Table C3Methodological quality of prospective cohort studies (PCS) and retrospective cohort studies (RCS) using Newcastle-Ottawa Quality Assessment Scale, by Key Question and design

Author, YearRepresentativeness of exposed cohortSelection of non-exposed cohortAscertainment of exposureOutcome of interest not present at start of studyComparability of cohorts (Study controls)Assessment of outcomeFollow up long enough for outcomes to occurAdequacy of cohort follow upTotal stars (quality rating)
known factors*additional factor
KQ2 – PCS
Solomon, 1996Selected group of usersSame community as exposed cohortStructured interviewYesNoNoRecord linkageYesSubjects lost unlikely to introduce bias6
(fair)
KQ2 – RCS
Chanprapaph, 2004Truly representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesComplete follow up7
(good)
KQ3 – PCS
Ardawi, 2000Truly representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesSubjects lost unlikely to introduce bias7
(good)
Lao, 2001Somewhat representativeSame community as exposed cohortSecure recordYesNoNoNo descriptionYesSubjects lost unlikely to introduce bias6
(fair)
Lapolla, 2007Truly representativeSame community as exposed cohortSecure recordYesYesYesNo descriptionYesComplete follow up8
(good)
Metzger/HAPO, 2008Truly representativeSame community as exposed cohortSecure recordYesYesYesIndependent blind assessmentYesSubjects lost unlikely to introduce bias9
(good)
Retnakaran, 2008Somewhat representativeSame community as exposed cohortSecure recordYesYesYesNo descriptionYesSubjects lost unlikely to introduce bias8
(good)
Sacks, 1995Somewhat representativeSame community as exposed cohortSecure recordYesYesYesNo descriptionYesSubjects lost unlikely to introduce bias8
(good)
Sermer, 1995
RCT
Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesSubjects lost unlikely to introduce bias9
(good)
Shirazian, 2008Somewhat representativeSame community as exposed cohortSecure recordYesYesYesNo descriptionYesSubjects lost unlikely to introduce bias8
(good)
KQ3 – RCS
Aberg, 2001Truly representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Adams, 1998Truly representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Berggren, 2011Truly representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesSubjects lost unlikely to introduce bias9
(good)
Berkus, 1995Somewhat representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesComplete follow up7
(good)
Biri, 2009Truly representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesComplete follow up7
(good)
Black, 2010Truly representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Bo, 2004Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Cheng, 2009Truly representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Chico, 2005Somewhat representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesComplete follow up7
(good)
Chou, 2010Somewhat representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesComplete follow up7
(good)
Cok, 2011Somewhat representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesComplete follow up7
(good)
Corrado, 2009Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesSubjects lost unlikely to introduce bias9
(good)
Hillier, 2007Truly representativeSame community as exposed cohortSecure recordYesYesYesIndependent blind assessmentYesSubjects lost unlikely to introduce bias9
(good)
Jensen, 2002Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesSubjects lost unlikely to introduce bias9
(good)
Kim, 2002Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesSubjects lost unlikely to introduce bias8
(good)
Kwik, 2007Truly representativeSame community as exposed cohort cohortSecure recordYesNoNoRecord linkageYesSubjects lost unlikely to introduce bias7
(good)
Langer, 2005Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Lao, 2003Somewhat representativeSame community as exposed cohortSecure recordYesYesNoRecord linkageYesSubjects lost unlikely to introduce bias8
(good)
Lapolla, 2011Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Morikawa, 2010Somewhat representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesComplete follow up7
(good)
Nord, 1995Somewhat representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesSubjects lost unlikely to introduce bias7
(good)
Pennison, 2001Truly representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Ricart, 2005Truly representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesComplete follow up7
(good)
Rust, 1996Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesSubjects lost unlikely to introduce bias9
(good)
Schwartz, 1999Truly representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesFollow up rate <75% and no description of those lost6
(fair)
Stamilio, 2004Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Tan, 2008Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Vambergue, 2000Truly representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
Yang, 2002Somewhat representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
KQ4/5 – PCS
Malcolm, 2006Somewhat representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesFollow up rate <75% and no description of those lost6
(fair)
KQ4/5 – RCS
Adams, 1998Truly representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesSubjects lost unlikely to introduce bias9
(good)
Bonomo, 1997Selected group of usersSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up8
(good)
Fassett, 2007Somewhat representativeSame community as exposed cohortSecure recordYesNoNoRecord linkageYesComplete follow up7
(good)
Langer, 2005Truly representativeSame community as exposed cohortSecure recordYesYesYesRecord linkageYesComplete follow up9
(good)
*

Controls for known factors: age, race, BMI, history of GDM, family history of DM

Quality rating based on EPC Methods Guide (good, fair, poor): total scores of 7-9 were considered good, 4-6 fair, and 0-3 poor.

BMI = body mass index; GDM = gestational diabetes mellitus; DM = diabetes mellitus; PCS = prospective cohort study; RCS = retrospective cohort study

From: Appendix C, Methodological Quality of Included Studies

Cover of Screening and Diagnosing Gestational Diabetes Mellitus
Screening and Diagnosing Gestational Diabetes Mellitus.
Evidence Reports/Technology Assessments, No. 210.
Hartling L, Dryden DM, Guthrie A, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.