Collaborative longitudinal research on alcohol problems.
Collaborative longitudinal research on alcohol problems.
复制标题
关于酒精问题的合作纵向研究。
DOI：
10.1111/j.1360-0443.1988.tb00493.x
复制
发表时间：
1988
期刊：
影响因子：
--
通讯作者：
Temple,MT
中科院分区：
文献类型：
--
作者：
Fillmore,KM;Grant,M;Hartka,E;Johnstone,BM;Sawyer,SM;Speiglman,R;Temple,MT
纵向研究设计构成了描述和预测整个生命过程中饮酒模式和问题的最有力策略。尽管自 20 世纪 20 年代以来已经对非临床人群进行了近 100 项纵向研究，但它们的潜在价值从未得到充分认识。在这个研究领域，人们很少关注跨时间和文化界限复制这些研究结果。现在已经开始尝试纠正这种不平衡。本文描述了一个新的研究项目，旨在评估不同时间段和文化条件的研究结果相互复制的程度。这种复制将通过使用荟萃分析提供的研究综合协调框架重新分析现有的多个纵向数据集来确定。荟萃分析是一组统计技术的总称，旨在确定不同研究的研究结果的重复程度。如果研究结果没有被重复，荟萃分析提供了工具来检查差异的潜在解释（Hedges & Olkin，1985）。该项目解决了一系列研究问题，这些问题对于当代酒精研究至关重要，并且对与酒精滥用有关的社会和健康计划的发展具有广泛的影响。在这一领域开展工作的最初建议源于对研究优先事项的广泛讨论，该讨论是在 1984 年医学研究咨询委员会生物行为科学和心理健康小组委员会工作组在日内瓦举行会议期间进行的。该研究项目源于世界卫生组织心理健康计划委托对世界酒精相关纵向文献的回顾（Fillmore，1985）。该评论指出，尽管许多纵向研究具有非常相似的测量结果，但很少有人努力综合各个研究的研究结果。因此，任何给定的发现仍然可能受到历史测量时期、受试者出生队列成员身份和文化条件的影响的影响。该评论认为，综合这些文献的努力将是确定研究结果可推广程度的唯一手段，或者相反，确定研究结果受历史或文化条件影响的程度。
Longitudinal research designs constitute the most powerful strategy for describing and predicting drinking patterns and problems across the life course. Although almost 100 longitudinal studies of non-clinical population have been undertaken since the 192O's, their potential value has never been fully realized. Too little attention in this research arena has been paid to the replication of findings across the temporal and cultural boundaries within which these studies were conducted. An attempt has now begun to redress that imbalance. This paper describes a new research project which seeks to assess the extent to which study findings from different temporal periods and cultural conditions replicate each other. This replication will be determined by re-analyzing existing multiple longitudinal data sets using the co-ordinated framework for research synthesis provided by meta-analysis. Meta-analysis is the general designation for a body of statistical techniques developed to determine the degree to which research findings from different studies replicate. If findings are not replicated, meta-analysis provides tools to examine potential explanations for differences (Hedges & Olkin, 1985). The project addresses a series of research questions which are critical in contemporary alcohol research and which have broad implications for the development of social and health programs pertaining to alcohol abuse. The original suggestion to undertake work in this area arose from extensive discussions on research priorities, which took place on the occasion of the meeting in Geneva in 1984 of a task force of the Advisory Committee on Medical Research subcommittee on biobehavioural science and mental health. This research project had its origins in a review of the world's alcohol-related longitudinal literature commissioned by the Mental Health Program of the World Health Organization (Fillmore, 1985). The review noted that although many longitudinal studies possess quite similar measurements, there has been little effort to synthesize research findings across studies. As a result, any given finding remains potentially confounded by the effects of historical period of measurement, birth cohort membership of subjects, and cultural conditions. The review argued that an effort to synthesize this literature would be the only means to determine the degree to which research findings are generalizable or, by contrast, the degree to which research findings are mediated by historical or cultural conditions.