We thank you for the interest in the article and for the questions raised about the method.
As we detailed, our sample was relatively homogeneous: all GPs were located in the UK, 82% were female, and over half have specialist postgraduate dementia education. As we did not aim to correlate findings with participant characteristics, collecting further participant information was unnecessary. We acknowledge that using convenience sampling can attract participants interested in the research topic. We do not suggest within the article that findings are transferable to other populations or indeed outside the UK. As identified, further study in other populations would be required.
The scope of the study was discrete, defined to answer the identified research questions only, with interview questions limited to this focus. Unstructured questions about prevention may have yielded further insight, as you suggest, but this was not the approach for this study. We do not suggest we exhausted all views of GPs about dementia prevention, but those identified in the scope of the study.
We acknowledge there is no universal approach to thematic analysis (TA), or indeed identifying saturation. Braun and Clarke suggest that when themes are conceptualised as analytic inputs, using this type of coding approach to TA, possible or actual themes will ‘saturate’ early. We had a rigorous process of coding data, reviewing codes and themes within the research team and, while themes remained somewhat semantic, no new codes were identified in the final two interviews.
We appreciate your time in reading and commenting on the paper.
REFERENCE
- 1.Braun V, Clarke V. To saturate or not to saturate? Questioning data saturation as a useful concept for thematic analysis and sample-size rationales. Qual Res Sport Exerc Health. 2021;13(2):201–216. doi: 10.1080/2159676X.2019.1704846. [DOI] [Google Scholar]