**3.1 Differential local Moran's I analyses**

LDMI analytics identified different cluster patterns for the different time periods evaluated (**Tables 2–4**).

**Table 2** represents LDMI results between time 0 (2000) and time 1 (2005). Algeria, Burkina Faso, and Senegal had significant clusters at different p-values (Supplementary Material—C).

**Figure 3.**

*Clusters and significance levels of TB treatment completion rates in year 2010.*

#### **Figure 4.**

*Clusters and significance levels of TB treatment completion rates in year 2015.*


#### **Table 2.**

*Differential local Moran's I estimations of TB treatment completion rates between time 0 (2000) and time 1 (2005).*

*Geospatial Clustering of Mobile Phone Use and Tuberculosis Health Outcomes among African… DOI: http://dx.doi.org/10.5772/intechopen.98528*


#### **Table 3.**

*Differential local Moran's I estimations of TB treatment completion rates between time 0 (2000) and time 2 (2010).*


#### **Table 4.**

*Differential local Moran's I estimations of TB treatment completion rates between time 0 (2000) and time 3 (2015).*

**Table 3** shows the result between time 0 (2000) and time 2 (2010). Niger, Senegal, Gambia, Namibia, Lesotho, Djibouti, Algeria, Cameroon, South Africa, Democratic Republic of Congo (DRC), Kenya, and Sierra Leone had significant clusters at different p-values (Supplementary Material—C).

**Table 4** represents estimation results between time 0 (2000) and time 3 (2015). Niger, Burkina Faso, Senegal, South Africa, Algeria, Cameroon, and DRC had significant clusters at different p-values (Supplementary Material—C).

*Spatial Correlation Analyses.*

Spatial relationship between TB treatment completion rates and mobile phone use identified High-High clusters, Low-Low clusters, and Low-High outliers (**Figure 5**).

**Figure 5.**

*Spatial correlation result between TB treatment completion rates and Mobile phone use.*
