Applied research design

Crossover design

Where participants switch between intervention and the control condition

Buffering

Where a treatment condition just prevents participants from getting worse, but no difference shows up

Ways of reducing bias in research

Random allocation to conditions
Blinding

Blinding

Concealing information from individuals as to who is in what condition
Can be single, double etc

Intention to treat

Including data from all participants regardless of dropout, adherence or changes
Last point of data collection is carried forward to make up for dropouts

Per protocol analysis

Includes only data from those who completed to the trial and adhered properly

Sensitivity analysis

Test the robustness of the final model by re-running your analyses differently (eg without outliers or missing data, controlling for other variables)

Triangulation mixed methods

Using methods with complementary strengths and weaknesses to fully inform a conclusion

Complementary mixed methods

Seeking a more complete understanding using methods that complement each other to explore multiple aspects of something

Development mixed methods

Using one method to inform the development of research involving a second method

Initiation mixed methods

Uncovering a paradox or divergence using complementary methods

Expansion mixed methods

Using different methods to expand the focus of an investigation to different phenomena

Sequential mixed method designs

Where one method happens after the other

Concurrent mixed method designs

Where both methods occur at the same time

Pilot (feasibility) studies

A small scale test of methods and procedures before larger scale studies

Service evaluations

Involving stakeholders to understand processes involved in a service and how they can be improved

Yardley et al. (2015) - person based approach

A systematic approach to incorporate views of user experience into interventions

Steps in a person-based approach

Developmental qualitative research at every stage of development
Identification of guiding principles to make sure the interventions addresses context-specific issues identified

PPIE

Patient and Public Involvement and EngagementW

What does PPIE do?

Involves the public in shaping research, where research carried out with/by members of the public rather than to/for them

Yardley et al. (2023) - ACE framework

Agile Coproduction Evaluation

What does the ACE frameowrk do?

Guides the production of intervention in times of dire need, uses quick research, co-production and evaluation

Advantage of the ACE framework

Helps to create effective intervention times of crisis by incorporating the groups at highest risk