The authors have declared that no competing interests exist.
Identifying and tackling the factors that undermine regulation of unhealthy commodities is an essential component of effective noncommunicable disease (NCD) prevention. Unhealthy commodity producers may use rules in US and EU Free Trade Agreements (FTAs) to challenge policies targeting their products. We aimed to test whether there was a statistical relationship between US and EU FTA participation and reduced implementation of WHO-recommended policies.
We performed a statistical analysis assessing the probability of at least partially implementing 10 tobacco, alcohol, and unhealthy food and drink policies in 127 countries in 2014, 2016, and 2019. We assessed differences in implementation of these policies in countries with and without US/EU FTAs. We used matching to conduct 48 covariate-adjusted quasi-experimental comparisons across 27 matched US/EU FTA members (87 country-years) and performed additional analyses and robustness checks to assess alternative explanations for our results. Out of our 48 tests, 19% (9/48) identified a statistically significant decrease in the predicted probability of at least partially implementing the unhealthy commodity policy in question, while 2% (1/48) showed an increase. However, there was marked heterogeneity across policies. At the level of individual policies, US FTA participation was associated with a 37% reduction (95%CI: −0.51 to −0.22) in the probability of fully implementing graphic tobacco warning policies, and a 53% reduction (95%CI: −0.63 to −0.43) in the probability of at least partially implementing smoke-free place policies. EU FTA participation was associated with a 28% reduction (95%CI: −0.45 to −0.10) in the probability of fully implementing graphic tobacco warning policies, and a 25% reduction (95%CI: −0.47 to −0.03) in the probability of fully implementing restrictions on child marketing of unhealthy food and drinks. There was a positive association with implementing fat limits and bans, but this was not robust. Associations with other outcomes were not significant. The main limitations included residual confounding, limited ability to discern precise mechanisms of influence, and potentially limited generalisability to other FTAs.
US and EU FTA participation may reduce the probability of implementing WHO-recommended tobacco and child food marketing policies by between a quarter and a half—depending on the FTA and outcome in question. Governments negotiating or participating in US/EU FTAs may need to establish robust health protections and mitigation strategies to achieve their NCD mortality reduction targets.
Pepita Barlow and Luke Allen investigate the association between free trade agreement participation and implementation of WHO-recommended policies targeting unhealthy commodities.
Identifying and attending to the factors that inhibit the proper regulation of unhealthy commodities is a pressing priority for governments seeking to accelerate progress towards reducing noncommunicable diseases (NCDs).
US and EU Free Trade Agreements (FTAs) may play a significant role in stalling policy progress by incentivising and empowering unhealthy commodity producers to challenge policies targeting their products in FTA partner countries.
However, these agreements also acknowledge governments’ right to regulate and protect public health, and previous studies were unable to establish whether countries with US/EU FTAs are typically less successful at implementing unhealthy commodity policies.
We conducted a global statistical analysis assessing the relationship between US and EU FTA participation and implementation of WHO-recommended policies targeting unhealthy commodities.
Our large-scale quantitative approach allows for the incorporation of data from many more countries and time periods than previous approaches while addressing key alternative explanations in our main models and >30 additional analyses and robustness checks.
We identified a substantial reduction in the predicted probability of implementing select WHO-recommended policies in countries participating in US FTAs and EU FTAs, with the probability of implementing tobacco and child food marketing restrictions reducing by between a quarter and a half depending on the FTA and regulation in question; other associations were not significant.
Our findings indicate that participating in US and EU FTAs is associated with reduced implementation of select unhealthy commodity policies that are crucial to achieving global targets to prevent and reduce NCD-related mortality, morbidity, and associated treatment costs.
For countries currently negotiating US/EU FTAs, there is now a potential opportunity to ensure these agreements do not empower unhealthy commodity producers to challenge unhealthy commodity policies and instead empower governments to accelerate NCD policy progress.
For countries already participating in US/EU FTAs, governments will need to ensure their policies are not unduly influenced by vested interests that are often concealed in technical discussions about trade rules.
Noncommunicable diseases (NCDs) are responsible for more than 70% of global deaths [
Notes: Data from Allen and colleagues based on WHO NCD Country Capacity Surveys [
Policy-makers, academics, and civil society have long noted that producers of unhealthy commodities play a major role in stalling and undermining policy progress by forcefully opposing policies targeting their products [
US and EU Free Trade Agreements (FTAs) create opportunities for the large, multinational tobacco, food, and alcohol companies headquartered in these jurisdictions to oppose unhealthy commodity policies, as summarised in
A substantial body of scholarship documents the various ways that firms can and have used FTA clauses to influence policies targeting their products, summarised in
US/EU FTAs contain written recognition that governments have a legitimate right to protect public health. They also establish trade rules, investor protections, and dispute settlement procedures that unhealthy commodity producers can use to pressure governments to modify or abandon a policy initiating or threatening a trade
US/EU FTAs also enable and incentivise direct influence and
A recent systematic review of FTAs and health did not identify any quantitative studies that systematically examined the relationship between participation in US and EU FTAs and unhealthy commodity policy implementation [
Here we conduct a statistical analysis assessing the relationship between US/EU FTA participation and the implementation of WHO-recommended policies on tobacco, alcohol, and unhealthy food and drinks.
We assessed the relationship between US/EU FTA participation and the achievement of (i) partial or full implementation; and (ii) full implementation of policies targeting tobacco, alcohol, and unhealthy food and drinks in 127 countries with available covariate data in 2014, 2016, and 2019. We analysed 10 categories of WHO-recommended policies including taxes and restrictions on marketing, sales, and consumption, as described in
Tobacco
Tobacco taxes
Smoke-free place policies
Graphic warnings on cigarette packages
Tobacco advertising bans
Alcohol
Alcohol sales or advertising restrictions
Alcohol taxes
Unhealthy food and nonalcoholic beverages
Legislation implementing the International Code of Marketing of Breastmilk Substitutes.
Policies to reduce salt/ sodium consumption
Policies to limit saturated fatty acids and eliminate
Policies targeting the marketing of foods and nonalcoholic beverages to children
Notes: All categories listed above correspond to those originally captured in WHO country surveys and categorised therein, with the exception of alcohol sales and advertising restrictions. We grouped these into a single category as very few countries had implemented these policies and we sought to ensure there was variation in implementation across FTA partners. For each policy above, we create 2 dichotomous indicators capturing at least partial (i.e., partial or full) implementation of the regulation, and full implementation of the regulation.
Following a published protocol [
Our final models used full matching on the Mahalanobis distance, a composite measure of the differences in the characteristics of countries with and without US/EU FTAs [
We first estimated logistic regression models using unmatched data with controls for covariates, which may influence US or EU FTA participation and unhealthy commodity policies (see Appendix A in
Our baseline regression models are as follows:
Equation 1. US agreements.
Equation 2. EU agreements.
Finally, we use the estimated models to calculate average marginal effects (AMEs): differences in the predicted probability of implementation according to US/EU FTA participation status [
US FTAs | EU FTAs | |||||||
---|---|---|---|---|---|---|---|---|
Variable | Without |
With | Difference | Without | With | Difference | ||
GDP per capita ($) | 22,705.79 (21,053.97) | 26,201.55 (17,956.99) | −3,496.00 | 0.23 | 24,387.44 (21,672.89) | 15,660.89 (8,527.76) | 8,727.00 | <0.001 |
Polyarchy score (0–1) | 0.61 (0.25) | 0.62 (0.24) | −0.01 | 0.79 | 0.60 (0.26) | 0.70 (0.13) | −0.11 | <0.001 |
Domestic market liberalisation index (0–100) | 67.42 (14.56) | 71.16 (11.02) | −3.70 | 0.041 | 68.32 (14.60) | 65.78 (10.49) | 2.50 | 0.17 |
Proportion with secondary education (%) | 87.06 (30.16) | 94.12 (21.68) | −7.10 | 0.053 | 88.36 (30.23) | 86.56 (20.55) | 1.80 | 0.62 |
KOF Political globalisation index (0–100) | 73.54 (17.76) | 75.37 (12.63) | −1.80 | 0.39 | 74.36 (17.50) | 70.23 (13.64) | 4.10 | 0.081 |
WTO membership (0 or 1) | 0.59 (0.49) | 1.00 (0.00) | −0.41 | <0.001 | 0.93 (0.25) | 1.00 (0.00) | −0.07 | <0.001 |
FTA with another country or countries with large food company HQ(s) (0 or 1) | 0.55 (0.50) | 0.71 (0.46) | −0.17 | 0.02 | 0.60 (0.49) | 0.74 (0.44) | −0.14 | 0.053 |
FTA with another country or countries with large tobacco company HQ(s) (0 or 1) | 0.55 (0.50) | 0.84 (0.37) | −0.28 | <0.001 | 0.56 (0.50) | 0.63 (0.49) | −0.07 | 0.38 |
FTA with another country or countries with large alcohol company HQ(s) (0 or 1) | 0.59 (0.49) | 1.00 (0.00) | −0.41 | <0.001 | 0.56 (0.50) | 1.00 (0.00) | −0.44 | <0.001 |
Region | ||||||||
East Asia and Pacific | 0.08 (0.27) | 0.12 (0.33) | −0.05 | 0.36 | 0.10 (0.29) | 0.00 (0.00) | 0.10 | <0.001 |
Europe and Central Asia | 0.45 (0.50) | 0.00 (0.00) | 0.45 | <0.001 | 0.42 (0.49) | 0.14 (0.35) | 0.28 | <0.001 |
Latin America and Caribbean | 0.08 (0.28) | 0.57 (0.50) | −0.49 | <0.001 | 0.06 (0.24) | 0.79 (0.41) | −0.73 | <0.001 |
Middle East and North Africa | 0.05 (0.22) | 0.24 (0.43) | −0.19 | 0.003 | 0.09 (0.29) | 0.00 (0.00) | 0.09 | <0.001 |
North America | 0.01 (0.10) | 0.06 (0.24) | −0.05 | 0.16 | 0.02 (0.13) | 0.02 (0.15) | −0.01 | 0.82 |
South Asia | 0.07 (0.26) | 0.00 (0.00) | 0.07 | <0.001 | 0.07 (0.26) | 0.00 (0.00) | 0.07 | <0.001 |
Sub-Saharan Africa | 0.25 (0.44) | 0.00 (0.00) | 0.25 | <0.001 | 0.24 (0.43) | 0.05 (0.21) | 0.19 | <0.001 |
Sample size (country-years) | 276 | 48 | 282 | 43 |
FTA, Free Trade Agreement.
aMean or proportion (%).
b
For consistency with our statistical models, all variables are lagged by 1 year apart from secondary education, which is lagged by 2 years to ensure sufficient data availability.
Notes: Figure shows estimated average marginal effects with 95% CIs. AME, average marginal effect; CI, confidence interval; FTA, Free Trade Agreement.
US FTAs | EU FTAs | |||
---|---|---|---|---|
Partial or full implementation | Full implementation | Partial or full implementation | Full implementation | |
Tobacco taxes | 0.05 (−0.08 to 0.18) | 0.05 (−0.20 to 0.30) | −0.06 (−0.17 to 0.05) | −0.05 (−0.25 to 0.14) |
Smoke-free places | −0.05 (−0.16 to 0.07) | 0.14 (−0.11 to 0.39) | ||
Graphic warnings | ||||
Tobacco ad bans | 0.01 (−0.16 to 0.18) | 0.05 (−0.19 to 0.29) | −0.07 (−0.31 to 0.16) | −0.04 (−0.27 to 0.18) |
Alcohol ad and sales restrictions | −0.03 (−0.29 to 0.23) | 0.03 (−0.02 to 0.07) | −0.22 (−0.48 to 0.05) | −0.30 (−0.67 to 0.06) |
Alcohol taxes | −0.01 (−0.19 to 0.17) | −0.05 (−0.23 to 0.14) | 0.01 (−0.24 to 0.25) | 0.20 (−0.01 to 0.40) |
Salt reduction | −0.01 (−0.23 to 0.21) | −0.04 (−0.27 to 0.19) | 0.11 (−0.08 to 0.31) | 0.14 (−0.03 to 0.30) |
Fat limits and bans | −0.04 (−0.24 to 0.16) | −0.07 (−0.26 to 0.12) | 0.11 (−0.06 to 0.28) | |
Child marketing restrictions | 0.12 (−0.06 to 0.30) | 0.12 (−0.06 to 0.30) | − |
− |
Breast milk code | 0.18 (−0.03 to 0.40) | 0.19 (−0.06 to 0.45) | 0.14 (−0.07 to 0.34) | 0.09 (−0.13 to 0.30) |
Boldface indicates statistically significant results. Figures show the difference in the predicted probability of achieving partial/full or full implementation of a given policy among countries with and without either US FTAs (Columns 2–3) or EU FTAs (Columns 4–5). 95% CIs are shown in parentheses.
AME, average marginal effect; CI, confidence interval; FTA, Free Trade Agreement.
There were, however, substantial differences in the association between both US FTA and EU FTA participation and the probability of implementing each policy according to the specific policy and FTA in question. US FTA participation was associated with a 37% lower predicted probability of fully implementing graphic warning policies (95% CI: −0.51 to −0.22), a 24% lower (95% CI: −0.39 to −0.08) probability of fully implementing smoke-free place policies, and a 53% lower (95% CI: −0.63 to −0.43) probability of at least partially implementing the same.
EU FTA participation was associated with a 28% (95%CI: −0.45 to −0.10) lower probability of achieving full implementation of graphic warning policies, and a 25% (95% CI: −0.47 to −0.03) lower probability of fully implementing child marketing restrictions. Comparable associations were observed for partial implementation of graphic warning and child food marketing policies. EU FTA participation was also associated with a 16% (95% CI: 0.01 to 0.03) higher probability of partially implementing fat limits and bans, although this was not robust in additional analyses presented below.
Prematching regression models are presented in Tables D and E in Supporting Information.
In line with our protocol, we conducted 35 additional analyses and robustness checks to assess whether our results may be explained by alternative processes. We first conducted placebo analyses in which we reestimated our models examining implementation of 2 NCD policies that we would not expect to be affected by FTAs: whether or not a country has implemented risk factor surveys or time-bound national targets to address NCDs. Table F in
We further examined whether our results may apply to all FTAs with countries where large producers of unhealthy commodities are headquartered, rather than US and EU FTAs specifically. Table G in
We additionally examined whether our results may apply to Bilateral Investment Treaties (BITs) with the US and EU members, and whether our results were consistent when adjusting for participation in these BITs, as they contain some of the clauses included in FTAs that can be used to challenge policies, although their scope is heterogeneous and the pathways to influence may differ. Tables H-K in
Tables L and M in
Our analysis has shown that US and EU FTA participation is associated with a substantial reduction in the predicted probability of implementing several WHO-recommended NCD policies that target unhealthy commodities. Approximately 19% (9/48) of tests showed a statistically significant decrease in the predicted probability of at least partial unhealthy commodity policy implementation, while 2% (1/48) showed an increase, and the latter was not robust in additional analyses. On average, US FTA participation and EU FTA participation were associated with a 5% decline in the predicted probability of at least partially implementing the unhealthy commodity policy in question, but these averages were not statistically significant. However, we identified substantial changes in the probability of implementation for some policies. Specifically, US FTA participation was associated with a 37% reduction in the predicted probability of fully implementing graphic tobacco warning policies and a 53% reduction in the probability of at least partially implementing smoke-free place policies. The probability of fully implementing of smoke-free place policies was also 24% lower in countries with US FTAs. EU FTA participation was similarly associated with a reduced probability of implementing select policies, including a 28% reduction in the probability of fully implementing graphic tobacco warning policies and a 25% reduction in the probability of fully implementing restrictions on child marketing of unhealthy food and drinks. These findings were consistent in a large number of additional analyses and robustness checks.
Our study provides new insight into the relationship between US/EU FTA participation and (non)implementation of WHO-backed policies that seek to restrict the marketing, sale, and consumption of unhealthy commodities. US/EU FTAs acknowledge that governments have a legitimate right to regulate to protect public health. Furthermore, high-profile trade and investment disputes were raised against Australia and Uruguay’s tobacco packaging legislation, but the policies were ultimately deemed consistent with the treaties cited in each case [
Notably, no formal disputes related to WHO-recommended NCD policies were initiated during the time period of our study. However, a number of countries rescinded policies relating to tobacco and child food marketing restrictions [
Our study has important limitations. Our results should not be interpreted as definitively causal, as our quasi-experimental comparisons have important underlying assumptions. One is that the associations we identify are unconfounded after matching and incorporating regression controls [
Further research is needed to evaluate whether our results apply to other FTAs, and whether domestic political prioritisation of unhealthy commodity policies may counteract any influence of US/EU FTAs and associated industry pressure. Our results also indicate a need to investigate sources of heterogeneity in the associations we identified. For example, we identified variation in the relationship of US/EU FTA participation and the implementation of similar policies (e.g., advertising or sales restrictions) across different commodities. This variation might be explained by a wide range of factors, such as differences in the degree of contentiousness of a particular policy where it targets different commodities, and the novelty of the policy and existing implementation levels prior to our study period. We also identified variation in the association between FTA participation and the implementation of different policies within the same commodity category. This may be explained, for example, by differences in the ability of industry actors to craft arguments that relate different policies to FTA rules, and differences in the visibility of economic benefits of the policies in addition to health benefits.
Our findings have important implications for policy-makers seeking to accelerate progress towards regulating unhealthy commodities and achieving global targets to reduce NCD mortality. Our results suggest that FTAs currently under negotiation may constrain efforts to achieve NCD-related global health targets in partner countries. For example, several US and EU FTAs are now under negotiation, including a US–Kenya agreement; UK accession to the Comprehensive and Progressive Agreement for Trans-Pacific Partnership, which was heavily influenced by the US (UK-CPTPP); a potential future US–UK deal; and EU agreements with New Zealand, the Philippines, Indonesia, China, and Australia [
Our findings also have implications for existing US/EU FTA participants. Countries with these FTAs appear to encounter difficulties in regulating tobacco and child food marketing. However, industry references to clauses in US/EU FTAs can be invalid and may constitute attempts to limit policies affecting their products by appealing to aspects of trade law that are poorly understood by policy-makers. Governments should be aware of this potential conflation of vested interests with the interpretation of FTA clauses. Governments may also be better able to implement unhealthy commodity policies despite opposition from industry where they have access to legal experts that can identify invalid trade-related claims at an early stage, and where they minimise industry involvement in policy-making processes via regulatory cooperation and lobbying. Finally, the risk of industry threats might be minimised by strategically designing unhealthy commodity policies in ways that accord with US/EU FTA rules while maximising efficacy. Whether countries are seeking to mitigate impacts of existing US/EU FTAs or negotiating new agreements, effective cross-government cooperation between legal, trade, and public health officials will be essential to accelerate progress to implement unhealthy commodity policies.
(DOCX)
(DOCX)
average marginal effect
Bilateral Investment Treaty
Comprehensive and Progressive Agreement for Trans-Pacific Partnership
Free Trade Agreement
noncommunicable disease
Sustainable Development Goal
Dear Dr Barlow,
Thank you for submitting your manuscript entitled "US and EU Free Trade Agreements and implementation of policies to control tobacco, alcohol, and unhealthy food and drink: a quasi-experimental analysis" for consideration by PLOS Medicine.
Your manuscript has now been evaluated by the PLOS Medicine editorial staff and I am writing to let you know that we would like to send your submission out for external peer review.
However, before we can send your manuscript to reviewers, we need you to complete your submission by providing the metadata that is required for full assessment. To this end, please login to Editorial Manager where you will find the paper in the 'Submissions Needing Revisions' folder on your homepage. Please click 'Revise Submission' from the Action Links and complete all additional questions in the submission questionnaire.
Please re-submit your manuscript within two working days, i.e. by Aug 08 2022 11:59PM.
Login to Editorial Manager here:
Once your full submission is complete, your paper will undergo a series of checks in preparation for peer review. Once your manuscript has passed all checks it will be sent out for review.
Feel free to email us at
Kind regards,
Callam Davidson
Associate Editor
PLOS Medicine
Dear Dr. Barlow,
Thank you very much for submitting your manuscript "US and EU Free Trade Agreements and implementation of policies to control tobacco, alcohol, and unhealthy food and drink: a quasi-experimental analysis" (PMEDICINE-D-22-02643R1) for consideration at PLOS Medicine.
Your paper was evaluated by an associate editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:
[LINK]
In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.
In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at
In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements:
We hope to receive your revised manuscript by Oct 19 2022 11:59PM. Please email us (
***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***
We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here:
Please use the following link to submit the revised manuscript:
Your article can be found in the "Submissions Needing Revision" folder.
To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at
Please ensure that the paper adheres to the PLOS Data Availability Policy (see
We look forward to receiving your revised manuscript.
Sincerely,
Callam Davidson,
PLOS Medicine
-----------------------------------------------------------
Comments from the Academic Editor:
My main comment was that out of 10 policies and 4 contexts (40 different tests), only 7 were significant in the direction that the paper focuses on (one significant in opposite direction, all others null). Whilst the authors note in the discussion that "a need to investigate why impacts may vary across regulatory domains and country-partners", I'd like to see a bit more evidence/theory-informed speculation here. Why might it be that FTAs impeded child food marketing restrictions, but not tobacco ad bans or alcohol ad and sales restrictions (ie why does the impact on essentially the same policies differ between commodity)? Similarly, why might FTAs impede smoke free places, but not tobacco ad bans/taxes (ie why do they impact on some but not all tobacco control efforts)? I suspect this is something to do with both how different commodities are 'valued' in society (ie tobacco universally considered 'bad', alcohol less so); and the co-effects of different policies (ie taxes are revenue raising, not JUST health promoting).
Requests from the editors:
Please structure your abstract using the PLOS Medicine headings (Background, Methods and Findings, Conclusions).
Please avoid assertions of primacy (“We conducted the first…”) in your Abstract Background. Instead, the final sentence should clearly state the study question.
Assertions of primacy are also present in the Discussion and should be tempered accordingly (e.g., “to our knowledge”, or similar).
In the Abstract Methods and Findings, please include the study design and main outcome measures.
In the last sentence of the Abstract Methods and Findings section, please describe the main limitation(s) of the study's methodology.
Please remove the Funding section from your Abstract.
Please include continuous line numbering throughout your manuscript to facilitate further review.
At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information:
Please place citations within square brackets and preceding punctuation (e.g., [1]). This also applies to your appendices.
Please organise and cite your appendices as outlined here:
The footnotes in Appendix 4 appear to be labelled incorrectly.
Please ensure that the study is reported according to the STROBE guideline, and include the completed STROBE checklist as Supporting Information. Please add the following statement, or similar, to the Methods: "This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 Checklist)."
The STROBE guideline can be found here:
When completing the checklist, please use section and paragraph numbers, rather than page numbers.
Table 1: Please define Free Trade Agreement (FTA) in the legend.
Figure 4: Please define the 95% confidence interval bars in the legend.
Reference 26 (your protocol) is a preprint: please add [preprint] to the reference to identify it as such. Further details can be found here:
Comments from the reviewers:
Reviewer #1: Thanks for the opportunity to review your manuscript. My role is as a statistical reviewer, so my review concentrates on the study design, data, and analysis that are presented. I have put general questions first, followed by queries relevant to a specific section of the manuscript (with a page/paragraph reference).
This manuscript examines whether participation in a free-trade agreement is associated with the implementation of policies that target tobacco/alcohol/unhealthy food and drinks. Data is at the country level from three time points (2014, 2016, 2019), policy implementation data comes from WHO country surveys (based on WHO recommended policies) and information on free trade agreements is from collaborative database. A range of additional covariates is available for secondary analysis of heterogeneity of the TIA effect. The WHO specific recommended policies includes tobacco and alcohol taxes and restrictions, graphic warnings on tobacco packing, legislation on breast-milk substitutes and policies that target sodium, sat and trans fats, and food marketing to children. In the main analyses, the outcomes are assessed as unimplemented or fully implemented (and with alternative coding considered in sensitivity analyses). The 'exposure' variable for the main analysis is binary (part of TIA/not part of TIA). In the main analysis a matching process (based on Manholobis distance of the matching variables) to limit confounding is used. Where balance wasn't achieved, the covariate was included in the main logistic regression model. The protocol is very detailed (and well explained), there are some deviations from this (detailed in app 5), these are quite reasonable and reflect changes required once features of the final dataset were apparent (e.g. not enough variation to get sufficient sub-groups for some secondary analyses). Limitations of the study design are well articulated in the discussion.
How is Mahalanobis distance calculated for differences in the region each country belongs to? I am more familiar with Mahalanobis distance calculated for continuous variables and I wasn't sure how it was done here.
With the Mahalanobis matching approach, is there a straightforward way to assess common support, i.e. that there is sufficient overlap between the FTA groups in the covariates or are the countries with FTAs fundamentally different from those without?
My usual area of expertise is far away from free-trade agreements - so the following question is probably more for my own understanding than any issue with the manuscript. Are the FTAs similar enough to be that we can represent them as a universal, binary covariate? I ask because it seems that mainstream media reporting around FTAs always seems to emphasise differences between FTAs (for my country at least the reporting is always that 'we were ripped off compared to country XX') and I am curious if perhaps FTAs are broadly similar and I have a distorted impression of reality.
P6, Paragraph 2. What criteria were used to decide between purely regression adjustment and matching (once it was clear first order models weren't appropriate)?
P7, Paragraph 3. This could be my reading of it but it seems like the part:
"FTAit is one of two FTA indicators in country i with coefficient B1: i) US FTA participation, or ii) EU FTA participation, lagged by 1 year (t-1) to allow for a delayed effect; we estimate separate models for US FTAs and EU FTAs."
is contradictory, if there is an indicator variable for EU or US participation woudn't that make the variable unvarying if separate models for US/EU are run?
With this model each country will contribute multiple 'rows' of data regression model, i.e. there are repeated measurements (of different times) of the same countries. Is this correlation accounted for by the block bootstrap procedure? Also - is this the 'simple' version of block/cluster bootstrapping where an entire block/country is selected in the sampling, or is this the more complex procedure where there is second level of sampling of years within the selected blocks?
Also to confirm, is country the 'block' for the block bootstrapping?
P19. Table 1. I would consider removing the p-values here, the rest of the table provides everything necessary to understand the differences.
Should the title of the table include "EU FTAs"?
Under the EU FTAs there is a typo - "differesnce" should be difference.
P18. What was the minimum number per category you considered to be appropriate allow for sufficient statistical power?
Reviewer #2: This is an excellent and informative paper. Well done. Before publishing there are some exceedingly minor changes to make:
p. 5 'first quant study' is a bold claim that needs nuance. A google search shows there have been lots of (non-peer reviewed?) 'studies', impact assessments and so forth in this space. Also next section suggests a few of these studies. Suggest paring back this claim or adding nuance.
Typo p. 10 'All results are consistent in sign for all models'. 'Significance?'
P. 12 'window of opportunity' is vague. What is the window and why has this come about?
Reviewer #3: This paper uses natural experiment methods to assess the association of US or EU Free Trade Agreement participation with the implementation of various health policies. The paper thoroughly assembles what data is available for countries as the unit of analysis, and uses a matching procedure to try and approximate an unconfounded comparison. Unfortunately, not many matched country-pairs make it through to the final analytical sample - a necessary cost of trying to remove confounding, and increase internal validity.
There are 48 'tests' of whether FTA participation effects health policies: 12 policies, by 4 combinations of EU/US FTA and Full or partial/Full classifications. Figure 4 shows these 48 'tests' as a forest plot, correctly portraying the mixed findings and lack of an overwhelming pattern. That said, there is a pattern - 9 of the 48 'tests' (19%) have 95% CI excluding the null AND a finding of worse health policy implementation if the country is a participant of a FTA, whereas only 1 'test' shows better health policy. Averaged across all 48 tests, the average percentage difference was -4.4%.
The study is clearly worthy of publication. And it will generate substantial discussion and debate - rightly so. This is the first paper to quantitatively test this hypothesis in a comprehensive manner - testing a hypothesis long suspected.
I have only a few recommendations for the authors in revision.
1. It is not clear on reading the abstract that 48 tests were run, than many were null, etc. An indication of the PATTERN of 19% of tests showing a statistically significant decreases, compared to only one (2%) an increase, would help. Moreover, as a more thorough sense of the whole pattern, giving the average effect (-4.4%) with a 95% CI would be useful (which may require some careful math to allow for correlations of the partial / partial or full classification).
2. Perhaps point out in the paper that pulling out the 9 'positive' tests for profiling is a bit risky. To many this falls foul of a statistical testing approach. Hence my suggestions above to give the average effect size. Or alternatively, the percentage of 'tests' with the effect size stronger than -20%, between --20% to -10%, etc - to convey a focus on magnitude of the associations.
3. The matching in Table 1 does not look as good as the text of the paper implies, or the absolute standardized means in Fig 3. For example, the absolute standardized mean differences in GDP in Fig 3 nicely moves almost to nil after matching - but not in Table 1. (Actually in Table 1, the differences in GDP remain, just flipped in sign.) Either I have misunderstood Table 1, or there is perhaps an error in the data in Table 1?
4. Box 2's title implies this is a WHO categorization - but it is actually the categorization used for analysis in this paper. Perhaps make this clearer.
Congratulations to the authors on an important paper.
Reviewer #4: This paper shines a light on an important issue and methodologically speaking, is executed superbly. The authors write in an accessible and clear way about a highly complex and technical public health issue. My one, very minor, comment is that the authors might consider including the UK-CPTPP deal as an important US led agreement under negotation ( bottom of pg. 4 )
Any attachments provided with reviews can be seen via the following link:
[LINK]
Submitted filename:
Dear Dr. Barlow,
Thank you very much for re-submitting your manuscript "US and EU Free Trade Agreements and implementation of policies to control tobacco, alcohol, and unhealthy food and drink: a quasi-experimental analysis" (PMEDICINE-D-22-02643R2) for review by PLOS Medicine.
I have discussed the paper with my colleagues and the academic editor and it was also seen again by two reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.
The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:
[LINK]
***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***
In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.
Please also check the guidelines for revised papers at
We hope to receive your revised manuscript within 1 week. Please email us (
We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.
Please ensure that the paper adheres to the PLOS Data Availability Policy (see
To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at
Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript.
Please note, when your manuscript is accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you've already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at
If you have any questions in the meantime, please contact me or the journal staff on
We look forward to receiving the revised manuscript by Nov 11 2022 11:59PM.
Sincerely,
Callam Davidson,
Associate Editor
PLOS Medicine
------------------------------------------------------------
Requests from Editors:
The URL in your Data Availability Statement leads to an error page - please correct or confirm this link will be made active upon publication.
Please update your Author Summary to use 2-3 single sentence bullet points per question (please see here for further guidelines:
Please check the formatting of your References (including Supplementary) is consistent (e.g., list first six authors before use of et al.) and follows our guidelines (
Comments from Reviewers:
Reviewer #1: Thanks for the revised manuscript and responses to my review. The updates to the manuscript resolve all my original queries. A great piece of work.
Only one very small change - 'per-protocol' and 'protocol deviations' both have specific meaning to in clinical trials which are different to how they are used here. It might make S2 Appendix clearer to describe these as 'changes from pre-registered protocol' or 'according to pre-registered protocol'.
Reviewer #3: My comments have been satisfactorily addressed
Any attachments provided with reviews can be seen via the following link:
[LINK]
Dear Dr Barlow,
On behalf of my colleagues and the Academic Editor, Professor Jean Adams, I am pleased to inform you that we have agreed to publish your manuscript "US and EU Free Trade Agreements and implementation of policies to control tobacco, alcohol, and unhealthy food and drink: a quasi-experimental analysis" (PMEDICINE-D-22-02643R3) in PLOS Medicine.
Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Once you have received these formatting requests, please note that your manuscript will not be scheduled for publication until you have made the required changes.
In the meantime, please log into Editorial Manager at
PRESS
We frequently collaborate with press offices. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximise its impact. If the press office is planning to promote your findings, we would be grateful if they could coordinate with
We also ask that you take this opportunity to read our Embargo Policy regarding the discussion, promotion and media coverage of work that is yet to be published by PLOS. As your manuscript is not yet published, it is bound by the conditions of our Embargo Policy. Please be aware that this policy is in place both to ensure that any press coverage of your article is fully substantiated and to provide a direct link between such coverage and the published work. For full details of our Embargo Policy, please visit
To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at
Thank you again for submitting to PLOS Medicine. We look forward to publishing your paper.
Sincerely,
Callam Davidson
Associate Editor
PLOS Medicine