Nationwide study of chronic codeine use and its impact on cough related diseases in South Korea

Abstract

Codeine is widely used to control coughs, although concerns about its overuse arise due to its side-effects. This study aimed to evaluate the status of codeine usage according to various medical conditions. The Korean National Health Insurance Service sample cohort was analyzed. Subjects with more than continuous sixty days of antitussive and codeine were defined as chronic users. It was evaluated according to age, smoking status, chronic obstructive pulmonary disease (COPD), asthma, allergic rhinitis (AR), bronchiectasis, chronic cough (CC), gastroesophageal reflux disease (GERD), and lung cancer. A total of 89,289 chronic antitussive users were identified, of whom 589 were chronic codeine users. The chronic codeine users were older, more likely to be smokers, and more likely to have multimorbidity (P < 0.001, all). After adjusting age, chronic codeine use showed a positive correlation with lung cancer (adjusted odds ratio [aOR]: 6.99), COPD (aOR: 2.04), GERD (aOR: 1.93), and CC (aOR: 1.60). Multimorbidity also revealed positive correlations, increasing as the number of comorbidities rose (P < 0.001). Our findings highlight that chronic codeine usage is associated with underlying cough-inducing diseases, emphasizing the need for monitoring and guidelines to ensure safer use, especially among older adults and those with chronic respiratory conditions.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-80506-y.

Introduction

Cough is one of the most common reasons for hospital visits throughout a person’s lifetime, affecting approximately 9–33% of the population at some point¹. Chronic cough (CC), defined as a cough lasting more than eight weeks, accounts for 5–10% of these cases, with many not showing clinical improvement, leading to a diagnosis of unexplained chronic cough (UCC)². Despite various medications being used to manage cough, there remain significant limitations in controlling symptoms effectively³.

In this context, opioid analgesics such as codeine are employed to control cough⁴. Codeine, a prodrug of morphine used to treat both pain and cough, is naturally found in the sap of the opium poppy (Papaver somniferum)⁵. However, the use of codeine for chronic cough management is highly debated due to its side effects^6,7. Short-term use can result in respiratory depression, drowsiness, and constipation, while long-term use may lead to more serious issues such as dependence and withdrawal symptoms^4,7.

Conditions like asthma, chronic obstructive pulmonary disease (COPD), and bronchiectasis frequently cause coughing^8–11. While antitussive medications are widely prescribed, codeine is sometimes prescribed for long-term control^12–14. Nevertheless, its side effects necessitate careful consideration, especially given the lack of comprehensive real-world usage data.

To date, no studies have investigated the national patterns of opioid antitussive use across various conditions. This research examines the prevalence and correlation between codeine usage and cough-related illnesses using a population-based cohort in South Korea. This study aims to highlight the actual usage patterns of codeine and emphasize the need for guidelines and monitoring systems to ensure safer use, especially among older adults and those with chronic respiratory conditions.

Results

Characteristics of chronic antitussive user in South Korea

Within the overall cohort, 8.46% (n = 89,441) used antitussives chronically. Chronic antitussive users were most prevalent among those under 40 years of age. Among older adults (≥ 70 years old), 17.4% were chronic antitussive users, with the trend increasing with age. COPD, asthma, AR, bronchiectasis, CC, GERD, and lung cancer were higher in chronic antitussive users than in non-chronic users (all, P < 0.001) (Supplementary Table 1). In the case of COPD and bronchiectasis, over 40% were prescribed antitussives chronically. Higher proportions of chronic antitussive usage were observed in lung cancer (31.3%), asthma (28.3%), and CC (21.1%). The proportion of chronic users also increased with the number of overlapping comorbidities. (Fig. 1).

Fig. 1.

Trends of chronic antitussive usage (≥ 60 consecutive days/year) according to variables.

Characteristics of chronic codeine user compared to nonuser in chronic antitussive usage

Codeine usage has consistently increased over the decades, with the ratio rising in correlation to the total number of prescription days per year. (Supplementary Fig. 1). Most prescriptions for codeine were for fewer than 50 days in a year (Fig. 2). The chronic codeine user group had the highest age distribution in the older age group (≥ 70 years old, 48.0%), with an increasing trend with age. Common comorbidities among chronic codeine users included GERD (80.3%), CC (37.0%), COPD (33.6%), lung cancer (25.6%), and bronchiectasis (8.0%). Both asthma and AR were more prevalent in the non-codeine group than in the codeine user group. The average number of diseases was higher in codeine users than in non-users (3.2 ± 1.4 vs. 2.4 ± 1.0, P < 0.001) (Table 1).

Fig. 2.

Proportion and frequency of the total number of consecutive codeine prescription days within a year.

Table 1.

Baseline characteristics of participants according to chronic codeine usage (≥ 60 consecutive days/year).

	Chronic user (n = 598)	Non-chronic user (n = 88,691)	Total (n = 89,289)	P-value
Age, years				< 0.001
< 40	24 (4.0)	47,404 (53.5)	47,428 (53.1)
40–49	35 (5.9)	6,145 (6.9)	6,180 (6.9)
50–59	96 (16.0)	9,067 (10.2)	9,163 (10.3)
60–69	156 (26.1)	10,306 (11.6)	10,462 (11.7)
≥ 70	287 (48.0)	15,769 (17.8)	16,056 (18.0)
Age (mean ± SD)	66.63 (± 12.83)	34.77 (± 29.9)	34.98 (± 29.93)	< 0.001
Smoking status				0.001
None	314 (62.6)	28,132 (69.7)	28,446 (69.6)
Former	85 (16.9)	6,059 (15.0)	6,144 (15.0)
Current	103 (20.5)	6,154 (15.3)	6,257 (15.3)
Comorbidities
COPD				< 0.001
No	397 (66.4)	80,654 (90.9)	81,051 (90.8)
Yes	201 (33.6)	8,037 (9.1)	8,238 (9.2)
Asthma				< 0.001
No	267 (44.6)	33,720 (38.0)	33,987 (38.1)
Yes	331 (55.4)	54,971 (62.0)	55,302 (61.9)
Allergic rhinitis				< 0.001
No	100 (16.7)	5,309 (6.0)	5,409 (6.1)
Yes	498 (83.3)	83,382 (94.0)	83,880 (93.9)
Bronchiectasis				< 0.001
No	550 (92.0)	86,158 (97.1)	86,708 (97.1)
Yes	48 (8.0)	2,533 (2.9)	2,581 (2.9)
Chronic cough				< 0.001
No	377 (63.0)	69,793 (78.7)	70,170 (78.6)
Yes	221 (37.0)	18,898 (21.3)	19,119 (21.4)
GERD				< 0.001
No	118 (19.7)	49,599 (55.9)	49,717 (55.7)
Yes	480 (80.3)	39,092 (44.1)	39,572 (44.3)
Lung cancer				< 0.001
No	445 (74.4)	86,808 (97.9)	87,253 (97.7)
Yes	153 (25.6)	1,883 (2.1)	2,036 (2.3)
Number of diseases (mean ± SD)	3.23 (± 1.35)	2.35 (± 1.04)	2.36 (± 1.05)	< 0.001
Number of diseases (group)				< 0.001
0–1	50 (8.4)	16,473 (18.6)	16,523 (18.5)
2	140 (23.4)	37,850 (42.7)	37,990 (42.6)
3	166 (27.7)	23,035 (25.9)	23,201 (26.0)
≥ 4	242 (40.5)	11,333 (12.8)	11,575 (12.9)

SD, standard deviation; COPD, chronic obstructive pulmonary disease; GERD, gastroesophageal reflux disease.

Comparison of codeine use by comorbidities

Logistic regression analysis revealed that codeine use is positively associated with age, smoking, and certain comorbidities (Table 2). Odds ratios (ORs) for codeine use increased with age: 40–49 years (OR [95% CI]: 11.20 [6.68–18.91]), 50–59 years (20.90 [13.36–32.71]), 60–69 years (29.88 [19.43–45.94]), and ≥ 70 years (35.93 [23.68–54.50]). Current smokers were more likely to use codeine than non-smokers (OR 1.50 [95% CI 1.20–1.88]). Positive associations with chronic codeine usage were observed for lung cancer (15.85 [13.12–19.15]), GERD (5.16 [4.22–6.32]), COPD (5.08 [4.28–6.03]), bronchiectasis (2.97 [2.21–4.00]), and CC (2.17 [1.83–2.56]), while asthma (0.76 [0.65–0.89]) and AR (0.32 [0.26–0.39]) were negatively associated (all, P < 0.001) (Supplementary Table 2). Even after adjusting for age, positive associations remained for lung cancer (OR 6.99; 95% CI 5.76–8.48), COPD (OR 2.04; 95% CI 1.71–2.44), GERD (OR 1.93; 95% CI 1.57–2.38), and CC (OR 1.60; 95% CI 1.36–1.90), with a negative association for AR (OR 0.65 [95% CI 0.52–0.81]). The ORs increased with the number of comorbid diseases, showing ORs of 1.44 (95% CI 1.04–1.99) for two, 1.67 (95% CI 1.22–2.30) for three, and 3.09 (95% CI 2.27–4.20) for four or more comorbidities. Similar results were demonstrated after adjusting for age, sex, and smoking history (Supplementary Table 3).

Table 2.

Comparison of codeine used by comorbidities in univariate and multivariate logistic regression analyses.

	Unadjusted	Adjusted*
	OR (95% CI)	OR (95% CI)
COPD
No	Ref.	Ref.
Yes	5.08 (4.28–6.03)	2.04 (1.71–2.44)
Asthma
No	Ref.	Ref.
Yes	0.76 (0.65–0.89)	0.98 (0.83–1.15)
Allergic rhinitis
No	Ref.	Ref.
Yes	0.32 (0.26–0.39)	0.65 (0.52–0.81)
Bronchiectasis
No	Ref.	Ref.
Yes	2.97 (2.21–4.00)	1.32 (0.98–1.78)
Chronic cough
No	Ref.	Ref.
Yes	2.17 (1.83–2.56)	1.60 (1.36–1.90)
GERD
No	Ref.	Ref.
Yes	5.16 (4.22–6.32)	1.93 (1.57–2.38)
Lung cancer
No	Ref.	Ref.
Yes	15.85 (13.12–19.15)	6.99 (5.76–8.48)
Number of diseases
0–1	Ref.	Ref.
2	1.22 (0.88–1.68)	1.44 (1.04–1.99)
3	2.37 (1.73–3.26)	1.67 (1.22–2.30)
≥ 4	7.04 (5.18–9.55)	3.09 (2.27–4.20)

OR, odds ratio; CI, confidence intervals; COPD, chronic obstructive pulmonary disease; GERD, gastroesophageal reflux disease.

*Adjusted by age.

Discussion

The present study analyzed the usage patterns of antitussives, including codeine, among patients with CC. We investigated the association between codeine use and the use of other cough remedies, as well as its relationship with underlying diseases. Our findings indicated that codeine use was more prevalent among patients with various comorbid conditions associated with coughing. Although most prescriptions for codeine were for durations of less than 50 days, long-term use exceeding this period was more common than anticipated, and this trend increased over time and with the chronicity of the cough. Notably, the significance of codeine use persisted for conditions such as lung cancer, chronic obstructive pulmonary disease (COPD), gastroesophageal reflux disease (GERD), and chronic cough, even after adjusting for confounding variables. Among these patients, 11.8–46.3% were prescribed cough remedies chronically, with 1.16–7.51% using codeine long-term. This raises significant concerns regarding the management of UCC. Unlike previous studies that primarily focused on the occasional use of codeine- or dihydrocodeine-containing medications, our research is the first to demonstrate the chronic prescription of codeine-containing remedies for a variety of causes. This underscores the necessity for vigilant monitoring of codeine usage not only in cases of chronic cough but also in chronic respiratory diseases^12–15. The results highlight the critical need for developing comprehensive guidelines to ensure the safe use of codeine, particularly among patients with chronic respiratory conditions.

Recently, the concept of cough hypersensitivity has emerged as a theory for understanding the mechanism of cough^16,17. This theory suggests that CC is triggered and intensified by mechanisms similar to those in pain, involving sensitization, reflexes, and enhancement in both the peripheral and central nervous systems^18,19. Due to these underlying mechanisms, various clinical guidelines recommend pharmacological options such as neuropathic drugs (e.g., pregabalin and gabapentin), amitriptyline, and opioid-based cough suppressants (e.g., codeine or hydrocodone) for the treatment of UCC. modulating neurotransmitter activity in the cough reflex arc These medications work by modulating neurotransmitter activity in the cough reflex arc, providing relief for patients^3,20–22. Codeine, one of the most popular medications for controlling UCC, remains indispensable yet controversial due to its side effects. The recent understanding of cough hypersensitivity underscores the need for careful selection and monitoring of pharmacological treatments to ensure both efficacy and safety.

Codeine is a selective agonist of the µ-opioid receptor and serves as a prodrug, which is metabolized into active metabolites such as morphine and codeine-6-glucuronide²³. Upon binding to these receptors, codeine causes neuron hyperpolarization and inhibits the release of nociceptive neurotransmitters²⁴. thereby reducing neuronal excitability and increasing pain tolerance, which are key mechanisms in cough control⁶. In the liver, codeine is metabolized into morphine, norcodeine, and hydrocodone, with most metabolites being excreted by the kidneys, primarily as glucuronic acid conjugates²⁴. The side effects of codeine are often due to its metabolism into morphine. Common side effects include drowsiness and constipation, while rarer side effects include nausea and dry mouth. Critical side effects such as respiratory depression have also been documented^4,15. The cytochrome P450 enzyme CYP2D6 plays a crucial role in converting codeine to morphine. Some patients, known as ultrarapid metabolizers, can convert codeine to morphine more quickly, leading to potentially dangerous blood levels²⁵. For instance, 4 mg of morphine can be produced from a standard 30 mg dose of codeine in ultrarapid metabolizers. Prolonged use of codeine can lead to serious complications, including respiratory depression and potentially fatal overdose consequences²⁶. Chronic use may also result in dependence and withdrawal symptoms such as drug craving and irritability⁴. Due to these risks, the chronic use of codeine should be carefully monitored, supervised, and controlled by healthcare professionals to prevent misuse, overuse, and abuse⁴. These considerations are particularly important for patients with vulnerable conditions such as lung cancer or COPD, who may carry the CYP2D6 genotype or are prescribed codeine for extended periods^27,28.

This study offers several notable insights. Firstly, it is pioneering in utilizing nationwide data to comprehensively describe codeine usage characteristics, revealing an upward trend in its use, particularly among individuals with cough-related diseases. This trend underscores the importance of monitoring real-world codeine utilization. Secondly, the study sheds light on the association between codeine use and various underlying conditions. Notably, high codeine usage in lung cancer and COPD suggests a greater likelihood of UCC in these conditions. Although CC and GERD patients showed lower chronic antitussive use rates, age-adjusted analyses indicated a significant link with chronic codeine use. While not statistically significant after adjusting for age, the high usage of antitussives and codeine in bronchiectasis was evident. Asthma and AR showed lower codeine use, logically attributed to primary treatments like inhalers and antihistamines/intranasal corticosteroids, highlighting distinct cough characteristics in these conditions compared to others with more common codeine use. Thirdly, the study reveals a strong correlation between codeine use and aging, with 48% of chronic codeine users being over 70 years old. The conditions linked to codeine use are predominantly associated with or related to aging, indicating that older adults, especially those with multimorbidity, are the most frequent and affected group. This emphasizes the need for discussions on the susceptibility of older adults to medication misuse or abuse and the development of programs and health policies promoting appropriate codeine use. This includes non-pharmacological treatments and education on managing refractory coughs.

This study has several limitations. First, it relied on a sample dataset of one million individuals, not the entire population. Nonetheless, NHIS sample data is known to accurately reflect the nationwide demographics of Korea, as previously described²⁹. Additionally, to define chronic medication use, we adopted a two-month criterion, minimizing selection bias in the analysis. Second, the study relied on ICD-10 codes to identify conditions such as asthma and COPD, but we did not have access to clinical spirometric data, which are critical for staging these diseases. As a result, it was not possible to classify patients by disease severity or control, such as severe or uncontrolled asthma, which limits the depth of comparison between asthma and COPD patients in relation to codeine use. This represents a key limitation, as the dataset did not include detailed clinical measurements. Third, while we observed notable trends regarding codeine use among younger age groups, we were unable to perform more detailed subgroup analyses due to the constraints of the available data. This limits our ability to provide more granular insights into the demographics of younger patients. Finally, since all conditions were identified using ICD-10 codes, there may be some underrepresentation or conservative classification of certain conditions. Although ICD-10 codes are widely used for defining chronic diseases, future research incorporating more detailed clinical data could offer deeper insights into codeine use across different chronic disease stages.

In conclusion, our study demonstrated the nationwide pattern of chronic codeine use and its association with various underlying diseases. These findings highlight the critical importance of systematic monitoring and exercising caution regarding codeine usage, particularly in vulnerable populations such as older adults, individuals with specific chronic conditions, and those with multiple comorbidities. By recognizing these associations, healthcare providers can better manage and mitigate the risks associated with long-term codeine use, ensuring safer and more effective treatment for patients with chronic cough and related conditions.

Methods

Database

This study utilized a sample cohort from the Korean National Health Insurance Service (NHIS) database, covering the period from 2002 to 2015. The NHIS database (NHIS-2021-2-185)) is a comprehensive source of health information, encompassing medical data for approximately 97% of the Korean population^30,31. The dataset includes demographic information, diagnosis codes, prescription records, and health examination results, making it a valuable resource for population-based health research. The cohort consists of a representative sample of one million individuals, stratified by gender, age, insurance type, income level, and geographic region. This stratification ensures that the sample accurately reflects the broader population demographics. It includes an eligibility DB (basic information about the subjects), a medical information DB (diagnosis and prescription codes), a birth and death DB, a health examination DB (physical examinations and medical histories), and a long-term care insurance DB³². This study was approved and waived the informed consent by the Institutional Review Board of the National Medical Center (NMC-2101-001). The informed consent was waived due to using secondary data for unidentifiable from NHIS. All methods were performed in accordance with the relevant guidelines and regulations.

Definition of chronic antitussive and codeine users

Long-term antitussive or codeine users were identified based on the use of medication for more than 60 days based on the definition of chronic cough (cough lasting 8 weeks or longer). Consequently, participants were classified into two groups as follows: chronic antitussive users and non-chronic antitussive users and chronic codeine users and non-chronic codeine users.

Study population

The study population was derived from the Korean National Health Insurance Service (NHIS) sample cohort as of 2014, consisting of 1,057,454 individuals. From this cohort, we identified patients who had been prescribed antitussive medications. To qualify as chronic antitussive users, patients must have had 60 or more consecutive prescription days within a single calendar year from 2014 to 2015. This criterion was established to align with the definition of chronic cough, characterized by a duration of at least eight weeks. (n = 89,441). However, we excluded 152 patients due to incomplete information such as missing age or disease history. As a result, the final study population comprised 89,289 patients. Among these, 598 were classified as chronic codeine users, having received continuous prescriptions for codeine for more than 60 days within the specified period. (Fig. 3).

Fig. 3.

Schematic flow of this study.

Other variables

Age was classified into < 40 years, 40–49 years, 50–59 years, 60–69 years, or ≥ 70 years. Smoking status was divided as none, former, and current smoker according to the health examination questionnaire responses, utilizing past responses from the same or the closest year to the year of antitussive use. Comorbidities, defined as COPD, asthma, allergic rhinitis (AR), bronchiectasis, chronic cough (CC), gastroesophageal reflux disease (GERD), and lung cancer, which are prevalent among patients with chronic cough, were diagnosed between 2010 and 2015. The diagnosis was coded according to the International Classification of Diseases (ICD)−10, with details provided in Supplementary Table 4. The number of diseases was tallied from the listed comorbidities and categorized as 0–1, 2, 3, or ≥ 4, with a minimum of 0 and a maximum of 7^33,34.

Statistical analysis

In this study, the t-test and χ2 test were performed for continuous variables and categorical variables, respectively. Frequency analysis was performed to see the baseline characteristics of the participants by the χ2 test. Univariate logistic regression analysis was utilized to determine the relationships between codeine use and other variables, such as age, smoking status, comorbidities, and the number of comorbid diseases. Additionally, by adjusting for the effect of age, multiple logistic regression analysis was conducted to further explore the relationship between codeine use and these variables. P < 0.05 was considered statistically significant. All statistical analyses were performed with SAS Enterprise Guide version 7.1 (SAS Institute, Cary, NC, USA) and R-Studio version 3.5.1 (RStudio: Integrated Development for R. RStudio, PBC, Boston, MA, USA; http://www.rstudio.com/).

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: J Joh, JP Myong. Data curation: Y Lee. Formal analysis: Y Lee. Investigation: J Joh, JP Myong. Validation: J Joh, JP Myong. Methodology: JP Myong. Resources: J Joh, JP Myong. Supervision: JP Myong. Visualization: TJ An, Y Lee. Writing – original draft: TJ An. Writing – review & editing: All authors.

Data availability

The present DB is available under permission of NHIS Korea. Requests for access to the data should be directed to Dr. Jun-Pyo Myong.

Declarations

Competing interests

The authors declare no competing interests.