1)     The above strategy yielded 151  documents deemed “potentially relevant”.  To be rated as relevant, the article had to
measure both ECC and breastfeeding as  well as meet other criteria outlined in Table I. Two readers independently rated
20 randomly selected articles for relevance and obtained a high level of agreement  (Kappa of 0.9). Remaining articles were then divided equally among two readers and were rated independently. Of the 151 articles that were retrieved, 28 were judged
to be relevant.
      2)     The articles (n=28) were then rated for their quality or validity (Table I). Validity (quality) considers the extent to which the design and conduct of the study are likely to bias the results.
       
      3.   In this review, validity criteria were modified to exclude criteria  related to controlled trials including blinding
of researchers. The tool included appraisals of study design, confounders, and data collection methods as per the
Cochrane systematic review protocol
    
  ปลขออนุญาติเเทกหลายห้องครับ เเทกผิดขออภัยด้วยครับ