Systematic review vs. meta-analysis:one is a process, the other is a tool

A systematic review (SR) is a type of literature review. It refers to the entire disciplined process of specifying your question, search strategy, selection criteria, and appraisal methods in advance, then carrying out the review according to that plan and documenting every step.

Notice that nothing in that definition involves statistics. An SR is a methodology for how the evidence was gathered and how studies were selected.

A meta-analysis (MA) is a method for statistically combining the results of multiple studies. It weights each study's effect size, pools them, and typically presents the result as a forest plot.

Notice what this definition leaves out: it says nothing about how the studies were found. An MA is a statistical technique for how to combine the data once you have it.

Because one is a process and the other is a tool, they can be combined in different ways.

First, an SR without an MA: the studies were gathered systematically, but they were too heterogeneous to pool, so the findings were synthesized narratively—still a perfectly valid systematic review. Second, an SR plus an MA: the most trustworthy form, where systematically gathered studies are also statistically pooled. Third, an MA without an SR: studies collected in a non-systematic way are pooled anyway—a pattern that calls for caution, as the next section explains.

A meta-analysis can look statistically rigorous, but if the studies fed into it were selected with bias, the pooled result inherits that bias. Cherry-pick the studies that suit your argument, average them, and you will simply arrive at the conclusion you started with.

As a reader, then, the right order is to check whether the studies were gathered systematically before you check whether a meta-analysis was performed. Even when "meta-analysis" appears in the title, go to the Methods section and read how the search and selection were done.

When you pick up an SR/MA paper, working through it in this order will keep you from getting lost.

First: what is the question (the PICO)? Second: were the search and selection systematic (Methods)? Third: were the results pooled, or only described? Fourth: if they were pooled, how much heterogeneity was there? An AI summary is useful for previewing these four points, but for the second one—whether the process was systematic—go to the Methods section of the original paper. A summary can compress away exactly the details that tell you whether the process was rigorous.

The same order applies when you run a review yourself. First design the question, search, and selection as a systematic reviewer; then look at what you've gathered and decide whether pooling (a meta-analysis) is even feasible.

The healthy sequence is not "collect studies because I want to run a meta-analysis" but "collect studies systematically, then discover that pooling looks possible." Consult a statistician or your supervisor before deciding to pool. Our separate articles on systematic review methods and meta-analysis methods walk through the full process.

Whether you stop at an SR or go on to an MA, the underlying work is identical: read each paper, verify it, record your judgments, and stay able to return to the original text.

Paperfy can serve as scaffolding for that shared groundwork. For every paper, the AI summary, extracted figures, and original PDF sit on a single page, and you can edit the summary and leave notes. Whichever path your review takes, your reading record accumulates in one place.