Validity and reliability of the Turkish version of the chrono-med diet score (CMDS) for adult participants

Abstract

Background

This study aimed to examine the validity and reliability of the Chrono-Med Diet Score (CMDS) to assess adherence to the Mediterranean diet in Turkish adults.

Methods

The methodological research conducted in Gümüşhane and Ordu from January to June 2024, including 592 individuals (62.3% women, 37.7% men, and the mean age 43.4 ± 10.6 years). Data were collected through face‒to‒face interview that contained socio-demographic characteristics, the CMDS, the Mediterranean Diet Adherence Screener (MEDAS), and 24-h dietary recall. Internal consistency reliability was assessed using Cronbach’s alpha coefficient, while repeatability was examined using the test–retest method. Construct validity was assessed by the MEDAS, and components of the adapted CMDS obtained from the 24-h dietary recall were analyzed using confirmatory factor analysis. The content validity was assessed by the Paired samples t-test and Wilcoxon test. The Bland–Altman test with 95% limits of agreement was used to evaluate the agreement between sum of CMDS. The results were statistically evaluated at a p < 0.05 significance level.

Results

The content validity index of CMDS was 0.87. The overall Turkish version of CMDS had acceptable internal consistency (Cronbach’s α = 0.853), thus indicating that the score was reliable. The Paired samples t test coefficients between each item and the overall questionnaire ranged from 0.234 to 1.000. A higher intake of olive oil and fish obtained from a 24-h dietary recall was associated with a higher CMDS quartile (p < 0.05). The MEDAS score was correlated with a higher CMDS quartile (p < 0.001). A moderate positive correlation was determined between total CMDS and MEDAS (r = 0.467; p < 0.001).

Conclusions

Our assessments of the CMDS, consisting of 13 items, in Türkiye demonstrate that it is a viable and reliable instrument to measure adherence to the Mediterranean diet for the adult population. The CMDS is more accurate than other Mediterranean diet scores as it questions about the amount of physical activity and grain products, which helps determine each participant’s eating patterns and general health.

Introduction

The Mediterranean diet (MD) is an eating pattern scientifically characterized in the twentieth century, rooted in a long-standing tradition among communities residing along the Mediterranean Sea [1]. The MD has received recognition from United Nations Educational, Scientific and Cultural Organization as an intangible cultural asset, intricately linked to its geographical origins and defined by agricultural and dietary practices that engage responsibly with the environment [2]. The conventional Mediterranean Diet (MD) is characterized by a significant intake of vegetables and fruits, along with a balanced weekly consumption of whole grains, legumes, nuts, seeds, and aromatic herbs. Extra virgin olive oil serves as the primary source of fat, and moderate alcohol consumption, typically in the form of red wine, is encouraged with meals. It is also recognized for consuming little to no butter or whole-fat dairy products, as well as a modest amount of meat (with a preference for lean meats such as chicken, turkey, or rabbit) and its byproducts, mainly processed meats [3, 4].

Demographic characteristics, unhealthy eating habits, sedentary lifestyles, and interactions within family and social environments influence diet quality and adherence to the MD [5, 6]. Identifying typical meals such as breakfast, lunch, and dinner has become challenging because of the increased frequency of meal skipping and snacking. The consumption of most daily calories earlier in the day while ensuring regular overnight rest and reduced caloric intake is advisable because of its beneficial effects on risk factors associated with cardiovascular disease and the development of type 2 diabetes [7]. The Mediterranean dietary pattern has been associated with a reduced risk of chronic noncommunicable diseases, including cardiovascular diseases [8], diabetes [9], and cancer [10]. Several studies indicate that the MD is an effective preventive measure for decreasing morbidity and mortality rates in the general population [11, 12]. Moreover, MD is considered a dietary pattern that promotes environmental sustainability [13].

Various methods exist to assess adherence to the MD, beginning with the original pyramid and extending to general descriptions, a priori scoring systems, a posteriori dietary patterns, or classifications on the basis of food and nutrient composition [14]. These indices include the Mediterranean Diet Scale [15], the Italian Mediterranean Index [16], the Mediterranean Diet Scoring System to assess adherence to the Mediterranean dietary pattern [17], and the Mediterranean Diet Adherence Scale [18]. Several studies have employed comprehensive food frequency questionnaires and dietary recalls to assess adherence to dietary patterns [19, 20]. However, these methods require significant time, specific expertise, skilled personnel, and appropriate equipment. Moreover, their margin of inaccuracy is higher [21]. Recently, a priori scores have become the most prevalent, as they are the simplest to associate with primary dietary outcomes. The adherence score is calculated as the total points awarded for improved consumption of well-being products, with negative or no points assigned for health-detrimental foods or behaviors. The MD score (MDS), developed by Panagiotakos et al. (2007), is regarded as one of the most significant scores due to its ease of application [22]. Sofi et al. (2017) developed a novel MEDI-LITE score, utilizing the MDS as a reference point [23]. The Mediterranean lifestyle (MEDLIFE) was created by Sotos-Prieto et al. to evaluate how well an individual follows the Mediterranean lifestyle, which encompasses an analysis of dietary habits, physical activity levels, and social interactions [24]. MDS and MEDI-LITE showed strong correlations with the MD pyramid and the likelihood of following the MD. MEDLIFE assesses an individual’s comprehensive adherence to a healthy Mediterranean lifestyle.

The MD holds significant importance globally and within Turkish society, and it is recognized for its health benefits and sustainable dietary model. The Mediterranean Diet Adherence Screener (MEDAS) is a 14-item questionnaire including 2 questions about dietary habits and 12 questions concerning the frequency of food intake, giving it a viable and reliable instrument for evaluating adherence to the Mediterranean diet within the Turkish population [18, 25]. It was utilized in several research assessing individual adherence to the MD in Türkiye [26–28]. The Chrono-Med Diet Score (CMDS) developed by De Matteis et al. is a new score for assessing adherence to the MD. In addition to other scores, categories for the duration of farinaceous products and physical activity were incorporated to enhance the understanding of each participant’s dietary habits and overall health status [29]. However, the Turkish version of the CMDS is not valid or reliable in the Turkish-speaking population. Therefore, the objective of this study was to adapt the CDMS to Turkish society, considering cultural and linguistic differences.

Methods

Study group

This methodological study was conducted between March 2024 and July 2024 with 592 participants (223 men, 369 women) aged between 19 and 64 years with the snowball sampling method who were living in Gümüşhane and Ordu, located in the Black Sea region of Türkiye. Individuals were excluded from the study if they had any psychological disorders and/or chronic diseases requiring a specific diet, had any eating disorders, were pregnant and/or breastfeeding, did not have the mental health to answer questions correctly or did not agree to participate in the study. The sample size for methodical studies should be calculated by considering items five to ten times the scale’s size [30]. While a sample size of 130 (13 items*10) was required for the necessary study power, we included 592 participants. The sociodemographic data of individuals were evaluated via a questionnaire via face‒to-face interviews.

Chrono-med diet score

The CMDS was developed by De Matteis et al. (2023) to measure adherence to the MedDiet and identify increased risk for metabolic diseases [29]. It includes eleven food categories evaluated to determine the total score predicated on daily to weekly consumption. (1) Fruits, (2) vegetables, (3) legumes, (4) farinaceous products (e.g., bread, pasta, cookies), (5) cereal grains, (6) fish, (7) meat and meat products, (8) milk and dairy products, (9) olive oil, (10) butter, margarine, and lard, and (11) alcohol intake. Adding categories for the time of farinaceous product intake and physical activity improved the ability to characterize an individual’s dietary habits and overall health. The total CMDS ranges from -13 to 25 points, where a higher score reflects greater adherence to the MD. More details of the CMDS have been explained in the study of De Matteis et al. [29], and the final CMDS is presented as Supplemental Material in that article.

Translation of the CMDS to Turkish and cultural adaptation

The first step involved obtaining permission from the authors to translate and validate the questionnaire into Turkish. The translation technique used adhered to the standard procedure outlined by Brislin and Prieto’s method, in which the scale was translated from English to Turkish by researchers proficient in both languages [31, 32]. Two experts who are fluent in two languages translated the questionnaire into Turkish, and the food items of the CMDS were modified to reflect Turkish food culture. The standard translation–back translation method was used for language adaptation to minimize discrepancies in expression. Native Turkish speakers who were proficient in both languages and had not seen the scale’s original English version translated it from English. The translators discussed the discrepancies until they reached a consensus on a version that was more concise, syntactically accurate, better suited for the intended audience, and closer to the original text. The researchers conducted a review of the scale to confirm its clarity and appropriateness for both the research context and the cultural setting. The researchers analyzed the consistency of meaning, and the Turkish text was derived from the most suitable expressions.

Content validity evaluation

Following completion of the translation process, the questionnaire was presented to a panel of experts consisting of six academic dietitians. The experts reached out via e-mail and evaluated the clarity, comprehensibility, cultural relevance, and significance of each question. The Davis method calculates the content validity index for each item [33]. In this context, the experts assessed the items of the scale as noncompliant (1), requiring appropriate revisions (2), suitable but needing minor adjustments (3), and highly suitable (4) on the basis of Davis’ method. The total of the initial two ratings was divided by the number of experts, resulting in the calculation of the content validity index (CVI) following this assessment. The CVI of an item is regarded as sufficient when the CVI is more significant than 0.80.

Pilot study

The adapted CMDS was applied to twenty individuals who met the inclusion criteria, and they were requested to evaluate the items on the basis of comprehensibility, fluency, and other relevant aspects observed at this stage. The data collected during this preapplication step were not used in the subsequent stages of the study. Minor suggestions (e.g., synonym alteration) for increasing the understanding were accepted and considered for one item. The final Turkish version of the tool was produced.

Content reliability evaluation

The Turkish CMDS was administered face-to-face to fifty-two participants who consented to take part and were provided a contact number 15 days later for the purpose of assessing test–retest reliability. The test–retest analysis included determining the intraclass correlation coefficient (ICC) via a two-way mixed effect model, accompanied by a 95% confidence interval (CI). The classification of ICCs was as follows: excellent reliability above 0.90, good reliability between 0.75 and 0.90, moderate reliability between 0.50 and 0.75, and poor reliability below 0.50 [34]. The internal consistency of the questionnaire items was evaluated using Cronbach’s alpha, with a coefficient of ≥ 0.60 regarded as acceptable [35]. The agreement between the sum of the CMDS from the two administrations was examined via Bland–Altman plots. The 95% limits of the agreement lines were also plotted [36].

Data collection tools

Data were collected via questionnaires and face‒to‒face interviews with participants by researchers. The questionnaire consisted of five sections: (1) general information (age, sex, education level, marital status, and employment status), (2) health information and nutritional habits (smoking status, alcohol consumption, and the number of meals eaten per day), (3) anthropometric measurements (body weight and height), (4) the CMDS, (5) the Mediterranean Diet Adherence Screener (MEDAS), and (6) 24-h dietary recall. Body weight and height measurements were self-reported by the subjects. The participants were categorized into four groups according to the World Health Organization classification of body mass index (BMI). The BMI categories are underweight (under 18.5 kg/m²), normal weight (18.5–24.9 kg/m²), overweight (25.0–29.9 kg/m²), and obese (30.0 kg/m² or more) [37].

Mediterranean diet adherence screener

Adherence to the MD was assessed via the MEDAS, which was developed by Schröder et al. [18] and subsequently validated in Turkish by Bekar and Goktas [25]. The MD adherence score evaluates dietary habits through two questions regarding food types and twelve questions concerning the frequency of food consumption. Each question received a score of either 0 or 1. One point was provided for choosing white meat over red meat or for using olive oil when preparing or consumption: Four or more tablespoons of olive oil per day, three or more pieces of fruit per day, two or more servings of vegetables per day, seven or more servings of red wine per week, three or more servings of pulses per week, three or more servings of fish per week, three or more servings of nuts per week, two or more servings per week of sofrino (tomato, onion, garlic in olive oil) or less than one serving of animal fat (butter, margarine, cream) per day, less than one serving of red meat or sausage per day, less than one cup of sugar-sweetened beverages per day, and fewer than three commercial pastries per week. The category received zero points if the requirement was not met. The total score varies from 0 to 14. A score of 0 indicates low adherence, whereas a score of 14 reflects the highest adherence to the MD.

Physical activity level

The physical activity levels of individuals were assessed in accordance with the Physical Activity Guideline of Türkiye. The guideline recommends a minimum of 150 min of moderate-intensity or 75 min of vigorous-intensity weekly for every adult [38]. The questionnaire inquired about the intensity and length of weekly physical exercise, and the subjects’ replies were documented. Individuals who performed at least 150 min of moderate-intensity exercise (such as brisk walking, cycling, light jogging, swimming, or tennis) or at least 75 min of vigorous-intensity activities (including walking with weights, uphill walking, running, volleyball, or basketball) per week were classified as sufficiently active. In contrast, those not meeting these criteria were classified as inactive.

24-h dietary recall

A 24-h dietary recall is a methodical interview designed to obtain comprehensive information regarding all foods and beverages the respondent consumes over the preceding 24 h. Typically, the recalled period is defined as extending from when the respondent wakes up one day until the time when they wake up the following day [39]. Using a photographic atlas that shows food portion sizes and quantities, researchers were able to determine the amount of each food consumed [40]. From dietary recall, each of the CMDS component (excluding physical activity and time of intake of farinaceous products) intakes was calculated by the quartile distribution of the CMDS. The Nutrition Information System (BeBis v9.0) was employed to compute individuals’ overall intake of fruits, vegetables, legumes, cereals, fish, milk and dairy products, extra virgin olive oil, butter, margarine, lard, and alcohol utilizing 24-h dietary recall data [41]. Quartiles were calculated by dividing the CMDS into four equal parts, each part containing 25% of the CMDS. Based on the CMDS obtained from the 24-h dietary recall, the following quartile divisions were established: -0.4 to 3.0 CMDS for the first quartile, 4.0 to 6.0 CMDS for the second quartile, 7.0 to 8.0 CMDS for the third quartile, and 9.0 to 15.0 CMDS for the fourth quartile.

Statistical analysis

The data were analyzed via IBM SPSS V26 (SPSS Inc., Chicago, IL, USA). Data were examined via descriptive statistics, including the mean, standard deviation, median, minimum, maximum, number and percentage. The variables were assessed for normality via the Shapiro–Wilk test, skewness and kurtosis, and visual indicators such as probability and histogram diagrams. Nonnormally distributed data were analyzed via the Kruskal‒Wallis test, with subsequent multiple comparisons conducted via the Bonferroni correction. The Wilcoxon test was used to compare nonnormally distributed test‒retest scores, whereas the paired sample t test was used for normally distributed scores. The associations between nonnormally distributed quantitative variables were assessed via the Spearman correlation coefficient.

The participants were divided into four equal sections using four quartiles according to the CMDS, each of which represents one-fourth of the total. The cutoff points of the CMDS were determined according to quartiles, and the 4th quartile was evaluated as a high CMDS consumption pattern. The accepted significance level was p < 0.05.

Results

A total of 592 participants completed the study. The sociodemographic characteristics of the individuals who participated in the study are shown in Table 1. The mean age of the individuals was 43.4 ± 10.6 years. A total of 47.5% of the individuals were between the ages of 36 and 49, 62.3% were women, 31.3% had a normal BMI, and 65.0% were inactive. A total of 32.4% of the participants were university/college graduates, and 75.5% were married. Table 2 shows the CMDS’s face and content validity scores. The CVI values for the items varied from 0.83 to 1, and the Questionnaire Content Validity Index (Q‐CVI) was established at 0.87.

Table 1.

Sociodemographic characteristics of the participants (n = 592)

Variable	n	%
Age (years)
18–35	132	22.3
36–49	281	47.5
50–64	179	30.2
Sex
Women	369	62.3
Men	223	37.7
Education level
Elementary	146	24.7
Secondary	95	16.0
High school	144	24.3
University	192	32.4
Postgraduate	15	2.6
Employment status
Employed	344	58.1
Unemployed	248	41.9
Occupation
Civil servant	145	37.3
Employee	91	23.4
Self-employed	109	28.0
Retired	44	11.3
Marital status
Single	145	24.5
Married	447	75.5
Smoking status	152	25.8
Alcohol consumption	32	5.4
BMI classification
Underweight	9	1.5
Normal	185	31.3
Overweight	238	40.2
Obese	160	27.0
Physical activity
Sufficiently active	207	35.0
Inactive	385	65.0
	Mean	SD
Age (years)	43.4	10.6
Number of main meals	2.5	0.5
Number of snacks	1.4	0.8
BMI (kg/m2)	27.4	5.0

Data are expressed as the means (SD) or n (%)

BMI Body mass index

Table 2.

Results for the content validity of the Turkish version of the Chrono-Med Diet Score

Items	CVI	Q-CVI
Q1 = Fruit	0.83	0.87
Q2 = Vegetables	0.83
Q3 = Legumes	1.00
Q4 = Farinaceous products	1.00
Q5 = Time of farinaceous product intake	1.00
Q6 = Cereals	0.83
Q7 = Fish	0.83
Q8 = Meat and meat products	0.83
Q9 = Milk and dairy products	0.83
Q10 = Extra virgin olive oil	0.83
Q11 = Butter, margarıne, and lard	0.83
Q12 = Alcohol	0.83
Q13 = Physical activity	0.83

CMDS Chrono-Med Diet Score, CVI Content validity index, Q-CVI Questionnaire content validity index

The internal consistency reliability of the adapted CMDS was specified via Cronbach’s alpha coefficient. The coefficient was 0.850 for the adapted CMDS and ranged from 0.707–1.000. For each of the thirteen items, the coefficient values were as follows: fruit (0.856), vegetables (0.727), legumes (0.806), farinaceous products (0.791), time of farinaceous product intake (0.707), cereals (0.752), fish (0.816), meat and meat products (0.816), milk and dairy products (0.789), extra virgin olive oil (0.935), butter, margarine, and lard (0.860), alcohol (1.000), and physical activity (0.919). Moreover, the reliability analysis of the CMDS is shown in Table 3. ICCs were examined in the context of the test–retest analysis and the ICC result was 0.853 for the study population. Based on these values, fruit (0.858), vegetables (0.730), legumes (0.782), farinaceous products (0.788), time of farinaceous product intake (0.710), cereals (0.755), fish (0.819), meat and meat products (0.815), milk and dairy products (0.790), butter, margarine, and lard (0.858) had good reliability. Extra virgin olive oil (0.936), alcohol (1.000), and physical activity (0.921) had excellent reliability. The Paired samples t-test and Wilcoxon test correlation coefficients between each items and the overall score ranged from 0.234 ~ 1.000, especially the coefficients of fruit, vegetables, time of farinaceous product intake, fish, extra virgin olive oil, alcohol, and physical activity, were more than 0.6, which showed a strong correlation with the overall score.

Table 3.

Reliability analysis of the chrono-med diet score

		Mean	SD	Median (min–max)	p	95%CI
Q1 = Fruit	Test	0.8	0.7	1.0 (0.0–2.0)	1.000*	0.858 [0.752–0.919] p < 0.001
	Retest	0.8	0.7	1.0 (0.0–2.0)
Q2 = Vegetables	Test	0.9	0.7	1.0 (0.0–2.0)	0.637*	0.730 [0.529–0.845] p < 0.001
	Retest	1.0	0.6	1.0 (0.0–2.0)
Q3 = Legumes	Test	1.2	0.6	1.0 (0.0–2.0)	0.555*	0.782 [0.593–0.880] p < 0.001
	Retest	1.0	0.6	1.0 (0.0–2.0)
Q4 = Farinaceous products	Test	0.6	1.2	1.0 (−1.0–2.0)	0.234*	0.788 [0.633–0.878] p < 0.001
	Retest	0.8	1.2	1.0 (−1.0–2.0)
Q5 = Time of farinaceous product intake	Test	−1.7	1.8	1.0 (−4.0–1.0)	0.675*	0.710 [0.494–0.834] p < 0.001
	Retest	−1.8	1.7	1.0 (−4.0–1.0)
Q6 = Cereals	Test	0.6	0.5	1.0 (0.0–1.0)	0.527*	0.755 [0.572–0.859] p < 0.001
	Retest	0.6	0.5	1.0 (0.0–1.0)
Q7 = Fish	Test	0.5	0.7	0.0 (0.0–2.0)	0.782*	0.819 [0.593–0.880] p < 0.001
	Retest	0.4	0.6	0.0 (0.0–2.0)
Q8 = Meat and meat products	Test	1.6	0.6	2.0 (0.0–2.0)	0.248*	0.815 [0.680–0.894] p < 0.001
	Retest	1.5	0.6	2.0 (0.0–2.0)
Q9 = Milk and dairy products	Test	1.3	0.6	1.0 (0.0–2.0)	0.405*	0.790 [0.635–0.879] p < 0.001
	Retest	1.3	0.6	1.0 (0.0–2.0)
Q10 = Extra virgin olive oil	Test	0.1	1.3	−1.0 (−1.0–2.0)	0.739*	0.936 [0.888–0.963] p < 0.001
	Retest	0.1	1.3	−1.0 (−1.0–2.0)
Q11 = Butter, margarine, and lard	Test	0.1	1.1	1.0 (−2.0–1.0)	0.160*	0.858 [0.753–0.918] p < 0.001
	Retest	−0.1	1.1	1.0 (−2.0–1.0)
Q12 = Alcohol	Test	3.0	0.3	3.0 (1.0–3.0)	1.000*	1 (1–1)
	Retest	3.0	0.3	3.0 (1.0–3.0)
Q13 = Physical activity	Test	−0.7	2.1	−1.0 (−3.0–3.0)	0.891*	0.921 [0.861–0.954] p < 0.001
	Retest	−0.7	2.1	−1.0 (−3.0–3.0)
Total CMDS	Test	8.0	4.4	8.0 (−3.0–19.0)	0.859**	0.853 [0.743–916] p < 0.001
	Retest	7.9	4.3	8.0 (−2.0–16.0)

*Paired-samples t test

**Wilcoxon test, ICC 95% confidence interval: intraclass correlation coefficient

The Bland‒Altman plot of the total difference in scores was used to estimate the agreement between applications of the score (Fig. 1). The mean difference between the two CMDS administrations (0.08 ± 3.11, with a 95% limit of agreement of -6.02, 6.17) demonstrated that both CMDS administrations showed similar total adherence MD scores.

Fig. 1.

Bland‒Altman plot showing the agreement for the two CMDS administrations. CMDS, Chrono-Med Diet Score

Table 4 displays the distribution of food intake obtained from 24-h dietary recall, MEDAS score, and BMI categorized by the quartiles of the CMDS. The food intake, MEDAS score, and BMI of the study participants according to the CMDS quartiles (quartile 1: n = 171, -0.4 ≤ CMDS ≤ 3.0; quartile 2: n = 160, 4.0 ≤ CMDS ≤ 6.0; quartile 3: n = 120, 7.0 ≤ CMDS ≤ 8.0; quartile 4: 9.0 ≤ CMDS ≤ 15.0) are shown. Participants in the fourth CMDS quartile tended to have higher intakes of fruit, fish, meat and meat products, and extra virgın olive oil compared to those in the first, second, and third CMDS quartiles.There were significant differences in the olive oil and fish intake for the CMDS quartiles (p < 0.05). Moreover, an important difference was observed in the MEDAS scores across the CMDS quartiles (p < 0.001) [fourth quartile versus first quartile; fourth quartile versus second quartile; fourth quartile versus third quartile; second quartile versus first quartile]. As shown in Fig. 2, a moderate correlation was noted between the MEDAS score and the CMDS (r = 0.467, p < 0.001). Specifically, a higher MEDAS score corresponded to a higher CMDS.

Table 4.

Food intake, MEDAS scores and BMI according to the CMDS quartiles

	Quartiles of the CMDS												Total (n = 592)		p
	1st Quartile (n = 171)			2nd Quartile (n = 160)			3rd Quartile (n = 120)			4th Quartile (n = 141)			Mean	SD
	Mean	SD	Min–Max	Mean	SD	Min–Max	Mean	SD	Min–Max	Mean	SD	Min–Max
CMDS	1.3	1.6	−4.0–3.0	5.1	0.8	4.0–6.0	7.5	0.5	7.0–8.0	11.0	1.9	9.0–15.0	5.9	3.8
Fruits	84.1	108.1	0.0–510.0	96.6	119.9	0.0–695.0	101.2	120.0	0.0–600.0	105.9	129.8	0.0–700.0	96.1	119.1	0.345
Vegetables	306.1	280.5	0.0–1331.0	367.0	372.9	0.0–1713.0	344.5	425.2	0.0–2371.0	289.4	268.1	0.0–1470.0	326.3	338.3	0.636
Legumes	20.5	38.7	0.0–188.0	17.4	34.2	0.0–200.0	20.0	38.1	0.0–180.0	18.9	32.9	0.0–170.0	19.2	36.0	0.789
Farinaceous products	166.2	154.3	0.0–800.0	143.6	131.0	0.0–750.0	138.0	134.2	0.0–525.0	142.7	164.8	0.0–970.0	148.8	147.1	0.183
Grain cereal	76.0	91.1	0.0–730.0	70.6	78.3	0.0–550.0	71.2	80.8	0.0–672.0	66.1	88.9	0.0–645.0	71.2	85.1	0.193
Fish	11.8	52.9	0.0–375.0	4.9	31.2	0.0–300.0	4.7	30.5	0.0–300.0	17.1	55.5	0.0–300.0	9.8	44.8	0.020
Meat and meat product	44.1	55.8	0.0–250.0	46.6	68.5	0.0–400.0	39.0	69.4	0.0–500.0	49.3	105.0	0.0–1000.0	45.0	75.8	0.163
Milk and dairy products	149.3	137.8	0.0–660.0	143.2	124.3	0.0–710.0	164.7	138.2	0.0–604.0	155.7	158.1	0.0–950.0	152.3	139.4	0.545
Extra virgın olive oil	3.7	10.3	0.0–85.0	4.1	8.9	0.0–50.0	4.0	8.2	0.0–41.0	6.1	12.2	0.0–80.0	4.3	10.1	0.027
Butter, margarine, and lard	17.9	15.8	0.0–70.0	15.3	13.9	0.0–75.0	15.3	13.3	0.0–61.0	15.4	15.5	0.0–80.0	16.1	14.8	0.388
Alcohol	0.0	0.0	0.0–0.0	0.0	0.0	0.0–0.0	0.0	0.0	0.0–0.0	0.0	0.0	0.0–0.0	0.0	0.0	NA
MEDAS	5.7	1.8	1.0–10.0	6.5	2.0	2.0–13.0	7.0	1.9	2.0–13.0	7.6	2.1	2.0–13.0	6.6	2.1	< 0.001
BMI	27.8	5.0	17.9–39.9	27.5	5.4	17.0–45.0	26.9	4.6	16.1–38.2	27.1	4.8	18.3–41.4	27.1	5.0	0.607

The data are expressed as the mean (SD) and min–max

BMI Body mass index, CMDS Chrono Med-Diet Score, MEDAS Mediterranean Diet Adherence Screener

p values were obtained with ANOVA and the Kruskal‒Wallis test between quartiles

Fig. 2.

Association between the CMDS and MEDAS score. CMDS, Chrono-Med Diet Score; MEDAS, Mediterranean Diet Adherence Screener. The p value was obtained via Spearman correlation

Discussion

The MD, characterized by the traditional eating habits of countries close to the Mediterranean Sea, is globally acknowledged as a healthy dietary pattern and has been frequently recommended in various publications as a crucial lifestyle factor for promoting good health [26]. Research indicates that dietary scores serve as effective instruments for assessing adherence to the MD and its associated health benefits. Various scores are employed to assess the level of agreement with MD. The scores present composite constructs based on dietary components, combining foods and nutrients to establish valid operational variables that assess the relationship between diet quality and health outcomes [42, 43]. However, these scores exhibit certain limitations. Consequently, it is essential to develop a user-friendly adherence questionnaire that includes details on dietary and lifestyle practices.

In 2023, the CMDS was developed by De Matteis with the aim of creating a useful tool for clinicians to screen adherence to MD [29]. The CMDS is a novel questionnaire designed to assess adherence to the MD; it contains eleven food categories, including the chronobiology of carbohydrate intake and physical activity, and serves as an easy-to-use instrument [29]. Nevertheless, this comprehensive, valid, and reliable measurement instrument has not been examined in Turkish. This study aimed to evaluate the validity and reliability of the CMDS among Turkish adults, thereby facilitating its broader application in both research and clinical environments. According to the findings of our study, the Turkish version of the CMDS, which has 13 items, is a valid and reliable tool for the adult population.

In our study, the CVI value of total CMDS was found to be 0.87. Ten items received a score of 0.83 in the evaluation, while the remaining three items received a full score of 1.00. All items were retained in the scale as they all met the requirement of > 0.80 as outlined in the literature [44]. The reliability of a measurement indicates the degree to which a score is precise, consistent, and reproducible. The evaluation of a scale’s reliability depends significantly on its internal consistency. A variety of parameters are employed to evaluate internal consistency, with Cronbach’s α being the most frequently utilized measure. The CMDS’s internal reliability coefficients ranged from 0.707 to 1.000. The original study did not provide Cronbach’s alpha values; however, in this study the Cronbach alpha value for the total score was 0.850, indicating that the score is highly reliable [45]. In another scale adaption assessing adherence to the MD in Türkiye, the Cronbach’s α value was 0.954 [46]. Similarly, in other scale adaptation studies conducted in Türkiye, a Cronbach’s α greater than 0.60 was considered reliable [47–49].

The test‒retest method provides an additional criterion for assessing reliability [50]. This study indicates that the pre- and postcorrelations of the score items range from 0.710 to 1.000, thus confirming the test–retest reliability of the scale. Specifically, the correlations for butter, margarine, and lard, fruit, physical activity, extra virgin olive oil, and alcohol score were 0.858, 0.921, 0.936, and 1, respectively. Also, the test–retest analysis revealed that the CMDS has a good reliability (ICC = 0.853). Although correlations are commonly utilized to evaluate reliability in diet validation techniques, supplementary assessments are beneficial since they offer a restricted measure of the concordance between measurements. Therefore, a Bland Altman graph was plotted total difference in score to predict the agreement between administrations of the score. Both administrations of the CMDS created a similar mean total score (8.02 ± 4.37 vs 7.94 ± 4.25) and relative agreement was good (r = 0.740, p < 0.001). Furthermore, calculating the mean CMDS difference between CMDS tests (0.08 ± 3.11, 95% limit of agreement -6.02, 6.17) confirmed that the two CMDS tests resulted in similar overall CMDS. As a result, the adapted CMDS is highly reliable and repeatable.

The MEDAS score and food intake data obtained from 24-h dietary recall were utilized to evaluate construct validity. The intake of extra virgin olive oil and MEDAS scores were significantly higher in the participants in the highest quartile of the CMDS than in the participants in the lowest CMDS quartile. Compared with those in the other CMDS quartiles, fish intake was significantly greater in the participants in the highest quartile of CMDS. However, butter, margarine and lard presented the highest intake in the first quartile, whereas fruit intake gradually increased across quartiles; however, no significant differences were detected. In the present study, we observed a medium adherence to the MD (mean score of 6.6), which aligns with previous studies performed in Türkiye [51, 52]. A moderate correlation was found between the MEDAS score and the CMDS (r = 0.467; p < 0.001). The original study analyzed the link between the CMDS and MDS in evaluating adherence to the MD, revealing a strong correlation between the two measures (r = 0.680; p < 0.001). CMDS also showed a significant negative correlation with BMI [29]. Despite that individuals in the highest quartile of the CMDS had a lower BMI than those in the lowest quartile, the difference was not statistically significant in the present study.

Due to the new development of CMDS, there is a lack of research utilizing CMDS in the literature. The adaption of CMDS to various the population is expected to increase the amount of research conducted. The initial research related the CMDS to MD adherence and many diseases, including visceral obesity, assessed by waist circumference, dyslipidemia, glucose intolerance, heightened cardiovascular risk, and hepatic steatosis [29]. A recent study including individuals who have gastrointestinal cancer indicated that low CMDS levels may serve as a legitimate marker for heightened cancer risk and adherence to the MD [53].

Limitations

This study has strengths and several limitations. First, a significant strength is that this study was the first to assess the validity and reliability of the CMDS in the Turkish community. Second, incorporating an extensive national sample and using stringent statistical techniques, including Cronbach’s alpha for internal consistency and test–retest reliability, together with corroboration from 24-h dietary recall, enhances the reliability of our findings. Third, the CMDS was validated compared with adherence to the MD and is acknowledged as a healthy and sustainable nutritional model. Finally, the validity and reliability of the CMDS in Turkish adults have important implications for epidemiological research and interventions related to adherence to the MD, nutrition, and health. In addition, this study has several limitations. Despite the substantial sample size, the findings exhibit restricted generalizability because of the use of a nonprobability sampling method. Second, this study was conducted in two cities (adults residing in Gümüşhane and Ordu), which may have resulted in bias. Third, we included only healthy adults in this study and excluded special groups (pregnant and/or breastfeeding women and those with chronic diseases) to minimize specific factors that may influence adherence to the MD. Given these limitations, further studies utilizing the CMDS with more diverse and representative samples are necessary to achieve more generalizable results and enhance external validity.

Conclusions

This study presents a novel tool in the Turkish language that contributes to the current literature on adherence to the MD. The results of our analyses evaluating the CMDS in Türkiye indicate that it is a valid and reliable tool for use in the adult population. Owing to its brevity and clarity, it may enhance adherence to usage. Future studies should evaluate the validity and reliability of the CMDS in both clinical and research contexts, utilizing various modes of administration, such as face‒to‒face interviews, and including an expanded population, including clinical groups.

Abbreviations

BMI

Body mass index

CMDS

Chrono-med diet score

CVI

Content validity index

ICC

Intraclass correlation coefficient

MD

Mediterranean diet

MEDAS

Mediterranean diet adherence screener

MEDLIFE

Mediterranean lifestyle

Author contributions

E.K. and T.K.: conceptualization, methodology, investigation, writing-review and editing; E.K.: investigation, writing-original draft preparation; E.K., T.K., and D.A.: writing-review and editing. All the authors approved the final submitted and published version.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Availability of data and materials

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The Ordu University Social Sciences and Humanities Ethics Committee (protocol number of meetings: 9, decisions: 2023-209, date: 07.12.2023), and the Helsinki Declaration principles were applied in the research. Before the survey commenced, all participants were briefed on the study’s details and provided their signatures on an informed consent form, confirming their voluntary involvement in the research.

Competing interests

The authors declare no competing interests. Duygu Ağagündüz is a Section Editor for Journal of Health, Population and Nutrition.