View as “PDF version”
One of our most popular columns has been our October 16, 2012 Patient Safety Tip of the Week “What is the Evidence on Double Checks?”. In that column and several subsequent ones (listed below) we pointed out the lack of high-quality studies addressing the efficacy of double checks, though we continue to recommend them in certain circumstances, with the caveat that they are not infallible.
In our August 27, 2019 Patient Safety Tip of the Week “Double Check on Double Checks” we discussed a systematic review of studies evaluating evidence of the effectiveness of double checking to reduce medication administration errors (Koyama 2020). Those authors again cited the paucity of high-quality evidence and concluded that there is insufficient evidence that double versus single checking of medication administration is associated with lower rates of medication administration errors or reduced harm. They called for higher-quality studies to determine if, and in what context (eg, drug type, setting), double checking produces sufficient benefits in patient safety to warrant the considerable resources required.
Now another new study raises questions about how double checks are actually done in practice and what their real impact is on medication errors. Westbrook and colleagues (Westbrook 2020) published results of direct observation of 298 nurses administering 5140 medication doses to 1523 patients in a pediatric hospital. Details of administrations and double-checking were recorded by independent observers in real time. For each dose administration, observers recorded whether nurses performed a truly independent double check, or one that was “primed” by the other nurse, or one that was incomplete, or no double-check.
Processes at this pediatric hospital were somewhat unusual in that hospital policy required independent double-checking by registered nurses (RN’s) for all medication administrations except for a select group. 69.3% of all medication administrations required double-checking according to hospital policy. Time taken (for the checking nurse alone) to double-check averaged 6.4 minutes (7.9 minutes for IV and 5.5 min for non-IV administrations).
The researchers found no association between double-checking and the occurrence of a medication administration error (OR 0.89) or the potential severity of MAE’s (OR 0.86). For double-checked administrations the error rate was 72/100 administrations and for those not double-checked 71/100.
For those medication administrations where double-checking was mandated, observations were:
For those medications for which double-checking was optional (such as topical creams, vitamins, oral antibiotics, inhaled medications), nurses did double checks in 26.4%. However, in only 1.7% was an independent double-check performed.
Essentially what they found was that mandatory double-checking conferred no additional safety benefit compared with single-checking. Nurses were highly compliant but almost never did truly independent double-checks. Instead, most double checks resulted from priming. (A “primed” double check is one in which a nurse shares information which may influence the checking nurse).
Interestingly, for those medication administrations where double-checking was optional but used, there was a significantly lower odds of the occurrence of a medication administration error (OR 0.71) and MAE severity (OR 0.75).
Westbrook and colleagues do discuss some of the factors that render double-checking potentially ineffective, including diffusion of responsibility, ritualization, deference to those with higher authority, false reassurance, and rushing the process.
Reduction or diffusion of responsibility refers to the complacency that tends to occur when someone feels that someone else will catch any mistakes that they made. We know from other industries that the error rate when a supervisor checks someone else’s work may be 10% or higher, though that usually does not involve truly independent double checks. More of a diffusion of responsibility today is deferring to technology (“the computer says it’s ok so it must be ok”). Somewhat related to this concept is one of the most common cognitive biases, confirmation bias, in which we accept a response or evidence as confirming our assessment even in the presence of disconfirming evidence.
Westbrook et al. go on to discuss the costs of double-checking, in both human and financial terms. They do note, appropriately, that medications for pediatric patients are often more complex than for adults, especially since dosing may be based upon patient weights (or other parameters, like age or body surface area). However, many of these same considerations would apply to certain adult populations, like an oncology ward.
Our biggest problem with this study is that double checks were mandated for such a high percentage of their medication administrations. We know few hospitals (or other healthcare venues) where double checks are so widely mandated. That certainly is a factor that probably leads to reducing the importance of the double check in the minds of many healthcare professionals. It likely tends to make the double-check perfunctory in many cases. We usually recommend double checks be required only for a few high-alert medications or situations where the probability of error is very high, such as those involving calculations.
In our July 2020 What's New in the Patient Safety World column “Medication Dosage Miscalculations” we discussed typical human error rates for a variety of industries. Error rates in simply “doing arithmetic wrongly” range from 0.01 to 0.03. That’s why having a second person independently perform a calculation can be valuable.
In our August 27, 2019 Patient Safety Tip of the Week “Double Check on Double Checks” we cited ISMP’s practical article on double checks (ISMP 2019a). Though it acknowledges the paucity of high-quality evidence on double checks, it summarizes findings from many of the observational studies that have demonstrated a benefit from correctly performed double checks. ISMP remains an advocate of truly independent double checks in special circumstances. Notably, it does not recommend them for all high alert medications. It cites the time-consuming nature of independent double checks as a downside.
ISMP suggests your decision about which processes should require an independent double check should be based upon analysis of 4 key considerations:
1. Processes and medications that pose the greatest risk of harm if an error occurs (e.g., intravenous [IV]/epidural opioids, IV insulin, IV heparin, IV chemotherapy)
2. The primary reason for the independent double check (what you are trying to catch) and what specifically needs to be verified to achieve that goal
3. Whether an independent double check is the best strategy to detect a specific risk or prevent a specific error
4. How the independent double check fits in with other risk-reduction strategies that might address the same or a similar safety concern
ISMP notes you might use findings from hazard and event analysis, or a FMEA (Failure mode and effects analysis) to help inform your decisions about what processes should require double checks. We’ve noted in our several columns on accidents with NMBA’s (neuromuscular blocking agents) that, if you store these in an automated dispensing cabinet, you need a way of verifying that the patient is intubated and ventilated (or that the NMBA is being used to facilitate intubation and ventilation) before dispensing the NMBA from the ADC. But we’ve also recommended that a second qualified healthcare professional verify that dispensing of the NMBA from the ADC is correct. But you need to be wary that the second person is truly acting independently and not simply ”signing off” complacently. How many times have you seen a transfusion reaction occur after two individuals supposedly “verified” the unit of blood was correct for that patient?
ISMP notes that bedside barcode verification may be more reliable than double checks for getting the correct patient, medication, and dose. But that independent double check at the bedside may be a better strategy for avoiding infusion pump programming errors and line confusion errors.
ISMP has always stressed that the double check needs to be a truly independent double check. That means that two qualified healthcare professionals need to assess the question(s) separately and only compare their conclusions after each has completed their assessments.
One study (Douglass 2018) randomized emergency department nurses into single- and double-check groups in exercises on a simulated patient. Errors were intentionally introduced into the simulation, including weight-based dosage errors and wrong medication vial errors. In the single-check group, 9% of nurses detected the weight-based dosage error compared with 33% of nurses in the double-check group (odds ratio 5.0) Fifty-four percent of nurses in the single-check group detected the wrong vial error compared with 100% of nurses in the double-check group (odds ratio 19.9). Overall, the researchers found that use of a double check increased certain error detection rates in some circumstances, but not others. Both techniques missed many errors. In some cases, the second nurse actually dissuaded the first nurse from acting on the error. In cases in which the second nurse dissuaded the first confused nurse and the error was missed, the second nurse tended to be older and more experienced (so likely “deference to authority”). They also noted in several cases that the second nurse rushed the first nurse. That study was conducted in a relatively controlled environment. One could imagine that such “rushing” might be more prevalent in high intensity environments like the ER, OR, or ICU.
The authors did note that use of a double check may promote cooperation between the 2 nurses to detect an error. They found that, in regard to the first nurse, the double-check condition may have made him/her more comfortable engaging the second nurse in analyzing the confusing weight-based order. If the first nurse was confused, he/she always involved the second nurse during the double-check condition. In contrast, in the single-check condition, the first nurse would often dismiss concerns without involving the other nurse.
In our April 14, 2020 Patient Safety Tip of the Week “Patient Safety Tidbits for the COVID-19 Pandemic” we noted the potential hazards of performing a double check in the COVID-19 era. ISMP (ISMP 2020a) recently addressed the need to balance the benefits of double checks versus the risks of exposure and the need to conserve personal protective equipment (PPE) when doing double checks in the COVID-19 environment. They found that most organizations are establishing ways to conduct critical parts of independent double checks without entering a patient’s room. For example, a hard stop in EHR’s requiring dual documentation of verification before proceeding now reflects only those components of the check that can be accomplished outside the patient’s room. They also describe how some hospitals became innovative. Where infusion pumps remain in the patient’s room, the nurse who enters the room takes a picture of the pump screen using a mobile phone device left in the room, and sends the picture to a nurse outside the room via a secure messaging system. This allows most components of the independent double check to occur.
In our March 2020 What's New in the Patient Safety World column “ISMP Smart Infusion Pump Guidelines” we noted that ISMP’s Smart Infusion Pump Guidelines (ISMP 2020b) recommend performance of double checks at certain points in the clinical workflow. When starting selected facility-defined high-alert medication infusions and at additional facility-defined steps (e.g., change of shift/handoffs, change in the rate/dose of infusion, change in bag/bottle/syringe) require that a double check be performed and documented to verify the following before starting the infusion:
ISMP notes that technology (eg. barcoding) is the preferred method for double checking most of these items but use of a second practitioner is necessary to perform and document an independent double check for line attachment.
Double checks may also be important when other medication safety techniques are not available. In our November 19, 2019 Patient Safety Tip of the Week “An Astonishing Gap in Medication Safety” we discussed several incidents in which community vaccination programs inadvertently gave dangerous medications to recipients rather than the intended vaccines. We noted that one of our most important medication safety tools we use in the hospital setting – barcoding – was not available in most of those cases. Given that there was no barcoding as a layer of defense, we suggested falling back on an old technique – the independent double check. Perhaps in settings lacking barcoding it may make sense to have a second person confirm the vaccine/medication being given. In at least one of the incidents there was only one healthcare professional present. In hospitals, having a second set of eyes is useful even if independent double checks are not being used. Typically, a pharmacist prepares and dispenses the medication and then a nurse checks it before administering it. In the vaccine incident there was only one person preparing and administering the vaccine.
In our January 1, 2019 Patient Safety Tip of the Week “More on Automated Dispensing Cabinet (ADC) Safety” we also recommended that independent double checks should be required for ADC overrides. Failure to incorporate double checks into the ADC override process was an important contributing factor in some of the serious accidents involving neuromuscular blocking agents (NMBA’s). But you need to be wary that the second person is truly acting independently and not simply ”signing off” complacently.
In our August 4, 2020 Patient Safety Tip of the Week “Intravenous Issues” we discussed ISMP Canada’s (ISMP Canada 2020) analysis of over 1500 incidents related to intravenous issues. One very important theme they found was over-reliance on technology, such as smart pump and bar-code technology (see our Patient Safety Tips of the Week for August 23, 2016 “ISMP Canada: Automation Bias and Automation Complacency” and June 11, 2019 “ISMP’s Grissinger on Overreliance on Technology”). They found that failure to perform an independent double check before IV medication administration (to verify and document the patient’s name, the drug and its concentration, the prescribed infusion rate, the line attachments and labels), was a key contributing factor in many of these incidents.
The “primed” double check prevalent in the Westbrook study clearly differs from the truly independent double check which we are usually referring to. Pfeiffer et al. point out that the definition of double checks is varied in much of the literature (Pfeiffer 2020) and offer a framework for future research on double checks. ISMP also pointed out the lack of standardization of terms in the literature.
Despite our concerns about the applicability of the Westbrook study to other hospitals, the study is important in several regards:
Lastly, we always take the opportunity to revive what seems to be a forgotten form of double check in the era of computerized medicine, the “mental” double check. When we talk about double checks, we are usually talking about two separate individuals independently checking something. The mental double check simply involves one person employing a separate cognitive process to double check. Or, it may be a nurse or pharmacist doing a quick mental double check of an order from a physician. In the “old days”, when a nurse looked at an order for a medication, he/she would do a quick mental calculation of the ordered dose and decide if the result fell into a reasonable range (this would be considered a “plausibility check” in the Pfeiffer framework). In the era of CPOE and e-prescribing, it’s often assumed that whatever the computer says is correct and the step of “mental double checking” seems to be a lost art.
We agree with several points made by ISMP (ISMP 2019b) after recent publication of studies that seemingly downplay the effectiveness of double checks:
Every time we come away from doing a column on double checks, we have great feelings of ambivalence. The lack of a solid evidence base remains disconcerting. We lament that the process of double checking is flawed and may lead to a false sense of security. But we hold out hope that, if double checks are limited to just a few very critical circumstances and are done in a truly independent fashion, we might see fewer errors.
Some of our other columns on double checks:
January 2010 “ISMP Article on Double Checks”
October 26, 2010 “Confirming Medications During Anesthesia”
October 16, 2012 “What is the Evidence on Double Checks?”
December 9, 2014 “More Trouble with NMBA’s”
April 19, 2016 “Independent Double Checks and Oral Chemotherapy”
December 11, 2018 “Another NMBA Accident”
January 1, 2019 “More on Automated Dispensing Cabinet (ADC) Safety”
March 5, 2019 “Infusion Pump Problems”
August 27, 2019 “Double Check on Double Checks”
November 19, 2019 “An Astonishing Gap in Medication Safety”
April 14, 2020 “Patient Safety Tidbits for the COVID-19 Pandemic”
March 2020 “ISMP Smart Infusion Pump Guidelines”
August 4, 2020 “Intravenous Issues”
KoyamaMaddoxLi, et al. Effectiveness of double checking to reduce medication administration errors: a systematic review.
Westbrook JI, Li L, Raban MZ, et al Associations between double-checking and medication administration errors: a direct observational study of paediatric inpatients. BMJ Quality & Safety Published Online First: 07 August 2020
Douglass AM, Elder J, Watson R , et al. A randomized controlled trial on the effect of a double check on the detection of medication errors. Ann Emerg Med 2018: 71(1): 74-82.e1
ISMP (Institute for Safe Medication Practices). Independent Double Checks: Worth the Effort if Used Judiciously and Properly. ISMP Medication Safety Alert! Acute Care Edition 2019. June 6, 2019
ISMP (Institute for Safe Medication Practices). Suspending independent double checks. ISMP Medication Safety Alert! Acute Care Edition 2020; Special Edition 25(7): 1-2 April 9, 2020
ISMP (Institute for Safe Medication Practices). Guidelines for Optimizing Safe Implementation and Use of Smart Infusion Pumps. ISMP 2020; February 10, 2020
ISMP Canada. Intravenous Medication Safety: A Multi-Incident Analysis. ISMP Canada Safety Bulletin 2020; 20(7): 1-4 July 16, 2020
Pfeiffer Y, Zimmerman C, Schwappach DLB. What are we doing when we double check? BMJ Quality & Safety 2020; Published Online First: 18 February 2020
ISMP (Institute for Safe Medication Practices). Published Review of Independent Double Checks Shouldn’t Dissuade Providers from Using Them Judiciously. ISMP Medication Safety Alert! Acute Care Edition 2020; September 26, 2019
Print “PDF version”