Empirical evaluation of the presence of a label containing standard drinks on pour accuracy among US college students.

PURPOSE: Alcohol concentration has traditionally been labeled in the form of alcohol by volume (ABV). This format can cause difficulty in evaluating accuracy of a pour because it doesn't directly connect with recommendations related to "standard drinks," the approach used by the US CDC and others organizations which intend to facilitate responsible drinking behaviors. Strategies which more directly connect guidelines related to healthy drinking behaviors to alcohol labeling are needed.
OBJECTIVE: Assess how a label identifying the number of standard drinks per container impacts the ability of undergraduate students to accurately pour a standard drink.
DESIGN: This study employed a 3 x 2 x 2 experimental design. Undergraduates were asked to pour a standard drink from mock products from three alcohol categories (beer, wine and liquor); products were presented in two types of label (traditional ABV vs. standard drinks/container) at two concentrations of alcohol content (high and low).
RESULTS: We calculated standardized pour errors (pour errors in standard drink units). Analysis of these standardized pour errors suggested that 1) people tended to underpour beverages of low concentration across product categories and overpour those high in concentration. 2) When the standard drink label was present, pour accuracy was improved, when compared with pours from containers affixed with ABV labels in low alcohol concentrations across all product categories (beer, wine and liquor). 3) For treatments that comprised high concentrations of alcohol, the standard drink label significantly increased accuracy only for beer. However, it is worth noting that beer with an ABV label was the condition with the most dramatic overpours, and these problematic overpours were dramatically reduced by the addition of a standard drink label.
CONCLUSIONS: Our work empirically supports the notion that Undergraduate students are better able to accurately assess and pour a standard drink of alcohol from bottles incorporating a label which includes standard drinks/container vs. those with traditional ABV labeling. That said, the effect is quite different for each alcohol category: beer, wine, and liquor and depends on whether the product is high or low in concentration of alcohol for its category; as such, policy makers should consider alcohol categories and concentrations from a public health perspective when recommending changes to labeling.

Introduction and background

The US government mandates specific information content and formatting for the labels of many consumer products, including: foods, over-the-counter medications (OTC), and alcohol. Regulations related to OTC and food labels are promulgated by the US Food and Drug Administration (FDA), an Agency within the Department of Health and Human Services (DHHS). By contrast, the labeling of alcohol is regulated by the Alcohol and Tobacco Tax and Trade Bureau (TTB), an Agency within the US Department of Treasury. Differing missions within the Departments, and the Agencies that they oversee, result in unique motivations which express themselves in the form of the labels that they require across product categories.

While food and drug labeling is intended to motivate informed, healthful choices by consumers, alcohol labeling, with roots in the abrogation of prohibition, has tended to focus on ensuring that consumers receive what they believe that they purchased (type, origin and content) and that producers pay appropriate taxes on the products that they sell. As such, the labeling of alcohol content has been traditionally expressed as alcohol by volume (ABV) or proof (2x ABV), measures which don’t easily lend themselves to informed decision-making for consumption. This is despite the fact that most guidance related to (un)/healthy drinking behaviors, in the US and throughout the world, are not defined on the basis of ABV, but in “standard drinks” (SD) (see Table 1) [1]. A standard drink is defined as a drink containing 14 g (0.6 fl oz) of pure alcohol. The standard drink concept has been indicated as “foundational knowledge” for understanding health policy aimed at reducing harm from alcohol [2]. It is an attractive approach to those lobbying more comprehensive labels for consumer products (food and drug) and also appeals to health policy makers because it accounts for both the strength and the volume of a beverage, allowing for comparisons within and across beverage categories [2].

Table 1

Standard drinking behaviors as recommended by the National Institute on Alcohol Abuse and Alcoholism (NIAAA) [1].

Drinking Behavior	Definition	For Men	For Women
Moderate Drinking	Standard drinking behavior	2 standard drinks per day	1 standard drink per day
Binge Drinking	Pattern of drinking that brings blood alcohol concentration (BAC) levels to 0.08 g/dL	5 standard drinks in 2 hours	4 standard drinks in 2 hours
Heavy/Excessive Alcohol Use	Pattern of excessive binge drinking	4+ standard drinks on any given day	3+ standard drinks on any given day

The disconnect between the labeling standards and consumption recommendations becomes painfully apparent in the following example. Consider what is required to determine the number of standard drinks in a 10 oz glass of wine which is labeled as having 14% ABV. The formula for conversion requires one to convert oz into grams (10 oz X 29.574 g/oz = 295.74g), then multiply that result by the ABV (295.74 g*(14/100) = 41.4), subsequently one must multiply by the specific gravity of alcohol (~0.7936 x 41.4 g = 32.858 g) to determine that there are 32.858 grams of pure alcohol per 10 oz glass. Finally, to convert this to standard drinks one would have to know that a standard drink is equivalent to consuming 14 grams of pure alcohol (see Table 1); as such, you divide the number (32.858 g) by 14 g/standard drink, to discover that the 10 oz glass of wine is the equivalent of 2.3 standard drinks.

Clearly, the conversion of ABV to standard drinks is not intuitive. As a result, the National Institute on Alcohol Abuse and Alcoholism (NIAAA) has promoted more intuitive estimates to guide consumers. NIAAA suggests estimating a standard drink as equivalent to approximately 12 oz of beer, 8–9 oz of malt liquor, 5 oz of wine, and 1.5 oz of liquor [3]. These estimates are based on the assumption of 5%, 7%, 12%, and 40% ABV for each drink type, respectively. Although this simplifies things considerably, it disregards the wide variation that can be present in ABV for a single category of alcohol, particularly beer and liquor. Consider 12 oz cans of Budweiser (5% ABV), and Sam Adams’s Triple Bock (17.5% ABV). While the NIAA heuristic (a 12 oz can of beer is a standard drink) is true for the Budweiser, the Triple Bock is equivalent to 3.5 standard drinks, yielding the potential for significant over-consumption; the consumer may be getting 2.5 more drinks than they interpreted from a single pour.

College students and young adults are frequently the focus of consumption studies because they represent a specific, at-risk group. Over-consumption can be particularly problematic for this cohort, which is susceptible to binge drinking [4] and resultant tribulations. In the US, binge drinking has become a serious problem, especially among those aged 18 to 25 [5]; one study found that a large portion of college students drink at peak levels well beyond the binge threshold (see Table 1) [6].

Although we recognize that decision making is multifactorial in nature and no single strategy can eliminate overconsumption behaviors, we were interested in whether or not a change in the US approach to labelling could better equip people motivated to consume appropriately to do so than the current labelling standard (ABV).

Research supports the idea that drinkers have difficulty identifying a serving size as it relates to the amount of alcohol they choose to consume. One study of U.S college students asked participants to complete an alcohol survey and pour a variety of alcoholic beverages to a volume they deemed to be a standard drink. The study found that the students did not know how to define standard drinks accurately and overestimated volume when pouring [7]. Other studies of young-adult drinkers and college students have also surmised that they are generally unaware of the alcohol content in beverages, as well as national recommendations regarding responsible drinking behaviors [8-10]. A study focusing on US college students and bartenders, suggests that both tend to over-pour liquor when utilizing shot glasses [11].

Revising the labelling of alcohol content to be more aligned with recommendations for consumption (i.e. standard drinks) (see Table 1) represents one change with the potential to enable informed decisions related to drinking behaviors. Even so, recent US regulatory action in the form of a proposed rule focused on alcohol labeling entitled, “Modernization of the Labeling and Advertising Regulations for Wine, Distilled Spirits, and Malt Beverages” [12], allows for, but does not mandate, such a change. The proposed rule (§ 7.65(b)(1)) states, “Other truthful, accurate, and specific, factual representations of alcohol content, such as alcohol by weight, may be made, as long as they appear together with, and as part of, the statement of alcohol content as a percentage of alcohol by volume [ABV].” Although other (truthful) information is allowed related to alcohol content, only the ABV remains as a requirement. As a result, the rule, as proposed, has been criticized as falling “dramatically short” of what is needed to “modernize” alcohol labeling [13]. The criticism, levied by three national consumer advocacy groups, indicates that the rule fails to require uniform disclosure of key information required for consumers to make informed choices for purchase and consumption of these products (e.g. alcohol content, serving size, calories, ingredients and allergen information).

Here, we focus specifically on the presentation of alcohol content, with the goal of investigating whether presenting information about the number of standard drinks per container enables young adults to make more accurate assessments of the serving size which constitutes one standard drink. A systematic review of the peer-reviewed and grey literature published between 1990 and 2016 concludes that overall knowledge related to standard drinks is low, but that standard drink label use can assist consumers to more accurately identify and pour a serving [14]; however, none of the published studies were conducted with US consumers. That said, studies conducted in Canada [15, 16], Australia [17, 18] and Europe were consistent with regard to these conclusions, and Australia has since adopted a standard drink labeling practice for alcoholic beverages.

We tested a standard drink label positioned on the front of the package using US college students, to see if this labeling strategy enabled young adults to better assess the quantity of alcohol comprising a standard drink as compared with the current US labelling standard (ABV). Additionally, we postulated that product category (beer, wine and liquor) and alcohol concentration (high vs low) had the potential to impact pour volumes assessed by young adults.

Materials and methods

Recruitment and consent

We based our power calculation on the most conservative comparison performed; a single sample t-test that compared a particular pour condition’s mean pour error to zero. Results suggested that to achieve the power to detect a moderate effect size of with power of 0.95 required 84 participants [19, 20]. Ultimately, we recruited and tested 84 undergraduate participants in accordance with methods approved by the Institutional Review Board (IRB) at Michigan State University as STUDY0000083.

Recruitment was conducted using the student subject pool administered by the College of Communication Arts and Sciences at Michigan State University. To participate subjects had to be: 18 or older, enrolled as a college student, have transportation to the study site (a campus laboratory) and be willing to share contact information for the purpose of scheduling. All subjects provided informed, written consent, and were paid $20 in exchange for their time.

Study stimulus

This study utilized a 3 x 2 x 2 experimental design, with 3 alcohol categories (beer, wine and liquor), 2 label types (ABV and standard drinks/container), and 2 alcohol concentrations (high and low in each of the alcohol categories- See Table 2). This resulted in a total of 12 different labels (3 categories x 2 label types x 2 concentrations of alcohol). For each alcohol category (beer, wine, and liquor), one original brand was created and used for each of the label types and alcohol concentrations (See Fig 1 for an example); information related to the alcohol concentration was positioned in the lower right corner of the label and the bottle capacity in the lower left and presented in accordance with current US regulation. The standard drinks per container were calculated using the standard drink calculator available at “Rethink Drinking” published by the NIAA [21]. Table 2 presents information related to stimulus design, pouring container and receiving container, which varied by alcohol category.

Table 2

Summary of stimulus design.

Category	Container	Container Size*	Receptacle	Receptacle Size*	Brand	Alcohol concentration	Standard Label (Alc. By Vol.)	Standard Drinks per container	Standard Drink Size*
Beer	Bottle	12 fl oz	Solo cup	16 oz	“High Seas Brewing Co.”	Low	5%	1	12 fl oz
							(A)	(B)	355 mL
		355 mL		473 mL		High	10%	2	6 fl oz
							(C)	(D)	177 mL
Wine	Bottle	25.4 fl oz	Stemless wine glass	14 oz	“Simply Divine Fine Wines”	Low	10%	4.2	6 fl oz
							(E)	(F)	177 mL
		750 mL		414 mL		High	15%	6.3	4 fl oz
							(G)	(H)	118 mL
Liquor	Bottle	25.4 fl oz	Bar shot glass	9 oz	“Caribbean Dreams”	Low	20%	8.4	3 fl oz
							(I)	(J)	89 mL
		750 mL		266 mL		High	40%	16.9	1.5 fl oz
							(K)	(L)	44 mL

Bold font indicates the 12 different label variations used in the study. The ABV and standard drink labels indicate the same amount of alcohol per container in low and high formats. Each letter corresponds to the stimulus labels located in Table 2.

*Volume provided in fl oz and mL due to differences in labeling requirements among alcoholic beverage categories (beer, wine, liquor).

Fig 1

Example stimulus design.

Data collection

Participants began with a brief, computer-based questionnaire related to their knowledge of consumption recommendations, as well as their drinking behaviors. Once complete, each participant engaged in a series of 24 pouring tasks, in two blocks of 12 unique treatments. They were instructed, “Pour a standard drink into the glass in front of you. Feel free to use any information present on the bottle to help inform your pour” (see Table 2 for bottles and glasses by treatment type). Order of presentation was randomized for each participant within the first block of 12. This order was repeated for the second block of twelve, comprised of a single replicate of each treatment. The volume of liquid each subject poured was measured (post-hoc) for every trial.

Data preparation & analyses

Each participant produced two pours within each condition of the 3x2x2 design. We began by averaging the volume (ml) of the two pours within a condition for each subject, producing a “mean pour volume” dependent variable (see Table 3). Six participants produced mean pour volumes for at least one condition that were more than three standard deviations from the overall mean for that condition. Data from those participants were eliminated from further analysis, resulting in analysis that included data collected from 78 participants.

Table 3

Summary of variables.

Variable	Explanation
Observed Pour	Participant pour recorded (in mL) during experiment as measured using a graduated cylinder with a capacity of 250 mL or 500 mL, and a sensitivity of 2 mL and 5 mL, respectively
Mean Pour Volume (MPV)	Average of the volume (in mL) of the two pours that a participant made within a condition.
Accurate Volume (AV)	The volume (in mL) that would accurately represent a standard drink for a given condition, as calculated using NIAA’s drink calculator [21].
Mean Standardized Pour Error (SPE)	Each subject’s pour error for a condition expressed in standard drink units. The formula to calculate it is SPE = (MPV-AV)/AV, where the MPV and AV values are for a specific condition.

We also converted the mean pour volume measure for each subject into a “standardized pour error” (SPE) variable (see Table 3). To do this we subtracted the “accurate volume” representing a standard drink (AV) for a given condition from each subject’s mean pour volume (MPV) for that condition, and divided the result by the accurate volume representing a standard drink (AV). Mathematically, this conversion is SPE = (MPV-AV)/AV, converting the pour errors into standard drink units. The rationale for transforming data into standard drink units is that it is the most relevant metric from the standpoint of public policy, is how much alcohol a person is pouring as opposed to the volume of the pour. That is, from a health perspective overpouring 30 milliliters of liquor is more concerning that overpouring 30 milliliters of beer. Once converted to standardize pour errors (SPEs), an overpour of 0.5 SPEs for beer and an overpour of 0.5 SPEs for liquor are equivalent in terms of the amount of alcohol that was overpoured; in both cases the pour was equivalent to 1.5 standard drinks or 21 grams of pure alcohol for each pour. Finally, we note that given the method of subtraction, positive standardize pour errors represent overpours, while negative values represent underpours.

An initial review of the data revealed them to be non-normally distributed; attempts to transform the data via standard transformations (e.g., log, double log) did not produce normal distribution. That said, we note that the ANOVA is fairly robust to violations of non-normality [22]. Given this robustness, we performed ANOVA omnibus tests, but in follow-up analyses, performed both paired t-tests and Wilcoxon Signed Rank Tests, a non-parametric analysis that does not have an assumption of normality. All statistical analyses were performed using SPSS [23].

Results

Data collected from our survey support previous findings utilizing college-aged participants which conclude them to be poorly acquainted with the idea of standard drinks [7-10]. This was true of even questions which utilized simple heuristics published by the NIAA. Specifically, only 60% (n = 50) of our respondents were able to correctly identify 12 oz of beer at 5% ABV as a standard serving. This dropped to 27% (n = 23) for wine (5 oz of wine at 12% ABV) and 17% (n = 14) for liquor (1.5 oz of liquor at 40% ABV).

The average mean pour volumes (MPV) by treatment are presented in Fig 2 juxtaposed against the volume representing the volume of an accurate standard drink (AV) for each treatment (presented in black). While we present these mean pour volumes for completeness, our statistical analyses were performed on the standardize pour errors (SPE).

Fig 2

Mean pour volume by category, label type, and alcohol concentration level.

The mean standardized pour errors across participants are presented in Fig 3. An omnibus 3 (alcohol category: Beer, Wine, Liquor) x 2 (concentration: high and low) x 2 (label type: ABV and standard drink label) within subjects ANOVA on the standardize pour errors revealed significant main effects of alcohol category F(2, 154) = 13.35, p<0.001, ηp2 = 0 .148, and concentration, F(1,77) = 194.67, p<0.001, ηp2 = 0 .717. While the main effect of label type did not approach significance, F(1, 77) = 0.681, ηp2 = 0.002, it was qualified by significant two way interactions of alcohol category x alcohol concentration, F(2, 154) = 13.97, p<0.001, ηp2 = 0.16; alcohol category x label type, F(2, 77) = 7.93, p<0.001, ηp2 = 0.093; and alcohol concentration by label type, F(1, 77) = 51.91, p<0.001, ηp2 = 0.40. The three-way interaction was also significant, F(2, 154) = 9.87, p<0.001, ηp2 = 0.114. One clear source of these interactions was that the direction of mean standardize pour errors differed as a function of alcohol concentration, with consistent overpours for high concentration conditions, and consistent underpours for low concentration conditions (see Fig 3). Thus, to unpack the three-way interaction we ran two 3 (alcohol category) x 2 (label type) ANOVAs–one for high concentrations and one for low concentration conditions. We believe that analyzing the high and low conditions separately is ideal since the types of errors for the two conditions are qualitatively distinct with different health consequences (over pouring has the potential to lead to adverse effects where underpouring carries less risk).

Fig 3

Mean standardized pour error by condition.

The pattern of data with for the high content conditions revealed significant main effects of alcohol category, F(2, 154) = 6.693, ηp2 = 0.080, p = .002, label type, F(1, 77) = 11.915, ηp2 = 0.134, p = 0.001, with a significant interaction related to alcohol category x label type, F(1, 77) = 12.128, ηp2 = 0.136, p<0.001. The cause for all three significant results appears to be that there were very high error rates for beer labeled with the ABV label, and these errors were mitigated by the use of a standard drink label. To examine this issue further, for each alcohol category, planned within subjects tests comparing SPEs between the standard drink labels and ABV labels were performed. These tests reveal that pours were significantly more accurate with the standard drink label for beer, with no apparent significant difference in accuracy when two label types were compared for wine or liquor (see Table 4 for statistics). The standard drink labels significantly reduce overpours for high content beer as compared to the traditional labeling approach (ABV). Importantly, high content beer labeled with an ABV label was the condition with the largest overpours, representing about an additional third of a standard drink (see Fig 3). By contrast, even with the ABV label, there were relatively small overpours for high content wine and liquor, with overpours of only about a tenth of a standard drink; as such, including a standard drink label yielded no evidence of a difference for these small overpours.

Table 4

Paired sample t-tests and Wilcoxon signed rank tests comparing ABV to standard drink errors for each type of drink.

	High Content	Low Content
Beer	t = 5.07, p < .001	t = -5.33, p < .001
	Wilcoxon p < .001	Wilcoxon p < .001
Wine	t = 1.10, p = .28	t = -4.56, p < .001
	Wilcoxon p = .183	Wilcoxon p < .001
Liquor	t = .26, p = .80	t = -2.82, p < .006
	Wilcoxon p = .90	Wilcoxon p = .004

All t-tests df. = 77. Bold values represent significant errors. Negative t-values represent underpours; positive t-values represent overpours.

Turning to the low content conditions, the ANOVA suggests main effects of alcohol category, F(2, 154) = 27.615, ηp2 = 0.264, p <0.001, and label type, F(1, 77) = 42.515, ηp2 = 0 .356, p<0.001, with no evidence of a significant interaction, F(2, 154) = 1.218, p = 0.299, ηp2 = 0.016. This pattern suggests that the inclusion of the standard drink label significantly reduced errors and did so to a similar extent for each alcohol type. To confirm, we ran a set of paired t-tests and Wilcoxon Signed Rank Tests comparing ABV label pour performance to that when standard drink labels were present for each alcohol category. The standard drink label significantly increased pour accuracy for all alcohol categories (see Table 4 for statistics). Even so, it is important to keep in mind that the errors in these low content conditions represent underpours, so the improved accuracy related to the standard drink label represents pouring additional alcohol.

Finally, to determine whether these observed errors were meaningful (i.e. were significantly different from zero), we performed a series of single sample t-tests and Wilcoxon Tests comparing each condition’s mean standardized pour error to zero (See Table 5). These analyses show that almost all conditions result in significant errors. Specifically, the only ABV condition that did not produce a significant error was the high content liquor condition, suggesting that participants were somewhat conservative in their pours of liquor. In addition, although the standard drink label reduced the errors in most conditions, only in the low content wine did this reduction result in a complete elimination of the error. That is, for most conditions, people continued to significantly deviate from the AV, having residual errors, even with the standard drink labels.

Table 5

Single sample t-tests and Wilcoxon signed rank tests.

	High Content		Low Content
	ABV	Standard Drink	ABV	Standard Drink
Beer	t = 7.44, p<0.001	t = 3.66, p<0.001	t = -7.10, p < .001	t = -3.89, p < .001
	Wilcoxon p<0.001	Wilcoxon p = 0.005	Wilcoxon p < .001	Wilcoxon p < .001
Wine	t = 3.49, p = 0.001	t = 2.98, p = 0.004	t = -4.883, p < .001	t = -.212, p = .833
	Wilcoxon p = 0.002	Wilcoxon p = 0.009	Wilcoxon p < .001	Wilcoxon p = .986
Liquor	t = 1.47, p = 0.15	t = 1.73, p = 0.09	t = -8.91, p<0.001	t = -6.95, p<0.001
	Wilcoxon p = 0.45	Wilcoxon p = 0.45	Wilcoxon p<0.001	Wilcoxon p<0.001

Each condition’s error was compared to zero. All t-tests df. = 77. Bold values represent significant errors. Negative t-values represent underpours; positive t-values represent overpours.

Discussion

The results of this experiment suggest that providing a standard drinks label, in place of the traditional ABV, enables participants to more accurately assess a serving size that is equivalent to recommended standards based on the amount of alcohol. This conclusion is generally in line with prior reports on the efficacy of standard drink labels [11-16]; however, our results incorporate more nuances. Specifically, not only did we investigate how this labeling strategy is influenced by both the concentration of alcohol in the product and its product category, we analyzed pour errors in units that considered the concentration of the alcohol, standard drink units. As such, results are more directly useful for those interested in health policy. Further, while the use of a standard drinks labels significantly reduced errors, there tended to be small residual errors (in the original direction of error), even in the presence of standard drink labels. That is, even with the improvement in pour accuracy acquired with the standard drink label, people continued to slightly overpour the drinks that were at the higher concentration and underpour those at the low.

One potential explanation for this pattern of over and under pouring (as a function of alcohol concentration) is that our participants have a preconceived notion of how large of a pour represents a standard drink for each alcohol category, and they adjust from that preconceived notion based on the information on the label. However, with the ABV label their adjustment is insufficient. That is, while they reduce the size of the of the pour for high concentration drinks, they under-adjust, resulting in an overpour. Similarly, when they are confronted with a low concentration, they increase the pour but fail to fully adjust, resulting in an underpour. The failure to adequately adjust from their preconceived notion of the size of a standard pour is reminiscent of the classic “Anchoring and Adjustment” heuristic [24, 25]. This heuristic suggests that people have a starting point, or anchor, that represents an initial evaluation of the value or quantity of an item, and then they adjust from that anchor based on additional information. However, a systematic bias has been noted whereby people will not adjust far enough, resulting in a final outcome that is too close to the anchor, not the end goal. While that decision-making heuristic may be at play in these judgements, our results suggest that providing a standard drink labels provides more (or better) information so that it leads to a larger adjustment from the anchor than the traditional labeling illicits.

It is also worth noting that, from a public health perspective, the benefit of the standard drink label were differential for high and low concentration examples of a drink type. While the standard drink label led to more accurate pours for all three low concentration drinks, the improvement in accuracy moved the person pouring to pour more alcohol. For the high concentration drinks, the standard drink label only produced more accurate pours than the ABV label for the beer category. Even so, it is worth noting that the high concentration beer with ABV labelling was the condition that had the largest overpours—the condition that is most problematic from a health and safety perspective. The fact that the standard drinks label greatly reduced overpours in this most troubling condition suggests that the label may be most effective in the area of the greatest public good. In light of these findings, policy makers should consider alcohol categories and concentrations from a public health perspective when recommending changes to labeling.

Limitations

One limitation of the current study is that it remains unclear why the efficacy of the standard label differed as a function of alcohol concentration within a class. One possible explanation is that our young adult participants may have different levels of familiarity and experience with different types of alcoholic drinks and that familiarity may influence the effectiveness of the standard drink label. For instance, it appears that our subjects were somewhat cautious with liquor, only slightly overpouring in the ABV high content condition and dramatically underpouring in the ABV low content condition. This initial wariness may have limited the ability of the standard drinks to significantly moderate behavior. However, this reasoning is purely speculative and more research that directly assessed the relationship between drink familiarity and the effectiveness of a standard drink label would be required before drawing conclusions about the potential explanation.

A second limitation is that the current study was a pour study rather than a consumption study. Although our pour results are consistent with the existing literature and suggest that labeling incorporating standard drink information improves consumers’ ability to identify the volume of liquid comprising a standard drink, it is unclear how this ability impacts actual consumption behavior. In fact, it has been suggested that those with the best recall of label information are, in fact, the biggest drinkers [26]; it is possible that providing accurate information has the potential to further enable high risk drinking behaviors. This notion has presented itself in historical policy. After prohibition was repealed, Congress passed a law which banned labeling beer with ABV [27], fearing its presence would result in “strength wars” between brewers attempting to increase demand for their product. However, in 1995 Coors successfully challenged the ban on ABV labeling for beer in the US Supreme Court [28] and it is now allowed. Even so that fear still exists, and there is recent evidence which seems supports the assumption. Following the implementation of an standard drink label in Australia [29], research found that college students were aware of these labels and reported using them to help “choose the strongest drinks for the lowest cost.” Thus, it seems that, particularly for young people, providing easily accessible information about alcohol concentration has the potential to enable overindulgence more effectively.

A final limitation was that the volume of the standard drink for our low content beer condition was the full receptacle (12 ounces), limiting the ability of people to overpour in this condition. We chose this as an ecologically valid condition (most low ABV beers cannot be overpoured in realistic conditions based on bottle (and glass) sizes, and glass size has been indicated to be a contributing factor to pour volume [30-33], nonetheless, this present an experimental imperfection.

Conclusion

Our results show clear evidence that a standard drink label is more effective than a standard ABV label at communicating accurate information about the serving size that represents a standard drink. In a majority of our conditions the standard drink label led to more accurate pours of a standard drink serving size, and in no cases did it perform significantly worse than the current ABV label. Given that a vast majority of the public health messaging around responsible drinking has focused on the standard drink metric, rather than the ABV metric, labels that allow one to better estimate the serving size that represents a standard drink are important to the efficacy of those public health campaigns. As such, we believe strongly that adopting a standard drink label, as has been done in Australia, can be an important contributor to effective public health messaging about responsible drinking. Further, providing such a label is consistent with calls of consumer groups for labelling that contains key information required for consumers to make informed choices for purchase and consumption of these products [13].

That said, the label alone will not accomplish these public health goals. An effective label can communicate accurate information about the product, but the responsible use of that information requires education and public health campaigns that convince people to utilize that information to make responsible decisions. In short, for such campaigns to be successful requires two elements. First, one must effectively convey the information that can empower consumers to make more healthy decisions. Second, the campaign must convince people to utilize that information to do so. The standard drink labels we tested here can assist with providing accurate alcohol content information and doing so is a necessary precondition for consumer to be able to make more healthy decisions. However, accurate product information alone is not sufficient to necessarily impact behavioral change. Instead this improved information needs to be coupled with effective public health campaigns to convince people to make healthy decisions.

(XLSX)

Click here for additional data file.

(XLSX)

Click here for additional data file.

3 Aug 2020

PONE-D-20-14913

Effects of a novel labeling strategy: Does presentation of standard drinks per container impact pour accuracy of alcohol among US college students?

PLOS ONE

Dear Dr. Bix,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Thank you for your submission to PLOS one. I apologise for the length of time this paper was under-review and thank them for their patience, it took a long time to find suitable and available reviewers. The paper makes an important contribution to the literature in terms of enhancing our understanding of beverage labelling strategies that can help improve the accuracy of pouring a standard drink (which could potentially make it easier for people to moderate their intake in relation to recommended guidelines). I am very grateful for reviewers careful consideration of the paper, its strengths and limitations and hope that their suggestions are helpful in revising the paper. One omission I noted was a limitations section, which should be added to the discussion as this will likely help guide future research in this area.

Please submit your revised manuscript by Sep 17 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Victoria Manning

Academic Editor

PLOS ONE

Additional Editor Comments:

I apologise for the length of time this paper was under-review and thank the authors for their patience as it took a long time to find suitable reviewers. The paper makes an important contribution to the literature in terms of enhancing our understanding of beverage labelling strategies that can help improve the accuracy of pouring a standard drink (which could potentially make it easier for people to moderate their intake in relation to recommended guidelines). I am very grateful for reviewers careful consideration of the paper, its strengths and limitations and hope that their suggestions are helpful in any revisions. One omission I noted was a limitations section, which should be added to the discussion as this will likely help guide future research in this area.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please ensure that you refer to Figure 7 in your text as, if accepted, production will need this reference to link the reader to the figure.

3. Please upload a copy of Figure 9, to which you refer in your text by line 480. If the figure is no longer to be included as part of the submission please remove all reference to it within the text.

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is an interesting study that explores the impact of standard drink labels, compared to existing ABV labels, on accuracy of drink pouring across different types of alcoholic drink. However, the manuscript is very long and in places could be edited to reduce length and improve clarity. In addition, the stated objectives and discussion of results should reflect the outcomes examined in the study (i.e., pouring accuracy) rather than consumption, and the discussion of results should be placed within the context of previous literature. Outlined below are some specific comments that may be helpful for improving clarity. The manuscript should also be checked for grammatical and typographical errors.

Introduction

• The introduction is long and could be edited to make the points more succinct. For example, the first paragraphs provide quite a lot of detail of the current regulations around alcohol content labelling that could be summarised in a few sentences.

• The introduction could also be structured more clearly. For example, it would be useful to define ‘standard drinks’ (i.e., 14g pure alcohol in US – para. 7, line 147 and para. 8, line 163) and how these are used in drinking guidelines (Table 1) when they are first mentioned (para. 3, line 110). Followed by why standard drink labelling may be preferable to ABV information, given the wide variation both within (para. 6, line 136 and para. 9) and across drink categories and sizes (para. 7, line 150).

• Para. 8, lines 164-171: The explanation of the calculation is quite long and complex, and misses out some elements (e.g., dividing g of alcohol by 100, line 167). While this makes the point about the difficulty using ABV to determine number of standard drinks, it may be easier and more intuitive for readers to provide a few example of drink types, volumes and ABV alongside the number of standard drinks.

• The objective outlined at the end (lines 225-7) may be too broad – while the impact of standard drinks labelling has implications for understanding consumption relative to health guidelines, the study’s examination of pouring does not address this directly.

Methods

• Was a sample size calculation conducted to inform the recruitment of 84 participants?

• This section could be edited in places to remove unnecessary details (e.g., type of measuring cylinder used to fill bottles, exact task wording for participants).

• Data collection (lines 325-8): could the authors clarify whether all participants completed the drink pours in the same random order, or whether order was randomised for each participant.

• As the 5%ABV beer container equalled one standard drink it was not possible to overpour a standard drink in this condition; therefore, it is difficult to compare the effect of labels on overpouring with other drink types.

• It would be useful to provide a link to the study protocol if this has been, or could be, uploaded to an open access repository.

Results

• Check for grammatical and typographical errors

• I think it would be helpful for readers if descriptive results were presented (e.g., mean and standard deviation) for the different measures across conditions.

• Exact p values should be given, rather than using ‘n.s.’

• A number of similar analyses and figures are presented – was there a prespecified analysis plan for the study? It may be clearer to present figure(s) for the primary analysis and move others to supplementary materials.

• It would be useful to make it clear in the data analyses section which is the primary outcome; explanations of the measures (e.g., reference to Weber’s Law) should be moved to the methods section too.

Discussion

• Details of the mean pour errors (lines 520-2) should be moved up to the results section.

• The assumption that drinkers would reduce their consumption in line with increasingly accurate standard drink pours (lines 522-7) may be too optimistic. While improved accuracy of pouring is valuable, and greater understanding of standard drinks may support people to monitor their alcohol consumption, it does not necessarily mean that they will monitor their drinking or change their overall consumption. There is evidence to suggest that, particularly young people, may use standard drinks to facilitate increased alcohol consumption and it is important to consider possible unintended consequences (e.g., JONES, S.C. and GREGORY, P. (2009), The impact of more visible standard drink labelling on youth alcohol consumption: Helping young people drink (ir)responsibly?. Drug and Alcohol Review, 28: 230-234. doi:10.1111/j.1465-3362.2008.00020.x)

• The results of the study should be put in the context of previous literature (which is absent in the discussion).

Data files: It would be helpful to have a data dictionary to help navigate the data

Reviewer #2: Clearly written and thoroughly analysed. The manuscript clearly illustrates the multiple steps currently required for individuals to estimate SDs. The public health benefit of SD labelling could also be enhanced if the authors new the % of all alcohol consumed from off-premise outlets, which relies of self-pouring compared to on-premise drinking which for liquors and wines are poured by bar staff. I would have liked to see a return to the discussion around legislation and policy in the Discussion and what policy implications the paper has, and whether any current move towards introducing SDs in the U.S.

Small errors:

- line 404 ‘There is an accuracy..’

-line 424 ‘categories of alcohol’

- line 446 ‘reasons’

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

21 Sep 2020

Author responses are presented with [AU] preceding them.

Reviewer comments are presented with reviewer number in front of them

[AU] We were very pleased to hear that editors felt that “The paper makes an important contribution to the literature in terms of enhancing our understanding of beverage labelling strategies that can help improve the accuracy of pouring a standard drink (which could potentially make it easier for people to moderate their intake in relation to recommended guidelines).” - -- Additionally, at the suggestion of the editor, we have incorporated a limitations section into the document which was absent in the previous version.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please ensure that you refer to Figure 7 in your text as, if accepted, production will need this reference to link the reader to the figure.

[AU] This paper has undergone major streamlining and revision, which included the figures. To address this issue, we have conducted a careful read through of the document to make sure that all figures are appropriately acknowledged within the document text. There is no longer a Figure 7

3. Please upload a copy of Figure 9, to which you refer in your text by line 480. If the figure is no longer to be included as part of the submission please remove all reference to it within the text.

[AU] Please refer to the comment immediately preceding this one. To address reviewer’s comments regarding the need for more streamlined prose and more concise manuscript, the entire document has gone through major revision and several figures have been removed. There is no longer a Figure 9.

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

5. Review Comments to the Author

Reviewer #1: This is an interesting study that explores the impact of standard drink labels, compared to existing ABV labels, on accuracy of drink pouring across different types of alcoholic drink. However, the manuscript is very long and in places could be edited to reduce length and improve clarity. In addition, the stated objectives and discussion of results should reflect the outcomes examined in the study (i.e., pouring accuracy) rather than consumption, and the discussion of results should be placed within the context of previous literature. Outlined below are some specific comments that may be helpful for improving clarity. The manuscript should also be checked for grammatical and typographical errors.

Introduction

• [REVIEWER 1] The introduction is long and could be edited to make the points more succinct. For example, the first paragraphs provide quite a lot of detail of the current regulations around alcohol content labelling that could be summarised in a few sentences.

[AU]- We have SIGNIFICANTLY streamlined the manuscript, particularly the Background, Methods and Results, in light of this comment. The background no longer incorporates the depth of explanation regarding the historical and legal frame which undergirds alcohol labelling. The Methods have been significantly truncated to the most relevant points and the results now include only the raw pour data and a single variable that we perform analyses on, the one that is most germane to public/health policy (pour error as measured in standard drinks). We believe that this results in a document that is not only easier to read, also more easily interpreted for meaningful change.

•[REVIEWER 1] The introduction could also be structured more clearly. For example, it would be useful to define ‘standard drinks’ (i.e., 14g pure alcohol in US – para. 7, line 147 and para. 8, line 163) and how these are used in drinking guidelines (Table 1) when they are first mentioned (para. 3, line 110). Followed by why standard drink labelling may be preferable to ABV information, given the wide variation both within (para. 6, line 136 and para. 9) and across drink categories and sizes (para. 7, line 150).

[AU] In the radically streamlined version of the introduction we have tried to do so, while emphasizing the disconnect between ABV labelling and the standard drink units used in health messaging.

•[REVIEWER 1] Para. 8, lines 164-171: The explanation of the calculation is quite long and complex, and misses out some elements (e.g., dividing g of alcohol by 100, line 167). While this makes the point about the difficulty using ABV to determine number of standard drinks, it may be easier and more intuitive for readers to provide a few example of drink types, volumes and ABV alongside the number of standard drinks.

[AU] We agree that it is long. We have tried to streamline it a little, but the point we are trying to make is converting from ABV (which is what is present on current labels) to standard drinks (which are the units used in health messaging) is extremely difficult. We think showing how difficult makes this point well.

• [REVIEWER 1] The objective outlined at the end (lines 225-7) may be too broad – while the impact of standard drinks labelling has implications for understanding consumption relative to health guidelines, the study’s examination of pouring does not address this directly.

[AU] Understood. We have worked throughout to reframe the language to indicate the ability to correctly identified a standard drink (which serves to inform consumption behavior). We have also revised the language such that it is more transparent in this regard. Further, the revision of the Results, which now only report a single dependent variable (pour error as measured by standard drinks) improves the document’s comprehensibility. Further, we now acknowledge in our limitations that consumption may actually be negatively impacted by imparting more accurate knowledge of alcohol content.

Methods

• [REVIEWER 1] Was a sample size calculation conducted to inform the recruitment of 84 participants?

[AU]Power calculations which support the sample size have been incorporated into the beginning of the Methods section of the document.

• This section could be edited in places to remove unnecessary details (e.g., type of measuring cylinder used to fill bottles, exact task wording for participants).

[AU] Understood. Much of the technical detail has been removed from this section.

• [REVIEWER 1] Data collection (lines 325-8): could the authors clarify whether all participants completed the drink pours in the same random order, or whether order was randomised for each participant.

[AU] The randomization was conducted for the first set of trials for each individual participant. Replicants were performed in the same order as the first randomized set (which was (theoretically) unique to ever participant. So if participant ones randomization first set indicated the order of presentation to be 12, 1, 4, 3…. There second set (the replicates) were also 12, 1, 4, 3…. In other words, randomization was done by participant. We have modified the text within the Methods to enhance the clarity around this important point of process.

• [REVIEWER 1] As the 5%ABV beer container equalled one standard drink it was not possible to overpour a standard drink in this condition; therefore, it is difficult to compare the effect of labels on overpouring with other drink types.

[AU]- This is correct. Although an ecologically valid condition (most low ABV beers cannot be overpoured in realistic conditions based on bottle (and glass) sizes), it does present an experimental imperfection. In an attempt to better address this, we have revised the language in the limitations to more directly address it (and also explain its presence).

• [REVIEWER 1] It would be useful to provide a link to the study protocol if this has been, or could be, uploaded to an open access repository.

[AU] We have taken advantage of the ability to upload Laboratory protocols in protocol.io as a supplemental material. This has not only allowed us to address the reviewer comments by streamlining the methods significantly, but also affords the opportunity to present the Methods with significant detail to interested parties.

Results

• [REVIEWER 1] Check for grammatical and typographical errors

[AU]- We have tried to give the document a thorough read through to be sure that it is not only more concise, but that grammatical and typographical errors have been corrected throughout.

•[REVIEWER 1] I think it would be helpful for readers if descriptive results were presented (e.g., mean and standard deviation) for the different measures across conditions.

[AU] Those data for the raw pour volumes are presented in Figure 2. For the standardized pour error variable they are presented in Figure 3.

• [REVIEWER 1] Exact p values should be given, rather than using ‘n.s.’

[AU] We have now done so, but note that the only time we used n.s., is when the calculated F was less than 1. The expected value for F under the null is 1, so statistical significance is impossible if F<1, thus there is a convention to report ns in those cases.

•[REVIEWER 1] A number of similar analyses and figures are presented – was there a prespecified analysis plan for the study? It may be clearer to present figure(s) for the primary analysis and move others to supplementary materials.

[AU] We revisited the Results upon receiving the reviewer’s comments and whole heartedly agree. We have removed redundant analyses and reworked the section significantly in an attempt to meaningfully synthesize the finding in a complete, yet more concise, form. We appreciate the feedback and believe that this change has significantly enhanced the readability of the document without altering the Conclusions that can be drawn from the data set.

• [REVIEWER 1] It would be useful to make it clear in the data analyses section which is the primary outcome; explanations of the measures (e.g., reference to Weber’s Law) should be moved to the methods section too.

[AU] In streamlining the paper we have presented the error in standard drink units and have motivated doing so because those units are most relevant from a health perspective. This framing made the discussion of Weber’s Law superfluous and it has been removed.

Discussion

•[REVIEWER 1] Details of the mean pour errors (lines 520-2) should be moved up to the results section.

[AU] With the new analyses this has been removed from the discussion.

• [REVIEWER 1] The assumption that drinkers would reduce their consumption in line with increasingly accurate standard drink pours (lines 522-7) may be too optimistic. While improved accuracy of pouring is valuable, and greater understanding of standard drinks may support people to monitor their alcohol consumption, it does not necessarily mean that they will monitor their drinking or change their overall consumption. There is evidence to suggest that, particularly young people, may use standard drinks to facilitate increased alcohol consumption and it is important to consider possible unintended consequences (e.g., JONES, S.C. and GREGORY, P. (2009), The impact of more visible standard drink labelling on youth alcohol consumption: Helping young people drink (ir)responsibly?. Drug and Alcohol Review, 28: 230-234. doi:10.1111/j.1465-3362.2008.00020.x)

[AU] Although making informed choices does depend on the ability to accurately assess the amount of alcohol in a standard drink, it is up to the drinker to use it. As the reviewer points out, this information could be used appropriately, ignored or used inappropriately. We have revised the language throughout the document to better reflect the complexities of human behavior and also worked to incorporate it into the state of knowledge specific to drinking behaviors. In addition, we explicitly acknowledge this point and reference the Jones & Gregory paper in the limitation section.

• [REVIEWER 1] The results of the study should be put in the context of previous literature (which is absent in the discussion).

[AU] Please view the aforementioned comment. We now state that our results are generally in line with others who have examined this issue and cite that work. We also point out ways in which our approach was more nuanced than some of the prior work.

[REVIEWER 1] Data files: It would be helpful to have a data dictionary to help navigate the data

[AU] With our simplified reporting this might now be overkill, but we included a data dictionary in table 3.

•

•

[Reviewer #2]: Clearly written and thoroughly analysed. The manuscript clearly illustrates the multiple steps currently required for individuals to estimate SDs. The public health benefit of SD labelling could also be enhanced if the authors new the % of all alcohol consumed from off-premise outlets, which relies of self-pouring compared to on-premise drinking which for liquors and wines are poured by bar staff. I would have liked to see a return to the discussion around legislation and policy in the Discussion and what policy implications the paper has, and whether any current move towards introducing SDs in the U.S.

Small errors:

- line 404 ‘There is an accuracy..’

-line 424 ‘categories of alcohol’

- line 446 ‘reasons’

[AU] We appreciate that reviewer two believed our work to be “clearly written and thoroughly analysed”. That said, we have provided major revisions based on weakeness pointed out by Reviewer one. As part of those, we have reanalyzed our data such that the dependent variable is “normalized” so that errors are no longer volume based, but based on the # of standard drinks that one over or under pours. We believe that this not only addresses some of the concerns of Reviewer #1, but is also far more relevant to the desire that this reviewer has with regard to policy implications. Additionally, we have worked to revise the conclusions to focus more solidly on the same.

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

In compliance with data protection regulations, you may request that we remove your personal registration details at any time. (Remove my information/details). Please contact the publication office if you have any questions.

Submitted filename: PONE Feedback for reviewers.docx

Click here for additional data file.

19 Oct 2020

Empirical evaluation of the presence of a label containing standard drinks on pour accuracy among US College students

PONE-D-20-14913R1

Dear Dr. Bix,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Victoria Manning

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Thank you for your careful consideration of the reviewers comments and for the thorough revisions made the manuscript which is substantially improved as a result.

Reviewers' comments:

26 Oct 2020

PONE-D-20-14913R1

Empirical evaluation of the presence of a label containing standard drinks on pour accuracy among US College students

Dear Dr. Bix:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Victoria Manning

Academic Editor

PLOS ONE