Indunil Karunarathna Systematic reviews are more detailed and it's protocol (PRISMA) is registered before even writing. This is considered gold standard because it possess all the information about the topic and it pinpoints each and every aspect of the topic. All the articles are assessed in depth. It requires more than 2 researchers. Even when assessed by 2 separate researchers, the articles are again reassessed by a 3rd person to reduce biasness. First the abstracts are checked, then whole articles are selected based on authors' exclusion inclusion criteria.
On the contrary, narrative review as the name suggests provide description about the topic and does not require any kind of registration. This does not consume as much time as seen in systematic review. Researchers can write a narrative review if they want to describe any topic and later have a team put on a same topic for a systematic review.
There are many layers to the answer to this question but I'd suggest reading this paper (or just digesting the abstract) it compared 'expert recommendation' (for which we might read a selective narrative review) to available evidence in a meta-analysis. https://jamanetwork.com/journals/jama/article-abstract/398415
I would say that these days the term systematic review is much used and abused and in some cases, the emphasis on process and a preplanned protocol leads to uninformative reviews for some complex topics (see Article A Comparison of Results of Meta-analyses of Randomized Contr...