- Title
- Statistical analysis and modelling of malaria relapse risk factors using Cox Proportional Hazards model approach : a case study of material data collected from Democratic Republic of Congo
- Creator
- Mutambayi, Ruffin Mpiana
- Subject
- Malaria -- Congo (Democratic Republic) -- Mathematical models Statistics -- Congo (Democratic Republic) -- Econometric models
- Date Issued
- 2017
- Date
- 2017
- Type
- Thesis
- Type
- Doctoral
- Type
- PhD
- Identifier
- http://hdl.handle.net/10353/15001
- Identifier
- vital:40110
- Description
- The field of survival analysis has experienced tremendous growth during the latter half of the 20th century. The objective of this study was to model the relapsing time of uncomplicated falciparum malaria using biographical, sanitation, environmental and preventive information as covariate risk factors. The methodological developments of survival analysis that have had the most profound impact are the Kaplan-Meier method for estimating the survival function, the log-rank test for comparing the equality of two or more survival distributions, and the Cox proportional hazards (PH) model for examining the covariate effects on the hazard function. This thesis presents basic concepts, nonparametric methods (the Kaplan-Meier method and the log-rank test), semi parametric methods (the Cox PH model, and Cox model with time-dependent covariates) and parametric methods (Parametric PH model and the AFT model) for analysing survival data. An improved method was applied to randomized uncomplicated falciparum malaria patients to investigate the risk factors related to the relapsing time in RD Congo. Firstly, the Cox PH model , inf 1 0 ( ) ( )exp 1.91613 0.49633 0.81466 dump spr t X h h t X X X where is the instantaneous risk of experiencing the event at time , without the influence of any covariate was found to better fit the data, as confirmed by the results of the global test that present reasonable and significant results: (Likelihood Ratio: 18.2264 with p-value < 0.0004; Score: 17.6569 with p-value < 0.0005 and Wald: 19.3975 with p<0.0002). Hence from the model above, it can be noticed that ‘dumping site’ (p-value <0.0106; 95percent HR C.I: 1.545, 29.451), ‘spray used’ (p-value < 0. 0.0220; 95percent C.I: 0.391, 0.915), and ‘information related to source of malaria’ (p-value < 0. 0.0012; 95percent C.I: 1.380, 3.725), have a significant impact on the relapsing time of the patients under investigation. It should be noted that many 0() h tt other important covariates such as “usage of mosquito nets” were excluded because of the non-proportionality of the hazards. Secondly, the improvement concept was applied to the Cox model as proposed, and the improved model was as follows: h h t t X t X t X t X t X t X t X t r dump_inf1 spr_inf1 Where h0 (t) is the instantaneous risk of experiencing the event at time t , without the influence of any covariate. The model was found to better fit the data than the standard model — as confirmed by the results from the convergence criterion comparison, where the AIC value of the improved model (754.460) was lower than that of the standard model (767.428). Moreover, the improved model was found to be adequate as the results from the global null hypothesis were significant: (Likelihood Ratio: p-value is 0.0001; Score: p-value is 0.0001; Wald: p-value is 0.0001) and the goodness-of-fit test was confirmed by the plotting of Coxresiduals where the 45 degrees’ line confirmed the results. In conclusion, it was found that the covariates ‘pit’ (p-value is 0.0009; 95percent HR C.I: 11.4133, 1.243e+04), ‘tdwel_spr’ (pvalue = 0.0257; 95percent HR C.I: 1.0016, 1.030e+00), ‘tdwel_inf1 (p-value = 0.0122; 95percent HR C.I: 0.9691, 9.964e-01), ‘maint_dump’ (p-value is 0.0005; 95percent HR C.I: 1.0239, 1.087e+00),‘maint_pit’ (p-value is 0.0007; 95percent HR C.I: 0.9179, 9.774e-01), ‘dump_spr’ (p-value is 0.0027; 95percent HR C.I: 0.9308, 9.853e-01), ‘dump_inf1’ (p-value <0.0001; 95percent HR C.I: 0.9849, 9.947e-01) and ‘spr_inf1’ (p-value is 0.0005; 95percent HR C.I: 1.0119, 1.043e+00) affected the time to relapse from uncomplicated falciparum malaria.
- Format
- 206 leaves
- Format
- Publisher
- University of Fort Hare
- Publisher
- Faculty of Science and Agriculture
- Language
- English
- Rights
- University of Fort Hare
- Hits: 565
- Visitors: 545
- Downloads: 116
Collections
UFH Department of Statistics
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | SOURCE1 | Ruffin Mpiana Mutambayi.pdf | 3 MB | Adobe Acrobat PDF | View Details Download |