Life Course Dietary Insulinemic and Inflammatory Potential and Risk of Type 2 Diabetes in U.S. Women: A Prospective Cohort Study

Yin Zhang; Mingyang Song; Molin Wang; Emily Riseberg; Dong D Wang; Deirdre K Tobias; Fred K Tabung; A Heather Eliassen; Stephanie A Smith-Warner; Qi Sun; Eric B Rimm; Meir J Stampfer; Walter C Willett; Frank B Hu; Edward L Giovannucci

doi:10.1016/j.jand.2026.156831

. Author manuscript; available in PMC: 2026 Sep 27.

Published before final editing as: J Acad Nutr Diet. 2026 Aug 20:156831. doi: 10.1016/j.jand.2026.156831

Life Course Dietary Insulinemic and Inflammatory Potential and Risk of Type 2 Diabetes in U.S. Women: A Prospective Cohort Study

Yin Zhang ¹, Mingyang Song ^1,^2,^3,⁴, Molin Wang ^2,^5,⁶, Emily Riseberg ¹, Dong D Wang ^1,⁶, Deirdre K Tobias ^1,⁷, Fred K Tabung ^1,^8,⁹, A Heather Eliassen ^1,^2,⁶, Stephanie A Smith-Warner ^1,², Qi Sun ^1,^2,⁶, Eric B Rimm ^1,^2,⁶, Meir J Stampfer ^1,^2,⁶, Walter C Willett ^1,², Frank B Hu ^1,^2,⁶, Edward L Giovannucci ^1,²

PMCID: PMC13615478 NIHMSID: NIHMS2206984 PMID: 42624368

Abstract

Background

The timing of exposure to diet across the lifespan may be critical in the development of T2D. However, no previous study has deciphered the influence of dietary insulinemic and inflammatory potential on the risk of T2D across the lifespan from a life course perspective.

Objective

This study aimed to evaluate the associations of dietary insulinemic and inflammatory potential with the risk of T2D from a life course perspective.

Design

This was a prospective cohort study.

Participants and setting

Data from 40,135 eligible, female participants in the Nurses’ Health Study II were analyzed. Adulthood diet was assessed quadrennially since 1991 using 131-item food frequency questionnaires (FFQ), and adolescent diet was recalled in 1997 using a 124-item high-school FFQ. The main exposures were empirical dietary index for hyperinsulinemia (EDIH) and empirical dietary inflammatory pattern (EDIP) scores across different life stages (adolescence, premenopausal adulthood, postmenopausal adulthood) and changes and cumulatively over the lifetime.

Main outcome measures

The main outcome was incident T2D.

Statistical analyses performed

Cox models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI).

Results

Higher EDIH and EDIP scores (highest vs lowest quintiles) were associated with increased lifetime risk of T2D as a lifetime average (HR, 95%CI: 2.72, 2.38–3.11 and 2.04, 1.81–2.30), during premenopausal adulthood (3.18, 2.52–4.01 and 2.31, 1.88–2.82), and postmenopausal adulthood (2.67, 2.15–3.33 and 1.70, 1.41–2.05), but not during adolescence (1.07, 0.95–1.20 and 1.10, 0.98–1.24). The HR, 95%CI associated with higher lifetime averages for both EDIH and EDIP (vs. low lifetime averages for both, based on tertiles) was 2.64 (2.32–3.01). Individuals with high adulthood EDIH or EDIP had similar magnitudes of lifetime risk elevation, regardless of their adolescent EDIH and EDIP status. Adolescent EDIH and EDIP were associated with a slightly increased premenopausal T2D risk (1.24, 1.02–1.51, and 1.24, 1.02–1.50). In additional analyses estimating the time window during which adulthood dietary insulinemic and inflammatory potential influences T2D risk, higher adulthood EDIH or EDIP was associated with an increased risk of T2D with a very short time lag.

Conclusions

Over the life course in women, high dietary insulinemic and inflammatory potential in both premenopausal and postmenopausal adulthood were independently associated with a substantially increased lifetime risk of T2D. Adulthood offers the most critical time window for dietary interventions to reduce lifetime T2D risk, though adolescent diet may influence the risk of premenopausal T2D.

Keywords: Dietary insulinemic potential, Dietary inflammatory potential, Life course, Type 2 diabetes, Nurses’ Health Study II

Introduction

As the leading preventable cause of type 2 diabetes (T2D), suboptimal diet contributes to over 70% of incident cases globally.¹ The American Diabetes Association emphasizes the importance of adopting healthy eating patterns for T2D prevention in the 2026 Standards of Care.²

Mounting evidence has solidified the robust connections between many dietary patterns and the risk of T2D.^1,3–6 Both mechanistically and epidemiologically, dietary insulinemic and inflammatory potential has been suggested to play a pivotal role.^5,7–13 Dietary insulinemic and inflammatory potential refer to the capacity of whole diets to stimulate hyperinsulinemia and promote systemic inflammation.^14–17 Over time, different dietary indices have been developed using either empirically derived, data-driven approaches (based on food-group–level prediction models of circulating insulinemic and inflammatory biomarkers) or hypothesis-driven, literature-derived approaches.^16–18 Dietary patterns associated with markers of hyperinsulinemia and inflammation have been reported to have a considerably stronger association with T2D than a spectrum of other major dietary patterns.⁵

A life course perspective in chronic disease epidemiology conceptualizes health and disease risk as the consequence of exposures acting across the lifespan.¹⁹ Life course dietary exposures have been implicated in the etiology of major chronic conditions, such as cardiovascular disease, cancer, and diabetes more broadly.^20–22 The timing of exposure to diet across the lifespan may be critical in the development of T2D.^23–26 However, no previous study has deciphered the influence of dietary insulinemic and inflammatory potential on the risk of T2D across the lifespan from a life course perspective.

This study aimed to address the following unanswered questions: What are the associations of dietary insulinemic and inflammatory potential with T2D risk across major life stages (adolescence, premenopausal adulthood, postmenopausal adulthood)? What are their joint associations across major life stages? What are the associations between changes in dietary insulinemic and inflammatory potential from adolescence to adulthood and the risk of T2D? Does lowering dietary insulinemic and inflammatory potential have immediate benefits in reducing the risk of T2D in adulthood, or is there a time-lag?

Methods

Study Population

The Nurses’ Health Study II (NHSII) is a well-established large U.S. longitudinal cohort initiated in 1989, enrolling 116,429 female registered nurses aged 25 to 42 years.²⁷ Demographics were collected at cohort enrollment. Throughout follow-up, participants updated information on a wide spectrum of health-related topics biennially or quadrennially, attaining a follow-up rate exceeding 90%. We used 1997 as the analytical baseline, as this was the follow-up cycle when participants were asked to provide comprehensive information on their adolescent diet. Participants were excluded from analyses if they had a prior diagnosis of T2D, cardiovascular disease, or cancer (except nonmelanoma skin cancer) before/at baseline (major changes in diet that often occur following the diagnosis of these conditions may reflect the disease rather than an individual’s long-term or habitual diet, potentially introducing reverse causation), reported no information on their baseline diet or adolescent diet, or with implausible energy intakes (<600 or >3500 kcal/day for baseline diet, <500 or >5000 kcal/day for high school diet ^28–31), leaving 40,135 eligible participants for analysis. (Supplementary Figure 1)

Ethical Approval

The study protocol was approved by the Institutional Review Boards of the Brigham and Women’s Hospital and Harvard T.H. Chan School of Public Health (Boston, MA) and those of participating registries (as required). Informed consent was implied by participants’ completion and return of the questionnaires. Written informed consent was obtained from participants to retrieve medical records.

Ascertainment of Diet

Beginning in 1991 in the NHSII, adulthood diet was assessed quadrennially throughout follow-up using semiquantitative 131-item food frequency questionnaires (FFQs) that inquired about participants’ diet during the previous year.³² Participants additionally completed a comprehensive supplemental 124-item food frequency questionnaire (HS-FFQ) in the 1997 follow-up cycle about their diet during high school (between the ages of 13–18 years).³² Cohort participants reported their consumption frequency of each food based on a specified standardized portion size (or a commonly used unit), with 9 response categories ranging from “never or <1 time per month” to “≥6 times per day”. Average daily nutrient intake was estimated by multiplying the consumption frequency of each food by its nutrient content (based primarily on the U.S. Department of Agriculture Nutrient Database and supplemented by other published sources).^33,34 The validity and reproducibility of FFQ ^32,35–37 and HS-FFQ^32,38,39 in the NHSII have been comprehensively reported. There were no major differences between participants who completed and who did not complete the HS-FFQ across a range of lifestyle factors.⁴⁰

Ascertainment of Dietary Insulinemic and Inflammatory Potential over the Life Course

The exposures of interest included the empirical dietary index for hyperinsulinemia (EDIH) ^14,15 and empirical dietary inflammatory pattern (EDIP) ^16,17 scores during adolescence (calculated based on the HS-FFQ), premenopausal adulthood (calculated as the cumulative average intake from the baseline FFQ in 1991 to the last FFQ completed prior to the onset of menopause or the diagnosis of type 2 diabetes), postmenopausal adulthood (calculated as the cumulative average intake from the first FFQ after menopause to the diagnosis of type 2 diabetes or the end of follow-up), lifetime average exposures (calculated as the average of adolescent, premenopausal average, and postmenopausal average [if applicable] intakes), and change in exposures from adolescence to adulthood (calculated as the difference between adolescent and the cumulative average of adulthood intakes).^41–43 (Supplementary Figure 2) Menopause was used as the cutoff to specifically categorize adulthood into premenopausal and postmenopausal stages, aiming to better tailor the physiological status of women.⁴¹ Repeated measurements were leveraged to better depict long-term dietary intake and reduce measurement error from random within-subject variation.^44,45

The EDIH score is a data-driven index assessing the potential of whole diets to influence long term insulin level derived based on food groups’ ability in predicting fasting plasma C-peptide concentrations, the development and validation of which have been elaborated previously.^14,15 In brief, using stepwise linear regression, a dietary pattern most predictive of the concentrations of fasting plasma C-peptide was identified. The EDIH score was defined as the weighted sum of these food groups, with higher scores indicating higher insulinemic potential. Components contributing to higher EDIH scores with higher intake include processed meat, red meat, fish other than dark-meat fish (including canned tuna, shrimp, lobster, scallops, and other fish or seafood varieties that are not classified as dark-meat fish, such as cod, haddock, and halibut), poultry, eggs, high-energy beverages (including cola with sugar, other carbonated beverages with sugar, and fruit punch drinks), low-energy beverages (including low-energy cola and other low-energy carbonated beverages), tomatoes, cream soups, margarine, butter, French fries, and low-fat dairy products, while components contributing to lower EDIH scores with higher intake include coffee, whole fruits, high-fat dairy products, and green leafy vegetables. (Supplementary Figure 3) The EDIH is superior to the dietary II (the dietary insulin index) in predicting both fasting and non-fasting C-peptide concentrations.¹⁵ Ranges and standard deviations (SDs) of EDIH scores across life stages: life course: −0.29 to 2.32 (0.23); adolescence: −0.42 to 4.29 (0.33); premenopausal adulthood: −0.38 to 2.31 (0.23); postmenopausal adulthood: −1.25 to 5.24 (0.25).

Similarly, the EDIP score is a data-driven index reflecting overall inflammatory potential of whole diets derived according to food groups’ ability in predicting plasma concentrations of three inflammatory markers (interleukin-6, C-reactive protein, and tumor necrosis factor α receptor 2). Details of the development and validation of EDIP have been described elsewhere.^16,17 Briefly, using reduced rank regression and stepwise linear regression, a dietary pattern most predictive of the concentrations of inflammatory markers was identified, based on which EDIP score was calculated as the weighted sum of these food groups, with higher scores indicating higher inflammatory potential. Processed meat, red meat, organ meat, fish other than dark-meat fish, vegetables other than green leafy or dark yellow vegetables, refined grains, high-energy beverages, low-energy beverages, and tomatoes are components that contribute to higher EDIP scores with higher intake, while tea, coffee, dark yellow vegetables, green leafy vegetables, snacks, fruit juice, and pizza are components that contribute to lower scores with higher intake. (Supplementary Figure 3) The EDIP has a greater ability than the DII (a literature-derived dietary inflammatory index) in predicting concentrations of plasma inflammatory markers in this cohort.¹⁷ Ranges and SDs of EDIP scores across life stages: life course: −1.83 to 1.74 (0.22); adolescence: −2.26 to 2.76 (0.30); premenopausal adulthood: −1.90 to 2.68 (0.26); postmenopausal adulthood: −4.80 to 2.26 (0.28).

Ascertainment of Incident T2D

Physician-diagnosed incident T2D was initially identified through biennial self-administered questionnaires. A supplementary questionnaire was then administered to confirm the diagnosis by gathering information on symptoms, diagnostic tests, and hypoglycemic therapy.⁴⁶ Cases identified prior to 1998 were confirmed if they satisfied a minimum of one criterion outlined by the National Diabetes Data Group:⁴⁷ 1) presence of elevated glucose concentration with fasting plasma glucose ≥140 mg/dL (≥7.8 mmol/L), random plasma glucose ≥200 mg/dL (≥11.1 mmol/L), or 2-h plasma glucose ≥200 mg/dL (≥11.1 mmol/L) after oral glucose load, accompanied by at least one related symptom (excessive thirst, polyuria, weight loss, or hunger); 2) absence of symptoms, but with elevated glucose concentrations on at least two occasions; 3) treatment with insulin or other hypoglycemic medications. Cases identified from 1998 onward were confirmed using the American Diabetes Association criteria,⁴⁸ with the cut-off point for elevated fasting plasma glucose adjusted to 126 mg/dL (7.0 mmol/L). Cases identified after 2010 were confirmed based on the updates in American Diabetes Association criteria,⁴⁹ with further consideration given to HbA1c ≥ 6.5%. Previous validation study within the NHS showed that 98.4% of self-reported cases of T2D, confirmed through the supplementary questionnaire, were reconfirmed through a review of medical records.⁵⁰

Ascertainment of Covariates

Self-reported adulthood height, race and ethnicity, as well as recalled adolescent characteristics, were collected once at enrollment. Throughout the cohort follow-up, adulthood information on current body weight (which was used to calculate adulthood body mass index [BMI] with height), smoking behavior, multivitamin use, menopausal status, postmenopausal hormone use, and oral contraceptive use were self-reported biennially. Physical activity (frequencies of engaging in common recreational activities, based on which metabolic equivalent of tasks [MET] scores were assigned and total physical activity in MET-hours per week was calculated), diet, and family history of diabetes were self-reported quadrennially. Waist circumference was self-reported in 1993 and 2005. Neighborhood socioeconomic status (nSES) was assessed using U.S. Census tract-level data from the Neighborhood Change Database, which provides harmonized Census data across multiple decades.⁵¹ Seventeen variables reflecting education, employment, housing, income, and demographic composition were selected to characterize neighborhood environments.⁵¹ Principal component analysis was used to derive an nSES score reflecting neighborhood affluence.⁵¹ The validity and reproducibility of reported information on anthropometrics, diet, lifestyle, menopausal status, and oral contraceptive use have previously been elaborated.^{32,35–37,52–56}

Statistical Analysis

Participants contributed person-years of follow-up from the return of the predefined analytical baseline questionnaire (1997) until the date of diagnosis of T2D, death recorded, or predefined follow-up completion (2019), whichever was earliest.

The proportional hazards assumption was verified using interactions between the exposures and log-transformed follow-up time. Given that no violation was detected, Cox models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) of T2D across quintiles of lifetime average, adolescent, premenopausal adulthood, and postmenopausal adulthood EDIH and EDIP, as well as HRs and 95% CIs per 1-SD increase in EDIH and EDIP across various life stages. The associations between adolescent EDIH and EDIP and the risk of premenopausal T2D were additionally quantified. HRs and 95% CIs by comparing lifetime consistent high (high in both adolescence and adulthood) vs. consistent low (low in both adolescence and adulthood) EDIH and EDIP, as well as the joint associations of lifetime EDIH and EDIP with the outcome using collapsed categories (based on tertiles), were further examined. Moreover, the associations between change in EDIH and EDIP from adolescence to adulthood and T2D risk were assessed. Multivariable-adjusted cumulative incidence of T2D in individuals with consistently high vs. low EDIH or EDIP throughout adolescence, premenopausal adulthood, and postmenopausal adulthood were also estimated, as well as stratified by BMI status.

Covariates were selected a priori. Fully-adjusted multivariable analyses were stratified by age and follow-up cycle; adjusted for basic characteristics (race, and family history of diabetes), adolescent characteristics (BMI at age 18, smoking status, physical activity, alcohol intake, total energy intake, multivitamin use, age at menarche, and oral contraceptive use in adolescence), and adulthood characteristics (smoking status, pack-years of smoking, physical activity, alcohol intake, total energy intake, multivitamin use, neighborhood socioeconomic status, menopausal status and postmenopausal hormone use, and oral contraceptive use). Specifically, in analyses of EDIH and EDIP across different life course periods, each dietary score in adolescence, premenopausal adulthood, and postmenopausal adulthood were further mutually adjusted to better disentangle their independent effect.

Recognizing the potential role of weight change since age 18 and waist circumference as mediators in the associations of interest, these factors were additionally adjusted in secondary analysis. Sensitivity analysis with a 4-year latency period were also performed. Besides, subgroup analyses of lifetime EDIP and EDIH according to cumulative-averaged BMI, and BMI at age 18 were conducted. Additive interaction was assessed using relative excess risk due to interaction (RERI) based on the delta method.⁵⁷

Finally, by setting the cutoffs for the ‘remote’ and ‘recent’ periods at 6, 8, 10, and 12 years,⁵⁸ a series of additional analyses were performed to comprehensively examine the potential time window in the associations between dietary insulinemic and inflammatory potential and the risk of T2D in adulthood based on predefined ‘hard’ thresholds on the adulthood analytical timeline. Taking the analysis based on the ‘remote’ and ‘recent’ time periods using the 10-year hard threshold as an example: The first “remote period” was 1997–1999, the first “subsequent 10-year period” was 1999–2009. The second “remote period” was 1997–2001, the second “subsequent 10-year period” was 2001–2011. The last “remote period” was 1997–2007, then the last “subsequent 10-year period” was 2007–2017. The follow-up period was from the end of the first “subsequent 10-year period” (i.e. 2009) to 2019. The same analytical strategy was used in analysis based on the hard thresholds of 6, 8, 10, and 12 years. This analytical approach rigorously accounts for the independent effects of the ‘remote’ and ‘recent’ period exposures. The cutoffs employed, based on a spectrum of candidate ‘hard’ thresholds on the analytical timeline, allowed us to reduce the confounding of remote exposure from recent exposure through mutual adjustment, thereby approximating the potential time window. (Supplementary Figure 2)

Data analyses were performed using SAS statistical software, version 9.4 for UNIX.⁵⁹ All tests were 2-sided, with P values <.05 indicating statistical significance.

Results

Population Characteristics

A total of 3,201 incident T2D cases were documented. Populations characteristics across lifetime and adolescent dietary insulinemic/inflammatory potential were reported in Tables 1–2. Participants with higher lifetime dietary insulinemic potential tended to have higher BMI and waist circumference. They also tended to use fewer multivitamins but more oral contraceptives during both adolescence and adulthood. Additionally, they were more likely to have a family history of diabetes or cardiovascular disease. We observed essentially the same variations across quintiles of adolescent dietary insulinemic potential. Similar differences in population characteristics were also noted when comparing individuals with higher versus lower dietary inflammatory potential. Individuals with a higher dietary insulinemic/inflammatory potential during adolescence were more likely to have a higher dietary insulinemic/inflammatory potential in adulthood.

Table 1.

Adolescent and adulthood characteristics by lifetime dietary insulinemic and inflammatory potential in the Nurses’ Health Study II (N = 40,135)

Characteristics ^a	Lifetime Empirical Dietary Index for Hyperinsulinemia (EDIH)					Lifetime Empirical Dietary Inflammatory Pattern (EDIP)

	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5

No. of participants (N)	6,256	7,026	7,815	8,685	10,353	6,390	6,868	7,817	8,887	10,173
Person-years	159,971	160,341	160,493	160,483	160,836	159,636	160,021	160,417	160,889	161,159
Age, years	53.2 (7.7)	52.8 (7.8)	52.4 (7.8)	51.7 (7.8)	50.6 (7.7)	53.5 (7.7)	53.1 (7.8)	52.4 (7.8)	51.4 (7.8)	50.4 (7.7)
White, %	95.0	95.0	95.0	94.8	95.0	95.7	95.9	95.4	94.5	93.0
Black, %	0.4	0.5	0.6	0.9	0.9	0.3	0.4	0.6	0.9	1.1
Hispanic, %	0.7	0.6	0.8	0.9	0.8	0.4	0.5	0.7	0.9	1.3
Asian, %	1.2	1.1	1.0	0.7	0.5	0.4	0.6	0.7	1.1	1.9
Height, cm	165 (6.6)	165 (6.5)	165 (6.6)	165 (6.6)	165 (6.5)	165 (6.6)	165 (6.5)	165 (6.5)	165 (6.6)	165 (6.6)
Family history of diabetes, %	29.6	32.5	34.6	36.4	37.9	31.3	32.0	33.6	35.8	38.4
Family history of cardiovascular disease, %	37.5	40.2	41.7	43.0	44.2	39.5	40.9	41.0	41.7	43.5
*Adolescent characteristics*
EDIH ^b, %	6.2	22.4	47.7	77.1	94.2	30.2	38.7	46.7	57.9	74.9
EDIP ^b, %	23.6	37.4	49.0	62.2	76.2	11.1	26.1	46.8	71.6	91.6
Body mass index at age 18 ^c, kg/m²	20.7 (2.7)	20.8 (2.7)	20.9 (2.9)	21.1 (3.1)	21.5 (3.5)	20.9 (2.7)	20.8 (2.7)	20.9 (2.8)	21.1 (3.1)	21.5 (3.7)
Ever smokers, %	21.1	21.8	22.1	23.5	24.0	26.7	23.6	22.0	20.0	20.1
Physical activity ^d, metabolic equivalent tasks-hours/week	53.1 (36.4)	51.7 (36.0)	51.2 (35.9)	52.3 (36.8)	53.7 (37.4)	54.0 (37.1)	51.9 (35.6)	51.3 (35.8)	51.8 (36.7)	53.1 (37.3)
Alcohol intake, g/day	1.0 (3.4)	1.0 (3.1)	1.0 (3.1)	1.0 (3.4)	1.2 (4.0)	1.2 (3.8)	1.0 (3.4)	1.0 (3.2)	0.9 (3.1)	0.9 (3.5)
Total energy intake, kcal/day	2878 (791)	2654 (768)	2616 (765)	2666 (768)	2879 (786)	2844 (805)	2667 (766)	2638 (761)	2684 (764)	2863 (797)
Multivitamin use, %	21.5	18.1	15.8	13.4	11.7	20.5	17.8	16.0	14.1	12.2
Age at menarche, years	12.5 (1.4)	12.5 (1.4)	12.4 (1.4)	12.4 (1.4)	12.3 (1.4)	12.4 (1.4)	12.5 (1.4)	12.5 (1.4)	12.4 (1.4)	12.3 (1.4)
Oral contraceptive use, %	19.1	20.3	20.7	22.5	26.0	21.6	21.5	21.2	21.2	23.4
*Adulthood characteristics*
EDIH ^b, %	7.0	26.1	51.9	74.8	90.2	19.1	33.6	49.1	65.6	83.0
EDIP ^b, %	22.8	38.1	50.1	62.9	76.4	8.1	25.1	51.8	75.2	89.7
Body mass index ^c, kg/m²	23.5 (4.0)	24.4 (4.4)	25.1 (4.7)	25.9 (5.2)	27.3 (6.0)	24.1 (4.2)	24.5 (4.4)	24.9 (4.8)	25.7 (5.2)	27.0 (6.0)
Waist circumference, cm	80.9 (12.0)	83.3 (13.0)	85.2 (13.4)	87.4 (14.3)	90.7 (15.8)	82.5 (12.7)	83.5 (13.0)	84.8 (13.6)	86.6 (14.3)	89.8 (15.9)
Ever smokers, %	31.3	32.0	32.5	34.0	34.4	37.9	34.0	32.3	30.0	29.9
Pack-years of smoking ^e	11.3 (10.1)	12.2 (10.6)	13.5 (11.4)	14.4 (12.1)	16.8 (13.6)	13.0 (11.4)	12.5 (10.9)	13.0 (11.0)	13.9 (11.7)	16.3 (13.5)
Physical activity ^d, metabolic equivalent tasks-hours/week	30.1 (25.8)	24.6 (22.6)	22.0 (20.3)	19.9 (18.7)	17.6 (17.6)	27.9 (24.7)	24.1 (21.4)	22.5 (21.0)	21.1 (20.9)	18.7 (18.7)
Alcohol intake, g/day	5.1 (6.9)	4.8 (6.7)	4.5 (6.7)	4.2 (6.5)	3.8 (6.6)	5.6 (7.2)	5.1 (6.8)	4.5 (6.6)	3.9 (6.4)	3.2 (6.2)
Total energy intake, kcal/day	1915 (450)	1775 (441)	1743 (446)	1749 (452)	1844 (480)	1916 (460)	1789 (439)	1752 (446)	1756 (453)	1814 (478)
Multivitamin use, %	61.4	60.4	58.6	57.6	54.4	61.0	60.1	59.0	57.5	54.9
Menopausal status and postmenopausal hormone use, %
Premenopausal	49.0	49.5	49.9	50.5	51.4	45.8	48.1	50.5	52.6	53.7
Postmenopausal-never	24.0	21.9	21.1	20.1	18.5	24.4	22.4	20.8	19.5	18.2
Postmenopausal-current	13.1	13.4	13.3	14.0	14.6	14.3	13.6	13.7	13.1	13.7
Postmenopausal-past	13.9	15.3	15.7	15.5	15.5	15.4	15.8	15.0	14.8	14.4
Oral contraceptive use, %	82.7	85.2	86.3	87.1	88.7	84.7	85.9	86.1	86.2	87.1

Open in a new tab

Data are expressed as mean (standard deviation) for continuous variables and percentage for categorical variables. All variables are age-standardized except age.

Percentage of participants with EDIH/EDIP scores above the 50^th percentile.

Calculated as weight in kilograms divided by the square of height in meters.

Weekly energy expenditure in metabolic equivalent tasks-hours/week from recreational and leisure time physical activity.

Cumulative among smokers.

Table 2.

Adolescent and adulthood characteristics in 1997 by adolescent dietary insulinemic and inflammatory potential in the Nurses’ Health Study II (N = 40,135)

Characteristics ^a	Adolescent Empirical Dietary Index for Hyperinsulinemia (EDIH)					Adolescent Empirical Dietary Inflammatory Pattern (EDIP)

	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5

No. of participants (N)	7,900	7,907	8,014	8,071	8,243	7,901	7,903	8,003	8,079	8,249
Person-years	160,389	160,497	160,507	160,366	160,363	160,504	160,468	160,455	160,444	160,252
Age, years	42.3 (4.6)	42.4 (4.7)	42.6 (4.6)	42.8 (4.7)	42.8 (4.6)	41.8 (4.6)	42.5 (4.7)	42.8 (4.7)	42.9 (4.7)	42.9 (4.6)
White, %	94.4	94.9	94.9	94.6	94.9	95.4	95.7	95.4	94.6	92.8
Black, %	0.6	0.5	0.7	1.0	0.8	0.4	0.5	0.8	1.0	0.9
Hispanic, %	0.8	0.6	0.9	0.9	0.9	0.6	0.4	0.5	0.8	1.7
Asian, %	1.4	1.2	0.9	0.7	0.5	0.5	0.5	0.7	1.1	1.7
Height, cm	165 (6.6)	165(6.7)	165 (6.5)	165 (6.5)	165 (6.6)	165 (6.6)	165 (6.5)	165 (6.6)	165 (6.6)	165 (6.6)
Family history of diabetes, %	21.1	22.8	24.9	26.5	27.4	21.8	22.1	23.7	25.9	28.7
Family history of cardiovascular disease, %	29.6	31.0	32.9	33.8	36.1	31.2	30.7	32.7	33.4	35.4
Adulthood EDIH ^b, %	37.7	49.5	55.8	63.5	72.2	42.5	49.3	56.4	61.9	68.6
Adulthood EDIP ^b, %	44.3	49.4	52.2	55.1	58.5	38.3	47.1	53.3	58.2	62.3
*Adolescent characteristics*
Body mass index at age 18 ^c, kg/m²	20.7 (2.9)	20.9 (2.8)	21.1 (3.0)	21.3 (3.2)	21.6 (3.6)	20.8 (2.7)	20.9 (2.8)	21.0 (2.9)	21.2 (3.2)	21.6 (3.7)
Ever smokers, %	17.8	20.1	22.1	25.1	28.8	23.9	22.4	21.7	22.6	23.7
Physical activity ^d, metabolic equivalent tasks-hours	55.6 (37.4)	52.3 (36.2)	51.8 (36.3)	50.7 (35.8)	52.7 (37.4)	55.9 (37.3)	52.6 (36.1)	51.9 (36.6)	50.8 (35.9)	52.0 (37.1)
Alcohol intake, g/day	0.8 (3.1)	0.9 (3.1)	1.0 (3.0)	1.1 (3.5)	1.4 (4.3)	1.2 (3.7)	1.0 (3.4)	1.0 (3.5)	0.9 (3.0)	1.0 (3.5)
Total energy intake, kcal/day	2972 (770)	2628 (764)	2571 (763)	2632 (764)	2902 (786)	2952 (800)	2631 (760)	2584 (758)	2637 (753)	2905 (785)
Multivitamin use, %	21.5	17.7	15.5	13.5	11.5	21.3	18.5	15.9	13.3	11.0
Age at menarche, years	12.5 (1.4)	12.5 (1.4)	12.4 (1.4)	12.4 (1.4)	12.3 (1.4)	12.5 (1.4)	12.5 (1.4)	12.4 (1.4)	12.4 (1.4)	12.3 (1.4)
Oral contraceptive use, %	17.4	19.3	20.7	23.8	27.8	20.9	21.4	21.4	20.7	24.8
*Adulthood characteristics*
Body mass index ^c, kg/m²	23.8 (4.5)	24.1 (4.6)	24.6 (4.8)	25.1 (5.2)	25.9 (5.8)	23.8 (4.4)	24.1 (4.5)	24.5 (4.8)	25.0 (5.2)	26.0 (5.9)
Waist circumference, cm	83.6 (13.8)	84.4 (13.8)	85.9 (14.3)	87 (14.9)	89.2 (15.8)	83.8 (13.8)	84.4 (13.8)	85.7 (14.2)	86.8 (14.8)	89.3 (16.0)
Ever smokers, %	26.9	30.0	32.9	35.9	40.2	33.3	32.3	32.2	32.9	35.5
Pack-years of smoking ^e	10.8 (9.1)	11.6 (9.2)	12.1 (9.6)	13.1 (10.2)	14.6 (10.4)	11.9 (9.6)	11.8 (9.6)	12.1 (9.5)	12.8 (9.8)	14.4 (10.5)
Physical activity ^d, metabolic equivalent tasks-hours	25.4 (25.6)	22.9 (22.9)	21.8 (22.7)	20.7 (22.6)	19.5 (21.3)	25.4 (25.3)	23.1 (23.2)	22.0 (23.3)	20.2 (21.6)	19.5 (21.9)
Alcohol intake, g/day	3.1 (5.3)	3.4 (5.7)	3.4 (5.9)	3.5 (6.0)	3.7 (6.5)	3.8 (6.0)	3.7 (6.2)	3.4 (5.8)	3.2 (5.8)	3.0 (5.9)
Total energy intake, kcal/day	1882 (492)	1779 (477)	1760 (489)	1761 (491)	1806 (505)	1869 (488)	1777 (478)	1755 (486)	1767 (491)	1819 (512)
Multivitamin use, %	55.3	53.9	52.0	50.8	49.2	56.3	53.9	51.3	50.5	49.3
Menopausal status and postmenopausal hormone use, %
Premenopausal	91.5	91.0	90.9	90.1	88.4	90.7	91.5	90.8	90.5	88.4
Postmenopausal-never	1.3	1.4	1.2	1.4	1.3	1.6	1.0	1.2	1.3	1.6
Postmenopausal-current	6.6	6.9	7.2	7.8	9.4	7.0	6.7	7.4	7.6	9.2
Postmenopausal-past	0.6	0.8	0.7	0.7	0.8	0.7	0.7	0.7	0.7	0.9
Oral contraceptive use, %	81.6	84.1	85.9	87.0	89.1	83.7	84.8	85.8	86.1	87.4

Open in a new tab

Data are expressed as mean (standard deviation) for continuous variables and percentage for categorical variables. All variables are age-standardized except age.

Percentage of participants with EDIH/EDIP scores above the 50^th percentile.

Calculated as weight in kilograms divided by the square of height in meters.

Weekly energy expenditure in metabolic equivalent tasks-hours/week from recreational and leisure time physical activity.

Cumulative among smokers.

Lifetime, Adolescent, Premenopausal Adulthood, and Postmenopausal Adulthood Dietary Insulinemic Potential and Risk of T2D

In fully-adjusted analyses, increased lifetime average EDIH was associated with significantly higher lifetime risk of T2D (HR, 95%CI: 2.72, 2.38–3.11; comparing the highest vs. lowest quintiles). Greater EDIH during premenopausal and postmenopausal adulthood was associated with a substantial increase in the risk of T2D (3.18, 2.52–4.01 and 2.67, 2.15–3.33, respectively). The association between adolescent EDIH and lifetime T2D risk was less clear (1.07, 0.95–1.20). The HR (95% CI) per 1-SD increase in EDIH was 1.38 (1.34, 1.43) for lifetime average intakes, 1.06 (1.02–1.10) for adolescence, 1.44 (1.36, 1.52) for premenopausal adulthood, and 1.23 (1.17, 1.29) for postmenopausal adulthood. (Figures 4–5; Tables 3–6)

Figure 4. — Lifetime dietary insulincmic and inflammatory potential and risk of type 2 diabetes in the Nurses’ Health Study H (N = 40,135)

Figure 5. — Adolescent, premenopausal adulthood, and postmenopausal adulthood dietary insulinemic and inflammatory potential and risk of type 2 diabetes in the Nurses’ Health Study II (N = 40,135)

Table 3.

Hazard ratios (95% confidence intervals) for lifetime dietary insulinemic and inflammatory potential and risk of type 2 diabetes in the Nurses’ Health Study II (N = 40,135)

	Lifetime Empirical Dietary Index for Hyperinsulinemia (EDIH)							Lifetime Empirical Dietary Inflammatory Pattern (EDIP)

	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a

No. of participants (40135 in total)	6256	7026	7815	8685	10353			6390	6868	7817	8887	10173
No. of cases (3201 in total)	288	412	575	770	1156			391	434	593	684	1099
Age-adjusted ^b	1 [Ref]	1.45 (1.24–1.68)	2.04 (1.77–2.35)	2.78 (2.43–3.19)	4.38 (3.85–4.99)	1.60 (1.56–1.66)	<.001	1 [Ref]	1.13 (0.98–1.29)	1.58 (1.39–1.79)	1.87 (1.65–2.12)	3.15 (2.80–3.54)	1.50 (1.45–1.55)	<.001
Adjusted for basic and adolescent characteristics ^c	1	1.41 (1.21–1.64)	1.89 (1.64–2.18)	2.45 (2.14–2.81)	3.61 (3.16–4.11)	1.50 (1.45–1.55)	<.001	1	1.12 (0.98–1.29)	1.54 (1.35–1.75)	1.72 (1.51–1.95)	2.59 (2.30–2.92)	1.38 (1.33–1.42)	<.001
Adjusted for basic, adolescent, and adulthood characteristics ^d	1	1.28 (1.10–1.49)	1.63 (1.41–1.88)	2.00 (1.75–2.30)	2.72 (2.38–3.11)	1.38 (1.34–1.43)	<.001	1	1.06 (0.93–1.22)	1.38 (1.22–1.58)	1.48 (1.30–1.68)	2.04 (1.81–2.30)	1.27 (1.23–1.32)	<.001
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e	1	1.11 (0.95–1.29)	1.26 (1.09–1.45)	1.37 (1.19–1.58)	1.58 (1.38–1.81)	1.15 (1.11–1.20)	<.001	1	1.01 (0.88–1.16)	1.19 (1.04–1.35)	1.18 (1.04–1.34)	1.39 (1.23–1.57)	1.12 (1.08–1.16)	<.001

Open in a new tab

P-value for trend was calculated using dietary insulinemic/inflammatory potential score modeled as a continuous variable

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for basic characteristics (including race, and family history), and adolescent characteristics (including body mass index at age 18, smoking status, physical activity, alcohol intake, total energy intake, multivitamin use, age at menarche, and oral contraceptive use in adolescence).

Stratified by age and follow-up cycle; adjusted for above-mentioned basic and adolescent characteristics; and additionally adjusted for adulthood characteristics (including smoking status, pack-years of smoking, physical activity, alcohol intake, total energy intake, multivitamin use, neighborhood socioeconomic status, menopausal status and postmenopausal hormone use, and oral contraceptive use).

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

Table 6.

Hazard ratios (95% confidence intervals) for postmenopausal adulthood dietary insulinemic and inflammatory potential and risk of type 2 diabetes in the Nurses’ Health Study II (N = 34,892)

	Postmenopausal adulthood Empirical Dietary Index for Hyperinsulinemia (EDIH)							Postmenopausal adulthood Empirical Dietary Inflammatory Pattern (EDIP)

	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a

No. of participants (34892 in total)	7916	6572	6398	6538	7468			7841	6561	6409	6564	7517
No. of cases (1725 in total) ^b	134	230	333	400	628			213	269	308	384	551
Age-adjusted ^c	1 [Ref]	1.75 (1.41–2.16)	2.60 (2.12–3.18)	3.18 (2.61–3.87)	5.06 (4.19–6.11)	1.50 (1.44–1.56)	<.001	1 [Ref]	1.30 (1.08–1.55)	1.46 (1.22–1.74)	1.85 (1.56–2.19)	2.63 (2.24–3.08)	1.41 (1.35–1.47)	<.001
Adjusted for basic and adolescent characteristics ^d	1	1.75 (1.41–2.16)	2.55 (2.08–3.12)	3.07 (2.52–3.74)	4.56 (3.77–5.51)	1.44 (1.38–1.50)	<.001	1	1.31 (1.09–1.57)	1.49 (1.25–1.78)	1.85 (1.56–2.19)	2.43 (2.07–2.85)	1.36 (1.30–1.42)	<.001
Adjusted for basic, adolescent, and adulthood characteristics ^e	1	1.62 (1.31–2.02)	2.21 (1.80–2.70)	2.56 (2.10–3.13)	3.44 (2.83–4.17)	1.33 (1.28–1.39)	<.001	1	1.28 (1.07–1.53)	1.42 (1.19–1.69)	1.67 (1.41–1.98)	2.02 (1.71–2.38)	1.27 (1.21–1.33)	<.001
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^f	1	1.38 (1.11–1.71)	1.68 (1.37–2.07)	1.74 (1.42–2.13)	1.93 (1.59–2.35)	1.15 (1.10–1.20)	<.001	1	1.19 (0.99–1.43)	1.25 (1.05–1.50)	1.35 (1.14–1.61)	1.40 (1.19–1.65)	1.12 (1.07–1.17)	<.001
Adjusted for basic, adolescent, and adulthood characteristics ^e + adolescent EDIH/P and premenopausal adulthood EDIH/P ^g	1	1.49 (1.20–1.86)	1.94 (1.57–2.39)	2.16 (1.75–2.66)	2.67 (2.15–3.33)	1.23 (1.17–1.29)	<.001	1	1.21 (1.01–1.46)	1.30 (1.09–1.56)	1.49 (1.25–1.78)	1.70 (1.41–2.05)	1.21 (1.14–1.28)	<.001
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^f + adolescent EDIH/P and premenopausal adulthood EDIH/P ^g	1	1.33 (1.07–1.66)	1.61 (1.30–1.98)	1.65 (1.33–2.03)	1.78 (1.43–2.22)	1.11 (1.05–1.17)	<.001	1	1.16 (0.96–1.39)	1.21 (1.01–1.45)	1.29 (1.07–1.54)	1.30 (1.07–1.57)	1.10 (1.04–1.17)	<.001

Open in a new tab

P-value for trend was calculated using dietary insulinemic/inflammatory potential score modeled as a continuous variable

Participants who experienced both their premenopausal adulthood and postmenopausal adulthood during the follow-up were included in this analysis.

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

Additionally adjusted for adolescent EDIH/P and premenopausal adulthood EDIH/P.

Lifetime, Adolescent, Premenopausal Adulthood, and Postmenopausal Adulthood Dietary Inflammatory Potential and Risk of T2D

A higher lifetime average EDIP was associated with a significant increase in the lifetime risk of T2D (HR, 95%CI: 2.04, 1.81–2.30; comparing the highest vs. lowest quintiles). The association between adolescent EDIP and lifetime risk of T2D was not statistically significant (1.10, 0.98–1.24). Increased EDIP during premenopausal and postmenopausal adulthood was associated with substantially higher risk of T2D (2.31, 1.88–2.82 and 1.70, 1.41–2.05, respectively). The HR (95% CI) per 1-SD increase in EDIP was 1.27 (1.23–1.32) for lifetime average intakes, 1.05 (1.01–1.09) for intakes in adolescence, 1.32 (1.25–1.40) for premenopausal adulthood, and 1.21 (1.14–1.28) for postmenopausal adulthood. (Figures 4–5; Tables 3–6)

Adolescent Dietary Insulinemic and Inflammatory Potential and Risk of Premenopausal T2D

Adolescent EDIH was associated with a slightly increased premenopausal T2D risk (HR 1.24, 95% CI: 1.02–1.51; comparing the highest vs. lowest quintiles). Similarly, we detected a slight increase in the risk of premenopausal T2D in relation to adolescent EDIP (HR 1.24, 95% CI: 1.02–1.50). The HR of premenopausal T2D per 1-SD increase in adolescent EDIH was 1.10 (95% CI: 1.04–1.17), and for adolescent EDIP, it was 1.07 (95% CI: 1.01–1.13). (Table 7)

Table 7.

Hazard ratios (95% confidence intervals) for adolescent dietary insulinemic and inflammatory potential and risk of premenopausal type 2 diabetes in the Nurses’ Health Study II (N = 37,528)

	Adolescent Empirical Dietary Index for Hyperinsulinemia (EDIH)							Adolescent Empirical Dietary Inflammatory Pattern (EDIP)

	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a

No. of participants (37528 in total)	7493	7462	7513	7503	7557			7505	7455	7490	7535	7543
No. of cases (1211 in total)	163	182	230	252	384			172	177	239	235	388
Age-adjusted ^b	1 [Ref]	1.13 (0.91–1.40)	1.43 (1.17–1.75)	1.58 (1.29–1.92)	2.37 (1.97–2.85)	1.34 (1.27–1.41)	<.001	1 [Ref]	1.07 (0.87–1.32)	1.44 (1.18–1.75)	1.42 (1.17–1.73)	2.33 (1.95–2.80)	1.34 (1.27–1.41)	<.001
Adjusted for basic and adolescent characteristics ^c	1	1.10 (0.89–1.36)	1.31 (1.07–1.60)	1.36 (1.11–1.66)	1.91 (1.58–2.30)	1.25 (1.19–1.32)	<.001	1	1.06 (0.86–1.31)	1.36 (1.12–1.66)	1.26 (1.03–1.54)	1.78 (1.48–2.14)	1.20 (1.14–1.26)	<.001
Adjusted for basic, adolescent, and adulthood characteristics ^d	1	1.11 (0.90–1.38)	1.30 (1.06–1.60)	1.31 (1.07–1.60)	1.73 (1.43–2.10)	1.21 (1.14–1.27)	<.001	1	1.04 (0.84–1.28)	1.28 (1.05–1.56)	1.13 (0.93–1.38)	1.51 (1.25–1.82)	1.13 (1.07–1.19)	<.001
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e	1	1.04 (0.84–1.29)	1.10 (0.89–1.35)	1.07 (0.87–1.31)	1.32 (1.09–1.60)	1.10 (1.04–1.16)	<.001	1	1.04 (0.84–1.28)	1.21 (0.99–1.48)	0.96 (0.78–1.17)	1.22 (1.01–1.48)	1.06 (1.01–1.12)	.03
Adjusted for basic, adolescent, and adulthood characteristics ^d + premenopausal adulthood EDIH/P ^f	1	0.96 (0.78–1.19)	1.04 (0.85–1.28)	0.99 (0.81–1.21)	1.24 (1.02–1.51)	1.10 (1.04–1.17)	.001	1	0.94 (0.76–1.17)	1.10 (0.90–1.34)	0.94 (0.76–1.15)	1.24 (1.02–1.50)	1.07 (1.01–1.13)	.02
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e + premenopausal adulthood EDIH/P ^f	1	0.97 (0.78–1.20)	0.98 (0.80–1.21)	0.93 (0.75–1.14)	1.13 (0.93–1.37)	1.05 (0.99–1.12)	.08	1	0.97 (0.79–1.21)	1.09 (0.89–1.34)	0.84 (0.69–1.04)	1.08 (0.89–1.31)	1.03 (0.97–1.09)	.35

Open in a new tab

P-value for trend was calculated using dietary insulinemic/inflammatory potential score modeled as a continuous variable

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

Additionally adjusted for premenopausal adulthood EDIH/P.

Lifetime Consistent High vs. Low Dietary Insulinemic and Inflammatory Potential and Risk of T2D

Individuals with consistently high EDIH or EDIP throughout their life course faced a significantly higher lifetime risk of T2D (HR, 95%CI: 2.64, 2.25–3.08 and 1.84, 1.60–2.12, respectively) than those with consistently low EDIH or EDIP in both adolescence and adulthood (comparing the highest vs. lowest tertiles). Notably, similar effect estimates were observed for individuals who were low in adolescence but high in adulthood (2.60, 2.17–3.11 and 1.72, 1.45–2.04, respectively), as well as for those who were intermediate in adolescence but high in adulthood (2.64, 2.24–3.12 and 1.78, 1.53–2.08, respectively). (Figure 6;Tables 8–9)

Figure 6. — Lifetime consistent high versus low dietary insulinemic and inflammatory potential and risk of type 2 diabetes in the Nurses’ Health Study II (N = 40,135)

Table 8.

Hazard ratios (95% confidence intervals) for lifetime consistent high versus low dietary insulinemic potential and risk of type 2 diabetes in the Nurses’ Health Study II (N = 40,135)

	Lifetime Empirical Dietary Index for Hvoerinsulinemia (EDIH) ^a (adolescence - adulthood)

	Low – Low	Low – Intermediate	Low – High	Intermediate – Low	Intermediate – Intermediate	Intermediate – High	High – Low	High – Intermediate	High – High

No. of participants (40135 in total)	5281	4520	3365	3541	4725	5082	2442	4064	7115
No. of cases (3201 in total)	205	262	302	190	305	530	174	349	884
Age-adjusted ^b	1 [Ref]	1.92 (1.60–2.30)	3.56 (2.98–4.26)	1.34 (1.10–1.63)	1.95 (1.63–2.33)	3.95 (3.36–4.64)	1.73 (1.41–2.12)	2.54 (2.14–3.02)	4.47 (3.84–5.21)
Adjusted for basic and adolescent characteristics ^c	1	1.83 (1.52–2.20)	3.27 (2.73–3.91)	1.24 (1.02–1.52)	1.77 (1.48–2.12)	3.42 (2.90–4.02)	1.53 (1.24–1.87)	2.16 (1.82–2.58)	3.52 (3.02–4.11)
Adjusted for basic, adolescent, and adulthood characteristics ^d	1	1.56 (1.30–1.88)	2.60 (2.17–3.11)	1.19 (0.98–1.45)	1.53 (1.28–1.82)	2.64 (2.24–3.12)	1.42 (1.16–1.74)	1.81 (1.52–2.16)	2.64 (2.25–3.08)
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e	1	1.29 (1.07–1.56)	1.65 (1.38–1.98)	1.13 (0.92–1.38)	1.21 (1.01–1.46)	1.63 (1.38–1.93)	1.21 (0.98–1.49)	1.36 (1.14–1.63)	1.58 (1.35–1.85)

Open in a new tab

‘Low’, ‘Intermediate’, and ‘High’ were defined based on tertiles.

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

Table 9.

Hazard ratios (95% confidence intervals) for lifetime consistent high versus low dietary inflammatory potential and risk of type 2 diabetes in the Nurses’ Health Study II (N = 40,135)

	Lifetime Empirical Dietary Inflammatory Pattern (EDIP) ^a (adolescence - adulthood)

	Low – Low	Low – Intermediate	Low – High	Intermediate – Low	Intermediate – Intermediate	Intermediate – High	High – Low	High – Intermediate	High – High

No. of participants (40135 in total)	5425	4316	3422	4208	4408	4718	3566	3894	6178
No. of cases (3201 in total)	271	234	271	232	291	476	306	327	793
Age-adjusted ^b	1 [Ref]	1.13 (0.95–1.34)	2.05 (1.73–2.43)	1.10 (0.92–1.31)	1.28 (1.08–1.51)	2.37 (2.04–2.76)	1.76 (1.49–2.08)	1.66 (1.41–1.95)	2.93 (2.55–3.37)
Adjusted for basic and adolescent characteristics ^c	1	1.16 (0.97–1.38)	1.99 (1.67–2.35)	1.06 (0.89–1.27)	1.26 (1.06–1.48)	2.18 (1.88–2.54)	1.55 (1.31–1.82)	1.48 (1.25–1.74)	2.37 (2.05–2.72)
Adjusted for basic, adolescent, and adulthood characteristics ^d	1	1.09 (0.91–1.30)	1.72 (1.45–2.04)	0.99 (0.83–1.19)	1.11 (0.94–1.32)	1.78 (1.53–2.08)	1.33 (1.12–1.57)	1.24 (1.05–1.46)	1.84 (1.60–2.12)
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e	1	1.10 (0.92–1.31)	1.39 (1.17–1.65)	0.99 (0.82–1.18)	1.04 (0.88–1.23)	1.37 (1.18–1.60)	1.11 (0.94–1.31)	1.09 (0.92–1.28)	1.34 (1.16–1.55)

Open in a new tab

‘Low’, ‘Intermediate’, and ‘High’ were defined based on tertiles.

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

On the other hand, among individuals with low EDIH or EDIP during adulthood, high levels in adolescence were associated with a modestly increased lifetime risk of T2D (1.42, 1.16–1.74 and 1.33, 1.12–1.57), compared to individuals with consistently low EDIH or EDIP throughout their life course. (Figure 6; Tables 8–9)

Joint Association of Lifetime Dietary Insulinemic and Inflammatory Potential and Risk of T2D

Individuals with higher lifetime averages for both EDIH and EDIP experienced a substantially higher lifetime risk of T2D (HR, 95%CI: 2.64, 2.32–3.01) compared with those with low lifetime averages for both factors (comparing the highest vs. lowest tertiles). (Table 10)

Table 10.

Hazard ratios (95% confidence intervals) for joint association of lifetime dietary insulinemic and inflammatory potential and risk of type 2 diabetes in the Nurses’ Health Study II (N = 40,135)

	Lifetime Empirical Dietary Index for Hyperinsulinemia (EDIH) and Empirical Dietary Inflammatory Pattern (EDIP) ^a

	Low EDIH – Low EDIP	Low EDIH – Intermediate EDIP	Low EDIH – High EDIP	Intermediate EDIH – Low EDIP	Intermediate EDIH – Intermediate EDIP	Intermediate EDIH – High EDIP	High EDIH – Low EDIP	High EDIH – Intermediate EDIP	High EDIH – High EDIP

No. of participants (40135 in total)	5849	3655	1388	3333	5462	4179	1661	3973	10635
No. of cases (3201 in total)	307	177	56	241	399	311	124	367	1219
Age-adjusted ^b	1 [Ref]	1.24 (1.03–1.49)	1.55 (1.17–2.07)	1.70 (1.43–2.01)	1.90 (1.64–2.21)	2.53 (2.15–2.96)	2.89 (2.34–3.57)	3.01 (2.59–3.51)	4.24 (3.74–4.81)
Adjusted for basic and adolescent characteristics ^c	1	1.24 (1.03–1.49)	1.46 (1.10–1.95)	1.63 (1.37–1.93)	1.81 (1.55–2.10)	2.22 (1.89–2.60)	2.64 (2.14–3.26)	2.70 (2.32–3.15)	3.45 (3.04–3.93)
Adjusted for basic, adolescent, and adulthood characteristics ^d	1	1.18 (0.98–1.42)	1.36 (1.02–1.81)	1.47 (1.24–1.74)	1.57 (1.35–1.83)	1.89 (1.61–2.22)	2.24 (1.81–2.77)	2.22 (1.90–2.60)	2.64 (2.32–3.01)
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e	1	1.11 (0.92–1.34)	1.15 (0.86–1.53)	1.17 (0.99–1.39)	1.23 (1.06–1.43)	1.44 (1.23–1.70)	1.55 (1.25–1.92)	1.52 (1.30–1.78)	1.58 (1.39–1.81)

Open in a new tab

‘Low’, ‘Intermediate’, and ‘High’ were defined based on tertiles.

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

Change in Dietary Insulinemic and Inflammatory Potential from Adolescence to Adulthood and Risk of T2D

From adolescence to adulthood, individuals who increased their EDIH or EDIP (compared to those who remained stable) had an elevated lifetime risk of T2D, generally regardless of whether their initial EDIH or EDIP status in adolescence was low (HR, 95%CI: 1.37, 1.15–1.62 and 1.27, 1.07–1.50, respectively), intermediate (1.46, 1.26–1.68 and 1.50, 1.30–1.74, respectively), or high (1.32, 1.11–1.57 and 1.18, 0.99–1.40). (Tables 11–12)

Table 11.

Hazard ratios (95% confidence intervals) for change in dietary insulinemic potential from adolescence to adulthood and risk of type 2 diabetes in the Nurses’ Health Study II (N = 40,135)

	Lifetime Empirical Dietary Index for Hyperinsulinemia (EDIH) ^a (adolescence - change from adolescence to adulthood)

	Low – Decrease	Low – Stable	Low – Increase	Intermediate – Decrease	Intermediate – Stable	Intermediate – Increase	High – Decrease	High – Stable	High – Increase

No. of participants (40135 in total)	1555	4506	7105	3563	5429	4356	8700	3442	1479
No. of cases (3201 in total)	62	205	502	204	357	464	821	368	218
Age-adjusted ^b	0.86 (0.65–1.15)	1 [Ref]	1.54 (1.30–1.82)	0.89 (0.74–1.05)	1 [Ref]	1.58 (1.37–1.82)	0.91 (0.80–1.03)	1 [Ref]	1.45 (1.22–1.72)
Adjusted for basic and adolescent characteristics ^c	0.82 (0.62–1.10)	1	1.54 (1.30–1.83)	0.86 (0.72–1.03)	1	1.57 (1.36–1.81)	0.92 (0.81–1.04)	1	1.39 (1.17–1.65)
Adjusted for basic, adolescent, and adulthood characteristics ^d	0.90 (0.67–1.21)	1	1.37 (1.15–1.62)	0.95 (0.80–1.13)	1	1.46 (1.26–1.68)	0.99 (0.87–1.13)	1	1.32 (1.11–1.57)
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e	0.84 (0.62–1.13)	1	1.10 (0.93–1.32)	1.00 (0.83–1.20)	1	1.17 (1.01–1.36)	1.10 (0.97–1.25)	1	1.14 (0.95–1.36)

Open in a new tab

‘Low’, ‘Intermediate’, and ‘High’ were defined based on tertiles.

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

Table 12.

Hazard ratios (95% confidence intervals) for change in dietary inflammatory potential from adolescence to adulthood and risk of type 2 diabetes in the Nurses’ Health Study II (N = 40,135)

	Lifetime Empirical Dietary Inflammatory Pattern (EDIP) ^a (adolescence – change from adolescence to adulthood)

	Low – Decrease	Low – Stable	Low – Increase	Intermediate – Decrease	Intermediate – Stable	Intermediate – Increase	High – Decrease	High – Stable	High – Increase

No. of participants (40135 in total)	1591	4390	7182	4201	5597	3536	8921	3314	1403
No. of cases (3201 in total)	78	200	498	225	365	409	831	382	213
Age-adjusted ^b	1.08 (0.83–1.41)	1 [Ref]	1.41 (1.20–1.67)	0.92 (0.78–1.09)	1 [Ref]	1.74 (1.50–2.01)	0.89 (0.78–1.00)	1 [Ref]	1.33 (1.12–1.58)
Adjusted for basic and adolescent characteristics ^c	0.97 (0.74–1.27)	1	1.38 (1.17–1.63)	0.89 (0.75–1.06)	1	1.65 (1.42–1.91)	0.88 (0.77–0.99)	1	1.27 (1.07–1.51)
Adjusted for basic, adolescent, and adulthood characteristics ^d	0.96 (0.73–1.26)	1	1.27 (1.07–1.50)	0.93 (0.78–1.10)	1	1.50 (1.30–1.74)	0.89 (0.79–1.01)	1	1.18 (0.99–1.40)
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e	0.87 (0.65–1.15)	1	1.14 (0.96–1.36)	0.99 (0.83–1.18)	1	1.28 (1.10–1.49)	0.91 (0.80–1.04)	1	1.00 (0.83–1.19)

Open in a new tab

‘Low’, ‘Intermediate’, and ‘High’ were defined based on tertiles.

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

Additional Analyses Adjusting for Weight Change and Waist Circumference

All observed associations were partially attenuated after further adjusting for weight change since age 18 and waist circumference in secondary analysis, suggesting the mediating role of metabolic factors in the observed relationships, as expected. (Tables 3–12)

Sensitivity Analyses

In sensitivity analysis of lifetime average dietary insulinemic and inflammatory potential with 4-year latency, all effect estimates remained robust, though attenuated slightly. (Table 13, available at https://github.com/drzhangyin/life_course_edih_edip_t2d/)

Interaction Analyses

A generally robust additive interaction was detected between BMI and EDIH and EDIP at different life stages (except for adolescence) on the lifetime risk of T2D. (Tables 14–17, available at https://github.com/drzhangyin/life_course_edih_edip_t2d/)

Estimated Cumulative Incidence of T2D in Individuals with Consistently High vs. Low Dietary Insulinemic and Inflammatory Potential

Comparing individuals with consistently high EDIH or EDIP throughout adolescence, premenopausal adulthood, and postmenopausal adulthood to those with consistently low EDIH or EDIP, their estimated multivariable-adjusted cumulative incidence of T2D by age 70 was: 17.3% vs. 7.1% and 16.4% vs. 9.2% in the total study population; and 26.8% vs. 15.9% and 25.6% vs. 17.9% among individuals with cumulative-averaged BMI ≥25 kg/m². (Figure 7)

Figure 7. — Estimated cumulative incidence of type 2 diabetes in individuals with consistently high vs. low dietary insulinemic and inflammatory potential throughout adolescence, premenopausal adulthood, and postmcnopausal adulthood in the Nurses’ Health Study II (N = 40,135)

Estimated Time Window for Adulthood Dietary Insulinemic and Inflammatory Potential and Risk of T2D

During adulthood, higher EDIH and EDIP were associated with an increased risk of T2D with a very short time lag. When using 10 to 12 years as the hard threshold for cutoffs, the associations of ‘recent’ period EDIH and EDIP with T2D risk were the strongest. (Figure 8)

Figure 8. — A. Hazard ratios and 95% confidence intervals of type 2 diabetes associated with dietary insulinemic potential in the remote and recent periods. X indicates the cutoff for the remote and recent periods (in years) in each of the analyses. Multivariable models were additionally mutually adjusted for dietary insulinemic potential in the remote/recent periods.

B. Hazard ratios and 95% confidence intervals of type 2 diabetes associated with dietary inflammatory potential in the remote and recent periods. X indicates the cutoff for the remote and recent periods (in years) in each of the analyses. Multivariable models were additionally mutually adjusted for dietary inflammatory potential in the remote/recent periods.

Discussion

No previous study has quantified the long-term excess risk of T2D associated with dietary insulinemic and inflammatory potential across various major life stages from a life course perspective. Existing population-based evidence is limited to adulthood exposures. Two large cohort studies in postmenopausal women or in both sexes observed an increased risk of T2D associated with higher adulthood EDIH and EDIP.^7,8 A recent meta-analysis reported a significant association between adulthood DII (a literature-derived dietary inflammatory index) and T2D risk in the analysis of high-quality investigations.¹¹ The EDIP outperforms the DII in predicting plasma inflammatory marker concentrations.¹⁷ The EDIP is based on foods while the DII is based primarily on nutrients.^16–18

This study presents the first epidemiological evidence from a life course perspective on this topic. After rigorously controlling for a wide range of potential confounders along with mutual adjustments for different life course periods when applicable for each dietary score, robust lifetime T2D risk elevations independently associated with higher EDIH and EDIP in premenopausal and postmenopausal adulthood, but not in adolescence, were detected. This suggests a predominant role of adulthood dietary insulinemic/inflammatory potential in driving the relationship with T2D. Additionally, this finding is reinforced by our joint association analysis and analysis of dietary changes, which revealed essentially similar effect estimates for individuals who were high in adulthood EDIH or EDIP, regardless of their adolescent EDIH or EDIP status.

Nonetheless, in individuals who had a low EDIH or EDIP during adulthood, those who experienced high levels during adolescence showed a modest increase in the risk of T2D, when compared to those who sustained low levels. This suggests the potential of residual risk from high levels of EDIH or EDIP during adolescence even if the adulthood EDIH or EDIP status has been improved. Although adulthood dietary insulinemic/inflammatory potential is of predominant importance, individuals with a higher EDIH or EDIP during adolescence were more likely to have a higher EDIH or EDIP in adulthood. Therefore, adolescent EDIH and EDIP may still contribute to the development of T2D through a carryover effect to adulthood. Possibly, in those with high EDIH or EDIP in adulthood, any effect of adolescent EDIH and EDIP is obscured, but it is observable in those with low adulthood EDIH and EDIP who are otherwise at low risk. A slightly increased risk of premenopausal T2D associated with higher adolescent EDIH and EDIP was observed.

Of note, in our analysis, the EDIH generally showed stronger associations with T2D risk than the EDIP, which is consistent with previous studies.^7,8 This difference is biologically plausible. The EDIH was empirically derived to capture the insulinemic potential of the overall diet, which is more directly related to pathways of chronic hyperinsulinemia and insulin resistance that are proximal to T2D development.^14,15 For example, food groups unique to EDIH either have strong postprandial insulin-stimulating effects (such as eggs, poultry, butter, margarine, cream soups, French fries, and low-fat dairy products) or possess well-established insulin-sensitizing properties (such as whole fruits and high-fat dairy products).^14,15 By contrast, the EDIP reflects the inflammatory potential of the diet based on circulating inflammatory markers, which, although important for cardiometabolic health, operates through broader pathways.^16,17

In analysis approximating the potential time window, an increased risk of T2D associated with higher EDIH or EDIP in adulthood was detected with a short time lag. This conveys the important message that it is never too late to reduce T2D risk by reducing dietary insulinemic and inflammatory potential. In addition, all the observed associations were substantially attenuated after additional adjustments for weight change since age 18 and waist circumference, supporting the important role of adiposity in mediating the mechanism. Of note, significant associations persisted even after accounting for these adiposity-related factors, suggesting that dietary insulinemic and inflammatory potential may influence T2D risk not only through pathways related to metabolic factors associated with energy balance but also through additional metabolic or inflammatory mechanisms.

Notable strengths of this study include the prospective study design, large sample size, long-term follow-up, high follow-up rates, repeated and extensively validated assessments of information on diet and a wide spectrum of covariates, and rigorous control for confounding. Especially, this study extends current knowledge on these topics from a life course perspective by disentangling the independent and joint influence of dietary insulinemic and inflammatory potential on T2D risk across three major life stages of women. Moreover, using menopause as a cutoff to separate women’s adulthood ensures tailoring to their physiological status. Lastly, the nature of cohort participants (all female healthcare professionals) further enhances internal validity.

This study has several limitations. First, the potential of residual and unmeasured confounding cannot be completely ruled out due to the observational nature. Second, information on adolescent diet was recalled by participants after their high school age. Some degree of measurement error is thus inevitable. Nevertheless, the assessment of adolescent diet in the NHSII cohort has been demonstrated to be reasonably valid and reproducible.^32,38,39 Third, blood samples (biomarkers) from participants during adolescence were not available, and therefore EDIP and EDIH previously derived and validated based on adulthood diet and biomarkers were utilized.^14–17 Potential difference in adolescent and adulthood dietary patterns might exist if there is heterogeneity in the influences of diet during these periods on insulinemic/inflammatory biomarker levels. Fourth, FFQ data are subject to measurement error and may be limited by the range of food items captured. Lastly, the EDIH and EDIP scores were originally developed within the NHS, raising the possibility of cohort-specific calibration. Although this may limit generalizability, these scores showed even stronger associations with T2D among Hispanics and African Americans than among European Americans in the more diverse Women’s Health Initiative cohort.⁸ Nevertheless, future investigations and consortium efforts across diverse populations are needed.

Conclusions

Over the life course in women, premenopausal and postmenopausal adulthood represent pivotal periods for implementing interventions on mitigating dietary insulinemic and inflammatory potential to reduce the risk of lifetime T2D, though adolescent diet may influence the risk of premenopausal T2D.

Supplementary Material

Supplementary Figure 3

NIHMS2206984-supplement-Supplementary_Figure_3.jpg^{(1.5MB, jpg)}

NIHMS2206984-supplement-2.pdf^{(9.2MB, pdf)}

Table 4.

Hazard ratios (95% confidence intervals) for adolescent dietary insulinemic and inflammatory potential and risk of type 2 diabetes in the Nurses’ Health Study II (N = 40,135)

	Adolescent Empirical Dietary Index for Hyperinsulinemia (EDIH)							Adolescent Empirical Dietary Inflammatory Pattern (EDIP)

	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a

No. of participants (40135 in total)	7900	7907	8014	8071	8243			7901	7903	8003	8079	8249
No. of cases (3201 in total)	468	477	610	702	944			465	491	587	705	953
Age-adjusted ^b	1 [Ref]	1.02 (0.90–1.16)	1.28 (1.14–1.45)	1.47 (1.31–1.65)	1.99 (1.78–2.22)	1.30 (1.25–1.34)	<.001	1 [Ref]	1.02 (0.89–1.15)	1.19 (1.06–1.35)	1.42 (1.27–1.60)	1.94 (1.74–2.17)	1.29 (1.24–1.33)	<.001
Adjusted for basic and adolescent characteristics ^c	1	1.00 (0.88–1.14)	1.19 (1.05–1.34)	1.30 (1.16–1.47)	1.66 (1.49–1.86)	1.22 (1.18–1.26)	<.001	1	1.00 (0.88–1.14)	1.14 (1.01–1.29)	1.29 (1.15–1.45)	1.57 (1.40–1.76)	1.18 (1.14–1.22)	<.001
Adjusted for basic, adolescent, and adulthood characteristics ^d	1	1.01 (0.89–1.15)	1.17 (1.03–1.32)	1.24 (1.10–1.39)	1.48 (1.32–1.66)	1.16 (1.12–1.20)	<.001	1	0.98 (0.86–1.12)	1.07 (0.95–1.22)	1.16 (1.03–1.31)	1.32 (1.17–1.48)	1.10 (1.07–1.14)	<.001
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e	1	0.96 (0.84–1.09)	1.03 (0.91–1.17)	1.04 (0.92–1.17)	1.16 (1.03–1.30)	1.07 (1.04–1.11)	<.001	1	0.97 (0.85–1.10)	1.01 (0.89–1.15)	1.01 (0.90–1.14)	1.08 (0.96–1.21)	1.04 (1.00–1.08)	.04
Adjusted for basic, adolescent, and adulthood characteristics ^d + premenopausal adulthood EDIH/P and postmenopausal adulthood EDIH/P ^f	1	0.89 (0.78–1.01)	0.96 (0.85–1.09)	0.96 (0.85–1.08)	1.07 (0.95–1.20)	1.06 (1.02–1.10)	.001	1	0.91 (0.80–1.03)	0.95 (0.84–1.08)	1.00 (0.89–1.13)	1.10 (0.98–1.24)	1.05 (1.01–1.09)	.01
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^e + premenopausal adulthood EDIH/P and postmenopausal adulthood EDIH/P ^f	1	0.90 (0.79–1.03)	0.94 (0.83–1.07)	0.93 (0.82–1.05)	1.01 (0.89–1.13)	1.04 (1.00–1.07)	.06	1	0.93 (0.82–1.06)	0.95 (0.84–1.08)	0.94 (0.83–1.06)	0.99 (0.88–1.11)	1.01 (0.98–1.05)	.50

Open in a new tab

P-value for trend was calculated using dietary insulinemic/inflammatory potential score modeled as a continuous variable.

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

Additionally adjusted for premenopausal adulthood EDIH/P and postmenopausal adulthood EDIH/P.

Table 5.

Hazard ratios (95% confidence intervals) for premenopausal adulthood dietary insulinemic and inflammatory potential and risk of type 2 diabetes in the Nurses’ Health Study II (N = 37,528)

	Premenopausal adulthood Empirical Dietary Index for Hyperinsulinemia (EDIH)							Premenopausal adulthood Empirical Dietary Inflammatory Pattern (EDIP)

	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a	Quintile 1	Quintile 2	Quintile 3	Quintile 4	Quintile 5	Per 1-standard deviation	P-trend ^a

No. of participants (37528 in total)	6369	6605	7187	7963	9404			7804	6912	6989	7476	8347
No. of cases (1317 in total) ^b	95	152	232	312	526			134	167	219	320	477
Age-adjusted ^c	1 [Ref]	1.54 (1.19–2.00)	2.40 (1.89–3.05)	3.27 (2.60–4.13)	5.58 (4.48–6.96)	1.71 (1.63–1.79)	<.001	1 [Ref]	1.19 (0.94–1.49)	1.51 (1.22–1.88)	2.23 (1.82–2.74)	3.36 (2.77–4.08)	1.54 (1.46–1.62)	<.001
Adjusted for basic and adolescent characteristics ^d	1	1.48 (1.14–1.91)	2.23 (1.75–2.83)	2.94 (2.33–3.70)	4.51 (3.61–5.64)	1.58 (1.51–1.66)	<.001	1	1.24 (0.99–1.56)	1.54 (1.24–1.92)	2.15 (1.75–2.64)	2.92 (2.40–3.55)	1.42 (1.35–1.50)	<.001
Adjusted for basic, adolescent, and adulthood characteristics ^e	1	1.33 (1.02–1.72)	1.88 (1.48–2.40)	2.38 (1.88–3.02)	3.41 (2.71–4.28)	1.46 (1.39–1.54)	<.001	1	1.19 (0.95–1.50)	1.42 (1.14–1.77)	1.89 (1.54–2.32)	2.39 (1.96–2.91)	1.33 (1.26–1.40)	<.001
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^f	1	1.12 (0.86–1.46)	1.47 (1.15–1.88)	1.66 (1.31–2.10)	1.81 (1.44–2.29)	1.19 (1.13–1.26)	<.001	1	1.19 (0.94–1.50)	1.37 (1.10–1.71)	1.59 (1.29–1.96)	1.72 (1.40–2.10)	1.17 (1.11–1.23)	<.001
Adjusted for basic, adolescent, and adulthood characteristics ^e + adolescent EDIH/P ^g (main model)	1	1.28 (0.99–1.66)	1.80 (1.41–2.30)	2.25 (1.77–2.85)	3.18 (2.52–4.01)	1.44 (1.36–1.52)	<.001	1	1.17 (0.93–1.47)	1.38 (1.11–1.72)	1.83 (1.48–2.25)	2.31 (1.88–2.82)	1.32 (1.25–1.40)	<.001
Adjusted for basic, adolescent, and adulthood characteristics + weight change since age 18 + waist circumference ^f + adolescent EDIH/P ^g	1	1.10 (0.85–1.43)	1.44 (1.13–1.85)	1.61 (1.27–2.05)	1.75 (1.38–2.22)	1.18 (1.12–1.26)	<.001	1	1.18 (0.94–1.50)	1.36 (1.09–1.70)	1.57 (1.27–1.94)	1.70 (1.38–2.09)	1.17 (1.10–1.23)	<.001

Open in a new tab

P-value for trend was calculated using dietary insulinemic/inflammatory potential score modeled as a continuous variable

Participants who remained in their premenopausal adulthood and did not enter postmenopausal adulthood during the follow-up were included in this analysis.

Stratified by age and follow-up cycle.

Stratified by age and follow-up cycle; adjusted for above-mentioned basic, adolescent and adulthood characteristics; and additionally adjusted for weight change since age 18 and waist circumference.

Additionally adjusted for adolescent EDIH/P.

Practice Implications.

What is the current knowledge on this topic?

Dietary patterns associated with markers of hyperinsulinemia and inflammation have been reported to have a considerably stronger association with type 2 diabetes (T2D) than a spectrum of other major dietary patterns. The timing of exposure to diet across the lifespan may be critical in the development of T2D. However, no previous study has deciphered the influence of dietary insulinemic and inflammatory potential on the risk of T2D across the lifespan from a life course perspective.

How does this research add to knowledge on this topic?

Our study provides the first epidemiological evidence on dietary insulinemic and inflammatory potential and T2D risk from a life course perspective. Leveraging data from 40,135 eligible, female participants in the Nurses’ Health Study II with decades of follow-up, we comprehensively investigated the associations of dietary insulinemic and inflammatory potential across different major life stages (adolescence, premenopausal adulthood, and postmenopausal adulthood) as well as changes and cumulatively over the lifetime, with the risk of T2D. Our study addresses the following unanswered questions: What are the associations of dietary insulinemic and inflammatory potential with T2D risk across major life stages? What are their joint associations across major life stages? What are the associations between changes in dietary insulinemic and inflammatory potential from adolescence to adulthood and the risk of T2D? Does lowering dietary insulinemic and inflammatory potential have immediate benefits in reducing the risk of T2D in adulthood, or is there a time-lag?

How might this knowledge impact current dietetics practice?

Over the life course in women, high dietary insulinemic and inflammatory potential in both premenopausal and postmenopausal adulthood, but not in adolescence, are independently associated with a substantially increased lifetime risk of T2D. Adulthood offers the most critical time window for dietary interventions to reduce lifetime T2D risk, though adolescent diet may influence the risk of premenopausal T2D. Higher dietary insulinemic and inflammatory potential during adulthood are associated with an increased risk of T2D with a very short time lag.

Research Snapshot.

Research Question:

What are the associations of dietary insulinemic and inflammatory potential with type 2 diabetes (T2D) risk across major life stages?

Key Findings:

Over the life course in women, high dietary insulinemic and inflammatory potential during premenopausal and postmenopausal adulthood, but not during adolescence, were independently associated with a substantially increased lifetime risk of T2D. Adolescent dietary insulinemic and inflammatory potential were associated with a slightly increased risk of premenopausal T2D. Higher adulthood dietary insulinemic and inflammatory potential were associated with an increased risk of T2D with a very short time lag.

Acknowledgments

The authors thank all participants and staff of the Nurses’ Health Study II for their contributions to this research. The authors assume full responsibility for analyses and interpretation of these data.

Funding

The Nurses’ Health Study II was supported by NIH grant U01 CA176726 to WCW and AHE. YZ was supported by NIH training-grant T32 DK 007703, Irene M. & Fredrick J. Stare Nutrition Education Fund Doctoral Scholarship, and Mayer Fund Doctoral Scholarship. ELG was supported as an American Cancer Society Clinical Research Professor (CRP-23–1014041). The funding sources played no role in the study design, data collection, data analysis, and interpretation of results, or the decisions made in preparation and submission of the article.

Footnotes

Conflict of Interest Disclosures

The authors declare no potential conflicts of interest.

Publisher's Disclaimer: This is a PDF of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability. This version will undergo additional copyediting, typesetting and review before it is published in its final form. As such, this version is no longer the Accepted Manuscript, but it is not yet the definitive Version of Record; we are providing this early version to give early visibility of the article. Please note that Elsevier’s sharing policy for the Published Journal Article applies to this version, see: https://www.elsevier.com/about/policies-and-standards/sharing#4-published-journal-article. Please also note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

O’Hearn M, Lara-Castor L, Cudhea F, et al. Incident type 2 diabetes attributable to suboptimal diet in 184 countries. Nat Med. Apr 2023;29(4):982–995. doi: 10.1038/s41591-023-02278-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
American Diabetes Association Professional Practice Committee for Diabetes. 3. Prevention or Delay of Diabetes and Associated Comorbidities: Standards of Care in Diabetes-2026. Diabetes care. Jan 1 2026;49(Supplement_1):S50–S60. doi: 10.2337/dc26-S003 [DOI] [PMC free article] [PubMed] [Google Scholar]
Chatterjee S, Khunti K, Davies MJ. Type 2 diabetes. Lancet. Jun 3 2017;389(10085):2239–2251. doi: 10.1016/S0140-6736(17)30058-2 [DOI] [PubMed] [Google Scholar]
Neuenschwander M, Ballon A, Weber KS, et al. Role of diet in type 2 diabetes incidence: umbrella review of meta-analyses of prospective observational studies. BMJ. Jul 3 2019;366:l2368. doi: 10.1136/bmj.l2368 [DOI] [PMC free article] [PubMed] [Google Scholar]
Wang P, Song M, Eliassen AH, et al. Optimal dietary patterns for prevention of chronic disease. Nat Med. Mar 2023;29(3):719–728. doi: 10.1038/s41591-023-02235-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
Jannasch F, Kroger J, Schulze MB. Dietary Patterns and Type 2 Diabetes: A Systematic Literature Review and Meta-Analysis of Prospective Studies. J Nutr. Jun 2017;147(6):1174–1182. doi: 10.3945/jn.116.242552 [DOI] [PubMed] [Google Scholar]
Lee DH, Li J, Li Y, et al. Dietary Inflammatory and Insulinemic Potential and Risk of Type 2 Diabetes: Results From Three Prospective U.S. Cohort Studies. Diabetes care. Nov 2020;43(11):2675–2683. doi: 10.2337/dc20-0815 [DOI] [PMC free article] [PubMed] [Google Scholar]
Jin Q, Shi N, Aroke D, et al. Insulinemic and Inflammatory Dietary Patterns Show Enhanced Predictive Potential for Type 2 Diabetes Risk in Postmenopausal Women. Diabetes care. Mar 2021;44(3):707–714. doi: 10.2337/dc20-2216 [DOI] [PMC free article] [PubMed] [Google Scholar]
Farhadnejad H, Mokhtari E, Teymoori F, et al. Association of the insulinemic potential of diet and lifestyle with risk of diabetes incident in Tehranian adults: a population based cohort study. Nutr J. Apr 23 2021;20(1):39. doi: 10.1186/s12937-021-00697-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
Omrani M, Hosseinzadeh M, Shab Bidar S, et al. Insulinaemic potential of diet and lifestyle and risk of type 2 diabetes in the Iranian adults: result from Yazd health study. BMC Endocr Disord. Jul 3 2023;23(1):136. doi: 10.1186/s12902-023-01364-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
Motamedi A, Askari M, Mozaffari H, et al. Dietary Inflammatory Index in relation to Type 2 Diabetes: A Meta-Analysis. Int J Clin Pract. 2022;2022:9953115. doi: 10.1155/2022/9953115 [DOI] [PMC free article] [PubMed] [Google Scholar]
Tan QQ, Du XY, Gao CL, Xu Y. Higher Dietary Inflammatory Index Scores Increase the Risk of Diabetes Mellitus: A Meta-Analysis and Systematic Review. Front Endocrinol (Lausanne). 2021;12:693144. doi: 10.3389/fendo.2021.693144 [DOI] [PMC free article] [PubMed] [Google Scholar]
Laouali N, Mancini FR, Hajji-Louati M, et al. Dietary inflammatory index and type 2 diabetes risk in a prospective cohort of 70,991 women followed for 20 years: the mediating role of BMI. Diabetologia. Dec 2019;62(12):2222–2232. doi: 10.1007/s00125-019-04972-0 [DOI] [PubMed] [Google Scholar]
Tabung FK, Wang W, Fung TT, et al. Development and validation of empirical indices to assess the insulinaemic potential of diet and lifestyle. Br J Nutr. Nov 8 2016:1–12. doi: 10.1017/S0007114516003755 [DOI] [PMC free article] [PubMed] [Google Scholar]
Tabung FK, Nimptsch K, Giovannucci EL. Postprandial Duration Influences the Association of Insulin-Related Dietary Indexes and Plasma C-peptide Concentrations in Adult Men and Women. J Nutr. Feb 1 2019;149(2):286–294. doi: 10.1093/jn/nxy239 [DOI] [PMC free article] [PubMed] [Google Scholar]
Tabung FK, Smith-Warner SA, Chavarro JE, et al. Development and Validation of an Empirical Dietary Inflammatory Index. J Nutr. Aug 2016;146(8):1560–70. doi: 10.3945/jn.115.228718 [DOI] [PMC free article] [PubMed] [Google Scholar]
Tabung FK, Smith-Warner SA, Chavarro JE, et al. An Empirical Dietary Inflammatory Pattern Score Enhances Prediction of Circulating Inflammatory Biomarkers in Adults. J Nutr. Aug 2017;147(8):1567–1577. doi: 10.3945/jn.117.248377 [DOI] [PMC free article] [PubMed] [Google Scholar]
Shivappa N, Steck SE, Hurley TG, Hussey JR, Hebert JR. Designing and developing a literature-derived, population-based dietary inflammatory index. Public Health Nutr. Aug 2014;17(8):1689–96. doi: 10.1017/S1368980013002115 [DOI] [PMC free article] [PubMed] [Google Scholar]
Wagner C, Carmeli C, Jackisch J, et al. Life course epidemiology and public health. Lancet Public Health. Apr 2024;9(4):e261–e269. doi: 10.1016/S2468-2667(24)00018-5 [DOI] [PubMed] [Google Scholar]
Pedamallu H, Zmora R, Perak AM, Allen NB. Life Course Cardiovascular Health: Risk Factors, Outcomes, and Interventions. Circ Res. Jun 9 2023;132(12):1570–1583. doi: 10.1161/CIRCRESAHA.123.321998 [DOI] [PubMed] [Google Scholar]
Wright RJ, Hanson HA. A tipping point in cancer epidemiology: embracing a life course exposomic framework. Trends Cancer. Apr 2022;8(4):280–282. doi: 10.1016/j.trecan.2022.01.016 [DOI] [PubMed] [Google Scholar]
Marshall SM. A life course perspective on diabetes: developmental origins and beyond. Diabetologia. Oct 2019;62(10):1737–1739. doi: 10.1007/s00125-019-4954-6 [DOI] [PubMed] [Google Scholar]
Viner R, White B, Christie D. Type 2 diabetes in adolescents: a severe phenotype posing major clinical challenges and public health burden. Lancet. Jun 3 2017;389(10085):2252–2260. doi: 10.1016/S0140-6736(17)31371-5 [DOI] [PubMed] [Google Scholar]
Magliano DJ, Sacre JW, Harding JL, Gregg EW, Zimmet PZ, Shaw JE. Young-onset type 2 diabetes mellitus - implications for morbidity and mortality. Nat Rev Endocrinol. Jun 2020;16(6):321–331. doi: 10.1038/s41574-020-0334-z [DOI] [PubMed] [Google Scholar]
Arslanian S, Bacha F, Grey M, Marcus MD, White NH, Zeitler P. Evaluation and Management of Youth-Onset Type 2 Diabetes: A Position Statement by the American Diabetes Association. Diabetes care. Dec 2018;41(12):2648–2668. doi: 10.2337/dci18-0052 [DOI] [PMC free article] [PubMed] [Google Scholar]
Lascar N, Brown J, Pattison H, Barnett AH, Bailey CJ, Bellary S. Type 2 diabetes in adolescents and young adults. Lancet Diabetes Endocrinol. Jan 2018;6(1):69–80. doi: 10.1016/S2213-8587(17)30186-9 [DOI] [PubMed] [Google Scholar]
Colditz GA, Hankinson SE. The Nurses’ Health Study: lifestyle and health among women. Nat Rev Cancer. May 2005;5(5):388–96. doi: 10.1038/nrc1608 [DOI] [PubMed] [Google Scholar]
Farvid MS, Eliassen AH, Cho E, Chen WY, Willett WC. Adolescent and Early Adulthood Dietary Carbohydrate Quantity and Quality in Relation to Breast Cancer Risk. Cancer Epidemiol Biomarkers Prev. Jul 2015;24(7):1111–20. doi: 10.1158/1055-9965.EPI-14-1401 [DOI] [PMC free article] [PubMed] [Google Scholar]
Farvid MS, Cho E, Chen WY, Eliassen AH, Willett WC. Adolescent meat intake and breast cancer risk. Int J Cancer. Apr 15 2015;136(8):1909–20. doi: 10.1002/ijc.29218 [DOI] [PMC free article] [PubMed] [Google Scholar]
Farvid MS, Chen WY, Michels KB, Cho E, Willett WC, Eliassen AH. Fruit and vegetable consumption in adolescence and early adulthood and risk of breast cancer: population based cohort study. BMJ. May 11 2016;353:i2343. doi: 10.1136/bmj.i2343 [DOI] [PMC free article] [PubMed] [Google Scholar]
Farvid MS, Eliassen AH, Cho E, Chen WY, Willett WC. Dairy Consumption in Adolescence and Early Adulthood and Risk of Breast Cancer. Cancer Epidemiol Biomarkers Prev. May 2018;27(5):575–584. doi: 10.1158/1055-9965.EPI-17-0345 [DOI] [PMC free article] [PubMed] [Google Scholar]
Hu FB, Satija A, Rimm EB, et al. Diet Assessment Methods in the Nurses’ Health Studies and Contribution to Evidence-Based Nutritional Policies and Guidelines. Am J Public Health. Sep 2016;106(9):1567–72. doi: 10.2105/AJPH.2016.303348 [DOI] [PMC free article] [PubMed] [Google Scholar]
Food Data Central, U.S. Department of Agriculture. U.S. Department of Agriculture Nutrient Database. FoodData Central, U.S. Department of Agriculture. Accessed August 11, 2026. https://fdc.nal.usda.gov/ [Google Scholar]
Nutrition Questionnaire Service Center, Harvard TH. Chan School of Public Health. Nutrient Tables. Accessed August 11, 2026. https://hsph.harvard.edu/department/nutrition/nutrition-questionnaire-service-center/ [Google Scholar]
Salvini S, Hunter DJ, Sampson L, et al. Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption. Int J Epidemiol. Dec 1989;18(4):858–67. doi: 10.1093/ije/18.4.858 [DOI] [PubMed] [Google Scholar]
Rimm EB, Giovannucci EL, Stampfer MJ, Colditz GA, Litin LB, Willett WC. Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals. Am J Epidemiol. May 15 1992;135(10):1114–26; discussion 1127–36. [DOI] [PubMed] [Google Scholar]
Yuan C, Spiegelman D, Rimm EB, et al. Relative Validity of Nutrient Intakes Assessed by Questionnaire, 24-Hour Recalls, and Diet Records as Compared With Urinary Recovery and Plasma Concentration Biomarkers: Findings for Women. Am J Epidemiol. May 1 2018;187(5):1051–1063. doi: 10.1093/aje/kwx328 [DOI] [PMC free article] [PubMed] [Google Scholar]
Maruti SS, Feskanich D, Rockett HR, Colditz GA, Sampson LA, Willett WC. Validation of adolescent diet recalled by adults. Epidemiology. Mar 2006;17(2):226–9. doi: 10.1097/01.ede.0000198181.86685.49 [DOI] [PubMed] [Google Scholar]
Maruti SS, Feskanich D, Colditz GA, et al. Adult recall of adolescent diet: reproducibility and comparison with maternal reporting. Am J Epidemiol. Jan 1 2005;161(1):89–97. doi: 10.1093/aje/kwi019 [DOI] [PMC free article] [PubMed] [Google Scholar]
Linos E, Willett WC, Cho E, Colditz G, Frazier LA. Red meat consumption during adolescence among premenopausal women and risk of breast cancer. Cancer Epidemiol Biomarkers Prev. Aug 2008;17(8):2146–51. doi: 10.1158/1055-9965.EPI-08-0037 [DOI] [PMC free article] [PubMed] [Google Scholar]
Riseberg E, Wu Y, Lam WC, et al. Lifetime dairy product consumption and breast cancer risk: a prospective cohort study by tumor subtypes. Am J Clin Nutr. Nov 30 2023;doi: 10.1016/j.ajcnut.2023.11.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
Sotos-Prieto M, Bhupathiraju SN, Mattei J, et al. Changes in Diet Quality Scores and Risk of Cardiovascular Disease Among US Men and Women. Circulation. Dec 8 2015;132(23):2212–9. doi: 10.1161/CIRCULATIONAHA.115.017158 [DOI] [PMC free article] [PubMed] [Google Scholar]
Willett W Nutritional epidemiology. 2nd ed. Monographs in epidemiology and biostatistics. Oxford University Press; 1998:xiv, 514 p. [Google Scholar]
Hu FB, Stampfer MJ, Rimm E, et al. Dietary fat and coronary heart disease: a comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements. Research Support, U.S. Gov’t, P.H.S. Am J Epidemiol. Mar 15 1999;149(6):531–40. [DOI] [PubMed] [Google Scholar]
Bernstein AM, Rosner BA, Willett WC. Cereal fiber and coronary heart disease: a comparison of modeling approaches for repeated dietary measurements, intermediate outcomes, and long follow-up. European journal of epidemiology. Nov 2011;26(11):877–86. doi: 10.1007/s10654-011-9626-x [DOI] [PubMed] [Google Scholar]
Hu Y, Zhang X, Ma Y, et al. Incident Type 2 Diabetes Duration and Cancer Risk: A Prospective Study in Two US Cohorts. J Natl Cancer Inst. Apr 6 2021;113(4):381–389. doi: 10.1093/jnci/djaa141 [DOI] [PMC free article] [PubMed] [Google Scholar]
National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes. Dec 1979;28(12):1039–57. doi: 10.2337/diab.28.12.1039 [DOI] [PubMed] [Google Scholar]
Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes care. Jul 1997;20(7):1183–97. doi: 10.2337/diacare.20.7.1183 [DOI] [PubMed] [Google Scholar]
American Diabetes Association. Standards of medical care in diabetes--2010. Diabetes care. Jan 2010;33 Suppl 1:S11–61. doi: 10.2337/dc10-S011 [DOI] [PMC free article] [PubMed] [Google Scholar]
Manson JE, Rimm EB, Stampfer MJ, et al. Physical activity and incidence of non-insulin-dependent diabetes mellitus in women. Lancet. Sep 28 1991;338(8770):774–8. doi: 10.1016/0140-6736(91)90664-b [DOI] [PubMed] [Google Scholar]
DeVille NV, Iyer HS, Holland I, et al. Neighborhood socioeconomic status and mortality in the nurses’ health study (NHS) and the nurses’ health study II (NHSII). Environ Epidemiol. Feb 2023;7(1):e235. doi: 10.1097/EE9.0000000000000235 [DOI] [PMC free article] [PubMed] [Google Scholar]
Hunter DJ, Manson JE, Colditz GA, et al. Reproducibility of oral contraceptive histories and validity of hormone composition reported in a cohort of US women. Contraception. Dec 1997;56(6):373–8. doi: 10.1016/s0010-7824(97)00172-8 [DOI] [PubMed] [Google Scholar]
Troy LM, Hunter DJ, Manson JE, Colditz GA, Stampfer MJ, Willett WC. The validity of recalled weight among younger women. Int J Obes Relat Metab Disord. Aug 1995;19(8):570–2. [PubMed] [Google Scholar]
Wolf AM, Hunter DJ, Colditz GA, et al. Reproducibility and validity of a self-administered physical activity questionnaire. Int J Epidemiol. Oct 1994;23(5):991–9. doi: 10.1093/ije/23.5.991 [DOI] [PubMed] [Google Scholar]
Colditz GA, Stampfer MJ, Willett WC, et al. Reproducibility and validity of self-reported menopausal status in a prospective cohort study. Am J Epidemiol. Aug 1987;126(2):319–25. doi: 10.1093/aje/126.2.319 [DOI] [PubMed] [Google Scholar]
Rimm EB, Stampfer MJ, Colditz GA, Chute CG, Litin LB, Willett WC. Validity of self-reported waist and hip circumferences in men and women. Epidemiology. Nov 1990;1(6):466–73. doi: 10.1097/00001648-199011000-00009 [DOI] [PubMed] [Google Scholar]
Hosmer DW, Lemeshow S. Confidence interval estimation of interaction. Epidemiology. Sep 1992;3(5):452–6. doi: 10.1097/00001648-199209000-00012 [DOI] [PubMed] [Google Scholar]
Zhang Y, Chan AT, Meyerhardt JA, Giovannucci EL. Timing of Aspirin Use in Colorectal Cancer Chemoprevention: A Prospective Cohort Study. J Natl Cancer Inst. Jul 1 2021;113(7):841–851. doi: 10.1093/jnci/djab009 [DOI] [PMC free article] [PubMed] [Google Scholar]
SAS statistical software, version 9.4 for UNIX. Version version 9.4 for UNIX. SAS Institute Inc [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Figure 3

NIHMS2206984-supplement-Supplementary_Figure_3.jpg^{(1.5MB, jpg)}

NIHMS2206984-supplement-2.pdf^{(9.2MB, pdf)}

[R1] O’Hearn M, Lara-Castor L, Cudhea F, et al. Incident type 2 diabetes attributable to suboptimal diet in 184 countries. Nat Med. Apr 2023;29(4):982–995. doi: 10.1038/s41591-023-02278-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R2] American Diabetes Association Professional Practice Committee for Diabetes. 3. Prevention or Delay of Diabetes and Associated Comorbidities: Standards of Care in Diabetes-2026. Diabetes care. Jan 1 2026;49(Supplement_1):S50–S60. doi: 10.2337/dc26-S003 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R3] Chatterjee S, Khunti K, Davies MJ. Type 2 diabetes. Lancet. Jun 3 2017;389(10085):2239–2251. doi: 10.1016/S0140-6736(17)30058-2 [DOI] [PubMed] [Google Scholar]

[R4] Neuenschwander M, Ballon A, Weber KS, et al. Role of diet in type 2 diabetes incidence: umbrella review of meta-analyses of prospective observational studies. BMJ. Jul 3 2019;366:l2368. doi: 10.1136/bmj.l2368 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] Wang P, Song M, Eliassen AH, et al. Optimal dietary patterns for prevention of chronic disease. Nat Med. Mar 2023;29(3):719–728. doi: 10.1038/s41591-023-02235-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] Jannasch F, Kroger J, Schulze MB. Dietary Patterns and Type 2 Diabetes: A Systematic Literature Review and Meta-Analysis of Prospective Studies. J Nutr. Jun 2017;147(6):1174–1182. doi: 10.3945/jn.116.242552 [DOI] [PubMed] [Google Scholar]

[R7] Lee DH, Li J, Li Y, et al. Dietary Inflammatory and Insulinemic Potential and Risk of Type 2 Diabetes: Results From Three Prospective U.S. Cohort Studies. Diabetes care. Nov 2020;43(11):2675–2683. doi: 10.2337/dc20-0815 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] Jin Q, Shi N, Aroke D, et al. Insulinemic and Inflammatory Dietary Patterns Show Enhanced Predictive Potential for Type 2 Diabetes Risk in Postmenopausal Women. Diabetes care. Mar 2021;44(3):707–714. doi: 10.2337/dc20-2216 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R9] Farhadnejad H, Mokhtari E, Teymoori F, et al. Association of the insulinemic potential of diet and lifestyle with risk of diabetes incident in Tehranian adults: a population based cohort study. Nutr J. Apr 23 2021;20(1):39. doi: 10.1186/s12937-021-00697-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] Omrani M, Hosseinzadeh M, Shab Bidar S, et al. Insulinaemic potential of diet and lifestyle and risk of type 2 diabetes in the Iranian adults: result from Yazd health study. BMC Endocr Disord. Jul 3 2023;23(1):136. doi: 10.1186/s12902-023-01364-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R11] Motamedi A, Askari M, Mozaffari H, et al. Dietary Inflammatory Index in relation to Type 2 Diabetes: A Meta-Analysis. Int J Clin Pract. 2022;2022:9953115. doi: 10.1155/2022/9953115 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R12] Tan QQ, Du XY, Gao CL, Xu Y. Higher Dietary Inflammatory Index Scores Increase the Risk of Diabetes Mellitus: A Meta-Analysis and Systematic Review. Front Endocrinol (Lausanne). 2021;12:693144. doi: 10.3389/fendo.2021.693144 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] Laouali N, Mancini FR, Hajji-Louati M, et al. Dietary inflammatory index and type 2 diabetes risk in a prospective cohort of 70,991 women followed for 20 years: the mediating role of BMI. Diabetologia. Dec 2019;62(12):2222–2232. doi: 10.1007/s00125-019-04972-0 [DOI] [PubMed] [Google Scholar]

[R14] Tabung FK, Wang W, Fung TT, et al. Development and validation of empirical indices to assess the insulinaemic potential of diet and lifestyle. Br J Nutr. Nov 8 2016:1–12. doi: 10.1017/S0007114516003755 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] Tabung FK, Nimptsch K, Giovannucci EL. Postprandial Duration Influences the Association of Insulin-Related Dietary Indexes and Plasma C-peptide Concentrations in Adult Men and Women. J Nutr. Feb 1 2019;149(2):286–294. doi: 10.1093/jn/nxy239 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] Tabung FK, Smith-Warner SA, Chavarro JE, et al. Development and Validation of an Empirical Dietary Inflammatory Index. J Nutr. Aug 2016;146(8):1560–70. doi: 10.3945/jn.115.228718 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] Tabung FK, Smith-Warner SA, Chavarro JE, et al. An Empirical Dietary Inflammatory Pattern Score Enhances Prediction of Circulating Inflammatory Biomarkers in Adults. J Nutr. Aug 2017;147(8):1567–1577. doi: 10.3945/jn.117.248377 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] Shivappa N, Steck SE, Hurley TG, Hussey JR, Hebert JR. Designing and developing a literature-derived, population-based dietary inflammatory index. Public Health Nutr. Aug 2014;17(8):1689–96. doi: 10.1017/S1368980013002115 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] Wagner C, Carmeli C, Jackisch J, et al. Life course epidemiology and public health. Lancet Public Health. Apr 2024;9(4):e261–e269. doi: 10.1016/S2468-2667(24)00018-5 [DOI] [PubMed] [Google Scholar]

[R20] Pedamallu H, Zmora R, Perak AM, Allen NB. Life Course Cardiovascular Health: Risk Factors, Outcomes, and Interventions. Circ Res. Jun 9 2023;132(12):1570–1583. doi: 10.1161/CIRCRESAHA.123.321998 [DOI] [PubMed] [Google Scholar]

[R21] Wright RJ, Hanson HA. A tipping point in cancer epidemiology: embracing a life course exposomic framework. Trends Cancer. Apr 2022;8(4):280–282. doi: 10.1016/j.trecan.2022.01.016 [DOI] [PubMed] [Google Scholar]

[R22] Marshall SM. A life course perspective on diabetes: developmental origins and beyond. Diabetologia. Oct 2019;62(10):1737–1739. doi: 10.1007/s00125-019-4954-6 [DOI] [PubMed] [Google Scholar]

[R23] Viner R, White B, Christie D. Type 2 diabetes in adolescents: a severe phenotype posing major clinical challenges and public health burden. Lancet. Jun 3 2017;389(10085):2252–2260. doi: 10.1016/S0140-6736(17)31371-5 [DOI] [PubMed] [Google Scholar]

[R24] Magliano DJ, Sacre JW, Harding JL, Gregg EW, Zimmet PZ, Shaw JE. Young-onset type 2 diabetes mellitus - implications for morbidity and mortality. Nat Rev Endocrinol. Jun 2020;16(6):321–331. doi: 10.1038/s41574-020-0334-z [DOI] [PubMed] [Google Scholar]

[R25] Arslanian S, Bacha F, Grey M, Marcus MD, White NH, Zeitler P. Evaluation and Management of Youth-Onset Type 2 Diabetes: A Position Statement by the American Diabetes Association. Diabetes care. Dec 2018;41(12):2648–2668. doi: 10.2337/dci18-0052 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R26] Lascar N, Brown J, Pattison H, Barnett AH, Bailey CJ, Bellary S. Type 2 diabetes in adolescents and young adults. Lancet Diabetes Endocrinol. Jan 2018;6(1):69–80. doi: 10.1016/S2213-8587(17)30186-9 [DOI] [PubMed] [Google Scholar]

[R27] Colditz GA, Hankinson SE. The Nurses’ Health Study: lifestyle and health among women. Nat Rev Cancer. May 2005;5(5):388–96. doi: 10.1038/nrc1608 [DOI] [PubMed] [Google Scholar]

[R28] Farvid MS, Eliassen AH, Cho E, Chen WY, Willett WC. Adolescent and Early Adulthood Dietary Carbohydrate Quantity and Quality in Relation to Breast Cancer Risk. Cancer Epidemiol Biomarkers Prev. Jul 2015;24(7):1111–20. doi: 10.1158/1055-9965.EPI-14-1401 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R29] Farvid MS, Cho E, Chen WY, Eliassen AH, Willett WC. Adolescent meat intake and breast cancer risk. Int J Cancer. Apr 15 2015;136(8):1909–20. doi: 10.1002/ijc.29218 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] Farvid MS, Chen WY, Michels KB, Cho E, Willett WC, Eliassen AH. Fruit and vegetable consumption in adolescence and early adulthood and risk of breast cancer: population based cohort study. BMJ. May 11 2016;353:i2343. doi: 10.1136/bmj.i2343 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] Farvid MS, Eliassen AH, Cho E, Chen WY, Willett WC. Dairy Consumption in Adolescence and Early Adulthood and Risk of Breast Cancer. Cancer Epidemiol Biomarkers Prev. May 2018;27(5):575–584. doi: 10.1158/1055-9965.EPI-17-0345 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] Hu FB, Satija A, Rimm EB, et al. Diet Assessment Methods in the Nurses’ Health Studies and Contribution to Evidence-Based Nutritional Policies and Guidelines. Am J Public Health. Sep 2016;106(9):1567–72. doi: 10.2105/AJPH.2016.303348 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] Food Data Central, U.S. Department of Agriculture. U.S. Department of Agriculture Nutrient Database. FoodData Central, U.S. Department of Agriculture. Accessed August 11, 2026. https://fdc.nal.usda.gov/ [Google Scholar]

[R34] Nutrition Questionnaire Service Center, Harvard TH. Chan School of Public Health. Nutrient Tables. Accessed August 11, 2026. https://hsph.harvard.edu/department/nutrition/nutrition-questionnaire-service-center/ [Google Scholar]

[R35] Salvini S, Hunter DJ, Sampson L, et al. Food-based validation of a dietary questionnaire: the effects of week-to-week variation in food consumption. Int J Epidemiol. Dec 1989;18(4):858–67. doi: 10.1093/ije/18.4.858 [DOI] [PubMed] [Google Scholar]

[R36] Rimm EB, Giovannucci EL, Stampfer MJ, Colditz GA, Litin LB, Willett WC. Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals. Am J Epidemiol. May 15 1992;135(10):1114–26; discussion 1127–36. [DOI] [PubMed] [Google Scholar]

[R37] Yuan C, Spiegelman D, Rimm EB, et al. Relative Validity of Nutrient Intakes Assessed by Questionnaire, 24-Hour Recalls, and Diet Records as Compared With Urinary Recovery and Plasma Concentration Biomarkers: Findings for Women. Am J Epidemiol. May 1 2018;187(5):1051–1063. doi: 10.1093/aje/kwx328 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] Maruti SS, Feskanich D, Rockett HR, Colditz GA, Sampson LA, Willett WC. Validation of adolescent diet recalled by adults. Epidemiology. Mar 2006;17(2):226–9. doi: 10.1097/01.ede.0000198181.86685.49 [DOI] [PubMed] [Google Scholar]

[R39] Maruti SS, Feskanich D, Colditz GA, et al. Adult recall of adolescent diet: reproducibility and comparison with maternal reporting. Am J Epidemiol. Jan 1 2005;161(1):89–97. doi: 10.1093/aje/kwi019 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R40] Linos E, Willett WC, Cho E, Colditz G, Frazier LA. Red meat consumption during adolescence among premenopausal women and risk of breast cancer. Cancer Epidemiol Biomarkers Prev. Aug 2008;17(8):2146–51. doi: 10.1158/1055-9965.EPI-08-0037 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R41] Riseberg E, Wu Y, Lam WC, et al. Lifetime dairy product consumption and breast cancer risk: a prospective cohort study by tumor subtypes. Am J Clin Nutr. Nov 30 2023;doi: 10.1016/j.ajcnut.2023.11.017 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R42] Sotos-Prieto M, Bhupathiraju SN, Mattei J, et al. Changes in Diet Quality Scores and Risk of Cardiovascular Disease Among US Men and Women. Circulation. Dec 8 2015;132(23):2212–9. doi: 10.1161/CIRCULATIONAHA.115.017158 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R43] Willett W Nutritional epidemiology. 2nd ed. Monographs in epidemiology and biostatistics. Oxford University Press; 1998:xiv, 514 p. [Google Scholar]

[R44] Hu FB, Stampfer MJ, Rimm E, et al. Dietary fat and coronary heart disease: a comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements. Research Support, U.S. Gov’t, P.H.S. Am J Epidemiol. Mar 15 1999;149(6):531–40. [DOI] [PubMed] [Google Scholar]

[R45] Bernstein AM, Rosner BA, Willett WC. Cereal fiber and coronary heart disease: a comparison of modeling approaches for repeated dietary measurements, intermediate outcomes, and long follow-up. European journal of epidemiology. Nov 2011;26(11):877–86. doi: 10.1007/s10654-011-9626-x [DOI] [PubMed] [Google Scholar]

[R46] Hu Y, Zhang X, Ma Y, et al. Incident Type 2 Diabetes Duration and Cancer Risk: A Prospective Study in Two US Cohorts. J Natl Cancer Inst. Apr 6 2021;113(4):381–389. doi: 10.1093/jnci/djaa141 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R47] National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes. Dec 1979;28(12):1039–57. doi: 10.2337/diab.28.12.1039 [DOI] [PubMed] [Google Scholar]

[R48] Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes care. Jul 1997;20(7):1183–97. doi: 10.2337/diacare.20.7.1183 [DOI] [PubMed] [Google Scholar]

[R49] American Diabetes Association. Standards of medical care in diabetes--2010. Diabetes care. Jan 2010;33 Suppl 1:S11–61. doi: 10.2337/dc10-S011 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R50] Manson JE, Rimm EB, Stampfer MJ, et al. Physical activity and incidence of non-insulin-dependent diabetes mellitus in women. Lancet. Sep 28 1991;338(8770):774–8. doi: 10.1016/0140-6736(91)90664-b [DOI] [PubMed] [Google Scholar]

[R51] DeVille NV, Iyer HS, Holland I, et al. Neighborhood socioeconomic status and mortality in the nurses’ health study (NHS) and the nurses’ health study II (NHSII). Environ Epidemiol. Feb 2023;7(1):e235. doi: 10.1097/EE9.0000000000000235 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] Hunter DJ, Manson JE, Colditz GA, et al. Reproducibility of oral contraceptive histories and validity of hormone composition reported in a cohort of US women. Contraception. Dec 1997;56(6):373–8. doi: 10.1016/s0010-7824(97)00172-8 [DOI] [PubMed] [Google Scholar]

[R53] Troy LM, Hunter DJ, Manson JE, Colditz GA, Stampfer MJ, Willett WC. The validity of recalled weight among younger women. Int J Obes Relat Metab Disord. Aug 1995;19(8):570–2. [PubMed] [Google Scholar]

[R54] Wolf AM, Hunter DJ, Colditz GA, et al. Reproducibility and validity of a self-administered physical activity questionnaire. Int J Epidemiol. Oct 1994;23(5):991–9. doi: 10.1093/ije/23.5.991 [DOI] [PubMed] [Google Scholar]

[R55] Colditz GA, Stampfer MJ, Willett WC, et al. Reproducibility and validity of self-reported menopausal status in a prospective cohort study. Am J Epidemiol. Aug 1987;126(2):319–25. doi: 10.1093/aje/126.2.319 [DOI] [PubMed] [Google Scholar]

[R56] Rimm EB, Stampfer MJ, Colditz GA, Chute CG, Litin LB, Willett WC. Validity of self-reported waist and hip circumferences in men and women. Epidemiology. Nov 1990;1(6):466–73. doi: 10.1097/00001648-199011000-00009 [DOI] [PubMed] [Google Scholar]

[R57] Hosmer DW, Lemeshow S. Confidence interval estimation of interaction. Epidemiology. Sep 1992;3(5):452–6. doi: 10.1097/00001648-199209000-00012 [DOI] [PubMed] [Google Scholar]

[R58] Zhang Y, Chan AT, Meyerhardt JA, Giovannucci EL. Timing of Aspirin Use in Colorectal Cancer Chemoprevention: A Prospective Cohort Study. J Natl Cancer Inst. Jul 1 2021;113(7):841–851. doi: 10.1093/jnci/djab009 [DOI] [PMC free article] [PubMed] [Google Scholar]

[R59] SAS statistical software, version 9.4 for UNIX. Version version 9.4 for UNIX. SAS Institute Inc [Google Scholar]

PERMALINK

Life Course Dietary Insulinemic and Inflammatory Potential and Risk of Type 2 Diabetes in U.S. Women: A Prospective Cohort Study

Yin Zhang, PhD, MS, MPH

Mingyang Song, MD, ScD

Molin Wang, PhD

Emily Riseberg, MPH

Dong D Wang, MD, ScD

Deirdre K Tobias, ScD

Fred K Tabung, PhD

A Heather Eliassen, ScD

Stephanie A Smith-Warner, PhD

Qi Sun, MD, ScD

Eric B Rimm, ScD

Meir J Stampfer, MD, DrPH

Walter C Willett, MD, DrPH

Frank B Hu, MD, PhD

Edward L Giovannucci, MD, ScD

Abstract

Background

Objective

Design

Participants and setting

Main outcome measures

Statistical analyses performed

Results

Conclusions

Introduction

Methods

Study Population

Ethical Approval

Ascertainment of Diet

Ascertainment of Dietary Insulinemic and Inflammatory Potential over the Life Course

Ascertainment of Incident T2D

Ascertainment of Covariates

Statistical Analysis

Results

Population Characteristics

Table 1.

Table 2.

Lifetime, Adolescent, Premenopausal Adulthood, and Postmenopausal Adulthood Dietary Insulinemic Potential and Risk of T2D

Figure 4.

Figure 5.

Table 3.

Table 6.

Lifetime, Adolescent, Premenopausal Adulthood, and Postmenopausal Adulthood Dietary Inflammatory Potential and Risk of T2D

Adolescent Dietary Insulinemic and Inflammatory Potential and Risk of Premenopausal T2D

Table 7.

Lifetime Consistent High vs. Low Dietary Insulinemic and Inflammatory Potential and Risk of T2D

Figure 6.

Table 8.

Table 9.

Joint Association of Lifetime Dietary Insulinemic and Inflammatory Potential and Risk of T2D

Table 10.

Change in Dietary Insulinemic and Inflammatory Potential from Adolescence to Adulthood and Risk of T2D

Table 11.

Table 12.

Additional Analyses Adjusting for Weight Change and Waist Circumference

Sensitivity Analyses

Interaction Analyses

Estimated Cumulative Incidence of T2D in Individuals with Consistently High vs. Low Dietary Insulinemic and Inflammatory Potential

Figure 7.

Estimated Time Window for Adulthood Dietary Insulinemic and Inflammatory Potential and Risk of T2D

Figure 8. Analysis of adulthood dietary insulinemic and inflammatory potential and type 2 diabetes risk based on the ‘remote’ and ‘recent’ periods (using hard thresholds of 6, 8, 10, and 12 years) in the Nurses’ Health Study II (N = 40,135).

Discussion

Conclusions

Supplementary Material

Figure 3.

Table 4.

Table 5.

Practice Implications.

What is the current knowledge on this topic?

How does this research add to knowledge on this topic?

How might this knowledge impact current dietetics practice?

Research Snapshot.

Research Question:

Key Findings:

Acknowledgments

Funding

Footnotes

References