Introduction:

Poly-pharmacy refers to the use of multiple medicines by the patient. It is associated with a significant morbidity, mortality, and disability especially in elderly population^{1, 2}. It has become a norm in current medical practice due to  the advent of challenges posed by chronic diseases, rise in elderly population, and availability of preventive medicines². Poly-pharmacy can be sometimes deem appropriate. However, there is a strong association with morbidity and mortality³. It is considered as a significant issue which continues to draw the attention of clinicians and policy and guideline developers⁴. 

Poly-pharmacy has been identified to impose negative health effects that contribute to potential side effects of a drug, drug-drug interaction, reduced patient compliance, and poor drug adherence leading to cognitive impairment and impaired balance resulting in falls^{5, 6}. Moreover, it contributes to drug adherence which involves numerous clinical concerns that may cause confusion⁷. The risk of an adverse drug effect is 13% with two medicines, while it increases to 58% with multiple (poly-pharmacy) prescription⁸. Similarly, the incidence of adverse drug interactions (ADIs) also increases⁹. Medication errors are a common observation in all areas of the health care system¹⁰.

Prescriptions in the ED are usually processed in a rush which have the potential to cause medication errors, incorrect medication dose, and failure to detect patient allergy records¹¹. The drug delivery sequence begins from receiving prescription followed by transcription, dispensing from a registered pharmacist, administration and monitoring by nursing staff in the end¹². Certain drugs are considered potentially inappropriate in older patients not only because of the higher risk of intolerance related to pharmacokinetics, pharmacodynamics or drug–disease interactions but also because they are prescribed at too high doses or for an extended time period¹³. It is vital to identify patients with inappropriate poly-pharmacy which has a potential to cause adversity and poor health outcomes².

A recent study on the incidence of poly-pharmacy in Malaysia reported 12.5% cases which is an alarmingly high level in terms of medication regimen¹⁴. Another study from India reported  that poly-pharmacy is common in more than 50% of the prescription having three or more medications¹⁵. Although a number of epidemiological studies investigating the role of poly-pharmacy among patients presenting to different hospital has been almost exclusively conducted for the geriatric populations, they have  been limited to  the developed countries.

There is little evidence of poly-pharmacy in Asian countries including Pakistan that caters to a large portion of patients in the EDs. The aim of this study was to understand the current practice of prescribing medication in ED in a low-income setting and to determine the factors associated with poly-pharmacy.

Methodology:

The present research was a retrospective study conducted at Aga Khan University Hospital (AKUH) Karachi, Pakistan, a 700 plus bedded, Joint Commission International (JCI) accredited tertiary care center providing health care facilities.

Study Design and Population:

 The required data of this study was collected from patients who visited the ED of Aga Khan University Hospital, Karachi (AKUH) during January to December 2012. Using a simple random sampling method, the detailed clinical records were retrieved out of 51,000 patients who visited ED during that period.  the prescription from admission till discharge of all the patients were reviewed through physician order entry (POE). The data was extracted and filtered for 5 or more medications. Other required data which included patient’s demographic information, reason for visit, vitals, clinical presentation, medication information (dosage and duration of treatment), triage category, ED length of stay and their disposition (admitted, discharge, LAMA or expired) were extracted from medical records as well.  

A representative sample of 267 was employed in this study 50% of whom used poly-pharmacy as a prescribing norm for the patient (adult and pediatric). The study assumed a confidence level of 95% and was bound on 5% of error. We further inflated sample size to 35% in order to handle the missing data in multiple variables. So the final sample size was increased to 370.

Patients with Poly-pharmacy:

In this study, we used the definition of poly-pharmacy proposed by Sabzwari who defined the term as the concurrent use of five or more different medications¹⁶. For each patient, the previous and current medications may collectively be five or more which will be considered as poly-pharmacy. Previous medications are those medicines that the patient was prescribed earlier before coming to the ED. While current medicines are those that are prescribed in the ED during present visit. Topical preparations are excluded from the defined poly-pharmacy criteria. For instance, ointments, ear/eye drops, and enemas, since they are non-systemic drugs that act within the intestinal lumen without reaching systemic circulation. Also, the data of patients with inaccurate or illegible medication information in the clinical records was excluded from the current study.    

Medication Classification by Anatomical Therapeutic Chemical (ATC):

Drugs involved in the poly-pharmacy criteria were coded into various drug classes according to Anatomical Chemical (ATC) classification based on WHO-ATC Index 2005¹⁷. The ATC classification system has divided all drugs into several groups according to their specific action on different organs or systems and their (chemical, pharmacological and therapeutic) properties.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 

Analysis

Descriptive statistics were used to describe the patient’s demographics. Chi-square (χ2) was used for categorical data to see the association between groups and t-test for continuous data. Moreover, predictors like Gender, Area, Age, Weight, Height and vital signs etc. were used to see their relationship with the dichotomous outcome that is the use of poly-pharmacy.

Results:

The data collected from a sample of 372 patients were analyzed meticulously. Male and females were equally distributed (n= 193, 52% versus n= 179, 48%) and the mean age of the participants was (42.1 ± 25.2) years. Triaging of the patients was done using Emergency Severity Index (ESI) which is a five-level ED triage algorithm. It provides clinically relevant stratification of patients into five groups from 1 (most urgent) to 5 (least urgent) on the basis of acuity and resource needs (18). Most of the patients came under P1 category (n= 100, 40.2%) followed closely by P3 (n= 95, 38.2%) and the rest in P2 and P4 (n= 46, 18.5% and n=8, 3.2%), respectively. See Table 1.

The medicines prescribed to the patients were between five to seven or more, in number (n= 251, 70%). The mean hospital stay was 9.71 ± 6.94 hours and the majority stayed greater than six hours in the hospital (n= 123, 59% versus n=85, 41%). Most of them were sent home after the treatment from ED (n= 194, 53.6) followed by admission to hospital (n= 99, 27.3%). There was a significant number of patients which was left against medical advice (LAMA) n= 47, 13% and the remaining were expired (n= 14, 3.9%) as depicted in Table 1.

The results indicated that the most frequent diagnosis made in ED was respiratory tract infection (n= 61, 16.4%) followed by gastroenteritis and others (See Table 2). Antibiotics were prescribed in 91.6 % of the cases followed by, Pain killers, and proton pump inhibitors were almost equally distributed among thirty percent of the cases (See figure 1)[MK1] . 5-6 medicines were prescribed in almost half of the patients (n=188, 52.4%) followed by 3-4 medicines n=83, 23.1%) as shown in Table 1.

Table 3 shows the use of poly-pharmacy among patients based on their age, gender, number of complaints, and triage level. Poly-pharmacy found to be common in all groups with no difference between male (n=130, 71%) and female (n=121, 69%). There was a significant difference noted with respect to age groups and triage level. Poly-pharmacy was higher in patients with age group of thirty and above (n= 179, p 0.002) and those who triaged P1-P2 category as (n= 114, p 0.037).

Discussion:

The emergency department (ED) of the tertiary care hospitals are usually overcrowded with sick patients and this problem gets worse when the ED is in a developing country where both human and financial resources are very limited. Similar situation applies to our ED which is a state of art department of tertiary care hospital located in Karachi, Pakistan, a developing country, where patients across the country come to seek medical care. 

In the present research found very limited data for the incidence of poly-pharmacy from Asian region in our study. The degree of poly-pharmacy (69.9%) was near to that reported by Abrams et al [U2] [MK3] in 1998(72.8%)^19. However, the rate of poly-pharmacy was 72% in patients examined by Toshikazu Abe et al. Research shows that excessive poly-pharmacy is associated with worse outcomes related to patient’s condition²¹. 

With regard to gender, we found that there was no significant difference between males and females concerning poly-pharmacy. It means that gender did not significantly affect the presence of poly-pharmacy. 69% of females in our study were found with poly-pharmacy. This finding was in tune with the percentage reported by Toshikazu Abe et al (i.e. 61%)²⁰.

Moreover, when we compared poly-pharmacy with different age groups, we observed that poly-pharmacy was most common in elderly group (31%) as shown in Table 1. This finding was also proven by international studies²². In our study, most patients with poly-pharmacy were elderly >60 years of age²³. The risk of poly-pharmacy causing Drug-Drug Interaction (DDI) are even higher in older population as most of elderly are suffering from multiple diseases and they need to take multiple medicine for treating these diseases. Some of these medicines were prescribed by physicians and some may be taken as over the counter by elderly patients and that increases the risk of DDI²². The results of this study are comparable to those of Baumgartner in which elderly population received the greatest number of medicines²⁴.

As we conducted our study in the ED,  we examined the relationship between presenting complaints and Triage category with poly-pharmacy as well¹⁸. As for the number of complaints, no association was found with poly-pharmacy. But it was found to have a strong association with triage category as patients who are critical and belong to triage category P1-P2 114 (79.2 %) were identified to have a strong association with poly-pharmacy as shown in Table 3. To the best of our knowledge, to date, these important factors were not reported or mentioned in the available literature in this domain. Future studies may require to investigate these factors in more detail.

In our study, three most common ED diagnoses were respiratory tract infection 61(16%) followed by acute gastro enteritis 32 (9%) and acute coronary syndrome (ACS) 27 (7%). The percentage of respiratory cases were almost similar to the study conducted by Krishna Pandey where 14% of cases had respiratory tract infection although in his study the respiratory cases were not the most common disease reported²⁵. The reason for more respiratory cases were probably due to smoking and exposure to air pollution of susceptible individuals. Moreover, acute gastro enteritis 32 (9%) was the second most common diagnosis found in our study, The most likely reason would be an increase in number of GI related diseases due to consumption of contaminated water, unhealthy living life style such as eating junk food and anxiety^{25, 26}. The third significant disease reported in our study was acute coronary syndrome and that was in contrast with Barot PA et al²⁷  and  Krishna Pandey who reported ACS as the number one diagnosis as raised by 21.79% and 26% of the patients in their studies. The reason for that difference is unclear, yet one probable reason might be that  the patients with ACS in our study preferred to go to government-based cardiac tertiary care centers for their care where all care is provided free of charge.

The top three drugs prescribed in our study were antibiotics, pain killers, and PPI which is almost the same as those found in a regional study in India by Balushi K et al. The most likely reason would be that similar complaints in the same socioeconomic setup of patients in both studies where infection is most common reason to visit ER followed by Gastritis/gastroenteritis and body-ache /musculoskeletal pain^{28, 29}.

In this study, it was also found that antibiotics were the most frequently prescribed medicines to patients admitted to ED (91.6%). It was very close to the percentage reported by Karishna Pandey (93.6%)²⁵. The most probable reason behind this is that in ED antibiotics are empirically prescribed to manage the presumed sepsis/infection and to control severity of illness as the diagnosis was not confirmed during initial couple of hours after ED arrival, while the percentage is 30% in developed countries which is much lower in comparison to our study. This may be due to more systematic and protocol driven care and/or less number of infectious cases in ED in developed countries^{30, 31}. This highlights the need to develop some strategies to decrease the unnecessary use of antibiotics which ultimately decreases the prevalence of antibiotics resistance in developing countries which has been alarming over the past decade. To optimize medicine use in ED, it is necessary to implement antibiotic stewardship to avoid the irrational use of broad-spectrum antibiotics which ultimately affects the cost of therapy.

Our study reported that 8.6% of patients visited ED due to GI related problems like Gastritis /Gastroenteritis but interestingly 31% of patients received PPI in ED. This may be because beside gastritis, PPI was given as prophylaxis against peptic ulcer diseases in patients who were on NG feeding or receiving NSAID or Aspirin or it may be used for the patients who did not need such cares^{25, 26, 31}.

Pain killers were among the top three prescribed medicines in our ED. a significant number of the patients in our study were elderly, and we knew that most of the elderly do suffer from chronic pain and have habits to take multiple pain killers including NSAIDS at home to control their symptoms. This is quite possible that many of them either visits ED due to onset of new pain or an exaggeration of chronic pain as their primary complaints; they may have also come to ED with some other reasons with body aches as secondary complaint and received analgesia during their ED visits^{28, 32}. Another important reason due to which the patients receive analgesia in ED is injury and trauma such as road traffic accidents, falls, animal bite etc.³³. As per the US[U4] [MK5] A data, injuries and trauma are top most reasons to visits ED. Although  this number has dropped in the last few years, a significant percentage of cases are elderly who have taken multiple comorbid and and multiple medications at home. This approves that providing safe and effective analgesia at times is not an easy task^{34, 35}.

The majority of the patients with poly-pharmacy in our study stayed in ED for more than 6 hours but further analysis did not find any correlation between higher numbers of drugs with LOS. This finding is in contrast with that of Rakesh Patidar where LOS was directly related to poly-pharmacy^{28, 31}.

Our study highlights the need to pay due attention to the implementation of poly-pharmacy by medical practitioners. Physicians need continuous education and knowledge about the drug, its dosing, interactions, and side effects. They should check already existing medication before prescribing new ones. Evaluation of patient’s compliance and adherence with the therapy is also one of the important considerations.

The limitations of this study were that it was retrospective chart review and patients with inaccurate or illegible medication information in the clinical records were excluded from analysis. Also, the data gathered from those patients who were shifted from emergency to other wards/ICU/CICU were  excluded.

Conclusion:

Nowadays, poly-pharmacy is a widespread problem all around the world. The results of this study showed the importance of poly-pharmacy as independent risk factor for adverse health outcomes after an ED visit. Health care systems are now becoming more complex and it is very important to have strong coordination with physicians, pharmacists, and patients to ensure more patient-centered approach.

The result of this study highlights the importance of strategies that must be implemented to optimize medication use at the EDs. In the current climate of ED overcrowding, boarding of admitted patients in ED and more elderly patients with multiple comorbidities, the need of pharmacist in ED would be of potential benefit to the process of identifying drug interactions.

There is a lack of data in the region regarding poly-pharmacy in Emergency patients. Our study highlighted that poly-pharmacy is the subject of interest among all ED patients. Moreover, further studies are needed to disseminate the findings of the problems associated with poly-pharmacy.

Abbreviations: