Application of Geoinformation Technology to the Management of Community-Based Natural Resources for Tourism Development in Northern Tanzania

The approaches to natural resources management have evolved. Disparities in their adoption are likely to produce a long-lasting negative impact on the resources and the livelihood security of the community depending on them. The use of geoinformation by the local community is a critical measure to the sustainability of its resources. Nonetheless, the application of geoinformation technologies to the community-based natural resources for the tourism industry is highly unknown. This chapter reviewed the application of geoinformation technology to the management of community-based natural resources in the Pangani District of Northern Tanzania. It considers how geoinformation technology is used in the management of tourism activities for community development. Specifically, the chapter discusses community developments resulting from that as well as challenges associated with the use of geographical information systems and remote sensing technologies. The chapter concludes with key recommendations for improving those challenges.

Incidence Estimates of Acute Q Fever and Spotted Fever Group Rickettsioses, Kilimanjaro, Tanzania, from 2007 to 2008 and from 2012 to 2014

Q fever and spotted fever group rickettsioses (SFGR) are common causes of severe febrile illness in northern Tanzania. Incidence estimates are needed to characterize the disease burden. Using hybrid surveillance—coupling case-finding at two referral hospitals and healthcare utilization data—we estimated the incidences of acute Q fever and SFGR in Moshi, Kilimanjaro, Tanzania, from 2007 to 2008 and from 2012 to 2014. Cases were defined as fever and a four-fold or greater increase in antibody titers of acute and convalescent paired sera according to the indirect immunofluorescence assay of Coxiella burnetii phase II antigen for acute Q fever and Rickettsia conorii (2007–2008) or Rickettsia africae (2012–2014) antigens for SFGR. Healthcare utilization data were used to adjust for underascertainment of cases by sentinel surveillance. For 2007 to 2008, among 589 febrile participants, 16 (4.7%) of 344 and 27 (8.8%) of 307 participants with paired serology had Q fever and SFGR, respectively. Adjusted annual incidence estimates of Q fever and SFGR were 80 (uncertainty range, 20–454) and 147 (uncertainty range, 52–645) per 100,000 persons, respectively. For 2012 to 2014, among 1,114 febrile participants, 52 (8.1%) and 57 (8.9%) of 641 participants with paired serology had Q fever and SFGR, respectively. Adjusted annual incidence estimates of Q fever and SFGR were 56 (uncertainty range, 24–163) and 75 (uncertainty range, 34–176) per 100,000 persons, respectively. We found substantial incidences of acute Q fever and SFGR in northern Tanzania during both study periods. To our knowledge, these are the first incidence estimates of either disease in sub-Saharan Africa. Our findings suggest that control measures for these infections warrant consideration.

Validating machine learning models for the prediction of labour induction intervention using routine data: a registry-based retrospective cohort study at a tertiary hospital in northern Tanzania

ObjectivesWe aimed at identifying the important variables for labour induction intervention and assessing the predictive performance of machine learning algorithms.SettingWe analysed the birth registry data from a referral hospital in northern Tanzania. Since July 2000, every birth at this facility has been recorded in a specific database.Participants21 578 deliveries between 2000 and 2015 were included. Deliveries that lacked information regarding the labour induction status were excluded.Primary outcomeDeliveries involving labour induction intervention.ResultsParity, maternal age, body mass index, gestational age and birth weight were all found to be important predictors of labour induction. Boosting method demonstrated the best discriminative performance (area under curve, AUC=0.75: 95% CI (0.73 to 0.76)) while logistic regression presented the least (AUC=0.71: 95% CI (0.70 to 0.73)). Random forest and boosting algorithms showed the highest net-benefits as per the decision curve analysis.ConclusionAll of the machine learning algorithms performed well in predicting the likelihood of labour induction intervention. Further optimisation of these classifiers through hyperparameter tuning may result in an improved performance. Extensive research into the performance of other classifier algorithms is warranted.