Frequency of ADRs and their Economic Impact in a Tertiary Care Public Sector Hospital in South India

Inclusion Criteria

• Patients of either sex who developed ADR during treatment and whoever gets admitted due to ADR.

Exclusion Criteria

• Patient who developed ADR due to intentional or accidental poisoning.
• Patient with drug abuse, intoxication and over dose.
• ADR due to fresh blood or blood products transfusion.
• Patients from outpatient department

Data Collection Procedure

The study investigator and pharmacovigilance associate (PvPI) reviewed the patient’s drug treatment charts, clinician and nursing notes of all the patients admitted in general medicine wards during the study period (Monday to Saturday). The review was conducted to screen the case records and the patient was interviewed for the presence of any ADRs. The objective markers of ADRs, e.g., laboratory results were identified from the case notes and hospital information system (HIS). The subjective marker of ADRs like headache, rash and nausea were identified through the patient progress notes, patient interview and finally there was a discussion with the concerned medical team. The ADRs defined according to World Health Organization (WHO)¹³ and as per study inclusion, exclusion criteria, after explaining the study purpose and details to each subject, informed written consent was obtained and included in the study.

Economic Burden due to ADRs

The direct cost associated with the management of ADRs (from the beginning of the ADR to end of treatment) was calculated by considering the actual purchasing cost of drugs and medical devices born by the institute. The length of stay extended due to ADRs and laboratory investigations for management of ADRs were discussed with the medical team and then it was included. The extended hospital stay of bed charge per day and patient food cost per day and laboratory investigation charge was calculated according to hospital information type C class charge (bed charge fifty rupees per day, food cost sixty rupees per day and laboratory investigation charge were minimal). Since the hospital was public sector the cost of physician, nursing, administration charge was not taken into account. The indirect costs of ADR treatment were calculated according to the human capital approach 2013 (actual loss to the patient/caretaker) like traveling expenses, daily wages of patient and caretaker (one per patient) and for the miscellaneous expenses, the patients or caretaker were interviewed the actual cost was calculated. However, the intangible cost like pain, sadness and depression of patients due to ADR were not considered in the study.

Data Documentation

The data were included either sex who developed ADR during treatment, whoever gets admitted due to ADRs. Who developed ADR due to intentional or accidental poisoning, drug abuse, intoxication, overdose and ADR due to fresh blood or blood product transfusion was excluded. The identified ADRs were documented in the suspected ADR documentation form provided by the Indian Pharmacopoeia Commission (IPC). The name and quantity of drug prescribed for ADR management, number of hospital stay days extended for ADR management and indirect cost interviewed from patient or caretaker were documented, separately designed data collection proforma.

Constitution of the Independent Clinical Panel

The documented ADRs were assessed by a panel of experts consisted of two clinical pharmacologists, one clinician, a research scholar and one pharmacovigilance associate. The panel evaluated the causality,^14,15 severity,¹⁶ types of reactions¹⁷ and seriousness¹⁸ of ADR using the appropriate scale. The reactions associated with the system involved were coded using World Health Organization Adverse Drug Reaction Terminology (WHO-ART). Drugs associated with ADRs were classified using WHO-Anatomical Therapeutical Classification (WHO-ATC). When there is a disagreement between members of the clinical panel, a consensus has arrived after discussion. The opinion of treating clinicians was also considered in arriving at a consensus. Information required for the assessment of ADR was obtained from standard drug information resources.

Evaluation of Results

All the identified ADRs were evaluated to identify the pattern the concerning patient demographic characteristics, reaction characteristics, medication usage of adverse drug reaction. The data was analyzed to determine the incidence, type of reaction, causality, severity, the seriousness of reaction, organ system affected, drugs implicated ADRs, risk factors for the development of ADRs, the cost associated with the treatment of ADRs. The frequency of ADRs to individual drugs calculated according to Hurwitz et al.,¹⁹ number of patients developed ADR for particular drug/total number of a patient exposed for particular drug × 100.

Statistical Analysis

All the data analyzed descriptively using Graph Pad Instat Version 3.0 and SPSS version 19.0 (IBM PASW Statistics; 19.0). The distribution of all the categorical data related to the patient characteristics such as gender, type of drug use, organ affected, etc., was presented on the frequencies and percentage. The logistic regression analysis was used to find out the association of risk factors for the development of ADRs like gender, diagnosis, age group, number of drug intake, number of hospital stays. All statistical analysis was carried out for two-tailed significance and p %3C; 0.05 were considered as statistically significant.

General Demographic Characteristic of the Study Population

A total of 3,012 patients were intensively monitored during the study period. Among them, 317 patients identified 382 ADRs. Amidst 317 patients, 265 (8.8%) patients developed ADRs during the hospital stay, 52 (1.7%) patients admitted to hospital due to ADRs and death due to ADRs was 1 (0.32%). The death was due to warfarin-induced intracranial bleed. The overall incidence of the ADR was found to be 10.5% and the average number of ADRs in a patient was 1.2, among the patients who suffered from ADR. The demographic details of the study population are presented in Table 1.

Characteristics and Assessment of Adverse Drug Reactions

System Organ Class (SOC) Affected due to ADRs

The most commonly affected system organ class were skin and appendage disorder [n = 76 (19.9%)] followed by gastrointestinal disorder [n = 58 (15.8%)] and neurology disorder [n = 40 (10.47%)]. Rash and vomiting (each n = 19) were the most commonly identified ADRs followed by giddiness and hepatocellular damage (each n = 18). The details of the system organ class affected by the ADRs and commonly reported ADRs are presented in Table 2.

Anatomical and Therapeutic Class (ATC) of Medication Implicated in ADRs

Anatomical class of medication frequently implicated in the ADRs were anti-infective system (J) (n = 116, 30.37%) followed by alimentary tract and metabolism (A) (n = 81, 21.2%). Among the anti-infective systems, anti-bacterial (J01) (n = 72, 18.85%) and antimycobacterial (J04) (n = 29, 7.59%) were the common drugs causing ADRs. Anatomical and therapeutical classes of medication implicated in ADRs are presented in Table 3.

Frequency of ADRs for Individual Drugs and their Incidence

The higher incidence of the adverse drug reactions was observed with methotrexate (33.33%), followed by dapsone (23.8%) and antitubercular drugs (ATT) (22.58%). The incidence of ADRs of cyclophosphamide and vancomycin observed in our study were 21.05% and 18.42%, respectively. The top five drugs causing the higher number ADRs were ATT (25), ceftriaxone (22), vancomycin (21), phenytoin (17), and prednisolone (16) (Table 4).

Risk Factors for the Occurrence of Adverse Drug Reactions

Multivariate regression analysis revealed that factors like gender, presence of comorbid medical conditions, polypharmacy and length of hospital stay were the risk factors for the development of adverse drug reaction.

Economic Burden Imposed by Adverse Drug Reactions

During the study period, 382 ADRs were identified. Of the total ADRs, 293 (76.7%) ADRs from 248 patients cost incurred in the management of ADRs. Amidst 293 ADRs, 209 ADRs have increased the length of stay (LOS) with 172 patients. The extended hospital stay was accounted for 1,002 bed days and the median increase in the length of stay was 5.8 days. The total economic loss from 248 patients of ADR was found to be ₹835,133, which includes total direct cost ₹296,093 (includes drug cost, food cost, bed charge and lab charges only) and total indirect cost ₹539,040 [includes traveling expenses, daily wages of patient and caretaker (one per patient) and miscellaneous expenses]. Considering the direct and indirect cost imposed by ADR per patient was ₹3,367 (₹835,133/248). The severity-based costs imposed by ADRs were shown in Table 5.

Participant’s Description

Adverse drug reactions are a serious health concern for society. The mortality and morbidity associated with hospital admissions and during hospitalization related to adverse drug reactions result in economic burden to the country.²⁰ Based on the “Health at a Glance 2017” from the organization of economic cooperation and development (OECD) report stated that India’s, spending on health care is very less compared to the developed countries,²¹ so it is very important to utilize the money very effectively. Identifying and preventing the ADRs by effective monitoring of the patients is very important in any healthcare setting. Most of the advanced countries have good infrastructure for adverse drug reaction reporting systems at the national level. Adverse drug reaction reporting programs on an institutional basis can provide valuable information about potential problems in drug usage in that institution. So we have taken up a study to evaluate the incidence and pattern of ADRs, its economic impact in our hospital setting.

In our study, a total of 3,012 patients were intensively monitored for ADRs in the general medicine ward. Three hundred and eighty two ADRs from 317 patients were considered for final analysis. In our study, 1.7% of patients were admitted to hospital due to ADRs which is less when compared to the other published reports from South India which reports 3.4% and 14%.^22,23 Most of the study reported that 2.9–6.7% of all hospital admissions are due to ADR which is higher compared to our study findings.^24,25 This difference could be attributed to different sample sizes, study setting, methods of monitoring and different study designs. However in the present study, 8.8% of patients developed ADRs during the hospitalization which is high compared to the other South Indian studies, which reported 3.7% of patients developed ADRs during hospitalization.^4,22 This study findings emphasize the need for critical monitoring of ADRs so that it can be prevented.

Incidence of Adverse Drug Reactions

The overall incidence of ADRs was 10.5% which is consistent with the findings from Arulmani et al.²² and Peter et al.²⁶ They reported incidence of 9.8% and 10.42%. However, studies from Mysuru and Gujarat reported the incidence rate of a hospitalized patient as 4.4%, 2.4%.^4,27 This could be attributed to under-reporting of ADRs whereas in our study intensive monitoring was done which resulted in more numbers of ADR observed.

In this study, the occurrence of ADRs was more in females when compared to the male group which was similar to that of other studies.^4,22Some studies have reported male patients have more ADRs than females.^24,28and this difference can be due to more male patients getting admitted and monitored in the hospital. In our study, the higher incidence of ADRs reported in females compared to men. This may be due to female patients have lower body weight and more body fat than men and the lower concentration of hepatic enzymes.²⁹ So rational dose titration may lead to the minimization of ADRs in females. Among the patients who experienced ADRs 78.5% of patients have more than two comorbid conditions. As the multiple diseases need various drugs for the treatment it leads to polypharmacy. The more the number of drugs prescribed, the higher is the probability of the occurrence of ADRs. The frequencies of ADR were higher in elderly patients (11.1%) and followed by 19–59 years (10.8%). Our findings were consistent with the studies reported from UK² and India.²³ which showed that geriatrics patients experience ADRs more compared to other age groups. In the present study, 70% of the patients admitted were in the age group of 19–59 years followed by 23% of them were geriatrics patients. This is in agreement with reported from Ramesh et al.⁴ which noticed that 67% of the patients were adults and nearly 30% were geriatrics patients.

Assessment of Adverse Drug Reactions

In our study type A reactions accounted for 72% of ADRs and followed by a type H reaction of 19.6%. These findings were in agreement with other study reports from South India whereas type A reactions were commonly observed.³⁰ Another South Indian study found a predominance of type B reactions, which is contradictory to our study findings.³¹ Arulmani et al. have reported type H reactions were most commonly observed which contradictors our findings.²² Type A reactions are mostly drug related and preventable. Type B reactions were pharmacologically predictable and type H reactions are unpredictable and unpreventable. In our study type A reactions were mostly observed which can be prevented from their known pharmacology and its intended effects so it was potentially avoidable. Critical monitoring of drug usage in patients can help in preventing type A reactions.

In our study 54.2% of ADRs were serious and among them, 8.4% was life-threatening followed by ADRs causing hospital prolongation and admission were 27% and 15.2%, respectively. Study findings from Gujarat reported that life-threatening ADRs were 8.3% which is similar to our study findings. In addition to that 42.1% of ADRs leads to initial or prolongation of hospitalization which is also in agreement with our findings. Our study revealed that only 3.4% ADRs required intervention to prevent permanent damage which is contradictory to study reported from Gujarat which reveals 43% of ADRs required intervention to prevent permanent damage.³² This may be due to higher antitubercular and antiretroviral drugs prescribed in their study which resulted in higher immunologically mediated hypersensitivity reactions. The study reported from Italy revealed 28% of ADRs were serious which is less compared to our study findings and further 26% of ADRs lead to hospitalization which is also less compared to our study.³³ This attributed difference may be due to the study was conducted in children’s whereas in our study all age groups of patients admitted in general medicine wards were monitored. In the current study, 54% were serious ADRs, so it is prudent to vigilantly monitor and follow-up the patients for early identification and prevention of serious ADRs. These findings suggest that the prescriber should remain vigilant and also educate the patient about the possible side effects of drugs.

Causality assessment of ADRs as per WHO scale reveals the majority of the ADRs, belong to possible (56.54%) followed by probable (29.32%) and certain (12%). Using Naranjo’s algorithm 66% of ADRs were classified as probable followed by 29.3% of them as possible and 4.7% of them were definite. Our study findings were in agreement with an Arulmani et al.²² which reported that as per Naranjo scale 62% of ADRs were probable and 31% of them were possible. Another study from Telangana³⁰ about the causality assessment as per Naranjo’s scale reveals 58% of ADRs were probable which is slightly less than our findings and followed by 30% were possible which is higher than our study findings.

In our study as per severity, 68% of ADRs were moderate followed by mild 24% and only 8% of ADRs were found to be severe. Study findings of severity assessment using the Hartwig and Siegel scale from Telangana also revealed that most of the ADRs were moderate and only 8% of them were severe. These findings were comparable to our study results.³⁰ Our study findings the majority of ADRs were moderate in severity in accordance with other studies.⁴

Most Commonly Affected Organ Systems due to ADRs and Medication-implicated ADRs

The most commonly affected system organ classes were skin and appendage disorder (19.9%) followed by gastrointestinal disorder (15.8%)] and neurology disorders (10.47%). The results were comparable with the study reported from the USA which revealed that the system most badly affected was the gastrointestinal system and dermatological.³⁴ Our study findings also consistent with studies reported from south India which reveals gastrointestinal, dermatological and nervous system were predominantly involved reported organ class of system involved in adverse drug reactions.^22,24 Rash and vomiting was the most commonly identified ADRs followed by giddiness and hepatocellular damage. Our study findings were in agreement with Manipal-based study which reported dermatology was the most commonly affected organ system with skin rash as the most frequently reported reaction.³⁵ In the present study anti-infective system followed by the alimentary tract and metabolism, cardiovascular system was commonly involved. Among the therapeutic class of drugs, antibacterial followed by anti-inflammatory and anti-diabetics therapeutic classes of drugs were commonly implicated in ADRs. Our finding is consistent with the studies reported by Suh et al.³⁶ and Prosser et al.³⁴ They have reported anti-infective and anti-inflammatory drugs were predominantly involved in the development of ADRs. So antibiotics usage should be monitored and an antibiotic sensitivity test should be done in suspected resistance cases to prevent the antibiotic resistance.

Risk Factors for the Occurrence of Adverse Drug Reactions

The median number of drugs prescribed per ADR developed patient was 6.5 which was consistent with the study reported from France and Spain.^28,37 The median hospital stay per ADR developed patient was 6 days which was consistent with findings from other studies.³⁸ Studies reported from the USA and England had 3 median days per patient which is less compared to our study.³⁹ Predisposing factors were assessed for the risk of ADRs based on various factors like age, gender, no of drugs, length of hospital stay, etc. Multivariate regression analysis identified female gender, presence of comorbid medical conditions, polypharmacy and length of stay as the risk factor for the development of adverse drug reaction. Length of stay and polypharmacy were strong indicators for the risk of adverse drug reactions. Those who take more than two drugs have a higher risk of ADR and more than or equal to 6 drugs have 2.5 times more likely to experience ADRs as compared to patients taking 1–2 drugs. Similarly, patients who stay more than or equal to 5 days stay in hospital have 4.95 times more risk for the development of ADRs. So it’s important to have medication review and focused monitoring to avoid unnecessary drugs.

Economic Burden Imposed by Adverse Drug Reactions

The results revealed that total economic loss from 248 patients with ADR was ₹835,133 and the cost imposed (total direct and indirect cost) by ADR per patient was ₹3,367. The direct cost for the treatment per ADR was ₹1,010. Our study results were comparable to study findings of economic assessment of ADRs in a private hospital which revealed the average direct cost per patient hospitalized with an ADR was ₹4,945. The north-Indian study revealed the total direct cost (₹56,620) and indirect cost (₹262,880) for the total cost for treatment of ADRs was ₹319,500.⁹ These differences may be due to study setting and other factors such as public sector hospital and private hospital. Another study from Pune¹⁰ reported the average cost incurred due to ADR related hospitalization was found to be ₹578.55 and ADR cost occurred in hospitalized inpatient was ₹441.86. We didn’t assess separately cost of ADRs in hospitalized patient and ADRs related hospitalization. This was one of the limitations of our study. In our study total, direct cost was ₹296,093 includes drug cost, food cost, bed charge and lab charge only. The total indirect cost was ₹539,040 includes traveling expenses, daily wages of patient and caretaker and miscellaneous expenses. Most of the studies in India only reported direct costs for the treatment of the ADRs.^4,7,10,22 We have assessed both the direct and indirect cost associated with ADRs. The direct cost of ADRs was about the management of ADRs. Assessing the indirect cost was also very important as it reveals the economic burden of ADRs. The average direct cost incurred in each mild, moderate and severe reactions were ₹171.5, ₹898, ₹2,073, respectively which was higher than the study reported by Arulmani et al.²² This difference may depend upon the nature of the hospital, the year of study the conducted and other factors. Our findings suggest that vigilant monitoring of ADRs and prevention of ADRs was the need of the hour. Establishment of standardized operating approaches and voluntary reporting of suspected ADR’s by all healthcare professionals working in the hospital can result in better monitoring and prevention of adverse drug reactions. Computerized prescribing may lead to a reduction in ADRs and if possible it can be implemented.