Our systematic review was not restricted to any language or geographic region. In the CPRD, patients of all ethnicities are registered and were included in the prognostic modelling. Consequently, the results of health economic modelling are derived from and relate to people from all ethnicities in the UK population. In the acceptability study, the initial invitation letter, participant information sheet and interview guide were all developed with PPI input. We attempted to recruit non-White populations to the interview study but were unsuccessful. Further work is needed to ascertain the reasons underlying this and it is possible that this could be related to the initial invitation letter and participant information sheet only being in English language.
Our systematic review did not observe any racial risk factors for immune-suppressing drug toxicity, and we did not include ethnicity as a prognostic factor. This approach is also in line with removing race from the eGFR calculation so as not to disadvantage Blacks. As we used data from the CPRD and its characteristics/demographics are like the UK population, our research does not suffer from non-inclusivity bias in prognostic modelling and health economic evaluation. There is no published evidence that the target of research (disease, disorder, or intervention) impacts or affects different groups in a different way based on our systematic review and there is no evidence to suggest that the existing preventative, screening or diagnostic strategies and treatments impact or affect different groups in a different way. The results of systematic review, prognostic modelling and health economic modelling are generalisable and transferable to the entire UK population as this was a broad review and included patients from the entire UK. However, the interview study only recruited White people and has limited generalisability to non-White populations. Nevertheless, it recruited people from across the UK, and from both sexes and all age groups. Further research is needed to find out if the intervention is feasible, acceptable, equitably applicable to all populations regardless of any protected characteristics. We advertised the study through patient networks and recruited from three NHS hospitals from a range of diverse patient populations. We could have engaged with community advisory boards and navigators, community health workers to improve the diversity of participants in the interview study. We kept burden of participation in study 3 to a minimum by just having one remote interview study and gave the participants option to consent over the call which could be done on telephone or by video conference depending on participant preference. As fewer than 15% patients in CPRD Gold or Aurum are non-White, it was not possible to disaggregate the results of prognostic modelling on ethnicity due to lack of power. Similarly, data on religion, sexual orientation etc. are not available tin the CPRD.
Our research team includes people from Black, Asian and White ethnicities; both male and female gender; and a good spread of age from your early career researchers to senior researchers. Junior members of the team were given the opportunity to be first authors and joint first authors on the manuscript and to lead on respective WPs.