Skip to main content

Advertisement

Table 2 Completeness and logical consistency of data records collected using smartphone and pen-and-paper data collection systems, Kenya, 2011-2012

From: A comparison of smartphones to paper-based questionnaires for routine influenza sentinel surveillance, Kenya, 2011–2012

  Smart phone Pen and paper  
Question type Total number of questions assessed Number of errors/incomplete/inconsistent records (%) Total number of questions assessed Number of errors/incomplete/inconsistent records (%) Adjusted IRR* p value 95% CI
Questions requiring responses without programmed checks on the smartphone version
Incomplete records        
Questions on clinical symptoms 11,440 11 (0.1) 11,440 54 (0.5) 4.2 <0.01 2.1-8.3
Questions assessing past medical history 1,760 6 (0.3) 1,760 5 (0.3) 0.9 0.82 0.3-2.9
Questions on exposure history 3,520 6 (0.2) 3,520 108 (3.1) 17.4 <0.01 7.6-39.8
Questions on vital signs 1,760 2 (0.1) 1,760 14 (0.8) 5.6 0.03 1.2-26.3
Total 18,480 25 (0.1) 18,480 181 (1.0) 7 <0.01 4.4-10.3
Errors and inconsistent answers        
Questions assessing past medical history 243 1 (0.4) 227 178 (78.4) 200 <0.01 28.0-1431.2
Questions on exposure history 6 2 (33.3) 6 3 (50.0) 1 0.86 0.2-8.2
Questions on vital signs 1,760 5 (0.3) 1,760 5 (0.3) 1 0.48 0.2-2.4
Total 2,009 8 (0.4) 1,993 186 (8.3) 25 <0.01 12.5-51.8
Questions requiring responses with programmed checks on the smartphone version
Date of interview - - 880 0 (0) - - -
Case classification - - 880 2 (0.2) - - -
Gender - - 880 2 (0.2) - - -
Pregnant if female - - 9 6 (66.7) - - -
Total - - 2,649 10 (0.4) - - -