Abstract
Background
Dietary patterns have been associated with altered risk of diabetes mellitus (DM) in the general population. These patterns can estimate habitual intake of food groups associated with decreased or increased (e.g., greater intake of red meats) risk of DM.
Objective
This analysis aimed to examine the underexplored associations of four dietary patterns in patients presenting with an acute pancreatitis (AP) diagnosis based on the presence of pre-existing DM.
Methods
Study participants were selected from the Diabetes RElated to Acute Pancreatitis and its Mechanisms (DREAM) study, an ongoing, multicenter study of adults (18–75 years) with AP in the US. This is a cross-sectional analysis of baseline data collected using the VioScreen computer-administered Food Frequency Questionnaire. We examined four dietary patterns: the Alternate Mediterranean Diet (AMED), the Mediterranean-DASH Diet Intervention for Neurodegenerative Delay (MIND), the Healthy Eating Index-2020 (HEI-2020), and the Alternate Healthy Eating Index-2010 (AHEI-2010). Multivariable logistic regression was used to estimate the odds ratios (ORs) and 95% confidence intervals (CIs).
Results
The study population included 647 individuals with AP (mean (SD) baseline age, 49.1 (14.5) years). Participants in the highest tertile of dietary index scores were more likely to be older, female, non-Hispanic White, and to have higher educational attainment. In multivariable analyses, greater adherence to the AMED and the MIND dietary patterns was associated with lower odds of pre-existing DM. Multivariable ORs for the odds of pre-existing DM comparing the highest to lowest tertile of diet scores were 0.44 (95% CI: 0.23, 0.84; P-trend = 0.01) for AMED, 0.46 (95% CI: 0.26, 0.82; P-trend = 0.01) for MIND, indicating lower odds for participants with a healthier dietary pattern.
Conclusions
In this analysis, greater adherence to selected healthy dietary patterns was associated with lower odds of pre-existing DM at the time of AP presentation.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12937-026-01294-x.
Keywords: Alternate Mediterranean diet, Mediterranean-DASH diet intervention for neurodegenerative delay, Healthy eating index, Alternate healthy eating index
Introduction
Acute pancreatitis (AP) is a common gastrointestinal condition, responsible for more than 300,000 hospitalizations in the U.S. each year [1, 2]. In the absence of effective treatment for AP, there is much interest in modifiable risk factors for primary and/or secondary prevention. The relationship between type 2 diabetes mellitus (DM) and AP is complex and potentially bidirectional. Historically, DM was believed to be a transient phenomenon observed in AP; however, recent meta-analyses have shown that approximately one-quarter of individuals with AP develop DM within 3 years of discharge [3, 4], making DM a relatively common metabolic complication in AP. Modifiable risk factors, such as alcohol consumption, cigarette smoking, and physical activity [5–7] and intake of red or processed meats, fried foods, fish, and added sugars [8–11] have been associated with AP and DM [12–14]. However, there is a lack of detailed epidemiological data to examine dietary contributors, including their associations with DM among the high-risk AP population.
DM is also commonly recognized at the time of initial AP diagnosis. The frequency of preexisting DM among patients with AP has been reported in several epidemiological studies to be approximately 20% [15–19]. Although some studies have indicated that the presence of pre-existing DM may negatively influence the pathogenesis, severity, and/or mortality of AP, the evidence remains mixed, with at least one study reporting no significant association between pre-existing DM and AP [20, 21]. Moreover, several DM-related biomarkers, such as elevated serum HbA1c levels, have been suggested as indicators of poor prognosis in AP [22]. Individuals with DM are more likely to have hypertriglyceridemia, a common cause of AP [23]. As it relates to dietary contributors to the disease, a small, cross-sectional study of 108 AP participants found that higher intake of fiber, vegetables, and nuts was associated with lower fasting plasma glucose and HbA1c [24]. However, there is a lack of research studies that have investigated detailed dietary patterns and pre-existing DM among patients with AP, which constitutes a persistent gap in our knowledge.
Several dietary indices, such as the Mediterranean diet and the Mediterranean-DASH Diet Intervention for Neurodegenerative Delay (MIND), have been associated with a lower risk of DM in the general population [25–27]. These diets emphasize higher consumption of antioxidant-rich, nutrient-dense foods, including fruits, vegetables, and fiber. In contrast, pro-inflammatory diets, such as greater intake of red and processed meats, ultra-processed foods, and added sugars, are associated with an increased risk of DM [3, 4, 28, 29]. However, it remains uncertain whether dietary patterns affect the risk of pre-existing DM among AP patients.
This study aims to examine the association between four dietary patterns (Alternate Mediterranean Diet (AMED); MIND; Healthy Eating Index-2020 (HEI-2020); and Alternate Healthy Eating Index-2010 (AHE-2010)) and pre-existing DM among patients presenting with AP in a large, multicenter study [30]. These dietary patterns were selected based on their anti-inflammatory properties, adherence to the U.S. Dietary Guidelines, and relevance to chronic disease risk, including DM. We hypothesized that patients with an episode of AP in Diabetes RElated to Acute Pancreatitis and its Mechanisms (DREAM) study who adhere to healthier dietary patterns are less likely to have pre-existing DM.
Methods
Study design and study population
Participants for this analysis were selected from the DREAM study, an ongoing prospective cohort study of AP in the U.S. that seeks to enroll over 1,000 participants [30]. The DREAM study is registered on ClinicalTrials.gov under the identifier NCT05197920, which provides additional information about its objectives, study plan, collaborators, investigators, and participation criteria. The DREAM study aims to determine the incidence and mechanisms of DM that occur following an episode of AP. Enrollment for the study is ongoing across 10 Clinical Centers. Following enrollment, adult patients (18–75 years) without pre-existing diabetes are followed for up to five years to complete longitudinal study assessments [30]. Patients with AP who had DM prior to enrollment are also invited to participate by completing a baseline assessment without additional follow-up. It is anticipated that approximately 1,200 participants will complete a baseline visit. Of these, about 200 participants are expected to have pre-existing diabetes at enrollment, and approximately 800 participants are projected to complete the V3 follow-up visit. For the current cross-sectional analysis of baseline data, we included participants irrespective of their diabetes status at the time of enrollment. The analysis focuses on pre-existing diabetes at enrollment rather than on incident diabetes during follow-up.
The present analysis included (n = 867) adult (18–75 years) participants with AP who were enrolled as of April 21, 2025. All DREAM study participants were invited to complete a baseline dietary assessment at the enrollment visit using the VioScreen computer-administered Food Frequency Questionnaire (FFQ) [31]. Participants were excluded from the current analysis if they did not complete the FFQ (n = 123), had the FFQ completion of over 90 days (n = 2), or reported extreme daily energy intake levels (< 800 kcal or > 4,200 kcal for men and < 500 kcal or > 3,500 kcal for women) (n = 94) (Fig. 1).
Fig. 1.
Flowchart of the study participants
Dietary patterns assessment
VioScreen’s dietary analysis utilizes food and nutrient information from the Nutrition Coordinating Center (NCC) at the University of Minnesota Food and Nutrient Database [31]. The FFQ has been validated for use in adults [32]. The FFQ included standard portion sizes for each item as well as the frequency of intake. Food and nutrient components were used to derive dietary scores. Computation of AMED scores [33], MIND [34], HEI-2020 [35], and AHEI-2010 [25] in these studies has been described previously in detail. The MIND and HEI-2020 scores were directly calculated by VioCare Nutrition Software Solution for Healthcare and Research. AMED scores include 9 components. For seven of these components (fruits, vegetables, legumes, nuts, whole grains, seafood, and Monounsaturated Fatty Acids to Saturated Fatty Acids ratio (MUFA-to-SFA), intake over the median receives 1 point. In contrast, red and processed meat receives 1 point for intake below the median. Moderate alcohol use also receives 1 point [36]. The AMED score ranges from 0 to 9 points, which is calculated by adding the component scores, with higher scores indicating greater adherence to a Mediterranean pattern (Supplementary Table 1) [37]. The MIND diet score reflects a combination of elements of the Mediterranean and DASH diets, emphasizing components shown to protect against cognitive decline, such as green leafy vegetables and berries [38]. The MIND score ranges from 0 (lowest) to 15 points (highest) (Supplementary Table 2) [39]. The HEI score is a measure of overall diet quality used to assess how well a set of foods aligns with key recommendations and dietary patterns published in the Dietary Guidelines for Americans. The 13 dietary components of the HEI-2020 include total fruit, whole fruit, total vegetables, greens & beans, whole grains, dairy, total protein foods, seafood and plant proteins, fatty acids, refined grains, sodium, added sugars, and saturated fats [40]. The HEI-2020 total score ranges from 0 to 100, with higher scores indicating greater adherence to the Dietary Guidelines for Americans (Supplementary Table 3). The HEI-2020 score was also calculated by VioCare and included in the Nutrients Vector Export (NVE) data. Lastly, the AHEI is a measure of diet quality that assigns ratings to foods and nutrients that help reduce the risk of chronic diseases [25]. The AHEI-2010 score is based on the intake of 11 dietary components, including fruits, vegetables, whole grains, sugar-sweetened beverages and fruit juice, nuts and legumes, red and processed meat, trans fat, long-chain n-3 fats, Polyunsaturated Fatty Acids (PUFA), sodium, and alcohol [25]. Each component is scored from 0 (lowest) to 10 (best), with partial adherence receiving proportionally scaled scores between 0 and 10 based on intake. The total AHEI-2010 score is the sum of all component scores and ranges from 0 to 110 (Supplementary Table 4) [36].
Ascertainment of pre-existing diabetes status and covariates
Clinical history was obtained through both abstraction from medical records and direct history from participants taken by the study team. For the DREAM study, pre-existing DM was defined as: a history of physician-diagnosed DM, use of antidiabetic medication for the treatment of DM, or an HbA1c level of ≥ 6.5% prior to the onset of the AP episode that qualified them for study participation. Potential confounders including age, sex, race/ethnicity, body mass index (BMI), smoking status, Charlson comorbidity index (CCI), hyperlipidemia, history of prior AP, etiology and severity of AP, metabolic dysfunction-associated fatty liver disease (MAFLD), hypertension, total calorie intake (from FFQ), and time from AP diagnosis to FFQ completion were selected a priori based on prior literature and clinical expertise regarding potential associations with both DM and AP. Although not included in the multivariable regression models due to high levels of missing data, information on education, annual income, employment status, and food insecurity status was also extracted. Physical activity was not collected in the DREAM study, so it could not be included as a covariate in this analysis.
Statistical analysis
Univariable analyses were performed to assess percentages for categorical variables and means (SD) for continuous variables. Baseline characteristics of participants were described across tertiles of each dietary index. A correlation matrix was calculated to determine the Pearson and Spearman correlation coefficients between each of the four dietary indices. Multivariable logistic regression was used to determine associations between dietary patterns and pre-existing diabetes (yes/no), adjusting for potential confounders. The goodness-of-fit of the multivariable logistic regression models was assessed using the Hosmer–Lemeshow test. A non-significant result (P > 0.05) indicates that the model fits the data well. Odds ratios (ORs) and 95% confidence intervals (95% CIs) across tertiles of dietary patterns were calculated. Linear trends were tested for significance using the median intake within each dietary score tertile, modeled as a single continuous variable. We performed subgroup analyses to test the robustness of our results. First, an effect modification was tested by including the multiplicative interaction term between each dietary pattern and a history of prior AP, and the interaction was tested using the −2log-likelihood ratio (−2 LL) test. Second, to evaluate how unmeasured confounding such as socioeconomic status index (SEI), social determinants of health (SDOH), and physical activity levels might have influenced the observed association between the four dietary patterns and pre-existing DM, we calculated the E-value, which is an alternative sensitivity analysis method used in epidemiology and causal inference in observational studies introduced by VanderWeele and Ding [41]. More details on the concept of E-values are described elsewhere [41, 42]. The E-values formula below is based on the odds ratio for a common outcome (prevalence > 15%).
 |
 |
All summary tables were annotated with the total population size relevant to each table. Data were extracted and analyzed using SAS software (version 9.4, SAS Institute), and R statistical software (version 4.5.1) was used to calculate AHEI-2010 scores and generate the forest plot. All statistical tests were 2-sided, and P < 0.05 was considered to indicate statistical significance.
Results
The final study population consisted of 647 participants with AP, with a mean (SD) baseline age of 49.1 (14.5) years. Table 1 presents the baseline characteristics by tertiles of the four dietary patterns. Participants with higher dietary scores, indicative of healthy dietary patterns, were more likely to be older, female sex, non-Hispanic White, with higher educational attainment (i.e., holding a college degree or higher), have an annual income exceeding $100,000, and be employed (Table 1). Furthermore, participants with higher dietary patterns were more likely to have a lower BMI, a CCI of 0, and no MAFLD. Participants with AP with higher dietary scores had lower rates of prior AP and pre-existing DM and were more likely to have mild AP (Table 1). The correlations between the four dietary patterns ranged from moderately high to high (r = 0.67–0.80) (Supplemental Table 5).
Table 1.
Baseline characteristics of DREAM study participants according to tertiles of the AMED, MIND, HEI-2020, and AHEI-2010 scores (N = 647)a
| Characteristics | AMED | MIND | HEI-2020 | AHEI-2010b | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Tertile 1 N = 271 |
Tertile 2 N = 197 |
Tertile 3 N = 179 |
Tertile 1 N = 206 |
Tertile 2 N = 183 |
Tertile 3 N = 258 |
Tertile 1 N = 219 |
Tertile 2 N = 215 |
Tertile 3 N = 213 |
Tertile 1 N = 213 |
Tertile 2 N = 220 |
Tertile 3 N = 214 |
|
| Score median (range) | 2 | 4 | 6 | 5 | 6.5 | 8.5 | 53 | 66.1 | 76.6 | 47 | 59.8 | 74.4 |
| (0, 3) | (4,5) | (6,9) | (1,5.5) | (6,7) | (7.5,12.5) | (30.4,59.9) | (59.9,71.2) | (71.2,90.5) | (19.7,54.5) | (54.5,66.8) | (66.8,94.4) | |
| Age, mean (SD), years | 48.4 (14.6) | 49.1 (14.6) | 50.1 (14.2) | 47.2 (14.6) | 48.9 (14.9) | 50.7 (14.0) | 47.1 (14.3) | 48.9 (14.8) | 51.3 (14.1) | 47.0 (14.5) | 48.2 (14.7) | 52.0 (13.9) |
| Age group, n (%) | ||||||||||||
| 18–30 | 38 (14.0) | 20 (10.2) | 19 (10.6) | 31 (15.0) | 23 (12.6) | 23 (8.9) | 33 (15.1) | 26 (12.1) | 18 (8.5) | 34 (16.0) | 29 (13.2) | 14 (6.5) |
| 31–60 | 167 (61.6) | 121 (61.4) | 110 (61.5) | 127 (61.7) | 107 (58.5) | 164 (63.6) | 141 (64.4) | 129 (60.0) | 128 (60.1) | 135 (63.4) | 134 (60.9) | 129 (60.3) |
| 61–75 | 66 (24.4) | 56 (28.4) | 50 (27.9) | 48 (23.3) | 53 (29.0) | 71 (27.5) | 45 (20.5) | 60 (27.9) | 67 (31.5) | 44 (20.7) | 57 (25.9) | 71 (33.2) |
| Sex, n (%) | ||||||||||||
| Male | 132 (48.7) | 77 (39.1) | 88 (49.2) | 103 (50.0) | 86 (47.0) | 108 (41.9) | 124 (56.6) | 81 (37.7) | 92 (43.2) | 104 (48.8) | 100 (45.5) | 93 (43.5) |
| Female | 139 (51.3) | 120 (60.9) | 91 (50.8) | 103 (50.0) | 97 (53.0) | 150 (58.1) | 95 (43.4) | 134 (62.3) | 121 (56.8) | 109 (51.2) | 120 (54.5) | 121 (56.5) |
| Race, n (%) | ||||||||||||
| White/Caucasian | 204 (75.6) | 146 (75.3) | 141 (79.7) | 147 (71.4) | 138 (76.7) | 206 (80.8) | 162 (74.3) | 166 (78.3) | 163 (77.3) | 156 (73.2) | 173 (80.1) | 162 (76.4) |
| Black/AA | 44 (16.3) | 33 (17.0) | 19 (10.7) | 46 (22.3) | 27 (15.0) | 23 (9.0) | 37 (17.0) | 30 (14.2) | 29 (13.7) | 39 (18.3) | 29 (13.4) | 28 (13.2) |
| Others | 22 (8.1) | 15 (7.7) | 17 (9.6) | 13 (6.3) | 15 (8.3) | 26 (10.2) | 19 (8.7) | 16 (7.5) | 19 (9.0) | 18 (8.5) | 14 (6.5) | 22 (10.4) |
| Missing | 1 (0.4) | 3 (1.5) | 2 (1.1) | 0 (0.0) | 3 (1.6) | 3 (1.2) | 1 (0.5) | 3 (1.4) | 2 (0.9) | 0 (0.0) | 4 (1.8) | 2 (0.9) |
| Education, n (%) | ||||||||||||
| College graduate or above | 14 (6.8) | 7 (4.4) | 2 (1.6) | 9 (5.6) | 6 (4.3) | 8 (4.3) | 10 (6.5) | 8 (4.5) | 5 (3.2) | 12 (7.5) | 6 (3.6) | 5 (3.1) |
| High school graduate | 56 (27.1) | 26 (16.5) | 15 (12.2) | 44 (27.5) | 31 (22.1) | 22 (11.7) | 45 (29.2) | 32 (18.0) | 20 (12.8) | 45 (28.1) | 34 (20.6) | 18 (11.0) |
| College or AA degree | 63 (30.4) | 59 (37.3) | 40 (32.5) | 57 (35.6) | 48 (34.3) | 57 (30.3) | 49 (31.8) | 60 (33.7) | 53 (34.0) | 48 (30.0) | 68 (41.2) | 46 (28.2) |
| College graduate or above | 74 (35.7) | 66 (41.8) | 66 (53.7) | 50 (31.3) | 55 (39.3) | 101 (53.7) | 50 (32.5) | 78 (43.8) | 78 (50.0) | 55 (34.4) | 57 (34.5) | 94 (57.7) |
| Missing | 64 (23.6) | 39 (19.8) | 56 (31.3) | 46 (22.3) | 43 (23.5) | 70 (27.1) | 65 (29.7) | 37 (17.1) | 57 (26.6) | 53 (24.9) | 55 (25.0) | 51 (23.8) |
| Annual income, n (%) | ||||||||||||
| Under 50,000 | 55 (32.4) | 45 (32.1) | 29 (25.9) | 49 (37.7) | 40 (32.3) | 40 (23.8) | 46 (37.1) | 46 (29.7) | 37 (25.9) | 44 (33.8) | 54 (37.2) | 31 (21.1) |
| 50,000–99,999 | 50 (29.4) | 37 (26.4) | 26 (23.2) | 40 (30.8) | 30 (24.2) | 43 (25.6) | 36 (29.0) | 42 (27.1) | 35 (24.5) | 38 (29.2) | 40 (27.6) | 35 (23.8) |
| Over 100,000 | 65 (38.2) | 58 (41.4) | 57 (50.9) | 41 (31.5) | 54 (43.5) | 85 (50.6) | 42 (33.9) | 67 (43.2) | 71 (49.7) | 48 (36.9) | 51 (35.2) | 81 (55.1) |
| Missing | 101 (37.3) | 57 (21.0) | 67 (37.4) | 76 (36.9) | 59 (32.2) | 90 (34.9) | 95 (43.4) | 60 (27.8) | 70 (32.7) | 83 (39.0) | 75 (34.1) | 67 (31.3) |
| Employment, n (%) | ||||||||||||
| Employed | 130 (64.4) | 100 (64.1) | 76 (62.3) | 102 (64.6) | 87 (63.5) | 117 (63.2) | 99 (65.6) | 114 (65.1) | 93 (60.4) | 100 (63.7) | 106 (64.6) | 100 (62.9) |
| Unemployed for other reasons | 47 (23.3) | 44 (28.2) | 40 (32.8) | 39 (24.7) | 34 (24.8) | 58 (31.4) | 34 (22.5) | 48 (27.4) | 49 (31.8) | 37 (23.6) | 43 (26.2) | 51 (32.1) |
| Unemployed for medical reasons | 25 (12.4) | 12 (7.7) | 6 (4.9) | 17 (10.8) | 16 (11.7) | 10 (5.4) | 18 (11.9) | 13 (7.4) | 12 (7.8) | 20 (12.7) | 15 (9.1) | 8 (5.0) |
| Missing | 69 (25.5) | 41 (20.8) | 57 (31.8) | 48 (23.3) | 46 (25.1) | 73 (28.3) | 68 (30.9) | 40 (19.0) | 59 (27.6) | 56 (26.3) | 56 (25.5) | 55 (25.7) |
| Smoking status, n (%) | ||||||||||||
| Never smoker | 156 (57.6) | 108 (55.1) | 121 (68.8) | 110 (53.4) | 103 (56.6) | 172 (67.5) | 115 (52.5) | 136 (63.8) | 134 (63.5) | 116 (54.7) | 132 (60.0) | 137 (64.9) |
| Former smoker | 68 (25.1) | 58 (29.6) | 38 (21.6) | 55 (26.7) | 50 (27.5) | 59 (23.1) | 58 (26.5) | 52 (24.4) | 54 (25.6) | 52 (24.5) | 59 (26.8) | 53 (25.1) |
| Current smoker | 47 (17.3) | 30 (15.3) | 17 (9.7) | 41 (19.9) | 29 (15.9) | 24 (9.4) | 46 (21.0) | 25 (11.7) | 23 (10.9) | 44 (20.8) | 29 (13.2) | 21 (10.0) |
| Missing | 0 (0.0) | 1 (0.5) | 3 (1.7) | 0 (0.0) | 1 (0.5) | 3 (1.2) | 0 (0.0) | 2 (0.9) | 2 (0.9) | 1 (0.5) | 0 (0.0) | 3 (1.4) |
| Food insecurity, n (%) | ||||||||||||
| No | 158 (75.6) | 132 (82) | 106 (86.9) | 123 (75) | 112 (79.4) | 161 (86.1) | 117 (73.9) | 145 (81.9) | 134 (85.4) | 119 (73.9) | 133 (79.2) | 144 (88.3) |
| Yes | 51 (24.4) | 29 (18.0) | 16 (13.1) | 41 (25.0) | 29 (20.6) | 26 (13.9) | 41 (26.1) | 32 (18.1) | 23 (14.6) | 42 (26.1) | 35 (20.8) | 19 (11.7) |
| Missing | 62 (22.9) | 36 (18.3) | 57 (31.8) | 42 (20.4) | 42 (23.0) | 71 (27.5) | 61 (27.9) | 38 (17.7) | 56 (26.3) | 52 (24.4) | 52 (23.6) | 51 (23.8) |
| BMI, mean (SD), kg/m2 | ||||||||||||
| 31.4 (8.3) | 30.7 (8.4) | 29.7 (6.9) | 32.2 (8.7) | 30.4 (7.8) | 29.8 (7.4) | 31.2 (8.4) | 32.1 (8.6) | 28.9 (6.4) | 31.7 (8.2) | 31.3 (8.8) | 29.3 (6.6) | |
| Missing | 1 (0.4) | 0 (0.0) | 0 (0.0) | 1 (0.5) | 0 (0.0) | 0 (0.0) | 1 (0.5) | 0 (0.0) | 0 (0.0) | 1 (0.5) | 0 (0.0) | 0 (0.0) |
| BMI, n (%) | ||||||||||||
| Normal weight | 62 (23.0) | 50 (25.4) | 44 (24.6) | 42 (20.5) | 42 (23.0) | 72 (27.9) | 47 (21.6) | 50 (23.3) | 59 (27.7) | 46 (21.7) | 49 (22.3) | 61 (28.5) |
| Overweight | 74 (27.4) | 57 (28.9) | 65 (36.3) | 55 (26.8) | 57 (31.1) | 84 (32.6) | 66 (30.3) | 50 (23.3) | 80 (37.6) | 53 (25.0) | 71 (32.3) | 72 (33.6) |
| Obese | 134 (49.6) | 90 (45.7) | 70 (39.1) | 108 (52.7) | 84 (45.9) | 102 (39.5) | 105 (48.2) | 115 (53.5) | 74 (34.7) | 113 (53.3) | 100 (45.5) | 81 (37.9) |
| Missing | 1 (0.4) | 0 (0.0) | 0 (0.0) | 1 (0.5) | 0 (0.0) | 0 (0.0) | 1 (0.5) | 0 (0.0) | 0 (0.0) | 1 (0.5) | 0 (0.0) | 0 (0.0) |
| Charlson CI, n (%) | ||||||||||||
| 0 | 222 (81.9) | 160 (81.2) | 149 (83.2) | 164 (79.6) | 148 (80.9) | 219 (84.9) | 173 (78.8) | 187 (87.0) | 171 (80.3) | 172 (80.8) | 189 (85.9) | 170 (79.4) |
| 1 | 38 (14.0) | 26 (13.2) | 25 (14.0) | 32 (15.5) | 27 (14.8) | 30 (11.6) | 32 (14.6) | 23 (10.7) | 34 (16.0) | 30 (14.1) | 22 (10.0) | 37 (17.3) |
| > = 2 | 11 (4.1) | 11 (5.6) | 5 (2.8) | 10 (4.9) | 8 (4.4) | 9 (3.5) | 14 (6.4) | 5 (2.3) | 8 (3.8) | 11 (5.2) | 9 (4.1) | 7 (3.3) |
| AP etiology, n (%) | ||||||||||||
| Gallstones | 96 (35.8) | 68 (35.6) | 66 (37.9) | 68 (33.3) | 65 (36.3) | 97 (38.8) | 72 (33.3) | 81 (38.2) | 77 (37.6) | 60 (28.6) | 85 (39.4) | 85 (41.1) |
| Alcohol | 55 (20.5) | 45 (23.6) | 33 (19.0) | 48 (23.5) | 47 (26.3) | 38 (15.2) | 56 (25.9) | 38 (18.4) | 38 (18.5) | 51 (24.3) | 49 (22.7) | 33 (15.9) |
| Others | 117 (43.7) | 78 (40.8) | 75 (43.1) | 88 (43.1) | 67 (37.4) | 115 (46.0) | 88 (40.7) | 92 (43.4) | 90 (43.9) | 99 (47.1) | 82 (38.0) | 89 (43.0) |
| Missing | 3 (1.1) | 6 (3.0) | 5 (2.8) | 2 (1.0) | 4 (2.2) | 8 (3.1) | 3 (1.4) | 3 (1.4) | 8 (3.8) | 3 (1.4) | 4 (1.8) | 7 (3.3) |
| History of prior AP, n (%) | ||||||||||||
| No | 167 (61.9) | 134 (68.7) | 130 (73.4) | 136 (66.3) | 111 (61.0) | 184 (72.2) | 143 (65.6) | 144 (67.6) | 144 (68.2) | 126 (59.4) | 146 (66.7) | 159 (75.4) |
| Yes | 103 (38.1) | 61 (31.3) | 47 (26.6) | 69 (33.7) | 71 (39.0) | 71 (27.8) | 75 (34.4) | 69 (32.4) | 67 (31.8) | 86 (40.6) | 73 (33.3) | 52 (24.6) |
| Missing | 1 (0.4) | 2 (1.0) | 2 (1.1) | 1 (0.5) | 1 (0.5) | 3 (1.2) | 1 (0.5) | 2 (0.9) | 2 (0.9) | 1 (0.5) | 1 (0.5) | 3 (1.4) |
| Severity of AP, n (%) | ||||||||||||
| Mild | 230 (85.8) | 165 (83.8) | 155 (87.6) | 169 (82.4) | 161 (89.0) | 220 (85.9) | 187 (85.8) | 183 (86.3) | 180 (84.9) | 178 (84.8) | 189 (85.9) | 183 (86.3) |
| Moderately | 27 (10.1) | 24 (12.2) | 19 (10.7) | 24 (11.7) | 14 (7.7) | 32 (12.5) | 22 (10.1) | 23 (10.8) | 25 (11.8) | 22 (10.5) | 22 (10.0) | 26 (12.3) |
| Severe | 11 (4.1) | 8 (4.1) | 3 (1.7) | 12 (5.9) | 6 (3.3) | 4 (1.6) | 9 (4.1) | 6 (2.8) | 7 (3.3) | 10 (4.8) | 9 (4.1) | 3 (1.4) |
| Missing | 3 (1.1) | 0 (0.0) | 2 (1.1) | 1 (0.5) | 2 (1.1) | 2 (0.8) | 1 (0.5) | 3 (1.4) | 1 (0.5) | 3 (1.4) | 0 (0.0) | 2 (0.9) |
| MAFLD, n (%) | ||||||||||||
| No | 218 (80.4) | 169 (85.8) | 152 (84.9) | 170 (82.5) | 155 (84.7) | 214 (82.9) | 178 (81.3) | 179 (83.3) | 182 (85.4) | 172 (80.8) | 188 (85.5) | 179 (83.6) |
| Yes | 53 (19.6) | 28 (14.2) | 27 (15.1) | 36 (17.5) | 28 (15.3) | 44 (17.1) | 41 (18.7) | 36 (16.7) | 31 (14.6) | 41 (19.2) | 32 (14.5) | 35 (16.4) |
| Hypertension, n (%) | ||||||||||||
| No | 153 (56.5) | 123 (62.4) | 108 (60.3) | 115 (55.8) | 114 (62.3) | 155 (60.1) | 133 (60.7) | 129 (60.0) | 122 (57.3) | 121 (56.8) | 141 (64.1) | 122 (57.0) |
| Yes | 118 (43.5) | 74 (37.6) | 71 (39.7) | 91 (44.2) | 69 (37.7) | 103 (39.9) | 86 (39.3) | 86 (40.0) | 91 (42.7) | 92 (43.2) | 79 (35.9) | 92 (43.0) |
| Hyperlipidemia, n (%) | ||||||||||||
| No | 194 (71.6) | 144 (73.1) | 121 (67.6) | 147 (71.4) | 145 (79.2) | 167 (64.7) | 170 (77.6) | 154 (71.6) | 135 (63.4) | 152 (71.4) | 170 (77.3) | 137 (64.0) |
| Yes | 77 (28.4) | 53 (26.9) | 58 (32.4) | 59 (28.6) | 38 (20.8) | 91 (35.3) | 49 (22.4) | 61 (28.4) | 78 (36.6) | 61 (28.6) | 50 (22.7) | 77 (36.0) |
| Preexisting diabetes, n (%) | ||||||||||||
| No | 204 (75.3) | 163 (82.7) | 153 (85.5) | 152 (73.8) | 150 (82.0) | 218 (84.5) | 167 (76.3) | 180 (83.7) | 173 (81.2) | 159 (74.6) | 187 (85.0) | 174 (81.3) |
| Yes | 67 (24.7) | 34 (17.3) | 26 (14.5) | 54 (26.2) | 33 (18.0) | 40 (15.5) | 52 (23.7) | 35 (16.3) | 40 (18.8) | 54 (25.4) | 33 (15.0) | 40 (18.7) |
| Total calorie (kcal) | ||||||||||||
| Mean (SD) | 1600.1 (713.5) | 1941.4 (759.0) | 2132.6 (788.9) | 1853.1 (834.9) | 1774.8 (751.8) | 1904.2 (754.9) | 1909.5 (837.9) | 1870.1 (762.6) | 1772.6 (734.5) | 1637.5 (731.3) | 1899.7 (831.4) | 2014.5 (729.3) |
Abbreviations: AMED Alternate Mediterranean Diet, MIND Mediterranean-DASH Diet Intervention for Neurodegenerative Delay, HEI-2020 Healthy Eating Index-2020, AHEI-2010 Alternate Healthy Eating Index-2010, DREAM Diabetes RElated to Acute Pancreatitis and its Mechanisms, BMI Body Mass Index, AP Acute Pancreatitis, FFQ Food Frequency Questionnaire, Charlson CI Charlson Comorbidity Index, AA African American, MAFLD Metabolic dysfunction-associated fatty liver disease
aMean (SD) for continuous variables and proportion for categorical variables
bThe median (range) of AHEI-2010 values for each tertile is as follows: Tertile 1–47.0 (19.7–54.47); Tertile 2–59.8 (54.5–66.79); Tertile 3–74.4 (66.81–94.4). Due to space constraints, values in Table 1 were rounded to one decimal place
In multivariable-adjusted analyses, greater adherence to the AMED and MIND dietary patterns was associated with lower odds of pre-existing DM at enrollment. For example, compared to the lowest tertile, participants in the highest tertile of the AMED dietary pattern had significantly lower odds of pre-existing DM (adjusted OR, 0.44; 95% CI, 0.23–0.84; P for trend = 0.01). Similarly, greater adherence to the MIND dietary pattern was associated with lower odds of pre-existing DM (OR, 0.46; 95% CI, 0.26–0.82; P for trend = 0.01). There was no linear dose–response observed for HEI-2020. Compared with the lowest tertile of the HEI-2020 dietary pattern, participants with AP in the middle tertile had lower odds of pre-existing DM (adjusted OR, 0.51; 95% CI, 0.29–0.92; P for trend = 0.04), whereas those in the highest tertile did not differ. Additionally, there was no evidence of a non-linear relationship between HEI-2020 dietary pattern and pre-existing diabetes (P for non-linearity = 0.82). When assessed as a continuous variable, the AHEI-2010 dietary pattern showed a non-statistical trend toward lower odds of pre-existing DM with each 1-standard deviation increment (adjusted OR, 0.84; 95% CI, 0.66–1.07) (Table 2). In all analyses, the Hosmer–Lemeshow test yielded a P-value > 0.05, indicating that the models fit the data well. We did not observe any significant interaction between the four dietary patterns and a history of prior AP (P-interaction > 0.05, for all) (Supplemental Fig. 1).
Table 2.
Multivariable logistic regression results showing the ORs (95% CI) for the association between dietary pattern scores and pre–existing diabetes in the DREAM study (n = 647)
| Dietary Patterns & Models* | Tertile 1 | Tertile 2 | Tertile 3 | Per 1-SD increment | P for trend§ |
|---|---|---|---|---|---|
| AMED | |||||
| Case subjects/non-case subjects | 67/204 | 34/163 | 26/153 | ||
| Median score (range) | 2 (0–3) | 4 (4–5) | 6 (6–9) | 4 (0–9) | |
| AMED—Model 1 | 1.00 (ref) | 0.59 (0.36, 0.95) | 0.45 (0.26, 0.77) | 0.71 (0.57, 0.88) | 0.002 |
| AMED—Model 2 | 1.00 (ref) | 0.74 (0.42, 1.29) | 0.43 (0.23, 0.82) | 0.74 (0.57, 0.95) | 0.01 |
| AMED—Model 3 | 1.00 (ref) | 0.74 (0.42, 1.30) | 0.44 (0.23, 0.84) | 0.75 (0.58, 0.97) | 0.01 |
| MIND | |||||
| Case subjects/non-case subjects | 54/152 | 33/150 | 40/218 | ||
| Median score (range) | 5 (1–5.5) | 6.5 (6–7) | 8.5 (7.5–12.5) | 6.5 (1–12.5) | |
| MIND Diet—Model 1 | 1.00 (ref) | 0.56 (0.34, 0.93) | 0.44 (0.27, 0.71) | 0.77 (0.63, 0.94) | 0.001 |
| MIND Diet—Model 2 | 1.00 (ref) | 0.70 (0.39, 1.26) | 0.45 (0.26, 0.79) | 0.78 (0.62, 0.99) | 0.005 |
| MIND Diet—Model 3 | 1.00 (ref) | 0.71 (0.40, 1.27) | 0.46 (0.26, 0.82) | 0.80 (0.62, 1.01) | 0.01 |
| HEI-2020 | |||||
| Case subjects/non-case subjects | 52/167 | 35/180 | 40/173 | ||
| Median score (range) | 53.1 (30.4–59.9) | 66.1 (60–71.2) | 76.7 (71.3–90.5) | 66.1 (30.4–90.5) | |
| HEI-2020—Model 1 | 1.00 (ref) | 0.56 (0.35, 0.92) | 0.62 (0.38, 1.00) | 0.81 (0.66, 0.98) | 0.04 |
| HEI-2020—Model 2 | 1.00 (ref) | 0.52 (0.29, 0.93) | 0.54 (0.30, 0.95) | 0.80 (0.63, 1.02) | 0.03 |
| HEI-2020—Model 3 | 1.00 (ref) | 0.51 (0.29, 0.92) | 0.57 (0.32, 1.01) | 0.82 (0.64, 1.04) | 0.04 |
| AHEI-2010 | |||||
| Case subjects/non-case subjects | 54/159 | 33/187 | 40/174 | ||
| Median score (range) | 47 (19.7–54.47) | 59.8 (54.5–66.79) | 74.4 (66.81–94.4) | 60.0 (19.7–94.4) | |
| AHEI-2010—Model 1 | 1.00 (ref) | 0.47 (0.29, 0.78) | 0.54 (0.33, 0.88) | 0.75 (0.61, 0.92) | 0.02 |
| AHEI-2010—Model 2 | 1.00 (ref) | 0.70 (0.40, 1.24) | 0.69 (0.39, 1.24) | 0.83 (0.65, 1.05) | 0.22 |
| AHEI-2010—Model 3 | 1.00 (ref) | 0.70 (0.39, 1.24) | 0.74 (0.41, 1.33) | 0.84 (0.66, 1.07) | 0.31 |
Model 1: Age and total calorie intake adjusted
Model 2: Adjusted for the covariates in Model 1 + further adjustment of sex, race/ethnicity, smoking, Charlson comorbidity index, hyperlipidemia, history of prior AP, etiology of AP, severity of AP, MAFLD, hypertension, duration from AP diagnosis to FFQ completion, and total calorie intake
Model 3: Adjusted for the covariates in Model 2 + further adjustment of BMI
Abbreviations: AMED Alternate Mediterranean Diet, MIND Mediterranean-DASH Diet Intervention for Neurodegenerative Delay, HEI-2020 Healthy Eating Index-2020, AHEI-2010 Alternate Healthy Eating Index-2010
§P values for linear trend were estimated using the original scale median intake of each dietary pattern within each tertile entered into the models as continuous variables
*Each multivariable logistic regression model's fit was evaluated using the Hosmer and Lemeshow Goodness of Fit test. There was no evidence of poor fit in any of the models' results, which were non-significant (p > 0.05 for all)
E-value
The E-value for the AMED Tertile 3 vs. Tertile 1 was 2.38, and for the MIND Tertile 3 vs. Tertile 1 was 2.31 (Supplemental Table 6). This indicates that an unmeasured confounder would need to have an OR of at least 2.38 for AMED and 2.31 for MIND to negate the observed association between these dietary patterns and pre-existing DM in this study. These high E-values suggest that our findings were robust to unmeasured confounding.
Discussion
In the largest study to date using detailed dietary data in a population of adults with AP, we observed that greater adherence to selected healthy dietary patterns was associated with lower odds of pre-existing DM. These findings were observed across various dietary patterns (AMED and MIND), and the differences persisted after adjusting for sociodemographic and lifestyle factors, including smoking, comorbidities, and other dietary factors. In addition to demonstrating the feasibility of prospectively collecting detailed dietary information in a large study population with AP, our results have potential implications for future research and clinical management for this common metabolic complication.
The observational design of our cross-sectional analysis does not allow us to make causal inferences; however, we observed that healthy eating patterns were associated with lower odds of pre-existing DM. These findings are consistent with previous prospective cohort studies conducted in the general population that have reached similar conclusions. For example, a study conducted by Ahmad and colleagues found that higher Mediterranean diet (MED) intake scores were associated with a 30% lower risk of DM using data from the Women’s Health Study [12]. A nationwide cohort study in China also found an inverse association between the MED adherence and new-onset DM in the Chinese population [43]. A meta-analysis based on 16 prospective cohort studies also found that the greater adherence to the MED was significantly associated with a reduced risk of DM [44]. Our finding that higher MIND dietary pattern is associated with lower odds of preexisting DM is consistent with a previous study suggesting a preventive effect of the MIND diet against DM [27, 45]. Furthermore, a comprehensive systematic review and meta-analysis of diet quality indices, including 113 studies with over 3 million participants, found that adherence to higher diet quality (assessed by the HEI, AHEI, and Dietary Approaches to Stop Hypertension), was associated with lower risk of developing T2D [46] Lastly, the Singapore Chinese Health Study, which included over 45,000 participants, also found that higher adherence to various dietary patterns, including the AMED and AHEI-2010, was associated with a 16% to 29% lower risk of developing T2D [47].
A potential biological mechanism underlying the association between dietary patterns and DM risk may involve oxidative stress. Oxidative stress results from an imbalance between the production of reactive oxygen and nitrogen species (RONS) and the body's antioxidant defenses, which play a significant role in the pathogenesis of many major chronic diseases, including DM [48, 49]. Furthermore, other essential components, such as dietary fiber and fat quality, may influence insulin sensitivity and DM risk [24, 50, 51]. A lower risk of DM in the general population has been linked to several dietary patterns, including the AMED and MIND, which are based on the adherence to foods high in antioxidants, such as fruits, vegetables, extra virgin olive oil, and whole grains, and consuming fewer red meats, sweets, and processed foods [25, 26]. In addition, the higher saturated fat and heme–iron content [52, 53] in red meat, as well as the high sodium levels [52] in processed meat, may contribute to adverse health effects and increase the risk of DM. These dietary components are known to be positively associated with endothelial dysfunction, inflammation, insulin resistance, and oxidative stress that can cause tissue damage and impair pancreatic β-cell function [53–55]. While dietary recommendations for the general population (such as reducing red meat and sodium intake and increasing fruit, vegetable, and fiber consumption) remain relevant to DM prevention, patients with AP may require additional or tailored dietary approaches. At this time, we are only able to speculate, but dietary recommendations in this patient population may potentially need to focus on measures that preserve β-cell function and/or lower the risk of recurrent AP episodes; considerations that may not be fully addressed by general population–based dietary guidance.
To the best of our knowledge, this is the first study to concurrently examine the associations of multiple dietary patterns with pre-existing DM among adults with AP using an approach that adjusts for a wide range of potential confounders. However, our study has several limitations that should be considered. First, although we did not observe an interaction based on prior AP history, the group sizes did not allow us to perform analyses restricted to those experiencing a first lifetime episode of AP. Second, due to a high level of missing values for the SEI and SDOH variables, we could not include them in the multivariable logistic regression models. Third, as in any observational study, residual confounding may persist. To minimize this risk, we carefully controlled a large set of potential confounders, including major lifestyle factors such as smoking and other dietary risk factors. In addition, individuals with healthy dietary patterns may also make other healthy choices, such as physical activity, which we did not adjust for in the current analysis. While future studies should consider this and other factors, the relatively high E-values in this study suggest robustness to potential unmeasured confounding. We also acknowledge the possibility of reverse causation that AP patients with preexisting DM may have improved their diet quality, which could have underestimated the true associations. The cross-sectional design of the study does not allow us to rule out this consideration and prevents us from examining causation beyond associations. Furthermore, as with all dietary pattern assessments, there is a risk of recall bias. Lastly, although each dietary pattern was modeled independently, this strategy entails numerous hypothesis tests, which may increase the risk of chance findings, so our results should be carefully interpreted.
Conclusions
In the largest study to date incorporating detailed dietary data in a population of adults with AP, greater adherence to selected healthy dietary patterns, including the AMED and MIND, was consistently associated with lower odds of pre-existing DM. This study represents an early step toward understanding the role of diet in the bidirectional relationship between DM and AP. Future prospective studies with repeated dietary assessments are warranted to confirm the observed associations. Instead of post-onset prevention, such studies could support risk assessment, early glycemic metabolic monitoring, and proactive therapeutic care for this specific patient population.
Supplementary Information
Supplementary Material 1: Supplemental Figure 1. Dietary patterns and pre-existing diabetes stratified by a history of prior AP. Supplementary Table 1. Alternate Mediterranean Diet (AMED) components and criteria for scoring. Supplementary Table 2. Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) components and criteria for scoring. Supplementary Table 3. Healthy Eating Index (HEI)-2020 components and criteria for scoringa. Supplementary Table 4. Alternate Healthy Eating Index (AHEI)-2010 components and criteria for scoring. Supplemental Table 5.Pearson and Spearman correlation coefficients between diet scores. Supplementary Table 6: Reporting E-values for the association between dietary pattern scores and pre-existing diabetes.
Acknowledgments
Collaborators (who contributed but not included in authorship): The following individuals should be listed in PubMed as collaborators in connection to the corporate author’s name, “on behalf of the Type 1 Diabetes in Acute Pancreatitis Consortium (T1DAPC)”: Darwin L Conwell; Jeffrey Easler; Steven Hughes; Rajesh Keswani; Ece Mutlu; Temel Tirkes.
IRB name
Office for Research Protections
Human Research Protection Program
The Pennsylvania State University
Abbreviations
- AMED
Alternate Mediterranean Diet
- MIND
Mediterranean-DASH Diet Intervention for Neurodegenerative Delay
- HEI-2020
Healthy Eating Index-2020
- AHEI-2010
Alternate Healthy Eating Index-2010
- DREAM
Diabetes RElated to Acute Pancreatitis and its Mechanisms
- AP
Acute pancreatitis
- DM
Diabetes mellitus
- T2D
Type 2 diabetes
- RONS
Reactive oxygen and nitrogen species
- FFQ
Food frequency questionnaire
Authors’ contributions
1. designed research (project conception, development of overall research plan, and study oversight): DMB, PH, SQ, NRK, VMC 2. analyzed data: DMB, PH, SQ, FT, TQ, NRK, VMC 3. performed statistical analysis: DMB 4. wrote first draft of the manuscript: DMB and PH 5. had primary responsibility for final content. PH, VMC, DMB 6. All authors have contributed to patient enrollment and data generation and have read and approved the final manuscript.
Funding
Research reported in this publication was supported by funding from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) for the Type 1 Diabetes in Acute Pancreatitis Consortium (T1DAPC) under award numbers U01 DK127367, U01 DK127377, U01 DK127378, U01 DK127382, U01 DK127384, U01 DK127388, U01 DK127392, U01 DK127395, U01 DK127400, U01 DK127403, and U01 DK127404. D.M.B. was supported by the NIDDK training grant 1K01DK140625-01. M.O.G. was supported by the Eris M. Field Chair in Diabetes Research and P30-DK063491. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Data availability
Data described in the manuscript, codebook, and analytic code will not be made available because enrollment in the DREAM study is ongoing across 10 Clinical Centers. Upon completion of the study, data will be shared in accordance with NIH data-sharing policies.
Declarations
Ethics approval and consent to participate
All participants provided informed consent in accordance with an IRB-approved informed consent document.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Phil A. Hart, Email: philip.hart@osumc.edu
Type 1 Diabetes in Acute Pancreatitis Consortium (T1DAPC):
Darwin L. Conwell, Jeffrey Easler, Steven Hughes, Rajesh Keswani, Ece Mutlu, and Temel Tirkes
References
- 1.Fagenholz PJ, del Fernández- Castillo C, Harris NS, Pelletier AJ, Camargo CA Jr. Direct medical costs of acute pancreatitis hospitalizations in the United States. Pancreas. 2007;35(4):302–7. 10.1097/MPA.0b013e3180cac24b. [DOI] [PubMed] [Google Scholar]
- 2.Peery AF, Murphy CC, Anderson C, Jensen ET, Deutsch-Link S, Egberg MD, et al. Burden and cost of gastrointestinal, liver, and pancreatic diseases in the United States: update 2024. Gastroenterology. 2025;168(5):1000–24. 10.1053/j.gastro.2024.12.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zhi M, Zhu X, Lugea A, Waldron RT, Pandol SJ, Li L. Incidence of new onset diabetes mellitus secondary to acute pancreatitis: a systematic review and meta-analysis. Front Physiol. 2019;10:637. 10.3389/fphys.2019.00637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Das SL, Singh PP, Phillips AR, Murphy R, Windsor JA, Petrov MS. Newly diagnosed diabetes mellitus after acute pancreatitis: a systematic review and meta-analysis. Gut. 2014;63(5):818–31. 10.1136/gutjnl-2013-305062. [DOI] [PubMed] [Google Scholar]
- 5.Yadav D, O’Connell M, Papachristou GI. Natural history following the first attack of acute pancreatitis. Am J Gastroenterol. 2012;107(7):1096–103. 10.1038/ajg.2012.126. [DOI] [PubMed] [Google Scholar]
- 6.Bertilsson S, Swärd P, Kalaitzakis E. Factors that affect disease progression after first attack of acute pancreatitis. Clin Gastroenterol Hepatol. 2015;13(9):1662-9.e3. 10.1016/j.cgh.2015.04.012. [DOI] [PubMed] [Google Scholar]
- 7.Yang H, Wang L, Shi YH, Sui GT, Wu YF, Lu XQ, et al. Risk factors of acute pancreatitis in the elderly Chinese population: a population-based cross-sectional study. J Dig Dis. 2014;15(9):501–7. 10.1111/1751-2980.12168. [DOI] [PubMed] [Google Scholar]
- 8.Dugum M, Gougol A, Paragomi P, Gao X, Matta B, Yazici C, et al. Association of dietary habits with severity of acute pancreatitis. Curr Dev Nutr. 2018;2(12):nzy075. 10.1093/cdn/nzy075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Setiawan VW, Pandol SJ, Porcel J, Wei PC, Wilkens LR, Le Marchand L, et al. Dietary factors reduce risk of acute pancreatitis in a large multiethnic cohort. Clin Gastroenterol Hepatol. 2017;15(2):257-65.e3. 10.1016/j.cgh.2016.08.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Oskarsson V, Orsini N, Sadr-Azodi O, Wolk A. Fish consumption and risk of non-gallstone-related acute pancreatitis: a prospective cohort study. Am J Clin Nutr. 2015;101(1):72–8. 10.3945/ajcn.113.076174. [DOI] [PubMed] [Google Scholar]
- 11.Oskarsson V, Sadr-Azodi O, Orsini N, Andrén-Sandberg Å, Wolk A. Vegetables, fruit and risk of non-gallstone-related acute pancreatitis: a population-based prospective cohort study. Gut. 2013;62(8):1187–92. 10.1136/gutjnl-2012-302521. [DOI] [PubMed] [Google Scholar]
- 12.Ahmad S, Demler OV, Sun Q, Moorthy MV, Li C, Lee IM, et al. Association of the Mediterranean diet with onset of diabetes in the Women’s Health Study. JAMA Netw Open. 2020;3(11):e2025466. 10.1001/jamanetworkopen.2020.25466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Gu X, Drouin-Chartier J-P, Sacks FM, Hu FB, Rosner B, Willett WC. Red meat intake and risk of type 2 diabetes in a prospective cohort study of United States females and males. Am J Clin Nutr. 2023;118(6):1153–63. 10.1016/j.ajcnut.2023.08.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Würtz AML, Jakobsen MU, Bertoia ML, Hou T, Schmidt EB, Willett WC, et al. Replacing the consumption of red meat with other major dietary protein sources and risk of type 2 diabetes mellitus: a prospective cohort study. Am J Clin Nutr. 2021;113(3):612–21. 10.1093/ajcn/nqaa284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Huang J, Shen Q, Tang W, Ji F, Liu Y, Zhou J, et al. The clinical significance of serum HbA1c to diagnose diabetes mellitus during acute pancreatitis. Expert Rev Gastroenterol Hepatol. 2023;17(4):385–94. 10.1080/17474124.2023.2192477. [DOI] [PubMed] [Google Scholar]
- 16.Mikó A, Farkas N, Garami A, Szabó I, Vincze Á, Veres G, et al. Preexisting diabetes elevates risk of local and systemic complications in acute pancreatitis: systematic review and meta-analysis. Pancreas. 2018;47(8):917–23. 10.1097/mpa.0000000000001122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Girman CJ, Kou TD, Cai B, Alexander CM, O’Neill EA, Williams-Herman DE, et al. Patients with type 2 diabetes mellitus have higher risk for acute pancreatitis compared with those without diabetes. Diabetes Obes Metab. 2010;12(9):766–71. 10.1111/j.1463-1326.2010.01231.x. [DOI] [PubMed] [Google Scholar]
- 18.Lai SW, Muo CH, Liao KF, Sung FC, Chen PC. Risk of acute pancreatitis in type 2 diabetes and risk reduction on anti-diabetic drugs: a population-based cohort study in Taiwan. Am J Gastroenterol. 2011;106(9):1697–704. 10.1038/ajg.2011.155. [DOI] [PubMed] [Google Scholar]
- 19.Pahomeanu MR, Ojog D, Nițu DT, Diaconu I, Nayyerani H, Negreanu L. Acute pancreatitis and type 2 diabetes mellitus: the chicken-egg paradox-a seven-year experience of a large tertiary center. J Clin Med. 2024;13(5). 10.3390/jcm13051213. [DOI] [PMC free article] [PubMed]
- 20.Paragomi P, Papachristou GI, Jeong K, Hinton A, Pothoulakis I, Talukdar R, et al. The relationship between pre-existing diabetes mellitus and the severity of acute pancreatitis: report from a large international registry. Pancreatology. 2022;22(1):85–91. 10.1016/j.pan.2021.10.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Huh JH, Jeon H, Park SM, Choi E, Lee GS, Kim JW, et al. Diabetes mellitus is associated with mortality in acute pancreatitis. J Clin Gastroenterol. 2018;52(2):178–83. 10.1097/mcg.0000000000000783. [DOI] [PubMed] [Google Scholar]
- 22.Han F, Shi XL, Pan JJ, Wu KY, Zhu QT, Yuan CC, Xiao WM, Ding YB, Yu XP, Jiao XP, et al. Elevated serum HbA1c level, rather than previous history of diabetes, predicts the disease severity and clinical outcomes of acute pancreatitis. BMJ Open Diabetes Res Care. 2023;11(1). 10.1136/bmjdrc-2022-003070. [DOI] [PMC free article] [PubMed]
- 23.Hart PA, Bellin MD, Andersen DK, Bradley D, Cruz-Monserrate Z, Forsmark CE, et al. Type 3c (pancreatogenic) diabetes mellitus secondary to chronic pancreatitis and pancreatic cancer. Lancet Gastroenterol Hepatol. 2016;1(3):226–37. 10.1016/s2468-1253(16)30106-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Li X, Kimita W, Cho J, Ko J, Bharmal SH, Petrov MS. Dietary fibre intake in type 2 and new-onset prediabetes/diabetes after acute pancreatitis: a nested cross-sectional study. Nutrients. 2021;13(4). 10.3390/nu13041112. [DOI] [PMC free article] [PubMed]
- 25.Chiuve SE, Fung TT, Rimm EB, Hu FB, McCullough ML, Wang M, et al. Alternative dietary indices both strongly predict risk of chronic disease. J Nutr. 2012;142(6):1009–18. 10.3945/jn.111.157222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sobiecki JG, Imamura F, Davis CR, Sharp SJ, Koulman A, Hodgson JM, et al. A nutritional biomarker score of the Mediterranean diet and incident type 2 diabetes: integrated analysis of data from the MedLey randomised controlled trial and the EPIC-InterAct case-cohort study. PLoS Med. 2023;20(4):e1004221. 10.1371/journal.pmed.1004221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Tison SE, Shikany JM, Long DL, Carson AP, Cofield SS, Pearson KE, et al. Differences in the association of select dietary measures with risk of incident type 2 diabetes. Diabetes Care. 2022;45(11):2602–10. 10.2337/dc22-0217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Tabung FK, Wang W, Fung TT, Hu FB, Smith-Warner SA, Chavarro JE, et al. Development and validation of empirical indices to assess the insulinaemic potential of diet and lifestyle. Br J Nutr. 2016;116(10):1787–98. 10.1017/s0007114516003755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tabung FK, Smith-Warner SA, Chavarro JE, Wu K, Fuchs CS, Hu FB, et al. Development and Validation of an Empirical Dietary Inflammatory Index123. J Nutr. 2016;146(8):1560–70. 10.3945/jn.115.228718. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hart PA, Papachristou GI, Park WG, Dyer A-M, Chinchilli VM, Afghani E, et al. Rationale and design for the Diabetes RElated to Acute Pancreatitis and Its Mechanisms Study: a prospective cohort study from the Type 1 Diabetes in Acute Pancreatitis Consortium. Pancreas. 2022;51(6):568–74. 10.1097/mpa.0000000000002079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Viocare Technologies. VioScreen. https://www.viocare.com/vioscreen-research.html. Accessed 22 May 2023.
- 32.Kristal AR, Kolar AS, Fisher JL, Plascak JJ, Stumbo PJ, Weiss R, et al. Evaluation of web-based, self-administered, graphical food frequency questionnaire. J Acad Nutr Diet. 2014;114(4):613–21. 10.1016/j.jand.2013.11.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Fung TT, Hu FB, McCullough ML, Newby PK, Willett WC, Holmes MD. Diet quality is associated with the risk of estrogen receptor-negative breast cancer in postmenopausal women. J Nutr. 2006;136(2):466–72. 10.1093/jn/136.2.466. [DOI] [PubMed] [Google Scholar]
- 34.Morris MC, Tangney CC, Wang Y, Sacks FM, Barnes LL, Bennett DA, et al. MIND diet slows cognitive decline with aging. Alzheimers Dement. 2015;11(9):1015–22. 10.1016/j.jalz.2015.04.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Shams-White MM, Pannucci TE, Lerman JL, Herrick KA, Zimmer M, Meyers Mathieu K, et al. Healthy Eating Index-2020: review and update process to reflect the Dietary Guidelines for Americans,2020–2025. J Acad Nutr Diet. 2023;123(9):1280–8. 10.1016/j.jand.2023.05.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Petimar J, Smith-Warner SA, Fung TT, Rosner B, Chan AT, Hu FB, et al. Recommendation-based dietary indexes and risk of colorectal cancer in the Nurses’ Health Study and Health Professionals Follow-up Study. Am J Clin Nutr. 2018;108(5):1092–103. 10.1093/ajcn/nqy171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Ahmad S, Moorthy MV, Demler OV, Hu FB, Ridker PM, Chasman DI, et al. Assessment of risk factors and biomarkers associated with risk of cardiovascular disease among women consuming a Mediterranean diet. JAMA Netw Open. 2018;1(8):e185708. 10.1001/jamanetworkopen.2018.5708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Morris MC, Tangney CC, Wang Y, Sacks FM, Bennett DA, Aggarwal NT. MIND diet associated with reduced incidence of Alzheimer’s disease. Alzheimers Dement. 2015;11(9):1007–14. 10.1016/j.jalz.2014.11.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Liu X, Morris MC, Dhana K, Ventrelle J, Johnson K, Bishop L, et al. Mediterranean-DASH intervention for neurodegenerative delay (MIND) study: rationale, design and baseline characteristics of a randomized control trial of the MIND diet on cognitive decline. Contemp Clin Trials. 2021;102:106270. 10.1016/j.cct.2021.106270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Krebs-Smith SM, Pannucci TE, Subar AF, Kirkpatrick SI, Lerman JL, Tooze JA, et al. Update of the Healthy Eating Index: HEI-2015. J Acad Nutr Diet. 2018;118(9):1591–602. 10.1016/j.jand.2018.05.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-Value. Ann Intern Med. 2017;167(4):268–74. 10.7326/m16-2607. [DOI] [PubMed] [Google Scholar]
- 42.Haneuse S, VanderWeele TJ, Arterburn D. Using the E-Value to assess the potential effect of unmeasured confounding in observational studies. JAMA. 2019;321(6):602–3. 10.1001/jama.2018.21554. [DOI] [PubMed] [Google Scholar]
- 43.Ying Z, Fu M, Fang Z, Ye X, Wang P, Lu J. Mediterranean diet lowers risk of new-onset diabetes: a nationwide cohort study in China. Nutr J. 2024;23(1):131. 10.1186/s12937-024-01036-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Sarsangi P, Salehi-Abargouei A, Ebrahimpour-Koujan S, Esmaillzadeh A. Association between adherence to the Mediterranean diet and risk of type 2 diabetes: an updated systematic review and dose-response meta-analysis of prospective cohort studies. Adv Nutr (Bethesda, Md). 2022;13(5):1787–98. 10.1093/advances/nmac046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Song Y, Chang Zg, Song C, Cui K, Shi B, Zhang R, et al. Association between MIND diet adherence and mortality: insights from diabetic and non-diabetic cohorts. Nutr Diabetes. 2023;13(1):18. 10.1038/s41387-023-00247-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Morze J, Danielewicz A, Hoffmann G, Schwingshackl L. Diet quality as assessed by the Healthy Eating Index, Alternate Healthy Eating Index, Dietary Approaches to Stop Hypertension Score, and health outcomes: a second update of a systematic review and meta-analysis of cohort studies. J Acad Nutr Diet. 2020;120(12):1998-2031.e15. 10.1016/j.jand.2020.08.076. [DOI] [PubMed] [Google Scholar]
- 47.Chen GC, Koh WP, Neelakantan N, Yuan JM, Qin LQ, van Dam RM. Diet quality indices and risk of type 2 diabetes mellitus: the Singapore Chinese Health Study. Am J Epidemiol. 2018;187(12):2651–61. 10.1093/aje/kwy183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Willcox JK, Ash SL, Catignani GL. Antioxidants and prevention of chronic disease. Crit Rev Food Sci Nutr. 2004;44(4):275–95. 10.1080/10408690490468489. [DOI] [PubMed] [Google Scholar]
- 49.Liguori I, Russo G, Curcio F, Bulli G, Aran L, Della-Morte D, et al. Oxidative stress, aging, and diseases. Clin Interv Aging. 2018;13:757–72. 10.2147/CIA.S158513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Marshall JA, Bessesen DH. Dietary fat and the development of type 2 diabetes. Diabetes Care. 2002;25(3):620–2. 10.2337/diacare.25.3.620. [DOI] [PubMed] [Google Scholar]
- 51.Rice Bradley BH. Dietary fat and risk for type 2 diabetes: a review of recent research. Curr Nutr Rep. 2018;7(4):214–26. 10.1007/s13668-018-0244-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Männistö S, Kontto J, Kataja-Tuomola M, Albanes D, Virtamo J. High processed meat consumption is a risk factor of type 2 diabetes in the Alpha-Tocopherol, Beta-Carotene Cancer Prevention study. Br J Nutr. 2010;103(12):1817–22. 10.1017/s0007114510000073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Rajpathak SN, Crandall JP, Wylie-Rosett J, Kabat GC, Rohan TE, Hu FB. The role of iron in type 2 diabetes in humans. Biochim Biophys Acta. 2009;1790(7):671–81. 10.1016/j.bbagen.2008.04.005. [DOI] [PubMed] [Google Scholar]
- 54.Peppa M, Goldberg T, Cai W, Rayfield E, Vlassara H. Glycotoxins: a missing link in the “relationship of dietary fat and meat intake in relation to risk of type 2 diabetes in men.” Diabetes Care. 2002;25(10):1898–9. 10.2337/diacare.25.10.1898. [DOI] [PubMed] [Google Scholar]
- 55.White DL, Collinson A. Red meat, dietary heme iron, and risk of type 2 diabetes: the involvement of advanced lipoxidation endproducts. Adv Nutr (Bethesda, Md). 2013;4(4):403–11. 10.3945/an.113.003681. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material 1: Supplemental Figure 1. Dietary patterns and pre-existing diabetes stratified by a history of prior AP. Supplementary Table 1. Alternate Mediterranean Diet (AMED) components and criteria for scoring. Supplementary Table 2. Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) components and criteria for scoring. Supplementary Table 3. Healthy Eating Index (HEI)-2020 components and criteria for scoringa. Supplementary Table 4. Alternate Healthy Eating Index (AHEI)-2010 components and criteria for scoring. Supplemental Table 5.Pearson and Spearman correlation coefficients between diet scores. Supplementary Table 6: Reporting E-values for the association between dietary pattern scores and pre-existing diabetes.
Data Availability Statement
Data described in the manuscript, codebook, and analytic code will not be made available because enrollment in the DREAM study is ongoing across 10 Clinical Centers. Upon completion of the study, data will be shared in accordance with NIH data-sharing policies.