Chapter 12. Procurement Support Case Study
Contents
1. Buttons on second screen should be carefully placed so that selecting a button quickly twice in a row either a selects nothing i e no button underneath on second screen or b selects a safe second action that is easy to exit from if desired No clear way to exit the Bolus Reversible Provide a clear exit add Exit key to bottom of Dose screen without giving a actions screen with a screen asking user to confirm that patient a bolus The user must they do not want proceed with a bolus Allow the press Cancel twice to exit The Users in control user to enter 0 mg as a dose and again confirm that pump will not accept 0 mg they are not giving any dose before asking user to Minimize close and lock door memory load 195 Section 12 4 2 Usability Testing Following the heuristic analysis of each of the three PCA pumps the human factors specialist biomedical technology professional and acute care pain nurse prepared for and conducted usability testing Chapter 8 They conducted the usability test in an empty patient room in the general ward at the healthcare organization The environment doubled as both a ward environment and a post anaesthetic care unit Figure 37 Figure 37 Usability testing environment for PCA usability testing Section 12 4 2 1 In Preparation for Usability Testing Data gathered during the process of 1 establishing the needs and wants of the user observations 2 conducting the task analysis2. operative pain 6 10 at rest and hasn t tried moving He needs an IV PCA started His morphine orders are in his chart I m just going to be over there with Ms Wu so holler if you need anything but hopefully you ll be okay on your own Participant Sure Nurse Actor Great thanks l Il be back in a little bit Figure 40 Script developed based on the usability test scenarios Pre and post questionnaires Figure 41 and Figure 42 were then designed in order to collect direct feedback and information from each representative user group 199 Demographics 1 What best describes your role in the hospital 0 Medical doctor O Registered nurse QO Other 2 What is your age 18 29 years old 30 39 years old 40 49 years old 50 64 years old 65 years old and over oOoooo How long have you been in your current role Less than a year 1to 4 years 5 to 9 years 10 to 20 years More than 20 years 3 a Q Q a m OR PACU General Ward ER SICU PICU Outpatient Other OOOOcCOcO Which unit s do you typically work in check all that apply 5 How long have you been working in your current clinical area 0 Less than a year 0 1to4 years O 5to9 years 0 10 to 20 years 0 More than 20 years Knowledge and Experience 1 Do you know what a PCA pump is O Yes O No Have you ever used a PCA pump before Yes No Not sure Have you ever received training about how to use a PCA pump Ye
3. Finally assistance may be required to upload customized drug library software and to download specified drug libraries into the test pumps 1Heuristic Evaluation A heuristic evaluation is a structured comparison of a user interface or product design with established human factors design principles or heuristics 3 This method is generally carried out by one or more usability experts A heuristic evaluation is helpful for identifying design issues likely to lead to use errors 2Usability Testing A usability test is a human factors evaluation method where representative end users interact with a system in a simulated environment Data from representative end users can be collected about safety appropriateness and ease of use of a system prior to its selection and implementation in the real world Figure 36 Description of human factors methods to be applied to shortlisted technologies Section 12 4 Evaluation The PIP team received four responses to the RFP After reviewing each vendors proposal against the RFP and re examining the needs and wants of users established earlier in the FPIP it was found that three out of the four vendor submissions met the requirements outlined in the RFP Thus the PCA pumps from each of these three vendors were shortlisted and moved forward into the evaluation phase of the PIP while the PCA pump from the vendor that did not meet the requirements did not progress any further as part of the 4F
4. The pump was not running but the user thought it was infusion depending on the concentration to dose ratio Patient does not receive medication Patient does not receive the increased dosage Patient receives an unintended dose that is not included in the 4 hour limit and is not prescribed Patient receives an unintended dose that is not included in the 4 hour limit and is not prescribed Patient does not receive medication Issue It is not clear to users what Anaesthesia Mode is or how it affects the pump Pump buttons are difficult to press Visual parallax effect on pump Description of Error User incorrectly selects Options when trying to review the setup parameters User selects Anesthesia Mode and enables it without knowing what it does User entered a dose of 0 2 mg but the pump only registered 2 because the buttons are hard to press User noticed and mitigated the error User selected the wrong drug three times because ofa Impact to Safety Patient could receive an overdose since medication limits are broadened in this mode If a dose of 2 mg is inside the programming limit a 10 fold overdose would occur each time the patient requested a dose Pump programmed incorrectly resulting in over under 204 screen parallax effect on the screen If infusion the user is not standing directly in front of and at the same height as the screen the buttons do not li
5. after it is connected to the patient while trying to start the pump Impact to Safety Patient does not receive medication but user thinks it s running Patient receives an unintended dose that is not included in the 4 hour limit and is not prescribed Patient receives an unintended dose that is not included in the 4 hour limit and is not prescribed Mitigating Strategy No effective mitigating strategy from within the organization Recommend that vendor review the alerts No effective mitigating strategy from within the organization Recommend that vendor reviews the menu structure of the user options No effective mitigating strategy from within the organization Recommend that vendor review technology design to ensure users are aware of which modality has been selected 206 Issue It is not clear to users what Anaesthesia Mode is or how it affects the pump Pump buttons are difficult to press Visual parallax effect on pump screen Pump response to key press is delayed Description of Error User incorrectly selects Options when trying to review the setup parameters User selects Anesthesia Mode and enables it without knowing what it does User entered a dose of 0 2 mg but the pump only registered 2 because the buttons are hard to press User noticed and mitigated the error User selected the wrong drug three times because of a parallax effect
6. representatives with the following areas of expertise Table 24 PCA Procurement Decision Team Direct Stakeholders Indirect Stakeholders Anaesthesia Biomedical technology Post anaesthetic care unit Human Factors General ward Clinical education Pain management Risk management Pharmacy Informatics Cleaning and maintenance Legal Central stores Hospital administration The clinical representatives included on the PIP team included a combination of front line staff and managers In many cases it is also advantageous to include a patient representative The list of stakeholders above included both direct and direct stakeholders Section 12 2 Establish Needs and Wants Once the procurement team was established the biomedical technology professional human factors specialist and post anaesthesia care unit nurse on the 4FPIP team conducted observations Chapter 4 of the current PCA pumps in use to learn more about the types of users and how those staff were interacting with the pump Informal interviews Chapter 5 were held with staff as observations were being done to gather information about the types of tasks they perform features they rely on features they wish they had and any general frustrations with their current devices A list of all user groups who interacted with the current PCA pump was created and a task analysis Chapter 6 was completed to describe the tasks conducted with the current pump by each user group Follo
7. 0 5 mg mL and meperidine 10 mg ml The nurse in this case was required to reject the first default option and select the second option but most likely selected the first option The concentration selection error is more likely in this particular case because the hospital stocks 1mg mL morphine and 5mg mL morphine but the 1 mg mL concentration was unavailable so the nurse had to obtain a 5mg mL drug container It is likely that 1mg mL morphine was the standard concentration for obstetric patients and the nurse was conditioned to programming the pump for 1mg mL morphine although these facts were not confirmed in the reports What makes Danielle s death even more upsetting is that this particular model of PCA pump had been implicated in several other morphine concentration programming errors due to incorrect selection of the default concentration ultimately leading to several patient deaths Further three years prior to Danielle s death a medical device alert was issued for the pump because of the default concentration issue 85 87 Since then human factors researchers have estimated this model of PCA pump which is no longer available for purchase was responsible for between 65 and 667 deaths due to programming errors 84 A human factors analysis of the user interface found the PCA pump programming sequence to be complex and confusing requiring as many as 27 distinct programming steps for proper operation A redesigned interface propos
8. Chapter 12 Procurement Support Case Study Biomedical technology professionals are often involved when a healthcare organization decides to procure a new medical technology As such technical professionals in these roles are well positioned to apply human factors methods throughout the procurement process to ensure that whatever technology is chosen by the healthcare organization fits well with the people who will use it given the context of use To illustrate how biomedical technology professionals can approach procurement using human factors methods the human factors informed procurement and implementation process 4 PIP will be applied to a case study see Case Study 2 that occurred in the United States in 2000 Human Factors Not Considered in Design of Patient Controlled Analgesia Pump On February 27 2000 at 2 34 AM nineteen year old Danielle McCray was admitted to the Tallahassee Memorial Hospital in Florida to have her baby After a long labour a healthy baby girl was delivered by Caesarean section at approximately 4 30 PM About two hours later Danielle complained of pain and at 7 00 PM she was connected to a patient controlled analgesia PCA pump A PCA pump is a special type of infusion pump that delivers small doses of pain medication at the request of a patient via a remote button A nurse programmed the PCA pump so Danielle could self administer small doses of morphine as prescribed by her physician At 8 30 PM Danie
9. O Undecided a Disagree a Strongly disagree Additional comments 7 lam confident completed all the tasks correctly during the scenario aA Strongly agree a Agree O Undecided aA Disagree OA Strongly disagree Additional comments The training received was comprehensive enough to allow me to complete all of the scenarios a Strongly agree a Agree OA Undecided QO Disagree Strongly disagree Additional comments Figure 42 Sample of post questionnaire for PCA pump usability testing The training protocol for the testing was developed by members of the PIP team First the PIP team members received training from each of the vendors at the level of depth that would typically be given by the vendor during an inservice The PIP team then developed a training protocol for the usability testing that included all the information required to complete each of the scenarios The length and depth of the training protocol for the usability testing was similar across all three pumps Introductory scripts and consent forms were also created to ensure each participant would receive all the necessary information prior to the start of the usability test session For testing a total of five nurses from the post anaesthesia care unit and five nurses nurse practitioners from the acute pain team were recruited Two pilot usability tests were completed one with the acute pain nurse from the HFPIP team and one with the general war
10. PIP As outlined in the RFP the PIP team requested 3 PCA pumps and 50 primary tubing sets from each shortlisted vendor so the heuristic analysis and usability testing could take place 193 Section 12 4 1 Heuristic analysis The human factors specialist conducted an independent heuristic analysis Chapter 7 using the Zhang et al heuristics for assessing the usability of medical technology Table 1 and then led several clinical representatives in turn through a heuristic analysis by sitting with them and asking them to complete a wide range of tasks on the PCA pump while they identified anything that seemed unclear awkward or difficult The task list was created based on the user s manual for each pump and the observation data collected during the user needs assessment Since not every PCA pump had the same features the task lists were somewhat different for each pump although there was a common set of basic tasks across each pump The human factors specialist documented the underlying design issue and the heuristics that were violated Once all the tasks were complete the human factors representative reviewed the list of issues with the clinician and asked the clinician to identify the impact worst possible outcome as a result of the issue The severity of each outcome was rated using pre defined scoring criteria similar to that shown in Table 2 A list of usability issues considered to be severe based on the scoring criteri
11. a in Table 2 were compiled from across the heuristic analyses completed by each team member A list of recommended changes or actions was identified for each of these issues A sample of some of the severe issues identified on each of the three PCA pumps is shown in Table 25 Table 25 A sample of the severe usability issues found by five independent reviewers during a heuristic analysis of the three shortlisted PCA pumps Issue Heuristic Recommended Changes or Actions Violated Pump can start infusing with Prevent Errors Pump should have a sensor on the lock and not just the cover closed but not the cover to ensure safety given the high risk locked Feedback nature of medications such as narcotics Cannot change the delivery Flexibility and Pump should allow users to adjust parameters in the parameters once the pump is efficiency setup menu once the pump is running A code or programmed and running key should be required Users have to re program all Users in control information and any shift information is lost There are two different task Minimize Task sequences for changing a syringe should be sequences associated with memory load consistent regardless of whether a new syringe changing a syringe depending contains the same drug on whether the same drug is continued or not 194 Issue Heuristic Recommended Changes or Actions Violated Pump does not require the Prevent Errors An option should be available to make the barcode barc
12. and 3 conducting the heuristic analysis were used as a basis for developing the usability testing scenarios Section 8 5 2 Each participant was required to complete four different scenarios on each pump to ensure the following list of tasks was completed on each pump 1 Setting up and programming a PCA for a new patient 2 Replacing an empty medication container and restarting the pump 3 Changing to anew medication and re programming the pump 4 Titrating the dose and checking the medication history Three sets of different but equivalent scenarios were developed so that participants would be able to complete the same set of tasks on each pump but with a different context so that testing each of the three products would not seem repetitive or familiar 196 The assignment of each set of scenarios to each pump was counterbalanced as was the order that each participant tested each pump The counterbalancing schedule is shown in Figure 38 For each participant i e Participant 1 10 the schedule illustrates the order each pump is tested e g First A Second B Third C and the scenario group applied to each pump i e 1 S2 S3 O PI P2 Pans Pont Pans Pane Pan Pans Pano PO A S1 B S3 C S2 A S1 B S3 C S2 A S1 B S3 C S2 ASS B S2 C S1 A S3 B S2 Cc S1 A S3 B S2 C S1 A S3 B S2 Third C S3 A S2 B S1 C S3 A S2 B S1 C S3 A S2 B S1 C S3 A B C pump A pump B pump C S1 S2 S3 scenario group 1 scenario group 2 scenario grou
13. cal and financial reviews Since there were no major technical or cost constraints the most significant differentiating factors were the results of the human factors evaluations Each of the three PCA pumps being considered had safety issues that could not be effectively mitigated by the healthcare organization Since these issues had the potential to cause serious patient harm discussion was initiated with each of the three vendors to determine whether software and other design changes could be made to address the concerns None of the three vendors were able to make the requested changes and so the HFPIP team decided not to purchase any of the three PCA pumps evaluated 208 The available funds for the capital expenditure were held by the healthcare organization and the purchase decision was deferred for about a year During this time another vendor introduced a new PCA pump to the marketplace that was licensed for sale and that met the criteria set out in the RFP The FPIP team evaluated the new PCA pump using the same process described in this chapter The results of the human factors evaluations showed fewer usability issues and none of the issues identified had potentially serious safety implications The PIP team selected this pump Pump D for procurement It is recognized that many hospitals are required to make purchasing decisions based primarily on cost In these cases it is even more important to ensure that user needs particularl
14. care unit nurses from pain management and nurses from the general ward were included in the study The vendors of each of the three PCA pumps were contacted for assistance with creating appropriate drug libraries for each of these four services 198 Nurse Actor Hi I m Carol nice to meet you Thanks for coming in on short notice we ve been swamped all morning and our resource nurse didn t show up We just got 3 patients back from the OR and all of them need meds started right away Since you haven t worked on this unit before there are a couple of things I ll show you before I get you start one of our PCA s First here is our medication administration cart where you can find the patient s records and medication orders You ll be responsible for administering the IV medications but I can take care of any documentation There is a formulary of IV medications in the binder on the cart if you need it and you can also access it online Any questions Participant No not that I can think of or answer whatever question they have Nurse Actor Great lets get started Our first patient is Mr Ricardi He is a 76 year old male who underwent a total hip replacement No complications or relevant medical history to report His heart rate is 62 his respiratory rate is 14 his temperature is 37 8 C and his blood pressure is 110 70 He is conscious but is currently sleeping and has been complaining of post
15. d nurse from the PIP team to ensure everything had been organized properly and was ready to go The human factors specialist and biomedical technology professional filled out the usability test checklist Section 8 5 13 prior to conducting the first official usability test session 201 Section 12 4 2 2 Conducting Usability Testing Upon arrival each participant was welcomed by the facilitator The facilitator delivered the introductory script Section 8 5 8 and reviewed the consent form Appendix A with the participant and then asked the participant to complete the pre usability test survey Figure 41 The facilitator then delivered the training associated with the first pump being tested to the participant Following the training session for that pump the facilitator introduced the participant to the first scenario acting as the actor confederate nurse and the participant completed the four test scenarios associated with each pump The testing was videotaped using a video camera on a tripod A biomedical technology student helped to do the video recording during the sessions During the test the human factors specialist documented any issues that were observed or actions that were unexpected in the data documentation sheet Following the completion of the scenarios the facilitator asked the participant to fill out the post training questionnaire Figure 42 and conducted an informal debrief session This process was repeated on the othe
16. d potentially result in incorrect tracking of drug volume administered which could lead to inappropriate changes to the medication order New drug may run using the previous drug protocol Patient would receive an over under infusion depending on the concentration to dose ratio Protocol parameters could be incorrect Severity of impact depends on the range of values allowed for the protocol Issue Barcode scanner difficult to activate Description of Error User is unsuccessful at scanning the barcode not holding it in the right position after pressing the top knob Impact to Safety Manual selection of drug allows the potential for the wrong drug to be selected Patient could receive over under 203 User is not alerted that the pump is running Pump terminology is inconsistent with user experience Unclear how to clear shift totals Easy to confuse modalities since they are selected using one knob User is not alerted that pump is not running User manually selects drug Pump is stopped and user does not restart User thinks the pump is running User increases bolus dose to 1 5 mg instead of the PCA dose because she interprets bolus to mean PCA dose User gives an accidental clinician bolus when trying to clear the shift totals User accidentally purges the pump after it is connected to the patient while trying to start the pump User programmed the pump but did not press start
17. ed by human factors researchers required a maximum of only 12 programming steps in comparison A controlled experiment comparing the two designs showed the human factors informed design led to fewer errors faster task completion times and lower mental workload 88 This incident serves to highlight that adverse events can be expected when devices have not been designed selected and implemented using human factors principles 84 89 91 Perhaps Danielle and others would be alive today if the manufacturer of the PCA pump had incorporated human factors methods when designing the PCA pump or if the hospital had been able to incorporate human factors into the procurement process To illustrate how a biomedical technology professional can apply the 4 PIP presented in Chapter 11 this chapter presents a case study applying the framework to the procurement of a new type of PCA pump to replace a hospital s existing PCA devices The rationale for the purchase was that the current PCA devices were very old and it was becoming increasingly difficult to get replacement parts to service the pumps Additionally 190 the current pumps had usability issues that were implicated in several medication incidents that resulted in patient harm Section 12 1 Create Team Once the hospital made a decision to move forward with procuring a new PCA pump a multidisciplinary team was created to undertake the PIP Table 24 In this case the team consisted of
18. hts of medication management 92 93 and the storage locations of intravenous tubing for PCA pumps and large volume infusion pumps were changed because of a near miss during usability testing where the wrong tubing was almost used by a participant Training and Education Several training sessions were provided to staff beginning a few months prior to the date the PCA pump was to be implemented Members of the PIP team including the clinical educator created a training program tailored to the needs of each representative 209 user group The human factors representative was also very instrumental in the design of the training and communicated what aspects of the pump needed to be highlighted in the training based on the results of the usability testing The training was administered jointly by the vendor and the clinical educator Information was presented not only about the knobology e g what button to press to start the PCA pump but also the underlying principles governing how the PCA pump worked so staff would understand why a process had to be done in a certain way Hands on training clinics were held regularly leading up to the implementation date so staff could practice using the PCA pump in a simulated setting Prior to receiving sign off to use the PCA pump in practice each end user had to successfully complete a set of hands on tasks to demonstrate their ability to perform each required task Implementation The
19. lementation considerations Section 12 3 Write and Distribute RFP Once the needs and wants of each user group were established the procurement team wrote and distributed a request for proposal RFP to several vendors In addition to standard RFP components the PIP team also included a request for documentation from each vendor outlining how the human factors standard HE75 73 had been interpreted and incorporated into the design of the PCA pump Table 23 In this case rather than highlighting specific parts of the standard in the RFP a more general inquiry was made using the following request Please indicate how the design of the technology has fulfilled the AAMI ANSI Standard HE75 Additionally the RFP requested that the vendor provide 192 evidence of how usability testing results Chapter 8 were incorporated into the design of the product The RFP also included a description of the human factors evaluation methods that were planned as part of the PIP Figure 36 Each product that is shortlisted for evaluation will undergo human factors testing which will include a heuristic evaluation and usability testing Vendors will not be invited to participate directly in these evaluations but will be expected to provide the healthcare organization with 3 PCA pumps and 50 primary tubing sets to enable this testing Additional support may be required to train procurement team staff members who will be running these evaluations
20. lle was awake alert and feeding her newborn Six hours later following a 30 minute resuscitation effort Danielle had died The autopsy results showed Danielle had experienced a morphine overdose with almost four times the lethal dose of morphine in her bloodstream Upon further investigation the cause of the overdose was found to be a programming error Specifically the nurse programmed the pump for a 1 mg mL morphine concentration but Danielle was receiving morphine at a concentration of 5 mg mL This meant that each time Danielle requested morphine she received a 5 fold overdose compared with the prescribed amount Case Study 2 Morphine Overdose of Danielle McCray The events of this case study are described in the book The Human Factor 9 anda journal article published in the Canadian Journal of Anesthesia 84 After reviewing this case it may not seem obvious how an error like this could have occurred A concentration of 5 mg mL is obviously more potent than a concentration of 1 mg mL and programming the pump for a lower concentration than is loaded means that 189 more volume of the drug is released with each dose which given the concentration used will result in a medication overdose One of the likely contributing factors to this error is that in the factor preset configuration the pump sequentially offers four default concentration settings during the programming sequence morphine 1mg mL morphine 5mg mL morphine
21. ne up with the screen options Pump response to key pressis User pressed the same button Unintended over dose of delayed several times because the medication pump did not respond quickly to the initial key presses Subsequent presses were applied to the next screens without the user knowing what was selected In one case the user inadvertently gave the patient a 1 5 mg bolus because of this design issue The HFPIP team reviewed each of the issues in Table 26 and discussed potential mitigating strategies for each one to see whether the risks could be addressed in a systematic way prior to implementation Risks that could not be suitably mitigated i e mitigated using a systems approach rather than a person approach as described in the hierarchy of effectiveness were compared across the three pumps to identify which of the three pumps was the safest and best fit for the organization Table 27 shows the mitigating strategies that were identified for the issues associated with each of the three pumps Table 27 Mitigating strategies strategies for the severe issues identified during usability testing Issue Description of Error Impact to Safety Mitigating Strategy Pump terminology User sets the Patient receives more Change wording on inconsistent with background rate at1 0 pain medication than the pre printed terminology used at mg hr instead of the prescribed medication order to healthcare bolus dose Consequences depend ma
22. new PCA pump was implemented by the healthcare organization and although the transition was somewhat stressful for staff highly trained clinical champions on each unit were available to support staff during the implementation Transitional and Ongoing Support Even after the PCA pumps had been implemented occasional training clinics were held where staff could come in to practice on the new PCA pump in a simulated environment A competency training program was established at the healthcare institution so staff could regularly brush up on the requirements when using these pumps and new staff could be trained in a systematic way The implementation was highly successful Nurses transitioned to the new devices with ease and the pumps were used safely and effectively by all user groups 210
23. ode to be scanned Manual scan mandatory prior to loading the syringe drug selection is always Scanner should detect and indicate a faulty available barcode After the barcode is scanned Prevent Errors Once a barcode has been scanned the drug menu the drug name and should only show the scanned drug All other drug concentration can be manually names should be eliminated changed Pump does not prompt users to Prevent Errors On the Run screen users should be forced to systematically review the confirm each setting settings before starting the pump There is only one lock level Prevent Errors Create at least two lock levels One that unlocks all that provides access to all functions and one that unlocks everything but the functions clinician bolus so that ward nurses cannot accidentally give boluses When the pump is unlocked Prevent Errors Pump should automatically re lock once it starts with a code it remains running or after 30 seconds of being idle during unlocked for one minute programming Patients could tamper with the pump during this time The 4 hour limit does not have Prevent Errors Units for the 4 hour limit should be fixed as part of fixed units Users can select the drug protocol mcg kg mcg mg kg mg Issue Heuristic Recommended Changes or Actions Violated Button key press is not visible Informative CPU should update screens much faster to prevent to user right away Feedback users from selecting the wrong button
24. on the screen If the user is not standing directly in front of and at the same height as the screen the buttons do not line up with the screen options User pressed the same button several times because the pump did not respond quickly to the initial key presses Subsequent presses were applied to the Impact to Safety Anaesthesia Mode removes many of the safety limits built into the drug templates This mode could be removed from the pump so this option would never be inadvertently selected If a dose of 2 mg is inside the programming limit a 10 fold overdose would occur each time the patient requested a dose Pump programmed incorrectly resulting in over under infusion Incorrect selections made which could result in giving drug to a patient inadvertently Mitigating Strategy Remove Anaesthesia mode from drug library template to prevent this use error since it is not needed for PCA No effective mitigating strategy from within the organization Recommend that vendor reviews button design to reduce force required to register key press No effective mitigating strategy from within the organization Recommend that vendor reviews screen design to reduce parallax effect No effective mitigating strategy from within the organization Recommend that vendor reviews button 207 next screens without design the user knowing what was selected In one case the user inadvertently ga
25. p 3 Figure 38 Assignment of the order of pumps and scenario groups to each participant to achieve counterbalancing Usability scenarios were reviewed with clinical members of the PIP team including those from anaesthesia the post anaesthesia care unit pain management the general ward and clinical education to ensure they were as realistic as possible Two of the usability test scenarios that were developed are included in Figure 39 as an example 197 Scenario Group 1 Scenario 1 Setting up a new PCA Infusion New patient Mr Ricardi arrives to the post anesthetic care unit PACU following a total hip replacement surgery The patient is 76 years old and stable His pain score is 6 10 when resting Task 1 Program the PCA pump so the patient receives morphine 2mg mL at a continuous dose of Smg hr with a bolus dose of 2mg The bolus lockout time is 5 min and the 4 hour limit is 40mg Scenario 2 Replacing an empty medication container Mr Ricardi is ready to move from the PACU to the ward Upon arrival at the ward you notice the morphine container in his pump is almost empty Task 2 Remove the empty drug container replace with a new one re start the pump Scenario 3 Changing to a new medication Mr Ricardi is starting to complain he is itchy You notice he is developing hives on his torso and they are spreading You speak to his wife and she says that he had a similar reaction to morphine when he fell and bruised his hi
26. p several years ago Although his chart has no allergies listed you consult with his doctor and she decides to switch him to hydromorphone Task 3 Change the morphine drug container to hydromorphone re program and start the pump Scenario 4 Titrating the infusion and documenting shift totals About an hour after the hydromorphone was started Mr Ricardi pain s score is still 5 10 His heart rate is elevated and he is visibly uncomfortable His medication orders allow the continuous dose of hydromorphone to be increased from 1mg hr to 1 5mg hr Since you are at the end of your shift you decide to first document your shift totals and then titrate the hydromorphone Task 4 Document the shift totals in the medication administration record and then titrate the hygromorphone Figure 39 Scenarios in scenario group 1 for PCA usability testing Once the scenarios were finalized scripts were written to facilitate each task A sample script based on the first scenario in Figure 39 is shown in Figure 40 A data documentation tool Section 8 5 4 was also developed based on the tasks and subtasks of each scenario Each of the three PCA pumps was customized to suit the usability scenarios and user groups In this case the drug library for each pump was customized to match the drugs and concentrations given by PCA on wards and in the post anaesthetic care unit Representative end users including anaesthetists nurses from the post anaesthesia
27. r two PCA pumps using the other two scenario groups to minimize familiarity with the tasks At the very end of the usability test after testing all three pumps a slightly longer debrief session was held with the participant to gather more general thoughts about the test session and the three PCA pumps used The usability test team analyzed the data collected from the usability test sessions by evaluating user performance and preferences for each of the three PCA pumps Section 8 7 1 The performance data was used to determine issues and their severity The preference data was used to identify any additional potential issues or user needs not previously captured Questionnaires and notes taken during each usability session were compiled across participants Data documentation spreadsheets that were completed during testing were also compiled across participants and a determination of which tasks were passed and failed was made Tasks that participants had difficulty completing were considered in further depth by the PIP team Results from usability testing uncovered several new issues and validated many of the findings from the heuristic analysis Each error was rated using a severity scale similar to that in Table 11 with the most severe errors being extracted and compiled for each of the three PCA pumps tested Table 26 Descriptive statistics were used to summarize user performance on each pump for example how many errors occurred on each ta
28. s No Not sure In general what is your preferred method to learn about how to use a new medical device rank options from 1 to 6 with 1 being most preferred and 6 being least preferred Read about the device Attend a hands on demonstration Attend a lecture or seminar Have access to the device to practice on your own and ask questions as needed Work side by side with an expert colleague Watch a training video Figure 41 Sample of pre questionnaire for PCA pump usability testing 200 Participant Feedback Using this PCA pump was intuitive 0 Strongly agree O Agree 0 Undecided 0 Disagree Q 1 Strongly disagree Additional comments 2 Ihave concerns about using this PCA pump in my clinical environment QO Strongly agree 0 Agree O Undecided 0 Disagree QO Strongly disagree Additional comments 3 This PCA pump will make my job easier as a healthcare practitioner O Strongly agree 0 Agree 0 Undecided O Disagree QO Strongly disagree Additional comments Using this PCA pump will change how think about delivering infusions O Strongly agree QO Agree O Undecided QO Disagree O Strongly disagree Additional comments 5 Using this PCA pump in my unit will help to improve patient safety a Strongly agree a Agree O Undecided O Disagree Strongly disagree Additional comments 6 I found the set up tasks during the scenario to be difficult to complete a Strongly agree a Agree
29. sk for each pump and of those how many had potentially harmful consequences 202 Table 26 Summary of issues with potentially severe safety consequences identified during usability testing Issue Input mechanism scroll wheel is not intuitive Pump terminology inconsistent with terminology used at healthcare organization Technical software glitch Users cannot remember the task sequence for changing a syringe to a new drug since it is different that changing a syringe when the drug remains the same User is not forced to verify settings Description of Error User selects the wrong drug either because they press the wheel trying to scroll or they accidentally press and turn the wheel at the same time User sets the background rate at 1 0 mg hr instead of the bolus dose When the pump is first turned on totals showing are 0 0 mg 42 2 mL They should both be zero When changing the syringe toa new drug users forgot to stop the pump in order to access the drug list One user changed a syringe to a new drug but kept the old drug protocol User did not verify each parameter selected They scrolled directly to Confirm Impact to Safety Patient receives over under infusion depending on the concentration to dose ratio if the user does not detect the error Patient receives over infusion Consequences depend on the rate entered but could be severe Consequences unclear but coul
30. tch the pump organization onthe rate entered but terminology could be severe Technical software When the pump is first Consequences unclear No effective turned on totals but could potentially mitigating strategy 205 glitch Task sequence for changing a syringe toa new drug is different that changing a syringe when the drug remains the same showing are 0 0 mg 42 2 mL when they should both be zero When changing the syringe to a new drug users forgot to stop the pump in order to access the drug list One user changed a syringe toa new drug but kept the old drug protocol result in incorrect tracking of drug volume administered New drug may run using the previous drug protocol Patient would receive an over under infusion depending on the concentration to dose ratio from within the organization Ensure vendor fixes technology glitch No effective mitigating strategy from within the organization Recommend that the vendor review the workflow and improve consistency Issue User is not alerted if the pump is programmed but not running Task sequence for clearing the shift totals is not intuitive Easy to confuse modalities since they are selected using one knob Description of Error Pump is stopped and user does not restart User thinks the pump is running User gives an accidental clinician bolus when trying to clear the shift totals User accidentally purges the pump
31. ve the patient a 1 5 mg bolus because of this design issue During this PIP neither an FMEA Chapter 9 nor in use trials Section 11 6 2 4 were conducted Since the healthcare organization s policy allowed it the PIP team shared some feedback with each vendor to highlight design issues that were uncovered on their product during the evaluation process The team was extremely careful when sharing results ensuring that information was kept confidential between the PIP team and each individual vendor In addition to the aforementioned human factors informed evaluations the biomedical technology department conducted technical assessments of each pump to confirm that each PCA pump operated according to specification All three products met the technical specifications and no products were eliminated on this basis The hospital s purchasing department also conducted a financial review of each product Pump C was more expensive than either of Pump A or Pump B and fell outside of the hospital s budget however this vendor indicated a willingness to negotiate on price in exchange for being a beta test site if there was interest in purchasing the product No products were eliminated from the evaluation based on financial concerns Section 12 5 Decide on Product s To make the final decision about which PCA pump to procure the PIP team weighed several factors including findings from the human factors evaluations and the techni
32. wing data collection and documentation another round of interviews and focus groups Chapter 5 were done with staff to validate the task analysis 191 and to collect information about other desired features and capabilities that an ideal PCA pump would have The biomedical technology professional from the PIP team reviewed a previous HFRCA Chapter 10 that had been completed following an incident involving the hospital s current PCA pumps Internal and external incident reporting systems were also reviewed to search for PCA pump related incidents For each outlined user group established needs and wants were sorted into functional requirements what the pump must do and implementation considerations what the hospital must consider and or adapt prior to implementing A small subset of the user needs were identified as a result of this process Figure 35 User Group 1 Post anaesthesia care unit nurses at the hospital Functional Requirements Must be able to administer 1 A PCA only dose 2 A PCA and continuous dose and 3 A continuous only dose Must be able to easily attach and detach the PCA pump from an IV pole Must have a robust lockbox Implementation Considerations To ensure other medications are not inadvertently bolused there must be a way to keep track of which IV catheter the PCA pump is infusing through Figure 35 Selection of user needs and wants divided into functional requirements and imp
33. y those associated with product features and functions that can impact safety are translated directly into the request for proposals so that products can be eliminated from contention that do not support safe use If the final decision is primarily determined by cost it is recommended that a usability evaluation be conducted on that product to identify potential issues and training requirements so that mitigating strategies and appropriate training can be developed as part of the implementation strategy Section 12 6 Configure Product s and Environment Findings from the human factors evaluations conducted on Pump D were used to help inform how to configure the pump for each representative user group For example alarm settings were adjusted based on the number of air in line alarms experienced during the usability testing Additionally the pre printed medication orders were re designed to ensure the wording matched the wording used on the pump and the order of information on the pre printed order form was consistent with the programming sequences to minimize data entry errors Section 12 7 Plan and Implement Product s Making Changes to the Work System Findings from the human factors evaluations conducted on Pump D were used to help inform the types of changes required at the work system level For example a worksheet was implemented to help support and guide nurses through some new documentation steps that were required for verifying the rig
Download Pdf Manuals
Related Search