Abstract
Purpose
The incidence of prediabetes has been on the rise, indicating a growing public health concern, as individuals with prediabetes are at higher risk of developing type 2 diabetes. This study aimed to determine the effects of simple interventions on the regression of pre-diabetes status into normoglycemia and also prevent progression to diabetes in a pragmatic community trial.
Methods
A total of 2073 (761 intervention; 1,312 controls) participants with pre-diabetes were included in the present secondary data analysis; cases with diabetes or normoglycemia were identified during nine years of follow-up. We utilized multinomial logistic regression to calculate relative risk reductions (RRR, 95% CIs) for educational interventions targeting lifestyle changes in both men and women. Additionally, we employed a linear regression model that considered the ordinal outcomes ranging from normal to prediabetes and diabetes.
Results
In men, after adjusting for confounders, the intervention group had a 53% (95% CI = 1.11–2.10) more significant chance of returning to normoglycemia than the control group after three years of follow-up. In addition, men in the intervention group also had an increased risk of developing type 2 diabetes than men in the control group (RRR = 1.53, 95% CI = 1.02–2.31) after three years of follow-up. These findings among men remained consistent even after a six-year follow-up period. In women, after adjusting for age, the chance of returning to normoglycemia after three years in the intervention group was 1.30 times higher than in women in the control group (95% CI = 1.00-1.69), which disappeared after adjusting for other covariates or after six years of follow-up. The results of the regression analysis showed that the intervention had no effect on changing the status of the outcome from normal to prediabetes and diabetes.
Conclusion
We could not demonstrate any effect of a simple intervention in improving prediabetes. This high-risk population may require more gender-specific intensive interventions and attention.
Keywords: Diabetes, Prediabetes, Pragmatic clinical research, Normoglycemia, Middle-income
Introduction
Type 2 diabetes is accompanied by excess risks of all-cause and cardiovascular mortality, which varies widely depending on age, glycaemic control, and other complications [1]. Therefore, preventing type 2 diabetes and its complications is necessary to reduce the burden of type 2 diabetes across the world [2]. Clinical trials have shown that convenient and inexpensive lifestyle interventions including a healthy diet, physical activity, and weight control in high-risk groups, effectively prevent type 2 diabetes and control cardiovascular risk factors [3, 4]. The Finnish Diabetes Prevention Study and the US Diabetes Prevention Program both reported a 58% reduction in diabetes incidence after about three years of lifestyle intervention among high risk individuals [5]. However, such evidence is scarce in low-resource populations [6, 7]. Given the limited resources of low- and middle-income countries, more cost-effective and equitable interventions to combat the spread of diabetes should be identified and prioritized [8].
Pre-diabetes, an intermediate stage where blood glucose levels are higher than normal but not high enough to be diagnosed as type 2 diabetes, is a significant risk factor for developing type 2 diabetes; studies have highlighted that lifestyle modifications can play a vital role in reversing pre-diabetes to normoglycemia [9]. In Iran, one out of every ten adults has pre-diabetes with an incidence rate of 40 per 1000 person-years; therefore, the increase in the incidence of pre-diabetes and the subsequent progression to diabetes is becoming a severe problem [10]. Clinical trials among the prediabetes population are limited, with most of them being conducted among Western populations. This is particularly true in low- and middle-income countries and the MENA region, where the burden of metabolic disorders and prediabetes is substantial [11]. These circumstances underscore the need for more research in these regions to address the growing public health challenge of prediabetes and its associated complications.
Previously, the results of a community trial in the Tehran Lipid and Glucose Study (TLGS) have shown that these pragmatic interventions have reduced the incidence of metabolic syndrome and diabetes [12, 13]. The aim of the current study, as a secondary data analysis of the TLGS results, is to answer whether these simple pragmatic interventions can play a role in the regression of pre-diabetes status into normoglycemia and prevent the progression to diabetes.
Materials and methods
Study population
The present study used longitudinal data from the Tehran Lipid and Glucose Study (TLGS). The rationale and design of TLGS have been reported before [14]. In brief, TLGS was started in 1999–2001 in the 13th district of Tehran, and its purpose was to study the prevalence, incidence, and risk factors of non-communicable diseases. All subjects were followed up and examined every three years.
Three medical health centers were selected in Tehran’s District 13, and 15,005 residents aged ≥ 3 years were recruited. Pragmatic lifestyle intervention was performed in one out of these three medical centers, and two other medical centers were used as the control. Among 10,368 (6,437 Intervention; 3,931 control) participants aged ≥ 20 years at baseline examination, 2,073 (761 intervention; 1,312 control) people with pre-diabetes were eligible for the present secondary data analysis. We considered the 2nd, 3rd, and 4th examinations after 3, 6, and 9 years of follow-up to find the incident cases of diabetes or normoglycemia. Figure 1 shows the flowchart of the study population.
Fig. 1.
Study flowchart
The Ethical Committee of the Research Institute for Endocrine Sciences approved this study. In the original study, informed consent was obtained from each participant. The TLGS has been registered at Iran Registry for Clinical Trials (http://irct.ir; IRCTID: IRCT138705301058N1).
Intervention
The primary education interventions on lifestyle changes in the intervention area were initiated with a view to encouraging individuals to quit smoking, enhance their physical activity levels and improve dietary patterns. Interventions were implemented within three categories including family-based, community-based, and school-based strategies. The details of this community trial have been published before [12, 15]. In brief:
The family-based intervention included educational sessions (a 2-hour session with video and slide presentation) and publications (“Courier of Health,” every three months).
Community-based programs educate key persons by socially significant figures (law enforcers, clergymen, etc.) and public/group meetings (i.e., lectures during religious ceremonies, e.g., Ramadan, 2–4 events annually, and large attendance seminars to present healthy lifestyles, 2–4 conferences annually).
School-based programs include classroom curriculum, peer education, anti-smoking policies, and some general policies like labelling snacks sold at school shops regarding their healthiness; educating school principals and volunteer teachers on lifestyle modification; educational sessions for parents regarding healthy lifestyles.
Measurements
A trained physician got medical histories and performed a brief physical examination for blood pressure and anthropometrics. Physical activity was assessed with the Lipid Research Clinic questionnaire. Laboratory measurements of fasting plasma glucose (FPG), 2-hour plasma glucose (2 h-PG), triglycerides (TGs), and high-density lipoprotein cholesterol (HDL-C) have been reported previously [16]. All measurements were carried out using the same protocols in both intervention and control areas.
Individuals who did not have a past medical history of glucose-lowering medications were categorized as having pre-diabetes and enrolled in the study if their fasting plasma glucose levels ranged from 100 to 125 mg/dL, or their two-hour postprandial plasma glucose levels after consuming an oral dose of 75 g of glucose fell within the range of 140 to 199 mg/dL. Type 2 diabetes was ascertained based on glucose-lowering medications use, or FPG ≥ 126, or 2 h-PG ≥ 200 mg/dL. Normoglycemia was ascertained when FPG fell < 126 and 2 h-PG < 200 mg/dL without glucose-lowering medications use.
Statistics for secondary data analysis
Baseline characteristics are described for the intervention and control groups as mean (SD) or frequencies (%); the variables are compared between groups using Student’s t-test or chi-square test as appropriate. Missing data for 492 individuals (186 interventions; 306 controls; 24% of all data regarding baseline or follow-up measurements) were imputed and all 2073 subjects (889 men, 1184 women) considered for intention-to-treat (ITT) analysis. The transition rate for progress to diabetes or regress to normoglycemia is reported separately for 3, 6, and 9 years of follow-up.
We used multinomial logistic regression to calculate the intervention’s relative risk ratios (RRR) and 95% confidence intervals (CI). Three categories were defined according to a participant’s glycaemic status at follow-up as the dependent variable; those who remained in the pre-diabetic status (which considered the reference group), those who regressed to normoglycemia, and those who progressed to diabetes. We run the multinomial logistic regression firstly adjusted for age and secondly adjusted for all possible confounders measured at the baseline, including age, BMI, systolic and diastolic blood pressure, FPG, 2 h-PG, HDL-C, TG, education levels (≤ 6, 6–12 and > 12 years of education), low physical activity (< 3 days of performing sports or heavy physical activity per week), current smoking, anti-hypertensive drug use, and family history of type 2 diabetes. This statistical analysis was reported for 3, 6, and 9 years of follow-up. All analyses are sex-specific because sex modified the effect of intervention (p for interaction of sex and intervention < 01). Since the outcomes are not independent and have a certain order, we also applied regression models as a sensitivity analysis, considering the outcome as 0 (regress), 1 (stable), and 2 (progress). All statistical analysis was carried out using STATA version 14 and significancy p-values was considered as < 0.05.
Results
Table 1 shows that at baseline, the women in the control group and the intervention group had the same basic characteristics, but there was a significant difference in the percent of people with low physical activity in the control and intervention groups (26% vs. 18%) (P = 0.001). As shown in Table 2, at the beginning of the study, the men in the intervention group had older age (50.8 vs. 48.8 years), higher BMI (27.2 vs. 26.7 kg/m2), and also lower educational levels compared to control group individuals. All other baseline variables had no significant difference between intervention and control groups’ men.
Table 1.
Baseline Characteristics by intervention group in women: Tehran lipids and glucose study
| Total (N = 1184) | Intervention (N = 433) | Control (N = 751) | P-value | |
|---|---|---|---|---|
| Age, years | 47.6 ± 13.4 | 48.2 ± 13.4 | 47.3 ± 13.3 | 0.27 |
| BMI (kg/m2) | 29.2 ± 5.1 | 29.1 ± 5.2 | 29.2 ± 5.05 | 0.70 |
| SBP (mmHg) | 126.2 ± 20.3 | 126.5 ± 20.8 | 126.0 ± 20.0 | 0.65 |
| DBP (mmHg) | 81.2 ± 10.7 | 81.2 ± 11.0 | 81.2 ± 10.5 | 0.96 |
| FPG (mg/dL) | 99.5 ± 10.8 | 99.0 ± 11.0 | 99.7 ± 10.7 | 0.23 |
| 2 h-PCG | 143.9 ± 27.6 | 143.8 ± 26.9 | 144.0 ± 28.0 | 0.90 |
| HDL-C (mg/dL) | 44.02 ± 10.8 | 44.0 ± 11.0 | 44.03 ± 10.7 | 0.96 |
| TG (mg/dL) | 193 ± 121 | 193 ± 124 | 193 ± 117 | 0.92 |
| Education levels | 0.31 | |||
| - ≤ 6 | 670 (56.5%) | 252 (58.2%) | 418 (55.6%) | |
| - 6–12 | 445 (37.5%) | 152 (35.1%) | 293 (39.01%) | |
| - > 12 | 69 (5.8%) | 29 (6.7%) | 40 (5.3%) | |
| Low physical activity, yes | 275 (23.2%) | 78 (18%) | 197 (26.2%) | 0.001 |
| Current Smoking, yes | 36 (3%) | 9 (2.8%) | 27 (3.6%) | 0.14 |
| Hypertension, yes | 438 (36.9%) | 163 (37.6%) | 275 (36.6%) | 0.71 |
| Anti-hypertensive drug use, yes | 181 (15.2%) | 60 (13.8%) | 121 (16.11%) | 0.29 |
| Family history of T2DM, yes | 380 (32%) | 139 (32.1%) | 241 (32%) | 0.99 |
Values are mean (SD) or n (%) for normal distributed covariates and median (Interquartile range) for skewed (e.g., TG) variables
BMI = Body mass index; SBP = Systolic blood pressure; DBP = Diastolic blood pressure; FPG = Fasting plasma glucose; 2 h-PCG = 2-hour post challenge plasma glucose; T2DM; Type 2 diabetes mellitus; TG = Triglycerides; HDL-C = High density lipoprotein cholesterol
Table 2.
Baseline Characteristics by intervention group in men: Tehran lipids and glucose study
| Total (N = 889) | Intervention (N = 328) | Control (N = 561) | P-value | |
|---|---|---|---|---|
| Age, years | 49.5 ± 14.5 | 50.8 ± 14.9 | 48.8 ± 14.2 | 0.05 |
| BMI (kg/m2) | 26.9 ± 4.08 | 27.2 ± 3.9 | 26.7 ± 4.1 | 0.05 |
| SBP (mmHg) | 126.2 ± 18.9 | 126.2 ± 18.6 | 126.2 ± 19.2 | 0.98 |
| DBP (mmHg) | 80.3 ± 11.5 | 80.2 ± 11.91 | 80.5 ± 11.3 | 0.77 |
| FPG (mg/dL) | 101.7 ± 9.3 | 101.5 ± 9.6 | 101.8 ± 9.3 | 0.72 |
| 2 h-PCG | 134.5 ± 34.5 | 135.2 ± 35.3 | 134.1 ± 34.2 | 0.62 |
| HDL-C (mg/dL) | 38.2 ± 9.4 | 37.8 ± 9.4 | 38.4 ± 9.5 | 0.41 |
| TG (mg/dL) | 206 ± 149 | 212 ± 160 | 202 ± 143 | 0.59 |
| Education levels | 0.001 | |||
| - ≤ 6 | 364 (38.9%) | 154 (46.9%) | 192 (34.2%) | |
| - 6–12 | 410 (46.1%) | 129 (33.3%) | 281 (50.0%) | |
| - > 12 | 133 (15%) | 45 (13.7%) | 88 (15.6%) | |
| Low physical activity, yes | 199 (22.3%) | 74 (22.5%) | 125 (22.2%) | 0.92 |
| Current Smoking, yes | 236 (26.5%) | 83 (25.3%) | 153 (27.2%) | 0.52 |
| Hypertension, yes | 281 (31.6%) | 103 (31.4%) | 178 (31.7%) | 0.91 |
| Anti-hypertensive drug use, yes | 73 (8.2%) | 28 (8.5%) | 45 (8.02%) | 0.78 |
| Family history of T2DM, yes | 244 (27.4%) | 90 (27.4%) | 154 (27.4%) | 0.99 |
Values are mean (SD) or n (%) for normal distributed covariates and median (Interquartile range) for skewed (e.g., TG) variables
BMI = Body mass index; SBP = Systolic blood pressure; DBP = Diastolic blood pressure; FPG = Fasting plasma glucose; 2 h-PCG = 2-hour post challenge plasma glucose; T2DM; Type 2 diabetes mellitus; TG = Triglycerides; HDL-C = High density lipoprotein cholesterol
During the three years of follow-up in men, 35% (95% CI = 33-37%) of the control group and 42% (95% CI = 40-44%) of the intervention group returned to normoglycemia, and 16% (95% CI = 15-17%) of the control group and 19% (95% CI = 18-20%) of the intervention group progressed to type 2 diabetes. In these three years of follow-up in women, 32% (95% CI = 31-33%) of the control group and 39% (95% CI = 38-40%) of the intervention group returned to normal glycemia, and 18% (95% CI = 17-19%) of the control group and 15% (95% CI = 14-16%) of the intervention group progressed to type 2 diabetes.
Table 3 shows the effect of the intervention after three years on the incidence of type 2 diabetes and return to normoglycemia state in men and women separately. After three years of follow-up in men, after adjustment for age, the probability of returning to normoglycemia in the intervention group was 1.50 times that of the control group (95% CI = 1.10–2.04). In addition, the probability of developing type 2 diabetes in men in the intervention group was 1.54 times higher than in the control group (95% CI = 1.04–2.27). After adjusting for other variables, the results did not change meaningfully. Meanwhile after six years of follow-up, the effect of intervention was diminished slightly for regression to normoglycemia (RRR = 1.47, 95% CI = 1.02–2.12). However, the relative risk ratio for progression to diabetes was 1.82 (95% CI = 1.20–2.76)., in 6 years of follow-up. As shown in Table 3, after adjustment for age, in women, the probability of returning to normoglycemia in the intervention group was 1.30 times higher than women in the control group (95% CI = 1.00-1.69). After adjusting for other variables, this effect decreased to 1.25, and its significance was borderline (p = 0.11). As a sensitivity analysis, the results of the regression analysis showed that the intervention had no effect on changing the status of the outcome from normal to prediabetes and diabetes (Table 4).
Table 3.
Multivariable Multinomial Logistic Regression Model for 3, 6 and 9 years of follow-up by male and female: Tehran lipids and glucose study
| Male | Female | |||||||
|---|---|---|---|---|---|---|---|---|
| Regress to normal | Progress to T2DM | Regress to normal | Progress to T2DM | |||||
| RRR (95% CI) | P-value | RRR (95% CI) | P-value | RRR (95% CI) | P-value | RRR (95% CI) | P-value | |
| After 3 years | ||||||||
| Model 1 | ||||||||
| Intervention | 1.50 (1.10–2.04) | 0.009 | 1.54 (1.04–2.27) | 0.02 | 1.30 (1.00-1.69) | 0.05 | 0.87 (0.61–1.23) | 0.44 |
| Model 2 | ||||||||
| Intervention | 1.53 (1.11–2.10) | 0.008 | 1.53 (1.02–2.31) | 0.03 | 1.25 (0.95–1.64) | 0.11 | 0.86 (0.60–1.24) | 0.50 |
| After 6 years | ||||||||
| Model 1 | ||||||||
| Intervention | 1.42 (0.99–2.02) | 0.05 | 1.78 (1.19–2.65) | 0.005 | 1.11 (0.83–1.49) | 0.47 | 0.87 (0.62–1.22) | 0.43 |
| Model 2 | ||||||||
| Intervention | 1.47 (1.02–2.12) | 0.04 | 1.82 (1.20–2.76) | 0.005 | 1.03 (0.75–1.40) | 0.79 | 0.85 (0.60–1.21) | 0.45 |
| After 9 years | ||||||||
| Model 1 | ||||||||
| Intervention | 1.20 (0.76–1.88) | 0.41 | 1.31 (0.83–2.07) | 0.23 | 1.20 (0.76–1.88) | 0.29 | 1.31 (0.83–2.07) | 0.64 |
| Model 2 | ||||||||
| Intervention | 1.32 (0.83–2.10) | 0.29 | 1.41 (0.88–2.26) | 0.14 | 1.15 (0.78–1.70) | 0.48 | 1.11 (0.75–1.65) | 0.60 |
T2DM = Type 2 Diabetes; RRR = Relative Risk Ratio
Model 1 adjusted for age and model 2 further adjusted for all possible confounders measured at the baseline including BMI, systolic and diastolic blood pressure, FPG, 2 h-PG, HDL-C, TG, Education levels (≤ 6, 6–12 and > 12 years of education), Low physical activity (< 3 days of performing sports or heavy physical activity per week), current smoking, anti-hypertensive drug use, and Family history of T2DM
Table 4.
Multivariable Linear Regression Model for 3, 6 and 9 years of follow-up by male and female: Tehran lipids and glucose study
| Male | Female | |||
|---|---|---|---|---|
| β coefficients (95% CI) | P-value | β coefficients (95% CI) | P-value | |
| After 3 years | ||||
| Model 1 | ||||
| Intervention | -0.03 (-0.13-0.07) | 0.529 | -0.10 -0.18–0.02) | 0.016 |
| Model 2 | ||||
| Intervention | -0.03 (-0.12-0.06) | 0.504 | -0.08 (-0.15-0.00) | 0.051 |
| After 6 years | ||||
| Model 1 | ||||
| Intervention | 0.05 [-0.06-0.16) | 0.373 | -0.08 (-0.18-0.02) | 0.106 |
| Model 2 | ||||
| Intervention | 0.04 (-0.06-0.15) | 0.429 | -0.05 (-0.14-0.04) | 0.311 |
| After 9 years | ||||
| Model 1 | ||||
| Intervention | 0.04 (-0.09-0.16) | 0.578 | -0.05 (-0.16-0.06) | 0.375 |
| Model 2 | ||||
| Intervention | 0.03 (-0.09-0.15) | 0.655 | -0.01 (-0.11-0.09) | 0.801 |
T2DM = Type 2 Diabetes
Model 1 adjusted for age and model 2 further adjusted for all possible confounders measured at the baseline including BMI, systolic and diastolic blood pressure, FPG, 2 h-PG, HDL-C, TG, Education levels (≤ 6, 6–12 and > 12 years of education), Low physical activity (< 3 days of performing sports or heavy physical activity per week), current smoking, anti-hypertensive drug use, and Family history of T2DM
Discussion
This study is one of the few studies, especially in middle-income populations, that has examined the effects of simple pragmatic interventions on short-term (3 years) and long-term (up to 9 years) people at high risk of diabetes. According to sex-stratified analysis, the intervention led to an around 50% rise in achieving normoglycemia among men with prediabetes after 3 years of intervention. Nevertheless, men with prediabetes in the intervention program showed also an around 50% increase in the likelihood of progressing to type 2 diabetes after three years compared to men with prediabetes who did not receive the intervention. These results remained consistent even after 6 years of follow-up. For women with prediabetes, a 30% higher percentage of regression to normoglycemia was achieved within 3 years compared to those who did not receive the intervention; there was no significant impact on reducing progression to diabetes among women.
It has been reported that around 5–10% of pre-diabetic individuals progress to diabetes annually, with a similar proportion reverting to normoglycemia [17]. In our study, during three years of follow-up, based on the control and intervention groups and men and women, the percentage of regression to normoglycemia and progression to diabetes ranged from 32 to 42% and from 15 to 19%, respectively. Some studies in line with our findings showed that the regression rate to normoglycemia is more frequent than progression to diabetes [17]. After nine years of follow-up, the rate of progression to diabetes has increased compared to the three-year follow-up period. However, regression to normoglycemia still occurred more frequently than progression to diabetes in our study population. A study conducted by Kohansal et al. as part of the TLGS study found that the ten-year cumulative incidence of normoglycemia and diabetes were approximately equal in the prediabetes population [18]. Paprott et al. reported that 33.8% of the participants reverted to normoglycemia, 20% progressed to diabetes, and 46.2% remained pre-diabetes during a mean 12-year follow-up [19]. Lazo-Porras et al. also confirmed similar results in a study in 2020, which showed that returning to a normoglycemia state is more common than progression to diabetes [20].
In a German cohort study, the female gender was one of the factors affecting the increase in the chance of returning to normal glycemia, which is consistent with the present study findings and probably due to women’s interest in participating in health issues [19]. Our previous study found that the intervention program decreased diabetes risk by 30% in the general population [13]. However, our current study did not show very significant results in the prediabetic population. This suggests that more intensive lifestyle modifications may be necessary for favorable outcomes in high-risk populations. Our study yielded an unexpected finding, revealing that lifestyle intervention may actually increase the risk of progression to diabetes among men. The mechanisms that underlie this phenomenon are not fully clear.
Fritsche et al. examined the effect of lifestyle intervention in high-risk and low-risk subgroups of pre-diabetes and performed the interventions according to DPP protocols. During a 3-year follow-up, intensified interventions were more likely to improve the dysglycemic status of pre-diabetic patients than conventional lifestyle interventions, which was consistent with our study [21]. Guangwei et al. observed a decrease in the incidence of diabetes over six years by examining the findings of the CDQDPOS (China Cohort Study). After adjusting for age and intervention clinics, the incidence of diabetes decreased by 43%. This study also showed that blood glucose levels would remain high in these people if they did not intervene. This study’s results align with the present study’s findings. However, in this study, the effect of the intervention on men and women was not expressed separately [5, 22]. Previous clinical trials have shown that intensive lifestyle or glucose-lowering interventions adversely affect subsequent health outcomes in different subgroups of people with diabetes [23, 24]. The Look AHEAD trial revealed that Intensive Lifestyle Intervention (ILI) was associated with a higher risk of cardiovascular outcomes solely among individuals with poorly controlled diabetes [24]. Moreover, the ACCORD study demonstrated that, compared to standard therapy, the utilization of intensive therapy to target normal glycated hemoglobin levels over a 3.5-year period resulted in increased mortality and did not significantly reduce major cardiovascular events [23]. However, the underlying mechanisms driving these outcomes in both studies remain incompletely understood or clarified. This collective body of research underscores a clinical gap in the effectiveness of interventions among high-risk individuals. However, regarding pragmatic and community trial lifestyle intervention data are limited especially in populations with prediabetes, and further research is needed, especially in gender-specific based structures and also in middle and low-income countries. One possible explanation for this surprising outcome is that the specific components of the lifestyle intervention employed in our study were not well-suited to the male participants. Another contributing factor could be the unique sociocultural aspects and lifestyle patterns observed in our study population. Cultural norms and societal expectations surrounding diet, physical activity, and health behaviors may have influenced the effectiveness of the lifestyle intervention in men. It is possible that the interventions were not adequately tailored to the specific needs and preferences of our male participants, leading to suboptimal results and potentially even exacerbating their risk of developing diabetes. It is important to acknowledge the limitations of our study, including the relatively small sample size and the potential for confounding variables that were not fully accounted for in our analysis.
In a previous study within our community trial, adherence to the intervention was assessed at three levels: community, school, and family. The findings showed that over 80% of households participated in public gatherings during national or religious holidays between examinations. Additionally, the monitoring process indicated successful implementation of 60–70% of planned lifestyle interventions in school settings during follow-ups [12]. While participant-specific adherence data is unavailable for individual analysis, the previous study offered valuable insights into adherence patterns and overall compliance rates in the community trial setting. This broader understanding helps assess adherence and the effectiveness of the intervention across different levels.
The findings should be interpreted in light of several limitations, such as the non-randomized study design. Our study has limitations regarding unmeasured socioeconomic variables and reliance on self-reported data, which may introduce non-randomization bias. However, the geographical proximity and similarity in socioeconomic status among our participants, along with appropriate adjustment in our analysis, help mitigate the potential biases. In addition, the TLGS study lacked data on diabetes knowledge, limiting our ability to investigate its impact on the long-term effectiveness of our intervention. Although increasing diabetes knowledge is a key goal of community-based lifestyle programs, the absence of this information prevented us from understanding whether barriers or reduced knowledge contributed to the lack of efficacy. However, given the relatively consistent knowledge levels among the general population with prediabetes and similar educational backgrounds among study groups, we believe the influence of diabetes knowledge as a confounding factor is minimal.
Since the intervention increases regression to the normal status, it decreases the probability of staying at the prediabetes status. This phenomenon may explain our relative risk ratio (RRR) being greater than 1 for progression to diabetes in men. To address this issue, we employed a linear regression model and assigned an order to the outcome categories from normal (0) to prediabetes (1) and diabetes (2). The regression results indicated no significant effects of the intervention on changing the status from normal to prediabetes and diabetes. It is important to note that this type of analysis is not typically performed for outcomes of this kind, as the weight of change from normal to prediabetes is not equal to the weight of changing from prediabetes to diabetes. Additionally, our study did not include a normal population at baseline, nor did we have a sample transitioning from normal to prediabetes. Therefore, this type of study may not conform to standard analysis protocols.
In conclusion, our findings challenge the prevailing assumptions about the universal benefits of lifestyle interventions in preventing diabetes, emphasizing the need for personalized and gender-specific approaches to diabetes prevention and management. Our study results demonstrate that the generalized interventions are insufficient, highlighting the imperative for more intensive and precise measures for the prediabetes population. Therefore, it is crucial to consider implementing more intensive lifestyle interventions for this high-risk population.
Acknowledgements
We express our thanks to the participants of District 13 of Tehran for their enthusiastic support in this study.
Author contributions
D.K. and Z.D. contributed to the study concept and design; D.K. performed literature research and wrote the manuscript. S.A. created the figures and performed the statistical analysis. Z.D., F.H., and F.A. contributed to the acquisition and interpretation of the data and critical revision of the manuscript. D.K. is the guarantor of this work, has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Funding
This study was supported by the Shahid Beheshti University of Medical Sciences (SBMU).
Data Availability
The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.
Declarations
Conflict of interest
The author(s) has/have no competing interests to declare.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Tancredi M, Rosengren A, Svensson A-M, et al. Excess mortality among persons with type 2 Diabetes. N Engl J Med. 2015;373(18):1720–32. [DOI] [PubMed] [Google Scholar]
- 2.Nwaneri C, Cooper H, Bowen-Jones D. Mortality in type 2 Diabetes Mellitus: magnitude of the evidence from a systematic review and meta-analysis. Br J Diabetes Vascular Disease. 2013;13(4):192–207. [Google Scholar]
- 3.Pan X-R, Li G-w, Hu Y-H, et al. Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance: the Da Qing IGT and Diabetes Study. Diabetes Care. 1997;20(4):537–44. [DOI] [PubMed] [Google Scholar]
- 4.Ramachandran A, Snehalatha C, Mary S, Mukesh B, Bhaskar A, Vijay V. The Indian Diabetes Prevention Programme shows that lifestyle modification and metformin prevent type 2 Diabetes in Asian Indian subjects with impaired glucose tolerance (IDPP-1). Diabetologia. 2006;49(2):289–97. [DOI] [PubMed] [Google Scholar]
- 5.Gong Q, Zhang P, Wang J, et al. Morbidity and mortality after lifestyle intervention for people with impaired glucose tolerance: 30-year results of the Da Qing Diabetes Prevention Outcome Study. The Lancet Diabetes & Endocrinology. 2019;7(6):452–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Viitasalo K, Hemiö K, Puttonen S, et al. Prevention of Diabetes and Cardiovascular Diseases in occupational health care: feasibility and effectiveness. Prim Care Diabetes. 2015;9(2):96–104. [DOI] [PubMed] [Google Scholar]
- 7.Gilis-Januszewska A, Szybinski Z, Kissimova-Skarbek K, et al. Prevention of type 2 Diabetes by lifestyle intervention in primary health care setting in Poland: Diabetes in Europe Prevention using Lifestyle, physical activity and nutritional intervention (DE-PLAN) project. Br J Diabetes Vascular Disease. 2011;11(4):198–203. [Google Scholar]
- 8.Bakhtiari A, Takian A, Majdzadeh R, Haghdoost AA. Assessment and prioritization of the WHO best buys and other recommended interventions for the prevention and control of non-communicable Diseases in Iran. BMC Public Health. 2020;20(1):1–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sallar A, Dagogo-Jack S. Regression from prediabetes to normal glucose regulation: state of the science. Experimental Biology and Medicine. 2020;245(10):889–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hadaegh F, Derakhshan A, Zafari N, et al. Pre-diabetes tsunami: incidence rates and risk factors of pre‐Diabetes and its different phenotypes over 9 years of follow‐up. Diabet Med. 2017;34(1):69–78. [DOI] [PubMed] [Google Scholar]
- 11.Azizi F, Hadaegh F, Hosseinpanah F, et al. Metabolic health in the Middle East and North Africa. The Lancet Diabetes & Endocrinology. 2019;7(11):866–79. [DOI] [PubMed] [Google Scholar]
- 12.Khalili D, Asgari S, Lotfaliany M, et al. Long-term effectiveness of a lifestyle intervention: a pragmatic community trial to prevent metabolic syndrome. Am J Prev Med. 2019;56(3):437–46. [DOI] [PubMed] [Google Scholar]
- 13.Lotfaliany M, Mansournia MA, Azizi F, et al. Long-term effectiveness of a lifestyle intervention on the prevention of type 2 Diabetes in a middle-income country. Sci Rep. 2020;10(1):1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Azizi F, Zadeh-Vakili A, Takyar M. Review of rationale, design, and initial findings: Tehran lipid and glucose study. Int J Endocrinol Metabolism. 2018;16(4 Suppl). [DOI] [PMC free article] [PubMed]
- 15.Khalili D, Azizi F, Asgari S et al. Outcomes of a longitudinal population-based cohort study and pragmatic community trial: findings from 20 years of the Tehran lipid and glucose study. Int J Endocrinol Metabolism. 2018;16(4 Suppl). [DOI] [PMC free article] [PubMed]
- 16.Azizi F, Ghanbarian A, Momenan AA, et al. Prevention of non-communicable Disease in a population in nutrition transition: Tehran lipid and glucose study phase II. Trials. 2009;10(1):1–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Tabák AG, Herder C, Rathmann W, Brunner EJ, Kivimäki M. Prediabetes: a high-risk state for Diabetes development. The Lancet. 2012;379(9833):2279–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kohansal K, Ahmadi N, Hadaegh F, et al. Determinants of the progression to type 2 Diabetes and regression to normoglycemia in people with pre-diabetes: a population-based cohort study over ten years. Prim Care Diabetes. 2022;16(6):797–803. [DOI] [PubMed] [Google Scholar]
- 19.Paprott R, Scheidt-Nave C, Heidemann C. Determinants of change in glycemic status in individuals with prediabetes: results from a nationwide cohort study in Germany. Journal of diabetes research. 2018;2018. [DOI] [PMC free article] [PubMed]
- 20.Lazo-Porras M, Bernabe-Ortiz A, Ruiz-Alejos A, et al. Regression from prediabetes to normal glucose levels is more frequent than progression towards Diabetes: the CRONICAS Cohort Study. Diabetes Res Clin Pract. 2020;163:107829. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Fritsche A, Wagner R, Heni M, et al. Different effects of Lifestyle intervention in high-and low-risk prediabetes: results of the Randomized Controlled prediabetes Lifestyle intervention study (PLIS). Diabetes. 2021;70(12):2785–95. [DOI] [PubMed] [Google Scholar]
- 22.Li G, Zhang P, Wang J, et al. The long-term effect of lifestyle interventions to prevent Diabetes in the China Da Qing Diabetes Prevention Study: a 20-year follow-up study. The Lancet. 2008;371(9626):1783–9. [DOI] [PubMed] [Google Scholar]
- 23.Gerstein H, Miller M, Byington R et al. Action to Control Cardiovascular Risk in Diabetes Study, Group. Effects of intensive glucose lowering in type. 2008;2:2545–59. [DOI] [PMC free article] [PubMed]
- 24.Bancks MP, Chen H, Balasubramanyam A, et al. Type 2 Diabetes subgroups, risk for Complications, and differential effects due to an intensive lifestyle intervention. Diabetes Care. 2021;44(5):1203–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.