Table 1.
Characteristics of the 65 included studies
| Characteristic | Europe | Asia | North America | Africa | Oceania | All Regions |
|---|---|---|---|---|---|---|
| Number of studiesa | 29 | 13 | 10 | 8 | 4 | 65 |
| Sample size | ||||||
| <5000 | 3 (10.3 %) | 1 (7.7 %) | 0 | 2 (25.0 %) | 1 (25.0 %) | 7 (10.8 %) |
| 5000–9999 | 3 (10.3 %) | 4 (30.8 %) | 1 (10.0 %) | 0 | 1 (25.0 %) | 9 (13.8 %) |
| 10,000–49,999 | 6 (20.7 %) | 2 (15.4 %) | 1 (10.0 %) | 4 (50.0 %) | 1 (25.0 %) | 14 (21.5 %) |
| 50,000–99,999 | 4 (13.8 %) | 1 (7.7 %) | 0 | 2 (25.0 %) | 0 | 7 (10.8 %) |
| 100,000+ | 13 (44.8 %) | 5 (38.5 %) | 8 (80.0 %) | 0 | 1 (25.0 %) | 28 (43.1 %) |
| Study design | ||||||
| Cohort | 16 (55.2 %) | 4 (30.8 %) | 8 (80.0 %) | 0 | 4 (100 %) | 32 (49.2 %) |
| Cross-sectional | 13 (44.8 %) | 9 (69.2 %) | 2 (20.0 %) | 8 (100 %) | 0 | 33 (50.8 %) |
| Publication date | ||||||
| 1990–2002 | 17 (58.6 %) | 2 (15.4 %) | 2 (20.0 %) | 5 (62.5 %) | 2 (50.0 %) | 28 (43.1 %) |
| 2003–2013 | 12 (41.4 %) | 11 (84.6 %) | 8 (80.0 %) | 3 (37.5 %) | 2 (50.0 %) | 37 (56.9 %) |
| Assessment yearb | ||||||
| <1990 | 10 (34.5 %) | 1 (7.7 %) | 0 | 2 (25.0 %) | 0 | 13 (20.0 %) |
| 1990–1999 | 10 (34.5 %) | 3 (23.1 %) | 4 (40.0 %) | 4 (50.0 %) | 2 (50.0 %) | 23 (35.4 %) |
| 2000–2009 | 7 (24.1 %) | 9 (69.2 %) | 6 (60.0 %) | 1 (12.5 %) | 1 (25.0 %) | 25 (38.5 %) |
| Not reported | 2 (6.9 %) | 0 | 0 | 1 (12.5 %) | 1 (25.0 %) | 4 (6.2 %) |
| Prevalence period typec | ||||||
| Point | 6 (20.7 %) | 4 (30.8 %) | 1 (10.0 %) | 2 (25.0 %) | 1 (25.0 %) | 14 (21.5 %) |
| 12 months | 11 (37.9 %) | 4 (30.8 %) | 5 (50.0 %) | 1 (12.5 %) | 1 (25.0 %) | 22 (33.8 %) |
| Other period | 7 (24.1 %) | 2 (15.4 %) | 2 (20.0 %) | 2 (25.0 %) | 0 | 13 (20.0 %) |
| Lifetime | 13 (44.8 %) | 6 (46.2 %) | 3 (30.0 %) | 5 (62.5 %) | 2 (50.0 %) | 30 (46.2 %) |
| Quality score | ||||||
| Mean ± SD | 9.0 ± 2.6 | 8.8 ± 3.4 | 7.7 ± 2.1 | 10.1 ± 1.6 | 8.8 ± 1.5 | 8.8 ± 2.6 |
aOne study (Nuevo et al.) reported on multiple regions, and is counted only in the “All regions” category. Therefore, the number of studies in each region sum to only 64 studies
bFor studies that assessed prevalence over a range of years, the most recent year was used
cMultiple types of prevalence could be reported within the same study; therefore, percents do not add up to 100