Skip to content

Episode Notes

Source / episode info

  • Episode: 230
  • Title: Divine Intervention Episode 230 – USMLE Nov 2020 Changes Series 2: Quality and Safety.
  • Published: 2020-04-16
  • Source: Episode page

One-liner

This episode reviews essential quality and safety buzzwords—including QC, QA, QI, PDSA, DMAIC, Lean, Six Sigma, FMEA, and the Swiss Cheese Model—to teach systematic approaches for improving clinical processes and patient care.

High-yield summary

  • Quality Control (QC): Daily, immediate checks on a process or product; highly frequent and retrospective (e.g., resident checking notes every day).
  • Quality Assurance (QA): Less frequent, periodic audits of a system to ensure standards are maintained; retrospective (e.g., weekly audit of notes).
  • Quality Improvement (QI): The proactive design and implementation of an intervention based on identified problems; can be both prospective and retrospective.
  • DMAIC vs PDSA: DMAIC (Define, Measure, Analyze, Improve, Control) is a structured, data-driven process improvement cycle, considered superior to the general Plan-Do-Study-Act (PDSA) cycle because it mandates rigorous measurement and analysis of baseline data.
  • Lean vs Six Sigma: The key difference is their target: Lean focuses on eliminating waste (non-value-added steps/resources), while Six Sigma focuses on reducing the defects or variation in a process (aiming for 3 defects per million).
  • Error Analysis Models: FMEA is a prospective method used to predict potential failure modes before they occur, whereas Root Cause Analysis (RCA) is a retrospective method used after an adverse event has happened.

Learning objectives

  • Differentiate between Quality Control (QC), Quality Assurance (QA), and Quality Improvement (QI) in terms of frequency and timing (retrospective vs. prospective).
  • Apply the principles of Lean thinking to identify and eliminate non-value-added waste in clinical processes.
  • Distinguish the goals and methodologies of Six Sigma (defect reduction) versus DMAIC/PDSA (process optimization).
  • Select the appropriate error analysis tool: FMEA for prospective risk assessment, or RCA for retrospective failure investigation.
  • Understand that high value care is achieved by maximizing service quality while minimizing cost.

Board exam buzzwords

ConditionKey FindingAssociationBoard Exam Tip
Quality Control (QC)Daily/Immediate checksRoutine adherence to protocols; mandatory checklistsThink daily and mandatory. It's the most frequent form of quality check.
DMAICDefine -> Measure -> Analyze -> Improve -> ControlData-driven process improvement; Six Sigma frameworkRemember it is a structured, data-intensive cycle, superior to general PDSA for complex problems.
FMEAFailure Mode and Effects AnalysisProspective risk assessment of potential failure pointsAlways remember FMEA looks into the future ("What could go wrong?").
Swiss Cheese ModelMultiple layers of safeguards requiredSystem safety; Error prevention in complex processes (e.g., medication administration)Failure only occurs when holes align across multiple independent systems/checks.

Rapid review table

TopicKey PointContextExam Relevance
QC vs QAQC is daily, QA is periodicQC checks compliance; QA audits standards adherenceTest question may ask if the process is routine (QC) or periodic (QA).
Lean ModelEliminate waste (Muda)Identifying non-value-added steps in a workflow (e.g., unnecessary paperwork, waiting time)Focus on waste, not defects. Example: Redundant forms are waste.
Six SigmaReduce variation/defectsStatistical process control; Goal of 3 defects per million (DPM)Focus on fidelity and defect rate. The goal is near-perfection.
FMEA vs RCAFMEA = Prospective; RCA = RetrospectiveFMEA predicts failure modes; RCA investigates actual failuresThis distinction is a common trap question in quality improvement testing.

Board-speak -> diagnosis

Board-speak / Vignette phraseDiagnosis / ConceptWhy it fits
A hospital implements mandatory electronic checklists and daily sign-offs for all surgical procedures to ensure compliance with institutional standards.Quality Control (QC)QC involves immediate, frequent checks on adherence to established protocols; the checklist enforces this day-to-day standard.
After a series of adverse drug events were identified, the hospital initiated a project using statistical process mapping and root cause analysis to determine systemic failures in the medication dispensing workflow.Root Cause Analysis (RCA) / QIRCA is used retrospectively after an event has occurred to find the underlying system failure, which falls under Quality Improvement.
A manufacturing plant aims to reduce its product defect rate from 10 defects per million to less than 3 defects per million using statistical process control.Six SigmaThe goal of reducing variation and achieving a specific low defect count (e.g., 3/million) is the hallmark of Six Sigma methodology.
A team redesigns a complex patient intake process by mapping out every step, identifying redundant forms, unnecessary waiting times, and excessive paperwork.Lean MethodologyThis approach focuses specifically on eliminating waste (time, resources, steps) to streamline efficiency, which is the core principle of Lean thinking.
Before implementing new safety protocols for high-risk procedures, a multidisciplinary team systematically brainstorms every way the procedure could fail and analyzes the potential consequences of each failure point.Failure Mode and Effects Analysis (FMEA)FMEA is inherently prospective; it predicts failures before they happen by analyzing "failure modes."
A new hospital initiative mandates that all staff must complete a structured training module on identifying process bottlenecks and improving workflow efficiency using data visualization tools.DMAIC CycleThis represents the systematic, data-driven approach (Define -> Measure -> Analyze -> Improve -> Control) required for comprehensive quality improvement.

Differential diagnosis / distinguishing features

FMEA vs Root Cause Analysis

Key FeaturesDistinguishing FindingsNext Step
FMEA: Failure Mode and Effects Analysis; Identifies potential ways something can fail.Highly prospective; focuses on predicting failure modes (e.g., "What if the pump fails?").Use during process design or risk assessment to prevent future errors.
RCA: Root Cause Analysis; Investigates actual failures that have already occurred.Highly retrospective; requires an adverse event as a starting point (e.g., "Why did the patient fall?").Use after an incident report or sentinel event has been filed to determine systemic failure points.

Management pearls

  • Process Improvement Hierarchy: Always remember that QI is the umbrella term; DMAIC, PDSA, Lean, and Six Sigma are specific tools used within the QI framework.
  • Swiss Cheese Model Application: When designing safety protocols (e.g., medication administration), implement multiple independent checks (computer hard stops -> pharmacist check -> nurse verification) to prevent error accumulation.
  • High Value Care Calculation: To improve value, focus on increasing service quality and efficiency while simultaneously reducing unnecessary costs (waste).
  • Hawthorne Effect Mitigation: When implementing a new protocol, be aware that initial compliance may spike simply because staff know they are being observed; sustained change requires systemic redesign, not just monitoring.

Don't miss

🚨
DMAIC is data-driven: Unlike PDSA, DMAIC forces the team to Measure and Analyze baseline data before proposing improvements, making it statistically robust.
🚨
Lean = Waste Elimination: The goal is removing non-value-added steps (e.g., waiting time, redundant documentation).
🚨
Six Sigma = Defect Reduction: The goal is achieving near-perfect process fidelity by minimizing variation and defects (3 DPM).
🚨
FMEA is Prospective: It predicts failure; RCA is retrospective.

Integration & clinical reasoning

  • Systems Thinking: All these models emphasize that medical errors are rarely due to single human failures but rather systemic breakdowns across multiple processes, best visualized using the Swiss Cheese Model.
  • Accountability vs. System Failure: While individual accountability (QC) is necessary, true safety improvement requires addressing system flaws (QI/RCA).
  • Value in Medicine: The concept of high value care forces clinicians to move beyond simply providing care and instead focus on providing the best possible outcome per unit cost .

Concept connections / cross-references

  • For a deeper understanding of process mapping and quality metrics, review general biostatistics principles covered in [ Episode 12 ].
  • The principle of multiple layers of safety is related to infection control protocols discussed in [ Episode 45 ].

High-yield association table

ConditionAssociationMechanismClinical Significance
DMAICDefine -> Measure -> Analyze -> Improve -> ControlStructured, data-driven process improvement cycle.Provides a systematic framework for solving complex clinical workflow problems.
Lean ModelWaste Elimination (Muda)Identifying and removing non-value-added steps in a process.Improves efficiency by streamlining workflows (e.g., reducing unnecessary patient transfers).
Six SigmaDefect Reduction; 3 Defects Per Million (DPM)Statistical process control aimed at minimizing variation.Used to achieve extremely high levels of fidelity and reliability in critical processes (e.g., blood bank testing).
FMEAFailure Mode AnalysisProspective identification of potential failure points and their consequences.Essential for risk management during the design phase of new medical technologies or protocols.

Key terms glossary

TermDefinitionContextExample
Quality Control (QC)Routine, immediate checks to ensure adherence to established standards.Daily clinical practice; mandatory checklists.A resident checking that all required elements (HPI, ROS, etc.) are present in a daily progress note.
DMAICDefine -> Measure -> Analyze -> Improve -> Control.Structured methodology for process improvement using data.Using patient fall rates (Define/Measure) to analyze the root cause and implement new bed alarms (Improve).
FMEAFailure Mode and Effects Analysis.Prospective risk assessment of potential system failures.Before implementing a new surgical robot, running an FMEA to predict mechanical or procedural failure points.
Swiss Cheese ModelMultiple layers of independent safeguards are required; error occurs when holes align across multiple processes.System safety and patient care protocols.A medication order requires the physician's sign (Layer 1), the computer hard stop (Layer 2), and the pharmacist's final check (Layer 3).

Study optimization

TopicStudy ApproachPriorityResources
Process Improvement ModelsCreate a comparison chart: Goal, Timing (Prospective/Retrospective), Key Action.HighReview definitions of DMAIC vs PDSA; Lean vs Six Sigma.
Error Analysis ToolsPractice distinguishing between FMEA and RCA based on the timeline (before vs after).Medium-HighUse vignettes that describe either a predicted failure or an actual adverse event.
Value AssessmentMemorize the formula: Value Service Quality / Cost.HighApply this concept to resource allocation decisions in healthcare policy questions.

Question pattern recognition

  • The "Which is Best?" Trap: Questions often ask which methodology is superior for a given scenario (e.g., Is it better to use PDSA or DMAIC?). The answer depends on whether the problem requires deep statistical analysis (DMAIC) or simple, iterative testing (PDSA).
  • Temporal Distinction: Always determine if the process being described is predicting failure (FMEA/Prospective) or investigating failure (RCA/Retrospective).
  • Goal Differentiation: Never confuse Lean's goal of eliminating waste with Six Sigma's goal of eliminating defects .

Test yourself

Common mistakes to avoid

🚫
Confusing QC and QA: Remember, QC is daily compliance checking (the "doing"), while QA is periodic auditing (the "checking the check").
🚫
Mixing up FMEA and RCA: Never use FMEA to investigate a past event; it must be used to predict future risks. Conversely, never use RCA to design a new process; it requires an existing failure.
🚫
Assuming PDSA = DMAIC: While related, remember that DMAIC is superior because its structure mandates rigorous data collection (Measure/Analyze) before improvement can occur.

Common traps

⚠️
The "All Three Must Be Positive" Trap (Light's Criteria): Do not assume all three criteria for an exudative effusion must be positive; meeting any one criterion is sufficient.
⚠️
Primary vs Secondary AI: Remember that primary adrenal insufficiency causes hyperkalemia/Type 4 RTA because aldosterone production is deficient, while secondary AI preserves aldosterone and does NOT cause hyperkalemia.
⚠️
The "Process" Trap: When asked to improve a process, the answer must be systemic (e.g., DMAIC or Swiss Cheese Model), not simply blaming an individual person's error.

Original transcript with highlights

Original transcript with highlights

Okay, welcome. My name is Devine, I'm a resident. This is episode 230 of the Divine Intervention Podcasts. And in this podcast, I'm going to be continuing the Clean SP curriculum, right? The curriculum that essentially addresses the changes that are coming to the US Emily exam study next month. And this will, so this will be series two of the Clean SP curriculum. And essentially, I'm going to be introducing quality and safety. Quality and safety is one of those things where it's important to understand almost like the buzzwords in a sense. So God willing, what I'm going to be doing today is I'm going to be talking about a lot of the buzzwords for quality and safety. I'll see how far I can get you know, if I can finish it, I'll just pick it up in the in the next podcast. So yeah, so I'm going to define these buzzwords and where possible again, I will try to I will try to give examples because I feel like examples really buttress these points because sometimes it kind of gets a little easy to jumble these points in one's head, right? So and some things I may discuss to you, maybe like, huh, this is kind of weird. I've never heard of this before. Well, these are things you could potentially see on an exam. So those are things I will spend a lot of my time talking about. So let's just jump right into it. So the first term I would want to define today is something that is called quality control, right? Quality control. So for example, right, let me give you a classic example.

So let's say you're a med student, you know, you're in your third year rotations, let's say you're in your I'm rotation, and you already know something that happens on I'm rotations, right? You know, you write notes a lot, right? Right notes a lot. So there's nothing wrong with writing notes, but you know, every day, let's say like the resident or the attending checks your notes, you know, to kind of make sure it meets certain standards, right? That's an example of quality control. It's almost something that happens every pretty much every day, right? It's something that happens pretty much every day. So you're like, oh, the look at your notes and you know, they try to make sure that everything that should be in that note is there, right? So they want to make sure you have like a HPI, a pass medical history, pass surgical history, social history, family history, drugs, meds, allergies, assessment and plan, right? You know, they're essentially trying to check the quality of your product just to make sure that you know, it meets like some acceptable parameters. I mean, this is something that is done by the like the records and building the department in most hospitals, right? When you become a resident, you'll see that if you don't put certain things in your notes, those people will send you like, you know, some flip-punt emails. And when they say those emails, they essentially tell you, this is what you need to correct in your notes, right?

So that the hospital can build up appropriately, right? So this is something the key thing I want to remember is this one of those things that almost happens like multiple times every day. The reason that I'm being very specific in describing this is that quality control is something that people easily confuse with quality assurance. Quality assurance is not the same thing as quality control. They are very closely related, but they're not exactly the same quality assurance. Essentially is something where you just periodically, right? Checking on a product just to make sure. So let's say, you know, you've, let's say, you know, you're, you've, you studied your medicine rotation like, you know, a month ago, right? You know, you studied like early, mid-March, something like that, right? And you're going to be working with the assay, my tendon and resident for like eight weeks, right? So, you know, for the first four weeks, let's say the resident kind of checks every single thing you do because they're like, ugh, this person is kind of studying out 30 years, probably don't know the other from the right yet, right? So, you know, they are checking your work every day, make sure that you have the HPI, the family history, the best, that, right? So let's say, you know, that checking every day, that's quality control. Quality assurance is basically less frequent quality quality control. Essentially, something that you do is almost like an audit, right?

Like you periodically audit like the notes. So let's say the resident is like, instead of checking like every day, right? He starts checking like once a week, just to make sure that your notes are still up to up to standard, right? That's quality assurance, right? He's just assuring himself that, oh, okay, this means you're actually doing what is correct, right? So, he's almost like, oh, like how a company gets audited, right? Like on a, you know, like on a yearly basis, not like a company is audited, like every single day of its life, right? But I mean, most companies have like internal control mechanisms, they're just going to check, make sure, you know, people are like, you know, doing bad behavior, you know, they're going to do in the right thing, right? So that's what a quality assurance, that's what quality assurances, quality assurance is just a less frequent form of quality control, right? So now what is then quality improvement? You see, these terms are similar enough to where your friends at the end of the year would like to test it because they know people are probably like, you know, kind of mixed this up on a test, right? So quality improvement is where you're essentially, you've identified a problem, right? And then you are designing an intervention, it's almost like an interventional thing, right? You're designing an intervention to kind of improve things, right? Just remember that I in QI for improvement, okay? So the I in QI for improvement, right?

So essentially, right? Like, so again, quality control, it's almost like a retrospective kind of thing, right? If you're looking at it from like, almost like the perspective of a research study, quality control is where you're looking at things in retrospect, like, oh, the product is already been made. Well, let's make sure that it meets certain standards, right? But again, it's something that happens on a regular basis, right? Again, imagine like an attendee or a resident checking your notes every day. And then quality assurance is also a retrospective kind of thing, right? It's kind of retrospective kind of thing, but it's basically a less frequent form of quality control. You're just making sure that, oh, the fact that you're not checking in on someone all the time doesn't mean that, you know, they've studied like, being poorly right? Because you know, people like, if someone, this is the whole principle of accountability, right? So if someone knows that, oh, they're going to be checked from every day, right? They tend to do the right thing, right? But if a person knows that, ooh, they just check on me like once a month or something, right? You know, they're not, they're not, they're less apt to doing the right thing, right? So that's why, like, for example, let's say a person goes into like alcoholics and animus or whatever, you know, they're sponsoring, you know, they kind of call them like every day or something at the beginning, right? That's quality control.

Just checking in, making sure the process is good, right? But you know, after a while, let's say you're like 500 days sober or something, you know, that's quality assurance. You know, that's checking in or you make like once a week, just to make sure that, you know, this person isn't going to the reservation, right? But quality improvement is, is something that can be retrospective, right? You retrospectively analyze something that's bad, but you can also be prospective, right? You can identify errors, prospectively before a, before a product is even made, right? And then you try to design an intervention to change things. That's why it's known as quality improvement. Remember that I in QI for improvement and the IQI for interventional, right? So quality control quality assurance, they are purely retrospective processes, but quality improvement can be a retrospective process and can also be a prospective process. Okay? And the reason I guess I'm trying to talk about this QCQQI, sometimes in an exam question, they could try to integrate this with something called the Hawthorne effect, right? So the Hawthorne effect, it's something again, you've probably seen tested many times on exams, right? The Hawthorne effect is just basically the fact that, oh, a person acts differently, right? When they know they're being watched, right? If you know you're being watched, you're going to act differently, right? So think of it this way, right?

That's why some people, you know, kind of desired desire privacy. They don't want to have room meets because they know that you can do whatever they want at home, right? So say, for example, right? Let's say in the work room, right? Let's say like the resident work room or the met student work room, you know, the met students, they kind of talk, they, you know, they can maybe talk crap about other people, talk crap about other medical students, talk crap about other residents, talk crap about this attending that is treating them unfairly, right? But if, for example, your school designs an intervention and they come and put like a camera or like a recording device, right? Or let's say they say, you know what? Instead of having the met student work room like like on another, in another building of the hospital, let's put it right next to the attending work room. I promise you, all those met students will start behaving properly, right? That's the fourth on effect. They know they are being watched, right? So they begin to behave in a somewhat, somewhat different way. And then another quality improvement buzzword, you want to be aware of for tests is something called the Weber effect. I don't if it's Weber, I think if you're German, you probably pronounce it as Weber, but it's W E B E R, right? W E B E R. And basically, it just means that if something, if you're tracking something, then the incidence of that thing is more likely to calm down, right?

So what do I, let me give you an example, right? So let's say, for example, right? Let's say, um, divine checks, you know, you everyday to see if you study or not, right? You know, divine checks, you know, you everyday, right? Um, if, for example, you're a chronic like procrastinator, you don't study, and let's say whenever divine checks, you know, you, divine tells you like, dude, come on, really? You know, studying what kind of terrible person are you? blah, blah, blah, blah, right? So it's like, when I'm tracking, initially, I'll see like a lot of bad behavior, like not studying, not studying, not studying, but you know, after a while, you kind of feel like crap, you're like, okay, you know, I don't want to disappoint divine blah, blah, blah, blah, blah. So let me start studying, right? That's the Weber, the Weber effect, right? So if there is like, if you start tracking adverse events, in a hospital, over time, the incidence of those adverse events will go down, because people know that, ooh, they're actually tracking this thing now. Maybe if the notice that this thing is so high, they'll reduce my pay or something like that, right? So by tracking those adverse events, right? People are beginning to get, you know, people get a little more responsible, right? And they start doing their jobs better, right? So essentially, like in the first year, when you start like tracking an adverse event, that's probably where you'll find the highest incidence.

But as time goes on, it begins to taper off, okay? That's what's known as the Weber effect. That's also known as the Weber effect, depending on the part of the world, or I guess the country you're from. And then there is something called the safety champion, right? The safety champion is just essentially the person that's placed in charge of quality and safety. Is the person that's placed in charge of quality and safety at like an institution, like a hospital in this case, since we're dealing with the USMLA exams. And then there is this equation that you want to keep at the back of your mind. It's an equation that kind of determines our value, right? Like you'll probably hear this term thrown around a lot of medicine. Again, it's a high yield buzzword to know for the USMLA exams. And this is the thing that's known as a high value care, right? High value care, right? High value care, right? And I'll give you an example that would really like just ingrain this into your mind on a permanent basis, right? So this equation, I'll say twice, its value is equal to service times quality divided by cost. So value is equal to service times quality divided by cost, right? So it's like value is directly proportional to service and quality, but value is inversely proportional to the cost, right? For example, residents are very valuable to a hospital. Why do you think that's the case? Residents for the most part, you know, they tend to be pretty smart, right?

So they provide very good service, right? So they're providing good value. They tend to provide, well, maybe you know, as you study now, you know, you may not provide the best quality service, but as time goes on, as you'll learn your craft, you get better, you begin to provide higher, higher quality service. I mean, as a third year next to it, right? You know, you're probably pretty crappy at the beginning with writing notes and with round E, maybe it takes you forever, right? But as time goes on, let's say by the end of your 30 year, you know, you're kind of like an expert on things. And then maybe you start residency and then you notice that, oh man, I'm crappy again, right? So basically, right? So residents are good because they provide very good service. They provide it at very high quality. And the salaries are fixed, right? So a resident who is not going to be paid per hour, no resident, if you put in 40 hours of work, you're going to be paid the exact same as if you're putting 80 hours of work, right? So they provide good service at high quality for very low cost, right? So because the cost is low, right? The value is really high, right? The value is really high. I'm not going to expand on this any further, but I can imagine that there are certain problems you can think about of health care, think about with regards to health care that arises from this mantra, but that's probably a more like an ethical discussion that is beyond the scope of this of this podcast.

Now, one of the quality and safety buzzword to keep in mind, for example, is something called the PDSA cycle, right? The PDSA cycle, the PDSA cycle is essentially, essentially, it's just a fancy term that someone just slapped on something that is essentially, it's essentially the scientific method. It's something that we've known for decades and decades and decades. Sometimes I feel like in education, we just make things a little more complicated than they need to be. This is essentially the scientific method, but it just adapted it to quality and safety. And then, you know, give it this shining new term, PDSA cycle to make it sound all fun and games, right? But it's pretty much the scientific method, but you don't want to know what PDSA stands for, right? So P stands for plan, right? So basically, right, you know, you plan an intervention, right? So let's say, for example, your a Met School is like, okay, you know, what we want to ask students to, we are noticing that the US Emily averages that are met school as pretty crappy. Let's say like the average is like 200, right? You know, that doesn't very consistent with being able to like, you know, like match, stuff like that. So you're like, okay, let's, let's do a PDSA cycle on this then, right? So, you know, you kind of like set a plan in place.

You're like, you know, we're going to give every student on the first day of Met School, we're going to give them a shining new laptop with Anki installed and with a copy of her state and with subscriptions to your world, US Emily RX, Ambals and Kaplan, right? You set that plan in place, right? And then you actually do it. You actually, you know, kind of put money together, which will, you know, probably come from like increased tuition, right? Where they will, you know, give you all the stuff, give you the laptops and everything you need, right? And then let's say that class, you know, they follow them for two years, they go ahead and take the test, right? And they study the results, right? You know, they ask you, you know, just fill out this questionnaire when you get your US Emily scores. And then the A stands for act, right? So plan do study act, act is you're like, rule, okay? Wow, in just two years, the average US Emily scores that this Met School went from like a 200 to a 290, you're like, oh, this didn't really work, right? So you just keep doing it. Right? So it's almost like, you then go back, it's like a cycle, right? Just keeps going and going and going and going, right? So that's the PDSA cycle. Now, one other quality and safety buzzword you want to keep at the back of your mind is something called lean, like L, E, A, N, right? Just being lean, right? Essentially, lean is, essentially something that was, I think, developed by like the Toyota company, right?

The Toyota company. Essentially, it's something where you try to find inefficiencies in a process and you just go ahead and eliminate them, right? It's almost like you are all pretty in the cutting edge at every single step in the process, right? So for example, right, it just basically means you're trying to like reduce waste, you're trying to eliminate waste, right? So, I mean, this is something that, you know, a lot of work I'm probably doing their lives to, you know, kind of make themselves more financially stable, you know, just kind of look at the expenses for a month. I just say, hmm, I spend 10 on this. I spend 30 on this. I spend 50 on that. And many of these things are like, man, I really don't need this thing. Like, there are some subscriptions you probably have to like some like subscriber thing that, you know, you don't really need to maybe use like once a month, right? You know, you can maybe even need those inefficiencies. Basically, I applied that to a healthcare system. That's the lean, that's the lean model, okay? That's the lean model. So again, that's something that you really want to make sure you know and understand for purposes of the of the USML exams. Now, one other thing that you don't commonly find in many resources was actually really high up to know for exams. Again, it's a very common quality and safety terminology. It's something that I call like D mic. Like, let me say what it is. It's like D-M-A-I-C, okay? D-M-A-I-C.

It's essentially the P-D-S-A cycle. It's just another shiny turf. When you never see things like this, you're just like, come on, what are you guys doing, right? But it's essentially the P-D-S-A cycle, but it's just a different term, right? So the MBM is sort of putting like P-D-S-A cycle as an answer. They can put DMA, they can put all these acronyms as answers and then get all confused, right? Basically, the DMA I-C just says, again, it's another fancy scientific method. Basically, the D stands for define, right? The M stands for measure, the A stands for analyze, the A stands for improve, and then the C stands for control. And you'll see what I'm talking about this because it kind of leads into another concept that you miss you on a test, right? So basically, the D right define, right? You essentially like, okay, well, it's almost like, let's say a person has a, let's say a common problem is like, met students are writing crappy notes, right? So you put that down, you define the problem. Crapy notes on the parts of met students, right? And then, oh, you know, let me not put that, let me say met students take four hours to round on one, or to pre-round on two patients, right? So that's the problem, right? So you measure it, you're like, and these met students have really spent like four hours, right? And then you kind of analyze like, okay, why do these spent four hours rounding on, pre-rounding on people?

Maybe like, you know, they don't have like a checklist that they use or maybe they, they look at certain parts of the medical record that, you know, probably doesn't provide as much yield for them, you know, you kind of analyze the problem, right? And then, you know, you try to figure out where the problems are, and then the items like improve, right? So like, so define measure, analyze improve, right? So improve is you're like, okay, let me perform an intervention, let me make like a checklist that met students can use to pre-round on this service, right? And then, you then see if it works, right? So that's the control part, you see, if it works, if it works, right? Then you essentially put it as like a control measure in place. So people just stop screwing up, right? Now, the thing is DMIC, it's a little better than PDSA. So this is a subtle, but I guess somewhat higher difference from PDSA. DMIC is driven by data. You actually study in the process versus PDSA where you're just like, let me just make a plan and see if it works, right? But in the DMIC approach, right? You're actually like, okay, let me study this problem. It's almost like you're doing research on the problem and then you then propound a solution, right? So why is DMIC important to know for exams? DMIC is important to know for exams because it plays into this concept that is known as six sigma, okay? It plays into this concept known as six sigma.

And I feel like in general, when people hear this term six sigma, at least that was definitely me until I kind of understood what it meant, you know, people just have shoulders go down their spine. You don't need to have shoulders go down your spine, right? So let me explain like what in the world of six sigma mean to start off? Basically, six sigma is a process where you try to remove like defects, right? So let's say you're a company that mass produces stuff, right? Let's say you produce like a million products, right? And you're like, you know what? I want to put processes in place so that out of that, out of those are one million products I produce, I want to only have like, let's say like N95s, right? So let's say you're making N95 masks, that's your whole duty as a company. You're like, you know what? Of all these N95 masks of per million N95 masks, I want to make sure that no more than three are defective, right? No more than three are defective. If you said that as your goal, you're essentially following six sigma principles, right? And people are like, when in the world of six sigma stands for basically it's sigma. I remember many of us already know about sigma. We know that sigma is like, oh, like if you're dealing with like a standard curve, right? Remember that you're like, oh, you know, one standard deviation about the mean, right? Is like 68% of the population. Two standard deviations is like 95%. Three standard deviations is like 99. Something percent, right?

The thing is if you go all the way to six standard deviations, that's essentially like three parts per million, right? So you're essentially shooting for a goal of like three defects per million, right? So that's why you do like that DNA IC process. So you can create a process that has such fidelity, right? I mean, DNA is pretty good, right? DNA makes what is like six errors per billion or something like that, right? DNA is definitely like almost like six sigma on steroids, right? So that's one thing you want to understand. But here's one thing I want to try to make sure you don't scrub. The lean model is different from the six sigma model. In the lean model, you're trying to remove inefficiencies. You're trying to eliminate waste, okay? In the six sigma model, you're trying to eliminate defects. You're trying to not make defective products, right? So there's a big difference between wasting resources and making defective products, right? So it depends on what you're trying to eliminate. That's the key differentiator between the lean model and the six sigma model, right? So in the lean model, you're trying to eliminate waste in the six sigma model, you're trying to eliminate defects, okay? That's something that is super, super high yield to understand. Now, the next thing I want to talk about is something that's called like FMEA, right? So FMEA just means like failure mode effects analysis.

Basically failure mode effect analysis is essentially like mechanism that you put in place to like find errors in a process before those errors are going to happen, right? So let me break down the term because again, that's one thing I love doing. I feel like when you break down terms for people, you know, it kind of sticks in their minds a lot easier to remember and understand, right? So FMEA, what does mean? It means failure mode, or failure mode and effects analysis. Basically, the failure mode part of things, the FM part of things just says, oh, what are the different ways? What are the different modes that something might feel, right? So like if for example, you're like, oh, before a student, he won't start symmetrical, you're like, hmm, what are the different ways that this student can fail? Step one or fail their class exams. You then say, oh, you know, maybe they could fail by not studying, by putting all the time and all that stuff, right? So you figure that out. And then you do an effect analysis to just say, okay, what are the consequences of making these failures? You know, like not matching is an effect, right? Or not graduating from med school is an effect, or you know, getting frustrated in med school and like, quitting med school is an effect, right? So that's the effects analysis. So the failure mode is like, this is a prospective thing. That's very high, you know, this is a prospective thing.

Essentially, you're like looking at the ways something could potentially fail, right? And then you're looking at the effects of that thing, right? That's what's known as FMEA, right? That's failure modes and effect analysis. Now unfortunately, some weird things that your friends at the MBN is still can expect you to know with these failure modes and effect analysis thing is essentially, you'll try to find, you'll try to find like the different ways, like this thing can fail, right? And you'll find like the biggest problems, right? Like the biggest causes of, you find like the biggest causes of the failures, right? And you then try to rank them, right? And then as you rank them, you'll try to say, okay, you know what? Maybe let's let's work on the biggest, let's work on the biggest potential error because I mean, you cannot necessarily eliminate every single error in a process, right? That takes forever, right? So you know, you may want to try to like, rank those things, but I feel like it's probably more important to just understand what the buzzword actually means. Fereon modes and effects analysis. And then a close causing of the FMEA approach is root cause analysis. root cause analysis is essentially the same thing. root cause analysis essentially the same thing, right? It's just instead of looking at things prospectively before they happen, you're looking at things retrospectively after they have happened, right?

You look at things retrospectively after they have happened. When you do stuff like that, that's what's known as, that's what's known as a root cause analysis. A root cause analysis is essentially FMEA, but you're doing it in retrospect, okay? That's very important to know, right? Now, another thing that's another buzzword you want to know is the Swiss cheese model, right? The Swiss cheese model is just that you essentially just means that multiple steps and it actually kind of goes both ways, right? So multiple steps in a process is not like, oh, it's like because of this one thing that happened, this patient got harmed. No, it typically means that there are multiple steps, like, you know, something bad happened and then that screwed up another process and then that screwed up another process and then that screwed up another process and then that screwed over the patient, right? That's the Swiss cheese model, right? But the thing is from that Swiss cheese model by analyzing things with that mechanism, you can essentially try to create like multiple layers of safeguards to prevent like almost like a multi-layered safety system, right? Like putting like many slices of cheese so that if one error happens in one process, well, there's another thing that can block it, right? So for example, it's like a classic test question can be like, you know, or a classic example of this that you can make into a test question is, right? In the hospital, right?

If you are ordering the wrong dose of something, most EM Rs will kind of be like, you're only like a thousand milligrams of like, I don't know, like, like fentanyl, for example, you're like, oh, I know about that, right? So it's like, you put something almost like something in place. I'll talk about forced functions in the future, right? But it's almost like you put something in place to be like, like, let's say basically the computer doesn't let you order like a thousand milligrams of fentanyl, right? So that's like something you put in place, right? That's like one layer of cheese. But let's say, for example, by some mechanism, the hospital, kind of like, you know, the computer has like fault or virus or something and you're mistakenly able to order like a thousand milligrams of fentanyl. Well, another layer of cheese on top of that is the pharmacist, right? A pharmacist, right? A thousand makes like, I tried to like, I leave it painting like two million people, no, right? The pharmacist will check, that's why pharmacists check prescriptions, right? Or check orders, at least for the case of hospital pharmacists, right? So that's essentially the thing that's known as the, as the Swiss cheese, as the Swiss cheese model, right? As the Swiss cheese model. So I think since I want to keep this on the 30 minutes, because again, I want this to be something that's all kind of listened to and not get over well, right? So I'm going to go ahead and stop here.

As I do at the, at the end of every podcast, right? I do offer one or one two to many exams for step four, step one for step two, CK for step two, Cs and for step three, right? I offered one or one two to do for all those things. And then I do one or one two to do for pre-clinical medical exams, 30-ish off exams. If I also offer booster courses, right? So it's 20 hours for step one or step two, CK for step three. And again, basically in those courses, I review the most notes, the highest of the high yields. For those respective exams, we cannot touch on every subject pretty much, right? Like the high yield stuff. It's rapid fire. But again, the vast majority of people have done this. We they found it to be extremely, extremely, extremely helpful. And then there is a new course that I've started offering to a pharmacology course. It's 30 hours. And essentially in 30 hours, I will cover with you, all the pharmacology that is tested on the USML step one exam. So if that is something you are interested in, just send me an email. I will give you information on like the cost. But basically, and I'll probably make a podcast where I just kind of talk about all the things I offer because people keep asking me one question after the other. So essentially, it's a pharmacology course. This is something you can do before you're dedicated period, or you can even do during your dedicated period. I'll just meet you for like an hour every day for 30 days.

And after that, our everyday for 30 days, all the pharmacology infested. And the thing is, obviously, right, you could just read first it on your own. But essentially what I do in the courses, I will show you how because the thing is your friends at the end being, if you don't ask, like, just straight up, like, oh, what is the making of a function of this role? No, they integrate from a college with pathology, with physiology, with anatomy, and with many other things. So the pharmacology course is 30 hours long, but it integrates all these things together. So if that's something you're interested in, feel free to reach out to me. And then if you're medicine resident, or a piece resident, I need student for the entry and exam or the board exams for those specialties, I reach out to me, I do offer tutoring for those. And then if you know someone that needs to be for the MCAT, I do offer tutoring for the MCAT as well. Or basically, I'm also most of the pre-meds subject. And then finally, if you, I can offer consulting services, right? So if you're medicine applying to a residency, so like an ERAS application, or a college student applying to med school, so like an AMCA application, I might offer like one or one like advising coaching, like, you know, rec letters, editing, slash writing, personal statements, editing applications, mock interviews for a diverse range of, I've worked with a lot of people from pretty much most of the common specialties.

Again, the vast majority of people have worked with them pretty much all much that their first choices. And I actually have admissions committee experience. I've been on an admissions committee for a year, right? So if you have like a tricky application like you graduated from med school a long time ago, you have low scores, you have nonexistent research. I can have worked with those kinds of applications and being successful with them. So if you want to put your best foot forward with those kinds of things, again, reach out to me and I'll be more than happy to point you out in the right direction. And then my life lesson for today is being understanding, being understanding, right? This is something that will help us as much students, well, I'm not a much anymore, but as residents as attendants and stuff, the thing is never say that you understand where a person is coming from, right? Well, I mean, you can say there's nothing wrong with that. But the thing is when people are going through a problem, right? Like, even if you've gone through the same problem, they may have a unique way. They are facing that problem that you don't really understand, right? So it's just typically prudent to just before you just judge someone, before you're like super judgmental and a person, you know, and again, this is Lisa. I mean, I admit this myself, you know, definitely I've been judgmental in the past and I'm still probably still judgmental today.

But that's something that I, you know, I keep working on, right? But don't judge people. Just try to listen to them first. Don't always be the person that just all immediately chips into a conversation. You know, just sometimes all people want is just a listening ear. You know, just go ahead, you know, just sit down, listen to the person's complaints, listen to everything that's going on. Many times if you're practicing listening, you can actually be, you'll actually be better at giving like a good response to their problem. And sometimes people just want to just hear them like just sell their problem, say it out loud and you just feel better, just from listening, right? So you know, don't judge people. Don't like keep interjecting. Just go ahead and listen to them. Remember, the Bible says that you should judge not so that you yourself will not be judged, right? And I mean, there's a part of the Bible that also says before you see the spec in another month's eye, right? Remove the plank that you're having your eye, right? So, you know, just don't be judgmental. Listen to people. Like even let's say a person is an added to you like, ah, you're smoking, you're using drugs, you're drinking a ton of alcohol all the time. No, no, no, don't judge them. Just listen to them, right? That listening is very effective.

In fact, sometimes I think of it as almost like motivational listening, like listening, in fact, this is probably something I'll unpack as my life lesson in the next podcast. But just listening, if you're a good listener, there are many life benefits you can get from it. So thank you for listening to those podcasts. Sorry, I run board for a bit, but again, please make sure you know everything in this podcast is super, super high yield, for example. And if you want to reach out to me, you can either go to the website, divine intervention. Divine intervention podcasts with an S at the end.com, right? You can reach out to me through that medium or you can send me an email, divine intervention podcasts with an S at the end at gmail.com. And please subscribe to the Word Press website, subscribe to the podcasts. They are all on Apple podcasts on Google Play and on Spotify. And, you know, leave any feedback. That's always appreciated. And also the You Tube channel is called Divine intervention, USMLE podcasts and videos. So thank you for listening to me. God bless you. I'll see you in the next podcast. Thank you.

Practice questions — USMLE style

Question 1 — Quality Improvement

A hospital quality committee is investigating a recent increase in medication errors during patient handoffs. The team decides to implement a structured process improvement initiative. They begin by defining the scope of the problem, collecting data on current error rates and contributing factors, analyzing the collected data to identify root causes (e.g., poor communication protocols), designing a new standardized checklist intervention, and finally, implementing and monitoring the new protocol to ensure sustained compliance. Which quality improvement methodology is best represented by this systematic, data-driven approach?

  • A) Plan-Do-Study-Act (PDSA) cycle
  • B) Failure Mode and Effects Analysis (FMEA)
  • C) Define-Measure-Analyze-Improve-Control (DMAIC)
  • D) Root Cause Analysis (RCA)

Answer: C. The DMAIC framework is a structured, data-driven methodology used for process improvement. It specifically requires the initial steps of defining the problem and measuring existing performance metrics before analyzing causes and implementing improvements. While PDSA is also an iterative cycle, DMAIC emphasizes the quantitative analysis of data (Measure and Analyze) to drive the intervention, making it superior for large-scale quality initiatives like this one.

Question 2 — Patient Safety

A patient suffers a severe adverse event due to a combination of factors: first, the electronic medical record (EMR) system failed to flag an incorrect drug dosage; second, the pharmacy technician processed the order without verifying the dose against standard protocols; and third, the attending physician who reviewed the chart was distracted by another task. Which model best describes this scenario, suggesting that multiple independent failures contributed to the patient's harm?

  • A) The Weber effect
  • B) The Hawthorne effect
  • C) The Swiss cheese model
  • D) The high-value care equation

Answer: C. The Swiss cheese model posits that adverse events rarely result from a single point of failure. Instead, they typically occur when multiple layers of safeguards (the "slices" of cheese) fail simultaneously or sequentially. In this case, the EMR system, the pharmacy technician's verification process, and the attending physician's attention all represent separate layers of defense that failed, leading to harm.

Question 3 — Process Analysis

A medical team is tasked with preventing potential complications in a new surgical procedure before it is ever performed on a patient. They systematically brainstorm every possible way the procedure could fail (e.g., equipment malfunction, human error, incorrect timing) and then assess the clinical consequences of each failure mode to prioritize risk mitigation strategies. This proactive approach falls under which quality improvement tool?

  • A) Root Cause Analysis (RCA)
  • B) Quality Assurance (QA)
  • C) Failure Mode and Effects Analysis (FMEA)
  • D) Six Sigma methodology

Answer: C. FMEA is a prospective analysis tool used to identify potential failure modes in a process before they occur. The team is systematically looking forward ("what if") to predict how the procedure could fail and what those consequences would be, which is the definition of Failure Mode and Effects Analysis. RCA, conversely, looks backward at events that have already happened.

Question 4 — Quality Improvement Models

A hospital aims to improve its patient throughput efficiency. They decide between two models: Model A focuses on eliminating all unnecessary steps in the current workflow (e.g., redundant paperwork, waiting times) to reduce resource waste; and Model B focuses on reducing the rate of surgical site infections from 10 per million procedures to less than three per million. What is the key difference between these two approaches?

  • A) Model A represents Quality Assurance, while Model B represents Quality Improvement.
  • B) Model A aligns with Six Sigma principles by eliminating defects, whereas Model B uses Lean methodology by removing waste.
  • C) Model A aligns with Lean principles by eliminating waste/inefficiencies, while Model B aligns with Six Sigma principles by reducing the rate of defects.
  • D) Both models are examples of Quality Control and require daily monitoring to be effective.

Answer: C. The key differentiator is the target of elimination. The Lean model focuses on identifying and removing waste (non-value-added steps or inefficiencies), which aligns with Model A's goal of streamlining workflow. Six Sigma, conversely, is a methodology focused specifically on reducing the rate of defects to an extremely low level (e.g., 3 defects per million), aligning with Model B's goal of minimizing infections.

Quick fire review

What is the key difference between Quality Control (QC) and Quality Assurance (QA)?

QC involves frequent, daily checks on a product or process output (e.g., checking notes every day). QA is less frequent, periodic auditing to ensure standards are maintained (e.g., checking notes once a week).

What does the 'I' stand for in Quality Improvement (QI)?

Improvement. QI focuses on designing an intervention to actively improve a process or outcome.

Which quality improvement model is considered more data-driven and comprehensive than PDSA?

DMAIC (Define, Measure, Analyze, Improve, Control).

What is the primary goal of Lean methodology in healthcare?

To eliminate waste and inefficiencies from a process.

If you are looking at why an adverse event occurred after it has already happened, what analysis method should you use?

Root Cause Analysis (RCA), which is retrospective.

What does the Hawthorne Effect describe in a clinical setting?

The tendency for people to change their behavior when they know they are being observed or monitored.

In the context of patient safety, what concept suggests that multiple layers of safeguards must be in place because no single safeguard is foolproof?

The Swiss Cheese Model.

What does 'V' represent in the Value Equation (Value = Service $\times$ Quality / Cost)?

Value is directly proportional to Service and Quality, but inversely proportional to Cost.

Which quality improvement tool is prospective—meaning it predicts potential failures before they happen?

Failure Mode and Effects Analysis (FMEA).

What does the acronym DMAIC stand for in process improvement?

Define $\rightarrow$ Measure $\rightarrow$ Analyze $\rightarrow$ Improve $\rightarrow$ Control.

If a system is designed to reduce defects to 3 parts per million, what quality standard principle is being applied?

Six Sigma principles.

What distinguishes the Lean model from the Six Sigma model?

Lean aims to eliminate waste (inefficiencies), while Six Sigma aims to eliminate defects.

If a resident checks your notes every single day, which quality process is this?

Quality Control (QC).

What term describes the decrease in adverse event reporting over time due to increased awareness and tracking?

Weber Effect.

Quick recall / Anki-style questions

What does 'V' represent in the Value Equation (Value = Service $\times$ Quality / Cost)?

Value is directly proportional to Service and Quality, but inversely proportional to Cost.

Which quality improvement tool is prospective—meaning it predicts potential failures before they happen?

Failure Mode and Effects Analysis (FMEA).

What does the acronym DMAIC stand for in process improvement?

Define $\rightarrow$ Measure $\rightarrow$ Analyze $\rightarrow$ Improve $\rightarrow$ Control.

If a system is designed to reduce defects to 3 parts per million, what quality standard principle is being applied?

Six Sigma principles.

What distinguishes the Lean model from the Six Sigma model?

Lean aims to eliminate waste (inefficiencies), while Six Sigma aims to eliminate defects.

If a resident checks your notes every single day, which quality process is this?

Quality Control (QC).

What term describes the decrease in adverse event reporting over time due to increased awareness and tracking?

Weber Effect.