in open systems, in biology and in wider domains including the psychological and the social, and Engel refers to von Bertalanffy's general system theory as a key theoretical driver for the new BPSM, anticipating impacts on healthcare science and practice (Engel, Reference Engel1977, pp. 391–392). The paradigm shift was, however, still in the early stages in the 1970s – the detailed work in theory development and new associated research paradigms has been underway in the decades since and continues.

The regulatory mechanisms that are central in the new biology have several core features that change the theoretical foundations of the life sciences in ways critical to explicating the BPSM. First, they are causal, but they are not, and are not reducible to, the energy-related equations of physics and chemistry. Second, and connected, regulatory mechanisms can break down, allowing foundational distinctions between life and death, health and disease, that are unavailable in physics and chemistry. Third, the same kind of theoretical apparatus used in biology (function, organization, regulation and dysregulation, information, production, and distribution) is also used in the psychological and social sciences – as reviewed below.

The new psychology

As noted in the Introduction, Engel supposed that the BMM assumed body/mind dualism and that this was an obstacle to accounting for psychological factors in health and disease. I outline in this section two general ways in which post-dualist, cross-disciplinary theories have been developed over recent decades, critical to formulating a biopsychological model.

Post-dualist models can be characterized as having two working assumptions: one is that mental processes regulate behavior and the other is that mental processing is a function of brain processing. The first point underpins cognitive psychology, while the second merges psychology with neuroscience. Both points are connected with the theoretical innovations in biology in the 1960s and 1970s outlined in the previous section: biology, psychology, and neuroscience all shared interests in new information-based models of the regulation of biological systems and behavior – the paradigm shift went across the life sciences.

The clearest expression of dualist assumptions in psychology was in behaviorism, which explicitly excluded mental processes from explanations of behavior – a position much like Engel attributed to the BMM. From around the 1960s onwards, however, behaviorism was swept away in the cognitive revolution (Miller, Reference Miller2003; Xiong & Proctor, Reference Xiong and Proctor2018).

The cognitive revolution influenced many specialties in psychology, not only learning theory. Cognitive models in clinical psychology emphasized the causal role of personal beliefs in the regulation of affect and behavior, such as Beck's cognitive model of depression (Beck, Rush, Shaw, & Emery, Reference Beck, Rush, Shaw and Emery1979). Further, there emerged around this period a class of psychological models focused on the role of expectancies and beliefs about personal control over events – or personal agency – and their implications for well-being. Main examples include Julian Rotter's locus of control theory (Rotter, Reference Rotter1966), Martin Seligman and colleagues’ learned helplessness theory (Seligman & Maier, Reference Seligman and Maier1967) and associated model of depression (Miller & Seligman, Reference Miller and Seligman1975), Albert Bandura's self-efficacy theory (Bandura, Reference Bandura1982, Reference Bandura2006), and Richard Lazarus and Susan Folkman's work on stress, appraisal, and coping (Lazarus & Folkman, Reference Lazarus and Folkman1984).

Importantly, this class of models spanned many psychology specialty areas, across many domains – physiology, learning, personality, and social – and interactions between them. In this sense, they already constituted a theorized BPSM within the broad psychological tent.

The appearance of personal processes in the new psychological science – beliefs, about the world and our agency, personal goals, emotions, and behavior – has substantial relevance to the question whether a broader BPSM is needed in health science and healthcare. Engel gave a long list of important issues the BMM could not account for, and top of the list was ‘the person who has the illness’ (Engel, Reference Engel1977, p. 131). Here the point is, at least, that biomedicine can theorize diseased or otherwise dysfunctional organs or systems, but has nothing to say, over and above that, about the person who has the illness. Equally, it can be added, cognitive psychological models of specific systems such as memory and attention, need a wider, person-level framework to theorize how lowered function affects the person, for example, or typically, by compromising agency.

To sum up, the cognitive revolution in psychology endorsed the relevance of mind to science by constructing causal explanatory models of behavior in terms of mental (or cognitive-affective) states. Within that overall framework, diverse psychology specialty areas focused on personal processes – beliefs, about the world and their own agency, personal goals, emotions, and behavior – in interaction with biological and social processes. These developments in psychology have wide implications and they surface again when considering biopsychosocial models, such as of impacts of social disadvantages on health, and of pain and service use, considered in Part 2, and models of clinical care, in Part 3.

The second aspect of post-dualism models mentioned above is that psychological processing is regarded as a function of, or implemented by, brain processing, hence merging psychology with neuroscience. Cognitive (or cognitive-affective) neuroscience (as the merger can be called) has developed alongside cognitive psychology (Albright, Kandel, & Posner, Reference Albright, Kandel and Posner2000).

The new post-dualist constructs of mind and body, further, accommodate crosstalk between neuroscience/psychology and biomedicine, in both directions. This is evident in the new fields of psychoneuroendocrinology (Fink, Pfaff, & Levine, Reference Fink, Pfaff and Levine2012) and psychoneuroimmunology (Moraes, Miranda, Loures, Mainieri, & Mármora, Reference Moraes, Miranda, Loures, Mainieri and Mármora2018), as well as in specific models such as of chronic stress and pain to be considered later. These interdisciplinary research programs, involving neuroscience, psychology, and biomedicine, were inconceivable in mind–body dualism. They are examples of the rationale for expanding the BMM to the BPSM, in effect contributing content to the concept of ‘biopsychology’ or ‘psychological medicine’ within the BPSM.

A relatively new class of theories known as ‘embodied mind’, ‘embodied cognition’, or ‘4E cognition’, explicitly overturns dualism and are, therefore, potentially relevant to a revitalized BPSM. They are less familiar than theories discussed above, however, and for reasons of space I do not consider them here – for details of the theories and controversies, see e.g. Newen, Gallagher, and De Bruin (Reference Newen, Gallagher, De Bruin, Newen, Bruin and Gallagher2018), Carney (Reference Carney2020), and for current applications in clinical psychology and psychiatry (e.g. Allen & Friston, Reference Allen and Friston2018; Gjelsvik, Lovric, & Williams, Reference Gjelsvik, Lovric and Williams2018).

Social determinants of health

Social factors can be accommodated within the conceptual framework of the new biopsychology because the social sciences have always employed comparable concepts, such as organization, rules and regulations, control (power), communication, and production and distribution of resources (e.g. Lasswell, Reference Lasswell1936). In this sense, it is psychology and biology that made the theory changes critical to the BPSM, thereby becoming more aligned with concepts familiar in the social sciences.

The concept of socioeconomic status is closely connected to an individual's or group's access to resources, and the immediate relevance to health is that resources include what promotes good health (Bickel, Moody, Quisenberry, Ramey, & Sheffer, Reference Bickel, Moody, Quisenberry, Ramey and Sheffer2014; McGowan & Shahab, Reference McGowan, Shahab, Llewellyn, McManus, Weinman, Petrie, Newman, Ayers and Revenson2019). Over the past few decades, a substantial range of epidemiological studies have established that there are social determinants of health, that is, a positive correlation between higher social status and better health, the so-called social gradient in health, which underpins health inequalities (Marmot, Reference Marmot2006). This applies to both physical health and mental health (Bell & Marmot, Reference Bell, Marmot, Dinesh Bhugra and Craig2022).

The resources that we need for biological health are well-known. Consistent with Schrödinger's insight into what life is, they include conditions of biological energy production. What we need for good psychological health is less well understood, probably because mind/body dualism never provided a useful definition of psychological life. In the new conceptualization of psychological life reviewed in the previous section, its conditions may be framed in terms of having sufficient agency (or autonomy). The extent of agency, like access to the conditions of physical well-being, depends on socioeconomic status. As Michael Marmot puts it: ‘The lower individuals are in the social hierarchy, the less likely it is that their fundamental human needs for autonomy and to be integrated into society will be met’ (Marmot, Reference Marmot2006, p. 1304). Autonomy is facilitated by social integration especially into dominant power structures, and conversely is downgraded by exclusion, by denial of voice, civil rights and protections, and other means of oppression. These issues have been explored mainly in and across feminist (Biana, Reference Biana2020), postcolonial (Fanon, Reference Fanon1968), critical race theory (Delgado & Stefancic, Reference Delgado and Stefancic2001), and increasingly in Lesbian, Gay, Bisexual, Trans, and Queer (LGBTQ) literatures (Lee & Brotman, Reference Lee and Brotman2013). Possible pathways linking social factors to health outcomes will be reviewed below as examples of models of biopsychosocial causal mechanisms.

Investigating psychological and social impacts on health

Research designs relevant to the BPSM are those that examine the effects of psychological and social, as well as biological factors, on health outcomes (e.g. Lacombe, Armstrong, Wright, & Foster, Reference Lacombe, Armstrong, Wright and Foster2019). Immediate findings are typically of correlations or associations, and control conditions of varying levels of stringency increase confidence in inference to causation. Large-scale group studies are necessary to identify small effects, and the methodology relies on statistical analytic methods such as regression that can estimate the effects of one or more variables on a health condition-dependent variable, estimating independent effects, and interactions moderating the effects of one independent variable by another. Multivariable regression models are applicable within the BMM, including biological variables only, but the expanded BPSM framework also accommodates inclusion of psychological and social variables, estimating their independent, additive, and interaction effects (e.g. Guloksuz et al., Reference Guloksuz, Pries, Delespaul, Kenis, Luykx, Lin and van Os2019).

BPSM compatible research studies were barely available when Engel proposed the new model in 1977. In fact, they began to appear at around the same time. The first clinical trials of psychological therapies appeared in the 1970s, heralding what has become a very large-scale research program of developing and evaluating psychological interventions for a wide range of health conditions and their complications. The early finding that cognitive therapy for depression was effective, and moreover, more effective than an antidepressant medication (Rush, Beck, Kovacs, & Hollon, Reference Rush, Beck, Kovacs and Hollon1977), reinforced the signal that the BMM was not enough, at least not for modeling and treating depression. At the same time, there was another major anomaly for the BMM, the emerging findings of social epidemiology, noted in the previous section, that social status affects a wide range of physical health and mental health outcomes, in the Whitehall Studies by Michael Marmot and colleagues (Marmot, Rose, Shipley, & Hamilton, Reference Marmot, Rose, Shipley and Hamilton1978; Marmot et al., Reference Marmot, Stansfeld, Patel, North, Head, White and Davey Smith1991).

The findings that are anomalous for the BMM but consistent with the BPSM are empirical data, related to specific influences on specific conditions at specific stages. There is nothing a priori in the empirical data. It is possible that a specific health condition at a particular stage may turn out to be primarily caused by only one kind of factor – biological, psychological, or social – and of course biomedical models of infectious diseases have had stunning successes in exactly this way. Such findings may be called a scientific-explanatory reduction to biological processes. This type of ‘reduction’ is different from theory-reduction of, for example, biology to physics and chemistry. Both types of ‘reduction’ are relevant to the relation between the BMM and the BPSM and both are in play in Engel's Reference Engel1977 paper. The BMM would predict scientific-explanatory reduction to primary biological causes only across the whole of health, like the biomedical models of infectious diseases (or of effects of lesions or of genes of major effect). But this is an empirical, not a theoretical matter, and the emerging picture across health is that, especially for the noncommunicable diseases, and including all or practically all mental health conditions, the etiological picture is of at most a ‘patchy’ reductionism, and is more typically diverse, with multiple causal factors (Kendler, Reference Kendler2005, Reference Kendler2012).

Notwithstanding evidence of influence of psychological and social factors on health and disease, there remains a tendency, possibly attributable to long-standing reductionist assumptions in the science, to roll everything up into the biological. There are several examples of this option in the theoretical psychiatry literature. Samuel Guze's highly influential paper over 30 years ago, ‘Biological psychiatry – is there any other kind?’ (Guze, Reference Guze1989) is an example, as indicated by the rhetorical nature of the title question. More recently, Peter White and colleagues proposed that mental disorders are brain disorders, without for a moment being unaware of the research showing the influence of psychosocial factors in the onset and course of many psychiatric conditions (White, Rickards, & Zeman, Reference White, Rickards and Zeman2012). Likewise National Institute of Mental Health (NIMH's) Research Domains Criteria framework, which regards psychiatric conditions as disorders of brain circuitry (Insel et al., Reference Insel, Cuthbert, Garvey, Heinssen, Pine, Quinn and Wang2010).

To the extent that biology, neurology, and neuroscience are being broadly conceived to acknowledge the causal role psychosocial factors in some conditions, in etiology and course, and in prevention and intervention, these theoretical proposals are not reductionist and, albeit unhelpfully expressed, they are compatible with a broad biopsychosocial framework. However, proposals to roll the psychosocial up into the biological appear to be, in name, a kind of reductionism, so far by-passing the need to theorize the acknowledged causal role of psychosocial factors or biopsychosocial causal mechanisms.

Theorized models of causal mechanisms

As well as findings of correlations (or associations) in well-controlled studies, the scientific picture benefits from also having a plausible theory that would explain apparent causal connections. Empiricism in science, relying on observation alone, controlled or otherwise, is well-known to be so far theory-free. In the present case of determining psychological and social causal connections, however, the problem has long been at exactly this point: the absence of a plausible theory of either psychological or social causation, still less theory as to how either could have material effects on biological processes, which were assumed determined by physics and chemistry alone. The assumed impossibility of psychological and social causation and the resulting downwards reductionist pressure inevitably encouraged skepticism toward any apparent empirical demonstration of the impossible. So while empirical evidence for psychosocial causes may accumulate, still, the theory problem is not yet solved and skepticism can persist.

Theory is necessary as well as data, of the sort outlined in the first part of the paper. In brief, psychological causation, implemented in brain processes, involves regulation of behavioral functioning toward attaining or maintaining some state. Social factors can causally interact with psychological processes, for example by regulating task demands and available resources. Psychological and social causal processes are both causal in the sense of regulatory, as is one kind of causation in biology, the other being energy transformations and exchanges covered by physicochemical laws. As to dysfunction, this has to involve disruption to regulation (however caused), because physicochemical laws cannot be disrupted. Models in which regulation/dysregulation are prominent are now to be found not only in biomedicine, but also in clinical psychology and psychiatry (Kendler & Woodward, Reference Kendler and Woodward2021; Liu, Chua, Chong, Subramaniam, & Mahendran, Reference Liu, Chua, Chong, Subramaniam and Mahendran2020). Two well-known illustrations of theorized biopsychosocial causal mechanisms are given below.

The first example is a set of models of chronic stress, applied to epidemiological findings on the social determinants of health, aiming to explain pathways between unfavorable social status and unfavorable health outcomes. The key point for the present context is that the hypothesized causal mechanisms are biopsychosocial, indeed they have to be because the pathways implicated by the epidemiological research run across the three domains. Common hypothesized causal mechanisms and pathways include the following: low levels of social resources (e.g. working poverty and other forms of social exclusion) lead to chronic lack of control over salient outcomes, leading to chronic psychological stress, raising risk of anxiety and depression, while the chronic physiological arousal associated with chronic psychological stress raises the risk of dysregulation and damage across multiple biological systems and hence poor health outcomes. There is not space here nor is it the intention to review the large literature on chronic stress models (O'Connor, Thayer, & Vedhara, Reference O'Connor, Thayer and Vedhara2021; Roberts & Karatsoreos, Reference Roberts and Karatsoreos2021), but simply to give an illustration of a substantial research program on biopsychosocial causal mechanisms.

The second example of a new causal explanatory biopsychosocial theory with wide application comprises new models of pain perception, implicating neurobiological–psychological processing as well as peripheral physiological or structural damage. The new models implicate the person's negative appraisals of the meaning of pain and expected adverse effects on their lives and task demands, and associated Central Nervous System (CNS) pain-processing mechanisms (Garland, Reference Garland2012; Ong, Stohler, & Herr, Reference Ong, Stohler and Herr2019).

This new understanding of pain perception is directly relevant to conditions dominated by pain, but there is a broader point that is relevant to the health sector as a whole, specifically to drivers of service use. The complex of pain, distress about pain, with associated impairment of functioning, is close to ‘feeling unwell’ and is a main driver of referral and service use. It is increasingly recognized that significant proportions of patients with such presentations turn out to have medically unexplained symptoms. ‘Medically unexplained’ here actually means biomedically unexplained, and biomedically orientated clinics typically have no biopsychosocial management protocols in place; consequently needs are not met, and the patient journey can be potentially long and costly; for example, in general practice and medical clinics (Jadhakhan, Lindner, Blakemore, & Guthrie, Reference Jadhakhan, Lindner, Blakemore and Guthrie2019), cardiology (Lenderink & Balkestein, Reference Lenderink and Balkestein2019), neurology (Carson et al., Reference Carson, Best, Postma, Stone, Warlow and Sharpe2003), and surgery for some pain presentations (Louw, Diener, Fernández-de-Las-Peñas, & Puentedura, Reference Louw, Diener, Fernández-de-Las-Peñas and Puentedura2017).

Neuroscience and genetics are biopsychosocial

This overview of BPSM research paradigms with examples of major research programs has to briefly mention that the two life sciences that have accelerated the most in recent decades – genetics and neuroscience – are suited to a biopsychosocial theoretical framework. Indeed, it's more than that; they have been instrumental in making the new BPSM compatible core theory reviewed in Part 1.

It was advanced in genetics that introduced into biology theoretical ideas of a new kind of science involving coding, information-transfer, error, regulation and control, additional to energy-transfer and -exchanges covered by physical–chemical laws (equations). Further, theories of genetics have always been thoroughly interactional across domains, in evolutionary theory, and recently in the new field of epigenetics, including in psychiatry (Campanile, Fanelli, Fabbri, Serretti, & Mendlewicz, Reference Campanile, Fanelli, Fabbri, Serretti, Mendlewicz, Bhugra, Moussaoui, Craig, Bhugra, Moussaoui and Craig2022; Cecil, Reference Cecil, Savulescu, Roache, Davies and Loebel2020).

The same theory-shift that transformed biology also transformed neuroscience and cognitive psychology, enabling a coherent biopsychology. As to the domain of social interactions, there is no shortage of research programs on its major importance to our biopsychology in phylogenesis (Barrett, Henzi, & Barton, Reference Barrett, Henzi and Barton2022) and ontogenesis (Blakemore, Reference Blakemore2008).

Clinical trials and guidelines

The clearest clinical implications of the BPSM, in contrast as always with the narrower BMM, is accommodation of psychological and social factors as well as biological factors relevant to clinical management and treatment. The importance of this broader scope has been substantially supported in the clinical trials literature, appearing mainly after Engel wrote his 1977 main paper. Large-scale clinical therapeutics research programs in the decades since have studied and shown the effectiveness of some psychological therapies for a large range of health conditions (Barkham & Lambert, Reference Barkham, Lambert, Barkham, Lutz and Castonguay2021), of combination therapy, medication plus psychotherapy, for some conditions such as depression (Breedvelt et al., Reference Breedvelt, Brouwer, Harrer, Semkovska, Ebert, Cuijpers and Bockting2021), and of social treatments such as social support (Brown et al., Reference Brown, Luderowski, Namusisi-Riley, Moore-Shelley, Bolton and Bolton2020; Wang, Mann, Lloyd-Evans, Ma, & Johnson, Reference Wang, Mann, Lloyd-Evans, Ma and Johnson2018). The details of exactly what helps what are always specific to details of treatment, condition(s) and stage, but the overall picture that has emerged from the clinical trials literature is consistent with the broad biopsychosocial framework, in the precise sense that any narrower framework – envisaging treatments that are biological only, psychological only, or social only – omits some effective treatments for some health conditions at particular stages.

Now that there is a substantial clinical trials literature, summarized and adapted in clinical guidelines, it is of major importance in clinical decision making. And the broad message, as above, is that the broad biopsychosocial framework is required to accommodate it.

Like the BMM, the BPSM covers not only causes of onset (etiology) and treatments, but also post-onset maintaining causes that adversely affect prognosis. The models of chronic stress and of pain outlined above are examples of maintaining mechanisms across the biopsychosocial domains. Social determinants of health associated with chronic stress, for example, are typically on-going, affecting not only illness onset but also prognosis, for example, by adversely affecting access to treatment (Schneider, Roots, & Rathmann, Reference Schneider, Roots, Rathmann, Immergut, Anderson, Devitt and Popic2021).

Theorizing personal, interpersonal, and institutional factors in clinical care

A lot more is going on in clinical care than decisions as to what treatments to recommend, including personal, interpersonal, and institutional processes. Specific issues include the role of the person as patient – in determining what is wrong, whether anything is wrong, in collaborating on a treatment plan – the imperative of ‘compassionate’ care (Hodges, Paech, & Bennett, Reference Hodges, Paech and Bennett2020), and institutional/professional factors supporting or jeopardizing good clinical care (Mannion et al., Reference Mannion, Davies, Powell, Blenkinsopp, Millar, McHale and Snowden2019). Engel says a lot of interesting things about all these things in his 1997 paper and others around that time (Engel, Reference Engel1980, Reference Engel1982), and they can be considered as part of what is covered by the BPSM.

While persons, interpersonal relations, and institutions can be accommodated within the BPSM, the contrast with the BMM is, as always, clear. These matters are simply out of the scope of the narrower BMM: they are not in its ontology, and therefore, it has no idea of them at all, still less their causes and effects. Therefore, adherence to the BMM (or to any model or line of thought, by whatever name, which regards biological processes alone as being causally relevant to health and disease) will have to construe these other matters in another way, broadly not as science, but as ‘art’. There is of course a grand tradition of this approach in medical theorizing – see, for example, Nassir Ghaemi (Ghaemi, Reference Ghaemi2010) – and Engel argued against it, favoring rather a broadly scientific-investigatory approach to understanding and improving, for example, receptive and expressive clinical communication skills, and institutional supports of clinical professional care (Bolton, Reference Bolton2020; Engel, Reference Engel1978, Reference Engel1980).

The revitalized, cross-disciplinary BPSM proposed here can be used to theorize personal and institutional factors relevant to clinical care and highlight their role as critical and not merely discretionary considerations. For example, the fundamental importance of personal agency in psychology is consistent with the central role of the person as patient, and the fundamental importance of socio-political factors in regulating recognition and access to resources can help theorize and highlight interpersonal, institutional, and wider political processes that affect clinical care.

The key added value of the BPSM, in contrast with BMM, is that it accommodates personal, interpersonal, and institutional factors in clinical care within the causal systems affecting health and disease. That said, it should be noted that these issues have been most theorized by other models of care, for example patient or person-centered models (Epstein & Street, Reference Epstein and Street2011; Nolte, Reference Nolte2017), and the more recent Recovery Model (Hare-Duke, Ng, & Slade, Reference Hare-Duke, Ng, Slade, Bhugra, Moussaoui and Craig2022), and in reports and studies on healthcare institutional failure (Reader & Gillespie, Reference Reader and Gillespie2013). There is the further important point that the increasing voice of the person as patient has been substantially a consequence of activism and wider socio-political movements, not a matter of healthcare theory and research (Brown, Reference Brown1981; Rashed, Reference Rashed2019).