Behaviour and Health Research Unit, Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK.
School of Sciences, Bath Spa University, Bath, UK.
Addiction. 2023 Dec;118(12):2327-2341. doi: 10.1111/add.16288. Epub 2023 Aug 1.
To estimate the impact on selection and actual purchasing of (a) health warning labels (text-only and image-and-text) on alcoholic drinks and (b) calorie labels on alcoholic and non-alcoholic drinks.
Parallel-groups randomised controlled trial.
Drinks were selected in a simulated online supermarket, before being purchased in an actual online supermarket.
Adults in England and Wales who regularly consumed and purchased beer or wine online (n = 651). Six hundred and eight participants completed the study and were included in the primary analysis.
Participants were randomized to one of six groups in a between-subjects three [health warning labels (HWLs) (i): image-and-text HWL; (ii) text-only HWL; (iii) no HWL] × 2 (calorie labels: present versus absent) factorial design (n per group 103-113).
The primary outcome measure was the number of alcohol units selected (with intention to purchase); secondary outcomes included alcohol units purchased and calories selected and purchased. There was no time limit for selection. For purchasing, participants were directed to purchase their drinks immediately (although they were allowed up to 2 weeks to do so).
There was no evidence of main effects for either (a) HWLs or (b) calorie labels on the number of alcohol units selected (HWLs: F = 0.406, P = 0.666; calorie labels: F = 0.002, P = 0.961). There was also no evidence of an interaction between HWLs and calorie labels, and no evidence of an overall difference on any secondary outcomes. In pre-specified subgroup analyses comparing the 'calorie label only' group (n = 101) with the 'no label' group (n = 104) there was no evidence that calorie labels reduced the number of calories selected (unadjusted means: 1913 calories versus 2203, P = 0.643). Among the 75% of participants who went on to purchase drinks, those in the 'calorie label only' group (n = 74) purchased fewer calories than those in the 'no label' group (n = 79) (unadjusted means: 1532 versus 2090, P = 0.028).
There was no evidence that health warning labels reduced the number of alcohol units selected or purchased in an online retail context. There was some evidence suggesting that calorie labels on alcoholic and non-alcoholic drinks may reduce calories purchased from both types of drinks.
评估（a）酒精饮料的健康警示标签（纯文本和图文）和（b）酒精和非酒精饮料的卡路里标签对选择和实际购买的影响。
平行组随机对照试验。
在模拟的在线超市中选择饮料，然后在实际的在线超市中购买。
经常在网上购买和饮用啤酒或葡萄酒的英格兰和威尔士成年人（n=651）。608 名参与者完成了研究并被纳入主要分析。
参与者被随机分配到一个 6 组的组间三因素[健康警示标签（HWL）（i）：图文 HWL；（ii）纯文本 HWL；（iii）无 HWL]×2（卡路里标签：有或无）因子设计（每组 103-113 名参与者）。
主要结局指标为所选酒精单位数量（意向购买）；次要结局指标包括购买的酒精单位和选择和购买的卡路里。选择没有时间限制。对于购买，参与者被要求立即购买他们的饮料（尽管他们被允许最多 2 周的时间这样做）。
在选择的酒精单位数量上，HWL 或卡路里标签均没有证据表明存在主要影响（HWL：F=0.406，P=0.666；卡路里标签：F=0.002，P=0.961）。HWL 之间也没有证据表明存在交互作用，并且在任何次要结局上也没有证据表明存在总体差异。在比较“仅卡路里标签”组（n=101）和“无标签”组（n=104）的预先指定的亚组分析中，没有证据表明卡路里标签减少了所选卡路里的数量（未调整的平均值：1913 卡路里与 2203 卡路里，P=0.643）。在继续购买饮料的 75%的参与者中，“仅卡路里标签”组（n=74）购买的卡路里少于“无标签”组（n=79）（未调整的平均值：1532 卡路里与 2090 卡路里，P=0.028）。
在在线零售环境中，没有证据表明健康警示标签减少了选择或购买的酒精单位数量。有一些证据表明，酒精和非酒精饮料上的卡路里标签可能会减少两种饮料的购买量。
