Recall Outcomes of Aspirin in Saudi Arabia: An Observational Study

Globally, drugs are often withdrawn from the market when issues are identified by manufacturers or regulatory authorities. Approximately, 4,500 recalls of drugs and devices have been recorded yearly by US FDA [1]. In Saudi Arabia, records of drug recalls since 2010 have shown 60% increase by 2018 [2]. Ideally, physicians switch treatment prescription of patients using the recalled drug to alternative medications whether generic agents or other therapeutic groups used for the same treatment indication of the recalled drug [3, 4]. The timeline of switching medications is critical in cases of serious conditions where treatment interruption, such as in antiplatelet therapy, subjects high-risk patients to life-threatening complications.

Acetylsalicylic acid (ASA), is an antiplatelet cyclooxygenase inhibitor medication used in low doses to prevent blood clot formation in patients at risk of coagulation [5]. ASA is one of the most commonly used drugs worldwide [6]. It is estimated that 40% of adults over 50 years of age in the USA are prescribed ASA [7]. Moreover, ASA is enlisted on the World Health Organization’s List of Essential Medicines [8].

Discontinuation of antiplatelet therapy, such as ASA, allows progressive reactivation of platelet function leading to increased rate of blood clot formation [9, 10]. ASA discontinuation is associated with significantly increased risk of major adverse cardiac events (MACEs) up to three-folds higher than with its continued use. In patients with perioperative aspirin withhold, the cardiovascular events may occur after ASA interruption within an average period of 8 days [11]. In fact, cardiovascular events have been reported after only 5 days of ASA withdrawal [11].

In January 2019, a certain generic of ASA was recalled from the Saudi market by the marketing authorization holder due to quality issues [12]. Recall of critical medications is often incidental and varies in recall specifications such as affected batch and type of product defect. Apparently, there is a scarcity in the literature describing the impact of similar drug recalls on patient management. Therefore, the aim of this study was to describe management characteristics following ASA generic recall in terms of alternative therapy used, medication switching timeline and patient health outcomes.

This is a retrospective cohort study conducted among patients who were using the ASA generic (recalled on 21st January 2019) during the pre-recall period (21st October 2018 to 20th January 2019). Antiplatelet medication use was assessed after the drug recall to explore therapy continuation status among patients. Data on medication use, patients’ diagnoses, and mortality were based on patients’ electronic health records available from Real-world Evidence Research Network at Saudi Food and Drug Authority [13]. This electronic database consists of in-patient and out-patient health information provided from a multicenter and multi-region healthcare facility.

The study timeline (21st Oct 2018 to 1st May 2019) was based on the recall announcement date which was 21st January 2019. In this study, we considered a 1-week period following the recall announcement date as a washout period to ensure full execution of drug recall by healthcare institution. Accordingly, all healthcare providers would have not been allowed to prescribe the recalled medication, and patients would have been instructed through healthcare providers and/or SFDA announcements to discontinue using it. Therefore, the pre- and post-recall periods are defined as the following:

• Pre-recall period: 3 months prior to the recall announcement date (21st October 2018 to 20th January 2019).

• Recall period: 1-week period following the recall announcement date as a washout period to ensure full execution of drug recall by healthcare centers (22nd January to 28th January 2019).

• Post-recall period: 3 months following the recall announcement date (29th January to 1st May 2019).

Three-month duration for each pre- and post-recall periods was considered sufficient to measure health outcomes, while not too far from recall announcement that may affect the extent of recall attribution to measured outcomes.

Patients who were prescribed ASA during the pre-recall period (21st October 2018 to 21st January 2019) were included. Then we excluded those who did have follow-up data in the post-recall period.

Patients were followed up in the post-recall period and divided into three groups based on alternative antiplatelet therapy:

• (a)

  Patients who switched to other ASA (other generics) as a replacement to the recalled drug.

• (b)

  Patients who switched to alternative antiplatelets (clopidogrel, prasugrel, and ticagrelor) as a replacement to ASA.

• (c)

  Patients who did not continue on any of the antiplatelet therapy described above (a) and (b).

This duration is recorded for all patients starting from the latest ASA prescription date prior to the public recall announcement (21st January 2019). The timeline of switching to alternative antiplatelet medication was calculated in days for all patients from the latest prescription date (21st January 2019) plus days of supply to the date of the new alternative prescription.

Outcomes include:

• (1)

  Demographics: (a) age of all individuals that met inclusion criteria and (b) gender (male and female).

• (2)

  Alternative antiplatelet medications used to replace recalled medication and medication switch timeline.

• (3)

  MACEs which include acute myocardial infarction (AMI), stroke, and post-switch MACE occurrence timeline. MACE outcomes were selected in relation to the scope of this study, based on the most recent systematic review on MACE selection in observational studies [14].

We employed descriptive analysis of the identified outcomes. Data are presented as percentages for categorical variables, and as sample midpoint quantification (mean, median, standard deviation, and interquartile range) for continuous variables. As well, cross tabulation of MACE and antiplatelet therapy in the post-recall period is presented. The analysis was conducted using Statistical Package for the Social Sciences (SPSS), version 26.0 (year 2022).

A total of 31,019 patients had a prescription for ASA during the pre-recall period. Of these patients, a total of 20,490 were included in the study based on occurrence of patient follow-up visits during the post-recall period. Patients’ data that were not available in the post-recall period (N = 11,079) are either deaths (n = 504) or no record of follow-up visits in the database (n = 10,575). Illustration of the study sample size and outcomes distribution is shown in Figure 1.

The average age of enrolled patients is 63 years. The study shows gender equal distribution in patients (11,130 [53% male]), see Table 1. Most patients were from the Central region (12,747 [60.9%]), followed by the Eastern region (4006 [19.1%]). Time period values for medication switch and MACE occurrence are abnormally distributed; therefore, the median and interquartile range are used to describe sample midpoint measurement, see Figure 2.

The vast majority of the enrolled patients were switched to another ASA in the post-recall period (20,522 [98%]). Only (410 [1.9%]) of patients were switched to clopidogrel, and (8 [0.04]) patients were not prescribed alternative antiplatelet therapy in the post-recall period. The time period between the last ASA prescription and the alternative antiplatelet therapy prescription dates ranged from 0 day (minimum period) to 93 days (maximum period). The median medication switch period is 31 (±40) days, see Table 2 and Figure 3.

None of the patients had an incident of MACE during the pre-recall period; almost all the enrolled patients were not diagnosed with MACE during the post-recall period (20,843 [99.5%]). Of the recorded MACE in the post-recall period, (95 [0.4%]) were stroke and only (2 [0.01%]) patients were diagnosed with AMI, see Table 3. Most of the patients who were diagnosed with MACE continued on ASA as alternative anti-platelet therapy (switched to ASA other than Aspirin) in the post-recall period (20,441 [98%], see Table 2. Moreover, patients who had MACE incidents included those that were not switched to alternative antiplatelet therapy (8 [0.04%]), as shown in Table 3.

A total of (123 [0.6%]) patients had MACE in post-recall period (9 had MACE before antiplatelet switch, 106 had MACE after antiplatelet switch, 8 had MACE while not prescribed alternative antiplatelet medication). The median time period of MACE occurrence after medication switch is (6.5 [±40]) days. More details are shown in Table 3.

Fortunately, almost all patients were managed during the 3 months following the recall as they were switched to alternative antiplatelet therapy (ASA other than the recalled generic), except for 8 out of 20,490 patients. This may reflect efficient patient follow-up by healthcare centers concerning the drug recall and patients’ adherence to instructions, despite the apparent delayed medication switch period from recalled to alternative drug.

A reassuring outcome of this study was management continuation using the same anti-platelet therapeutic group (another ASA generic), which was achieved in the vast majority of patients (20,522 [98%]). Same-therapeutic group medication switch is in accordance with the literature, which described that same-class therapeutic interchange is often practiced in hospitals to allow medication substitution without compromising patient care [15, 16].

Regarding the lag time period between ASA recall and alternative antiplatelet prescription date, the overall switching time period appears to be delayed considering the clinically expected period for health complications from therapy discontinuation, especially, for high-risk patients. In patients with perioperative aspirin withhold, the average time for cardiovascular events occurrence after aspirin interruption was 8 days [11]. On the other hand, the number of patients who were prescribed alternative antiplatelet medication was highest within the beginning of the pre-recall period and starts decreasing toward the end of the post-recall period (Fig. 2). This study may be considered a real-world demonstration of the potential effect of critical drug recalls on patient management that subsequently depends on alternative drug availability.

Clinically, the major indicators of successful management are positive health outcomes. Results of this study show no negative health outcomes following Aspirin recall as there were mostly no MACE incidents in the 3-month post-recall period (20,843 [99.5%]). This may be attributed to the timely switch for a large number of patients with a medication switch with same-therapeutic group alternative. A recent study describing prescription switching rationale and risks indicated that medications of the same-therapeutic group are considered equivalent, hence, switching between these medications is mostly recommended. Not only such medication substitution is safe, but also considered in many cases to improve health outcomes through higher efficacy, fewer side effects and/or lower cost [17].

As for the time period of MACE occurrence after ASA recall, almost 40% of patients had their first alternative anti-platelet prescription on the same day of MACE occurrence. In other words, it seems that patients had come to the hospital with symptoms of MACE were diagnosed and prescribed the alternative anti-platelet on the same visit. MACE incidents were dispersed throughout the 3 months of post-recall period, however, half the patients reportedly had MACE within 6 days following medication switch. This may reflect high-risk conditions such as perioperative cases that are predicted to be affected by antiplatelet management interruption within an average period of 8 days, as described in the literature [11]. Expectedly, all patients who were not prescribed alternative antiplatelet therapy (n = 8) had MACE incidents. Possible causes of management discontinuation include uninformed patients/caregivers and patient behavior concerning adherence to drug recall announcement. According to the literature, disturbance in management of long-term conditions such as switching medications even to other same-active ingredient products may negatively affect patients’ adherence [18, 19]. In fact, multiple studies indicated that older patients on multiple medications demonstrated anxiety from changing medications, even within same-therapeutic group, due to change in product appearance leading to increased risk of errors and nonadherence [20‒24].

This study is one of few studies to assess outcomes of immediate recall of a critical medication in Saudi Arabia. On the other hand, there were limitations concerning the uncertainty of patient management timing that mainly arise from missing information regarding the timely delivery of recall announcement and therapy change requirement to patients by healthcare providers, patients’ compliance and adherence to alternative management and patients’ hospital visitation patterns. Meaning, measured outcomes of this study may be affected by patient management received from outside of the healthcare facility from which the data of this study is derived. Such confounding factors often exist in studies using real-world data where settings may not be controlled such as the case in clinical studies.

In terms of generalizability of this study outcomes, key factors that may vary from one healthcare center to another should be considered: (a) availability of alternative ASA (other than the recalled ASA generic) in the healthcare center; (b) policy and procedures employed by the healthcare center in case of drug recall and/or shortage; (c) the number of high risk patients (e.g., perioperative patients) who may develop health complications from treatment discontinuation. Accordingly, measured outcomes in terms of medication switch to alternatives, time periods of the switch and patients’ health outcomes may vary across healthcare centers. Therefore, findings of this study may not be generalized on other healthcare centers. Rather, this study illuminates patient management characteristics after the ASA generic recall through assessing the study outcomes in a multicenter/region healthcare facility that mainly procured the recalled ASA.

Recommendations to avoid critical therapy delay or discontinuation after drug recall may include developing communication procedure among stakeholders before drug recall regarding ensuring alternative drug availability, conducting risk assessment before the drug recall. Moreover, healthcare facilities may be encouraged to develop contingency plan concerning alternative drug use within clinically recommended timeline in case of critical drug recalls.

Post-recall patient management appears to be considerably longer than the period clinically estimated for health complications occurrence after therapy discontinuation. Meaning, health complications may be expected to occur due to prolonged period of discontinued treatment that took place before patients were switched to alternative therapy. However, the vast majority of patients in this study have not been affected by health complications after the recall, and almost all patients have been switched to another ASA generic.

We would like to express our gratitude to the following colleagues and researchers for their valuable support in the data collection for this study; Ahmed Alonazi, Dr. Rana Almotawa and Dr. Ghadeer Alhamid.

The study procedures were performed in accordance with relevant guidelines and regulations. The study was approved by the Institutional Review Boards (IRB) Committee of Saudi Food and Drug Authority, which provided IRB number for this study. Due to the retrospective nature of the study and the use of de-identified data, this consent protocol was reviewed and the need for written and informed consent was waived by Institutional Review Boards Committee of Saudi Food and Drug Authority, which provided IRB number for this study.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author 1: writing – original draft, methodology, formal analysis, data curation, and conceptualization. Author 2: writing – review and editing, methodology, and conceptualization. Author 3: writing – review and editing, validation, and methodology. Author 4: writing – review and editing, methodology, conceptualization, and supervision.

The datasets collected for analyzed during the current study was from hospital electronic health records provided by the Research Informatics Department database at the Saudi Food and Drug Authority. Further inquiries may be directed to the corresponding author. The data that support the findings of this study are not publicly available due to the nature of data that may compromise the privacy of research participants and/or healthcare center. The data used in this study may be available from the Executive department of Research and Studies (e-mail: [email protected]) upon reasonable request.

Disclaimer.

The views expressed in this paper are those of the author(s) and not do not necessarily reflect those of the SFDA or its stakeholders. Guaranteeing the accuracy and the validity of the data is the sole responsibility of the research team.