May 2022, Volume 72, Issue 5

Research Article

Development of a protocol for conducting a randomized control trial on effects of artificial intelligence on nutritional status of children post cardiac surgery

Maryam Zahid  ( Al Shifa Trust Eye Hospital, School of Public Health, Rawalpindi, Pakistan. )
Ume Sughra  ( Al Shifa Trust Eye Hospital, School of Public Health, Rawalpindi, Pakistan. )

Abstract

Objectives: To assess the effect of diet-related mobile application based on artificial intelligence on the nutritional status of children post-cardiac surgery, and to compare their diet-related problems with their counterparts having the usual diet.

 

Method: The two-arm double-blind randomised controlled trial will be conducted at a tertiary care hospital in Rawalpindi, Pakistan, over an expected duration of 6 months from March to August 2021. Intervention group A will be given a diet-related mobile application based on artificial intelligence, while usual-care group B will be handed a pamphlet with instructions related to post-operative diet on discharge.

 

Results: The findings will improve perception about the influence of artificial intelligence on nutritional status of children post-cardiac surgery. If proven to be effective, this mobile application can be used in other hospitals.

 

Conclusion: The study protocol will give an indication that whether diet-related mobile application can contribute to improving the nutritional status of children post-cardiac surgery. As the pandemic has forced people to minimise hospital visits, this is the right time to evaluate the utility of such an application.

 

Keywords: Artificial intelligence, Diet-related mobile application, Nutritional status, Children post-cardiac surgery, Randomised controlled trial.

Study was registered on clinicaltrial.gov with trial identity number NCT04782635.

(JPMA 72: 908; 2022) DOI: https://doi.org/10.47391/JPMA.3751

 

 

Introduction

 

Malnutrition is found in 15-50% children, and is the most common problem in paediatric patients having congenital heart diseases (CHDs).1 Even in the developed nations, underfeeding is common in critically ill children,2 whereas in the developing countries, the incidence of malnourishment may be even higher.3

In Pakistan, 17.7% children suffer from wasting.4 In a study,5 21% patients had in-hospital mortality. Mortality in normal-weight babies was 3.4%, but increased to 47% in underweight, and 49% in severely underweight patients.

When devoted nutrition support teams have focussed on energy and protein intake, it produced better postoperative outcomes in high-income countries.6 In Pakistan, there is shortage of workforce to provide such knowledge in hospitals, so a mobile application would be helpful in creating nutritional awareness.6

Health and fitness-based mobile applications have recently emerged in the smartphone market in large numbers.7 They play an important role in tracking the intake of foods and drinks, which may help individuals improve understanding of their dietary patterns.8 Despite being economically beneficial, dietetic counselling requires  plenty of time.9 Mobile health applications have become accessible,10 and present opportunities to increase the reach of dietetic services to those who may be unable to reach dietary clinics.11

The proposed clinical trial will assess the effectiveness of a diet-related mobile application based on artificial intelligence (AI) on the nutritional status of cardiac postoperative children, and to compare their diet-related problems with their counterparts having the usual diet and care.

 

Methods

 

The two-arm double blind randomised controlled trial (RCT) will be conducted at a tertiary care hospital in Rawalpindi, Pakistan, over an expected duration of 6 months from March to August 2021. The RCT was registered with the clinicaltrial.gov (NCT04782635), and the recommendations of the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT)12 will be followed as shown in Table-1.

 

 

The patient's enrollment will take about 10-14 weeks at the rate of 7-8 patients per week. The intervention is expected to be given for at least 8 weeks per patient. A specific inclusion/exclusion criteria will be followed (Table-2).

 

 

Patients using usual care will be blinded from those using the application. The random allocation of enrolled patients into the two groups will be performed using permuted block randomisation, making sure 1:1 ratio in intervention group A and usual-care group B. Recruitment of the patients will be carried out using the chronologically numbered sealed envelopes.

The primary objective of the diet-related mobile application would be to improve or maintain the nutritional status of the children. Children/caregivers will be able to choose their goals and challenges, daily intake of calories, carbohydrates, proteins, fats, minerals and vitamins. Nutritional intake will be tracked and daily and monthly insight regarding nutrients intake will be provided to the user. Meal and supplement reminders will be given to the users as per their requirements. Recipes will be available and meals can be planned according to the food choices of the child.

On discharge from hospital, the guardian/caregivers of children in group A will be explained the intervention and written consent will be requested. The mobile application will be installed in their mobile phones and they will be given the basic knowledge regarding the working of the application. Personal information will be entered in the mobile application, which will produce a diet plan for the child.

The child will then use the mobile application for 4 weeks after which, during the routine visit, the caregiver will be asked about the effectiveness of the application, and the nutritional status of the child will be checked. The patient will follow the diet plan provided in the mobile application and will visit the doctor for routine check-up after another 4 weeks. During the hospital visit, the patient will be asked about the effectiveness of the application, and nutritional status of the patient will be checked.

in group B, the children's weight, height, clinical signs will be noted on discharge. A pamphlet with dietary information will be given to the caregivers. During the routine hospital visit after 4 weeks, the nutritional status of the patient will be checked. The same procedure will be followed when the patient will visit doctor after another 4 weeks.

The sample-size will be calculated using results from a previous study13 in which wasting was reduced by the end of the study in 12% subjects in the intervention group, but there was no marked difference between the intervention and control groups at the end of the study. The sample size is likely to be 88; 44(50%) in both groups, with 80% power, 5% significance level and predicted dropout 25%.

In terms of primary outcomes, weight changes in pounds (lbs), average caloric intake in kcal and average protein intake will in gm will be observed / calculated in both the groups for 8 weeks.

Patients will be requested to provide input regarding the usefulness of mobile application on Likert scales. The language used in the application as well as the usage of application itself will be rated from 'easy' to 'difficult'; financial effectiveness of the app will be rated from 'yes' to 'very costly'; insight about calories and nutrients will be rated from 'useful' to 'difficult to understand'; the recipes provided in the application will be rated from 'being useful' to 'not useful'; meal and supplement reminders will be rated from 'being useful' to 'irritating'.

Patients will be inquired about any nutritional issues faced during that period, like episodes of vomiting, diarrhoea, loss of appetite and other nutrition-related problems.

Secondary outcome measures would include average carbohydrates and fats consumed in both groups which will be recorded in gm for 8 weeks, and average vitamins and minerals consumed which will be recorded in mg and µg for 8 weeks.

 

Results

 

Ethical and institutional permission is being pursued. The recruitment of the subjects is planned, and the data-collection process is expected to be completed by July 2021. Results will be analysed subsequently. The research work is anticipated to be published within three months post- analysis.

 

Discussion

 

Malnutrition is common in CHD patients. Mehta et al. defined malnutrition as the difference between the nutrients requirement and the intake, which results in a collective insufficiency of energy, protein and micronutrients that may adversely affect growth and development.14

According to Becker et al., assessment of nutritional status and examinations on the role of nutrition in patient's health management should include data over a period of time. The data should include anthropometrics, actual dietary intake and reason for malabsorption.15

Application-based mobile health interventions are a predominantly favourable method of modifying nutrition behaviours and health outcomes related to nutrition due to increased universal smartphone penetration and the easiness of installing apps in the mobile devices.16

 

Conclusion

 

The use of mobile health (mHealth) technology in the dietetic field has not been studied comprehensively and there has not been any RCT planned in Pakistan involving the effect of diet-related mobile application on nutritional status of children after cardiac surgery. The study protocol will give an indication that whether a diet-related mobile application can improve the nutritional status of children post-surgery. As the pandemic has forced the people to minimise hospital visits, this is the right time to evaluate the usefulness of such an application.

 

Disclaimer: The text is based on an academic research.

Conflict of Interest: None.

Source of Funding: None.

 

References

 

1.       Blasquez A, Clouzeau H, Fayon M, Mouton JB, Thambo JB, Enaud R, et al. Evaluation of nutritional status and support in children with congenital heart disease. Eur J Clin Nutr 2016; 70: 528-31.

2.       Mehta NM, Bechard LJ, Zurakowski D, Duggan CP, Heyland DK. Adequate enteral protein intake is inversely associated with 60-d mortality in critically ill children: a multicenter, prospective, cohort study. Am J Clin Nutr 2015; 102: 199-206.

3.       Hassan BA, Albanna EA, Morsy SM, Siam AG, Elsaadany HF, Shehab M, et al. Nutritional Status in Children with Un-Operated Congenital Heart Disease: An Egyptian Center Experience. Front Pediatr 2015; 3: 53.

4.       National Nutrition Survey, Key findings report, nutrition wing, ministry of national health services, Regulations and Coordination Government of Pakistan; 2018

5.       Zahid M, Khalil H, Shah N, Nasim S, Majeed K, Sadiq T. Association between weight for age and clinical outcomes in children undergone cardiac surgery. Pak Armed Forces Med J 2019; 69 (Suppl-3): S409-13

6.       Martinez EE, Bechard LJ, Mehta NM. Nutrition algorithms and bedside nutrient delivery practices in pediatric intensive care units: an international multicenter cohort study. Nutr Clin Pract 2014; 29: 360-7.

7.       Pohl M. Research Guidance. research2guidance; 2017.

8.       Azar KM, Lesser LI, Laing BY, Stephens J, Aurora MS, Burke LE, et al. Mobile applications for weight management: theory-based content analysis. Am J Prev Med 2013; 45: 583-9.

9.       Health and Care Professions Council Dietitians; 2020.

10.     Research2guidance. mHealth App Developer Economics; 2015.

11.     Martin CK, Gilmore LA, Apolzan JW, Myers CA, Thomas DM, Redman LM Smartloss: A Personalized Mobile Health Intervention for Weight Management and Health Promotion JMIR Mhealth Uhealth 2016; 4: e18.

12.     Chan AW, Tetzlaff JM, Altman DG, Laupacis A, Gøtzsche PC, Dickersin K, et al. SPIRIT 2013 Statement: Defining standard protocol items for clinical trials. Ann Intern Med 2013; 158: 200-7.

13.     Saleem, AF, Mahmud S, Baig-Ansari N, Zaidi AK. Impact of maternal education about complementary feeding on their infants' nutritional outcomes in low- and middle-income households: a community-based randomized interventional study in Karachi, Pakistan. J Health Popul Nutr 2014; 32: 623-33.

14.     Mehta NM, Corkins MR, Lyman B, Malone A, Goday PS, Carney LN, et al. American Society for Parenteral and Enteral Nutrition Board of Directors. Defining pediatric malnutrition: a paradigm shift toward etiology-related definitions. JPEN J Parenter Enteral Nutr 2013; 37: 460-81.

15.    Becker P, Carney LN, Corkins MR, Monczka J, Smith E, Smith SE, et al; Academy of Nutrition and Dietetics; American Society for Parenteral and Enteral Nutrition. Consensus statement of the Academy of Nutrition and Dietetics/American Society for Parenteral and Enteral Nutrition: indicators recommended for the identification and documentation of pediatric malnutrition (undernutrition). Nutr Clin Pract 2015; 30: 147-6.

16.     de Morais AM, Machado LS, Valenc¸a AMG. [Serious games in dentistry: applications, features and possibilities]. XII Brazilian Congress of Health Informatics; 2010.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: