Pharmacoeconomics of medicines used for geriatric individuals in a tertiary care hospital in Delhi

As per Census 2011, India is home to almost 100 million people over 60 yr, comprising the geriatric age group1. The share of India’s geriatric population is projected to increase to 19 per cent of the global geriatric population by 2050, almost 300 million in number according to the United Nations Population Division2. Most patients in this age group have comorbidities requiring multiple medicines3. The elderly people are also at a greater risk of adverse drug reactions (ADRs) and drug interactions which occur four to seven times more frequently in the geriatric population compared with that in middle-aged individuals45678.

Healthcare costs are of particular concern for elderly population who may be economically dependent and physically less able. In India, 70 per cent of the healthcare expenses are incurred by people from their pockets, of which 70 per cent is spent on medicines alone. It has been estimated that healthcare costs alone put approximately 63 million people at risk of poverty every year9.

It has been observed that households with elderly family members spend 3.8 times more on health than families with no elderly persons. The allocation to overall healthcare spending is maximum in elderly households, 13 per cent, followed by housholds with both elderly and non elderly who spend 7 percent and least in families with no elderly persons who spend five percent.10

It has been projected that by 2030, 45 per cent of healthcare burden will be borne by the elderly in India.

Since the elderly constitutes a large proportion of the population, which is increasing, it will be useful to know the expenditure incurred on medicines utilized among the elderly in in India.

Drug utilization studies, as defined by the World Health Organization (WHO), explore the marketing, distribution, prescription, and use of pharmaceuticals within a society, laying particular emphasis on the medical, social, and economic outcomes.11

The present drug utilization study was carried out with the objective of assessing expenditure patterns on medicines utilized in geriatric inpatients admitted to a medicine department of a tertiary care hospital. The impact of comorbidity, number of medicines prescribed, days of hospitalization, ADRs and potentially inappropriate medications (PIM) prescribed on expenditure on medicines were evaluated.

Material & Methods

A cross-sectional, hospital-based observational study was conducted in the Departments of Medicine and Pharmacology, Maulana Azad Medical College and associated Lok Nayak Hospital, New Delhi. The study was undertaken after obtaining an approval from the Institute Ethics Committee. A written informed consent was taken from all the study partcipants. prior to the start of the study.

Study participants: The study included 1000 individuals prospectively selected using systematic sampling (every fifth case) from the weekly case records of eligible inpatients every month for 12 months. Individuals who were on intensive care support were excluded.

Data about prescribed medicines, occurrence of ADRs and demographic details were obtained from inpatient case records. Comorbidity, cost of medicines in case patients purchased them and follow up information about the occurrence of ADRs were obtained from the individuals. Data were collected using a pretested proforma. The proforma was divided into the following sections:

Demographic characteristics: Demographic characteristics such as age, sex, residence, marital status, diagnosis and duration of hospital stay was obtained from inpatient case record.

Medicines prescribed: Number of the medicines, name (brand/generic), dosage form, strength (amount), route, frequency and duration were recorded. Medicine utilization was quantified in the defined daily doses (DDDs). The total number of DDDs of each medicine prescribed was assessed as follows:

Total number of DDDs for a drug=Total drug dose (in mg) in 1 day×number of days of drug use/1 DDD equivalent (in mg)

The prescribed medicines were classified based on the Anatomical Therapeutic Classification (ATC)12. The International Classification of Diseases (ICD)-10 was used to classify the diagnosis of all patients13.

Appropriateness of medicines prescribed: Appropriateness of the prescribed medicines was determined using the American Geriatrics Society Updated Beers Criteria14. The criterion involves three categories of medicines. The first category, which includes medicines to be totally avoided, was used for assessing the PIM prescribed.

Adverse drug reactions (ADRs): All individuals were monitored for ADR during their stay in the hospital. After discharge, they were monitored telephonically at monthly intervals for one, two and three months. The standard ADR report form of the Central Drugs Standard Control Organisation, Ministry of Health and Family Welfare, Government of India, was used15. Causality assessment was carried out using the Naranjo algorithm16.

Presence or absence of comorbidity: Diseases other than primary diagnosis were considered comorbidity.

Expenditure incurred on medicines: The following were analyzed:

1. Total expenditure on prescribed medicine for the entire duration of admission was calculated as follows:

   Total expenditure= total units of medicine prescribed×unit cost of that medicine

2. The per capita cost of medicines: (unit cost of medicine 1×number of units of medicine consumed)+(unit cost of medicine 2×number of units of medicine consumed)+………+(unit cost of medicine (n)×number of units of medicine(n) consumed) for each patient (where n is the number of individual medicines prescribed to each patient) divided by the number of patients (1000).

   The per capita expenditure was compared for inpatients: (a) With comorbidity versus inpatients with no comorbidity, (b) between 60 and 80 yr versus inpatients over 80 yr, (c) with hospital stay of less than seven days versus inpatients who stayed for equal to or more than seven days, (d) who were clinically treated versus those who died/left against medical advice (LAMA) and (e) with ADR versus inpatients with no ADR.

3. Average cost-effectiveness ratio (ACER)17 was calculated as follows:

   ACER=Total expenditure/number of successfully managed inpatients

The total expenditure for all inpatients who were successfully managed or who LAMA or who died in a disease category was calculated. Successfully managed inpatients are those who are labelled as recovered/recovering or improved at the time of discharge by the treating physician. An activity-based costing (ABC) analysis of the expenditure on consumed medicine was carried out18.

The cost of the medicines was obtained from the hospital purchase list available in the hospital supply, and for medicines not available in the hospital, the cost was obtained from the current edition of Drugs Today (2017)19 a commercial medicine compendium that has the prices of the medicines available in the market. The hospital purchases medicines which are on the hospital’s essential medicines list. The inpatient’s expenditure was verified by bills if available or the brand of medicine purchased by the individual was asked and the price of same was assessed using the compendium.

Statistical analysis: The data were entered in MS Excel version 2010 and analyzed using the statistical software SPSS (IBM Corp., Chicago, Illinois, USA). The graphs were created and analyzed using the GraphPad Prism (7.0d) software (Dotmatics, California, USA). The continuous data are presented as mean±standard deviation (SD) and categorical data are presented in percentages. For economic assessment non-parametric tests, the Wilcoxon rank-sum test was used for intragroup analysis and the Mann–Whitney U test was used for intergroup comparison. The Kruskal–Wallis test with Dunn post hoc analysis was used for multiple group comparison of non-parametric data among multiple groups. P<0.05 was considered significant at a confidence interval (CI) of 95 per cent.

Results

Geriatric inpatients comprised 41.3 per cent (1586 of a total of 3840) of the total individuals admitted to the ward during the study period. The mean age (±SD) of the patients was 65.6±7.4 yr. There were more male individuals (56.4%). Majority of them belonged to urban areas (74.1%) and were from Delhi (66.3%). Most were married (84.1%) and stayed in joint families (63.8%) (Table I).

The inpatients were admitted to the hospital for durations ranging from 1 to 18 days (average: 6.8 days); 29.1 per cent of the individuals were admitted for a week. The inpatients were diagnosed with a total of 62 different disorders as per the ICD-10 classification. 74.7 per cent had associated comorbidity ranging from one to six diseases (mean±SD) : 2.2±0.98). Cardiovascular diseases (CVD) were most common, with almost one in four individuals presenting with a CVD (319), followed by pulmonary diseases (175) and other miscellaneous diseases (164) (Fig. 1).

Close

Fig. 1

Morbidity profile of geriatric inpatients.

Export to PPT

Prescribing pattern and consumption of medicines: A total of 8366 medicines were prescribed in 127 formulations. Maximum number of medicines were prescribed as oral formulations (54) followed by parenteral route (51) (Table II).

Individuals were prescribed medicines ranging from 3 to 16 in number (mean±SD: 8.3±2.33). Nearly 30 per cent of the inpatients received 10 or more medicines per prescription. Individuals with respiratory disorders had the highest number of medicines prescribed per patient (9.2 medicines per prescription). Inpatients with genito-urinary diseases had the highest duration of stay (7.4 days per patient) and the highest number of diseases per inpatient (2.8 diseases per patient).

Pantoprazole under the brand name of Pantocid (923), ondansetron under the brand name of Emset (719) and ceftriaxone under the brand name of Monocef (625) were the most prescribed medicines. A total of 65,536 DDDs of medicines were consumed. Ondansetron constituted the highest consumed medicine in terms of DDDs, followed by dexamethasone and pantoprazole (Supplementary Table I).

Appropriateness of medicines: As per the Beers criteria of PIM use group one, 281 (28.1%) prescriptions were found to be inappropriate. Twenty-nine prescriptions (10.3%) had two or more inappropriate medications prescribed. Five medicines [metoclopramide (134), insulin sliding scale (55), alprazolam (53), digoxin (38) and nitrofurantoin (4)] accounted for all the PIM. Individuals with PIM had a longer stay in the hospital and more number of medicines prescribed compared to those without it (P<0.001) (Supplementary Table II).

Adverse drug reactions: ADRs were observed in 13.9 per cent of the inpatients. Fifty-three medicines were responsible for 140 ADRs in the study. Maximum number of ADRs were attributed to ceftriaxone (8) and pantoprazole (8), followed by aspirin and ondansetron. Nausea (14) was the most common ADR, followed by bleeding (13), headache (11), constipation (7) and vertigo (7). As per Naranjo’s causality assessment, 94 (67.2%) ADRs were in probable and 46 (32.86%) were in possible category. Individuals reporting an ADR had a longer stay in the hospital (6.84±2.03 vs. 6.78±2.04 days, P<0.0001) and had more number of medicines prescribed (8.88±2.7 vs. 8.28±2.3, P <0.001) (Supplementary Table III).

Expenditure on medicines: The total expenditure on prescribed medicines was INR 1,087,175 with a per capita expenditure of INR 1087.17 (Table III). Parenteral medicines accounted for 91 per cent of expenditure on medicines. Maximum expenditure was incurred on intravenous mannitol and tazobactam–piperacillin combination. Expenditure on antimicrobials was 26.6 per cent of the total expenditure. Intravenous (IV) tazobactam–piperacillin and IV ceftriaxone contributed to almost 70 per cent of the antimicrobial expenditure (Table III). The average per capita expenditure (INR 1800.55) and the ACER were highest for individuals with gastrointestinal disorders (Table IV).

The per capita expenditure was significantly higher in individuals with comorbidities (P=0.0327) and those who had a longer duration of hospital stay (P<0.0001) (Table V). The average per capita expenditure in individuals with ADR was higher, although not significant (P=0.1028). Average per capita expenditure correlated positively with number of diseases (r=0.1001), duration of hospital stay (r=0.3107) and number of medicines prescribed (r=0.3369) (Fig. 2). Majority of the medicines were supplied by the hospital free of cost. Only seven (6.4%) medicines contributed towards the out-of-pocket (OOP) expenditure, which was 5.85 per cent of total expenditure (Fig. 3).

Close

Fig. 2

Correlation of per capita expenditure on prescribed medicines with various parameters: (A) number of disease, (B) duration of stay, (C) number of medicines prescribed.

Export to PPT

Close

Fig. 3

Medicines contributing to out of pocket (OOP) expenditure.

Export to PPT

The ABC analysis showed that 13 medicines contributed to 70 per cent of the total expenditure (Category A). All these 13 medicines were parenteral medicines. Eighty medicines contributed to only 10 per cent of the expenditure (Category C) (Table VI).

Discussion

Cost of healthcare, including that of medicines, across the world is increasing2021. This assumes greater significance for the elderly. In the developing world, public healthcare systems play a vital role in providing low-cost, accessible healthcare to the elderly. However, to ensure this, national governments spend approximately three times more on the elderly than the general population22. Medicines constitute a very important component of overall healthcare costs and are responsible for a very high OOP expenditure232425. These high costs are often associated with decreased compliance and premature termination of therapies26. Specific policies to take care costs of the healthcare, including medicines, are needed for the elderly.

In this study, since most of the medicines were provided free by the hospital, the OOP expenditure was fairly low (5.75%) compared to that of some international studies where OOP expenditure on prescription medicines was higher (18%)23. A study carried out in Korea showed that one in 10 elderly people spent more than 10 per cent of their income on medications, and the probability of having an expenditure burden amongst elderly persons was 3.8 times as high as that amongst the non-elderly27. The Government of Delhi has a drug policy based on the essential medicines concept28. In this, the medicines fulfilling the healthcare needs of majority of the patients are included in the hospital essential medicines list and are provided free of cost to the participants29. Such a policy improves access to essential medicines for patients. This is of great importance in India, where 70 per cent of healthcare expenditure is OOP of which 70 per cent is on medicines930.

In this study, the per capita expenditure on medicines was INR 1087.17. Individuals with gastrointestinal disorders had the highest per capita expenditure on medicines, followed by those with genito-urinary disorders. This could be because of a greater number of associated comorbidity, longer duration of hospital stay and a high number of medicines prescribed. There were many individuals with upper gastrointestinal tract bleeding for which IV ondansetron, pantoprazole, ceftriaxone along with terlipressin and octreotide were prescribed. Most of these medicines are expensive and were not included in the hospital list of essential medicines. They had to be bought from outside at market prices.

The expenditure incurred on parenteral medicines (91%) contributed to majority of the expenditure. A high load of cerebrovascular accident patients coupled with the high cost of medicines resulted in intravenous mannitol contributing to the highest expenditure in the study. The use of broad-spectrum antimicrobials as prophylaxis against hospital-acquired infections resulted in almost a fourth of the total expenditure on antimicrobial medicines, of which parenteral antimicrobials accounted for 70 per cent. Antimicrobial stewardship programme to assess the appropriateness of antimicrobials prescribed is required31. A high burden of respiratory diseases and diabetes patients leads to significant expenditure on medicines such as inhalational steroids and insulin. Enoxaparin is a costly medicine used in ischaemic cardiovascular events. The high burden of CVDs, with a large number of patients being prescribed enoxaparin (clexane) an expensive drug lead to a total high expenditure on in this study. Prescribing relatively newer efficacious medicines tend to be associated with higher per-unit costs leading to an overall increased expenditure on medicines7.

The cost of therapy in individuals with ADRs was higher, although not significantly. The expenditure increased with comorbidity, number of medicines prescribed and longer duration of hospital stay. Such associations have been previously established32. An emphasis on minimizing the hospital stay and the number of medicines prescribed therefore can serve as a potential strategy for expenditure reduction.

In the elderly, the appropriateness of prescribed medicines assumes great importance because of the risk of ADR and cost considerations. In the current study, five medicines, including metoclopramide, insulin sliding scale, alprazolam, digoxin and nitrofurantoin, accounted for all the PIM. These medicines become inappropriate in the elderly due to manifestations of various adverse effects in geriatric individuals owing to altered pharmacokinetics and pharmacodynamics. The high risk of extrapyramidal side effects makes metoclopramide inappropriate, but its longstanding availability in the market and the resulting experience of healthcare professionals with metoclopramide contribute to its high use33. Risk of hypoglycaemia with insulin, arrhythmia with digoxin, decrease in renal function with nitrofurantoin and sedation with alprazolam make them potentially inappropriate. Such a high level of inappropriate prescribing indicates a lack of awareness about the possible problems associated with such commonly used medications in geriatric populations.

Marks Beer created the Beers PIM criterion in 199134 to optimize medicine selection and minimize ADRs among the elderly. The criteria have been revised multiple times. We used the Beers Criteria 2015 in this study14. The latest criteria is Beers Criteria 201935. This differs from 2015 criteria in Category I medicines in that three more medicines (glimepiride, methscopolamine and pyrilamine) have been added and two removed (ticlopidine and pentazocine) as both are not available in the USA36.

ABC analysis revealed that only 13 medicines contributed to 70 per cent of the total expenditure incurred on medicines. All 13 drugs were parenteral medicines. Since this expenditure is highly localized on a few parenteral medicines, including expensive antimicrobials, any strategy for cost reduction in pharmaceuticals in the hospital should focus on this group first.

As per one study, relative lifetime per capita health spending is the highest for the age group of 65-84 yr (36.5%) as compared to the younger age group of 20-39 yr (12.5%). The higher per capita health expenditure in older ages reveals a positive association between age and poor health status with more health spending37. In India, over the decades, the annual hospitalization rates have been increasing along with their associated expenditure. As the number of elderly persons is also increasing in India, this is projected to increase the per capita health expenditure38.

The current pension scheme covers only 35 per cent of senior citizens and will leave around 61 per cent of elderly population without income security by 205039. India’s elderly population is increasing expeditiously from eight per cent in 2013 to nearly 18.3 per cent by 205040. Two third of this elderly population lives in villages and nearly half of them belong to poor socio-economic status and are dependent on their families41.

The National Health Policy 201742 recognized the healthcare needs of the rural geriatric population and specified that primary healthcare should include geriatric care. The government, both central and states, has started social insurance schemes and government-based voluntary insurance schemes. The world’s largest social insurance launched by the Government of India as part of Ayushman Bharat is Pradhan Mantri Jan Arogya Pariyojana in 201843. In this scheme, all families belonging to the poorest, lowest 40 per cent of the population are eligible to get a benefit of up to INR five lakhs each family per year for secondary and tertiary care. Till 2019, 17.96 lakh beneficiaries had availed benefits of the scheme since inception of the same. Ayushman Bharat replaced two prior schemes: the centrally funded Rashtriya Swasthya Bima Yojana and the Senior Citizens Health Insurance Scheme (2016). The disaggregated data about these and other social insurance schemes, especially with regard to the geriatric population, are not available40. In the absence of data, it is difficult to comment on the impact of the benefits of these schemes on the geriatric population. What is available is that the households’ share, including insurance contributions, constitutes 71 per cent (INR 320,262 crore) of the Current Health Expenditure (CHE) share. Moreover, the total pharmaceutical expenditure is 37.9 per cent of CHE40.

In view of the above, a multi-pronged strategy is required to take care of the health and, specifically, medicine needs of geriatric patients. Financial aid through enhanced pension schemes, social insurance schemes, access to low cost, quality essential medicines is one such multi component approach.

A drug policy specifically focussing on geriatric pharmacotherapy would take care of the medicine needs as well as costs for the rising number of geriatric individuals. The following are recommended: (i) a drug policy with special considerations for geriatric individuals be formulated; (ii) healthcare providers be made aware of rational prescribing for the elderly; (iii) a system for monitoring drug use with an economic assessment of the same be put in place in the public health systems; (iv) pharmacovigilance for monitoring of ADRs in the elderly.

The present cross sectional investigation had several limitations as we took into consideration only the direct cost of prescribed medicines for calculating expenditure. Moreover, the study was in only one public hospital. The generalizability of the study results is therefore limited. Furthermore, we used only Beers Criteria for assessing PIM prescribed.

Comorbidity, prolonged hospital stays and prescribing of drugs, specifically injectables, contribute to increased expenditure on medicines in geriatric individuals. Polypharmacy and prolonged hospital stay are associated with the prescription of PIMs and ADRs. While studies with a larger sample size with inclusion of more and different types of health facilities would generate evidence with wider relevance. The observations of this study are able to recommend a specific drug policy for the elderly to improve the utilization of expenditure on medicines.