Skip to main content
PMC home page
PLOS ONE logoLink to PLOS ONE
. 2021 Jul 13;16(7):e0254451. doi: 10.1371/journal.pone.0254451

Association between type 2 diabetes and osteoporosis risk: A representative cohort study in Taiwan

Hsin-Hui Lin 1, Hsin-Yin Hsu 2,3, Ming-Chieh Tsai 2,4,5,6, Le-Yin Hsu 2, Kuo-Liong Chien 2,7, Tzu-Lin Yeh 2,8,*
Editor: Robert Daniel Blank9
PMCID: PMC8277062  PMID: 34255803

Abstract

Although previous studies have investigated the relationship between fracture risk and type 2 diabetes (T2D), cohort studies that estimate composite osteoporosis risk are lacking. This retrospective cohort study sought to determine the risk of osteoporosis in Taiwanese patients with T2D. Patients diagnosed with T2D between 2002 and 2015 identified through the 2002 Taiwan Survey of Hypertension, Hyperglycemia, and Hyperlipidemia were included. A total of 1690 men and 1641 women aged ≥40 years linked to the National Health Insurance Research Database (NHIRD) were followed up to the end of 2015 to identify the incidences of osteoporosis through ICD9-CM codes for osteoporosis or osteoporotic fractures or usage of anti-osteoporotic agents according to Anatomical Therapeutic Chemical codes determined from NHIRD. The person year approach and Kaplan–Meier analysis were then used to estimate the incidences and cumulative event rates, whereas the Cox proportional hazard model was used to calculate adjusted hazard ratios (HR) for osteoporosis events. A total of 792 new osteoporosis events were documented over a median follow-up duration of 13.6 years. Participants with T2D had higher osteoporosis risk [adjusted HR: 1.37, 95% confidence interval (CI): 1.11–1.69] compared with those without T2D. Subgroup analyses revealed that age had a marginally significant effect, indicating that T2D had a more pronounced effect on osteoporosis risk in younger population (<65 years old). No difference was found between patients stratified according to sex. In conclusion, T2D was significantly associated with increased osteoporosis risk, especially in younger participants.

Introduction

Osteoporosis, one of the most common metabolic diseases, is characterized by low mineral bone mass and microarchitectural deterioration of bone tissue, which consequently increases bone fragility and susceptibility to fractures [1]. Osteoporosis can occur in both men as well as in women, particularly aged [2, 3], and it is a significant health burden due to its high morbidity, mortality, and healthcare costs. A nutrition and health survey in Taiwan from 2013 to 2015 revealed an overall prevalence of osteoporosis at 12.3% [4]. Several risk factors for osteoporosis have been identified, some of which are complex due to the multiple mechanisms involved, such as type 2 diabetes (T2D).

T2D is also a considerably prevalent health burden both in Taiwan and worldwide [5]. According to the Nutrition and Health Survey, the overall prevalence of T2D in Taiwan is 9.3% diabetes [6]. T2D affects sugar, fat, and protein metabolism while also causing dysregulation of calcium, phosphorus, and magnesium, and subsequently promoting a series of complications, such as neuropathy, and cardiovascular, peripheral vascular, retinal, and metabolic bone diseases [7–9].

The effects of diabetes on the bone are complex. Although most studies agree that diabetes increases fracture risk [10–14], the association between T2D and risk for bone mineral density (BMD) loss have been inconsistent in prior studies, with abundant data showing a greater baseline BMD in individuals with T2D [15–17]. Osteoporosis is clinically diagnosed using both the World Health Organization criteria based on BMD [18] and the incidence of fragility fractures without T-score data [19]. Although numerous studies have assessed the risk of fractures or BMD change, only a few cross-sectional studies have explored the composite risk of osteoporosis in diabetes [20].

Given the growing prevalence of T2D and osteoporosis as well as the lack of information from cohort-based studies, this study aimed to determine the relationship between T2D and osteoporosis among a Taiwanese population.

Materials and methods

Study design and participants

This nationwide, representative community-based cohort study included all patients aged ≥ 40 years from the 2002 Taiwanese Survey on Prevalence of Hypertension, Hyperglycemia, and Hyperlipidemia (TwSHHH). The exclusion criteria were osteoporosis diagnosis, use of medication for osteoporosis, and a history of non-traumatic fractures diagnosed on the basis of International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes (S1 Table) at least twice as per the outpatient department note or once as per the hospitalization discharge note before the index date. Those who were pregnant within 1 year of the 2002 TwSHHH and those with type 1 diabetes diagnosed using ICD-9-CM codes (250 ×1 and 250 × 3) at least twice as per the outpatient department note or once as per hospitalization discharge note were also excluded.

The 2002 TwSHHH followed the 2001 National Health Interview Survey (NHIS) of Taiwan, a nationwide health survey, between March 2002 and October 2002 [21]. Half of the primary sampling units of the 2001 NHIS were randomly selected, and all members aged ≥15 years were interviewed for the 2002 TwSHHH study [22].

All participants of the 2002 TwSHHH were interviewed using a questionnaire that include questions on sociodemographic characteristics, dietary characteristics, menopause status in women, and personal histories, including smoking, exercise status, medical history (hypertension, diabetes, etc.), medication history, and family history of cardiovascular diseases. Moreover, participants underwent physical examinations, to measure body mass index (BMI), waist and hip circumferences, and blood pressure (BP), and laboratory tests to determine parameters such as fasting plasma glucose (FPG), hemoglobin A1c (HbA1c), uric acid, renal and liver function, and lipid profiles. All laboratory data were obtained using standard protocols. The coefficients of variations for these laboratory measurements were approximately 5%. A total of 7578 participants in the 2002 TwSHHH completed the questionnaires, and 6941 of them had BP data (634 refusals) and 6602 provided a blood sample. Finally, 6600 participants completed the survey, and the data of 5786 participants (2927 men and 2959 women) were successfully linked with their data from the 2001 NHIS database [23].

Taiwan’s National Health Insurance (NHI) program, which covers >99% of its 23 million residents [24], is a mandatory single-payer medical insurance program that was implemented in 1995 and it provides comprehensive health care services. The National Health Insurance Research Database (NHIRD) includes all data on NHI resource utilization, including outpatient visits, hospital care, prescribed medications, and National Death Registry. All data in the NHIRD are anonymous and encrypted for security purposes. Researchers using the NHIRD must state that they have no any intention of violating the insurants’ privacy. The data of the 2002 TwSHHH populations are linked with that of the 2001 NIHS and NHIRD.

The datasets generated and analyzed herein are not publicly available due to the terms of consent to which the participants agreed. However, such data can be accessed from the authors upon reasonable request and with permission from the Health Promotion Administration at the Ministry of Health and Welfare in Taiwan at https://dep.mohw.gov.tw/DOS/cp-2516-3591-113.html (approved number: H108044). All interested researchers can direct their data inquiries to ntuepm@ntu.edu.tw, an institutional, non-author point of contact. Data are also available from the National Taiwan University Hospital Institutional Data Access/Ethics Committee (contact via ntuepm@ntu.edu.tw) for researchers who satisfy the criteria for accessing confidential data.

Our protocol was reviewed and approved by the Research Ethics Committee of National Taiwan University Hospital. The committee was organized under and operates in accordance with the Good Clinical Practice Guidelines [NTUH-REC Number: 201901103W (Institutional Review Board reference)]. Written informed consent was obtained from all participants.

Definition of diabetes

Diabetes was defined as an FPG concentration of ≥ 126 mg/dL or HbA1c of ≥ 6.5% determined from the 2002 TwSHHH or as the use of antidiabetics according to the Anatomical Therapeutic Chemical (ATC) codes determined from the NHIRD (≥ 28 tablets 1 year before the index date) (S2 Table).

Definition of outcome

This study used a composite diagnosis of osteoporosis comprising at least two outpatient diagnosis or one hospital discharge diagnosis based on ICD9-CM codes for osteoporosis or non-traumatic fractures (S1 Table) or the use of ≥14 tablets or one injectable dose of anti-osteoporotic agents (excluding calcium, vitamin D, or hormone replace therapy) between 2002 and 2015 according to the ATC codes (S2 Table) determined from the NHIRD.

Covariates

Patient characteristics, including age, sex, menopause status, calcium consumption status, exercise habits, smoking status, daily alcohol consumption, and socioeconomic status (including marital status, income, and education level) were obtained from the 2001 NHIS and 2002 TwSHHH. Medication- and diagnosis based comorbidities were determined from the NHIRD (S3 Table).

Statistical analyses

Descriptive analyses were performed. Categorical and continuous variables were analyzed using the chi-square test and analysis of variance, respectively. Kaplan–Meier survival curves for osteoporosis events were plotted according to baseline diabetes status, after which the log rank test was performed to determine the differences between both groups. Person-years were calculated for each participant from the date at which the 2002 TwSHHH questionnaire was competed until the occurrence of an event, death, or till December 31, 2015, whichever came first. Osteoporosis incidence rates were expressed as the number of cases divided by the number of 1000-person years of follow-up.

Cox proportional hazards regression analysis was performed to determine multivariable adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) after testing the proportionality assumption using the time-dependent covariate method [25]. Potential confounders were then adjusted using three models. Baseline model 1 adjusted for sex (male, female) and age (40–64, ≥65 years old); model 2 for factors in model 1 plus BMI category (<18.5,18.5–23.9, ≥24–26.9, and 27 kg/m2), smoking status (current, non-current smoker), exercise (regular, non-regular exercise), daily alcohol consumption (yes, no), menopause status (yes, no), marriage status (living with spouse, etc.), educational level (≥9 years, <9 years), and income [≥40,000 New Taiwan dollars (NTD), <40000 NTD]; and model 3 for factors in model 2 plus serum creatinine, alanine aminotransferase (ALT), hyperthyroidism, hypertension, hyperlipidemia, high calcium diet, long-term systemic steroid use, and oophorectomy.

Potential effect modifiers, such as sex and age (cutoff 65 years), were assessed using the likelihood ratio test to compare the goodness-of-fit of the models with and without the interaction terms in the fully adjusted model 3.

Sensitivity analysis was performed to test the robustness of our results, excluding events in the first year to enhance causal relationships and excluding fractures from the endpoint. To determine the correlation between glycemic control and osteoporosis risk, participants were also divided into two groups according to HbA1 level (cutoff 7%), after which Cox proportional hazards regression analysis was conducted. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC) with a two-tailed alpha level of <0.05 indicating statistical significance.

Results

Descriptive analyses

After excluding 3375 participants at baseline who were pregnant within 1 year of the 2002 TwSHHH (n = 18), ≦40 years old (n = 3079), diagnosed with osteoporosis before the 2002 TwSHHH (n = 227), had missing data (n = 1), and were diagnosed with type 1 diabetes (n = 50), a total of 3331 participants [1690 men and 1641 women; mean (standard deviation) age of 55.1 (11.8) years] were included. The study flow chart is presented in Fig 1.

Fig 1. Flow diagram of participant enrollment.

Fig 1

Open in a new tab

Baseline characteristics of the study population are detailed in Table 1. Overall, 390 individuals were identified to have had diabetes, yielding a prevalence rate of 11.7%. Expectedly participants with diabetes tended to be older (mean age 59.5 years old) and had significantly higher BMI, waist circumference, systolic BP, and diastolic BP compared to those without diabetes. The diabetes group had more participants living with their spouse as well as a higher education level, greater monthly income, and higher proportion of participants with menopause, long-term systemic steroid use, hypertension, and hyperlipidemia, compared to the nondiabetes group. Diabetes was significantly associated with higher levels of FPG, triglyceride, low-density lipoprotein cholesterol, ALT, and creatinine and lower levels of high-density lipoprotein cholesterol. No difference in lifestyle factors, such as smoking status, daily alcohol consumption, and exercise habits, were observed between both groups.

Table 1. Baseline characteristics of the study participants.

Without diabetes (n = 2941) With diabetes (n = 390) p value
Mean (SD) Mean (SD)
Age (years old) 54.5 (11.8) 59.5 (11) <0.05*
BMI (kg/m2) 23.8 (3.3) 25.2 (3.7) <0.05*
Waist circumference (cm) 82.3 (9.8) 87.8 (10.3) <0.05*
Systolic blood pressure (mmHg) 120.9 (18.6) 129.9 (19.2) <0.05*
Diastolic blood pressure (mmHg) 77.9 (11.5) 80.4 (11.3) <0.05*
Fasting plasma glucose (mg/dL) 90.6 (9.7) 157.8 (65.8) <0.05*
Triglycerides (mg/dL) 133.6 (82.5) 189.1 (124.7) <0.05*
LDL-C (mg/dL) 122.1 (26.3) 129.1 (31.6) <0.05*
HDL-C (mg/dL) 57.1 (15.8) 52 (18.8) <0.05*
ALT (mg/dL) 20.3 (14.4) 22.8 (19.0) <0.05*
Creatinine (mg/dL) 0.9 (0.3) 1.0 (0.6) <0.05*
HbA1C (%) 5.2 (0.4) 7.9 (1.9) <0.05*
n (%) n (%)
40–64 (years) 2310 (78.5) 258 (66.2) <0.05*
≥ 65 (years) 631 (21.5) 132 (33.9)
Women 1455 (49.5) 186 (47.7) 0.51
BMI < 18.5 (kg/m2) 95 (3.5) 10 (2.9) <0.05*
18.5 ≤ BMI < 24 kg/m2 1432 (52.4) 124 (35.6)
24 ≤ BMI < 27 kg/m2 818 (29.9) 120 (34.5)
BMI ≥ 27 kg/m2 388 (14.2) 94 (27)
Current smokers 691 (23.5) 103 (26.4) 0.20
Alcohol use almost every day 156 (5.3) 19 (4.9) 0.72
Regular exercise habit 776 (26.4) 105 (26.9) 0.82
Living with spouse 2369 (80.6) 297 (76.2) <0.05*
Educational level (≥9 years of schooling) 1125 (38.3) 89 (22.8) <0.05*
Average month income ≥ 40,000 NTD 644 (21.9) 56 (14.4) <0.05*
Menopause 730 (24.8) 148 (38) <0.05*
Oophorectomy 65 (2.2) 8 (2.1) 0.84
High calcium diet 1353 (46) 164 (42.1) 0.14
Long-term systemic steroid use 224 (7.6) 41 (10.5) <0.05*
Hypertension 637 (21.7) 148 (38) <0.05*
Hyperlipidemia 1271 (43.2) 254 (65.1) <0.05*
Hyperthyroidism 619 (21.1) 71 (18.2) 0.75
Open in a new tab

ALT, Alanine Aminotransferase; BMI, Body mass index; HDL-C, High-density lipoprotein cholesterol; LDL-C, Low-density lipoprotein cholesterol; NTD, New Taiwan Dollars; SD: standard deviation

*p < 0.05.

Over a median follow-up duration of 13.6 years, 792 incident cases of osteoporosis were documented, with the diabetes and nondiabetes groups having an incidence rate of 34.20 and 20.55 per 1000 person-years, respectively.

Kaplan–Meier survival curves for osteoporosis according to the presence of diabetes showed that participants with diabetes had significantly lower osteoporosis event-free rates compared with those without diabetes (Fig 2).

Fig 2. Kaplan–Meier survival curves for osteoporosis according to the presence of diabetes.

Fig 2

Open in a new tab

Inferential analyses

Table 2 shows the HRs and 95% CIs for osteoporosis according to the presence of diabetes. After adjusting for age and sex, the HR for osteoporosis in the diabetes group was 1.44 (95% CI: 1.18–1.75; p < 0.001). After additional adjustment for BMI, smoking status, daily alcohol consumption, regular exercise habits, menopause, marital status, educational level, and income, the HR for osteoporosis in the diabetes group was 1.37 (95% CI: 1.11–1.69; p = 0.003). After additional adjustment for serum creatinine level, ALT, hyperthyroidism, hypertension, hyperlipidemia, high calcium diet, long-term systemic steroid use, oophorectomy, the HR for osteoporosis in the diabetes group was 1.37 (95% CI: 1.11–1.69; p = 0.004).

Table 2. Risk of osteoporosis according to the presence of diabetes.

Variables With diabetes Without diabetes
Participants 390 2,941
Person years 3596.43 32,550.45
Events 123 669
Incidence rate (per 1,000 person years) 34.20 20.55
Model 1 1.44 (1.18, 1.75)* 1
Model 2 1.37 (1.11, 1.69)* 1
Model 3 1.37 (1.11, 1.69)* 1
Open in a new tab

Data are presented as hazard ratios with 95% confidence intervals.

Model 1: Adjusted for sex and age (40–64 and ≥65 years old).

Model 2: Additional adjustment for body mass index (<18, 18–24, 24–27, and ≥27kg/m2), current smoking (yes/no), daily alcohol consumption (yes/no), regular exercise habit (yes/no), menopause (yes/no), living with spouse (yes/no), educational level (</≥9 years of schooling), average monthly income (</≥40,000 New Taiwan Dollars).

Model 3: Additional adjustment for serum creatinine level, alanine aminotransferase, hyperthyroidism, hypertension, hyperlipidemia, high calcium diet, long-term systemic steroid use, and oophorectomy.

Subgroup analysis according to age categories (<65/≥65 years old) found that the association between the presence of diabetes and risk for osteoporosis was marginally modified by age, which suggested that diabetes plays a more significant role in the risk of osteoporosis among non-elderly individuals (interaction p = 0.06).

No difference was found between subgroups stratified according to sex (interaction p = 0.97). After stratification, a significantly higher risk of osteoporosis was still observed among women (adjusted HR: 1.42, 95% CI: 1.09–1.85) but not men (adjusted HR: 1.34, 95% CI:0.94–1.92) (Table 3).

Table 3. Subgroup analyses for the risk of osteoporosis according to the presence of diabetes.

Variables With diabetes Without diabetes pinteraction
Age 0.06
 40–65 years old 1.34 (0.94, 1.92) 1
 ≥65 years old 1.08 (0.77, 1.51) 1
Sex 0.97
 Women 1.42 (1.09, 1.85) 1
 Men 1.34 (0.94, 1.92) 1
Open in a new tab

Adjustment based on model 3. Data are presented as hazard ratios and 95% confidence intervals.

Sensitivity analyses demonstrated that results excluding events within 1 year after the 2002 TwSHHH were consistent with those obtained in the main analyses (Table 4).

Table 4. Sensitivity analyses for the risk of osteoporosis according to the presence of diabetes.

Sensitivity analyses With diabetes Without diabetes
Excluding events in the first year 1.39 (1.1, 1.74)* 1
Osteoporosis without fracture 1.52 (1.11, 2.08)* 1
Open in a new tab

Adjustment based on model 3. Data are presented as hazard ratios and 95% confidence intervals.

Participants with HbA1C ≥7 had significantly higher osteoporosis risk than those with HbA1C <7 (adjusted HR:1.49, 95% CI: 1.15–1.92; p = 0.002) (Table 5).

Table 5. Risk of osteoporosis according to HbA1c level.

HbA1c level HbA1c <7 HbA1c ≥7
HR 1 1.49 (1.15, 1.92)*
Open in a new tab

Adjustment based on model 3. Data are presented as hazard ratios and 95% confidence intervals.

Discussion

The current study suggested a positive association between the presence of T2D and incidences of osteoporosis in a Taiwanese population, with such an association being more pronounced among those aged <65 years old. This result indicated the presence of several important risk factors apart from T2D that should not be neglected among older individuals. Our results showed no difference on the basis of sex, suggesting that T2D may be associated with osteoporosis in both men and women. Moreover, the current study found that there was positive correlation between poor glucose control and osteoporosis risk.

Osteoporosis is clinically diagnosed using BMD or the presence of non-traumatic fractures. Although numerous studies have demonstrated an association between T2D and the risk of osteoporotic fractures [10, 11, 14, 26–28], the results of previous studies on osteoporosis prevalence among patients with T2D are inconsistent. Accordingly, some studies showed that those with T2D had a lower, if not similar, of prevalence of osteoporosis determined using BMD compared to controls [29–31]. One possible explanation could be the degenerative changes and diffuse idiopathic skeletal hyperostosis frequently found in patients with T2D [32]. In contrast, a systemic review in China reported a higher prevalence of osteoporosis in patients with T2D [33]. A cross-sectional Korean population study also found an association between reduced BMD and diabetes duration [34]. Despite the lack of BMD data in our database, we attempted to use a composite diagnosis of osteoporosis through ICD9-CM codes of both osteoporosis and non-traumatic fractures and ATC codes of anti-osteoporotic agents to establish a more comprehensive osteoporosis diagnosis. Our data consistently suggested a significantly greater risk of osteoporosis among Taiwanese patients with T2D.

Long-term exposure to diabetes promotes changes in bone metabolism and impairs bone micro-architecture through multiple mechanisms, including elevated insulin levels, hypercalciuria, reduced renal function, obesity, more advanced glycation end products in collagen, angiopathies, neuropathies, and inflammation [7, 35]. In particular, insulin stimulates osteoblast proliferation and differentiation, whereas high glucose levels directly affect osteoblast metabolism and maturation by altering gene expression and diminishing bone mineral quality (7). Meanwhile, BMI is believed to be positively associated with BMD due to increased loading, adipokines, and higher aromatase activity [36]. Increased risk for falls caused by diabetic retinopathy, advanced cataracts, hypoglycemia, peripheral neuropathy, foot ulcers, polyuria, and decreased reflexes may also increase the rate at which osteoporosis is diagnosed in patients with T2D [37].

The goal of osteoporosis treatment has always been fracture prevention. As such, apart from those diagnosed with osteoporotic fractures, patients diagnosed with osteoporosis without fracture should also be monitored. Our findings highlight the need for clinicians to monitor bone health, including BMD and fracture conditions, in patients with T2D. Early interventions aimed at enhancing bone health have to be implemented to prevent osteoporosis events, even in younger populations and regardless of sex.

This study has several strengths. First, although evidence of a higher prevalence of osteoporosis had been found in Asian populations, prospective cohort studies have not been available. To the best of our knowledge, this has been the first cohort study in Asia to explore the relationship between T2D and composite osteoporosis risk. Second, this study was a representative cohort study with validated outcomes based on the TwSHHH database. Moreover, the community-based populations from the TwSHHH database may reduce selection bias. The TwSHHH database includes important clinical laboratory variables and socioeconomic and lifestyle factors needed to adjust for potential confounding factors. Lastly, unlike most studies, our study surveyed younger populations, although marginal differences according to age had been noted.

Several potential limitations of the current study are worth noting. First, no BMD data for outcome evaluation had been available. To detect most events, both ICD9-CM codes for osteoporosis or osteoporotic fractures and ATC codes of anti-osteoporotic drugs determined from the NHIRD were used. Second, relatively few incident cases of osteoporosis had been observed for risk estimations. A median follow-up duration of 13.6 years already makes our result significant; however, osteoporosis may take longer to develop. If we follow a longer duration, more prominent results can be seen. Third, limited information was available regarding the effects of antidiabetic medications, calcium, or vitamin D on osteoporosis given that our study focused on primary prevention. Fourth, several of the covariates assessed only at baseline can change over time. Some individuals without T2D at baseline may eventually develop diabetes, whereas those diagnosed with T2D at baseline will have retained their diagnosis, which may be of particular significance. Considering the already significant positive result before adjustment, no further analysis was conducted on those who were newly diagnosed with diabetes.

Conclusions

T2D was significantly associated with the risk of osteoporosis, especially among non-elderly participants, regardless of sex.

Supporting information

S1 Table. ICD-9-CM diagnostic codes for osteoporosis.

(DOCX)

Click here for additional data file. (14.5KB, docx)
S2 Table. Anatomical therapeutic chemical codes used to define the medications in the study cohort.

(DOCX)

Click here for additional data file. (15.2KB, docx)
S3 Table. The definition of the covariates in the study cohort.

(DOCX)

Click here for additional data file. (15.6KB, docx)
S1 File. Raw data.

(DOCX)

Click here for additional data file. (36.9KB, docx)

Acknowledgments

The authors would like to thank Enago (www.enago.tw), the editing brand of Crimson Interactive Pvt., Ltd. for the English language review.

Data Availability

The datasets generated and analyzed herein are not publicly available due to the terms of consent to which the participants agreed. However, such data can be accessed from the authors upon reasonable request and with permission from the Health Promotion Administration at the Ministry of Health and Welfare in Taiwan at https://dep.mohw.gov.tw/DOS/cp-2516-3591-113.html (approved number: H108044). All interested researchers can direct their data inquiries to ntuepm@ntu.edu.tw, an institutional, non-author point of contact. Data are also available from the National Taiwan University Hospital Institutional Data Access/Ethics Committee (contact via ntuepm@ntu.edu.tw) for researchers who satisfy the criteria for accessing confidential data.

Funding Statement

The authors received no specific funding for this work.

References

  • 1.Seeman E, Delmas PD. Bone quality—the material and structural basis of bone strength and fragility. N Engl J Med. 2006;354(21):2250–2261. doi: 10.1056/NEJMra053077 [DOI] [PubMed] [Google Scholar]
  • 2.Gullberg B, Johnell O, Kanis JA. World-wide projections for hip fracture. Osteoporos Int. 1997;7(5):407–413. doi: 10.1007/pl00004148 [DOI] [PubMed] [Google Scholar]
  • 3.Burge R, Dawson-Hughes B, Solomon DH, Wong JB, King A, Tosteson A. Incidence and economic burden of osteoporosis-related fractures in the United States, 2005–2025. J Bone Miner Res. 2007;22(3):465–475. doi: 10.1359/jbmr.061113 [DOI] [PubMed] [Google Scholar]
  • 4.Health Promotion Administration MoHaW (2015) Metabolic syndrome. [updated 12-27-2017. https://www.hpa.gov.tw/Pages/Detail.aspx?nodeid=639&pid=1219.
  • 5.World Health Organization, Diabetes [https://www.who.int/health-topics/diabetes.
  • 6.Health Promotion Administration MoHaw. Statistical yearbook of health promotion (2018).
  • 7.Murray CE, Coleman CM. Impact of diabetes mellitus on bone health. Int J Mol Sci. 2019;20(19). doi: 10.3390/ijms20194873 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Rangel EB, Rodrigues CO, de Sa JR. Micro- and macrovascular complications in ciabetes mellitus: preclinical and clinical studies. J Diabetes Res. 2019;2019:2161085. doi: 10.1155/2019/2161085 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Zheng Y, Ley SH, Hu FB. Global aetiology and epidemiology of type 2 diabetes mellitus and its complications. Nat Rev Endocrinol. 2018;14(2):88–98. doi: 10.1038/nrendo.2017.151 [DOI] [PubMed] [Google Scholar]
  • 10.Bai J, Gao Q, Wang C, Dai J. Diabetes mellitus and risk of low-energy fracture: a meta-analysis. Aging Clin Exp Res. 2020;32(11):2173–2186. doi: 10.1007/s40520-019-01417-x [DOI] [PubMed] [Google Scholar]
  • 11.Fan Y, Wei F, Lang Y, Liu Y. Diabetes mellitus and risk of hip fractures: a meta-analysis. Osteoporos Int. 2016;27(1):219–228. doi: 10.1007/s00198-015-3279-7 [DOI] [PubMed] [Google Scholar]
  • 12.Formiga F, Freitez Ferreira MD, Montero A. [Diabetes mellitus and risk of hip fracture. A systematic review]. Rev Esp Geriatr Gerontol. 2020;55(1):34–41. doi: 10.1016/j.regg.2019.08.009 [DOI] [PubMed] [Google Scholar]
  • 13.Koromani F, Oei L, Shevroja E, Trajanoska K, Schoufour J, Muka T, et al. Vertebral Fractures in Individuals with type 2 diabetes: more than skeletal complications alone. Diabetes Care. 2020;43(1):137–144. doi: 10.2337/dc19-0925 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Moayeri A, Mohamadpour M, Mousavi SF, Shirzadpour E, Mohamadpour S, Amraei M. Fracture risk in patients with type 2 diabetes mellitus and possible risk factors: a systematic review and meta-analysis. Ther Clin Risk Manag. 2017;13:455–468. doi: 10.2147/TCRM.S131945 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Leslie WD, Morin SN, Majumdar SR, Lix LM. Effects of obesity and diabetes on rate of bone density loss. Osteoporos Int. 2018;29(1):61–67. doi: 10.1007/s00198-017-4223-9 [DOI] [PubMed] [Google Scholar]
  • 16.Holloway-Kew KL, Marijanovic N, De Abreu LLF, Sajjad MA, Pasco JA, Kotowicz MA. Bone mineral density in diabetes and impaired fasting glucose. Osteoporos Int. 2019;30(9):1799–1806. doi: 10.1007/s00198-019-05108-1 [DOI] [PubMed] [Google Scholar]
  • 17.Li KH, Liu YT, Yang YW, Lin YL, Hung ML, Lin IC. A positive correlation between blood glucose level and bone mineral density in Taiwan. Arch Osteoporos. 2018;13(1):78. doi: 10.1007/s11657-018-0494-9 [DOI] [PubMed] [Google Scholar]
  • 18.Group TWS. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. In: Geneva: WHO, editor. 1994. [DOI] [PubMed] [Google Scholar]
  • 19.Kanis JA, Gluer CC. An update on the diagnosis and assessment of osteoporosis with densitometry. Committee of Scientific Advisors, International Osteoporosis Foundation. Osteoporos Int. 2000;11(3):192–202. doi: 10.1007/s001980050281 [DOI] [PubMed] [Google Scholar]
  • 20.Sugimoto T, Sato M, Dehle FC, Brnabic AJ, Weston A, Burge R. Lifestyle-related metabolic disorders, osteoporosis, and fracture risk in Asia: a systematic review. Value Health Reg Issues. 2016;9:49–56. doi: 10.1016/j.vhri.2015.09.005 [DOI] [PubMed] [Google Scholar]
  • 21.Hsu CC, Hwang SJ, Wen CP, Chang HY, Chen T, Shiu RS, et al. High prevalence and low awareness of CKD in Taiwan: a study on the relationship between serum creatinine and awareness from a nationally representative survey. Am J Kidney Dis. 2006;48(5):727–738. doi: 10.1053/j.ajkd.2006.07.018 [DOI] [PubMed] [Google Scholar]
  • 22.Wang WS, Wahlqvist ML, Hsu CC, Chang HY, Chang WC, Chen CC. Age- and gender-specific population attributable risks of metabolic disorders on all-cause and cardiovascular mortality in Taiwan. BMC Public Health. 2012;12:111. doi: 10.1186/1471-2458-12-111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Fan YC, Chou CC, You SL, Sun CA, Chen CJ, Bai CH. Impact of worsened metabolic syndrome on the risk of dementia: a nationwide cohort study. J Am Heart Assoc. 2017;6(9). doi: 10.1161/JAHA.116.004749 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Chang HJ. “National Health Insurance in Taiwan” (Presentation at DoH Seminar on the NHI, 6 December 2002).
  • 25.Ng’andu NH. An empirical comparison of statistical tests for assessing the proportional hazards assumption of Cox’s model. Stat Med. 1997;16(6):611–626. [DOI] [PubMed] [Google Scholar]
  • 26.Ni Y, Fan D. Diabetes mellitus is a risk factor for low bone mass-related fractures: A meta-analysis of cohort studies. Medicine (Baltimore). 2017;96(51):e8811. doi: 10.1097/MD.0000000000008811 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Li CI, Liu CS, Lin WY, Meng NH, Chen CC, Yang SY, et al. Glycated hemoglobin level and risk of hip fracture in older people with type 2 diabetes: a competing risk analysis of Taiwan diabetes cohort study. J Bone Miner Res. 2015;30(7):1338–1346. doi: 10.1002/jbmr.2462 [DOI] [PubMed] [Google Scholar]
  • 28.Janghorbani M, Van Dam RM, Willett WC, Hu FB. Systematic review of type 1 and type 2 diabetes mellitus and risk of fracture. Am J Epidemiol. 2007;166(5):495–505. doi: 10.1093/aje/kwm106 [DOI] [PubMed] [Google Scholar]
  • 29.Picke AK, Campbell G, Napoli N, Hofbauer LC, Rauner M. Update on the impact of type 2 diabetes mellitus on bone metabolism and material properties. Endocr Connect. 2019;8(3):R55–R70. doi: 10.1530/EC-18-0456 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Gilbert MP, Pratley RE. The impact of diabetes and diabetes medications on bone health. Endocr Rev. 2015;36(2):194–213. doi: 10.1210/er.2012-1042 [DOI] [PubMed] [Google Scholar]
  • 31.Leidig-Bruckner G, Grobholz S, Bruckner T, Scheidt-Nave C, Nawroth P, Schneider JG. Prevalence and determinants of osteoporosis in patients with type 1 and type 2 diabetes mellitus. BMC Endocr Disord. 2014;14:33. doi: 10.1186/1472-6823-14-33 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Sencan D, Elden H, Nacitarhan V, Sencan M, Kaptanoglu E. The prevalence of diffuse idiopathic skeletal hyperostosis in patients with diabetes mellitus. Rheumatol Int. 2005;25(7):518–521. doi: 10.1007/s00296-004-0474-9 [DOI] [PubMed] [Google Scholar]
  • 33.Si Y, Wang C, Guo Y, Xu G, Ma Y. Prevalence of osteoporosis in patients with type 2 diabetes mellitus in the Chinese mainland: a systematic review and meta-analysis. Iran J Public Health. 2019;48(7):1203–1214. [PMC free article] [PubMed] [Google Scholar]
  • 34.Jang M, Kim H, Lea S, Oh S, Kim JS, Oh B. Effect of duration of diabetes on bone mineral density: a population study on East Asian males. BMC Endocr Disord. 2018;18(1):61. doi: 10.1186/s12902-018-0290-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Pouresmaeili F, Kamalidehghan B, Kamarehei M, Goh YM. A comprehensive overview on osteoporosis and its risk factors. Ther Clin Risk Manag. 2018;14:2029–2049. doi: 10.2147/TCRM.S138000 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Walsh JS, Vilaca T. Obesity, type 2 diabetes and bone in adults. Calcif Tissue Int. 2017;100(5):528–535. doi: 10.1007/s00223-016-0229-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Kumeda Y. [Osteoporosis in diabetes]. Clin Calcium. 2008;18(5):589–599. [PubMed] [Google Scholar]

Decision Letter 0

Robert Daniel Blank

20 Apr 2021

PONE-D-21-09780

Association between type 2 diabetes and risk of osteoporosis: a nationwide cohort study

PLOS ONE

Dear Dr. Yeh,

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.

Reviewer 2 has raised important questions about the cohort, the method of identifying exposure, and endpoints that must be addressed.  Data availability must be addressed.  Reviewer 1 wishes clarification about how models 2 and 3 differ.

Please submit your revised manuscript by Jun 04 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,

Robert Daniel Blank, 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. Thank you for including your ethics statement: 

"Our protocol was reviewed and approved by the Research Ethics Committee of National Taiwan University Hospital. The committee was organized under and operates in accordance with the Good Clinical Practice Guidelines [NTUH-REC Number: 201901103W (Institutional Review Board reference)].

The data were analyzed anonymously.".   

Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information

Once you have amended this/these statement(s) in the Methods section of the manuscript, please add the same text to the “Ethics Statement” field of the submission form (via “Edit Submission”).

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research.

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

4. Please include captions for ALL 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: Partly

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: No

**********

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: Model 3 adjusts for oophorectomy. Unsure of added value given that menopause status was already adjusted for in Model 2.

The study collects HbA1c values. The authors could augment their study by looking the correlation of glycaemic control to risk of osteoporosis.

Reviewer #2: There is substantial evidence, giving rise to many meta-analyses, for increased risk of fracture in patients with type 2 diabetes (T2D) not explained from bone mineral density (BMD) which is paradoxically higher in T2D. The current report claims that “cohort studies estimating composite osteoporosis risk have been lacking”. The composite outcome proposed by the authors includes incident fractures, new diagnosis of osteoporosis (ICD-9-CM codes) or anti osteoporosis medications (NHIRD), but is problematic as noted below. Using a modest-sized cohort from a cardiovascular study (2002 Taiwanese Survey on Prevalence of Hypertension, Hyperglycemia, and Hyperlipidemia) the authors were able to identify 3,331 eligible participants including 390 with T2D at baseline. During median follow-up of 13.6 years, the authors noted an increased risk for the composite end point (adjusted HR 1.37) with perhaps slightly greater risk in younger individuals (HR 1.34 age 40 – 60 years, 1.08 age ≥ 65 years, p–interaction 0.06). The authors conclude that T2D was associated with increased risk for osteoporosis.

Major comments:

1. The composite end point chosen will conflate opposing effects seen in the individual end points. Fracture risk would be expected to be increased in T2D, whereas diagnosis and treatment of osteoporosis (from BMD) would be lower. The authors need to analyze and report the individual end points separately and, if they are divergent, reconsider the use of the composite measure.

2. The absence of BMD measurements limits the authors ability to draw conclusions about what triggers a diagnosis of osteoporosis. The use of osteoporosis diagnosis ICD-9-CM codes is an uncertain surrogate for BMD. Diagnosis of osteoporosis from BMD is usually dependent upon referral for screening DXA. Could more frequent medical contacts for T2D patients lead to more frequent DXA (differential ascertainment bias)?

3. From Table S1, the fracture end points are limited to vertebral (difficult to assess from ICD-9-CM codes), humerus and radio-ulnar. Hip fractures are the most strongly associated with T2D yet are not reported. Why?

Minor comments:

1. The title refers to a “nationwide” cohort study. This is incorrect since the cohort is only a small sample of the Taiwanese population.

2. The Introduction refers to risk factors for osteoporosis and “…multiple mechanisms involved, such as type 2 diabetes”. A reference is required. The Introduction refers to BMD loss (References #15, 16) but needs to highlight the abundant data showing greater baseline BMD in individuals with T2D.

3. Given the long follow-up many covariates assessed only at baseline will change over time. Some individuals without T2D at baseline will develop diabetes. Although the authors may not be able to address this in their analyses, it needs to be discussed.

4. Table S3 shows that oral corticosteroid use was defined from 2002-2015 data. This creates a situation where the exposure may actually follow the outcome. Time-varying methods or using a fixed covariate from baseline data avoids this.

5. The authors explain "The TwSHHH database contains data on the 2002 and 2007 TwSHHH populations linked together with the 2001 NIHS and NHIRD." It is unclear how the 2007 data were used, if at all, since all outcome measures seem to come from the NHIRD.

6. It is unclear why the sensitivity analysis excluding events within the first year was performed (Table 4). Please explain.

7. Duration alone does not overcome the many limitations. Please reword "duration of 13.6 years makes our results convincing."

8. The age stratification in Table 3 is probably incorrect (40 – 60 years, elsewhere < 65 years).

9. There are many grammatical errors that need to be corrected. Editing by a native English language speaker is required. The authors use sex and gender interchangeably. These are slightly different concepts and consistent language should be used.

**********

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.

PLoS One. 2021 Jul 13;16(7):e0254451. doi: 10.1371/journal.pone.0254451.r002

Author response to Decision Letter 0

27 May 2021

Responses to each point raised by the academic editor:

Thank you for having taken your time to provide us with reminders and suggestions which allow us to improve the quality of our manuscript. We have carefully made revisions according to your reminders.

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

Response: We appreciate for editor’s friendly reminder. We have corrected the format according to the PLOS ONE style templates including the capitalization of title, some level 2 heading in Methods and Results sections and the font size of table 1. We also adjust the file name of “raw code” to “S1 file” according to the style requirements.

2. Thank you for including your ethics statement: "Our protocol was reviewed and approved by the Research Ethics Committee of National Taiwan University Hospital. The committee was organized under and operates in accordance with the Good Clinical Practice Guidelines [NTUH-REC Number: 201901103W (Institutional Review Board reference)].The data were analyzed anonymously.". Please provide additional details regarding participant consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

Response: Thanks for the kindly reminder. We added the information accordingly in the Methods section: “Written informed consent was obtained from all the participants.” (Page 6)

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

Response：Thanks for your advice. The datasets generated and analyzed during the current study are not publicly available due to the terms of consent to which the participants agreed but data are however available from the authors upon reasonable request and with permission of the Health Promotion Administration at the Ministry of Health and Welfare in Taiwan. We revised our cover letter and provided contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

4. Please include captions for ALL 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.

Response: Thanks for your reminder. We added the raw data file at the end of our manuscript as “S1 File. Raw data.” (Page 22) as the guidelines.

Response to each point raised by reviewer #1:

Thank you for having taken your time to provide us with valuable feedback and comments which allow us to improve the quality of our manuscript. We have carefully made revisions according to your comments and suggestions.

1. Model 3 adjusts for oophorectomy. Unsure of added value given that menopause status was already adjusted for in Model 2.

Response: Thank you so much for your valuable comments. Oophorectomy is one of a well-known osteoporosis factor. According to our original questionnaire, oophorectomy may include partial oophorectomy. Partial oophorectomy may lead to early menopause. This condition is not included in menopause status when answering questionnaire. (E.K. Bjelland, P. Wilkosz, T.G. Tanbo, A. Eskild, Is unilateral oophorectomy associated with age at menopause? A population study (the HUNT2 Survey), Human Reproduction, Volume 29, Issue 4, April 2014, Pages 835–841, https://doi.org/10.1093/humrep/deu026, M. Rosendahl, M. K. Simonsen & J. J. Kjer (2017) The influence of unilateral oophorectomy on the age of menopause, Climacteric, 20:6, 540-544, DOI: 10.1080/13697137.2017.1369512)

2. The study collects HbA1c values. The authors could augment their study by looking the correlation of glycaemic control to risk of osteoporosis.

Response: Thank you so much for your valuable comments. We have analyzed the correlation of HbA1c and risk of osteoporosis as your opinion (Method, Page 8 and Results, Page 14, Discussion, Page 15). We divided the participants into 2 group according to HbA1c value. The group with HbA1C ≥7 has significantly higher osteoporosis risk than the group with HbA1C <7 (adjusted HR:1.46, 95% CI: 1.23-1.90).

Response to reviewer #2:

Thank you for having taken your time to provide us with valuable feedback and comments which allow us to improve the quality of our manuscript. We have carefully made revisions according to your comments and suggestions.

Major comments:

1. The composite end point chosen will conflate opposing effects seen in the individual end points. Fracture risk would be expected to be increased in T2D, whereas diagnosis and treatment of osteoporosis (from BMD) would be lower. The authors need to analyze and report the individual end points separately and, if they are divergent, reconsider the use of the composite measure.

Response: Thank you so much for your valuable comments. After stratification, a significantly higher osteoporosis risk was still observed among participants without fracture (adjusted HR: 1.52, 95% CI: 1.11–2.08) but not participants with fracture (adjusted HR :1.22, 95% CI:0.92–1.63). The end points are consistent in direction. We have added one more sensitivity analysis to clarify the question (Method, Page 8 and revised Table 4, Page 14).

Table 4. Sensitivity analyses for the risk of osteoporosis according to the presence of diabetes

 Sensitivity analyses With diabetes Without diabetes

Excluding events in the first year 1.39 (1.1, 1.74)* 1

Osteoporosis without fracture 1.52 (1.11, 2.08)* 1

Adjustment based on model 3, presented with hazard ratios and 95% confidence intervals

2. The absence of BMD measurements limits the authors ability to draw conclusions about what triggers a diagnosis of osteoporosis. The use of osteoporosis diagnosis ICD-9-CM codes is an uncertain surrogate for BMD. Diagnosis of osteoporosis from BMD is usually dependent upon referral for screening DXA. Could more frequent medical contacts for T2D patients lead to more frequent DXA (differential ascertainment bias)?

Response: Thank you so much for your valuable comments. More frequent medical contacts for T2D patients may introduce possible bias. However, in Taiwan, DXA is a self-paid exam, so T2D patient would not be advised to do DXA in current routine care. Also, accessibility to medical care is high in Taiwan, DXA is easily arranged for non T2D patients. Most osteoporosis diagnosis were made during health exam or after fracture. Therefore, the effect was supposed to be small.

3. From Table S1, the fracture end points are limited to vertebral (difficult to assess from ICD-9-CM codes), humerus and radio-ulnar. Hip fractures are the most strongly associated with T2D yet are not reported. Why?

Response: Thank you so much for your valuable comments. We have added back the missing ICD-9-CM codes for femoral neck fractures in our revised Table S1 as below.

Femoral neck fractures 820, 820.0, 820.2, 820.8, 820.0x, 820.2x

Minor comments:

1. The title refers to a “nationwide” cohort study. This is incorrect since the cohort is only a small sample of the Taiwanese population.

Response: Thank you so much for your valuable comments. We have revised our title to “A Representative Cohort Study in Taiwan” and also deleted “nationwide” in abstract. (Abstract, Page 2)

2. The Introduction refers to risk factors for osteoporosis and “…multiple mechanisms involved, such as type 2 diabetes”. A reference is required. The Introduction refers to BMD loss (References #15, 16) but needs to highlight the abundant data showing greater baseline BMD in individuals with T2D.

Response: Thank you so much for your valuable comments. We have added references for “…multiple mechanisms involved, such as type 2 diabetes” on page 3. We have revised “While most studies agree that diabetes increases fracture risk,(10-14), the association between T2D and risk for bone mineral density (BMD) loss in prior studies have been inconsistent.(15, 16) “ to “Although most studies agree that diabetes increases fracture risk (10-14), the association between T2D and risk for bone mineral density (BMD) loss have been inconsistent in prior studies, with abundant data showing a greater baseline BMD in individuals with T2D (15-17).” (Page 3)

3. Given the long follow-up many covariates assessed only at baseline will change over time. Some individuals without T2D at baseline will develop diabetes. Although the authors may not be able to address this in their analyses, it needs to be discussed.

Response: Thank you for your valuable comment. The control group may develop diabetes during follow-up period, but T2D patient at baseline won’t become normal. The significance may be more prominent. Due to already significant positive result before adjustment, we did not further analysis the newly develop diabetes condition. We have added the discussion of the limitation of our statistic method as “Fourth, several of the covariates assessed only at baseline can change over time. Some individuals without T2D at baseline may eventually develop diabetes, whereas those diagnosed with T2D at baseline will have retained their diagnosis, which may be of particular significance. Considering the already significant positive result before adjustment, no further analysis was conducted on those who were newly diagnosed with diabetes.” (Discussion, Page 17)

4. Table S3 shows that oral corticosteroid use was defined from 2002-2015 data. This creates a situation where the exposure may actually follow the outcome. Time-varying methods or using a fixed covariate from baseline data avoids this.

Response: Thank you for your valuable comment. According to previous studies, glucocorticoids related osteoporosis need long term steroid use and is reversible if stop steroid. So baseline data may not reflect the possible effect. We have excluded the oral corticosteroid use after event day (revised S3 table) and reanalysis. The new descriptive data is shown in Table 1 (Page 11). There is no change in other results after change the definition of oral corticosteroid use.

Long-term systemic steroid use  224 (7.6) 41 (10.5) <0.05*

5. The authors explain "The TwSHHH database contains data on the 2002 and 2007 TwSHHH populations linked together with the 2001 NIHS and NHIRD." It is unclear how the 2007 data were used, if at all, since all outcome measures seem to come from the NHIRD.

Response: Thank you for your valuable comment. We did not use 2007 data in the research this time. We have removed the statement about 2007 TwSHHH (Page 5).

6. It is unclear why the sensitivity analysis excluding events within the first year was performed (Table 4). Please explain.

Response: Thank you so much for your valuable comments. We excluded events within the first year to enhance causal relationship. The development of osteoporosis takes more than one year. We want to remove the possible bias as we can to make the causal relationship of T2D and osteoporosis stronger. We have added “to enhance causal relationships” in Method, page 8.

7. Duration alone does not overcome the many limitations. Please reword "duration of 13.6 years makes our results convincing."

Response: Thank you so much for your valuable comments. We have revised “Although osteoporosis may take longer to develop, a median follow-up duration of 13.6 years makes our results convincing.“ to “A median follow-up duration of 13.6 years already makes our result significant; however, osteoporosis may take longer to develop. If we follow a longer duration, more prominent results can be seen.” (Page 17)

8. The age stratification in Table 3 is probably incorrect (40 – 60 years, elsewhere < 65 years).

Response: Thank you so much for your valuable comments. We have revised the mistake in Table 3.

Table 3. Subgroup analyses for the risk of osteoporosis according to the presence of diabetes

Variables With diabetes Without diabetes pinteraction

Age 0.06

  40–65 years old 1.34 (0.94, 1.92) 1

  ≥65 years old 1.08 (0.77, 1.51) 1

Sex 0.97

  Women 1.42 (1.09, 1.85) 1

  Men 1.34 (0.94, 1.92) 1

9. There are many grammatical errors that need to be corrected. Editing by a native English language speaker is required. The authors use sex and gender interchangeably. These are slightly different concepts and consistent language should be used.

Response: Thank you so much for your valuable comments. We have edited the grammatical errors and inconsistent language according to a native English language speaker’s suggestion.

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file. (37.3KB, docx)

Decision Letter 1

Robert Daniel Blank

28 Jun 2021

Association between type 2 diabetes and osteoporosis risk: a representative cohort study in Taiwan

PONE-D-21-09780R1

Dear Dr. Yeh,

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,

Robert Daniel Blank, 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

**********

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

**********

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

Reviewer #1: 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

**********

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

**********

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: (No Response)

**********

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

Acceptance letter

Robert Daniel Blank

5 Jul 2021

PONE-D-21-09780R1

Association between type 2 diabetes and osteoporosis risk: a representative cohort study in Taiwan

Dear Dr. Yeh:

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

Professor Robert Daniel Blank

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Table. ICD-9-CM diagnostic codes for osteoporosis.

    (DOCX)

    Click here for additional data file. (14.5KB, docx)
    S2 Table. Anatomical therapeutic chemical codes used to define the medications in the study cohort.

    (DOCX)

    Click here for additional data file. (15.2KB, docx)
    S3 Table. The definition of the covariates in the study cohort.

    (DOCX)

    Click here for additional data file. (15.6KB, docx)
    S1 File. Raw data.

    (DOCX)

    Click here for additional data file. (36.9KB, docx)
    Attachment

    Submitted filename: Response to Reviewers.docx

    Click here for additional data file. (37.3KB, docx)

    Data Availability Statement

    The datasets generated and analyzed herein are not publicly available due to the terms of consent to which the participants agreed. However, such data can be accessed from the authors upon reasonable request and with permission from the Health Promotion Administration at the Ministry of Health and Welfare in Taiwan at https://dep.mohw.gov.tw/DOS/cp-2516-3591-113.html (approved number: H108044). All interested researchers can direct their data inquiries to ntuepm@ntu.edu.tw, an institutional, non-author point of contact. Data are also available from the National Taiwan University Hospital Institutional Data Access/Ethics Committee (contact via ntuepm@ntu.edu.tw) for researchers who satisfy the criteria for accessing confidential data.

    Articles from PLoS ONE are provided here courtesy of PLOS

    RESOURCES