The Behaviour Change Wheel (BCW) is a framework that supports systematic development of interventions [27, 29]. It is designed to facilitate systematic, evidence-based progression from behavioural analysis of a problem to intervention design employing behaviour change theory to bring about desired change in three stages as shown in Fig. 2.
Each of the COM-B components maps onto the Theoretical Domains Framework (TDF)-a synthesis of 33 theories and 84 theoretical constructs of behaviour change organized into 14 domains [21]. The domains that are thought to be relevant to health workers’ change in behaviour include: knowledge; skills; memory, attention and decision processes; behavioural regulation; social/professional role and identity; beliefs about capabilities; optimism; beliefs about consequences; intentions; goals; reinforcement; emotion; environmental context and resources; and social influences [28, 35]. The TDF therefore provides a theoretical basis for implementation research, to aid understanding of which interventions are likely to work and why. Behaviour Change Techniques are the active, observable and replicable components that make up an intervention. COM-B/BCW have been used successfully for behavioural analysis and to design interventions in both health and non-health-related fields [26, 36–55], but to our knowledge so far has been used in only one study of TB on contact tracing in a low-resource setting, which enabled them to successfully identify barriers and facilitators to tailor interventions to improve contact investigation in Kampala [26].
Data Collection (Stage 1: Understanding the behaviour)
We used a mixed-methods strategy (Additional File 1) to collect empirical data to identify challenges in case detection of TB in children to enable behavioural analysis. For the quantitative arm, we analysed national TB programme data as well as data from children admitted to 13 county hospitals in Kenya to describe the burden of childhood TB and diagnostic practices and these have been reported elsewhere [6, 11]. Our study results show at national level, there is under-detection of TB in children and underuse of available TB diagnostic tests. At hospital level, we found more than half of all paediatric admissions in Kenyan county hospitals had signs and symptoms suggestive of TB, but in most, TB was not considered as a differential diagnosis. Only 1% of these children meeting criteria for diagnostic testing had an Xpert® MTB/RIF assay performed, which was available in all the hospitals.
In the qualitative arm, to understand the challenges in recognising and testing for TB in admitted children we analysed data from: i) semi-structured interviews, small-group discussions and key informant interviews with front line health workers and mid-level managers; ii) observations of TB trainings, sensitisation meetings, policy meetings, and hospital practices, and iii) desk review of guidelines, job aides and policy documents, which have been reported elsewhere [31]. We used the COM-B framework to interpret emerging themes. At individual level, we found that knowledge, skill, competence and experience, as well as beliefs and fears impacted on capability (physical & psychological) as well as motivation (reflective) to think of TB as a differential diagnosis in children and use diagnostic tests. Hospital level influences included hospital norms, processes & patient flows and resources which affected how individual health workers attempted to diagnose TB in children by impacting on their capability (physical & psychological), motivation (reflective & automatic) and opportunity (physical & social). At the wider system level, community practices & beliefs, and implementation of TB programme directives impacted some of the decisions that health workers made through capability (psychological), motivation (reflective & automatic) and opportunity (physical).
Behavioural Analysis and intervention design: Identifying intervention options, content and implementation options (Stage 2 & 3)
We used an iterative process, going back and forth from the quantitative and qualitative empiric data to reviewing literature, and applying the BCW guide [27]. The key questions reflected on included: i) what is the problem we are trying to solve; ii) what behaviours are we trying to change and in what way; iii) what will it take to bring about desired change; iv) what types of interventions are likely to bring about desired change; v) what should be the specific intervention content and how should this be implemented?
The empiric data helped identify the gaps in case detection of TB in children and use of diagnostic tests. We used COM-B and TDF to map out these gaps in behavioural terms. Depending on where they were situated on the BCW, we linked these to evidence-based intervention functions like education, persuasion, environmental restructuring and these were in turn linked to policy categories.
We used the experience of the research team including implementation scientists, epidemiologists, social scientists, clinicians and clinician educators, together with feedback from clinical colleagues to select potential interventions (see Table 1). We focused on those behaviour change techniques and modes of delivery that would yield results at low cost and that could most easily be taken up by the National TB programme to support hospital teams. We presented findings from our data to key paediatric TB stakeholders and had informal discussions with them during technical working group meetings to gain their ideas on what could work.
Table 1
Linking gaps in empiric data for behavioural analysis to intervention design (Stages 1 & 2)
Summary of gaps identified in empiric data | COM-B | TDF constructs linked to COM-B | Relevance of the theoretical domain | Proposed intervention function |
Under-detection of TB in children, 60–70% thought to be missed (QUAN) Nearly 60% of all paediatric admissions met guideline-criteria for suspected TB but < 3% got a diagnosis (QUAN) | Capability-psychological | Knowledge Behavioural regulation | Awareness of steps in diagnosing TB in children; of the available tests. Do they know what they should do and when and why? Self-monitoring; how to break a habit e.g. missed diagnosis. Anything in place to prompt them to make a diagnosis and to self-monitor? | Training: Imparting skills on how to correctly diagnose TB in children Modelling: Providing an example for people to aspire/imitate e.g. via champions/clinical leaders Persuasion: Using communication to stimulate action e.g. via audit & feedback |
Some reported that they did consider a TB differential diagnosis but sometimes forgot to document (QUAL) Some reported they do tests but forgot to document (QUAL) | Capability-psychological | Memory attention and decision processes Behavioural regulation | Ability to retain information, to consistently remember to document what is done Self-monitoring; how to break a habit e.g. failure to document. Anything in place to prompt them to always document? | Environmental restructuring: Changing the physical context e.g. availability of record forms for better documentation, job aides Persuasion: Using communication to induce positive or negative feelings or stimulate action e.g. via audit & feedback; shared goals with peers |
Some health workers fear/are reluctant to make a diagnosis of TB in children sometimes due to stigma in caregivers of TB-HIV association (QUAL) | Capability-psychological Motivation-automatic | Knowledge Reinforcement Emotion | Awareness of steps in diagnosing TB in children; of the available tests. Do they know what they should do and when and why? Anything to motivate or demotivate them? Does it evoke an emotional response e.g. some got uncomfortable when babies cried during specimen collection; some were reprimanded harshly by caregivers | Education: Increasing knowledge or understanding of TB in children Persuasion: Building communication skills to better counsel families Modelling: by the champions to demonstrate how best to de-stigmatise |
Underutilisation of TB diagnostic tests, 1% get Xpert done (QUAN) Health workers generally seem to have a challenge in collecting specimen for children (QUAL) | Capability-psychological Capability-physical Motivation-reflective Motivation-automatic | Knowledge Physical skills Beliefs about capability Reinforcement | Awareness of steps in diagnosing TB in children; of the available tests. Do they know what they should do, when and why? Are they physically able/proficient in diagnosing TB; collecting specimen; using diagnostic tests? Acquired through practice Are they confident diagnosing TB in children; collecting specimen? How difficult or easy? Increasing likelihood of TB tests being used appropriately | Training: Imparting skills to use available diagnostic tests and specimen collection Modelling: Champions/clinical leaders demonstrating correct procedures Environmental restructuring: identifying who is responsible for ensuring TB tests get done; job aides to serve as reminders of procedures |
Health workers report consistently negative Xpert test results (QUAL) | Capability-psychological Motivation-reflective | Knowledge Beliefs about consequences | Do they know how to respond to negative test results? How and when to make a clinical diagnosis? Do they believe doing it or not makes a difference? | Education: increasing understanding on making a clinical diagnosis and the epidemiology and natural course of TB in children Persuasion: communication to pass on the value of TB tests |
Some facilities had good teamwork and mentorship that helped model the correct way to diagnose TB in children (QUAL) | Opportunity-social Motivation-reflective | Social/professional role & identity Optimism | Do they think it is part of their job e.g. to collect specimen (senior doctors struggled) Do they think it’s something that can be done? How confident are they of this? | Modelling and social environment restructuring: Providing an example for people to aspire/imitate and encouraging teamwork Persuasion: communication to pass on the value of diagnosing TB in children |
Most facilities had long and unclear processes that contributed to TB being missed in children (QUAL) Some reported frequent stock-outs of some reagents and XPert cartridges (QUAL) | Opportunity-physical | Environmental context & resources | Organisational processes and patient flows; resources like job aides, PPE, reagents. Aspects of the environment that influence whether or not they diagnose TB in children | Environmental restructuring: Changing the physical context to ensure better work flows and availability of equipment, reagents |
Lack of skilled human resource to interpret some test results like Chest X-rays (QUAL) | Opportunity-physical Capability-psychological | Environmental context & resources Knowledge | Aspects of the environment that influence whether or not they diagnose TB in children Awareness of steps in diagnosing TB in children; of the available tests. How to make a clinical diagnosis? | Environmental restructuring: e.g. job aides to guide clinical diagnosis; remote decision-support for X-ray interpretation Training: Imparting skills of reading X-rays looking for TB-specific features; making a clinical diagnosis |
Some policies and directives including selection of participants for training disadvantaged front-line health workers (QUAL) | Opportunity-physical Motivation-automatic | Environmental context & resources Reinforcement | Aspects of the environment that influence whether or not they diagnose TB in children Anything to motivate or demotivate? (Lack of training was a demotivator) | Education: increasing policy makers’ understanding of the need of rethinking how TB training is done Persuasion: Using communication to stimulate action e.g. feedback to policy makers on the impact of training |
TB programme policy of doing quarterly audits and supervisory visits helped (QUAL) | Motivation-reflective | Intentions Goals | Feedback to enable health workers to make a conscious decision to improve case detection Visualise what they want to achieve | Persuasion: Using communication to stimulate action e.g. via audit & feedback |
Using information gathered from our empirical data, literature on interventions likely to be successful, [56, 57], our understanding of the context and taking the perspective of what would be feasible for hospital managers and NTP officers to implement, we came up with a list of possible interventions to address the gaps in diagnosing TB in children. We then further selected options linked to the predicted mechanism of action of change according to the TDF constructs and used the APEASE criteria to finally rationalise in terms of affordability, practicability, effectiveness, acceptability, safety and equity [27]. Table 1 summarises the process of linking the gaps in empiric data through the major behaviour change wheel design steps.
Relevant aspects of The Standard for Reporting Implementation Studies (STaRI) tool [58] were used to help ensure key elements needed when developing and evaluating implementation strategies have been covered to enhance adoption and sustainability of effective interventions. A summary of the completed checklist is included in Additional File 2.
[1] APEASE Criteria
A- Affordability
P- Practicability (can be delivered as designed through the means intended to target population)
E- Effectiveness and cost-effectiveness
A- Acceptability (judged to be appropriate by relevant stakeholders)
S- Side-effects/Safety- minimal unintended consequences
E- Equity (reduces or increases disparities in standard of living or wellbeing)