This study evaluated the incidence of colorectal cancer (CRC) according to the number of metabolic syndrome (MetS) components.
Methods
Using health checkup and insurance claims data of 6,365,409 subjects, the occurrence of CRC according to stage of MetS by sex was determined from the date of the health checkup in 2009 until December 31, 2018.
Results
Cumulative incidence rates (CIR) of CRC in men and women was 3.9 and 2.8 per 1000 (p < 0.001), respectively. CIR of CRC for the normal, pre-MetS, and MetS groups in men was 2.6, 3.9, and 5.5 per 1000 (p < 0.001) and CIR in women was 2.1, 2.9, and 4.5 per 1000 (p < 0.001), respectively. Compared with the normal group, the hazard ratio (HR) of CRC for the pre-MetS group was 1.25 (95% CI 1.17–1.33) in men and 1.09 (95% CI 1.02–1.17) in women, and the HR of CRC for the MetS group was 1.54 (95% CI 1.43–1.65) in men and 1.39 (95% CI 1.26–1.53) in women after adjustment.
Conclusions
We found that MetS is a risk factor for CRC in this study. Therefore, the prevention and active management of MetS would contribute to the prevention of CRC.
Hinweise
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Abkürzungen
CIR
Cumulative incidence rates
CRC
Colorectal cancer
ID
Incidence density
MetS
Metabolic syndrome
NCEP-III
National Cholesterol Education Program Adult Treatment Panel III
NHIS
National Health Insurance Service
Introduction
Colorectal cancer (CRC) is the third most common cancer worldwide and accounts for 10.2% of all cancers (approximately 1.8 million people a year) [1]. CRC is the second most common cancer after stomach cancer in Korea [1], so identifying and managing risk factors is the first step in preventing CRC. In Korea, the incidence of CRC has increased over the past decade, and the age-adjusted incidence per 100,000 men and women has increased from 26.2, 16.4 in 1999 to 40.4, 22.4 in 2016, respectively [2]. Major risk factors for CRC include genetic predisposition, Western dietary habits, lifestyle (smoking, drinking, physical activity, etc.), and metabolic diseases (obesity, insulin resistance, etc.) [3].
Metabolic syndrome (MetS) is a cluster of metabolic risk factors that includes abdominal obesity, hypertension, hyperglycemia, and dyslipidemia; several definitions have been suggested using different criteria [4, 5]. More than 20% of adults are known to have MetS [6], but its prevalence worldwide varies depending on race, environmental factors, the age and gender composition of the population, genetic differences, physical activity level, eating habits, and differences in measurement standards [7, 8]. The estimated total prevalence of MetS for adults in Korea is 26.9%: 30.0% in males and 24.6% in females [9].
Anzeige
Some studies on the correlation between metabolic syndrome and CRC have been reported recently, but many studies have been conducted in Western countries and races [10‐13]. However, studies on Asian races are still limited to East Asian countries such as Taiwan and Japan [14, 15]. In addition, the study population is limited to those who are screened at only university hospital screening centers, so it is not easy to generalize to the entire population [14]. Also there is a limitation that it is difficult to produce a meaningful result because of the short observation period and the lack of cancer cases [15].
In order to complement the limitations of previous research conducted on Asian races, this study evaluated the incidence of CRC according to the number of MetS components using the health checkup data and insurance claims data from the National Health Insurance Service (NHIS) in Korea.
Methods
Study design and population
This was a retrospective observational study that used health checkup and insurance claims data from the NHIS (Fig. 1). NHIS is constructing and providing health database that includes medical history, treatment, type of disease, and prescription history of all Koreans who use health checkup or medical service under national health insurance. The database has been operating continuously, and about 50 million people’s medical data are monitored from 2002 to 2018. In detail, the National Health Information database is divided into Qualification database, Treatment database, and Health check-up database. The National Health Information database is significant in terms of representation because it monitors medical history of all Koreans, and it is also useful because it can be linked and analyzed with other administrative data through the social security number. The Health check-up database used in this study is very suitable for the purpose of research because it contains the actual measurement value such as blood pressure, blood sugar, triglycerides, HDL, and health habits for those who undergo health checkups every 2 years.
Fig. 1
Study design and selection of the study population
×
The source population was defined as those who had had health checkups in 2009. While this comprised 15,036,607 people, only 9,927,538 people were actually examined. Of these, 3,562,129 were excluded for the following reasons:
1.
Age < 30 years old or ≥ 70 years old.
2.
Missing MetS checkup items (fasting blood sugar, systolic and diastolic blood pressure, triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), abdominal obesity); height < 120 or > 200 cm, weight < 20 or > 200 kg.
3.
If the subject has a history of cardiovascular disease, cerebrovascular disease, or cancer; cardiovascular diseases (ischemic heart disease I20-I25, cardiac failure I42 or I50), atrial fibrillation (I48), cerebrovascular disease (I60-I69), circulatory system disease (I00-I99), and beneficiaries for cancer (C00-C97).
4.
Colorectal cancer found within 3 years after health checkups in 2009.
Anzeige
Consequently, the final study population came to 6,365,409; they were divided into three groups depending on the number of MetS components. There were 1,858,419, 3,358,496, and 1,148,494 in the normal, pre-MetS, and MetS groups, respectively. Using insurance claims data, the occurrence of CRC was checked in the study population from the date of the health checkups in 2009 to December 31, 2018. Mean follow-up period is 9.3 years.
Measurements
Dependent variable
The dependent variable was the occurrence of CRC, as identified operationally from insurance benefit claims data, with codes C18-C20 as a primary diagnosis from the International Classification of Diseases-Tenth revision and V193 as a special code. A special code is a system whereby the economic burden on patients diagnosed with cancer, severe burns, cerebrovascular and heart diseases, or intractable diseases is reduced by decreasing the copayment. If a disease is claimed with this special code, it means that the diagnosis is more accurate than a claim disease without a special code.
Independent variable
The independent variable was the stage of MetS, which was classified into three groups according to the criteria of modified National Cholesterol Education Program-Adult Treatment Panel III (NECP-ATP III, which are the most agreed-upon criteria [16]. MetS refers to subjects with at least three of the following five factors: (1) abdominal obesity (waist circumference ≥ 90 cm for men, ≥ 85 cm for women); (2) elevated blood pressure (systolic BP ≥ 130 mmHg or diastolic BP ≥ 85 mmHg or treatment of previously diagnosed hypertension); (3) elevated fasting glucose (≥ 100 mg/dL or treatment of previously diagnosed diabetes mellitus); (4) high TG (≥ 150 mg/dL or drug treatment for high TG); and (5) low HDL-C (< 40 mg/dL for men, < 50 mg/dL for women or drug treatment for low HDL-C). The subjects were placed in the pre-MetS group with one or two MetS factors and the Normal group with no MetS factors.
Adjusted variables
The study population was divided into two subsets by gender. Next, age, health behaviors, family history, and laboratory data from the health checkup were used as adjustment variables. Age was divided into 10-year intervals, and health behaviors were selected from smoking, drinking, and physical activities in the questionnaire. Smoking was categorized into three groups: never smoked, smoked in the past but no longer, and currently smoking. Alcohol consumption was categorized into four groups based on the drinking frequency: non-drinker, 2–3 times per month, 1–4 times per week, and almost every day. Physical activity group was categorized into three groups based on frequency: never, 1–4 times per week, and almost every day. Family history of cancer was divided into two groups of yes or no. Height, weight, body mass index (BMI), hemoglobin, serum creatinine, total cholesterol, and alanine aminotransferase (ALT), which were health checkup items, were used as laboratory findings, and as continuous variables. BMI (kg/m2), hemoglobin (g/dL), and serum creatinine (mg/dL) are categorized according to the criteria of Korean health screening [17]. Total cholesterol (mg/dL) and ALT (IU/L) were used in the model as continuous variables, and the original data values were increased by ten times to improve the readability of the hazard ratio.
Height (cm) was divided into four groups according to the quartiles of height distribution for each sex (< 167, 167–171, 172–174, ≥ 175 cm for men; < 154, 154–156, 157–160, ≥ 161 cm for women). Weight (kg) also was divided into four groups according to the quartiles of weight distribution for each sex (< 64, 64–69, 70–76, ≥ 77 cm for men; < 52, 52–55, 56–61, ≥ 62 kg for women). BMI was categorized into five groups: underweight (< 18.5), normal (18.5–22.9), overweight (23.0–24.9), obesity (25–29.9), and altitude obesity (≥ 30). Hemoglobin was categorized into three groups and differed by sex: normal (men: > 12.0, women: > 10.0), mild or moderate anemia (men: 10.0–12.0, women: 8.0–10.0), severe anemia (men: < 10.0, women: < 8); serum creatinine was categorized into two groups: normal (<1.5), abnormal (≥ 1.5). Total cholesterol and ALT were also divided into four groups for according to the quartiles of their distribution for each sex (total cholesterol < 173, 173–194, 195–218, ≥ 219, ALT < 18, 18–24, 25–35, ≥ 36 for men; total cholesterol < 171, 171–192, 193–217, ≥ 218, ALT < 13, 13–15, 16–21, ≥ 22 for women) for descriptive analysis.
Statistical analysis
We summarized the frequency of study population for related variables according to the number of MetS components, and calculated the cumulative incidence rate (CIR) and incidence density (ID) of CRC. CIR is the incidence per 1000 person and ID is the incidence per 10,000 person-years (PY). PY was calculated as the time from the baseline examination to the date of CRC diagnosis, the date of death, or December 31, 2018 when who survives without diagnosis of CRC.
Next, we compared the frequency between variables using the chi-square test. To analyze the risk of developing CRC due to MetS, Cox proportional hazard regression was applied. All variables affecting the incidence of CRC were considered stepwise and five Cox proportional hazard models were fitted, including one unadjusted model for comparison.
Model 1 was adjusted for age. Model 2 was adjusted for health behavior (smoking, exercise). Model 3 was adjusted for family history cancer. Model 4 was adjusted for the laboratory findings (height, weight, hemoglobin, and other relevant values). Model 5 was adjusted for the laboratory findings (body mass index, creatinine, hemoglobin, and other relevant values). The results are summarized as the hazard ratio (HR) and 95% confidence interval (95% CI).
A level of α = 0.05 was used to determine the significance of the models and variables. All statistical analyses were performed using SAS ver. 9.1 (SAS Institute, Cary, NC, USA).
Ethics considerations
The study was approved by the institutional review board of Konkuk University (7001355-201909-E-100).
Results
CIR and ID of CRC according to the number of Mets components in men
The number of study population and the CIR of CRC in men were 3,695,923 and 3.9 (Table 1). The CIR of CRC by MetS stage was 2.6, 3.9, and 5.5 for the normal, pre-MetS, and MetS groups, respectively (p < 0.001).
Table 1
Cumulative incidence rate and incidence density of colorectal cancer according to the progression of metabolic syndrome in men
Characteristics
Category
No. of study population
No. of colorectal cancer
Cumulative incidence rate (per 1000)
Incidence density (per 10,000 person-years)
Normal
Pre-MetS
MetS
Total
Normal
Pre-MetS
MetS
Total
Normal
Pre-MetS
MetS
p value
Total
p value
Normal
Pre-MetS
MetS
p value
Total
p value
Total
841,887
2,037,811
816,225
3,695,923
2174
7858
4517
14,549
2.6
3.9
5.5
3.9
2.8
4.2
6.0
4.2
Age
30–39
52,090
78,689
19,451
150,230
31
62
11
104
0.6
0.8
0.6
< 0.001
0.7
< 0.001
0.6
0.8
0.6
< 0.001
0.7
< 0.001
40–49
403,228
802,049
267,504
1,472,781
457
1066
422
1945
1.1
1.3
1.6
< 0.001
1.3
1.2
1.4
1.7
< 0.001
1.4
50–59
259,358
695,206
294,199
1,248,763
771
2722
1387
4880
3.0
3.9
4.7
< 0.001
3.9
3.2
4.2
5.1
< 0.001
4.2
60–69
127,211
461,867
235,071
824,149
915
4008
2697
7620
7.2
8.7
11.5
< 0.001
9.2
7.8
9.4
12.5
< 0.001
10.0
Smoking
Non-smoker
254,976
551,062
198,274
1,004,312
599
1998
1098
3695
2.3
3.6
5.5
< 0.001
3.7
<0.001
2.5
3.9
6.0
< 0.001
4.0
< 0.001
Ex-smoker
191,807
502,695
209,573
904,075
506
2112
1330
3948
2.6
4.2
6.3
< 0.001
4.4
2.8
4.5
6.8
< 0.001
4.7
Smoker
390,833
975,701
405,252
1,771,786
1061
3718
2076
6855
2.7
3.8
5.1
< 0.001
3.9
2.9
4.1
5.6
< 0.001
4.2
Alcohol consumption
No drink
266,110
561,954
205,282
1,033,346
722
2099
1121
3942
2.7
3.7
5.5
< 0.001
3.8
< 0.001
2.9
4.0
5.9
< 0.001
4.1
< 0.001
2–3/per month
435,666
1,015,825
391,247
1,842,738
965
3469
1882
6316
2.2
3.4
4.8
< 0.001
3.4
2.4
3.7
5.2
< 0.001
3.7
1–4/per week
100,885
330,563
157,013
588,461
309
1479
984
2772
3.1
4.5
6.3
< 0.001
4.7
3.3
4.8
6.8
< 0.001
5.1
5/per week
29,429
110,628
56,183
196,240
149
745
503
1397
5.1
6.7
9.0
< 0.001
7.1
5.5
7.3
9.8
< 0.001
7.7
Physical exercise,
per week
No exercise
343,189
841,263
346,095
1,530,547
897
3244
1948
6089
2.6
3.9
5.6
< 0.001
4.0
< 0.001
2.8
4.2
6.1
< 0.001
4.3
< 0.001
1–4/per week
177,100
419,415
167,585
764,100
452
1489
854
2795
2.6
3.6
5.1
< 0.001
3.7
2.7
3.8
5.5
< 0.001
3.9
5/per week
314,767
762,500
297,226
1,374,493
807
3072
1694
5573
2.6
4.0
5.7
< 0.001
4.1
2.7
4.3
6.2
< 0.001
4.4
FHx of cancer
No
494,125
1,199,780
486,757
2,180,662
1,176
4376
2520
8072
2.4
3.6
5.2
< 0.001
3.7
< 0.001
2.6
3.9
5.6
< 0.001
4.0
< 0.001
Yes
95,222
217,852
83,206
396,280
234
816
467
1517
2.5
3.7
5.6
< 0.001
3.8
2.6
4.0
6.1
< 0.001
4.1
Height
(cm)
< 167
239,064
636,453
236,248
1,111,765
822
3205
1723
5750
3.4
5.0
7.3
< 0.001
5.2
< 0.001
3.7
5.4
7.2
< 0.001
5.6
< 0.001
167–171
220,547
534,838
209,648
965,033
588
2131
1193
3912
2.7
4.0
5.7
< 0.001
4.1
2.9
4.3
5.6
< 0.001
4.4
172-174
204,006
469,535
192,522
866,063
435
1485
950
2870
2.1
3.2
4.9
< 0.001
3.3
2.3
3.4
4.8
< 0.001
3.6
≥ 175
178,270
396,985
177,807
753,062
329
1037
651
2017
1.8
2.6
3.7
< 0.001
2.7
2.0
2.8
3.6
< 0.001
2.9
Weight
(kg)
< 64
372,420
572,899
83,536
1,028,855
1,106
2,672
646
4,424
3.0
4.7
7.7
< 0.001
4.3
< 0.001
3.2
5.0
8.5
< 0.001
4.6
< 0.001
64-69
235,461
530,335
128,262
894,058
555
2157
871
3583
2.4
4.1
6.8
< 0.001
4.0
2.5
4.4
7.4
< 0.001
4.3
70–76
171,877
534,264
213,956
920,097
397
1852
1277
3526
2.3
3.5
6.0
< 0.001
3.8
2.5
3.7
6.4
< 0.001
4.1
≥ 77
62,129
400,313
390,471
852,913
116
1177
1723
3016
1.9
2.9
4.4
< 0.001
3.5
2.0
3.2
4.8
< 0.001
3.8
BMI
(kg/m2)
< 18.5
33,171
29,275
1,844
64,290
76
144
21
241
2.3
4.9
11.4
< 0.001
3.7
< 0.001
2.5
5.4
13.1
< 0.001
4.1
< 0.001
18.5–22.9
445,140
656,729
83,735
1,185,604
1,138
2617
572
4327
2.6
4.0
6.8
< 0.001
3.6
2.7
4.3
7.4
< 0.001
3.9
23.0–24.9
234,677
621,458
159,332
1,015,467
631
2384
977
3992
2.7
3.8
6.1
< 0.001
3.9
2.9
4.1
6.6
< 0.001
4.2
25.0–29.9
128,038
690,021
477,677
1,295,736
327
2580
2537
5444
2.6
3.7
5.3
< 0.001
4.2
2.7
4.0
5.7
< 0.001
4.5
≥ 30.0
861
40,328
93,637
134,826
2
133
410
545
2.3
3.3
4.4
< 0.001
4.0
2.5
3.6
4.7
< 0.001
4.4
Total cholesterol
(mg/dL)
< 173
287,648
499,111
154,244
941,003
650
1819
886
3355
2.3
3.6
5.7
< 0.001
3.6
< 0.001
2.4
3.9
6.2
< 0.001
3.8
< 0.001
173–194
239,234
504,393
176,051
919,678
615
1886
961
3462
2.6
3.7
5.5
< 0.001
3.8
2.8
4.0
5.9
< 0.001
4.1
195–218
189,771
516,797
210,313
916,881
530
2,044
1,159
3,733
2.8
4.0
5.5
< 0.001
4.1
3.0
4.3
6.0
< 0.001
4.4
≥ 219
125,234
517,510
275,617
918,361
379
2,109
1,511
3,999
3.0
4.1
5.5
< 0.001
4.4
3.3
4.4
5.9
< 0.001
4.7
ALT
(IU/L)
< 18
354,952
530,427
94,775
980,154
880
2,154
628
3,662
2.5
4.1
6.6
< 0.001
3.7
< 0.001
2.7
4.4
7.2
< 0.001
4.0
< 0.001
18–24
241,797
549,667
159,960
951,424
648
2,144
948
3,740
2.7
3.9
5.9
< 0.001
3.9
2.9
4.2
6.4
< 0.001
4.2
25–35
154,775
500,574
221,678
877,027
417
1967
1277
3661
2.7
3.9
5.8
< 0.001
4.2
2.9
4.2
6.2
< 0.001
4.5
≥ 36
90,363
457,143
339,812
887,318
229
1593
1664
3486
2.5
3.5
4.9
< 0.001
3.9
2.7
3.8
5.3
< 0.001
4.2
Hemoglobin
(g/dL)
> 12
836,812
2,025,415
811,497
3,673,724
2,153
7,772
4,476
14,401
2.6
3.8
5.5
< 0.001
3.9
< 0.001
2.8
4.1
6.0
< 0.001
4.2
< 0.001
10–12
4,173
9,967
3,769
17,909
15
64
31
110
3.6
6.4
8.2
< 0.001
6.1
3.9
7.2
9.4
< 0.001
6.9
< 10
624
1,624
623
2,871
5
17
6
28
8.0
10.5
9.6
< 0.001
9.8
8.8
11.9
11.2
< 0.001
11.1
Serum creatinine
(mg/dL)
≤ 1.5
815,023
1,962,815
783,300
3,561,138
2,112
7599
4339
14,050
2.6
3.9
5.5
< 0.001
3.9
< 0.001
2.8
4.2
6.0
< 0.001
4.3
< 0.001
> 1.5
26,813
74,886
32,883
134,582
61
259
178
498
2.3
3.5
5.4
< 0.001
3.7
2.4
3.7
5.8
< 0.001
4.0
MetS metabolic syndrome, FHx family history, BMI body mass index, ALT alanine aminotransferase
The ID of CRC in men was 4.2 (Table 1). The ID of CRC by MetS stage was 2.8, 4.2, and 6.0 for the normal, pre-MetS, and MetS groups, respectively (p < 0.001).
Anzeige
The number of study population and the CIR and the ID of CRC for the other variables in men were summarized in Table 1.
CIR and ID of CRC according to the number of Mets components in women
The number of study population and the CIR of CRC in women were 2,669,486 and 2.8 (Table 2). The CIR of CRC by MetS stage was 2.1, 2.9, and 4.5 for the normal, pre-MetS, and MetS groups, respectively (p < 0.001).
Table 2
Cumulative incidence rate and incidence density of colorectal cancer according to the progression of metabolic syndrome in women
Characteristics
Category
No. of study population
No. of colorectal cancer
Cumulative incidence rate (per 1000)
Incidence density (per 10,000 person-years)
Normal
Pre-MetS
MetS
Total
Normal
Pre-MetS
MetS
Total
Normal
Pre-MetS
MetS
p value
Total
p value
Normal
Pre-MetS
MetS
p value
Total
p value
Total
1,016,532
1,320,685
332,269
2,669,486
2119
3827
1493
7439
2.1
2.9
4.5
2.8
2.2
3.1
4.8
3.0
Age
30–39
44,462
24,310
1909
70,681
29
14
1
44
0.7
0.6
0.5
< 0.001
0.6
< 0.001
0.7
0.6
0.6
< 0.001
0.7
< 0.001
40–49
355,151
277,454
33,864
666,469
363
323
46
732
1.0
1.2
1.4
< 0.001
1.1
1.1
1.3
1.5
< 0.001
1.2
50–59
424,813
555,494
114,030
1,094,337
1030
1464
416
2910
2.4
2.6
3.6
< 0.001
2.7
2.6
2.8
3.9
< 0.001
2.9
60–69
192,106
463,427
182,466
837,999
697
2026
1030
3753
3.6
4.4
5.6
< 0.001
4.5
3.9
4.7
6.0
< 0.001
4.8
Smoking
Non-smoker
958,586
1,249,613
312,149
2,520,348
1,997
3581
1393
6971
2.1
2.9
4.5
< 0.001
2.8
< 0.001
2.2
3.1
4.8
< 0.001
3.0
< 0.001
Ex-smoker
20,223
21,067
5,102
46,392
45
69
23
137
2.2
3.3
4.5
< 0.001
3.0
2.4
3.5
4.8
< 0.001
3.2
Smoker
31,594
42,661
13,015
87,270
71
155
68
294
2.2
3.6
5.2
< 0.001
3.4
2.4
3.9
5.6
< 0.001
3.6
Alcohol consumption
No drink
695,610
956,081
255,189
1,906,880
1,460
2792
1167
5419
2.1
2.9
4.6
< 0.001
2.8
< 0.001
2.3
3.1
4.9
< 0.001
3.0
< 0.001
2–3/per month
267,878
291,847
58,767
618,492
549
816
245
1610
2.0
2.8
4.2
< 0.001
2.6
2.2
3.0
4.5
< 0.001
2.8
1–4/per week
30,509
41,508
9,949
81,966
76
132
42
250
2.5
3.2
4.2
< 0.001
3.1
2.7
3.4
4.5
< 0.001
3.3
5/per week
8,547
13,865
3,936
26,348
18
42
17
77
2.1
3.0
4.3
< 0.001
2.9
2.3
3.3
4.6
< 0.001
3.1
Physical exercise,
per week
No exercise
529,972
692,901
179,742
1,402,615
1092
1998
781
3871
2.1
2.9
4.3
< 0.001
2.8
< 0.001
2.2
3.1
4.7
< 0.001
3.0
< 0.001
1–4/per week
191,856
241,698
59,221
492,775
412
721
270
1403
2.1
3.0
4.6
< 0.001
2.8
2.3
3.2
4.9
< 0.001
3.1
5/per week
288,778
379,098
91,511
759,387
611
1092
432
2135
2.1
2.9
4.7
< 0.001
2.8
2.3
3.1
5.1
< 0.001
3.0
FHx of cancer
No
554,083
725,060
189,415
1,468,558
1,103
2069
854
4026
2.0
2.9
4.5
< 0.001
2.7
< 0.001
2.1
3.1
4.8
< 0.001
2.9
< 0.001
Yes
121,122
147,564
34,493
303,179
297
441
174
912
2.5
3.0
5.0
< 0.001
3.0
2.6
3.2
5.4
< 0.001
3.2
Height
(cm)
<154
264,445
431,540
124,997
820,982
628
1419
613
2657
2.4
1.5
4.9
< 0.001
3.2
< 0.001
2.5
3.5
5.2
< 0.001
3.5
< 0.001
154–156
210,014
285,243
73,169
568,426
458
873
360
1691
2.7
4.0
5.7
< 0.001
4.1
2.3
3.3
5.3
< 0.001
3.2
157–160
285,846
344,154
80,677
710,677
606
946
302
1854
2.1
3.2
4.9
< 0.001
3.3
2.3
2.9
4.0
< 0.001
2.8
≥ 161
256,227
259,748
53,426
569,401
427
589
218
1234
1.8
2.6
3.7
< 0.001
2.7
1.8
2.4
4.4
< 0.001
2.3
Weight (kg)
< 52
413,406
340,565
29,049
783,020
770
943
121
1834
1.9
2.3
4.2
< 0.001
2.3
< 0.001
2.0
3.0
4.5
< 0.001
2.5
<0.001
52–55
255,743
285,512
38,948
580,203
542
806
174
1522
2.1
1.9
4.5
< 0.001
2.6
2.3
3.0
4.8
< 0.001
2.8
56–61
245,247
375,870
83,951
705,068
578
1127
377
2082
2.4
1.5
4.5
< 0.001
3.0
2.5
3.2
4.8
< 0.001
3.2
≥ 62
102,136
318,738
180,321
601,195
229
951
821
2001
2.2
0.7
4.6
< 0.001
3.3
2.4
3.2
4.9
< 0.001
3.6
BMI (kg/m2)
< 18.5
69,872
34,585
1,015
105,472
113
69
4
186
1.6
3.3
3.9
< 0.001
1.8
< 0.001
1.7
2.1
4.3
< 0.001
1.9
< 0.001
18.5–22.9
650,264
583,930
51,625
1,285,819
1,270
1518
203
2991
2.0
2.2
3.9
< 0.001
2.3
2.1
2.8
4.2
< 0.001
2.5
23.0-24.9
199,271
340,173
69,405
608,849
488
1055
309
1852
2.4
1.4
4.5
< 0.001
3.0
2.6
3.3
4.8
< 0.001
3.3
25.0–29.9
95,519
328,761
164,397
588,677
242
1083
770
2095
2.5
0.7
4.7
< 0.001
3.6
2.7
3.5
5.0
< 0.001
3.8
≥ 30.0
1,606
33,236
45,827
80,669
6
102
207
315
3.7
0.2
4.5
< 0.001
3.9
4.0
3.3
4.8
< 0.001
4.2
Total cholesterol (mg/dL)
<171
297,122
339,711
56,297
693,130
506
699
218
1423
1.7
1.5
3.9
< 0.001
2.1
< 0.001
1.8
2.2
4.2
< 0.001
2.2
< 0.001
171-192
285,233
308,510
65,401
659,144
528
852
289
1669
1.9
1.7
4.4
< 0.001
2.5
2.0
3.0
4.7
< 0.001
2.7
193–217
250,903
323,283
84,889
659,075
598
1033
382
2013
2.4
1.8
4.5
< 0.001
3.1
2.6
3.4
4.8
< 0.001
3.3
≥ 218
183,274
349,181
125,682
658,137
487
1243
604
2334
2.7
1.4
4.8
< 0.001
3.5
2.8
3.8
5.2
< 0.001
3.8
ALT (IU/L)
<13
414,706
370,487
39,262
824,455
752
884
158
1794
1.8
2.0
4.0
< 0.001
2.2
< 0.001
1.9
2.6
4.3
< 0.001
2.3
< 0.001
13–15
219,168
259,979
43,677
522,824
453
754
183
1390
2.1
1.7
4.2
< 0.001
2.7
2.2
3.1
4.5
< 0.001
2.8
16-21
230,941
354,345
88,877
674,163
554
1094
407
2055
2.4
1.6
4.6
< 0.001
3.0
2.6
3.3
4.9
< 0.001
3.3
≥ 22
151,717
335,874
160,453
648,044
360
1095
745
2200
2.4
1.1
4.6
< 0.001
3.4
2.5
3.5
5.0
< 0.001
3.6
Hemoglobin (g/dL)
> 10
986,791
1,281,381
325,615
2,593,787
2,060
3742
1464
7266
2.1
1.6
4.5
< 0.001
2.8
< 0.001
2.2
3.1
4.8
< 0.001
3.0
< 0.001
8–10
25,669
32,557
5,620
63,846
51
74
25
150
2.0
1.6
4.4
< 0.001
2.3
2.1
2.4
4.8
< 0.001
2.5
< 8
4,005
6,650
1,005
11,660
8
11
4
23
2.0
1.2
4.0
< 0.001
2.0
2.1
1.8
4.3
< 0.001
2.1
Serum creatinine
(mg/dL)
≤ 1.5
999,732
1,298,976
326,608
2,625,316
2,090
3774
1466
7330
2.1
1.6
4.5
< 0.001
2.8
< 0.001
2.2
3.1
4.8
< 0.001
3.0
< 0.001
> 1.5
16,732
21,632
5,649
44,013
29
53
27
109
1.7
1.3
4.8
< 0.001
2.5
1.9
2.6
5.1
< 0.001
2.7
MetS metabolic syndrome, FHx family history, BMI body mass index, ALT alanine aminotransferase
The ID of CRC in women was 3.0 (Table 2). The ID of CRC by MetS stage was 2.2, 3.1, and 4.8 for the normal, pre-MetS, and MetS groups, respectively (p < 0.001).
The number of study population and the CIR and the ID of CRC for the other variables in women were summarized in Table 2.
Anzeige
Risk of the number of Mets components on CRC in men and women
In men, the HR of CRC for the pre-MetS group compared with the normal group before adjustment (unadjusted model) was 1.50 (95% CI 1.43–1.57), and the HR for the MetS group was 2.16 (95% CI 2.06–2.28) (Table 3). After full adjustment (model 5), the HR for the pre-MetS group and MetS group was 1.25 (95% CI 1.17–1.33) and 1.54 (95% CI 1.43–1.65).
Table 3
Hazard ratios of the cumulative incidence rate of colorectal cancer according to the progression of metabolic syndrome in men
Characteristics
Category
HR (95% CI)
Non–Adjusted
aModel 1
bModel 2
cModel 3
dModel 4
eModel 5
Metabolic syndrome
(MetS) stage
Normal
Ref.
Ref.
Ref.
Ref.
Ref.
Ref
Pre-MetS
1.50 (1.43 1.57)
1.24 (1.19 1.30)
1.22 (1.17 1.28)
1.26 (1.19 1.34)
1.24 (1.17 1.32)
1.25 (1.17 1.33)
MetS
2.16 (2.06 2.28)
1.58 (1.50 1.66)
1.54 (1.46 1.62)
1.59 (1.49 1.70)
1.54 (1.43 1.65)
1.54 (1.43 1.65)
Age
30–39
Ref.
Ref.
Ref.
Ref.
Ref.
40–49
1.86 (1.53 2.26)
1.85
(1.52
2.26)
1.88
(1.48
2.39)
1.88
(1.48
2.40)
1.88
(1.48
2.40)
50–59
5.36 (4.42 6.52)
5.35 (4.39 6.51)
5.31 (4.19 6.73)
5.31 (4.20 6.74)
5.35 (4.22 6.79)
60
12.49 (10.29 15.16)
12.48 (10.26 15.18)
12.48 (9.85 15.80)
12.52 (9.87 15.89)
12.67 (10.00 16.06)
Smoking
Non-smoker
Ref.
Ref.
Ref.
Ref.
Ex-smoker
1.05 (1.00 1.10)
1.07 (1.01 1.13)
1.07 (1.01 1.13)
1.07 (1.01 1.13)
Smoker
1.17 (1.12 1.22)
1.19 (1.13 1.26)
1.19 (1.13 1.26)
1.19 (1.13 1.25)
Alcohol consumption
No drink
Ref.
Ref.
Ref.
Ref.
2–3/per month
1.15 (1.01 1.10)
1.05 (1.00 1.10)
1.05 (1.00 1.11)
1.05 (1.00 1.11)
1–4/per week
1.19 (1.13 1.25)
1.16 (1.09 1.24)
1.16 (1.10 1.24)
1.16 (1.10 1.24)
5/per week
1.38 (1.29 1.47)
1.40 (1.29 1.51)
1.39 (1.29 1.51)
1.39 (1.29 1.51)
Physical exercise, per week
No exercise
Ref.
Ref.
Ref.
Ref.
1–4/per week
0.97 (0.93 1.02)
0.97 (0.97 1.02)
0.97 (0.92 1.03)
0.97 (0.92 1.03)
5/per week
0.97 (0.94 1.01)
0.98 (0.94 1.03)
0.99 (0.94 1.04)
0.99 (0.94 1.04)
FHx of cancer
No
Ref.
Ref.
Ref.
Yes
0.99 (0.93 1.05)
0.99 (0.93 1.05)
0.99 (0.93 1.04)
Hemoglobin (g/dL)
> 12
Ref
Ref
10–12
1.24 (0.97 1.60)
1.24 (0.97 1.59)
< 10
2.03 (1.24 3.30)
2.02 (1.24 3.30)
Serum creatinine
(mg/dL)
≤ 1.5
Ref
Ref
> 1.5
0.92 (0.84 1.02)
0.92 (0.84 1.02)
fTotal cholesterol (mg/dL)
1.02 (1.01 1.03)
1.02 (1.01 1.03)
fALT (IU/L)
1.01 (1.00 1.02)
1.01 (1.00 1.02)
Height (cm)
< 167
Ref
167–171
0.98 (0.93 1.04)
172-174
0.95 (0.89 1.01)
≥ 175
0.97 (0.90 1.05)
Weight (kg)
< 64
Ref
64–69
1.00 (0.94 1.05)
70–76
0.97 (0.91 1.03)
≥77
1.05 (0.98 1.13)
BMI
(kg/m2)
< 18.5
1.11 (0.94 1.32)
18.5–22.9
Ref.
23.0–24.9
0.98 (0.93 1.03)
25.0–29.9
1.00 (0.95 1.06)
≥ 30.0
1.16 (1.04 1.30)
Values are presented as β (95% confidence interval)
MetS metabolic syndrome, Ref reference, FHx family history, ALT alanine aminotransferase, BMI body mass index
aModel 1: adjusted for age
bModel 2: adjusted for age, smoking, alcohol consumption, and exercise
cModel 3: adjusted for age, smoking, alcohol consumption, exercise, FHx of cancer
dModel 4: adjusted for age, smoking, alcohol consumption, exercise, FHx of cancer, hemoglobin, serum creatinine, total cholesterol and ALT, hight, weight
eModel 5: adjusted for age, smoking, alcohol consumption, exercise, FHx of cancer, hemoglobin, serum creatinine, total cholesterol and ALT, BMI
fTotal cholesterol, fALT: These continuous variables were analyzed by increasing 10 units in the original data to improve the readability of the hazard ratio analysis
In women, the HR of CRC for the pre-MetS group compared with the normal group before adjustment (unadjusted model) was 1.39 (95% CI 1.32–1.46), and the HR for the MetS group was 2.15 (95% CI 2.01–2.30) (Table 4). After full adjustment (model 5), the HR for the pre-MetS group and MetS group was 1.09 (95% CI 1.02–1.17) and 1.39 (95% CI 1.26–1.53).
Table 4
Hazard ratios of the cumulative incidence rate of colorectal cancer according to the progression of metabolic syndrome in women
Characteristics
Category
HR (95% CI)
Non–Adjusted
aModel 1
bModel 2
cModel 3
dModel 4
eModel 5
Metabolic syndrome
(MetS) stage
Normal
Ref.
Ref.
Ref.
Ref.
Ref
Ref.
Pre-MetS
1.39 (1.32 1.46)
1.14 (1.08 1.20)
1.13 (1.07 1.19)
1.12 (1.05 1.20)
1.09 (1.01 1.16)
1.09 (1.02 1.17)
MetS
2.15 (2.01 2.30)
1.50 (1.40 1.61)
1.49 (1.39 1.59)
1.52 (1.40 1.66)
1.39 (1.26 1.52)
1.39 (1.26 1.53)
Age
30-39
Ref.
Ref.
Ref.
Ref.
Ref.
40-49
1.74 (1.28 2.36)
1.77 (1.30 2.41)
1.66 (1.15 2.40)
1.62 (1.12 2.34)
1.64 (1.13 2.37)
50-59
4.03 (2.99 5.43)
4.19 (3.10 5.66)
3.89 (2.71 5.59)
3.71 (2.58 5.33)
3.76 (2.62 5.40)
60
6.38 (4.74 8.60)
6.74 (4.99 9.12)
6.37 (4.44 9.15)
5.97 (4.15 8.60)
6.04 (4.20 8.68)
Smoking
Non-smoker
Ref.
Ref.
Ref.
Ref.
Ex-smoker
1.25 (1.06 1.49)
1.32 (1.08 1.60)
1.31 (1.07 1.59)
1.32 (1.08 1.60)
Smoker
1.25 (1.10 1.41)
1.23 (1.06 1.43)
1.23 (1.06 1.43)
1.23 (1.06 1.43)
Alcohol consumption
No drink
Ref.
Ref.
Ref.
Ref.
2–3/per month
1.11 (1.05 1.18)
1.12 (1.05 1.21)
1.12 (1.05 1.21)
1.12 (1.05 1.20)
1–4/per week
1.20 (1.05 1.36)
1.13 (0.96 1.33)
1.13 (0.96 1.33)
1.13 (0.96 1.33)
5/per week
0.99 (0.79 1.24)
1.07 (0.82 1.40)
1.07 (0.82 1.40)
1.07 (0.82 1.40)
Physical exercise, per week
No exercise
Ref.
Ref.
Ref.
Ref.
1–4/per week
1.04 (0.98 1.10)
1.03 (0.96 1.11)
1.03 (0.96 1.11)
1.03 (0.96 1.12)
5/per week
0.96 (0.91 1.01)
0.96 (0.89 1.02)
0.95 (0.89 1.02)
0.96 (0.90 1.02)
FHx of cancer
No
Ref.
Ref.
Ref.
Yes
1.09 (1.01 1.17)
1.09 (1.01 1.17)
1.09 (1.01 1.17)
Hemoglobin
(g/dL)
> 10
Ref
Ref
8–10
1.15 (0.94 1.40)
1.15 (0.94 1.40)
< 8
0.91 (0.55 1.51)
0.90 (0.54 1.50)
Serum creatinine
(mg/dL)
≤ 1.5
Ref
Ref
> 1.5
0.92 (0.75 1.13)
0.92 (0.75 1.13)
eTotal cholesterol(mg/dL)
1.03 (1.01 1.04)
1.03 (1.01 1.04)
fALT (IU/L)
1.00 (0.99 1.02)
1.00 (0.99 1.02)
Height (cm)
< 154
Ref
154–156
1.05 (0.97 1.13)
157–160
0.91 (0.88 1.03)
≥161
0.94 (0.85 1.03)
Weight (kg)
< 52
Ref
52–55
1.03 (0.94 1.12)
56–61
1.09 (1.01 1.19)
≥ 62
1.18 (1.08 1.29)
eBMI (kg/m2)
< 18.5
1.01 (0.84 1.22)
18.5–22.9
Ref
23.0–24.9
1.09 (1.01 1.17)
25.0–29.9
1.13 (1.05 1.22)
≥ 30.0
1.20 (1.03 1.39)
Values are presented as β (95% confidence interval)
MetS metabolic syndrome, Ref reference, FHx family history, ALT alanine aminotransferase, BMI body mass index
aModel 1: adjusted for age
bModel 2: adjusted for age, smoking, alcohol consumption, and exercise
cModel 3: adjusted for age, smoking, alcohol consumption, exercise, FHx of cancer
dModel 4: adjusted for age, smoking, alcohol consumption, exercise, FHx of cancer, hemoglobin, serum creatinine, total cholesterol and ALT, hight, weight
eModel 5: adjusted for age, smoking, alcohol consumption, exercise, FHx of cancer, hemoglobin, serum creatinine, total cholesterol and ALT, BMI
fTotal cholesterol, fALT: These continuous variables were analyzed by increasing 10 units in the original data to improve the readability of the hazard ratio analysis
Discussion
To investigate the effect of MetS on the incidence of CRC, we analyzed about 6 million medical claim and check-up data that had the high external validity with large sample size and average follow-up period of 9.3 years per person from NHIS in Korea.
The results of the study revealed there is a positive association between MetS and the incidence of CRC. This supports previous findings that MetS is the main factor expediting tumor growth [18‐20].
Anzeige
After analyzing 18 studies (687,413 people) of MetS and CRC, Jinjuvadia et al. [18] reported that MetS increases the occurrence of CRC [relative rate (RR), 1.30; 95% CI 1.18–1.43] and colorectal adenoma (RR, 1.37; 95% CI 1.26–1.494). The Atherosclerosis Risk in Communities (ARIC) follow-up observational study of 14,000 Americans identified MetS as the main factor (RR, 1.49; 95% CI 1.0–2.4) responsible for the occurrence of CRC [19]. With 12 years of follow-up, the Metabolic Syndrome and Cancer Project determined that the risk of CRC in men and women increased by 1.25 (95% CI 1.18–1.32) and 1.14 (95% CI 1.06–1.22), respectively [20]. Numerous other studies also identify MetS as increasing the risk of CRC [21, 22].
Although the mechanism of MetS in the development of CRC is not clear, it is thought to be related to hyperinsulinism and insulin resistance [23, 24], which increases insulin-like growth factor-1 levels. In addition, adipocyte-secreted hormones such as adiponectin, leptin, and resistin [25]; the greater proportion of Firmicutes and lower proportion of Bacteroidetes within the large intestine [26, 27]; and a high-fat, low-fiber diet [28] are all related to the occurrence of CRC. Since it is important to understand the pathological mechanism of CRC related to MetS [29], additional studies need to examine how to prevent CRC as MetS progresses [30].
In this study, pre-MetS group had a 25% higher risk of CRC and MetS group had a 54% higher risk of CRC than Normal group in men based on the full adjustment model, and pre-MetS group had a 9% higher risk of CRC and MetS group had a 39% higher risk of CRC than Normal group in women. The risk of Mets on incidence of CRC was slightly higher in men than in women, and all were significant.
However, some previous studies showed different risk ratios by gender [10, 12, 31, 32]. The reason for this is difference in the number of study subjects, study design, and fundamental biological differences between males and females. Although this study showed the positive association between MetS and the incidence of CRC, further investigation is needed as to the question of how men and women may be affected by metabolic abnormalities in terms of CRC risk [13].
Lifestyle has been known to be a major factor associated with CRC, and it is known that the risk of cancer gradually decreases if healthy lifestyles are practiced in stages [33]. This study found that drinking alcohol and smoking were related to the risk of CRC. Other studies also shown that smoking [34] and drinking alcohol have carcinogenic effects and promote cancer, especially in the case of drinking alcohol, moderate amounts of drinking alcohol can lead to an increase in CRC [35]. Education programs for lifestyle improvements should be conducted to reduce the incidence of colorectal cancer [36]. It has been suggested as an effective way to raise awareness, knowledge, and screening rate for colorectal cancer screening [37, 38].
There are some studies [39‐41] that consider height in the analysis as risk factors for colorectal cancer, and the results of the analysis are also reported to be significant. Based on these similar previous studies, this study considered the relationship between height and colorectal cancer but found no significant correlation. There was also no significant association between weight and colorectal cancer.
Previous studies of Asian races have limits, such as a lack of representativeness or only a few variables were examined. Our study identified MetS as a risk factor for CRC after adjusting for various variables and determined the magnitude of the risk using national health checkup and insurance claims data in Korea. Because there are ethnic differences in the relationship between MetS and cancer [13, 25, 42], the presentation of risk ratios for Asians using large data and long follow-up period sets is a significant research achievement. Study on the relationship between metabolic and disease occurrence for Asian races [43] should be conducted and other approaches to mitigate health risks need to be reviewed [44]. In addition, this study has some newly informative knowledge as follows. First, this study is meaningful in that it attempts to represent the entire Korean population using the National Health Insurance Corporation's claim data. Second, this study is the most recent long-term data taken from the initial establishment of the data source to the latest data. Third, unlike other studies, this study considered the incubation period sufficiently to closely observe the association of colorectal cancer risk due to metabolic syndrome. The incubation period was set to 3 years, and the analysis was performed except for patients with colorectal cancer who developed within 3 years of the observation. Fourth, this study was able to understand the risk of colorectal cancer in more detail by classifying subjects according to the number of factors of metabolic syndrome rather than whether they were metabolic syndrome.
This study has some limitations. First, we did not consider changes in the risk of the number of metabolic components on CRC in men and women in the study population after it was diagnosed, since we used only the health check-up data for 2009. This should be examined in a future study. Second, we did not adjust for some variables, such as the consumption of meat, which is a known risk factor for CRC, because we used secondary data. Lastly, health screening programs are being provided to all Koreans regardless of income, and the participation rate is also increasing from 72.9% in 2012 to 78.5% in 2017 [17]. However, it is mandatory for industrial workers to be screened, and the rate of participation of industrial workers is relatively higher than that of self-employees [45]. Therefore, this study cannot exclude the possibility of health workers effect.
Conclusions
This study investigated the relationship between MetS and CRC using national health check-up and insurance claims data for Korea. It showed that MetS was a risk factor for the occurrence of CRC.
From a clinical and public health perspective, Mets has emerged as an important disease that requires early management and more thorough management and prevention of Mets are needed to prevent CRC based on the results of this study with long-term follow-up and large-scale of Asian subjects.
Acknowledgments
We thank the NHIS for permission to use the NHIS data (NHIS-2019-1-404).
Ethics approval and consent to participate
This study was reviewed and approved by the Institutional Review Boards of Konkuk University(7001355-201909-E-100). This study used the secondary data of NHIS with omitting individual’s information, so there was no need of the informed consents from the study population.
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
