Real-world evidence of constipation and laxative use in the Korean population with chronic kidney disease from a common data model

Constipation is a highly prevalent gastrointestinal disorder in patients with chronic kidney disease (CKD). However, our understanding of its epidemiology and management in CKD is limited. We aimed to explore real-world data on constipation and laxative use in patients with CKD in a nationwide population-based cohort from the Korean Health Insurance Review and Assessment-National Patient Sample database. This study analyzed retrospective health claims data in Korea from 2012 to 2017 that were transformed into the Observational Medical Outcomes Partnership Common Data Model. The pooled proportion of constipation diagnoses was 30.5% in all patients with CKD and 15.9%, 16.5%, 17.4%, 29.9%, and 43.3% in patients with CKD stages 1–5, respectively, suggesting a higher prevalence in advanced CKD. Patients receiving peritoneal dialysis or hemodialysis had the highest prevalence of constipation, while transplant recipients showed a prevalence comparable to that of patients with early CKD. Patients with CKD had a significantly higher risk of constipation than age- and sex-matched non-CKD individuals (range of odds ratio [OR]:1.66–1.90). Laxative prescribing patterns differed by CKD severity. Osmotic agents were prescribed in more than half of patients with advanced CKD, while magnesium salts and bulking agents were prescribed less frequently. The CKD patients with constipation were more likely to be prescribed constipation-inducing medications, including antipsychotic and neurological medications. Our findings provide real-world constipation and laxative prescription status in the Korean CKD population, revealing a significantly higher risk of constipation and different laxative prescribing patterns in patients with CKD.

Constipation is one of the leading gastrointestinal symptoms commonly encountered in the general population, particularly among older adults 1 .Constipation significantly impairs the health-related quality of life (QoL) of affected individuals and poses considerable economic and healthcare burdens 2 .Recent evidence suggests a close association between constipation and clinical outcomes such as cardiovascular disease, chronic kidney disease (CKD) progression, and mortality [3][4][5][6] .In particular, numerous observational studies have reported a higher prevalence of constipation in patients with CKD.Although the global prevalence of constipation in the general population has been estimated at approximately 14%, the prevalence in patients with CKD is reported to be much higher 7 .A recent metaanalysis reported a constipation prevalence of 38.8% in patients with advanced non-dialysis CKD 8 .Patients with end-stage kidney disease (ESKD) have been shown to have a substantially higher constipation prevalence, with some studies reporting over 50% 9 .CKD is associated with constipation both clinically and pathophysiologically. Dietary and fluid restrictions, high pill burdens, comorbidities, gut dysbiosis, and uremic milieu may all contribute to chronic constipation 10 .

Data source
We used the HIRA-NPS database, which is a nationwide standard cohort based on claims data from 2012 to 2017.In Korea, national health insurance covers almost 98% of the national population 16 ; HIRA-NPS data are representative datasets generated during the reimbursement for healthcare services, sampling approximately 3% of the total population.HIRA-NPS was sampled using a proportionate stratified sampling method by dividing total population into 32 strata (2 gender strata and 16 age strata) 17 .Estimates from HIRA-NPS demonstrates a high level of representativeness for the total population 17 .The Korean Health Insurance Review and Assessment (HIRA) and HIRA-NPS data have enabled diverse real-world data analyses in previous publications [18][19][20][21] .HIRA-NPS data from 2012 to 2017 were transformed into the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM, version 5.0) developed by the Observational Health Data Sciences and Informatics (OHDSI) network with standard vocabulary and ontology 22 .

Study design and cohort definition
Definitions of the concept sets used are presented in Supplementary Table 1.Descendant concepts of each concept set were also used in the analysis.CKD concepts were defined according to CKD stages.The CKD stage was determined by the highest CKD stage code among patient condition occurrences.We only included patients with 2 or more inpatient or outpatient ICD codes for each condition, similar to other studies using HIRA database 23,24 .Patients who underwent kidney transplantation (KT) were investigated separately from the total CKD population.Patients undergoing peritoneal dialysis (PD) and hemodialysis (HD) were identified using procedure-occurrence concepts.Constipation medications included 8 types of laxatives (psyllium, polycarbophil, polyethylene glycol [PEG], magnesium salts, lactulose, lactitol, bisacodyl, and docusate).Some constipation drugs, such as PEG, are often used for bowel preparation for colonoscopy; thus, cases with a drug era length of less than 3 days were excluded.Over-the-counter drugs such as senna and non-benefit drugs were not included in the analysis.This study was conducted in accordance with the World Medical Association Declaration of Helsinki and was approved by the Institutional Review Board of Inha University Hospital (IRB No. 2023-08-034).The requirement for informed consent from the study subjects was waived by the IRB of Inha University Hospital due to the retrospective study design.All CDM-converted data used in the analysis were fully anonymized.

Statistical analysis
Data analysis was performed using an interactive analysis tool ATLAS (version 2.12.0) provided by the FEEDER-NET platform (https:// feede rnet.com) 25 .We used the characterization, cohort pathways, and estimation functions of ATLAS and R (version 4.2.1;R Foundation for Statistical Computing, Vienna, Austria).Fisher's exact test was used to compare differences in drug prescriptions between CKD patients with and without constipation.The risk of constipation and laxative use was evaluated in patients with CKD compared to age-and sex-matched individuals without CKD using logistic regression.The cohort entry event was the initial occurrence of diagnostic codes, and the time-at-risk was defined as the time between 1 day after cohort start date and cohort end date.Demographic covariates were included in the model fitting, and regularization was applied.KT recipients and patients receiving HD or PD were also compared in the same manner.Patients with stage-unspecified CKD were only included in the analyses where CKD stage and severity were not used.P-values of < 0.05 were considered statistically significant, and the Bonferroni correction was applied for multiple testing.

Characteristics of the study population
A total of 39,293 individuals diagnosed with CKD between 2012 and 2017 were included in this study.Table 1 presents the characteristics of the participants.The overall number of patients with CKD increased noticeably from 2012 to 2017.This trend reflects the age distribution of the Korean population and the recent rapid population aging 26 .Recent data on CKD and ESKD populations in Korea also provide consistent findings 27,28 .Patients with CKD stage 1 and 2 were less likely to be captured, possibly due to low rates of screening or followup.Patients with CKD stage 5 were more likely to be recruited based on the Korean Classification of Disease

Treatment pathways for laxative prescriptions
The treatment pathway of laxative prescriptions was examined in the different subgroups (Fig. 2 and Supplementary Fig. 1-5).In individuals without CKD, the commonly prescribed laxatives are lactulose, magnesium salts, and polycarbophil.However, as CKD progressed from the mild to advanced stages, the use of magnesium salts and polycarbophil decreased markedly, while the use of lactulose gradually increased.In a sunburst plot, the outer rings represent the second-line and subsequent therapies in individual patients.Sunburst plots of patients with CKD had a larger second ring area, indicating a greater need for second-line laxatives.In patients with ESKD, lactulose accounts for over 60% of all laxatives, followed by polycarbophil.Magnesium salts are rarely www.nature.com/scientificreports/used in patients with ESKD.Sunburst plots of patients on HD showed longer spikes in the third and fourth outer rings, suggesting more frequent changes in the laxatives.Most of the second-or higher-line therapies are osmotic agents, such as lactulose and lactitol.KT recipients were prescribed mainly lactulose and bisacodyl, but the proportion of lactulose was similar to patients with mild CKD.

Differences in drug prescriptions
We evaluated differences in medication prescriptions between CKD patients with and without constipation.Excluding laxatives, the top 10 drug groups sorted by ORs are shown in Table 3 (individual drugs are presented in Supplementary Table 5).The CKD patients with constipation were more frequently prescribed medications known to cause constipation, such as antipsychotics (risperidone, quetiapine, and haloperidol), antidepressants (nortriptyline, amitriptyline, and escitalopram), antiepileptics (valproate and clonazepam), hypnotics and sedatives (zolpidem), anxiolytics (diazepam and lorazepam), opioids (fentanyl, oxycodone, and tramadol), nonsteroidal anti-inflammatory drugs (celecoxib and diclofenac), and antimuscarinics (solifenacin and propiverine) 29 .Of note, the findings suggest that drug categories such as antipsychotics may pose the greatest risk for constipation in patients with CKD, and that special caution is needed when prescribing these drugs.In addition, drugs commonly used in patients with CKD have been identified, such as diuretics (furosemide and spironolactone), potassium binders (calcium polystyrene sulfonate), and iron supplements (ferrous sulfate).While most medications with high ORs may be directly related to constipation from their adverse effects, others are possibly linked to specific medical conditions, such as hypokalemia (potassium chloride) and poor oral intake (megestrol).

Discussion
In this study, we present real-world data on constipation and laxative use in patients with CKD in a nationwide cohort based on health insurance claims data.Consistent with previous observations, our data show a high prevalence of constipation in patients with CKD.In particular, constipation and laxative prescriptions were notably higher in the later stages of CKD, with ESKD being the highest.The risks of constipation and laxative use in patients with CKD were also significantly higher than those in age-and sex-matched populations without CKD.Second-line laxative use increases with the progression of CKD.The most frequently prescribed laxative is the osmotic agent lactulose, which accounts for nearly half of all laxative prescriptions for patients with advanced  www.nature.com/scientificreports/CKD.We also found differences in the prescription patterns of constipation-inducing drugs, including antipsychotic and neurological medications, between CKD patients with and without constipation.While previous studies on constipation have mostly focused on patients at specific stages of CKD, such as ESKD, this study explored constipation and laxative use across all stages of CKD in a population-based cohort, which is a major strength of our study.The prevalence of constipation largely depends on the diagnostic tools used.Early studies used a self-reported definition of constipation; however, recent studies have favored the more standardized Rome criteria based on defecation symptoms and stool forms.In general, the prevalence based on the Rome criteria was lower than the patient-reported prevalence.To date, only a small number of studies have examined the prevalence of constipation in patients with non-dialysis CKD.A recent meta-analysis reported the pooled prevalence of self-reported and functional constipation in different CKD stages 8 .The prevalence of self-reported and functional constipation in patients with CKD stage 4-5 was 38.78% and 21.97%, respectively.Therefore, the prevalence of 29.9% (CKD stage 4), and 43.3% (CKD stage 5) in this study was close to the selfreported prevalence.The self-reported prevalence of constipation in CKD stage 3 was 29.75% in metaanalysis, which was mainly driven by two recent studies 30,31 .However, these studies only included patients with CKD stage 3B (GFR 30-44 ml/min/1.73m 2 ), while we included all stage 3 patients, resulting in a relatively low prevalence of 17.4%.For patients with CKD stages 1 and 2, small studies with sample sizes of 22 or less have reported a prevalence of 12.5-31.8% 32,33.In our study, the estimated prevalence of constipation in CKD stages 1 and 2 was 15.9% and 16.5%, respectively, which was much lower than in advanced CKD.
On the other hand, numerous reports on constipation in patients with CKD have mainly focused on ESKD populations, which show a significantly higher prevalence.The prevalence of constipation has been reported to be higher in HD patients than in PD patients, which may be due to fluid restriction during the interdialytic period and low physical activity 34 .A study by Yasuda et al. compared the prevalence of constipation between patients on PD (n = 204) and HD (n = 268) and found a prevalence defined using a questionnaire of 28.9% in PD patients and 63.1% in HD patients 35 .Studies using the Rome criteria and with a sample size of 100 or more have reported a prevalence of 30.5-71.7% in HD patients and 14.2% in PD patients [36][37][38] .In this study, the pooled prevalence of constipation in HD and PD patients was estimated to be 45.0% and 40.5%, respectively.We found that constipation was more prevalent in patients on HD than those on PD, but the difference was not as pronounced as in previous studies.Notably, patients receiving PD had a higher proportion of laxative use than expected from their constipation diagnosis.This may indicate chronic use of laxatives in patients on PD regardless of the severity of their symptoms.Indeed, constipation can cause catheter dislocation, leading to drainage failure, which can lead clinicians to prescribe more laxatives.Of note, KT recipients showed a substantially lower prevalence of constipation and laxative use compared to patients with advanced CKD or ESKD.The prevalence of self-reported constipation in KT recipients is reported to be 37-43% in recent questionnaire-based studies 39,40 .However, in contrast to patients with ESKD, diarrhea rather than constipation is the most common www.nature.com/scientificreports/and problematic gastrointestinal symptom in KT recipients 40 .These differences are likely related to immunosuppressive agents such as mycophenolate mofetil, low pill burden, normalization of uremic milieu, and changes in the gut microbiota after KT 41 .We found that lactulose was the most frequently prescribed laxative in Korean patients with CKD.Lactulose is a relatively safe laxative, even in patients with advanced CKD.Also, there is some evidence that lactulose has a beneficial effect on gut microbiota in CKD, suppressing uremic toxins 42,43 .The use of magnesium salts and polycarbophil were greatly reduced in patients with advanced CKD.Magnesium salts act as osmotic agents but are avoided due to the risk of hypermagnesemia in patients with CKD 44 .Half of the patients on PD taking a polycarbophil as first-line therapy switched to the osmotic agent lactulose.Polycarbophil is a bulk-forming agent that can induce abdominal fullness and have a relatively weaker efficacy than osmotic agents.Moreover, polycarbophil is typically administered in the form of a calcium salt, which produces free calcium ions, posing a burden on calcium pools in patients with CKD 45 .It remains uncertain which laxatives are more beneficial for patient outcomes.To date, no direct comparison of mortality between laxatives in patients with CKD has been reported.This topic deserves to be explored further in future well-designed prospective studies.In addition, there is no consensus regarding the management of constipation in patients with CKD.The KDIGO guidelines do not address the management of constipation in CKD 11 , and a recent draft of the KDIGO clinical practice guideline only mentions the role of constipation in terms of hyperkalemia treatment.Recently published algorithms for constipation treatment in patients on HD recommend to avoid magnesium-or phosphate-containing laxatives 46 .The authors of the algorithm suggested nonpharmacological therapies such as fiber, fluid intake, and exercise, followed by pharmacological therapy using PEG, lactulose, senna, and bisacodyl 46 .
Constipation in patients with CKD is speculated to be multifactorial 9 .Fluid and dietary restrictions are among the main causes of constipation.Decreased physical activity and medications commonly used in CKD also affect bowel movements.As expected, our study showed that CKD patients with constipation were frequently prescribed constipation-inducing medications related to their CKD condition, including diuretics, potassium binders, iron supplements, antihistamines for pruritus, and analgesics for chronic pain.However, we also found the greatest risk of constipation in CKD patients who are prescribed psychotropic drugs such as antipsychotics, antiepileptics, and antidepressants.Psychotropic drugs and medications for neurologic disorders such as Parkinson's disease and dementia are known to cause constipation, but the risk of constipation when prescribing these medications in patients with CKD has been relatively unrecognized 47,48 .Therefore, our data suggest that clinicians should be aware of the potential adverse effects of these drug categories in managing patients with CKD.Constipation could also be attributed to multiple comorbidities that are highly prevalent in CKD, such as diabetes, cardiovascular disease, depression, and insomnia, which is consistent with our findings 49,50 .Prior to pharmacotherapy, identifying medications and conditions associated with constipation should be prioritized.In addition, several studies have supported that uremic toxins and alterations in the gut microbiome are the main mechanisms responsible for constipation in CKD, but the causal role and clinical implications of dysbiosis remains elusive [51][52][53] .
Our study has some limitations.We used retrospective data obtained from the health insurance reimbursement process.We determined CKD based on the diagnostic codes of treated patients, not on GFR or albuminuria.There may be some discrepancies between the HIRA data and actual patient conditions.Therefore, we only included patients with 2 or more inpatient or outpatient ICD codes for CKD per year to reduce false positive cases.Given that patients with early-stage CKD have low health care utilization rates, this method could be particularly effective for selecting true mild CKD patients.On the other hand, we were able to identify nearly all ESKD patients with insurance claim codes because all patients receiving HD or PD in Korea are registered with the differential co-payment system of the National Health Insurance Service 54 .The definition of constipation is also based on diagnostic codes and not clinical diagnostic criteria.Therefore, the estimated prevalence may be higher than that estimated using the Rome criteria.In addition, because data after 2017 were not available, newly developed drugs for constipation, such as serotoninergic agonists and chloride channel activators, were not included.To overcome these limitations, more controlled prospective studies, including recently developed drugs, are needed.Nevertheless, HIRA contains enormous amounts of data, including demographics, diagnoses, prescriptions, and outcomes, extracted from almost the entire Korean population.The findings are not specific to any one region or institution but rather reflect the overall treatment patterns and characteristics of constipation in patients with CKD.To our knowledge, this is the first study to investigate the prevalence of constipation in Korean patients with CKD, including pre-dialysis patients.Therefore, our findings add to the understanding of the epidemiology of constipation in patients with CKD and could serve as a useful reference for future studies.
In conclusion, this study provides real-world evidence on the prevalence of constipation and laxative use according to the CKD stage in a nationwide population-based cohort.We found a significantly higher risk of constipation in patients with CKD than in non-CKD individuals and different prescribing patterns of laxatives.

Figure 1 .
Figure 1.Proportion of constipation and laxative prescriptions in patients with chronic kidney disease (CKD).(A) Constipation diagnosis according to the CKD stages.(B) Laxative use according to the CKD stages.(C) Constipation diagnosis in patients with end-stage kidney disease (ESKD) including kidney transplant (KT) recipients.(D) Laxative use in patients with ESKD including KT recipients.

Table 2 .
Risks of constipation diagnosis and laxative use in patients with CKD.All odd ratios were adjusted for age group, sex, and condition occurrence at any prior time.Patients with stage-unspecified CKD were included in the analysis.The non-CKD and CKD groups were matched according to age groups and sex.CKD, chronic kidney disease; OR, odds ratio; CI, confidence interval.

Table 3 .
Drug groups prescribed significantly more often in CKD patients with constipation.Laxatives were excluded.Potassium-sparing agents include spironolactone and amiloride.Drugs for acid-related disorders include histamine-2 receptor antagonists and proton pump inhibitors.Vasodilators include nicorandil, molsidomine, glyceryl trinitrate, isosorbide mononitrate, and isosorbide dinitrate.BPH, benign prostatic hypertrophy; OR, odds ratio; CI, confidence interval.