Table 1. Quality assessment criteria used for cross-sectional studies through a modified version of Newcastle-Ottawa Scale for case-control studies.
| CASE-CONTROL STUDIES AND THEIR ASSESSMENT RATINGS | |||||||||||||
| Dynesen et al., 2008 [21] | Ohrn et al., 1999 [35] | Järvinen et al., 1991 [28] | Johansson et al., 2012 [4] | Emodi-Perlman et al., 2008 [22] | Rytömaa et al., 1998 [20] | Robb et al., 1995 [19] | Milosevic & Slade, 1989 [32] | Jones & Cleaton- Jones, 1989 [31] | Howat et al., 1990 [18] | Sivolella et al., 2000 [33] | Greenwood et al., 1988 [30] | Touyz et al., 1993 [34] | |
| SAMPLE SELECTION CRITERIA | |||||||||||||
| 1) Diagnosis of eating disorders a) Clinical examination or medical record with validated instrument, or referred from a hospital ★ b)Without clinical examination, or based on self-reports c) No description | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ |
| 2) Representativeness and selection of the patients suffering from eating disorders (cases) a) Patients with eating disorders selected from a defined catchment area, in a defined hospital or clinic, health maintenance organization, communities or random sample, sample calculation ★ b) Potential for selection biases or not satisfying requirements in part (a) c) No description | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | c |
| 3) Selection of participants without eating disorders (controls) a) Participants selected from a defined catchment area, communities or random sample, sample calculation ★ b) Not satisfying requirements in part (a) c) No description | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | c | ★ | c |
| 4) Definition of controls a) no history of disease (eating disorder) ★ b) no description of source/ not stated if the patient was healthy/ self-report of eating disorders | ★ | ★ | ★ | ★ | ★ | ★ | ★ | b | ★ | ★ | ★ | b | b |
| COMPARABILITY OF CASES/ CONTROLS ON THE BASIS OF THE DESIGN OR ANALYSIS | |||||||||||||
| 1) Control for confounders a) The exposure of interest (tooth erosion) is adjusted for the one confounder ★ b) The exposure of interest (tooth erosion) is adjusted for two or more confounders ★★ c) No description related to the adjustment analysis for confounding factors | ★★ | ★★ | ★★ | ★ | c | c | c | c | c | c | c | c | c |
| EVALUATION OF TOOTH EROSION | |||||||||||||
| 1) Diagnosis of tooth erosion a) Clinical examination reporting the use of a tooth erosion index/ report of observer agreement – kappa ★ b) Satisfied requirements in part (a) and the examiner was blinded to case/ control status ★★ c) Based on self-reports or not satisfying requirements in part (a/b) d) No description | ★ | ★ | ★ | ★ | ★★ | ★★ | ★ | ★★ | ★ | c | c | c | c |
| 2) Same method of evaluation for cases and controls a) Yes ★ b) No | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ |
| 3) Response rate a) Rate of sample loss ≤20% ★ b) Rate of sample loss >20% c) Not stated | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ | ★ |
| SUMMARY SCORE (Stars) | 9/10 (high) | 9/10 (high) | 9/10 (high) | 8/10 (high) | 8/10 (high) | 8/10 (high) | 7/10 (high) | 7/10 (high) | 8/10 (high) | 6/10 (high) | 5/10 (low) | 5/10 (low) | 3/10 (low) |
Not all studies described all variables. Confounders were extracted and described as if they were evaluated in multivariate analysis.