Table 2.
Quality appraisal of retained publications
| First author (year) | # CASP criteria satisfied | # unclear criteria | # CASP criteria unmet | Proportion of satisfied criteria n (%) | Assessment | Main unmet criteria |
|---|---|---|---|---|---|---|
| Adams (2013) [54] | 9 | 1 | 0 | 9/10 (90 %) | Good | Relationship between researcher and participants not mentioned |
| Almerich-Silla (2008) [28] | 6 | 2 | 4 | 6/10 (60 %) | Good | Reliability and validity of questionnaire not mentioned No Confidence Interval calculated |
| Angelillo (1996) [43] | 8 | 2 | 2 | 8/12 (66.7 %) | Good | No Confidence Interval calculated |
| Blackwell (2002) [40] | 8 | 1 | 3 | 8/12 (66.7 %) | Good | Statistical significance of results not assessed No Confidence Interval calculated |
| Cote (2004) [29] | 9 | 1 | 9/10 (90 %) | Good | n/a | |
| Davidson (2006) [5] | 8 | 2 | 2 | 8/12 (66.7 %) | Good | Statistical significance of results not assessed No Confidence Interval calculated |
| Davidson (2007) [15] | Review article Excluded | |||||
| el Barbari (1993) [30] | 8 | 2 | 2 | 8/12 (66.7 %) | Good | No Confidence Interval calculated |
| Fox (2010) [52] | Not satisfactory | Screening criteria not satisfied | ||||
| Geltman (2014) [7] | 6 | 3 | 3 | 6/12 (50 %) | Satisfactory | Selection of participants not clearly described. No Confidence Interval calculated |
| Ghiabi (2014) [3] | 8 | 2 | 2 | 8/12 (66.7 %) | Good | No Confidence Interval calculated |
| Gibbs (2014) [27] | Screening criteria for MM not met | |||||
| Gunaratnam (2013) [35] | 6 | 3 | 3 | 6/12 (50.0 %) | Satisfactory | Selection of participants not clearly described. Statistical significance of |
| results not assessed No Confidence Interval calculated | ||||||
| Hayes (1998) [32] | 6 | 3 | 3 | 6/12 (50 %) | Satisfactory | Statistical significance of results not assessed No Confidence Interval calculated |
| Hjern (1991) [31] | 8 | 2 | 2 | 8/12 (66.7 %) | Good | No Confidence Interval calculated |
| Honkala (1992) [48] | 8 | 2 | 2 | 8/12 (66.7 %) | Good | No Confidence Interval calculated |
| King (2012) [36] | 9 | 1 | 2 | 9/12 (75 %) | Good | No Confidence Interval calculated |
| Lamb (2009) [9] | 8 | 2 | 8/10 (80 %) | Good | Relationship between researcher and participants not mentioned | |
| Mahajan (2013) [4] | 8 | 2 | 2 | 8/12 (66.7) | Good | No Confidence Interval calculated |
| McNabb (1992) [47] | 7 | 2 | 3 | 7/12 (58 %) | Satisfactory | Statistical significance of results not assessed No Confidence Interval calculated |
| Mickenautsch (1999) [50] | 8 | 1 | 3 | 8/12 (66.7 %) | Good | Statistical significance of results not assessed No Confidence Interval calculated |
| Nair (1996) [51] | 8 | 1 | 3 | 8/12 (66.7 %) | Good | Statistical significance of results not assessed No Confidence Interval calculated |
| Nicol (2014) [53] | 10 | 10/10 (100 %) | Good | n/a | ||
| Ogunbodede (2000) [34] | Field Report Excluded | |||||
| Okunseri (2008) [39] | 9 | 2 | 1 | 9/12 (75 %) | Good | n/a |
| Prowse (2014) [26] | 8 | 2 | 8/10 (80 %) | Good | Relationship between researcher and participants not mentioned | |
| Puertes-Fernandez (2011) [37] | 8 | 2 | 2 | 8/12 (66.7 %) | Good | Response rate of participants not mentioned |
| Redwood-Campbell (2008) [13] | 8 | 1 | 3 | 8/12 (66.7 %) | Good | No Confidence Interval calculated |
| Riggs (2014) [2] | Did not satisfy screening criteria for MM studies | |||||
| Roucka (2011) [16] | 6 | 4 | 2 | 6/12 (50.0 %) | Satisfactory | Potential for bias in sample selection Statistical significance of results not assessed No Confidence Interval calculated |
| Singh (2008) [14] | 8 | 1 | 3 | 8/12 (66.7 %) | Good | Statistical significance of results not assessed No Confidence Interval calculated |
| Smith (2000) [41] | 6 | 2 | 4 | 6/12 (50.0 %) | Satisfactory | Statistical significance of results not assessed No Confidence Interval calculated |
| Smith (1998) [42] | 7 | 3 | 2 | 7/12 (58 %) | Satisfactory | Statistical significance of results not assessed No Confidence Interval calculated |
| Todd (1990) [33] | 8 | 2 | 2 | 8/12 (66.7 %) | Good | No Confidence Interval calculated |
| Umamaheswaran-Mahara (2010) [38] | 9 | 1 | 2 | 9/12 (75.0 %) | Good | No Confidence Interval calculated |
| Willis (2005) [56] | 10 | 10/10 (100 %) | Good | n/a | ||
| Willis (2008) [55] | 10 | 10/10 (100 %) | Good | n/a | ||
| Willis (2011) [8] | 8 | 1 | 3 | 8/12 (66.7 %) | Good | Statistical significance of results not assessed No Confidence Interval calculated |
| Wolf (1996) [6] | 8 | 2 | 2 | 8/12 (66.7 %) | Satisfactory | No Confidence Interval calculated |
| Zimmerman (1993) [45, 46, 57] | 9 | 2 | 1 | 9/12 (75 %) | Good | n/a |
| Zimmerman (1990) [49] | 9 | 1 | 2 | 9/12 (75 %) | Good | n/a |
| Zimmerman (1993a) [45] | 8 | 3 | 1 | 8/12 (66.7 %) | Good | n/a |
| Zimmerman (1993b) [46] | 9 | 2 | 1 | 9/12 (75 %) | Good | n/a |
| Zimmerman (1995) [44] | 10 | 1 | 1 | 10/12 (83.3 %) | Good | n/a |