Repository logo

Impact of bottle size on in-home consumption of wine: feasibility and acceptability randomised cross-over study.

Published version



Change log


Mantzari, Eleni 
Galloway, Catherine 
Zupan, Zorana 


Background: Reducing alcohol consumption across populations would prevent many non-communicable diseases. Large packages increase food and non-alcoholic drink consumption and large glasses increase wine consumption. Smaller bottles may reduce alcohol consumption but their impact is uncertain. This study aims to (i) explore the feasibility and acceptability of conducting a large-scale randomised study to assess the impact of bottle size on in-home wine consumption and (ii) estimate the effect size and variance of the intervention on consumption to inform the design of future studies. Methods: Cross-over randomised study in which 16 households in Cambridge, England, consuming at least two 750-ml bottles of wine each week, received a pre-set volume of wine biweekly for 4 weeks, in 750-ml and 375-ml bottles, in random order. Consumption was assessed by recording the number of empty and partially full bottles at the end of each biweekly period. At the end of the study, household representatives were interviewed about their experiences of participating in the study. Results: The study procedures proved feasible. Comparable to similar trials, 14% of identified eligible households (18/125) consented to participate in the study. Attrition between consent and study completion was 11% (2/18) and 0% between study periods and 13% of households (2/16) correctly identified the study aim. The study procedures were considered acceptable. After adjusting for guest and out-of-home consumption, the difference in consumption between the 750-ml (3385.2 ml; SD = 1698.5) and 375-ml bottles (3376.7 ml; SD = 1719.0) was 8.4 ml (SD = 1235.4; 95%CI - 596.9, 613.8). Results suggest a possible order effect, with households receiving the 375-ml bottles first consuming more wine out of the 750-ml bottles and vice versa. This might also reflect an increase in consumption with study duration. Households receiving the 375-ml bottles first (6315.9 ml; SD = 3293.5) also drank less wine overall than those receiving the 750-ml bottles first (7335.4 ml; SD = 3735.4). Discussion: The findings support the feasibility and acceptability of running a large-scale randomised study to assess the impact of bottle size on in-home wine consumption. Due to the heterogeneous patterning of results, a future study will be powered using the variance observed in the current study to detect a meaningful reduction of 250 ml of wine when consumed from smaller compared with larger bottles. Trial registration: Open Science Framework (OSF): rmk43; May 23, 2017.



Alcohol, Bottle size, Consumption, Portion size, Wine

Journal Title

Pilot and Feasibility Studies

Conference Name

Journal ISSN


Volume Title



BioMed Central
Department of Health (PRP number 107001)
Wellcome Trust (106679/Z/14/Z)
This study was supported by a grant from the National Institute for Health Research, Policy Research Programme (Policy Research Unit in Behaviour and Health [PR-UN-0409-10109] PI: Theresa Marteau). The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research, the Department of Health and Social Care or its arm’s length bodies, and other Government Departments. Rachel Pechey is supported by a Wellcome Trust Research Fellowship in Society and Ethics (106679/Z/14/Z)