Characteristics of recruited licensed premises.

Background
Interventions that alter aspects of the physical environments in which unhealthy behaviours occur have the potential to change behaviour at scale, i.e., across populations, and thereby decrease the risk of several diseases. One set of such interventions involves reducing serving sizes, which could reduce alcohol consumption. The effect of modifying the available range of serving sizes of wine in a real-world setting is unknown. We aimed to assess the impact on the volume of wine sold of removing the largest serving size by the glass from the options available in licensed premises.
Methods and findings
The study was conducted between September 2021 and May 2022 in 21 licensed premises in England that sold wine by the glass in serving sizes greater than 125 ml (i.e., 175 ml or 250 ml) and used an electronic point of sale till system. It used an A-B-A reversal design, set over 3 four-weekly periods. “A” represented the nonintervention periods during which standard serving sizes were served and “B” the intervention period when the largest serving size for a glass of wine was removed from the existing range in each establishment: 250 ml (18 premises) or 175 ml (3 premises). The primary outcome was the daily volume of wine sold, extracted from sales data. Twenty-one premises completed the study, 20 of which did so per protocol and were included in the primary analysis. After adjusting for prespecified covariates, the intervention resulted in −420·8 millilitres (ml) (95% confidence intervals (CIs) −681·4 to −160·2 p = 0·002) or −7·6% (95% CI −12·3%, −2·9%) less wine being sold per day. There was no evidence that sales of beer and cider or total daily revenues changed but the study was not powered to detect differences in these outcomes. The main study limitation is that we were unable to assess the sales of other alcoholic drinks apart from wine, beer, and cider, estimated to comprise approximately 30% of alcoholic drinks sold in participating premises.
Conclusions
Removing the largest serving size of wine by the glass from those available reduced the volume of wine sold. This promising intervention for decreasing alcohol consumption across populations merits consideration as part of alcohol licensing regulations.
Trial registration
ISRCTN https://doi.org/10.1186/ISRCTN33169631; OSF https://osf.io/xkgdb.

研究背景：对引发不健康行为的物理环境进行干预，有望在群体层面实现行为改变，进而降低多种疾病的发病风险。其中一类干预措施为缩减饮品供应分量，该手段或可减少酒精摄入。然而，在真实场景中调整葡萄酒可提供的杯售分量范围所产生的效应尚不明确。本研究旨在评估：在持牌酒类经营场所中，移除最大杯售葡萄酒分量后，对葡萄酒销量的影响。

研究方法与结果：本研究于2021年9月至2022年5月间，在英格兰21家提供杯售葡萄酒（分量大于125ml，即175ml或250ml）且配备电子收银系统（electronic point of sale till system）的持牌酒类经营场所开展。研究采用A-B-A逆转设计（A-B-A reversal design），分为3个为期4周的阶段：阶段“A”为非干预期，各场所维持标准分量供应；阶段“B”为干预期，移除每家场所原有最大杯售葡萄酒分量——18家场所移除250ml分量，3家场所移除175ml分量。本研究的主要结局为从销售数据中提取的每日葡萄酒销量。共有21家场所完成研究，其中20家符合方案要求，被纳入主要分析。在校正预先设定的协变量后，干预使每日葡萄酒销量减少420.8毫升（ml）（95%置信区间(CI)：-681.4至-160.2，p=0.002），相对降幅达7.6%（95%CI：-12.3%至-2.9%）。未观察到啤酒与苹果酒销量或每日总营收出现显著变化，但本研究统计效能不足，无法检测此类结局的差异。本研究主要局限性在于：仅能评估葡萄酒、啤酒与苹果酒的销量，而据估算，参与场所销售的酒精饮品中，除上述三类外的其他品类占比约为30%。

研究结论：移除持牌场所可提供的最大杯售葡萄酒分量，可降低葡萄酒销量。这一有望在群体层面减少酒精摄入的干预措施，值得作为酒类经营许可监管条例的组成部分加以考量。

试验注册：ISRCTN：https://doi.org/10.1186/ISRCTN33169631；OSF：https://osf.io/xkgdb。