Introduction

The world is undergoing a major type 2 diabetes health problem in adults. Its prevalence is on the peak in India in sync with the inactive life conditions adopted and other environmental factors. The number of people living with diabetes in India is calculated as the second most in the world as stated by the International Diabetes Federation (2017). Lately, number of people with type 2 diabetes with in a population-based sample in India was 10.4% among 20 to 79 year olds.¹

Diabetes management involves adoption of lifelong healthy lifestyle measures like healthy dietary practice and physically active days, in addition to any pharmacological treatment. Significant regimen applications play a major role in paving the way for the reduction of the disease impediments.² The main aim of quality regimen applications is to enhance life standard, status of wellness, and reduce the incidence of other noncommunicable disease risk components and so as to block adverse obstacles. Substantiation is well done and healthy dietary practice can help to achieve glycemic control³ and it could decrease glycosylated hemoglobin (HbA1c) by 1.0% to 2.0% and can further improve clinical and metabolic outcomes.⁴ Glycemic control can be improved by both degree of caloric restriction and weight reduction.⁵ Healthy dietary practice can enhance various factors which include obesity, hypertension, insulin release, and responsiveness. Absence of healthy dietary practice is a crucial declining factor in attaining glycemic control in type 2 diabetes. Despite the priority of healthy dietary practice in the regulation of type 2 diabetes, diabetic patients are generally not aware of its value in assuring better glycemic control.^6,7 Communities with educational qualities and social improvement also fail to acknowledge the importance of healthy dietary applications.4-7 Healthy dietary practice alone may suffice to assist the diabetic patient perpetuate normal glycemic levels. Various studies found that diet involving low carbohydrate and high protein will help to achieve the glycemic control in untreated diabetes.^8,9

Achieving glycemic control and healthy dietary practice in patients has demonstrated an actual challenge to health practitioners in many South Asian countries. According to Benoit et al,¹⁰ it was depicted that only 38% of type 2 diabetes patients had good glycemic control and mean HbA1c is 8.3%.¹⁰ A study from South Kerala demonstrated that only 27% patients ingested within 35% of their prescribed carbohydrate intake.⁷ Further studies in Mexico and Thailand portrayed noncompliance rates to healthy dietary practice of 62% and 45.7%, respectively.^11,12 To evaluate the present scenario in accordance to healthy dietary practice and glycemic control in India, we committed ourselves to conduct a survey within diabetic patients of Kerala, the “diabetic capital of India.”¹³ As far as we researched, no prior studies have been performed from Kerala marking these affairs.

Subjects and Methods

This was a cross-sectional survey. We conducted this survey in Malappuram, the most populous district of Kerala state; it is considered as a typical representative of the districts of Kerala in terms of human development index (score of the district was 0.67, the same as average of the state in 2011), life expectancy (70.5 years compared with the state average of 70.6 years), and literacy rate (89.4% compared with the state average of 90.9%).^14,15

A study was conducted in Tamil Nadu (2012) and resulted in a good dietary practice of 29% across diabetic patients.¹⁶ The sample size was calculated accordingly. The procedure of sample selection followed was cluster sampling and hence an effect of 2 (arbitrarily) was taken. Sample size was calculated by N = 4P × Q/D2 (P = 0.29 Q = 0.71) and D was arbitrarily taken as 8% or 0.08. By considering the design effect, 257.38 was the sample size. By examining a nonresponse rate of 20%, 308.85 was the sample size and it got rounded to 310. But we included only 304 subjects in the study due to missing entries and incomplete data.

We used multistage cluster sampling technique to identify the self-reported diabetic patients and depicted in Figure 1. Of the 15 block panchayaths in Malappuram district, Malappuram block panchayath was arbitrarily selected for the study. Block panchayath is local self-government body at tehsil level and is a link between gram panchayath and zila panchayath of the 3-tier panchayath raj system in India. The block panchayath was further divided into electoral wards. Average population of a ward was 4321, ranging from 628 to 8457. We selected 24 wards randomly from a total of 117 wards. Each electoral ward was considered as a cluster.

Figure 1.Sample Selection Procedure.

We enrolled 13 self-reported diabetic patients from each of the divisions based on the following inclusion criteria: (a) age 25 years or more, (b) having a diagnosis of diabetes by a modern medical practitioner, (c) diabetic for minimum 1 year at the data collection date (from the date of first prescription till the date of data collection). Individuals with severe physical or cognitive impairments, gestating women, and subjects with medical conditions with dietary restrictions (eg, chronic kidney disease) were excluded from the study. The main area of the ward was identified with the assistance of a local guide/person. Using a spin bottle method, house visits were guided in a fine manner. The first house along the selected path was recognized by selecting a number between 1 and 10 in random. Commencing from the first house, successive houses were paid visited until 13 diabetic patients were included. If more than 1 qualified member was available in the household, the subject for the study was taken by lottery method. Eligible patients were spoken from consecutive households till a cluster of 13 patients were acquired from each ward.

Ethics

Prevailing guidelines and standards of ethical conduct of research on human subjects were followed in the study. The ethical approval of the study was granted from the Institutional Ethics Committee of Sree Chitra Tirunal Institute for Medical Science and Technology, Trivandrum (Ref no: SCT/IEC-480/2013). We obtained written informed consent from all the subjects in the study with the freedom to deny from the study at any time during the interview.

Evaluation of Dietary Pattern

We used a pretested interview schedule to acquire demographic and clinical factors and to identify the dietary pattern. Socio economic status (SES) was determined in 3 groups placed on the type of floor of the household. We also used international physical activity questionnaire-short form (IPAQ) to estimate the physical activity and general self-efficacy scale evolved by Matthias Jerusalem and Ralf Schwarzer to measure the general self-efficacy of study subjects. Information on the following variables was also obtained: sociodemographic variables (age, sex, family type, socioeconomic status, and marital status), factors related to patient empowerment (responsible person in cooking, decision-making role, completed years of education, and occupation), diabetes and treatment-related factors (system of medicine, regularity in blood sugar monitoring and medication pattern, type of health facility used) and structural or facilitating factors (type of information, source of information, barriers to purchase healthy diet caused by cost, availability, taste, obstacles in work place, and family assistance).

Dietary practice was measured by 5 dietary behaviors namely fruits and vege table  intake, fat, salt, and sugar in diet and snacking behaviors. A composite outcome variable of healthy dietary practice was defined as “having not more than one undesirable dietary practice.”^6,13,14 By undesirable dietary practice, we meant low fruits and vegetable platter, high fat, salt and sugar diet, and unsound refreshment behaviors. Definitions of unhealthy behaviors we used, are as follows.^6,13,14

• Low fruits and vegetable serving: Patients with intake of less than or equal to 3 servings weekly. (The World Health Organization [WHO] recommends 5 servings for fruits and vegetables per week. The study considers 3 as the number of servings since number of applicants met the WHO criteria was very few.) 
• High fat diet: Patients with intake of high fat food for more (fried nonvegetarian dishes) more than 1 day weekly or who reported intake of fried food items with main course “often”/“always.” 
• High salt diet: Patients with high salt intake like papad, pickles, and dried fish more than 1 day weekly or adding salt often at table. 
• High sugar diet: Patients with intake of high levels of starch (like maida and white rice) or sugar contented beverage more than 1 day weekly.
• Unsound snacking behavior: Patients with intake of junk food or snacks more than a day in a week.⁶

Statistical Analysis

Descriptive statistics were used to outline the sociodemographic and clinical characteristics and their dietary practices. Bivariate and multivariate logistic regression analyses were performed using dietary practices as an outcome variable. Chi-square test was used to examine the relationship between the predictor variables and dietary practices in bivariate analysis. The variables were evident to be remarkably linked (P < .05) with deficient dietary practices in the bivariate analysis and was encompassed in a multiple logistic regression model. All statistical analyses were executed using Statistical Package for Social Sciences for Windows (version 17.0, IBM SPSS).

Results

Total sample size was 304. Distribution of sociodemographic variables was described in Table 1. The mean age of the sample population was 55 years (28 years-86 years), 56.2% were women, and only 5.6% were having more than 10 years of schooling. Mean duration of diabetes was 7.44 ± 5.84 years (1 month-16 years). Most patients got some form of dietary advice, mainly from a doctor (86.8%). Majority of the patients have trouble following a low salt, sugar, and fat diet (46.1%) and the main barrier for not adhering to this regimen was low taste and not culturally acceptable. Nonadherence in consuming more fruits and veg etables  were reported by 13.2% and the main barrier was the cost of the fruits and veg etables . Nearly 27% of patients have no family support for following good dietary practices. Forty-seven percent of study subjects have other family members to help control the diet. Forty-one percent of the study subjects have good family support to follow good dietary practice. Mean self-efficacy score of the study population was 28.84 ± 8.67 and median 30 with a Cronbach’s alpha of 0.97. The least score was 10 and utmost score was 40. A total of 36 (11.8%) subjects described that diet had no crucial role in the diabetic control. Quality understanding about difficulties of diabetes was reported by 204 (67.1%). Mean self-efficacy score of the study subjects was 28.84 ± 8.67, given a possible range of 16.2-34.64. A total of 14% of study subjects had free treatment and the remaining carried out the expenses by themselves. Among the subjects, 19.7% (n = 60) had sedentary in their lifestyle and 38.2% (n = 118) had sitting time more than 6 h. Sixty-two percent of the subjects were unemployed at the time of the study. Forty-nine percent of the participants considered themselves the decision-makers in purchasing the food items for the family. The factors contributing the patient empowerment in deciding the dietary choices were explained in Figure 2.

 Table  1.Distribution of Sociodemographic Variables Among the Sample.

Source: Crossectional survey among the households of Malappuram, 2017.

Figure 2.Distribution of Patient Empowerment Factors.

Seventy-nine percent of the subjects are not taking medications regularly. Only 49% of the subjects are monitoring the blood sugar regularly. Nearly three-fourths of the study subjects (75.3%) having low fruits and vegetable intake (explained in Figure 3). Only 20.4% of the subjects met our criteria of good dietary practice.

 Figure  3.Distribution of Dietary Practices.

Source: Crossectional survey among the households of Malappuram, 2017.

 Older age group (more than 55 years), currently married (married within 1 year), lower level of education (completed years of education less than 10 years), no considerable difficulty in consuming low salt/sugar/fat diet, not having frequent moving for work or other purposes (travelling outside from home by vehicle or walk), having a family member to follow or support healthy dietary practice, lower general self-efficacy score, sitting time more than 6 h and duration of diabetes more than 6 years, and frequent blood sugar monitoring (5 or more) in the 6 months prior to the date of data collection are the variables significantly associated with good dietary practice in bivariate analysis (P < .05). In addition to all variables significant at the bivariate level, sex, availability of another person who had a role in diet control of the patient, and medication consumption pattern as per advice were considered, as crude P values were <.10 and there could be underlying relationships of these variables to other potential predictors like education, occupation, self-efficacy, and so on. SES was included as it was expected to influence the effect of self-efficacy on dietary practice.

On multivariate analysis, having the family member to remind on healthy dietary practice (adjusted odds ratio [AOR]: 2.89 [1.24-6.75]), having peer in controlling or advising for healthy dietary practice (AOR: 2.23 [1.07-4.66]), and undertaking blood sugar monitoring more than once in a month (AOR: 2.74 [1.27-5.94]) were appreciably related with healthy dietary practice among type 2 diabetes patients.

Study subjects were asked to report on food items and behaviors that they thought as unhealthy for them. The consumption of foods like sugar and red meat were generally perceived as unhealthy. However, there was no association between this perception and good dietary practice. Approximately 10 percentage of the subjects trusted that fruits are unhealthy for diabetic patients. Around half of the patients accepted that they were keeping a healthy diet all over the week. The factors determining the good dietary factors are explained in Figure 4.

Figure 4.Factors Promoting Good Dietary Practice Among Diabetic Patients/Diagrammatic Representation of the Study Results.

Source: Crossectional survey among the households of Malappuram, 2017.

Discussion

Dietary practice is a complex concept and here we used 5 components to define and measure the concept comprehensively. In this cross-sectional study, we found that only a small proportion of type 2 diabetes patients (20.4%) had healthy dietary practice. The intricate definition for healthy regimen practice could be the cause for the low prevalence witnessed.

Diabetic patients with family assistance or with someone to look after their diet were following a healthy dietary application. Various studies from India and other countries depicted the positive influence of family and other coordinators in following a better diet.^12,17 Difficulty in intake of low salt, low sugar, and low-fat foods was evident to be associated in particular with healthy dietary practice. The reason for this may be the struggle to compromise the taste and some studies showed the influence of taste in the selection of foods.^10,16,18,19

According to American Diabetes Association’s guidelines for the management of diabetes, all diabetic patients at time of diagnosis must be provided with an access to a dietitian/diet counselor or other health-care professional trained in nutrition literacy who will offer the basics of nutritional content of various food stuffs an initial consultation with 2 or 3 follow-up sessions, either individually or family counseling.³ Such a service may help overcome some of the difficulties patients felt in achieving a healthy diet. Healthy dietary practice was seen among subjects with frequent monitoring of their blood sugar levels. Such persons may have more knowledge on disease and its management in turn may have more control on their diet. Therefore, we should give more importance to patient-oriented education on their available and affordable healthy diet choices in general.^16,20,21

Preceding studies proved that there is a distinction in dietary pattern among various SES group and the effectual management of diabetes is established by socioeconomic status of the subject. But this study did not prove any  notable a ssociation of various SES group with quality dietary practice. The expected reason could be the result of lifestyle modification and adverse dietary behavior transition of people in Kerala traversing all socioeconomic strata, consequent to macro-level influences of globalized and liberalized economies.²²

Most of the diabetic patients believed that fruits are a source of sugar. This misleading information and the cost of fruits were the barriers for the required servings of fruits and vegetable consumption.^6,12 The study findings showed a prevalence of healthy dietary practice of around 20%, but over half of the subjects perceived their current diet as healthy. Adequate nutritional knowledge for diabetes patients is very important instead of vague saying like “diet control” and absence  of eatable f ruits for diabetic patients. This gap suggests the nutritional illiteracy among the type 2 diabetes patients. When diabetic patients seemed to have some degree of knowledge on good dietary practice, it did not seem translate into healthy dietary practice.²³ The blind belief in the oral hypoglycemic agents to control diabetes rather than its chronic nature and lifestyle management observed in previous studies might be the reason for poor dietary management among type 2 diabetic patients.^20,21

The frequency of other behaviors like use of tobacco (14.8%) and alcohol (1%) use among diabetic patients was low in comparison to reported levels in an average population.¹⁷ The diabetic repute and social norms of use of alcohol among women and in Muslim population might be the reason for the low prevalence of alcohol use in the study. Most subjects reported doctors as their main source of dietary advice, which was consistent with other studies. However, it was found that some of the studied diabetic patients did not receive dietary advice from any source since the diagnosis.^16,24

This study discloses certain questions in addressing a difficult element of diabetes regulation, specifically regimen practices, in Malappuram, a proportionately backward district of Kerala and the challenges are nearly similar in most of the rural Kerala. Among the individual components which affect the regimen practice, the study has also observed the social elements, some health system-oriented components, and health attracting behavior of the patients with diabetes. We did not consider the comorbidities of patients, stringent estimates of nutritional literacy of the patient, and quality and quantity of actual diet in the study group, and these may have influenced our results. However, it is unlikely that more rigorous assessments will lead to finding a higher proportion of people having healthy dietary practice. Hence, despite these limitations in our approach, we feel our findings are cogent enough to suggest the perfunctory nature of dietary advice and support given to diabetic patients. This part of diabetes management in Kerala through the National Program for Control and Prevention of Cancer, Diabetes, Cardiovascular Disease, and Stroke) as well as through other systems of care needs more attention, given that about 1 in 5 persons in Kerala are likely to have type 2 diabetes.

Conclusion

This was a cross-sectional study to find out the prevalence of healthy dietary practice, related elements, and risk acknowledgement of frequently consumed foods amid the self- announced patients. Among type 2 diabetes patients in Malappuram, the prevalence of good dietary practice was 20.4%. The study recognized the role of factors beyond the individual level such as family support and control over ones diet by another person as determinants of good dietary practice. This suggests the importance of considering interventions that focus on family and other individuals connected with diabetes patients in the emerging control strategies of our NCD control program.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.