Estimating the attributable risk of vascular disorders in different ranges of fasting plasma glucose and assessing the effectiveness of anti-diabetes agents on risk reduction; questioning the current diagnostic criteria

Abstract

Introduction

Attributable risk of cardiovascular disorders (CVDs) and chronic kidney disease (CKD) in association with diabetes and pre-diabetes is under debate. Moreover, the role of anti-diabetes agents in risk reduction of such conditions is obscure. The purpose of this work is to define the population attributable fraction (PAF) of CVDs and CKD in different rages of plasma glucose.

Method

Iranian stepwise approach for surveillance of non-communicable disease risk factors (STEPs) was used to calculate PAF in four subsequent phases. Phase 0: whole population regardless of diagnosis; Phase I: in three CVD risk groups: minimal risk (FPG < 100 mg/dL), low risk (FPG 100–126 mg/dL), and high risk (FPG ≥ 126 mg/dL) groups; Phase II: three diagnostic groups: normal, pre-diabetes, and diabetes; Phase III: diabetes patients either receiving or not receiving anti-diabetes agents.

Result

A total of 19,503 participants [female-to-male ratio 1.17:1] had at least one FPG measurement and were enrolled. Phase 0: PAF of young adults was lower in the general population (PAF range for CVDs 0.05 ─ 0.27 [95% CI 0.00 ─ 0.32]; CKD 0.03 ─ 0.41 [0.00 ─ 0.62]). Phase I: High-risk group comprised the largest attributable risks (0.46 ─ 0.97 [0.32 ─ 1]; 0.74 ─ 0.95 [0.58 ─ 1]) compared to low-risk (0.16 ─ 0.41 [0.04 ─ 0.66]; 0.29 ─ 0.35 [0.07 ─ 0.5]) and minimal risk groups (negligible estimates) with higher values in young adults. Phase II: higher values were detected in younger ages for diabetes (0.38 ─ 0.95 [0.29 ─ 1]; 0.65 ─ 0.94 [0.59 ─ 1] and pre-diabetes patients (0.15 ─ 0.4 [0.13 ─ 0.45]; 0.26 ─ 0.35 [0.22 ─ 0.4]) but not normal counterparts (negligible estimates). Phase III: Similar estimates were found in both treatment (0.31 ─ 0.98 [0.17 ─ 1]; 0.21 ─ 0.93 [0.12 ─ 1]) and drug-naïve (0.39 ─ 0.9 [0.27 ─ 1]; 0.63 ─ 0.97 [0.59 ─ 1]) groups with larger values for younger ages.

Conclusion

Globalized preventions have not effectively controlled the burden of vascular events in Iran. CVDs and CKD PAFs estimated for pre-diabetes were not remarkably different from normal and diabetes counterparts, arguing current diagnostic criteria. Treatment strategies in high-risk groups are believed to be more beneficial. However, the effectiveness of medical interventions for diabetes in controlling CVDs and CKD burden in Iran is questionable.

Introduction

Diabetes mellitus (DM) accounts for more than 463 million cases worldwide in 2019 [1]. It is well emphasized that DM is inter-related with both micro and macrovascular complications, especially coronary artery disease, cerebrovascular accidents, and chronic kidney disease (CKD), and they pertain main reasons of death in DM [2]. Intensive glycemic control is believed to decrease this risk and serial assessment of organs functioning to be beneficial [3].

Disability attributed to vascular complications and diminished life-expectancy place huge burden on every society and understanding effects of metabolism deviation on vascular events is crucial [4, 5]. Medical professionals have long been entangled with cases lying outside the 95% of ‘normal’ glucose distribution as high-risk strategy of prevention of vascular consequences [6]. Pre-diabetes ‘borderliners’ with glucose measurements close to 95% cut-off, though are at lower risk of CVDs and CKD than diabetes counterparts, because of their vast population size make up major number of cases [7]. Also, little is known about the risk and pattern of CVDs and CKD in younger diabetes patients. One main gap here is the lack of knowledge in regard to attributable risk of vascular consequences to diabetes and pre-diabetes, especially in different age groups. On the other hand, the extent of efficiency of anti-diabetes regimens in controlling CVDs and CKD is not well understood [8].

In this study we hypothesized that attributable risk of CVDs and CKD be higher in DM and participants with worse glycemia control. Also, we hypothesized that participants receiving anti-diabetes agents will have lower attributable risk of vascular events. Herein we used the Iranian stepwise approach for surveillance of non-communicable disease (STEPs) and estimates of population attributable fraction (PAF) of CVDs and CKD in different ranges of plasma glucose and potential effectiveness of anti-diabetes agents.

Methods and material

Study design and participants

Participants recruited in this study were obtained from Iran STEPs 2016 study [9], based on 30,540 enrolled participants. STEPs is a cross-sectional, nation-wide survey, performed to estimate the extent and distribution of metabolic risk factors in Iran. Cluster random sampling method of participants was done to enroll candidates from urban and rural populations of 31 provinces of Iran. Fasting plasma glucose measurement was performed for 19,503 of participants. Venous blood samples were collected after 12h of fasting. An autoanalyzer [Cobas C311 Hitachi, Tokyo, Japan] was used to assess for levels of fasting plasma glucose (FPG) from serum samples. All the laboratory tests and measurements were carried out in a single center with same protocol. For more detailed information please visit the STEPs study protocol [9].

Definitions

Population attributable fraction (PAF)

PAF is used for estimating the fraction of relative risk being directed to a risk condition. It assumes a situation in which if a particular risk (hyperglycemia in current study) could be ideally omitted from a population, extent of decrement in the upcoming disease occurrence (CVDs and CKD) is regarded as PAF. Generally, many risk factors correspond to the occurrence of disorders, elimination of each of them will result in modification of the overall risk of a disease. It should be reminded that sharing of many risk factors in a single individual will lead to cumulative PAFs more than 100% in a population. In other words, multiple risk factors are in action for occurrence of a disease and many of them are not independent and are closely entangled to each other. For instance, diabetes PAF of 0.2 for ischemic stroke does not indicate that 20% of ischemic strokes are caused by DM, but it illustrates that if we eliminate DM cases from that particular population the rate of ischemic stroke will reduce by 20% either by direct effect of DM or indirect contribution of DM in other risk factors such as atherosclerosis.

PAF can be calculated with the following equation [10].

$$ {\displaystyle \begin{array}{c}\ \\ {}\frac{\int_{R^p} RR\left({X}_i\right).f\left({X}_i\right)d{X}_i-1}{\int_{R^p} RR\left({X}_i\right).f\left({X}_i\right) dX}\ \end{array}} $$

Where f(Xi) being the showcase of population proportion being exposed to the risk factor and RR(Xi) representing the relative risk for the hazard. RRs were retrieved from Global Burden of Disease database [11]. In order to better understand the effects of gender on the risk and medical coverage, stratification was carried out in priori.

Phase 0

In this phase CVDs and CKD PAFs were estimated for the total population in different age groups regardless of their health status. This phase will highlight the overall pattern of CVDs and CKD in general population.

Phase I

In this phase, participants were evaluated only by their baseline FPG assessments. Due to normal distribution of FPG, measurements higher than 126 mg/dL were defined as high-risk group for CVDs and CKD [11]. Based on the proposed theoretical minimum value of 88 mg/dL and current practice, participants ranging in between 100 and 126 mg/dL were categorized as low-risk and < 100 mg/dL as minimal risk groups [11]. This phase enables us to apprehend the rough effect of random-sample FPG condition on CVDs and CKD.

Phase II

In this phase, participants were categorized into DM (consisting of those with FPG ≥ 126 mg/dL, or those with values less than this cut-off but prior medical history of DM), pre-diabetes (FPG 100–126 mg/dL, or those with values less than this cut-off but prior medical history of pre-diabetes), and normal (FPG < 100 mg/dL and no prior history of glycemic derangement) based on American Diabetes Association (ADA) guidelines [3]. This phase provides insight on the impacts of diabetes diagnosis and risk of CVDs and CKD.

Phase III

In this phase, a comparison was made between CVDs and CKD PAF of patients with approved diagnosis of DM (either by medical history of diabetes or random FPG measurement of higher than 126 during STEPs study) who never have received treatment and those receiving treatment regardless of their FPG levels, to evaluate treatment coverage effectiveness.

Wealth index

The wealth was quantified based on each household’s expenditure coming from STEPs 2016 survey. Principle component analysis (PCA) was utilized on these measures. Quintiles of the primary component of PCA were named from 1 (first quintile; the poorest) to 5 (fifth quintile; the richest).

Statistical analysis

Descriptive statistics were reported with their count (percentage), mean (±standard deviation [SD]), or median (interquartile range [IQR]), accordingly. Simulation was performed considering normal distribution of FPGs to estimate mean and confidence intervals with random permutation of subjects for each subgroup. PAF values are reported with a 95% confidence interval. Non-overlapping of CIs were considered significant. All the statistical analyses, plots and numbers created in this study were performed by R for windows v 3.6.1 and RStudio v 1.0.136 (http://www.r-project.org/, RRID: SCR_001905).

Results

Participants

Demographic and characteristics of participants enrolled in this study is summarized in Table 1. Total number of samples being available in the STEPs survey was 30,540 that 19,503 had at least one FPG measurement and were eligible to enter upcoming phases. 10,534 (54.01%) of them were female participants. Twelve 5-year age groups were created with similar sample sizes. Considering education, 35-level education assessment scale indicated that most respondents were educated untill the secondary school with median (IQR) of 7 (9) [9]. Participants were evenly dispersed in the 5-level wealth assessing index. 66.32% of respondents lived in the rural areas and 94.32% had basic insurance. Moreover, 20.85% of study population had full-covering insurance. Further demographic and characteristics of participants in each phase can be found in Supplementary Material.

Table 1 demographic and characteristics of enrolled participants

Phase 0

In this phase, attributable risk of vascular events were considered in all the sample population, regardless of their underlying lab findings or clinical background. The PAF estimates of different CVDs layed in a wide range from 0.05 [95% CI 0.03 ─ 0.07] for young females attributable to ischemic heart disease to 0.27 [0.21 ─ 0.32] in middle-aged females due to ischemic stroke. These values for CKD are found to follow a similar pattern as 0.03 [0.00 ─ 0.04] in the youngest female age group [25–29] to 0.48 [0.31 ─ 0.62] in older male population. It is worth notifying the wide confidence intervals especially in younger ages were detected due to lower sample sizes. From the clinical point of view, these findings are common sense, as CVDs and CKD risk were found less likely during younger ages, and age advancement led to signified increase in attributable risk. PAF estimates were similar between both sexes. Fig. 1 summarizes results of phase 0.

Fig. 1
figure1

Population attributable fraction (PAF) of macrovascular (ischemic stroke, hemorrhagic stroke, and ischemic heart disease) and microvascular (chronic kidney disease) complications of diabetes stratified by sex (F: females, M: males) and age

Phase I

In this phase, regardless of the clinical history of participants, the lab results were used for categorization. As illustrated in Fig. 1, participants with FPG ≥ 126 comprised the largest PAF, extending from 0.46 [0.32 ─ 0.59] in older male participants to 0.97 [0.90 ─ 1] for younger ones. Most of the top PAF estimates were found in the younger age subgroups. Similarly, participants with intermediate blood glucose measurement, though being categorized as normal, had intermediate PAFs ranging from 0.16 to 0.41, following similar age and sex pattern as those with FPG ≥ 126. On the other hand, participants with FPGs less than 100 mg/dL had very low PAFs. CKD is an important sequel of uncontrolled blood glucose. It has been calculated that FPG ≥ 126 is responsible for PAFs as large as 0.74 [0.58 ─ 0.89] dispersing almost evenly in different age and sex groups (Fig. 1). Estimated PAF value for participants with 100 < FPG < 126 was in range of 0.29 to 0.35. On the contrary, PAF of CKD in glucose measures less than 100 is negligible for the minimal-risk group. Clinically, a member of population has an imperative attributable risk of vascular events, unless he/she has FPG less than 100 mg/dL, highlighting the need for screening systems.

Phase II

This phase groups participants based on their diagnosis. Attributed risk of CVDs among diagnosed patients with DM is highly affected by their age and sex (Fig. 1). The highest PAF estimate is 0.95 [0.85 ─ 1] for male diabetes cases in their 30s, being attributed to hemorrhagic stroke. Similar finding has been detected for diabetes female patients aged 25–29. On the other hand, elders had lower PAFs. Similar pattern was found for the pre-diabetes group but with lower calculated PAFs. Considering normal diagnosis group, PAF estimates were negligibly close to zero. PAF of CKD in diabetes patients is fundamentally high, ranging from 0.65 [0.59 ─ 0.70] in elderly to as high as 0.94 [0.88 ─ 0.99] in middle age males. This notion is considerably high in pre-diabetes cases, ranging between 0.26 [0.22 ─ 0.29] and 0.35 [0.31 ─ 0.38], indicating higher values in elder groups. This condition is found to be tangentially close to zero in normal population. Practical aspect of this phase is that proper interventions are required for pre-diabetes cases, especially during their elderly.

Phase III

In this phase effects of medication in diabetes patients is evaluated. As is illustrated in Fig. 1, attributable risk of CVDs is almost equal in the two treatment groups, opposed to our primary hypothesis. Highly overlapping CIs highlights the indifference between calculated PAF values, providing insight that patients in both groups having similar attributable risks. PAF of CVDs for diabetes patients receiving medication lies in between 0.31 [0.17 is 0.44] and 0.98 [0.96 ─ 1], greater numbers more commonly found in younger patients. Otherwise, in the group not receiving treatment, PAF ranged from 0.39 to 0.90. Similarly, larger numbers are directed to younger population. Also, PAF of CKD showed similar estimates between two groups of study with highly overlapping CIs, ranging from 0.63 [0.59 ─ 0.66] in 25–29 year-old drug-naïve females to 0.97 [0.93 ─ 1] in 40–44 year-old drug-naïve males. Age and sex seemingly did not have much of effect on the attributable risk of CKD. The PAF estimates of treatment group and drug-naïve patients is excruciatingly similar, questioning the potential efficacy of DM treatment in preventing CVDs and CKD.

Discussion

As we hypothesized, the attributable risk of CVDs and CKD in diabetes patients was higher than pre-diabetes and normal participants. But, opposed to our hypothesis, anti-diabetes agents have not modified the risk of vascular events in diabetes patients. The main finding of current study can be divided into two categories: A) Preventive actions: high-risk group of population and patients with known history of diabetes are at critical risk of CVDs and CKD. Low-risk division of population and cases with approved history of pre-diabetes having lower PAFs compared to the first mentioned groups, but remarkably higher than minimal-risk group. B) Treatment actions: medicinal interventions in order to dampen the blood glucose content seems to pertain not enough effect on the CVDs and CKD risk. The upcoming sections will review and scrutinize these two main findings.

Prevention paradigm or paradox?

Rose offers two ways of approaching a disease: ‘high-risk’ vs ‘mass’ strategies. In the ‘high-risk’ strategy, laboratory cut-offs coming from deviation from ‘normal range’ are employed to categorize a population into high-risk vs low risk (and minimal risk) groups. In the next step, medical interventions are employed to bring back high-risk group members into the normal range and modify their risk. On the other hand, the ‘mass’ strategy applies massive interventions globally in order to not only controlling derangement of high-risk subpopulation but also for borderline and even minimal-risk groups. For instance, while the first strategy deploys anti-diabetes commodities to patients with glucose measures exceeding the laboratory cut-off, the other notion exploits life-style and dietary modification for all community members. Rose indicates that the mass strategy is more successful for controlling disease in population-level and it has not enough effects on patients with prior diagnosis of diabetes. Therefore, while the mass strategy is amply adequate for the general population, a mix of both is the suitable strategy for high-risk groups; a contemplate exemplified as ‘prevention paradox’ [12,13,14,15].

Iran is in fast transition from communicable diseases toward non-communicable conditions [16,17,18]. Iranian health system provides health goods for the community based on ADA’s guideline [19]. This guideline recommends population-level actions in accompaniment of proper and regular screening methods to diagnose high-risk patients and utilization of individual-based treatment approaches for them [20]. Effective individual-level treatment approaches are expected to decrease the gap between PAF in diabetes and pre-diabetes subpopulations. This annotation comes from this belief that effective medical interventions in diabetes cases theoretically leads to decrement of PAFs while the pre-diabetes subpopulation, considering their underlying risk and vast population size, will be responsible for the highest attributed risk. On the other hand, a mix of both population-level and individual-level strategies is responsible for CVDs and CKD risk attenuation in all population, more pronouncedly in high-risk subgroups. From the current survey results, it is evident that the active strategy in Iran has not yielded effective preventive results.

Previous works on the attributed risk of CVDs have indicated that hypertension and hypercholesterolemia are the main factors responsible for such conditions, while diabetes being placed after. Prior experience on the CVDs PAF in diabetes has revealed values between 7 and 13 for Iranian population [21, 22]. These values are lower than the calculated PAFs of current work. However, PAF values are not comparable as it depends on the prevalence of the disease as well as relative risk estimates [10]. The mentioned works focused on data coming from Tehran (capital of Iran) and all age groups together while in the current study whole Iranian population was evaluated. Diabetes care is highly affected by literacy, especially ‘health literacy’ [23]. Lower education level of STEPs sample may be another reason for worse PAF estimates, compared to the mentioned studies. Also, relative risks of our survey were retrieved from the Global Burden of Disease database [11] while other works used values coming from their original collected samples. Unfortunately, experience on the attributable risk of CVDs and CKD in Iran and other developing countries is limited and highly heterogeneous and the time trend in the last decade, considering the transition of burden toward non-communicable diseases, is not clear. Nevertheless, it is conferred that death rate due to diabetes is decreasing contradicting the rising of non-communicable diseases [21]. Altogether, PAF is highly susceptible to variation and its pure value may not convey enough information, while its trend in a community is more enlightening and further work on a prospective cohort is immensely valuable.

Effectiveness of treatment

From the third phase of this study, it is illuminated that the potential effectiveness of anti-diabetes medications on CVDs and CKD risk modification is dubious. The role of medication in vascular diseases reduction is not fully understood. No strong relation has been detected between glycemic control and vascular events [24]. Furthermore, it is suggested that older medications modulate the macrovascular events but not the microvascular ones. Also, there is this chance that hypoglycemia corresponded to older medicines might paradoxically increase the cardiovascular accidents [25]. From previous studies we know that diabetes treatment in Iran is suboptimal and general knowledge of community is deficient [26]. In recent years, programs for identification of high-risk individuals and actions for affordability of medications has resulted in better care, although the rising burden of non-communicable diseases, prominently diabetes, is outpacing these efforts [27]. Use of newer anti-diabetes drugs in recent years has increased substantially but still older ones are the most prescribed medications and have failed in enhancing glycemic control of patients [28].

Diabetes treatment requires multidisciplinary teams and multi organ-system approaching [19]. It is a wise critique to stipulate that FPG is not mechanistically the major contributor of vascular injury and reduction of FPG per se might lack the great intended impact on CVDs and CKD compared to anti-hypertensive and statins [21]. Although, we considered anti-diabetes medications – being easily found in even distant rural areas and modest expense – as a proxy for access to health goods. In other words, we postulated that a diabetic patient not taking medication for his/her diabetes is unlikely to be taking anti-hypertensive drugs or dyslipidemia pills as well.

American Diabetes Association (ADA) and World Health Organization (WHO) have controversial definitions of pre-diabetes [3, 29]. As the WHO’s criteria are more sensitive in diagnosis and less technically demanding, it is believed that this definition is more suitable for developing countries with more limited resources [30]. Therefore, the WHO’s definition will diagnose less cases of pre-diabetes and categorize them as diabetes. Our results suggest that CVDs and CKD PAF of pre-diabetes with definition of ADA is not much higher than the normal population and WHO’s criteria might be more feasible for preventive programs.

Strengths, limitations, and future prospective

One main limitation of current work is low number of cases in certain strata leading to unreliable estimates and wide confidence intervals. Diabetes is a heterogeneous disorder with many etiologies. Here we were not able to analyze the effects of diabetes subtypes, especially type 1 and 2, on the outcome measures. As type 1 diabetes is mostly found in younger patients and treatment approach is not the same as type 2. CVDs are preventable with not only anti-diabetes medications, but necessitating blood pressure, lipid profile, nutrients and dietary regulation, and obesity management. Here we only were limited to glycemia controlling measures. Beside these limitations, there are many strengths directed to this survey. One main strength of the current work is use of a robustly performed nation-wide survey with lowest rate of data heterogeneity. Large sample size, inclusion of many age groups and both sexes provide sufficient evidence to reduce the confounding effects of age and sex. Results of this work empower policy makers about the effectiveness of ongoing plans and modify their acts, accordingly. For the aim of better understanding the epidemiologic aspects of CVDs prevention, it is suggested to run sophisticated analyses with consideration of diabetes type and multi organ-systems physiologic metrics in future studies. Also, it is claimed that newer anti-diabetes medications have better efficacy for controlling CVDs and CKD compared to older ones, further inclusion of different medication and assessment for potential efficacy of CVDs and CKD is an active matter of debate and is encouraged. Prospective real-time analysis of a condition is a luxury but proper clinical practice and legislative acts can be best settled by such measures.

Conclusion

Based on findings of current study, it sounds like population-level preventive measures have almost failed to control the risk of CVDs and CKD related to uncontrolled glycemia. On the other hand, individual-level interventions are seemingly futile to controlling CVDs and CKD burden in large scale. Challenges of DM treatment and care delivery and use of older anti-diabetes medications are supposedly the main limiting factors. As CVDs and CKD place huge burden and expenditure on the health-system, it is suggested to major action plans and paradigm shifts being executed toward utilization and affordability of new generation anti-diabetes drugs, better educational approaches for both physicians and patients, and fertile registries. Moreover, ADA’s definition for pre-diabetes considering FPG cut-off of 100 mg/dL may be over-estimating this condition, as PAFs of this group did not remarkably differ from normal population.

References

  1. 1.

    Saeedi P, Petersohn I, Salpea P, Malanda B, Karuranga S, Unwin N, et al Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the international diabetes federation diabetes atlas, 9th edition. Diabetes Res Clin Pract 2019;157.

  2. 2.

    Rahman S, Rahman T, Ismail AA, Rashid AR. Diabetes-associated macrovasculopathy: pathophysiology and pathogenesis. Diabetes Obes Metab. 2007;9(6):767–80.

    CAS  Article  Google Scholar 

  3. 3.

    American Diabetes Association. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2020. Diabetes Care. 2020;43(Supplement 1):S66.

  4. 4.

    Garofolo M, Gualdani E, Giannarelli R, Aragona M, Campi F, Lucchesi D, et al. Microvascular complications burden (nephropathy, retinopathy and peripheral polyneuropathy) affects risk of major vascular events and all-cause mortality in type 1 diabetes: a 10-year follow-up study. Cardiovasc Diabetol. 2019;18(1):159.

    Article  Google Scholar 

  5. 5.

    Strain WD, Paldánius PM. Diabetes, cardiovascular disease and the microcirculation. Cardiovasc Diabetol. 2018;17(1):57.

    CAS  Article  Google Scholar 

  6. 6.

    Shaye K, Amir T, Shlomo S, Yechezkel S. Fasting glucose levels within the high normal range predict cardiovascular outcome. Am Heart J. 2012;164(1):111–6.

    CAS  Article  Google Scholar 

  7. 7.

    Færch K, Vistisen D, Johansen NB, Jørgensen ME. Cardiovascular risk stratification and management in pre-diabetes. Curr Diab Rep. 2014;14(6):493.

    Article  Google Scholar 

  8. 8.

    Paneni F, Lüscher TF. Cardiovascular protection in the treatment of type 2 diabetes: a review of clinical trial results across drug classes. Am J Cardiol. 2017;120(1s):S17–s27.

    CAS  Article  Google Scholar 

  9. 9.

    Djalalinia S, Modirian M, Sheidaei A, Yoosefi M, Zokaiee H, Damirchilu B, et al. Protocol Design for Large-Scale Cross-Sectional Studies of surveillance of risk factors of non-communicable diseases in Iran: STEPs 2016. Arch Iran Med. 2017;20(9):608–16.

    PubMed  Google Scholar 

  10. 10.

    Farzadfar F, Danaei G, Namdaritabar H, Rajaratnam JK, Marcus JR, Khosravi A, et al. National and subnational mortality effects of metabolic risk factors and smoking in Iran: a comparative risk assessment. Lancet. 2013;381:S47.

    Article  Google Scholar 

  11. 11.

    Gakidou E, Afshin A, Abajobir AA, Abate KH, Abbafati C, Abbas KM, et al. Global, regional, and national comparative risk assessment of 84 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990–2016: a systematic analysis for the global burden of disease study 2016. Lancet. 2017;390(10100):1345–422.

    Article  Google Scholar 

  12. 12.

    Rose G. Strategy of prevention: lessons from cardiovascular disease. Br Med J (Clin Res Ed). 1981;282(6279):1847–51.

    CAS  Article  Google Scholar 

  13. 13.

    Rose G. Sick individuals and sick populations. Int J Epidemiol. 2001;30(3):427–32 discussion 33-4.

    CAS  Article  Google Scholar 

  14. 14.

    Rose G. Sick individuals and sick populations. Int J Epidemiol. 1985;14(1):32–8.

    CAS  Article  Google Scholar 

  15. 15.

    Rose G. Sick individuals and sick populations. 1985. Bull World Health Organ. 2001;79(10):990–6.

    CAS  PubMed  PubMed Central  Google Scholar 

  16. 16.

    Danaei G, et al. Iran in transition. The Lancet. 2019;393(10184):1984–2005.

  17. 17.

    Mohammadi E, et al. Epidemiologic pattern of cancers in Iran; current knowledge and future perspective. Journal of Diabetes & Metabolic Disorders. 2020:1–5.

  18. 18.

    Aminorroaya A, et al. Burden of non-communicable diseases in Iran: past, present, and future. Journal of Diabetes & Metabolic Disorders. 2020:1–7.

  19. 19.

    American Diabetes Association. 9. Pharmacologic Approaches to Glycemic Treatment: Standards of Medical Care in Diabetes-2020. Diabetes Care. 2020;43(Supplement 1):S98.

  20. 20.

    American Diabetes Association 1. Improving Care and Promoting Health in Populations: Standards of Medical Care in Diabetes-2020. Diabetes Care. 2020;43(Supplement 1):S7.

  21. 21.

    Pirani N, Khiavi FF. Population attributable fraction for cardiovascular diseases risk factors in selected countries: a comparative study. Mater Soc. 2017;29(1):35–9.

    Google Scholar 

  22. 22.

    Sardarinia M, Akbarpour S, Lotfaliany M, Bagherzadeh-Khiabani F, Bozorgmanesh M, Sheikholeslami F, et al. Risk Factors for Incidence of Cardiovascular Diseases and All-Cause Mortality in a Middle Eastern Population over a Decade Follow-up: Tehran Lipid and Glucose Study. PLoS One. 2016;11(12):e0167623-e.

    Article  Google Scholar 

  23. 23.

    Cavanaugh KL. Health literacy in diabetes care: explanation, evidence and equipment. Diabetes Manag (Lond). 2011;1(2):191–9.

    Article  Google Scholar 

  24. 24.

    Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet. 1998;352(9131):837–53.

  25. 25.

    Acharya T, Deedwania P. Cardiovascular outcome trials of the newer anti-diabetic medications. Prog Cardiovasc Dis. 2019;62(4):342–8.

    Article  Google Scholar 

  26. 26.

    Mohseni M, Shams Ghoreishi T, Houshmandi S, Moosavi A, Azami-Aghdash S, Asgarlou Z. Challenges of managing diabetes in Iran: meta-synthesis of qualitative studies. BMC Health Serv Res. 2020;20(1):534.

    Article  Google Scholar 

  27. 27.

    Noshad S, Afarideh M, Heidari B, Mechanick JI, Esteghamati A. Diabetes Care in Iran: where we stand and where we are headed. Ann Glob Health. 2015;81(6):839–50.

    Article  Google Scholar 

  28. 28.

    Davari M, Bayazidi Y, Esteghamati A, Larijani B, Kebriaeezadeh A. The prescription pattern of anti-diabetic medication and glycemic control in type 2 diabetes in Iran. A patient-level stud. Diabetes Management. 2020;10(1):1–9.

  29. 29.

    World Health Organization. Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia: report of a WHO/IDF consultation. 2006.

  30. 30.

    Bansal N. Prediabetes diagnosis and treatment: a review. World J Diabetes. 2015;6(2):296–303.

    Article  Google Scholar 

Download references

Acknowledgements

The authors thank the other investigators, the staff, and the participants of STEPs 2016 study for the valuable contribution.

Availability of data and material

Not applicable.

Author information

Affiliations

Authors

Contributions

EM performed statistical analysis, interpreted the data and drafted the manuscript. FSM reviewed the literature. SR performed statistical analysis. SA, MA, EG, SSM, MK, and NR critically revised the manuscript and interpreted the data. NF, KJ, and SN reviewed the literature and revised the manuscript. NE and MRK prepared tables and figures and critically revised the manuscript. BL and FF conceptualized the study and are guarantor of this work and had full access to all the data and take responsibility for the integrity and accuracy of all steps.

Corresponding authors

Correspondence to Bagher Larijani or Farshad Farzadfar.

Ethics declarations

Conflict of interest

No conflicts of interest relevant to this study is reported.

Ethics approval

This study has been approved by the ethics committee of Tehran University of Medical Sciences.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Code availability

Not applicable.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

ESM 1

(XLSX 28 kb)

ESM 2

(DOCX 22 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Mohammadi, E., Morasa, F.S., Roshani, S. et al. Estimating the attributable risk of vascular disorders in different ranges of fasting plasma glucose and assessing the effectiveness of anti-diabetes agents on risk reduction; questioning the current diagnostic criteria. J Diabetes Metab Disord (2020). https://doi.org/10.1007/s40200-020-00663-5

Download citation

Keywords

  • Diabetes mellitus
  • Population attributable fraction
  • Prevention
  • Cardiovascular diseases