研究背景： / 2009年起，中國政府發起並資助了一項覆蓋全國31個省221個鄉村、針對100萬名農村婦女的細胞學及肉眼觀察宮頸癌篩查試點項目。國家及地方政府需要對可行的篩查策略進行衛生經濟學評估，為下一步擴大規模的篩查提供政策依據。 / 研究目標： / 應用人群特異性Markov模型，對巴氏塗片及肉眼觀察的宮頸癌篩查策略進行成本效果及成本效用兩方面的衛生經濟學評估，進而為中國農村婦女宮頸癌篩查政策的制定提供依據。 / 研究方法： / 本論文工作建立了Markov人群動態擬合模型，該模型能夠整合與中國農村宮頸癌流行情況相吻合的成本及健康狀況的數據，進而用於擬合20年內35-59歲中國農村婦女在有/無篩查幹預下的成本、效用和效果。本文分析的八個備選篩查策略包括：採用醋酸染色肉眼觀察（VIA）或傳統細胞學（巴氏塗片）分別進行10年，5年，3年及1年一次的篩查。 / 本文從社會學角度出發，成本數據涵蓋篩查、診斷及治療過程中產生的直接及間接成本。模型在結構上綜合了已被廣泛認可的宮頸癌自然發展史模型，以及宮頸癌及其癌前病變（CIN）在中國農村進行篩查和治療的標準臨床路徑。模型輸入參數盡可能地使用了能夠反映中國農村婦女人群特異性的數據。通過對比國家報告數據與模型預測結果，本文從全死因死亡率、宮頸癌死亡率及宮頸癌發病率三個方面驗證了模型的可信度。 / 模型的結局變量包括：累計成本、累計生命年(LYs)、累計質量調整生命年(QALYs)、預期宮頸癌死亡率及發病率降低百分比(%)、CIN 相對風險、宮頸浸潤癌相對風險，增量成本效用比(ICUR, 表述為每挽救一個質量調整生命年消耗的成本)及增量成本效果比(ICER, 表述為每挽救一個生命年消耗的成本)等。與無篩查幹預相比，我們界定ICUR及ICER小於三倍人均國內生產總值（76,824元，2009年)的優勢策略為‘具有成本效益’的選擇，並將其中ICUR和ICER最低的策略，定義為‘最具成本效益’的策略，將具有最大健康效益的策略（挽救最多質量調整生命年或生命年的策略），定義為‘最有效’的策略。同時，我們對可能影響決策的不確定因素進行了敏感性分析。 / 結果： / 與無篩查幹預相比，肉眼觀察及巴氏塗片篩查均能夠減少宮頸癌患病例數，進而顯示出一定的健康效益。較短的篩查間隔具有更高的健康效益。模型預測在不同的篩查策略幹預下，宮頸癌死亡率和發病率分別有望降低6.67-31.95%和5.12-24.71%，預期CIN發病相對風險為0.89-0.98，預期宮頸癌發病相對風險為0.73-0.95。篩查幹預對健康的保護作用在本研究中得到了證實。 / 成本效用分析顯示，10年一次的肉眼觀察策略最具成本效益，其次為5年一次、3年一次、1年一次的肉眼觀察篩查策略及1年一次的巴氏塗片篩查策略。與無篩查幹預相比，如上策略每挽救一個質量調整生命年消耗的成本為11,921至26,069元（1,892-4,138美元，2012年）。同時成本效果分析也顯示，10年一次的肉眼觀察策略最具成本效益，其次為5年一次的肉眼觀察策略及5年一次的巴氏塗片篩查策略。同樣與無篩查幹預相比，如上策略每挽救一個生命年消耗的成本為37,211至68,226元（5,906-18,830美元，2012年）。 / 對於某一既定策略，相應的ICUR和ICER受當地經濟狀況相關因素的影響最大，這些因素包括治療成本、篩查成本和成本貼現率。從檢測技術水平上看，肉眼觀察對分析結果的影響小於巴氏塗片，原因是前者敏感度範圍較小。篩查覆蓋率、初篩陽性隨訪率、診斷陽性治療率也都與相應的ICUR和ICER呈負相關性。敏感性分析結果顯示本文中模型對於健康結局的預測，及相關的衛生經濟學分析，受自然史模型中HPV感染和CIN之間轉移概率的不確定性的影響最大。HPV感染與CIN間的進展和逆轉概率是該項模型研究的核心參數。 / 結論： / 本文中成本效用和成本效果分析均顯示，相較於傳統的細胞學篩查策略，採用間隔時間較長（10年或5年）的肉眼觀察篩查策略，對一般發病地區的35-59歲的農村婦女來說，是更具‘成本效益’的選擇。對於宮頸癌高發地區，其篩查頻率可以提高到1年一次。1年一次的巴氏塗片篩查策略，是最有效的篩查策略，可以挽救最多的生命。但採用該策略時，應在財政預算允許的前提下，確保篩查技術和項目完成的質量。 / 篩查項目的高覆蓋率，對篩查陽性患者良好的隨訪和診治，初篩檢測技術平均水平以上的表現，以及較低的篩查和治療成本是確保篩查項目具備成本效益優勢的核心因素。本文完成的成本效用及成本效果分析，能夠為公共衛生決策提供重要的輔助作用。 / Background: / A Chinese government-sponsored cytology/visual inspection pilot cervical cancer screening program covered 10 million rural women in 221 counties of 31 provinces was initiated in 2009. Both the local and national governments in China need health economic evaluations of feasible strategies so as to make better policies for the next-step enlarging screening. / Objectives: / To perform health economic evaluations of Pap smear and visual inspection cervical cancer screening strategies using population-specific Markov modeling cost-utility (CUA) and cost-effectiveness (CEA) analyses, in order to assist screening policy making for women in rural China. / Methods: / Markov simulation models were developed to synthesize the evidence on costs and health outcomes related to cervical cancer epidemiology in rural China, and applied to predict the long-term utility, effectiveness and costs for hypothetical cohorts of 35-59 years old rural Chinese women, with or without the presence of screening over 20 years. The eight alternative screening strategies assessed were visual inspection with acetic acid (VIA) or traditional cytology (Pap smear) each with ten-year, five-year, three-year and one year screening intervals. / The study was conducted from the societal perspective, thus both directed and non-direct costs related to screening, diagnosis and treatment interventions were considered. The model structures incorporated with the well-accepted the natural history model of cervical cancer and the standard clinical pathway of screening and treatment interventions for precancerous lesions (CIN) and cervical cancer in real practice in rural China. Population-specific data were used as much as possible to be the model inputs. The model estimates were validated by comparison of our predictions of all-cause mortality, cervical cancer mortality and cervical cancer incidence with the national reported data. / Outcome variables included cumulative cost, life years (LYs), quality-adjusted life years (QALYs), predicted reduction(%) in cervical cancer mortality and incidence, relative risk of CIN, relative risk of cervical cancer, incremental cost-utility ratio (ICUR, presented as cost per QALY saved) and incremental cost-effectiveness ratio (ICER, presented as cost per life year saved). Compared with no screening, not-dominated strategies with ICUR and ICER less than three times China’s GDP per capita (76,824 CNY, 2009) were considered to be ‘cost-effective’ options. Among the identified ‘cost-effective’ options, the strategy with lowest ICUR or ICER was defined as the most cost-effective strategy, and the strategy with the highest health benefit (largest QALY saved or life year saved) was defined as the most effective strategy. Sensitivity analyses were conducted to test the effect of uncertainties on decision making. / Results: / All of the VIA and Pap smear screening strategies of showed certain benefits due to the decreased number of women developing cervical cancer, when compared with no screening. A trend for shorter screening interval to have greater benefit was also found. Cervical cancer mortality and incidence were expected to be reduced by 6.67-31.95% and 5.12-24.71% with different screening strategies. And the predicted relative risks of CIN and invasive cervical cancer of 0.89-0.98 and 0.73-0.95, respectively, also demonstrated the protective effect of screenings. / Modeling cost-utility analysis identified ten years VIA screening as the most cost-effective strategy followed by VIA screening with five-, three- and one year interval and Pap smear screening with a one year interval. Compared with no screening, the incremental costs per QALY saved of these strategies ranged from 11,921 to 26,069 Yuan (1,892-4,138 US dollars, 2012). In the meanwhile, modeling cost-effectiveness analysis also identified ten-years VIA screening as the most cost-effective strategy followed by VIA screening with five-year intervals and Pap smear screening with five-year intervals. Compared with no screening, the incremental costs per life year saved of these strategies ranged from 37,211 to 68,226 Yuan (5,906-18,830 US dollars, 2012). / Both ICUR and ICER of a selelected strategy were greatest influnced by factors related to variations in local economies , including treatment cost, screening cost and discounting rate of the cost. The influence of primary test performance of VIA was rather less than that of Pap smear due to the narrower ranges of the VIA sensitivities. Screening coverage, follow-up rate and treatment rate were also negatively associated with ICUR and ICER. Health outcome predictions and health economic analyses were mostly influenced by the uncertainties in HPV infection and CIN transitions in the natural history. Progression and regression probabilities between HPV infection and CIN were considered to be the key parameters of the simulation models. / Conclusions: / Baseline CUA and CEA results suggested that in comparison with traditional cytology screening strategies, organized VIA screening with long intervals (ten or five years) were more cost-effective options than for 35-59 years old women in normal incidence areas of rural China. The VIA screening interval can be shorten to one year in high incidence areas. Pap smear strategy with one year interval can be utilized as the most effective strategy with most lives saved when budget allows and the performances of program and test are ensured. / High coverage of the screening program, good management of screening positives, average or above performance of primary test, and lower screening and treatment costs are key elements for a cost-effective screening program. Cost-utility and cost-effectiveness analyses, such as the one conducted in this thesis study, can be considered important adjuncts to policy decision-making about public health objectives. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Li, Xue. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2013. / Includes bibliographical references (leaves 388-401). / Abstracts also in Chinese; appendixes includes Chinese. / Abstract of thesis --- p.i / 中文摘要 --- p.v / ACKNOWLEDGEMENTS --- p.viii / TABLE OF CONTENTS --- p.1 / LIST OF TABLES --- p.8 / LIST OF FIGURES --- p.11 / ABBREVIATIONS --- p.12 / Chapter CHAPTER 1 --- INTRODUCTION --- p.14 / Chapter 1.1 --- Epidemiological patterns and disease burden of cervical cancer --- p.14 / Chapter 1.1.1 --- Cervical cancer incidence and mortality worldwide --- p.14 / Chapter 1.1.2 --- Risk factors for cervical cancer --- p.15 / Chapter 1.1.2.1 --- Human Papillomavirus (HPV) --- p.15 / Chapter 1.1.2.2 --- Parity --- p.16 / Chapter 1.1.2.3 --- Smoking --- p.16 / Chapter 1.1.2.4 --- Human Immunodeficiency Virus (HIV) --- p.17 / Chapter 1.1.2.5 --- Contraception --- p.17 / Chapter 1.1.2.6 --- Sexual behavior, nutrition and other factors --- p.18 / Chapter 1.1.3 --- Disease burden of cervical cancer in China --- p.18 / Chapter 1.1.3.1 --- Epidemiology of Cervical Cancer in China --- p.18 / Chapter 1.1.3.2 --- Cervical cancer in different geographic areas of China --- p.20 / Chapter 1.2 --- The need for cost-effectiveness analysis of cervical screening strategies in China --- p.21 / Chapter 1.2.1 --- Cervical cancer prevention in China --- p.21 / Chapter 1.2.2 --- Why do we need a modeling cost-effectiveness analysis? --- p.23 / Chapter 1.3 --- Natural history of cervical cancer --- p.25 / Chapter 1.3.1 --- Terminology --- p.25 / Chapter 1.3.2 --- Natural history of cervical cancer --- p.27 / Chapter 1.4 --- Secondary prevention strategies of cervical cancer --- p.29 / Chapter 1.4.1 --- Screening tests --- p.29 / Chapter 1.4.1.1 --- Cervical cytology --- p.29 / Chapter 1.4.1.2 --- Visual Inspection --- p.32 / Chapter 1.4.1.3 --- HPV testing --- p.36 / Chapter 1.4.2 --- Summary of different screening strategies all over the world --- p.37 / Chapter CHAPTER 2 --- LITERATURE REVIEW --- p.40 / Chapter 2.1 --- Background --- p.40 / Chapter 2.2 --- Objectives of the literature review --- p.41 / Chapter 2.3 --- Search strategies and results --- p.41 / Chapter 2.3.1 --- Search strategies --- p.41 / Chapter 2.3.2 --- Inclusion and exclusion criteria --- p.42 / Chapter 2.4 --- Literature results summary --- p.44 / Chapter 2.4.1 --- Methodology, target population and analytical perspective --- p.44 / Chapter 2.4.2 --- Screening test and program performance --- p.47 / Chapter 2.4.3 --- Cost and utility estimation --- p.49 / Chapter 2.4.4 --- Model parameter sources and validation --- p.53 / Chapter 2.4.5 --- Alternatives and identified cost-effective strategies --- p.58 / Chapter 2.5 --- Conclusions --- p.63 / Chapter CHAPTER 3 --- OBJECTIVES --- p.64 / Chapter 3.1 --- General Objectives --- p.64 / Chapter 3.2 --- Alternative cervical cancer screening strategies in this study --- p.64 / Chapter 3.3 --- Decision rules for recommended cost-effective options --- p.65 / Chapter 3.4 --- Analytical perspective and time horizon --- p.65 / Chapter 3.5 --- Objectives --- p.66 / Chapter 3.6 --- Analytical scenario in this study --- p.66 / Chapter 3.6.1 --- Patterns of cervical screening program delivery in rural China --- p.67 / Chapter 3.6.2 --- Demographic profile of the simulated hypothetical cohort --- p.67 / Chapter 3.6.3 --- Summary of model assumptions --- p.68 / Chapter 3.6.3.1 --- Assumptions related to screening performance and clinical practice --- p.68 / Chapter 3.6.3.2 --- Assumptions related to epidemiological characteristics of cervical cancer --- p.68 / Chapter 3.6.3.3 --- Assumptions related to economic evaluation --- p.69 / Chapter CHAPTER 4 --- METHODOLOGY --- p.70 / Chapter 4.1 --- Alternative strategies in this study --- p.70 / Chapter 4.2 --- Markov Model Developments and Applications --- p.72 / Chapter 4.2.1 --- General introduction of Markov Transition Model --- p.72 / Chapter 4.2.2 --- Structure of Markov models --- p.76 / Chapter 4.2.2.1 --- Natural history model of cervical cancer --- p.76 / Chapter 4.2.2.2 --- Structure of Pap smear and Visual Inspection screening models --- p.82 / Chapter 4.2.2.3 --- Structure of precancerous lesion and invasive cancer treatment models --- p.83 / Chapter 4.2.2.4 --- Interaction of the models --- p.85 / Chapter 4.2.3 --- Demographic profile of the hypothetical cohort --- p.86 / Chapter 4.2.4 --- Probabilities --- p.88 / Chapter 4.2.4.1 --- Identification and converting between rate and probability --- p.89 / Chapter 4.2.4.2 --- Initial probabilities --- p.90 / Chapter 4.2.4.3 --- Transition probabilities --- p.91 / Chapter 4.2.5 --- Screening, diagnosis and treatment characteristics --- p.101 / Chapter 4.2.5.1 --- Screening program characteristics --- p.101 / Chapter 4.2.5.2 --- Diagnosis test performance --- p.104 / Chapter 4.2.5.3 --- Precancerous lesions treatment characteristics --- p.104 / Chapter 4.2.5.4 --- Invasive cancer and treatment characteristics --- p.106 / Chapter 4.2.6 --- Model validation --- p.111 / Chapter 4.3 --- Cost data collection --- p.112 / Chapter 4.3.1 --- Perspective of study --- p.112 / Chapter 4.3.2 --- Selection of study sites --- p.113 / Chapter 4.3.3 --- Screening cost data collection --- p.113 / Chapter 4.3.4 --- Treatment cost data collection --- p.115 / Chapter 4.4 --- Cost-utility analysis and cost-effectiveness analysis --- p.117 / Chapter 4.4.1 --- General introduction of these two analyses --- p.117 / Chapter 4.4.2 --- Utility Estimates --- p.118 / Chapter 4.4.3 --- Screening utility and effectiveness evaluation --- p.120 / Chapter 4.4.4 --- Cost-effectiveness and cost-utility analysis method --- p.122 / Chapter 4.5 --- Time horizon and discounting rate --- p.125 / Chapter 4.6 --- Summary of modeling assumptions --- p.126 / Chapter 4.6.1 --- Assumptions related to screening performance and clinical practice --- p.126 / Chapter 4.6.2 --- Assumptions related to epidemiological characteristics of cervical cancer --- p.127 / Chapter 4.6.3 --- Assumptions related to economic evaluation --- p.128 / Chapter 4.7 --- Sensitivity analysis --- p.128 / Chapter 4.8 --- Ethical approval --- p.129 / Chapter CHAPTER 5 --- RESULTS --- p.130 / Chapter 5.1 --- Model validation --- p.130 / Chapter 5.2 --- Cost analysis results --- p.134 / Chapter 5.2.1 --- Screening costs results --- p.134 / Chapter 5.2.2 --- Treatment cost results --- p.136 / Chapter 5.2.3 --- The proportional costs breakdown for different screening strategies --- p.139 / Chapter 5.3 --- Utility estimation results --- p.141 / Chapter 5.4 --- Cost-utility analysis results --- p.144 / Chapter 5.4.1 --- Baseline analysis --- p.144 / Chapter 5.4.2 --- Influence of screening program performance --- p.148 / Chapter 5.4.2.1 --- Coverage of the screening program --- p.148 / Chapter 5.4.2.2 --- Follow up rate and treatment rate of positives --- p.155 / Chapter 5.4.3 --- Influence of screening test performance --- p.159 / Chapter 5.4.4 --- Influence of costs --- p.165 / Chapter 5.4.4.1 --- Influence of screening costs --- p.165 / Chapter 5.4.4.2 --- Influence of treatment costs --- p.168 / Chapter 5.4.5 --- Influence of discounting --- p.171 / Chapter 5.4.6 --- Summary of factors and their influences on the baseline CUA results --- p.174 / Chapter 5.5 --- Cost-Effectiveness analysis results --- p.180 / Chapter 5.5.1 --- Baseline analysis --- p.180 / Chapter 5.5.1.1 --- Life year saved --- p.181 / Chapter 5.5.1.2 --- Cervical cancer mortality reduction --- p.185 / Chapter 5.5.1.3 --- Cervical cancer incidence reduction --- p.187 / Chapter 5.5.1.4 --- Relative risk of CIN and cervical cancer --- p.189 / Chapter 5.5.1.5 --- Effectiveness summary of alternative screening strategies on the hypothetical 100,000 rural Chinese women --- p.191 / Chapter 5.5.2 --- Factors that influence the CEA results --- p.195 / Chapter 5.5.2.1 --- Best scenario analysis --- p.196 / Chapter 5.5.2.2 --- Worst scenario analysis --- p.201 / Chapter 5.5.2.3 --- Summary of the possible ranges of costs and effectiveness in different scenarios --- p.206 / Chapter 5.6 --- Sensitivity analysis --- p.209 / Chapter 5.6.1 --- Sensitivity analysis of Cost-Utility analysis results --- p.209 / Chapter 5.6.1.1 --- Tornado analysis --- p.209 / Chapter 5.6.1.2 --- One-way sensitivity analysis --- p.213 / Chapter 5.6.2 --- Sensitivity analysis of Cost-Effectiveness analysis results --- p.220 / Chapter 5.6.2.1 --- Tornado analysis --- p.220 / Chapter 5.6.2.2 --- One-way sensitivity --- p.224 / Chapter 5.6.3 --- Summary of sensitivity results --- p.236 / Chapter CHAPTER 6 --- SUMMARY, DISSICUSSION AND CONCLUSIONS --- p.240 / Chapter 6.1 --- Summary of Markov model development and validation --- p.240 / Chapter 6.1.1 --- Category and source summary of input parameters --- p.240 / Chapter 6.1.2 --- Model validation --- p.244 / Chapter 6.2 --- Summary of modeling results --- p.245 / Chapter 6.2.1 --- Summary of Cost-Utility Analysis --- p.245 / Chapter 6.2.1.2 --- Baseline analysis findings --- p.245 / Chapter 6.2.1.2 --- Influential factors on the cost-effective manner of alternative strategies --- p.246 / Chapter 6.2.2 --- Summary of Cost-Effectiveness Analysis --- p.250 / Chapter 6.2.2.1 --- Baseline analysis findings --- p.251 / Chapter 6.2.2.2 --- Possible ranges for cost and effectiveness of alternative strategies under different scenarios --- p.253 / Chapter 6.2.3 --- Summary of CUA and CEA findings --- p.257 / Chapter 6.2.4 --- Summary of sensitivity analysis --- p.259 / Chapter 6.2.4.1 --- Important variables on health outcome predictions --- p.259 / Chapter 6.2.4.2 --- Sensitive variables to the baseline CUA and CEA recommendations --- p.260 / Chapter 6.2.4.3 --- Overview of the sensitivity analysis --- p.263 / Chapter 6.3 --- Discussion --- p.264 / Chapter 6.3.1 --- Alternative strategies of cervical cancer screening in rural China --- p.264 / Chapter 6.3.1.1 --- Target ages --- p.265 / Chapter 6.3.1.2 --- Screening intervals --- p.266 / Chapter 6.3.1.3 --- Feasible primary screening tests --- p.267 / Chapter 6.3.1.4 --- Service delivering patterns --- p.269 / Chapter 6.3.1.5 --- Time horizon of this thesis study --- p.270 / Chapter 6.3.2 --- Transition probability estimation --- p.271 / Chapter 6.3.3 --- Screening and treatment cost estimation --- p.276 / Chapter 6.3.3.1 --- Representativeness of the selected counties --- p.276 / Chapter 6.3.3.2 --- Screening costs of VIA and Pap smear --- p.277 / Chapter 6.3.3.3 --- Treatment costs --- p.279 / Chapter 6.3.4 --- Utility estimation --- p.280 / Chapter 6.3.4.1 --- Instrument selection --- p.280 / Chapter 6.3.4.2 --- Utility estimation between studies --- p.281 / Chapter 6.3.5 --- Baseline cost-utility and cost-effectiveness analyses --- p.283 / Chapter 6.3.6 --- Sensitivity Analysis --- p.284 / Chapter 6.3.7 --- Strengths and limitations --- p.286 / Chapter 6.3.7.1 --- Limitations --- p.286 / Chapter 6.3.7.2 --- Strengths --- p.288 / Chapter 6.4 --- Policy implications --- p.289 / Chapter 6.4.1 --- How to manage a cost-effective cervical cancer screening program? --- p.289 / Chapter 6.4.2 --- How can VIA screening be adopted? --- p.290 / Chapter 6.4.3 --- How can Pap smear screening be adopted? --- p.291 / Chapter 6.4.4 --- Framework for policy decision making --- p.292 / Chapter 6.5 --- Conclusions --- p.295 / Chapter APPENDIX --- p.300 / Chapter Appendix 1-1 --- The 2001 Bethesda System* --- p.300 / Chapter Appendix 1-2 --- The FIGO Staging for cervical cancers* --- p.301 / Chapter Appendix 1-3 --- Cervical Cancer Screening Program in different countries --- p.302 / Chapter Appendix 4-1 --- WHO World Standardized Population Distribution (%) --- p.305 / Chapter Appendix 4-2 --- Summary of transition probabilities literature review --- p.306 / Chapter Appendix 4-3 --- Price Indices from 1978 to 2010 --- p.326 / Chapter Appendix 4-4 --- Screening Cost Questionnaire --- p.327 / Chapter Appendix 4-5 --- Programmatic Cost Survey Questionnaire --- p.339 / Chapter Appendix 4-6 --- Treatment Cost Survey Questionnaire --- p.342 / Chapter Appendix 4-7 --- EQ-5D Algorism (UK) --- p.344 / Chapter Appendix 4-8 --- Chinese Version of EQ5D----HQOL score questionnaire --- p.345 / Chapter Appendix 5-1 --- Calibrated variables and its final settings --- p.348 / Chapter Appendix 5-2 --- Cervical cancer new cases and deaths all over the world in 2008 --- p.349 / Chapter Appendix 5-3 --- Data distribution of CIN2-3 and cervical cancer treatment costs --- p.350 / Chapter Appendix 5-4 --- Relative risk of CIN and cervical cancer by age groups of alternative screening strategies --- p.361 / Chapter Appendix 5-5 --- Influence of discounting rate of life years on the CEA results --- p.363 / Chapter Appendix 5-6 --- Tornado analysis results based on the effect on QALYs predictions --- p.367 / Chapter Appendix 5-7 --- Tornado analysis results based on the effect on life-year predictions --- p.372 / Chapter Appendix 6-1 --- Summary of Markov Model Inputs and Sources --- p.377 / REFERENCE --- p.388

A model was developed in this study to understand women's cervical screening behaviour. It revealed that the interaction among institutional factors, risk appraisal, coping appraisal, and health beliefs and cultural factors contributed to the complex nature of screening behaviour among Chinese women. The institutional component provided the contextual factors within which women perceived the risk of cervical cancer, perceived the practice of cervical screening, and decided to take or not to take cervical screening. Risk appraisal provided the premise factor that induces women to seek coping strategies to reduce or remove the risk. During the process of coping appraisal, women's motivation to have cervical screening could be increased or decreased as the perceived benefits and costs of screening interacted with each other. The importance of the women's health beliefs and cultural factors was reflected in the way that they were affected by their notions of health behaviour and their cultural beliefs about cervical cancer risk and cervical screening participation. Commitment to participate in screening was a reinforcing factor inducing women to take up an offer of cervical screening. / Aim: To explore the knowledge and the perception of the risk of cervical cancer, identify the factors determining cervical screening behaviour, and develop a model to understand cervical screening behaviour among women in mainland China. / Background: Cervical cancer is the most common type of cancer, and is the second most common cause of cancer death in women in mainland China. Cervical screening is the most important intervention for the secondary prevention of cervical cancer. Theories of health behaviour and empirical research highlight risk perception as a significant factor motivating people to opt for cancer screening. However, little is known about the risk perception of cervical cancer and the factors influencing the screening participation of women in mainland China. / Conclusion: This study provides evidence of the complex factors influencing cervical screening behaviour and contributes new knowledge to the understanding of cervical screening behaviour within the Chinese cultural context. It further informs programmes for the promotion of cervical screening among this population. / Methods: A mixed method design consisting of two phases was used, employing both quantitative and qualitative methods of data collection. First, a cross-sectional survey was conducted to collect a baseline assessment of women's knowledge of cervical cancer and screening, their perceptions of the risk of cervical cancer, and the relationship between these factors and their cervical screening behaviour. Findings from this phase also guided the purposive sampling of participants in phase two. / Results: The findings from phase one demonstrated that the availability of an organized screening programme was a major motivator for women to opt for cervical screening. Multivariate analysis shows that having children (OR=2.57, p=0.026), a perception that visiting doctors regularly is important for health (OR-2.66, p=0.025), average (OR-4.84, 1)=0.006) and high levels of knowledge about cervical screening (OR-9.66, p=0.001) were significantly associated with having been screened in the previous three years. / Then in phase two, qualitative research was conducted using semi-structured interviews of 27 women, 16 of whom had been screened and 11 had not. The interview structure was based on an initial analysis of the data from phase one and from a review of the related literature. The data from the interviews were analyzed using latent content analysis, involving an interpretative reading of the symbolism underlying the surface structure in the text. The audio recordings of the interviews were transcribed verbatim in Chinese, and then the key phrases which were important for the objectives of the study were identified. The key phrases and words were grouped according to their commonality of meaning. Then, these groups of data were sorted and classified to create categories and sub-categories, which were mutually exclusive, explicit and accurate without overlapping. / Two themes emerged from the qualitative data from phase two. Theme I was that perceptions of cervical cancer and cervical screening included five categories: the perceived effects of suffering from cervical cancer; the perception of cervical screening; a lack of understanding about cervical cancer and screening; the perceived risk of cervical cancer; and factors related to the cultural beliefs system. Theme II was that the institutional and health care practitioner system included two categories: availability of an organised physical examination programme and the role of the health care practitioner in encouraging cervical screening utilization. / Gu, Can. / Adviser: Chan, Carmen. / Source: Dissertation Abstracts International, Volume: 73-02, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 244-267). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.