First author name | Year | Study design | Study setting | Study population | Sample size | Measurement of glycemic control | Definition of glycemic control | Generic factor | Specific factors | Measure of association | Point estimate | Lower bound | Upper bound | Association with glycemic control |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Mash [55] | 2014 | Pragmatic randomized controlled trial | South Africa | Type 2 diabetes | 422 | HbA1c | No threshold | Educational intervention | Effectiveness of group education | Mean difference | 0.01 | − 0.2 | 0.28 | No significant difference between the intervention and control groups in reduction of HbA1c level by 1% |
Thuita [81] | 2020 | Randomized controlled trail | Kenya | Type 2 diabetes; 20–79 years | 153 | HbA1C and FBG | Good control (HbA1c < 7%); poor control (HbA1c > 7%) | Effect of a nutrition education programme on the Metabolic syndrome in type 2 diabetes patients | Odds Ratio | 2.04 | 0.84 | 4.92 | No significant difference in the prevalence of high HbA1c between groups at six months post intervention (Nutrition education group versus Control) | |
2.08 | 0.85 | 5.09 | No significant difference in the prevalence of high HbA1c between groups at six months post intervention (Nutrition education peer to peer support group (NEP) versus Control) | |||||||||||
Gathu [41] | 2018 | Randomized controlled trial | Kenya | Sub-optimally controlled Type 2 diabetes; 18–65 years | 140 | HbA1c | No threshold | Diabetes self-management education | Mean difference | 0.37 | − 0.45 | 1.19 | No significant difference in the primary outcome (HbA1c) between the two groups | |
Muchiri [63] | 2016 | Randomized controlled trial | South Africa | type 2 diabetes; 40–70 years | 82 | HbA1c | No threshold | Effect of a participant-customised nutrition education programme | Mean difference | − 0.64 | − 0.19 | 1.15 | No significant difference between the intervention and control groups for HbA1c at 6 months (p = 0·13) | |
− 0.63 | − 0.26 | 1.5 | No significant difference between the intervention and control groups for HbA1c at 12 months (p:0.16) | |||||||||||
Hailu [43] | 2018 | Controlled before-and-after study | Ethiopia | Type 2 diabetes mellitus; > 30 years | 220 | FBS | No threshold | Nurse-led diabetes self-management education | Mean difference | 27 | 17 | 37 | Significant difference in the intervention group compared to the control group at 9 months | |
Fseha [40] | 2017 | Cross-sectional | Ethiopia | Type 2 diabetes; 22–60 years | 200 | FBS | Good (FBS 70–130 mg/dl), Poor (FBS > 130 mg/dl) | Self-management of diabetes | Home glucose monitoring | Odds Ratio | 1.697 | 0.852 | 3.353 | Home glucose monitoring was not associated with glycaemic control |
Fseha [40] | 2020 | Cross-sectional | Malawi | Type 2 diabetes; ≥ 25 years | 428 | HbA1c | Poor control (HbA1c clinically elevated ≥ 8%) | Additional blood glucose monitoring | ß coefficient | − 0.359 | − 0.609 | − 0.149 | Additional blood glucose monitoring at private clinic/home/diabetes peer groups was not associated with glycaemic control | |
Mwavua [64] | 2016 | Cross-sectional | Kenya | Type 2 diabetes; ≥ 18 years | 200 | HbA1c | Good control (HbA1c < 7%); poor control (HbA1c ≥ 7%) | Monthly self-monitoring | Odds ratio | 2.6 | 0.6 | 10.7 | Monthly self-monitoring was not significantly associated with glycaemic control | |
Absence of self-monitoring | Odds Ratio | 3.3 | 0.8 | 12.6 | Absence of self-monitoring was not associated with glycaemic control | |||||||||
Maharaj [52] | 2016 | randomized controlled trial | Nigeria | Non-insulin dependent type 2 diabetes; 30–58 years | 90 | HbA1c | No threshold | Exercise programs | Effect of rebound exercise | Mean difference | 0.904 | 0.832 | 0.985 | Significant improvements at 9 weeks post-intervention in mean HbA1c in the exercise group |
Yan [84] | 2014 | Randomized controlled trial | Mozambique | Type 2 diabetes; 40–70 years | 41 | HbA1c | No threshold | Effect of Aerobic Training | Mean difference | − 1.0 | − 1.3 | − 0.7 | Significant reduction of plasma glucose at 120 min (Glu 120) following glucose load in the exercise group after training | |
Siddiqui [76] | 2018 | Quasi-experimental study | South Africa | Type 2 diabetes; 18–65 years | 95 | HbA1C | No threshold | Level of physical activity measured with a pedometer | Mean difference | − 1.0420 | − 1.2225 | − 0.86 | Change in HbA1c over the three-month period was significant in the intervention group | |
Fayehun [38] | 2018 | Randomized trial | Nigeria | Type 2 diabetes; 33–64 years | 46 | HbA1c | No threshold | Physical activity with a 10,000 steps each day | Mean difference | − 0.74 | − 1.32 | − 0.02 | Endline HbA1c was lower in the intervention group than in the control group | |
Ezema [37] | 2014 | Randomized trial | Nigeria | Type 2 diabetes; 40–55 years | 54 | FBS | No threshold | Aerobic exercise training (V02 max) | Pearson product moment correlation test | − 0.252 | Significant effect of the exercise training program on FBS (p:0.001); 95% CI not provided | |||
Mayet [57] | 2012 | Quasi-experimental study | South Africa | Type 2 diabetes | 600 | HbA1c | No threshold | Medical intervention | Insulin therapy initiation | Paired t-test | Mean HbA1c at insulin initiation was 10.29% (± 2.42), and 10.63% (± 1.93) after adjustment of insulin dose (p-value > 0.05) | |||
Rambiritch [71] | 2014 | A 12-week, prospective, single-center, open-label, dose-escalation study | South Africa | Poorly controlled type 2 diabetes requiring oral antidibetic medications; > 20 years | 22 | FBG | No threshold | Dose escalation of Glibenclamide | Analysis of variance | 19.61 | Significant decrease in percentage of glucose from dose zero to 2.5 mg (P ≤ 0.001); no significant decrease for the 2.5–5 mg, 5–10 mg, and 10–20 mg doses. 95% CI provided | |||
Assah [24] | 2015 | Non-randomized controlled trial | Cameroon | Type 2 diabetes | 192 | HbA1c | No threshold | Multi level peer support | Effectiveness of a community-based multilevel peer support intervention | Mean difference | − 1.7 | − 2.2 | − 1.3 | Significant reduction in HbA1c in the intervention group compared with controls |
Mash [54] | 2016 | Quasi-experimental study | South Africa | Type 2 diabetes; > 18 years | 600 | HbA1c | No threshold | Laboratory testing | Introducing point-of-care (POC) testing for HbA1c | Mean difference | 0.00 | -1.5 | 1.5 | No significant difference in Mean difference in HbA1c result (%) at 18 months in the intervention group |