Background

Engineering safety into the overall health care system to reduce errors and improve patient outcomes is an important paradigm that is being actively embraced by the designers of healthcare products and systems, caregivers and policy makers [1]. But research investigating the relationship of the design of packaged products and their impact on the behaviour of healthcare providers and patient care has not been widely conducted. This is likely because, historically, the regulators of medications and medical supplies have required the submission of objective evidence that manufacturing processes produce packaging that preserves, enables, and maintains the product’s safety and efficacy throughout the expected shelf life. Although preserving, protecting and manufacturing products used for patient care is critically important, what has not been assessed is how the package design impacts the ability of the healthcare provider (or patient) to accomplish critical tasks (e.g. properly identifying, aseptically opening and transferring, and properly dosing and administering). Only recently have a host of standards and regulatory documents started to proliferate which require the objective evaluation of how packaging performs in the hands of users (both healthcare providers and patients) [2–6] (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Objective evaluation and regulatory documents by standards.

https://doi.org/10.1371/journal.pone.0255099.t001

Even as standards and requirements for packaging change in an attempt to engineer safety into healthcare, little evidence exists describing the relationship of packaging design to provider behaviours, and ultimately, patient outcomes. This could be because the errors associated with packaging and labelling design are latent in nature, with the mistake occurring upstream from the presentation of the problem. Further, latent errors tend to be “systems related factors;” [16] that is, one factor among several that contribute to an event. Papers that review root-cause analysis of medical errors and adverse events suggest that similar packaging and labelling are fundamental reasons for medication errors [16–19], though other potential problems related to poor packaging design (e.g. difficulty opening or transferring cleanly) have yet to be thoroughly investigated.

And it isn’t just packaging that hasn’t been thoroughly researched, understood, and optimized; the healthcare environment itself, undoubtedly, impacts how care givers interact with products to deliver patient care. Of particular interest to us was how packaging performs within the prehospital setting. Emergency Medical Services (EMS) are administered during prehospital care to stabilize patients and transport them to a location better equipped to provide comprehensive care. Despite the extreme conditions that may be present (e.g., poor lighting, extreme heat or cold, noise, chaos, emergency vehicle movement, interloping friends and family), they are infrequently (if ever) considered by product and package designers.

As evidence of this, we offer data collected in healthcare transport environments, specifically intrahospital transports of critically ill patients, where the rates of adverse events have been noted to be as high as 70% [20–22]. Bergman et al. identified five broad categories of hazards contributing to transport-related adverse events and identified the most common to be errors associated with the category tools and technology, within this category a common problem was “products that were not designed in ways that consider the context/environment of care and support the needs [of heathcare providers] [23].”

Anecdotally, paramedics have been observed to employ flashlights to identify products, or use one hand, and scissors (or teeth) to open items required urgently. These coping strategies, used to deal with shortcomings of designs which don’t appropriately consider the myriad of contexts where paramedics deliver care, have the potential to play a role in patient outcomes. For example, tools and/or inappropriate opening techniques may serve as an indirect mechanism for the transfer of microbes [17,24], a source of healthcare associated infections (HAIs). Although precise sources of microbes affiliated with HAIs are difficult to identify, a surveillance study investigating prehospital infection rates suggests that patients treated by advanced life support (ALS) providers prior to being admitted to the hospital have greater rates of HAIs than those transported by other means [25]. Supporting the idea that the prehospital environment is a fertile place for systemic improvement is the work of Orellana et al [24]. Among the factors that their research team identified as significant risk factors associated with Methicillin-resistant Staphylococcus aureus (MRSA) detection on EMS personnel were infrequent hand hygiene after glove use and low frequency of hand washing. Researchers found a high prevalence of MRSA among Ohio EMS personnel, noting this was both an occupational hazard and a patient safety concern [24].

It is unclear whether conditions present in prehospital environments affect the use of packaged products in ways that provide a source of risk for patients. In light of this, we evaluate the notion that prehospital work creates unique challenges for paramedics as they use packaged products to deliver care. In doing so, our objectives were to:

1. describe difficulties associated with packaging design in prehospital settings;
2. investigate the coping strategies used by paramedics when difficulties occurred; and
3. assess the potential impacts these difficulties had on patient care.

Methods

The study was conducted in accordance with methods approved by the Institutional Review Board at the American Institutes for Research (AIR (#EX00411)), the review board affiliated with the National Registry of Emergency Medical Technicians (NREMT). The transfer of de-identified data to MSU for analysis was determined to not meet the definition of human subjects as defined by the US Department of Health and Human Services) by the MSU Institutional Review Board Manager (application #x16-1412e).

Results

Characteristics of study subjects

A total of 1,912 responses were received (response rate = 16%). Respondents were excluded because they had not provided care in the prehospital setting in the past 12 months (n = 193) or due to incomplete responses (n = 17). The majority of the survey respondents were male (79.3%), white (88.7%), had a more than 10 years’ experience with EMS (61.5%) and served the role of care provider (74.7%). Table 2 presents an overview of the results related to participant characterization for the 1,702 responses included for analysis.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Characteristics of participants.

https://doi.org/10.1371/journal.pone.0255099.t002

Difficulties and coping strategies

Frequencies and proportions related to all three objectives are found in Table 3 (medications) and Table 4 (medical supplies). Nearly 20% of the paramedics indicated that they had encountered difficulty related to packaging within the previous 12 months for both tasks (identifying and opening) across product categories (medications and medical supplies); specifically: 359 respondents (21.1%) reported that they had difficulty identifying medications within the previous 12 months (Table 3); while 290 respondents (17.0%) reported difficulty identifying medical supplies (Table 4); 349 (20.5%) reported difficulty opening a medication and 399 (23.4%) had difficulty opening a medical supply during this same time frame.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Frequencies and proportion of respondents self-reporting difficulty with each task (identifying and opening), resulting coping strategies and negative impact on care related to medication use within the previous 12 months.

https://doi.org/10.1371/journal.pone.0255099.t003

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Frequencies and proportion of respondents self-reporting difficulty with each task (identifying and opening), resulting coping strategies and negative impact on care related to use of medical supplies within the previous 12 months.

https://doi.org/10.1371/journal.pone.0255099.t004

The most commonly reported reason for difficulty identifying a medication was that “different medications have similar packaging” (n = 246, 68.5%, Table 3), while the most commonly reported reason leading to difficulty in identifying a medical supply was “crowded label made it difficult to read” (n = 189, 65.2%, Table 4).

The top reason for difficulty opening was the same across both product categories, “product required two hands to open” (n = 172, 49.3%; n = 270, 67.7%). Frequencies of reported, specific reasons associated with difficulty opening products are presented in Table 3 for medications and Table 4 for medical supplies.

“Flashlight use” was reported as the most common coping strategy employed when paramedics reported difficulty identifying both medications (n = 211/359, 58.8%) and medical supplies (n = 156/290, 53.8%). “Scissors” were the most commonly reported strategy employed when opening difficulties were encountered; this was the case for both medications (n = 189/349, 54.2%; Table 3) and medical supplies (n = 307/399, 76.9%; Table 4). Of additional interest, was the reported use of teeth to open both medications (n = 103/349, 29.5%) and medical supplies (n = 156/399, 39.1%) when difficulties were encountered.

We fit two multivariable logistic regression models with the two outcomes of interest: difficulty opening medication (Model 1, Table 5) and difficulty opening medical supplies (Model 2, Table 6). The following covariates were considered for both models: age, sex, race, years of experience in EMS, primary role, the level of education, and size of community. In addition, difficulty in identifying medication and difficulty in identifying medical supplies were included for Model 1 and Model 2, respectively.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Univariable and multivariable regression analysis for difficulty opening medication.

https://doi.org/10.1371/journal.pone.0255099.t005

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 6. Univariable and multivariable regression analysis for difficulty opening medical supplies.

https://doi.org/10.1371/journal.pone.0255099.t006

Tables 4 and 5 present the logistic regression analysis for difficulties opening medications and difficulties opening medical supplies, respectively. The odds of difficulty opening medications for those who had experienced difficulty identifying medications within the past year was 2.91 times that of the odds of difficulty opening medications for those who had not experienced difficulty identifying medications within the past year (OR: 2.91, 95% CI: 2.21–3.83). Significant effects of sex and primary role on the odds of reporting difficulty opening medications were also observed (Table 5). Specifically, the odds of reporting a difficulty opening medication was reduced by approximately 49% for males when compared to that of female respondents (OR: 0.51, 95% CI: 0.38–0.68). Additionally, the odds of reporting difficulty with opening medications were reduced by 29% for those with other roles such as dispatcher, instructor and administrator, compared to patient care providers (OR: 0.71, 95% CI: 0.52–0.97).

When difficulties associated with opening medical supplies were examined (Table 6), results were similar to those reported for opening medications (Table 5). The odds of difficulty opening medical supplies for participants who reported a difficulty identifying medical supplies within the past 12 months were 3.67 times higher than the odds for those who had not reported difficulty identifying a medical supply within the past 12 months (OR: 3.67, 95% CI: 2.80–4.85). For every one-year increase in age, there was a 2% increase in the odds of difficulty opening medical supplies (OR: 1.017, 95% CI: 1.00–1.03).

Discussion

Although other researchers have indicated healthcare products not designed in ways which consider care context as among the most common hazard associated with medical errors [23], there is a dearth of work characterizing how packaged products perform in realistic contexts of use. The most oft-studied healthcare environment is the perioperative context, an environment that is arguably less demanding for product design than that of prehospital environment. Further, the performance of package design, a critical and ubiquitous component of all products, has, to our knowledge, never been investigated in prehospital environments, despite the fact that similar packaging and labelling have been identified as fundamental reasons for medication errors [27].

Our analysis provides some initial evidence that paramedics have difficulties both identifying and opening packaged products and that these difficulties have the potential to negatively impact patient outcomes (Tables 3 and 4). Several factors were found to be significantly associated with the reported difficulty with opening for both medications and medical supplies. For both categories of product (medications and medical supplies), respondents had significantly higher odds of reporting difficulty with opening if they had also reported difficulty with identifying within the past year (Tables 5 and 6). These results could indicate that respondents who were more comfortable reporting difficulties were more likely to report for each of the tasks studied (identifying and opening); however, it could also suggest that when manufacturers do not engage in a thoughtful design process, poor design affects multiple, critical tasks. If the latter is the case, paramedics could encounter serial difficulties among the tasks to be accomplished (first in identifying and then, subsequently, in opening needed products), potentially delaying patient care. The potential and impact for serialized difficulties associated with packaging in prehospital contexts was beyond the scope of the work but presents an interesting idea that needs study.

Results also advocate for the incorporation of users likely to have difficulties associated with packaging during the iterative design process. Findings indicate that females were significantly more likely to report difficulty opening medications than their male counterparts (Table 5), potentially because of documented differences in hand strength [28–31]. That said, this could also be indicative of a willingness to express vulnerability that men do not share [32]; if this is the case, it would support the idea that difficulties are actually underreported. We also found older paramedics to be significantly more likely to report difficulties opening medical supplies than their younger counterparts (Table 6); declines in the strength of the hand are well-documented as a natural process that occurs with aging [29,33] and offer an explanation of this result.

The limited work that has been done to understand how packaging performs in healthcare contexts has focused on perioperative environments [27,34,35]. Many of the published findings conducted in these more controlled contexts of care indicate improvements in packaging are needed, parallel to the findings we present here. Published studies using perioperative contexts have concluded that packages are: “hard to open” [35,36] and employ “crowded label contents” which make crucial information difficult to locate [27,34,35]. While our survey of paramedics echoed the difficulties voiced by perioperative personnel, they also expressed problems, and coping strategies, that were uniquely their own. The indication or implication that paramedics only have one hand available to physically manipulate packaging was apparent throughout the results. The need for designs which enable one-handed use was evident in this sample, with the “need for two hands [when opening the packages] [32]” reported as a common difficulty and “use of teeth” or the need for a “partner’s assistance” as common coping mechanisms when difficulties were encountered. It is not beyond fathom that the use of instruments or teeth have the potential to serve as an indirect mechanism for the transfer of microbes, one possible contributor among many which could help to explain the high rates of infection among patients who have received advanced life support (ALS) prior to hospital admission as compared to those transported by other means [25].

If this is the case, it would suggest that the rates of negative impact on patient outcome (Tables 3 and 4) are actually under-reported, further supporting this as an issue in need of careful consideration.

Limitations

One of the more interesting contributions of our study is the number of participants who reported having difficulties that resulted in negative impact on patient care (1.2% associated with identifying a medication; 1.9% associated with opening a medication;1.8% associated with identifying a medical supply; and 3% associated with opening a medical supply). Readers are urged to interpret results related to negative patient outcomes cautiously. The survey itself did not specifically define the term “negative outcome,” instead leaving the survey respondent to interpret its meaning (see Supplemental files for survey). The open-ended nature of this approach enables a wide array of potential outcomes to be considered as negatively impacting care (from delay of care to death of patient). A lack of questions that probed details surrounding the negative outcome precludes detailed interpretation of these results and implores the need for future research.

A further limitation of the survey is the binary nature of the question about difficulties affiliated with packaging use. Specifically, a single question asks respondents about difficulties (identifying or opening) by product type (medications or medical device) and is only presented once (for each combination). A “yes response” related to the difficulty question triggered a cascade of questions probing the difficulty and coping strategies that were employed. Although this provides us with an indication of the frequency across participants, the way that the survey was framed we cannot deduce the number of occurrences for each respondent (across the last year) or whether or not the coping mechanisms that they reported arose from multiple incidences or a single event.

Conclusions

Our findings suggest that paramedics encountered difficulties identifying and opening packaged products and provides preliminary evidence that suboptimal designs potentially negatively impact patient outcomes. In light of this, the designers of healthcare products should incorporate insights gathered in prehospital environments to enhance package designs in ways that enable the performance of tasks critical to care. This becomes an imperative as healthcare evolves to better integrate the complex components that must work in synergy toward common, patient-centered goals, and is requisite to improved outcomes [37].

Supporting information

Acknowledgments

The author would like to thank Hope Akaeze who assisted in her role with the MSU CSTAT organization. Ms. Akaeze provided useful resources for statistics concepts presented herein.