Référence - NCCHPP



Extraction table used during the literature review conducted for the following document:

Morestin, F., Hogue, M.-C., Jacques, M. & Benoit, F. (2011). Public policies on nutrition labelling: Effects and implementation issues — A knowledge synthesis. Montreal: National Collaborating Centre for Healthy Public Policy.

The documents are presented in reverse chronological order and alphabetically by first author.

Reference |Characteristics of Document |Status |Effectiveness |Unintended Effects |Equity |Cost |Feasibility |Acceptability | |Canadian Food Inspection Agency, 2009 |Grey literature; source = CFIA website; description of CFIA activities; authors = government |“Health Canada and the Canadian Food Inspection Agency (CFIA) carry joint responsibility for federal food labelling policies.” “Health Canada is responsible for setting food labelling policies with respect to health and safety matters. CFIA is responsible for the development of non-health and safety food labelling regulations and policies.” | | | | | | | |Center for Science in the Public Interest, 2009 |Grey literature; source = CSPI website; fact sheet (primary study, cross-sectional survey, 755 persons surveyed); authors = lobbyists |In 2008, New York City was the first American city to require posting calories on the menus and menu boards of fast-food restaurants and other chain restaurants. |89% of respondents were surprised by the number of calories posted (higher than they thought).

Of those who frequented restaurants, 82% said that the postings had an impact on their selections. Among those who took the nutrition information into account, 71% looked for lower-calorie options and 51% no longer ordered some items. | | | | |89% of the residents surveyed felt that the labelling was a positive change. 76% felt that governments at all levels should play a more active role regulating nutritional issues in restaurants. | |Centre for Science in the Public Interest – Canada, 2009a |Grey literature; source = CSPI website; press release; authors = lobbyists |- April 2009, Bill 156 tabled by NDP member (for posting calorie counts on the menus and menu boards of large chain restaurants) was adopted in its second reading in the Ontario Legislature.

- Similar measures exist in the UK, New York, California and other American jurisdictions. | | | | | |- Earlier in the week, the Ontario Medical Association asked the Ontario government for calorie counts to be posted in chain restaurants and school cafeterias. | |Centre for Science in the Public Interest – Canada, 2009b

(cont’d) |Grey literature; source = CSPI website; press release; authors = lobbyists |- In 2000, Health Canada decided to exempt restaurants from the requirement to provide nutrition information.

- In 2006, the federal parliament also dropped the issue in exchange for promises that the large chain restaurants would voluntarily present this information.

- Bill 156 (proposing mandatory posting of calories on menus by the large chain restaurants) was tabled in the Ontario legislature in 2009 by the NDP member from Sudbury

- In the U.S., in October 2008, Yum! Brands (which owns KFC, Pizza Hut and Taco Bell) promised to provide calorie counts on its menus.

- In April 2009, the Minister of Health of the UK and the Food Standards Agency announced the names of 18 large chain restaurants that were ready to post calorie counts in the following few months. | | | | |- In 2006, the federal parliament also dropped the issue in exchange for promises that the large chain restaurants would voluntarily present this information.

- In the Fall of 2007, two thirds of chain restaurants had not applied the (already weak) standards of the nutrition information program of the Canadian Restaurant and Food Service Association. |The Ontario Medical Association asked the Government of Ontario to require the posting of calorie counts in chain restaurants and school cafeterias. | |Coalition québécoise sur la probléma-tique du poids, 2009a |Grey literature; source = environmental scan; press release; authors = lobbyists |- Logos appearing on the front of packages are unregulated. 

- The criteria set by industry.  | | | | |Presence of logos on certain products is highly questionable, given their low nutritional value |- The Coalition and Options Consommateurs (a consumer protection organization) have asked Health Canada to regulate logos so that consumers will be able to trust them. | |Coalition québécoise sur la probléma-tique du poids, 2009b

(cont’d) |Grey literature; source = environmental scan; presentation at a press conference (nutrition analyses on products with health logos); authors = lobbyists |- In Canada, there are no regulations governing front-of-package logos. 

- The industry sets the criteria for logos.  | | | | |Study of several products with “health” logos (Smart Selections by PepsiCo, Snack Wise by Old Dutch, Sensible Solution by Kraft): shows that they are only healthy with respect to some ingredients, or do not merit the logo when compared with even less healthy products |- The Coalition has asked for regulatory control of front-of-package logos.  | |Engelhard, Garson, & Dorn, 2009

(cont’d) |Grey literature; source = website; comment; authors = an independent, non-profit research centre (study funded by a health insurance company) |- The NLEA (the Nutrition Labelling and Education Act) was proclaimed in 1990: labelling of packaged foods

- Most developed countries are planning front-of-package labelling to complete nutrition facts table; already widespread on a voluntary basis in Europe and the UK.

- System of stars in Hannaford supermarkets, implemented in Sept. 2006. 28% of their 25,500 products receive 1 or more stars. After testing, program extended to over 1,300 stores.

- Laws on posting of nutrition info on menus in chain restaurants adopted in Philadelphia, NYC, California, Maine, Massachusetts, and Oregon, and are proposed in two bills before Congress. |- Half of the food budget and a third of calories are consumed in the U.S. = in food services.

- NLEA appears to have shifted consumption toward more healthy products.

- Consumers say that they often use these labels as guidance on purchases, but studies show that actual use is much lower than reported use, and fewer consumers use labels now than in the past.

- Studies show that the nutrition facts table often creates confusion (too much information).

- Many studies show that front-of-package labels significantly improve consumers’ ability to understand and use nutrition information.

- More effective format (studies in the UK and Australia in 2009): multiple traffic lights with % of daily intake.

- Sales figures showed shift to healthier prepared foods after logos introduced at Tesco.

- After stars introduced at Hannaford: packaged foods and frozen meals with stars were chosen 2.5 to 4.5 times more often than those without the stars.

- The many types of labels on packages creates confusion among consumers in the UK and the U.S. | | | |Restaurants: In June 2009, one of the bills before Congress received the support of the National Restaurant Association (see efforts made to have a single national standard for nutrition labels).

Food:

- The industry needs some time to implement the standard; more time may be given to small companies. |- Consumers have shown interest in Hannaford’s star system.

| |Health Canada, 2009a

(cont’d)

(cont’d) |Grey literature; source = Health Canada website; description of Health Canada’s actions; authors = the government |- Before adoption of regulations, nutrition regulations were optional. Not uniformly applied.

- “Through the Food and Drug Act, Health Canada regulates the labelling of food products in Canada. Regulations published on January 1, 2003 make nutrition labelling mandatory on most food lables.” Since December 2005, mandatory labelling for most pre-packaged foods. Small businesses had to comply with the new regulation starting December 12, 2007.

- Content of the nutrition facts table: caloric value and content in 13 core nutrients: fats, saturated fats, trans fats, cholesterol, sodium, carbohydrates, fibre, sugars, protein, vitamin A, vitamin C, calcium and iron.

- Inspired by the model used in the U.S., particularly its graphic design, which earned prizes.

- The nutrition facts table is not mandatory for other foods, including those served or sold in restaurants. However, consumers need to be encouraged to look for and demand nutrition information. For example, pamphlets are sometimes available upon requests in restaurants.

- “Front of package (FOP) labelling is the term sometimes used for simplified nutrition or health-related claims, symbols and logos that are highlighted and positioned on the front of a food package.”

- Manufacturers sometimes use symbols or logos on the front of packages to communicate these messages, implicitly and simply.

- Nov. 2007 to March 2008, Health Canada carried out consultations on health claims with researchers, consumer groups, the federal gov’t, provincial and municipal gov’ts, the industry, institutions and health professionals in order to develop a new frame of reference. Mailings, workshops held in Vancouver, Winnipeg, St-Hyacinthe, Halifax, Edmonton and Toronto. | | | | |- Regulation on nutrition labelling for foods was developed following extensive consultations with consumers, health-related interest groups and the food industry.

- The Canadian Food Inspection Agency was charged with enforcing the regulation.

- Nov. 2007 to March 2008, Health Canada carried out consultations on health claims with researchers, consumer groups, the federal gov’t, provincial and municipal gov’ts, the industry, institutions and health professionals in order to develop a new frame of reference. Mailings, workshops held in Vancouver, Winnipeg, St-Hyacinthe, Halifax, Edmonton and Toronto. | | |Health Canada, 2009b

(cont’d) |Grey literature; source = environmental scan; action plan; authors = government |- Consultations on managing health claims for food: online publication of a discussion paper; meetings with 286 stakeholders in 6 cities; written feedback using a questionnaire.

- Following the consultations, Health Canada developed a five-year plan.

- Regarding front-of-package labelling and implicit health claims: “Health Canada [proposes] to undertake consumer research on the interpretation of FOP labelling in concert with the Nutrition Facts table and to examine the standardization of the nutritional criteria underlying FOP symbols and claims.”

- Current measure: analysis at the national and international scale of the nutritional criteria used, the definitions of “healthy” foods, the potential obstacles to implementing this type of measure and its potential impacts on public health. 

- Next step: targeted consultations |- “Health Canada [proposes] to undertake consumer research on the interpretation of FOP labelling in concert with the Nutrition Facts table.”  | | | |- Consultations on managing health claims for foods: online publication of a discussion paper; meetings with 286 stakeholders in 6 cities; written feedback using a questionnaire.

- Following the consultations, Health Canada developed a five-year plan.

- Next step: targeted consultations | | |Heart and Stroke Foundation, 2009a

(cont’d)

|Grey literature; source = Heart and Stroke Foundation website; presentation of activities; authors = Foundation |- To receive the Health Check logo, a food or meal must satisfy the nutritional criteria of Canada’s Health Guide. Different criteria for each product category. Neither the food industry nor restaurants participate in developing these criteria.

- Once accepted in the Health Check program, companies pay a modest annual fee to cover program operating costs.

- Participating restaurants must make nutrition info available for each meal carrying the logo.

- Companies submit products on a voluntary basis.

- Over 1,700 products, including fresh products, carry the Health Check logo. It also appears on the menus of several chain restaurants.

- But there are also healthy products that do not carry the logo: the nutrition facts table remains the best way to compare products.

- H&S dieticians provide companies with suggestions on how to modify recipes to meet program criteria.

- Participating restaurants must create food preparation standards and procedures.

- Informative material provided by H&S to help train staff. | | | | |[Recap]

- Companies submit products on a voluntary basis.

- Over 1,700 products, including fresh products, carry the Health Check logo. It also appears on the menus of several chain restaurants.

| | |Heart and Stroke Foundation, 2009b |Grey literature; source = Health Check website (Canada); fact sheet (presenting opinion surveys); authors = NGO | |- 90% of consumers had seen the Health Check logo on products.

- 74% said that the logo helps them to make good choices.

- 74% are more likely to buy a product with the logo. | | | | |80% trust Health Check because it’s a Heart and Stroke Foundation program. Consumers also believe that the program is based on science/research/input from dieticians | |Kiesel & Villas-Boas, 2009

(cont’d) |Grey literature (thesis – University of California, Berkeley); source = PsycINFO; primary study (controlled randomized trial, 5 supermarkets in the test, 27 control subjects); authors = academics |In response to the declining use of labels, the FDA is considering changing their format and content. |- Other studies: Consumer interest in the labels; 50% use them in their purchase decision making; use associated with weight loss and probability of obesity; decline in sales of high-fat vinaigrettes after NLEA is passed. But declining use from 1995-2006, inability to interpret the labels, increased sales of only certain types of healthy products.

- This study: labels placed on shelves rather than on products; experiment with popcorn; sales data measures.

- The “No trans fat” label significantly increases sales (even if info was already available on the NLEA label and sometimes in the company’s written notices; the effect could therefore be attributable to the standardized info format, which allows consumers faster product comparisons).

- “Low fat” label reduces sales (consumers may perceive that low-fat products are less tasty, and popcorn is a treat: taste issue may be more important than health issue).

- “Low calorie” label increases sales.

- No apparent evidence that consumers infer that unlabelled products are of inferior quality (since in this case only healthier products, in terms of calories, fat and trans fat, are labelled).

- Labels showing information on several nutrients have no effect on sales (consumers may be less likely to interpret info when there is too much of it).

- Strongest impact seen right after implementation (4-week experiment, data collected 5 weeks before and after); impact dissipates after the study period for calories and fat, but not for trans fat (probably because it is easier to verify products for this nutriment). | |- Less use of NLEA labels, particularly among young adults and less educated consumers. | | |- Consumers prefer having logos on the front of the package (vs. labels). | |Kohn, 2009 |Grey literature; source = environmental scan; reporting; authors = health professionals |A law on menu labelling in the State of California comes into force in 2011. |- Calorie labelling in restaurants led 25% of clients to make healthier choices. |Restaurants changed their menus to include lower-calorie options. | | | |Big Cities Health Coalition (group of 18 municipal or county health departments representing 14% of the U.S. population) asked the federal government to support their efforts to prevent chronic illnesses. Proposed interventions include requiring chain restaurants to post easy-to-understand nutrition info on their menus. | |Robert Wood Johnson Foundation, 2009

(cont’d)

(cont’d)

|Grey literature; source = CSPI website; comments (non-systematic review of the literature); authors = Johnson & Johnson Foundation |NLEA 1990 applies to packaged foods. Restaurants are not subject to federal laws unless they make nutritional claims. Few other regulations have been adopted at this time. Examples: NYC, first state law (California) requiring posting of calories in chain restaurants starting January 2011.

- In 2005, the National Restaurant Association (U.S.) launched the Ask us! program to help restaurants provide nutrition info. |- Cites Wootan studies: number of American restaurants providing nutrition info has increased over the last 10 years, but most still do not present info at point of sale.

- Nutrition info is less available in restaurants not belonging to chains and not providing fast food.

- If nutrition info is not apparent enough, it is not frequently used.

- 63% of adults use nutrition labels when choosing packaged foods.

- Consumers who read labels on packaged foods tend to have healthier diets.

- From 1991 to 1995, sales of low-fat products increased.

- Cites LA County study: annual weight gain will be reduced 39% if restaurant labelling made 10% of clients make choices that reduce consumption by 100 calories.

- Posting reduces intentions to buy high-calorie and high-fat meals (many studies of teenagers 11-18 years old: 29% changed their choice when they saw nutrition info, and 46% of them made choices with fewer calories), above all if there is a large difference between perceived and actual content. But intentions are subject to social desirability bias, as much as respondents were not interested in paying the difference.

- Reality of purchases: modest improvement in selections (according to 14 studies); 6 studies found no evidence of a positive impact.

- Restaurants’ health claims made consumers underestimate the number of calories in main courses and order higher-calorie side dishes, drinks and desserts.

- Other important factors in selections (price, taste, convenience) often run counter to a healthy diet. In a randomized trial that tested 4 different labelling (or non-labelling) conditions and value pricing (or lack thereof) (lower per-unit prices on larger portions), no differences found in the food ordered by different groups.

- Many consumers have problems interpreting calorie info in terms of the recommended daily amounts (implies that labelling needs to be accompanied by nutrition education).

- Lack of studies on repeated exposure to labels. |- From 1991 to 1995, the number of low-fat products increased. This suggests that a similar change may occur in restaurants (no research on this subject).

- Possible impacts of labelling: theoretically, restaurant revenues could fall, particularly those with high-calorie menus, but no evidence currently available on this subject. Desire to spend in restaurants is mainly determined by disposable income, convenience and taste. On the other hand, purchases could be transferred to other restaurants. Few studies have measured the impact of labelling on consumer spending.

- Labelling could reduce demand for menu items that generate higher profits. |- Labelling could lead to men (particularly young men) consuming more calories. Other studies suggest that men use nutrition info less than women. According to a study of college students, men use nutrition info mainly to gain weight (as compared to women). | |- Chefs are not accustomed to following standardized menus. Custom orders would change nutrition values from those posted.

- Labelling would reduce menu planning flexibility.

- Labelling is difficult when menu has many variations.

- Little space on menus for nutrition labelling.

- May be difficult to train staff to answer questions on the nutrition info provided. |- Most consumers (over 60% in 4 different surveys) would like a law passed on info in restaurants (even if they do not use the info each time). Cites O’Dougherty: Those in favour believe it would help them make informed decisions and would make restaurants accountable; opponents believe that the law would place too great a burden on restaurants and not be effective in changing consumers’ decisions. | |Center for Science in the Public Interest, 2008a

(cont’d) |Grey literature; source = CSPI website; fact sheet; authors = lobbyists | |- Half of the chain restaurants already post nutrition info on their websites.

- Labels on packaged foods: ¾ of adults say that they use them. Use associated with healthier choices.

- Info provided in restaurants helps consumers choose low-calorie products. |Labels on packaged foods led to the reformulation of some products. | |- Half of chain restaurants already post nutrition info on their websites and will not incur additional costs to analyze products. Others: analysis software costs $500; the cost of a dietician or laboratory analysis is similar to the cost of a new commercial dishwashing machine.

- In most chain restaurants, analysis and printing costs are assumed by head office.

- Printed and illuminated menus are already often changed for marketing reasons.

- Labelling requirements apply to chains only, not to small restaurants. |- Concerns regarding feasibility of special orders and combinations of items, but labelling only applies to standard items on the menu.

| | |Center for Science in the Public Interest, 2008b

(cont’d) |Grey literature; source = CSPI website; literature review (non-systematic? Not specified); authors = lobbyists | |PACKAGED FOODS

Nutrition info has an effect on attitudes toward products.

Use of labels has a positive impact on food quality.

Lower sales of high-fat vinaigrettes after enactment of NLEA.

One study (Variyam 2006) associated NLEA with lower body weight and lower probability of obesity in the general public.

The FDA believes that listing trans fats on labels will save 2,100 to 5,600 lives per year.

RESTAURANTS

Labelling changes the attitudes of customers and their purchase intentions toward more healthy alternatives.

Nutrition info is associated with healthier, lower-calorie purchases, particularly when the levels listed exceed consumers’ estimates.

It has been estimated that in California, if 80% of customers see the info on labels in fast food restaurants, the weight loss would be one pound per person per year (vs. the one-pound gain currently observed).

Los Angeles County Public Health estimates that in LA County, labelling could prevent 39% of current weight gains.

NYC Department of Health estimates that, over 5 years, 150,000 fewer people would be obese, resulting in at least 30,000 fewer cases of diabetes.

|Between the intro. of NLEA and 1996, more new products were introduced with fewer negative nutrients. More positive nutrients were also added to existing products (without more negative nutrients being removed). |A study by Variyam (2006) found that the NLEA had a greater impact on body weight reduction and the probability of obesity among white non-Hispanic women.

- Labels in restaurants used more by women (students); students used labels to identify food with higher levels of proteins.

- One study found that nutrition info had a positive impact on persons who were not dieting, but no change among people who were dieting. |- The FDA has estimated that listing trans fats on labels would save $3 billion to $8 billion per year.

- The economic benefit of the NLEA over 20 years due to reduced weight and risk of obesity is $63 billion to $166 billion (Variyam 2006). | |- 78% of Americans support restaurant labelling policies.

- 84% of respondents in NY found the new labelling regulation useful. | |Center for Science in the Public Interest, 2008c

|Grey literature; source = CSPI website; fact sheet; authors = lobbyists |Since 1994, the NLEA deals with labels on packaged foods but explicitly excludes restaurants. |- Half of chain restaurants provide no nutrition info.

- Of those that do: often by means that are not very visible to consumers when they place their orders (website, which requires access to a computer and the Internet; on tray liners or food packaging, which are only visible after the order; brochures and posters – but less than 5% of consumers see this info).

- Americans consume one third of their calories outside the home.

- Labels on packaged foods: ¾ of adults say that they use them. Use associated with healthier choices.

- Info in restaurants helps consumers make lower-calorie choices. |Labels on packaged foods have led to the reformulation of some products (particularly in the case of trans fats). | | | |- [Vs. coercion]: The authors argue that having labels on menus addresses the consumer’s right to be informed and allows consumers to exercise personal responsibility.

- 78% of Americans support labels on menus.

- Nutrition info in restaurants is recommended by the Institute of Medicine, the Food and Drug Administration, the Surgeon General, the U.S. Department of Health and Human Services, the National Cancer Institute and the American Medical Association. | |Center for Science in the Public Interest, 2008d

(cont’d) |Grey literature; source = CSPI website; fact sheet; authors = lobbyists |Trans fat labelling requirements came into effect in January 2006.

Policies on menu labelling have been adopted in California, Philadelphia and Multnomah County (Portland, Oregon) and applied in NYC. |Immediately following enactment of the NLEA, the demand for low-fat products and products low in sugar grew by 65%. | |- Faced with the post-NLEA change in demand, 79% of firms surveyed said that they had introduced new products.

- Since trans fat labelling was introduced, the use of partially hydrogenated oils in North America has fallen 50%.

- Following the intro of labelling in NYC restaurants, calories were significantly reduced in many dishes. | | | | |Center for Science in the Public Interest, 2008e |Grey literature; source = CSPI website; fact sheet (summary of 4 national surveys, 1 in Connecticut and 1 in California); authors = lobbyists | | | | | | |Surveys found that 67% to 84% of respondents favoured nutrition labelling in restaurants. | |Centre for Science in the Public Interest – Canada, 2008a

(cont’d)

|Grey literature; source = CSPI Canada website; press release (on a cross-sectional observation survey of 136 Canadian chain restaurants); authors = lobbyists |- The Canadian Restaurant and Foodservice Association created a program in 2006 (commitment to make nutrition info available in brochures in restaurants and on websites).

- In December 2007, Health Canada began consultations on more detailed labelling provisions for packaged foods, but did not discuss labelling in restaurants.

- In the U.S.: NYC and Seattle have adopted restaurant labelling laws. Governor Schwarzenegger vetoed a bill adopted by the California Assembly and Senate. |. | | | |- The Canadian Restaurant and Foodservice Association’s program was created to convince members of Parliament to defeat Bill C-283 on nutrition labels for chain restaurant menus.

- Among the 27 chain restaurants participating in the program, 18 have not met their commitment, and none post the info on menus.

- In 2006 the Conservative Party voted unanimously against Bill C-283, and it was not supported by a sufficient number of opposition members. |Among the opponents to Bill C283: 4 MPs with personal or family interests in large fast food chains and 7 MPs who raise livestock.

- The CSPI called on provincial and municipal governments to adopt menu labelling laws as long as they were not adopted at the federal level.

| |Centre for Science in the Public Interest – Canada, 2008b

(cont’d)

(cont’d) |Grey literature; source = CSPI Canada website; letter of notification sent to Health Canada; authors = lobbyists |- Companies have developed their own logos: Smart Spot by Pepsico, Sensible Solutions by Kraft, Blue Menu by President’s Choice and Goodness Corner by General Mills.

- The companies’ logos are based on criteria developed in-house.

- Hannaford’s star program: stars are placed on the shelf label next to the price. Hannaford applies its program to all of its 25,500 products; 28% of them are awarded stars (criteria are stricter than Health Check). |- Many American consumers use the nutrition labels. Its use is associated with healthier eating.

| |Front-of-package logos would be a strong incentive to reformulate products. |- Mandatory nutrition labelling on packaged foods (2003 regulation) should save $5 billion CAN over 20 years through healthier food choices that reduce the risk of premature death or disability due to cardiovascular disease, cancers and diabetes (source = The Canada Gazette).

- Cost (non recurring) of label changes: $263 million CAN over 3 years (source = The Canada Gazette).

- By comparison, revenues on food sales over 3 years = $120 billion CAN. |- In 9 years, Canada’s Health Check program has included 1,500 products, which is impressive for an NGO, but this represents only 3% of the products available in the market.

- 130 participants in Health Canada’s consultation on nutritional claims in Toronto, including health and food industry experts.

- The criteria for the companies’ logos vary and are sometimes weak.

- The Health Check criteria (Heart and Stroke Foundation of Canada) are stricter than those of the industry, but some products carrying the logo are only healthy in comparison to other products in the same category.

- The use of simple symbols would solve the problem of the lack of space on the menus. |- CSPI Canada recommends that Health Canada: establish nutrition criteria for a symbol that would be mandatory on packaged foods and in chain restaurants; to conduct research and ask for comments from scientists, health and communications experts, the industry and consumers on the best possible system; determine whether the current logos promoted by industry and third parties do not correspond with the proposed official system and should be eliminated. Believes that the federal government is responsible for labelling (even if an NGO can provide nutrition education).

- The Standing Committee on Health of the House of Commons recommends mandatory and standardized front-of-package logos.

- Health Canada’s Advisory on Healthy Children and Youth makes the same recommendation. Also recommends posting nutrition info on food packaging and menu boards in chain restaurants.

- In November 2006, the B.C. Select Standing Committee on Health recommended that the B.C. government develop a front-of-package logo system (of the traffic light and/or caloric value type).

- In its opposition to Bill C-283, the restaurant industry argued that there isn’t enough space on menus.

- The U.S. Institute of Medicine believes that the industry’s different logos may create confusion, particularly among young consumers, so a classification and symbol system needs to be developed (by the FDA) and regulated, one that would apply to the entire industry and could be understood by children. | |Friedman, 2008

(cont’d) |Grey literature; source = Yale Rudd Center website; non-systematic review of the literature; authors = academics |Policies on labelling in restaurants have been adopted in several large American cities and have been proposed at the federal level and in many states. |- 75% of American adults read the labels on packaged foods.

- 48% of American adults say that reading labels has changed their purchasing habits.

- Americans spend half of their food budget on food prepared outside the home and consume a third of their calories in restaurants.

- Many chain restaurants already publish nutrition info, but on their websites or other supports not directly accessible when ordering (posters, brochures, tray liners, etc.). Only a small minority of clients consult this information, since they usually don’t notice it.

- There is an association between the fact that restaurant customers receive info and the probability that they will select lower-calorie dishes.

- Impact study conducted in Los Angeles County: if 10% of chain restaurant customers reduced the content of their meals by 100 calories, this would prevent 40% of the 6.75 million pounds gained each year in the county. |- Following mandatory listing of trans fat on the labels of packaged foods, many products were reformulated.

- [Theory] restaurants risk losing clients. |- Fast food is consumed disproportionately by low-income people, a group at a higher risk for obesity. |- Analysis of nutrition content is a one-time cost; the current policies proposed do not apply to small restaurants, which are unable to assume these costs. |- Many chains already analyze the nutrition content of their menus.

- Feasibility problems and counter-arguments:

- Special offers are constantly changing, and customers often make special orders (answer: labelling only applies to dishes offered on a permanent basis).

- The reading of this info slows down service in fast food restaurants (answer: If the info is provided in large letters, clients can read it while they wait in line).

- The menus should be changed too often because of special offers (answer: this is already the case for marketing reasons and for price changes). |- According to 5 national surveys and 4 local surveys, from 62% to 84% of consumers support a policy for labelling in restaurants.

- One argument against labelling laws (opponents not identified): they are an infringement on our freedom to eat what we want (answer: it is only info, it creates no obligation).

- According to opponents, labelling will not result in fewer calories consumed.

- Many associations and public institutions recommend caloric labelling in restaurants (see list in the document). | |Government Accountabi-lity Office, 2008

(cont’d)

(cont’d)

(cont’d) |Grey literature; source = CSA; audit report; authors = Government Accountability Office (independent agency that works for Congress and audits federal government activities) |- The UK: traffic lights on the front of packages (voluntary)

- The Netherlands: voluntary “healthy choice” logo developed by industry and supported by the Department of Health

- Sweden: voluntary “keyhole” logo

- U.S.: consumer health associations and the industry have developed many logos (American Heart Association, Whole Grains Council, Kraft, Hannaford) |- Other studies: consumers find the daily intake values useful, but find the labels difficult to understand. Some studies suggest that many read the labels when they buy something the first time, but other studies suggest that they don’t interpret the info

- The UK: sales data suggests that the traffic lights influence purchases.

- The U.S.: many brands and supermarkets report that logos increase sales (Sensible Solutions, Hannaford’s stars program) |- The UK: the industry develops new products or reformulates products to be able to have green or amber lights.

- Sweden: idem. | | |[Implication for imple- mentation]

- The FDA inspects industry compliance with labelling obligations and may sign contracts with the states to have them make inspections. The FDA may take samples and have them analyzed.

- Over the last few years the number of food companies has grown but the number of inspections and corrective actions has declined or not changed.

- The FDA deplores its lack of resources (financial and human) and power (for example, the power to have offending firms pay the cost of re-inspection and to certify third parties for some inspections) to apply the labelling regulations and conduct research on consumer perceptions.

- The U.S.: Grocery Manufacturers / Food Products Association opposes mandatory front-of-package labelling (wants a voluntary system) as well as having nutrition labelling for the entire content of packages (vs. smaller portions)

- In April 2008, Health Canada indicated that it would consult with stakeholders and propose research on front-of-package labels. |- According to the Institute of Medicine, there is little evidence that labelling has a significant impact on food and purchases.

- The Institute of Medicine has recommended more research on labelling (and, in particular, logos), by the FDA and others.

- The Institute of Medicine has recommended that industry, government, researchers, public health authorities and consumer associations work together to develop and implement a front-of-package labelling system across the industry.

- The Institute of Medicine has recommended that food consumed in one sitting have a label indicating the number of calories for the entire package.

- According to research financed by the industry: consumers prefer nutrition labelling for the whole package (rather than by portion).

- The American Medical Association and the American Heart Association recommend a mandatory front-of-package labelling system.

- In 2006 the Center for Science in the Public Interest asked the FDA to develop such a system.

- The U.S.: The Grocery Manufacturers / Food Products Association is opposed to a mandatory front-of-package labelling system (wants a voluntary system); it believes that many companies use logos and have reformulated their products. The Association is also opposed to labels with nutrition info for the entire package contents (vs. smaller portions), arguing that this will encourage consumers to consume the entire product.

- Canada: in 2007, the Health Committee of the House of Commons recommended that Health Canada introduce a mandatory, standardized and simple front-of-package labelling system. Recommendation supported by the Chronic Disease Prevention Alliance of Canada.

- The European Commission has proposed a law that would ask for front-of-package labelling.

- The European Heart Network (an alliance of health organizations) and the European Consumers’ Organization support mandatory front-of-package labelling.

- The UK: the National Heart Forum supports the use of traffic lights. | |Rudd Center for Food Policy and Obesity, 2008

(cont’d) |Grey literature; source = Yale Rudd website; fact sheet (case study); authors = academics |- Subway already uses labelling as a sign of the attention it pays to customers.

- In January 2007, a California senator introduced a bill for labelling in the state’s chain restaurants (calories on menu boards, calories + saturated fats, trans fats, sugar and sodium on menus). Approved by the Assembly and Senate but vetoed by the Governor in October 2007. The bill was re-introduced in 2008. | |2nd objective of the bill (after informing consumers): to encourage product reformulations | |- Study of tax impact: According to the California Senate, the restaurant labelling act would cost the state nothing; according to the Senate, it would cost $100,000. |- The bill (California) was drafted based on the model provided by the CSPI and reviewed by lawyers to prevent lawsuits.

- The bill was approved by the Assembly and Senate but vetoed by the Governor in October 2007. |- Bill proposed to a senator by the California Center for Public Health Advocacy, the American Heart Association, the American Cancer Society and the California Optometric Association.

- Bill moved quickly through the two houses due to a positive political climate: obesity was on the agenda, senators were increasingly informed on the subject, and the senator behind the bill was truly committed.

- Survey of the public in March 2007: 84% in favour of labelling; 78% of Republicans and 89% of Democrats; percentages were similar across different education levels and income groups and different ethnicities.

- The media gave labelling considerable coverage, except some very conservative local papers.

- Greatest opponents: the California Restaurant Association, the Grocery Manufacturers Association, the State Chamber of Commerce, the California Independent Grocers Association.

- According to the NRA, the public doesn’t need labelling because people already know how to make healthier choices (false, according to testing of the subject).

- Governor’s arguments for using his veto: some restaurants incur costs and charges, but not others; some restaurants already provide info, particularly on the Internet (and called on them to continue this work); labelling is not an effective way to educate the public. | |Centre for Science in the Public Interest – Canada, 2007

(cont’d)

(cont’d)

|Grey literature; source = CSPI Canada website; testimony to the House of Commons’ Standing Committee on Health; authors = lobbyists |- Some companies have developed their own logos: Smart Spot by PepsiCo, Sensible Solutions by Kraft, Blue Menu by President’s Choice, Goodness Corner by General Mills.

- Companies’ logos are based on their own criteria.

- Voluntary traffic lights system in the UK was established in March 2006; some large companies followed suit while others have their own systems.

- Lock symbol in Sweden: identifies the healthier choices in a given category; use is voluntary.

- The Hannaford Star Program: stars placed on shelf label next to price. Hannaford applies its program to all its products; three quarters of them do not obtain stars.

|- Many American consumers use the nutrition label. Its use is associated with a healthier diet.

- The nutrition label could be more effective through simple presentation of the info on the front of the package, which requires less calculating and time to interpret.

- Logos are even more important in chain restaurants, where the information is often available on the website or not at all.

- The companies’ logos are based on their own criteria, which vary (risk of confusing consumers).

- In Sainsbury’s supermarkets, use of traffic lights had a positive impact (in terms of healthy food) on sales.

- In Sweden and Norway, sales of products carrying the logo rose 15% to 20%. |Front-of-package labelling is a stronger incentive to reformulate products. | |- According to Health Canada, mandatory nutrition labelling on packaged foods (2003 regulation) should result in $5 billion in savings over 20 years.

- The cost (non-recurring) of label changes (figures from Agriculture and Agri-Food Canada): $263 million CAN over 3 years.

- By comparison, revenues from food sales over 3 years = $120 billion CAN. |- Companies’ logos are based on their own criteria, which are sometimes weak.

- Even some Health Check criteria (Heart and Stroke Foundation) are weak.

- The restaurant industry was opposed to Bill C-283.

- Use of simple symbols would solve the problem of lack of space on menus. |- The restaurant industry’s argument against Bill C-283 was that space was lacking on menus.

- The U.S. Institute of Medicine believes that the industry’s different logos may create confusion, particularly among young consumers, so a classification and symbol system needs to be developed (by the FDA) and regulated that would apply to the entire industry and could be understood by children.

- In November 2006, the B.C. Select Standing Committee on Health recommended that the B.C. government develop a front-of-package logo system (of the traffic light and/or caloric value type).

. CSPI Canada recommended that Health Canada conduct research and ask for comments from scientists, health and communications experts, the industry and consumers on the traffic light system. It should then determine whether the selected system (with a symbol and nutrition criteria) should be voluntary or mandatory (CSPI favours the mandatory approach, since otherwise companies will not use the logos on marginally healthy foods). | |Health Canada, 2007

(cont’d)

(cont’d)

|Grey literature; source = Health Canada website; document for discussion/ consultation; authors = Health Canada |- “With respect to foods, Health Canada is responsible for the development of policies, regulations and standards that relate to health and safety”; “The Canadian Food Inspection Agency (CFIA) is responsible for their related compliance and enforcement,” as well as their application.

- Definition of “a health claim for food”: “any representation in labelling and advertising that states, suggests or implies that a relationship exists between the consumption of foods or food constituents and health.”

- These relationships “may be stated explicitly with words or represented implicitly (implied) through slogans, graphics, logos, symbols.”

- Health claims can be specific or general.

- No rule specifies or governs the use of general health claims (of the “healthy choice” type), except that under the Food and Drug Act, claims should not be false, misleading or deceptive. There is no standardized nutrition criterion for making this type of claim on a package. But the CFIA and Health Canada have developed guidelines (in 2003) to foster proper use.

- The Food and Drug Regulations do not directly address implicit claims.

- Various simplified messages on the front of packages are based on different criteria (some are based on a nutrient, others on several nutrients; still others are more complex, since they take into account dietary guidance). Some criteria are set by manufacturers, others by third parties.

- “Health claims are distinct from nutrient content claims, which are representations, expressed or implied, that characterize the energy value of the food or the amount of a nutrient contained in the food. Provisions for nutrient content claims in Canada were updated as part of the regulations on nutrition labelling and claims introduced in the Food and Drug Regulations (FDR) in December 2002 and are outside the scope of this document and initiative.”

- CONSULTATION DOCUMENT: Several options proposed to reduce confusion around the proliferation of logos: consumer education, industry guidelines on the conditions formulations, and improved nutrition labelling regulations, if required.

- In 2007, the FDA announced initiatives concerning consumer reactions to claims and symbols/logos. Public hearings on the use of symbols in September 2007. |- Confusion due to the proliferation of implicit health claims.

- Problem of implicit claims [particularly theory]: difficult for consumers to determine what criteria have been used. When the simplification is exaggerated, the info may become misleading. Too much importance is given to positive aspects, which may divert attention from other, less healthy characteristics. Some studies [non-Canadian] have shown that implicit claims may lead to reduced use of the nutrition facts table. Because of the different criteria used, products cannot be compared. There is a risk that consumers turn away from products with no claims that may nevertheless be healthy and less expensive.

- Canadian consumer surveys indicate significant reported use of food labels [sic: not only nutrition labels] to select healthier foods, but some (non-Canadian) studies suggest less actual use.

- Health Canada believes that policy development should be backed by research on consumers’ interpretations of claims (including symbols), their relationship with use of the nutrition facts table, the interpretation when the food also has negative characteristics, and attitudes toward products with no claims.

|- Health Canada’s proposals include monitoring the long-term impact of health claims on the nature of food supply in Canada and investments in research by Canadian companies for food innovation. |- Health Canada wants to collect info on how health claims are understood and used by different social groups (including purchase behaviour). | |- Consultation on claims:

- Health Canada seeks the participation of consumers, health professionals, the industry and organizations interested in specific illnesses.

- The use of symbols or slogans (vs. explicit claims made in longer statements) helps solve the problem of too little space on labels (including the challenge of bilingual labels on small packages).

- “With respect to foods, Health Canada is responsible for the development of policies, regulations and standards that relate to health and safety”; the Canadian Food Inspection Agency is “responsible for their related compliance and enforcement,” as well as their application. |- If the claims are not regulated or standardized, consumers may become sceptical about them.

- But consumers appear to appreciate simplified representations of nutrition profiles.

- Health Canada recommends research on consumers’ assessments of claims (including symbols), the sources of scepticism.

- In March 2007, the Standing Committee on Health of the House of Commons recognized the usefulness of having clear and simple front-of-package labels based on standard criteria, and recommended that the federal government act quickly.

- Health Canada recommends research on consumers’ expectations about the roles, responsibilities and accountability obligations of government, industry and other stakeholders concerning front-of-package claims.

- In this consultation document, Health Canada recommends (for discussion purposes) several options for reducing the confusion around the proliferation of front-of-package claims: consumer education, industry guidelines on the conditions formulations, and improved nutrition labelling regulations, if required.

| |Heart and Stroke Foundation, 2007 |Grey literature; source = Heart and Stroke Foundation of Canada website; position statement; author = NGO | | | | | | |H&S Foundation recommends that the industry improve the availability of nutrition info in fast-food restaurants, on menu boards and on menus. | |Jauregui, 2007

(cont’d)

(cont’d) |Grey literature (thesis – University of Florida); source = PsycINFO; secondary analyses (data from two chronological series, 1984-2003, 30,414 and 13,150 observations); author = academic |- In 1989: a committee of the National Academy of Sciences, under contract with the FDA and the USDA, examines how to improve labels.

- 1990: the NLEA

- 1991: the FDA makes over 20 proposals on how to apply the NLEA. The FDA issues a rule establishing a voluntary labelling program for fruits, vegetables and fish. The USDA’s Food Safety and Inspection Services make proposals on labels for meat and poultry.

- 1992: Voluntary labelling program for fruits, vegetables and fish in force.

- 1994: The NLEA is enacted. |- [Logical model] in order for the labels to be useful, consumers must be aware that the info exists, pay attention to it and understand it. The availability of substitute products influences changes after info becomes available (if there are no substitutes, the cost of changing is too high).

- Other studies: new labels (introduced in 1994) improved understanding and had a modest impact on purchases, but (soon after 1994) only the most motivated and sceptical consumers used them, and the percentage who never used them did not change.

- [Note: The study dealt with label use in general – not only nutrition info, even if, implicitly, it plays a large role.]

- This study: 59% use labels for purchase decisions; 55% use them to identify harmful ingredients. Label reading peaked at the start of the 1990s (application of the NLEA), then declined. |- [Theory] Even if there are few informed consumers, competition over this minority may lead companies to make changes that are beneficial to all consumers. |- [Theory] Even the poor and less educated, who are less able to use nutrition info, must pay for it.

- Other studies: those who make more use of labels = women, better educated, older, households with children, married people, higher income, Caucasians, people on special diets, non-smokers, people engaged in regular physical activity, people who consume less fat, people who believe that they have more knowledge on nutrition and believe there is a link between food and health.

- This study: education has no effect (author notes that this is unusual). Variables with greatest impact on the probability of reading labels (in descending order): calorie awareness, nutrition knowledge, warning from doctor on diet, consumption of fried chicken, attention to additives, attention to cholesterol, consumption of hot dogs, attention to fat, trust in major brands as a indication of better quality, age of the woman heading the household, women on diets, avoiding foreign food. |- [Theory]: Direct costs of labelling = generating the info, testing whether claims are true, printing labels, applying/ enforcing the labelling rules. Indirect costs due to loss of flexibility in production (due to need to comply with what is indicated on the label, even if there are short-term changes in input availability or costs). Costs are relatively higher for small companies.

- [Theory] Even the poor and less educated, who are less able to use nutrition info, must pay for it. | |70% want easy-to-understand labels.

- Spanish study: the groups that see mandatory labels as more beneficial are label users, are older and are more educated. | |Centre for Science in the Public Interest – Canada, 2006a

(cont’d) |Grey literature; source = CSPI Canada website; fact sheet; authors = lobbyists |Content of Bill C-283: in the large chain restaurants (others are exempt), have calorie counts of standard items posted on menu boards, as well as sodium and saturated and trans fats listed on the menus of restaurants with table service. Also, nutrition labelling on meat, poultry and fresh seafood products. | | | |- The cost of C-283 will be minimal because: almost all large chain restaurants already calculate or can easily calculate nutrition info for their standard items; the menus can be changed sometime during the next three years when menus are updated.

- The cost of C-283 will be minimal because the nutritional profiles of meat, poultry and seafood products are already provided on certain websites of the meat industry; the labels can be changed sometime during the next three years when label inventories are replenished.

- According to CSPI estimates, the cost of nutrition analysis is 1/10 of 1% of retail sales.

- According to estimates, the nutrition facts table would save $5 billion over the next 20 years (due to reduced health costs and higher productivity); this is 20 times the cost of changing the labels. |- The voluntary labelling program of the Canadian Restaurant and Foodservice Association does not encourage placing info on menus, rather it encourages placing it where it is less visible (websites, brochures, tray liners). |- Labelling in restaurants and/or on fresh products is supported by health and consumer defence groups representing over 2 million Canadians [see list in document].

- Labelling in restaurants is recommended in reports by the CIHI’s Canadian Population Health Initiative, Ontario’s Chief Medical Officer of Health (in 2004), the Heart and Stroke Foundation of Canada, and the Institute of Medicine in the U.S.

- Labelling on products other than packaged products recommended in reports by Ontario’s Chief Medical Officer of Health (in 2004) and the B.C. Healthy Living Alliance

- Associations in the food industry believe that there is not enough space on labels and menus. | |Centre for Science in the Public Interest – Canada, 2006b |Grey literature; source = CSPI Canada website; press release; author = lobbyist |Bill C-283 was defeated in a vote in Parliament. | | | | |- The restaurant industry mounted a lobbying campaign against C-283.

- All Conservative members voted against the Bill, and it was not supported by enough members of the opposition. |- [According to the CSPI], federal government scientists and economists recognize that mandatory point-of-sale info can efficiently reduce the burden of diet-related disease if it is visible prior to purchase.

- Some members of Parliament opposing the law had family links to the food industry. | |Center for Science in the Public Interest, n.d.

(cont’d) |Grey literature; source = CSPI website; fact sheet; authors = lobbyists | |[Everything suggests the theoretical effectiveness of different ways of providing info.]

- Available at the decision point: menus and menu boards, food labels (on displays), posters and brochures (occasionally), alternative menus or appendices including nutrition info, printed documents on tables (occasionally), electronic kiosks (occasionally).

- Accessible and easily visible: menus and menu boards, food labels (on displays).

- Allows food comparisons: menus and menu boards, food labels (on displays), posters and brochures, alternative menus or appendices including nutrition info, printed documents on tables, electronic kiosks, websites, tray liners.

- Presented with the price (see arbitration in the purchase decision): menus and menu boards, food labels (on displays).

- Specific notes: websites: consumers must see them before ordering. Kiosks: only one person at a time can use it. Represents a problem for persons with little computer experience. | | | |[Theoretical] Ways of providing info that do not impede the ordering process: menus and menu boards, websites, tray liners, food packaging, cashier receipts | | |

List of documents referenced in this table

Canadian Food Inspection Agency (2009). Food labelling. Consulted on August 3, 2009:

Center for Science in the Public Interest (2008a). Myth vs. reality: Nutrition labeling at fast-food and other chain restaurants. Retrieved from:

Center for Science in the Public Interest (2008b). Summary of findings: Influence of nutrition information provision. Retrieved from:

Center for Science in the Public Interest (2008c). Nutrition labeling at fast-food and other chain restaurants. Retrieved from:

Center for Science in the Public Interest (2008d). Product reformulation: A beneficial outcome of menu labeling. Retrieved from:

Center for Science in the Public Interest (2008e). Summary of polls on nutrition labeling in restaurants. Retrieved from:

Center for Science in the Public Interest (2009). New York City. Rave reviews for menu labeling. Retrieved from:

Center for Science in the Public Interest (n.d.). Approaches to providing nutrition information in restaurants. Retrieved from:

Centre for Science in the Public Interest - Canada (2006a). 10 reasons to support Tom Wappel's Bill C-283 on September 18, 2006. Retrieved from:

Centre for Science in the Public Interest - Canada (2006b). Statement on Parliament's vote denying consumers of nutrition info. Retrieved from:

Centre for Science in the Public Interest - Canada (2007). Testimony of Bill Jeffery, national coordinator of the Centre for Science in the Public Interest, before the House of Commons Standing Committee on Health on measures to reduce rates of childhood obesity. Ottawa: CSPI. Retrieved from:

Centre for Science in the Public Interest - Canada (2008a). Most "volunteers" in restaurant industry's nutrition information program stepped back two paces. Retrieved from:

Centre for Science in the Public Interest - Canada (2008b). Simplifying nutrition labels and preventing deceptive claims (Response to Health Canada's "Managing health claims for foods in Canada: Towards a modernized framework"). Retrieved from:

Centre for Science in the Public Interest - Canada (2009a). "Healthy decisions for healthy eating Act" moves forward in Ontario legislature! Retrieved from:

Centre for Science in the Public Interest - Canada (2009b). Ontario Medical Association's menu labelling advice. Just what the doctor ordered. Retrieved from:

Coalition québécoise sur la problématique du poids (September 30, 2009a). Des logos "santé" qui sèment la confusion: Une réglementation s'impose. Press release. Retrieved from:

Coalition québécoise sur la problématique du poids (2009b). Des logos "santé" qui sèment la confusion: Une réglementation s'impose. Presentation made during a press conference held on September 30, 2009 [PowerPoint slides]. Retrieved from: .

Engelhard, C. L., Garson, A. & Dorn, S. (2009). Reducing obesity: Policy strategies from the tobacco wars. Washington DC: The Urban Institute. Retrieved from:

Friedman, R. (2008). Menu labeling in chain restaurants. Opportunities for public policy. New Haven: Rudd Center for Food Policy and Obesity, Yale University. Retrieved from:

Government Accountability Office (2008). Food labeling: FDA needs to better leverage resources, improve oversight, and effectively use available data to help consumers select healthy foods. Washington DC: United States Government Accountability Office. Retrieved from:

Health Canada (2007). Managing health claims for foods in Canada: Towards a modernized framework. Ottawa. Retrieved from:

Health Canada (2009a). Nutrition labelling. Consulted on August 2, 2009:

Health Canada (2009b). Health Canada's action plan in response to stakeholder feedback from consultations on modernizing Canada's framework for health claims on food. Ottawa. Retrieved from:

Heart and Stroke Foundation (2007). Position statement. Overweight, obesity, and heart disease and stroke. Retrieved from:

Heart and Stroke Foundation (2009a). Heart&Stroke Health CheckTM... Helping you eat well. Consulted on August 4, 2009:

Heart and Stroke Foundation (2009b). Health Check TM. What consumers say. Consulted on August 4, 2009:

Jauregui, C. E. (2007). Consumers' use of food labels: An application of ordered probit models (doctoral thesis). University of Florida, Gainesville.

Kiesel, K. & Villas-Boas, S.B. (2009). Can information costs confuse consumer choice? Nutritional labels in a supermarket experiment (doctoral thesis). California State University, Sacramento. To be published as: Kiesel, K. & Villas-Boas, S.B. (in press). Can information costs affect consumer choice? Nutritional labels in a supermarket experiment. International Journal of Industrial Organization.

Kohn, D. (2009). Better health, not just better health care: A prescription for progress from the Big Cities Health Coalition. New York: Milbank Memorial Fund. Retrieved from:

Robert Wood Johnson Foundation (2009). Menu labeling: Does providing nutrition information at the point of purchase affect consumer behavior? Princeton: Robert Wood Johnson Foundation. Retrieved from:

Rudd Center for Food Policy and Obesity (2008). A case study of California's menu labeling legislation. New Haven: Rudd Center for Food Policy and Obesity, Yale University. Retrieved from:

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download