This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in Journal of Participatory Medicine, is properly cited. The complete bibliographic information, a link to the original publication on http://jopm.jmir.org, as well as this copyright and license information must be included.
The number of academic papers referring to patient engagement or to related terms has been rising sharply for at least 20 years; several review articles have recently been published enumerating a wide variety of situations of patient involvement in research and partnership with health professionals.
As no standardized keywords and no shared classifications exist to facilitate comparative studies of situations where patients and their organizations are recognized as coresearchers, this paper purports to create a typology to analyze those situations.
Based on 8 already existing meta-reviews or related studies, this work is achieved using a template based on Claude Bernard’s conceptualization about experimental medicine.
This typology allows differentiating between modes of involvement and levels of patients reflexivity mobilized in evidence-based medicine (EBM) trials. Screening through a first set of various meta-reviews using this typology shows that a high level of reflexivity is seldom observed and seen only when a patient organization (PO) is involved in the process. This suggests that such an organization can play several roles essential to high reflexivity trials; the PO is capable not only of grouping singular approaches but also of synchronizing and correlating them. However, as nowadays health researchers and POs give more attention to syndromes or troubles for which EBM clinical trials are not relevant due to lack of biomedical indicators (eg, fibromyalgia, chronic fatigue syndrome, or psychiatric disorders), a supplementary mediation category is added to take into account action-research, community-based participatory research, and grounded theories.
With this new category, this typology should be able to classify most of the cooperation schemes and thus be a useful tool for the next systematic reviews.
The number of publications mentioning patient engagement and related terms such as patient involvement or patient participation reported through PubMed has strongly increased for at least 2 decades, as shown on
Screening those publications shows that the uses of the terms patient engagement, patient involvement, and patient participation have developed concomitantly over the last few years without any clearly defined differences in acceptance. Most works use one of these terms without specifying why they have chosen it, either because authors consider them synonymous or because they do not feel the need to justify their choice. More generally [
It results from this situation that, although the recognition of patient experiential knowledge (PEK) has reached the level of a social fact attested by diplomas, jobs, laws, and academic concepts [
In this context, this paper aims at creating a classification embracing patient opinions and approaches as well as those in which patient contributions based on their PEK are accepted in their own right (ie, where patients are fully accepted as coresearchers).
From a methodological point of view, without shared keywords, ascendant clustering converging to a semantic-based taxonomy is not possible. Therefore, our aim cannot be to create a taxonomy based on a bibliometric or lexical study but rather to formalize a complete classification (a set of rational categories)—that is to say, a categorization which allows us to specify major types corresponding to the main ways in which patients and academic researchers associate today.
Ratios of publications mentioning patient commitment, patient involvement and patient participation compared to those mentioning pneumothorax and psoriasis in their titles and abstracts according to PubMed data. The slower growth of the patient empowerment ratio was also indicated as a reference. All curves are approximated by 4th degree polynomials (Microsoft Excel).
Thanks to previous studies [
Step 1: Look for a template or grid able to describe the phases in which the cooperation would be mobilized or not (we used a very simplistic description of Claude Bernard's experimental medicine [
Step 2: Using this model, screen a corpus of papers embracing the broadest types of patient cooperation with 2 aims: validate the best set of categories capable of accounting for the diversity of cooperation encountered and be sure that any kind of cooperation could fit in one of those categories.
Due to the keywords issue, instead of creating a new minute, even pernickety, review of primary literature, we decided for step 2 to use already published meta-reviews such as those quoted above. Our first idea was to select them through a systematic search on PubMed. However, similar problems of terminology relevance arose. Queries on PubMed titles or abstracts for review and patient engagement (152 reviews found), patient involvement (246) or patient participation (275) provide too numerous results; 642 different reviews are found with 1 of those 3 terms. Therefore, we opted for a pragmatic, targeted selection and chose a small set of 8 complementary meta-reviews that were already often quoted:
Four general syntheses and descriptions concerning situations of coresearch found in medical literature [
Three more specific meta-reviews: autoethnography [
An eighth review paper dedicated to inventory and description of the European Patients’ Organizations in Knowledge Society (EPOKS) [
We added this last review paper because we wished to take both collective and individual cooperation into account.
As explained previously, our first goal was to be able to distinguish different levels of reflexivity recognized by patients; therefore, an easy-to-use description of current research protocols was needed. To build a first version of the typology itself, we elected to use a system of description inspired from a simplification of Claude Bernard’s formalization of experimental medicine [
This OHERIC grid is introduced only to be used in a quite pragmatic way as an ideal type description or a computational intermediate (ie, a background against which to set the practices we aim at describing). In view of the critic of linearity, we give no specific chronological significance to OHERIC phases, which can then be considered as aspects of the same process that can overlap or even interpenetrate.
The OHERIC pattern can be used to build grids in which arrows specify who exerts the main reflexivity at each stage of the research. Up arrows specify a nonacademic origin (bottom-up process), while down arrows designate an academic origin. This appears close to the stages of research process as described in the
Phases of an investigation described with the stages of Bernard, Wallcraft, and Dewey.
OHERIC phases | Oa | Hb | Ec | Rd | Ie | Cf | ||||
Bernard’s research phase [ |
Initial data or observation leading to research | Putting forward a research hypothesis | Building an experimental device, inventing a test | Collecting observations, realizing a test | Processing results and data | Interpreting results | Scientific conclusion | |||
Walcraft et al “Involvement in Mental Health Research” | Identifying a research topic | Designing research | Outcomes measures | Data collection | Data analysis | Interpretation | Write up and dissemination | |||
Dewey’s phenomenology “How we think” | A felt difficulty; its location and definition | Suggestion of a possible solution | Development by reasoning of the bearings of the suggestion | Further observation and experiment | Leading to its acceptance or rejection (1) | Leading to its acceptance or rejection (2) | That is, the conclusion of belief or disbelief |
aO: initial observation.
bH: hypothesis.
cE: experiment proper.
dR: result.
eI: interpretation.
fC: conclusion.
Comparison between 2 protocols for chronic diseases.
Such a template can easily be applied to the clinical trials in EBM because they are based on protocols that are close to the OHERIC phase sequence. Therefore, it is quite easy to create a first classification defining different levels of mobilization for the patients’ reflexivity. Concerning patients associated with this kind of trial, OHERIC grids allow us to distinguish 2 main types of roles: (1) patients and their relatives were mainly considered as mere data collectors and (2) more diverse situations that all had a fact in common—patient reflexivity (ie, PEK) was at the heart of the research process.
More precisely, category 1 corresponds to an involvement of patients and relatives as data collectors in the E phase of OHERIC. Collected data may consist of personal opinions and biomedical parameters (eg, blood pressure, glycemic measurements) or self-evaluation according to one’s own perception (eg, pain, anxiety, discomfort) or opinion [
In some cases belonging to this first category, patients are also associated with part of the processing of the data collected (OHERIC R). As this has the effect of reinforcing their reflexive activity, we decided to characterize these situations through a new category we named 1+. At this step, we can mention that it could be relevant to distinguish the collection of opinions from the collection of biomedical parameters (either quantitative or qualitative) through 2 more subcategories 1OP (opinions) and 1PA (parameters). We will see later that this first subcategory, 1OP (opinions), has to be merged into a larger one (ie, M category).
Category 2 collects situations where patients and relatives contribute with their reflexive capacities to other phases besides data collection or initial processing. We first distinguished 2 subcategories in it: 2 for participatory EBM and 2+ for a full popular epidemiology process.
The subcategory 2 or 2ParEBM was created to characterize situations corresponding to what we can call participatory EBM where academics decide to involve—in parallel to the main EBM process—lay people’s reflexivity for all research phases following the initial set-up (ie, all OHERIC but O or perhaps E).
The second subcategory, 2+ or 2+PopEpi, is inspired by Brown’s research in medical sociology [
Categories and subcategories of types of involvement forpatients and relatives as coresearchers in medical research projects.
Screening the publications quoted in meta-reviews shows that individual patients are seldom recognized as contributing fully to the reflexive production of new knowledge. As a matter of fact, in category 2, real research responsibilities are entrusted to collective community actors rather than to individual patients, who find themselves restricted to type 1 functions (ie, data collection). Such intermediary collective actors may be whole communities, mutual assistance groups, or patient organizations (POs).
By providing a framework for collective action and investigation, these POs not only help patients develop their individual reflexivity and ability to compare their situations but also lead patients to synchronize them. Collective, synchronized patients’ reflexivity and investigations become then more readily describable in the language of a collective protocol taking on the form of a succession of stages akin to the OHERIC formalism.
If we adopt the spiral-shaped representation of the individual pragmatic forms of thinking (in the sense of the curls proposed—in addition to Dewey’s work—by Ashby [
Furthermore, once started, such a PO producing an OHERIC-type protocol aggregates the incoming patients by putting them literally in sync with those already included. And it is precisely because the work accomplished by the PO is then collectively turned into hypothetical-deductive parlance thanks to this unfurling that it is made acceptable in type 2 projects without endangering the criteria of the academic patterns according to which new medical knowledge is produced. POs then fulfill a role of socializing and reformulating each individual patient’s metacognition and reflexivity, allowing them to be taken into account by academic research teams as intellectual inputs. Furthermore, these POs may also be places where collective inquiries can be decided, either invented by the patients themselves or suggested by relatives, caregivers, clinicians, or researchers.
The collectivization of reflexivity operated by a PO can act during different OHERIC phases: Observation and Hypotheses, through the collective problematization or formalization of the issue (this includes issuing a hypothesis that can be tested through collectively taking/acquiring a critical distance from situations experienced individually); Experience and Results, as a self-training framework in which patients learn how to observe and tend to themselves (and sometimes as a furnisher of note-taking tools, multiple choice questionnaires, or quantified self-tools); and Interpretation and Conclusion, by organizing the formalization of conclusions.
Potential effect of a patient organization according to research categories.
From an epistemological point of view, it can be said that the multiple interactions between various patients through a PO bring about a reshaping of their singular pragmatic phenomenology into a kind of investigation relevant for EBM researchers (eg, in a collective process that fits with Bernard’s experimental medicine). Therefore, we may represent interactions between the patients and the academic researchers using a 3-layer model: pragmatic phenomenology (Dewey’s level) is linked to experimental method (Bernard’s level) through PO interaction, and these 2 levels are themselves linked to the third one (EBM clinical trials).
Level of global scientific research:
Clinical trials developed in evidence-based medicine epistemology [
Level of rational work, local or in group:
OHERIC-describable protocols [
Level of individual experiential knowledge:
Dewey’s pragmatic phenomenology of each patient [
In the EBM clinical trials point of view [
However, when researchers following EBM protocols realize they need to take into account not only biomedical data but patients’ reflection as well, they themselves may try to foster the creation of POs, either in cooperation with the patients or with economic actors as manufacturers of health products or health insurers. Meta-reviews show that with the great expansion of interest in mobile phones and other connected objects, numerous cooperation programs have now been established between researchers, mHealth companies, and POs, either preexisting or specially created for this purpose.
Of course, cooperation between patients and academic researchers is not limited to the pattern of clinical trials in EBM. The existence of other cooperation schemes is easy to verify through an analysis of more specific corpora (ie, mental health [
In fact, the POs have different postures regarding cooperation with health researchers, and EPOKS’s evidence-based activism [
In an intermediate posture between what we can call allegiance to the EBM and radical opposition to it, many POs are focused on individual strategies and folk theories and on developing intersubjectivity through the transformation of individual experiences into narratives or accounts that can be shared.
As a matter of fact, when data consist only of isolated patients’ dispersed and unsynchronized lived accounts, EBM (particularly clinical trials) has the effect of drastically limiting the potential contributions of patient reflexivity to the construction of new health knowledge (see, for instance, Faulkner and Thomas [
We thought it necessary to distinguish between these different cases while still retaining the already mentioned 2, in which we had specifically included EBM research projects organized beforehand to entrust patients with specific functions that can be described through a OHERIC framework (eg, a patients’ group organizing and analyzing specific biographic workshops to collect qualitative data in order to compare various evaluation processes of the effectiveness of a long-term treatment on pain perception).
Hence we decided to define, besides the main categories 1 and 2, a third category called M (as for mediator in the sense of facilitators) to collect cases where patients were used as intermediaries, facilitators, or even multipurpose interpreters [
For that purpose, M ethnological protocols offer an alternative to the Dewey-OHERIC-EBM translation. It relates to the patients in both a more heuristic and comprehensive manner, as is shown in
Practically, patients are no longer seen only as witnesses bringing data to hypothetical-deductive epistemologies. Constructions of folk and academic types of knowledge can then be put into relationship under the condition that other tools are brought to the core of the protocols: tools of description of patients’ relationships to their illness as well as tools of production of innovating knowledge, using not only hypothetical-deductive models but also more comprehensive methods, such as grounded theory [
The recourse to ethnological protocols is not the only approach found by researchers to short-circuit the Dewey-OHERIC-EBM translation and allow the setting up of other bottom-up chains of production of knowledge. Other disciplines are used in M category to blend with the benefits available from patients’ and relatives’ pragmatic phenomenology in order to achieve more efficiency and hence more confidence and observance from its beneficiaries. This is, for instance, the case with educational science for research on PEK [
Three layers model used to compare 2 regimes of connection between pragmatic phenomenology and global knowledge production.
Coupling model | EBMa clinical trials regime through patient organizations and Claude Bernard | Full participatory action research non-EBM regimes |
Types | 1, 1+, 2, 2+ | M |
Level of global scientific research | Epistemology of the EBM [ |
Epistemologies of comprehensive research (such as action research [ |
Level of rational work, local or in group | OHERICb-describable protocols [ |
Comprehensive organization through participatory M category action research |
Level of individual experiential knowledge | Dewey’s pragmatic phenomenology of each patient | Pragmatic phenomenology of each patient |
aEBM: evidence-based medicine.
bOHERIC: initial Observation, Hypothesis, Experiment proper, Result, Interpretation, and Conclusion.
Examples of subcategories for M type based on disciplines contributing to knowledge production in complement to evidence-based medicine.
M sub- categories | M anthropology | M psycho-sociology | M politics and economics | M education | M info-communication | M ergonomics |
Concepts, facts studied | Culture, ethics, knowledge, and beliefs | Social representations | Health governance, users’ representation, health democracy, activism | Knowledge, abilities, self-education, learning | Connected tools, networks, quantified self, patient 2.0, eHealth | Adaptation of artifacts, prosthesis, customer made |
Completed typology tree.
Building such a typology also addresses the question of the social representations of research. Our hypothesis is that these representations are dependent on the need to associate lay people as producers of data, producers of PEK, and even as coresearchers: the greater the need to involve communities to obtain scientific results by using their (individual and collective) reflexivity, the greater the resulting shift by research bodies toward epistemologies more open to taking into account lay people’s pragmatic phenomenology.
Health research as a social construct negotiated among stakeholders: with the rise of impure science [
The main question is neither to determine whether academic knowledge obtained via clinical trials in EBM is worth more than patient knowledge experienced from synchronization of singular phenomenologies nor to choose the best model of knowledge production between Bernard, Dewey, or Lewin; on the contrary, it is now to find ways to make all those contributions converge. The more researchers must steer between the taken-for-granted representations of the world of lived experience and the ideal types of proof-finding, the more they will have to balance multiple ways to define how health research can be done.
evidence-based medicine
European Patients’ Organization in Knowledge Society
initial Observation, Hypothesis, Experiment proper, Result, Interpretation, Conclusion
patient experiential knowledge
patient organization
None declared.
Lexical analysis illustration of PubMed papers about "patient engagement" obtained by the author (Olivier Las Vergnas) using Iramuteq and R open source softwares.