社區整合式篩檢世代之創新擴散研究

Abstract

背景：癌症與慢性病對人類健康帶來極大的威脅，早期篩檢可以達到降低罹病率與死亡率的目的。然而，無法提高民眾的篩檢涵蓋率及意願，是社區護理師所面臨最大的問題。而建立於1999年至2009年的「基隆社區整合式篩檢世代（Keelung Community-based Integrated Screening Cohort, KCIS）」強調以家庭為基礎整合六項癌症與六項慢性病一次完成的篩檢，相較於過去的單一疾病高危險群篩檢，強調能整合多項疾病篩檢及提供衛生教育、與篩檢後直接轉介的優勢，正是一項創新的健康促進策略。 目的：本論文研究主旨在應用創新擴散理論 (Diffusion of Innovation Theory, DOI Theory）探討KCIS創新之擴散經歷。創新擴散理論是從宏觀的角度以社區政策推廣的立場來達到行為改變的理論，因此透過本研究可以瞭解此創新服務的擴散過程。所以本論文之研究目的在於瞭解創新的KCIS策略在十年間如何被民眾接受與擴散；透過篩檢世代資料分析，探討影響社區民眾對於接受此健康創新篩檢模式的時間屬性因素、個人因素與代謝症候群和糖尿病的疾病因素間之關係。 方法：本研究共計納入79,489位社區民眾進行分析，透過創新擴散理論的S型時間曲線，將民眾首次接受KCIS篩檢世代時間進行分組，就篩檢世代之各項問卷資料、篩檢結果與生理數據為依據進行研究。將以各種統計分析模式進行流行病學分析，包括：連續變項的描述性統計，並應用迴歸模式分析各種變數與社區整合式創新篩檢間之關係，以及Cox proportional hazards regression model估計不同時間點接受篩檢民眾的特質差異，再以Kaplan-Meier method估計不同的接受時間點分組的差異，此外在應用以加速失敗時間模式（Accelerated failure time model , AFT Model）之Weibull分佈模式估算參加篩檢時間中位數，調整其他相關因素後估計首次參加社區篩檢之中位數時間（median time）。 結果:此創新擴散在十年間篩檢涵蓋率由1999年的4.2%逐年上升至2009年的84.3%。女性（aHR = 1.23, 95% CI:1.21-1.25）接受早於男性、年長者（aHR = 1.012, 95% CI:1.012-1.013）接受早於年輕者，已婚者（aHR = 1.33, 95% CI:1.30-1.36）、低教育程度者（aHR = 1.15, 95% CI:1.12-1.18）早於高教育程度者，健康生活型態健康者，包括：不吸菸（aHR = 1.16, 95% CI:1.14-1.17）、不喝酒（aHR =1.10, 95% CI:1.08-1.12）、有規律運動（aHR = 1.36, 95% CI:1.29-1.35）者早於沒有者。罹患代謝症候群（aHR = 1.34, 95% CI: 1.31-1.36）、糖尿病（aHR = 1.16, 95% CI: 1.12-1.21）等疾病會阻礙民眾接受創新篩檢服務。沒有罹患代謝症候群者會提早0.82年接受創新篩檢，疾病嚴重度最高者與健康者平均接受創新時間差距有高達2.25年。 結論：創新的KCIS整合式篩檢計畫成功的在十年內擴散，促進了民眾對於自身健康的重視，個人因素、不良生活型態與罹患慢性病會影響民眾接受創新。在進行健康政策的創新擴散時，本結果建議護理師可以應用此理論擴散創新的健康促進策略，應針對不同時程介入、從個人逐步擴散到家庭及至社區、關懷不同性別年齡的受檢者的需求進行篩檢，建檔管理有健康問題的群眾，優先邀請規律篩檢並提供疾病相關衛生教育，達到社區民眾行為改變之目的。

Background: Because the risk of cancer and chronic diseases are increasing, early health screening can reduce morbidity and mortality. However, developing strategies to improve the screening coverage rate and increase willingness to adopt screening programmes is challenging for public health nurses. The Keelung Community-based Integrated Screening (KCIS), a new health promotion policy, integrated six cancers and six chronic diseases in a screening programme, which was conducted between 1999 and 2009. The KCIS is an innovative health promotion strategy for community populations because it combines outreach screening, health education, and a direct referral system. Aims: In this study, we applied the diffusion of innovation (DOI) theory, a macro-level human behaviour change theory, and analysed participants who were enrolled in the KCIS programme between 1999 and 2009 to investigate the relationship between time to adoption KCIS programme and demographic characteristics, lifestyle factors, and diabetes- and metabolic syndrome-related factors. Methods: A total of 79,489 participants participated in the KCIS programme between 1999 and 2009. According to the definition of the DOI theory, the ‘S-time shaped’ curve was used to categorise the participants on the basis of their time to adoption. The Kaplan–Meier method was used to plot the curve for the time of enrolment to estimate time in each category. The Cox proportional hazards regression model was used to calculate hazard ratios (HRs) for the time of enrolment in the KCIS with respect to demographic characteristics, lifestyle factors, and diabetes- and metabolic syndrome-related factors. In addition, the Weibull distribution method was used for the accelerated failure time model to estimate the median time of enrolment. Results: The KCIS coverage rate increased from 4.2% in 1999 to 84.3% in 2009. After adjustment for all variables, demographic factors such as age (aHR = 1.012, 95% CI = 1.012, 1.013), sex (aHR = 1.23, 95% CI = 1.21–1.25), low education level (aHR = 1.15, 95% CI = 1.12, 1.18), and early marriage (aHR = 1.32, 95% CI = 1.30, 1.36) as well as lifestyle factors such as not smoking (aHR = 1.16, 95% CI = 1.14–1.17), not consuming alcohol (aHR = 1.10, 95% CI = 1.08–1.12), and regular exercise (aHR = 1.36, 95% CI = 1.34–1.38) were significantly associated with early enrolment in the programme. The rate of early enrolment was higher in the participants without diabetes (aHR = 1.16, 95% CI = 1.12–1.21) and metabolic syndrome (aHR = 1.34, 95% CI = 1.341–1.36) than in those with diabetes and metabolic syndrome. Compared with the participants with metabolic syndrome, the adjusted median time of enrolment of the participants without metabolic syndrome was 0.82 years earlier. Furthermore, compared with the participants with severe metabolic syndrome, the adjusted median time of enrolment of the participants without metabolic syndrome was up to 2.25 years earlier. Conclusions: The innovative KCIS programme had successful diffusion within a decade and improved screening adoption behaviour. Demographic characteristics, unhealthy lifestyle, and chronic disease diagnosis affected the time of enrolment in the programme. The results suggest that nurses should apply the DOI theory to disseminate the health promotion strategy from the individual to the community level. Moreover, nursing intervention should be provided in different time zones and should be disseminated from the personal to the family level rather than the community level, with a focus on the specific screening needs of people in different age groups and of different sexes. The accurate documentation of screening results, early invitation to regular screenings, and health education should be provided for patients with chronic diseases.