Table 2 Results of QUADAS-2 tool
From: Glucose prediction by analysis of exhaled metabolites – a systematic review
Study | Risk of Bias | Applicability Concerns | ||||||
|---|---|---|---|---|---|---|---|---|
Patient selection | Index test | Reference standard | Flow and timing | Patient selection | Index test | Reference standard | Comments | |
Righettoni (2013) [17] | ? | ? | × | ✓ | ✓ | ✓ | ✓ | Single measurements in morning and in afternoon make prediction of trend impossible. Possible verification bias because of incorrect reference standard. |
Storer (2011) [18] | ? | × | × | ✓ | × | × | ✓ | Single measurement makes prediction of trend impossible. Test review bias because reference standard is used for index test. Possible verification bias because of incorrect reference standard |
Minh (2011) [22] | ? | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | Clamp study design possibly lowers clinical relevance because of lack of generalizability. Test review bias because reference standard is used for index test. |
Turner (2009) [21] | ? | × | ✓ | ✓ | × | × | ✓ | Small sample size. Clamp study design possibly lowers clinical relevance because of lack of generalizability. Test review bias because reference standard is used for index test. |
Lee (2009) [20] | ? | × | ✓ | ✓ | ✓ | ✓ | ✓ | Small sample size. Clamp study design possibly lowers clinical relevance because of lack of generalizability. Test review bias because reference standard is used for index test. |
Fritsch (2008) [19] | ? | ✓ | × | ✓ | ✓ | × | ✓ | Small sample size. OGTT study design possibly lowers clinical relevance because of lack of generalizability. Test review bias because reference standard is used for index test. Possible verification bias because of incorrect reference standard. |
Novak (2007) [39] | ? | × | ✓ | ✓ | × | ✓ | ✓ | Clamp study design possibly lowers clinical relevance because of lack of generalizability. Test review bias because reference standard is used for index test. Possible reporting error, results of only one subject mentioned in detail. |
Galassetti (2005) [32] | ? | × | ? | ✓ | ✓ | ✓ | ✓ | Small sample size. OGTT study design possibly lowers clinical relevance because of lack of generalizability. Test review bias because reference standard is used for index test. Possible verification bias because of incorrect reference standard |
Paredi(1999) [27] | ? | ✓ | × | ✓ | ✓ | × | ✓ | Small sample size. OGTT study design possibly lowers clinical relevance because of lack of generalizability. Possible verification bias because of incorrect reference standard. |