The prevalence of and demographic factors associated with radiographic knee osteoarthritis in Korean adults aged ≥ 50 years: The 2010-2013 Korea National Health and Nutrition Examination Survey.

BACKGROUND: To reduce the social burden of knee osteoarthritis (OA) by addressing it in the early stages in the population at greatest risk, the epidemiology of knee OA needs to be understood and associated demographic factors need to be identified.
OBJECTIVES: We evaluated the weighted prevalence of and demographic factors associated with radiographic knee OA in Korean adults.
METHODS: We analyzed data from 12,287 individuals aged ≥ 50 years who had radiographs of the knee taken in the 2010-2013 Korea National Health and Nutrition Examination Survey (KNHANES). Radiographic knee OA was defined based on the Kellgren-Lawrence grade, as follows: 0: No abnormal finding 1: Mild degenerative changes, minute osteophytes 2: Mild knee OA, definite osteophytes 3: Moderate knee OA, moderate joint-space narrowing and definite osteophytes 4: Severe knee OA, severe joint-space narrowing with subchondral sclerosis.
RESULTS: We found that the prevalence of radiographic knee OA in the Korean adult population was 35.1%. Logistic regression analyses were performed to identify factors associated independently with radiographic knee OA, with age, sex, area of residence, education level, household income, and obesity serving as covariates. Women were at greater risk than men of having knee OA (OR 2.12, 95% CI 1.90-2.37, p < 0.001). Compared with subjects aged 50-59 years, adults aged ≥ 80 years were at 8.87-fold (95% CI 7.12-11.06, p < 0.001) greater risk of having knee OA. Residence in a rural area was associated with a greater risk of having radiographic knee OA than was residence in an urban area (OR 1.26, 95% CI 1.08-1.48, p = 0.004), regardless of knee OA severity (Kellgren-Lawrence grades ≥2, ≥3, and 4). Elementary school graduates had 1.71-fold (p < 0.001) greater risks of having knee OA than did college graduates. Household incomes ≤24th percentile were associated with a greater risk of having knee OA compared with those ≥75th percentile (OR 1.28, 95% CI 1.08-1.52, p = 0.004). Obesity was associated with an approximately two-fold greater risk of knee OA, regardless of knee OA severity (Kellgren-Lawrence grades ≥2, ≥3, and 4).
CONCLUSIONS: Using data from the 2010-2013 KNHANES and defining knee OA as Kellgren-Lawrence grade ≥ 2, we found that the prevalence of radiographic knee OA was 35.1% (24.4% in men, 44.3% in women) in a representative sample of Korean adults aged ≥ 50 years, with the highest prevalence (78.7%) observed in women aged ≥ 80 years. Low socioeconomic status and traditional factors, including age, female sex, and obesity, were associated with the risk of knee OA.

Introduction

Osteoarthritis (OA) of the knee is a complex peripheral joint disorder with multiple risk factors that results in progressive loss of function, pain, and stiffness [1]. Frequent knee pain affects approximately 25% of adults, and OA is the most common cause of knee pain in people older than 50 years [2,3].

Clinically, knee OA consists of joint symptoms and evidence of structural change, usually demonstrated radiographically [3]. According to the European League Against Rheumatism’s recommendations, as no guideline for the clinical diagnosis of knee OA is currently available, plain radiography is often used as the gold standard for the assessment of knees with clinical evidence of OA [4]. Most studies have involved radiographic assessment as the primary means of identifying OA, with the Kellgren–Lawrence scale used to grade OA based on the definite presence of osteophytes [5]. The presence of osteophytes in knee OA correlates well with its symptoms [6].

Given that aging and obesity are major risk factors for knee OA, this disorder and total knee replacement have become substantially more common in recent decades [7,8]. According to the 2010 Global Burden of Diseases study, the burden of OA is increasing, most rapidly among musculoskeletal disorders in terms of disability-adjusted life years; it will impose new challenges on health systems, along with mental disorders and diabetes [9]. To reduce the social burden of knee OA by addressing it in the early stages in the population at greatest risk, the epidemiology of knee OA needs to be understood and associated demographic factors need to be identified.

In this study, we investigated the prevalence of and demographic factors associated with radiographic knee OA based on the Kellgren–Lawrence grade in Korean adults aged ≥ 50 years using data from the 2010–2013 Korea National Health and Nutrition Examination Survey (KNHANES).

Methods

Study population and data collection

This study was based on data from the 2010–2013 KNHANES, a cross-sectional, nationally representative survey conducted by the Korean Center for Disease Control for Health Statistics. As described in detail previously [10,11], the KNHANES is independent dataset obtained from the general population of Korea, similar to data from the National Health and Nutrition Examination Survey (NHANES) in the United States. The KNHANES has been conducted periodically since 1998 to assess the health and nutritional status of the civilian, noninstitutionalized population of Korea. Participants are selected using proportional-allocation systematic sampling with multistage stratification. Standardized interviews are conducted in the homes of the participants to collect information on demographic variables, family and medical histories, medications used, and various other health-related variables. The interviewers use an established questionnaire to record the demographic and socioeconomic characteristics of the subjects, including age, education level, occupation, household income, marital status, smoking status, alcohol consumption, exercise habits, previous and current diseases, and family disease history.

Of the 33,552 participants in the 2010–2013 KNHANES, data from 12,287 individuals aged ≥ 50 years who had radiographs of the knee taken were analyzed in this study.

Assessment of radiographic knee OA

Bilateral anteroposterior and lateral plain radiographs of the knees were taken using a DigiRAD-PG 9M. Two radiologists performed OA examinations and independent assessments by webhard uploading and downloading using the Kellgren–Lawrence grading system. For differences of one grade between radiologists, the higher grade was accepted. For discrepancies exceeding one grade, a third radiologist was consulted, and the grade concordant with the third assessment was accepted. Radiographic OA grading agreement rates for 2010–2013 were 87.96%, 95.18%, 89.62%, and 85.19%, respectively. Weighted kappa coefficients for inter-rater reliability in 2010–2013 were 0.6522, 0.7407, 0.8383, and 0.6842, respectively, indicating fair to very high degrees of agreement.

Radiographic knee OA was defined based on the Kellgren–Lawrence grade, as follows:

0: No abnormal finding

1: Mild degenerative changes, minute osteophytes

2: Mild knee OA, definite osteophytes

3: Moderate knee OA, moderate joint-space narrowing and definite osteophytes

4: Severe knee OA, severe joint-space narrowing with subchondral sclerosis

Ethical considerations

The institutional review board of Ilsan Paik Hospital, Republic of Korea, approved this study. After the study proposal had been approved, the KNHANES dataset was made available at the request of the investigator. The study was exempt from the requirement for consent because the dataset did not include personal information and KNHANES participants had already given consent.

Statistical analyses

The KNHANES participants were not sampled randomly. The survey was designed using a complex, stratified, multistage probability-sampling model; consequently, individual participants were not equally representative of the Korean population. To obtain representative prevalence rates from the dataset, consideration of the power of each participant (sample weight) as a representative of the Korean population was necessary. Following approval from the Korea Centers for Disease Control and Prevention, we received a survey dataset that included information on the survey location, age, sex, and various other factors and the sample weight for each participant. The survey sample weights, which were calculated using the sampling and response rates and age/sex proportions of the reference population (2005 Korean National Census Registry), were used in all of the analyses to produce representative estimates of the noninstitutionalized Korean civilian population. The statistical analyses were performed using SPSS ver. 21.0 for Windows (SPSS, Chicago, IL, USA). To compare the weighted prevalence of radiographic knee OA by sociodemographic factors, chi-squared tests and general linear model were performed. The prevalence of radiographic knee OA was analyzed using age (50–59, 60–69, 70–79, ≥80 years old), sex (men/women), area of residence (urban [Dong]) /rural [Eup/Myeon]), education level (elementary school/junior high school/senior high school/college graduated), number of family members (1/2/3/≥4), household income (≤ 24th, 25-49th, 50-74th,≥75th percentile), occupation (managers and professionals/clerical support workers/service and sales workers/skilled agricultural, forestry and fishery workers/craft, plant, or machine operators and assemblers/laborers/unemployed (including students and house wives), and obesity [body mass index (BMI) ≥ 25 kg/m2] as covariates. Logistic regression analyses were used to calculate the odds ratio (OR) for radiographic knee OA with age(50–59, 60–69, 70–79, ≥80 years old), sex(men/women), area of residence(urban/rural), education level(elementary school/junior high school/senior high school/college graduated), household income(≤ 24th, 25-49th, 50-74th,≥75th percentile), and obesity(no/yes) serving as covariates. All tests were two sided and p < 0.05 was considered to be indicative of statistical significance.

Results

Weighted demographic and clinical characteristics of the study population

The weighted demographic and clinical characteristics of the study population are shown in Table 1 shows the weighted demographic and clinical characteristics of the study population. The mean weighted age was 62.5 years [95% confidence interval (CI) 62.2–62.7], and 54% of the participants were female. The weighted percentage of obesity (BMI ≥ 25 kg/m2) was 35.2%. The weighted percentages of Kellgren–Lawrence grades 0–4 were 40.8%, 24.0%, 14.3%, 14.6%, and 6.2%, respectively. Based on Kellgren–Lawrence grade ≥ 2, we found that the prevalence of radiographic knee OA in the Korean adult population was 35.1%.

Table 1

Demographic and clinical characteristics of the study population.

Variables	Unweighted Number (%)	Weighted Number (%)
Total	12,287	14,837,279
Sex
Men	5,231 (42.6)	6,846,055 (46.1)
Women	7,056 (57.4)	7,991,224 (53.9)
Age (years)
50–59	4,513 (36.7)	7,006,966 (47.2)
60–69	3,967 (32.3)	4,079,939 (27.5)
70–79	3,075 (25.0)	2,922,903 (19.7)
80-	732 (6.0)	827,471 (5.6)
Area of residence
Urban	8,966 (73.0)	10,889,741 (73.4)
Rural	3,321 (27.0)	3,947,538 (26.6)
Education
Elementary school graduated	6,093 (49.6)	7,048,684 (47.5)
Junior high school graduated	2,027 (16.5)	2,629,140 (17.7)
Senior high school graduated	2,793 (22.7)	3,534,363 (23.8)
College graduated	1,374 (11.2)	1,625,093 (11.0)
Family member (n)
1	1,497 (12.2)	1,561,237 (10.5)
2	5,132 (41.8)	5,455,879 (36.8)
3	2,841 (23.1)	3,852,005 (26.0)
≥ 4	2,817 (22.9)	3,968,122 (26.7)
Household income
≤ 24th percentile	4,048 (32.9)	4,511,878 (30.4)
25-49th percentile	3,132 (25.5)	3,748,286 (25.3)
50-74th percentile	2,484 (20.2)	3,181,774 (21.4)
≥ 75th percentile	2,623 (21.3)	3,395,841 (22.9)
Occupation
Managers and professionals	632 (5.1)	845,042 (5.7)
Clerical support workers	321 (2.6)	451,428 (3.0)
Service and sales workers	1,176 (9.6)	1,665,965 (11.2)
Skilled agricultural, forestry and fishery workers	1,496 (12.2)	1,661,755 (11.2)
Craft, plant, or machine operators and assemblers	960 (7.8)	1,542,072 (10.4)
Laborers	1,373 (11.2)	1,674,336 (11.3)
Unemployed(including students and house wives)	6,329 (51.5)	6,996,681 (47.2)
Obesity (BMI ≥ 25kg/m2)	4,269 (34.7)	5,225,644 (35.2)
Kellgren-Lawrence grade
0	4,690 (38.2)	6,060,519 (40.8)
1	2,944 (24.0)	3,563,758 (24.0)
2 (mild)	1,848 (15.0)	2,121,220 (14.3)
3 (moderate)	1,929 (15.7)	2,167,619 (14.6)
4 (severe)	876 (7.1)	924,165 (6.2)

Weighted prevalence of radiographic knee OA according to age and sex

The weighted prevalences of Kellgren–Lawrence grades ≥2, ≥3, and 4 in the study population were 35.1% (33.7–36.6%), 20.8% (19.8–22.0%), and 6.2% (5.7–6.8%), respectively (Table 2). The weighted prevalence of radiographic knee OA increased with age, irrespective of knee OA severity (Kellgren–Lawrence grades ≥2, ≥3, and 4). The overall weighted prevalence of radiographic knee OA in adults aged ≥ 80 years was 71.6% (67.6–75.3%).

Table 2

Weighted prevalence of radiographic knee osteoarthritis in the Korean adults (≥ 50 years).

	Number (unweighted/weighted)	Prevalence of mild to severe knee osteoarthritis (Kellgren-Lawrence grade ≥ 2)	Prevalence of moderate to severe knee osteoarthritis (Kellgren-Lawrence grade ≥3)	Prevalence of severe knee osteoarthritis (Kellgren-Lawrence grade = 4)
Total
Total	12,287/14,837,279	35.1 (33.7–36.6)	20.8 (19.8–22.0)	6.2 (5.7–6.8)
50–59 years old	4,513/7,006,966	18.9 (17.4–20.6)	8.3 (7.3–9.4)	1.0 (0.8–1.4)
60–69 years old	3,967/4,079,939	39.6 (37.4–41.8)	22.0 (20.3–23.8)	5.5 (4.7–6.5)
70–79 years old	3,075/2,922,903	57.5 (55.2–59.7)	39.2 (36.9–41.4)	13.8 (12.2–15.5)
≥ 80 years old	732/827,471	71.6 (67.6–75.3)	56.4 (52.0–60.7)	26.9 (23.1–31.1)
Men
Total	5,231/6,846,055	24.4 (22.7–26.2)	10.2 (9.2–11.3)	2.1 (1.7–2.6)
50–59 years old	1,872/3,467,680	14.3 (12.4–16.4)	4.8 (3.8–6.2)	0.6 (0.3–1.1)
60–69 years old	1,780/1,934,143	27.8 (25.1–30.7)	10.5 (8.8–12.5)	1.8 (1.2–2.7)
70–79 years old	1,326/1,191,503	41.8 (38.4–45.3)	20.5 (18.0–23.3)	5.4 (4.2–7.0)
≥ 80 years old	253/252,729	55.5 (57.5–63.1)	33.9 (26.9–41.7)	10.7 (6.3–17.5)
Women
Total	7,056/7,991,244	44.3 (42.7–46.0)	29.9 (28.4–31.5)	9.7 (8.9–10.6)
50–59 years old	2,641/3,539,286	23.5 (21.4–25.7)	11.7 (10.2–13.4)	1.5 (1.1–2.0)
60–69 years old	2,187/2,145,796	50.2 (47.5–52.8)	32.4 (29.9–35.0)	8.9 (7.6–10.5)
70–79 years old	1,749/1,731,400	68.2 (65.6–70.8)	52.0 (49.1–54.9)	19.6 (17.3–22.1)
≥ 80 years old	479/574,743	78.7 (74.5–82.4)	66.3 (61.3–71.0)	34.0 (29.2–39.2)

Data are expressed as mean (95% CI)

In men, the weighted prevalences of Kellgren–Lawrence grades ≥2, ≥3, and 4 were 24.4% (22.7–26.2%), 10.2% (9.2–11.3%), and 2.1% (1.7–2.6%), respectively. The weighted prevalence of radiographic knee OA in men increased with age, irrespective of knee OA severity (Kellgren–Lawrence grades ≥2, ≥3, and 4). The weighted prevalence of radiographic knee OA in men aged ≥ 80 years was 55.5% (57.5–63.1%).

In women, the weighted prevalences of Kellgren–Lawrence grades ≥2, ≥3, and 4 were 44.3% (42.7–46.0%), 29.9% (28.4–31.5%), and 9.7% (8.9–10.6%), respectively. The weighted prevalence of radiographic knee OA was higher in women than in men, irrespective of knee OA severity (Kellgren–Lawrence grades ≥2, ≥3, and 4). The weighted prevalence of radiographic knee OA in women aged ≥ 80 years was 78.7% (74.5–82.4%).

Weighted prevalence of radiographic knee OA according to demographic and clinical characteristics

Table 3 shows the weighted unadjusted and adjusted prevalences of radiographic knee OA according to demographic and clinical variables after adjustment for age, sex, area of residence, education level, number of family members, household income, occupation, and obesity.

Table 3

Weighted prevalence of radiographic knee osteoarthritis (Kellgren-Lawrence grade ≥ 2) according to the demographic and clinical factors.

Variables		Unadjusted		Adjusted for age, sex,		Adjusted for age, sex, and other variables*
Area of residence	Urban	32.4 (30.8–33.9)	Reference	33.7 (32.2–35.2)	Reference	34.0 (32.6–35.5)	Reference
	Rural	42.8 (39.9–45.8)	<0.001	39.1 (36.3–41.9)	0.001	38.2 (35.5–40.9)	0.008
Education			<0.001		<0.001		<0.001
Elementary school graduated		47.5 (45.6–49.5)	Reference	38.4 (36.5–40.3)	Reference	37.6 (35.7–39.5)	Reference
Junior high school graduated		28.9 (26.3–31.4)	<0.001	34.4 (31.9–37.0)	0.007	34.3 (31.7–36.8)	0.020
Senior high school graduated		23.5 (21.5–25.3)	<0.001	32.4 (30.4–34.4)	<0.001	33.4 (31.4–35.5)	0.002
College graduated		16.9 (14.2–19.6)	<0.001	28.0 (25.3–30.8)	<0.001	29.5 (26.7–32.3)	<0.001
Family member (n)			<0.001		0.200		0.855
1		51.0 (47.6–54.4)	Reference	36.7 (33.7–39.6)	Reference	35.6 (32.7–38.6)	Reference
2		39.6 (37.7–41.6)	<0.001	36.1 (34.3–37.9)	0.718	35.5 (33.7–37.3)	0.941
3		28.5 (26.2–30.8)	<0.001	33.6 (31.5–35.8)	0.099	34.4 (32.3–36.5)	0.527
≥ 4		29.2 (26.8–31.5)	<0.001	34.7 (32.5–36.9)	0.293	35.2 (33.0–37.3)	0.806
Household income			<0.001		<0.001		0.010
≤ 24th percentile		50.0 (47.9–52.1)	Reference	38.8 (36.8–40.9)	Reference	37.6 (35.6–39.7)	Reference
25-49th percentile		33.2 (31.0–35.3)	<0.001	33.7 (31.7–35.7)	<0.001	33.4 (31.4–35.4)	0.002
50-74th percentile		29.3 (26.8–31.8)	<0.001	35.2 (32.7–37.6)	0.016	35.5 (33.0–37.9)	0.168
≥ 75th percentile		23.0 (20.7–25.3)	<0.001	31.8 (29.5–34.1)	<0.001	33.4 (31.1–35.7)	0.007
Occupation			<0.001		0.001		0.192
Managers and professionals		14.7 (11.4–17.9)	Reference	31.4 (28.1–34.7)	Reference	33.8 (30.2–37.3)	Reference
Clerical support workers		17.4 (12.3–22.6)	0.357	35.1 (30.0–40.2)	0.203	37.6 (32.6–42.6)	0.182
Service and sales workers		25.1 (22.1–28.1)	<0.001	33.7 (30.8–36.5)	0.290	33.9 (31.2–36.7)	0.942
Skilled agricultural, forestry and fishery workers		39.0 (35.1–42.8)	<0.001	40.4 (36.8–44.1)	<0.001	37.8 (34.4–41.3)	0.112
Craft, plant, or machine operators and assemblers		20.3 (17.1–23.6)	0.016	38.2 (35.0–41.5)	0.002	37.7 (34.4–40.9)	0.104
Laborers		36.3 (33.2–39.5)	<0.001	35.7 (32.6–38.8)	0.065	35.5 (32.5–38.5)	0.464
Unemployed(including students and house wives)		43.2 (41.5–45.0)	<0.001	33.9 (32.1–35.6)	0.203	34.1 (32.4–35.9)	0.853
Obesity (BMI ≥ 25kg/m2)	No	30.2 (28.7–31.8)	Reference	30.0 (28.6–31.4)	Reference	30.1 (28.6–31.5)	Reference
	Yes	44.2 (42.1–46.3)	<0.001	44.6 (42.7–46.5)	<0.001	44.5 (42.6–46.4)	<0.001

Data are expressed as mean (95% CI) *Other variables include area of residence, education level, number of family members, household income, occupation, and obesity.

The unadjusted and adjusted weighted prevalences of radiographic knee OA were lower in urban areas than in rural areas [adjusted, 34.0% (32.6–35.5%) vs. 38.2% (35.5–40.9%), p = 0.008].

Education level was correlated inversely with the prevalence of radiographic knee OA, before and after adjustment. Elementary school graduates had a higher prevalence of radiographic knee OA than did college graduates [adjusted, 37.6% (35.7–39.5%) vs. 29.5% (26.7–32.3%), p < 0.001].

The number of family members was associated negatively with the unadjusted prevalence of radiographic knee OA (p < 0.001). However, after adjustment, the statistical significance did not persist.

Household income was also associated negatively with the prevalence of radiographic knee OA before and after adjustment. Subjects with household incomes ≤24th percentile had higher unadjusted and adjusted prevalences of knee OA than did subjects with household incomes in the 25–49th percentiles, 50–74th percentiles, and ≥75th percentile (p < 0.001).

Regarding occupation, with managers and professionals serving as controls, service and sales workers (p < 0.001); skilled agricultural, forestry, and fishery workers (p < 0.001); assemblers (p = 0.016); laborers (p<0.001); and unemployed subjects (p<0.001) had higher prevalences of knee OA. After adjustment for age and sex, with managers and professionals serving as controls, only skilled agricultural, forestry, and fishery workers (p < 0.001) and assemblers (p = 0.002) had a higher prevalences of knee OA. However, the statistical significance did not persist after adjustment for age, sex, area of residence, education level, number of family members, household income, and obesity.

Obesity was associated positively with a higher prevalence of radiographic knee OA, before and after adjustment (p < 0.001).

Logistic regression analyses of radiographic knee OA

Logistic regression analyses were performed to identify factors associated independently with radiographic knee OA, with age, sex, area of residence, education level, household income, and obesity serving as covariates (Table 4).

Table 4

Logistic regression analyses for radiographic knee osteoarthritis.

		Kellgren-Lawrence grade ≥ 2		Kellgren-Lawrence grade ≥ 3		Kellgren-Lawrence grade = 4
Variables		Odd ratio (95% CI)	P	Odd ratio (95% CI)	P	Odd ratio (95% CI)	P
Sex
Men		Reference		Reference		Reference
Women		2.12 (1.90–2.37)	<0.001	3.10 (2.72–3.54)	<0.001	3.30 (2.67–4.08)	<0.001
Age (years)	50–59	Reference	<0.001	Reference	<0.001	Reference	<0.001
	60–69	2.54 (2.23–2.90)	<0.001	2.60 (2.21–3.07)	<0.001	4.20 (2.97–5.95)	<0.001
	70–79	4.90 (4.25–5.64)	<0.001	5.47 (4.57–6.54)	<0.001	9.70 (6.74–13.97)	<0.001
	80	8.87 (7.12–11.06)	<0.001	10.81 (8.59–13.61)	<0.001	21.21 (14.53–30.97)	<0.001
Area of residence	Urban	Reference		Reference		Reference
	Rural	1.26 (1.08–1.48)	0.004	1.31 (1.11–1.54)	0.001	1.35 (1.09–1.67)	0.005
Education
College graduated		Reference	<0.001	Reference	<0.001	Reference	<0.001
Senior high school graduated		1.32 (1.05–1.66)	0.016	1.21 (0.89–1.66)	0.229	1.08 (0.52–2.21)	0.842
Junior high school graduated		1.45 (1.14–1.85)	0.003	1.37 (1.00–1.89)	0.053	1.83 (0.92–3.65)	0.086
Elementary school graduated		1.71 (1.36–2.14)	<0.001	1.84 (1.36–2.49)	<0.001	2.49 (1.30–4.76)	0.006
Household income
≥ 75th percentile		Reference	0.004	Reference	0.015	Reference	0.026
50-74th percentile		1.15 (0.96–1.36)	0.121	1.07 (0.87–1.32)	0.527	0.83 (0.59–1.18)	0.306
25-49th percentile		1.03 (0.87–1.22)	0.753	1.21 (0.97–1.50)	0.091	0.88 (0.64–1.20)	0.419
≤ 24th percentile		1.28 (1.08–1.52)	0.004	1.36 (1.11–1.66)	0.003	1.17 (0.87–1.58)	0.299
Obesity (BMI ≥ 25kg/m2)
No		Reference		Reference		Reference
Yes		2.10 (1.89–2.34)	<0.001	2.21 (1.97–2.48)	<0.001	2.12 (1.78–2.51)	<0.001

Women were at greater risk than men of having knee OA (Kellgren–Lawrence grade ≥ 2; OR 2.12, 95% CI 1.90–2.37, p < 0.001). Women were at 3.30-fold greater risk of having severe knee OA (Kellgren–Lawrence grade = 4) than were men (p < 0.001).

Compared with subjects aged 50–59 years, adults aged ≥ 80 years were at 8.87-fold (95% CI 7.12–11.06, p < 0.001) greater risk of having knee OA (Kellgren–Lawrence grade ≥ 2). For severe knee OA (Kellgren–Lawrence grade = 4), the risk was more than 20 times (95% CI 14.53–30.97, p < 0.001) greater among adults aged ≥ 80 years than among those aged 50–59 years.

Residence in a rural area was associated with a greater risk of having radiographic knee OA than was residence in an urban area (OR 1.26, 95% CI 1.08–1.48, p = 0.004), regardless of knee OA severity (Kellgren–Lawrence grades ≥2, ≥3, and 4).

Elementary school graduates had 1.71-fold (p < 0.001) and 2.49-fold (p = 0.006) greater risks of having knee OA (Kellgren–Lawrence grade ≥ 2) and severe knee OA (Kellgren–Lawrence grade = 4), respectively, than did college graduates.

Household incomes ≤24th percentile were associated with a greater risk of having knee OA compared with those ≥75th percentile (OR 1.28, 95% CI 1.08–1.52, p = 0.004).

Obesity was associated with an approximately two-fold greater risk of knee OA, regardless of knee OA severity (Kellgren–Lawrence grades ≥2, ≥3, and 4).

Additionally, logistic regression analyses were performed to identify factors associated independently with radiographic knee OA according to sex with age, area of residence, education level, household income, and obesity serving as covariates (Table 5).

Table 5

Logistic regression analyses for radiographic knee osteoarthritis (Kellgren-Lawrence grade ≥ 2) according to sex.

		Men		Women
Variables		Odd ratio (95% CI)	P	Odd ratio (95% CI)	P
Age (years)	50–59	Reference	<0.001	Reference	<0.001
	60–69	2.20 (1.80–2.69)	<0.001	2.79 (2.36–3.30)	<0.001
	70–79	4.08 (3.26–5.10)	<0.001	5.58 (4.64–6.70)	<0.001
	80	7.09 (4.90–10.25)	<0.001	10.33 (7.82–13.65)	<0.001
Area of residence	Urban	Reference		Reference
	Rural	1.16 (0.92–1.47)	0.207	1.33 (1.13–1.58)	0.001
Education
College graduated		Reference	0.002	Reference	0.008
Senior high school graduated		1.42 (1.08–1.86)	0.012	1.23 (0.87–1.73)	0.249
Junior high school graduated		1.51 (1.11–2.06)	0.008	1.40 (0.97–2.01)	0.072
Elementary school graduated		1.74 (1.31–2.32)	<0.001	1.59 (1.14–2.23)	0.007
Household income
≥ 75th percentile		Reference	0.758	Reference	0.001
50-74th percentile		1.09 (0.84–1.41)	0.536	1.19 (0.97–1.48)	0.101
25-49th percentile		1.03 (0.80–1.34)	0.797	1.04 (0.84–1.28)	0.755
≤ 24th percentile		1.14 (0.87–1.49)	0.351	1.39 (1.14–1.70)	0.001
Obesity (BMI ≥ 25kg/m2)
No		Reference		Reference
Yes		1.80 (1.52–2.14)	<0.001	2.30 (2.00–2.65)	<0.001

In men, age, education level, and obesity were associated with radiographic knee OA. However, area of residence and household income were not.

In women, age, area of residence, education level, household income, and obesity were all associated with radiographic knee OA

Discussion

Using data from the 2010–2013 KNHANES, we found that the prevalence of radiographic knee OA was 35.1% (24.4% in men, 44.3% in women) in a representative sample of Korean adults aged ≥ 50 years, with the highest prevalence (78.7%) observed in women aged ≥ 80 years.

Based on the NHANES III, the prevalences of radiographic knee OA and symptomatic knee OA were 37.4% and 12.1%, respectively, among adults aged > 60 years [12]. In the 2002–2005 Framingham Osteoarthritis Study, the age- and BMI-adjusted prevalences of radiographic knee OA in women and men were 35.4% and 35.1%, respectively [7].

Although we could not directly compare the prevalences of radiographic knee OA in general populations among countries because of the use of different data collection and analysis methodologies, the prevalence of radiographic knee OA (Kellgren–Lawrence grade ≥ 2) in South Korea seems to be similar to the global estimate.

Our data also suggest that sociodemographic factors, such as low education level and low household income, are associated with the risk of radiographic knee OA, in addition to the traditional factors of age, female sex, and obesity. However, the number of family members and occupation were not associated with radiographic knee OA after adjusting for age, sex, area of residence, education level, household income, and obesity.

Increasing age and female sex are well-known risk factors for knee OA in all regions [13]. Our findings are in agreement: women had a 2.1-fold greater risk of radiographic knee OA than did men, and persons aged ≥ 80 years had an approximately 9-fold greater risk of radiographic knee OA than did those aged 50–59 years.

Rural residence was also a risk factor for radiographic knee OA in this study, even after adjustment for age, sex, education level, number of family members, household income, occupation, and obesity.

Population-based surveys conducted in urban Beijing and rural Wuchuan County, China, also showed that men and women in Wuchuan had roughly double the prevalence of knee OA compared with their Beijing counterparts [14]. In elderly Japanese population-based cohorts, residents of mountainous areas had a greater risk of radiographic knee OA (Kellgren–Lawrence grade ≥ 2) than did urban residents, indicating the involvement of environmental factors such as nutrition or occupation (e.g., farming, forestry), which demands physical activity and repetitive laborious use of the knee joints [15].

In this study, the weighted prevalence of obesity, defined as BMI ≥ 25 kg/m2, was approximately 35%. Obesity was associated with an approximately two-fold greater risk of knee OA, regardless of knee OA severity (Kellgren–Lawrence grades ≥2, ≥3, and 4). This result was similar to the findings of a meta-analysis, which yielded a pooled OR of 2.1 (95% CI 1.82–2.42), indicating an increased risk of knee OA, in overweight (BMI 25–30 kg/m2) and obese (BMI > 30 kg/m2) individuals [13]. Obesity plays a role in the development and progression of knee OA through variable combinations of mechanical, humeral, and metabolic factors, including elevated adipocytokine levels and associated pro-inflammatory responses, as well as mechanical loading of the knee joint during weight bearing [16].

We found that a low education level and low household income were associated with radiographic knee OA. A few studies have revealed associations between low socioeconomic status and knee OA [17-20]. Callahan et al. reported that low educational attainment, but not occupation, was associated significantly with radiographic knee OA.[18]. According to the China Health and Retirement Longitudinal Study, knee OA was more prevalent in subjects who had received less education than in those who had received more education [21]. A low education level could lead to reduced health literacy and health-promoting activities. Jorgensen et al. suggested that lifestyle differences are responsible, at least in part, for the reduced risk of knee OA in persons with more education and higher than average incomes, based on finding from their study of a Danish cohort [22].

Occupational activity, which includes kneeling, squatting, lifting, and climbing stairs at work, is a modifiable risk factor for the development and progression of knee OA [23]. One study showed that male farmers, construction workers, and firefighters had increased risks of knee OA [24]. Based on 2010–2012 KNHANES data, Kim et al. reported that male low-level workers (skilled agricultural and fishery workers) and blue-collar workers (technicians and device and machine operators) aged ≥ 50 years were at greater risk of knee OA and chronic knee pain [25]. In our study, weighted prevalences of radiographic knee OA were higher in skilled agricultural, forestry, and fishery workers and in craft, plant, or machine operators and assemblers compared with managers and professionals, after adjustment for age and sex. However, after adjustment for age, sex, area of residence, education level, number of family members, household income, and obesity, the statistical significance did not persist.

This study has several strengths. First, we examined a large, nationally representative sample of adult Koreans. To our knowledge, few other studies have been based on national surveillance of knee OA in the general population that included >10,000 subjects using sampling weights. Second, we excluded subjective self-reported knee pain and focused on the radiographic findings of knee OA. The knee OA grading agreement rate was high, and coefficients of inter-rater reliability between radiologists showed fair to very high degrees of agreement. Third, we identified sociodemographic factors associated with radiographic knee OA from nation-representative sample in Korea. An enhanced understanding of the demographic factors associated with knee OA provides information on the population at high risk of knee OA, which is useful for prevention and management in the early stages.

Nevertheless, our study has some limitations. First, radiographic findings of knee OA are usually, but not always, correlated with patient symptoms; radiographic OA changes are not always associated with knee pain [26,27]. As we did not consider knee pain, the prevalence of knee OA may have been over- or underestimated in this study. Second, although we adjusted for many covariates, the effects of residual or hidden confounding variables cannot be excluded, similar to other cross-sectional studies.

In conclusion, using data from the 2010–2013 KNHANES and defining knee OA as Kellgren–Lawrence grade ≥ 2, we found that the prevalence of radiographic knee OA was 35.1% (24.4% in men, 44.3% in women) in a representative sample of Korean adults aged ≥ 50 years, with the highest prevalence (78.7%) observed in women aged ≥ 80 years. Low socioeconomic status and traditional factors, including age, female sex, and obesity, were associated with the risk of knee OA. To reduce inequalities in knee OA prevalence, interventions and policies should target low-socioeconomic-status groups.

10 Oct 2019

PONE-D-19-15672

The Prevalence of and Demographic Factors Associated with Radiographic Knee Osteoarthritis in Korean Adults Aged ≥ 50 years: The 2010–2013 Korea National Health and Nutrition Examination Survey

PLOS ONE

Dear Dr. Kim,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Nov 24 2019 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Young Dae Kwon, M.D., Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Please see my Major comments below:

1. This reader does not agree with the authors’ statement “Evidence for the sociodemographic factors affecting radiographic knee OA, other than traditional factors such as age, female sex, and obesity, is limited.”

a. First, this statement is vague in providing the rationale for this current study.

b. Second, many of the factors in the current analysis have been examined previously in NHANES in the US, as well as studies in China, and other countries worldwide. For example, factors such as income, education, occupation, geographical area of residence in addition to the more “traditional” factors such as sex, age, and BMI have been examined as evidenced by the authors citing of previous work in their Discussion section. In fact, many factors were also examined in the 2009 Korean NHANES by Lee KM et al. Yonsei Med J 56(1):124-131, 2015.

c. Thus, the novelty of and rationale for this study need to be better defined.

2. Methods section is not clearly written with respect to statistical analysis and variables of interest.

a. First, it is unclear the methods used to account for sampling weights?

b. Second, please clarify the statements “To compare the weighted prevalence of radiographic knee OA by sociodemographic factors, chi-squared tests and analysis of covariance were performed. The prevalence of radiographic knee OA was analyzed using age, sex, area of residence, education level, number of family members, household income, occupation, and obesity [body mass index (BMI) ��25 kg/m2] as confounding variables.”

I may be mistaken, but my understanding of analysis of covariance (ANCOVA) is that it is used to test the main and possible interaction effects of categorical variables on a continuous dependent variable, controlling for other continuous variables. It is unclear what this main dependent variable is? Was Radiographic OA as measured with KL grade used as a continuous/ordinal variable in order to justify use of ANCOVA? If so, this cannot be a measure of prevalence? Therefore, it is unclear how prevalence of radiographic knee OA (which I am assuming is a proportion) was defined. If proportion of those with KL >=2 vs. those <2; or >=3 vs those <3; or those =4 vs. <4, was defined and what analyses were used to adjust for “confounders” in the prevalence of ROA?

c. My understanding of a confounding variable is that it is an “extraneous” risk factor for ROA other than the exposure of interest and needed to be controlled for in the analysis. We usually control for confounding in relation to an exposure of interest. It seems to me that age, sex, area of residence, education, # family members, household income, occupation, BMI >= 25, are not necessarily confounding variables in this research project, and that they are used more like predictors of, or covariates in a prediction model with ROA (defined one of 3 ways: as KL >=2 vs. <2, >=3 vs. <3, or =4 vs. <4) as the outcome?

3. Methods and Tables should stand alone? In other words, a reader should not have to study the tables to understand how variables of interest were defined in a study. For example, I believe that categories of age, education, income, occupation, etc.. should be defined in the Methods section. Also, how was rural vs. urban defined?

4. Results section:

a. Please clarify what “irrespective of severity” means.

b. As for the Tables such as Table 3, what does “other variables” mean? Is Table 4 based on multivariable logistic regression? If so, need to mention these variables are mutually adjusted? is used in the models. Is it used as a continuous variable or as categorical variable? How was rural vs. urban defined?

c. Given separate models for different definitions of ROA and separate models where different sets of covariates were used, it would be clearer to understand what the authors mean by prevalence(s) of radiographic knee OA or “after adjustment” in the Results.

d. Since prevalence and incidence of OA are very different in men compared with women, and factors associated with OA may be very different in the two groups, we tend to stratify by men and women in analyses? Why did the authors NOT conduct the analyses stratified by men and women? Would these factors, whether sociodemographic or lifestyle factors differ in men and women, including that of occupation?

5. In the ABSTRACT, the last sentence in the Results section related to the use of Logistic regression analyses should be under the Methods section? Again, because the authors mentioned that ROA was defined as KL >=2 vs. < 2, then several times mentioned “regardless of OA severity”—this was a bit confusing because data for “severity” were not presented in abstract, as the Abstract is a stand-alone document I read before I read the results in the main paper.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

21 Nov 2019

Reviewer #1: Please see my Major comments below:

1. This reader does not agree with the authors’ statement “Evidence for the sociodemographic factors affecting radiographic knee OA, other than traditional factors such as age, female sex, and obesity, is limited.”

a. First, this statement is vague in providing the rationale for this current study.

b. Second, many of the factors in the current analysis have been examined previously in NHANES in the US, as well as studies in China, and other countries worldwide. For example, factors such as income, education, occupation, geographical area of residence in addition to the more “traditional” factors such as sex, age, and BMI have been examined as evidenced by the authors citing of previous work in their Discussion section. In fact, many factors were also examined in the 2009 Korean NHANES by Lee KM et al. Yonsei Med J 56(1):124-131, 2015.

c. Thus, the novelty of and rationale for this study need to be better defined.

� Thank you so much for your valuable comments.

a. We totally agree with you. We deleted the sentences you mentioned (“Evidence for the sociodemographic factors affecting radiographic knee OA, other than traditional factors such as age, female sex, and obesity, is limited.”) in abstract and introduction section.

b. We fully understand your concern. However, there are some differences between previous and our study in Korea.

Lee KM et al. used data from 1,728 subjects aged > 65 years from the overall dataset and reported risk factors for OA (unspecified site) based on self-reports. And they investigated the association in 1,728 subjects, not in weighted-sample. On the other hand, we excluded subjective self-reported knee pain and focused on the radiographic findings of knee OA. Our study used data from 12,287 individuals aged ≥ 50 years (weighted number of subjects, 14,837,279 persons) and demonstrated the weighted prevalence and weighted associated factors of knee OA based on radiographic finding.

c. We understand your concern about the novelty of and rationale for this study. However, to our knowledge, few epidemiologic studies have been based on national surveillance of knee OA in the general population that included >10,000 subjects using sampling weights. The most important meaning of this study will be the presentation of radiographic finding-verified knee OA prevalence in nation-representative samples of Korea, because the prevalence of knee OA could be different among nations and the presentation of knee OA prevalence in whole population of one country could be useful for further understanding and future studies in the field of OA. Although there is no novel associated factor with knee OA in this study compared to previous studies, the investigation of demographic factors associated with the presence of knee OA could be meaningful because the results of this study were drawn from nation-representative sample in Korea.

2. Methods section is not clearly written with respect to statistical analysis and variables of interest.

a. First, it is unclear the methods used to account for sampling weights?

� Sampling weights were provided by Korea CDC (Center for Disease Control) when we download raw data from the Korea National Health and Nutrition Examination Survey database website. Details of methods used to account for sampling weights are presented in ‘the guideline for analyses of Korea NHANES dataset by Korea CDC (in Korean)’, which can be downloaded from the Korea NHNES database website.

b. Second, please clarify the statements “To compare the weighted prevalence of radiographic knee OA by sociodemographic factors, chi-squared tests and analysis of covariance were performed. The prevalence of radiographic knee OA was analyzed using age, sex, area of residence, education level, number of family members, household income, occupation, and obesity [body mass index (BMI) ��25 kg/m2] as confounding variables.”

I may be mistaken, but my understanding of analysis of covariance (ANCOVA) is that it is used to test the main and possible interaction effects of categorical variables on a continuous dependent variable, controlling for other continuous variables. It is unclear what this main dependent variable is? Was Radiographic OA as measured with KL grade used as a continuous/ordinal variable in order to justify use of ANCOVA? If so, this cannot be a measure of prevalence? Therefore, it is unclear how prevalence of radiographic knee OA (which I am assuming is a proportion) was defined. If proportion of those with KL >=2 vs. those <2; or >=3 vs those <3; or those =4 vs. <4, was defined and what analyses were used to adjust for “confounders” in the prevalence of ROA?

-> We are very sorry for our terrible mistakes. As you pointed out, ANCOVA is an analysis for continuous variable as an outcome. We analyzed by using GLM (general linear model), not ANCOVA, and corrected mistake in method section. We used GLM analysis of complex analysis in SPSS.

“To compare the weighted prevalence of radiographic knee OA by sociodemographic factors, chi-squared tests and general linear model were performed.”

c. My understanding of a confounding variable is that it is an “extraneous” risk factor for ROA other than the exposure of interest and needed to be controlled for in the analysis. We usually control for confounding in relation to an exposure of interest. It seems to me that age, sex, area of residence, education, # family members, household income, occupation, BMI >= 25, are not necessarily confounding variables in this research project, and that they are used more like predictors of, or covariates in a prediction model with ROA (defined one of 3 ways: as KL >=2 vs. <2, >=3 vs. <3, or =4 vs. <4) as the outcome?

� Thank you for your comment. We revised “the confounding variables” to “covariates” in abstract, method, and discussion section.

3. Methods and Tables should stand alone? In other words, a reader should not have to study the tables to understand how variables of interest were defined in a study. For example, I believe that categories of age, education, income, occupation, etc.. should be defined in the Methods section. Also, how was rural vs. urban defined?

�We modified all the following sentences as your recommendation.

“ The prevalence of radiographic knee OA was analyzed using age (50-59, 60-69, 70-79, � 80 years old), sex (men/women), area of residence (urban[Dong]/rural[Eup/Myeon), education level (elementary school/junior high school/senior high school/college graduated), number of family members (1/2/3/�4), household income (� 24th, 25-49th, 50-74th, ≥75th percentile), occupation (managers and professionals/clerical support workers/service and sales workers/skilled agricultural, forestry and fishery workers/craft, plant, or machine operators and assemblers/laborers/unemployed (including students and house wives), and obesity [body mass index (BMI) ≥ 25 kg/m2] as covariates.”

According to administrative district of South Korea, “dong” is classified into the urban area, and “eup” / “myeon” are classified into to the rural area.

We added these administrative districts in Method section.

“area of residence (urban[Dong]/rural[Eup/Myeon)”

4. Results section:

a. Please clarify what “irrespective of severity” means.

� We clarified the “severity” to “knee OA severity (Kellgren–Lawrence grades ≥2, ≥3, and 4)”

b. As for the Tables such as Table 3, what does “other variables” mean? Is Table 4 based on multivariable logistic regression? If so, need to mention these variables are mutually adjusted? is used in the models. Is it used as a continuous variable or as categorical variable? How was rural vs. urban defined?

� We added specific variables “area of residence, education level, number of family members, household income, occupation, and obesity.”, instead of “other variables” in Result section and Table 3.

Table 4 is analyzed based on multivariable logistic regression as categorical variable. We added the categorical definition at each variables in Method section.

“Logistic regression analyses were used to calculate the odds ratio (OR) for radiographic knee OA with age(50-59, 60-69, 70-79, ≥80 years old), sex(men/women), area of residence(urban[Dong]/rural[Eup/Myeon), education level(elementary school/junior high school/senior high school/college graduated), household income(�24th, 25-49th, 50-74th,≥75th percentile), and obesity(no/yes) serving as covariates.”

According to administrative district of South Korea, “dong” is classified into the urban area, and “eup” / “myeon” are classified into to the rural area.

We added these administrative districts in Method section.

c. Given separate models for different definitions of ROA and separate models where different sets of covariates were used, it would be clearer to understand what the authors mean by prevalence(s) of radiographic knee OA or “after adjustment” in the Results.

� We revised the results from unadjusted to adjusted prevalence of radiographic knee OA

“The unadjusted and adjusted weighted prevalences of radiographic knee OA were lower in urban areas than in rural areas [adjusted, 34.0% (32.6–35.5%) vs. 38.2% (35.5–40.9%), p = 0.008].

Education level was correlated inversely with the prevalence of radiographic knee OA, before and after adjustment. Elementary school graduates had a higher prevalence of radiographic knee OA than did college graduates [adjusted, 37.6% (35.7–39.5%) vs. 29.5% (26.7–32.3%), p < 0.001].”

d. Since prevalence and incidence of OA are very different in men compared with women, and factors associated with OA may be very different in the two groups, we tend to stratify by men and women in analyses? Why did the authors NOT conduct the analyses stratified by men and women? Would these factors, whether sociodemographic or lifestyle factors differ in men and women, including that of occupation?

� We agree with your opinion.

We added table 5 which showed logistic regression analyses for radiographic knee osteoarthritis (Kellgren-Lawrence grade ≥ 2) according to sex.

5. In the ABSTRACT, the last sentence in the Results section related to the use of Logistic regression analyses should be under the Methods section? Again, because the authors mentioned that ROA was defined as KL >=2 vs. < 2, then several times mentioned “regardless of OA severity”—this was a bit confusing because data for “severity” were not presented in abstract, as the Abstract is a stand-alone document I read before I read the results in the main paper.

To avoid confusion, we added the definition of Kellgren–Lawrence grade in methods section in the Abstract.

Submitted filename: Reply to Reviewers 20191120.docx

Click here for additional data file.

5 Mar 2020

The Prevalence of and Demographic Factors Associated with Radiographic Knee Osteoarthritis in Korean Adults Aged ≥ 50 years: The 2010–2013 Korea National Health and Nutrition Examination Survey

PONE-D-19-15672R1

Dear Dr. Kim,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Young Dae Kwon, M.D., Ph.D.

Academic Editor

PLOS ONE

9 Mar 2020

PONE-D-19-15672R1

The Prevalence of and Demographic Factors Associated with Radiographic Knee Osteoarthritis in Korean Adults Aged ≥ 50 years: The 2010–2013 Korea National Health and Nutrition Examination Survey

Dear Dr. Kim:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Young Dae Kwon

Academic Editor

PLOS ONE