Becoming Science : Science of Becoming.

Improvement science as a science of practice is not about ‘getting things done’ but about ‘doing things’. Batalden characterized the moral engagement required for professional ‘becoming’ as not only ‘doing ones work’ but also ‘improving ones work’. It is not the case of changing systems and keeping individuals constant.  Improvement as a practice science collapses the distinction between ‘doing’ and ‘becoming’ by lowering the threshold for action and increasing our sensitivity to day-to-day experience. The paradigm of learning in improvement science is radically/’rigorously’ empiricistic: Experimental and Experiential.

Often times the debate is framed as one between ‘mere ideas’ or opinion versus that of ‘indubitable evidence’. This gives entrenched health service researchers their pound of flesh and conveniently reduces ‘improvement science’ to ‘evidence based practice’ or more ostensibly ‘implementation science’. This paradigmatic clap-trap forces the progressives into the subaltern quicksand of ‘practice based evidence’ or even more fragile ‘question- generating research’. Rigour and Relevance have their own masters: fidelity and productivity respectively. What is hidden is the ideological regression underlying implementation from ‘doing things ‘ to ‘getting things done’. The latter is underpinned by an apriorism: evidence resides in an lofty Platonic realm of the unchanging world of the absolute or a Cartesian Theatre of subjective certainty. In this characterization what lies outside ‘intellectual’ evidence is wrongly treated as ‘embodied’ ignorance. It mandates that without ‘necessity’ there cannot be ‘sufficiency’; without ‘thought’ there cannot be ‘action’; without ‘efficacy’ there cannot be ‘effectiveness’.

A science of practice must answer this ‘causal’ conundrum: doing the right thing is a prerequisite for doing things right. The more fundamental question is can you ‘do’ the right thing without doing it right. The problem arises ‘in theory’ because the emphasis is on the ‘right thing ‘ and ‘right way’, while ‘in practice’ what matters most  is the ‘doing’. Evidence doesn’t walk the ward. The evidence- practice gap is material and contingent not trans-historical. It is discerned empirically in ‘practice’, there is no evidence of it in theory. Evidence known and evidence not practiced can be the same thing. In an individual it is failure of the will or akrasia, but evidence based medicine is epistemology not gnosis. Evidence based medicine is a peuditic hand maiden to practice: it assists the birth of practice. It is not a theory of practice. It exists but is not ‘real’: If there is no formalism i.e. evidence -practice gap, there is no evidence- based medicine. On the other hand, one can negate the need for improvement only by negating the need for practice itself.

The argument that there is no effectiveness without prior demonstration of efficacy hides totalitarian ideological assumptions  of the evidence- based medicine movement. It not only confuses evidence with science but equates evidence with practice. To say all practice must be evidence- based is admissible only if there was evidence for everything and yet it is not being implemented or if there is ‘indubitile’ evidence of harm and it is being implemented. This is ‘justifiable territory’. It is in the edges or margins of knowledge that there is contestation between implementation and improvement science.  The absence of evidence is not evidence of absence. Here binary thinking breaks down. In this ‘justiciable territory’, casuitic  reason has a role and both implementation and improvement science have equal locus standi. The one who doesn’t know the way cannot forbid passage to another. This creeping of evidence-based medicine from ‘quid facti’ to ‘quid juris’ creates an unfair moratorium on purposeful, ’justiciable’ creativity or technological poiesis. [Heidegger] in practice.

Improvement science is not merely ‘evidence based practice ‘. It is the ‘justiciable’ progress of practice. Where there is no evidence available there is no evidence based medicine , but practice exists and is real: not as an organism or organization but a ‘body without organs’, a ’grass root’ , a prostatic ‘rhizome’. Effectiveness is inherent to the process of all clinical practices. Practice is carefully motivated: the essence of practice is creativity-aesthetic, machinic, productive, destructive, consumptive, informatics etc. Improvement augments that creativity. Practice effectuates, Improvement meta-effectuates: Improvement systemizes effect ie simple [visible knowledge],significant [valuable],stable [predictable],sustainable [standardized] effects. Effectiveness unlike efficacy is not an universalizing testimonial function but is expressed as a range or degree or progression or simply ‘qualities’. The recognition of this ‘differential’ nature of effect  allows us to identify Positive , Neutral and Negative states of practice :improvement, needlessness/waste and harm. Needless/waste and negative/ risky processes are governed by long-tail logic and are not readily amenable to methods based on the central limit theorem. The opposite of improvement is not ‘evidential status quo’ [whether for or against] but systemic entropy, waste and harm. This is the reason improvement comes prior to quality. There is no status quo in becoming.

Whitehead’s idea of ‘singular causation’ plays a predominant role in improvement thinking. The textbook example of singular causation as posited in Whitehead’s process philosophy is ‘I woke up this morning because my alarm clock rang’. Additional singular causes may contribute such as ‘I was lying close to the clock when the alarm rang’.This is contrasted with ‘nomic’ causality which is expressed as ‘alarm clock generally wakes people in the morning’. Practice is also an abstraction that can be theorized. To become a science it is necessary, though only minimally,  that the practice under study be a knowledge- producing practice.  It is impossible to separate experience from practice. It is the sequence of experience that constitutes practice. Experience per se is unchanging but practice as an enduring object is changeable as it is constituted by a ‘complex nexus of occasions of experience’: open and always in a flux. Singular- causes structure ‘artificial knowledge’ or knowledge that emerges from practice. Practice knowledge is useful but complex, malleable and decomposable into overlapping ‘territories’or paths of knowing, affect and significance.

When practice is viewed  as inherently positive there is no role for ‘hyper-effectiveness’ ie, the testimonial function of effectiveness or insisting on confirmation of nomic causality where singular causality is sufficient. When justiciability is fore-grounded in practice, such as the creative use of the scientific method / improvement , then it is meta- effectiveness. So the real debate between implementation science and improvement science  or between evidence and ‘real time data ‘ is actually between justification and justiciability,between testimony and eunomics,between analytic and synthetic reasoning;between maximization and optimization,  between hyper-effectiveness and meta- effectiveness. The onus is on implementation science to justify hyper-effectiveness: calling for evidence where ‘structural modeling’ is sufficient. Not using a parachute when jumping off an aeroplane heading for the moon, because there are no randomised trials to support ; checking the drivers licence  before boarding a public transport or insisting every police man should be protected by another policemen, ‘administering a survey to a spouse to improve familial bliss’   are all examples of excessive ‘formalisation of effect ‘ or hyper-effectiveness.

Improvement science is closer to research than evidence- based medicine. Knowledge classically is defined [in spite of Gettier] as Justified True Belief. Research and Improvement are about ‘justification’ whereas evidence based medicine is about belief. If one denies possibility of ‘Truth’ and accepts verisimilitude as natural then there are degrees of belief and degrees of justification. The evidence pyramid conflates the two. It is in this ‘dynamics of thought’ [ not merely thought- action gap or evidence-practice gap ] that trial and learning of improvement science thrives. Agential ‘Knowing’ and not merely disembodied  ‘knowledge’ are preconditions of ‘deep practice’: a practice that acknowledges epistemic and metaphysical uncertainity in knowledge, the flux and flow that is inherent in practice ie dynamism,  as well as risk in the role or identities of improvement agents.

The ‘definition’ of quality is always ‘qualitative’: Quality is not fixed but is judged/expressed in degrees.  The IOM defines it as a degree and Berwick following Ishikawa as the degree of match between demand and provision. Thus we can map degrees of belief and degrees of justification with degrees of effect. Thus belief, Justification and quality are ‘not simple’ . Not static.  Acknowledging the ‘differential’ constitution of belief, Justification and effect or quality provides the locus for the pragmatic emergence and growth of knowledge. Effectiveness accommodates a ‘historicity in the present’ in which knowledge evolves through cumulative degrees of usefulness, through graduating degrees of justiciability. While hyper-effectiveness mandates the best option first, effectiveness through ‘creative experimentation’ [Steve Spear] allows starting with the first best option.

Efficacy is being, effectiveness is ‘becoming’. In theory one cannot become without being, but in practice where is the being in becoming: Heraclitus warned us about the river’s flow. The necessary- sufficient conditions of deterministic reasoning are diluted in ‘change’. Research is reflective, Improvement is reflexive.[in the George Soros version though it’s provenance  can be traced back to Kant via Popper;]. It searches the future. It epitomizes ‘possibility’ not ‘necessity’.  Atleast since Deming, it is recognized that improvement requires epistemology. My reason for invoking the metaphysical distinction between being and becoming is to assign equivalent epistemic value to ‘knowledge’ and ‘knowing’.   Improvement science affords metaphysical [modal], epistemic, moral and aesthetic spaces besides a psychological one: learning- to- be or becoming, rather than merely learning- to–do/not-to-do [adaptive capacity; Rashad Massoud calls implementation as adaptive science] or even learning-to-learn [absorptive capacity; Cohen and Levanthal 1996]. Creativity or innovation or ‘insight’ [‘rapid solution to a problem’] should not be reduced to ‘functions’ of the last two categories of learning alone.

Not all research is useful to practice. Evidence based medicine is temporally prior to improvement science not metaphysically so.  The vast majority of research is not ‘utility’ driven. Much of basic research is done for research’s sake [absence of a practical goal] with the hope of indirect benefit or utility in the long run. Discouraging such ‘academic’ research would be detrimental to the research enterprise as a whole. Thus the epidemiology of medical information estimates offered by David Eddy et al answers the question what percentage of practice is evidence based.[evidence -of]. This question is not the same as asking what percentage of research is used in practice [evidence -for]. Mc Glynn et al’s work doesn’t ask this question either. Her work asks the question how wide spread is evidence- based best- practice or basic quality standards or recommended processes.

The general reification of evidence in the post-EBM era has masked the fact that the ‘concept of evidence’ is not the same in efficacy and effectiveness. ‘Evidence of’ is different from ‘Evidence for’. Evidence is judged not just by rigor but also by relevance. Evidence-of is generation of practice-rules[metaphysical]; evidence-for is generation of practice tools[material]. Production of effect is methodologically different from production of evidence: one is a ‘recipe’ theory and evidence is about ‘ingredients’; one is about ‘change’ and the other is about ‘choice’, one aims for ‘arete’ and the other for ‘episteme’. It is the ‘identity’ of ‘evidence’ in efficacy and   effectiveness narrative that hinders progress on theory of improvement. To paint efficacy and effectiveness with the same brush of ‘evidence’ is a category- mistake. Hyper-effectiveness is the mis-use of analytical reasoning for synthetic purposes.

In comparing efficacy and effectiveness there is a misleading assumption about context. The idea of context is treated differently given the controlled and real-time assumptions underlying research and practice. But methodologically study designs make a distinction between independent and dependent variables and between causal and confounding variables. The positivism that is associated with underpinning methodological empiricism of efficacy studies is that variables are simple. Yet in both efficacy and effectiveness the confounder cannot be qualified. The confounder is a confounder. There is no scope for distinction between ‘good confounder’ and ‘bad confounder’. Context is not a passive condition but a driver of adaptation in the micro-sociology of practice knowledge.  Effect is a posteriori result not just apriori goal.

There is also a confusion between ‘history’ and ‘theory’ that requires disentanglement. The theory of improvement is not about the origins of these ideas which quite often gravitates towards the ‘intuitions’ and ‘experiences’ of management consultants., when it should be about clarifying the ‘ explanatory’ power of these ideas. The idea of ‘explanation’ in improvement and the philosophical assumptions underpinning it, is the theoretical starting point for improvement as a science: How and why does improvement work? How does knowledge grow in practice? How does practice knowledge become practical value? To encourage such a search I offer a theoretical manifesto below:

 

Efficacy Effectiveness
Being Becoming
Focus on Rigour [reliability] Focus on Relevance [validity]
Indubitable Knowledge: evidence- of Useful Knowledge: evidence-for
Methodological Individualism/ action-theoretic Methodological Pluralism and system-theoretic
Reductionism Holism
Assumes Homogeneity of field conditions Recognizes Heterogeneity of field conditions
Inclusion/Exclusion of study variables are Theory led Inclusion/Exclusion of study variables are Practice driven
Organic Mechanism and Clinical Feasibility Inorganic Mechanics and Organizational Feasibility
Ingredient Theory Recipe Theory
Universalism: True under all skies Universal Localism: Particular practices
Best option first First best option
Certain Knowledge in principle Knowledge –over- time: Emergence, Growth and Spread
Inorganic/disruptive spread or Plug’n Play Organic/adaptive spread: local re-invention
God’s eye view of practice Grounded: Patients view of Practice
Nomic causality Singular/ efficient causality
Reflective Reflexive
Enumerative Statistics Analytical Statistics
Analysis Synthesis
Control Accountability
Substance/ identity metaphysics Systemic[process]/ Differential  metaphysics
Intentional agent Volitional agent
Linear system Circular[iterative] and interactive
‘Testimonial’ Function ‘Eunomic’ Function
Striated Territory Smooth Territory
Statist Science Nomadic Science
Second Person Perspective [Off-line or MetaResearch] First and Third Person Perspective [Real-time or ChronoResearch]
Post Baconian Induction Baconian Induction and Piercean Abduction [model-based]
Standalone Necessity Cumulative Sufficiency
Representational Interventional

The philosophy of pragmatism in the development of improvement science is well known but the role of ‘process philosophy’ is not acknowledged. The idea of becoming is central to process philosophy especially since the attempts of Alfred Whitehead  to explicate a broader foundations for science.  Process philosophy is not only possible and desirable but is constitutive of the theoretical underpinning of improvement as a ‘rigorous’ science.  Call for broadening the ‘evidence-base’ of practice requires additional meta –theoretical commitments: to include process, ideas, change, structures, interests, agency , power, empathy , rights, models, functions, imagination, risk etc and their interrelations as objects of study. Theoretical pluralism gives us the scope to go beyond any stipulation of Deming’s ‘system of profound knowledge ‘as the only foundation of improvement science.

Note : Descriptions of improvement as a science of change [Greenlagh],of action, of execution, of implementation etc are prevalent. My intention in this post is to go beyond the comfort zone of positivism and flag the possibility of exploring the ontology of practice as ‘becoming’ and the implications for the epistemology of improvement. While I have thematically borrowed from process philosophers, my intention is not to defend a standpoint as my  audience is not philosophers. I have not made any definitive arguments as my intention is to merely to ‘open a window to let some fresh air into the room.’       Process philosophy’s culmination in the French avante garde thinker Deleuze  does not recognize ‘systems with aims’ but intensities and affinities. Process philosophy and process thinking [as viewed in systems theory ] are not co-terminus. For eg. In Deleuze’s theory there are no solutions, only ‘new problems’.

It is generally accepted that natural sciences demand explanation, mathematical propositions call for proofs and action calls for justification. Yet it is possible to bring them all under the common denomination of understanding. My reason for delineating ‘Justification’ itself into ‘justifiability’ and ‘justiciability’ is to nuance the emphasis on ‘answerability’ and ‘questionability’; of ‘reflection’ and ‘reflexity’; of ‘research’ and ‘future search’  respectively.Knowing as growth of belief and justification.

The aesthetics of improvement is in the creative use of the scientific method. To progress, improvement science must stake its rightful claim to effectiveness. But to do that it must first clarify itself to itself: its aim, its functional domain or boundaries, its object of study ,its list of problems,  its definition of knowledge, its sources of explanatory power, the identity of its agents and its own limitations

This entry was posted in Theory. Bookmark the permalink.