Tele‐Medical Nutrition Therapy to Promote Healthy Eating and Nutrition Adherence of Chronic Patients in the COVID‐19 Pandemic: A Pragmatically Before‐After Trial

ABSTRACT

Background and Aims

Patients find it difficult, during the pandemic, to receive appropriate nutrition services because of the limited access to health services. This study assessed the effectiveness of a tele‐consulting nutrition intervention on the nutrition adherence of patients with diabetes and hypertension.

Methods

A tele‐medical nutrition therapy (MNT) intervention was conducted as a one‐group before‐after trial. The estimated sample size was 314. Of these, 183 patients were selected from Hefdah‐e‐Shahrivar and 131 from Farabi Hospital. Patients were asked to complete an adapted version of the nutrition adherence self‐reported instrument to compare nutrition adherence when they received face‐to‐face therapy versus tele MNT.

Results

In total, 166 completed all phases of the study. The results showed that the nutrition adherence score significantly increased compared to before the intervention. Hedges's g effect size also confirmed the high efficacy of telenutrition therapy. The magnitude of differences was within a high range (effect size 3.76 [CI: 3.40–4.12]).

Conclusions

The present findings showed that providing teleconsulting nutrition services to diabetic and hypertensive patients can positively affect their nutrition adherence. Telenutrition to promote a healthy diet can contribute to effectively controlling diseases in chronic patients during/after the pandemic with no gaps or failures.

Trial Registration

This study was a tele‐nutrition intervention. In this study, we provided consultation on regular food consumption and did not order any chemical substances or nutritional supplements. Furthermore, we did not perform any treatment or invasive intervention, we assumed that a registration number was unnecessary, so we did not apply for clinical trial registration.

Abbreviations

BMI

body mass index

CI

confidence interval

ICC

intra‐class correlation coefficient

MNT

medical nutrition therapy

NA

nutrition adherence

NASI

nutrition adherence self‐report instrument

NIMAD

National Institute for Medical Research Development

RDN

registered dietitian nutritionists

1. Background

Noncontagious diseases such as type 2 diabetes and hypertension are the main causes of mortality in the world and the main health challenge in the 21st century. These diseases account for 70% of mortalities on a global scale [1]. In Iran, the prevalence of diabetes and hypertension has been reported at 11.3 and 22.6, respectively [2, 3]. The World Health Organization (WHO) reported that the prevalence of hypertension was higher in developing than developed countries [4].

Diabetes and hypertension are associated with COVID‐19 mortality and morbidity. The results of a retrospective cohort study on COVID‐19 patients in China showed 48% of patients had at least one comorbidity; 30% had hypertension and 19% had diabetes [5]. Similarly, in New York, among 5700 patients hospitalized with COVID‐19, 56.6% and 33.8% had hypertension and diabetes, respectively [6]. In Italy, the prevalence of hypertension and diabetes among 1591 patients with COVID‐19 in the ICU was 49% and 17%, respectively [7]. The prevalence of COVID‐19 in hypertensive and diabetic patients in Iran has been reported at 21.1% and 16.3% [8]. A point of note is that diabetic patients are at a higher risk of different complications when infected with the virus [9].

Dietary interventions are among the best and most cost‐effective ways to manage diseases, such as diabetes and hypertension [10]. A complementary and safe strategy to help diabetic and hypertensive patients is to change their dietary habits and train them to eat healthily or deliver counseling interventions by registered dietitians (rather than other specialists) [11, 12, 13]. Furthermore, they provide evidence‐based and cost‐effective treatments for patients [14, 15].

Nutrition is key to a healthy immune system and can reduce the risk of COVID‐19 infection. However, in the COVID‐19 pandemic, many chronic patients especially those with diabetes and hypertension that a healthy diet is a key factor in their disease management [16, 17] found it difficult to adhere to their nutrition due to limited access to health services. Now more than ever, widespread access to healthy food should be prioritized, and people should develop healthier eating habits to reduce the long‐term complications of COVID‐19 [18]. It is essential to raise the middle‐aged and the elderly's awareness of a healthy diet and use new methods to easily convey and give unequivocal information to people [19]. Considering the limited access to health services and concerns about visiting health centers infected with COVID‐19, it could be helpful to provide tele‐consulting services to manage diseases during the COVID‐19 pandemic. Tele‐medical nutrition therapy (tele‐MNT) is not only beneficial during pandemic situations but also proves useful in other situations. For instance, it can be a valuable approach for individuals residing in remote areas, far from major health centers, and for the elderly who face physical constraints in commuting to health centers.

During the COVID‐19 pandemic, telehealth has been widely used as an innovative approach, so that patients can better access services such as tele‐MNT without the attending risk of places infected with the virus [20, 21, 22, 23]. This method also provides continuous and regular patient–physician communication to check on weight loss, blood sugar, and so forth, as was common in the pre‐pandemic era [20].

Due to the safety measures for the COVID‐19 pandemic, chronic patients were restricted from visiting health centers to lower the risk of infection. Thus, patients with chronic diseases tended to visit health centers less often. Therefore, Services such as tele‐MNT have been developed to help these patients continue their follow‐up care without worrying about the risk of infection. The present study aimed to assess the effectiveness of a tele‐MNT intervention in chronic patients’ nutrition adherence (NA) before and after the intervention.

2. Methods

The present study was registered in the National Institute for Medical Research Development (NIMAD) (#994347). It was conducted based on the Transparent Reporting of Evaluations with Nonrandomized Designs (TREND) Checklist [24] (see File S1).

2.1. Study Design, Setting, and Sample Size

A tele‐MNT intervention was conducted as a before‐after trial on diabetic and hypertensive patients during the COVID‐19 pandemic. This study was conducted in two hospitals affiliated with the Social Security Organization in Mashhad, Iran. More information about the setting of the study is provided in another relevant academic paper [25]. These two hospitals are the most frequently visited by patients with chronic diseases for follow‐up care and counseling for nutrition therapy and drugs. Among patients with chronic diseases visiting social security hospitals in Mashhad (i.e., Farabi and Shahrivar‐17 Hospitals), those with social security insurance were recruited. A stratified random sampling was used for sample selection. Hefdah‐e‐Shahrivar Hospital and Farabi Hospital have provided health services to many patients with chronic diseases. In the past 2 years, Hefdah‐e‐Shahrivar Hospital and Farabi Hospital catered to a total of 3500 and 2500 patients, respectively. To estimate the sample size, it was assumed that the present study could improve 2.5% of NA in patients with chronic diseases [26]. The sample size was estimated at a significance level of 95%, β = 0.2, and N = 6000. The estimated sample size was 314. Of these, 183 patients were selected from Hefdah‐e‐Shahrivar Hospital and 131 from Farabi Hospital. Given that this study was a single‐group before‐after trial, the patients were randomly selected based on their ID numbers recorded in electronic patient records (EPRs).

2.2. Tele‐Nutrition: Counseling Description

The interactive tele‐MNT was conducted by three registered dietitian nutritionists (RDNs). These consultations were provided for adults with Type 2 diabetes and hypertension according to the Nutrition Guidelines by the Academy of Nutrition and Dietetics [27, 28]. These guidelines helped nutritionists to give evidence‐based MNT for adults with diabetes and hypertension besides, provided recommendations for the treatment and management of these diseases. In addition, the MNT needed by patients during the COVID‐19 pandemic was also provided based on suggestions by the Ministry of Health and Medical Education of Iran [29].

Before the tele‐MNT intervention, patients received MNT services by visiting the hospital in person. During the study, RDNs provided consultation when they were present in the hospital, but the patients received the services at home or anywhere else they preferred. Before the counseling, the patients were contacted and asked to take part in the study. A member of the research team explained the purpose of the study and made an appointment for the tele‐MNT. During the study, patients underwent tele‐MNT on the phone once for 30–60 min within a month. During the phone‐mediated counseling, RDNs gave individualized MNT, which included dietary assessment, education, and nutrition counseling. The MNT was individualized based on the participants' dietary history, perceived barriers, medical diagnoses, and laboratory test results. During the counseling, patients asked the experts if they had any questions or doubts about the recommendations offered. To compare the results, the NA self‐reported instrument (NASI) was completed before the intervention and 1 month later after patients received the tele‐MNT. Patients answered these questions over the phone and their NA level was evaluated based on their self‐reporting responses.

2.3. Participants and Eligibility Criteria

Adult patients with hypertension or diabetes were included in the study if they met the following criteria:

1. Having an EPR.

2. Regular follow‐up visits to receive face‐to‐face nutrition services.

3. Having an active phone or mobile phone number.

4. Being able to answer the phone or mobile phone.

Patients who did not receive any services within the last year were excluded from the study. Patients who had difficulty understanding the questionnaire content were also excluded from the study.

2.4. Study Instrument

An adapted version of the NASI was used to collect the data. This instrument was developed based on the related literature and was modified for the Iranian nutrition culture [25, 30, 31, 32]. This instrument consists of three parts, the first one contains the patients' demographic information, derived from their EPRs. This information included age, sex, cigarette smoking, height, weight, body mass index (BMI), type of chronic disease, duration of chronic disease, education level, marital status, comorbidities, and occupation.

The second part contained five questions about lifestyle and routine dietary habits. The answers to these questions were binary, three‐point or four‐point. The third group of questions enquired about NA based on a 1‐to‐4‐week dietary recall. The answers to the questions in this section were rated on an 8‐point Likert scale (see File S2). The maximum and minimum scores of this questionnaire were 93 and 0, respectively. Higher scores indicate a better NA.

Before the data collection, to increase the readability of the instrument, its face validity was assessed by four medical informatics experts with medicine and health information technology backgrounds. The validity of the instrument was also confirmed by two nutritionists. Finally, according to the experts' comments, the present researchers modified the instrument in a panel discussion. The reliability was tested using the intra‐class correlation coefficient (ICC). ICC was estimated using a two‐way random effect. To this aim, 30 participants completed the questionnaire before the tele‐MNT intervention. After 2 weeks, the same participants completed the questionnaire again (ICC = 0.75; 95% CI: 0.64–0.83). In the final analysis, these were excluded.

2.5. Data Analysis

All the analyses were done in SPSS 24 and STATA 14.2. The normality of distribution was checked using histogram plots. Paired sample t‐test was run to test the within‐group difference. Hedges' g effect size (95% CI) was used to determine the magnitude of within‐group difference and the size of treatment efficacy. It is interpreted as follows: ≤ 0.19 (trivial zone), 0.2 ≤ effect size ≤ 0.49 (low efficacy), 0.5 ≤ effect size ≤ 0.79 (medium efficacy), and ≥ 0.8 (high efficacy) [33].

A multiple linear regression analysis was run to estimate the relationship between the dependent variable (the NASI score after the intervention) and independent variables (e.g., the NASI score before the intervention, sex, and age). Standardized Beta was used to determine the magnitude of these relationships. it is interpreted as follows: 0 ≤ stz‐beta ≤ 0.09 (trivial), 0.1 ≤ stz‐beta ≤ 0.29 (small effect), 0.3 ≤ stz‐beta ≤ 0.49 (medium effect), and ≥ 0.5 (large effect) [34]. The statistical assumptions of a linear regression analysis including the normality, linearity, and homoscedasticity of data were checked. If a covariate was not significantly correlated with the independent variable or the correlation between a covariate and independent variables was not high, the variable was excluded from the main model.

Demographic variables were described as frequency and percentage. The significance level was considered at p ≤ 0.05.

3. Results

3.1. Participants

The tele‐MNT was given between November 1, 2021, and April 20, 2022. Among the patients, 314 met the inclusion criteria and, thus, entered the study. Out of the 314 eligible patients, 148 did not participate in the second phase (a participation rate of 53%). The reasons for withdrawal were unwillingness to continue with the study, mortality, failure to answer the phone or mobile phone calls, or failure to comprehend the questionnaire (Figure 1).

Figure 1.

The consolidated standards of reporting trials (CONSORT) Guidelines; flow chart of participation in the before and after tele‐MNT.

3.2. Demographic Characteristics

The patients' baseline characteristics are reported in Table 1. Among the participants, 110 (66.3%) patients were over 60 years old. More than half were female (68.7%). Almost all (98.8%) were non‐smokers, and all were married (100%). Most participants had an education level of below diploma (84.9%). Over 60% were housewives. More than half of the subjects were overweight (65.7%). The information reported includes patients who participated in all phases of the study.

Table 1.

Baseline data.

Variables	Frequency (N = 166)
Age (years)
≤ 50	8 (4.8%)
51–60	48 (28.9%)
61–70	63 (38%)
≥ 71	47 (28.3%)
Sex
Female	114 (68.7)
Male	52 (31.3)
Length of the disease
≤ 5	66 (39.8)
6–10	59 (35.5)
11–15	19 (11.4)
≥ 16	22 (13.3)
Chronic disease(s)
Diabetes	66 (39.76%)
Hypertension	56 (33.73%)
Diabetes and hypertension	44 (26.5%)
Smoking
Yes	2 (1.2%)
No	164 (98.8%)
Education Level
Illiterate	48 (28.9%)
< diploma	93 (56%)
≥ diploma	25 (15.1%)
Occupation
Housewife	109 (65.7%)
Employee	43 (25.9%)
Worker	14 (8.4%)
BMI
Normal (≤ 24.99)	57 (34.3%)
Overweight (25.00–29.99)	75 (45.2%)
Fat (≥ 30)	34 (20.5%)
Marital status
Married	166 (100%)
Not married	0 (0%)

3.3. Primary Result: NASI Score

Within‐group comparisons showed that, after the tele‐MNT intervention, the NASI score significantly increased compared to before the intervention (p < 0.001) (Table 2). Hedges' g effect size also confirmed the high efficacy of the tele‐MNT in increasing NASI score after the intervention compared to before the intervention (effect size > 0.8). The magnitude of differences was within a high range (effect size = 3.76 [CI: 3.40–4.12]).

Table 2.

Paired‐sample t‐test results, within‐group comparisons, and Hedges' g with 95% CI of the primary result.

Primary outcome		Mean (SD)	t	p‐value	Mean difference (95% CI)	Hedges' g (95% CI)
NASI score	Before	48.81 (12.44)	33.93	< 0.001	33.60 (31.64–35.55)	3.76 (3.40–4.12)
	After	82.41 (1.97)

Abbreviations: CI = confidence interval, NASI = Nutrition Adherence Self‐Report Instrument, SD = standard deviation.

3.4. Secondary Results

A multiple linear regression model was used to identify factors independently associated with the NASI score after the intervention (Figure 2). Independent variables included age, sex, BMI, length of the disease, type of chronic disease, and education level. To predict the NASI score after the intervention, the NASI score before the intervention was also added as an independent variable. Different models are presented in Table 3. Considering the statistical significance and high standardized beta, Model 1 was specified as the final model.

Figure 2.

Scatter plot of linearity: Response 2: NASI score after the intervention, Response 1: NASI score before the intervention. CI = confidence interval, NASI = Nutrition Adherence Self‐Report Instrument.

Table 3.

Linear regression models.

Model	Independent variable	Beta‐coefficient (95% CI)	p‐value	Standardized beta
Model 1	NASI score	−0.01 (−0.04 to 0.01)	0.24	−0.089
Model 2	NASI score	−0.02 (−0.04 to 0.01)	0.16	−0.109
Model 3	NASI score	−0.02 (−0.04 to 0.01)	0.21	−0.099
Model 4	NASI score	−0.02 (−0.04 to 0.01)	0.23	−0.094
Model 5	NASI score	−0.01 (−0.04 to 0.01)	0.25	−0.089
Model 6	NASI score	−0.02 (−0.04 to 0.01)	0.22	−0.096

Note: Model 1 adjusted by sex, Model 2 adjusted by sex and age, Model 3 adjusted by sex and length of the diseases, Model 4 adjusted by sex and type of chronic disease, Model 5 adjusted by sex and BMI, Model 6 adjusted by sex and Education level.

Abbreviations: CI = confidence interval, NASI = Nutrition Adherence Self‐Report Instrument.

In addition, adding the new covariates, the p‐value was non‐significant (p > 0.09) and the standardized beta was insufficient (standardized beta < −0.257). A comparison of all models led to the selection of Model 1 as the final model (adjusted R ² = 0.011) (see Table 3 and File S3).

4. Discussion

The present study assessed the effect of tele‐MNT services on chronic patients' NA during the COVID‐19 pandemic, before and after the intervention. The findings showed a considerable difference in the NA of chronic patients using tele‐MNT versus face‐to‐face MNT. In this regard, a systematic review and meta‐analysis has confirmed the effectiveness of the tele‐MNT program in chronic patients' nutrition quality and adherence compared to the face‐to‐face MNT [35]. The present findings revealed a significant relationship between participants' NA and sex. However, this study did not show any significant relationship between NA and age, duration of disease, type of chronic diseases, BMI, and education level.

The present study showed a high level of NA in chronic patients after the tele‐MNT intervention. A pilot randomized controlled trial also showed that a tele‐nutrition intervention was able to increase the NA of cardiovascular patients in the intervention group up to a satisfactory level [36], which is consistent with the present study's main finding, though there is a certain difference. In the above‐mentioned study, NA was a secondary objective in the trial while this study conducted a pragmatic trial in a real environment and reached conclusive results (Hedges' g: 3.76 [3.40–4.12]). Overall, both studies led to similar findings.

The present study did not reveal a considerable difference between NA in men and women, though the NA score of men was 0.54 higher than women. A similar study in Nepal on diabetic patients' NA also confirmed the present findings [37]. However, a study in Brazil and Nigeria contradicted the present findings [38, 39]. In these studies, diabetic women had a higher NA than diabetic men. Due to cultural differences in dietary patterns, living conditions, and the complexity of variables involved in NA, the results regarding sex and NA in different studies are inconclusive.

Although the present study showed no significant relationship between NA and age, after adjusting for age and sex, for each unit of increase in NA before the intervention, the patient's NA decreased by 0.017 after the intervention. Also, in the 61–70‐year age group, the NA score decreased by 0.27 compared to the reference group (≤ 50) after the study. In the age group over 71 years, the NA score was 0.12 lower than the reference group. These results show that with increasing age, the NA decreased. In line with these findings, Norton et al.'s cohort study on older Irish adults has confirmed this result [40]. However, the study conducted on the adult population in Spain showed that with increasing age, NA increased [41]. Some studies found that the reasons for the increased NA with age were cultural. The ease of preparing and consuming food at home can be another reason [42, 43, 44]. This finding differed from the results of the present study. Besides cultural factors, the time of study is another reason for the difference. As the present study was conducted during the COVID‐19 pandemic, this particular inconsistency may have been caused by social restrictions, lockdown, and lacking support from other family members [45].

In this study, no significant correlation was found between NA and BMI. Yet, as Hedges' g showed, when age and BMI were adjusted, for every one‐unit increase in NA before the intervention, the patients' NA after the intervention decreased by 0.01. Also, the NA of the overweight decreased by 0.03 after the study, compared to the normal peers (BMI < 25). In the fat group, the results were quite the opposite. There was an increase in NA for people with BMI ≥ 30 by 0.2. These results are inconsistent with the study of Norouzi et al. and Maskarinec et al. [17, 46]. They showed a relationship between increasing BMI and consuming high‐calorie foods such as fats and oils. It implies that those with a high BMI fail to success adhere to their disease‐specific diet. Though the result of the present study could not provide conclusive evidence, however, one reason for this conflict could be the different methodologies of the studies. The present study's design was a before‐after trial, while that of Norouzi et al and Maskarinec et al. was observational (cross‐section). In addition, in the present study, patients were provided with individualized nutrition therapy based on their dietary history, medical diagnoses, and lab test results by registered nutritionists, which can be a reason for divergent findings.

The present study also failed to find a significant relationship between education level and NA. However, patients with less than a diploma level of education had an increase of 0.15 in their NA compared to the reference group (the illiterate) after the intervention. While in the high school diploma group, a decrease of 0.09 NA was observed compared to the reference group after the intervention. In line with this finding, a systematic review also showed that health literacy is rarely effective in adherence to nutrition and other nutritional behaviors, especially in patients [47].

Patients with chronic diseases play a key role in the self‐management of disease. During the COVID‐19 pandemic, chronic patients may have concerns about visiting centers infected with COVID‐19. One way to change the patients' lifestyles and nutrition is to use telehealth. A strength of the present study is that using tele‐MNT during the COVID‐19 pandemic helped patients keep the continuity of care. In this regard, a study on the importance of telehealth for patients with chronic diseases during the COVID‐19 pandemic showed one benefit of telehealth during the pandemic is to provide continuity of care for patients after the pandemic [20]. This study argued that successful disease control in patients with chronic diseases in the pre‐pandemic era, mediated by telehealth, can hopefully continue to be effectively used with no problem or failure after the pandemic.

5. Limitations and Future Studies

Similar to other research, the present study faced some limitations. The study's design was the first limitation (one group before‐after trial). One reason for choosing this design was that this study was conducted during the COVID‐19 pandemic, and the participants were the elderly with chronic diseases, at a high risk of virus infection. Therefore, we could not have a parallel face‐to‐face group for comparison. To compensate for this limitation, we used a pragmatic trial and included a large sample size from two general hospitals in a big city. As a wide range of patients from different social classes often visit these two general hospitals, the findings can be generalized to other cultures and geographical areas. Another limitation was that some patients could not understand the advice given due to their age. To solve this problem, nutritionists created a warm and friendly relationship with patients and devoted more time to providing easy and comprehensive advice to them. The third limitation was the high attrition rate of patients in the first phase of completing the NASI. One reason for the high attrition rate in this study was the timing of recruitment. The recruitment for this trial coincided with the peak of the Covid‐19 pandemic in Iran. It is important to note that many trial interventions during this period were impacted by a high attrition rate [48]. Unfortunately, due to the timing of the recruitment, this study also faced this challenge that could not be avoided. Furthermore, another reason was that these patients may have considered the MNT as a luxury that they found unnecessary. Thus, they did not contribute well after the first phase.

As the present study was conducted in a developing country, internet access faced numerous issues. Therefore, to provide telehealth services, the intervention was made on the phone as the simplest means of telecommunication for doctor–patient communication [49, 50]. Future research is suggested to focus on other telehealth media, such as video‐conference, which can facilitate video communication. Moreover, the study's design was a before‐after trial with only one group. Other researchers can use another design, such as a randomized controlled trial.

6. Conclusion

Considering the present findings, it can be concluded that there is a considerable difference in NA of chronic patients before and after the tele‐MNT intervention. In addition, the present study showed a considerable difference in NA between men and women. During the COVID‐19 pandemic, telehealth programs such as tele‐MNT could contribute to the continuity of care and the gains made for disease control in chronic patients in the pre‐pandemic era with no gaps or failures. The present findings can help health policy‐makers provide better healthcare services to improve the continuity of care for chronic patients during the pandemic. Future research could take advantage of this study to focus on other telehealth platforms, such as video‐conference, which can facilitate video communication. As a primary objective in the other studies, they can adopt different aims, including the effect of attendance rate and minutes per visit on NA.

Author Contributions

Somaye Norouzi, Fateme Arefi Majd, Somaye Rostami, Samane Sistani, and Leila Ahmadian conceptualized, designed, and conducted the study. Somaye Norouzi drafted the manuscript with significant intellectual input from Leila Ahmadian and assisted with revising the article. The analysis and interpretation were done by Somaye Norouzi and Moghadameh Mirzaee.

Ethics Statement

The study was conducted following the Helsinki Declaration and Ethics Publication on Committee (COPE). Before starting the intervention, this study was approved by the ethics committee of the National Institute for Medical Research Development (NIMAD) (#IR.NIMAD.REC.1399.140). Eligible patients were contacted by phone and asked to participate in the study. A research team member explained the study's purpose and obtained oral informed consent. This oral informed consent was approved by the NIMAD. Appointments were made with patients who were willing to participate in the study. Participation was completely voluntary and the patients could withdraw from the study at any time and receive face‐to‐face MNT services at the hospital. All questionnaires were completed anonymously, and the patients were assured that the confidentiality of data would be preserved. The counseling was offered in a quiet and peaceful place, where nobody else was present. The questionnaires were also completed in similar conditions.

Conflicts of Interest

The authors declare no conflicts of interest.

Transparency Statement

The lead author Leila Ahmadian affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Supporting information

Supporting information.

Supporting information.

Supporting information.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.