Chronic illnesses and polypharmacy in elderly patients: A hospital-based study in Northern Sri Lanka

Population aging and multiple pathology among older people have increased medicines use in elderly. Aim of this study was to evaluate chronic illnesses and polypharmacy and to determine factors associated with polypharmacy among hospitalized elderly at T eaching Hospital-Jaffna. This was a cross sectional, descriptive study. Elderly and polypharmacy were defined as ≥65years and concomitant use of ≥6 medications for ≥1 month respectively. Data were collected from bedhead-tickets of 288 elderly patients using data extraction sheet. Chi-squared and paired t-test were performed to determine the level of significance and unadjusted odd ratios were calculated to determine the association between polypharmacy and common chronic illnesses. A p value <0.05 was set as statistically significance. Mean age was 72 (SD ±6.2) and majority were males (53.5%). Relatively higher proportion, 58.7% (n=169) of patients had polypharmacy. Seven chronic illnesses were prevalent in >10% of the patients. Lipid-modifying (80.9%) and antithombotic agents (76.4%) were the top two subgroups prescribed to elderly. Polypharmacy was substantially increased with increasing number of chronic illnesses (p<0.05). Significant association ( p <0.05) was found between polypharmacy and coronary heart disease, anaemia, heart failure, asthma and diabetes mellitus. Significantly greater number of medications were prescribed on discharge compared to admission ( p <0.05). This study concluded that polypharmacy was common among hospitalized elderly and increases with hospitalization. There was increased tendency to prescribe antithrombotic and lipid-modifying agents. These findings indicate the need for medication reconciliation and review in elderly patient. Further studies targeting wider population are needed to determine the appropriateness of polypharmacy in elderly.


Introduction
Sri Lanka is one of the fastest aging countries (1). According to the 2012 census, 7.84% of the population of Sri Lanka is 65 years and above (2).
One of the major challenges of population aging is chronic non-communicable diseases which commonly affect adults and elderly. Rapidly growing aging population will increase the demands of health care system and there is a need to address the health issues of older people. Because of multiple co-morbidities, elderly suffer from more symptoms and illnesses than the young which contribute to polypharmacy (3,4). Increasing tendency to use more than one medication for optimal treatment of a single condition causes further increase in the medicines use in elderly (5).
Polypharmacy is the use of multiple medications and is one of the major and common health-related issues of elderly patients (6,7). The definition of Chronic illnesses and polypharmacy in elderly patients: A hospital-based study in Northern Sri Lanka 1 Thiyahiny S N, 1 Kumanan T, 1 Surenthirakumaran R 1 Faculty of Medicine, University of Jaffna polypharmacy varies according to the population and study setting (8). Though there is no universally accepted formal definition for polypharmacy, it generally refers to the simultaneous use of multiple medications (more than 4-6 medications) (9,10,11).
Although elderly patients have the greatest potential to benefit from multiple drug therapy, they are more susceptible to the adverse consequences of polypharmacy than younger people because of age-related physiological changes, decline in organ functions and various pathological changes which affect the pharmacokinetics and pharmacodynamics of drugs (12,13,14).
Like many developing countries geriatric health needs are emerging problem in Sri Lanka. This study was carried out with the aim of evaluating the common chronic illnesses and polypharmacy and to determine the factors associated with polypharmacy among elderly patients treated in the medical wards of Teaching Hospital-Jaffna.

Case definitions
In this study polypharmacy was defined as simultaneous use of six or more medications for the duration of one month or more. Chronic illness that were prevalent in 10% or more of the study population and medications prescribed to 10% or more of the study population were considered as common chronic illnesses and commonly prescribed medications respectively. Medications are classified at level 3 of the Anatomical Therapeutic Chemical (ATC) classification (15).

Data analysis
Analysis of medications was done with medications prescribed on discharge except the analysis between admission and discharge where medications prescribed on admission and discharge were taken for comparison. Descriptive statistics such as frequencies, percentage mean, median and range were used to present the results. Chi-square test was used to determine the association between polypharmacy and age, sex and number of illness.
Paired t-test was used to determine the level of significance between the number of medications prescribed on admission and on discharge and unadjusted odd ratio (OR) was used to identify the common chronic illnesses that were associated with polypharmacy. Data were analyzed using 95% confidence interval (CI) and a p value less the 0.05 was considered significant.

Results
Data from 288 elderly patients were anaylsed. The mean age of the study population was 72 years (SD±6.2), ranging from 65 to 90 years and males were slightly higher (53.5%) than females (46.5%). Among the 48 chronic illnesses that were prevalent in 288 elderly patients, 7 occurred in more than 10% of patients. Hypertension was the most prevalent chronic illness, seen in 53.1% patients followed by coronary heart disease (48.6%) diabetes mellitus (41.7%), asthma (22.6%), cerebrovascular disease (12.8%), anaemia (11.8%) and heart failure (11.5%). Peptic ulcer disease, chronic kidney diseases, dyslipidaemia, valvular heart illness, hypothyroidism, osteoarthritis and chronic obstructive pulmonary disease were observed in more than 5% of the patients. On average, all the patients with common chronic illnesses had three or more chronic illnesses and prescribed with more than 5 medications (ranging from 5.4 to 7.4). Table 2 shows the distribution of common chronic illnesses among different age groups and sex and Table 3 shows the distribution of comorbidities  Out of 57 pharmacological subgroups (level 3 ATC classification) of medications prescribed for the elderly, 20 subgroups were prescribed to more than 10% of the patients (Table 4).       Distribution of polypharmacy among common chronic illnesses was shown in Table 6. Mean number of medications prescribed on admission and discharge were 4.3 (range 0 -14) and 6.0 (range 1-12) respectively (Table 7). prescribed on discharge (p < 0.001).

Discussion
The three main findings of this cross-sectional study among hospitalized elderly in medical wards of Teaching Hospital-Jaffna are; 1) polypharmacy was common among hospitalized elderly and increases with hospitalization; 2) significant association between polypharmacy and number and types of chronic illnesses was observed; 3) there was a tendency to prescribe antithrombotic and lipid-modifying agents.
Prevalence of polypharmacy varies widely because of the differences in defining polypharmacy and criteria for patient selection. A review on multiple diseases and polypharmacy in the elderly reported that the prevalence of polypharmacy among hospitalized elderly varied from 20% to 60%. (16) In the present study prevalence of polypharmacy (59%) was at the higher end of the above range.

Mean number of medications prescribed for
elderly in the present study was 6.03 and similar finding was reported in a study conducted among elderly in an outpatient setting in United States (11). A higher mean number of medications (8.7) was reported in a study conducted among community-dwelling elderly in Australia (17).
Findings of this study showed that significantly greater number of medications were prescribed on discharge, compared to admission (mean number of medications on admission vs on discharge = 4.4 vs 6.03). Similar findings were reported in studies conducted among hospitalized elderly in Australia and Italy (18,19).
Unlike most of the other studies, the present study did not show significant association between polypharmacy and age or sex (11,17,20). Two studies that were conducted in primary care elderly patients in Germany and hospitalized elderly patients in Italy support the above findings of this study (19,21).
The present study found that there was significant association between number and types (coronary heart disease, diabetes mellitus, asthma, anaemia and heart failure) of chronic illnesses and previous studies that investigated the factors associated with polypharmacy supported these findings (20)(21)(22) (6,10,17,25).
A tendency to prescribe antithrombotic and lipidmodifying agents was observed in the present study.
Proportion of elderly patients prescribed with lipidmodifying (81%) and antithrombotic (76%) agents in this study was higher when compared to other studies (37-64% and 50-70% respectively) (11,25). In this study many patients (25% of the study population) received more than one antithrombotic agent. Among this study population 82% had cardiovascular disease (one or more) and this could be the reason for higher prescription rates for lipidmodifying and antithrombotic agents. However, further evaluation in this regard is needed to comment on the appropriateness of this prescribing tendency.

Limitations
This study was conducted in hospitalized