Results of CA using NDNS RP (food group and timeslots)
Chaochen Wang
8/7/2019
1 Pre-CA analyses
1.1 Summary statistics
In this section, we look at some of the summary statistics for our NDNS RP (9 year) dataset in adults (age >= 19 y.o.). After excluding food recordings from young participants, there were 749,026 recordings of food entry written by the 6802 pariticipants.
1.1.1 numbers of food entries by time slots (7 or 3 groups)
load("../Food1_9_adlt_labl.Rdata")
Food1_9_adlt %>%
ungroup() %>%
group_by(MealTimeSlot) %>%
summarise(n = n()) %>%
mutate(relfreq = n/sum(n)) %>%
mutate(rel.freq = paste0(round(100 * n/sum(n), 3), "%")) %>%
mutate(cum.freq = paste0(round(100 * cumsum(relfreq), 3), "%")) %>%
print(n=Inf)## # A tibble: 7 x 5
## MealTimeSlot n relfreq rel.freq cum.freq
## <fct> <int> <dbl> <chr> <chr>
## 1 6am to 8:59am 107144 0.143 14.304% 14.304%
## 2 9am to 11:59am 110614 0.148 14.768% 29.072%
## 3 12 noon to 1:59pm 138183 0.184 18.448% 47.521%
## 4 2pm to 4:59pm 94606 0.126 12.631% 60.151%
## 5 5pm to 7:59pm 180498 0.241 24.098% 84.249%
## 6 8pm to 9:59pm 81716 0.109 10.91% 95.158%
## 7 10pm to 5:59am 36265 0.0484 4.842% 100%Food1_9_adlt %>%
ungroup() %>%
group_by(Time3g) %>%
summarise(n = n()) %>%
mutate(relfreq = n/sum(n)) %>%
mutate(rel.freq = paste0(round(100 * n/sum(n), 3), "%")) %>%
mutate(cum.freq = paste0(round(100 * cumsum(relfreq), 3), "%")) %>%
print(n=Inf)## # A tibble: 3 x 5
## Time3g n relfreq rel.freq cum.freq
## <fct> <int> <dbl> <chr> <chr>
## 1 Morning 217758 0.291 29.072% 29.072%
## 2 Afternoon 413287 0.552 55.177% 84.249%
## 3 Evening 117981 0.158 15.751% 100%1.1.2 Days of week collected in the dataset
Food1_9_adlt %>%
mutate(DayofWeek = factor(DayofWeek, levels = c("Monday",
"Tuesday",
"Wednesday",
"Thursday",
"Friday",
"Saturday",
"Sunday"))) %>%
ungroup() %>%
group_by(DayofWeek) %>%
summarise(n = n()) %>%
mutate(relfreq = n/sum(n)) %>%
mutate(rel.freq = paste0(round(100 * n/sum(n), 3), "%")) %>%
mutate(cum.freq = paste0(round(100 * cumsum(relfreq), 3), "%")) %>%
print(n=Inf)## # A tibble: 7 x 5
## DayofWeek n relfreq rel.freq cum.freq
## <fct> <int> <dbl> <chr> <chr>
## 1 Monday 104922 0.140 14.008% 14.008%
## 2 Tuesday 102039 0.136 13.623% 27.631%
## 3 Wednesday 100621 0.134 13.434% 41.064%
## 4 Thursday 104898 0.140 14.005% 55.069%
## 5 Friday 112669 0.150 15.042% 70.111%
## 6 Saturday 111405 0.149 14.873% 84.984%
## 7 Sunday 112472 0.150 15.016% 100%1.1.3 Sex and age
The participants were:
- 2810 men (sex = 1) and
- 3992 women (sex = 2) aged 19 or above.
load("../Indiv1_9_adlt.Rdata")
Indiv1_9_adlt$Sex <- factor(Indiv1_9_adlt$Sex)
Indiv1_9_adlt %>%
group_by(Sex) %>%
summarise(Meanage = mean(age),
SDage = sd(age))## # A tibble: 2 x 3
## Sex Meanage SDage
## <fct> <dbl> <dbl>
## 1 1 50.4 17.2
## 2 2 49.6 17.8ggplot(Indiv1_9_adlt, aes(x=age, fill = Sex, color = Sex)) +
geom_histogram(aes(y=..density..), colour="black", fill="white", binwidth = 2)+
geom_density(alpha=.2) 
1.1.4 Diabetes status
In all 9 years data combined, there were:
- 227 diabetes patients (DM4cat = 3);
- 99 undiagnosed diabetes (DM4cat = 2);
- 133 prediabetes (DM4cat = 1);
- 2626 non-diabetes participants (DM4cat = 0);
- 3717 missing values.
load("../Indiv1_9_adlt.Rdata")
Indiv1_9_adlt %>%
group_by(DM4cat) %>%
summarise(n = n()) %>%
mutate(rel.freq = paste0(round(100 * n/sum(n), 2), "%")) %>%
print(n=Inf)## # A tibble: 5 x 3
## DM4cat n rel.freq
## <dbl> <int> <chr>
## 1 0 2626 38.61%
## 2 1 133 1.96%
## 3 2 99 1.46%
## 4 3 227 3.34%
## 5 NA 3717 54.65%1.2 Food groups
All 60 food groups sorted by the percentage contributed to the total calories consumed by participants. We can see 13 food groups contributed 50% of total energy that these pariticipants consumed, 28 food groups contributed 80% of total energy that these people consumed.
TableFoogGroup <- Food1_9_adlt %>%
ungroup() %>%
group_by(mfgLab) %>%
summarise(n = n(), meanHpoint = mean(H_points, na.rm = T), mfgCalories = sum(Energykcal)) %>%
arrange(-mfgCalories) %>%
mutate(n.freq = paste0(round(100 * n/sum(n), 2), "%")) %>%
mutate(cal.Prop = paste0(round(100 * mfgCalories/sum(mfgCalories), 2), "%")) %>%
mutate(calprop = mfgCalories/sum(mfgCalories)) %>%
mutate(calcumprop = paste0(round(100 * cumsum(calprop), 3), "%")) %>%
select(-calprop) %>%
print(n=Inf)## # A tibble: 60 x 7
## mfgLab n meanHpoint mfgCalories n.freq cal.Prop calcumprop
## <chr> <int> <dbl> <dbl> <chr> <chr> <chr>
## 1 Pasta & Rice 18353 -0.301 3512070. 2.45% 7.36% 7.362%
## 2 White Bread 18434 3.61 3245641. 2.46% 6.8% 14.166%
## 3 Chips 6749 0.0935 1884059. 0.9% 3.95% 18.116%
## 4 Cakes & Pastri… 7806 17.3 1710594. 1.04% 3.59% 21.701%
## 5 Veg not raw 51317 -7.03 1665474. 6.85% 3.49% 25.193%
## 6 Biscuits 13200 18.3 1662598. 1.76% 3.49% 28.678%
## 7 Fruit 33903 -2.81 1641675. 4.53% 3.44% 32.12%
## 8 Misc./Vending 48597 9.49 1639025. 6.49% 3.44% 35.555%
## 9 Chicken/turkey 8863 0.534 1617820. 1.18% 3.39% 38.947%
## 10 Cheese 10983 21.5 1492015. 1.47% 3.13% 42.074%
## 11 Beer lager 8199 0.990 1484001. 1.09% 3.11% 45.185%
## 12 2% milk 57611 1.02 1302650. 7.69% 2.73% 47.916%
## 13 Potatos other 10113 -1.49 1291448. 1.35% 2.71% 50.623%
## 14 JamsSpreads 37960 14.7 1215279. 5.07% 2.55% 53.171%
## 15 Beef 4987 2.34 1124560. 0.67% 2.36% 55.528%
## 16 HiFi cereals 8215 2.78 1072814. 1.1% 2.25% 57.777%
## 17 WMeal Bread 7193 2.08 1070696. 0.96% 2.24% 60.022%
## 18 Chocolate 6495 24.9 1046113. 0.87% 2.19% 62.215%
## 19 Wine 6967 1.33 1027793. 0.93% 2.15% 64.369%
## 20 Brown Bread 6183 1.11 1009075. 0.83% 2.12% 66.484%
## 21 Butter 10203 24.3 965901. 1.36% 2.02% 68.509%
## 22 Eggs 7554 3.70 964769. 1.01% 2.02% 70.532%
## 23 Reg soft drinks 11387 3.31 940517. 1.52% 1.97% 72.503%
## 24 Spreads less-f… 12620 23.4 848835. 1.68% 1.78% 74.283%
## 25 Crisps 5664 13.0 835672. 0.76% 1.75% 76.035%
## 26 Sausages 3025 15.3 775004. 0.4% 1.62% 77.659%
## 27 Meat pastries 1979 14.8 744640. 0.26% 1.56% 79.22%
## 28 Bacon and ham 8467 15.5 738727. 1.13% 1.55% 80.769%
## 29 Yogurt 6776 3.22 665485. 0.9% 1.4% 82.164%
## 30 LoFi cereals 4303 11.6 560296. 0.57% 1.17% 83.338%
## 31 Nuts and seeds 6259 4.37 559874. 0.84% 1.17% 84.512%
## 32 Oily fish 2610 6.21 550425. 0.35% 1.15% 85.666%
## 33 Whole Milk 13628 2.24 530449. 1.82% 1.11% 86.778%
## 34 White fish, sh… 1597 3.99 498929. 0.21% 1.05% 87.824%
## 35 Puddings 2291 8.40 459785. 0.31% 0.96% 88.788%
## 36 Other Milk Cre… 6605 11.6 434239. 0.88% 0.91% 89.698%
## 37 Pork 1832 3.88 420504. 0.24% 0.88% 90.579%
## 38 Fruit juice 6960 1.79 419867. 0.93% 0.88% 91.459%
## 39 Margarine 8742 19.6 410108. 1.17% 0.86% 92.319%
## 40 Coated Chicken 1170 6.43 359773. 0.16% 0.75% 93.073%
## 41 Other white fi… 3703 1.28 355118. 0.49% 0.74% 93.818%
## 42 Lamb 1251 6.42 342289. 0.17% 0.72% 94.535%
## 43 Burgers/kebabs 939 14.6 328396. 0.13% 0.69% 95.224%
## 44 Ice cream 1816 15.4 309564. 0.24% 0.65% 95.873%
## 45 Salad and raw … 33263 -7.03 297085. 4.44% 0.62% 96.495%
## 46 Spirits and li… 2698 4.03 286589. 0.36% 0.6% 97.096%
## 47 Other meat 1571 17.0 271445. 0.21% 0.57% 97.665%
## 48 Tea/Coffee/Wat… 152108 -0.245 175659. 20.31% 0.37% 98.033%
## 49 Skimmed Milk 9683 -0.545 168679. 1.29% 0.35% 98.387%
## 50 Sugar confecti… 1820 14.5 164392. 0.24% 0.34% 98.732%
## 51 Oth Bread 932 4.96 149552. 0.12% 0.31% 99.045%
## 52 LowFat Spreads 3654 18.1 134527. 0.49% 0.28% 99.327%
## 53 Polyunsatu mar… 3179 20.0 126714. 0.42% 0.27% 99.593%
## 54 Liver 454 13.6 67330. 0.06% 0.14% 99.734%
## 55 Dietary supple… 13016 5.16 44916. 1.74% 0.09% 99.828%
## 56 Diet soft drin… 14155 0.321 32855. 1.89% 0.07% 99.897%
## 57 1% milk 1263 0.973 27534. 0.17% 0.06% 99.955%
## 58 Smoothies 137 2.93 17061. 0.02% 0.04% 99.99%
## 59 Artificial Swe… 7518 1.60 2689. 1% 0.01% 99.996%
## 60 Commercial tod… 66 18.5 1862. 0.01% 0% 100%1.2.1 Healthy point definition by (Pechey et al, 2013)
1.2.1.1 Healthy foods (n = 3) meanHpoint < -2
TableFoogGroup <- TableFoogGroup %>%
mutate(healthy = meanHpoint < -2,
lesshealthy = meanHpoint > 4,
neutral = (meanHpoint <= 4) & (meanHpoint >= -2))
TableFoogGroup %>%
filter(healthy) %>%
select(-lesshealthy, -neutral,-calcumprop)## # A tibble: 3 x 7
## mfgLab n meanHpoint mfgCalories n.freq cal.Prop healthy
## <chr> <int> <dbl> <dbl> <chr> <chr> <lgl>
## 1 Veg not raw 51317 -7.03 1665474. 6.85% 3.49% TRUE
## 2 Fruit 33903 -2.81 1641675. 4.53% 3.44% TRUE
## 3 Salad and raw veg 33263 -7.03 297085. 4.44% 0.62% TRUE1.2.1.2 Neutral foods (n = 26); meanHpoint between (-2, 4)
TableFoogGroup %>%
filter(neutral) %>%
select(-lesshealthy, -healthy, -calcumprop) %>%
print(n=Inf)## # A tibble: 26 x 7
## mfgLab n meanHpoint mfgCalories n.freq cal.Prop neutral
## <chr> <int> <dbl> <dbl> <chr> <chr> <lgl>
## 1 Pasta & Rice 18353 -0.301 3512070. 2.45% 7.36% TRUE
## 2 White Bread 18434 3.61 3245641. 2.46% 6.8% TRUE
## 3 Chips 6749 0.0935 1884059. 0.9% 3.95% TRUE
## 4 Chicken/turkey 8863 0.534 1617820. 1.18% 3.39% TRUE
## 5 Beer lager 8199 0.990 1484001. 1.09% 3.11% TRUE
## 6 2% milk 57611 1.02 1302650. 7.69% 2.73% TRUE
## 7 Potatos other 10113 -1.49 1291448. 1.35% 2.71% TRUE
## 8 Beef 4987 2.34 1124560. 0.67% 2.36% TRUE
## 9 HiFi cereals 8215 2.78 1072814. 1.1% 2.25% TRUE
## 10 WMeal Bread 7193 2.08 1070696. 0.96% 2.24% TRUE
## 11 Wine 6967 1.33 1027793. 0.93% 2.15% TRUE
## 12 Brown Bread 6183 1.11 1009075. 0.83% 2.12% TRUE
## 13 Eggs 7554 3.70 964769. 1.01% 2.02% TRUE
## 14 Reg soft drinks 11387 3.31 940517. 1.52% 1.97% TRUE
## 15 Yogurt 6776 3.22 665485. 0.9% 1.4% TRUE
## 16 Whole Milk 13628 2.24 530449. 1.82% 1.11% TRUE
## 17 White fish, shell… 1597 3.99 498929. 0.21% 1.05% TRUE
## 18 Pork 1832 3.88 420504. 0.24% 0.88% TRUE
## 19 Fruit juice 6960 1.79 419867. 0.93% 0.88% TRUE
## 20 Other white fish 3703 1.28 355118. 0.49% 0.74% TRUE
## 21 Tea/Coffee/Water 152108 -0.245 175659. 20.31% 0.37% TRUE
## 22 Skimmed Milk 9683 -0.545 168679. 1.29% 0.35% TRUE
## 23 Diet soft drinks 14155 0.321 32855. 1.89% 0.07% TRUE
## 24 1% milk 1263 0.973 27534. 0.17% 0.06% TRUE
## 25 Smoothies 137 2.93 17061. 0.02% 0.04% TRUE
## 26 Artificial Sweete… 7518 1.60 2689. 1% 0.01% TRUE1.2.1.3 Less Healthy foods (n = 31); meanHpoint > 4
TableFoogGroup %>%
filter(lesshealthy) %>%
select(-neutral, -healthy, -calcumprop) %>%
print(n=Inf)## # A tibble: 31 x 7
## mfgLab n meanHpoint mfgCalories n.freq cal.Prop lesshealthy
## <chr> <int> <dbl> <dbl> <chr> <chr> <lgl>
## 1 Cakes & Pastri… 7806 17.3 1710594. 1.04% 3.59% TRUE
## 2 Biscuits 13200 18.3 1662598. 1.76% 3.49% TRUE
## 3 Misc./Vending 48597 9.49 1639025. 6.49% 3.44% TRUE
## 4 Cheese 10983 21.5 1492015. 1.47% 3.13% TRUE
## 5 JamsSpreads 37960 14.7 1215279. 5.07% 2.55% TRUE
## 6 Chocolate 6495 24.9 1046113. 0.87% 2.19% TRUE
## 7 Butter 10203 24.3 965901. 1.36% 2.02% TRUE
## 8 Spreads less-f… 12620 23.4 848835. 1.68% 1.78% TRUE
## 9 Crisps 5664 13.0 835672. 0.76% 1.75% TRUE
## 10 Sausages 3025 15.3 775004. 0.4% 1.62% TRUE
## 11 Meat pastries 1979 14.8 744640. 0.26% 1.56% TRUE
## 12 Bacon and ham 8467 15.5 738727. 1.13% 1.55% TRUE
## 13 LoFi cereals 4303 11.6 560296. 0.57% 1.17% TRUE
## 14 Nuts and seeds 6259 4.37 559874. 0.84% 1.17% TRUE
## 15 Oily fish 2610 6.21 550425. 0.35% 1.15% TRUE
## 16 Puddings 2291 8.40 459785. 0.31% 0.96% TRUE
## 17 Other Milk Cre… 6605 11.6 434239. 0.88% 0.91% TRUE
## 18 Margarine 8742 19.6 410108. 1.17% 0.86% TRUE
## 19 Coated Chicken 1170 6.43 359773. 0.16% 0.75% TRUE
## 20 Lamb 1251 6.42 342289. 0.17% 0.72% TRUE
## 21 Burgers/kebabs 939 14.6 328396. 0.13% 0.69% TRUE
## 22 Ice cream 1816 15.4 309564. 0.24% 0.65% TRUE
## 23 Spirits and li… 2698 4.03 286589. 0.36% 0.6% TRUE
## 24 Other meat 1571 17.0 271445. 0.21% 0.57% TRUE
## 25 Sugar confecti… 1820 14.5 164392. 0.24% 0.34% TRUE
## 26 Oth Bread 932 4.96 149552. 0.12% 0.31% TRUE
## 27 LowFat Spreads 3654 18.1 134527. 0.49% 0.28% TRUE
## 28 Polyunsatu mar… 3179 20.0 126714. 0.42% 0.27% TRUE
## 29 Liver 454 13.6 67330. 0.06% 0.14% TRUE
## 30 Dietary supple… 13016 5.16 44916. 1.74% 0.09% TRUE
## 31 Commercial tod… 66 18.5 1862. 0.01% 0% TRUE1.2.2 Healthy point definition by tertiles of meanHpoint
TableFoogGroup <- TableFoogGroup %>%
mutate(HealthPoints3g = ntile(meanHpoint, 3))1.2.2.1 Healthy foods Group 1 (n = 20)
TableFoogGroup %>%
filter(HealthPoints3g == 1) %>%
select(-lesshealthy, -neutral,-calcumprop, -healthy, -HealthPoints3g) %>%
print(n=Inf)## # A tibble: 20 x 6
## mfgLab n meanHpoint mfgCalories n.freq cal.Prop
## <chr> <int> <dbl> <dbl> <chr> <chr>
## 1 Pasta & Rice 18353 -0.301 3512070. 2.45% 7.36%
## 2 Chips 6749 0.0935 1884059. 0.9% 3.95%
## 3 Veg not raw 51317 -7.03 1665474. 6.85% 3.49%
## 4 Fruit 33903 -2.81 1641675. 4.53% 3.44%
## 5 Chicken/turkey 8863 0.534 1617820. 1.18% 3.39%
## 6 Beer lager 8199 0.990 1484001. 1.09% 3.11%
## 7 2% milk 57611 1.02 1302650. 7.69% 2.73%
## 8 Potatos other 10113 -1.49 1291448. 1.35% 2.71%
## 9 WMeal Bread 7193 2.08 1070696. 0.96% 2.24%
## 10 Wine 6967 1.33 1027793. 0.93% 2.15%
## 11 Brown Bread 6183 1.11 1009075. 0.83% 2.12%
## 12 Whole Milk 13628 2.24 530449. 1.82% 1.11%
## 13 Fruit juice 6960 1.79 419867. 0.93% 0.88%
## 14 Other white fish 3703 1.28 355118. 0.49% 0.74%
## 15 Salad and raw veg 33263 -7.03 297085. 4.44% 0.62%
## 16 Tea/Coffee/Water 152108 -0.245 175659. 20.31% 0.37%
## 17 Skimmed Milk 9683 -0.545 168679. 1.29% 0.35%
## 18 Diet soft drinks 14155 0.321 32855. 1.89% 0.07%
## 19 1% milk 1263 0.973 27534. 0.17% 0.06%
## 20 Artificial Sweeteners 7518 1.60 2689. 1% 0.01%1.2.2.2 Healthy foods Group 2 (n = 20)
TableFoogGroup %>%
filter(HealthPoints3g == 2) %>%
select(-lesshealthy, -neutral,-calcumprop, -healthy, -HealthPoints3g)## # A tibble: 20 x 6
## mfgLab n meanHpoint mfgCalories n.freq cal.Prop
## <chr> <int> <dbl> <dbl> <chr> <chr>
## 1 White Bread 18434 3.61 3245641. 2.46% 6.8%
## 2 Misc./Vending 48597 9.49 1639025. 6.49% 3.44%
## 3 Beef 4987 2.34 1124560. 0.67% 2.36%
## 4 HiFi cereals 8215 2.78 1072814. 1.1% 2.25%
## 5 Eggs 7554 3.70 964769. 1.01% 2.02%
## 6 Reg soft drinks 11387 3.31 940517. 1.52% 1.97%
## 7 Yogurt 6776 3.22 665485. 0.9% 1.4%
## 8 LoFi cereals 4303 11.6 560296. 0.57% 1.17%
## 9 Nuts and seeds 6259 4.37 559874. 0.84% 1.17%
## 10 Oily fish 2610 6.21 550425. 0.35% 1.15%
## 11 White fish, shellfish 1597 3.99 498929. 0.21% 1.05%
## 12 Puddings 2291 8.40 459785. 0.31% 0.96%
## 13 Other Milk Cream 6605 11.6 434239. 0.88% 0.91%
## 14 Pork 1832 3.88 420504. 0.24% 0.88%
## 15 Coated Chicken 1170 6.43 359773. 0.16% 0.75%
## 16 Lamb 1251 6.42 342289. 0.17% 0.72%
## 17 Spirits and liqueurs 2698 4.03 286589. 0.36% 0.6%
## 18 Oth Bread 932 4.96 149552. 0.12% 0.31%
## 19 Dietary supplements 13016 5.16 44916. 1.74% 0.09%
## 20 Smoothies 137 2.93 17061. 0.02% 0.04%1.2.2.3 Healthy foods Group 3 (n = 20)
TableFoogGroup %>%
filter(HealthPoints3g == 3) %>%
select(-lesshealthy, -neutral,-calcumprop, -healthy, -HealthPoints3g)## # A tibble: 20 x 6
## mfgLab n meanHpoint mfgCalories n.freq cal.Prop
## <chr> <int> <dbl> <dbl> <chr> <chr>
## 1 Cakes & Pastries 7806 17.3 1710594. 1.04% 3.59%
## 2 Biscuits 13200 18.3 1662598. 1.76% 3.49%
## 3 Cheese 10983 21.5 1492015. 1.47% 3.13%
## 4 JamsSpreads 37960 14.7 1215279. 5.07% 2.55%
## 5 Chocolate 6495 24.9 1046113. 0.87% 2.19%
## 6 Butter 10203 24.3 965901. 1.36% 2.02%
## 7 Spreads less-fat 12620 23.4 848835. 1.68% 1.78%
## 8 Crisps 5664 13.0 835672. 0.76% 1.75%
## 9 Sausages 3025 15.3 775004. 0.4% 1.62%
## 10 Meat pastries 1979 14.8 744640. 0.26% 1.56%
## 11 Bacon and ham 8467 15.5 738727. 1.13% 1.55%
## 12 Margarine 8742 19.6 410108. 1.17% 0.86%
## 13 Burgers/kebabs 939 14.6 328396. 0.13% 0.69%
## 14 Ice cream 1816 15.4 309564. 0.24% 0.65%
## 15 Other meat 1571 17.0 271445. 0.21% 0.57%
## 16 Sugar confectionery 1820 14.5 164392. 0.24% 0.34%
## 17 LowFat Spreads 3654 18.1 134527. 0.49% 0.28%
## 18 Polyunsatu margarine 3179 20.0 126714. 0.42% 0.27%
## 19 Liver 454 13.6 67330. 0.06% 0.14%
## 20 Commercial toddlers foods 66 18.5 1862. 0.01% 0%2 CA analysis in all adults
2.1 All food group and 7 Eating Time slots
2.1.1 Contingency table of food groups and time slots:
Note that we have randomly splitted the food recording data sets into two subsets. From here we are using only the data set 1 for hypothesis generation.
library(FactoMineR)
library(factoextra)
load("../HFood.Rdata")
freqtab <- xtabs(~HFood$mfgLab + HFood$MealTimeSlot)
as.data.frame.matrix(freqtab) %>%
kable() %>%
kable_styling(bootstrap_options = c("striped", "hover"))| 6am to 8:59am | 9am to 11:59am | 12 noon to 1:59pm | 2pm to 4:59pm | 5pm to 7:59pm | 8pm to 9:59pm | 10pm to 5:59am | |
|---|---|---|---|---|---|---|---|
| 1% milk | 189 | 137 | 50 | 66 | 79 | 48 | 35 |
| 2% milk | 7513 | 6844 | 3262 | 3986 | 3124 | 2530 | 1533 |
| Artificial Sweeteners | 920 | 832 | 467 | 591 | 410 | 337 | 219 |
| Bacon and ham | 209 | 757 | 1707 | 453 | 906 | 213 | 57 |
| Beef | 3 | 28 | 435 | 254 | 1462 | 306 | 38 |
| Beer lager | 11 | 34 | 253 | 440 | 1014 | 1467 | 826 |
| Biscuits | 417 | 1173 | 1130 | 1405 | 813 | 1013 | 614 |
| Brown Bread | 540 | 567 | 1073 | 287 | 363 | 178 | 87 |
| Burgers/kebabs | 0 | 14 | 112 | 81 | 219 | 49 | 12 |
| Butter | 812 | 876 | 1243 | 548 | 1085 | 385 | 145 |
| Cakes & Pastries | 182 | 597 | 778 | 891 | 857 | 465 | 177 |
| Cheese | 131 | 381 | 1916 | 656 | 1563 | 644 | 182 |
| Chicken/turkey | 19 | 96 | 1011 | 571 | 2039 | 544 | 103 |
| Chips | 19 | 96 | 624 | 442 | 1727 | 376 | 83 |
| Chocolate | 52 | 349 | 470 | 728 | 522 | 834 | 303 |
| Coated Chicken | 0 | 6 | 115 | 91 | 300 | 64 | 20 |
| Commercial toddlers foods | 3 | 3 | 15 | 6 | 11 | 4 | 0 |
| Crisps | 11 | 270 | 875 | 520 | 443 | 501 | 228 |
| Diet soft drinks | 361 | 679 | 1167 | 1076 | 1916 | 1199 | 674 |
| Dietary supplements | 4685 | 1244 | 137 | 74 | 121 | 121 | 179 |
| Eggs | 424 | 810 | 915 | 378 | 960 | 262 | 56 |
| Fruit | 2631 | 3012 | 3700 | 2341 | 2937 | 1793 | 518 |
| Fruit juice | 992 | 592 | 536 | 337 | 650 | 294 | 134 |
| HiFi cereals | 2534 | 1255 | 102 | 41 | 43 | 69 | 95 |
| Ice cream | 0 | 8 | 93 | 143 | 399 | 210 | 44 |
| JamsSpreads | 4639 | 4477 | 2283 | 2504 | 2518 | 1673 | 912 |
| Lamb | 1 | 3 | 113 | 72 | 296 | 99 | 16 |
| Liver | 10 | 14 | 72 | 20 | 76 | 23 | 2 |
| LoFi cereals | 1212 | 704 | 80 | 37 | 40 | 48 | 75 |
| LowFat Spreads | 330 | 323 | 533 | 151 | 309 | 101 | 45 |
| Margarine | 137 | 317 | 802 | 448 | 1946 | 663 | 91 |
| Meat pastries | 11 | 68 | 308 | 140 | 349 | 104 | 29 |
| Misc./Vending | 877 | 1608 | 5795 | 2720 | 9386 | 2990 | 859 |
| Nuts and seeds | 645 | 528 | 466 | 401 | 498 | 394 | 153 |
| Oily fish | 26 | 47 | 414 | 147 | 572 | 151 | 17 |
| Oth Bread | 76 | 79 | 162 | 51 | 70 | 19 | 13 |
| Other meat | 40 | 86 | 257 | 100 | 207 | 78 | 18 |
| Other Milk Cream | 610 | 535 | 446 | 431 | 780 | 386 | 139 |
| Other white fish | 8 | 58 | 604 | 194 | 673 | 244 | 39 |
| Pasta & Rice | 694 | 545 | 1574 | 925 | 3934 | 1259 | 270 |
| Polyunsatu margarine | 51 | 95 | 264 | 143 | 753 | 228 | 39 |
| Pork | 4 | 15 | 169 | 102 | 475 | 107 | 7 |
| Potatos other | 13 | 44 | 982 | 536 | 2987 | 434 | 36 |
| Puddings | 13 | 39 | 246 | 147 | 498 | 155 | 53 |
| Reg soft drinks | 295 | 551 | 1052 | 969 | 1396 | 920 | 527 |
| Salad and raw veg | 165 | 620 | 5822 | 1878 | 6086 | 1814 | 301 |
| Sausages | 70 | 266 | 301 | 151 | 578 | 111 | 15 |
| Skimmed Milk | 1416 | 1047 | 509 | 626 | 510 | 436 | 309 |
| Smoothies | 12 | 14 | 20 | 10 | 4 | 4 | 1 |
| Spirits and liqueurs | 3 | 9 | 31 | 71 | 297 | 518 | 403 |
| Spreads less-fat | 983 | 1262 | 1776 | 549 | 1049 | 424 | 190 |
| Sugar confectionery | 40 | 125 | 113 | 243 | 164 | 165 | 82 |
| Tea/Coffee/Water | 14297 | 16103 | 10587 | 11780 | 10414 | 7327 | 5544 |
| Veg not raw | 203 | 669 | 5299 | 2559 | 13693 | 3022 | 314 |
| White Bread | 1195 | 1775 | 2748 | 908 | 1643 | 660 | 280 |
| White fish, shellfish | 1 | 6 | 148 | 72 | 448 | 85 | 5 |
| Whole Milk | 1654 | 1584 | 792 | 851 | 919 | 645 | 442 |
| Wine | 3 | 11 | 184 | 178 | 1317 | 1376 | 376 |
| WMeal Bread | 683 | 674 | 1178 | 296 | 474 | 185 | 92 |
| Yogurt | 524 | 435 | 803 | 326 | 808 | 372 | 114 |
2.1.2 Foods and 7 Eating Time Slots
res.ca <- CA(as.data.frame.matrix(freqtab), graph = FALSE)
fviz_screeplot(res.ca, addlabels = TRUE)
Scree plot – CA of the NDNS RP 9 year data, eating time slots and food goups.
First two dimensions represents 84.2% of the inertia (variation in time slots profile).
fviz_ca_biplot(res.ca, repel = TRUE, title = "Biplot of Correspondence analysis for 60 food groups.") 
The horizontal axis contrasts the early time (am, and earlier than 6 am included) with later time (noon, afternoon, till 10 pm); and breakfast foods with the others.
The vertical axis shows large contribution from alcohol against the others.
2.1.3 Biplot by food healthiness (1)
2.1.3.1 Three healthy food only
HealthyFoods <- as.data.frame.matrix(freqtab)[rownames(as.data.frame.matrix(freqtab)) %in% c("Veg not raw", "Fruit", "Salad and raw veg"),]
Healthy.ca <- CA(HealthyFoods, graph = FALSE)
fviz_ca_biplot(Healthy.ca,
repel = TRUE, title = "Biplot of Correspondence analysis for 3 healthy food groups.") 
2.1.3.2 with confidence region (1000 bootstrapping method)
Healthy Foods with bootstrap 95% confidence regions for Food groups.
Healthy Foods with bootstrap 95% confidence regions for Time Slots.
2.1.3.3 Neutral food (n = 26)
NeutralFoods <- as.data.frame.matrix(freqtab)[rownames(as.data.frame.matrix(freqtab)) %in% TableFoogGroup$mfgLab[TableFoogGroup$neutral],]
Neutral.ca <- CA(NeutralFoods, graph = FALSE)
fviz_ca_biplot(Neutral.ca,
repel = TRUE, title = "Biplot of Correspondence analysis for 26 neutral food groups.") 
2.1.3.4 with confidence region (1000 bootstrapping method)
Neutral Foods with bootstrap 95% confidence regions for Food groups.
Neutral Foods with bootstrap 95% confidence regions for Time Slots.
2.1.3.5 less healthy food (n = 31)
LessFoods <- as.data.frame.matrix(freqtab)[rownames(as.data.frame.matrix(freqtab)) %in% TableFoogGroup$mfgLab[TableFoogGroup$lesshealthy],]
Less.ca <- CA(LessFoods, graph = FALSE)
fviz_ca_biplot(Less.ca,
repel = TRUE, title = "Biplot of Correspondence analysis for 31 less healthy food groups.") 
2.1.3.6 with confidence region (1000 bootstrapping method)
Less Healthy Foods with bootstrap 95% confidence regions for Food groups.
Less Healthy Foods with bootstrap 95% confidence regions for Time Slots.
2.1.4 Biplot by food healthiness (2)
2.1.4.1 Food Healthiness Group 1
G1Foods <- as.data.frame.matrix(freqtab)[rownames(as.data.frame.matrix(freqtab)) %in% TableFoogGroup$mfgLab[TableFoogGroup$HealthPoints3g == 1],]
G1.ca <- CA(G1Foods, graph = FALSE)
fviz_ca_biplot(G1.ca,
repel = TRUE, title = "Biplot of Correspondence analysis for food group 1 (1st 20 healthy foods).") 
2.1.4.2 Food Healthiness Group 2
G2Foods <- as.data.frame.matrix(freqtab)[rownames(as.data.frame.matrix(freqtab)) %in% TableFoogGroup$mfgLab[TableFoogGroup$HealthPoints3g == 2],]
G2.ca <- CA(G2Foods, graph = FALSE)
fviz_ca_biplot(G2.ca,
repel = TRUE, title = "Biplot of Correspondence analysis for food group 2 (2nd 20 healthy foods).") 
2.1.4.3 Food Healthiness Group 3
G3Foods <- as.data.frame.matrix(freqtab)[rownames(as.data.frame.matrix(freqtab)) %in% TableFoogGroup$mfgLab[TableFoogGroup$HealthPoints3g == 3],]
G3.ca <- CA(G3Foods, graph = FALSE)
fviz_ca_biplot(G3.ca,
repel = TRUE, title = "Biplot of Correspondence analysis for food group 3 (3rd 20 healthy foods).") 
3 CA analysis by diabetes status
3.1 among diagnosed diabete patients
DiagDM <- as.logical(HFood$DM4cat.y == 3)
DiagDMtab <- xtabs(~HFood$mfgLab + HFood$MealTimeSlot, subset = DiagDM)
DiagDMmatrix <- matrix(data = DiagDMtab, nrow = 59, ncol = 7,
dimnames = list(rownames(DiagDMtab), colnames(DiagDMtab)))
DiagDM.ca <- CA(DiagDMmatrix, graph = FALSE)
fviz_ca_biplot(DiagDM.ca,
repel = TRUE, title = "Biplot of Correspondence analysis among diagnosed DM.") 
3.2 among non-diabete participants
DiagDM <- as.logical(HFood$DM4cat.y == 0)
DiagDMtab <- xtabs(~HFood$mfgLab + HFood$MealTimeSlot, subset = DiagDM)
DiagDMmatrix <- matrix(data = DiagDMtab, nrow = 60, ncol = 7,
dimnames = list(rownames(DiagDMtab), colnames(DiagDMtab)))
DiagDM.ca <- CA(DiagDMmatrix, graph = FALSE)
fviz_ca_biplot(DiagDM.ca,
repel = TRUE, title = "Biplot of Correspondence analysis among non-diabete participants.") 
3.3 among Pre-diabetes participants
DiagDM <- as.logical(HFood$DM4cat.y == 1)
DiagDMtab <- xtabs(~HFood$mfgLab + HFood$MealTimeSlot, subset = DiagDM)
DiagDMmatrix <- matrix(data = DiagDMtab, nrow = 59, ncol = 7,
dimnames = list(rownames(DiagDMtab), colnames(DiagDMtab)))
DiagDM.ca <- CA(DiagDMmatrix, graph = FALSE)
fviz_ca_biplot(DiagDM.ca,
repel = TRUE, title = "Biplot of Correspondence analysis among Prediabetes participants.") 
3.4 among undiagnosed-diabetes participants
DiagDM <- as.logical(HFood$DM4cat.y == 2)
DiagDMtab <- xtabs(~HFood$mfgLab + HFood$MealTimeSlot, subset = DiagDM)
DiagDMmatrix <- matrix(data = DiagDMtab, nrow = 59, ncol = 7,
dimnames = list(rownames(DiagDMtab), colnames(DiagDMtab)))
DiagDM.ca <- CA(DiagDMmatrix, graph = FALSE)
fviz_ca_biplot(DiagDM.ca,
repel = TRUE, title = "Biplot of Correspondence analysis among undiagnosed DM participants.") 
1.1.5 Social economic status
121 pariticipants’ socio-economic classification were missing.