DIRECT COST ASSOCIATED WITH ADVERSE DRUG REACTIONS AMONG HOSPITALISED CHRONIC KIDNEY PATIENTS IN A PUBLIC HEALTHCARE FACILITY IN MALAYSIA: A RETROSPECTIVE 3-YEAR STUDY

Adverse reactions which are clinically diverse increases the overall cost of care, as it often results in additional days of hospitalisation, clinical investigations and treatment drugs. Thus, the main objective of this study is to evaluate direct medical costs among chronic kidney disease (CKD) patients who experienced adverse drug reactions (ADRs) during hospitalisation and identification of associated drug classes and clinical symptoms. Individual direct (n = 23 [14.4%]), analgesic (n = 12 [7.5%]), statin (n = 10 [6.3%]) and anti-diabetic (n = 8 [5.0%]). Antibacterial constitutes the majority of the anti-infectives reactions. Vancomycin (n = 7 [13.7%]) tops the most ADRs contributing antibacterial. ADRs experienced during hospitalisation caused prolongation of hospitalisation and its associated investigational and treatment charges. The true value of the cost estimate could be much higher than the calculated value as the indirect costs were not included in the final estimates of this study and as a result of the Malaysian government’s waiver policy.


INTRODUCTION
Chronic kidney disease (CKD) is defined according to the presence or absence of kidney damage and level of kidney function. CKD patients have an increased mortality rate and higher risks of adverse events and toxicity from treatments (Webster et al. 2017), stressing that adverse drug reactions (ADRs) are important factors contributing to morbidity and mortality (Khan et al. 2016;Davies et al. 2009).
Management of ADRs can be costly which utilises a considerable amount of financial resources. ADRs are a major concern for morbidity and mortality in the public health sector, resulting in more than 100,000 deaths in the United States with annual treatment costs of USD136 billion (Tan et al. 2016). Globally, many countries use about 15% to 20% of their hospital budgets to manage drug-related complications (Bordet et al. 2001). Impact and administration of ADRs are complex and can cost up to USD30.1 billion annually in the United States. ADRs further increases treatment costs due to the prolongation of hospitalisation and additional clinical investigations. ADRs aggravate treatment cascades when new drugs are prescribed for conditions resulting from another drug usage, which is considered to be an unrecognised ADR. The cost per preventable ADRs was higher than for non-preventable ADRs for ADRs that require hospitalisation . Another in-patient study reported that the cost per ADR was USD2,262. Hospitalised ADR patients' costs, fluctuate depending on the hospital and wards . ADRs occurrence during hospitalisation represents an increase of 9% in the length of stay and a 20% increase in the cost of care for bed usage, laboratory and treatment (Khan et al. 2013).
Thus, the main aim of this study is to evaluate the individual direct medical costs among stages 3-5 CKD patients who experienced ADRs during hospitalisation from the perspective of the Ministry of Health (MOH) Malaysia and identification of associated drug classes that contributed to ADRs and its associated clinical symptoms. This is the first study in Malaysia to estimate the direct cost of treating ADRs among hospitalised CKD patients in a general healthcare facility.

Study Design and Patients
This is a retrospective observational study conducted in General Hospital Pulau Pinang, the second-largest General Hospital in Malaysia. A total of 1,070 medical records of CKD patients who experienced ADRs from various wards for 3 years 3

Direct Cost Associated with Adverse Drug Reactions
Malay J Pharm Sci, Vol. 19, No. 2 (2021):  (1 January 2014 and 31 December 2016) were screened for this study. CKD patients in stages 3-5 (estimated glomerular filtration rate [eGFR] < 60 mL/min/1.73m 2 ) with stable serum creatinine (sCr) values during the initial days of admission and experienced ADRs during hospitalisation were the primary inclusion criteria of this study. The sCr value obtained during the first day of admission were used to estimate the glomerular filtration rate (GFR). Additional inclusion criteria were patients aged ≥18 years old and admitted for more than 24 h. Medical records that were dubious and incomplete and ward admission due to ADRs or acute kidney injury (AKI) were excluded from this study. Only 160 patients were selected after subsequent screening and identification of records that met the inclusion and exclusion criteria. From the total number of the patient records finally selected, 132 patients survived and 28 patients did not survive from ADRs during hospitalisation. More detailed study methods outlined in this manuscript have been described extensively in our earlier published research articles (Danial et al. 2018;. Before study commencement, ethical approval was obtained from the Medical Research and Ethics Committee (MREC), MOH Malaysia.

DESCRIPTION OF STUDY OUTCOME
Individual direct medical costs from the perspective of the MOH Malaysia, among stages 3-5 CKD patients who experienced ADRs during hospitalisation and identification of associated drug classes and clinical symptoms were the main outcome evaluated in this study.

DATA COLLECTION
Reported drugs that caused ADRs and the affected organs due to ADRs were recorded from the ADRs reporting form acquired from Drug Information Unit General Hospital Pulau Pinang. The reported ADRs and the affected organs were cross-checked again for any discrepancies in the patients' medical reports. Once, no discrepancies observed in both documents, the drugs and the affected organs were classified into major drug classes and major symptom organ classes.
In this study, only the cost incurred during ADRs were considered. The duration ADRs for each patient were captured from the ADRs reporting form. Direct medical costs associated with ADRs during hospitalisation such as extra days of monitoring and laboratory costs plus treatment drug costs incurred were acquired from the hospital's revenue department database. Monitoring and laboratory costs included incurred ward charges, inpatient charges, laboratory tests charges, ultrasound tests performed and haemodialysis performed. Treatment drug costs were the price of drugs used in treating ADRs. The cost of treatment was considered to be zero for ADRs requiring discontinuation of the suspected drugs. The costs of all these were included in calculating the costs incurred by the individual patient when treating ADRs requiring additional medications, treatments, or laboratory tests. All the cost calculation in this study was done in Ringgit Malaysia (RM).

MINISTRY OF HEALTH, MALAYSIA HEALTHCARE POLICIES
In Malaysia, at all governments hospitals and clinics waivers of charges and/or a special discount of the hospital's bill are granted for patients infected with all types of infectious diseases or illness, for those who are earning below RM300 per month, the holder of Malay J Pharm Sci, Vol. 19, No. 2 (2021): 1-21 Social Welfare or National Islamic Council card, disabled person, students and government servants (MOH 2017b). Also, there is a 50% reduction with maximum charges of RM250 for each admission for the senior citizens aged 60 years old and above (MOH 2012). Furthermore, most of the monitoring, laboratory and drug costs charges were already at a subsidised rate by the Malaysian government (MOH 2017a). Most of the eligible patients included in this study were over 60 years old and were either waived or given a discount on their total hospital bill. The cost of treatment was considered to be zero for waived payments.

STATISTICAL ANALYSIS
Major symptom organ classes listed based on major drug classes were presented as frequency (n) and percentage (%). Patients with multiple ADRs were considered only once during cost analysis. Normally distributed data were reported as mean ± SD. Non-normally distributed data were reported as median (first quartile [Q1]-third quartile [Q3]). Categorical variables were presented as frequency (n) and percentage (%). Two-sided p-values of less than 0.05 were considered to be statistically significant. Statistical analysis was performed by using IBM SPSS version 22 (SPSS Inc., Chicago, IL).

Major drug class Patients, n (%) Symptom organ class Frequency of ADRs, n (%)
Anti-viral

DISCUSSION
ADRs contributed to a longer duration of hospitalisation coupled with a higher ward, laboratory and drug costs in this study. Our findings were supported by previously published studies (Classen et al. 1997;Phillips, Christenfeld and Glynn 1998;Budnitz et al. 2006). The differences in ADRs costing outcomes were influenced by factors such as disease prevalence, culture differences, ethnicity and economic status (Chan et al. 2015;Jung et al. 2017). The top five major drug classes that contributed to ADRs among our study patients were anti-infectives, anti-hypertensive, analgesic, statin and anti-diabetic medications.
Prolongation of hospitalisation days remains the parameter that best reflects the increase in ADRs' direct cost (Bordet et al. 2001). The estimated cost in this study is much lower than data from other published studies (Rottenkolber, Hasford and Stausberg 2012;Bates et al. 1997;Bates et al. 1999;Jha et al. 2001;Dartnell et al. 1996). However, these published studies also reported that the costs reported were very expensive. In their final cost calculations, ADRs-associated risks in ambulatory treatment were also included in the review of cost accountability concerning the origin of these studies (Thurmann 2001). Moreover, the higher cost reported by some previous studies may be due to the severity of the ADRs (Chan et al. 2008). In addition, lower cost estimation in this study may due to waiver policy implemented by the Malaysian government as the majority of the patients in this study were more than 60 years old. Also, since Malaysia has centralised its public sector administration, it is subjected to policy and programmes formulated and financed by its MOH. Also, lower-cost estimation in this study may due to the waiver policy implemented by the Malaysian government as the majority of the patients in this study were more than 60 years old. Also, since Malaysia has centralised its public sector administration, it is subjected to policy and programmes formulated and financed by its MOH. Therefore, patients will only be required to pay minimum fees as low as RM1 when treated at government health facilities which may influence the actual total payable amount (Rosli et al. 2017).
The highest number of days of hospitalisation and the estimated direct cost of monitoring and laboratory costs were associated with haematological reactions from this study. Treating haematological reactions often results in the continuation of hospital care coupled with tough clinical decisions when patients have advanced disease states such Malay J Pharm Sci, Vol. 19, No. 2 (2021): 1-21 as kidney disease and alternative treatment options are limited (Ganguli et al. 2015). The advanced state of kidney disease also emerges as an important complication that further complicates the treatment. Moreover, it was reported that hospitalised patients with kidney impairment with haematological reaction possess the highest mortality rate (Salahudeen et al. 2013;de Mendonca et al. 2000). Other associated risk factors were older age, mechanical ventilation and the use of nephrotoxic medications (Lahoti et al. 2010).
In this study, the electrolyte reaction was also one of the important factors that contributed to the prolongations of hospitalisation and higher monitoring, laboratory and drug costs. Instability of pH even for a short time might lead to a vast array of adverse effects (Dhondup and Qian 2017). The kidneys play a key role in electrolyte and acid-base balance regulation. In the case of kidney insufficiency, electrolyte and acid-base imbalances occur such as hyperkalaemia, metabolic acidosis and hyperphosphatemia which results in severe complications such as muscle loss, bone-mineral disorder, vascular calcification and mortality (Raphael et al. 2011;Kovesdy et al. 2009).
Moreover, another important disorder that contributed to the highest drug treatment cost and moderate prolongation of hospitalisation, monitoring and laboratory cost were psychiatric reactions. Psychiatric reactions such as depression and anxiety are associated with lower quality of life among CKD patients as it is associated with lower treatment adherence, frequent hospitalisations and increased mortality (Kimmel et al. 2019), which directly increases the cost of treatment.
Also, CKD patients with diabetes were significantly associated with higher morbidity and mortality, mainly due to high cardiovascular risk (Grandfils et al. 2013). Metformin is one of the common drugs for type 2 diabetes treatment (American Diabetes Association 2014). Metformin mainly acts by decreasing the production of hepatic glucose, increasing the peripheral absorption of glucose, improving glucose tolerance and lowering fasting and postprandial plasma glucose. Metformin should be prescribed with caution as it is renally excreted and it results in the development of lactic acidosis as the most serious adverse effect (Game 2014).
In addition to the direct financial costs, patients and their caregivers also have several indirect expenses incurred by ADRs, such as missed working days and/or morbidity, such as anxiety caused by the ADR episodes (Wu and Pantaleo 2003). Other related expenses are wages of the medical personnel, disposable goods and medications (Wasserfallen et al. 2001). Prevention efforts can be undertaken by implementing ADR evidence-based surveillance programs in healthcare facilities which may focus on ADR causative drug classes to reduce human consequences and economic costs which effectively leads to cost savings (Dormann et al. 2000;Tan et al. 2016).

CONCLUSION
To the best of our knowledge, this is the first study in Malaysia to estimate the direct cost of treating ADRs that occurs during hospitalisation among CKD patients in a general healthcare facility. Our finding showed that ADRs experienced during hospitalisation stay caused prolongation of hospitalisation and its' associated investigational and treatment charges. The true value of cost estimation might be far greater than the calculated value as the indirect costs were not included in the final estimation of the present study and due to the waiver policy implemented by the Malaysian government. This study outcome hoped to contribute some preliminary information on the cost of treating ADRs among hospitalised CKD patients in Malaysia for more in-depth future research.

STUDY LIMITATIONS
Estimation of the actual hospitalisation cost due to ADRs could not be determined as the indirect cost is not taken into account in this study and the estimation of the direct hospitalisation cost is impaired by waiver policy and subsidisation by the Malaysian government. Furthermore, since the cost was estimated from the perspective of the MOH, the study results may not reflect the cost in private hospitals as drugs are procured at subsidised rates for public hospitals in Malaysia.