Consumer food environment: NEMS-R, developed from the conceptual model of Glanz et al.
Trained raters used standard NEMS-R protocols from the training manual (http://nems.nursing.upenn.edu/) to complete the evaluation.
Descriptive characteristics (means and standard deviations (SD) for NEMS-R total and sub-scores (normally distributed)) were calculated to assess restaurant consumer food environments.
According to the NEMS-R results, the consumer food environments of each restaurant sub-category significantly differed in mean NEMS-R scores (Table 2).
As an observational tool, NEMS-R was designed to evaluate selected attributes that are readily observed as well as relevant to obesity and other chronic diseases .
This leads us to question the utility of the NEMS-R assessment tool for characterizing consumer food environments, particularly restaurants.
Although most NEMS-R items were reported in the original study  to have high kappa values (greater than 0.80) for inter-rater reliability, some Facilitator items (such as nutrition information on menus, healthy item labelling, and signage encouraging healthful eating) and some Barriers items (such as menus discouraging special requests and signs encouraging overeating) had lower values (less than 0.60).