以計劃行為理論為基礎之社群媒體性行為衛教介入：臺灣大學生運動員之隨機分派實驗

Social Media-Delivered Sexual Health Intervention Using the Theory of Planned Behavior: A Randomized Controlled Trial among College Athletes in Taiwan

張巧柔(Chiao-Lou Chang)

指導教授 ： 黃俊豪

國立臺灣大學/公共衛生學院/健康政策與管理研究所/碩士(2017年)

https://doi.org/10.6342/NTU201701092

摘要

目標：以計劃行為理論 (Theory of Planned Behavior, TPB) 為基礎設計危險性行為預防策略，透過Facebook對大學生運動員實際進行衛教介入，並進行效果評價，期能降低其從事危險性行為之行為意圖，並檢視影響後測時無套性交意圖改變之預測因子。
方法：招募知情同意之臺灣大學生運動員 (共109人)，以隨機分派實驗研究法，分成實驗組 (57人) 與對照組 (52人)，進行八週介入，分別為TPB為基礎之性健康衛教，以及一般健康議題之衛教，並於介入前及介入完成之二個月後，運用網路匿名問卷，收集其背景特質、性行為相關之TPB構念，及無套性交意圖之資料，以多變項線性自迴歸與多變項羅吉斯迴歸，進行統計分析。
結果：實驗介入之有效樣本數為48人 (實驗組24人、對照組24人)。雖實驗組相較於對照組，於校正前測資料後，在後測時並無顯著較低之無套性交意圖，然本研究發現，兩組於後測時，均顯著受前測時無套性交意圖 (β=.39–.50) 影響。此外，亦發現年級越高 (β=.38)、男性 (β=.42; AOR=.05–.07) 與非異性戀 (AOR=.06–.09)，均為影響後測時無套性交意圖之重要影響因子。於TPB構念的部分，發現不論是實驗組或是對照組，若其在前後測期間之正面行為結果評價轉趨負向 (β=-.27 ; AOR=25.23)，及有利情境下之自覺控制轉趨負向 (β=-.32)，則後測時之無套性交意圖較有顯著降低的可能，而主觀規範轉趨負向者 (β=.27)，其後測時之無套性交意圖反而會顯著上升。
結論：本隨機分派實驗研究，研發以TPB為基礎，適合大學生族群之性健康衛教介入方案，並透過社群媒體實際執行以評價其效果，此研究設計與進行方式，乃全新之嘗試，雖然組別效果皆未達統計顯著，但其數值所呈現之方向，均與研究預期之方向相同，有待未來較大樣本之實驗介入進一步驗證。值得一提的是，本研究發現，性別、性取向、年級與TPB構念之改變 (正向行為結果評價、有利情境下之自覺控制、主觀規範)，為影響後測時無套性交意圖降低的重要預測因子，此實證資訊，可做為未來性行為衛教策略發展之基礎，以優化其介入成效，並促進此年輕族群之性健康。

關鍵字

計劃行為理論 ； 大學生運動員 ； 無套性交意圖 ； 隨機分派實驗 ； 社群媒體 ； Facebook

並列摘要

Objectives: With a “dual role” and related stress, college athletes have been shown to have higher prevalence of health risk behaviors (e.g., sexual risk-taking) than their non-athlete peers. Rapid advances in social media have popularized Facebook use for communication in youth. Therefore, based on the Theory of Planned Behavior (TPB), this study developed a Facebook-delivered health intervention aiming to reduce condomless sex intention (CSI), assessed the intervention effectiveness employing a randomized controlled trial (RCT), and examined the predictors of change in CSI.
Methods: A total of 109 athletes consented to participate in an 8-week Facebook intervention to reduce CSI, with 57 and 52 randomly assigned to the experimental and control group, respectively. The experimental group participants received a TPB-based sexual health intervention via a Facebook fan page, whereas the control group participants were offered general health education information via Facebook as well. Pretest was conducted before the intervention began and posttest was conducted 8 weeks after the intervention was completed, using an anonymous online questionnaire to collect background characteristics, TPB constructs concerning CSI, and CSI among these college athletes. Multivariate linear autoregression and multivariate logistic regression were both used to evaluate the intervention effect.
Results: In total, 48 college athletes (24 in the experimental group and 24 in the control group) completed both pretest and posttest. While no statistically significant intervention effect to reduce CSI was found at 8-week posttest after controlling for the pretest, we found that, regardless of whether participants were in the experimental or control group, higher CSI at pretest predicted greater CSI at 8-week posttest (β=.39–.50). Further, more senior (β=.38), male (β=.42; AOR=.05–.07), and non-heterosexual (AOR=.06–.09) athletes were less likely to have reduced CSI at 8-week posttest. In addition, negative changes toward positive behavioral outcome evaluations (β=-.27; AOR=25.23) and negative changes toward perceived behavioral control under facilitating conditions (β=-.32) both predicted decreased CSI, whereas negative changes toward supportive subjective norms (β=.27) predicted increased CSI.
Conclusions: This RCT was designed to evaluate the effectiveness of the TPB-based sexual health intervention via Facebook among college athletes. To our knowledge, this theory-based RCT delivered through social media to reduce CSI was among the first of its kind. Although the intervention effect was not statistically significant, the directions of the parameter estimates derived from this study were as anticipated. Future studies on a larger scale are needed to confirm the effectiveness of such interventions. Notably, this study found that male gender, non-heterosexual orientation, year in university, and TPB construct changes (positive behavioral outcome evaluations, perceived behavioral control under facilitating conditions, and supportive subjective norms) were significant predictors of CSI at 8-week posttest. These findings can inform the development of future sexual health education strategies for optimal effectiveness to promote the sexual health of this young population.

並列關鍵字

Theory of Planned Behavior ； College athletes ； Condomless sex intention ； Randomized controlled trial ； Social media ； Facebook

參考文獻

張錦治、王瑞霞. (2007). 有性經驗大學生保險套使用行為及相關因素之探討－－跨理論模式之應用. 實證護理, 3(4), 261-270. doi: 10.6225/JEBN.3.4.261

晏涵文, 劉潔心, 李思賢, & 馮嘉玉. (2009). 臺灣國小、國中、高中職學校性教育成果調查研究. [A Study of the Outcomes on Sexuality Education of Elementary, Junior High, Senior High and Vocational Schools in Taiwan]. 臺灣性學學刊, 15(2), 65-80.

晏涵文, 劉潔心, & 馮嘉玉. (2009). 青少年網路交友與約會、婚前性行為影響因素探討. [Influential Factors in Internet friend-making, Dating and Premarital Sexual Behaviors of Adolescents]. 臺灣公共衛生雜誌, 28(4), 322-333.

梁曉藍, 黃俊豪. (2013). 台灣女同志之指險套使用盛行率與相關因素探討. 臺灣公共衛生雜誌，32(1)，31-41.

Ajzen, Icek. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179-211. doi: 10.1016/0749-5978(91)90020