3020

Quality Assurance/Quality Control

Reviewed by Standard Methods Committee, 2005. Additional revisions, 2017. Editorial revisions, 2021. Randy A. Gottler (chair), Rodger B. Baird, Andrew D, Eaton, William
C. Lipps.

3020 A. Introduction
Quality assurance (QA) is a laboratory operations program recommended; when must is used, the QC is required. Additional
that specifies the measures required to produce defensible data QC procedures should be used when necessary to ensure that
with known precision and accuracy. This program is defined in a results are valid. Some regulatory programs may require addi-
QA manual, written procedures, work instructions, and records. tional QC or have alternative acceptance limits. In those cases,
The manual must include a policy that defines the procedures to the laboratory should follow the more stringent requirements.
calculate statistical levels of confidence used to express data pre- The QC program consists of at least the following elements, as
cision and bias, as well as calculation of method detection levels applicable to specific methods:
(MDLs) and reporting limits. The overall system includes all QA • calibration,
policies and quality control (QC) processes needed to demon- • continuing calibration verification (CCV),
strate the laboratory’s competence and to ensure and document • operational range and MDL determination,
the quality of its analytical data. Quality systems are essential for • initial demonstration of capability (IDC),
laboratories seeking accreditation under state or federal labora- • ongoing demonstration of capability,
tory certification programs. Refer to Section 1020 for details on • method blank or reagent blank,
establishing a Quality Assurance Plan. • laboratory-fortified blank (LFB),
As described in Part 1000, essential QC measures may include • laboratory-fortified matrix (LFM),
method calibration, reagent preparation and standardization, • duplicate sample or laboratory-fortified matrix duplicate
assessment of each analyst’s capabilities, analysis of blind check (LFMD),
samples, determination of the method’s sensitivity [method • verification of MDL and MRL,
detection level (MDL, limit of detection (LOD), level of quan- • QC calculations,
tification level (LOQ), or minimum reporting level (MRL)], and • control charts,
daily evaluation of bias, precision, and laboratory contamination • corrective action,
or other analytical interference. • frequency of QC,
Some methods in Part 3000 include specific QC procedures, • QC acceptance criteria, and
frequencies, and acceptance criteria. These are considered the • definitions of prep and analytical batches.
minimum QCs needed to perform the method successfully. Sections 1010 and 1030 describe calculations for evaluating
When the word may or preferably is used, the QC is optional but data quality.

3020 B. Quality Control Practices

At a minimum, analysts must use the QC practices specified highest concentration standard defines the upper end of the cali-
here unless a method specifies alternative practices. Laborato- bration range. Ensure that the calibration range encompasses the
ries may save time and money by purchasing premade standards, analytical concentration values expected in samples or required
titrants, and reagents, but they still must perform the QC checks dilutions.
on these materials required by the analytical methods. Calibration curves may be linear through the origin, linear
not through the origin, or nonlinear through or not through the
1. Calibration origin. Some nonlinear functions can be linearized via mathe-
matical transformations (e.g., log). The following acceptance cri-
a. Instrument calibration (not applicable to non-instrumental teria are recommended for various calibration functions. If using
methods): Perform both instrument calibration and maintenance response factors or calibration factors, the calculated %RSD for
according to the manufacturer’s instructions and recommen- each analyte of interest must be less than the method specified
dations. Conduct instrument-performance checks according to value. Refer to the method for the calibration procedure and
method or standard operating procedure (SOP) instructions. acceptance criteria on the response or calibration factors for each
b. Initial calibration: Perform initial calibration using at least analyte. If linear regression is used, use the minimum correlation
3 concentrations of standards for linear curves, at least 5 concen- coefficient specified in the method. If the minimum correlation
trations of standards for nonlinear curves, or as specified by the coefficient is not specified, then a minimum value of 0.995 is
method. Set the lowest concentration at the reporting limit. The recommended. Compare each calibration point to the curve by

https://doi.org/10.2105/SMWW.2882.040 1
 3020 Quality Assurance/Quality Control - B. Quality Control Practices

recalculating its concentration. If any recalculated concentration blanks can be analyzed on multiple instruments, as long as at
is not within the method’s acceptance criteria, identify the source least 7 sets of spikes and blanks total are used. Alternatively,
of outliers and correct before sample quantitation. Use the initial determine instrument-specific MDLs.
calibration with any of the above functions (response factor, cali- Calculate the estimated sample standard deviation, ss, of the 7 rep-
bration factor, or calibration curve) to quantitate analytes of inter- licates, and multiply by 3.14 to compute the MDLs. Calculate MDLb
est in samples. Use the calibration verification (see ¶ c below) (MDL based on method blanks) using the following procedure.
only for initial-calibration checks, not for sample quantitation, If none of the method blanks give a numerical result (positive
unless otherwise specified by the method. Perform initial cali- or negative), then MDLb is not applicable, and MDL = MDLs.
bration when the instrument is set up and whenever calibration- If some give numerical results, then MDLb equals the highest
verification criteria are not met. method blank result. If all of the method blanks give numerical
c. Calibration verification: In calibration verification, analysts results, calculate MDLb as
periodically use a calibration standard to confirm that instrument
performance has not changed significantly since initial calibra- MDLb = X + 3.14Sb
tion. Base this verification on the number of samples analyzed
(e.g., after every 10 samples). Verify calibration by analyzing 1
where:
standard at a concentration near or at the midpoint of the cali-
bration range. Evaluate the calibration-verification analysis based X = mean of blank results (set negative mean value to 0), and
either on allowable deviations from the values obtained in the ini- Sb = standard deviation of blank results.
tial calibration or from specific points on the calibration curve. If
The MDL then equals whichever is greater: MDLs or MDLb.
the calibration verification is out of control, then take corrective
If using more than 7 replicates, adjust the t value from 3.14
action, including re-analysis of any affected samples. Refer to the
using student t tables with n–1 degrees of freedom.
method for the frequency of and acceptance criteria for calibra-
tion verification. The analytical results for this second-source
midrange standard must be within 10% of its true value, except 3. Initial Demonstration of Capability
for ICP-AES, which must be within 5% of its true value. If
not, determine the cause of the error, take corrective action, and Each analyst in the laboratory should conduct an IDC at least
reverify the calibration. If the reverification passes, continue the once before analyzing any sample to demonstrate proficiency in
analyses; otherwise, repeat the initial calibration. performing the method and obtaining acceptable results for each
See the individual method or manufacturer’s instructions for analyte. The IDC also is used to demonstrate that a laboratory’s
ISE methods. modifications to a method will produce results as precise and accu-
Refer to the method for CCV frequency and acceptance criteria; rate as those produced by the reference method. As a minimum,
if not specified, use the criteria given here. Other concentrations include a reagent blank and at least 4 LFBs at a concentration
(e.g., one near the MRL) may be used, but be aware that the accep- between 1 and 4 times the MRL (or other level specified in the
tance criteria may vary depending on the standard’s concentration. method). Ensure that the reagent blank does not contain any analyte
of interest at a concentration greater than half the lowest calibration
2. Operational Range and MDL Determination
point (or other level specified in the method). Ensure that precision
and accuracy (percent recovery) calculated for LFBs are within the
acceptance criteria listed in the method of choice or generated by
Before using a new method or instrument, determine its oper-
the laboratory (if there are no established mandatory criteria).
ational (calibration) range (upper and lower limits). Calibrate
To establish laboratory-generated accuracy and precision lim-
according to 3020 B.1, or verify the calibration by analyzing pre-
its, calculate the upper and lower control limits from the mean
pared standard solutions ranging from low to high concentrations.
and standard deviation of percent recovery for ≥20 data points:
Determine the maximum concentration that can be measured
within 10% of its true value based on the calibration curve: this is
the limit of linearity. Dilute all samples whose concentrations are Upper control limit = Mean + 3(Standard deviation)
above the limit of linearity. Lower control limit = Mean − 3(Standard deviation)
If reporting results < MRL, initially estimate the MDL as a
concentration about 3 to 5 times lower than the minimum cali- In the absence of established mandatory criteria, use labo-
bration standard. This method for determining the MDL is based ratory-generated acceptance criteria for the IDC or else obtain
on the procedure outlined by the U.S. Environmental Protection acceptance criteria from a proficiency testing (PT) provider on
Agency (EPA).1 PT studies and translate the data to percent recovery limits per
To determine an MDL, prepare and analyze at least 7 portions analyte and method of choice. Ensure that lab-generated criteria
of a solution spiked at or near the minimum calibration concen- are at least as tight as PT-study criteria, which are typically based
tration and an equal number of blanks. Analysts should prepare on either multiple lab results or PT-provider-fixed limits.
and analyze the spikes and blanks over 3 d rather than analyzing
them all in one batch. If 1 MDL will be used for multiple instru-
ments, then the MDL analysis must be performed across all of 4. Ongoing Demonstration of Capability (Laboratory Control
them (however, it is unnecessary to analyze all samples on all Sample)
instruments). Analysts must prepare and analyze at least 2 spikes
and 2 blanks on different calendar dates for each instrument. The ongoing demonstration of capability, sometimes called a
If evaluating more than 3 instruments, then 1 set of spikes and laboratory control sample (LCS), laboratory control standard,

https://doi.org/10.2105/SMWW.2882.040 2
 3020 Quality Assurance/Quality Control - B. Quality Control Practices

QC check sample, or laboratory-fortified blank, is used to ensure of the analyte(s) of interest has been added. The LFB may be
that the laboratory analysis remains in control while samples are used as the LCS (3020 B.4) if the method requires a preliminary
analyzed and separates laboratory performance from method per- sample extraction or digestion.
formance on the sample matrix. This standard should be preserved An LFB is used to evaluate laboratory performance and ana-
in accordance with method requirements and carried through the lyte recovery in a blank matrix. Its concentration should be high
entire procedure, including any digestions, extraction, or fil- enough to be measured precisely, but not high enough to be irrel-
tration. Purchase an external QC standard (if available) from a evant to measured environmental concentrations. The analyst
reputable supplier and use the certified acceptance limits as the should rotate LFB concentrations to cover different parts of the
laboratory acceptance criteria. calibration range. As a minimum, include 1 LFB with each sam-
Acceptance criteria vary depending on the method, matrix, and ple set (batch) or on a 5% basis, whichever is more frequent. (The
concentration. The concentration range should be either near the definition of a batch is typically project-specific.)
middle of the calibration range or near the maximum contami- Process the LFB through all sample preparation and analysis
nant level (MCL), whichever is lower. Alternatively, prepare your steps. Use an added concentration of at least 10 × MDL, or a
own QC standard and calculate acceptance limits as ±2 standard level specified in a project plan’s data quality objectives. Ideally,
deviations based on analysis of ≥20 replicates, unless the method the LFB concentration should be less than the MCL (if the con-
specifies acceptance limits. taminant has one). Depending on method requirements, prepare
The ongoing demonstration of capability may be one of the the addition solution from either the same reference source used
following: for calibration or an independent source. Evaluate the LFB for
• acceptable performance of a blind sample analysis (single percent recovery of the added analytes by comparing results to
blind to the analyst); method-specified limits, control charts, or other approved criteria.
• another IDC; If LFB results are out of control, take corrective action, including
• at least 4 consecutive LCSs with acceptable levels of preci- repreparation and re-analysis of associated samples if required.
sion and accuracy (the laboratory shall determine acceptable Use LFB results to evaluate batch performance, calculate recov-
precision and accuracy limits before analysis); or ery limits, and plot control charts.
• a documented analyst-review process using QC samples (QC
samples can be reviewed to identify individual or group pat- 7. Laboratory-Fortified Matrix
terns and determine whether corrective action or retraining
is necessary). A laboratory-fortified matrix (LFM) is an additional portion of
a sample to which a known amount of the analytes of interest is
5. Reagent Blank added before sample preparation. Some analytes (e.g., speciation
methods) are not appropriate for LFM analysis.
A reagent blank (method blank) consists of reagent water (see The LFM is used to evaluate analyte recovery in a sample matrix.
Section 1080) and all reagents (including preservatives) that nor- If an LFM is feasible and the method does not specify LFM fre-
mally are in contact with a sample during the entire analytical proce- quency requirements, then include at least 1 LFM with each sample
dure. The reagent blank is used to determine whether and how much set (batch) or on a 5% basis, whichever is more frequent. Add a
reagents and the preparative analytical steps contribute to measure- concentration that is at least 10 × MRL, less than or equal to the
ment uncertainty. As a minimum, include 1 reagent blank with each midpoint of the calibration curve, or method-specified level to the
sample set (batch) or on a 5% basis, whichever is more frequent. selected samples. The analyst should use the same concentration
Analyze a blank after the CCV standard and before analyzing sam- as for LFB (3020 B.6) to allow analysts to separate the matrix’s
ples. Evaluate reagent-blank results for contamination; if contam- effect from laboratory performance. Prepare LFM from the same
ination levels are unacceptable, identify and eliminate the source. reference source used for LFB. Make the addition such that sam-
Positive sample results are suspect if analytes in the reagent ple background levels do not adversely affect recovery (prefera-
blank are > ½ MRL, unless the method specifies otherwise. Sam- bly adjust LFM concentrations if the known sample is more than
ples analyzed with a contaminated blank must be reprepared and 5 times the background level). For example, if the sample contains
re-analyzed unless concentrations are ≥10 times those of the the analyte of interest, then add approximately as much analyte to
blank, concentrations are nondetect, or the data user will accept the LFM sample as the concentration found in the known sample.
qualified data. See the method for specific reagent-blank accep- Evaluate LFM results for percent recovery; if they are not within
tance criteria. General guidelines for qualifying sample results control limits, then take corrective action to rectify the matrix
with regard to reagent-blank quality are as follows: effect, use another method, use the method of standard addition,
• If reagent blank is < MDL, then no qualification is required. or flag the data if reported. See the method for specific LFM-ac-
• If reagent blank is > ½ MRL but < MRL and sample results ceptance criteria until the laboratory develops statistically valid,
are > MRL, then qualify results to indicate that analyte was laboratory-specific performance criteria. If the method does not
detected in the reagent blank. provide limits, use the calculated preliminary limits from the IDC
• If reagent blank is > MRL, then further corrective action and (3020 B.3). LFM control limits may be wider than for LFB or LCS,
qualification is required. and batch acceptance generally is not contingent upon LFM results.

6. Laboratory-Fortified Blank 8. Duplicate Sample/Laboratory-Fortified Matrix Duplicate

A laboratory-fortified blank (LFB) is a reagent-water sample Duplicate samples are analyzed to estimate precision. If an ana-
(with associated preservatives) to which a known concentration lyte is rarely detected in a matrix type, use an LFM duplicate.

https://doi.org/10.2105/SMWW.2882.040 3
 3020 Quality Assurance/Quality Control - B. Quality Control Practices

An LFM duplicate is a second portion of the sample described in a. Laboratory-fortified matrix (LFM) sample (matrix spike sample):
3020 B.7 to which a known amount of the analytes of interest is
added before sample preparation. If sufficient sample volume is LFM % Recovery =
collected, this second portion of sample is added and processed  LFM conc × (spike vol + sample vol ) − (sample conc × sample vol ) 
in the same way as the LFM. As a minimum, include 1 dupli-  
 spike solution conc × spike vol 
cate sample or 1 LFM duplicate with each sample set (batch) or  
on a 5% basis, whichever is more frequent, and process it inde- ×100
pendently through the entire sample preparation and analysis.
Evaluate LFM duplicate results for precision and accuracy b. Relative percent difference (RPD):
(precision alone for duplicate samples). If LFM duplicate results
are out of control, then take corrective action to rectify the matrix  
 
effect, use another method, use the method of standard addition,  LFM − LFMD 
or flag the data if reported. If duplicate results are out of control,   ×100 = RPD
  LFM + LFMD  
then reprepare and re-analyze the sample and take additional cor-    
  2  

rective action, as needed. When the value of one or both duplicate
samples is ≤5 × MRL, the laboratory may use the MRL as the
control limit for percent recovery, and the duplicate results are or
not used. See method for specific acceptance criteria for LFM
duplicates or duplicate samples until the laboratory develops sta-  
 
tistically valid, laboratory-specific performance criteria. If the  D1 − D2 
  ×100 = RPD
method does not provide limits, use the calculated preliminary   D1 + D2  
   

limits from the IDC. In general, batch acceptance is not contin-   2 
gent upon LFM duplicate results.
where:
9. Verification of MDL and MRL
LFM = concentration determined for LFM,
LFMD = concentration determined for LFMD,
With each analytical batch, analyze a reagent-water sample
D1 = concentration determined for first duplicate, and
spiked at MRL and ensure that it meets MRL acceptance crite-
D2 = concentration determined for second duplicate.
ria (generally ±50%). If not, re-analyze the entire batch or flag
results for all samples in the batch. If the MRL is biased high, c. Initial calibration: See Section 1020 B.12a.
nondetect (ND) samples can be reported with flags if the method d. Calibration verification: See Section 1020 B.12b.
or regulation allows. e. Laboratory-fortified blank recovery: See Section 1020 B.12c.
If reporting to the MDL, then verify the MDL at least quarterly f. Laboratory-fortified matrix: See Section 1020 B.12e.
by analyzing a sample spiked at the same level used to determine g. Standard additions: See Section 1020 B.12g.
the MDL and ensure that the result is positive. If 2 consecu-
tive MDL-verification samples do not produce positive results, 11. Control Charts
then recalculate the MDL using the most recent set of at least 7
blanks and MRL level spikes, following the protocols outlined in See Section 1020 B.13.
3020 B.2.
Reference
10. QC Calculations
1. U.S. Environmental Protection Agency. III.H. Changes to method
The following is a compilation of equations frequently used in detection limit (MDL) procedure. In: Clean Water Act Methods
QC calculations. Update Rule for the Analysis of Effluent. 40 CFR 136; 2016.

Published Online: August 27, 2018