https://orcid.org/0000-0001-8155-5144
Figures
Abstract
Introduction
The healthy plate model (HPM) is a practical guide to modulate the portion of staple food in main meals, subsequently affecting the risks associated with Non-communicable Diseases include type2 diabetes mellitus (T2DM).
Objective
This study investigated the effectiveness of health information and the healthy plate model on cardio-metabolic risk factors, knowledge and attitude towards T2DM prevention measures.
Methods
A pre-post analysis, as part of a cluster randomized trial with street food vendors and their customers, was implemented in three randomly selected districts in Dar es Salaam, Tanzania. Two vendor-customer clusters each with 15 and more vendors from each district were randomly assigned to receive either T2DM health information only (Intervention package1 [IP1]) or IP1 plus a subsidized meal with vegetables and fruits, following the principles of the HPM (Intervention package2 [IP2]). Within the clusters the participants were informed on the importance of the intervention they received. An intervention period lasted for three months from 1st April to 31st June 2019. We applied Generalized Linear Mixed Models and Bayesian Modelling (for sensitivity analysis) to assess the effectiveness of the interventions.
Results
Overall, 336 (IP2 = 175 and IP1 = 161) out of 560 (280/arm) previous study participants participated in evaluation. Diastolic BP was lower among IP2 participants in the evaluation than baseline AβC = -4.1mmHg (95%CI:-5.42 to -2.76). After adjusting for the interaction between IP2 and age of the consumers, the BMI was significantly lower among IP2 in the evaluation than baseline AβC = -0.7kg/m2 (95%CI: -1.17 to -0.23). With interaction between IP2 and income, BMI was higher in the IP2 in the evaluation than baseline AβC = 0.73kg/m2 (95%CI: 0.08 to 1.38). Systolic and diastolic BP were significantly lower among IP1 in the evaluation than baseline AβC = -3.5mmHg (95%CI:-5.78 to -1.24) and AβC = -5.9mmHg (95%CI:-7.34 to -4.44) respectively. Both the knowledge scores and positive attitudes towards T2DM prevention measures were higher in the evaluation than baseline in both interventions arms.
Conclusion
The positive effects on cardio-metabolic risk factors, knowledge and attitude were observed in both intervention arms. Due to interactions between IP2, age and income; designing interventions relating to food and cardio-metabolic risk factors, should consider combining socio-economic factors.
Citation: Kagaruki GB, Mahande MJ, Mayige MT, Kreppel KS, Ngadaya ES, Haydon D, et al. (2023) The effectiveness of interventions to reduce cardio-metabolic risk factors among regular street food consumers in Dar es Salaam, Tanzania: The pre-post findings from a cluster randomized trial (Registered by Pan African clinical trial registry with trial # PACTR202208642850935). PLoS ONE 18(11): e0289289. https://doi.org/10.1371/journal.pone.0289289
Editor: Aleksandra Klisic, University of Montenegro-Faculty of Medicine, MONTENEGRO
Received: August 9, 2022; Accepted: July 8, 2023; Published: November 15, 2023
Copyright: © 2023 Kagaruki et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The National Health Research Ethics Committee (NatHREC) has developed a guideline that before data are taken outside Tanzania, clearance is sought to NatHREC by completing a Data Transfer Agreement (DTA) which is signed by both parties, provider and recipient of data stating the reason for transfer and which data are being transferred even though they could be de-identified data. Data request should be sent to the head of the institution that conducted the research, in this case National Institute for medical research. The institution will initiate the DTA signing processes between the two parties before submitting it to the National Health Research Ethics Committee for approval. The contact details for the head of the institution, for requesting the data: Director General, National Institute for Medical Research, email: dg_office@nimr.or.tz.
Funding: This study was funded by DELTAS Africa Initiative [Afrique One-ASPIRE /DEL-15-008]. Afrique One-ASPIRE is funded by a consortium of donor including the African Academy of Sciences (AAS) Alliance for Accelerating Excellence in Science in Africa (AESA), the New Partnership for Africa’s Development Planning and Coordinating (NEPAD) Agency, the Wellcome Trust [107753/A/15/Z] and the UK government. Gibson B. Kagaruki received partial support from Novo Nordisk to travel to Glasgow and Liverpool Universities for a four months attachment course on Advanced Statistics analysis and Modelling; knowledge and skills gained from this course informed analysis of this paper. The authors also acknowledge their institutions and organizations that maintained their salary during the period of executing the study. These institutions include National Institute for Medical Research and Kilimanjaro Christian Medical University College. Others institutions include Centre Suisse de Recherches Scientifiques en Côte d’Ivoire, Abidjan, Côte d’Ivoire. Some of authors received fully or part of salary from Afrique One-ASPIRE these include Gibson B Kagaruki, Katharina S. Kreppel and Bassirou Bonfoh. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Gibson B. Kagaruki received partial support from Novo Nordisk to travel to Glasgow and Liverpool Universities for a four months attachment course on Advanced Statistics analysis and Modelling. There are no patents, products in development or marketed products associated with this research to declare. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
In urban areas, a rising number of people consume meals outside their homes, most often unhealthy foods, i.e., food high in fat, sugar, and salt, hence increasing their risks for developing non-communicable diseases (NCDs) including type 2 diabetes [1–5]. This change in landscape has resulted in a transition of disease epidemiology, as developing countries are facing a growing double burden of infectious diseases and NCDs [6, 7]. Presently, in Tanzania, NCDs are responsible for 33% of all deaths [8]. The cardio-metabolic risk factors, overweight and obesity, raised blood glucose, raised blood pressure (BP), raised triglycerides, and low high-density lipoprotein (HDL), are intermediate predictors of type 2 diabetes [9]. Intermediate risk factors, to a large extent, can be influenced by behavioural risk factors including unhealthy eating practices, physical inactivity, smoking and harmful alcohol consumption [10–14]. Findings from the baseline data collected among regular street food consumers in Dar es Salaam in a previous study linked to the current analysis, revealed high levels of cardio-metabolic risk factors including raised BP (42%), overweight/obesity (62%); raised triglycerides (13%) and raised blood glucose (6.6%) [15]. Healthy plate model (HPM) is a practical guide to modulate the portion of staple food in main meals [16]. Meals designed after the HPM have a protective effect from NCDs including type2 diabetes and cardiovascular diseases [16]. However, getting a healthy plate from street food vendors remains a challenge. Evidence shows that street food meals are served with only small amounts of vegetables and fruits or none at all. Servings of street food are also associated with larger amounts of carbohydrate and protein, thus exposing the consumers to excess calorie intake [17, 18] and increasing cardio-metabolic risks [19–22]. Despite an increase in cardio-metabolic risks factors, in Tanzania, effectiveness of available interventions on cardio-metabolic risks, knowledge, and attitude of the populations at risk of NCDs, including regular consumers of street food, have not been fully explored. These potential interventions include the HPM, which was validated by nutrition experts from the Harvard School of Public Health [23], and a type 2 Diabetes Mellitus (T2DM) health information package currently promoted by the Ministry of Health of Tanzania [24].
The current study involved implementation of two interventions namely: i) intervention package 1 (IP1) consisting of T2DM health information only and ii) Intervention Package 2 (IP2) consisting of T2DM health information and serving of a meal subsidized with vegetables and fruits following the principles of the HPM. Later, the study evaluated if IP1 and or IP2 are effective in reducing cardio-metabolic risk factors, increasing knowledge on T2DM and affecting attitudes towards T2DM prevention measures, as compared to a baseline (pre-post analysis).
Material and methods
Study design and setting
This pre-post analysis as part of a cluster randomized trial study was implemented from April 2018 to September 2019 in three out of five randomly selected districts in Dar es Salaam city. The study setting has been previously described in detail in our baseline study [15]. Dar es Salaam is one of the fastest growing urban centres in Africa [25, 26] and with a large population dependent on street food [15, 17, 27], it presented an ideal study site. Another selection criteria was the presence of the headquarters of all the country’s food regulatory authorities, including the Tanzania Food and Nutrition Centres (TFNC), Tanzania Bureau Standards (TBS) and Tanzania Medicines & Medical Devices Authority (TMDA) during the time of the study. In each district, two street markets were chosen for the study. These included: Kawe and Tandale in the Kinondoni district, Mbezi and Manzese in the Ubungo district and Ilala Boma and Kisutu in the Ilala district.
Study population and sample size calculation
A street food consumer is defined as a person who buys, and eats food or drink prepared in the street or at a supplier’s home, which is then sold in the street/public areas [28, 29]. Using its own operationalized definition, the study defines a regular street food consumer as a person who consumes at least three street vended lunches per week. The sample size for regular street food consumers was calculated by considering the following statistical parameters: level of significance (Z1-α) of 95%CI, power of the test (Z1-β) of 80, population standard deviation (δ) and population variance (δ2). The baseline indicator for the mean waist hips ratio of μ0 = 0.82 and a target of reducing the mean waist ratio by 5% after the intervention i.e., from 0.82 to μa = 0.78. A design effect (DE) of 1.5 to address variance between sites was applied and a non-response rate of 20% was estimated. With the above statistical parameters, and the applied sample size (n) calculation formula [30], the desired sample size was 280 participants. This sample size was doubled to include both arms (IP1 and IP2).
Eligibility criteria for market site, street food consumers and vendors
A market site was eligible for the study if it had at least 15 street food vendors. Regular street food consumers: A consumer was eligible for the study if he/she met the following criteria: being aged 25–64 years; consuming at least three lunches per week at the same street food vendor; having no plans to move out of the study area in the next 12 months; and have been consuming street food for not less than a year. Lactating mothers and pregnancy women were not eligible. Eligibility criteria for street food vendors included being aged 18 or above; has been vending food at the current site for at least 12 months; has at least seven customers who have been consuming at least three lunch meals per week for at least 3 months; is ready to implement the components required for an intervention; and will continue vending food at the same market site for at least another 12 months.
Allocation of study market sites/clusters
Three out of five districts included Kinondoni, Ubungo and Ilala, which were randomly selected from Dar es Salaam city. From each sampled district, the study mapped all markets sites (clusters) then two markets were selected from each district randomly. In each district, one cluster was randomly assigned to receive IP1 and the second cluster to receive IP2 using a simple randomization technique [1] by an independent statistician.
Recruitment of vendors and consumers
Street food vendors were mapped at each selected market site. From 152 food vendors in 6 selected markets (sites), 58 food vendors were randomly selected; Ilala (13), Ubungo (23) and Kinondoni (22). In each selected market, 90 regular street foods consumers were recruited with a minimum of 7 regular consumers per vendor. Finally, 280 consumers were allocated into IP1 market sites and 280 were allocated to IP2 market sites. Allocation of both the clusters and study participants in the two study arms was at a ratio of 1:1 (Fig 1).
*Number of food consumers considered were the regular street food consumers i.e. those who consumed three or more lunches per week from the same food vendors; # stands for “number”.
The recruitment of study participants including seeking consent took place from November 2018 to March 2019. The process involved awareness creation and sensitization of street food consumers and vendors with information education communication (IEC) such as brochures and banners. In this phase the principle investigator together with the food vendors and their leaders were also screening eligible participants. Eligible vendors were asked to provide the names and contact details of their regular clients. This was possible because regular consumers typically call vendors to have food delivered to their workplace. Vendors were asked to list all of their regular customers to avoid selection bias; vendors were informed in advance that only consented customers will be included in the study. With the customers’ list, the principal researcher approached each of the customers and gave information about the study and sought informed consent. Finally, a list of eligible and consented participants per cluster and vendor was obtained. All eligible and consenting participants were informed about the start and end date of the intervention. Specifically, the intervention period lasted for three months from 1st April to 31st June 2019.
Intervention packages
Intervention package 1 –IP1 (T2DM health information).
This study used the One Health approach [31] for delivering T2DM health information to participants; the interdisciplinary team involved a physiologist from the National Hospital-Muhimbili, an epidemiologist from the National Institute for Medical Research and a nutritionist from the Dar es Salaam Regional Office. The participants were informed on the behavioural, familial, and metabolic risk factors for type2 diabetes and on prevention measures of diabetes. These measures included the recommended level of metabolic equivalent task per week (MET/Week) of physical activities, the recommended limit of alcohol to drink per week for each sex, and the recommended amount of fruit and vegetable servings to be consumed each day. Furthermore, the participants were informed of the risk of smoking and eating excess fats, carbohydrates, and fruit high in calories. The signs, symptoms, and complications of type2 diabetes were also explained to the participants. The channels for delivering T2DM key health information included group meetings, text messages on their mobile phones, and the distribution of brochures and banners.
Intervention package 2-IP2 (T2DM health information and healthy plate model).
This package comprised two components, T2DM health information and a subsidized plate model, enabling vendors to sell subsidized meals following the plate model principles. The study adopted a plate model validated by nutrition experts from the Harvard School of Public Health [23]. The pictorial healthy plate model illustrates the recommended portions of the main food groups namely carbohydrate (1/4 of the plate), protein (1/4 of the plate), fruits and vegetables (1/2 of the plate), per person/meal [23]. To ensure optimal adherence to the plate model, the project subsidized costs (0.45USD/consumer/day) of fruits and vegetables in the three market sites receiving IP2. The consumers and vendors from the IP2 arm were informed on how to prepare the recommended food plate, and on its potential role in reducing cardio-metabolic risk factors. It was envisaged that knowledge of the healthy plate model would influence practice and increase uptake of an intervention among consumers. The benefits of serving dishes corresponding to the healthy plate model were explained to the vendors, and these benefits included serving smaller amounts of staple food commonly available in Tanzania [32], these groups include carbohydrate (1/4 of the plate of rice, banana, stiff porridge/ugali, sweet or round potatoes etc.) and balancing other food groups including fruit (sweet bananas, mangoes, oranges, pineapples, water melon, pawpaw etc.), vegetables (spinach, amaranth, pumpkin and cassava leaves etc. (1/2 of the plate) and protein such as fish, white and red meat, beans etc. (1/4 of the plate).
Implementation steps.
Implementation of this study involved four steps. In the first step, baseline data was collected to serve as control from August to September 2018. This phase involved collection of information of socio-economic and demographic characteristics, NCDs’ behavioural risks, cardio-metabolic risks, familial risks, knowledge on T2DM and attitude towards T2DM prevention measures. Phase 2 ran from November 2018 to March 2019, which involved awareness creation and sensitization of street food consumers and vendors with information education communication (IEC) such as brochures and banners. Awareness creation also involved group meetings, training of street food vendors on how to prepare a balanced diet and serving a “model plate” to the IP2 group. The messages included in the IEC materials were adapted from information and education communication materials developed and validated by TANCDA in collaboration with the Ministry of Health (MoH) of Tanzania TANCDA et al. [Unpublished]”) [24]. The third step lasted for three months from April to June 2019. In this phase, the interventions were implemented, with the health information provided to street food consumers on T2DM for one group, and health information on T2DM and the healthy plate model served to the other. The last step ran from August to September 2019 and involved the collection of evaluation data among study participants.
Study conceptual framework.
The study summary concept is described in Fig 2. The assumption is that the population in general is healthy. However, after being exposed to socio-economic and non-modifiable risks factors, that population becomes vulnerable to behavioural risks. These include smoking, excessive alcohol consumption, physical inactivity, and unhealthy eating practices, among others. The hypothesis is that if no strategic prevention measures are put in place to reduce these risks, behavioural risks trigger intermediate risks such as low HDL, raised triglycerides, overweight/obesity, raised blood pressure and blood glucose, and as a result the population becomes vulnerable to developing T2DM. The intervention packages (IP1 and IP2) aim to increase positive attitude towards behavioural prevention measures and improve knowledge on diabetes so as to reduce risk behaviours. This will work toward the ultimate goal to reduce intermediate risks factors and in turn aid the population in preventing the development of T2DM.
Study variables
Dependent variables.
In this study, the primary outcome was cardio metabolic risk factors (i.e., BMI, blood glucose, triglycerides, and waist circumference, systolic and diastolic BP). Correct knowledge on T2DM and positive attitude scores towards prevention measures of T2DM were the secondary outcomes of interest. All dependent variables were modelled as continuous variables.
Independent variables.
The independent variables include age, sex, marital status, income, education level and exposer status to intervention (baseline and IP1 or IP2); these were considered as fixed independent variables. The two random effects included were market sites and participant identification numbers.
Effectiveness/impact of intervention.
The goal was to run pre-post analysis to assess if there was a significant difference in primary and secondary outcomes between IP1 and baseline and IP2 against baseline (pre-post comparative analysis).
Data collection and management.
The data collection team involved three laboratory technicians, two nurses, two sociologists, one demographer, three medical doctors and a statistician. They had similar levels of experience on the studied topic. However, they participated in a three-day training conducted by principal investigator before the baseline and evaluation phase, which involved briefing them on the study protocol, data collection tools and ethical considerations. In the evaluation phase, the same procedures as for the baseline data collection were adhered to [15]. Information collected included socio-economic and demographic characteristics, cardio-metabolic risk factors: weight and height for body mass index, blood pressure (BP), fasting blood glucose and triglycerides. All tests and measurements were undertaken as per recommended standards and operating procedures [9, 33–40]. Blood samples were taken and stored in a cool box and shipped immediately after collection to St. Laurent Diabetes Centre, Dar-es-Salaam, for analysis on the same day by trained laboratory technicians. Data was collected electronically using the Open Data Kit (ODK) system [41]. Data was cleaned and later synchronized on the main server at the National Institute for Medical Research Muhimbili Centre in real-time. The anonymity of the participants was ensured by assigning a unique identifier to each participant. Further details regarding data collection procedures are published elsewhere [15].
Twenty-four (24) items were adopted from different grey literature and used to assess the knowledge level of each regular street food consumer [37, 42–46]. All items have been used previously in similar studies that were conducted with different populations (HIV positive clients and the general population) [39, 47] in Dar es Salaam. Possible responses for knowledge assessment under each sub domain were listed and if mentioned by the respondent, a “yes” response was recorded and “no” if not mentioned in each respective domain. A weight of “1” was given to each mentioned item and “0” if not mentioned. Participants were also assessed on their knowledge on the daily-recommended quantity of foods from different food groups (protein, fats, carbohydrates, fruits and vegetables, high-fibre food, red meat, glycaemic foods, and salted foods). They were asked to judge the quantity of food to be consumed per person per meal as “large,” “average,” “small” or “not sure.” A correct response was given a score of “1” and a wrong or absent response received “0”. This study also assessed the attitude of study participants towards prevention measures for cardio-metabolic and type2 diabetes. This study also involved reviewing of different published articles [37, 42–46] to generate 12 points to assess the attitude of street food consumers towards available prevention measures for NCDs including T2DM, cardiovascular diseases, and heart disease. The participants who agreed with a positive statement and those who disagreed with a negative statement were considered as having positive attitude and were given a score of “1” while those who reported disagreeing with a positive statement and agree with a negative statement were given a score of “0”. Individual absolute scores for both knowledge and attitude absolute scores were summed and converted into percentages. The items used to assess knowledge and attitude have been summarized in Appendices I & II of S2 File.
Statistical analysis
Cleaning and descriptive data analysis were performed using Stata version 15.0 (StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC). Comparisons of both explanatory and outcome variables between intervention arms was performed using both Chi-Square tests for categorical variables and T-tests for continuous variables to assess differences between IP1 and IP2 at the baseline phase. Explanatory variables were categorized into random and fixed variables, to overcome dependence problem resulting from repeated measures. The study applied Generalized Linear Mixed Models (GLMM) to accommodate fixed and random variables [48–50]. Assessment was done for the interaction between intervention(s) and age, sex, income, and education. Sensitivity analysis was also conducted to assess if variables which were statistically significant in the GLMM model, were also supported using a Bayesian modelling technique fitted with Markov Chain Monte Carlo (MCMC) but without consider prior information. The results reported in detail in this paper are those generated from the GLMM model while the Bayesian model results have been reported as in the Appendix V in S2 File. All advanced analyses were conducted using R-Software. Both lme4 and MCMCglmm were R packages used while lmer and glmer were the commands used to fit the mixed effects models. These packages and commands assume the random effects to be Gaussian distributed, and the random effects were modelled as independent [49, 50]. Model selection; to select the most parsimonious models, likelihood ratio tests were undertaken [51–53]. Correlation between the baseline measure and end time measures repeated on the same sites and the same subjects were expected, so this study also reported between site and within-subject Interclass Correlation Coefficients (ICCs) [54, 55]. In this study, low and high variability in the outcomes results was observed among sites and subjects, respectively (Appendix III in S2 File). To ensure reliable inference, residual analysis was performed and outputs are included as Appendix IV in S2 File, the findings from that analysis showed that there are not substantive deviations from the model assumptions that need concern this study.
Results
Characteristics of the participants
Recruitment and seeking consent from study participants started from November 2018 to March 2019. This stage was followed by intervention and follow up phase which lasted for three months from 1st April to 31st June 2019. Of the 560 enrolled participants, 60% (n = 336) participated in the evaluation. The comparison between the IP1 and IP2 arm by socio-demographic at baseline is shown in Table 1. There were no significant differences for compared variables between arms. The mean age of both IP1 and IP2 participants was equal (45 years) and more than half of the participants in both arms were aged 41–64 years. More than half of the participants in both arms were male and more than 88% had primary level of education. Above three quarters of participants in both arms were married/cohabiting and above 50%, in both arms their monthly income was above mean values. Table 2 shows the comparison of both primary (physical and biomedical measurements) and secondary (correct knowledge and positive attitude scores) outcomes between IP2 and IP1, again there was no significant difference between IP1 and IP2.
The end term evaluation conducted after three months of follow up ended up with 336 (175 form IP2 and 161 from IP1) consumers has been shown in Fig 1, however, at the enrolment each arm enrolled 280 consumers. A total of 224 consumers were registered as lost to follow up due to several reasons including changing business sites, shifting from Dar es Salaam city, and changing food vendors.
Effectiveness of the IP2 on cardio-metabolic risk factors of type2 diabetes (pre-post analysis)
GLMM results indicate that diastolic BP was significantly lower in the evaluation phase than baseline AβC = -4.09mmHq (95% CI: -5.42 to -2.76). Fasting blood glucose was significantly higher in the evaluation phase than the baseline phase AβC = 0.39mmol/L (95% CI: 0.23 to 0.55). After adjusting for the interaction between IP2 and age of the consumers, the BMI was significantly lower in the evaluation phase than baseline AβC = -0.7kg/m2 (95% CI: -1.17 to -0.23). With interaction between IP2 and income, BMI was higher in the evaluation than baseline phase AβC = 0.73kg/m2 (95% CI: 0.08 to 1.38). With interaction between IP2 and income, the waist circumference was significantly higher in the evaluation phase than baseline AβC = 2.07cm (95% CI: 0.15 to 4.00). Considering the interaction between IP2 and age, systolic BP was significantly higher in the evaluation phase than baseline AβC = 2.25mmHq (95% CI: 0.09 to 4.41) (Table 3).
Effectiveness of IP2 on knowledge and attitude (pre-post analysis)
Results from GLMM model showed that positive attitude scores towards NCDs/T2DM prevention strategic measures was significantly higher during evaluation compared to baseline AβC = 18.9%(CI 95%: 15.69 to 22.11). Similarly, knowledge scores on T2DM were significantly higher in the evaluation phase as compared to baseline phase AβC = 13.34% (CI 95%:10.75 to 15.93) (Table 4).
Effectiveness of the IP1 on cardio-metabolic risk factors of type2 diabetes (pre-post analysis)
Systolic and diastolic BP were significantly lower among IP1 consumers in the evaluation than in the baseline AβC = -3.5mmHg (95% CI: -5.78 to -1.24) and AβC = -5.9mmHg (95% CI: -7.34 to -4.44) respectively. Further, fasting blood glucose was higher in the evaluation phase than the baseline phase AβC = 0.49 mmol/L (95% CI: 0.31 to 0.67) (Table 3).
Effectiveness of IP1 on knowledge and attitude (pre-post analysis)
Results from GLMM model showed that positive attitude scores towards NCDs/T2DM prevention strategic measures was significantly higher during evaluation as compared to baseline AβC = 17.73%(CI 95%: 14.30 to 21.16). The results also showed that, positive attitude scores towards T2DM prevention measures was significantly higher in the evaluation phase as compared to baseline phase AβC = 12.63% (CI 95%:9.87 to 15.39) (Table 4).
Sensitivity analysis
Results for effectiveness of interventions on cardio-metabolic risks including overweight/obesity, systolic and diastolic BP, waist circumference and blood glucose as well as attitude towards NCDs/T2DM prevention strategic measures and knowledge scores on T2DM observed from GLMMs were consistent with Bayesian techniques Appendix V (Table A and B) in S2 File.
Discussion
Findings from the current pre-post analysis study show that both T2DM health information and subsidized meal with vegetables and fruits, following the principles of a healthy plate model (IP2) and T2DM health information only (IP1), had equal effect. It was observed that age and income were modulating BMI and blood pressure outcomes among participants in IP2. Positive attitude on T2DM prevention measures increased significantly among participants of both arms. Knowledge on risk factors, symptoms, prevention measures, complications and recommended amount of food groups to eat per meal, increased significantly among participants of both arms.
The current study shows that T2DM health information only as well as T2DM health information plus subsidized meal with vegetables and fruits, following the principles of a healthy plate model are effective in reducing cardio-metabolic risk factors including blood pressure, waist circumference and body mass index. Similar effect was also observed from South Africa [56]. Consumption of balanced fruits with moderate glucose and adequate amount of vegetables has been linked to reduction of risks of developing cardio-metabolic risk factors, due to high contents of fibre, folic acid, high potassium, magnesium, vitamin C and flavonoid in them [57–59]. These ingredients improve endothelial function, modulate baroreflex sensitivity, cause vasodilation, and increase antioxidant activities [57–59]. However, a study conducted in Nigeria to investigate effect of fruits and vegetable consumption on cardiovascular risks did not result any significant effect on the risk [58].
To demonstrate portion sizes with the right proportions of each food group in every meal and required ingredients, nutritionists developed the “healthy plate model” [23]. The plate model is a new, innovative idea in Africa for the prevention of NCDs, thus limited knowledge on its importance may function as barrier to compliance, therefore, it may take longer to be adopted and it may require multi-sectorial and multi-disciplinary promotion strategies in different settings and at various levels [60]. To encourage uptake of the plate model, incentives to vendors, such as certificates of recognition to those who will consistently continue to offer food following the plate model, and complete trainings on food nutritional values, can serve as positive re-enforcement [61]. Instituting a plate model in Tanzania is feasible since all plate model components, i.e., foods groups, are readily available in the country [32]. However, vendors do not yet comply with the recommended portion sizes due to costs, individual preference, and culture; hence, T2DM health information is highly required to the community and stakeholders [26]. Continued awareness creation on the benefits of a balanced healthy diet is of paramount importance [61].
Findings from this study show that age and income influenced the outcomes such as BMI and raised BP. Previous studies indicate that higher age and income accelerate cardio-metabolic risks due to long-term accumulation and increased control over resources. This in turn leads to increased risks of exposure to cardio-metabolic risk factors, including excessive alcohol consumption, low physical activity, and consumption of fatty and processed foods [15, 62–65]. The current findings confirm the previous study linked to the current pre-post analysis study which indicated that regular street food consumers who consume recommended/above recommended servings of fruit/vegetables per week had a higher prevalence of overweight/obese than those who were not [15]. A systematic meta-analysis that involved 23 cohort studies, documented an inverse association for intake of some fruits such as, apples and pears, blueberries, grapefruit and grapes and raisins against risks for type 2 diabetes mainly be due to the high glycaemic load (GL) and the added sugar in fruit drinks [66]. The review also documented direct effect of type 2 risks against the intake of fruits such as cantaloupe, brussels sprouts, cauliflower and potatoes. It is therefore, argued that, promotion of fruits consumption should consider glycaemic load.
Many studies have established a direct relationship between obesity and hypertension [19–22]. The current study shows that there was a reduction of both systolic and diastolic BP in both arms. Thus, significant reduction of both systolic and diastolic BP in both arms may partly be explained by the level of decrease in body mass index and waist circumference observed. These findings were in line with those reported in a meta-analysis involving 25 studies, to assess fruit and vegetable consumption and risk of hypertension, concluded that consumption of fruits and vegetables can be inversely associated with the risk of hypertension [67].
This study showed an increase in positive attitude towards prevention measures for cardio-metabolic risks. It has also shown increased knowledge regarding risk factors, signs and symptoms and complications of type 2 diabetes as well as knowledge on the recommended amounts of different food groups. It can, therefore, be argued that the observed positive change on diastolic and systolic BP, BMI as well as waist circumference can be explained by an increase in knowledge and positive attitude which may be was translated into practice, hence influencing behavioral change. Similar findings were observed in a study in South Africa [56].
In this study a negative effect was observed on blood glucose; this is contrary from what was observed in a meta-analysis that involved seven (7) cohort studies. Observed negative glucose results from this study also differ from another meta-analysis which concluded that optimal consumption of fruit or green leafy vegetables significantly reduces T2DM risks [68].
However, the findings documented in this study on glucose may be linked to poor adherence to prevention measures especially at home, where it is usually a challenge to prepare a balanced diet, hence accelerating pre-diabetes [69]. Even though there is increased chance of eating balance diet while an individual is at home, available evidence show that many people confirm that it is exceedingly difficult to avoid overfeeding if delicious foods are plenty [37]. To maximize the effect of intervention, future similar study should consider monitoring individual eating practices while at homes since people are more likely to prepare and consume preference food while at their home hence increase chance of overfeeding [70].
Limitation of the study
Although this study emphasized the importance of the consumption of a balanced diet while at home, no follow up information regarding the quantity and food groups consumed during weekend days and in evening meals was collected, the focus was on lunch meals consumed to the vendors only and not elsewhere, which are always eaten outside homes. It can therefore be hypothesized that follow up on what participants were eating at home, study would have gained further insights. This study had a considerable rate of loss to follow-up, which could have lowered the power of the study. However, based on the nature of the study design, it remained with a large enough sample size in both arms to allow for meaningful results.
Conclusion
Pre-post analysis shows that both interventions in this study had positive effects on cardio-metabolic risk factors, knowledge and attitude. However, interactions between IP2, age and income was observed in this study. It is therefore, recommended that designing interventions relating to food should consider combining socio-economic factors. Community awareness creation campaigns regarding the impact of unhealthy eating practices on cardio-metabolic risk factors are recommended.
Supporting information
S1 Checklist. CONSORT 2010 checklist of information to include when reporting a randomised trial*.
https://doi.org/10.1371/journal.pone.0289289.s001
(PDF)
Acknowledgments
We thank all participants for their full support of the study. The authors also acknowledge the team of research assistants, participants and ward, municipality, and regional administrative officers for the managerial support. These municipalities were Ilala, Ubungo and Kinondoni in the Dar es Salaam region. Special appreciation also goes to Professor Jim Tod for his invaluable technical support and guidance during data analysis and interpretation.
Ethical clearance
This study followed a set of ethical principles in accordance with the revised 1964 Declaration of Helsinki, including ethical approval and informed consent [71]. This study was approved by the National Medical Research and Coordinating Committee and given ethical approval number NIMR/HQ/R.8a/Vol.IX/2794, and by the Kilimanjaro Christian Medical University College Review Board and given approval number 2291. This trial is also registered with the Pan African Clinical Trial Registry with Trial number PACTR202208642850935. Registration of the study at the Pan African Clinical Trial Registry was done before the enrolment of participants; however, due to technical problems of the system it could not be finalized in time. Ethical approval had already been given, so we proceeded with the recruitment of the participants while the trial registration process was ongoing. Every participant was informed about the aim and objectives of the study and its risks and participation benefits. They were also assured that participation was voluntary and could be stopped at any time, and nobody would be affected negatively for not consenting to participate in the study. Each consenting participant was asked to sign a consent form. All participants were adult and literate. All study subjects found to show cardio-metabolic risk factors were advised according to the WHO guidelines [72] and those found with disease problems were informed and referred to a nearby health facility for further management.
References
- 1. FAO. Influencing food environments for healthy diets. Food and agriculture organization of the united nations. 2016; 1–36.
- 2. Caswell JA, Yaktine AL, Board N. Supplemental Nutrition Assistance Program: Examining the Evidence to Define Benefit Adequacy. 2013.
- 3. Roman-urrestarazu A, Mohamed F, Ali H, Reka H, Renwick MJ, Roman GD, et al. Structural equation model for estimating risk factors in type 2 diabetes mellitus in a Middle Eastern setting: evidence from the STEPS Qatar. 2016
- 4. Budreviciute A, Damiati S, Sabir DK, Onder K, Schuller-Goetzburg P, Plakys G, et al. Management and Prevention Strategies for Non-communicable Diseases (NCDs) and Their Risk Factors. Front Public Health. 2020;8(November):1–11. pmid:33324597
- 5. Weerdt D. From corn to popcorn? Urbanization and food consumption in sub-Sahara Africa: Evidence from rural-urban migrants in Tanzania. 2017.
- 6. Ciccacci F, Orlando S, Majid N, Marazzi C. Essay Epidemiological transition and double burden of diseases in low-income countries: the case of Mozambique. 2020;37: 1–8.
- 7. Gaye B, Diop M, Narayanan K, Offredo L, Reese P, Antignac M, et al. Epidemiological transition in morbidity: 10-year data from emergency consultations in Dakar, Senegal. 2019; 1–7. pmid:31406585
- 8.
WHO. World Health Organization–Non-communicable Diseases (NCD): Risk of premature death due to NCDS Country Profiles, 2018. 2018; 2018.
- 9.
IDF. The IDF consensus worldwide definition of the metabolic syndrome (Available: https://www.pitt.edu/∼super1/Metabolic/IDF1.pdf: Accessed on 20 Jan 2020). B-1000 Brussels, Belgium; 2006.
- 10. Anand SS, Hawkes C, Souza RJ De, Mente A, Nugent R, Zulyniak MA, et al. Fast food consumption and its impact on CVD: importance of solutions focused on the globalized food system. J Am Coll Cardiol. 2016;66: 1590–1614.
- 11. Casas R, Castro-Barquero S, Estruch R, Sacanella E. Nutrition and cardiovascular health. International Journal of Molecular Sciences. 2018. pmid:30544955
- 12. Dehghan M, Mente A, Teo KK, Gao P, Sleight P, Dagenais G, et al. Relationship between healthy diet and risk of cardiovascular disease among patients on drug therapies for secondary prevention a prospective cohort study of 31 546 high-risk individuals from 40 countries. Circulation. 2012;126: 2705–2712. pmid:23212996
- 13. Rosoff DB, Smith GD, Mehta N, Clarke TK, Lohoff FW. Evaluating the relationship between alcohol consumption, tobacco use, and cardiovascular disease: A multivariable Mendelian randomization study. PLoS Medicine. 2020;17: 1–20. pmid:33275596
- 14. Xiao J, Huang JP, Xu GF, Chen DX, Wu GY, Zhang M, et al. Association of alcohol consumption and components of metabolic syndrome among people in rural China. Nutrition and Metabolism. 2015;12. pmid:25745507
- 15. Kagaruki GB, Mahande MJ, T MM, Jaffar S. Prevalence and Correlates of Cardio-Metabolic Risk Factors Among Regular Street Food Consumers in Dar es Salaam, Tanzania. Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy. 2021; 1011–1024. pmid:33707960
- 16. Frank SM, Webster J, Mckenzie B, Geldsetzer P, Manne-goehler J, Andall-brereton G, et al. Consumption of Fruits and Vegetables Among Individuals 15 Years and Older in 28 Low- and Middle-Income Countries.: 1252–1259.
- 17. Lopes Nonato I, Oliveira De Almeida Minussi L, Pascoal GB, De-Souza DA, Br D. Nutritional Issues Concerning Street Foods. Journal of Clinical Nutrition & Dietetics J Clin Nutr Diet. 2016;2: 1–9.
- 18.
FAO. Improving the nutritional quality of street foods to better meet the micronutrient needs of urban populations. 2007.
- 19. Kotsis V, Stabouli S, Papakatsika S, Rizos Z, Parati G. Mechanisms of obesity-induced hypertension. Hypertension Research. 2010;33: 386–393. pmid:20442753
- 20. Jiang SZ, Lu W, Zong XF, Ruan HY, Liu Y. Obesity and hypertension. Experimental and Therapeutic Medicine. 2016;12: 2395–2399. pmid:27703502
- 21. Narkiewicz K. Obesity and hypertension ‐ The issue is more complex than we thought. Nephrology Dialysis Transplantation. 2006;21: 264–267. pmid:16311261
- 22. Kotchen TA. Obesity-related hypertension: Epidemiology, pathophysiology, and clinical management. American Journal of Hypertension. 2010;23: 1170–1178. pmid:20706196
- 23. Harvard T. H. Chan School of Public Health. Healthy Eating Plate & Healthy Eating Pyramid. Harvard School of Public Health. 2011; 1–6.
- 24.
TANCDA. MTINDO WA MAISHA NA MAGONJWA YASIYO YA KUAMBUKIZA Dalili, Athari na Kinga Elimu kwa Jamii (Unpublished IEC Materials). 2018.
- 25.
Wikipedia. The Free Encyclopaedia, (https://en.wikipedia.org/wiki/Dar_es_Salaam, Accessed on 25. 2020 p. 2020.
- 26. Kinabo J. A Case Study of Dar es Salaam City, Tanzania: Globalization of food systems: Impacts on food security and nutrition. 2003; 10–20.
- 27. Xu XW, Wu XX, Jiang XG, Xu KJ, Ying LJ, Ma CL, et al. Clinical findings in a group of patients infected with the 2019 novel coronavirus (SARS-Cov-2) outside of Wuhan, China: Retrospective case series. The BMJ. 2020;368: 1–7. pmid:32075786
- 28.
WHO. Eseential Safety Requirements for Street-Vended Foods (https://apps.who.int/iris/bitstream/handle/10665/63265/WHO_FNU_FOS_96.7.pdf?sequence=1&isAllowed=y: Accessed on 12th June 2019). World Health Organization. 1996;96.7: 36.
- 29. Bryan FL, Jermini M, Schmitt R, Chilufya EN, Michael M, Matoba A, et al. Harzards associated with holding and reheating foods at vending sites in a small town in Zambia. J Food Prot. 1997;60(4): 391–398. pmid:31195538
- 30.
Lemeshow S, Hosmer Jr DW, Klar J, Lwanga SK. Part 1: Statistical Methods for Sample Size Determination(Available:https://apps.who.int/iris/bitstream/handle/10665/41607/0471925179_eng.pdf: Accessed on 27th May 2018). Adequacy of Sample Size in Health Studies. 1990; 247. doi:10.1186/1472-6963-14-335
- 31. Mackenzie JS, Jeggo M. The one health approach-why is it so important? Tropical Medicine and Infectious Disease. 2019;4: 5–8. pmid:31159338
- 32. Lukmanji Z, Hertzmark E, Mlingi N, Assey V. Tanzania food composition tables. muhimbili university of allied science,Tanzania food and nutrition centre and Havard school of public health. 2008; 259.
- 33. Pereira MA, Kottke TE, Jordan C, Connor PJO, Pronk NP, Carreón R. Preventing and Managing Cardiometabolic Risk: The Logic for Intervention. Int J Environ Res Public Health. 2009;6: 2568–2584. pmid:20054455
- 34.
WHO. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee (Available:https://apps.who.int/iris/bitstream/handle/10665/37003/WHO_TRS_854.pdf?sequence=1&isAllowed=y: Accessed on 23rd July 2019). World Health Organization technical report series. 1995. pp. 1–452. doi:854
- 35.
Aspirateur DUL, Garantie LAP. Owner ‘ s Manual Guide d ‘ utilisation (Available:https://www.glucoplus.ca/images/glucoplus-manuel.pdf: Accessed on 25 May 2020). 2009.
- 36. Essack Y, Hoffman M, Rensburg M, Van Wyk J, Meyer CS, Erasmus R. A comparison of five glucometers in South Africa. Journal of Endocrinology, Metabolism and Diabetes of South Africa. 2009;14: 102–105.
- 37. Kagaruki GB, Mayige MT, Ngadaya ES, Kilale AM, Kahwa A, Shao AF, et al. Knowledge and perception on type2 diabetes and hypertension among HIV clients utilizing care and treatment services: A cross sectional study from Mbeya and Dar es Salaam regions in Tanzania. BMC Public Health. 2018;18: 1–9. pmid:30055591
- 38.
WHO. STEPs Survey Report of Risk factors of Non communicable diseases (Accessed on 10th January 2017:https://www.who.int/ncds/surveillance/steps/UR_Tanzania_2012_STEPS_Report.pdf). 2012.
- 39. Kagaruki GB, Mayige MT, Ngadaya ES, Kimaro GD, Kalinga AK, Kilale AM, et al. Magnitude and risk factors of non-communicable diseases among people living with HIV in Tanzania: a cross sectional study from Mbeya and Dar es Salaam regions. 2014; 1–9.
- 40. Nguma LK. Health seeking and health related behaviour for type 2 diabetes mellitus among adults in an urban community in Tanzania. 2010; 1–349.
- 41.
Chrissy R, Marks M. Open Data Kiti @ LSHTM:User Manual (Available:http://opendatakit.lshtm.ac.uk/files/2015/10/ODK-@-LSHTM-User-Guide_v1.pdf: Accessed on 22nd June 2019(. 2013.
- 42. Shah V, Kamdar P, Shah N. Assessing the knowledge, attitudes and practice of type 2 diabetes among patients of Saurashtra region, Gujarat. International Journal of Diabetes in Developing Countries. 2009;29: 118. pmid:20165648
- 43. Bano R, AlShammari E, Fatima SB, Norah Ayed A-S. A comparative study of Knowledge, Attitude, Practice of nutrition and non-nutrition student towards a balanced diet in. Journal of Nursing and Health Science. 2013;2: 29–36.
- 44. Fatema K, Hossain S, Natasha K, Chowdhury HA, Akter J, Khan T, et al. Knowledge attitude and practice regarding diabetes mellitus among Nondiabetic and diabetic study participants in Bangladesh. BMC Public Health. 2017;17: 1–10. pmid:28446194
- 45. Saleh F, Mumu SJ, Ara F, Ali L, Hossain S, Ahmed KR. Knowledge, attitude and practice of type 2 diabetic patients regarding obesity: Study in a tertiary care hospital in Bangladesh. Journal of Public Health in Africa. 2012;3: 29–32. pmid:28299081
- 46. Yang H, Gao J, Ren L, Li S, Chen Z, Huang J, et al. Association between Knowledge-Attitude-Practices and Control of Blood Glucose, Blood Pressure, and Blood Lipids in Patients with Type 2 Diabetes in Shanghai, China: A Cross-Sectional Study. Journal of Diabetes Research. 2017;2017. pmid:29464182
- 47. Health MOF, Medical NFOR. Tanzania Steps Survey Report Ministry of Health and Social National Institute for Medical in Collaboration with World Health. 2013.
- 48. Winter B. What These Models Are Doing … and Then We’Ll Spend Most of the Time. 2013; 1–42.
- 49. Hadfield JD. MCMC methods for multi-response generalized linear mixed models: The MCMCglmm R package. Journal of Statistical Software. 2010;33: 1–22.
- 50.
Hadfield J. Notes, MCMCglmm Course. 2021.
- 51.
Vuong QH. Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses Author (s): Quang H. Vuong Reviewed work (s): Published by: The Econometric Society Stable URL: http://www.jstor.org/stable/1912557. THE MAIN PURPOSE OF THIS PAPER is to propose. Econometrica;Journal of the Econometric Society. 1989;57: 307–333.
- 52. Zhang Z. Model building strategy for logistic regression: Purposeful selection. Annals of Translational Medicine. 2016;4: 4–10. pmid:27127764
- 53. Tredennick A. T., Hooker G., Ellner S. P., & Adler P. B. (2021). A practical guide to selecting models for exploration, inference, and prediction in ecology. Ecology, 102(6). pmid:33710619
- 54. Gulliford MC, Ukoumunne OC, Chinn S. Components of variance and intraclass correlations for the design of community-based surveys and intervention studies: Data from the Health Survey for England 1994. American Journal of Epidemiology. 1999;149: 876–883. pmid:10221325
- 55. Thompson DM, Fernald DH, Mold JW. Intraclass correlation coefficients typical of cluster-randomized studies: Estimates from the robert wood johnson prescription for health projects. Annals of Family Medicine. 2012;10: 235–240. pmid:22585888
- 56. Murphy KM, Mash R, Malan Z. The case for behavioural change counselling for the prevention of NCDs and improvement of self-management of chronic conditions. South African Family Practice. 2016;58: 249–252.
- 57. Wada K, Tsuji M, Nakamura K, Oba S, Nishizawa S, Yamamoto K, et al. Effect of dietary nori (dried laver) on blood pressure in young Japanese children: An intervention study. Journal of Epidemiology. 2021;31: 37–42. pmid:32201400
- 58. Adebawo O, Salau B, Ezima E, Oyefuga O, Ajani E, Idowu G, et al. Fruits and vegetables moderate lipid cardiovascular risk factor in hypertensive patients. Lipids in Health and Disease. 2006;5: 1–4. pmid:16753064
- 59. Aburto NJ, Hanson S, Gutierrez H, Hooper L, Elliott P, Cappuccio FP. Effect of increased potassium intake on cardiovascular risk factors and disease: Systematic review and meta-analyses. BMJ (Online). 2013;346: 1–19. pmid:23558164
- 60.
Epstein NE. Multidisciplinary in ‐ hospital teams improve patient outcomes: A review. 2014; 295–303. https://doi.org/10.4103/2152-7806.139612
- 61. Karg H, Drechsel P, Amoah P, Jeitler R. Facilitating the Adoption of Food-Safety Interventions in the Street-Food Sector and on Farms.
- 62. Wu F, Guo Y, Chatterji S, Zheng Y, Naidoo N, Jiang Y, et al. Common risk factors for chronic non-communicable diseases among older adults in china, ghana, mexico, india, russia and south africa: The study on global AGEing and adult health (SAGE) wave 1. BMC Public Health. 2015;15: 1–13. pmid:25885218
- 63. Hui L. Assessment of the role of ageing and non-ageing factors in death from non-communicable diseases based on a cumulative frequency model. Scientific Reports. 2017;7: 1–7. pmid:28811618
- 64. Allen L, Williams J, Townsend N, Mikkelsen B, Roberts N, Foster C, et al. Socioeconomic status and non-communicable disease behavioural risk factors in low-income and lower-middle-income countries: a systematic review. The Lancet Global Health. 2017;5: e277–e289. pmid:28193397
- 65. Hosseinpoor AR, Bergen N, Kunst A, Harper S, Guthold R, Rekve D, et al. Socioeconomic inequalities in risk factors for non communicable diseases in low-income and middle-income countries: results from the World Health Survey. BMC Public Health. 2012;12: 912. pmid:23102008
- 66. Halvorsen R. E., Elvestad M., Molin M., & Aune D. (2021). Fruit and vegetable consumption and the risk of type 2 diabetes: A systematic review and dose-response meta-analysis of prospective studies. BMJ Nutrition, Prevention and Health, 4(2), 519–531. pmid:35028521
- 67. Li B, Li F, Wang L, Zhang D. Fruit and Vegetables Consumption and Risk of Hypertension: A Meta-Analysis. The Journal of Clinical Hypertension. 2016;18: 468–476. pmid:26826021
- 68. Li M, Fan Y, Zhang X, Hou W, Tang Z. Fruit and vegetable intake and risk of type 2 diabetes mellitus: Meta-analysis of prospective cohort studies. BMJ Open. 2014;4. pmid:25377009
- 69. Breland JY, McAndrew LM, Gross RL, Leventhal H, Horowitz CR. Challenges to healthy eating for people with diabetes in a low-income, minority neighborhood. Diabetes Care. 2013;36: 2895–2901. pmid:23877980
- 70. Gibson B Kagaruki, Michael J Mahande, Godfather D Kimaro, Esther S Ngadaya MMT, Majige Selemani, Lindsay M Jaacks, et al. Prevalence and Correlates of Cardio-Metabolic Risk Factors Among Regular Street Food Consumers in Dar es Salaam, Tanzania. Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy. 2021.
- 71. Kori-Lindner C. Ethical principles for medical research involving human subjects: World medical association declaration of Helsinki. Klinische Pharmakologie Aktuell. 2000;11: 26–28.
- 72.
WHO. Pocket guidelines for assessment and management of cardiovascular risk: (WHO/ISH cardiovascular risk prediction charts for the African Region) (Available:https://www.who.int/cardiovascular_diseases/guidelines/PocketGL.ENGLISH.AFR-D-E.rev1.pdf: Accessed o. World Health Organization. 2007; 1–30. https://doi.org/10.1093/innovait/inr119