MRC (UK) Complex Interventions

Hi everyone... I'd like to have a bit of clarification on something...

I have developed a mobile app to guide patients preparing for colonoscopy in an ambulatory setting in India. Is such a product considered to be a "complex" healthcare intervention according to the UK's medical research guidance framework?

(The app is intended to be used as a patient education and engagement software)

You need to be a member of Health Informatics Forum to add comments!

Join Health Informatics Forum

Email me when people reply –


  • Hi Rohit, interesting question!

    I would say if you are evaluating this, I would probably treat it as a complex intervention (most digital interventions are in my opinion). The MRC guidance says to look at these dimensions of complexity to determine how complex your intervention is (as opposed to a 'simple' intervention like a drug):

    • Number of and interactions between components within the experimental and control interventions
    • Number and difficulty of behaviours required by those delivering or receiving the intervention
    • Number of groups or organisational levels targeted by the intervention
    • Number and variability of outcomes
    • Degree of flexibility or tailoring of the intervention permitted


    You could also look at the recent NICE framework for what kind of level of evidence you need to demonstrate effectiveness for different types of digital health apps:
    • Thank you very much Chris for the guidance and information on complex interventions!

      Also, really appreciate your pointing to this Evidence Standards Framework from NICE. I think mine would be a tier 2 digital health technology because it fits into the "inform" and "communicate" functional classifications and produces measurable patient outcomes. Since it is not a potentially higher risk clinical intervention, I just need to develop all of the "minimum evidence standards" for tier 1 and tier 2 cumulatively into my app. And if I'm not mistaken, a pilot RCT should suffice for its clinical validation or peer review by a medical journal, right?

      By the way, I'd be interested to know why you consider a drug as a simple intervention? Is it in terms of their mechanism of action.. please clarify.
      • Hi Rohit, it depends what you mean by 'clinical validation'. For something like this there is a range of research that you could conduct to ensure it meets the needs of users and is acceptable in the context where it is to be used. If you want to show effectiveness in improving health outcomes, you would likely need a trial at some point. The WHO guidelines on monitoring and evaluation of digital health systems might be more relevant for an app like this - they go through the research you can conduct at various stages of development and implementation: We also have a video on these here:
        • Hi Chris, that is a very useful resource suggested by you!

          I will certainly peruse the 'WHO Monitoring & Evaluation (M&E) guide for DHIs' that you advised for my app. By the way, the M&E document seems different from the one you actually linked to: 'WHO recommendations on DHIs for health system strengthening', the guidelines that your e-Seminar reviews. Thank you for posting all of them.

          WHO M&E guide:

          My objective is to demonstrate efficacy in improving the quality of bowel preparation (as measured by colonoscopists using a validated scale like the Boston Bowel Preparation Scale).

          I had tried to conduct a randomized controlled pilot study in September 2019, however, my exclusion criteria were excluding participants on the basis of language, literacy, socioeconomic group, internet access, and digital literacy skills to suit my app's design and ended up not finding enough eligible subjects in a teaching hospital in India.

          I will go through the WHO M&E toolkit and hopefully come up with a practical approach.
          • Yes, sorry got my links a bit mixed up - both should be useful! My advice would be to take a mixed methods approach and start with co-design and qualitative work. RCTs, while the gold-standard, may be too difficult/expensive to do properly and result in inconclusive evidence if not powered properly. You may be better doing a simpler type of study such as a time-series analysis for the quantitative work depending on the level of evidence you need.
            • Yeah, I will keep that in mind. Thank you so much!
This reply was deleted.