Protocol and standard operating procedures for common use in a worldwide multicenter study on reference values

2.2.1 Inclusion criteria

1. The participants should be feeling subjectively well.

2. The participants should be older than 18 years of age. At least 80% of subjects should be 18–65 years, with equal gender mix and age distributions, except for individuals over 65 years of age.

3. The participants ideally should not be taking any medication. Any subject taking medications or vitamin supplements should have them recorded (name, dose, and frequency) so that secondary exclusion after measurement can be done as required. The following medications are permitted but should be recorded: contraceptive pills or estrogens and thyroxine, if the subject is well replaced (i.e., TSH is lower than the upper reference limit), are permitted, but they should be recorded.

2.2.2 Exclusion criteria

The participant may not enter the study if any of the following applies:

1. Known diabetes on oral therapy or insulin (diet alone is acceptable).

2. History of chronic liver or kidney disease.

3. Blood test results that clearly point to a severe disease.

4. History of being a hospital in-patient or otherwise seriously ill during the previous 4 weeks.

5. Blood donation in the previous 3 months.

6. Known carrier state for HBV, HCV, or HIV.

7. Female participants who are pregnant, breastfeeding, or within 1 year after childbirth.

8. Any other significant disease or disorder that, in the opinion of the investigator, may either put the participants at risk because of participation in the study or may influence the results of the study.

9. Participation in another research study involving an investigational product in the past 12 weeks.

2.2.3 Ethnicity

Information on ethnicity can be collected from the volunteers, if relevant, using the classification shown in Appendix A.

2.2.4 Sample size

The practically attainable target sample size from each country is set at a minimum of 500 (male and female: 250×2) or more, which is greater than twice the minimum number recommended by C28-A3 [120×2 (men and women)], so that country-specific RIs can be obtained in a more reproducible manner. This number is adequate for making between-country comparisons of test results with a power of detecting a difference of two means equivalent to 0.25 times SD comprising the RI (SD_RI), which corresponds to a bias of 0.25 times between-individual variation, allowing errors of α<0.05 and β<0.2 in the statistical hypothesis testing done separately for each gender. If there is an interest in exploring regional within-country variations, it is recommended to obtain at least 120 (men and women, 60×2) samples from each local area to acquire adequate power to test for a difference of two means equivalent to 0.5×SD_RI by the above specification.

2.7.1 Requirements for the central laboratories

One or two laboratories in each country should be chosen as central laboratories, which will provide collective measurements. Requirements for a central laboratory are listed in SOP 1.

2.7.2 Quality control

Each central laboratory should prepare commutable specimens for QC monitoring as described in SOP 2.4. The appropriate batch size of the assay should be decided according to local requirements.

2.7.3 Standardization of the assay

For the standardized analytes, internationally qualified standards or CRMs should be measured to ensure traceability of test results.

2.7.4 Cross-comparison of values

A panel of sera prepared from healthy individuals should be used for cross comparison of values among the central labs based on linear regression analysis (reduced major major-axis regression). The magnitude of error in converting any RI from one assay platform to another can be estimated either by CV of slope, CV(b), or by SDR [the ratio of the standard error of the converted lower and upper limits, SE(LL) and SE(UL), relative to the SD comprising the converted RI, equal to SDR_LL and SDR_UL]. The optimal limit for CV(b) is 5.5% and that for SDR is 0.125 [18].

2.9.1 Declaration of Helsinki

The investigator should ensure that this study is conducted in full conformity with the current revision of the Declaration of Helsinki.

2.9.2 Ethics committee approval

The protocol, informed consent form, participant information sheet (questionnaire), and any proposed advertising material should be submitted to the ethics committee of each collaborator’s institution for written approval. The investigator should also submit and obtain approval for all substantial amendments to the original approved documents.

2.9.3 Informed consent

See SOP 1.

2.9.4 Participant confidentiality

The investigating team should ensure that the participants’ anonymity is maintained. The participants must be identified only by initials and an ID number on any electronic database sent outside the collecting institution. All documents should be stored securely and only accessible to the investigating team. The study should comply with the Data Protection Act, which requires data to be made anonymous as soon as is practical.

2.10.1 Validation of data

Those volunteers with overtly abnormal results should be secondarily excluded (e.g., hepatic or renal disease).

2.10.2 Analyses of source of variation of test results

Multiple regression analysis (MRA) should be performed analyte by analyte to identify factors related to variation of the test results [22]. The factors to be included in MRA are gender, age, BMI, smoking status, level of alcohol consumption, frequency and strength of physical exercise, dietary status, and, if available, ethnic group, the ABO blood group.

In the analysis, after combining all data across countries, dummy variables for the country or ethnic groups should be introduced to reveal regionality/ethnicity dependency. The dummy variables are also important to conduct regionality/ethnicity-adjusted analysis of sources of variation for the test results.

2.10.3 Partitioning criteria

The components of SD, i.e., between-country SD (SDcntr), between-sex SD (SDsex), between-age SD (SDage), and net between-individual SD (SDind) with removal of the component of within-individual variation, will be computed by 3-level nested ANOVA. The presence of significant regionality in the test results can be determined by taking a ratio (SDR) of SDcntr over SDindiv; an SDR >0.3 signifies the evidence of significant regionality in the test results [22]. However, its validity must be confirmed after adjusting for other possible confounding factors using MRA or similar procedures. SDR can be also computed for between-sex SD and between-age SD. The cut-off value of SDR=0.3 corresponds approximately to the midpoint of desirable and minimal proportion of analytical bias to the between-individual SD of 0.25 and 0.375 times the between-individual SD [18]. However, it is only a guide to consider whether to partition reference values according to that factor. It can be modified according to local need and policy in implementing the RIs. It is also necessary to note that, in dealing with regionality, if there is an isolated difference between only one or two countries with no differences among all other countries, ANOVA is insensitive for detecting regional differences. Therefore, it is recommended to apply an ad hoc analysis using the Harris-Boyd method be performed to evaluate the magnitude of the difference between any two groups.

If there are no apparent regional differences in test results, globally applicable common RIs may be derived by specifying conditions used for the derivation. Otherwise, reference values should be partitioned to derive region-, ethnicity-, or country-specific RIs.

2.10.4 Derivation of RI

The parametric method will be used after normalizing data based on a modified Box-Cox power transformation method [22] for derivation of RIs. Secondary exclusion of inappropriate subjects should be made to remove those with abnormal results attributable to highly prevalent conditions such as the metabolic syndrome, alcohol-related hepatic dysfunction, or diabetes mellitus. A multivariate iterative method called latent abnormal value exclusion (LAVE) [22] can be applied at the time for computing RIs as a method for tertiary exclusion.

When an analyte is not considered well-standardized, the RI derived by the centralized measurement may be converted to each laboratory’s value using the linear regression parameters derived from the cross-check test results if the scatter around the regression line is within the allowable limit [CV(b)≤11%] [18].

3.1.1 Invitation to the study

It is advised to advertise the study by posters displayed in the wards and out-patient areas and by electronic invitations sent to staff members within the participating health-care institution. It is also advisable to hold meetings to explain the clinical and scientific implications of the study and possible benefits for the laboratory and volunteers to obtain cooperation.

Give each volunteer the following:

• An invitation to the study.

• An explanation of the study (including/exclusion criteria).

• A consent form (written in the local language, in accordance with the guidelines of the local ethics committee).

• Procedures for participation.

3.1.2 Informed consent

During the introductory period, written and verbal information should be presented to the participants detailing the exact nature of the study and any risks involved in taking part. It should be clearly stated that the participant is free to withdraw from the study at any time for any reason with no obligation to give the reason for withdrawal. The participant must be allowed as much time as desired to consider the information and allowed an opportunity to question the investigator, their physicians or other independent parties to help decide whether they will participate in the study. Written informed consent should then be obtained, including the personal signature and signature date for both the participant and the person who presented and obtained the informed consent. The original signed form should be retained by the local representative.

3.2.3 Tabulation of volunteers by age and gender

In a practical flow of recruitment, one who agrees to participate in the study by reading the invitation is expected to contact the local representative. A balanced distribution of gender and age should be ensured by tabulating volunteers as shown below (Table 1). All participants must be older that 18 years. The main target range of ages is 18 to 65 years, for which an even distribution of age and gender is of the utmost importance to have comparability of test results across regions.

Individuals older that 65 years are also sought, as long as they match the inclusion criteria. Approximately 20% of the total number should be in this age group. The primary objective of including this age range is to evaluate age-related changes in test results, and thus, there is no need to balance the gender distribution of this group.

3.1.4 Appointment and preparation for blood sampling

It is advisable to make the appointment for the drawing of blood after volunteers for most of the subgroups in the above table have been recruited. Make appointments between 7:00 and 10:00 AM with 30-min intervals (i.e., 7:00, 7:30,…, 9:30, 10:00).

3.1.5 Procedures on the day of blood collection

3.1.6 Procedures just before blood collection

3.1.7 Procedures for drawing blood

• Apply the tourniquet 7–10 cm above the venipuncture site. The pressure should be set below the diastolic blood pressure for the smooth pooling of blood in the periphery. Never leave the tourniquet on for longer than 1 min.

• Do not clench the fist while drawing blood; this causes false elevation of serum potassium.

• Draw the volume of blood required.

• Invert each tube 180° (upside down) at least five times (in the presence of a clot activator in the sampling tubes).

• If the blood draw is interrupted before a tube is completely filled, the remaining vacuum should be removed by filling air to avoid vacuum-induced hemolysis [].

3.1.8 Preparation of the serum and its aliquots

• Do not place the blood-filled tubes in direct sunlight or in low temperatures.

• At 30–60 min after sampling, centrifuge the specimen at 1200 g for 10 min at room temperature.

• Be sure the ID label on the blood collection tubes matches the ID labels on the sample containers (2 mL each). Transfer the serum into each of the containers.

3.1.9 Storage and shipment of the specimens

The containers with the serum aliquots (1–2 mL) should be well sealed and stored at −80°C within 2 h. All aliquots that are not being retained for local use will be shipped on dry ice to the central laboratory for collective measurement.

3.1.10 Thawing of the specimens and preparation of the samples before measurement

On the day of analysis, the samples must be thawed by letting them stand at room temperature for at least 1 h, avoiding direct sunlight because of its effect on bilirubin concentrations. Homogenization is then achieved by inverting the samples 10 times. Analysis must be performed within 4 h from the start of thawing.

3.1.11 Measurements for the cross-check survey

3.2.1 Sample size

The target sample size from each country should be at least 500 so that country-specific RIs can be derived in a reproducible manner. This number is sufficient for between-country comparison of test results. If there is an additional interest in exploring regional variations within a country, it is recommended that at least 120 (men and women, 60×2) samples from each local area be assayed to acquire sufficient power to test for between area differences.

3.2.2 Target analytes

The standardized analytes to be measured in common and to be compared worldwide should be decided and listed. Additional analytes to be measured are determined in each country based on clinical need and research interest. At the time of combined analyses of reference values, comparison will be made between results of all analytes that are available for comparison.

3.2.3 Central laboratory

One or two laboratories in each country should be chosen as a central laboratory that will provide collective measurements. The requirements for the central laboratory should be specified so that it (1) implements reliable measures for both short- and long-term quality control, (2) ensures traceability of test results for the standardized analytes based on CRMs and value-assigned sera for enzymes, and (3) participates in the alignment of test results across central laboratories, using the reference panel of sera.

3.2.4 Quality control

In addition to standard quality control materials supplied by manufacturers, each laboratory will prepare multiple commutable specimens (mini-panel) for QC monitoring. The desirable limits of bias should be specified beforehand. The desirable limits for between- and within-day CV are set as 1/2 of CV_I (within-individual CV listed in the Westgard website:www.westgard.com/biodatabase1.htm).

3.2.5 Standardization of assays

To ensure essential traceability of test results, internationally qualified standards or CRMs should be measured for the standardized analytes.

3.2.6 Cross-comparison of values

A panel of sera composed of 40 sera prepared from healthy individuals should be divided into at least four aliquots each, to be tested on different days in order to determine between-day variation of the test results. The between-day variation of test results should also be monitored by the QC specimens and recorded.

3.3.1 Data analysis

3.3.2 Report of test results

Only after all the specimens have been analyzed and the RIs have been derived should the test results be returned to the volunteers. A sheet with a given participant’s printed test results should be sealed in an envelope with the ID label on the outside. The local representative should be asked to give this to the corresponding individual by referring to the name from the informed consent form with the same ID number. Each individual’s test results and the newly-derived, country-specific RIs will be reported to the volunteers after all the measurements in each country have been completed, together with an explanatory sheet for the interpretation of the test results.