Nearly all thirty most frequently used emergency department drugs experienced shortages from 2006–2019

Abstract

Background:

Drug shortages contribute to avoidable medication error and patient harm; these shortages are exacerbated in the Emergency Department due to the time-sensitive nature of acute care.

Methods:

We performed a cross-sectional study to describe the frequency and duration of drug shortages associated with the most frequent medications administered in the ED. We identified the most frequently used ED medications and calculated number of visits associated with these medications using the 2006–2019 National Hospital Ambulatory Medical Care Survey. We obtained the frequency and duration of shortages associated with these medications from the University of Utah Drug Information System. We calculated duration and total ED visits associated with shortages of the most frequently used ED medications.

Results:

From 2006 through 2019, the most frequently used drugs were ondansetron (255.1 million ED visits), 0.9% normal saline (251.3 million ED visits), and ibuprofen (188.5 million ED visits). All but two of the top thirty most frequently used medications experienced a shortage. The median shortage duration was 425 days, while the longest were for injectable morphine (3,202 days). The number of ED visits associated with drugs experiencing shortages increased from 2,564,425 (2.2% of U.S. ED visits) in 2006 to 67,221,968 (60.4%) in 2019. The most common reasons for shortage include manufacturing delays and increased demand.

Conclusions and relevance:

Drug shortages were more frequent and persistent from 2006 through 2019. Further studies on the clinical impact of these shortages are needed, in addition to policy interventions to mitigate shortages.

1. Background

Drug shortages contribute to preventable medication errors, delayed care, and increased morbidity [1–4]. The impact of shortages are amplified in the Emergency Department (ED) where care is most time-sensitive [1–4]. The U.S. Food and Drug Administration (FDA) has identified the root causes of persistent drug shortages as: 1) lack of financial incentive for manufacturers to produce less profitable drugs; 2) little market incentives to develop supply chains that detect shortages early; and 3) difficulty recovering from disruptions due to logistical and regulatory challenges [5]. Prior research suggests new drug shortages peaked in 2011; more recent evidence suggests recent shortages may be worsening again [1,4]. In a 2018 survey, 91% of ED physicians reported a shortage of a critical medication within the past month [6]. Early evidence suggests the COVID pandemic has exacerbated drug shortages given disruptions to the supply chain and manufacturing capacity and increased demand for analgesics and sedatives for patients requiring ventilatory support [5,7,8].

Little is known about the prevalence of drug shortages among the most commonly used ED medications. Therefore, we aim to quantify the frequency and duration of shortages of the most frequently used medications in U.S. EDs.

2. Methods

2.1. Data source

We performed a cross-sectional analysis of the National Health Ambulatory Medical Care Survey (NHAMCS) from 2006 through 2019 [8]. NHAMCS is a national sample of ambulatory visits to non-institutional, non-federal general and short-stay hospitals; detailed methods are described elsewhere [9].

We obtained drug shortage frequency and duration data from January 1, 2006 through September 30, 2019 from the University of Utah Drug Information Service (UUDIS) Drug Shortage Dataset [10]. The UUDIS defines a shortage as a supply issue that alters how the pharmacy prepares or dispenses a drug product, or influences patient care when prescribers must use an alternative agent, and the Government Accountability Office (GAO) considers UUDIS data these data to be the most comprehensive and reliable source of drug shortage information [11]. Detailed methods on UUDIS have been previously published [12]. Briefly, UUDIS receives voluntary reports of drug shortages online; then, clinical pharmacists verify each shortage by contacting each manufacturer to determine which medications are in shortage nationally, reasons for the shortage, and estimated release dates. This study was determined to be exempt from IRB review.

2.2. Analytic approach

We used the NHAMCS to identify the most common drugs administered during ED visits using the variable “medications administered in ED” and “medications administered in ED or prescribed at discharge” from 2006 through 2019. We excluded medications that were only prescribed at discharge in order to reflect shortages in the acute care setting [13]. For drugs with multiple administration routes and formulations, we used the most frequent formation in ED settings; when both injectable and oral routes were commonly used, we included both. We estimated the number of U.S. ED visits using survey weights to account for the NHAMCS’s weighted probability sampling survey design.

To calculate prevalence, we first calculated the number of days each drug experienced a shortage using the UUDIS dataset, then estimated the number of ED visits associated with each drug experiencing shortage by multiplying the number of weighted visits in NHAMCS for each medication by the duration of the shortage. Because the NHAMCS provides the annual average visits and drug shortages in number of days over a four-week period, and institutions and providers likely experience drug shortages (new or resolving) gradually over the course of weeks, we rounded shortages to the next nearest month after calculating the number of days on shortage and dividing by the total number of days in that month. We also tested the hypothesis that most frequently used medications in each year would not change substantially between years using the Spearman’s rank correlation to quantify the consistency of ranking among most frequently used medications between years.

3. Results

We examined 376,648 ED visits, representing 1,829,083,068 million visits from January 1, 2006 to September 30, 2019. The most frequently used drugs were ondansetron (255.1 million ED visits), 0.9% normal saline (251.3 million ED visits), and ibuprofen (188.5 million ED visits) (Table 1). A majority (60%, N=18) were injectable, and 83% (N=25) were available in generic form during the entire study period; all drugs were generic by 2013.

Table 1.

Most commonly used ED medications with associated shortages, 2006–2016.

Medication	Route1	Available generically as of 1/1/06? If not, date of FDA generic approval	Most recent shortage dates2	Reason for most recent shortage2	Total shortage duration 1/1/06–9/30/19 (days)2	Longest shortage duration (days)2
Ondansetron	Injection/Oral	8/1/07	3/7/18-ongoing (injection)	Unknown	3,703 (injection)	3,134
0.9% Normal saline	Injection	Y	1/28/13–6/26/19	Manufacturing delays	2,339	2,339
Ibuprofen	Oral	Y	n/a	n/a	0	0
Acetaminophen-hydrocodone	Oral	Y	11/24/08–12/15/09	Unknown	385	385
Acetaminophen	Oral	Y	6/4/18–8/8/18	Unknown	158	65
Ketorolac	Injection	Y	11/11/09-ongoing	Unknown	3,609	3,609
Morphine	Injection	Y	6/24/10-ongoing	Manufacturing delays	3,368	3,368
Hydromorphone	Injection	4/3/09	6/12/17-ongoing	Manufacturing delays	2,848	2,848
Promethazine	Oral/injection	Y	8/23/18-ongoing (injection)	Unknown	3,210 (injection)	2,809 (injection)
Albuterol	Inhaled	Y	10/26/15–11/11/15	Manufacturing delays	200	113
Acetaminophen-oxycodone	Oral	Y	3/14/11–6/16/11	Unknown	229	136
Ceftriaxone	Injection	Y	2/14/14-ongoing	Manufacturing delays	2,137	2,044
Diphenhydramine	Injection/oral	Y	8/13/18-ongoing (injection)	Manufacturing delays	1,264 (injection)	565 (injection)
Lidocaine	Injection	Y	9/7/11-ongoing	Supply/demand	2,994	2,944
Aspirin	Oral	Y	1/16/18-ongoing (325 mg)	Unknown	621 (325 mg)	621 (325 mg)
Lorazepam	Injection	Y	2/11/16-ongoing	Manufacturing delays	4,195 (injection)	2,685 (injection)
Prednisone	Oral	Y	3/22/19-ongoing	Manufacturing delays	1,184	1,054
Ciprofloxacin	Injection/oral	Y	2/14/18-ongoing (injection)	Unknown	2,484 (injection)	1,893 (injection)
Methylprednisolone sodium succinate	Injection	Y	1/28/16–9/26/16	Manufacturing delays	2,689	1,882
Azithromycin	Injection/Oral	Y	1/11/18-ongoing	Manufacturing delays	2,079 (injection)	948 (injection)
Cephalexin	Oral	Y	3/14/14–3/9/15	Unknown	360	360
Amoxicillin	Oral	Y	11/14/14–8/24/15	Unknown	427	283
Tramadol	Oral	Y	3/4/09–4/28/11	Production delays	784	117
Tetanus-diphtheria-acellular pertussis (Tdap)	Injection	N (brand name only: Adacel 6/13/05, Boostrix 12/13/08)	8/18/13–7/28/15	Manufacturing delays	832	709
Flexeril	Oral	1/31/13	n/a	n/a	0	0
Levofloxacin	Injection/Oral	6/21/11	7/14/14–9/18/15 (injection)	Business decision	924 (injection)	450 (injection)
Naproxen	Oral	Y	12/1/10–12/27/11	Unknown	391	391
Famotidine	Injection/Oral	Y	6/5/11–7/29/19 (injection)	Unknown	2,966 (injection)	2,966 (injection)
Trimethoprim-sulfamethoxazole	Injection/Oral	Y	4/28/10–4/16/14 (injection)	Unknown	1,488 (injection)	1,449 (injection)
Acetaminophen-codeine	Oral	Y	12/10/14–2/10/16	Supply/demand	427	427

¹Most common route in ED setting.

²Drug shortage dates, reason, duration data from the University of Utah Drug Information System (UUDIS).[1]

Of the top thirty most frequently used drugs over the study period, all but two (ibuprofen and cyclobenzaprine) experienced a shortage. The median shortage duration was 425 days, while the longest were for injectable morphine (3,202 days) and injectable lidocaine (2,762 days). The medication shortages potentially associated with the most ED visits were ondansetron (172 million ED visits) and 0.9% normal saline (138.8 million ED visits).

The number of ED visits associated with drugs experiencing a shortage increased from 2,564,425 (2.2% of U.S. ED visits) in 2006 to 68,221,968 (60.4% of U.S. ED visits) between January 1 and September 30, 2019 (Fig. 1). The Spearman’s rank correlation coefficient for the top thirty most frequently used ED medications ranged from 0.87 to 0.97 (P<0.001 for all pairwise comparisons), demonstrating high consistency between years.

Fig. 1.

ED Visits per Month Impacted by Drug Shortages, January 1, 2006 through September 30, 2019

The most common reasons for shortage include manufacturing delays and increased demand. Only one formulation of one medication (0.9% normal saline in small volume bags) was impacted by natural disaster.

4. Discussion

Shortages of the most frequently used ED medications have increased exponentially over time, affecting 60% of U.S. ED visits in 2019. To our knowledge, there has been no previous national study examining the prevalence and duration of shortages among the most frequently used ED medications. The high volume of visits associated with these medications amplifies the potential for harm when clinicians are forced to rely on less commonly used alternatives [2].

The majority of shortages are due to increased demand, manufacturer decisions, and underlying economic drivers, such as low profit margins for older, generic drugs, as reported by the GAO and Food and Drug Administration (FDA) [11,14]. Shortages may have worsened over time due to these lower financial incentives, resulting in fewer, more consolidated suppliers with more fragile supply chains. The increasing number of visits affected by shortages may also be due to the increase in ED visit volume over time [15]. Our findings appear to support a transient decrease in shortages immediately following the implementation of the Food and Drug Administration Safety and Innovation Act of 2012 [3]. However, the frequency and severity of shortages increase acutely after 2017, primarily due to shortages of the two most frequently used ED medications: injectable normal saline and ondansetron (Fig. 1). Further research is needed to understand why shortages have worsened in recent years, and whether legislative or regulatory reform is needed.

We also demonstrate there has been little change in the most frequently used ED medications over time; because demand is predictable, these shortages should be preventable. Potential solutions include enhanced communication regarding supply chain vulnerabilities; proactive mitigation of critical medication shortages; financial incentives and contracts using risk-adjusted returns; and public-private partnerships [11,14]. Future studies evaluating the impact on clinical outcomes are needed, especially in light of disruptions to the supply chain as a consequence of the COVID-19 pandemic [6,7].

4.1. Limitations

First, utilization data may be impacted by ongoing shortages; thus, our estimates are likely to underestimate actual numbers of ED visits associated with drugs experiencing a shortage. Second, individual hospital supply chains differ; therefore, it is not possible to determine which hospitals were affected, and how severe the shortage was at a given facility. Third, the UUDIS dataset was not designed specifically to assess ED shortages and relies on pharmacist reporting, however extensive verification of shortages via manufacturers is conducted, and the UUDIS dataset is considered the most reliable source of drug shortage information per the U.S. Government Accountability Office (GAO) and is more consistent and thorough than the Food and Drug Administration (FDA) database. Lastly, our study has limitations common to studies using administrative data; however, we followed suggested National Center for Health Statistics (NCHS) guidelines to limit potential shortcomings [8].

5. Conclusions

Drug shortages affecting ED medications in the U.S. are increasingly prolonged and severe, affecting all but two of the top thirty most frequently used medications over the past decade, and affected nearly 60% of ED visits annually in both 2018 and 2019. The minimal change in medications from year to year demonstrates predictable demand; thus, shortages should be preventable. Policy changes are urgently needed to ensure sufficient supply and maintain the quality and safety of ED care in the U.S.