RESEARCH The Impact of Menu Labeling on Calories

RESEARCH The Impact of Menu Labeling on Calories

RESEARCH The Impact of Menu Labeling on Calories Selected or Consumed: A Systematic Review and Meta-Analysis Susan E. Sinclair, MSc Marcia Cooper, PhD, RD Elizabeth D. Mansfield, PhD, RD Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9). RESEARCH Key criteria used to select studies for a systematic review and meta-analysis that measured the impact of menu labeling on calories selected or consumed Inclusion Criteria Exclusion Criteria A controlled experimental or quasiexperimental design was used A control or comparison group was not included (eg, before and after designs) The effect of nutrition information on menus was tested Outcomes measured included calories selected or consumed Participants nutrition literacy or awareness of nutrition information was measured rather than actual behavior Setting was Canada, or any other country with a similar nutrition labeling environment Consumers intentions were measured rather than their actual food selection or consumption behavior Participants were generally healthy adolescents or adults Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9). RESEARCH Key quality appraisal, data extraction, and data synthesis methods for a systematic review and meta-analysis that measured the impact of menu labeling on calories selected or consumed Quality appraisal The Scottish Intercollegiate Guidelines Network (SIGN) checklists for cohort studies and for controlled trials32 were used to assess the quality of the quasi-experimental and experimental studies, respectively. Sensitivity analysis was used to take into account the influence of studies judged to be of lower quality.

Data extraction and synthesis Authors were contacted by email to obtain any information needed for data extraction. To facilitate comparison between studies, proportional differences in calories consumed or selected were calculated. Subgroup analyses were planned to test the influence of: gender, age, socioeconomic status, setting, cost of food, type of label. Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9). RESEARCH Table 3. Summary of characteristics and results of five quasi-experimental studies that measured the association between the introduction of menu labeling and average calories purchased per transaction Lead author, year Elbel, 201149 Study location (test vs. comparison) Sample size (n) Calories purchased per Calories purchased per transaction transaction at the test location at the comparison location at the before vs. after mandatory menu same time points as the test location labeling was introduced where mandatory menu labeling (Calories) was introduced (Calories) Test difference vs. comparison difference in calories purchased per transaction (Calories) United States (New York City vs. Newark) Males: 82 747 vs. 814 878 vs. 795 +67a vs. -83a Females: 102 712 vs. 716

560 vs. 730 +4 vs. +170 Finkelstein, 201148 United States (King County, Washington vs. adjacent counties) >11,000 1211 vs. 1217 1391 vs. 1392 +5.7 vs. +0.9 Tandon, 201147 United States (Seattle/King county, Washington vs. San Diego county) 128 823 vs. 720 895 vs. 789 -103a vs. -106a Bollinger, 201046 United States (New York City vs. Boston and Philadelphia) 118,480 247 vs. 232.6a NRb vs. NRb -14.4 vs. NRb** Elbel, 200919 United States (New York City vs. Newark) 1156 825 vs. 846

823 vs. 826 +21a vs. +3a Value was imputed from other values provided in the study report (actual value was not reported). To protect the confidentiality of the data, the authors did not provide these data in the study report. **The regression coefficient estimate of the effect of mandatory calorie posting was statistically significant (p<0.01) a b Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9). RESEARCH Table 4. Summary of characteristics and results of two quasi-experimental studies that measured the association between the introduction of menu labeling and the percent frequency with which targeted food items were purchased Lead author, year Comparison Webb, 201145 Levin, 199644 a Test cafeteria vs. control cafeteria Test cafeteria vs. control cafeteria Sample size (n) 600 406 Frequency with which targeted items were purchased in the test cafeteria before vs. after Food items targeted with menu labeling was menu labeling introduced (%) Frequency with which targeted items were purchased in the comparison cafeteria at the same time points as in the test cafeteria where menu labeling was introduced (%)

Test difference vs. control difference in frequency with which targeted items were purchased (%) (p-value) Entrees 68.6 vs. 68.6a 79.2 vs. 79.3a +0.028 vs. +0.05 (p-value not significant actual value not reported) Sides 78.4 vs. 83.2a 69.2 vs. 64.4a +4.8 vs. -4.8 (p=0.0007) Snacks 40.3 vs. 41.6a 27.8 vs. 19.7a +1.3 vs. -8.1 (p=0.006) Entrees 4.3 vs. 11.9 (p<0.001) 5.4 vs. 5.1 (p=0.78) +7.6a vs. -0.3a (p-value not reported) Value was imputed from other values provided in the study report (actual value was not reported) Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9). RESEARCH Table 8, Part 1. Sensitivity and subgroup analysis results for the effect of menu labeling on calories selected Description of the sensitivity/subgroup analysis

Number of studies included in the sensitivity or subgroup analysis Number of Number of relevant menu participants in the Mean difference in calories labeling menu label (test) between the menu label group and comparisons group and in the no in no menu label group included in the menu label (control) (Calories) (95% confidence interval studies group conditions (CI); p-value) Overall (all conditions) 7 18 test: 1267 control: 799 -42.54 (95% CI: -81.90, -3.19; p=0.03) Conditions from studies rated as higher quality 3 6 test: 541 control: 409 -63.89 (95% CI: -166.86, 39.09; p=0.22) Calorie content only menu label conditions 6 8 test: 776 control: 583 -30.84

(95% CI: -95.85, 34.18; p=0.35) Calorie content and contextual or interpretive information menu label conditions 4 4 test: 351 control: 148 -67.39 (95% CI: -116.99, -17.79; p=0.008) Conditions from studies that took place in a natural setting 3 10 test: 580 control: 248 -52.57 (95% CI: -92.16, -12.98; p=0.009) Conditions from studies where participants believed they needed to pay for their meal 5 14 test: 1027 control: 683 -34.59 (95% CI: -73.11, 3.94; p=0.08) Results reported for women separately 2 3 test: 128 control: 110 -145.05 (95% CI: -235.70, -54.39; p=0.002) Results reported for men separately 1

1 test: 50 control: 61 +92.22 (95% CI: -35.28, 219.72; p=0.13) Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9). RESEARCH Table 8, Part 2. Sensitivity and subgroup analysis results for the effect of menu labeling on calories consumed Description of the sensitivity/subgroup analysis Number of studies included in the sensitivity or subgroup analysis Number of Number of participants in the relevant menu menu label (test) labeling group and in the comparisons no menu label included in the (control) group studies conditions Overall (all conditions) 5 22 test: 855 control: 544 -41.22 (95% CI: -78.98, -3.46; p=0.03) Conditions from studies rated as higher quality 3 6 test: 541 control: 409

-45.18 (95% CI: -129.22, 38.86; p=0.29) Calorie content only menu label conditions 4 8 test: 545 control: 383 -12.73 (95% CI: -62.29, 36.82; p=0.61) Calorie content and contextual or interpretive information menu label conditions 4 10 test: 310 control: 161 -80.67 (95% CI: -138.99, -22.36; p=0.007) Results reported for women separately 3 10 test: 207 control: 97 -72.38 (95% CI: -130.26, -14.50; p=0.01) Results reported for men separately 2 8 test: 147 control: 60 -21.82 (95% CI: -133.01, 89.37; p=0.70) Mean difference in calories between the menu label group and in no menu label group (Calories) (95% confidence

interval (CI); p-value) Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9). RESEARCH Figure 8. Forest plot showing pooled weighted mean difference with 95% confidence interval (CI) for 8 informative conditions or subgroups and 16 contextual or interpretive conditions or subgroups for which the calories consumed effect was estimated For each condition or subgroup, the shaded square represents the point estimate of the menu label effect. The horizontal lines join the lower and upper limits of the 95% CI of these effects. Arrows at the end of the line indicate that the CI extends beyond the scale shown at the bottom of the plot. The size of the shaded square reflects the relative weight of the study in the meta-analysis. The diamond at the bottom of each plot represents the pooled weighted mean difference with the 95% CI. a Standard deviation; binverse variance; cconfidence interval; dcalorie content menu label; erestrained eating participants subgroup; ffemale participant subgroup; gmale participant subgroup; hunrestrained eating participants subgroup; icalorie content and value pricing label; jcalorie content and recommended daily caloric intake for an average adult label; k calorie content and exercise equivalent label; lNutrition Facts table label; mlean participants subgroup; noverweight or obese participant subgroup; otraffic light label. Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9). RESEARCH Figure 9 Forest plot showing pooled weighted mean difference with 95% confidence interval (CI) for 8 informative conditions or subgroups and 10 contextual or interpretive conditions or subgroups for which the calories selected effect was estimated For each condition or subgroup, the shaded square represents the point estimate of the menu label effect. The horizontal lines join the lower and upper limits of the 95% CI of these effects. Arrows at the end of the line indicate that the CI extends beyond the scale shown at the bottom of the plot. The size of the shaded square reflects the relative weight of the study in the meta-analysis. The diamond at the bottom of each plot represents the pooled weighted mean difference with the 95% CI. a Standard deviation; binverse variance; cconfidence interval; dcalorie content menu label; efemale participant subgroup; fmale participant subgroup; gcalorie content and value pricing label; hcalorie content and traffic light symbol label; icalorie content and exercise equivalent label; jcalorie content and recommended daily caloric intake for an average adult label; k low-fat label on the beef item; llow-fat and preparation details label on the beef item; mlow-fat label on the haddock item; nlow-fat and preparation details label on the haddock item; o low-fat label on the pasta item; plow-fat and preparation details label on the pasta item. Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9). RESEARCH Conclusions from a systematic review and meta-analysis that measured the impact of menu labeling on calories selected or consumed The findings of this meta-analysis support menu labeling approaches that include contextual or interpretive nutrition information along with calories to help consumers select and consume fewer calories when eating in restaurants and other foodservice establishments. The labeling of menus with calories alone does not have a significant impact upon consumers selection or consumption of calories. The best approach for menu-based nutrition information, particularly, for those consumers who may be limited in their food and health literacy skills, merits further exploration. Sinclair SE , Cooper M, Mansfield ED. J Acad Nutr Diet. 2014;114(9).

Recently Viewed Presentations

  • Wireless M-Bus - Royal Holloway

    Wireless M-Bus - Royal Holloway

    Wireless M-Bus Sniffer Protocol sniffers display wireless M-Bus data record contents provided you know the key. The standard suggests "at least 8 bytes of the key shall be different for each meter"
  • Project Stakeholder Management

    Project Stakeholder Management

    Basic courtesies. Operating agreements. Problem Solving and decision making. Accountability. Conflict resolution. Leader's role. Define the meaning of dignity and respect for the team. Agree to hold each other accountable. 2016 Version
  • Diapositive 1 - Unité d&#x27;Enseignement Robert Desnos, CHU-Rennes

    Diapositive 1 - Unité d'Enseignement Robert Desnos, CHU-Rennes

    SESIMBRA PORTUGAL. ChapterChapitre CapítuloCapítuloCapítol#1. ENG There was once a seal he knew to fly and went on holiday with her friends the seagulls. To France, Portugal and Spain carry a suitcase and her sunglasses. ... The weather-vane-birds warn them: a...
  • WHY ADVOCA CY MATTERS By: Kathy Bates and

    WHY ADVOCA CY MATTERS By: Kathy Bates and

    Is a columnist for the NH Challenge . Started her own business . ... We are committed to supporting people who experience disabilities by tackling big issues that keep them from living quality lives in their communities. ... Big Brother,...
  • Airport Lease Types: Examples

    Airport Lease Types: Examples

    George Bush Intercontinental Airport created a Consolidated Rental Car Facility for eight rental car operators. An LLC was established to govern all operations. Stakeholders: Houston Airport System, city of Houston, and rental car operators. Benefits: Rental income to the airport,...
  • UNSWide Timetabling An Overview for Academic Staff Objective

    UNSWide Timetabling An Overview for Academic Staff Objective

    UNSWide Timetabling An Overview for Academic Staff Objective To provide staff with an understanding of the background and context for University-wide timetabling Provide staff with an understanding of how the changes may impact them and their academic unit To provide...
  • How Much Garbage?

    How Much Garbage?

    When biomass decomposes, it helps recycle energy and nutrients back into the environment. When dead things decompose, many things are released. The most common is a bad smell. Decomposing biomass also releases thermal energy (heat) as it breaks down.
  • Clayfighter Pitch - Callum Rourke

    Clayfighter Pitch - Callum Rourke

    Bernie's Low attack will involve Bernie throwing a Liquorice projectile underarm like a bowling ball and it will slide along the floor and send the opponent on the floor. His High attack involves Bernie Bassett taking off his top hat...