第五章 討論與結論
5.6 結論
當納入的試驗有偏差時，對於不同的網絡結構的影響程度不同。越是密集的 網絡受到的影響越低，偏差的效果會較為均勻的由每個估計值承擔，三種模型的
46
結果越接近。因此不論使用何種模型，結果差異不會太大。而越稀疏的網絡結 構，三種模型結果越容易根據模型的特性產生差異，特別是當網絡結構中有單獨 分支(例如圖 4.1(c)Network 3 的結構)。Lu & Ades model 依賴分支上的試驗結。若 這些試驗有偏差，偏差效果無法分散出去。研究者在結果解釋上需特別注意，避 免下錯誤的結論。
三種模型的特色中，Lu & Ades model 偏差主要影響排序靠後的治療。
Baseline model 在選擇的參考治療沒有偏差的情況下，偏差主要影響實際有偏差 的治療的估計值。少部分的偏差影響會透過參考治療散布出去，而影響到整個網 絡。散布出去的影響主要會在排序靠後的治療上觀察到。而Arm-based model 在 假設治療彼此獨立的情況下，偏差的效果只會反應在偏差治療的估計值上，其餘 的完全不受影響。
47
參考文獻
1. Guyatt, G., Cairns, J., Churchill, D., and et al. (1992). Evidence-based medicine: A new approach to teaching the practice of medicine. JAMA 268, 2420-2425.
2. Uman, L.S. (2011). Systematic Reviews and Meta-Analyses. Journal of the Canadian Academy of Child and Adolescent Psychiatry 20, 57-59.
3. Efthimiou, O., Debray, T.P., van Valkenhoef, G., Trelle, S., Panayidou, K., Moons, K.G., Reitsma, J.B., Shang, A., and Salanti, G. (2016). GetReal in network meta-analysis: a review of the methodology. Res Synth Methods 7, 236-263.
4. Biondi-Zoccai, G. (2016). Umbrella Reviews: Evidence Synthesis with Overviews of Reviews and Meta-Epidemiologic Studies.(Springer International Publishing).
5. Zhang, J., Fu, H., and Carlin, B.P. (2015). Detecting outlying trials in network meta-analysis. Stat Med 34, 2695-2707.
6. Mills, E.J., Bansback, N., Ghement, I., Thorlund, K., Kelly, S., Puhan, M.A., and Wright, J. (2011). Multiple treatment comparison meta-analyses: a step forward into complexity. Clinical Epidemiology 3, 193-202.
7. Pannucci, C.J., and Wilkins, E.G. (2010). Identifying and Avoiding Bias in Research.
Plastic and reconstructive surgery 126, 619-625.
8. Higgins, J.P., and Thompson, S.G. (2002). Quantifying heterogeneity in a meta-analysis. Stat Med 21, 1539-1558.
9. Higgins, J.P.T., Thompson, S.G., Deeks, J.J., and Altman, D.G. (2003). Measuring inconsistency in meta-analyses. BMJ : British Medical Journal 327, 557-560.
10. Haidich, A.B. (2010). Meta-analysis in medical research. Hippokratia 14, 29-37.
11. Lumley, T. (2002). Network meta-analysis for indirect treatment comparisons. Stat Med 21, 2313-2324.
12. Jansen, J.P., Fleurence, R., Devine, B., Itzler, R., Barrett, A., Hawkins, N., Lee, K., Boersma, C., Annemans, L., and Cappelleri, J.C. (2011). Interpreting indirect treatment comparisons and network meta-analysis for health-care decision making: report of the ISPOR Task Force on Indirect Treatment Comparisons Good Research Practices: part 1. Value in health : the journal of the International Society for Pharmacoeconomics and Outcomes Research 14, 417-428.
13. Dias, S., Welton, N.J., Sutton, A.J., Caldwell, D.M., Lu, G., and Ades, A.E. (2013).
Evidence synthesis for decision making 4: inconsistency in networks of evidence based on randomized controlled trials. Medical Decision Making 33, 641-656.
14. Lu, G., and Ades, A.E. (2006). Assessing Evidence Inconsistency in Mixed Treatment Comparisons. Journal of the American Statistical Association 101,
48
447-459.
15. Dias, S., J., W.N., M., C.D., and E., A.A. (2010). Checking consistency in mixed treatment comparison meta‐analysis. Statistics in Medicine 29, 932-944.
16. Higgins, J.P., Jackson, D., Barrett, J.K., Lu, G., Ades, A.E., and White, I.R. (2012).
Consistency and inconsistency in network meta-analysis: concepts and models for multi-arm studies. Res Synth Methods 3, 98-110.
17. Lin, L., Chu, H., and Hodges, J.S. (2016). Sensitivity to excluding treatments in network meta-analysis. Epidemiology (Cambridge, Mass) 27, 562-569.
18. Zhang, J., Chu, H., Hong, H., Virnig, B.A., and Carlin, B.P. (2017). Bayesian hierarchical models for network meta-analysis incorporating nonignorable missingness. Statistical methods in medical research 26, 2227-2243.
19. Egger, M., Davey Smith, G., Schneider, M., and Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. BMJ : British Medical Journal 315, 629-634.
20. Lu, G., and Ades, A.E. (2004). Combination of direct and indirect evidence in mixed treatment comparisons. Stat Med 23, 3105-3124.
21. Dias, S., Welton, N.J., Sutton, A.J., and Ades, A.E. (2013). Evidence Synthesis for Decision Making 5: The Baseline Natural History Model. Medical Decision Making 33, 657-670.
22. Zhang, J., Carlin, B.P., Neaton, J.D., Soon, G.G., Nie, L., Kane, R., Virnig, B.A., and Chu, H. (2014). Network meta-analysis of randomized clinical trials:
reporting the proper summaries. Clinical trials (London, England) 11, 246-262.
23. Lu, G., Welton, N.J., Higgins, J.P., White, I.R., and Ades, A.E. (2011). Linear inference for mixed treatment comparison meta-analysis: A two-stage approach.
Res Synth Methods 2, 43-60.
24. Hong, H., Chu, H., Zhang, J., and Carlin, B.P. (2016). A Bayesian missing data framework for generalized multiple outcome mixed treatment comparisons. Res Synth Methods 7, 6-22.
25. Gatsonis, C., and Morton, S. (2017). Methods in Comparative Effectiveness Research.(New York: Chapman and Hall/CRC).
26. Dias, S., and Ades, A.E. (2016). Absolute or relative effects? Arm‐based synthesis of trial data. Res Synth Methods 7, 23-28.
27. van Valkenhoef, G., Dias, S., Ades, A.E., and Welton, N.J. (2016). Automated generation of node‐splitting models for assessment of inconsistency in network meta‐analysis. Research Synthesis Methods 7, 80-93.
49