The common cold is a viral infection of the upper respiratory tract that is one of the most common reasons for seeing a doctor [1,2]. Despite being a non-serious condition, it imposes significant costs on society in terms of medical visits, medication expenses, and lost productivity [3,4]. In Korea, people are more likely to visit a physician for minor ailments such as the common cold due to better access to medical services [5].

Treatment for the common cold typically focuses on relieving symptoms such as nasal congestion, sneezing, and coughing, and is often done with decongestants, antihistamines, and other over-the-counter medications [1-3,6,7]. These medications have been found to have a low risk of causing gastrointestinal (GI) side effects when used temporarily. Additionally, the common cold itself does not typically cause GI symptoms [8-14]. Therefore, prescribing GI drugs for the treatment of a common cold is unnecessary and can lead to strain on healthcare finances without providing any additional therapeutic benefits, unless there are underlying or accompanying GI disorders [15]. This practice of prescribing unnecessary GI drugs has been identified as an area of concern in prescribing habits and has raised concerns in several studies [15-17].

This study aims to investigate the national prevalence of potentially inappropriate GI drug prescriptions for patients with the common cold in ambulatory settings and identify the associated factors in Korea.

Study Data

We analyzed data from the National Patients Sample (NPS) of the Korean Health Insurance Review and Assessment (HIRA) Service from 2018 (HIRA-NPS-2018-0078). HIRA is a government agency that collects claims data for reimbursement purposes [18]. The NPS data is a nationally representative sample of all beneficiaries that were extracted using random sampling methods stratified by sex and age from all patients who used medical services in 2018 [19]. The data consists of the complete insurance claim data of 3% of the general population [19].

The diagnostic information in the HIRA-NPS was coded using the International Classification of Diseases 10th Revision (ICD-10) code. The data also contains information on patient characteristics (age, sex, and type of insurance), medical institutions (type of institution and region), and outpatient and inpatient clinical management (medical procedures and medications).

Study Population

The study population consisted of patients who were diagnosed with a common cold as the primary diagnosis in December 2018, identified using ICD-10 code J00. To avoid duplication, we selected only the first instance of a common cold diagnosis per patient. We included patients aged 20 years and older who visited a primary care clinic. Patients diagnosed with GI disorders (ICD-10 code: K20, K21, K22, K23, K25, K26, K27, K28, K29, and K30) from September to November and patients without any drug prescribed for a common cold were excluded (Appendix 1 and 2). Patient characteristics, such as age, insurance type, and comorbidities, as well as institutional and physician characteristics, including regions and specialized areas of clinics (as registered by primary care clinics with administrative authority), were extracted for analysis [20].

Potentially Inappropriate GI Medication Utilization

We conducted a study to examine the utilization of GI medications in the outpatient prescriptions of individuals with the common cold. We estimated the percentage of unnecessary use of these medications, which included antiulcer drugs (such as antacids, H2-receptor blockers, proton pump inhibitors, and mucosal protective agents) and GI tract regulators (such as GI tract stimulants, anticholinergics, and others; Appendix 3).

Statistical Analysis

We used descriptive statistics and chi-square tests to analyze the data, and performed a multiple logistic regression analysis to identify predictors for prescribing GI medications for the common cold.

The factors considered in this analysis were sex, age, region and specialized areas of clinics, use of non-steroidal anti-inflammatory drugs (NSAIDs) use, and the total number of cold drugs prescribed. C-statistics and the Hosmer-Lemeshow test were used to check the goodness-of-fit of the models. We used R statistical software (version 4.0.3; R Foundation for Statistical Computing, Vienna, Austria) for data analysis and set statistical significance at p<0.05.

Subject Characteristics

During the study period, we identified 26,752 patients who had been diagnosed with a common cold. After excluding those were under the age of 20 and those who visited medical institutions other than primary care clinics, we selected a total of 10,776 patients for further analysis. Of these, 3,530 patients were included in the final analysis, after excluding those without any drug prescription associated with the common cold, and those diagnosed with GI disorders within three months prior to the diagnosis of colds (Fig. 1).

Figure 1. Case extraction diagram. GI=gastrointestinal, HIRA=health insurance review and assessment service, NPS=national patients sample.

The demographic and clinical characteristics of the study population are summarized in Table 1. The proportion of female patients was higher than that of male patients (51.76% vs. 48.24%), and almost all patients were covered by the National Health Insurance (98.47%). A significant proportion of patients (57.22%) were prescribed NSAIDs, and the most prevalent comorbid disorder was hypertension (16.09%), followed by diabetes mellitus (8.53%).

Table 1 . Demographic characteristics of study population.

Explanatory variables	N	(%)	Potentially inappropriate GI medication use	(%)	p-value
Overall	3530		1546	43.80
Sex					0.003
Male	1703	48.24	702	41.22
Female	1827	51.76	844	46.20
Age					<0.001
20-29	645	18.27	270	41.86
30-39	707	20.03	266	37.62
40-49	810	22.95	358	44.20
50-64	975	27.62	468	48.00
≥65	393	11.13	184	46.82
Insurance type					0.857
NHI	3476	98.47	1523	43.82
MedAid	54	1.53	23	42.59
PVI	-		-
Specialized area of clinics					<0.001
General practitioner	1101	31.19	486	44.14
Internal medicine	1514	42.89	677	44.72
Pediatrics	121	3.43	33	27.27
Otorhinolaryngology	432	12.24	186	43.06
Family medicine	147	4.16	59	40.14
Surgery	156	4.42	85	54.49
Others	59	1.67	20	33.90
Region of clinics					<0.001
Metropolitan areas	1818	51.50	723	39.77
Urban areas	648	18.36	349	53.86
Rural areas	1064	30.14	474	44.55
NSAIDs use					<0.001
No	1510	42.78	551	36.49
Yes	2020	57.22	995	49.26
Antibiotics use					0.550
No	3188	90.31	1391	43.63
Yes	342	9.69	155	45.32
Total number of cold drugs prescribed other than GI medication			<0.001
1	323	9.15	164	50.77
2	917	25.98	460	50.16
3	1367	38.73	568	41.55
≥4	923	26.15	354	38.35
Coexisting diseases
Hypertension					0.447
No	2962	83.91	1289	43.52
Yes	568	16.09	257	45.25
Diabetes mellitus					0.530
No	3229	91.47	1409	43.64
Yes	301	8.53	137	45.52
Asthma					0.998
No	3304	93.60	1447	43.80
Yes	226	6.40	99	43.81
COPD					0.698
No	3517	99.63	1541	43.82
Yes	13	0.37	5	38.46
Heart failure					0.571
No	3487	98.78	1529	43.85
Yes	43	1.22	17	39.54
Renal failure					0.522
No	3504	99.26	1533	43.75
Yes	26	0.74	13	50.00

COPD=chronic obstructive pulmonary disease, GI=gastrointestinal, MedAid=medical aid, NHI=National Health Insurance, NSAIDs=non-steroidal anti-inflammatory drugs, PVI=Patriots and Veterans Insurance..

Potentially Inappropriate GI Medication Utilization

Our study found that 43.8% of study subjects were prescribed potentially inappropriate GI medications (Fig. 1). There were regional variations in the rate prescriptions, with the lowest rate observed in metropolitan areas (39.77%) and the highest in urban areas (53.86%). Variations were also observed based on age group and use of NSAIDs. The highest rate of prescriptions was observed in the 50 to 64-year-old age group and among NSAID users (48.0% and 49.26%, respectively). Furthermore, in terms of specialized areas of clinics, the highest proportion of GI drug prescriptions were observed in orthopedic surgery, general surgery, neurosurgery, etc (54.49%). Additionally, we found that the total number of drugs prescribed for cold symptoms was inversely proportional to the use of potentially inappropriate GI medications (Table 1).

Analysis of Potentially Inappropriate GI Medicine Use

Among 1,546 patients who were taking potentially inappropriate GI drugs, three-quarters (74.15%) were prescribed antiulcer drugs, and a quarter (25.85%) were prescribed GI tract regulators. As shown in Fig. 2, the antiulcer drugs most used were mucosal protectives (i.e., rebamipide, 56.21%), followed by H2-receptor blockers (20.71%), antacids (18.18%), and PPIs (4.90%). Among GI tract regulators, GI tract stimulants such as mosapride accounted for 85.71%, followed by anticholinergics (13.38%), and others (0.91%).

Figure 2. Classification of gastrointestinal (GI) drugs. GI=gastrointestinal, H2=histamine 2.

Factors Influencing GI Medication Prescription

Our logistic regression analysis identified several factors influencing the prescription of potentially inappropriate GI medications (Table 2). Women were more likely be prescribed these medications than men (OR=1.314, 95% CI=1.144–1.508). Our analysis found that specialized areas of clinics played a role in the prescription of potentially inappropriate GI medications. In comparison to clinicians from specialized areas of general practice, clinicians from specialized areas of pediatrics were the least likely to prescribe these medications (OR=0.479, 95% CI=0.307–0.732), while clinicians from specialized areas of surgery were the most likely to prescribe them (OR=1.655, 95% CI=1.171–2.345).

Table 2 . Adjusted odds ratios and 95% confidence intervals from multiple logistic regression analysis of gastrointestinal medication prescription.

Explanatory variable	Potentially inappropriate GI medication use
	Adj. OR	95% CI	p-value
Sex
Male (R)
Female	1.314	1.144-1.508	<0.001
Age
20-29 (R)
30-39	0.859	0.686-1.075	0.184
40-49	1.036	0.835-1.286	0.745
50-64	1.194	0.971-1.470	0.094
≥65	1.212	0.933-1.575	0.150
Specialized area of clinics
General practitioner (R)
Internal medicine	1.019	0.867-1.199	0.818
Pediatrics	0.479	0.307-0.732	<0.001
Otorhinolaryngology	1.060	0.839-1.337	0.626
Family medicine	0.769	0.534-1.099	0.152
Surgery	1.655	1.171-2.345	0.004
Others	0.623	0.346-1.088	0.103
Region of clinics
Metropolitan areas (R)
Urban areas	1.742	1.446-2.100	<0.001
Rural areas	1.191	1.016-1.396	0.031
NSAIDs use
No (R)
Yes	1.903	1.648-2.199	<0.001
Total number of cold drugs prescribed other than GI medication
1 (R)
2	0.864	0.665-1.123	0.275
3	0.568	0.440-0.731	<0.001
≥4	0.471	0.360-0.616	<0.001
c statistic	0.633
p-value of Hosmer–Lemeshow test	0.275

Adj. OR=adjusted odds ratio, CI=confidence interval, GI=gastrointestinal, NSAIDs=non-steroidal anti-inflammatory drugs, (R)=reference..

There were also geographic variations, clinicians from urban areas having 1.74-fold greater odds of prescribing potentially inappropriate GI medications than those in metropolitan areas. The use of NSAIDs was directly related to the prescription of potentially inappropriate GI medications (OR=1.903, 95% CI=1.648–2.199). Additionally, patients who were prescribed three or more cold medicines were less likely to be prescribed GI drugs (OR=0.568, 95% CI=0.440–0.731 and OR=0.471, 95% CI=0.360–0.616, respectively).

This study aimed to explore the current prevalence of potentially inappropriate GI medication use in outpatient prescriptions for the common cold in Korea. Our findings revealed 43.8% of prescriptions included potentially inappropriate GI medications. GI drugs have been routinely included in numerous prescriptions to decrease GI symptoms such as heartburn, nausea, and dyspepsia in Korea [16,17]. Byeon [15] conducted a chart review study of a large city hospital and discovered that 58.6% of patients with the common cold who did not have symptoms or a history of GI diseases were prescribed GI drugs. Cho and Kim [16] conducted a study to analyze the prescription behaviors of 148 office-based doctors using data from standardized common cold patients in Korea. In this study, approximately 80% of the doctors prescribed GI medicines (such as H2-blockers and motility drugs) following analgesics and NSAIDs (89.2%) to patients with the common cold. The prescription rate of potentially inappropriate GI medications in our study was lower than that in previous studies, probably because of the differences in study designs.

The use of potentially inappropriate GI drugs can lead to various negative outcomes, including adverse drug reactions, drug interactions, and increased drug expenditure [17]. Additionally, certain drugs such as dopamine antagonists (i.e., domperidone and metoclopramide) have been linked to an increased risk of extrapyramidal symptoms, which can manifest as acute dystonic reactions and hyperprolactinemia, potentially leading to gynecomastia and impotence [21]. Furthermore, PPIs have been associated with an increased risk of bone fractures [22-25], pneumonia [26-29], Clostridioides difficile (C. diff) infection [30,31], and vitamin B12 deficiency [17,32,33].

Given the potential negative outcomes associated with the use of potentially inappropriate GI medications, it is important to consider deprescribing these drugs in cases where they may not be necessary. Studies evaluating the effect of deprescribing show its potential positive impact on improving health outcomes [34-37]. Research has shown that deprescribing, which involves identifying and discontinuing unnecessary medications, can lead to improvements in health outcomes such as cognitive function, reduced risk of falls, and lower risk of hospitalization [38]. McGrath et al. [39] reported that PPIs are a common target of deprescribing because of the few indications for long-term use; significant drug-drug interactions with other commonly used medications; and increased risk of bone fractures, pneumonia, C. diff infection, and vitamin B12 deficiency.

Our study identified variations in potentially inappropriate GI medication prescribing based on specialized areas of primary care clinics. Clinicians from specialized areas of pediatrics were the least likely to prescribe potentially inappropriate GI medications. This may be due to the fact that these drugs are not typically prescribed to children.

We found that NSAID use was a strong predictor of potentially inappropriate GI medication utilization (OR=1.903, 95% CI=1.648-2.199). NSAIDs effectively relieve pain in headaches, myalgias, and arthralgias and the fever-related discomfort experienced during a cold [2,6]. NSAIDs inhibit cyclooxygenase-1 and -2, converting arachidonic acid to prostaglandins, and thereby exert antipyretic, analgesic, and anti-inflammatory effects [40]. Meanwhile, these prostaglandins also protect the gastric mucosa, and therefore, the inhibitory actions of NSAIDs have adverse effects, mainly on the GI tract [41]. However, the majority of patients taking therapeutic doses of NSAIDs for a short duration, particularly those without underlying GI disorder, usually tolerate them well [42]. It is important to note, though, that there has been a study suggesting an increased risk of upper GI bleeding even with short-term NSAID use [43]. This highlights the complexity of assessing the appropriateness of GI medication use in the context of NSAID therapy.

Our study found a statistically significant difference between the prescription of GI drugs and the number of cold drugs prescribed. We discovered that patients who were prescribed three or more cold medicines received fewer GI drugs compared to those prescribed only one. This suggests that clinicians may be more likely to prescribe GI drugs when fewer cold medicines are prescribed.

The study has several limitations to consider. Firstly, the diagnoses recorded in the claims data may not be entirely accurate due to providers seeking higher reimbursement rates. This has been demonstrated in a previous study, which found that an average of 70% of diagnoses corresponded with those in medical charts [18]. Secondly, the claims data do not include information on healthcare services that are not covered by insurance or over-the-counter drugs [44]. Thirdly, patients diagnosed with the common cold who were included in the study may also have had other conditions, which may affect the relationship between the use of GI medication and a specific disease [45]. Lastly, the exclusion criteria did not explicitly account for indications that may legitimately require antacid therapy, such as esophageal varices, congenital stenosis or stricture of the esophagus, heartburn, gastrointestinal hemorrhage, Zollinger-Ellison syndrome, or Helicobacter pylori infection. This may affect the interpretation of our findings regarding antacid therapy.

Despite these limitations, the study’s results have significant implications. The study targeted a much broader population group compared to previous studies that only analyzed prescription patterns in certain medical institutions [15,16]. Therefore, the results could be generalized. In addition, it is meaningful that our study also investigated the influencing factors in comparison with the existing studies that only calculated the prescription rate [15-17].

It is crucial to address the perception that adding GI medication to symptom-relieving drugs for the common cold is beneficial or harmless, as this practice can lead to inappropriate prescriptions. Shin et al. [45] have demonstrated that promoting public health campaigns and implementing restrictive drug policies can help reduce such prescriptions.

Patel et al. [46] conducted an observational, cross-sectional, questionnaire-based study to assess the prescribing pattern of doctors for patients presenting with the common cold. In that study, the authors reported that inappropriate prescriptions by doctors is due to a lack of adequate training, lack of self-confidence, or both [46]. Therefore, it is essential to address this issue by providing proper training to prescribers during their formative years and reinforcing their education through continuing medical education programs.

To improve the quality of prescriptions, there needs to be an increased awareness of the need for deprescribing GI medications for the common cold in primary care settings and continued research on potentially inappropriate GI medication utilization. Public health campaigns and regulatory policies from healthcare stakeholders can also play a role in reducing these prescriptions.

This study highlights a significant issue of potentially inappropriate utilization of GI medications for the treatment of the common cold in Korea. The research identifies several factors that contribute to the prescription of these drugs, including sex, region and specialized areas of clinics, concurrent use of NSAIDs, and the number of drugs prescribed. The finding of this study can be used to develop targeted interventions to address this problem and improve the quality of care for patients with the common cold.

The study was approved by the Institutional Review Board (IRB) of Pusan National University (PNU IRB/2021_29_HR) and since it used secondary data obtained from the HIRA and contained no personal information, written consent was waived by the IRB.

None.

We used the National Patient Sample data collected by the Korean Health Insurance Review and Assessment Service (HIRA-NPS-2018-0078); however, the results do not concern the Ministry of Health and Welfare or HIRA.

No potential conflict of interest relevant to this article was reported.

M. Kim contributed to the design, analysis, and interpretation; drafted the manuscript, critically revised the manuscript; and gave final approval, N. Je contributed to the conception and design, acquisition, and interpretation; drafted the manuscript; critically revised the manuscript and gave final approval. Both authors agree to be accountable for all aspects of work ensuring integrity and accuracy. No assistance in the preparation of this article is to be declared.