In the current study, almost two-thirds (59.3%) of the HF patients had at least one ADR during their hospital stay. The current prevalence is comparable with 53.2%  and 67%  in India and 69% in Indonesia .
However, our finding of prevalence among hospitalized HF patients is higher than in previous studies done in high and middle-income countries including 8.6% in Italy , 7.74% in UAE , and 24.2% in Iran . The current higher prevalence may be attributed to our prospective study design and ADR assessment method. Accordingly, we detected ADRs at enrollment, reviewed patient records, interviewed patients, did the physical examination, and followed up until they were discharged, which provided ample time to detect the ADRs. In contrast, some of the former studies  were done observationally and some were solely based on patient records, retrospectively . In addition, the lack of active pharmacovigilance in our setting based on a study done in 2018  has contributed to the increased prevalence of ADRs, As the reported ADRs help to avoid known risk factors and monitor patients. Moreover, more than half (59.3%) of this study's patients were either elderly or pediatric, who are known for being vulnerable to ADRs. Also, the majority (78.8%), of our study participants had at least one comorbid condition that predisposes them to receive multiple medications and leads to ADRs. The incidence of new ADR was 106 ADRs/ 1000 person-days. If 1000 HF patients were followed up in a day, on average 106 (10.6%) would incur an ADR.
The Naranjo causality scale rated over two-thirds (71.9%) of the ADRs as probable. The probable causality was considerably higher than those in other studies including, 58.3% in UAE , 56.7% & 18.4% in India [21, 37], and 33.3% in Iran . Additionally, Lupitaningrum et al reported 41.9% of probable ADRs in Indonesia among hospitalized HF patients . The deviation with some of the studies could possibly be because of the method used to assess the causality, which was the WHO classification of causality [37, 40]. In the present study, however, daily monitoring of patients, laboratory investigation review, and physical assessment added an objective proof for the ADRs and increased the probability of ADRs.
GI system was the most frequently (27.5%) affected system by ADRs, followed by the nervous system (19.5%), Endocrine & metabolic systems (18.3%). These results are in line with previous findings that had shown the GI system as the commonly affected system among hospitalized HF patients [38, 39, 42]. The present study revealed that electrolyte imbalance, majorly hyponatremia (12.8%), is the most commonest specific ADR followed by hypotension and dizziness. Electrolyte imbalance was also noted as one of the frequent ADRs among hospitalized HF patients by Catananti et al . In our study, hypotension (11.5%) was mainly caused by the combination of anti-hypertensive agents and IV diuretics. This was in line with previous studies [21, 37, 43].
Over half (52.4%) of the ADRs were mild, while 40.2% and 7.3% were moderate and severe, respectively. The proportion of severe ADRs in this study was comparable with previous studies of 10.9% & 13.5% in India [21, 37], 9.1% in UAE  and 4.9% in Italy .
Additionally, this study revealed that over half (58.5%) of the ADRs were preventable: 50% were probably preventable, whereas 8.5% were definitely preventable. The preventable nature of the ADRs calls for attention by the health care team involved in prescribing and following up with patients. It accounts for demonstrating prevention strategies among at-risk patients for ADRs. Patients at high risk for ADR, including patients with comorbid conditions, who are on poly-pharmacy, elderly, and pediatrics, need special attention while prescribing, monitoring, and assessing them. It is fact that preventable ADRs are a significant burden to health care among hospitalized patients . The proportion of preventability in the current study is comparable with previous studies done among hospitalized HF patients, in which preventability was 65.9%  and 40% .
Almost three-quarters (72.5%) of the ADRs were caused by cardiovascular drugs. Furosemide, a diuretic agent, alone contributed to almost half (48%) of the ADRs caused by cardiovascular drugs. Previous findings showed that specific drug commonly implicated in ADRs among hospitalized HF patients was digoxin & furosemide , and Bisoprolol [38, 43].
The identification of associated factors for ADRs helps to identify the most susceptible patients who require close monitoring of drug therapy . Our study showed that ADR was less likely associated among younger adults (19–59 years old) by 85%, compared to the age group less than 18 (pediatrics). This finding is supported by a number of studies that revealed that age being very old or very young compared to younger adults, were susceptible to ADRs [21, 37, 46]. This may be is explained by the fact that there is a pharmacokinetic and pharmacodynamics change among elderly and pediatric populations .
The present study reported that patients who were taking more than 5 medicines during hospital stay were almost nine times more likely to experience ADRs. Former studies done among hospitalized HF patients [10, 21, 43] were in agreement with the current finding that poly-pharmacy was a significant risk factor and every additional treatment had an increased risk of increasing ADR by 8.6% − 9% [48, 49]. Poly-pharmacy has been associated with an increased risk of drug interaction and ADRs [10, 50]. This calls for attention to the use of medications among HF patients by providing only the necessary medications and avoiding the overuse of multiple medications, which might lead to ADRs. Patients who had used herbal drugs in the four weeks before admission were as well 3 times more likely to experience an ADR than those who did not. The safety and effectiveness of herbal drug use are not yet assured and the interaction with conventional medicine leads to ADRs .
Patients who had a significant drug-drug interaction (DDI) had six times the odds of ADRs as revealed in former studies that DDI was significantly associated with ADRs [50, 52]. The synergistic effect of the DDI may attribute to ADRs. Clinical pharmacists have a significant role in detecting and preventing DDI-related ADRs. Studies have reported a decreased occurrence of ADR and drug-related problems as clinical pharmacists are engaged in medication review as part of the multidisciplinary team for optimizing patient safety [53, 54].
Strength and limitation
This study provided detailed information on the types of ADR and associated factors among HF patients. There is a paucity of literatures done in this research area in LMIC. Hence, the present study will help to fill the unprecedented evidence gap in our setup and to come up with a solution for the problems in the future perspective.
The prospective nature of the study enabled us to gather complete information daily and assess and record the problem directly from the patients. Moreover, the data was collected by the PI (clinical pharmacy Masters Student), with the assistance of resident physicians, which increases the quality and accuracy of the data.
Despite the strength, our study had some limitations. For one, the study was single centered and conducted in a hospital serving referred patients who have severe illnesses and more comorbidities, which makes the finding slightly difficult to generalize to a larger population. In addition, objective measurements (laboratory investigation results) had a great impact in showing the disease progress, response to treatment, and ADRs caused by the initiated drug. Clinical examination and patient reports were mainly used as a method used to identify suspected ADRs.