Factors Contributing to Negative Outcomes Associated with Medications and Drug-Related Problems in Kidney Replacement Therapy—A Hospital-Based Prospective Observational Study

Background: Negative outcomes associated with medications (NOM) and drug-related problems (DRP) significantly impact individuals with kidney replacement therapy (KRT) given the complexities of managing kidney disease and associated comorbidities. The present study aims to assess the frequency of NOMs/DRPs among KRT patients and identify contributing factors. Methods: A cross-sectional study was conducted at Virgen de las Nieves University Hospital (Granada, Spain), involving 117 outpatient adults with KRT. Data were collected from February 2021 to July 2023 using electronic records, semi-structured interviews (Dáder Method), and discussions with nephrology specialists. NOMs/DRPs were identified following treatment guidelines. Binary logistic regression was used to determine associated factors (p-value < 0.05). Results: Across 117 patients, 2436 NOMs and 3303 DRPs were identified, averaging 20.82 NOMs and 28.23 DRPs per patient. Prevalent NOMs included untreated conditions (58.95%), quantitative ineffectiveness (35.43%), and non-quantitative safety problems (5.13%). Dominant DRPs were undertreated conditions (37.63%), wrong dose/posology/length (33.00%), risk of adverse drug reactions (ADR) (16.14%), and non-adherence (6.87%). Patients with ADR, undertreated conditions, and anemia were associated with quantitative ineffectiveness. Risk of ADR and vitamin D deficiency/insufficiency correlated with non-quantitative safety problems. Conclusions: KRT patients exhibited a substantial prevalence of NOMs/DRPs. Further research is needed to deepen our understanding of these complexities for improved patient care.

The intricacies inherent in KRT present formidable medical challenges, compelling the use of a plethora of medications to address associated comorbidities [6].These health conditions, in turn, impose substantial financial strains on healthcare systems.Individuals grappling with these ailments are susceptible to issues related to medications, encapsulated as drug-related problems (DRP), which, in turn, heighten the risks of morbidity, mortality, and healthcare expenditures [1,4,5,7,8].The resultant negative outcomes associated with medications (NOM) further compound these challenges [1,7,8].
Despite the critical nature of the subject, limited studies focusing on NOMs and associated factors in the context of KRT have been published [7].Recognizing NOMs and DRPs emerges as a pivotal responsibility that pharmacists can effectively manage in collaboration with other healthcare professionals through comprehensive medication reviews and subsequent follow-ups [1,3,[7][8][9].The evaluation of NOMs/DRP within the KRT population in Spain holds significance for a multitude of reasons, including the escalating prevalence of CKD within the local community.This increase in prevalence has the potential to detrimentally impact quality of life, prolong hospitalization durations, and elevate the likelihood of both mortality and morbidity [10].Consequently, this study aims to elucidate the frequency of NOMs/DRPs and associated factors among patients undergoing KRT.

Design and Setting
A cross-sectional and prospective observational study was carried out at the nephrology department of Virgen de las Nieves University Hospital (Granada, Spain).The research spanned from 2 February 2021 to 31 July 2023, encompassing a period of 29 months, and focused on patients with ESRD.

Study Population and Eligibility Criteria
All outpatient individuals meeting the following criteria were considered for inclusion in the study: being over 18 years old; undergoing KRT such as HD, PD, or KT at nephrology department during the study period; and expressing a willingness to participate.Patients with cognitive impairment and inpatients were excluded from the study due to their medical conditions.
No cluster randomization was performed during the study.All patients meeting the inclusion criteria for the study were recruited by nephrology experts in clinical practice.
The stage of kidney disease was determined by categorizing patients using the estimated Glomerular Filtration Rate, calculated from serum creatinine levels employing the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation, as documented in the electronic medical records.Laboratory tests were recorded in the electronic medical record as a standard component of routine clinical practice.

Methods
Data collection involved two main sources: electronic medical records and semistructured interviews following the Dáder Method developed by the Pharmaceutical Care Research Group at the University of Granada [11].Additionally, discussions with nephrology experts were facilitated by pharmacists.
The pharmacist conducted a thorough review of electronic medical records and conducted semi-structured interviews.These interviews aimed to assess the number of prescribed medications and identify baseline demographics, comorbidities, clinical laboratory data, allergies, and the number of NOMs/DRPs.
All the data for the analysis, extracted from the medical records, treatment chart, laboratory data, and by interviewing patients, were recorded onto a data collection form.

Study Outcomes
The primary outcome was the prevalence of NOMs/DRPs and associated factors among outpatients undergoing KRT.These included different types of NOMs/DRPs, as was reported in previous studies according to the Granada Third Consensus [11], and associated factors such as age, sex, number of comorbidities, number of medications, clinical laboratory data, time in KRT, time in KT and number of hospitalizations.
Additional secondary outcomes include types of medications and comorbidities.

Sample Size
We included all patients consulting the nephrology department of the Virgen de las Nieves University Hospital and meeting the eligibility criteria.
The sample size was calculated by using simple proportion formula with the estimated prevalence of DRP among CKD patients, p = 81.5% [8], 95% confidence interval, and sample error of 5%, n = 231.In total, 170 ESRD patients were admitted to the hospital.The final sample size, 117, was calculated using the correction formula.

NOMs
A NOM is a result affecting the health of the patient that is or may be associated with the use of medications [11].

DRPs
A DRP is an event or circumstance involving drug therapy that actually or potentially interferes with desired health outcomes [11].

Statistical Analysis
The prevalence of NOMs/DRPs is presented as a percentage, with a 95% confidence interval using an exact binomial test.We tabulated descriptive statistics by a count or percentage for categorical variables, and the mean ± standard deviation (SD), median, or interquartile range (IQR), when appropriate, was used for continuous data.
Spearman's Rank correlation (r s ) was used to measure the association or correlation between the number of NOMs as the dependent variable and the following independent variables: number of DRPs, number of comorbidities, number of medications, age, clinical laboratory data, time in KRT, time in KT and number of hospitalizations.A chi-square test of independence was performed to examine the relationship between categorical variables.A t-test was performed to compare means (baseline and final) for continuous data.
Binary logistic regression was used to analyze the association between independent and dependent variables, and variables with a p-value <0.05 were a candidate for multivariate analysis.

Ethics
Institutional review board approval was obtained from the Andalusian Biomedical Research Ethics Committee (FIS-IRB-2020-01) on 28 July 2020.Written informed consent was obtained from each study participant.

Data Management
The data were arranged into folders identified by patient ID, encompassing patient demographics and additional variables.Variable codes were documented in a separate file, and metadata were supplied to clarify the data format and abbreviations.The information was securely stored on a password-protected server, accessible exclusively to investiga-tors.Routine backups were performed, and access was restricted to the research team to uphold confidentiality.
In compliance with Spanish legislation, the data will be retained for five years post the study's completion.Personal identifiers were eliminated, and a distinct study ID was employed to de-identify patient records, guaranteeing anonymity.

Sociodemographic and Clinical Characteristics of the Study Population
From patient data obtained during a 29-month study period, 117 KRT patients were included in the study.Eleven patients died throughout the study due to complications of kidney disease (Table 1).In terms of the clinical characteristics of the patients, the cause of end-stage renal disease was unknown in 26.50% of cases, while glomerulonephritis accounted for 23.93% (refer to Table 2).
Additional clinical parameters are presented in Table 4.A t-test was conducted to examine the means (baseline and final) of the clinical parameters of interest throughout the follow-up period.The results indicated a statistically significant difference in the means (baseline and final) of urea (p < 0.01) and HDL cholesterol (p < 0.05).

Types of NOMs/DPRs
During the study duration, 117 patients with KRT exhibited a total of 2436 NOMs (Table 5) and 3303 DRPs (Table 6).This equates to an average rate of 20.82 NOMs and 28.23 DRPs per patient.A t-test was conducted to assess the means (baseline and final) of NOM occurrences in the patient group.The analysis revealed no statistically significant differences between the means (baseline and final) of NOM, including undertreated health problems, effect of unnecessary medicine, non-quantitative ineffectiveness, quantitative ineffectiveness, non-quantitative safety, and quantitative safety.
A t-test was conducted to assess the means (baseline and final) of DRP in the patient group, as indicated in Table 6.A statistically significant difference was observed between the means (baseline and final) of DRPs categorized under risk of adverse effects (p < 0.01).However, no statistically significant difference was noted between the means (baseline and final) of the following DRPs: personal characteristics, wrong dose/posology/length, lack of knowledge of medication use, unnecessary drug, non-adherence to medication, other conditions affecting the treatment, precautions, undertreated condition, and others.

Medications
At the baseline, patients used a total of 1398 medications, out of which 1222 were administered at home.By the end of the study period, the overall medication count increased to 1460, with 1279 medications administered at home (Table 7).No statistically significant differences were found between the medians of total medications administered at home and during the dialysis process at baseline and at the end of the study.The predominant instance of quantitative ineffectiveness, characterized by insufficient prescribed quantity or dosage of medication, was observed with darbepoetin alfa, accounting for 94 cases (97.9%).To examine the potential association between quantitative ineffectiveness in NOM and darbepoetin alfa, a chi-square test of independence was performed (X 2 (1, n = 117) = 6.27, p < 0.05) (OR = 7.62, 95% CI = 0.814-97.38).
Moreover, the number of the following DRPs: risk of adverse effects (r s = 0.451; p < 0.001), wrong dose/posology/length (r s = 0.215; p < 0.05), and undertreated conditions (r s = 0.184; p < 0.05); was significantly associated with the number of NOMs, specifically non-quantitative safety.On the contrary, albumin (r s = −0.186;p < 0.05) and magnesium (r s = −0.194;p < 0.05) were significantly associated with the number of non-quantitative safety.
A chi-square test of independence was conducted to explore the significant relationship between the undertreated condition in NOM and the following variables: the presence of wrong dose/posology/length (X 2 (1, n = 117) = 28.49,p < 0.05) and the presence of undertreated conditions (X 2 (1, n = 117) = 38.33,p < 0.05).
Similarly, a chi-square test of independence was used to investigate the significant association between non-quantitative ineffectiveness in NOM and hyperphosphatemia (X 2 (1, n = 117) = 7.68, p < 0.05).
We examined the relationship between independent variables and dependent variables (non-quantitative ineffectiveness, quantitative ineffectiveness, and non-quantitative safety) through the application of both univariate and multivariate logistic regression methods.
In the univariate logistic regression analysis, hyperphosphatemia was found to be associated with non-quantitative ineffectiveness.Additionally, the risk of adverse effects, undertreated conditions, and anemia showed associations with quantitative ineffectiveness.Moreover, the risk of adverse effects and vitamin D deficiency/insufficiency were linked to non-quantitative safety.
Variables with a p-value < 0.05 in bivariate analysis were subsequently included in the multiple logistic regression.The results of the multivariate analysis revealed that participants with hyperphosphatemia were 0.08 times more likely to experience nonquantitative ineffectiveness compared to those without hyperphosphatemia (OR = 0.0845, 95% CI: 0.0044-0.5192,p = 0.02).
Moreover, participants with a risk of adverse effects were 1.802 times more likely to have NOM compared to those without a risk of adverse effects (OR = 1.802, 95% CI: 0.076-18.631).Participants with undertreated conditions were 23.883 times more likely to have NOM compared to those without undertreated conditions (OR = 23.883,95% CI: 1.265-1062.966).Participants with anemia were 8.836 times more likely to experience NOM compared to those without anemia (OR = 8.836, 95% CI: 0.851-90.667)(Table 8).Participants with a risk of side effects were 14.625 times more prone to have NOM of non-quantitative safety compared to those without the risk of side effects (OR = 14.625, 95% CI: 2.646-273.914,p = 0.012).Moreover, participants with vitamin D deficiency/insufficiency were 2.177 times more likely to encounter NOM of non-quantitative safety compared to those without vitamin D deficiency/insufficiency (AOR = 2.177, 95% CI: 0.954-5.058,p = 0.06).

Types of Interventions for Preventing or Resolving NOMs/DRPs among KRT Patients
Throughout the 29-month study period, healthcare providers conducted a total of 3355 interventions aimed at addressing issues associated with medication usage.On average, there were 1. 4

Discussion
Individuals undergoing KRT face an escalated risk of encountering DRPs and NOMs.This heightened susceptibility can be attributed to the presence of comorbidities, complications, and the intricate nature of their medication regimens [1,4].
Our study, which examined 117 patients within the nephrology department, revealed a mean prescription of 12.14 medications and 18.64 comorbidities per patient.Notably, these findings align with those reported in studies conducted in Ethiopia [4], India [12], and the USA [1], where individuals are prescribed an average of 10-12 medications per day, contributing to complex dosing routines and an increased likelihood of encountering DRPs.
The investigation unveiled that antithrombotic agents (94.02%) and COVID-19 vaccines (98.07%) were the most frequently prescribed classes of drugs, while mineral and bone disorder (91.45%) and anemia (89.74%) emerged as the predominant comorbidities.
A total of 2436 NOMs and 3303 DRPs were identified, with the most common NOMs being untreated health problems (58.95%) and the prevalent DRPs being undertreated conditions (37.63%) and wrong dose/posology/length (33.00%).The rates of NOMs and DRPs were 20.82% and 28.23%, respectively, showcasing the substantial burden within this patient population.Notably, our study adds to the existing literature, as some studies primarily focused on DRPs [4,12] while others exclusively identified NOMs [1,7].
Our analysis of NOMs revealed a significant need for additional drug therapy (58.94%), deviating notably from rates reported in other regions.The most common NOM identified was an untreated health problem (99.15%), contrasting with the prevailing DRP, which was a poorly treated health problem (97.44%).Furthermore, ineffective drug therapy contributed to 35.92% of all NOMs identified, indicating a noteworthy divergence from studies conducted in Ethiopia, India, Canada, Beirut, and the USA [8].
Moreover, our study identified statistically significant correlations between various DRPs and NOMs, emphasizing the complex interplay between these factors.Notably, parameters such as urea, creatinine, potassium, calcium, phosphorus, parathyroid hormone, ferritin, erythrocytes, hemoglobin, and hematocrit exhibited significant associations with NOMs related to quantitative ineffectiveness.These findings enrich the understanding of the intricate relationships between patient parameters and NOMs, shedding light on potential avenues for targeted interventions.
While our study contributes valuable insights, it is essential to acknowledge its limitations.The identification of NOMs lacked the establishment of causal relationships, relying on information retrieved from medical records and patient interviews.Future research endeavors should strive to overcome such limitations to enhance the robustness of conclusions drawn from similar investigations.
In summary, our study underscores the multifaceted nature of NOMs and DRPs among KRT patients, emphasizing the need for tailored interventions, collaborative healthcare approaches, and continued research efforts to advance our understanding and enhance patient outcomes within this complex medical landscape.

Conclusions
In conclusion, our study reveals compelling findings regarding the prevalence of NOM and DRP among patients undergoing KRT within the nephrology department.The average rates of NOM and DRP per KRT patient were determined to be 20.82 and 28.23, respectively.Notably, factors such as ADR, undertreated conditions, anemia, and vitamin D deficiency were identified as associated contributors to these outcomes.
The intricate nature of the medication regimen, coupled with the unique disease status of these patients, emerged as a pivotal determinant contributing to the heightened incidence of NOMs.It is evident that this specific patient population demands meticulous attention, emphasizing the importance of a comprehensive healthcare approach.A thorough medication review, sustained follow-up, and collaborative care from all pertinent healthcare professionals are imperative to mitigate complications stemming from the administered therapies and healthcare interventions.These insights underscore the necessity for a multidisciplinary approach to optimize patient outcomes and enhance the overall quality of care for individuals undergoing KRT.

Table 1 .
Characteristics of KRT patients at the Virgen de las Nieves University Hospital (Granada, Spain).

Table 2 .
Underlying cause of End-Stage Renal Disease of KRT patients at the Virgen de las Nieves University Hospital (Granada, Spain).

Cause of End-Stage Renal Disease n = 117 a
a Unless otherwise indicated: numbers present n (%).Kidney replacement therapy (KRT).

Table 3 .
Comorbidities and risk factors for CKD of KRT patients at the Virgen de las Nieves University Hospital (Granada, Spain).
a Unless otherwise indicated: numbers present n (%).bThe list of comorbidities is exhaustive for our cohort and was identified as documented in the medical records.Chronic kidney disease (CKD); kidney replacement therapy (KRT).

Table 4 .
Serum clinical laboratory data of KRT patients at the Virgen de las Nieves University Hospital (Granada, Spain).

Table 5 .
Types of treatments for uncontrolled health problems identified as NOMs among KRT patients.

Table 6 .
Types of DRPs identified as potential causes of ineffectiveness or unsafety in the treatment of uncontrolled health problems among KRT patients.

Table 7 .
Overview of the types of medications used in the treatment of patients with KRT a .

Table 7 .
Cont.The list of medications is exhaustive for our cohort and was identified as documented in the medical records.
a b Unless otherwise indicated: numbers present n (%).

Table 8 .
Factors influencing quantitative ineffectiveness in KRT patients.
interventions per NOM and 1.0 intervention per DRP.49% of the interventions.On average, there were 1.0 pharmacist interventions per patient.These interventions were accepted in 85.5% of cases, totaling 100 accepted interventions.