Abstract
Background:
Rising prevalence and poor outcomes make the twin challenges of diabetes epidemiology. This study evaluates effect of 2-only-daily-meals with exercise (2-OMEX) for its effect on HbA1c, oral hypoglycaemic agents (OHA) usage, body-weight among type-2-diabetes (T2DM) subjects, compared with conventional management.
Material and Methods:
A quasi-experimental, multicentre study in 2-OMEX arm, and HbA1c by HPLC method. HbA1c and body-weight changes were analyzed by ‘Difference in Difference’ (DID) method. Meal frequency, exercise, energy intakes were based on recall. The required sample size was 20X2 for 1.1 difference in HbA1c with 95% CL and 80% power
Results:
Socio-demographic and risk profile of analysed and omitted subjects were similar. Studied arms were also similar in baseline features. The results in 2-OMEX and conventional arm are: complete records analyzed 201 and 120. Mean (sd) values as follows: observation days 234 and 236, age 52.03(8.84) and 52.45(9.48) years (P=0.6977), diabetes duration 4.6(3.05) and 4.9(2.97) years, BMI 27.28(5.27), 26.90(3.74) (P = 0.1859), baseline HbA1c gm% 7.46(1.52) and 7.55(1.58), end-line proportion of subjects attaining HbA1c ≤6.5gm% was 35.3% and 19.2% (P=0.002), bodyweight loss 2.57% and 1.26%. OHA count 1.6 (1.23) and 2.7(1.06), (P=0.0003). In 2-OMEX arm log-normal HbA1c declined significantly by 0.94 (95%CI: 1.60 to – 0.56, p=0.0333), weight loss difference 0.96 kg, and statistically not significant (P=0.595). Two subjects in 2-OMEX arm showed partial remission. Mean baseline Kcal intakes in 2-OMEX arm, were 1200.4(F) and 1437.3(M) were significantly higher than conventional arm (F) and 1430 (M)
Conclusion:
The 2-OMEX showed a sizeable and significant reduction in HbA1c and OHA use, in 7-months, with moderate intakes, compared to the conventional arm, possibly attributable to fewer insulin surges. More studies are required for its impact and pathways.
Keywords: Difference in difference analysis; HbA1c; hypoglycaemic agents; lifestyle,2-meal frequency; oral remission; type 2 diabetes
INTRODUCTION
The world faces the rising prevalence and poor outcomes of type 2 diabetes (T2DM). India has a high burden of T2DM (5.3% to 13.6%) and impaired glucose tolerance (8.3%–14.6%).[1] Indian diabetes risk score studies predict a high tide of T2DM in various Indian states.[2–5] The interest in T2DM remission or reversal is rather recent and implies euglycemia (HbA1c <6.5 gm%) for one year without the use of hypoglycemic agents.[6] Restriction of calorie intake and optimal physical activity is reported to lead to remission.[7] Some popular campaigns for reversal include intermittent fasting, vegan diets, low cereal/carb diets, and keto diets, but scientific literature on these is scant.[8,9]
A popular campaign started in Maharashtra in 2017 for weight loss through lifestyle modification, using a strategy of two-only meals a day and modest exercise (2-OMEX) and many diabetic subjects participated. Decline in weight and HbA1c was reported even in diabetic subjects.[10] The campaign got wide publicity.[11]
This quasi-experimental study aimed to compare the effect of the 2-OMEX intervention among the T2DM patients on (a) HbA1c, (b) count of oral hypoglycemic agents (OHA), (c) body weight, and (d) baseline energy intakes, with conventional management arm.
METHODS
Study design
This is a nonequivalent comparison group quasi-experimental study.[12,13] The 2-OMEX intervention group and the group following conventional management were compared. We used the TREND statement for reporting the study.[14] Ethical approval was obtained (SMBT/IEC/19/038 dated January 16, 2019).
Eligibility criteria
Following participants were included (a) age between 30-70 years, (b) current diabetic status, as defined by HbA1c ≥6.5gm% and/or current antidiabetic medication, (c) diabetes duration < 10 years. Exclusion criteria were: renal impairment, reported weekly alcohol intake, taking insulin, hemoglobin disorders, serious cardiovascular disease, and impending surgery. For conventional treatment arm, all these criteria remained the same but with meal frequency being more than two.
Study settings and recruitment of subjects
This study was conducted in the state of Maharashtra, where the 2-OMEX campaign had established weekly clinics in seven cities for Diabetes Reversal Center (DRC), with the help of diabetes specialist physicians on a free and charitable basis. Three private clinics offered as conventional clinics. All the eligible participants were enrolled by voluntary participation. Similarly, eligible subjects were recruited in the conventional arm. Written informed consent was obtained from all participants. There was no randomizaton of subjects in this study. Enrollment was done in the period starting June 2019 to August 2019 for such patients visiting DRCs or conventional clinics with one to two prior visits to minimize attrition. Patients were followed up till February 2020 till the COVID-19 pandemic disruption. Patients made monthly visits for the initial three months and then every three months.
Intervention
The intervention (2-OMEX) implied taking only two meals each day, without energy drinks or snacks in between, along with a daily walking/aerobic workout for 45 min for a minimum of six days per week. DRC participants were given a personal diary for noting daily meals and exercise. The study participants were supported by DRC volunteers on cell phones and social media at least once monthly, regarding meal frequency, exercise, HbA1c reports, and regularity of DRC visits. Those subjects who failed to adhere to 2-OMEX (more than one cheat day for meal restriction per week for four consecutive weeks) were de-listed from DRC and advised to seek conventional care. We also trained primary care assistants in the conventional arm for keeping case records. In both arms, medication was managed by respective physicians.
For energy intake, we asked for the usual count of chapatis and rice-watis consumed daily. Our average estimate of weight of chapati (dried on a pan) was 38 gm. Bhakri (millet-pancake) was a rare item and its dry weight (84gms) was about two chapatis. We estimated that raw milled rice of 35 gm contributed to one wati (bowl) rice. We estimated kcal using ICMR-NIN nutritive values at 3.5 kcal/gm of whole wheat (after adjusting for 10% moisture lost in dried chapati), and 124 kcal per wati cooked rice as per ICMR tables.[15] A recent ICMR survey for the region, estimated cereals and millets making 45.6% of all energy intake in urban western India.[16] Therefore estimation of average total energy intake from all foods consumed was arrived at by multiplying the pooled cereal-millet energy intake by a factor of 2.2 for both men and women.
Variables
The intervention arm and control arm were defined on meal frequency (2 only or ≥3 meals). Recommended exercise adherence was common to both arms. The main independent variables in analysis were age, sex, occupation, social category, type of diet, history of diabetes among parents, and duration of diabetes. The primary outcome variable was HbA1c while secondary outcomes were weight and the count of OHA medicines.
Data sources/measurement
In 2-OMEX arm, DRC case records were kept by trained volunteers and physicians, in successive monthly/quarterly visits, with laboratory test records. HbA1c reports (done by participants at convenient private labs using the HPLC method) were obtained within the current or the last week. Prescription information regarding OHA medication was deciphered as generic molecules in counts (0–6) irrespective of change in dose. Stadiometer for height machines of the same make (MCP-number 265M, 200cm) and batch were used at all centres. All 10 digital weight machines were initially tested against the one at Nashik (model WI/05/2016/014/El). All anthropometric measurements were done with ordinary clothes and no footwear.
We used the following Statistical software: Excel for data management, Epi-Info for most of data analysis work, and R software for difference-in-difference (DID) analysis and multiple linear regression.
Sample size
Based on the pilot study, to detect an effect size of 1.1 gm% in HbA1C with an SD of 1.22 gm%, the sample size needed was 20 in each arm, based on open-epi calculations. Since the study was conducted in campaign mode we could recruit 423 in 2-OMEX arm and 201 participants in conventional arm at baseline. Of these, 201 participants in 2-OMEX arm and 120 in control arm could complete the required follow up and were included in the analysis. The numbers (201 in 2-OMEX arm and 120 in conventional arm) yielded a power of 85.2%.
Sources of bias and efforts to overcome
Of the seven DRCs, two were dropped due to administrative difficulties (Thane, Hingoli). Further, some of the enrolled subjects were omitted due to incomplete records/follow up. From the conventional arm, the Pune clinic was a non-starter and hence dropped. Compliance with 2-OMEX was assessed on phone call enquiry as diaries were not physically accessible due to the COVID-19 pandemic. The initially planned observation period of one year was curtailed to minimum of 120 days due to the COVID-19 pandemic. About 10% of the subjects were randomly cross-checked during the study. Figure 1 shows the steps of selection and exclusion of participants in the two arms.
Figure 1.
Flow chart of inclusion exclusion of case records
Case records and data entry
A pre-tested physical case record sheet was kept and used for information retrieval. Data entry of eligible case records completing at least 120 days of observation between the recorded first visit to the last recorded visit was done in Excel. In case of missing data, subjects were contacted telephonically and an attempt was made to complete the data.
Statistical methods
Continuous variables (age, duration of diabetes, follow-up time, HbA1c, and BMI) were tested for normality and log transformation was used for HbA1c. Baseline comparisons between 2-OMEX and the conventional group were made by t-test and Chi-squared test for continuous and categorical variables, respectively. DID analysis was used to estimate the effectiveness of the intervention on HbA1c and BMI using multiple linear regression.[13,17,18] The regression technique allowed adjusting the DID estimate for confounders such as sex, type of diet, occupation, social category, and the number of OHAs used at baseline. Assumptions of multiple linear regression were met including multicolinearity. Chi-squared test was used to assess the trend in the number of OHAs used at baseline and end-line separately in each arm.
RESULTS
The 2-OMEX arm has shown a sizeable reduction in HbA1c over the conventional treatment arm, reduction in the mean count of OHA medication, and also better weight loss in an average period of 235 days. The 2-OMEX arm also showed lower estimates of current energy intakes compared with the conventional arm.
Baseline features of enrolled, omitted, and study subjects
Table 1 shows baseline features of the two arms for enrolled subjects and finally analyzed case records. The baseline features of enrolled, omitted, and analyzed subjects in both arms including age, gender, activity, social category, duration of T2DM, and first record of HbA1c and BMI had no significant difference [Table 1].
Table 1.
Baseline values of subjects
| Numbers analyzed | 2-OMEX arm at enrollment (n=419) | Conventional arm at enrollment (n=419) (n=198) | Test statistics for subjects from 2-arms | P | ||||
|---|---|---|---|---|---|---|---|---|
|
|
|
|||||||
| Analyzed records | Omitted Case Records | P | Analyzed records | Omitted case Records | P | |||
| n | 201 (47.52%) | 222 (52.48%) | 120 (59.7%) | 81 (40.30%) | 201 & 120 | |||
| Age (Years) (mean+SD) | 52.03 (8.84) | 53.06 (9.98) | 0.2627 | 52.45 (9.48) | 54.22 (8.33) | 0.1640 | 0.6977 | |
| Sex | ||||||||
| Female | 88 (50.3%) | 87 (49.7%) | 0.3382 | 38 (52.8%) | 34 (47.2%) | 0.1348 | χ2for gender difference=4.62 | 0.0131 |
| Male | 113 (45.6%) | 135 (54.4%) | 82 (63.6%) | 47 (36.4%) | ||||
| Diet Type | ||||||||
| Mixed | 82 (43.4%) | 107 (56.6%) | 0.088 | 61 (60.4%) | 40 (39.6%) | 0.8401 | χ2 for diet type=3.06 | 0.0800 |
| Vegetarian | 119 (51.7%) | 111 (48.3%) | 59 (59.0%) | 41 (41.0%) | ||||
| T2DM in parents | ||||||||
| No parental H/o T2DM | 135 (67.16%) | NA | 89 (74.17%) | NA | χ2 for parental T2DM=1.7473 | 0.1862 | ||
| Any parent with T2DM | 66 (32.84%) | NA | 31 (25.83%) | NA | ||||
| Occupation | ||||||||
| Farmer | 1 (16.7%) | 5 (83.3%) | 0.4297 | 10 (37.0%) | 17 (63.0%) | 0.0029 | 16.05 | 0.029 |
| Business & Service | 108 (48.2%) | 116 (51.8%) | 60 (69.8%) | 26 (30.2%) | ||||
| Domestic work | 53 (50.0%) | 53 (50.0%) | 31 (50.8%) | 30 (49.2%) | ||||
| Other | 26 (41.3%) | 37 (58.7%) | 16 (80.0%) | 4 (20.0%) | ||||
| Missing Information | 13 (54.2%) | 11 (45.8%) | 3 (42.9%) | 4 (57.1%) | ||||
| Social Category$ | ||||||||
| General | 111 (44.0%) | 141 (56.0%) | 0.0497* | 58 (56.9%) | 44 (43.1%) | 0.0080* | ||
| Underprivileged | 22 (66.7%) | 11 (33.3%) | 17 (54.8%) | 14 (45.2%) | ||||
| OBC+ | 45 (53.6%) | 39 (46.4%) | 16 (48.5%) | 17 (51.5%) | ||||
| NA | 23 (42.6%) | 31 (57.4%) | 29 (85.3%) | 5 (14.7%) | ||||
| Diabetes duration (mean+SD) | 5.00 (2.91) | 4.99 (3.02) | 0.9933 | 4.50 (1.87) | 4.54 (3.03) | 0.9746 | ||
| HbA1c at Visit 1$ | 7.46 (1.52) | 7.70 (2.05) | 0.5855 | 7.72 (1.82) | 7.55 (1.58) | 0.5222 | 0.6970 | 0.4867 |
| BMI at Visit 1 (rerun) | 27.28 (5.27) | 25.57 (5.05) | 0.5793 | 26.90 (3.74) | 26.81 (4.16) | 0.9706 | 1.3280 | 0.1859 |
| Period of observation | 234.4 (59.9) | 236.8 (51.0) | 0.7119 | |||||
$Information on these variables was limited in both arms
After the omission of case records for incompleteness, we had completed records of 201 subjects from 2-OMEX arm and 120 subjects from the conventional arm for analysis.
Estimates of current total kcal intakes were as follows: in 2-OMEX arm; women (n = 69): 1200.40 (436.9); men (n = 93):1437.38 (588.38). Current kcal intakes in the conventional arm: women (n = 34): 1513.21 (319.22); men (n = 78):1892.08 (548.50). The differences between the two arms were highly significant for both men and women (P = 0.003, 0.000). The 2-OMEX arm subjects were found to have already dropped their previously reported kcal intakes by 334 (F) and 297 (M) at the time of enrollment.
Effect size
Table 2 and Figure 2 show the effects of the 2-OMEX intervention, using the DID technique and adjusted for differences at baseline between the two arms (sex, type of diet, occupation, social category, and count of drugs used for treatment of diabetes). Table 3 shows baseline to end line changes in glycemic status and OHA count in the two arms.
Table 2.
Difference-in-Difference (DID) estimates of changes in HbA1c and body weight
| Variable | Time | Arm | DID estimate## | 95% Confidence Interval## | P | R-square | |
|---|---|---|---|---|---|---|---|
|
| |||||||
| 2-OMEX arm (n=201) | Conventional arm (n=121) | ||||||
| HBA1C (log Normal values) | Baseline | 2.02 (0.199)# | 2.01 (0.199)# | -0.94 | -1.60 to -0.56 | 0.033 | 0.156 |
| End line | 1.93 (0.151)# | 1.99 (0.146)# | |||||
| Body Weight | Baseline | 71.43 (13.01) | 67.96 (9.66) | -0.95 | -4.49 to 2.57 | 0.595 | |
| End line | 69.76 (10.01) | 69.79 (10.05) | |||||
#Log-transformed, ##Adjusted for baseline differences between intervention and conventional arm (sex, type of diet, occupation, social category and number of drugs used for treatment of diabetes)
Figure 2.
Baseline to End line Change in Mean LogNHBA1C and Medication (OHA count)
Table 3.
HbA1c status and generic drug use for diabetes from baseline to end line
| Status of HbA1c and OHA use | 2-OMEX arm (n=201) | Conventional arm (n=120) | ||
|---|---|---|---|---|
|
|
|
|||
| Baseline | End line | Baseline | End line | |
| HbA1c <6.5 gm%$ | 45 (22.4) | 71 (35.3) | 30 (25.0) | 23 (19.2) |
| HbA1c ≥6.5 gm%$ | 156 (77.6) | 130 (64.7) | 90 (75.0) | 97 (80.8) |
| 0 OHA generics | 38 (18.9) | 53 (26.4) | 0 (0) | 0 (0) |
| 1 OHA generics | 26 (12.9) | 37 (18.4) | 17 (14.2) | 17 (14.2) |
| 2 OHA generics | 71 (35.3) | 60 (29.9) | 37 (30.8) | 34 (28.3) |
| 3 OHA generics | 49 (24.4) | 41 (20.4) | 43 (35.8) | 41 (34.2) |
| 4 OHA generics | 14 (7.0) | 8 (4.0) | 20 (16.7) | 23 (19.2) |
| 5 OHA generics | 3 (1.5) | 2 (1.0) | 3 (2.5) | 5 (4.2) |
| P value using X2 for trend | 0.009* | 0.492 | ||
| OHA mean count$$ | 1.9 (1.24) | 1.6 (1.23) | 2.6 (1.00) | 2.7 (1.06) |
$, χ2 between baseline counts of 2-OMEX and Conventional arm for HbA1c<>than 6.5gm% P=0.592, and at end line P=0.002*. $$: Both arms had significant difference between baseline and end line mean count of OHA generics (P=0.000)
DISCUSSION
The study findings
We observed a noticeable and statistically significant change due to 2-OMEX lifestyle intervention of about seven months in all three outcomes as compared to the conventional arm. (a) The 2-OMEX intervention showed statistically significant reduction in the log-normal HbA1c by 0.94 gm% (95%CI: -1.60 to –0.56) at end line after adjusting for the counterfactual. Further, the baseline to end line change in proportion of subjects in 2-OMEX arm having HbA1c <6.5 gm% was 35.3%, and this was significantly higher than 19.2% seen in the conventional arm [Table 3]. (b) The highly significant reduction in OHA generic count in 2-OMEX arm is also prominent [Table 3] compared to no such change in the conventional arm. (c) Weight reduction was more pronounced in the 2-OMEX (F: 1.90 kg and M: 1.82 kg) as compared to the conventional arm (F: 0.88 kg and M: 0.83 kg), and both the within-arm-declines were statistically significant. The decline in body weight was 2.57% in the 2-OMEX arm, and 1.26% in the conventional arm. When tested with DID analysis, the DID for body-weight decline between two arms was 0.95 kg and this was statistically not significant (P = 0.562).
Although we have not analyzed baseline to end line trends in energy intakes in each arm during the observation period, the 2-OMEX arm subjects seem to have already dropped their kcal intakes by 334 (F) and 297 (M) by the time they were enrolled from their reported prior intakes, which amounted to 21% and 17%, respectively. The 2-OMEX arm subjects had a variable period of joining the 2-OMEX campaign before enrollment, from 0 (same day) to 530 days, with a mean of 104 days. The conventional arm subjects were also under treatment at the clinic for an average of 141.9 days (0–314 days) before enrollment. The energy intakes of 1150 (F) and 1430 (M) at enrollment of 2-OMEX subjects are not very low caloric diets as practiced in many remission trials.[7,19] This factor of modest reduction of intakes observed in our study favors sustainability of the campaign.
The partial remission (HbA1c <6.5 and OHA-free status at end visit) of 26 subjects from T2DM in 2-OMEX arm against none in the conventional arm shows a window of opportunity for T2DM subjects through an easy-to-follow 2-OMEX lifestyle. However, we could not combine reduction in the count with reduction in dose of OHA in a single analytic measure. Despite this limitation and the multicenter design involving 14 different physicians working on a common approach, this is a robust primary evidence that 2-OMEX strategy helped reduce OHA medication.
The one-year requirement of the study was not fulfilled due to pandemic-disrupting DRCs, and we had to settle with 120 days as the minimum period of observation in case records.
Plausible pathways for glycemic changes in 2-OMEX
The observed weight reduction in this intervention arm is small, and it is corroborated by a small reduction in energy intake (17% in men and 21% in women). The 2-OMEX intervention showing some association with HbA1c and OHA reduction could be working through either or both (a) modest calorie intake reduction and (b) fewer insulin secretion spikes due to fewer meals and hence less insulin resistance. This is in line with meal-related insulin secretion.[20] A systematic review compares the advantages of caloric restriction for reducing insulin resistance, among other benefits.[21] A study of two-meal pattern consisting of breakfast and lunch has shown better diabetes management outcomes, but this article does not mention HbA1c outcomes.[22]
2-OMEX in the context of conventional management and recent T2DM remission studies
Comparison of comprehensive intervention in T2DM against usual clinical management, as reported by a metanalysis in good health care settings, concluded that the former has significant but moderate edge over the latter, amounting to about 0.54 gm% difference in HbA1c.[23] A large Scottish study of T2DM patients concluded that only 47% of the subjects could achieve good glycemic control at the end of two years of conventional management and the overall weight loss in this study was 2.5%.[24] Eminent diabetes physicians in India emphasized in 2010 about lifestyle interventions because of large numbers of T2DM (over 58 million), unsatisfactory compliance and weak glycemic control, but also stated that lifestyle interventions were not enough and called for better medication.[25] Another study reports that even in advanced country settings, conventional management and lifestyle change did not achieve good glycemic control (<7gm%) in 60% patients.[26] These reports clearly suggest that conventional management combining three or more meals with exercise and medication, including insulin use has unsatisfactory compliance and outcomes even in the best health care settings. India already has unsatisfactory outcomes regarding diabetes management.[25] Hence T2DM becomes a greater challenge in Indian or similar health care settings. Remission or reversal of T2DM should be an important frontline against the global epidemic, as we not only need to prevent T2DM but also roll back T2DM and IGT to reduce the caseload and improve quality of life. Literature on remission of T2DM is now increasingly available and the 9th edition of World Diabetes Atlas has mentioned this possibility for some people.[27] Remission approaches essentially include weight loss through calories intake reductions.[28,29] Other less documented efforts are vegan diet, intermittent fasting, all with recommended physical activity.[13,25–29] In a recent article, Roy Taylor and colleagues review accumulated experience about T2DM remission attainments through weight control by substantial restriction of calories intake.[7] The DiRECT trial in Scotland endorsed calorie intake restricted to 800–850 kcal.[30] Indian experts on diabetes have suggested this restriction level to 600 kcal.[25] Veganism is still not well documented in research. The 2-OMEX campaign does not insist on veganism or severely reduced intake like 600–800 cal; it banks on whatever modest intake reduction happens with a change to two-meal frequency. Meal frequency is shown to be associated with BMI changes.[31] This modest approach gives 2-OMEX some edge as an acceptable and sustainable lifestyle change. The 2-OMEX pattern overlaps with intermittent fasting for the >12-hour dinner-to-meal overnight separation. From India, we have two reports stating restoration of glycemic control with lifestyle changes [32,33] A Ramadaan-fasting study of daylong fasting tradition suggested a decline of HbA1c. Ramadaan involves a dawn-to-dusk fasting period and the eating window at night and may involve multiple meals.[34] In contrast with this, the 2-OMEX is limited to just two meals. Further, this two-only meal and exercise approach may be a better campaign slogan than fasting.
Limitations to validity of the study
This study has the limitations of a quasi-experimental study. The COVID-19 pandemic curtailed our observation period. Possible laboratory variation in HbA1c estimation in a multicentric study, estimation of independent variables of meals and exercise on recall are other limitations. Further, the patients enrolled in both arms were not newly diagnosed T2DM patients but had undergone some medication for a variable period already. The 2-OMEX subjects had also practiced it some weeks before our enrollment. Also we did not account for changes in energy intake from baseline to end line and can only state rounded baseline estimates. The cost of HbA1c tests borne by campaign participants has also caused some attrition. The high proportion of lost-to-follow up cases and weak compliance is a threat too, but a chronic disease demanding meal pattern changes may face similar challenges. The large sample, multicentric design, DID analysis and consenus of the physician teams in this study mitigate some limitation posed by purposive/deliberate sampling.
Recommendations
This study suggests new paradigms for rollback and management of type 2 diabetes, with two-meal frequency as major plank for research, especially because it is culturally relevent at least for the Indian subcontinent and does not require drastic reductions in energy intake, hence sustainable as a lifestyle change.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
Acknowledgement
We are grateful to SMBT IMS&RC for offering institutional time and access to services for this study. This study was possible mainly due to several diabetes patients participating in a voluntary trial of 2-OMEX, and several vounteers who helped them sustain through the period. The study team is thankful to several organizations including Indian Medical Association at Nashik and Aurangabad who offered space for weekly diabetes counselling services. We are grateful to the ADORE trust for the 2-OMEX campaign for permitting us to access DRCs and train assistants and prepare case records. We thank the two private clinics that offered to serve as conventional arm clinics for this study. We are grateful to several research assistants who have helped in data collection. We also appreciate the help of Prof. Anil Gore and Prof. Sharayu Gore for statistical insights from time to time, and Prof. Sanjay Mehendale for valuable critical comments. Lastly we are indebted to donors who helped in raising funds including Bharatvaidyaka sanstha for handling financial resources.
REFERENCES
- 1.Anjana RM, Deepa M, Pradeepa R, Mahanta J, Narain K, Das HK, et al. Prevalence of diabetes and prediabetes in 15 states of India:Results from the ICMR–INDIAB population-based cross-sectional study. Lancet Diabetes Endocrinol. 2017;5:585–96. doi: 10.1016/S2213-8587(17)30174-2. [DOI] [PubMed] [Google Scholar]
- 2.Dudeja P, Singh G, Gadekar T, Mukherji S. Performance of Indian diabetes risk score (IDRS) as screening tool for diabetes in an urban slum. Med J Armed Forces India. 2017;73:123–8. doi: 10.1016/j.mjafi.2016.08.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gupta SK, Singh Z, Purty AJ, Vishwanathan M. Diabetes prevalence and its risk factors in urban Pondicherry. International journal of diabetes in developing countries. 2009;29:166. doi: 10.4103/0973-3930.57348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Khan MM, Sonkar GK, Alam R, Mehrotra S, Khan MS, Kumar A, et al. Validity of Indian Diabetes Risk Score and its association with body mass index and glycosylated hemoglobin for screening of diabetes in and around areas of Lucknow. J Family Med Prim Care. 2017;6:366–73. doi: 10.4103/2249-4863.220010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mohan A, Reddy SA, Sachan A, Sarma KVS, Kumar DP, Panchagnula MV, et al. Derivation &validation of glycosylated haemoglobin (HbA1c) cut-off value as a diagnostic test for type 2 diabetes in South Indian population. Indian J Med Res. 2016;144:220–8. doi: 10.4103/0971-5916.195035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Captieux M, Prigge R, Wild S, Guthrie B. Defining remission of type 2 diabetes in research studies:A systematic scoping review. PLoS Med. 2020;17:e1003396. doi: 10.1371/journal.pmed.1003396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Taylor R. Calorie restriction for long-term remission of type 2 diabetes. Clin Med. 2019;19:37–42. doi: 10.7861/clinmedicine.19-1-37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Furmli S, Ramos M, Fung J, Elmasry R, Ramos M, Fung J. Therapeutic use of intermittent fasting for people with type 2 diabetes as an alternative to insulin. BMJ Case Rep. 2018;2018:bcr2017221854. doi: 10.1136/bcr-2017-221854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ojo O, Ojo OO, Adebowale F, Wang XH. The effect of dietary glycaemic index on glycaemia in patients with type 2 diabetes:A systematic review and meta-analysis of randomized controlled trials. Nutrients. 2018;10:373. doi: 10.3390/nu10030373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dixit JV, Badgujar SY, Giri PA. Reduction in HbA1c through lifestyle modification in newly diagnosed type 2 diabetes mellitus patient:A great feat. Journal of Family Medicine and Primary Care. 2022;11:3312–7. doi: 10.4103/jfmpc.jfmpc_1677_21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. [Last accessed on 2022 Oct 17]. https://timesofindia.indiatimes.com/city/mumbai/two-meals-a-day-professor-to-drive-diabetes-fight/articleshow/66700564.cms.
- 12.Harris AD, Mcgregor JC, Perencevich EN, Furuno JP, Zhu J, Peterson DE, et al. the use and interpretation of quasi-experimental studies in medical informatics. [Last accessed on 2021 Dec 17]. Available from: https://academic.oup.com/jamia/article/13/1/16/780729 . [DOI] [PMC free article] [PubMed]
- 13.Reeves BC, Wells GA, Waddington H. Quasi-experimental study designs series—paper 5:A checklist for classifying studies evaluating the effects on health interventions—A taxonomy without labels. J Clin Epidemiol. 2017;89:30–42. doi: 10.1016/j.jclinepi.2017.02.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Des Jarlais DC. TREND (Transparent Reporting of Evaluations with Nonrandomized Designs) Guidel Report Heal Res A User's Man. 2014:156–68. [Google Scholar]
- 15.Longvah T, et al. Indian food Composition Tables, National Institute of Nutrition (ICMR), Table 1. Vol. 1980 Hyderabad: 2017. [Google Scholar]
- 16.R Hemlatha AL, et al. What India Eats 2020;ICMR-NIN, P 11 and 37. [Last accessed on 2022 Sep 22]. accessed online from: https://www.nin.res.in/nutrition2020/what_india_eats.pdf .
- 17.Wing C, Simon K, Bello-Gomez RA. Designing difference in difference studies:Best practices for public health policy research. Annu Rev Public Health. 2018;39:453–69. doi: 10.1146/annurev-publhealth-040617-013507. [DOI] [PubMed] [Google Scholar]
- 18.Fredriksson A, Magalhães De Oliveira G. Fredriksson A, Magalhães De Oliveira G, editors. Impact evaluation using Difference-in-Differences. Impact evaluation using Difference-in-Differences. https://doi.org/10.1108/RAUSP-05-2019-0112. Available from: https://doi.org/10.1108/RAUSP-05-2019-0112 .
- 19.Juray S, Axen K V, Trasino SE, Martínez-Hervás S, González-Navarro H. Remission of Type 2 Diabetes with Very Low-Calorie Diets-A Narrative Review. 2021. [Last accessed on 2021 Dec 11]. Available from: https://doi.org/10.3390/nu13062086 . [DOI] [PMC free article] [PubMed]
- 20.Papakonstantinou E, Oikonomou C, Nychas G, Dimitriadis GD. Effects of diet, lifestyle, chrononutrition and alternative dietary interventions on postprandial glycemia and insulin resistance. Nutrients. 2022;14:823. doi: 10.3390/nu14040823. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Napoleão A, Fernandes L, Miranda C, Marum AP. Effects of calorie restriction on health span and insulin resistance:Classic calorie restriction diet vs. Ketosis-inducing diet. Nutrients. 2021;13:1302. doi: 10.3390/nu13041302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kahleova H, Belinova L, Malinska H, Oliyarnyk O, Trnovska J, Skop V, et al. Eating two larger meals a day (breakfast and lunch) is more effective than six smaller meals in a reduced-energy regimen for patients with type 2 diabetes:A randomised crossover study. Diabetologia. 2014;57:1552–60. doi: 10.1007/s00125-014-3253-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Pimouguet C, Le Goff M, Thiébaut R, Dartigues JF, Helmer C. Effectiveness of disease-management programs for improving diabetes care:A meta-analysis. CMAJ. 2011;183:E115–27. doi: 10.1503/cmaj.091786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Aucott LS, Philip S, Avenell A, Afolabi E, Sattar N, Wild S, et al. Patterns of weight change after the diagnosis of type 2 diabetes in Scotland and their relationship with glycaemic control, mortality and cardiovascular outcomes:A retrospective cohort study. BMJ Open. 2016;6:e010836. doi: 10.1136/bmjopen-2015-010836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ramachandran A, Das AK, Joshi SR, Yajnik CS, Shah S, Kumar KMP. Current status of diabetes in India and need for novel therapeutic agents. J Assoc Physicians India. 2010;58:7–9. [Google Scholar]
- 26.Harris MI. Health Care and Health Status and Outcomes for Patients with Type 2. Diabetes Care. 2000;23:754–8. doi: 10.2337/diacare.23.6.754. [DOI] [PubMed] [Google Scholar]
- 27.WHO Global Report on Diabetes. Global Report on Diabetes. [Last accessed on 2022 Jun 12];Isbn [Internet] 2016 978:6–86. Available from: https://www.who.int/publications/i/item/9789241565257 . [Google Scholar]
- 28.Greenhill C. Therapy:Achieving T2DM remission in primary care. Nat Rev Endocrinol. 2018;14:66. doi: 10.1038/nrendo.2017.176. [DOI] [PubMed] [Google Scholar]
- 29.Dambha-Miller H, Day AJ, Strelitz J, Irving G, Griffin SJ. Behaviour change, weight loss and remission of Type 2 diabetes:A community-based prospective cohort study. Diabet Med. 2020;37:681–8. doi: 10.1111/dme.14122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hopkins MD, Taylor R, Lean MEJ. The DiRECT principles:Giving Type 2 diabetes remission programmes the best chance of success. Diabet Med. 2019;36:1703–4. doi: 10.1111/dme.14126. [DOI] [PubMed] [Google Scholar]
- 31.Kahleova H, Lloren JI, Mashchak A, Hill M, Fraser GE. Meal frequency and timing are associatedwith changes in body mass index in Adventist Health Study 2. J Nutr. 2017;147:1722–8. doi: 10.3945/jn.116.244749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Sarathi V, Kolly A, Chaithanya HB, Dwarakanath CS. High rates of diabetes reversal in newly diagnosed Asian Indian young adults with type 2 diabetes mellitus with intensive lifestyle therapy. J Nat Sci Biol Med. 2017;8:60–3. doi: 10.4103/0976-9668.198343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Bhatt AA, Choudhari PK, Mahajan RR, Sayyad MG, Pratyush DD, Hasan I, et al. Effect of a low-calorie diet on restoration of normoglycemia in obese subjects with type 2 diabetes. Indian J Endocrinol Metab. 2017;21:776–80. doi: 10.4103/ijem.IJEM_206_17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Siaw MY, Chew DE, Toh MP, Seah DE, Chua R, Tan J, et al. Metabolic parameters in type 2 diabetic patients with varying degrees of glycemic control during Ramadan:An observational study. J Diabetes Investig. 2016;7:70–5. doi: 10.1111/jdi.12374. [DOI] [PMC free article] [PubMed] [Google Scholar]