Abstract
Background
Hypocalcemia following severe injury is common. Current institution‐specific guidelines recommend calcium (Ca++) supplementation during blood product resuscitation. We hypothesize that a nudge intervention would improve clinician adherence to Ca++‐specific guidelines.
Methods
This study at an urban Level I trauma center evaluated compliance with Ca++‐specific guidelines during trauma resuscitations. A baseline assessment of guideline awareness preceded four‐month pre‐intervention and four‐month post‐intervention periods from July 2021 to February 2022. Nudge signs prompting clinicians to administer Ca++ were placed throughout all phases of care. Administration of ≥1 dose of Ca++ after four blood products was the primary endpoint. Aggregate ionized calcium (iCa++) levels and percent time within a normal iCa++ range were secondary outcomes.
Results
Baseline assessment of n = 41 trauma team members indicated 83% were uncertain or unaware of current Ca++ recommendations. Of 86 screened patients, 25 met inclusion criteria. These were median ([IQR] 34 [25–43]) years old with an Injury Severity Score of 18 [14‐26] and 84% penetrating injuries with n = 11 pre‐intervention and n = 14 post‐intervention. The absolute difference (95% CI) in Ca++ guideline adherence post‐intervention compared to pre‐intervention was 6.5% (−11.9% to 24.9%, p = 0.755). In aggregate, iCa++ levels were no different between groups, although the distribution of levels post‐intervention trended toward the normal range with less extreme hypocalcemia.
Conclusions
Signs prompting clinicians to give Ca++ were associated with a modest, non‐statistically significant increase in adherence to institution‐specific guidelines and a slight shift in the distribution of iCa++ values toward normal. Future work to optimize resuscitation should evaluate larger cohorts of acutely injured patients and more potent nudges.
Keywords: blood product resuscitation, guideline adherence, nudge intervention
1. BACKGROUND
In the setting of severe trauma, calcium (Ca++) dysregulation is common and contributes to mortality. 1 , 2 , 3 , 4 , 5 A significant proportion of trauma patients are hypocalcemic on presentation to the trauma bay. 1 Blood product resuscitation further compounds this dysregulation as blood products are anticoagulated with citrate, which binds or chelates Ca++. If the citrate is not metabolized quickly, this can further lower systemic Ca++ levels.
Multiple studies have demonstrated that patients who were given blood products in the prehospital setting are susceptible to developing hypocalcemia. 2 , 6 Moore et al. retrospectively reviewed two randomized control trials assessing how ionized Ca++ (iCa++) levels, and by extension mortality, are affected by prehospital plasma transfusion. Mortality was significantly higher in the hypocalcemia group compared to the normocalcemia group (18.6% vs. 12.2%). 7 However, when given supplemental Ca++, Kyle et al. found that 70% of patients maintained normal iCa++ levels during blood product resuscitation. 6
After a study conducted at our institution by MacKay et al. showed that extreme hypocalcemia, defined as iCa++ < 0.84 mmol/L, was significantly associated with mortality (60% vs. 4%, p < 0.01), 4 we modified our institutional Trauma Exsanguination Policy to include Ca++‐specific guidelines, which addressed both empiric administration (after the first fourunits of blood products) and dosing in response to low iCa++ on intermittent blood gas analysis. Despite education on this guideline update, abnormal iCa++ levels persisted, and adherence to the new guidance appeared limited.
To address this gap in knowledge and guideline adherence, we sought to leverage our institutional expertise in clinical nudges. 8 Nudge theory has been applied in healthcare previously, with many of the investigations focused on changing the choice architecture in the electronic health record (EHR) to guide clinicians in making the optimal choice. 9 , 10 , 11 With proper optimization for the clinical landscape, nudges, or changes to the choice environment that lead to predictable, altered behavior that neither forbids options nor alters incentives, have the potential to both improve healthcare delivery and patient outcomes. 8 , 9 , 10 , 11 , 12 Nudges can take many forms and lead to varying degrees of behavioral impact.9–11,13 In this study, we sought to quantify the baseline understanding of our institutional Ca++ guideline and the effect of a nudge intervention on both adherence to the guideline and on iCa++ levels. We hypothesized that a nudge intervention would improve adherence to Ca++‐specific guidelines during blood product resuscitation.
2. METHODS
This study was approved under Exempt Review by the University of Pennsylvania Institutional Review Board (Protocol #849600) with a waiver of informed consent and Health Insurance Portability and Accountability Act (HIPAA) Waiver of Authorization. The investigators adhered to the policies regarding the protection of subjects and data. This manuscript follows the Standards for Quality Improvement Reporting Excellence (SQUIRE) 2.0 reporting guideline. 14
2.1. Phase 1: Baseline assessment
The Trauma Exsanguination Policy at the University of Pennsylvania contains Ca++‐specific guidelines during massive transfusion, noting that 1 g of calcium chloride (CaCl) should be given empirically after the first 4 units of any blood product administration. In addition, blood gases should be run approximately every 30 min during massive transfusion protocol administration, and low iCa++ levels should be treated with additional doses of CaCl. This study focused exclusively on the empiric administration of CaCl.
Prior to the start of the study, a baseline performance‐improvement survey was sent via email to a multidisciplinary group of representative trauma team members that are routinely involved in massive transfusions of severely injured trauma patients. This included all current Trauma and ICU Registered Nurses (RNs), Certified Registered Nurse Anesthetists (CRNAs), Anesthesiologists, Critical Care Advanced Practice Providers (APPs), including both Nurse Practitioners and Physician Assistants, Trauma Fellows, and Trauma Attendings at Penn Presbyterian Medical Center (PPMC). These team members all participate in acute trauma resuscitations; physicians, CRNAs, and APPs write orders in collaboration with nurses who can prompt or suggest interventions. The survey contained questions to both assess 1) clinicians' awareness of the Ca++‐specific guidelines in the institutional Trauma Exsanguination Protocol and 2) acceptability, appropriateness, and feasibility of the institutional Ca++‐specific guidelines 15 (Supplemental Figure S1). Data were collected and managed using REDCap electronic data capture tools hosted at the University of Pennsylvania. 16 , 17
2.2. Phase 2: Pre‐ and post‐intervention
2.2.1. Nudge design and feasibility assessment
We worked with the Penn Medicine Nudge Unit to assess the feasibility of a variety of nudges. Ideas for nudges ranged from a sign providing information about the policy, offering Ca++ adjuncts in the massive transfusion protocol (MTP) coolers that follow the patient through all phases of care, and automating orders for Ca++ adjuncts in the EHR system after a certain number of blood products had been given. While the most effective nudges tend to be higher on the Nudge ladder (Figure 1), often making the optimal choice the default option or prompting an active choice in the decision workflow, these are not always feasible given the existing workflows in clinical spaces.
FIGURE 1.
Nudge ladder with examples specific to calcium administration during blood product resuscitation. Created at https://BioRender.com.
Our contextual inquiry identified low baseline awareness of the Ca++‐specific guidelines, but that the guidelines were acceptable. According to models of behavior change (i.e., COM‐B, EAST, Fogg), behavior change happens when the behavior is easy enough to do, when there is opportunity or a prompt to do the behavior, when there is sufficient motivation, and when it is socially acceptable. 18 , 19 , 20 Therefore, we focused our nudge solution on prompting the behavior in the environment in which decisions on resuscitation therapies were made. Our clinical observations revealed that resuscitation orders are often given verbally and executed by nurses in these critical situations and then later documented in the EHR. Therefore, we focused our initial nudge design efforts on prominent signage to prompt clinicians to remember to use Ca++ in conjunction with the transfusion of blood products rather than in the context of technological choice architecture.
Designing the graphic was an iterative process, informed by subpopulation interviews with the Trauma Program Medical Director, two Trauma RNs, one CRNA, one Anesthesiologist, and one Trauma Surgeon at PPMC (Supplemental Figure S2). Discussions during these interviews included which phrase should be placed at the top of the sign to best capture the attention of clinicians and guide behavior, whether to include the Ca++‐specific guidelines in text, as a graphic, or both, how best to visualize blood products and the Ca++, and whether a message about leadership recommendations or patient well‐being would best motivate clinicians. The final nudge intervention (Figure 2) contained a simple graphic and text highlighting the Ca++‐specific guidelines. It also highlighted that maintaining a normal iCa++ level during the trauma resuscitation is associated with improved survival.
FIGURE 2.
Final nudge intervention sign (graphic and text) following iterative design process.
Signs were placed in each of the three phases of care—the trauma bay, operating room (OR), and intensive care unit (ICU)—in high‐traffic locations or locations where they would be most likely to be seen. For example, signs were placed on the Belmont machines used in the trauma bay and OR. Additionally, signs were placed on the poles where blood products are hung during resuscitations and at locations where clinicians retrieve blood products and scan them into the EHR.
2.2.2. Study setting and data collection
PPMC is a Level 1 trauma center located in Philadelphia, PA, which conducts over 3500 trauma evaluations annually. The aim of the study was to evaluate the effectiveness of a nudge intervention to improve adherence to Ca++‐specific guidelines during massive transfusion in the clinical setting. The design of the study was informed by the RE‐AIM model, a framework for evaluating interventions that assesses five components: reach, efficacy, adoption, implementation, and maintenance. 21 For this study, we were primarily interested in reach (i.e., the proportion of resuscitations in which CaCl was administered empirically after four blood products). The 1‐year pilot study used a quasi‐experimental design with a pre‐intervention period (July 1, 2021–October 31, 2021) and a post‐intervention period (November 1, 2021–February 28, 2022). During the pre‐intervention period, baseline adherence was assessed as no intervention was performed at that time. The EHR was used to identify patients who were massively transfused and fit the study's inclusion criteria. The subset of patients who met the criteria had data merged with the Penn Trauma Databank, which has detailed trauma‐related outcomes.
2.2.3. Study participants
In this pre‐post study, all patients at PPMC receiving one blood product in the trauma bay from July 2021 to February 2022 were identified and initially screened for eligibility (Supplemental Figure S3). Patients were included if they were Critical Administration Threshold‐positive (CAT+), meaning that they received three or more units of red blood cells (RBC) within any 1‐h interval during the first 6 h, were 18 years or older, had Ca++ measurements, and survived beyond the trauma bay. The primary endpoint was adherence to Ca++‐specific guidelines listed in the Trauma Exsanguination Policy at PPMC. Patients meeting inclusion criteria were separated into pre‐ and post‐intervention groups depending on when they presented to the trauma bay. Post hoc power analysis indicated that a sample size of n = 14 in the intervention phase would detect an increase in guideline compliance from a baseline of 35% to 70% with a power of 0.8. To detect an increase in compliance to 85%, the level typically expected for most guidelines, an intervention sample size of n = 6 would be required.
2.2.4. Intervention
Signs were placed in the trauma bay, OR, and ICU in the morning of November 1, 2021 (Supplemental Figure S4). This change was implemented using our standard Trauma Performance Improvement Process of (1) soliciting input from trauma team participants, (2) proposing a revision to all stakeholders, (3) following the revision, and (4) educating stakeholders and resuscitation participants on the changes in advance of a mutually agreed upon “go live” date. There were a total of nine signs in the trauma bay (one on the Belmont, one on each of the two code carts, one on the blood product pole, one on the blood kiosk, two at the physician station, and two at the nurse station), two signs in the OR (one on the Belmont and one on the blood product pole), and 22 signs placed in the ICU (one on the Belmont, one by the notification board in the hall, and one in each of the 20 rooms on the floor).
2.2.5. Outcomes
The primary endpoint was the administration of at least one Ca++ dose following the administration of four blood product units. Aggregate iCa++ levels and percent time (% time) within a normal iCa++ range were secondary outcomes. Additionally, both hypocalcemia and hypercalcemia at the first Ca++ measurement were evaluated. Hypocalcemia and hypercalcemia were defined using the institution's laboratory reference range (low limit: iCa++ = 1.0 mmol/L, high limit = 1.24 mmol/L) and by assessing the extreme iCa++ levels as defined in the work by MacKay et al. (extreme hypocalcemia: iCa++ < 0.84 mmol/L, extreme hypercalcemia: >1.30 mmol/L). 4
2.2.6. Analysis
Baseline and demographic characteristics were summarized by standard descriptive statistics. Continuous variables were reported as median and interquartile range (IQR), whereas categorical variables were reported with counts (n) and percent (%) and as absolute difference (%) with a 95% confidence interval (CI). Differences in continuous variables were assessed using the nonparametric Mann–Whitney U test, given the small numbers. Categorical variables were assessed using the χ 2 test. The level of statistical significance was set at p < .05. Figures were created with R (R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/).
3. RESULTS
3.1. Phase 1: Baseline assessment
Respondents (n = 41) included Registered Nurses (RNs) 5/41 (12.2%), CRNAs 7/41 (17.1%), Anesthesiologists 9/41 (22%), Critical Care APPs 9/41 (22%), Trauma Fellows 4/41 (9.8%), and Trauma Attendings 7/41 (17.1%). This survey included questions to assess the awareness of Ca++‐specific guidelines. When asked, “What is the recommended practice for calcium administration during MTP at Penn?”, 34 of the 41 respondents (83%) were “unsure” or answered incorrectly. Participants were also queried on the acceptability, appropriateness, and feasibility of administering Ca++ according to the institutional policy. Questions were analyzed using a 5‐point Likert scale with Mean: 1, extremely unwilling; 2, unwilling; 3, neutral; 4, willing; 5, extremely willing “extremely unwilling,” 1; “neutral,” 3; “extremely willing,” 5) with Mean (SD). Respondents were willing to give Ca++ empirically without a blood gas measurement 3. 78 (1.17), agreed that administering 1 g CaCl empirically after four blood products seems suitable 4.15 (0.82), and welcomed administering 1 g of CaCl empirically after four blood products 4.05 (0.89).
3.2. Phase 2: Pre‐post intervention study
Twenty‐five patients met the inclusion criteria during the four‐month pre‐intervention and four‐month post‐intervention periods between July 2021–February 2022 and were included in the final analysis (Supplemental Figure S3). Eleven patients met the final inclusion criteria in the pre‐intervention period, and 14 patients met the inclusion criteria in the post‐intervention period. Of the total included patients, 84% had penetrating injuries, the median Injury Severity Score (ISS) was 18, and the median age was 34. Compared with the pre‐intervention group, post‐intervention patients were more likely to have a lower Abbreviated Injury Scale (AIS) chest score (p = 0.001) (Table 1). Other basic demographics did not vary between groups in a statistically significant way.
TABLE 1.
Patient demographics.
| Pre‐intervention (n = 11) | Post‐intervention (n = 14) | p | |
|---|---|---|---|
| Age (yr) | 29 [25–38] | 36 [25–49] | 0.096 |
| Men (%) | 11 (100.0) | 13 (92.9) | 0.387 |
| Penetrating (%) | 9 (81.8) | 12 (85.7) | 0.802 |
| ISS | 20 [17–28] | 17 [14–23] | 0.292 |
| AIS head | 2 [2–2] | 3 [2–3] | 0.272 |
| AIS chest | 4 [4–5] | 3 [3–3] | 0.001 |
| AIS abdomen | 3 [2–3] | 3 [2–4] | 0.869 |
| Systolic BP (mm Hg) | 101 [85–116] | 88 [80–94] | 0.688 |
| HR (beats/min) | 107 [79–120] | 94 [65–114] | 0.462 |
| Temperature (°C) | 37 [36.3–37.0] | 36.3 [36.2–36.5] | 0.924 |
| GCS | 15 [11–15] | 14 [12.0–15.0] | 0.439 |
| pH | 7.0 [7.1–7.3] | 7.3 [7.2–7.4] | 0.108 |
| Lactate (mg/dL) | 7.0 [4.0–11.3] | 6.3 [5.4–12.9] | 0.959 |
| Hemoglobin (g/dL) | 13.0 [11.0–13.2] | 11.6 [8.7–13.6] | 0.515 |
| PTT (s) | 29.0 [25.5–31.8] | 29.0 [25.4–34.1] | 0.935 |
| PT (s) | 14.0 [11.9–15.0] | 12.8 [11.8–15.0] | 0.411 |
| INR | 1.0 [1.0–1.3] | 1.1 [1.0–1.3] | 0.419 |
| iCa++ (mmol/L) | 1.0 [0.95–1.11] | 1.1 [0.95–1.26] | 0.643 |
Note: All values are presented as n (%) or median [IQR]. Ionized calcium (iCa++) and extreme hypercalcemia were determined based on the first iCa++ value. Timing ranged from within 30 min of arrival to the trauma bay to up to 2 h after arrival to the trauma bay.
Abbreviations: BP, blood pressure; GCS, Glasgow Coma Scale; HR, heart rate; INR, international normalized ratio; IQR, interquartile range; ISS, Injury Severity Score; PT, prothrombin time; PTT, partial thromboplastin time.
Post‐intervention resuscitations were more often adherent to Ca++‐specific guidelines 6/14 (42.9%) compared to resuscitations that took place in the pre‐intervention phase 4/11 (36.4%) reflecting an absolute difference (95% confidence interval) of 6.5% (−11.9% to 24.9%, p = 0.755) (Table 2). Adherence increased to 7/14 (50%) when evaluating Ca++administration at 6 min beyond the administration of the fourth blood product and 8/14 (57.1%) at 15 min beyond the administration of the fourth blood product (Table 2).
TABLE 2.
Adherence to Ca++ guidelines within our institutional Trauma Exsanguination Policy.
| Pre‐intervention (n = 11) | Post‐intervention (n = 14) | Absolute difference with 95% CI | p | |
|---|---|---|---|---|
| Before or immediately following 4th product | 4 (36.4) | 6 (42.9) | 6.5% (−11.9% to 24.9%) | 0.755 |
| Within 6 min of 4th product | 4 (36.4) | 7 (50.0) | 13.6% (−8.7% to 35.8%) | 0.516 |
| Within 10 min of 4th product | 5 (45.4) | 7 (50.0) | 4.6% (−14.6% to 23.8%) | 0.830 |
| Within 15 min of 4th product | 6 (54.5) | 8 (57.1) | 2.6% (−13.7% to 19.0%) | 0.902 |
Note: All values are presented as n (%) except as shown.
In aggregate, iCa++ levels were no different between groups, although there was a slight shift in the distribution toward normal levels post‐intervention (Figure 3). Over the course of the first 6 h of resuscitation, there were no differences in iCa++ levels at 30‐min intervals (Figure 4, Supplemental Table S1). Percent time within normal iCa++ levels could not be assessed due to sparse data (median of 4 [4‐6] iCa++ values out of 12 available time points per patient). Looking at the first iCa++ values, post‐intervention patients trended toward less extreme hypocalcemia (<0.84 mmol/L) compared to pre‐intervention patients (1/14, [7.1%] vs. 3/11 [27.3%], p = 0.187) but slightly more hypercalcemia (4/14, [28.6%] vs. 1/11 [9.1%], p = 0.244) (Table 3).
FIGURE 3.
Aggregate iCa++ in pre‐intervention and post‐intervention groups. Circles indicate raw data values. Boxes represent the median (bar) and interquartile range, with whiskers extending 1.5 times the interquartile range.
FIGURE 4.
Comparison of iCa++ values over time. Data are shown as median with error bars representing interquartile range at each time point. All p = NS.
TABLE 3.
Incidence of hypocalcemia and hypercalcemia.
| Pre‐intervention (n = 11) | Post‐intervention (n = 14) | Absolute difference with 95% CI | p | |
|---|---|---|---|---|
| Hypocalcemia (<1.0 mmol/L) | 5 (45.5) | 6 (42.9) | −2.6% (−22.0% to 16.8%) | 0.902 |
| Extreme Hypocalcemia (<0.84 mmol/L) | 3 (27.3) | 1 (7.1) | −20.2% (−39.7% to −0.8%) | 0.187 |
| Hypercalcemia (≥1.25 mmol/L) | 1 (9.1) | 4 (28.6) | 19.5% (3.2% to 35.8%) | 0.244 |
| Extreme Hypercalcemia (>1.30 mmol/L) | 1 (9.1) | 1 (7.1) | −2.0% (−19.0% to 15.0%) | 0.866 |
Note: All values are presented as n (%) except as shown. Values were determined based on the first Ca++ value. The timing of this assessment ranged from 30 min to up to 2 h after arrival to the trauma bay.
4. DISCUSSION
There are mixed opinions about whether to empirically give Ca++ during blood product resuscitations. The Joint Trauma System recommends giving 1 g of Ca++ IV empirically after the first unit of blood and every four subsequent units for patients in hemorrhagic shock. 22 However, other studies have noted that aggressive calcium supplementation can be risky and result in hyperglycemia. 23 In the Phase 1 baseline assessment of guideline awareness, we discovered that most respondents at our institution were comfortable with the idea of empirically administering Ca++ during blood product resuscitation for severely injured patients. Interestingly, we also discovered that most respondents were unaware of the exact institutional guidelines for Ca++ supplementation. It is difficult to say why exactly this was the case, but these Phase 1 findings informed our nudge design, prompting us to highlight the institutional guidelines as both text and as a visual representation on the sign. The iterative design process was crucial, and having input from a variety of clinicians provided differing yet equally important perspectives on where to place the signs, what kind of visuals to include, and what language would best motivate clinicians.
Nudges can take many forms and produce varying levels of behavioral impact (Figure 1). 9 , 13 For example, information framing, which passively influences members of the care team, exerts a lighter influence than guiding choices through default settings. 9 , 13 Examples of information framing include sending email reminders of existing guidelines to clinicians and strategically placing signs with information in various phases of care (i.e., trauma bay, OR, ICU) as we did in this study. In contrast, enabling active choice between two options or using default orders directly influences clinicians at the time of decision‐making. While the most effective nudges tend to be the ones that limit the set of choices or change the default options, they are not always feasible, whether due to time limitations, barriers to integration in the EHR, etc. For that reason, it is important to assess feasibility and to optimize the nudge for the environment. 11 , 24
Mertens et al. categorized interventions as (1) decision information (i.e., translation, visibility, social reference), (2) decision structure (i.e., default, effort, composition, consequence), and (3) decision assistance (i.e., reminder, commitment). 25 In their meta‐analysis, the authors found that all three categories were effective in inducing statistically significant behavior change; however, they noted varying effect sizes, with the average effect sizes ranging from d = 0.31 to 0.55. Given limited resources and time, an intervention altering decision structure was not possible for this pre‐post study. Instead, we created an intervention that increased the availability, comprehensibility, and relevance of information. Nudges displayed as signs have been previously used to prompt behavior in the healthcare setting. For example, signs have been a common and affordable intervention used for improving hand hygiene practices and compliance. 26 , 27 , 28 In this Phase 2 pre‐post study of severely injured patients at an urban Level I trauma center, a nudge sign providing information on Ca++‐specific guidelines during blood product resuscitation was selected as a simple and feasible option. While the sign did not lead to a statistically significant difference in the pre‐intervention and post‐intervention phases, we did see a modest increase in adherence to institutional guidelines. It is worth noting that the size of our study was small; thus, the study was not powered to detect a significant difference.
Future efforts might explore improving adherence using a more potent nudge via the EHR. Several studies have explored the utilization and effectiveness of interruptive alerts and default orders in the EHR. 29 , 30 , 31 , 32 , 33 In a stepped‐wedge randomized clinical trial by Mehta et al., embedding a default order for screening EHR substantially increased ordering and completion of testing compared to a conventional interruptive alert. 30 Moving forward, we might consider integrating a default order for Ca++ into the EHR in accordance with the institutional guidelines (i.e., following the scanning and administration of four blood products). We might also consider implementing an accountable justification strategy in the EHR, requiring the clinician to justify their rationale for not administering Ca++ prior to being able to order or administer additional blood products and adjuncts. If we pursue an EHR integrated nudge in the future, it is worth considering how having the nudge in the EHR would affect who bears the responsibility of reminding the team to administer Ca++. Blood product resuscitations for massively transfused patients at our institution require collaboration between nurses, CRNAs, surgeons, and anesthesiologists as the patient is moved through all three phases of care (trauma bay to OR to ICU). In the trauma bay, orders are approved by the physician trauma team leader, but any team member can prompt or suggest interventions; in the trauma OR, the Anesthesiologist and CRNAs work together to complete the resuscitation; in the ICU, the on‐call Fellow and APPs write the orders, but all team members can provide input. Therefore, everyone is or can be responsible for adherence to the Ca++‐specific guidelines unless a leader or champion is assigned. Having the nudge in the EHR may mean that the onus would be on the team member interfacing with the EHR to ensure that the team is thinking about Ca++ administration. This is an important consideration when designing and optimizing nudges for a given clinical environment.
5. LIMITATIONS
Our study has several limitations to consider. First, these results are from a single, urban academic health system at a Level I trauma center with high‐volume MTs; thus, the results may not be generalizable to all settings. In the future, we might consider identifying another center to use as a control. Second, it is possible that the REDCap survey that informed our nudge design primed the clinicians likely to be involved in the resuscitations over the eight‐month study period. This may have increased the likelihood of adhering to policy in the pre‐intervention phase and/or post‐intervention phase. Given this, there may be value in evaluating the perception of the nudge post hoc. Thirdly, it is impossible to say whether the nudge signs, while there were several placed in the trauma bay, OR, and ICU, were seen and read by the care team members making decisions during the blood product resuscitations. As this was a pilot study, and since our institution sees one to two MTPs per week, the sample size was small. As a result, the study was underpowered to detect small improvements in outcomes. Compliance would have had to improve by a large amount in the post‐intervention phase for us to detect a significant difference. Lastly, more aggressive nudges, such as a default order in the EMR, require more time. One reason for choosing the simplest nudge was our time limitation.
6. CONCLUSION
In this small study, signs reminding clinicians to administer Ca++ were associated with a modest, non‐statistically significant increase in adherence to institutional guidelines and a slight shift in the distribution of iCa++ values toward normal. Future work to optimize blood product resuscitation should evaluate nudges in more acutely injured patients and should assess more potent nudges that guide choice with active prompts or default orders in these high‐acuity situations.
CONFLICT OF INTEREST STATEMENT
Dr. Beidas is principal at Implementation Science & Practice, LLC. She is currently an appointed member of the National Advisory Mental Health Council and the National Academies of Science, Engineering, and Medicine study, “Blueprint for a national prevention infrastructure for behavioral health disorders.” She serves on the scientific advisory board for AIM Youth Mental Health Foundation and the Klingenstein Third Generation Foundation. She has received consulting fees from United Behavioral Health and OptumLabs. She previously served on the scientific and advisory board for Optum Behavioral Health and has received royalties from Oxford University Press. All activities are outside of the submitted work. The remaining authors do not have any other personal or institutional interest regarding the authorship and/or publications of this article.
FUNDING INFORMATION
This work was supported by the Measey Scholars Program at the University of Pennsylvania.
Supporting information
Figure S1. Calcium Administration During MTP baseline assessment survey.
Figure S2. Iterations of nudge intervention.
Figure S3. Study flow diagram.
Table S1. iCa++ values at 30‐minute intervals.
ACKNOWLEDGMENTS
The authors gratefully acknowledge the Penn Nudge Unit for their guidance and support throughout this project.
Schmulevich D, Joergensen SM, Zone AI, Bishop KE, Morlok AP, Colyar TA, et al. The simplest solution may be good, but is it good enough? Evaluating the effect of a nudge to administer calcium during blood product resuscitation for traumatic injuries. Transfusion. 2025;65(Suppl. 1):S113–S122. 10.1111/trf.18180
This work was presented at the American College of Surgeons Quality and Safety Conference in Minneapolis, MN, July 10–13, 2023.
REFERENCES
- 1. Ditzel RM, Anderson JL, Eisenhart WJ, Rankin CJ, DeFeo DR, Oak S, et al. A review of transfusion‐ and trauma‐induced hypocalcemia: is it time to change the lethal triad to the lethal diamond? J Trauma Acute Care Surg. 2020;88(3):434–439. [DOI] [PubMed] [Google Scholar]
- 2. Webster S, Todd S, Redhead J, Wright C. Ionised calcium levels in major trauma patients who received blood in the emergency department. Emerg Med J. 2016;33(8):569–572. [DOI] [PubMed] [Google Scholar]
- 3. Giancarelli A, Birrer KL, Alban RF, Hobbs BP, Liu‐DeRyke X. Hypocalcemia in trauma patients receiving massive transfusion. J Surg Res. 2016;202(1):182–187. [DOI] [PubMed] [Google Scholar]
- 4. MacKay EJ, Stubna MD, Holena DN, Reilly PM, Seamon MJ, Smith BP, et al. Abnormal calcium levels during trauma resuscitation are associated with increased mortality, increased blood product use, and greater hospital resource consumption: a pilot investigation. Anesth Analg. 2017;125(3):895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Kronstedt S, Roberts N, Ditzel R, Elder J, Steen A, Thompson K, et al. Hypocalcemia as a predictor of mortality and transfusion. A scoping review of hypocalcemia in trauma and hemostatic resuscitation. Transfusion. 2022;62(S1):S158–S166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kyle T, Greaves I, Beynon A, Whittaker V, Brewer M, Smith J. Ionised calcium levels in major trauma patients who received blood en route to a military medical treatment facility. Emerg Med J. 2018;35(3):176–179. [DOI] [PubMed] [Google Scholar]
- 7. Moore HB, Tessmer MT, Moore EE, Sperry JL, Cohen MJ, Chapman MP, et al. Forgot calcium? Admission ionized‐calcium in two civilian randomized controlled trials of prehospital plasma for traumatic hemorrhagic shock. J Trauma Acute Care Surg. 2020;88(5):588–596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Volpp K, Delgado MK. Behavioral nudges are used widely to steer clinicians and patients alike. NEJM Catal Innov Care Deliv. 2023;4(6). https://catalyst.nejm.org/doi/full/10.1056/CAT.23.0125 [Google Scholar]
- 9. Last BS, Buttenheim AM, Timon CE, Mitra N, Beidas RS. Systematic review of clinician‐directed nudges in healthcare contexts. BMJ Open. 2021;11(7):e048801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Patel MS, Volpp KG, Asch DA. Nudge units to improve the delivery of health care. N Engl J Med. 2018;378(3):214–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Harrison JD, Patel MS. Designing nudges for success in health care. AMA J Ethics. 2020;22(9):796–801. [DOI] [PubMed] [Google Scholar]
- 12. Thaler R, Sunstein C. Nudge: improving decisions about health, wealth and happiness. New Haven: Yale University Press; 2008. p. 293. [Google Scholar]
- 13. Schmulevich D, Hynes AM, Murali S, Benjamin AJ, Cannon JW. Optimizing damage control resuscitation through early patient identification and real‐time performance improvement. Transfusion. 2024;64(8):1551–1561. [DOI] [PubMed] [Google Scholar]
- 14. Ogrinc G, Davies L, Goodman D, Batalden P, Davidoff F, Stevens D. SQUIRE 2.0 (standards for QUality improvement reporting excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf. 2016;25(12):986–992. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Weiner BJ, Lewis CC, Stanick C, Powell BJ, Dorsey CN, Clary AS, et al. Psychometric assessment of three newly developed implementation outcome measures. Implement Sci. 2017;12(1):108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Research electronic data capture (REDCap) . A metadata‐driven methodology and workflow process for providing translational research informatics support – ScienceDirect. 2024. Available from: https://www.sciencedirect.com/science/article/pii/S1532046408001226 [DOI] [PMC free article] [PubMed]
- 17. Harris PA, Taylor R, Minor BL, Elliott V, Fernandez M, O'Neal L, et al. The REDCap consortium: building an international community of software platform partners. J Biomed Inform. 2019;95:103208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Michie S, van Stralen MM, West R. The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci. 2011;6(1):42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. BIT . EAST Framework: Four Simple Ways to Apply Behavioural Insights. 2025. Available from: https://www.bi.team/publications/east‐four‐simple‐ways‐to‐apply‐behavioural‐insights/
- 20. Fogg B. A behavioral model for persuasive design. In Proceedings of the 4th international Conference on Persuasive Technology (Persuasive '09). Association for Computing Machinery. New York, NY. 2009; Article 40:1‐7. Available from: 10.1145/1541948.1541999 [DOI] [Google Scholar]
- 21. Holtrop JS, Estabrooks PA, Gaglio B, Harden SM, Kessler RS, King DK, et al. Understanding and applying the RE‐AIM framework: clarifications and resources. J Clin Transl Sci. 2021;5(1):e126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Damage Control Resuscitation – Joint Trauma System. 2019. Available from: https://jts.health.mil/index.cfm/PI_CPGs/pi
- 23. Helsloot D, Fitzgerald M, Lefering R, Verelst S, Missant C, the TraumaRegister DGU . Trauma‐induced disturbances in ionized calcium levels correlate parabolically with coagulopathy, transfusion, and mortality: a multicentre cohort analysis from the TraumaRegister DGU®. Crit Care. 2023;27(1):267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Schmulevich D, Cacchione PZ, Holland S, Quinlan K, Hinkle A, Meador C, et al. Optimizing a decision support system for damage‐control resuscitation using mixed methods human factors analysis. J Trauma Acute Care Surg. 2021;91(2S):S154–S161. [DOI] [PubMed] [Google Scholar]
- 25. Mertens S, Herberz M, Hahnel UJJ, Brosch T. The effectiveness of nudging: a meta‐analysis of choice architecture interventions across behavioral domains. Proc Natl Acad Sci USA. 2022;119(1):e2107346118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Caris MG, Labuschagne HA, Dekker M, Kramer MHH, van Agtmael MA, Vandenbroucke‐Grauls CMJE. Nudging to improve hand hygiene. J Hosp Infect. 2018;98(4):352–358. [DOI] [PubMed] [Google Scholar]
- 27. Lawson A, Vaganay‐Miller M. The effectiveness of a poster intervention on hand hygiene practice and compliance when using public restrooms in a university setting. Int J Environ Res Public Health. 2019;16(24):5036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Harrison BL, Ogara C, Gladstone M, Carrol ED, Dusabe‐Richards J, Medina‐Lara A, et al. “We have to clean ourselves to ensure that our children are healthy and beautiful”: findings from a qualitative assessment of a hand hygiene poster in rural Uganda. BMC Public Health. 2019;19(1):1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Kerlin MP, Small D, Fuchs BD, Mikkelsen ME, Wang W, Tran T, et al. Implementing nudges to promote utilization of low tidal volume ventilation (INPUT): a stepped‐wedge, hybrid type III trial of strategies to improve evidence‐based mechanical ventilation management. Implement Sci. 2021;16(1):78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Mehta SJ, Torgersen J, Small DS, Mallozzi CP, McGreevey JD, Rareshide CAL, et al. Effect of a default order vs an alert in the electronic health record on hepatitis C virus screening among hospitalized patients: a stepped‐wedge randomized clinical trial. JAMA Netw Open. 2022;5(3):e222427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Delgado MK, Shofer FS, Patel MS, Halpern S, Edwards C, Meisel ZF, et al. Association between electronic medical record implementation of default opioid prescription quantities and prescribing behavior in two emergency departments. J Gen Intern Med. 2018;33(4):409–411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Halpern SD, Ubel PA, Asch DA. Harnessing the power of default options to improve health care. N Engl J Med. 2007;357(13):1340–1344. [DOI] [PubMed] [Google Scholar]
- 33. Halpern SD. Using default options and other nudges to improve critical care. Crit Care Med. 2018;46(3):460–464. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Calcium Administration During MTP baseline assessment survey.
Figure S2. Iterations of nudge intervention.
Figure S3. Study flow diagram.
Table S1. iCa++ values at 30‐minute intervals.