The help phase: how do you develop the intervention? - Chapter 5

Once the variables causing the outcome variable have been identified and mapped in the process model, the intervention can be developed. An intervention is a means to change the causal variables and thus the outcome variables in the desired direction. An adequate intervention targets one or more causal variables in the process model. However, sometimes it’s not feasible or even necessary to target all variables in the model. So the first step in the help stage is to determine which causal variables will be targeted in the intervention. The estimated modifiability of the causal variables and the expected effects of interventions will direct this choice. Once these variables have been identified, an intervention that targets these variables can be developed. Then decisions must be made about how the target group will be reached and what the content of the intervention will be. The content largely depends on empirical evidence. The last step in this phase concerns the implementation process.

How do you prepare an intervention development?

The essence of the help phase is that interventions must focus on changing causal variables in the explanatory model. It is not always necessary, appropriate, or possible to target all the causal variables in the explanatory model. Therefore, the practitioner chooses the variables that are modifiable and have the greatest effect on the outcome variable. To do so, it’s convenient to put all the causal variables from the process model into a balance table.

Modifiability

Six questions can help exclude causal variables that are difficult to change:

1. Does the variable concern a stable personality trait?
2. Is the variable related to deeply held political or religious values?
3. Is the variable related to stable environmental conditions?
4. Is the variable related to some chronic medical or psychiatric condition?
5. Is the variable related to a lack of intelligence or literacy?
6. Does the variable have strong biological basis?

Effects

Not all variables in the process model have an equally strong impact on the outcome variable, practitioners should focus on the ones that have the strongest effect. Socio-psychological literature should be consulted to determine this. If no information is available in the literature, there are other ways to estimate effect size. These include looking at past experiences with similar situations and other empirical evidence.

The balance sheet

A balance sheet helps in making the decision on which variables will be targeted in the intervention. A psychologist will evaluate all variables form the process model for their adaptability and effect size. It looks like this:

Variables from the process model modifiability effect size

1. Perceived negative outcomes of child being overweight +++
2. Knowledge on weight reducing parental behaviours +++
3. Perceived positive outcomes of weight reducing parental behaviours +++
4. Perceived control over child’s behaviour +++
5. Experience with structured parenting 0 +

Note:

• Regarding modifiability: ++ = high modifiability; + = medium modifiability; 0 = low modifiability; - = not modifiable; +/0 = depends on another variable.
• Regarding the effect size: ++ = large effect; + = moderate effect; 0 = small effect; - = no effect; +/0 = depends on another variable.

Using this table, the psychologist can determine which factors need to be changed to change the outcome variable.

How do you develop the intervention?

Once the practitioner has decided which variables to target, the intervention can be developed. Two tasks can be distinguished in the development of an intervention:

1. Choosing the right channel: the way in which the target group is reached.
2. Selecting the appropriate method: the way in which the changes will be brought about.

The channel and method must consider the target group for intervention. It is important to note that the development of an intervention is usually a dynamic process: choices for the channel and method are made in combination.

The channel

The channel is the means through which people are reached and the intended changes will only take place if people are exposed to the channel. In the table below is an overview of various channels and their characteristics.

Channels have several features, and may vary from simple (for example, sticker or label) to complex (for example, community intervention), each communicating a distinct type of information (e.g., text, picture, video) and a different volume of information (e.g., one simple message vs. a complex set of arguments). Some channels communicate with high intensity (e.g., group therapy) and others with low intensity (e.g., information signs).

Channels also differ in potential reach of the target group and some channels will only have small effects on the individual level (a sticker), while others have large effects (the example of group therapy). Channels will also bring about different types of effects. Say, a label is appropriate to increase people’s knowledge while counselling is more appropriate to change complex and deeply rooted behaviours. The channel is chosen based on information about the target group and the relevant variables, methods, and strategies. The following should be considered when choosing the channel:

• Is the channel an effective way to reach the target group?
• Is exposure through this channel intensive enough to change the variable?
• Is the channel appropriate for the method and strategy?
• What is the impact on the population level of an intervention using this channel?

The method

Intervention methods also require consideration. Methods are often derived from theoretical frameworks. Selection of a method depends, first, on consideration of the balance table. For each variable, an intervention method must be chosen. Additionally, selection of a method depends on the extent to which the method ‘fits’ the variable one aims to change. Whereas some channels can motivate people to show the desired behaviour, they cannot teach them how to change it. In recent decades many methods for psychological interventions have been developed and tested. These are often called ‘behavioural change techniques’ (BCTs).

The following methods (or BCTs) are often used in psychological interventions:

• Providing arguments: arguments can persuade people to change their attitude towards a particular topic. This type of persuasion can use both rational and emotional arguments. Rational arguments may cause people to cognitively re-evaluate their position on a particular topic. Emotional arguments can persuade people by appealing to their emotions.
• Goal setting: setting concrete and specific goals is important. Goals direct people’s attention and effort, provide them with expectations, and give the opportunity for feedback on goal accomplishment, thereby regulating motivation. Goal setting changes behaviour by defining goals that people must reach in a given period of time. Sub-goals can help people work on small but important steps in reaching the superordinate goal.
• Fear communication: can be effective to motivate certain behaviours by showing the adverse consequences of the behaviour that aren’t desired. Fear communication is only effective (and ethically justified) when it is accompanied by explicit guidelines on how to avert the health threat.
• Modelling: refers to learning through the observation of others. Watching others behave and showing the consequences can teach people to people to perform a new behaviour.
• Enactive learning: the most effective way of learning a skill is to try to accomplish it yourself. This is called enactive learning. In interventions, people can be stimulated to practice a certain skill and evaluate it.
• Social comparison: information on how others are doing may affect one’s mood and well-being.
• Implementation intentions: these are intentions to perform a particular action in a specified situation. People are sometimes asked to formulate their implementation intentions which may increase the occurrence of desirable behaviours.
• Reward and punishment: generally, people repeat behaviours that are followed by a positive experience (reward), while avoiding a negative experience (punishment). Punishment for undesirable behaviours work best when accompanied by rewards for desired behaviours.
• Feedback: feedback on accomplishments is essential in behavioural change. people want to know how far they are in achieving their goal to stay motivated.

How do you translate channels and methods in practical applications?

The channels and methods described above are general. To effectively influence a specific target’s group attitude, emotion, or behaviour they must be translated into practical applications that fit with the target group and its characteristics, context, and culture. A practitioner often has many options and decisions to make when translating channels and methods into specific applications. In this context it may be helpful to summarize what the intervention should be about, by shortly describing the methods, channels, target groups, and variables to be changed, as identified based on the balance table.

Based on these global intervention descriptions, the practitioner can think of different applications that may be used to make up the intervention for this specific target group. This usually happens in two phases, a divergent and a convergent phase. In the divergent phase, the practitioner lists as many applications as possible. In the convergent phase, he or she critically evaluates these applications.

The divergent phase: generating applications

There are many ways to generate applications for interventions:

• Direct intervention association: ideas for strategies can be based on all kinds of sources, like what you see on the internet, what makes intuitive sense, what proves to be effective in the literature, or what people have experienced themselves.
• Direct method approach: this approach consists of looking at applications that have been used in similar situations.
• Debilitating strategies: this approach is to come up with applications that have undesired effects on the problem.  By generating ideas of what would probably have no effect or reverse effects, we can learn about what would have an effect, about the relevant dimensions of an effective intervention, and about ways to specify the channel and method.
• Interviews: interviewing people from the target group could generate additional ideas for practical applications.
• Insights from theory: this approach consists of looking at relevant social psychological theories.
• Insights from research: this approach consists of looking at relevant social psychological literature.

The convergent phase: choosing applications

The divergent phase often results in a laundry list of possible applications. Therefore, a limited number of applications need to be selected that, together, form the intervention. The choice for a particular application or combination of applications must have both a theoretical and empirical basis. First, the application should consider the conditions underling the theory. Second, it is preferable that the application is based on empirical evidence from either laboratory experiments or field studies. Ideally, evidence ought to be available for the combination of the channel and the method, the variable to be changed, and the target group. Sometimes, evidence for the effectiveness of a certain application isn’t there. In that case, especially when the costs of an intervention programme are high, we recommend that the effectiveness of an application or intervention (combination of applications) should first be tested through research.

How do you develop and pre-test the intervention?

Once applications have been selected, the intervention can be shaped. If applications concern visual or video materials, this will often be done in collaboration with a professional graphic designer or editor who is acquainted with the technological possibilities. Here are some rules of thumb for designing and shaping applications and the materials needed for these applications:

1. Be as specific as possible in the provision of information.
2. When an intervention has several channels and applications or sequential elements, all parts must be fine-tuned, and a protocol must be written.
3. If other professionals are involved, it should be clear how much influence they can have over the end-product.

Each planned application or intervention or part thereof must be pre-tested. The primary function of a pre-test is to improve the intervention and to avoid major flaws in the design. A pre-test may but doesn’t necessarily include a measure of the outcome variable. It primarily ensures that the target group will attend to the message as well as understand the message. The format of a pre-test usually includes exposing a few target group members to the preliminary intervention and assessing their reactions. This assessment can be done in different ways:

1. Interview: free way of asking questions of the target audience.
2. Quantitative assessment: people of the target group answer closed questions about the intervention in a questionnaire.
3. Recall: recall tasks assess which aspects of the intervention people from the target group remember after having been exposed to the intervention.
4. Observation people from the target group may be observed when they are exposed to the intervention.
5. Expert opinions: one may also ask the experts involved in bringing about the effects of the intervention.

Generally, participants seldom agree completely about an intervention in a pre-test. So the practitioner must not only look at pre-test data, but also consider theoretical aspects as well as empirical evidence that may be relevant. After revisions have been made, the improved intervention can be pre-tested again. The final version of the intervention can now be. Developed and distributed.

How do you pilot the intervention?

In the case of a large-scale, expensive intervention or an intervention that makes use of a strategy for which little research evidence has been found yet, the pre-test may be followed up by a pilot stud. Whereas a pre-test merely aims to ensure that the target group attends to and understands the intervention, a pilot study aims to assess the intervention’s effect on the outcome variable and the causal variables from the process model that are the focus of the intervention. Making sure the intervention works reduces the chance of enrolling an expensive large-scale intervention that later shows disappointing outcomes, and in doing so may prevent a waste of time and money.

Ideally, the practitioner finds that both the outcome variable and the causal variables from the process model that the intervention focuses on are changed in the desired direction.

How do you implement the intervention?

The implementation process has one major goal: to ascertain that the intervention is sued as intended. Implementation is not about the effects of the intervention but about positioning the intervention in such a way that it can have its effects. The core challenge of the implementation phase is that the extent to which the target group is exposed to the intervention depends on the people and organizations that are involved in the distribution of the intervention. So, while psychologists can create a perfect intervention program, it is highly dependent on the people who need to expose the interventions to the target audience. If this is not or hardly ever done, the effects will be very small or even non-existent.

The implementation process

The implementation process can be time consuming as communication and information exchange can take a long time. Roger’s (2010) model of diffusion describes how the implementation of new, large-scale interventions takes place in time. This process is referred to as the diffusion process. This model also helps decide when and what the practitioner can do to stimulate the process of diffusion. Rogers distinguishes four phases which will be illustrated here with an example of a mobile app for battered women to get professional help.

• Dissemination phase: the general practitioner (GP) becomes aware of the existence of the app and leaflet on domestic violence and discusses it with colleagues.
• Adoption phase: GPs become motivated to use the innovation and hand out the leaflet to patients who are suspected of being victims of domestic violence.
• Implementation phase: the doctor engages in the behaviour that will expose the target group to the intervention: he/she hands out the flyer to the right patients.
• Continuation phase: handing out the leaflet becomes normal practice.