SIZE UP A STUDY

Women's Health (Malaysia) - - HEALTHY DOSE -

It’s my job to read the stud­ies be­hind health head­lines. If you want to fol­low suit, try Pubmed, a free on­line archive of tons of med­i­cal jour­nals. But if you’re pressed for time, or the sci­en­tific lingo is just too dizzy­ing, make sure these de­tails are ID’D in the news re­port. The more boxes ticked, the bet­ter.

It was pub­lished in a sci­en­tific jour­nal. That means the study was peer­re­viewed or checked out by other ex­perts be­fore pub­li­ca­tion. Con­fer­ence ab­stracts (sum­maries of re­search pre­sented at med­i­cal meet­ings) or un­pub­lished pa­pers haven’t been through this same sci­en­tific wringer. The study was on hu­mans. Re­search done in test tubes or on an­i­mals can yield im­por­tant find­ings, but it only tells you what hap­pens un­der ideal lab con­di­tions or how a mouse (not a hu­man) re­acted to a drug or treat­ment.

And plenty of hu­mans. The more peo­ple in the study sam­ple, and the more of them who are like you (women of a sim­i­lar age and health sta­tus), the more ap­pli­ca­ble it will be to you. Re­search done on nine Scan­di­na­vian men who’ve had heart at­tacks won’t tell you much about your own car­dio­vas­cu­lar health. “One of my big­gest gripes is with weight-loss stud­ies,” says WH re­searcher Sara Green. “Many are done on obese peo­ple, who are likely to lose weight when they’re put on any kind of diet. If you’re just try­ing to drop a few ki­los, you’re not go­ing to see the same re­sults.”

It shows cau­sa­tion, not just cor­re­la­tion. Two things can be linked, but that doesn’t mean one causes the other. Ice cream sales and sun­burn risk both rise in sum­mer; that doesn’t mean that buy­ing more ice cream causes sun­burn.

Who­ever funded it doesn’t have an agenda. The study should dis­close who paid for it, and thereby any po­ten­tial con­flicts of in­ter­est. So if a study that found eat­ing rutaba­gas will give you bet­ter or­gasms was paid for by The Na­tional Rutabaga So­ci­ety, ques­tion it.

Two things can be linked, but that doesn’t mean one causes the other.

Newspapers in English

Newspapers from Malaysia

© PressReader. All rights reserved.