Increasing levels of development assistance have been coupled with increased attention to its effectiveness, resulting in a series of international declarations outlining an agenda of five principles of aid effectiveness: ownership, alignment, harmonisation, management for results and mutual accountability. This PhD thesis examines whether the Tanzanian health Sector Wide Approach (SWAP) has achieved these principles. It uses a case study approach, mixing quantitative and qualitative methods. It first maps out the health policy and financial landscape of Tanzania since the introduction of the SWAP. The thesis then explores the international aid effectiveness agenda and develops a set of indicators to assess its achievement in the Tanzanian health SWAP. This includes analysing external and domestic health financing flows over the last ten years; document review of key processes and in-depth interviews. The application of this indicator framework shows mixed results. Better progress is found towards indicators from international declarations, which are based on having aid-management processes in place, than towards those developed as defined by local stakeholders. Institutional factors, including the incentives of the institutions and individuals involved in aid relationships, as well as the political context within which aid relationships take place, are found to be key in explaining these results. A political economy approach is then undertaken to characterise and explore these factors further. Individual and institutional incentives are found to be unaligned with aid effectiveness principles. Furthermore, the structure of the SWAP is technocratic, excludes important stakeholders and does not fully reflect the political context and power dynamics of aid relationships. This thesis finds fatigue and disengagement with the SWAP and the aid effectiveness agenda, and recommends that the international community engage in SWAP as a process of institutional reform rather than just a technocratic solution to development assistance. Principles of aid effectiveness should allow for greater adaptation to national 4 contexts. More research is needed to further integrate political and economic elements of frameworks to analyse aid relationships and deepen our understanding of how best to achieve institutional reform and improve aid effectiveness.

Considerable attention has been given over to the politicisation of life within the 21st century: the threat of new disease and the promise of new drugs; the advancement of technology capable of transforming live anew; and the recasting of biological citizenship. This thesis, however, responds to the growing calls, made by the likes of Kearns and Reid-Henry (2007), to consider the other side of our contemporary biopolitical regime and the avoidable suffering that is played out against this backdrop of possibilities. Utilising malaria as the disease specific entry point, the thesis aims to disclose the way in which health is mediated by (biological) events within the body as well as (political) events outside of the body and explore the dialogue that takes place across the body’s fleshy barrier. In doing so, I aim to interrogate the injustice and reveal the structural violence anonymously enacted through systems but personally embodied by certain individuals. Thus, the thesis contributes to, and moves forward, the on going work on the critical geographies of global health by traversing scales, bringing the critical conversations that have been predominantly focused at the all-too-impersonal global level down to those ‘at the sharp end’ (Dixon and Marston 2011, 445), ensuring such voices join the conversation and speak back to the global narrative. In doing so I provide a more geographically and personally attuned account of the ‘epidemiology of inequality’ (Sparke and Anguelov 2012) currently being sketched out within the discipline. By embedding personal experiences of (ill)health within a national and international context, I work to ensure that such episodes of illness are not framed as sad, unfortunate, biologically inevitable, or bad luck, but unequivocally as episodes of violence (after Craddock 2009). The thesis does so through a series of distinct chapters, each offering different perspectives yet threaded together with the themes of (structural) violence and the valuation and management of life today. From an initial focus on the (de)valuation of life implicit in an economic conceptualisation of the disease burden within the global health arena, the thesis goes on to focus on the politics of life from the perspectives of individuals themselves. Drawing on conversations with women in Dar es Salaam, Tanzania, the thesis seeks to recover the journeys travelled to and through the health system, pausing to reflect on the situations that influence the contours of this journey as well as the biological consequence of them.

Prostatic diseases are common; benign prostate hyperplasia (BPH) is almost ubiquitous in elderly men and 899,000 men were diagnosed with prostate cancer worldwide in 2008. The incidence of both is increasing and expected to continue to rise. Therefore, prostatic diseases represent a considerable economic burden, but there are currently no reliable markers available to accurately differentiate indolent from aggressive disease nor to predict who will benefit from treatment for either BPH or prostate cancer. This results in over and under-treatment of both diseases with consequent patient related morbidity and mortality. The molecular mechanisms underlying the natural history of prostatic diseases remain elusive. It is accepted that prostate cell growth and survival are exquisitely dependent upon activation of the androgen receptor (AR) by androgens. Following ligand binding, AR undergoes further phosphorylation at serine residues, which inhibit proteolytic degradation, stabilise AR and influence AR transactivation. It is therefore plausible that alterations in AR phosphorylation may drive prostatic disease progression. However, few studies have explored the significance of AR phosphorylation, or the kinases driving AR serine phosphorylation in the clinical setting. The over-riding objective of this study was to establish the clinical relevance of AR serine phosphorylation status in prostate tissue in both BPH and prostate cancer. The specific aims of the current study were: • To firstly establish and validate a panel of AR phosphospecific antibodies. • To evaluate site specific AR serine phosphorylation expression levels in prostate cancer and BPH patient cohorts, with full clinical data and follow-up. • To investigate the expression of candidate kinases mediating such phosphorylation. This involved establishing tissue banks with linked comprehensive clinical databases, and utilising this tissue to establish AR phosphorylation expression profiles for each patient. Six AR phosphospecific antibodies (pARS81, pARS94, pARS213, pARS515, pARS578, pARS650) were verified using peptide competition assays and western blotting. Cdk1, ERK1/2, Akt and PKC were identified as putative kinases mediating AR phosphorylation using the online kinase search tool Scansite 2.0. Immunohistochemistry was performed on hormone naïve diagnostic prostate cancer tissue relating to 90 patients. High expression levels of AR phosphorylation at serine sites 81, 515 and 578 were each associated with a poorer clinical outcome. Following cox regression analysis, cytoplasmic pARS515 expression (p=0.038, HR 4.5 (95% CI 1.1–20.6)) and pARS81 nuclear expression (p=0.030, HR 0.033 95% CI 0.002-0.721) were independently associated with shorter time to biochemical relapse and shorter disease specific survival respectively. Cdk1 and/or pCdk1161 were significantly associated with pARS81 and pARS515 as predicted by Scansite 2.0. Similarly, nuclear PKC expression was significantly associated with pARS578 expression both in the cytoplasm and the nucleus. In patients with PSA at diagnosis ≤20ng/ml, high cytoplasmic pARS515 expression was associated with significantly shorter time to biochemical relapse (p=0.019). This translated into significantly shorter disease-specific survival (p<0.001, 10y survival 38.1% vs 100%). Prostate cancer patients with a low serum PSA level at diagnosis may be suitable for delayed radical treatment via active surveillance. An investigation was therefore undertaken in 51 prostate cancer patients treated by active surveillance. Active surveillance is a deferred radical treatment approach which provides a potential solution to the problem of over treatment as a result of over-diagnosis. However some patients harbour occult aggressive disease and delay in treatment may result in disease progression and failure of radical therapy. Although none of the individual AR serine phosphorylation sites were associated with clinical outcome measures on univariate analysis, high expression of total AR in the cytoplasm (p=0.021, HR 4.6 (95% CI 1.3-16.8)) and presence of perineural invasion in the tumour specimen (p=0.003, HR 8.6 (95% CI 2.1-35.7)) were deemed independent with regards to shorter time to treatment intervention in a cox regression analysis. Validation of the results seen in the first active surveillance prostate cancer cohort was undertaken in a second prospectively collected cohort consisting of 84 active surveillance patients. The results in the first cohort were not replicated in the second. Although cytoplasmic pARS81 was associated with time to intervention (p=0.032) and pARS515 expression trended towards an association (p=0.072), an increase in patient numbers in both cohorts may have provided more reliable results. However even with the numbers available in contrast to the first active surveillance cohort, but in line with the pilot prostate cancer cohort, Cdk1 was associated with pARS515 expression, and pCdk1161 trended towards an association. BPH is also an androgen driven disease dependent upon the AR. Previous research into predictive and prognostic markers in BPH is scant. Therefore a comprehensive analysis of clinical and novel pathological factors, including markers of inflammation, was performed in 336 BPH patients. Following this a complete panel of AR serine phosphorylation sites, and associated kinases, was analysed with reference to clinical outcome measures in the BPH cohort. Low expression levels of total AR and AR phosphorylated at Ser-81, 515 and 650 were associated with poorer clinical outcomes. Low expression of smooth muscle pARS515 (p=0.029, HR 0.31 (95% CI 0.10-0.94)) and older age (p=0.004, HR 5.13 (95% CI 1.43-18.41)) were deemed independent on cox regression analysis with regards to shorter time to postoperative acute urinary retention (AUR). Furthermore, low expression of pARS515 in the smooth muscle was associated with increased incidence of postoperative AUR in patients over 70 years old (25.1% vs 2.8% at 10 years following transurethral resection of prostate (TUR)), (p=0.002, HR 0.20 (95% CI 0.06-0.62)). This may have important clinical implications in postoperative counselling of these patients. In addition it may influence the decision to commence early postoperative medical treatment (with 5-alpha-reductase inhibitors and/or alpha blockers) on a prophylactic basis in these patients. Cytoplasmic pARS650 expression (p=0.010, HR 0.50 (95% CI 0.29-0.86)) and PSA at diagnosis (p=0.018, HR 1.89 (95% CI 1.11-3.16)) were independently associated with time to failure of surgical intervention. Furthermore, low expression of pARS650 in the cytoplasm was associated with increased failure of surgical intervention in patients with PSA ≥4ng/ml at diagnosis (45.5% vs 13% at 5 years post TUR), (p=0.026, HR 0.52 (95% CI 0.29-0.93)). This comprehensive study on immunohistochemical expression of site specific AR serine phosphorylation and associated kinases fills a gap in the current literature. It has demonstrated the clinical significance of AR serine phosphorylation in prostate cancer and BPH and uncovered potentially exciting new avenues for future investigation. Site specific serine phosphorylation of the AR may serve as a prognostic and predictive biomarker in prostatic disease and has potential as a future target for therapeutic intervention.