Participants and energy and nutrient intake grouping
This study included blood samples and general data from 376 older residents in 3 rural communities in Ningxia. The age of the participants was 65~89 (72.06±5.95) years (Table 1). The grouping of participants’ energy and nutrient intakes is shown in Table 2. The grouping method is described in supplemental material Table 2.
Table 1 Baseline characteristics of the participants(N=376).
Variable
|
Numbers
|
Frequency (%)
|
Gender
|
Men
|
178
|
47.3
|
Women
|
198
|
52.7
|
Age( , years)
|
72.06±5.95
|
Education
|
Uneducated
|
204
|
54.3
|
Elementary
|
101
|
26.9
|
Intermediate and above
|
71
|
18.8
|
Monthly income
|
Below 1000
|
273
|
72.6
|
1000-2000
|
44
|
11.7
|
2001 and above
|
59
|
15.7
|
Marital status
|
Married with surviving spouse
|
293
|
77.9
|
Unmarried/divorced/widowed
|
83
|
22.1
|
Employment
|
Physical labor
|
297
|
79.0
|
Mental labor
|
51
|
13.6
|
Unemployment
|
28
|
7.4
|
Table 2 Grouping comparison among the participants’ nutrient intake and the recommendation in the Dietary Guidelines
Energy and nutrients
|
Men
|
Women
|
RNI a /AI b
|
Nutrients intake c
|
Occupy RNI%/AI%
|
Group
|
RNI a /AI b
|
Nutrients intake c
|
Occupy RNI%/AI%
|
Group
|
Energy/(kcal/d)
|
1900
|
1753.79±668.54
|
<80,insufficiency
80~110, moderateness
>110, excessiveness
|
1
2
3
|
1500
|
1576.83±586.92
|
<80,insufficiency
80~110, moderateness
>110, excessiveness
|
1
2
3
|
Protein RNI/(g/d)
|
65
|
54.28±27.98
|
55
|
45.78±24.62
|
Fat(%E d)
|
42
|
36.52±3.31
|
33
|
33.14±20.72
|
Carbohydrate(%E d)
|
238
|
224.15±87.39
|
188
|
205.35±76.48
|
Vitamin A (μg RAE e/d)
|
800
|
420.64±198.90
|
<80,insufficiency
80~110, moderateness
>110, excessiveness
|
1
2
3
|
700
|
439.85±212.56
|
<80,insufficiency
80~110, moderateness
>110, excessiveness
|
1
2
3
|
Vitamin C RNI/(mg/d)
|
100
|
75.31±40.71
|
100
|
77.32±34.19
|
Vitamin E AI/(mgα-TE f/d)
|
14
|
11.77±8.30
|
14
|
12.62±8.78
|
Note. RNI a: Recommended intake; AI b: Adequate Intake; c Average intake; d %E: The percentage of total energy; e Retinol activity equivalent (RAE, μg) =Dietary or supplement source all trans retinol(μg) +1/2 Supplement Pure All Trans β- Carotene(μg)+1/12 Dietary all trans β- Carotene(μg) +1/24 Other Dietary Vitamin A Procarotenoids(μg); fα- Tocopherol equivalent(α- TE, mg), total in diet α- TE equivalent (mg)=1 ×α- Tocopherol (mg)+0.5 ×β- Tocopherol (mg)+0.1 × γ – tocopherol (mg)+0.2 ×δ- Tocopherol (mg)+0.3 ×α- Triene tocopherol (mg); *Cholesterol intake should not be excessive, so only distinguishing between moderateness and excessiveness.
The participants of this study are all individuals with light physical activity
Oxidative stress results
The values and reference ranges of oxidative stress biomarkers among the participants are shown in Table 3.
Table 3 Measurement values and reference ranges of oxidative stress biomarkers(N=376).
Biomarkers of oxidative stress
|
Range
|
|
MDA (nmol/mL)
|
1.58~11.29
|
4.82 1.74
|
8-iso-PGF2α(pg/mL)
|
247.59~1788.34
|
782.13±245.82
|
SOD(U/mL)
|
9.21~72.12
|
38.59
|
T-AOC (mmol/L)
|
0.60~1.34
|
1.00±0.15
|
Abbreviations: T-AOC, alternate Total Antioxidant Capacity; 8-iso-PGF2α, 15-F2t-isoprostane 8-isoprostanes-F2α; MDA, Malondialdehyde; SOD, Superoxide Dismutase.
Latent class model selection
The fit statistics of the LCA models are presented in Table 4. The AIC and BIC values are the smallest in class 3, the BIC value begins to increase in class 4, the LMR-LRT and BLRT values support a three-class model, and the model is characterized by sufficient Entropy (0.932). The structure of the three-class model is presented in Fig. 1. Class 1 (C1: n = 141, 37.50% of the total samples) was characterized by an imbalance in the intake of energy and nutrients, and the probability of energy and protein intake was much greater than that of other nutrients. For the purposes of this analysis, the sample was assigned to the ‘imbalanced nutrient—high energy’ class. Class 2 (n = 69, 18.35% of the total samples) was characterized by sufficient and balanced levels of nutrients except for energy and protein. For this reason, the plants were assigned to the ‘sufficient nutrient—low energy and protein’ class. Class 3 (n = 166, 44.15% of the total samples) was characterized by low intake of energy with significant differences in intake of various nutrients and was designated the ‘low nutrient’ class.
Table 4 Results of the LCA – fit statistics(N=376).
Model
(Numbers of latent classes)
|
LL
|
AIC
|
BIC
|
aBIC
|
Entropy
|
LMR LR P-values
|
BLRT P-values
|
1
|
-2269.384
|
4566.768
|
3874.935
|
4577.363
|
-
|
-
|
-
|
2
|
-1851.488
|
3760.977
|
4431.29
|
3782.925
|
0.899
|
<0.001
|
<0.001
|
3
|
-1735.821
|
3559.641
|
3732.543
|
3592.942
|
0.932
|
<0.001
|
<0.001
|
4
|
-1692.886
|
3503.772
|
3735.618
|
3548.426
|
0.951
|
<0.001
|
<0.001
|
5
|
-1664.006
|
3476.011
|
3766.801
|
3532.018
|
0.947
|
0.684
|
<0.001
|
Note. LL = Loglikelihood; AIC = Akaike Information Criterion; BIC = Bayesian Information Criterion; aBIC = Ajusted BIC; LMR LR = Vuong-Lo-Mendell-Rubin Likelihood Ratio Test; BLRT = Parametric Bootstrapped Likelihood Ratio Test. Class chosen is shown in bold.
Oxidative stress across the latent class of energy and nutrient intake
The ANOVA results showed that the concentrations of 8-iso-PGF2α and SOD, which are oxidative stress biomarkers, were significantly different (P < 0.05). In other words, oxidative stress is significantly different among the classes, and the post test results revealed that for 8-iso-PGF2α, class 1 had significantly greater values than class 2 and class 3. For SOD, class 2 had significantly greater values than did class 1 and class 3. The results are shown in Table 5.
Table 5 Comparison of different classes of energy and nutrient intakes with oxidative stress biomarkers. (N = 376, x̄ ±s)
Oxidative stress biomarkers
|
Class
|
P
|
Difference between class
|
C1
|
C2
|
C3
|
MDA(nmol/mL)
|
4.88±1.62
|
4.72±2.03
|
4.82±1.71
|
0.847
|
-
|
8-iso-PGF2α(pg/mL)
|
834.59±280.37
|
721.03±211.65
|
762.96±218.85
|
0.004
|
1>2, 1>3
|
SOD(U/mL)
|
36.82±10.10
|
42.01±11.27
|
38.68±10.65
|
0.004
|
2>1, 2>3
|
T-AOC (mmol/L)
|
1.00±0.14
|
1.00±0.16
|
1.02±0.15
|
0.120
|
-
|
Abbreviations: T-AOC, alternate Total Antioxidant Capacity; 8-iso-PGF2α, 15-F2t-isoprostane