[Durgical](https://durgical.com/en/) › [DurgiPedia](https://durgical.com/en/durgipedia/) › [Epidemiology](https://durgical.com/en/epidemiology/) › [Health management and leadership](https://durgical.com/en/epidemiology/areas/management/) › Quality improvement methods: PDSA, run charts and root causes

# Quality improvement methods: PDSA, run charts and root causes

- Concept

Most care problems are fixed by small tests of change, measured week by week, not by one big plan. These tools tell you what to change and whether it worked.

Updated Nov 19, 2026

## Key points

- Model for improvement: what are we trying to accomplish, how will we know a change is an improvement, what change can we make?
- [PDSA](https://durgical.com/en/nursing/the-pdca-cycle-in-nursing-services/): Plan (prediction and measure), Do (test small), Study (compare with the prediction), Act (adopt, adapt or abandon).
- Run chart: data over time with the median; 6 or more consecutive points on one side of the median is a shift, a non-random signal.
- Pareto: rank causes by frequency; usually a few causes explain most of the problem (the 80/20 rule).
- Fishbone: causes grouped as people, methods, equipment, materials, environment and measurement.
- Clinical audit cycle: choose a standard, measure, compare, change, re-audit.

On this page

- The basics Key points
- What it is
- How to think about it
- Key facts
- Worked example
- Which tool answers which question
- Markers
- In practice
- Common pitfalls
- Common traps

- Study Test yourself

- Connections Connections

## What it is

Quality improvement is the systematic use of small, measured changes to make care safer, faster and more reliable. Its core tools are the Plan-Do-Study-Act (PDSA) cycle, run charts to see change over time, Pareto charts to rank causes, fishbone (Ishikawa) diagrams to list them, process maps to see the steps, and clinical audit to compare practice with a standard. In Colombia, the audit programme for quality improvement (PAMEC) is how each provider applies these methods within the [SOGCS](https://durgical.com/en/nursing/colombia-s-mandatory-health-care-quality-system-habilitacion-auditoria-and-acred/).

## How to think about it

Improve like a scientist with a small budget: guess, test on a few patients, look at the data, and adjust. Many small cycles that each teach something beat one large rollout that fails without telling you why.

## Key facts

- Measures come in three types: outcome (what happens to patients), process (what we do) and balancing (what might get worse).
- Start a PDSA on one shift, one clinician or five patients; scale up only when it works.
- A process map drawn with the people who do the work shows the delays, loops and [handoffs](https://durgical.com/en/epidemiology/staffing-fatigue-and-patient-safety/).
- A run chart needs no statistics software; a control chart adds limits to tell common from special-cause variation.
- In Colombia, PAMEC aims above the licensing level and is guided by accreditation standards; it is not a separate quality system.
- PAMEC works through self-assessment, prioritising processes, setting expected quality, measuring, an action plan, follow-up and organisational learning.

## Worked example

- Aim: chest pain patients get an ECG within 10 minutes of triage. Baseline: median 22 minutes; 6 of 20 patients (30 %) within 10.
- Fishbone and process map: the ECG machine sits in another room and only the doctor orders it after seeing the patient.
- PDSA 1: the triage nurse records the ECG without waiting for an order. Prediction: 60 % within 10 minutes.
- Study: 14 of 20 patients (70 %) within 10 minutes, median 12; the machine is often in use elsewhere.
- PDSA 2: a dedicated ECG cart at triage. Study: 18 of 20 (90 %), median 8 minutes.
- Run chart: 8 weekly points in a row below the old median of 22 minutes, a shift beyond chance.
- Act: adopt it, add a balancing measure (triage waiting time) and re-audit in 3 months.

## Which tool answers which question

| Tool | Question it answers | Typical use |
| --- | --- | --- |
| Process map | What actually happens, step by step? | Find delays from triage to the first ECG |
| Fishbone (Ishikawa) | What could be causing it? | Brainstorm the causes of missed follow-up |
| Pareto chart | Which few causes matter most? | Rank the causes of [glosas](https://durgical.com/en/epidemiology/how-eps-pay-ips-contracts-billing-and-glosas/) or cancelled surgery |
| PDSA cycle | Does this change work here? | Test a new triage step on 20 patients |
| Run chart | Is it improving over time? | Plot weekly door-to-ECG time against the median |
| Clinical audit | Do we meet the standard? | Compare records with a guideline, then re-audit |

## Markers

- PDSA
- Run chart
- Pareto
- Ishikawa
- Clinical audit
- PAMEC

## In practice

- Pick a problem the team feels every day; the first cycle should be small enough to run next week.
- Write the prediction before the test: comparing result and prediction is how the team learns.
- Plot the data weekly where the team can see it; a visible run chart keeps the change alive.
- In a Colombian IPS, link the project to the PAMEC plan so that it counts and is followed up.

## Common pitfalls

- Testing on the whole hospital first: if it fails, nobody knows which part failed.
- Comparing one before value with one after value instead of following the data over time.
- Skipping the balancing measure: faster triage that makes everyone else wait longer.
- Treating PAMEC as paperwork for the inspector rather than as the provider's improvement plan.

## Common traps

- **Two points before and after prove an improvement**: a single pair cannot separate a real change from ordinary week-to-week variation; follow the run chart rules
- **PDSA means planning carefully, then implementing everywhere**: it means testing small and fast, learning, and only then scaling
- **Clinical audit is research**: audit checks practice against a known standard; research creates new knowledge
- **PAMEC is a fifth component apart from the SOGCS**: it is one of its components, the provider's own audit for improvement above licensing

## Test yourself

Think of the answer, then tap the question to see it.

## A ward's weekly fall counts are below the median for 7 weeks in a row after a new bed alarm. What does the run chart show?

A shift: 6 or more consecutive points on one side of the median is a non-random signal, so the change is probably working.

## Your team lists 12 causes of cancelled surgery. Which tool helps you choose where to act first?

A Pareto chart: rank the causes by how often they occur and act on the few that explain most cancellations.

## Why should a PDSA test start with 5–20 patients rather than the whole service?

A small test is cheap, fast and safe to fail; it shows what to adapt before scaling, and several quick cycles teach more than one large rollout.

## After faster discharges, what balancing measure would you track?

Something that could get worse because of the change, such as 30-day readmissions or return visits to the emergency department.

[Add to my cards (4)](https://durgical.app/estudiar/?add=/en/epidemiology/quality-improvement-methods-pdsa-run-charts-and-root-causes/)

## Connections

## Related 6

- [The Mandatory Quality Assurance System (SOGC)](https://durgical.com/en/epidemiology/the-mandatory-quality-assurance-system/)
- [Quality of care: structure, process and outcome](https://durgical.com/en/epidemiology/quality-of-care-structure-process-and-outcome/)
- [Patient safety and adverse events](https://durgical.com/en/epidemiology/patient-safety-and-adverse-events/)
- [Clinical practice guidelines](https://durgical.com/en/epidemiology/clinical-practice-guidelines/)
- [Running a clinical service: planning, capacity and indicators](https://durgical.com/en/epidemiology/running-a-clinical-service-planning-capacity-and-indicators/)
- [Learning from incidents: reporting, root cause analysis and barriers](https://durgical.com/en/epidemiology/learning-from-incidents-reporting-root-cause-analysis-and-barriers/)

[← All of epidemiology](https://durgical.com/en/epidemiology/)

Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.

Sources: [Quality Improvement Methods (LEAN, PDSA, SIX SIGMA)](https://www.ncbi.nlm.nih.gov/books/NBK599556/) (StatPearls (NCBI Bookshelf), 2026); [Quality Improvement](https://www.ncbi.nlm.nih.gov/books/NBK556097/) (StatPearls (NCBI Bookshelf), 2026); [Pautas de auditoría para el mejoramiento de la calidad de la atención en salud](https://www.saludcapital.gov.co/DDS/Documentos_I/Pautas_PAMEC.pdf) (Ministerio de la Protección Social, Colombia (copia de la Secretaría Distrital de Salud de Bogotá), 2007).

## **Spotted a mistake?** We are always improving for the community. Tell us what is wrong and our team will review it. Report a mistake

Representative model, for educational purposes only.

---
Source: https://durgical.com/en/epidemiology/quality-improvement-methods-pdsa-run-charts-and-root-causes/
License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Cite as: DurgiPedia (durgical.com), https://durgical.com/en/epidemiology/quality-improvement-methods-pdsa-run-charts-and-root-causes/
