**3.**Generalcharacteristicsofthestudiesincludedinthereview.











socialization:

 Tercile 1; S: T3: variable analysis group socialization:

 Tercile 3.

#### *3.5. Components of the Mediterranean Lifestyle Evaluated in Each Study*

Table 4 includes the MLS components evaluated in each study. To analyze these components, the guidelines published in the study by Diolintzi et al. [2] were followed. The table shows that no article fully evaluated the MLS since none of them included participation in food preparation. MD adherence and the practice of physical activity (PA) were evaluated in all of them. Four studies focused on these two components [10,18,24,33], five also included night-time sleep [19–21,23,28], and three included MD, PA, and socialization [18,31,32]. Another study included MD, PA, and the use of locally grown, seasonal products [34]. Only six articles evaluated all of the MLS components, with the exception of the participation in food preparation and the consumption of locally grown and seasonal products [5,11,14,15,35,36]. Moreover, only two articles assessed all but one of the components, food preparation [16,17].


**Table 4.** Mediterranean Lifestyle components evaluated in each study.

**Table 4.** *Cont.*


#### *3.6. Evaluation Strategies Used to Analyze MLS Components*

Table 5 was created with the purpose of organizing all of the extracted information and summarizing the different evaluation strategies used by the authors to analyze each MLS component.

It shows all of the MLS components together with the questionnaires or self-reported ad hoc questions that were used to evaluate each dimension, with the exception of six studies, which exclusively used a tool to analyze the MLS components globally. The MEDLIFE questionnaire, the MedCOVID-19 score, the Total Lifestyle Index (TLI), MEDiLIFE-index, and MEDI-Lifestyle index were used [12,16,23,28,34,36]. It should be noted that, in one of the studies, instead of applying the previously validated 28-item MEDLIFE questionnaire, Bowden et al. [16] used an initial pilot 32-item questionnaire that was created by the original authors, which included 4 questions that could not be validated in a second study [35]. In addition to all the ad hoc questions used for the assessment of each MLS component, both the MEDiLIFE-index and MEDI-Lifestyle indices do not include the assessment of MD adherence, but require a specific questionnaire for their evaluation, and therefore include the KIDMED and PREDIMED questionnaires, respectively. In this sense, the TLI index is made up of different questionnaires that assess each MLS dimension separately, such as the MedDiet Score questionnaire for MD adherence, the Athens Physical Activity Questionnaire (APAQ), the sleep scale of the Medical Outcomes Study (MOS), and the Sleep Index II.

In addition, seven articles used the MEDLIFE tool in combination with other questionnaires to evaluate certain dimensions of the MLS separately [5,11,13–15,17,35]. Hershey et al. [17] modified the original MEDLIFE questionnaire, making variations in a total of nine items.

Regarding the other studies, each MLS component was evaluated using different, previously validated, and specific questionnaires [10,18–22,24–27,31,32].


**Table 5.**Assessment tools used to analyze MLS components.







#### *3.7. Indices for the Assessment of MLS*

#### **MEDiLIFE-index** [28]

This index relies on a 3-point scoring system (0-1-2) and has a maximum final score of 8 points (the sum of all questionnaire components). The higher the score, the better the adherence to the MLS.

To evaluate MD adherence, the KIDMED questionnaire was applied. KIDMED ≥ 8 (high adherence) received 2 points; KIDMED 4 to 7 received 1 point; and KIDMED ≤3 received 0 points (weak adhesion). PA was measured as follows: 2 points for PA ≥ 60 min/day; 1 point for PA ≥ 30 and < 60 min/day; and 0 points for PA <30 min/day. For a sedentary lifestyle, 2 points were given for <1 h/d watching TV, videos, screens, etc.; 1 point for ≥1 and ≤2 h/d of sedentary activities; and 0 points for >2 h/d. For sleep, the American Academy of Sleep Medicine Guidelines were followed, taking into account the different age ranges (6–12 years 9/12 h; 13–18years 8/10 h): 2 points were given if the optimal duration was achieved; 1 point if the duration was longer; and 0 points if the duration was shorter. As observed, this questionnaire used short questions, and another questionnaire was used to measure MD adherence (the KIDMED questionnaire).

#### **MEDI-Lifestyle** [12]

This index consists of seven short ad hoc dichotomous questions. Scores range from 0 to 7, with a score of 7 representing the best degree of adherence, and 0 the poorest. The individual's weight was evaluated via BMI, receiving 1 point for a BMI <30 kg/m2 and 0 points for a BMI ≥ 30 kg/m2. Tobacco consumption was also included: 1 point if the person had not smoked in the last 6 months, 0 points if they smoked. The PREDIMED questionnaire evaluated the MD, receiving 1 point for high MD adherence (≥9) and 0 points for poor adherence (≤9). PA was also evaluated, with physically active individuals (≥16 h/week) receiving 1 point and physically inactive individuals (<16 h/week) receiving 0 points. Time watching TV received 1 point for <2 h/d and 0 points for ≥ 2 h/d. For sleep, 1 point was given for 7–8 h/d and 0 points for sleep <7 h or >8 h/d. Regarding naps, 1 point was given if a nap was taken, and 0 points were given if no nap was taken. In this case, two factors were included in the index that are not included in the MLS: weight and smoking.

#### **Total Lifestyle Index** (TLI) [23]

This index evaluates four dimensions of LS and includes a specific questionnaire for each: diet (MedDiet Score), physical activity (APAQ), sleep quality (MOS and Sleep Index II), and Instrumental Activities of Daily Living (IADL). The results obtained in each questionnaire were divided into quartiles. Values were assigned from 0, for the first quartile (worst score), to 1, 2, and 3 for the other quartiles (higher scores). The total TLI score ranged from 0 to 12. Higher values indicated a more beneficial LS. As in the MEDI-Lifestyle index, a factor that was not part of the MLS was also included: the Instrumental Activities of Daily Living.

#### **MedCOVID-19 Score** [34]

This questionnaire assesses the current intake of nine foods from the Mediterranean Diet and five MLS-related behaviors, in terms of decreased, maintained, or increased intake, making comparisons between 2019 and autumn 2020.

To estimate the dietary rating, the following scores were assigned:


consumption of pre-cooked foods: −1 point for changes in undesired behaviors, and 0 points if they remained the same.

Diet and behavior scores were equaled to obtain a total mark ranging from −14 to 14. Once the score had been calculated, the population was classified as follows: stable population (score = 0), population with an improved MLS (>1), and population with a worsened MLS (score < 0).

**MEDLIFE** questionnaire [5,35]

This questionnaire was designed specifically to evaluate the MLS, without using other supplementary questionnaires to analyze each dimension separately.

It is divided into three blocks. The first block consists of 15 items and measures the consumption of Mediterranean foods. The second block is composed of seven items and measures the habits of the Mediterranean Diet, including hydration. The third block includes six items and measures PA, rest, social habits, and conviviality. The range varies from 0 (low MLS adherence) to 28 (high MLS adherence).

To analyze the reliability of the questionnaire, Cohen's Kappa coefficient, the intraclass correlation coefficient (ICC), and the limit of agreement (LOA) were used. A comparison was performed using a 142-item questionnaire (full-Q) from which the 28 items constituting the MEDLIFE questionnaire were derived [5,35].

According to the authors, the MEDLIFE questionnaire is a valid instrument to measure MLS adherence in middle-aged adults and can be used for clinical and epidemiological studies in this population. Its generalizability and predictive validity have yet to be examined [35].

In a study by Sotos-Prieto et al. [35], 4 additional questions were included to evaluate food seasonality and moderation, resulting in a 32-item instrument. Despite this, they were not included in the final questionnaire since a comparison with other tools was not possible. Thus, their validity could not be evaluated, and they were excluded from this questionnaire. However, this 32-item questionnaire was used in a recent study by Bowden et al. [16].

The original MEDLIFE questionnaire was also modified in another study. The following specific changes were made [17].


For the modified MEDLIFE questionnaire [17], scores varied from 0 to 26, following the same criteria as the original questionnaire. Although it is indicated that the modified MEDLIFE questionnaire presents a total of 26 items, it only includes 25.

#### *3.8. Statistical Analysis Conducted to Create a Mediterranean Lifestyle Score in the Studies Included in the Review*

Table 6 shows the different methods that researchers have suggested so far to create a questionnaire or index for assessing the MLS as a global dimension. Thus, most of the studies have employed different statistical analyses to integrate punctuations from diet adherence or dietary intake, sleep quality, or physical activity in a global MLS score. Moreover, the methodologies used to examine the structure of these questionnaires or indices are related to PCA, KMO, and also reliability and validity analyses [5,28,34,35].



188

#### **4. Discussion**

This systematic review focused on 26 studies that addressed the different means of assessing the Mediterranean Lifestyle [5,10–28,31–36]. This is the first review of its kind that identifies and analyzes the strategies used in the scientific literature to examine this lifestyle.

MLS is characterized by an adherence to the MD, proper hydration, the use of locally grown and seasonal products, participation in culinary activities, physical activity, and socialization, as well as adequate rest, both at night and through daytime napping [2]. Although numerous studies have supposedly analyzed the MLS, many of them have failed to examine its components as a general construct of lifestyle [10,19–21,23–28].

All of the reviewed articles assessed the adherence to the MLS components of MD and the practice of PA. The other most frequently evaluated components include socialization, sleep, and napping. The least commonly evaluated MLS dimensions were hydration and the use of seasonal/locally grown products, which were included in the MEDLIFE questionnaire. Moreover, not all of the components making up the MLS were fully evaluated, since none of the studies assessed participation in culinary activities.

Only six articles exclusively used specific strategies to evaluate the MLS [12,16,23,28,34,36], generating the MEDiLIFE-index, MEDI-Lifestyle, Total Lifestyle Index (TLI), MedCOVID-19 Score, and MEDLIFE questionnaires.

Despite the existence of three indices and two specific questionnaires for the evaluation of the MLS, the MEDLIFE questionnaire appears to be the only tool having adequate psychometric properties. However, it has only been validated for an adult population and not for young or elderly populations. While the studies included in this review mainly used validated tools to assess different MLS components, they did so in an independent manner. Only four of the studies relied on global indices that were created through statistical processes to integrate the different evaluated dimensions and perform a global analysis of the MLS [12,23,28,34]. However, after carrying out this systematic review, we have yet to find a tool having these qualities. It has been suggested, therefore, that MEDLIFE is the best instrument for providing an assessment of adherence to the MLS. Until a new tool is developed, this questionnaire appears to be the most appropriate one for analyzing adherence, even though it does not include all the components of this lifestyle [2,5]. Once a new tool with appropriate psychometric properties has been created, it should be validated for the general and clinical population.

Data on the psychometric properties of MLS questionnaires are almost nonexistent. Measures of reliability and validity are quite scarce. Sotos-Prieto et al. [35] and Bonaccio et al. [34] have offered some evidence regarding psychometric properties. The former reported intraclass correlation coefficients and Kappa coefficients that demonstrate the reliability and validity of the MEDLIFE questionnaire, respectively. Bonaccio et al. [34] revealed adequate reliability for the MedCOVID-19 score with a Cronbach's alpha coefficient of 0.83. However, both internal consistency (via Cronbach's alpha coefficient) and test–retest reliability and equivalence reliability must be considered in order to reveal the instrument's level of accuracy with regard to the construction of an assessment [56]. Moreover, scientific papers state that convergent, concurrent, predictive, and construct validity are different gold standards to evaluate this psychometric aspect [56]. They provide information on the relationship between new and validated tools that share the same construct. Katsagoni et al. [28] show a principal component analysis and KMO values, in relation to the structure of the instruments. The Confirmatory Factor Analysis (CFA) is a powerful statistical tool for the development of measurement instruments [57]. CFA and its analytic version, the Exploratory Factor Analysis (EFA), play an essential role in measurement model validation in this regard [57].

Concerning the methods used to create questionnaires for evaluating MLS, four studies included in this systematic review have merely described how researchers have attempted to generate MLS assessment tools. One of them showed a PCA and KMO to determine the intern structure of the questionnaire [28]. Moreover, both Bonaccio et al. and

Sotos-Prieto et al. [5,34] also run psychometric analysis such as the reliability of the questionnaire they created. However, it seems extremely difficult to find studies which indicate the psychometric properties of the tools they have created to assess the MLS. As far as psychometrics is concerned, Muñiz and Fonseca-Pedrero [58,59] stated that both qualitative and quantitative methods should be included when constructing a new assessment tool. These authors have indeed suggested that reliability, convergent validity, and factor structure are needed to demonstrate the accuracy and veracity of evaluating the construct [58,60]. It goes without saying that the test construction process needs to be explained in detail, considering all the theoretical and metric principles, since these kinds of studies do not appear to be automatic or universal [58,59,61,62]. Following strictly the guidelines for creating assessment instruments written by Muñiz and Fonseca-Pedrero [58,59], the psychometric model used, the type of item response, the application form, and the assessment context should be considered for the construction of a high-quality evaluation tool.

#### *Strengths and Limitations*

This study offers considerable advances in the examination of evaluation strategies used to analyze the MLS in different populations. However, it has certain limitations that should be taken into consideration. For instance, only four relevant databases were searched. Moreover, although a wide variety of keywords were used, some specific words may not have been identified and included in the search strategies. Moreover, the fact that only articles available in full text were evaluated for inclusion in the revision could also limit the search strategy. However, in this case, all the evaluated studies were found in full text and could be fully assessed for inclusion or non-inclusion in the review. Furthermore, although many studies aim to assess the MLS, they appear to only assess MD adherence or to not explicitly state that MLS is evaluated, potentially hindering the identification of the studies that assess the MLS as a whole. Some articles evaluate two or more components of the MLS but do not name them as such, assigning all of the benefits of this LS exclusively to MD adherence. Therefore, many benefits attributed to the MD may also be derived from MLS adherence, although they are not identified as such in the studies.

Despite these limitations, an exhaustive systematic review was carried out in this study, demonstrating that the MLS is being examined in an increasing number of works. The strengths and limitations of each evaluation strategy performed in the different studies were also evaluated. This analysis provides objective and reliable data on the importance of using a tool with adequate psychometric properties that is capable of performing a comprehensive assessment of all of the dimensions that constitute the MLS.

#### **5. Conclusions**

The MLS is considered to be a healthy lifestyle in which the frequency and quantity of the consumption of certain foods play a key role. It is also a lifestyle that refers to other dimensions rooted in traditional Mediterranean life, as well as the interrelation between parameters such as socialization, physical activity, leisure activities, proper rest, and diet.

This systematic review attempted to consider all of the methods used to evaluate the MLS. A total of four indices specifically designed for assessment were obtained. However, none of the methods evaluated all of the dimensions that constitute the MLS. Although MEDLIFE may be one of the most reliable and integrating questionnaires for the assessment of the MLS, other psychometric properties of this instrument should be analyzed in depth, such as its factorial structure and functioning in both clinical and healthy populations.

A notable limitation of the current evaluation strategies is the heterogeneity of the tools used to evaluate the MLS, since different authors use distinct methods of analysis. No consensus has yet been reached on a single instrument to comprehensively and reliably measure the MLS that has proven and adequate psychometric properties. Therefore, future studies should attempt to design a tool with appropriate psychometric properties for the general population and include all of the MLS dimensions.

This would allow professionals to carry out more accurate analyses of the level of adherence to a healthy lifestyle and would lead to the identification of populations at risk of developing different pathologies. Moreover, this tool would aid in the creation of comprehensive intervention programs that are aimed at improving health by promoting adherence to a healthy lifestyle, such as the MLS.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/nu14194179/s1, Table S1. Methodological quality assessment of the cross-sectional studies included in the review; Table S2. Methodological quality assessment of the randomized controlled trials included in the review; and Table S3. Methodological quality assessment of the cohort studies included in the review.

**Author Contributions:** Conceptualization, E.M.-S., R.F.-C., and N.R.-R.; methodology, E.M.-S., B.C.- L., R.F.-C., and N.R.-R.; investigation, E.M.-S., R.F.-C., C.A.-B., and N.R.-R.; writing—original draft preparation, E.M.-S., C.A.-B., B.C.-L., and N.R.-R.; writing—review and editing, E.M.-S., R.F.-C., N.R.-R., N.A.-B., and B.C.-L.; supervision, R.F.-C., N.R.-R., and N.A.-B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable since this study is a systematic review.

**Informed Consent Statement:** Not applicable since this study is a systematic review.

**Data Availability Statement:** Data sharing not applicable. No new data were created or analyzed in this study. Data sharing is not applicable to this article.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland www.mdpi.com

*Nutrients* Editorial Office E-mail: nutrients@mdpi.com www.mdpi.com/journal/nutrients

Disclaimer/Publisher's Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Academic Open Access Publishing

mdpi.com ISBN 978-3-0365-9157-5