Nigeria rolled out the 10-valent Pneumococcal Conjugate Vaccine (PCV10) into the routine childhood immunisation schedule through support from Gavi, the Vaccine Alliance. PCV10 was introduced in the context of the lack of pneumococcal disease data. Because PCV protects against nasopharyngeal carriage and pneumococcal diseases caused by included serotypes (vaccine serotypes - VT), monitoring the serotype distribution in carriage and disease is essential to evaluate vaccine impact fully. Based on the Gavi timelines, the country will transition to fully self-financing the PCV10 programme in a few years, translating to nearly half the total cost of fully immunising a child. Contextual evidence of PCV10 impact will benefit policymakers when deciding on financing and sustaining the PCV10 programme.Relying on the necessity of pneumococcal carriage for invasion, this PhD addresses the three aspects of evaluating PCV impact: Firstly, I conducted annual carriage (2017-2020) and vaccine coverage (2018-2020) surveys to assess the population-level impact of PCV10 introduction on pneumococcal carriage and its relationship to PCV10 coverage in children. I found a slow rise in PCV10 coverage accompanied by a significant reduction in carriage prevalence of VT in children aged <5 years and persons aged ≥5 years and a variable increase in carriage prevalence of non-VT. I also found evidence of direct and indirect protection against carriage demonstrated by an inverse relationship between coverage with two doses of PCV10 among children aged <5 years and VT carriage among children aged <5 years and persons ≥5 years. I found substantial residual VT carriage of 22% in the rural and 12% in the urban sites among children aged <5 years. Secondly, I assessed the applicability of three carriage-based models, where IPD surveillance is non-existent, to predict the impact of PCV10 introduction on IPD in children aged <5 years. These models predicted varying levels of decline in the overall IPD incidence. Analyses of underlying model assumptions and input data sources indicate the model predictions cannot be accurate because they discount the potential of residual VT carriage to cause IPD, thereby overestimating vaccine impact. Or they ignore the potential capacity of direct protection against VT invasion, thereby underestimating vaccine impact. Finally, I conducted a cost-of-illness study to assess the economic burden of pneumonia and IPD. I found that costs were substantial to the healthcare provider and households with significant variation by clinical syndrome and level of care. In addition, a third of households encountered catastrophic costs, ranging from 4% of the least poor to 53% of the poorest households. In conclusion, sustaining the PCV10 programme has reduced the burden of carriage and has also at a minimum reduced a proportionate fraction of invasive disease. Improved PCV10 coverage can improve herd immunity and further drive reductions in VT carriage and subsequent disease. Additionally, the programme will potentially save resources from treatment costs at both provider and household levels. Two adaptations will be required to reliably apply carriage-based models in settings that lack IPD surveillance. Firstly, model input data sources should be more representative to give a more accurate picture serotype distribution. Secondly, models should be adjusted to capture direct vaccine effects against IPD from persistent VT carriage.