Skip to main content

Table 2 Comparison of relative risk estimates in unadjusted model with unmatched sample to adjusted model with matched sample at each time point

From: Improved clinical and laboratory skills after team-based, malaria case management training of health care professionals in Uganda

Indicator Percentage correct at six weeks relative to baseline Percentage correct at 12 weeks relative to baseline Percentage correct at one year relative to baseline
  Unadjusted Adjusted Difference Unadjusted Adjusted Difference Unadjusted Adjusted Difference
  RR RR   RR RR   RR RR  
Number of patients observed 179 124   166 109   133 56  
Proper history taking 2.10 1.86 -0.24 2.96 3.35 0.39 4.37 5.63 1.26
Thorough physical exam 3.09 2.66 -0.43 4.51 4.09 -0.42 3.40 2.35 -1.05
Correct diagnosis 1.91 1.77 -.14 1.92 1.96 0.04 1.74 2.16 0.42
Correct treatment 2.03 1.96 -.07 2.18 2.42 0.24 2.14 2.21 0.07
Appropriate patient education 4.80 4.47 -.33 5.04 5.74 0.70 2.23 2.04 -0.19
  1. The unadjusted model was estimated with unmatched observations. The adjusted model was estimated with matched observations
  2. These results and their confidence intervals are reported in Table S2 of the Additional file 2. All increases were statistically significant with the exception of patient education from baseline to one year when estimated with the adjusted model
  3. The number of clinician as well as the number of patients observed was smaller in the matched than the unmatched sample. For the percentage correct at six weeks relative to baseline, 49 clinicians were observed in the matched compared to 94 in the unmatched sample. For the percentage correct at twelve weeks relative to baseline, 47 clinicians were observed in the matched compared to 95 in the unmatched sample. For the percentage correct at one year relative to baseline, 24 clinicians were observed in the matched compared to 84 in the unmatched sample