لخّصلي

METHODS
Data source
We analyzed the Korean nationwide cohort database of
the National Health Insurance Service (NHIS) from 2014 to
2020. A cohort database was requested, and data were
provided after the NHIS reviewed the data provision. This
cohort database includes information related to the medical
usage of the sample population for a 6-year period from
2014 to 2020. Korean NHIS is an insurance system that requires all citizens to subscribe to health insurance. The NHIS
collects the health data of all people enrolled in the National
Health Insurance and uses it for various research and policy decisions to improve national health. Information related to disease codes and drug use was confirmed using a
nationwide cohort database. All methods were performed
in accordance with relevant guidelines and regulations. The
requirement of informed consent was waived by the appropriate board. The study protocol was approved by the Institutional Review Board (IRB) of the Dongguk University Ilsan
Hospital (IRB no. DUIH 2020-05-010-002). To comply with
the privacy rules of health insurance transfers and other privacy laws, all personal identification numbers were encrypted by converting them into scrambled numbers before data
processing. Therefore, the use of anonymized information
exempted this study from the need for written informed
consent.
Study population
Patients who had been prescribed anti-peptic agents, such
as PPI, H2RA, or MPA, for more than 2 days between 2014
and 2020 were selected from the NHIS database. The cohort
entry date was defined as the date of the first prescription
of anti-peptic agents during the study period from 2014 to
2020. The end point date was defined as the time of osteoporotic fracture occurrence in the study group and the end
date of the NHIS claims data in the control group. Upon cohort entry, all patients were required to be at least 40 years
old and have at least 1 year of prior medical information.
Patients under the age of 50, those diagnosed with malignancy, acquired immunodeficiency syndrome, osteoporosis
and fracture, and those prescribed anti-peptic agents within
1 year prior to the entry date were excluded.
The finally enrolled patients were divided into case and
control groups based on the occurrence of osteoporotic
fractures. In the case group, osteoporotic fractures occurred
among anti-peptic agent users. Osteoporotic fractures were
confirmed according to International Classification of Diseases 10th Revision (ICD-10) codes included in the medical claims data of the NHIS. This included patients with a
diagnosis code of osteoporotic fracture (M80) and those
with fracture codes for specific locations (thoracic vertebrae
[S220-S221], lumbar vertebra [S320, S327], spine [M484,
M485], proximal humerus [S422, S423], femur [S720,
S721], and distal radius [S525, 526]), following treatment
and diagnosis of osteoporosis (M81, M82) [15]. The control
group was comprised of anti-peptic agent users who did
not experience osteoporotic fractures. Case-control matching was performed between both groups using age and sex
as variables.
Drug exposure and covariates
We confirmed the prescription and duration of PPI, H2RA,
and MPA using the drug code (main ingredient code) in
the NHIS claims database. The PPIs included rabeprazole,
pantoprazole, S-pantoprazole, lansoprazole, dexlansoprazole, omeprazole, esomeprazole, and ilaprazole. The H2RAs
included roxatidine, nizatidine, lafutidine, famotidine, cimetidine, and ranitidine. The MPAs included eupatilin, irsogladine, ecabet sodium, sodium alginate, misoprostol,
rebamipide, teprenone, and troxipide. During the study period, the groups of anti-peptic agents were classified into
groups PPI alone; H2RA alone; MPA alone; PPI and MPA;
H2RA and MPA; PPI and H2RA; and PPI, H2RA, and MPA
based on drug exposure. In each group, anti-peptic agents
could be either used simultaneously or independently.
We also identified concomitant drugs that may affect
osteoporotic fractures such as bisphosphonates, glucocorticoids, anticonvulsants, hormone replacement therapy,
warfarin, heparin, antacids, selective serotonin reuptake
inhibitors, benzodiazepines, and tricyclic antidepressants.
Concomitant diseases such as chronic obstructive pulmonary disease (COPD), testicular dysfunction, hypothalamic
dysfunction, hyperthyroidism, hyperparathyroidism, Cushing’s syndrome, hyperprolactinemia, vitamin D deficiency,
idiopathic hypercalciuria, diabetes, anorexia nervosa, systemic lupus erythematosus, hypertension, intestinal absorption disorder, inflammatory bowel disease, chronic kidney
disease, and secondary amenorrhea were also identified.
The Charlson Comorbidity Index (CCI) score was confirmed
by considering these diseases and comorbidities [16].
Study outcomes and statistical analysis
The primary outcome was to confirm the use of anti-peptic
agents and the risk of osteoporotic fracture during the study
period. The secondary outcomes were to determine the risk
of osteoporotic fracture according to the type of anti-peptic
agent used and duration of use. Categorical variables, such
as sex, types of health insurance, presence of comorbidities,
and presence of concomitant medications, are presented as
frequencies and percentages, while continuous variables,
such as age and CCI, are presented as mean and standard
deviation. Conditional logistic regression analysis was performed to determine the association between the use of anti-peptic agents and the incidence of osteoporotic fractures,
and the results are presented as hazard ratio (HR) and 95%
confidence interval (CI). If any diseases or drugs known to
be associated with osteoporotic fractures were identified
during the study period, they were considered major risk
factors. Logistic regression analysis was conducted for each
factor to determine its statistical significance in predicting the risk of osteoporotic fractures. A significance level of
p < 0.05 was used for each factor included as a covariate in
the regression model. Statistical analysis was performed on
all data using SAS version 9.3 (SAS Institute Inc., Cary, NC,
USA), and statistical significance was tested with a significance probability of 0.05 and a two-tailed test.