Fatty liver index as a predictive marker for the development of diabetes: A retrospective cohort study using Japanese health check-up data.

BACKGROUND & AIMS: Fatty liver is associated with incident diabetes, and the fatty liver index (FLI) is a surrogate marker for non-alcoholic fatty liver disease (NAFLD). We aimed to determine whether or not FLI was associated with incident diabetes in relation to obesity and prediabetic levels in the general Japanese population.
METHODS: This was a retrospective study using the Japanese health check-up database of one health insurance from FY2015 to FY2018. This study included 28,991 individuals with prediabetes. First, we stratified all participants into two groups: "high-risk," comprising patients with HbA1c >6.0%, and "standard," comprising the rest. Subsequently, we divided them into four groups according to FLI (<30 or not) and obesity (BMI <25 kg/m2 or not). Subsequently, the incidence rate of diabetes was compared among the groups after 3 years of follow-up using multiple logistic regression models after adjusting for potential confounders.
RESULTS: After 3 years of follow-up, 1,547 new cases of diabetes were found, and the cumulative incidence was 2.96% for the standard group and 26.1% for the high-risk group. In non-obese individuals, odds ratios (95% confidence interval) for FLI ≥30 versus FLI <30 were: 1.44 (1.09-1.92) for the standard group and 1.42 (0.99-2.03) for the high-risk group. In the high-risk group, obesity (BMI ≥25 kg/m2) but FLI <30 was not a risk factor for developing diabetes.
CONCLUSION: Although high FLI is generally considered to be a risk factor for developing diabetes, obesity might have been a confounding factor. However, the present study showed that high FLI is a risk factor for the development of diabetes, even in non-obese individuals. Our results include suggestion to develop a screening tool to effectively identify people at high risk of developing diabetes from the population (especially non-obese prediabetes) who are apparently at low health risk and are unlikely to be targeted for health guidance.

Introduction

Non-alcoholic fatty liver disease (NAFLD) refers to a fatty liver without a history of excessive alcohol consumption or liver disease. It is becoming a common chronic liver disease worldwide. Its prevalence in the general population is reported to be approximately 25% worldwide [1] and 18%–30% in Japan [2-4]. It is considered to be a hepatic phenotype of metabolic syndrome (MetS) and is closely related to obesity. Recent meta-analysis studies have shown that patients with NAFLD are approximately twice as likely to develop diabetes as those without NAFLD [5].

Liver biopsy is the gold standard for diagnosing NAFLD. Since it is an invasive procedure, abdominal ultrasonography is clinically used for the diagnosis. In 2006, Bedogni et al. introduced the fatty liver index (FLI) as a surrogate marker for NAFLD, which comprises the body mass index (BMI), waist circumference (WC), and gamma-glutamyl transferase (GGT) and triglyceride (TG) levels [6]. Movahedian et al. performed a meta-analysis of NAFLD defined by FLI (FLI-NAFLD) and the risk of developing diabetes and concluded that high FLI scores increased the risk of developing diabetes [7].

Cases of most previous studies on the association between FLI and the development of diabetes have been a mixture of normoglycemia and prediabetes, or “not diabetic.” However, normoglycemia and prediabetes might differ in the risk of diabetes and related factors. Heianza et al. reported that patients with prediabetes diagnosed based on impaired fasting glucose (IFG; fasting plasma glucose (FPG) ≥100 mg/dL) and/or hemoglobin A1c (HbA1c) (≥5.7%) according to ADA criteria [8] are six times more likely to develop diabetes than patients with normoglycemia [9]. In evaluating the risk of developing diabetes, subjects should be differentiated between normoglycemia and prediabetes.

In addition, only 20% of patients with prediabetes met both criteria (IFG and HbA1c) in the study by Heianza et al. [9]. Prediabetes should be extracted with criteria for both HbA1c and IFG because applying only one criterion often leads to a missed diagnosis. Among previous studies on prediabetes [10-13], only Nadal et al. extracted prediabetes using both HbA1c and FPG criteria [11].

Even in the prediabetic population, the risk of developing diabetes is strongly affected by an individual’s glucose tolerance level. Therefore, the statistical analysis should be performed after adjusting for baseline glucose tolerance levels. Of these four studies, only Wargny et al.’s study had adjusted glucose tolerance levels [13].

Recent reports have shown that NAFLD can occur in individuals who are not obese and have a normal BMI. These individuals are labeled as “lean NAFLD” or “nonobese NAFLD” [14]. Ye et al. reported that in the general population (comprising individuals with and without NAFLD), 12.1% of people had non-obese NAFLD and 5.1% had lean NAFLD [15]. Young et al. reported that the prevalence of lean NAFLD in the general population was 11.2% worldwide and 12% in Asia [16]. Metabolic risk factors associated with insulin resistance are relevant for non-obese and obese NAFLD [17]. In addition, lean or non-obese NAFLD is a risk factor for the development of diabetes [18-22]. Approximately 25% and 40% of all NAFLD cases are lean and non-obese, respectively [15,16]. Therefore, focusing only on obesity may miss patients with NAFLD and those at a high risk of developing diabetes.

Obesity, as defined by BMI, is a common risk factor for both NAFLD and diabetes. Since BMI is a component of FLI, the confounding effect of obesity must be considered when assessing the incidence of diabetes using FLI. Many previous studies have argued that FLI is a predictor of the development of new diabetes mellitus [7,10-13,23-29]. However, there have been no studies on FLI and the development of diabetes in non-obese individuals.

The aim of this study was to assess the risk of developing diabetes in FLI-NAFLD after strict classification of blood glucose status among the Japanese population. Therefore, we limited the subjects in this study to those with prediabetes, as assessed by both FPG and HbA1c criteria. Additionally, we aimed to evaluate whether or not FLI-NAFLD is associated with the risk of developing diabetes considering obesity and prediabetes levels.

Methods

Study design and data source

The present study was a retrospective study performed using the Japanese health check-up and administrative claims databases from FY2015 to FY2018. Data were obtained from one health insurance association, comprising annual health check-up and claims data collected from all prefectures in Japan other than Tokyo. The database comprises information on the age, sex, diagnosis, prescriptions, medical procedures, and regions.

Study subjects

Subjects were diagnosed with prediabetes using both the HbA1c criterion and FPG. Eligible subjects for this study were those who (1) underwent the annual health check-up at FY2015 and had data available; (2) had no missing data for weight, height, WC, HbA1c, FPG, TG, or GGT according to the questionnaire of the use of antidiabetics in the health check-up at FY2015 and FY2018; (3) had no cardiovascular disease, chronic kidney disease, or stroke according to the questionnaire of the health check-up at FY2015; (4) did not drink alcohol every day, with daily alcohol consumption not exceeding 20 g of ethanol according to the questionnaire of the health check-up at FY2015; (5) had no claims of ICD-10 codes for B18 (chronic viral hepatitis), C22 (malignant neoplasm of the liver and intrahepatic bile ducts), K743 (primary biliary cirrhosis), or K754 (autoimmune hepatitis) at FY2015; (6) had no outlier data for HbA1c or WC at FY2015 and 2018; (7) did not have diabetes (HbA1c ≥6.5% or FPG ≥126 mg/dL or use of antidiabetics) at FY2015; and (8) did not have normoglycemia (HbA1c <5.7% and FPG <100 mg/dL) at FY2015. Subjects who met all eligibility criteria are shown in Fig 1.

Fig 1

Flow of eligible subjects.

The definition of “prediabetes” is shown in Fig 2. According to ADA criteria, prediabetics with HbA1c >6.0% are considered to be at a high risk and require aggressive intervention and vigilant follow-up [8]. Therefore, in this study, HbA1c >6.0% was defined as “high-risk prediabetes” and HbA1c ≤6.0% as “standard prediabetes.” In the original Bedogni et al. study, FLI ≥60 was suggested to rule in FLD, but in Asians, the cutoff value for FLI in NAFLD diagnosed by ultrasonography is often approximately 30 [30,31]. Therefore, FLI ≥30 was used to define FLI-NAFLD in this study. The International Obesity Task Force recommended the lower cutoffs of BMI ≥23kg/m2 for overweight, and ≥25.0kg/m2 for obese for Asian people, according to the risk for type 2 diabetes and hypertension [32]. According to Japanese guidelines, obesity is defined as a BMI of 25 kg/m2 or higher [33]. Therefore, BMI≥25 kg/m2 was used to define obesity in this study. The patterns were grouped into four groups according to the following criteria: BMI <25 kg/m2 with FLI <30 as “non-obese without FLI-NAFLD”; BMI ≥25 kg/m2 with FLI <30 as “obese without FLI-NAFLD”; BMI <25 kg/m2 with FLI ≥30 as “non-obese with FLI-NAFLD”; and BMI ≥25 kg/m2 with FLI ≥30 as “obese with FLI-NAFLD.”

Fig 2

Definition of prediabetes.

Outcomes and variables

The primary outcome was the development of diabetes mellitus at the time of the 2018 health check-up. Diabetes mellitus was defined as HbA1c ≥6.5%, FPG ≥126 mg/dl, or use of antidiabetic medication in the questionnaire.

For background variables, the age, sex, FLI, high-density lipoprotein cholesterol (HDL-c), low-density lipoprotein cholesterol (LDL-c), systolic blood pressure (SBP), HbA1c (based on National Glycohemoglobin Standardization Program units), comorbidities (hypertension, hyperlipidemia, diabetes mellitus based on self-administered questionnaire), eating habits, smoking habits, and physical activities at FY2015 were extracted from the database. The FLI score was calculated as follows:

Statistical analysis

For descriptive statistics of baseline characteristics, mean (SD) or median (IQR) were calculated for continuous variables, and frequency and percentage were calculated for categorical variables for each group. The chi-squared test or analysis of variance was performed to compare groups for each aggregated background.

Univariate and multivariate analyses were performed to evaluate the background and risk of developing diabetes. Multivariate adjusted logistic regression models were then applied to evaluate the association between the development of diabetes and the combination of BMI and FLI. Multivariate analyses with three models were performed to calculate odds ratios (ORs) and their 95% confidence intervals (CIs). Models were adjusted for age (categorized into three age groups: 39–49, 50–59, and 60–71 years) and sex for Model-1; Model-1 plus FPG, SBP, HDL, and LDL for Model-2; Model-2 plus smoking, eating habit, physical activities, adequate sleeping, weight change within 1 year, and age at 20 for Model-3.

Yang et al. reported cutoff value for FLI that differed by sex (35 for males and 20 for females) [31], so we also performed a sensitivity analysis using the Yang et al. cutoff values.

The dose-dependent analysis between FLI and incident diabetes was performed. The FLI was categorized into seven FLI groups (0– <10, 10– <20, 20– <30, 30– <40, 40– <50, 50– <60, and ≥60) and multivariate adjusted logistic regression models were then applied. The model was adjusted for age, sex, BMI, FPG, SBP, HDL, LDL, smoking, eating habit, physical activities, adequate sleeping, weight change within 1 year, and age at 20.

To evaluate the fitness of the model, we performed the lack-of-fit test [34].

All tests were two-tailed, and the significance level was set to 0.05. For the statistical analysis, R version 3.63 (R Core [Team 2020] R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/) and JMP® version 15.0 (SAS Institute Inc., Cary, NC, USA) was used.

Results are reported in accordance with the recommendations of the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) [35].

Ethics

In this study, only anonymized data were used and we had no access to personal information. The Teikyo University Ethical Review Board for Medical and Health Research Involving Human Subjects approved this study after due ethical consideration (approval No.:18-200-3).

Results

Study population

Of the 213,652 beneficiaries, 28,991 eligible subjects were extracted from the database. For patients with normoglycemia, standard prediabetes, and high-risk prediabetes, the cumulative incidences of new-onset diabetes were 0.19% (92/47,487 cases), 2.96% (770/26,014 cases), and 26.1% (777/2977 cases), respectively.

Baseline characteristics

Descriptive analyses for baseline characteristics of the eligible subjects are shown in Table 1. Analyses of lifestyle factors, such as eating habits, physical activities, and daily sleeping, are provided in Table in S1 Table. Every variable significantly differed among the four groups.

Table 1

Baseline characteristics of eligible subjects.

	Non-obese without FLI-NAFLD		Non-obese with FLI-NAFLD		Obese without FLI-NAFLD		Obese with FLI-NAFLD
	n = 14,676		n = 3230		n = 1291		n = 6817
Standard prediabetes (HbA1c; 5.7–6.0%)
Sex (Male), n (%)	8808	60.0%	2919	90.4%	721	55.8%	5441	79.8%
Age (year), n (%)
39–49	7227	49.2%	1410	43.7%	683	52.9%	3521	51.7%
50–59	7181	48.9%	1754	54.3%	587	45.5%	3188	46.8%
60–71	268	1.8%	66	2.0%	21	1.6%	108	1.6%
Current smoker, n (%)	1504	10.3%	648	20.1%	115	8.9%	1246	18.3%
Metabolic syndrome*	107	0.7%	464	14.4%	62	4.8%	2329	34.2%
FLI	9.9	(5.1–17.2)	41.3	(34.8–51.6)	22.8	(18.1–26.2)	59.5	(44.9–75.7)
BMI (kg/m2)	21.6	(20.0–22.9)	23.8	(22.24.4)	25.7	(25.3–26.4)	27.4	(26.29.4)
WC (cm)	77.6	±6.2	85.1	±4.2	86.7	±4.1	94.4	±7.4
SBP (mmHg)	115.7	±15.7	122.5	±15.5	121	±15.3	126.8	±15.6
DBP (mmHg)	72.5	±11.4	78.6	±11.3	75.9	±11.4	81	±11.6
Laboratory tests
HbA1c (%)	5.7	±0.2	5.6	±0.2	5.7	±0.2	5.7	±0.2
FPG (mg/dl)	97.9	±8.4	101.5	±7.7	99.3	±7.9	101.3	±8.1
AST (U/L)	19	(17–23)	23	(20–28)	19	(17–23)	24	(20–30)
ALT (U/L)	17	(13–22)	28	(21–39)	18	(14–24)	31	(22–41)
GGT (U/L)	21	(16–30)	54	(36–87)	19	(15–25)	40	(28–61)
LDL-C (mg/dl)	123.6	±28.6	135.8	±31.5	124.8	±27.9	136.4	±30.5
HDL-C (mg/dl)	64	(55–75)	52	(44–61)	58	(51–68)	50	(44–58)
TG (mg/dl)	76	(57–100)	154	(118–206)	73	(59–89)	132	(100–181)
Hemoglobin (mg/dl)	14.0	±1.6	15.1	±1.1	14.0	±1.6	15.0	±1.3
	Non-obese without FLI-NAFLD		Non-obese with FLI-NAFLD		Obese without FLI-NAFLD		Obese with FLI-NAFLD
	n = 1080		n = 377		n = 133		n = 1387
High-risk prediabetes (HbA1c; 6.1–6.5%)
Sex (Male), n (%)	599	55.5%	327	86.7%	61	45.9%	1061	76.5%
Age (year), n (%)
39–49	379	35.1%	121	32.1%	57	42.9%	611	44.1%
50–59	664	61.5%	239	63.4%	71	53.4%	751	54.1%
60–71	37	3.4%	17	4.5%	5	3.8%	25	1.8%
Current smoker, n (%)	159	14.7%	92	24.4%	10	7.5%	280	20.2%
Metabolic syndrome*	27	2.5%	85	22.6%	8	6.0%	694	50.0%
FLI	11.8	(6.1–19.0)	41.7	(35.5–52.0)	22.9	(18.7–26.8)	66.8	(51.0–81.9)
BMI (kg/m2)	21.9	(20.4–23.1)	23.8	(23.1–24.5)	25.6	(25.3–26.4)	28.2	(26.5–30.5)
WC (cm)	78.3	±6.0	85.3	±4.1	86.2	±4.9	96.3	±8.5
SBP (mmHg)	116.1	±15.3	121.9	±13.9	123.2	±14.0	127.5	±15.0
DBP (mmHg)	72.7	±11.0	77.8	±10.7	76.1	±10.1	81.4	±11.4
Laboratory tests
HbA1c (%)	6.2	±0.1	6.2	±0.1	6.2	±0.1	6.2	±0.1
FPG (mg/dl)	101.2	±10.4	106.7	±9.5	103.3	±9.7	106.5	±9.5
AST (U/L)	20	(17–23)	24	(19–29)	21	(17–25)	26	(21–34)
ALT (U/L)	18	(14–23)	29	(22–42)	20	(16–28)	36	(25–55)
GGT (U/L)	22	(16–30)	53	(36–81)	20	(15–26)	44	(31–66)
LDL-C (mg/dl)	128.9	±30.2	136.8	±32.5	134	±34.2	137.9	±30.9
HDL-C (mg/dl)	62	(53–73)	50	(44–57)	59	(51–66)	49	(44–57)
TG (mg/dl)	80	(59–107)	157	(124–210)	75	(62–98)	138	(103–188)
Hemoglobin (mg/dl)	13.7	±1.7	14.9	±1.2	14	±1.4	15	±1.4

AST, aspartate aminotransferase; ALT, alanine aminotransferase.

* According to Japanese diagnostic criteria [36].

Risk of new-onset diabetes in patients with standard prediabetes and high-risk prediabetes

In the multivariate analysis, significant factors associated with an increased risk of developing diabetes in patients with standard prediabetes were FLI ≥30, BMI ≥25 kg/m2, female sex, current smoking, weight increase >3 kg within 1 year, and no walking or exercise for at least 1 h/day (Table 2). In patients with high-risk prediabetes, FLI ≥30, current smoking, and weight gain of 10 kg or more since age 20 were significant factors. In addition, there was no interaction between obesity and FLI-NAFLD (P-value for interaction: 0.69).

Table 2

Univariate and multivariate adjusted logistic regression models for the incidence of new-onset diabetes.

	Univariate model			Multivariate model*
	OR	95% CI	P-value	OR	95% CI	P-value
Standard prediabetes (HbA1c; 5.7–6.0%)
FLI ≥ 30	3.94	(3.37–4.60)	< .0001	1.40	(1.10–1.77)	0.006
BMI ≥ 25	3.57	(3.08–4.14)	< .0001	1.68	(1.35–2.09)	< .0001
Sex (male)	1.99	(1.66–2.39)	< .0001	0.76	(0.61–0.94)	0.013
Age
39–49	reference
50–59	1.09	(0.95–1.27)	0.224	1.03	(0.87–1.21)	0.762
60–71	1.31	(0.80–2.16)	0.280	1.25	(0.71–2.19)	0.443
Life stile
Current smoking	1.99	(1.68–2.36)	< .0001	1.81	(1.48–2.21)	< .0001
Weight change of more than ±3 kg within 1 year	1.96	(1.67–2.29)	< .0001	1.37	(1.15–1.64)	< .001
Weight gain of 10 kg or more since aged at 20	2.71	(2.31–3.18)	< .0001	1.10	(0.91–1.35)	0.327
Light exercise of at least 30 minutes per session	0.93	(0.77–1.12)	0.445	0.99	(0.80–1.22)	0.900
1 hour per day of walking or exercise in daily life	1.07	(0.91–1.30)	0.406	1.22	(1.01–1.48)	0.038
Walking speed is faster	0.88	(0.75–1.03)	0.100	0.86	(0.72–1.02)	0.079
Eating dinner within 2 hours before bedtime	1.26	(1.07–1.48)	0.005	1.04	(0.87–1.25)	0.632
Eating midnight snack	1.01	(0.83–1.23)	0.896	0.94	(0.76–1.16)	0.537
Skipping breakfast	1.54	(1.25–1.91)	< .0001	1.00	(0.79–1.26)	0.993
Eating speed
Slow	reference
Normal	1.43	(0.96–2.14)	0.082	1.07	(0.70–1.65)	0.741
Fast	2.29	(1.53–3.43)	< .0001	1.34	(0.87–2.06)	0.185
Adequate sleeping	1.08	(0.92–1.26)	0.350	1.02	(0.86–1.20)	0.863
High-risk prediabetes (HbA1c; 6.1–6.5%)
FLI ≥ 30	2.40	(2.00–2.88)	< .0001	1.56	(1.15–2.11)	0.004
BMI ≥ 25	2.00	(1.69–2.37)	< .0001	1.13	(0.86–1.49)	0.371
Sex (male)	1.78	(1.47–2.15)	< .0001	0.89	(0.69–1.15)	0.382
Age
39–49	reference
50–59	0.93	(0.79–1.10)	0.421	0.87	(0.71–1.07)	0.189
60–71	0.58	(0.33–1.04)	0.066	0.47	(0.22–1.00)	0.051
Life stile
Current smoking	1.62	(1.32–1.98)	< .0001	1.41	(1.10–1.82)	0.008
Weight change of more than ±3 kg within 1 year	1.76	(1.45–2.12)	< .0001	1.57	(1.26–1.95)	< .0001
Weight gain of 10 kg or more since aged at 20	1.64	(1.36–1.97)	< .0001	0.89	(0.70–1.13)	0.337
Light exercise of at least 30 minutes per session	0.99	(0.80–1.23)	0.919	1.02	(0.70–1.33)	0.865
1 hour per day of walking or exercise in daily life	0.92	(0.75–1.10)	0.420	0.99	(0.78–1.26)	0.955
Walking speed is faster	0.96	(0.80–1.14)	0.623	1.00	(0.81–1.23)	0.978
Eating dinner within 2 hours before bedtime	1.20	(0.99–1.46)	0.007	1.05	(0.84–1.32)	0.644
Eating midnight snack	0.94	(0.75–1.18)	0.621	0.95	(0.74–1.23)	0.692
Skipping breakfast	1.47	(1.12–1.93)	0.006	1.22	(0.89–1.67)	0.206
Eating speed
Slow	reference
Normal	1.26	(0.28–0.83)	0.279	1.17	(0.73–1.87)	0.508
Fast	1.38	(0.90–2.11)	0.142	1.04	(0.65–1.68)	0.859
Adequate sleeping	0.91	(0.76–1.08)	0.281	0.86	(0.70–1.06)	0.150

OR, odds ratio; CI, 95% confidence interval.

* Adjusted for age, sex, FPG, SBP, HDL, LDL, smoking, eating habit, physical activities, adequate sleeping, weight change within year and since aged at 20.

The results of the multivariate adjusted logistic regression analysis in the three models are shown in Table 3 and Fig 3. Patients with obesity with FLI-NAFLD, whether standard or high-risk prediabetes had a significantly higher risk of developing diabetes than non-obese subjects without FLI-NAFLD in all three models. In Model-3, non-obese patients with FLI-NAFLD and patients with obesity without FLI-NAFLD were at a higher risk in patients with standard prediabetes, with ORs of 1.44 (95% CI: 1.09–1.92; P = 0.011) and 1.79 (95% CI: 1.21–2.65; P = 0.003), respectively, but were not found in patients with high-risk prediabetes.

Table 3

Multivariate adjusted logistic regression model for the incidence of new-onset diabetes by the combination pattern of obesity and FLI-NAFLD.

	New-onset diabetes FY2018		Model 1†			Model 2‡			Model 3§
	n	%	OR	95% CI	P-value*	OR	95% CI	P-value*	OR	95% CI	P-value*
Standard prediabetes (n = 26,014)
Non-obese without FLI-NAFLD (n = 14,676)	190	1.29%	reference			reference			reference
Non-obese with FLI-NAFLD (n = 3230)	112	3.47%	2.45	(1.92–3.12)	< .0001	1.60	(1.24–2.06)	< .001	1.44	(1.09–1.92)	0.011
Obese without FLI-NAFLD (n = 1291)	38	2.94%	2.36	(1.66–3.36)	< .0001	1.85	(1.29–2.65)	< .001	1.79	(1.21–2.65)	0.003
Obese with FLI-NAFLD (n = 6817)	430	6.31%	4.81	(4.03–5.73)	< .0001	2.80	(2.30–3.41)	< .0001	2.36	(1.85–3.01)	< .0001
High-risk prediabetes (n = 2977)
Non-obese without FLI-NAFLD (n = 1080)	117	16.39%	reference			reference			reference
Non-obese with FLI-NAFLD (n = 377)	106	28.12%	1.79	(1.35–2.37)	< .0001	1.25	(0.92–1.71)	0.155	1.42	(0.99–2.03)	0.054
Obese without FLI-NAFLD (n = 133)	26	19.55%	1.29	(0.81–2.04)	0.278	0.99	(0.61–1.61)	0.973	0.89	(0.51–1.57)	0.693
Obese with FLI-NAFLD (n = 1387)	468	33.74%	2.40	(1.96–2.93)	< .0001	1.60	(1.27–2.03)	< .0001	1.73	(1.29–2.32)	< .001

OR, odds ratio; CI, 95% confidence interval.

* P-values are derived from multivariate logistic regression model.

† Adjusted for age and sex.

‡ Adjusted for age, sex, FPG, SBP, HDL and LDL.

§ Adjusted for age, sex, FPG, SBP, HDL, LDL, smoking, eating habit, physical activities, adequate sleeping, weight change within year and since aged at 20.

Fig 3

Risk of developing new diabetes for each combination pattern of FLI-NAFLD and obesity.

The bars represent each odds ratio for non-obese patients without FLI-NAFLD, and error bars represent 95% CI of the odds ratio.

The results of sensitivity analysis with a cutoff value of FLI of 35 for males and 20 for females are shown in Table 4, and descriptive analyses for baseline characteristics of the eligible subjects are shown in Table in S2 Table. The results were not remarkable changed, but one difference was that in high-risk prediabetes, the non-obese with FLI-NAFLD group had a significant OR from 1.42 (0.99–2.03) to 1.68 (1.18–2.39).

Table 4

Sensitivity analysis of the multivariate adjusted logistic regression model for the incidence of new-onset diabetes by the combination pattern of obesity and FLI-NAFLD, with a cutoff value of FLI of 35 for men and 20 for women.

	New-onset diabetesFY2018		Model 1†			Model 2‡			Model 3§
	n	%	OR	95% CI	P-value*	OR	95% CI	P-value*	OR	95% CI	P-value*
Standard prediabetes (n = 26,014)
Non-obese without FLI-NAFLD (n = 14,961)	209	1.40%	reference			reference			reference
Non-obese with FLI-NAFLD (n = 2945)	93	3.16%	2.17	(1.70–2.79)	< .0001	1.42	(1.09–1.84)	0.009	1.35	(1.01–1.81)	0.042
Obese without FLI-NAFLD (n = 1336)	45	3.37%	2.26	(1.63–3.14)	< .0001	1.72	(1.23–2.41)	0.002	1.58	(1.09–2.30)	0.015
Obese with FLI-NAFLD (n = 6772)	423	6.25%	4.49	(3.79–5.32)	< .0001	2.65	(2.19–3.20)	< .0001	2.31	(1.83–2.92)	< .0001
High-risk prediabetes (n = 2977)
Non-obese without FLI-NAFLD (n = 1104)	180	16.30%	reference			reference			reference
Non-obese with FLI-NAFLD (n = 353)	103	29.18%	2.03	(1.53–2.69)	< .0001	1.57	(1.15–2.14)	0.004	1.68	(1.18–2.39)	0.004
Obese without FLI-NAFLD (n = 132)	27	20.45%	1.22	(0.77–1.92)	0.392	1.09	(0.67–1.76)	0.735	1.06	(0.62–1.83)	0.826
Obese with FLI-NAFLD (n = 1388)	467	33.65%	2.46	(2.02–3.00)	< .0001	1.73	(1.37–2.18)	< .0001	1.81	(1.35–2.41)	< .0001

OR, odds ratio; CI, 95% confidence interval.

* P-values are derived from multivariate logistic regression model.

† Adjusted for age and sex.

‡ Adjusted for age, sex, FPG, SBP, HDL and LDL.

§ Adjusted for age, sex, FPG, SBP, HDL, LDL, smoking, eating habit, physical activities, adequate sleeping, weight change within year and since aged at 20.

The results of the dose-dependent analysis between FLI and incident diabetes is shown in Fig 4. There was a dose-dependent relationship between FLI and the development of diabetes in both standard and high-risk prediabetes, which was more marked in standard prediabetes.

Fig 4

Graph of FLI and risk of incident diabetes.

The dots represent each odds ratio for the categories classified by FLI values, and error bars represent 95% CI of the odds ratio.

Discussion

The present study showed that FLI-NAFLD was an independent risk factor for the development of diabetes in the middle-aged Japanese population. The association between FLI-NAFLD and diabetes incidence differs with obesity and prediabetes levels. Patients with obesity with FLI-NAFLD was a higher risk factor for the development of diabetes than non-obese patients without FLI-NAFLD.

In the non-obese population, FLI-NAFLD was an independent risk factor for incident diabetes in patients with standard prediabetes (OR, 1.44; P = 0.011). Patients with high-risk prediabetes had a moderately increased risk of developing diabetes (OR: 1.42; P = 0.054). These results suggest that FLI is an effective tool for screening high-risk prediabetic individuals for the development of diabetes, even in non-obese individuals, particularly those with HbA1c ≤6.0%.

Concerning prediabetic levels, the population of “high-risk prediabetes” was only approximately one-ninth of that of “standard prediabetes.” However, the number of new cases of diabetes was almost the same. HbA1c levels in 90% of patients with prediabetes were below 6.0% (standard prediabetes). In this majority population, 70% of new cases of diabetes were complicated with FLI-NAFLD, which accounted for only 38.6% of this population. The diabetes incidence was 3.0% in the standard prediabetes group and 26.1% in the high-risk prediabetes group, showing a nine-fold difference between the two groups. The prevalence of obesity significantly differed between standard prediabetes (31%) and high-risk prediabetes (51%) groups. In patients with high-risk prediabetes, neither obesity (BMI ≥25 kg/m2) nor FLI-NAFLD alone was a risk factor, but both were strong risk factors (OR = 1.73). In particular, obesity alone was not a risk factor for the development of diabetes (OR = 0.89). A quarter of this population developed diabetes whereas half was obese. HbA1c >6.0% itself is a strong risk factor, and since obesity accounts for half of the population, BMI ≥25 kg/m2 is probably not a risk factor.

In the present study, 1,547 new patients with diabetes were identified from 28,911 patients with prediabetes. Of the patients with newly diagnosed diabetic, 37.8% had a BMI <25 kg/m2. Furthermore, the incidence of diabetes was higher in non-obese patients with FLI-NAFLD than in those without FLI-NAFLD (6.0% vs. 4.5%).

The difference in the risk of diabetes in patients with obesity and FLI-NAFLD can be explained by “lipid spillover.” Asians, particularly East Asians, have a lower capacity for fat storage in subcutaneous adipose tissue than in other ethnic groups [37]. Therefore, lipid spillover, in which free fatty acid (FFA) overflows from adipose tissue, is thought to be more likely. Lipid spillover may result in the accumulation of ectopic fat, such as fatty liver, which may lead to insulin resistance. Kadowaki et al. evaluated fat distribution, adipose tissue insulin resistance, and skeletal muscle insulin resistance in non-obese Japanese men [38]. Even among non-obese individuals, visceral and hepatic fat accumulations were observed in some individuals, with various accumulation patterns. Even in the absence of visceral fat accumulation, muscle insulin resistance (metabolic disturbance) was observed in the presence of fatty liver, whereas no insulin resistance was observed in the absence of fatty liver, even in the presence of visceral fat accumulation. Non-obese prediabetes should not be neglected. In such cases, FLI is the best screening method because it is non-invasive, inexpensive, and can be calculated using health check-up data.

In this study, the NAFLD population was dominantly male. The first reason for this was that the subjects were 19,937 men and 9054 women, with men accounting for 69% of the total. Second, the prevalence of NAFLD, defined as FLI ≥ 30, was 49% (9748) in men and 23% (2063) in women. The FLI values were significantly higher in males than in females, which seemed to cause a gender difference in the prevalence of FLI-NAFLD. Though there was concern that the gender difference in the prevalence of FLI-NAFLD might be a selection bias, the results of sensitivity analysis using a cutoff value of FLI of 35 for males and 20 for females also showed no significant difference. Rather, the odds ratio of non-obese FLI-NAFLD in high-risk prediabetes changed from 1.42 (0.99–2.03) to 1.68 (1.18–2.39), which was significant. This result further supported the finding that a high FLI is a risk factor for developing diabetes, even in non-obese individuals. In addition, for gender risk, there seemed to be a strong confounding of FLI, laboratory values (especially SBP, FPG, HDL), and lifestyle (especially current smoking). After adjusting for these confounding factors, biologically speaking, women were at higher risk of developing diabetes than men in the present study.

The strength of this study was that it was the largest cohort study (n = 28,911) on the relationship between FLI and the risk of diabetes incidence among patients with prediabetes. Therefore, we could stratify and limit our analysis by glucose tolerance levels and obesity, which allowed us to prevent and address confounding factors. In addition, study subjects were civil servants from Japan, and more than 90% of insurance beneficiaries in this study had undergone a specific health checkup every year, leading to a small sampling bias.

This study had several limitations. First, we did not perform the 2-h oral glucose tolerance test, so the prevalence of diabetes might be underestimated. Second, the optimal cutoff FLI score for predicting NAFLD has been controversial in Asia; however, most cutoff scores reported by validation studies on FLI using ultrasonography in Asia are approximately 30. Third, the study population was limited to a single insurance member. Although it is a sampling bias, it is the same occupation and is less affected by bias due to the socioeconomic status. Finally, unmeasured confounders, such as a family history of diabetes, were not included in this study.

In Japan, to prevent MetS, all public health insurers are obliged to provide specific health checkups and health guidance. Obesity based on WC and BMI is a mandatory criterion to select subjects for health guidance. Therefore, there is a lack of evaluation of lifestyle-related diseases and health guidance for non-obese people. Even in non-obese individuals, the risks of developing insulin resistance and diabetes increase if they have fatty liver, so it is necessary to improve their lifestyle. FLI can be calculated using only health checkup test items and may be effective in identifying individuals at a high risk for lifestyle-related diseases, particularly diabetes, among non-obese individuals.

In conclusion, by assessing FLI in combination with obesity, an association between FLI and those at high risk of developing diabetes in a middle-aged Japanese population was observed. Validating these results, it is desirable to develop a screening tool to effectively identify people at high risk of developing diabetes from the population (especially non-obese prediabetes) who are apparently at low health risk and are unlikely to be targeted for health guidance.

Analysis of lifestyle factors.

(TIF)

Click here for additional data file.

Baseline characteristics of the subjects in a sensitivity analysis with the cutoff value of FLI as 35 for males and 20 for females.

(TIF)

Click here for additional data file.

24 May 2021

PONE-D-21-12023

Fatty liver index as a predictive marker for the development of diabetes: a retrospective cohort study using Japanese health check-up data

PLOS ONE

Dear Dr. Kitazawa,

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.

Please submit your revised manuscript by Jul 08 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're 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.

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). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ming-Lung Yu, MD, PhD

Academic Editor

PLOS ONE

Journal Requirements:

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

1. 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

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We suggest you thoroughly copyedit your manuscript for language usage, spelling, and grammar. If you do not know anyone who can help you do this, you may wish to consider employing a professional scientific editing service.

Whilst you may use any professional scientific editing service of your choice, PLOS has partnered with both American Journal Experts (AJE) and Editage to provide discounted services to PLOS authors. Both organizations have experience helping authors meet PLOS guidelines and can provide language editing, translation, manuscript formatting, and figure formatting to ensure your manuscript meets our submission guidelines. To take advantage of our partnership with AJE, visit the AJE website (http://learn.aje.com/plos/) for a 15% discount off AJE services. To take advantage of our partnership with Editage, visit the Editage website (www.editage.com) and enter referral code PLOSEDIT for a 15% discount off Editage services.  If the PLOS editorial team finds any language issues in text that either AJE or Editage has edited, the service provider will re-edit the text for free.

Upon resubmission, please provide the following:

The name of the colleague or the details of the professional service that edited your manuscript

A copy of your manuscript showing your changes by either highlighting them or using track changes (uploaded as a *supporting information* file)

A clean copy of the edited manuscript (uploaded as the new *manuscript* file)

3. This is a retrospective study observational study. As such, we do not feel that any conclusions on the effectiveness of FLI as a screening tool may be supported; thus, we ask that you revise the text (especially Conclusions) to avoid unsupported statements.

4. Thank you for stating the following in the Financial Disclosure section:

[The authors received no specific funding for this work.].

We note that one or more of the authors are employed by a commercial company: Kyorin Pharmaceutical

Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests) . If this adherence statement is not accurate and  there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

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: Yes

Reviewer #2: Partly

**********

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

Reviewer #1: Yes

Reviewer #2: Yes

**********

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

Reviewer #2: 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: Yes

Reviewer #2: Yes

**********

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: Non-alcoholic fatty liver disease (NAFLD) is an increasingly cause of chronic liver disease. Current studies have been shown that NAFLD contributed substantially to the development of insulin resistance and type 2 diabetes mellitus (DM). In this study, Kitazawa et al. investigated the predictive value of fatty liver index for the development of DM in patients with prediabetes. They found that high FLI is a risk factor for the development of diabetes. Although the results were clinically interesting, several points need be critically addressed.

1. In this study, obesity was defined as body mass index (BMI) ≥25 kg/m2. However, the cutoff for obesity is 23.0 kg/m2 in Asia-Pacific countries, including Japan. The authors may perform subgroups analysis by BMI <23 versus ≥23 kg/m2.

2. The authors should provide the data that how many patients have metabolic syndrome. The correlation between and the development of DM should be taken into consideration.

3. In this study, the cutoff value for FLI in NAFLD was set at 30. The authors should analyze the dose-dependent relationship between FLI and incidence of DM.

4. The cutoff value for FLI in NAFLD were different between male and female gender. Usually, the cut-off value was lower in females than in males. Regardless of gender, the cutoff value for FLI in NAFLD was set at 30 in this study. The authors should explain this important issue.

5. The inclusion criteria (8) did not have normoglycemia (HbA1c �6.5% or FPG �126 mg/dL or use of antidiabetics) at FY2015. Patients with normoglycemia should have HbA1c < 5.7% or FPG <100 mg/dL. Please correct.

Reviewer #2: Authors addressed to find a tool to screen and identify subjects who are at high risk of progressing to diabetes mellitus and FLI maybe the choice of the tool. There are many tools and scores to identify subjects who are progressing to diabetes mellitus, however, maybe FLI is more suitable for Asian population.

Here are some concerns, the most important one is the selection bias. We noted among the population of non-obese with FLI-NAFLD and obese with FLI-NAFLD, both in HbA1c 5.7-6.0% and HbA1c 61.-6.4%, the majority of gender is male.

Major consideration:

Table 1: the gender ratio of non-obese with FLI-NAFLD and obese with FLI-NAFLD were almost male (90.4% and 79.8%), why?

Minor consideration:

1. Table 1, current smoker: is the smokers almost male? What is the proportion?

2. Table 2: the male risk is 1.99 in univariate model, and 0.76 in multivariate model, can you explain it?

3. How to define walking and eating speed is faster?

4. Page 7, line 116: (3) had no outlier data: how to define “outlier” data?

**********

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

Reviewer #2: 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.]

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 PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

30 Jun 2021

We appreciate your comments on our manuscript. The comments have helped us significantly improve the paper.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

20 Jul 2021

PONE-D-21-12023R1

Fatty liver index as a predictive marker for the development of diabetes: a retrospective cohort study using Japanese health check-up data

PLOS ONE

Dear Dr. Kitazawa,

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.

Please submit your revised manuscript by Sep 03 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're 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.

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). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ming-Lung Yu, MD, PhD

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. 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: Yes

Reviewer #2: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. 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

Reviewer #2: No

**********

5. 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: Yes

Reviewer #2: Yes

**********

6. 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: This revised manuscript is much improved and all previous comments were responded on point-to-point basis. I have no additional comments.

Reviewer #2: Dear authors, you had replied my concerns and did a good work about this manuscript.

According to author's reply, you use a new cutoff value of FLI of 35 for males and 20 for females and it seems have better results in this study. "the odds ratio of non-obese FLI-NAFLD in high-risk prediabetes changed from 1.42 (0.99-2.03) to 1.68 (1.18-2.39), which was significant."

Can you upload these differences in fully supplement tables and why don't you use the new cutoff value instead of the original one?

Why the original enrolled subjects are male dominant? 19,937 in male v.s. 9054 in female?

**********

7. 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

Reviewer #2: 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.]

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 PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

6 Aug 2021

Reply to Reviewer #2

1.According to author's reply, you use a new cutoff value of FLI of 35 for males and 20 for females and it seems have better results in this study. "the odds ratio of non-obese FLI-NAFLD in high-risk prediabetes changed from 1.42 (0.99-2.03) to 1.68 (1.18-2.39), which was significant. Can you upload these differences in fully supplement tables and why don't you use the new cutoff value instead of the original one?

Reply.

First, because the protocol was initially determined with FLI 30 as the cutoff value. Secondly, there is no common consensus on the cutoff value of FLI in Asia. Therefore, we thought it should be easier to use the FLI as a screening tool in health examination results. Therefore, we used a common cutoff value for men and women.

The results of sensitivity analysis with a cutoff value of FLI of 35 for males and 20 for females are shown in Table 4, and descriptive analyses for baseline characteristics of the eligible subjects are shown in Table in S2 Table.

2.Why the original enrolled subjects are male dominant? 19,937 in male v.s. 9054 in female?

Reply.

This is because the gender distribution of the employers in this insurance association is significantly more male.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

31 Aug 2021

Fatty liver index as a predictive marker for the development of diabetes: a retrospective cohort study using Japanese health check-up data

PONE-D-21-12023R2

Dear Dr. Kitazawa,

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

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. 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 help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- 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.

Kind regards,

Ming-Lung Yu, MD, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. 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: Yes

Reviewer #2: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. 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

Reviewer #2: Yes

**********

5. 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: Yes

Reviewer #2: Yes

**********

6. 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: This revised manuscript is much improved and all previous comments were responded on point-to-point basis. I have no additional comments.

Reviewer #2: They showed cutoff value of FLI 35 for male and 20 for female in table 4 and detailed characteristics in S2 table, and they had fully responded my opinions. Good work!

**********

7. 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

Reviewer #2: No

6 Sep 2021

PONE-D-21-12023R2

Fatty liver index as a predictive marker for the development of diabetes: a retrospective cohort study using Japanese health check-up data

Dear Dr. Kitazawa:

I'm 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 let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, 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.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Ming-Lung Yu

Academic Editor

PLOS ONE