Abstract
Background
Randomised controlled trials are often criticised for excluding older people with multiple long-term conditions. This study used individual participant data (IPD) for 25 trials of sodium glucose co-transporter-2 inhibitors (SGLT2i) to compare baseline characteristics, comorbidities, and event rates between trial participants and community SGLT2i-treated people.
Methods
Trials were identified through a systematic review with subsequent application for IPD. Community SGLT2i-treated people in routine care were identified from SAIL databank. For each trial, we applied the eligibility criteria to the community SGLT2i-treated populations. We then assessed the proportion eligible/ineligible for each trial; compared age, sex and comorbidities between trial participants and those eligible/ineligible in routine care; compared rates of serious adverse events in the trials to the expected rate in community SGLT2i-treated participants; and compared the rate of major adverse cardiovascular events (MACE) and mortality between trial and community participants.
Results
Mean age was similar between trial participants and eligible community-treated participants. The number of comorbidities was consistently lower in trial populations compared people treated in the community. However, compared with other trial populations, participants in the large cardiovascular outcome trials (CANVAS, CANVAS-R, CREDENCE and EMPA-REG) had higher levels of comorbidity and rates of serious adverse event, MACE and mortality that were broadly similar to the expected rate in the community. For the remaining trials, the serious adverse event rate was lower in the trials than the expected rate based on community SGLT2i-treated participants.
Conclusion
While people with comorbidity are under-represented compared to routine care populations in most trials, the large cardiovascular outcome trials were more representative of SGLT2i-treated patients with similar rates of serious adverse events. While our findings support calls for caution regarding trial representativeness, the criticism that trials are not representative does not apply equally to all trials, some of which closely reflect current drug use.
Comments
Cardiovascular outcomes
As far as I know, CANVAS and EMPA-REG were landmark studies in forming current guidelines. Especially in the context of primary care, many people are now eligible for them if they are over 50 with a QRISK2 of greater than 10%. I have seen criticisms of the use of SGLT2 inhibitors in older people. Does your research show that for cardiovascular outcomes, these trials are representative for an older population?
Sort of!!
Thanks for the question… really important point.
It shows that they are representative of people currently treated. I think this is important caveat. There were few people aged >80 in the trials, but in the study period there were also relatively few people aged over 80 being treated with SGLT2’s. As treatment expands to more diverse patient groups (including much older people, for example) these trials will become less representative against that new benchmark.
But, on the other hand, they do appear more representative than that’s are sometimes criticised for being.
My suspicion is that they will still have excluded many with severe frailty or functional impairment… but unfortunately these domains are not measured even with access to the individual level data.
This is a really insightful…
This is a really insightful piece of work. It tackles an important question about how well clinical trials reflect the real patients we see every day, especially older adults with multiple conditions. I really like how clearly the comparison between trial populations and community-treated individuals is laid out. The finding that the large cardiovascular outcome trials are actually quite representative is particularly interesting and adds a lot of nuance to the usual criticism that trials exclude complex patients. Overall, it’s a thoughtful and well-structured study that adds real value to our understanding of trial generalisability.
Thanks!
Thank you for these comments. Very kind!