Age- and sex-matching in a 1:n ratio from a pre-existing cohort

Hi all,

I'm quite new to SPSS and am using it for some medical research. I have an existing cohort of ~20,000 participants, of which ~2,000 have a cancer diagnosis. I would like to assign controls from the remaining ~18,000 patients in an age- and sex-matched fashion. My issue is that I can't work out how to do this in a 1:n ratio (given the size of the cohort, I'd like to use at least a 1:4 ratio).

I've been playing around with the Case-Control Matching function on SPSS 23 but haven't had any luck in getting the 1:4 ratio. I've also had a look at various syntax suggested on some threads online but none of them quite seem to exactly fit my problem.

Any guidance would be hugely appreciated!



Less is more. Stay pure. Stay poor.
Why do you want to match them, what purpose will it serve? What is your study question and what analyses do you plan to conduct to answer your question.
Without revealing too much, we are looking at the impact of cancer on a number of measures including ischaemic heart disease and cerebrovascular disease. The outcomes we are measuring are heavily linked with age. Therefore, if the control cohort is considerably younger/older, we are going to have vastly different results. The same goes for sex - ischaemic heart disease for example generally occurs earlier in male patients.

We will be performing some sort of survival analysis (likely cox proportional hazard models) as we have some nice detailed data on events and event dates.