Abstract:
This research developed epidemiological models for the susceptible infected
recovered for tuberculosis epidemic in Kenya using stochastic and deterministic
models as well as spatial temporal models. Tuberculosis disease is transmitted within
and between communities when infected and susceptible individuals interact. Interest
in the epidemiology of TB was triggered by the re-emergence of tuberculosis in the
early 1990's with the advent of HIV and falling economic status of many people
which subjected them to poverty. In this study, we focused on the period 2000-2013
and all the notified data in Kenya was included. Data on estimates of TB incidence,
prevalence and mortality was extracted from the WHO global tuberculosis database.
The study was guided by the following objectives: to develop epidemic models for
TB progression; to estimate the expected number of individuals with the disease at
any given time t; to formulate small area estimation models for TB progression; and
to develop spatial-temporal models for the TB progression. The results showed that
there was an average decline of 5% over the last 8 years with the highest decline
being reported in the year 2012/13. TB continues to disproportionately affect the
male gender with 58% being male and 42% being female. Kenya has made
significant efforts to address the burden of HIV among TB patients with
cotrimoxazole preventive therapy (CPT) uptake reaching 98% with and ART at 74%
by the end of 2013. The gains in the decline of TB could be attributed in part to in
the outcomes of integrating TB and HIV services and these gains should be
sustained. What is equally notable is the clear epidemiologic shift in age indicating
reduced transmission in the younger age groups. The spatial reference regions
considered were the 47 Kenyan counties. The covariates considered were gender,
xviii
HIV positive proportion, directly observed Treatment (DOTs), average weight,
average Body Mass Index (BMI) and average age. From the results of all notified TB
cases, only average BMI was excluded from the spatial temporal model since it was
not statistically significant (p-value > 0.05). The estimated risk of case notification
rates per 100,000 were found to be highest in the following counties Marsabit, Isiolo,
Nairobi, Lamu, Mombasa, Machakos, Kajiado, Makueni, Kisumu, Siaya and
Homabay. The study recommends that efforts must be made in addressing the risk
factors for TB which is geographically differentiated.
