Internal Quality Control...1 2S violation Internal Quality Control Control Samples In principle any...

Internal Quality Control

Piet Meijer


Outline:

• The patient story

• Reliability of test results

• Quality Control

• Principals of Internal Quality Control

• Evaluation Quality Controls Results

• Use of Quality Control Results


The Patient Story


Dr. Marie Chantal Nzayisaba meets with a Congolese patient at the health center in Maratane refugee camp in Mozambique

https://www.unrefugees.org.au/our-stories/dr-marie-chantal-nzayisabas-story/


The story of the patient (1)


Laboratory testing and diagnosis

Qualitative

Microscopy

Tuberculoses

Yes / No

Semi-quantitative

Urine Dipstick

Urinary infection

Change color

QuantitativeViral load

HIV

CD4 cut-off


The story of the patient (2)

DiagnosisTreatmentMonitoring Effect

LABORATORY TEST RESULTS

Reliable test result:

Yes or No?


Qualitative test result

Tuberculoses

Correct treatment

Tuberculoses

Incorrect treatment

https://www.thebureauinvestigates.com/stories/2018-09-25/how-india-is-trying-but-failing-to-get-on-top-of-its-tb-crisis

TRUE

WRONG

Diagnosis

https://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=2ahUKEwjY4OGu8PrhAhWBYVAKHXfrArIQjRx6BAgBEAU&url=https://www.usaid.gov/news-information/frontlines/may-june-2016/hills-and-valleys-tuberculosis-philippines&psig=AOvVaw2cRssRqB3W_lbNvTJ8WdU4&ust=1556818756388568


Semi-quantitative test result

Urinary infection

Correct treatment

Urinary Infection

Incorrect treatment

Diagnosis

TRUE

WRONG


Quantitative test result

HIV positive

Correct treatment

HIV negative

No treatment

Diagnosis

TRUE

WRONG


TEST RESULT

CORRECT

CORRECT DIAGNOSIS AND TREATMENT

INCORRECT

INCORRECT DIAGNOSIS AND TREATMENT


Reliability of Test Results


Result

CORRECT TEST RESULT

PRECISION

ACCURACY

Each test result is affected by

both imprecision and inaccuracy

https://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=0ahUKEwj4zsuaj5bPAhWBCBoKHdD7DQMQjRwIBw&url=https://nl.wikipedia.org/wiki/Normale_verdeling&psig=AFQjCNE1-E81OBYoKKwsGm1bxUctsRWewA&ust=1474191917036460


Random errors are errors in measurement that lead to measurable

values being inconsistent when repeated measurements of a constant

attribute or quantity are taken.

PRECISION

1 5.2

2 5.4

3 5.5

4 5.3

5 5.4

6 5.3

Random errors

Wikipedia


ACCURACY

Systematic errors are errors that are not determined by chance but

are introduced by an inaccuracy

Wikipedia


PreciseAccurate

ImpreciseAccurate

PreciseInaccurate

ImpreciseInaccurate

CORRECT TEST RESULT



Precise and accurate

Low risk for misdiagnosis / mistreatment

Imprecise and accurate

Low risk for misdiagnosis

Medium risk for mistreatment



Precise and Inaccurate

Medium/high risk for misdiagnosis

Low risk for mistreatment

Imprecise and Inaccurate

Medium/high risk for misdiagnosis

Medium risk for mistreatment


Qualitative test result

• Microscope

• Blood smear

• Quality of staining

• Quality of staining reagent

• Experience technician


Semi-quantitative test result

• Expiry date of dipstick

• Storage

• Quality of urine



• Reagents

• Maintenance equipment

• Quality blood sample


Quality Control


Laboratory quality control is designed to detect, reduce, and correct

deficiencies in a laboratory's internal analytical process prior to the

release of patient results, in order to improve the quality of the results

reported by the laboratory.

QUALITY CONTROL

Wikipedia


IQC EQC

Control of Analytical Process

Does the analytical test

system provides reliable

result?

Does the analytical test

system provides accurate

result?


Direct Indirect

Type of Quality Control

In the case of direct quality control, a sample checked the quality of

the product itself.

E.g. the length of nails

With indirect quality control, the quality of a product is checked

indirectly.

E.g. the result of a QC sample instead of a patient sample


Quality control in a medical laboratory is by definition always indirect quality control.

The quality of the test result is never performed directly on the determination of a

patient sample, but with the help of separate quality control samples.

This means that a quality control sample must be representative of a patient sample.

The so-called like-versus-like principle.

There is always a risk that the result of the quality control sample does not reflect the

quality of the test result of a patient sample. This can e.g. by a difference in matrix or

by interfering components in the plasma.

Type of Quality Control in Medical Laboratory


Principles of



Qualitative testing

• Check with the same slide the magnitude of the microscope

• Check with a known strain of the bacteria the of staining reagent

and staining procedure

• Check with known slides the experience of the technician


Example Qualitative testing

• Ziehl–Neelsen staining

• New preparation of staining reagent

• First check with a known strain of the Mycobacterium tuberculosis

whether the reagent is working appropriate


Semi-quantitative testing

• Check of the quality of the urine strips

• Check of the quality of the urine


Example Semi-quantitative testing

• New batch of urine dipsticks

• With a known urine sample the quality of the dipsticks is tested


Example Semi-quantitative testing

1 √

2 √

3 √

4 √

5 √


Quantitative testing

Measurement of blood glucose

• Check of the quality of the reagents

• Check of the quality of the blood sample

• Check of the quality of the equipment


Example Quantitative testing


• New batch of reagents

• With a known serum or plasma sample the quality of the reagents

is tested


Example Quantitative testing


mMol/L

1 5.0

2 5.1

3 5.0

4 5.2

5 4.9


Evaluation


Results


Quality control results are used to consider whether a laboratory test

is working appropriate.

For example:

• Qualitative testing: record whether the staining reagent

give the correct result.

• Semi-quantitative testing: record whether a urine control

sample gives the same colour

change with a dipstick.

EVALUATION OF QUALITY CONTROL


EVALUATION OF QUALITY CONTROL

• Quantitative testing: Record wether the test result of a

control sample gives acceptable results.

Simple graphical and statistical tools


Example of evaluation quality control results

Results for parameter X:

205

185

195

205

195

205

200

215

210

185

Next results is 225.

Is this good yes or no?


Example of evaluation quality control results

Results for parameter X:

205

185

195

205

190

205

200

215

205

185

Next results is 225.

Is this good yes or no?

How could we know that the result of 225 is acceptable?

• Simple evaluation tools


Simple graphical and statistical tools

• Mean

• Standard deviation

• Levey-Jennings chart

• Control rules


Statistical Process Control

Statistical process control (SPC) is a method of quality control which employs statistical

methods to monitor and control a process. This helps to ensure that the process operates

efficiently, producing more specification-conforming products with less waste.

Wikipedia


Statistical Process Control

Probability for error detection (Ped)

The probability, in the case of a stable test system, to

detect an error in the analytical run.

Ped ≥ 0.90 (= at least 90% probability to detect an error)

Probability on false rejection (Pfr)

The probability, in the case of a stable test system, on

false rejection of an analytical run.

Pfr ≤ 0.05 (= equal or less than 5% probability on false

rejection)

Ped

Pfr


MeanResults for parameter X:

205

185

195

205

195

205

200

215

210

185

Sum: 2000

N = 10

Mean: 2000/10 = 200

Biorad: Basic Lessons in Laboratory Quality Control


Standard Deviation



Standard Deviation

Result Result - Mean (Result – Mean)2

205 5 25

185 -15 225

195 -5 25

205 5 25

195 -5 25

205 5 25

200 0 0

215 15 225

210 10 100

185 -15 225

900 SUM

SD = 900

9

= 10.0


Levey-Jennings Chart


200

210

220

230

170

180

190




200

210

220

230

170

180

190

?


Control rules

Control rules are used to define specific performance limits for a particular assay and can

be used to detect both random and systematic errors.


Single Rule Multi Rules

With multi-rules a combi-

nation of single rules are

used for the interpretation

of QC results

A single rule (e.g. 2 SD) is

used for the interpretation

of QC results

Control rules


12S

Westgard website



In the next part of the course the use of multi rules will be explained

200

210

220

230

170

180

190

12S violation


Control Samples

In principle any deviation, whether systematic or random, should

be reflected to the same extent in the results for both patient and

control samples. Thus, the control result is an indicator of the

quality of the run and, hence, of the quality of the patient sample.

Hyltoft Petersen et al Eur J Clin Chem Clin Biochem 1996; 34: 983 - 999


Control Samples

Sources of control samples

• Commercial samples

• Patient samples and sample of healthy donors (abnormal / normal)


WHO: Laboratory Quality Management System



Local Predetermination


Alternative Quality Control

In the case no control samples are available an alternative QC approach can be

used:

• Calculate on a daily basis the median value of normal and pathological patient

samples

• Make a graphical presentation of the mean values

• Evaluate trends


Use of Quality Control Results


The purpose of colleting quality control results is to see whether a laboratory test,

independent which type of test (qualitative or (semi) quantitative) is within or out of

control.

Quality Control Results


Out of Control

What to do when a test is out of control?

• The testing process should be stopped immediately and first the source of the

problem should be identified.

• When the problem is be identified, corrective actions should be performed.

• Quality control should be repeated and should be considered within control before

patient samples can be tested again.

DO NOT REPORT PATIENT RESULTS UNTILL THE PROBLEM IS SOLVED AND

CONTROLS INDICATE PROPER PERFORMANCE!



Potential causus

• Outdated reagents

• Inproper stored reagents / consumables

• Inproper control material

• Calibration error

• Equipment / analyzer failure

• Inproper use of test procedure

• Etc.


Documentation

• Document the problem identified and the corrective action(s) taken

• Use this information testing improvement processes and for training purposes


Summary


Design Quality Control

Processes

Implement Quality

Control Processes

Perform Quality

Control Processes

Evaluate Quality

Control Results

Perform corrective

actions when

necessary

Contineous

Improvement Testing

Processes


FOR THE BENEFIT

OF THE PATIENT!

Internal Quality Control...1 2S violation Internal Quality Control Control Samples In principle any...

Documents

Transcript of Internal Quality Control...1 2S violation Internal Quality Control Control Samples In principle any...