Table 3.
Findings for each diabetes comorbidity from included systematic reviews and meta-analyses
| Comorbidity | Author, year | Main findings* | Heterogeneity | Subgroup analysis† |
|---|---|---|---|---|
| Cardiovascular disease |
Einarson, 2018 [24] |
Weighted average prevalence | N/A |
N/A |
| Upper middle-income countries: | ||||
| Stroke: 6.9% | ||||
| Myocardial infarction: 3.4% | ||||
| Coronary artery disease: 17.7% | ||||
| Lower middle-income countries: | ||||
| Stroke: 6.3% | ||||
| Coronary artery disease: 27.3% | ||||
| Poorzand, 2019 [25] |
Premature coronary artery disease OR = 2.35 (95% CI = 1.71-3.21) |
I2 = 25.5%, P = 0.225 |
N/A |
|
| Chronic kidney disease |
Koye, 2017 [26] |
Annual cumulative incidence in two primary studies: 8.1% and 8.6% |
N/A |
N/A |
| Shiferaw, 2020 [27] |
Prevalence, stages 1-5: 36% (95% CI = 26%-45%) |
I2>90%, P < 0.001 |
Study design for stages 3-5 |
|
| Prevalence, stages 3-5: 15% (95% CI = 11%-19%) | ||||
| Depression |
Mendenhall, 2014 [28] |
Average prevalence: 36% | N/A |
N/A |
| Uphoff, 2019 [29] |
Prevalence: 40% (95% CI = 34%-45%) |
I2 = 97.5%, P < 0.001 |
N/A |
|
| Hussain, 2018 [30] |
Prevalence: 38% (95% CI = 31%-45%) |
I2>90%, P < 0.001 |
N/A |
|
| Khalighi, 2019 [31] |
Prevalence: 61% (95% CI = 55%-67%) |
I2 = 98%, P < 0.001 |
Income level, depression assessment tool |
|
| Wang, 2019 [32] |
Prevalence: 30.7% (95% CI = 16.6%-44.9%); OR = 3.05 (95% CI = 1.11-8.37) |
I2 = 99%, P < 0.001; I2 = 90.5%, P < 0.001 |
Study design, income level |
|
| Tuberculosis |
Bailey & Ayles, 2017 [33] |
Adjusted OR range: 2.14 (95% CI = 1.32-3.46) to 19.3 (95% CI = 6.1-61.0) |
N/A |
N/A |
| McMurry, 2018 [34] |
Prevalence range: 0.1% to 4.9% |
N/A |
N/A |
|
| Al-Rifai, 2017 [35] |
Risk estimate: 3.40 (95% CI = 1.53-7.56) |
I2 = 87.4%, P < 0.001 |
N/A |
|
| OR = 3.04 (95% CI = 2.29-4.03) |
I2 = 23.5%, P = 0.250 |
|||
| Lee, 2017 [36] |
OR = 1.16 (95% CI = 0.97-1.37) |
I2 = 0%, P = 0.608 |
N/A |
|
| Tegegne, 2018 [37] | OR = 1.78 (95% CI = 1.26-2.52) | I2 = 71.1%, P < 0.001 | Study design, income level, confounder adjustment |
*Prevalence estimates refer to the prevalence of each comorbidity among people with diabetes. Effect estimates refer to the risk or odds of the comorbidity in people with vs without diabetes.
†Subgroup analyses conducted as part of this umbrella review.