The association between dietary inflammation index and bone mineral density: results from the United States National Health and nutrition examination surveys

Figures & data

Figure 1. Study population.

Table 1. Baseline characteristics of the study population by tertiles of dietary inflammatory index.

		Dietary Inflammatory Index (DII)
	total	Tertile1 < 0.380	Tertile2 0.380–2.286	Tertile3 > 2.286	p-value
N	10312	3437	3437	3438
Age (years)	50.00 (37.00–63.00)	50.00 (38.00–62.00)	50.00 (37.00–63.00)	50.00 (36.00–65.00)	0.22
Gender, n(%)					<.001*
male	5343 (51.81%)	2168 (63.08%)	1783 (51.88%)	1392 (40.49%)
female	4969 (48.19%)	1269 (36.92%)	1654 (48.12%)	2046 (59.51%)
Race/ethnicity, n(%)					<.001*
Mexican American	1957 (18.98%)	687 (19.99%)	661 (19.23%)	609 (17.71%)
Non-Hispanic White	1858 (18.02%)	612 (17.81%)	635 (18.48%)	611 (17.77%)
Non-Hispanic Black	4623 (44.83%)	1645 (47.86%)	1522 (44.28%)	1456 (42.35%)
Other race	1874 (18.17%)	493 (14.34%)	619 (18.01%)	762 (22.16%)
Education level, more than high school, n(%)	5092 (49.44%)	1966 (57.22%)	1723 (50.19%)	1403 (40.90%)	.001*
Income, n(%)					<.001*
Poor	1887 (20.00%)	520 (16.56%)	581 (18.47%)	786 (24.94%)
Near poor	3914 (41.48%)	1171 (37.29%)	1327 (42.19%)	1416 (44.92%)
Not poor	3636 (38.53%)	1449 (46.15%)	1237 (39.33%)	950 (30.14%)
Smoker, n(%)	4575 (45.32%)	1458 (43.33%)	1526 (45.31%)	1591 (47.34%)	.004*
BMI, (kg/m^2)	28.25 ± 5.69	27.85 ± 5.42	28.32 ± 5.72	28.56 ± 5.91	<.001*
Systolic blood pressure (mm Hg)	123.55 ± 18.22	122.82 ± 17.34	123.22 ± 17.57	124.60 ± 19.62	<.001*
Diastolic blood pressure(mm Hg)	68.18 ± 13.91	69.06 ± 13.73	68.21 ± 13.41	67.26 ± 14.51	<.001*
Femoral neck BMD	0.83 ± 0.15	0.84 ± 0.15	0.83 ± 0.15	0.82 ± 0.16	<.001*
Femoral neck T score	−0.30 (−1.20 to 0.60)	−0.20 (−1.10 to 0.60)	−0.30 (−1.10 to 0.60)	−0.40 (−1.20 to 0.50)	<.001*^,**
Trochanter BMD	0.73 ± 0.13	0.75 ± 0.13	0.73 ± 0.13	0.71 ± 0.13	<.001*
Trochanter T score	0.20 (−0.70 to 1.10)	0.30 (−0.60 to 1.30)	0.20 (−0.70 to 1.10)	0.00 (−0.90 to 0.90)	<.001*^,**
Intertrochanter BMD	1.15 ± 0.19	1.17 ± 0.18	1.16 ± 0.19	1.13 ± 0.19	<.001*
Intertrochanter T score	0.40 (−0.50 to 1.30)	0.60 (−0.30 to 1.50)	0.50 (−0.40 to 1.30)	0.30 (−0.60 to 1.20)	<.001*^,**
Total femur BMD	0.97 ± 0.16	0.99 ± 0.16	0.98 ± 0.16	0.95 ± 0.16	<.001*
Total femur T score	0.20 (−0.60 to 1.10)	0.40 (−0.50 to 1.30)	0.30 (−0.60 to 1.20)	0.10 (−0.80 to 1.00)	<.001*^,**
eGFR(ml/min/1.73 m^2)	93.27 (78.38–109.78)	92.66 (78.75–109.04)	93.49 (78.78–109.47)	93.44 (77.53–111.00)	.50
UACR(mg/g)	0.69 (0.44–1.29)	0.62 (0.42–1.15)	0.69 (0.44–1.25)	0.76 (0.48–1.53)	<.001*,**
Serum albumin(g/dL)	42.62 ± 3.14	43.04 ± 3.13	42.67 ± 3.04	42.15 ± 3.18	<.001*
Fast serum glucose (mg/dL)	93.00 (85.00–104.00)	93.00 (86.00–104.00)	93.00 (85.00–104.00)	93.00 (85.00–105.000	.24
Glycohemoglobin (%)	5.50 (5.20–5.90)	5.50 (5.20–5.80)	5.50 (5.20–5.90)	5.50 (5.30–5.90)	<.001*
Serum calcium (mmol/L)	2.35 (2.30–2.42)	2.35 (2.30–2.42)	2.35 (2.30–2.42)	2.35 (2.30–2.42)	.003*
Serum phosphorus (mmol/L)	1.20 (1.10–1.32)	1.20 (1.10–1.32)	1.23 (1.10–1.32)	1.20 (1.07–1.32)	.32
Serum C-reactive protein(mg/dL)	0.17 (0.07–0.39)	0.14 (0.06–0.31)	0.18 (0.07–0.39)	0.21 (0.08–0.48)	<.001*
WBC count (103 cells/uL).	6.80 (5.60–8.20)	6.65 (5.60–8.00)	6.90 (5.70–8.20)	6.90 (5.70–8.40)	<.001*
NLR	1.91 (1.45–2.53)	1.89 (1.45–2.50)	1.92 (1.44–2.50	1.92 (1.45–2.57)	.22
CVD, n(%)	591 (7.79%)	147 (6.07%)	197 (7.73%)	247 (9.43%)	<.001*
Hypertension, n(%)	4139 (40.14%)	1270 (36.95%)	1347 (39.19%)	1522 (44.27%)	<.001*
DM, n(%)	6040 (58.57%)	1983 (57.70%)	1976 (57.49%)	2081 (60.53%)	.017*
Liver condition, n(%)	235 (3.10%)	91 (3.76%)	73 (2.87%)	71 (2.71%)	0.07
Hypercholesterolemia, n(%)	5708 (58.68%)	1895 (58.63%)	1924 (59.42%)	1889 (58.00%)	0.51
Calcium intake (mg)	815.00 (525.75–1185.00)	1128.00 (820.00–1559.00)	818.00 (569.00–1138.00)	541.00 (350.00–803.00)	<.001*
Arthritis, n(%)	2570 (25.51%)	791 (23.55%)	828 (24.61%)	951 (28.36%)	<.001*
Aspirin use, n(%)	161 (2.88%)	38 (2.13%)	52 (2.79%)	71 (3.65%)	.021*
Calcitonin use, n(%)	6 (0.11%)	2 (0.11%)	2 (0.11%)	2 (0.10%)	1.00
Biphosphonate use, n(%)	180 (3.22%)	53 (2.97%)	63 (3.39%)	64 (3.29%)	.76
DMARDs use, n(%)	59 (1.06%)	12 (0.67%)	18 (0.97%)	29 (1.49%)	.046*
Estrogen use, n(%)	87 (1.04%)	24 (0.82%)	26 (0.94%)	37 (1.39%)	.09

Mean ± standard deviation (SD)and ANOVA are for continuous variables with a normal distribution. Median (interquartile range) and Kruskal–Wallis H Test are for continuous variables that are not normally distributed. N and percentage (%) and weighted χ² test are for categorical variables.

DII: dietary inflammation index; BMI: body mass index; BMD: bone mineral density; eGFR: estimated glomerular filtration rate; UACR: urine albumin-creatinine ratio; WBC: white blood cell; NLR: neutrophil to lymphocyte ratio; CVD: cardiovascular disease; DM: diabetes mellitus; DMARDs, disease-modifying antirheumatic drugs.

*p < .05.

**p value was calculated by Fisher’s exact test.

Table 2. The association betwwen DII and femoral BMD.

		Dietary Inflammatory Index (DII)
			Tertile2 0.380–2.286		Tertile3 > 2.286
Region of interest	Model	Tertile1 < 0.380	β (95% CI)	p value	β (95% CI)	p value
Femoral neck	Crude model	Reference	−0.01 (−0.01, 0.00)	.07	−0.02 (−0.03, −0.01)	<.0001*
	Adjusted model 1	Reference	−0.00 (−0.01, 0.00)	.36	−0.01 (−0.02, −0.00)	.0008*
	Adjusted model 2	Reference	−0.01 (−0.02, 0.00)	.15	−0.02 (−0.03, −0.01)	.0036*
Trochanter	Crude model	Reference	−0.02 (−0.02, −0.01)	<.0001*	−0.03 (−0.04, −0.03)	<.0001*
	Adjusted model 1	Reference	−0.01 (−0.01, −0.00)	.006*	−0.02 (−0.02, −0.01)	<.0001*
	Adjusted model 2	Reference	<0.0001*	.0281*	−0.02 (−0.03, −0.01)	.0004*
Intertrochanter	Crude model	Reference	−0.02 (−0.02, −0.01)	.0004*	−0.04 (−0.05, −0.03)	<.0001*
	Adjusted model 1	Reference	−0.00 (−0.01, 0.00)	.30	−0.02 (−0.02, −0.01)	<.0001*
	Adjusted model 2	Reference	−0.01 (−0.02, 0.00)	.16	−0.02 (−0.03, −0.01)	.0043*
Total femur	Crude model	Reference	−0.02 (−0.02, −0.01)	<.0001*	−0.04 (−0.05, −0.03)	<.0001*
	Adjusted model 1	Reference	−0.01 (−0.01, 0.00)	.11	−0.02 (−0.02, −0.01)	<.0001*
	Adjusted model 2	Reference	−0.01 (−0.02, 0.00)	.06	−0.02 (−0.03, −0.01)	.0005*

Adjusted model 1: Adjust for age, gender, race/ethnicity.

Adjusted model 2: Adjust for age, gender, race/ethnicity, smoker, BMI, eGFR, UACR, serum C-reactive protein, WBC count, NLR, serum calcium, arthritis, aspirin use, calcitonin use, biphosphonate use, DMARDs use, calcium intake and estrogen use.

BMD: bone mineral density; DII: dietary inflammation index; OR: odds ratio; CI: confidence interval; BMI: body mass index; eGFR: estimated glomerular filtration rate; UACR: urine albumin-creatinine ratio; WBC: white blood cell; NLR: neutrophil to lymphocyte ratio; DMARDs: disease-modifying antirheumatic drugs.

*p<.05.

Table 3. The association between DII and femoral T score.

		Dietary Inflammatory Index (DII)
			Tertile2 0.380–2.286		Tertile3 > 2.286
Region of interest	Model	Tertile1 < 0.380	β (95% CI)	p value	β (95% CI)	p value
Femoral neck	Non-adjusted	Reference	−0.06 (−0.12, 0.00)	.07	−0.18 (−0.24, −0.12)	<.0001*
	Adjust I	Reference	−0.02 (−0.07, 0.03)	.37	−0.09 (−0.14, −0.04)	.0008*
	Adjust II	Reference	−0.06 (−0.14, 0.02)	.15	−0.13 (−0.22, −0.04)	.0042*
Trochanter	Non-adjusted	Reference	−0.16 (−0.23, −0.10)	<.0001*	−0.34 (−0.41, −0.28)	<.0001*
	Adjust I	Reference	−0.08 (−0.14, −0.02)	.0065*	−0.17 (−0.23, −0.11)	<.0001*
	Adjust II	Reference	−0.11 (−0.20, −0.01)	.0312*	−0.19 (−0.29, −0.08)	.0004*
Intertrochanter	Non-adjusted	Reference	−0.11 (−0.17, −0.05)	.0004*	−0.30 (−0.36, −0.23)	<.0001*
	Adjust I	Reference	−0.03 (−0.08, 0.03)	.30	−0.11 (−0.17, −0.06)	<.0001*
	Adjust II	Reference	−0.06 (−0.15, 0.02)	.16	−0.14 (−0.24, −0.05)	.0038*
Total femur	Non-adjusted	Reference	−0.13 (−0.19, −0.06)	<.0001*	−0.31 (−0.38, −0.25)	<.0001*
	Adjust I	Reference	−0.04 (−0.10, 0.01)	.11	−0.13 (−0.19, −0.08)	<.0001*
	Adjust II	Reference	−0.08 (−0.17, 0.00)	.06	−0.17 (−0.26, −0.07)	.0004

Adjusted model 1: Adjust for age, gender, race/ethnicity.

Adjusted model 2: Adjust for age, gender, race/ethnicity, smoker, BMI, eGFR, UACR, serum C-reactive protein, WBC count, NLR, serum calcium, arthritis, aspirin use, calcitonin use, biphosphonate use, DMARDs use, calcium intake and estrogen use.

DII: dietary inflammation index; CI: confidence interval; BMI: body mass index; eGFR: estimated glomerular filtration rate; UACR: urine albumin-creatinine ratio; WBC: white blood cell; NLR: neutrophil to lymphocyte ratio; DMARDs: disease-modifying antirheumatic drugs.

*p<.05.

Table 4. The association between DII and the possibility of the presence of osteoporosis.

		Dietary Inflammatory Index (DII)
			Tertile2 0.380–2.286		Tertile3 > 2.286
Region of interest	Model	Tertile1 < 0.380	OR (95% CI)	p value	OR (95% CI)	p value
Femoral neck	Non-adjusted	Reference	1.20 (0.89, 1.62)	.22	1.81 (1.37, 2.39)	<.0001*
	Adjust I	Reference	1.06 (0.77, 1.45)	.72	1.31 (0.97, 1.75)	.0774
	Adjust II	Reference	1.34 (0.80, 2.26)	.27	1.88 (1.11, 3.20)	.019*
Trochanter	Non-adjusted	Reference	1.82 (1.17, 2.83)	.0078*	2.58 (1.70, 3.93)	<.0001*
	Adjust I	Reference	1.55 (0.98, 2.43)	.06	1.65 (1.07, 2.55)	.0222*
	Adjust II	Reference	1.49 (0.80, 2.81)	.21	1.62 (0.85, 3.09)	.1457
Intertrochanter	Non-adjusted	Reference	1.41 (0.86, 2.32)	.17	2.62 (1.68, 4.10)	<.0001*
	Adjust I	Reference	1.20 (0.72, 1.99)	.48	1.70 (1.07, 2.70)	.0245
	Adjust II	Reference	1.06 (0.52, 2.18)	.87	2.10 (1.05, 4.20)	.0365*
Total femur	Non-adjusted	Reference	1.22 (0.79, 1.89)	.37	2.30 (1.56, 3.40)	<.0001
	Adjust I	Reference	1.02 (0.65, 1.60)	.94	1.47 (0.98, 2.20)	.065
	Adjust II	Reference	1.20 (0.62, 2.30)	.59	1.94 (1.02, 3.69)	.0428*

Adjusted model 1: Adjust for age, gender, race/ethnicity.

Adjusted model 2: Adjust for age, gender, race/ethnicity, smoker, BMI, eGFR, UACR, serum C-reactive protein, WBC count, NLR, serum calcium, arthritis, aspirin use, calcitonin use, biphosphonate use, DMARDs use, calcium intake and estrogen use.

DII: dietary inflammation index; OR: odds ratio; CI: confidence interval; BMI: body mass index; eGFR: estimated glomerular filtration rate; UACR: urine albumin-creatinine ratio; WBC: white blood cell; NLR: neutrophil to lymphocyte ratio; DMARDs: disease-modifying antirheumatic drugs.

*p<.05.

Table 5. Subgroup analysis of the association between DII and the possibility of the presence of osteoporosis.

			Dietary Inflammatory Index (DII)
Region of		Sample	Tertile1	Tertile2 0.380–2.286		Tertile3 > 2.286		p for
interest	Subgroup	size	< 0.380	OR (95% CI)	p value	OR (95% CI)	p value	interaction
Femoral neck	Age (year)							.248
	<65		Reference	0.62 (0.22, 1.73)	.3631	0.99 (0.37, 2.69)	.9854
	≥65		Reference	1.65 (0.90, 3.03)	.1072	2.20 (1.18, 4.09)	.013*
	Gender							.1458
	Male^a	5343	Reference	1.34 (0.56, 3.21)	.5179	0.90 (0.33, 2.49)	.843
	Female	4969	Reference	1.37 (0.71, 2.64)	.3452	2.31 (1.21, 4.42)	.0113*
	Race/ethnicity							.9379
	Non-Hispanic White	4623	Reference	1.34 (0.68, 2.66)	.3965	2.01 (1.00, 4.03)	.0486*
	Others	5689	Reference	1.38 (0.60, 3.17)	.4433	1.76 (0.76, 4.07)	.1832
	BMI(kg/m^2)							.0491*
	<25	3117	Reference	0.92 (0.41, 2.04)	.8354	2.28 (1.06, 4.88)	.0345*
	≥25	7164	Reference	1.73 (0.86, 3.49)	.1231	1.42 (0.68, 2.99)	.3486
	Calcium intake (mg)							.0749
	<815	5145	Reference	0.59 (0.25, 1.40)	.233	1.18 (0.56, 2.49)	.6575
	≥815	5167	Reference	2.12 (1.11, 4.05)	.0236*	2.62 (1.26, 5.47)	.0101*
	eGFR(mL/min/1.73m^2)							.4562
	<60	668	Reference	2.79 (0.65, 11.93)	.1669	3.84 (0.90, 16.32)	.0688
	≥60	9644	Reference	1.19 (0.67, 2.10)	.5539	1.67 (0.93, 2.99)	.0843
Trochanter	Age (year)							.0425
	<65	8005	Reference	0.43 (0.13, 1.40)	.1599	0.63 (0.21, 1.93)	.4224
	≥65	2307	Reference	2.58 (1.16, 5.72)	.02*	2.56 (1.13, 5.82)	.0245
	Gender							.2931
	Male^a	5343	Reference	1.22 (0.30, 5.02)	.7801	0.44 (0.06, 3.07)	.4083
	Female	4969	Reference	1.70 (0.83, 3.47)	.1437	2.00 (0.97, 4.12)	.06
	Race/ethnicity							.0425*
	Non-Hispanic White	4623	Reference	1.48 (0.53, 4.16)	.454	3.22 (1.18, 8.79)	.0225*
	Others	5689	Reference	1.54 (0.68, 3.49)	.302	0.95 (0.39, 2.27)	.9004
	BMI(kg/m^2)							.9582
	<25	3117	Reference	1.53 (0.63, 3.74)	.3475	1.67 (0.67, 4.18)	.2746
	≥25	7164	Reference	1.27 (0.51, 3.16)	.6071	1.48 (0.59, 3.74)	.4028
	Calcium intake (mg)							.6698
	<815	5145	Reference	1.06 (0.41, 2.73)	.9038	1.21 (0.52, 2.85)	.6545
	≥815	5167	Reference	1.90 (0.81, 4.45)	.1387	1.81 (0.68, 4.82)	.2318
	eGFR(mL/min/1.73m^2)							.9497
	<60	668	Reference	1.36 (0.26, 6.99)	.7153	1.89 (0.37, 9.61)	.4438
	≥60	9644	Reference	1.49 (0.75, 2.96)	.2517	1.48 (0.72, 3.02)	.2835
Intertrochanter	Age (year)							.1481
	<65	8005	Reference	0.35 (0.07, 1.69)	.1899	0.93 (0.23, 3.76)	.9215
	≥65	2307	Reference	1.36 (0.60, 3.10)	.46	2.46 (1.10, 5.52)	.0285*
	Gender							.2987
	Male^a	5343	Reference	0.77 (0.18, 3.34)	.7263	0.46 (0.06, 3.81)	.4741
	Female	4969	Reference	1.20 (0.52, 2.76)	.6756	2.95 (1.32, 6.60)	.0086*
	Race/ethnicity							.1481
	Non-Hispanic White	4623	Reference	0.67 (0.25, 1.80)	.4293	2.60 (1.06, 6.39)	.0375*
	Others	5689	Reference	1.78 (0.57, 5.58)	.3256	1.97 (0.61, 6.36)	.2552
	BMI(kg/m^2)							.2776
	<25	3117	Reference	0.66 (0.28, 1.57)	.3475	1.59 (0.70, 3.62)	.2726
	≥25	7164	Reference	2.43 (0.60, 9.86)	.2157	3.66 (0.93, 14.49)	.0642
	Calcium intake (mg)							.4335
	<815	5145	Reference	0.73 (0.21, 2.45)	.6064	1.90 (0.68, 5.28)	.2186
	≥815	5167	Reference	1.44 (0.58, 3.57)	.4332	1.60 (0.57, 4.50)	.3744
	eGFR(mL/min/1.73m^2)							.9386
	<60	668	Reference	0.91 (0.17, 4.96)	.9093	2.33 (0.44, 12.23)	.3170
	≥60	9644	Reference	1.07 (0.49, 2.37)	.86	1.94 (0.89, 4.24)	.0956
Total femur	Age (year)							.5776
	<65	8005	Reference	0.37 (0.11, 1.32)	.1266	0.75 (0.23, 2.41)	.6266
	≥65	2307	Reference	1.78 (0.81, 3.91)	.1506	2.86 (1.30, 6.32)	.0092*
	Gender							.1863
	Male^a	5343	Reference	0.52 (0.09, 3.07)	.4732	0.27 (0.02, 3.17)	.2969
	Female	4969	Reference	1.48 (0.72, 3.05)	.2823	2.49 (1.22, 5.09)	.0123
	Race/ethnicity							.5776
	Non-Hispanic White	4623	Reference	0.95 (0.39, 2.36)	.9167	2.24 (0.95, 5.30)	.0668
	Others	5689	Reference	1.53 (0.57, 4.11)	.4005	2.00 (0.74, 5.44)	.1748
	BMI(kg/m^2)							.5712
	<25	3117	Reference	0.84 (0.37, 1.91)	.6705	1.60 (0.72, 3.57)	.2508
	≥25	7164	Reference	1.73 (0.58, 5.16)	.323	2.48 (0.84, 7.36)	.1005
	Calcium intake (mg)							.3791
	<815	5145	Reference	0.63 (0.22, 1.83)	.3966	1.27 (0.52, 3.11)	.6014
	≥815	5167	Reference	1.64 (0.71, 3.75)	.2439	1.94 (0.77, 4.86)	.1591
	eGFR(mL/min/1.73m^2)							.8526
	<60	668	Reference	1.38 (0.27, 7.06)	.6979	2.09 (0.40, 10.88)	.3791
	≥60	9644	Reference	1.10 (0.54, 2.25)	.7968	1.85 (0.91, 3.75)	.087

Adjust for age, gender, race/ethnicity, smoker, BMI, eGFR, UACR, serum C-reactive protein, WBC count, NLR, serum calcium, arthritis, aspirin use, calcitonin use, biphosphonate use, DMARDs use, calcium intake and estrogen use, except for the variate for subgroup analysis.

DII: dietary inflammation index; OR: odds ratio; CI: confidence interval; BMI: body mass index; eGFR: estimated glomerular filtration rate; UACR: urine albumin-creatinine ratio; WBC: white blood cell; NLR: neutrophil to lyphocyte ratio; DMARDs: disease-modifying antirheumatic drugs.

^aMale was not adjusted with estrogen use.

*p<.05.

Data availability statement

All data used in the study are publicly available online. (https://wwwn.cdc.gov/nchs/nhanes/Default.aspx and https://www.cdc.gov/nchs/data-linkage/mortality-public.htm).