鄰里效應對於個人飲酒、吸菸、嚼檳榔與成癮藥物使用之影響：台灣人口代表性樣本之多階層分析

Abstract

研究背景：過去許多研究指出個人的飲酒、吸菸和藥物使用行深受居住環境鄰里所影響，然而，少有研究以系統性的方式檢視不同的物質與不同的使用模式與鄰里效應之間的關係，並且過往多數的研究是以西方國家為主，較少探討非西方地區環境的影響。 研究目的：以台灣全國代表性樣本的調查資料進行分析以探討飲酒、吸菸、嚼檳榔和成癮藥物/吸入劑使用與個人因子和社會環境鄰里效應之間的關聯，並比較不同物質使用模式與鄰里效應之間的關連。 研究方法：個人層次資料來源為103年全國物質使用調查中18歲到64歲的成人樣本，共計13,392人，分布於168個鄉鎮市區。鄰里層次資料以鄉鎮市區為分析單位，採用2010年全國普查和相關社會經濟指標資料，以探索性因素分析萃取出三項鄰里層次變數：弱勢、身心障礙和老年人群聚程度、鄰里富裕程度，以及社會零碎化程度，以多階層羅吉斯模型分別針對飲酒（危害性飲酒、低風險飲酒、未飲酒）、抽菸（尼古丁依賴、非依賴經常性使用、未抽菸）、嚼檳榔（有/無）和成癮藥物/吸入劑使用(有/無)，與鄰里層次之間的關聯性進行分析，最後加入性別的跨層次因素，探討鄰里效應對於個人物質使用行為之影響是否因為男女而有所不同。 研究結果：鄰里效應中的弱勢、身心障礙和老年人群聚程度越高，尼古丁依賴吸菸和檳榔使用的勝算較高；鄰里富裕程度越高則對於低風險的飲酒行為有正向影響；居住在社會零碎化程度較高的地區，則個人的危害性飲酒、尼古丁依賴抽菸、非依賴經常性吸菸和嚼檳榔的勝算較高，本研究並未發現鄰里效應對於成癮藥物/吸入劑之影響。此外，在性別的跨層次交互作用部分，發現女性受鄰里環境影響較男性強烈，特別是在社會零碎化程度與尼古丁依賴吸菸和檳榔使用，以及弱勢、身心障礙和老人群聚程度與檳榔使用的相關性。 結論：在台灣的成人群體中，不同的鄰里環境效應對於不同種類的物質使用，以及相異的物質使用形態有不同程度的關聯性。

Background: A growing body of research indicates the effect of neighbourhood characteristics on alcohol, tobacco, and substance use behaviours; however, few studies have systematically examined the difference in patterns across different substances. We have investigated the association of neighbourhood socioeconomic indicators with alcohol, tobacco, betel nut and illicit drugs/inhalants use in Taiwan Methods: Data were drawn from the adult sample aged 18-64 years (N=13,392) across 168 townships in the 2014 National Survey on Substance Use. Township-level socioeconomic data were from the 2010 Census and factor analysis was used to derive three neighbourhood factors: concentrations of disadvantaged, disable, and elderly people; neighbourhood affluence; and social fragmentation. Multilevel logistic regression models were used to examine the association of neighbourhood factors with the use of alcohol (harmful use, low risk use, or non-use), tobacco (nicotine dependence, non-dependent regular use, or non-use), betel nut (yes/no use), and illicit drugs/inhalants (yes/no use) separately, before and after controlling for individual-level characteristics. The cross-level interactions between sex and neighbourhood factors were also examined. Results: Neighbourhood concentrations of disadvantaged, disabled, and elderly people was associated with higher odds of nicotine dependence and betel nut use after controlling for individual-level characteristics. Neighbourhood affluence was associated with higher odds of low risk alcohol use compared to non-use. Social fragmentation was associated with harmful drinking, nicotine dependence, non-dependent regular smoking, and betel nut chewing. There was no statistical evidence for an association of illicit drugs/inhalants use with the three neighbourhood factors. There was some evidence for sex differences in neighbourhood effects: the associations appeared to be stronger in women than men between social fragmentation and nicotine dependence / betel nut use, and between concentrations of disadvantaged, disabled, and elderly people and betel nut use. Conclusions: Different neighbourhood factors were associated with the use of different types of substances and the pattern of use behaviours among adults in Taiwan.