You may want to check out the Newcastle Ottawa scale. It is not perfect, but it works just fine. It is a flexible scale adapted for different types of study designs, evaluating a variety of domains. It may not extensively evaluate epidemiologic biases, though http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp

Dear Basem

The question you asked is highly dependent on the type of the included studies.

For diagnostic, therapeutic, prognostic and virtually any kind of studies we have numerous checklists.

What I can recommend is to use the checklists not the scales (which give a number for quality of each study).

I personally use Oxford center for evidence based medicine checklists

http://www.cebm.net/?o=1040

Ramin

Thanks Ramin for sharing this link.

Hi

A lot of meta-analysis authors use the Jadad scale to assess the quality of RCTs- easy to use and well recognized

Hope this helps

Dear Basem,

I think you should explain more regarding the nature of your systematic review. If there is no case or control group how could the harm be quantified?

Hi Ramin,

Thanks for your response. My study is looking to use meta-analyses to identify predictors of lost-time injury the workplace (e.g. shift work, heavy lifting etc.) Thus, most of my studies are observational in nature and not RCT or experimental in design.

Dear Basem,

Although you mentioned that the studies included in your meta-analysis do not have control group, I think inherently you have control groups in your included studies. For example in a study evaluated the effect of heavy lifting on lost time injury, some are in the group of heavy lifting tasks and others are not. This makes your articles case control studies.

If you consider the main effect size you have in the studies, the problem would be soleved more easily. Are you pooling Odds Ratios or Risk Ratios, or Risk difference? If it is the case, then you definitely have two groups and you need to use quality assessment checklists of observational studies.

The attached files may help.

Best wishes,

Ramin

You may also check the program revman of Cochrane, which I found very capable. You can find it here: http://ims.cochrane.org/revman/download

Hi Ramin, this is exactly what I was looking for. Many thanks!

And yes, I am pooling RR

Agree with Ajibola, Newcastle-Ottawa scale has several limitations, but is easy and straightforward for study quality rating.

You may find various extension for CONSORT checklist at this address: http://www.equator-network.org/reporting-guidelines/consort

however, you may find scales such as Jadad or Newcastle-Ottawa easier to use and more feasible than CONSORT or its extensions. I also recommend finding a proper risk of bias assessment tool.