Planetary Health Diet adherence in relation to dietary cost and its variation by socio-economic factors among Iranian older adults

doi:10.21203/rs.3.rs-4919309/v1

Download PDF

Article

Planetary Health Diet adherence in relation to dietary cost and its variation by socio-economic factors among Iranian older adults

https://doi.org/10.21203/rs.3.rs-4919309/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Poor dietary patterns among older adults are significantly influenced by socioeconomic status (SES) and food prices, impacting both diet quality and food choice. This study examines how dietary cost relate to the Planetary Health Diet Index (PHDI) and assesses the role of SES in Iran's elderly population.

In this cross-sectional study, 398 elderly individuals from southern Tehran were sampled. Dietary data were collected using a validated food frequency questionnaire, and adherence to the PHDI was evaluated based on EAT-Lancet guidelines. food costs were derived from the Iranian Statistics Center and retail prices. Multivariable linear regression analyzed the relationship between PHDI adherence and dietary cost.

PHDI adherence varied significantly across demographics. Higher adherence was observed in females (OR = 0.82, 95% CI: 0.71-0.95) and those with a BMI ≤27.5 (OR = 0.84, 95% CI: 0.73-0.97), while single individuals had higher scores (OR = 0.85, 95% CI: 0.74-0.98). Higher PHDI tertiles were linked to lower daily dietary cost, with a significant decrease observed in the highest tertile compared to the lowest (β = -708,367 Rials, 95% CI: -1,060,371 to -356,362). However, this association was not significant among single and low-income participants.

The study reveals that higher adherence to PHDI is associated with reduced dietary cost, although this relationship does not hold for single or lower income individuals. Future research should focus on longitudinal studies to develop equitable public health strategies that promote affordable and sustainable dietary practices.

Health sciences/Health care/Nutrition

Health sciences/Health care/Public health

Planetary Health Diet Index

dietary cost

socioeconomic status

Poor dietary patterns are not solely the result of individual behaviors; they are significantly shaped by a wide range of contextual factors, including social, commercial, environmental, and cultural influences ¹. Food price is one of the primary factors impacting food choice and diet quality ². Some studies demonstrated that higher quality diets which tend to have lower energy content cost more than unhealthy diets ^3–6. Diets rich in whole grains, lean meats, fish, fresh fruits and vegetables tend to have a higher cost per kilocalorie compared to energy-dense diets containing higher levels of fats and sweets ⁷. However other studies have indicated that healthy diets are not more expensive than the unhealthy ones ^8–11.

In addition to food cost, the existence of a social gradient in diets and health has been widely established ^3,12–14. While individuals with higher SES typically gravitate towards higher-quality diets, those with lower SES often opt for energy-dense diets lacking in essential nutrients ¹⁵. Socioeconomic factors may have a greater impact on the dietary decisions of the elderly population, rendering them more susceptible to food insecurity ¹⁶. As demonstrated, poor nutritional quality, is prevalent among older adults with low SES ^17,18. Earlier studies have also demonstrated that the cost of food influences the food choices made by older adults ^19–21. Therefore, it is essential to recognize the factors that might hinder adherence to a high-quality diet within this demographic.

In the face of escalating global challenges such as climate change, there is an urgent imperative to transition towards more sustainable food systems. Hence, in addition to health of diet, we must consider the sustainability aspect while assessing diet quality. According to the Food and Agriculture Organization, sustainable healthy diets are “dietary patterns that promote all aspects of an individual’s health and wellbeing; have low environmental pressure and impact; are accessible, affordable, safe and equitable; and are culturally acceptable” ²². Expanding on this notion, the EAT-Lancet Commission introduced the concept of the "Planetary Health Diet" ²³. Numerous studies have demonstrated the co-benefits of a more sustainable diet on both planetary and human health ^23–26. Conversely, other studies highlight potential conflicts among the various dimensions of food sustainability. For example, it has been shown that Individuals with a more sustainable diet have slightly higher diet monetary cost ²⁷. While some studies have investigated the affordability of the EAT-Lancet diet, there is limited data available on the variations in the cost of diet based on adherence to the Planetary Health Diet ^28,29.

Given that research has shown individuals with higher SES tend to uphold higher-quality diets and experience lower disease risks, epidemiological studies frequently view SES as a potential confounding factor. The current study explored the relationship between diet and food cost stratified by socioeconomic and demographic variables. With the aging demographic and economic crisis contributing to higher food prices in Iran, our aim was to investigate how the cost of diet varies according to the sustainability and health, as measured by the PHDI, among older adults.

Study population and design

This study is a cross-sectional analysis conducted on 398 elderly individuals living in the southern region of Tehran. Participants were selected from healthcare centers affiliated with Tehran University of Medical Sciences between October 2022 and May 2023, utilizing the simple random sampling method. Individuals aged 60 and above, who reported no chronic illnesses, did not take specific medications, and did not alter their usual diet due to illness or dietitian recommendation, were included in our study.

This study was approved by the Tehran University of Medical Sciences Research Ethics Committee under reference number IR.TUMS.MEDICINE.REC.1401.588, and it adhered to the principles outlined in the Helsinki Declaration. Each participant in the study voluntarily provided written informed consent.

Dietary data

The dietary intake of participants was collected using a validated and reliable semi-quantitative 168-item Food Frequency Questionnaire (FFQ) ^30,31. Participants were asked to provide information about both the frequency and quantity of their consumption of each food item over the past year. This information was collected through a face-to-face interview conducted by a skilled and experienced researcher. The FFQ comprised an extensive list of food items, each with standardized serving sizes, allowing participants to indicate their consumption frequency on a daily, weekly, monthly, or yearly basis. Considering the specified portion size and consumption frequency for each food item, all foods were calculated on a daily basis and then converted to grams per day using household measures ³². Nutrient values, including energy and other nutritional components for each food item, were computed using Nutritionist 4 software (First Databank, Hearst Corp., San Bruno, CA, USA).

Calculation of PHDI

The Planetary Health Diet Index was derived based on the guidelines outlined in the reference diet proposed by the EAT-Lancet Commission ²³. The method used to calculate the PHDI was developed by Cacau et al. ³³. The PHDI comprises 16 components, with a maximum achievable score of 150. A higher PHDI score indicates a greater adherence to the Planetary Health Diet. Food groups were classified into adequacy, optimum, ratio, and moderation categories. Each food within the adequacy (including nuts and peanuts, legumes, fruits, vegetables, and whole cereals), optimum (including eggs, fish and seafood, tubers and potatoes, dairy, and unsaturated oils), and moderation (including red meat, chicken and substitutes, animal fats, and added sugars) categories can receive a maximum score of 10. Components in the ratio category (representing compositional distribution of dark green vegetables and red and orange vegetables relative to the total vegetable intake), however, are assigned a maximum score of 5 to prevent the overvaluation of a specific dietary aspect during the assessment process. The PHDI utilized the recommended intake ranges and midpoints outlined in the 2500 kcal/d reference Planetary Health Diet ²³. Subsequently, these values were converted into the percentage of calories contributed by each food group to the overall diet.

Food cost

The average food price data were derived from the prices listed on the website of the Iranian Statistics Center, which provided the average price of food items per 1000 grams in 2022 ³⁴. This dataset includes the average prices of each food per month in Iran. It offers the mean price for the region as a whole, including cities and rural areas. For food items not found on the Statistics Center website, we then went to retail centers and licensed stores situated in three distinct locations in southern Tehran. We took the average price of the 3 locations for the average price of the particular foods. Following this, the prices of these food items were adapted to correspond to the average price in 2022. This adaptation involved utilizing the Inflation Rate and Consumer Price Index available on the Central Bank website ³⁵. To ensure the consistency between the prices reported by the Statistical Center of Iran and those obtained from retail centers, a comparison was undertaken for the prices of three primary food items with the highest consumption (rice, pasta, and tea). The variation between the values from the two sources was negligible. Subsequently, the acquired prices, which pertain to the purchased forms, were converted into the prices of the edible amount of the food. To accomplish this, conversion factors outlined in the household scales book were utilized. Ultimately, the cost of food items was computed in Rials per gram for each item listed in the food frequency questionnaire. The cost of each food item consumed in the food frequency questionnaire for different individuals was determined by multiplying the grams consumed by the unit cost (the price of one gram). Finally, the sum of all food costs was regarded as the total daily diet price for each participant.

Sociodemographic characteristics

The general and sociodemographic characteristics of the households were asked through a face-to-face questionnaire. Each participant was interviewed regarding the following variables: gender, age, marital status (single or married), educational attainment (below high school, high school and above), working status (unemployed, retired, worker), household dimension (number of family members), house ownership (owner or rental), possession of specific items (including personal vehicle, washing machine, dishwasher, laptop, internet, microwave), and income (≤ 20 million Rials, 20 to 60 million Rials, and ≥ 60 million Rials).

Assessment of other variables

Participants height and body weight were measured by a trained researcher. Body weight was measured with the participant barefoot and wearing lightweight clothing, using a Seca model digital scale with a precision of 0.1 kg. Height measurement was taken while the individual stood without shoes, aligning their head, buttocks, and heels against the wall, and looking straight ahead horizontally. This measurement was conducted using a tape measure with an accuracy of 0.1 cm.

Lifestyle data incorporated details about smoking habits. Participants were specifically asked about their smoking status, indicating whether they were smokers or non-smokers. The International Physical Activity Questionnaire (IPAQ), utilized to evaluate the physical activity levels of the elderly, has undergone validation and reliability assessments in 12 countries including Iran, up to the year 2000. ^36,37. The final results of this assessment confirm the effectiveness of the questionnaire as a suitable tool for measuring physical activity in diverse contexts and languages. Consisting of seven inquiries, this survey evaluates the frequency and duration of an individual's involvement in physical activities. Participants report their engagement levels in four categories over the past week: 1) intense activity; 2) moderate activity; 3) walking; and 4) sitting. To precisely calculate MET (Metabolic Equivalent of Task) per minute, it is essential to express the duration of physical activity in minutes. In the end, by summing up the MET per minute for each person throughout a week, the total weekly physical activity can be ascertained.

Statistical analysis

First, the percentage of calories from all 16 food groups recommended by EAT-Lancet was determined. Then, the PHDI score was calculated as previously described. Participants were then grouped into tertiles based on their PHDI scores.

Mean (SE) values for food cost and PHDI scores were calculated across different demographic variables using analysis of variance (ANOVA) to identify differences in mean values of PHDI among categorical variables such as sex, BMI, smoking status, physical activity, marital status, household dimension, educational attainment, income, and socioeconomic status. Analysis of covariance (ANCOVA) was employed to compare dietary intakes across tertiles of food cost, adjusted for calorie intake (except for energy intake), age, and gender. Linear regression analyses were conducted to estimate the associations between PHDI adherence and food cost, with two models: Model 1 adjusted for energy intake, and Model 2 further adjusted for age, gender, physical activity, marital status, socioeconomic status, smoking status, supplement intake, and BMI. Additionally, linear regression was used to estimate the association between PHDI scores and food costs for every 10-point increase in PHDI. Adjusted mean (SE) values for diet cost were estimated across tertiles of PHDI, stratified by sociodemographic variables to understand the variation in food costs associated with PHDI adherence within different subgroups. The analyses were carried out using SPSS version 26 (IBM Corp, Armonk, NY, USA), and statistical significance was determined for P-values below 0.05.

In total, 398 participants, evenly split between males and females (50% each), were enrolled in the study. The average (standard deviation) age of the participants was 63.28 (3.58) years, and their average (SD) weight was 76.43 (10.39) kg.

Table 1 represents the mean (SD) of PHDI and food cost (Rials) across demographic variables. The PHDI and daily food costs varied across different sociodemographic characteristics among 398 elderly participants. Significant differences in PHDI scores were observed between males (55.46 ± 9.71) and females (57.81 ± 9.25), with females demonstrating higher adherence to the Planetary Health Diet (P value = 0.01). Participants with a higher BMI (> 27.5) showed a slightly lower PHDI score (55.84 ± 9.77) compared to those with BMI ≤ 27.5 (58.04 ± 8.99) (P value = 0.02). Marital status also influenced PHDI scores, with single participants scoring higher (58.61 ± 9.81) compared to married individuals (56.13 ± 9.42) (P value = 0.03). However, no significant differences in PHDI were found based on smoking status, physical activity levels, household dimension, education levels, income, or socioeconomic status. Regarding daily food costs, significant differences were observed only in relation to marital status (P value = 0.09), where single participants had lower mean food costs (2,613,420 ± 152,212) compared to married participants (2,917,889 ± 143,286). Other sociodemographic variables did not show statistically significant differences in daily food costs among the participants.

Table 1

Mean scores of PHDI (Planetary Health Diet Index), and individual diet cost (per day) according to sociodemographic characteristics (n = 398)
		PHDI		Food Cost
Variable	Percentage	Mean (SD)	P value¹	Mean (SD)	P value¹
Sex			0.01		0.01
Male	50.0	55.46 (9.71)		3,037,975 (138090)
Female	50.0	57.81 (9.25)		2,673,874 (145469)
BMI			0.02		0.98
≤ 27.5	36.2	58.04 (8.99)		2,857,465 (154345)
> 27.5	63.8	55.84 (9.77)		2,855,051 (140502)
Smoking Status			0.55		0.11
No	79.1	56.78 (9.56)		2,796,033 (147245)
Yes	20.9	56.09 (9.50)		3083222 (136997)
Physical Activity			0.21		0.57
Low	48.2	55.85 (10.14)		2847243 (137296)
Moderate	48.2	57.51 (8.86)		2835323 (153697)
High	3.5	55.42 (9.74)		325752 (142824)
Marital Status			0.03		0.09
Single	20.4	58.61 (9.81)		2613420 (152212)
Married	79.6	56.13 (9.42)		2917889 (143286)
Household Dimension			0.92		0.10
< 4	20.9	56.73 (8.86)		3088987 (181519)
≥ 4	79.1	56.61 (9.73)		2794514 (134048)
Education²			0.11		0.95
Low	58.8	56.01 (9.67)		2859483 (137955)
High	41.2	57.54 (9.31)		2850847 (155996)
Income			0.89		0.15
Low	4.8	57.63 (10.45)		2409204 (101104)
Moderate	65.1	56.57 (9.64)		2809098 (142303)
High	30.2	56.64 (9.54)		3027722 (156471)
Socioeconomic Status			0.41		0.36
Low	26.6	55.64 (9.53)		2688881 (126324)
Moderate	41.2	56.78 (9.82)		2891124 (155232)
High	32.2	57.28 (9.20)		2949158 (147283)
¹Obtained from the ANOVA
²Low=below high school, High = high school and above

Table 2 displays dietary intakes across tertiles of individual daily food cost among 398 elderly participants. Participants in the highest tertile of food cost (≥ 3,073,987 Rials) had significantly higher energy intake (2528.66 ± 52.56 kcal/day) compared to those in the lowest tertile (P < 0.001). Between macronutrients, significant differences were observed across food cost tertiles for carbohydrate (P = 0.04) and protein consumption (P = 0.001). Also, iron (14.91 ± 0.24) and magnesium (259.88 ± 4.97 mg/day) intakes were significantly lower in the highest food cost tertile compared to the lowest tertile (P < 0.001). No significant differences were observed for other dietary intakes across the tertiles of food cost.

Table 2

Nutrient intakes across tertiles of individual food cost (per day)
	Food Cost Tertiles			P-value¹
	1,961,460 ≥ T1 (n = 132)	1,961,460 < T2 < 3,073,987 (n = 133)	T3 ≥ 3,073,987 (n = 133)	P-value¹
Energy (kcal)	53.44 ± 1886.59	2166.87 ± 52.50	2528.66 ± 52.56	< 0.001
Macronutrients
Protein (percent of total calories)	13.15 ± 0.19	12.58 ± 0.19	12.13 ± 0.19	0.001
Carbohydrate (percent of total calories)	62.75 ± 0.58	64.43 ± 0.57	62.61 ± 0.57	0.04
Total fat (percent of total calories)	26.47 ± 0.57	26.16 ± 0.56	27.41 ± 0.56	0.26
W3 (gram/day)	0.42 ± 0.02	0.38 ± 0.01	0.39 ± 0.01	0.20
Total fiber (gram/day)	19.35 ± 0.43	18.37 ± 0.41	17.63 ± 0.38	0.43
Food groups
Vegetable (gram/day)	362.97 ± 14.72	321.27 ± 13.89	327.77 ± 14.60	0.10
Fruit (gram/day)	345.51 ± 15.52	353.33 ± 14.64	359.60 ± 15.39	0.82
Dairy (gram/day)	271.74 ± 18.09	258.42 ± 17.07	266.91 ± 17.94	0.86
Red meat (gram/day)	29.26 ± 1.97	33.32 ± 1.86	36.08 ± 1.95	0.06
Whole grain (gram/day)	64.39 ± 5.11	59.22 ± 4.82	57.17 ± 5.07	0.61
Micronutrients
Calcium (mg/day)	815.85 ± 25.01	770.93 ± 23.60	783.73 ± 24.81	0.42
Iron (mg/day)	16.49 ± 0.24	15.49 ± 0.23	14.91 ± 0.24	< 0.001
Magnesium (mg/day)	232.74 ± 5.01	230.75 ± 4.72	259.88 ± 4.97	< 0.001
Zinc (mg/day)	7.27 ± 0.15	7.15 ± 0.15	7.47 ± 0.15	0.33
Vitamin C (mg/day)	146.74 ± 5.40	134.28 ± 5.09	133.98 ± 5.35	0.17
Vitamin B12 (mg/day)	2.90 ± 0.11	2.85 ± 0.11	2.99 ± 0.11	0.70
¹Obtained from ANCOVA
All values are means ± standard error (SE); energy, protein, carbohydrate and fat are adjusted for age and gender; all other values are adjusted for age, gender and energy intake.

Table 3 presents the results of multivariable linear regression analyses examining variations in daily diet cost according to PHDI tertiles. After adjusting for energy intake, participants in PHDI tertiles showed a significant inverse relationship between PHDI score and daily diet cost. Specifically, for every 10-point increase in PHDI score, there was a corresponding decrease in daily diet cost by 343,870 Rials (95% CI: -495,299 to -192,442; p < 0.001). Across PHDI tertiles adjusted for energy intake, participants in the highest tertile had significantly lower daily diet costs (-698,638 Rials, 95% CI: -1,029,070 to -368,200; p < 0.001) compared to the reference tertile. In the fully adjusted model, this inverse relationship persisted. Participants in the third tertile had lower daily diet costs (-708,367 Rials, 95% CI: -1,060,371 to -356,362; p < 0.001) compared to the first tertile.

Table 3

Multivariable linear regression analyses estimating daily diet cost variations according to PHDI tertiles
			Daily Diet Cost
	n	Score (Range)	Mean (SEM¹)	β (95% CI)²	P value
PHDI
For every 10 points increase				-343,870 (-495,299, -192,442)	< 0.001
Tertiles (Energy adjusted)
T1	139	27 to 53	3,222,624 (118,696)	Reference
T2	121	54 to 60	2,813,249 (127,043)	-409,375 (-752,050, -66,700)	0.01
T3	138	61 to 82	2,523,986 (118,766)	-698,638 (-1,029,070, -368,200)	< 0.001
Tertiles (fully adjusted)³
T1	139	27 to 53	3,210,600 (123,375)	Reference
T2	121	54 to 60	2,830,272 (128,053)	-391,373 (-746,558, -36,188)	0.03
T3	138	61 to 82	2,521,171 (122,442)	-708,367 (-1,060,371, -356,362)	< 0.001
¹SEM = standard error of the mean (Obtained from ANCOVA)
²Obtained from linear regression
³Adjusted for age, gender, energy intake, physical activity, marital status, socio-economic status, smoking status, supplement intake and BMI.

Based on the adjusted mean (SE) values presented in Table 4, significant variations in daily diet costs across tertiles of the PHDI were observed within different sociodemographic subgroups. In the stratified analyses, inverse association between adherence to the PHDI and daily diet costs remained significant across various demographic subgroups. However, this inverse relationship did not maintain statistical significance within the single and low-income categories

Table 4

Adjusted means of daily diet cost according to PHDI tertiles stratified by demographic variables
		Food Cost
		Energy Adjusted		Fully Adjusted¹
Variable	n	Mean (SEM²)	P value³	Mean (SEM)	P value
Gender
Male			0.01		0.02
T1	83	3,371,454 (151,247)		3,367,805 (155,962)
T2	56	2,837,691 (181,787)		2,842,157 (183,826)
T3	60	2,763,595 (177,380)		2,764,475 (184,800)
Female			0.04		0.03
T1	56	2,982,995 (189,580)		3,006,682 (197,580)
T2	65	2,770,521 (176,697)		2,775,046 (180,461)
T3	78	2,371,401 (160,804)		2,350,624 (165,524)
Marital Status
Single			0.36		0.29
T1	18	3,012,722 (344,823)		3,015,409 (362,501)
T2	28	2,614,702 (275,783)		2,710,760 (279,472)
T3	35	2,407,038 (247,005)		2,328,810 (253,126)
Married			0.001		0.001
T1	121	3,248,149 (126613)		3275148 (131742)
T2	93	2,876,424 (143,927)		2,885,480 (145,130)
T3	103	2,567,355 (136,483)		2,527,460 (142,648)
Household Dimension
≥ 4			0.01		0.03
T1	26	3,856,476 (334,330)		3,892,108 (343,426)
T2	26	3,095,024 (334,880)		3,075,172 (352,858)
T3	31	2,440,224 (307,529)		2,426,989 (315,575)
< 4			0.01		0.01
T1	113	3,078,791 (122,369)		3,058,239 (128,066)
T2	95	2,736,165 (132,801)		2,746,470 (133,805)
T3	107	2,546,103 (125,149)		2,558,658 (130,471)
Education
Low			0.02		0.05
T1	88	3,161,447 (143,996)		3,133,423 (148,544)
T2	72	2,781,297 (160,166)		2,805,479 (163,009)
T3	74	2,576,464 (156,735)		2,586,261 (164,469)
High			0.009		0.01
T1	51	3,316,707 (206,381)		336,7079 (223,386)
T2	49	2,886,054 (209,987)		2,853,486 (215,312)
T3	64	2,452,659 (184,366)		2,437,454 (191,172)
Income
Low			0.43		0.18
T1	6	2,702,097 (425,212)		2,710,719 (417,505)
T2	7	2,553,562 (410,061)		2,731,973 (350,696)
T3	6	1,947,895 (430,082)		1,731,127 (364,764)
Moderate			0.001		0.001
T1	87	3,254,779 (147,385)		3,287,097 (155,168)
T2	85	2,592,080 (149,152)		2,594,427 (151,568)
T3	87	2,575,446 (146,504)		2,540,836 (153,796)
High			0.01		0.02
T1	46	3,237,821 (220,081)		3,137,395 (232,533)
T2	29	3,497,219 (276,684)		3,563,025 (286,898)
T3	45	2,510,389 (222,121)		2,570,638 (233,456)
Socioeconomic Status
Low			0.02		0.02
T1	42	3,053,676 (189,648)		3,015,490 (185,555)
T2	39	2,395,212 (198,488)		2,566,571 (195,777)
T3	36	2,410,841 (206,235)		2,269,751 (204,561)
Moderate			0.03		0.04
T1	61	3,284,200 (193,682)		3,288,265 (201,931)
T2	44	2,839,660 (228,240)		2,875,069 (233,744)
T3	60	2,569,959 (194,872)		2539860 (203421)
High			0.03		0.06
T1	36	3,309,410 (236,303)		3,279,312 (256,709)
T2	38	3,223,557 (227,296)		3,270,103 (235,944)
T3	42	2,549,697 (218,409)		2,533,382 (238,392)
¹Adjusted for age, gender, energy intake, physical activity, marital status, smoking status, supplement intake, BMI, socio-economic status with exception for the stratified variable.
²SEM = standard error of the mean
³Obtained from ANCOVA

The main aim of this study was to examine the relationship between adherence to the PHD) and dietary costs among older adults in Tehran, as well as to identify sociodemographic factors linked to these cost differences. Our findings show that higher PHDI scores are generally linked to lower dietary costs. However, this inverse relationship was not significant among single individuals and those with lower incomes. This suggests that while adherence to the Planetary Health Diet can be economically beneficial, these benefits may differ based on individual sociodemographic factors.

Most studies in the literature have found that healthier diets are generally more expensive. The first demonstration of a linear relationship between diet quality and cost was observed in the 1999 study in the UK Women’s Cohort Study. ³⁸. As the diet quality, assessed by the Healthy Diet Index, improved, the daily expenditure on food also increased. This positive relationship between diet quality and cost was reaffirmed in the same cohort of women in 2013 ³⁹. A systematic review of studies investigating the link between diet quality and cost found that healthier diets tend to be more costly ⁴⁰. However, it emphasizes that although potato chips, sweets, and biscuits are more costly per kilogram when priced in euros, compared to low-energy-density foods such as apples, tomatoes, and carrots, they are cheaper when evaluated in euros per 100 kcal. This difference arises because apples, tomatoes, and carrots contain more water compared to sweets and fats, as observed by Atwater in 1896.

Studies have demonstrated a direct correlation between higher nutrient density scores and increased cost per calorie ^41,42. Yet, there was significant variability in both nutrient density and cost within the different food groups ⁴³. For instance, it has been demonstrated that the cost of vegetables varied, with not all being equally expensive. Also, a meta-analysis of food prices collected after the year 2000 revealed a smaller gap of $1.5 (on average across 10 different countries) between the cost of the healthiest dietary patterns and the least healthy ones ⁶. In this study, when comparing nutrient-based patterns, the price per day did not show significant differences.

In contrast, our study aligns with a smaller body of research suggesting that it is possible to follow a healthy diet at a lower cost. One study demonstrated that the cost per calorie of a convenience diet was 24% higher than that of a healthy diet ⁴⁴. A Japanese study revealed that a higher monetary cost of dietary energy was linked to increased consumption of fat, oil, meat, and energy-containing beverages, while there was a decreased consumption of cereals ⁴⁵. Studies investigating the economic effects of enhanced diets have shown that improved diets can significantly reduce food expenditures ^46,47. This research demonstrated that the group that received nutrition education not only lowered their food costs but also boosted their intake of essential nutrients ⁴⁶. Furthermore, a study conducted in Spain found that women who were encouraged to adopt a Mediterranean-style diet in Canada did not report an increase in their food expenditures ⁴⁸.

One possible explanation for the inconsistent results could be differences in how food costs were measured across studies. A study examining the assumption that healthy diets are more expensive, compared prices of healthy and less healthy foods using three distinct metrics: the cost per food energy unit, the cost per edible weight unit, and the cost per average portion ⁴⁹. Except for the price of food energy, the study reveals that healthy foods are generally less expensive than less healthy options (defined as foods high in saturated fat, added sugar, and/or sodium, or having minimal nutritional value). Therefore, this research suggests that the choice of metric for assessing food prices significantly influences which foods appear more costly. Indeed, since the price per calorie fails to consider the quantity of food consumed, it is not an accurate proxy for actual out-of-pocket food expenses ^50,51.

Another factor contributing to the lower dietary cost associated with higher PHDI scores is the scoring system itself: consuming more meat results in a lower PHDI score, and while dairy intake is beneficial up to a certain threshold, excessive consumption can also reduce the score. In March 2020, fruits, red meat, and dairy products were identified as the top three contributors to the cost of a desirable food basket in Iranian society, accounting for more than half (56%) of its total cost ⁵². This suggests that replacing meat with plant proteins and limiting dairy consumption to some extent, as promoted by the PHDI, can result in significant reductions in overall dietary expenses. This hypothesis is supported by one study demonstrating that individuals who follow plant-based diets, especially vegans, incur lower food expenses compared to omnivorous individuals ⁵³.

By conducting subgroup analysis according to marital status and income, the association between PHDI and food cost was weakened, implying that socioeconomic and demographic factors may mediate the relationship between PHDI and food cost. Prior research has firmly established that socioeconomic status and demographic factors can influence the relationship between diet cost and diet quality ^4,5,54–57. In a study investigating the link between monetary value and diet quality, it was found that the positive correlation between the monetary value of the diet and dietary quality, was influenced by factors such as gender, race, and poverty status ⁵⁷.

One strength of our study is that it is the first to investigate the relationship between the Planetary Health Diet Index and dietary costs in Iranian population while also considering socioeconomic and demographic factors. Another key strength of this study is the use of the Planetary Health Diet Index (PHDI) to assess adherence to the EAT-Lancet dietary guidelines. This study's scoring system is incremental, setting it apart from other systems that rely on binary assessments ^58,59. Moreover, consumption data for 151 items from the FFQ were paired with both Iranian Statistics Center data and retail prices and the price data were aligned with the same period as the consumption data. However, there are some limitations worth noting. Firstly, the study utilized a cross-sectional analysis, which enables the identification of associations but does not allow for the establishment of causation. Secondly, food consumption was assessed using a Food Frequency Questionnaire (FFQ), which has inherent limitations, such as a restricted food list and the potential for dietary misreporting bias. Thirdly, we did not assess food consumed away from home, leading to potentially inaccurate food cost estimations. Our method assumes that all foods are purchased at retail prices and prepared at home, which may underestimate the true variability in food prices and the costs associated with individual food consumption. Also, variations in food costs due to home preparation, purchasing food away from home, differences between brand-name and generic or low-cost options, and seasonal price fluctuations are overlooked when using average prices. Also, we did not account for food waste, which could affect the accuracy of our calculations of food costs by potentially underestimating the true expenses associated with dietary habits ⁶⁰.

In conclusion, our study sheds light on the complex relationship between adherence to the PHDI, dietary costs, and socioeconomic factors among older adults in Tehran. We found that higher PHDI scores generally correlate with lower dietary costs, although this association was not significant among single individuals and those with lower incomes. This underscores the variability in economic benefits associated with adopting a Planetary Health Diet, influenced by individual sociodemographic characteristics. Future research should focus on longitudinal studies for assessing dietary behaviors to inform effective public health policies aimed at promoting affordable, sustainable and nutritious diets for all segments of the population.

SES, Socioeconomic status; FAO, Food and Agriculture Organization; FFQ, Food Frequency Questionnaire; PHDI, Planetary Health Diet Index; IPAQ, International Physical Activity Questionnaire; HEI, Health Eating Index

Acknowledgements

Authors’ contribution: LA, ADM and MKD designed the study. MKD, HA and PNH carried out the sampling. MKD, HA and PNH carried out interviews and imported the data for analysis. MKD performed the statistical analyses. MKD drafted the manuscript. LA and ADM read and commented on the manuscript. LA supervised the study. The final manuscript was approved by all the authors.

Disclosure Statement: None of the authors had any personal or financial conflicts of interest

Sources of support: This research was supported by Tehran University of Medical Sciences, Tehran, Iran (no: IR.TUMS.MEDICINE.REC.1401.588).

Data availability: Data supporting the findings are available from the corresponding author on reasonable request and under conditions that ensure the protection of participant privacy.

Conflicts of interest: Authors declared no personal or financial conflicts of interest.

Kickbusch, I., Allen, L. & Franz, C. The commercial determinants of health. The Lancet Global Health 4, e895-e896 (2016). https://doi.org/10.1016/S2214-109X(16)30217-0
Lee, J., Ralston, R. & Truby, H. Influence of food cost on diet quality and risk factors for chronic disease: A systematic review. Nutrition & Dietetics 68 (2011). https://doi.org/10.1111/j.1747-0080.2011.01554.x
Darmon, N. & Drewnowski, A. Contribution of food prices and diet cost to socioeconomic disparities in diet quality and health: a systematic review and analysis. Nutr Rev 73, 643-660 (2015). https://doi.org/10.1093/nutrit/nuv027
Monsivais, P. & Drewnowski, A. Lower-energy-density diets are associated with higher monetary costs per kilocalorie and are consumed by women of higher socioeconomic status. J Am Diet Assoc 109, 814-822 (2009). https://doi.org/10.1016/j.jada.2009.02.002
Pedroni, C., Castetbon, K., Desbouys, L., Rouche, M. & Vandevijvere, S. The Cost of Diets According to Nutritional Quality and Sociodemographic Characteristics: A Population-Based Assessment in Belgium. J Acad Nutr Diet 121, 2187-2200.e2184 (2021). https://doi.org/10.1016/j.jand.2021.05.024
Rao, M., Afshin, A., Singh, G. & Mozaffarian, D. Do healthier foods and diet patterns cost more than less healthy options? A systematic review and meta-analysis. BMJ Open 3, e004277 (2013). https://doi.org/10.1136/bmjopen-2013-004277
Palermo, C., McCartan, J., Kleve, S., Sinha, K. & Shiell, A. A longitudinal study of the cost of food in Victoria influenced by geography and nutritional quality. Aust N Z J Public Health 40, 270-273 (2016). https://doi.org/10.1111/1753-6405.12506
Clark, P. et al. A Healthy Diet Is Not More Expensive than Less Healthy Options: Cost-Analysis of Different Dietary Patterns in Mexican Children and Adolescents. Nutrients 13 (2021). https://doi.org/10.3390/nu13113871
Lee, A. J., Kane, S., Ramsey, R., Good, E. & Dick, M. Testing the price and affordability of healthy and current (unhealthy) diets and the potential impacts of policy change in Australia. BMC Public Health 16, 315 (2016). https://doi.org/10.1186/s12889-016-2996-y
Love, P. et al. Healthy Diets in Rural Victoria-Cheaper than Unhealthy Alternatives, Yet Unaffordable. Int J Environ Res Public Health 15 (2018). https://doi.org/10.3390/ijerph15112469
Andrea Carlson, E. F. Are Healthy Foods Really More Expensive? It Depends on How You Measure the Price. USDA-ERS Economic Information Bulletin 96 (18 Jan 2013).
Orr, J. B. Food, health and income. A report on a survey of adequacy of diet in relation to income. (1936).
Raffensperger, S. et al. Effect of Race and Predictors of Socioeconomic Status on Diet Quality in the HANDLS Study Sample. Journal of the National Medical Association 102, 923-930 (2010). https://doi.org/https://doi.org/10.1016/S0027-9684(15)30711-2
Alkerwi, A., Vernier, C., Sauvageot, N., Crichton, G. E. & Elias, M. F. Demographic and socioeconomic disparity in nutrition: application of a novel Correlated Component Regression approach. BMJ Open 5, e006814 (2015). https://doi.org/10.1136/bmjopen-2014-006814
Darmon, N. & Drewnowski, A. Does social class predict diet quality?1. The American Journal of Clinical Nutrition 87, 1107-1117 (2008). https://doi.org/https://doi.org/10.1093/ajcn/87.5.1107
Encalada-Torres, J. et al. Socioeconomic Status and Nutritional Status as Predictors of Food Insecurity in Older Adults: A Case Study from Southern Ecuador. Int J Environ Res Public Health 19 (2022). https://doi.org/10.3390/ijerph19095469
Nazri, N. S., Vanoh, D. & Leng, S. K. Malnutrition, low diet quality and its risk factors among older adults with low socio-economic status: a scoping review. Nutrition Research Reviews 34, 107-116 (2021). https://doi.org/10.1017/S0954422420000189
Atkins, J. L. et al. Diet quality in older age: the influence of childhood and adult socio-economic circumstances. Br J Nutr 113, 1441-1452 (2015). https://doi.org/10.1017/s0007114515000604
Host, A., Mcmahon, A.-T., Walton, K. & Charlton, K. ‘While we can, we will’: Exploring food choice and dietary behaviour amongst independent older Australians. Nutrition & Dietetics 73, 463-473 (2016). https://doi.org/https://doi.org/10.1111/1747-0080.12285
WHAM, C. A. & BOWDEN, J. A. Eating for health: Perspectives of older men who live alone. Nutrition & Dietetics 68, 221-226 (2011). https://doi.org/https://doi.org/10.1111/j.1747-0080.2011.01535.x
Kamphuis, C. B., de Bekker-Grob, E. W. & van Lenthe, F. J. Factors affecting food choices of older adults from high and low socioeconomic groups: a discrete choice experiment. Am J Clin Nutr 101, 768-774 (2015). https://doi.org/10.3945/ajcn.114.096776
FAO. Sustainable diets and biodiversity – directions and solutions for policy, research and action, <i3004e.pdf. http://www.fao.org/docrep/016/i3004e/i3004e.pdf> (2010).
Willett, W. et al. Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. Lancet 393, 447-492 (2019). https://doi.org/10.1016/s0140-6736(18)31788-4
Cambeses-Franco, C., Feijoo, G., Moreira, M. T. & González-García, S. Co-benefits of the EAT-Lancet diet for environmental protection in the framework of the Spanish dietary pattern. Sci Total Environ 836, 155683 (2022). https://doi.org/10.1016/j.scitotenv.2022.155683
Laine, J. E. et al. Co-benefits from sustainable dietary shifts for population and environmental health: an assessment from a large European cohort study. The Lancet Planetary Health 5, e786-e796 (2021). https://doi.org/10.1016/S2542-5196(21)00250-3
Tilman, D. & Clark, M. Global diets link environmental sustainability and human health. Nature 515, 518-522 (2014). https://doi.org/10.1038/nature13959
Baudry, J. et al. Associations between measures of socio-economic position and sustainable dietary patterns in the NutriNet-Santé study. Public Health Nutr 26, 965-975 (2023). https://doi.org/10.1017/s1368980022002208
Hirvonen, K., Bai, Y., Headey, D. & Masters, W. A. Affordability of the EAT-Lancet reference diet: a global analysis. Lancet Glob Health 8, e59-e66 (2020). https://doi.org/10.1016/s2214-109x(19)30447-4
Drewnowski, A. Analysing the affordability of the EAT–<em>Lancet</em> diet. The Lancet Global Health 8, e6-e7 (2020). https://doi.org/10.1016/S2214-109X(19)30502-9
Mirmiran, P., Esfahani, F. H., Mehrabi, Y., Hedayati, M. & Azizi, F. Reliability and relative validity of an FFQ for nutrients in the Tehran lipid and glucose study. Public health nutrition 13, 654-662 (2010).
Esfahani, F. H., Asghari, G., Mirmiran, P. & Azizi, F. Reproducibility and relative validity of food group intake in a food frequency questionnaire developed for the Tehran Lipid and Glucose Study. Journal of epidemiology 20, 150-158 (2010).
Ghafarpour, M., Houshiar-Rad, A., Kianfar, H. & Ghaffarpour, M. (Tehran: Keshavarzi Press, 1999).
Cacau, L. T. et al. Development and Validation of an Index Based on EAT-Lancet Recommendations: The Planetary Health Diet Index. Nutrients 13 (2021). https://doi.org/10.3390/nu13051698
<https://www.amar.org.ir/>
<https://www.cbi.ir/InflationCalculator/InflationCalculator_fa.aspx> (2024).
Craig, C. L. et al. International physical activity questionnaire: 12-country reliability and validity. Medicine & science in sports & exercise 35, 1381-1395 (2003).
Vasheghani-Farahani, A. et al. The Persian, last 7-day, long form of the International Physical Activity Questionnaire: translation and validation study. Asian journal of sports medicine 2, 106 (2011).
Cade, J., Upmeier, H., Calvert, C. & Greenwood, D. Costs of a healthy diet: analysis from the UK Women's Cohort Study. Public Health Nutr 2, 505-512 (1999). https://doi.org/10.1017/s1368980099000683
Morris, M. A., Hulme, C., Clarke, G. P., Edwards, K. L. & Cade, J. E. What is the cost of a healthy diet? Using diet data from the UK Women's Cohort Study. J Epidemiol Community Health 68, 1043-1049 (2014). https://doi.org/10.1136/jech-2014-204039
Darmon, N. & Drewnowski, A. Contribution of food prices and diet cost to socioeconomic disparities in diet quality and health: a systematic review and analysis. Nutrition Reviews 73, 643-660 (2015). https://doi.org/10.1093/nutrit/nuv027
Maillot, M., Darmon, N., Darmon, M., Lafay, L. & Drewnowski, A. Nutrient-dense food groups have high energy costs: an econometric approach to nutrient profiling. J Nutr 137, 1815-1820 (2007). https://doi.org/10.1093/jn/137.7.1815
Mendoza-Velázquez, A., Guzmán-Rodríguez, M., Lara-Arévalo, J. & Drewnowski, A. The Nutrient Rich Food Price Index: a nutrition-relevant adaptation of the Laspeyres price index to track the cost of affordable nutrient density. Frontiers in Nutrition 10 (2023). https://doi.org/10.3389/fnut.2023.1107573
Darmon, N., Darmon, M., Maillot, M. & Drewnowski, A. A nutrient density standard for vegetables and fruits: nutrients per calorie and nutrients per unit cost. J Am Diet Assoc 105, 1881-1887 (2005). https://doi.org/10.1016/j.jada.2005.09.005
McDermott, A. J. & Stephens, M. B. Cost of eating: whole foods versus convenience foods in a low-income model. Fam Med 42, 280-284 (2010).
Murakami, K. et al. Monetary costs of dietary energy reported by young Japanese women: association with food and nutrient intake and body mass index. Public Health Nutr 10, 1430-1439 (2007). https://doi.org/10.1017/s1368980007000213
Burney, J. & Haughton, B. EFNEP: a nutrition education program that demonstrates cost-benefit. J Am Diet Assoc 102, 39-45 (2002). https://doi.org/10.1016/s0002-8223(02)90014-3
Ottelin, A. M. et al. Costs of a self-selected, health-promoting diet among the participants of the Finnish Diabetes Prevention Study. Diabetes Care 30, 1275-1277 (2007). https://doi.org/10.2337/dc06-2444
Goulet, J., Lamarche, B. & Lemieux, S. A nutritional intervention promoting a Mediterranean food pattern does not affect total daily dietary cost in North American women in free-living conditions. J Nutr 138, 54-59 (2008). https://doi.org/10.1093/jn/138.1.54
Carlson, A. a. F., Elizabeth. Are Healthy Foods Really More Expensive? It Depends on How You Measure the Price. USDA-ERS Economic Information Bulletin 96
Frazao, E., Carlson, A. & Stewart, H. Energy-adjusted food costs make little economic sense. The American journal of clinical nutrition 93, 861; author reply 862-863 (2011). https://doi.org/10.3945/ajcn.110.009639
Frazao, E. Less-energy-dense diets of low-income women in California are associated with higher energy-adjusted costs but not with higher daily diet costs. The American Journal of Clinical Nutrition 90, 701 (2009). https://doi.org/10.3945/ajcn.2009.27960
Hojaji, E., Sadeghian, S. & Pouraram, H. Cost Estimating of Providing Desired Food Basket for the Iranian Society and Its Situation in the Last ۳ Years. (1401).
Pais, D. F., Marques, A. C. & Fuinhas, J. A. The cost of healthier and more sustainable food choices: Do plant-based consumers spend more on food? Agricultural and Food Economics 10, 18 (2022). https://doi.org/10.1186/s40100-022-00224-9
Schröder, H. et al. Monetary Diet Cost, Diet Quality, and Parental Socioeconomic Status in Spanish Youth. PLoS One 11, e0161422 (2016). https://doi.org/10.1371/journal.pone.0161422
Tong, T. Y. N. et al. Dietary cost associated with adherence to the Mediterranean diet, and its variation by socio-economic factors in the UK Fenland Study. Br J Nutr 119, 685-694 (2018). https://doi.org/10.1017/s0007114517003993
Okubo, H., Murakami, K. & Sasaki, S. Monetary value of self-reported diets and associations with sociodemographic characteristics and dietary intake among Japanese adults: analysis of nationally representative surveys. Public Health Nutr 19, 3306-3318 (2016). https://doi.org/10.1017/s1368980016001695
Beydoun, M. A. et al. Monetary Value of Diet Is Associated with Dietary Quality and Nutrient Adequacy among Urban Adults, Differentially by Sex, Race and Poverty Status. PLoS One 10, e0140905 (2015). https://doi.org/10.1371/journal.pone.0140905
Knuppel, A., Papier, K., Key, T. J. & Travis, R. C. EAT-Lancet score and major health outcomes: the EPIC-Oxford study. Lancet 394, 213-214 (2019). https://doi.org/10.1016/s0140-6736(19)31236-x
Trijsburg, L. et al. Method for the Development of WISH, a Globally Applicable Index for Healthy Diets from Sustainable Food Systems. Nutrients 13 (2020). https://doi.org/10.3390/nu13010093
Conrad, Z. Daily cost of consumer food wasted, inedible, and consumed in the United States, 2001–2016. Nutrition Journal 19, 35 (2020). https://doi.org/10.1186/s12937-020-00552-w

No competing interests reported.

Download PDF

Editor assigned by journal
11 Nov, 2024
Editor invited by journal
24 Aug, 2024
Submission checks completed at journal
23 Aug, 2024
First submitted to journal
15 Aug, 2024

You are reading this latest preprint version

Planetary Health Diet adherence in relation to dietary cost and its variation by socio-economic factors among Iranian older adults

Status:

Version 1

Abstract

Introduction

Methods

Study population and design

Dietary data

Calculation of PHDI

Food cost

Sociodemographic characteristics

Assessment of other variables

Statistical analysis

Result

Discussion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1