The Use of Dietary Supplements and Their Association with COVID-19-Related Anxiety among Non-Institutionalized Elderly in Northern Greece

Abstract The elderly constitute a vulnerable group for increased anxiety and poor diet during the COVID-19 pandemic. There is limited research on the levels of COVID-19-related anxiety and dietary habits including dietary supplementation practices among the elderly and very elderly in Greece. A cross-sectional survey was conducted in 364 non-institutionalized elderly (65–74 y) and very elderly (≥75 y) living in northern Greece, with the aim to investigate the use of dietary supplements and their association with other factors, particularly the COVID-19-related anxiety. Levels of anxiety were assessed with the use of the Coronavirus Anxiety Scale (CAS-5). The percentage of dietary supplement users was 62.6%. The most popular dietary supplements used were vitamin D followed by vitamin C and multivitamin and mineral supplements. Multivariate analysis showed that the very elderly and overweight individuals were less likely to consume vitamin D supplements. Approximately a third of the participants (33.8%) exhibited signs of COVID-19-related anxiety but only 8% showed dysfunctional levels of anxiety. Regression analysis indicated that women, former smokers, and people exhibiting any sign of COVID-19 anxiety were approximately two times more likely to consume dietary supplements of any kind (Gender: OR 2.34, 95% CI 1.30–4.19; Smoking: OR 2.15, 95% CI: 1.08–4.26; COVID-19 anxiety: OR 2.16, 95 % CI: 1.20–3.91). Our results provide useful insights into the current practices of dietary supplement use in this population group and could be used by dietetic and medical associations as well as public authorities in the formulation of targeted, safe, and effective interventions for the protection of public health.


Introduction
The COVID-19 pandemic has been reported to adversely impact mental health (1) particularly of vulnerable groups, such as the elderly (2)(3)(4). Studies comparing prewith peri-COVID-19 pandemic data have revealed that older adults experience more severe depressive, anxiety and stress symptoms during the COVID-19 pandemic (4)(5)(6). It is well established that advanced age has been identified as a risk factor for severe disease and increased COVID-19 mortality mainly due to age-related immune senescence and higher prevalence of chronic diseases, which can in turn contribute to increased anxiety in the elderly. Anxiety may in turn, result in higher risk of cardiovascular and cerebrovascular disease as well as other debilitating physical manifestations including headache, sleep disturbance, uncontrolled trembling, muscle tension, and aches (7,8).
In addition to increased anxiety, the COVID-19 pandemic has been associated with poor diet quality (2). This may have serious consequences in the elderly, since acute and chronic diseases, medication use (9), poor oral health (10), as well as diminished absorption and utilization of nutrients concomitant with aging or medication use (11), often result in suboptimal or inadequate nutrient intakes (12). While the energy needs of older adults decrease with advancing age and reduced physical activity, the requirement for essential nutrients, including calcium and vitamins D and B6 becomes more prominent (13). Previous research in elderly subjects (65-80 years) in the Netherlands has shown that the percentages of participants not meeting their average requirement were considerable for essential nutrients such as vitamin D and selenium, even when vitamin and mineral supplement use by this group was taken into account (14). Data from the most recent national dietary survey in Greece also confirm that a substantial percentage of Greek elderly subjects have micronutrient (such as vitamins A, D, E, K, C, and folate as well as calcium, magnesium, potassium, and zinc) intakes which are less than the estimated average requirement (15). Although deficiency and low levels of vitamin B12 are frequently observed among the elderly and constitute an important public health issue globally (16), there is lack of robust data on vitamin B12 status among the elderly in Greece using appropriate biomarkers.
The fear of a new disease and the fear of the inadequacy of the diet to meet nutritional needs for optimal immunity, may lead to increased use of food supplements (17) in an attempt to boost their immune system, prevent illness and hence reduce anxiety related to COVID-19. Reviews published in the scientific literature have suggested early nutritional supplementation with certain immunonutrients (such as zinc, selenium, and vitamin D) as a means for raising anti-viral resistance against progressive COVID-19 (18,19). In addition, a systematic review (8) concluded that nutritional and herbal supplementation could be an effective method for treating anxiety and anxiety-related conditions without the risk of serious adverse effects. Knowledge of dietary supplement use and level of COVID-19-related anxiety among the elderly in Greece during the COVID-19 pandemic is sparse. A recent Greek study that was conducted during the COVID-19 pandemic showed that 19% among 2258 participants 17 years and older have started or increased the frequency of taking dietary supplements that enhance the immune system (20). However, no studies have focused on dietary supplement use among the elderly and very elderly in Greece during the COVID-19 pandemic. Access to the results of such studies would provide valuable insights for public health professionals with regard to how the pandemic has affected the psychology and dietary behavior of this vulnerable and steadily-growing demographic group.
The aim of this study was to examine dietary supplement use during the COVID-19 pandemic among non-institutionalized elderly (including the very elderly i.e. ≥75 years old) individuals, evaluate their COVID-19 anxiety and assess whether there is any association between the use of supplements and covid-related anxiety or other behavioral and demographic factors. We focused on non-institutionalized elderly since those confined in elderly homes or other types of institutions in Greece are generally expected to have less or no control over food choices or use of dietary supplements. In addition, during the pandemic it was not easy to conduct surveys in such institutions.

Study design and participants
This was a cross-sectional study that was carried out in non-institutionalized elderly individuals aged 65 years and over, residing in urban areas of north Greece. Data collection started in October 2021 and was completed in January 2022. No exclusion criteria were applied, except that the participants had to be fluent in Greek in order to avoid any misunderstanding (i.e. immigrants were not included). No financial or any other incentive was offered to complete the survey. Elderly people were recruited outside various stores such as food stores and pharmacies. Four researchers (DM, EV, GKo, and LK) assisted the participants in answering the electronic questionnaire using tablets (i.e. read out the questions if they were any eyesight problems, check for any missing answers or clear out any confusion for those puzzled about the differences between dietary supplements and medicines), following a short explanation of the study and obtaining their consent to participate. Four hundred elderly were asked to participate, and 364 subjects agreed to complete the questionnaire (response rate: 91%).
Participation in this survey was voluntary and anonymous. All procedures carried out in this study were in line with Helsinki's statement. The study protocol was registered and approved by the Ethics Committee of the Department of Nursing of the University of Thessaly (Larissa 27-07-2021/Protocol no. 656).

Questionnaire
Data were collected with the use of a questionnaire that included questions related to (a) personal/demographic information, (b) dietary supplements use, and (c) covid-related anxiety using a previously validated Covid anxiety scale mini questionnaire (CAS-5) (21). The questionnaire was created electronically using Google Forms in the Greek language.
The part of the questionnaire that addressed demographic and personal information included information on age, gender, self-reported weight and height, smoking habit, type of diet followed (e.g. mixed diet, vegetarian, low in carbohydrates, etc.) and perception of their diet (i.e. balanced vs. unbalanced), self-evaluation of immune status, and vaccination status. Weight and height were used to calculate Body Mass Index (BMI) defined as the weight in kilograms divided by the square of his height in meters (kg/m 2 ). Classification of BMI was done according to the World Health Organization cutoff points for adults aged 18 years and over (22).
The questions on dietary supplements intake were developed by the researchers, based on previous experience (23). Initially, the questionnaire was circulated among experts (n = 6) for comments on the content and was tested before the collection of data for validity, presentation and formulation of questions as well as for the clarity of the instructions. Internal consistency of the specific questions related to supplement use was assessed using Cronbach's alpha statistic. Cronbach's alpha was 0.799 for questions related to dietary supplements use. A value above 0.7 is generally considered reliable. Covid-related anxiety was assessed with the use of a short, unidimensional 5-item scale, namely Coronavirus anxiety scale (CAS-5) (21) translated into Greek by the Medical School of the University of Patras, Greece (24). To ascertain the accuracy of translation, the questionnaire was also translated into Greek by a person fluent in English and then translated back into English by a different person who was also fluent in English. This scale was chosen because it is short, enabling easier administration and minimizing drop-out rates (25). Data from online surveys in 25 countries in 6 continents (26), in which 10,232 people aged 20-69 years old participated, has also indicated that CAS-5 has cross-cultural reliability. The participants were asked to indicate the frequency of certain physiological symptoms related to fear and anxiety (dizziness, sleep disturbances, tonic immobility, appetite loss, abdominal stress) over the last 2 weeks, on a 5-point Likert scale. (How often have you experienced any of the following in the past 2 weeks? 0 = not at all; 1 = rare, less than a day or two; 2 = several days; 3 = more than 7 days; 4 = nearly every day over the last 2 weeks). Patients with CAS scores equal to or greater than five were considered to be dysfunctionally anxious (27). Scores obtained were ex-post-facto (i.e. levels of anxiety had already been impacted by other factors prior to the moment of data collection, such as prior infection with COVID-19 or death in the family because of COVID-19) (26). Although this scale has its own limitations and does not encompass parameters such as personality or coping style behaviors, CAS has previously been found useful to assess the psychological aftermath of the pandemic in time with high internal consistency (α = 0.93). In our study, the internal consistency reliability using Cronbach's alpha was estimated to be slightly lower (α = 0.802) but acceptable.

Statistical analysis
Data analysis was performed with SPSS Version 23.0. Descriptive analysis was performed by calculating absolute and relative frequencies (N, %). For the univariate analysis of association between supplement consumption and demographics, perceived health and nutritional characteristics, chi-square tests were calculated. Binary logistic regression analysis was used to examine the multivariate associations between nutritional supplement consumption and independent variables (gender, age, BMI, smoking, perception of supplements necessary during the pandemic, COVID-19 anxiety, COVID-19 vaccination, following a mixed diet, having a balanced nutrition, perception of immune system stronger compared to others of same age). Post-hoc analyses to examine the association of demographics and health/nutritional characteristics to the most popular nutritional supplements, as reported by the sample of this study, were also performed with logistic regressions.

Results
The demographic, health and nutritional characteristics of the participants are depicted in Table 1. Women accounted for 56.6% of the sample. About half of the participants (51.1%) belonged to the age group 65-74 years, with the rest being 75 years of age and older. Based on their BMI values, nearly half of the participants (48.9%) were overweight and 28.0% were obese. With regard to smoking habits, 17.9% of participants reported being active smokers, 25.8% former smokers, and 56.3% were nonsmokers. About a third of the participants (33.8%) exhibited any sign of COVID-19-related anxiety (i.e. CAS-5 > 0), but only 8% were dysfunctionally anxious (i.e. CAS ≥ 5). Most of the participants (87.64%) were vaccinated against COVID-19. The vast majority (92.6%) followed a normal mixed diet that includes all food groups and only 0.8% reported being vegetarian. A small percentage (3.8%) followed a low-carbohydrate diet and the rest (2.8%) indicated other types of diet such as low-calorie diet. Most of the participants (80.5%) perceived their diet as balanced, while only 12.1% perceived their diet as unbalanced, with the rest being unable to evaluate it. Finally, 29.1% perceived their immune system as stronger compared to that of same-age peers, while only 7.1% believed that their immune system was weaker compared to that of same-age peers (with the rest of the participants perceiving that their immune system is the same as that of same-age peers). About a third of the participants (33.2%) believed that dietary supplements are not necessary during the COVID-19 pandemic. Even though 34.3% of the participants considered dietary supplements as necessary during the pandemic, 62.6% (N = 228) of participants reported taking nutritional supplements during the pandemic. Among supplement users, the majority (84.8%) reported using dietary supplements before the COVID-19 pandemic, whereas only 15.2% stated that they started consuming dietary supplements during the pandemic.
Multivariate analysis of the association of supplement consumption with demographic, health, and nutritional characteristics ( As presented in Table 3, the majority of supplement users reported always taking the recommended dose as indicated on the label (77.6%), whereas only eight individuals (3.5% among users) reported taking higher than recommended doses. Moreover, most of the participants purchased the supplements from a pharmacy store (77.6%) (only 5.7% of users from the Internet) and consulted a health professional before taking them (73.6%). Internet and social media were reported by 7.5% of users (i.e. 17 individuals) in this age group as sources of information before taking dietary supplements. Interestingly, more traditional sources of nutrition information such as TV, radio, newspaper, and magazines were only reported by 3.5% of users (i.e. eight individuals) as a source of information that could guide them in using dietary supplements.
Multivariate analyses for vitamin D, C and multivitamin/mineral consumption's association with demographic, health, and nutritional characteristics are presented  Multivariate analysis for the association of B-complex vitamins, magnesium and fish oil consumption with demographic, health, and nutritional characteristics are presented in Table 5. B-complex vitamin consumption was associated only with the perception of nutritional supplements as necessary during the pandemic (OR 2.17, 95% CI: 1.13-4.16). Magnesium consumption was associated with a positive view of the potential contribution of nutritional supplements during the pandemic, not following a mixed diet and having a strong immune system (Supplements necessary: OR 0.28, 95% CI: 0.10-0.75; Mixed diet: OR 0.30, 95% CI 0.11-0.81; Immune system: OR 2.34, 95% CI: 1.22-4.49). In the same frame, fish oil consumption was associated with a positive view of the potential contribution of nutritional supplements during the pandemic, not following a mixed diet and having a strong immune system (Supplements necessary: OR 0.20, 95% CI: 0.06-0.63; Mixed diet: OR 0.29, 95% CI 0.11-0.79; Immune system: OR 2.23, 95% CI: 1.15-4.32).

0.017
Bold values denotes statistically significant values.

Discussion
Higher age is a risk factor for both increased anxiety and nutrient deficiencies (28). Although there is no direct evidence to support the link between diet and prevention of COVID-19 infection and severe disease (29), the importance of optimal nutrition (e.g. adherence to Mediterranean dietary model (30)), achievement of optimal body weight (31), reduced salt intake (32), and appropriate dietary supplementation (e.g. with micronutrients, ω-3 fatty acids, probiotics, and/or phytochemicals) (33,34), can be inferred as a means to reduce the risk of COVID-19 infection and subsequent inflammation, oxidation and cytokine storm (35). Our study investigated the prevalence of dietary supplement use and its potential determinants, including COVID-19related anxiety, during the COVID-19 pandemic in 364 free-living elderly and very elderly individuals residing in northern Greece. A substantial percentage (62.6%) of individuals reported use of dietary supplements, with the prevalence being higher among women compared to men which is in line with the findings of previous studies (36). However, there was no correlation between gender and use of individual dietary supplements. The use of dietary supplements of any kind was positively associated with signs of COVID-19 related anxiety, as well as with a perceived necessity of dietary supplementation during the COVID-19 pandemic, thus implying that the use of supplements by this population group is probably not random but conscious and thoughtful.
The percentage of individuals consuming dietary supplements in Greece was found to be higher than previous studies conducted in Greek elderly individuals. The EPIC study showed that only 2.1% of those aged 65-74 years reported use of dietary supplement (37). More recently, among individuals aged 25-70 years who participated in the Epirus Health Study (38), 31.4% reported dietary supplement use. The authors of the Epirus Health study also observed a suggestive increase in dietary supplement use after the onset of the COVID-19 pandemic in men, which is in agreement with the relative widespread use of dietary supplements seen in our study. Similarly, other studies have also provided some evidence that the use of dietary supplements has increased during the COVID-19 pandemic (20,39,40), suggesting that individuals tend to resort to dietary supplementation as a complementary approach to boost their immune system in the fight against COVID-19 and improve their diet and quality of life. The fact that the majority of participants obtain the supplements from pharmacies and consult a doctor/dietitian shows that the level of awareness and involvement of health care professionals as sources of information about dietary supplements has increased in Greece compared to that reported a decade ago (41).

Types of dietary supplements used
The most popular dietary supplement used was vitamin D, with 35.7% of supplement users reporting its use. This was an unexpected finding since the Hellenic National Nutrition and Health Survey which was carried out in 2013-2015, showed that the vast majority of those ≥ 71 years old (98.4% of males and 92.9% of females) reported that they do not take any vitamin D supplement (42), while ninety nine percent of the Greek adult population (99.6% for males and 99.9% for females) had intake levels below the EAR (42). Our finding indicates a heightened awareness about the potential benefits of vitamin D during the pandemic as an easy-to-use, low-cost adjunct remedy in the fight against COVID-19. Vitamin D has long been known to participate in the activity of the immune system (both in terms of innate and acquired immunity) through its active form, 1-25 dihydroxycalciferol (calcitriol) and directs the immune response toward a reduction in the production of pro-inflammatory cytokines, while favoring the differentiation of T cells toward the production of subtypes that significantly attenuate the extent of inflammation (32). Hence vitamin D may prevent cytokine storms and reduce mortality (29). In fact, vitamin D has been identified among the three most important nutritional factors with high impact in this pandemic, suggesting the need for screening for potential inadequate vitamin D status in subjects at higher mortality risk from COVID-19 (32). Although participants in our study were not asked whether they had previously been infected with COVID-19 and how severe the infection was, those who used vitamin D supplements perceived their immune system as stronger compared to that of their same-age peers.
Interestingly, the very elderly (>75 y) as well as the overweight and obese individuals who are at greater risk of vitamin D deficiency and who might benefit more from vitamin D supplementation (43,44), were the least likely ones to consume vitamin D supplements in our study. Since most participants consult a health professional (e.g. doctor, dietitian) before taking the supplements and purchase the supplements from pharmacies, it is important that health care professionals evaluate vitamin D status in these individuals and recommend appropriate supplementation, if necessary.
Vitamin C was the second most popular supplement taken by elderly. Despite the evidence of beneficial effects of vitamin C supplementation for the common cold (45), a recent review concluded that the available evidence cannot recommend routine oral or intravenous vitamin C in COVID-19 (29). However, scientists have suggested that vitamin C in lower doses may help with COVID-19 prophylaxis and in higher doses may serve as adjunctive therapeutic treatment for COVID-19 (19). Interestingly, the two most commonly-used dietary supplements in our sample, are the two vitamins (C and D) whose relationship and importance for a well-functioning immune system are best documented (28), even though the evidence for their use in COVID-19 is still regarded as weak (29).
Apart from supplements of vitamins D and C, supplements of other individual micronutrients with immunomodulatory properties, especially zinc and selenium, did not appear to be popular and were used only by few participants. Zn deficiency has been implicated in smell and taste disorders related to COVID-19 (29), while zinc supplementation may shorten the course of viral illnesses because Zn can directly modulate antiviral immunity and block viral replication inside host cells (19,46). Supplementation with Se is also believed to prevent or relieve clinical complications associated with Covid-19 especially in elderly populations via multiple mechanisms (19). The evidence is mainly coming from observational studies which show that Se suboptimal levels may be a risk factor for COVID-19 infection and mortality (29). Alexander et al. (18) recognized that early intervention and adequate levels and availability of Zn, Se, and vitamin D are essential for resistance to viral infections and enhanced immune function, thereby protecting against SARS-CoV-2, and modifying the progression of COVID-19. It is possible that some participants were advised by their health professionals that a multivitamin and mineral supplement would be sufficient to cover their needs without the potential risk of overdosing, bearing in mind that there is no data suggesting that micronutrients at recommended doses would do any harm in COVID-19 (28). In general, very few individuals (3.5% among supplement users) reported exceeding the recommended supplement dose. This would allay concerns of the potential toxicity of large amounts of some micronutrients (47) and potential interactions between them and prescription or over-the-counter drugs (48).
In addition to supplementation with immunomodulatory micronutrients (vitamins C & D, Se & Zn), Berger et al. (49) suggested that during the COVID-19 pandemic, it may be prudent to consider supplementation with moderate doses of n-3 PUFA, in risk groups as a complement to a healthy and balanced diet. In our study, only 16% of supplement users reported taking fish oil supplements which was positively correlated with the self-perception of a strong immune system. So far, the exact type, amount, and duration of n-3 fatty acid supplementation needed as well as the patient populations that will benefit more from fish oil supplements with respect to COVID-19 and its complications such as CVD, remains to be established in further clinical studies (29).

Anxiety and dietary supplements
The COVID-19 pandemic has been associated with increased anxiety, particularly among the older adults (2)(3)(4). A previous study in Greece (50) indicated that increased anxiety was present during the pandemic in more than 45% of the sample. In our study, COVID-19 anxiety was evaluated with the Coronavirus Anxiety Scale (CAS-5) which showed that although a third of the participants exhibited some signs of COVID-19 anxiety, only 8% exhibited dysfunctional anxiety. The low level of COVID-19 related anxiety could be due to the fact that curfews and quarantine measures were lifted during the period of data collection. Furthermore, lower anxiety may be attributed to the fact that the majority of participants were vaccinated, even though in some cases, uncertainty shrouding the COVID-19 vaccine may as well contribute to increased anxiety. In line with our findings, a cross-sectional study in Greece and Spain during the COVID-19 pandemic showed that anxiety symptoms were also less prevalent in Greek compared to Spanish participants (51), possibly due to cultural differences.
In pro-covid period, a vegetarian type of diet appeared to be associated with less anxiety among Greek adult females according to Attica study (52). However, only 0.8% of our sample followed a vegetarian diet and therefore the type of diet was not further assessed as a confounding factor in the evaluation of the association between anxiety and supplement use. However, any association between anxiety levels and supplement use has not been previously evaluated. In our study, the use of dietary supplements was positively associated with signs of COVID-19 anxiety in the elderly. Since some participants started taking supplements only after the pandemic, it is reasonable to suggest that many supplements were taken in an effort to improve nutrition and immune status by those who suffered from higher levels of anxiety related to the pandemic, although a cause-and-effect relationship cannot be established. It is noteworthy that this positive correlation was not evident for individual dietary supplements, possibly due to the relatively small number of study participants consuming specific supplements. A systematic review (8) concluded that nutritional and herbal supplementation appears to be an effective method for treating anxiety and anxiety-related conditions, without the risk of serious side effects. Although magnesium-containing supplements appear to hold promise in managing stress caused by COVID-19 and post-traumatic stress disorder in survivors (19), more evidence from large-scale studies is needed before Mg supplements can be recommended to patients.
So far, there are no specific official recommendations regarding dietary supplementation strategies for the elderly and very elderly. This is possibly due to the lack of evidence from high-quality studies with regard to the necessity and the optimal and effective dosage of essential micro-and macro-nutrients as well as non-essential dietary components to protect or alleviate symptoms against COVID-19 (53) as well as improve the wellbeing and quality of life of these individuals. In addition, the risk of inadequate vitamin D status highlights this nutrient as a prime subject for screening of the elderly, so that more people can benefit from the potential health benefits of its supplementation, as suggested in the article by Iaccarino Idelson et al. (18).

Limitations of the study
This study was regional and hence a more geographically representative study is needed to confirm our findings. There might have been sample bias due to the recruitment method used. Data were obtained through self-reporting, which may be subject to recall bias and/or social desirability bias, thus potentially resulting in misreporting the use of dietary supplements. Information on the brand name or the amount of nutrients present in the dietary supplements used was not collected since the questionnaire was completed out of home and many elderly would be unable to recall such information, and this would result in potentially inaccurate estimates. However, such information is valuable if future studies are conducted via in-home visits since it would provide intake data of vitamins, minerals, or other substances via dietary supplementation for use in exposure assessment of nutrients and phytochemicals. Food consumption data or data from biomarkers (e.g. blood levels of certain vitamins) were not obtained that could substantiate the need for dietary supplement use.
We did not collect detailed data on any prescribed medication since the aim of the study was not to evaluate any interaction between prescription medicines and dietary supplements. The potential risks of overconsumption and interactions with prescription and/or over-the-counter medications need to be analyzed in future studies. Finally, this survey was cross-sectional and therefore questions about causality could not be addressed.

Conclusions
Despite the lack of robust scientific evidence and/or guidelines from national and international scientific bodies regarding the benefits or risks of the use of dietary supplements among the elderly during the pandemic, a number of scientific articles have encouraged the use of dietary supplements as a cost-effective strategy to reduce the burden of Covid-19 particularly for the vulnerable elderly population who suffer from a disproportionate burden of morbidity and mortality. In our study, a widespread use of dietary supplement was observed among free-living elderly individuals living in northern Greece, particularly in women. However, our sample did not exhibit high levels of COVID-19 anxiety. Nevertheless, signs of COVID-19 related anxiety appeared to be related to greater dietary supplement use in general, but were not associated with the use of any specific dietary supplement. Although the design of the study could not establish a cause-and-effect relationship, our results provide useful insights into the current practices of dietary supplement use in this population group and could be used by dietetic and medical associations as well as public authorities in the formulation of targeted, safe and effective interventions and communication strategies for proper use of dietary supplements and neutraceuticals. Future studies could investigate the use of supplements by the elderly suffering from long COVID. However, regardless of their history of infection with COVID-19, evaluation of the nutritional status of the study participants would greatly enhance similar studies.