Association of food habit with the COVID-19 severity and hospitalization: A cross-sectional study among the recovered individuals in Bangladesh

Abstract

Background: It was assumed that dietary habits might influence the status of COVID-19 patients. Aim: We aimed at the identification of association of dietary habits with the COVID-19 severity and hospitalization. Methods: It was a retrospective cross-sectional study (n = 1025). We used bivariate and multivariate analyses to correlate the association between self-reported dietary patterns and COVID-19 severity and hospitalization. Results: Dietary habits (black tea, milked tea, pickles, black caraway seeds, honey, fish, fruits, vegetables, garlic, onion and turmeric) were identified with lower risk of COVID-19 severity and hospitalization. Interestingly, the consumption frequency (one-, two- or three-times/day) of rice - the staple food in Bangladesh - was not associated with COVID-19 severity and hospitalization for comorbid patients. In contrast, a moderate rice-eating habit (two times/day) was strongly associated with the lower risk of severity and hospitalization for non-comorbid patients. However, for both comorbid and non-comorbid patients, consumption of black tea, milked tea, pickles and honey were associated with a lower likelihood of severity and hospitalization. Overall, a high consumption (three-times/day) of fish, fruits and vegetables, a moderate consumption of garlic, onion and turmeric spices and a daily intake of black/milked tea, and honey were associated with reduced risk of COVID-19 severity and hospitalization. Conclusions: To reduce the severity of COVID-19, a habitual practice of intaking black tea, milked tea, black caraway seeds and honey along with dietary habit (rice, fish and vegetables) and with a moderate consumption of ginger, garlic, onion, mixed aromatic spices (cinnamon + cardamom + cloves) and turmeric might be suggested.

Introduction

COVID-19 – a disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2 - has been a buzzing word across the world posing a severe stress to the global healthcare system. Since the first report in December 2019 in Wuhan, China, COVID-19 has caused over 265 million cases and over 5.24 million deaths globally as of 4^th December 2021 (Andersen et al., 2020; WHO, 2021). On the other hand, an unprecedented speed in developing COVID-19 vaccines around the world resulted in a handful of successful vaccine candidates; however, some big concerns still remain such as lasting of vaccine-imparted immunity, vaccine effectiveness against new variant or even facing more infectious variant and new pandemic (Dodd et al., 2021; Forni et al., 2021; Lai et al., 2021; Li et al., 2021). Additionally, the solid evidence of vaccine in terms of longer-term effectiveness, safety, and protection against severe COVID-19 is still missing (Krause et al., 2020). Of note, Delta variant is the most common variants available recently infected people in Bangladesh (Ghosh et al., 2021; Moona et al., 2021). Recently, a new variant designated as Omicron has surged over the world while Bangladesh is experiencing a rapid spreading of third wave from January 2022 (TBS, 2022).

So far, scientists and physicians have noticed that people exposed to SARS-CoV-2 are not equally infected with COVID-19 disease (asymptomatic to mild to severe to death) (Kim et al., 2020; Shi et al., 2020). Notable that, several factors have been found to be associated with the severity and mortality rate which may increase the risk in COVID-19 patients, including aged person (>60 years) and having comorbidities (Akbar and Gilroy, 2020; Martins-Filho et al., 2020; Richardson et al., 2020). However, elderly-aged and comorbidities are not enough to explain the vulnerability to this severe infection. Additionally, pre-existing dietary habits of people appear to be linked to variability in COVID-19 symptoms (Butler and Barrientos, 2020; Hull et al., 2020; Whittemore, 2020; Zbinden-Foncea et al., 2020). Nevertheless, it is poorly known exactly why some people are more vulnerable whilst others less.

It is obvious that balanced nutritional status of the individuals can fight against any disease. Recently, Alam et al. (2021) has carried out a comprehensive review where strong nutritional interventions serve as a therapeutic tool against COVID-19 (Alam et al., 2021). Abdulah and Hassan (2020) explored the relation between dietary factors and infection/mortality rates globally; briefly, higher intake of fruits/sugar-sweetened products and lower intake of legumes/beans had a positive and negative effect on infection and mortality rates (Abdulah and Hassan, 2020). Several other review and original articles forecasted that nutritional management/supplement/interventions could be considered as a therapeutic gun against pandemic (Brugliera et al., 2020; Cobre et al., 2021; Grant et al., 2020; Iddir et al., 2020; Lidoriki et al., 2020; Liu et al., 2020; Moscatelli et al., 2021; Singh et al., 2020). Particularly, very recently, Kim et al. (2021), Salazar-Robles et al. (2021), Tavakol et al. (2021), and Merino et al. (2021) reported that increased level of physical activity and some dietary patterns such as consuming plant-based diets, fruits and poultry were associated with less severity of COVID-19 and disease duration (Kim et al., 2021; Merino et al., 2021; Salazar-Robles et al., 2021; Tavakol et al., 2021). However, dietary habit considerably varies across cultures as well as within culture among people. All the reports mentioned earlier mainly focused on the Western and Mediterranean diets which are typically less spicy, though various spices are regarded as hidden treasures of numerous therapeutic components and related health benefits (Peter, 2006; Sharif et al., 2018).

Consequently, it is critical to decipher the impact of dietary habits including spices consumption on the susceptibility to COVID-19, hospitalization risk and recovery. In this scenario, various predictors such as pre-existing food habits of infected people, symptoms ranging from asymptomatic, mild to moderate to severe can be considered for the study. No doubt, the eating habit of Bangladeshi people is a very complex issue. Keeping this in mind, we aimed to focus on the specific food habits such as various spices and drinks, dietary carbohydrates, proteins and fats. Furthermore, biological attribution of variation in age and gender were considered to find out their association with COVID-19 severity and hospitalization (Hu et al., 2021; Kushwaha et al., 2021). The various benefitting predictors, of course will not terminate COVID-19 infection rate, but may help to reduce the pain and mortality rate of the patients. Therefore, this study will help to examine this association (beneficiary or worsening) so that a new combo approach can be developed to relieve patients from COVID-19 sufferings and saves as many lives as possible via reducing severity and prompting recovery during this deadly contagious disease. Though a few investigations have been done around the world (Bousquet et al., 2020; Ingram et al., 2020; Jayawardena and Misra, 2020; Maiorino et al., 2020; Martinez-Ferran et al., 2020), to our knowledge, this is the first survey-based findings in Bangladesh which deals with the investigation on the association between pre-existing food habits with COVID-19 severity and hospitalization.

Methods

Study design, participants, sites and data collection

The study included laboratory-confirmed COVID-19 patients who were diagnosed positive and later found to be negative at the COVID-19 test laboratory by RT-PCR assay. The risk factors such as age, sex, dwelling and comorbidities were considered to quantify the association between COVID-19 severity and hospitalization status of the recovered individuals. By following the guidelines from Directorate General of Health Services (DGHS), Ministry of Health and Family Welfare (MOHFW), Bangladesh and World Health Organization other than clinical data, the symptoms of recovered patients were divided into three categories (we did not find any moderate case among the respondents): asymptomatic, mild and severe (DGHS, MOHFW, 2020). One might raise a valid question about why the COVID-19 patients enrolled for the study were not clinically assessed to ascertain mild, moderate, and severe cases. Inclusion of clinical examination of cases could have been more ideal for this study, but was impracticable for recruiting a fairly large number of cases from a South Asian country like Bangladesh where access to quality health services (proper documentation, electronic medical data archive etc.) for the common people is insufficient. To overcome the limitation, we compared the patients self-descried reports, as received during the telephonic interviews, with the guidelines provided by the DGHS (Bangladesh) aligned with the COVID-19 categorizations described by WHO (DGHS, MOHFW, 2020), to discriminate among the severity of the cases enrolled. In our previous study, we followed the similar way to ascertain the epidemiological and symptom data (Ganguli et al., 2022). Of note, it was confirmed that the attending doctors had already diagnosed the patients during their infection period and the same was reported by the respondents (patients) through telephonic interviews. Additionally, physicians directed the patient’s severity (asymptomatic, mild, severe) of COVID-19 disease upon examination of laboratory reports based on the guidelines of DGHS, MOHFW and WHO in order to avoid unnecessary hospital admission load caused by the COVID-19. Huang C. (2020) and his group also directly communicated with patients or their relatives to ascertain the epidemiological and symptom data which were not available from electronic medical records (Huang et al., 2020). For asymptomatic, mild, and moderate cases, interviews were conducted over telephone following the list collected from city corporation booths in Dhaka, Bangladesh by Hawlader M. D. H. and his group (Mohsin et al., 2021). COVID-19 severities and hospitalization were compared between the recovered patients with or without comorbidity status. We randomly included 1025 recovered individuals who provided completed data (See Supplementary Information for details). The study sites were two districts (Dhaka and Jhenaidah) of Bangladesh. All data were collected through telephone interviews and the same were recorded in an electronic form. The data collection period was November 2020 to March 2021.

Statistical analysis

Bivariate and multivariate analyses were done to identify the association of the selected parameters with COVID-19 status. We used graphical presentation and tabular format to display the distribution of age, sex and dwelling for identifying the pattern of the patients who suffered from COVID-19. We performed the bivariate analysis to compare age, hospitalization, and comorbidities between male and female. Chi-square test was performed to test the association between the categorical variables (sex, dwelling, consumption habits, dietary foods and spices) while Student’s t-test was used to explore the significance of the mean difference of the continuous variables (age). In multivariate analysis, we performed three regression models to assess the influencing factors of two outcome variables such as hospitalization status (model 1) and degree of severity (model 2). The binary logistic regression model was used to estimate each model of interest It is noted that the outcome variable “degree of severity” was defined as “severe” if the patients reported the severity level of the symptoms as “severe” and as “not severe” if the patients reported the severity level as “mild” or they were asymptomatic. The explanatory variables were age (in years), sex (male vs. female), dwelling status (urban vs. village), consumption habits (black tea, milked tea, coffee, betel leaf, honey, black caraway seeds and pickles; consumption vs non-consumption), dietary foods (rice fish, meat, vegetables and fruits; consumption-once/twice/thrice per day vs non-consumption) and spices (ginger, garlic, onion, turmeric, bay leaf and mixed aromatic spices (cinnamon + cardamom + cloves); consumption vs non-consumption). To identify consumption vs non-consumption habits, at least 3 months period prior to COVID-19 infection was considered for analyzing risk association of dietary habits to hospitalization and severity. We considered the odds ratios (ORs), 95% confidence intervals (CI), and p values for the variables in the models. In the case of all statistical analyses, we assumed significance only if p < 0.05 (two-tailed). We performed the statistical analysis using SPSS (version 25).

Results

Demographic data

Demographic data suggested that the non-comorbid patients differed from the comorbid patients. Of the patients enrolled for the study, 45.6% (n = 467) had at least one comorbidity (Supplementary Information (SI) Figure 1). Overall, comorbid patients required more hospitalization (12.6%) (SI Figure 2).

In addition, comorbidity might have influenced the severity of the diseases. Though mild symptoms prevailed mostly for both types of patients, severe symptoms were found comparatively higher in the case of comorbid patients (Figure 1). Median age of the non-comorbid patients was 31 ± 10.7 year which was much less than that of comorbid patients (45 ± 12.8 year) (SI Figure 3). It is worth noting that male and urban patients outnumbered their counterparts in terms of numbers (SI Figure 4). On the other hand, regardless of gender and comorbidity status, a higher percentage of urban patients suffered from severe symptoms and required a greater rate of hospitalization compared to those of dwelling in villages (Figure 1). For instance, comorbid urban dwellers experienced a greater degree of disease severity (urban 38.6% vs village 32.1%) as well as a higher rate of hospitalization (urban 13.2% vs village 9.9%) than the villagers. A similar trend also was observed for non-comorbid patients (severity: urban 20.7% vs village 14.0% and hospitalization: urban 7.2% vs village 4.4%). The results of chi-square test showed that significant association between severity and socio-demographic characteristics (sex, dwelling). Alike severity, similar results were obtained in case of hospitalization as well (Supplementary Table 1).

Figure 1.

Demographic data of gender (a, b) and habitation (c, d) in terms of hospital admission and degree of severity.

Comparison among the consumption habits of tea, honey, black caraway seeds, pickles and betel leaf

Consumption of black tea was found as most popular habit in comparison to milked tea and coffee for both types of patients (SI Figure 5(a) and 5(b)). Of the patients with comorbidity, 88.4% consumed black tea (SI Figure 5(a)). Betel leaf was the least consumed of all the regular habits (28.8% for non-comorbid and 40.1% for comorbid), while honey, black caraway seeds, and pickles were consumed by more than 70% of non-comorbid patients (SI Figure 5(b)). Their corresponding consumption rates were a little lower among the comorbid patients (68–82%). As rice is the staple food of Bangladesh, a large number of patients consumed rice twice or thrice a day (SI Figure 5(c) and 5(d)). Fish are found adequately and cheaply compared with another source of animal protein in Bangladesh. Consequently, two times consumption of fish were the frequent phenomenon because of the traditional habit where they mostly consumed meat once a day. Vegetables (leafy or fleshy; fiber and vitamin-rich) were evenly popular for both types of the patients. Surprisingly, fruits were consumed thrice per day by non-comorbid patients (55.9%) but much less by comorbid patients (22.9%). The percentages of patients with moderate consumption of spices (70–79%) were higher than those of low and high consumption of spices for both types of the patients (SI Figure 5(e) and 5(f)).

Impact of dietary habits towards hospitalization and severity

Table 1 shows the correlation among food habits, severity and hospitalization for comorbid and non-comorbid patients. The trend of degree of severity and the hospitalization rate for non-comorbid patients were found decreasing with consumption of black and milked tea, honey, black caraway seeds and pickles. Hospitalization rates were less than 7%. Besides, severity decreased along with regular consumption of these habitual foods. However, mixed results were observed for the comorbid patients. Less hospitalization was observed for comorbid patients who had consumed pickles (12.1%), black tea (11.9%), milked tea (11.3%) and honey (11.3%). However, opposite trends for hospitalization of comorbid patients were found for coffee, betel leaf and black caraway seeds. The least severity was found in case of comorbid black tea consumers (29.7%) where the severity was increased up to 35.6% for black caraway seeds consumers. On the other hand, severity did not exceed 20% for non-comorbid patients when they consumed these foods (Table 1).

Table 1.

Correlation between food habit, severity and hospitalization.

		Comorbid					Non-comorbid
		Hospitalization		Severity			Hospitalization		Severity
		No	Yes	Asymptomatic	Mild	Severe	No	Yes	Asymptomatic	Mild	Severe
Black tea	Not Consumed	82.8%	17.2%	6.0%	55.3%	38.7%	92.6%	7.4%	11.3%	69.2%	19.5%
	Consumed	88.1%	11.9%	9.4%	60.9%	29.7%	93.5%	6.5%	12.3%	69.1%	18.5%
Milk tea	Not consumed	83.7%	16.3%	6.1%	54.9%	39.0%	84.6%	15.4%	11.4%	67.5%	21.1%
	Consumed	88.7%	11.3%	7.3%	59.3%	33.3%	95.9%	4.1%	11.5%	69.7%	18.9%
Coffee	Not consumed	88.1%	11.9%	4.9%	55.3%	39.8%	91.8%	8.2%	10.7%	69.5%	19.8%
	Consumed	86.8%	13.2%	8.5%	57.2%	34.3%	94.2%	5.8%	12.9%	68.6%	18.6%
Betel leaf	Not consumed	88.7%	11.3%	5.1%	55.1%	39.8%	92.5%	7.5%	13.4%	68.8%	17.8%
	Consumed	85.2%	14.8%	7.2%	56.7%	36.1%	95.5%	4.5%	10.7%	69.3%	20.0%
Honey	Not consumed	84.5%	15.5%	5.0%	56.1%	38.9%	89.9%	10.1%	11.6%	67.4%	21.0%
	Consumed	88.7%	11.3%	9.5%	56.1%	34.5%	94.5%	5.5%	11.4%	69.8%	18.8%
Black caraway seeds	Not consumed	89.9%	10.1%	5.0%	57.6%	37.4%	92.1%	7.9%	15.9%	63.4%	20.7%
	Consumed	86.3%	13.7%	7.0%	55.5%	37.5%	93.9%	6.1%	9.6%	71.6%	18.8%
Pickles	Not consumed	85.4%	14.6%	6.5%	54.4%	39.1%	91.2%	8.8%	10.3%	69.4%	20.3%
	Consumed	87.9%	12.1%	6.3%	62.5%	31.3%	93.8%	6.2%	17.6%	68.1%	14.3%
Ricea	Once	87.1%	12.9%	4.3%	58.6%	37.1%	89.2%	10.8%	2.7%	59.5%	37.8%
	Twice	86.7%	13.3%	7.6%	54.3%	38.1%	97.3%	2.7%	13.3%	70.1%	16.7%
	Thrice	88.2%	11.8%	5.9%	57.2%	36.9%	89.9%	10.1%	10.9%	69.6%	19.5%
Fisha	Once	89.3%	10.7%	7.1%	46.4%	46.4%	91.6%	8.4%	13.7%	59.5%	26.7%
	Twice	86.7%	13.3%	6.8%	59.0%	34.1%	95.0%	5.0%	12.2%	69.4%	18.3%
	Thrice	86.8%	13.2%	4.7%	59.4%	35.8%	91.9%	8.1%	8.1%	77.2%	14.8%
Meata	Once	85.8%	14.2%	7.1%	50.0%	42.9%	92.6%	7.4%	14.8%	59.3%	25.9%
	Twice	87.8%	12.2%	6.8%	63.3%	29.9%	93.7%	6.3%	9.4%	75.4%	15.2%
	Thrice	91.3%	8.8%	3.8%	61.3%	35.0%	94.4%	5.6%	8.1%	79.0%	12.9%
Vegetablesa	Once	89.9%	10.1%	9.0%	32.6%	58.4%	88.4%	11.6%	11.2%	62.9%	25.8%
	Twice	87.4%	12.6%	5.7%	58.3%	36.0%	96.1%	3.9%	13.2%	68.0%	18.9%
	Thrice	86.2%	13.8%	5.9%	64.5%	29.6%	100.0%	0.0%	10.0%	72.6%	17.4%
Fruitsa	Once	87.2%	12.8%	5.8%	52.9%	41.2%	90.7%	9.3%	12.7%	63.3%	24.1%
	Twice	88.4%	11.6%	7.0%	59.3%	33.7%	95.2%	4.8%	15.0%	67.1%	18.0%
	Thrice	86.9%	13.1%	7.5%	61.7%	30.8%	100.0%	0.0%	9.3%	71.8%	18.9%
	Low	77.6%	22.4%	7.5%	56.7%	35.8%	90.1%	9.9%	16.0%	59.3%	24.7%
	Medium	89.6%	10.4%	7.0%	57.7%	35.4%	94.5%	5.5%	11.4%	70.1%	18.5%
	High	85.5%	14.5%	1.8%	45.5%	52.7%	87.2%	12.8%	2.6%	79.5%	17.9%
Ginger	Low	78.3%	21.7%	0.0%	48.1%	51.9%	87.2%	12.8%	16.7%	59.0%	24.4%
	Medium	89.9%	10.1%	7.2%	56.9%	35.8%	94.5%	5.5%	11.3%	70.3%	18.4%
	High	82.7%	17.3%	7.2%	58.0%	34.8%	90.1%	9.9%	4.4%	75.6%	20.0%
Onion	Low	78.8%	21.2%	0.0%	55.3%	44.7%	87.5%	12.5%	15.8%	60.5%	23.7%
	Medium	89.8%	10.2%	7.1%	55.6%	37.3%	94.6%	5.4%	11.0%	70.4%	18.5%
	High	80.9%	19.1%	7.6%	59.1%	33.3%	90.8%	9.2%	8.9%	71.4%	19.6%
Turmeric	Low	79.7%	20.3%	0.0%	55.8%	44.2%	86.8%	13.2%	17.5%	61.3%	21.3%
	Medium	89.6%	10.4%	7.0%	56.1%	36.9%	94.1%	5.9%	11.1%	69.8%	19.1%
	High	81.4%	18.6%	7.2%	56.5%	36.2%	92.5%	7.5%	2.6%	78.9%	18.4%
Bay leaf	Low	77.8%	22.2%	0%	51.1%	48.9%	86.5%	13.5%	14.5%	59.1%	26.4%
	Medium	90.7%	9.3%	7.5%	56.3%	36.1%	94.2%	5.8%	11.4%	71.0%	17.5%
	High	82.2%	17.8%	5.6%	57.8%	36.7%	92.7%	7.3%	2.7%	78.4%	718.9%
Mixed Aromatic Spices	Low	78.9%	21.1%	0%	56.1%	43.2%	88.9%	11.1%	15.4%	59.0%	25.6%
	Medium	90.4%	9.6%	7.5%	56.8%	36.6%	94.1%	5.9%	11.1%	71.4%	17.5%
	High	81.8%	18.2%	5.6%	56.7%	37.8%	92.3%	7.7%	2.8%	77.8%	19.4%

^a Per day.

In case of non-comorbid patients, moderate (twice) to high (thrice) consumption of rice, fish, meat, vegetables, and fruits were associated with mild form of disease and lower rate of hospitalization. For instance, three times consumption of fish per day reduced severity from 26.7% (once/day) to 14.8%. Whereas two times consumption of fish per day reduced the rate of hospitalization from 8.4% (once/day) to 5.0%. Overall, a similar trend was observed for the patients with comorbidity. Interestingly, no hospitalization was observed in case of the non-comorbid consumers of vegetables and fish (thrice/day) where rice, fish and meat were associated with 5–10% hospitalization for the non-comorbid patients. On the contrary, hospitalization percentages (10–13%) for comorbid patients did not differ much along with the frequency of the consumption of these food.

In case of ginger, garlic, onion, turmeric and mixed aromatic spices (MAS) (cinnamon + cardamom + cloves), reduced degree of severity and hospitalization were observed in case of the moderate consumers of these spices for both types of patients (Table 1). For example, moderate consumption of ginger reduced the hospitalization rate to 10.1% from 21.7% (low consumption) and 17.3% (high consumption) for the comorbid patients, whereas to 5.5% from 12.8% (low consumption) and 9.9% (high consumption) for the non-comorbid patients. Overall, moderate consumption of spices was associated with least hospitalization for comorbid patients (<10%) which were quite similar to the trend for non-comorbid patients.

Risk association of dietary habits to hospitalization and severity

The results of chi-square test indicated the significant association between severity and food habits (consumption habits, dietary foods, or spices) and similar results were obtained in cases of hospitalization as well (Supplementary Table 1). Table 2 displays the odds ratio of the dominating factors of hospitalization and severity of symptoms due to COVID-19. It is manifest from a cursory glance at the habitual consumption of the respondents, those who did not consume any kinds of hot drinks such as black or milk tea or coffee, natural remedies such as honey or black caraway seeds had more likelihood of being hospitalized due to COVID-19 compared to the counterpart. Similar evidence has been observed for both non-comorbid and comorbid conditions.

Table 2.

Risk association for hospitalization and severity.

		Hospitalization						Severity
		Non-comorbid			Comorbid			Non-comorbid			Comorbid
	Category	OR (95%CI)	p-valuea		OR (95%CI)	p-valuea		OR (95%CI)	p-valuea		OR (95%CI)	p-valuea
Consumption Habits
Black tea	Not consumed	1.167 (.348–3.915)	.061		1.146 (.488–2.687)	.045		.602 (.270–1.340)	.014		1.271 (.644–2.511)	.489
Milked tea	Not consumed	3.317 (1.286–8.556)	<.001		1.597 (.777–3.280)	.003		1.129 (.619–2.060)	.032		1.191 (.689–2.031)	.022
Coffee	Not consumed	2.153 (.798–5.805)	.063		1.357 (.677–2.720)	.039		1.054 (.622–1.787)	.045		1.267 (.794–2.022)	.044
Betel leaf	Not consumed	2.407 (.786–7.371)	.197		.715 (.380–1.347)	.300		1.247 (.718–2.166)	.433		1.544 (.984–2.422)	.283
Honey	Not consumed	1.569 (.583–4.221)	.046		1.750 (.869–3.527)	.017		1.028 (.558–1.894)	.030		1.265 (.746–2.126)	.122
Black caraway seeds	Not consumed	1.295 (.469–3.578)	.027		2.238 (.958–5.227)	.043		1.031(.488–1.95)	.028		1.029 (.583–1.814)	.085
Pickles	Not consumed	1.583 (.558–4.489)	.065		1.016 (.453–2.279)	.269		1.605 (.783–3.292)	.197		1.232 (.680–2.232)	.047
Dietary Foods b
Rice	Once		.002			.399			.003			.521
	Twice	.386 (.073–2.046)	<.001		1.242 (.457–3.378)	.671		.242 (.103-.571)	.001		.785 (.411–1.500)	.108
	Thrice	.980 (.207–4.641)	.002		1.522 (.672–3.448)	.314		.465 (.193–1.123)	.089		1.157 (.571–2.344)	.086
Fish	Once		.318			.257			.710			.442
	Twice	.311 (.095–1.019)	.131		.342 (.095–1.230)	.100		1.344 (.654–2.764)	.421		1.326 (.700–2.514)	.387
	Thrice	.602 (.149–2.433)	.415		.547 (.183–1.639)	.282		1.090 (.339–3.507)	.884		1.920 (.685–5.383)	.215
Meat	Once		.791			.052			.012			.063
	Twice	3.508 (.847–14.526)	.517		5.003 (1.282–19.529)	.020		.995 (.855–8.010)	.017		.566 (.308–1.039)	.066
	Thrice	4.590 (.919–22.910)	.812		2.532 (1.672–9.537)	.170		.917 (.282–3.506)	.013		.569 (.205–1.578)	.279
Vegetable	Once		<.001			.017			.008			<.001
	Twice	.394 (.128–3.524)	.006		.751 (.249–2.271)	.012		.927 (.423–11.722)	.045		.444 (.225-.878)	.020
	Thrice	.182 (.020–1.690)	.034		.798 (.364–1.752)	<.001		.867 (.911–5.642)	.039		.201 (.091-.442)	<.001
Fruit	Once		.006			.049			.186			.080
	Twice	.893 (.324–1.723)	.044		.785 (.325–1.895)	.014		1.028 (.233–4.527)	.071		.958 (.516–1.778)	.089
	Thrice	.346 (.128–3.695)	.004		.755 (.293–1.945)	.036		2.081 (.428–10.130)	.064		.813 (.435–1.520)	.117
		Hospitalization						Severity
		Non-comorbid			Comorbid			Non-comorbid			Comorbid
	Category	OR (95%CI)	p-valuea		OR (95%CI)	p-valuea		OR (95%CI)	p-valuea		OR (95%CI)	p-valuea
Spices consumption
Ginger	Low		.048			.024			.035			.199
	Medium	.823 (.001–14.330)	.037		.420 (.037–4.729)	.016		.720 (.025–7.152)	.049		.226 (.024–2.136)	.194
	High	.774 (.621–3.249)	.014		.195 (.006–5.967)	.650		.683 (.324–4.287)	.027		.588 (.040–8.604)	.068
Garlic	Low		.005			.017			.011			.014
	Medium	.821 (.056–1.166)	.016		1.059 (.017–7.555)	.014		.957 (.554–11.166)	.013		.945 (.008–2.807)	.020
	High	.329 (.192-.736)	.002		.656 (.054–7.978)	.006		.859 (.099–7.392)	.030		.856 (.196–7.664)	.046
Onion	Low		.043			.017			.046			.333
	Medium	.915 (.082–3.219)	.036		.315 (.006–16.438)	.024		.864 (.568–7.256)	.049		7.152 (1.440–16.316)	.167
	High	.955 (.469–2.008)	.022		.746 (.058–9.586)	.005		.936 (.268–5.773)	.044		3.669 (.070–9.106)	.520
Turmeric	Low		.214			.036			.094			.031
	Medium	.646 (.109–2.563)	.186		.277 (.005–15.268)	.038		.260 (.193–2.430)	.041		2.220 (.361–13.645)	.038
	High	.869 (.571–3.438)	.094		.442 (.013–15.163)	.010		.988 (.307–1.811)	.080		.951 (.008–8.207)	.043
Bay leaf	Low		.190			.097			.013			.099
	Medium	1.670 (.050–16.143)	.209		3.011 (.031–295.372)	.082		.643 (.040–2.963)	.066		1.138 (.187–6.916)	.083
	High	.864 (.559–7.326)	.082		.794 (.009–72.150)	.051		.770 (.593–1.038)	045		.843 (.573–3.227)	.089
Mixed Aromatic Spices	Low		.026			.037			.013			.037
	Medium	.587 (.013–2.599)	.076		.637 (.008–48.027)	.089		.603 (.084 4.346)	.003		1.414 (.216–9.259)	.018
	High	.739 (.318–6.544)	.064		1.017 (.016–66.299)	.042		.898 (.495–1.428)	.002		.648 (.187–1.054)	.045

^a Two-tailed independent t-test, ^bPer day, OR: odds ratio.

COVID-19 patients without comorbidity had 57% (OR 1.57; p < 0.05) and 30% (OR 1.30; p < 0.05) higher risk of admitting hospital when they did not consume honey and black caraway seeds respectively than those patients who consumed these items. While these figures were 75% (OR 1.75; p < 0.05) and 124% (OR 2.24; p < 0.05) for COVID-19 patients with comorbid condition. The risk of hospitalization was surprisingly 232% (OR 3.32; p < 0.01) and 60% (OR 1.60; p < 0.01) higher respectively in case of non-comorbid and comorbid milked tea consumer. However, habit of milked tea consumption has less benefit compared to black tea. These figures were 13% and 19% respectively for non-comorbid and comorbid patients in developing the risk of severe symptoms. Honey (OR 1.028; p < .05) and black caraway seeds (OR1.038; p < .05) also had a significant effect on the occurrence of severe symptoms in COVID-19 patients without comorbidity; however, surprisingly, they had no significant effect (p > .05) on comorbid patients.

There was less likelihood of hospitalization for the non-comorbid patients who had consumed rice, vegetables, and fruits two or three times daily. The likelihood of being hospitalized for the non-comorbid patients decreased by 61 percent (OR 0.39; p < 0.01) with increased (twice daily) consumption of rice. For comorbid patients, there was no significant effect (p > .05) of rice consumption on their hospitalization and degree of severity.

Moreover, patients without comorbidity who ingested more (2 times/day) vegetables and fruits were 61% and 11% respectively less likely to be hospitalized compared to those who did not take such foods once a day. If these patients increased their daily consumption level of such foods, i.e., three times/day, the corresponding figures also increased to 82% and 65%, respectively. Vegetables and fruits consumption also reduced the likelihood of hospitalization and severity of the patients with comorbidity. Interestingly, the comorbid patients who took meat two times daily had 400% more likelihood of being hospitalized due to COVID-19 compared to the counterpart though the likelihoods for developing severe symptoms were slightly decreased for both two- and three-time consumption (OR 0.995 and 0.917 respectively). The likelihood of having severe symptoms decreased with an increased intake of vegetables. Specially, three-time consumption per day was associated with the least likelihood (with respect to one-and two-time consumers) of hospitalization and severe symptoms both for comorbid and non-comorbid patients.

According to the amount of spices intake, more consumption of ginger, garlic, and onion reduced the hospitalization risk for both comorbid and non-comorbid patients. The likelihood of being hospitalized due to COVID-19 for the non-comorbid patients decreased by 23% (OR 0.77; p < 0.01) and 67% (OR 0.33; p < 0.01) with the increased (high) levels of ginger and garlic intake, respectively. In addition to these spices, other spices such as turmeric and bay leaf also played a significant role in reducing the risk of hospitalization for patients with comorbidity. The comorbid patients who consumed a medium and high levels of turmeric had 72% (OR 0.28; p < 0.05) and 56% (OR 0.44; p < 0.01) respectively less likelihood of hospitalization. Medium and high levels of consumption of all the aforementioned spices had a significant effect on reducing the risk of severe symptom development for the patient without comorbidity. In the case of comorbid patients, only high-level intake of garlic (OR 0.86; p < 0.05), turmeric (OR 0.95; p < 0.05), and mixed aromatic spices (OR 0.65; p < 0.05) had a significant effect on developing the severity symptom, however, surprisingly, those patients who consumed medium level of turmeric (OR 2.22; p < 0.05), and mixed aromatic spices (OR 1.41; p < 0.05) had the opposite effect.

Discussion

Hospitalization requirements were relatively higher in case of comorbid patients which might be due to less immunity (Callender et al., 2020; Castle et al., 2005). In this study, comorbid patients experienced higher severity and had the fewest asymptomatic cases which are coherent with prior investigations (Davies et al., 2020). According to our data, males prevailed over females in number of cases where most of the patients were urban dwellers which also reflected the previous reports (Bwire, 2020; Girdhar et al., 2021; Peckham et al., 2020). Of note, hospitalization and severe symptoms were observed in higher percentages for comorbid and non-comorbid female patients than their male counterparts which might be observed for the first time in the South-Asian country. Females, specially, aged ones usually suffer from different post-menopausal difficulties which might trigger worseCOVID-19 outcomes (Colditz et al., 2010; Costeira et al., 2021). In addition, Delta variant of coronavirus (SARS COV-2) were predominant during our investigation, therefore, possibly females were vulnerable against Delta variant (Chen et al., 2021; Rangchaikul and Venketaraman, 2021).However, villagers required less hospitalization and suffered mostly from mild symptoms. This phenomenon can be explained by the fact that they were accustomed to more physical activities, greener abode and contamination-free foods than the urban peoples (Chen et al., 2017; Riva et al., 2009). However, easy access to hospital for urban peoples cannot be ruled out at this moment (Bain et al., 2014; Chatterjee and Sarkar, 2021; Lee et al., 2015). We found that consumption of black tea, milked tea, honey and pickles were responsible for decreasing the degree of severity and hospitalization requirement due to COVID-19 (Table 1). Particularly, consumption of black tea appeared to be liked with lesser severity (non-consumer 38.7% vs consumer 29.7%) and lower hospitalization (non-consumer 17.2% vs consumer 11.9%) for comorbid patients. It is worthwhile to mention that the role of natural products for defending against infectious disease is a widespread practice from the earlier days. Because of their preventative properties against different microbial infection, they are being suggested to use in case of COVID-19 as well, somewhere, they have been proved to be effective (Aman and Masood, 2020; Ayivi et al., 2021; Gasmi et al., 2021). It is well known that tea and coffee contain antioxidants which are regarded as the immune promoters (Açıkalın and Sanlier, 2021; Bhattacharyya et al., 2003; Hamer, 2007; Turkmen et al., 2007). Even tea has been reported as a bioactive modulator of innate immunity in cases of COVID-19 (Chowdhury and Barooah, 2020). Though coffee is considered as an effective beverage against different diseases, its impact on the recovery from COVID-19 is yet to be concluded (Belaroussi et al., 2020; Kennedy et al., 2021; Wierzejska, 2017).Black caraway seeds and pickles were also reported for their high medicinal values as they are well known for their anti-microbial actions (Chakraborty and Roy, 2018; Forouzanfar et al., 2014). Taken collectively, it can be recalled that due to antimicrobial, antiviral, antioxidant properties and having vitamins/minerals into tea, honey and pickles might play a role against COVID-19.

On the other hand, increased consumption of rice, fish, meat, fruits and vegetables were found to be associated with less severity and hospitalization (Table 1). Though direct relationship between rice consumption and COVID-19 severity and hospitalization is not evident, but, rice consumption might be beneficiary for avoiding massive damage by this disease. As carbohydrates are closely associated with immune components at molecular level, carbohydrate rich food is suggested for COVID-19 patients in different studies (Kumbhar et al., 2021). Fish and meat both contain protein which might be the factor for less hospitalization and severity among the consumers. This result is coherent with previous studies (Batiha et al., 2021; Fan et al., 2020). Fruits and vegetables contain different vitamins and minerals including vitamin-C which assist the immune system (Carr and Maggini, 2017). For this reason, most of the dietary guidelines recommended for taking vegetables and fruits to prevent COVID-19 (Jayawardena and Misra, 2020).

Besides, moderate consumption of spices (ginger, garlic, turmeric, onion, mixed aromatic spices and bay leaf) was linked with less hospitalization and severity comorbid and non-comorbid patients. Ginger was reported to modulate oxidative stress and prostaglandins as harmful factors in COVID-19 (Mashhadi et al., 2013; Mohamed et al., 2015). In addition, it is capable of modulating improper effect or T cell responses in COVID-19 (Jafarzadeh et al., 2021). Garlic is a potential therapeutic as it contains organosulfur and flavonoid compounds which can be used to fight with COVID-19 (Khubber et al., 2020). Turmeric contains curcumin which is regarded as a treatment for COVID-19 infection (Babaei et al., 2020). Meanwhile, onion and bay leaf have also high medicinal values which might be the reason for less damage among the users. High and low consumption of these spices were observed with higher hospitalization and severity, moderate consumption of these spices were mostly favorable.

Significant correlation between food habits with severity and hospitalization for comorbid and non-comorbid COVID-19 patients were observed. Among the pre-existing habits, consumption of black tea, milked tea, pickles, honey, and black caraway seeds were identified as positive factors of reducing the degree of severity and hospitalization for COVID-19 patients. Two- or three-times/day consumption of fish, fruits and vegetables may reduce severity and hospitalization requirements. On the other hand, comorbid patients with a daily meat intake (twice daily) were at a very high risk of hospitalization. Moderate consumption of ginger, garlic, onion, turmeric and MAS was found as favorable and hence recommended. Therefore, we hypothesized that a combo approach of nutritional management with the necessary medical diagnose and treatment might be suggested against COVID-19. However, cautions should be considered as it is first time survey-based reports among the recovered individuals in Bangladesh and sixth report in world.

Limitations

This study included recovered patients from only two selected cities in Bangladesh. No specific community population involvement in the present study. Therefore, the diversity of the sample size is questionable. We do not have clinical data such as oxygen level during infection, CBC, SGPT, blood glucose, ferritin, D-dimer, ESR, Creatinine, HbA1C, etc. Moreover, we could not record the duration of the underlying conditions. In addition, we collected data through a telephone call; there could be misinformation or biased information from the interviewer and participants. However, this study reconfirms the global findings that preexisting nutritional habits associated with low severity and hospitalization.