**3. (Bio)chemical information**

The main output of (bio)chemical measurement processes is analytical or chemical/biochemical information, which is used to describe objects and systems for a variety of purposes, but especially to (*a*) understand processes and mechanisms in multidisciplinary approaches; and (*b*) provide support for grounded, efficient decisionmaking in a great variety of scientific, technical and economic fields. "Information" is probably the most important keyword for Analytical Chemistry, which has been aptly defined as an "information discipline" [8]. As shown below, (bio)chemical information lies in between raw data and knowledge; also, it has evolved markedly over the past few centuries and eventually become highly influential on human life and the environment by virtue of the increasing importance attached to social responsibility in Analytical Chemistry.

"(Bio)chemical information" and "analytical information" are two equivalent terms in practice. In fact, the difference between chemical and biochemical analysis is irrelevant as it depends on the nature of the analyte (e.g. sodium or proteins), sample (e.g. soil or human plasma) and tools involved (e.g. an organic reagent or immobilized enzymes).

## **3.1. Contextualization**

100 Analytical Chemistry

As illustrated by Figure 6, quality in the results should go hand in hand with quality in the analytical process. In other words, capital analytical properties should rely on basic properties as their supports. It is a glaring error to deal with analytical properties in isolation as it has been usual for long. In fact, these properties are mutually related in ways that can be more consequential than the properties themselves. Their relationships are discussed in detail in Section 4. Each type of analytical problem has its own hierarchy of analytical properties, which materializes in the above-described "quality compromises".

**Figure 6.** Holistic view of analytical properties as classified into three major groups and of their

The main output of (bio)chemical measurement processes is analytical or chemical/biochemical information, which is used to describe objects and systems for a variety of purposes, but especially to (*a*) understand processes and mechanisms in multidisciplinary approaches; and (*b*) provide support for grounded, efficient decisionmaking in a great variety of scientific, technical and economic fields. "Information" is probably the most important keyword for Analytical Chemistry, which has been aptly defined as an "information discipline" [8]. As shown below, (bio)chemical information lies in between raw data and knowledge; also, it has evolved markedly over the past few

relationships with quality of the results and the analytical process. For details, see text.

**3. (Bio)chemical information** 

Information is the link between raw data and knowledge in the hierarchical sequence of Figure 7. *Primary* or *raw data* are direct informative components of objects and/or systems, whereas *information* materializes in a detailed description of facts following compilation and processing of data, and *knowledge* is the result of contextualizing and discussing information in order to understand and interpret facts with a view to making grounded, timely decisions. Einstein [9] has proposed *imagination* as an additional step for the sequence in critical situations requiring the traditional boundaries of knowledge to be broken by establishing new paradigms.

**Figure 7.** "Information" as an intermediate step between "raw data" and "knowledge", and their significance in the context of chemistry and biochemistry. For details, see text.

In a (bio)chemical context, "raw data" coincide with the primary "signals" provided by instruments (e.g. absorbance, fluorescence intensity, electrical potential readings). Also, "information" corresponds to the "results" of (bio)chemical measurement processes, which can be quantitative or qualitative. Finally, "knowledge" corresponds to "reports", which contextualize information, ensure consistency between the information required and that provided, and facilitate decision-making.

Analytical Chemistry Today and Tomorrow 103

similar structure/nature (e.g. greases, polyphenols, PAHs, PCBs) and/or exhibiting a similar operational behavior or effect (e.g. toxins, antioxidants, endocrine disruptors). More than 50% of the information required for decision-making is of this type. A large number of validated analytical methods produce this peculiar type of output. Probably, the greatest

Classification 2 in Figure 8 establishes two types of results: typical and atypical. Typical (ordinary) results can be quantitative (viz. numerical data with an associated uncertainty range) and qualitative (e.g. yes/no binary responses); the latter have gained increasing importance in recent times. There are also atypical results requiring the use unconventional metrological approaches in response to specific social or economic problems. Thus, so named "method defined parameters" (MDPs) [12] are measurands that can only by obtained by using a specific analytical method —which, in fact, is the standard— and differ if another method is applied to the same sample to determine the same analyte. Usually, MDPs are total indices expressed in a quantitative manner (e.g. 0.4 mg/L total phenols in water; 0.02 mg/L total hydrocarbons in water). In some cases, MDPs are empirical (e.g. bitterness in beer or wine). Some can be converted into yes/no binary responses (e.g. to state whether a threshold limit imposed by legislation or the client has been exceeded). Markers [13] are especially important analytes in

terms of information content (e.g. tumor markers, saliva markers to detect drug abuse).

**Figure 9.** Contradiction between the frequency of information demands and the level of quality required in a situation of growing demands for (bio)chemical information. For details, see text.

Classification 3 in Figure 8 is based on the quality level of the results required in response to the client's information needs and comprises (*a*) routine information provided by control laboratories analyzing environmental, industrial, clinical or agrifood samples, for example; and (*b*) information of a high scientific and technical level that can only be obtained by using sophisticated instrumentation in specialized laboratories usually involved in R&D activities.

problem to be solved here is to obtain appropriate metrological support.

#### **3.2. Types**

Figure 8 shows several classifications of (bio)chemical information according to complementary criteria such as the relationship between the analyte(s) and result(s), the nature of the results, the required quality level in the results in relation to the analytical problem and the intrinsic quality of the results [10].

**Figure 8.** Four complementary classifications of (bio)chemical information based on different criteria. For details, see text.

Based on classification 1 in Figure 8, results can be discriminated by analyte (one analyte– one result), which is the most frequent situation when a separation (e.g. chromatographic, electrophoretic) is involved or when the measurement process is highly selective (e.g. immunoassays). Of increasing interest in this context are "total indices" [11], which can be defined as parameters representing a group of (bio)chemical species (analytes) having a similar structure/nature (e.g. greases, polyphenols, PAHs, PCBs) and/or exhibiting a similar operational behavior or effect (e.g. toxins, antioxidants, endocrine disruptors). More than 50% of the information required for decision-making is of this type. A large number of validated analytical methods produce this peculiar type of output. Probably, the greatest problem to be solved here is to obtain appropriate metrological support.

102 Analytical Chemistry

**3.2. Types** 

For details, see text.

provided, and facilitate decision-making.

problem and the intrinsic quality of the results [10].

In a (bio)chemical context, "raw data" coincide with the primary "signals" provided by instruments (e.g. absorbance, fluorescence intensity, electrical potential readings). Also, "information" corresponds to the "results" of (bio)chemical measurement processes, which can be quantitative or qualitative. Finally, "knowledge" corresponds to "reports", which contextualize information, ensure consistency between the information required and that

Figure 8 shows several classifications of (bio)chemical information according to complementary criteria such as the relationship between the analyte(s) and result(s), the nature of the results, the required quality level in the results in relation to the analytical

**Figure 8.** Four complementary classifications of (bio)chemical information based on different criteria.

Based on classification 1 in Figure 8, results can be discriminated by analyte (one analyte– one result), which is the most frequent situation when a separation (e.g. chromatographic, electrophoretic) is involved or when the measurement process is highly selective (e.g. immunoassays). Of increasing interest in this context are "total indices" [11], which can be defined as parameters representing a group of (bio)chemical species (analytes) having a Classification 2 in Figure 8 establishes two types of results: typical and atypical. Typical (ordinary) results can be quantitative (viz. numerical data with an associated uncertainty range) and qualitative (e.g. yes/no binary responses); the latter have gained increasing importance in recent times. There are also atypical results requiring the use unconventional metrological approaches in response to specific social or economic problems. Thus, so named "method defined parameters" (MDPs) [12] are measurands that can only by obtained by using a specific analytical method —which, in fact, is the standard— and differ if another method is applied to the same sample to determine the same analyte. Usually, MDPs are total indices expressed in a quantitative manner (e.g. 0.4 mg/L total phenols in water; 0.02 mg/L total hydrocarbons in water). In some cases, MDPs are empirical (e.g. bitterness in beer or wine). Some can be converted into yes/no binary responses (e.g. to state whether a threshold limit imposed by legislation or the client has been exceeded). Markers [13] are especially important analytes in terms of information content (e.g. tumor markers, saliva markers to detect drug abuse).

**Figure 9.** Contradiction between the frequency of information demands and the level of quality required in a situation of growing demands for (bio)chemical information. For details, see text.

Classification 3 in Figure 8 is based on the quality level of the results required in response to the client's information needs and comprises (*a*) routine information provided by control laboratories analyzing environmental, industrial, clinical or agrifood samples, for example; and (*b*) information of a high scientific and technical level that can only be obtained by using sophisticated instrumentation in specialized laboratories usually involved in R&D activities.

The frantic recent changes in social and economic activities have promoted an impressive expansion of (bio)chemical information about objects and systems. As can be seen in Figure 9, the quality of (bio)chemical information increases from routine laboratories to specialized laboratories, whereas the frequency of information demands decreases in the same direction. A compromise must often be made between these two contradictory notions. The panoramic view of Figure 9 is essential to perceive all connotations of analytical information. Classification 4 in Figure 8 is based on the intrinsic quality of the results and is examined in detail in Section 4 of this chapter.

Analytical Chemistry Today and Tomorrow 105

procedure for determining antioxidants in foodstuffs, which involves time-consuming sample treatment and the use of sophisticated instruments (e.g. a liquid chromatograph coupled to a mass spectrometer). This trend is also related to simplification and is rendering the classical

*4. Increasing importance of productivity-related properties.* The holistic approach to analytical properties of Figure 6, which considers hierarchical, complementary and contradictory relationships between them, and systematically using information needs as the third basic analytical reference (Figure 4), provide solid support for the increasingly popular productivityrelated analytical properties ( expeditiousness, cost-effectiveness and personnel-related factors). These properties are in contradiction with capital and basic analytical properties. Thus, achieving a high accuracy is not always the primary target and, in some cases, productivity-related properties are more important than capital properties. Such is the case with so named "point of care testing" approaches [15], the best known among which is that behind the glucose meter used to monitor the glucose level in blood at home. Glucose meter

paradigm of Analytical Chemistry (viz. maximizing selectivity) obsolete.

readings are inaccurate but rapid and convenient enough to control diabetes.

**Figure 10.** Major trends in the characteristics of (bio)chemical information provided by routine

alternatives emphasizing positive aspects rather negative connotations.

*5. Use of positive approaches to produce reports from results*. Analysts tend to emphasize negative aspects in delivering results and reports. A dramatic impulse of their "marketing abilities" to communicate with clients is therefore needed. One case in point is the word "uncertainty", inherited from Metrology in Physics and introduced in Metrology in Chemistry during the last few decades. This word can lead to wrong interpretations in chemistry nonmajors (e.g. politicians, economists, managers, judges) and raise global doubts about results. Simply replacing "uncertainty" with "confidence interval", which has the same scientific and technical meaning, can facilitate interpretation and acceptance of the results [16]. One other typical case is the use of "false positives" and "false negatives" to describe errors in binary responses. There is an obvious need to revise the terms related with (bio)chemical information and find

laboratories. For details, see text.

#### **3.3. Evolution**

The routine information provided by control laboratories has evolved dramatically in the last decades. Figure 10 summarizes the most salient general trends in this context, which are commented on briefly below.

*1. Simplification.* Instead of delivering large amounts of high-quality information (a classical paradigm in Analytical Chemistry), there is a growing trend to delivering the information strictly required to make grounded decisions while avoiding time-consuming efforts to obtain oversized information that is useless in practice. Specially relevant here is the third basic standard supporting Analytical Chemistry (see Figure 4). The situation is quite common in routine laboratories but should be minimized or avoided altogether. Such is the case, for example, with the determination of hydrocarbons in tap water, the legal threshold limit for which is 0.1 ng/mL total hydrocarbons. Using a classical method involving several steps (e.g. filtration, cleanup, solvent changeover) and sophisticated equipment (e.g. a gas chromatograph and mass spectrometer) allows a long list of aliphatic and aromatic hydrocarbons with their concentrations —usually at the ppt or even lower level— to be produced which is utterly unnecessary to make grounded decisions, especially when a simplified method (e.g. one involving extraction into Cl4C and FTIR measurement of the extract) can be used instead to obtain a total index totally fit for purpose.

*2. Binary responses.* Qualitative Analysis has been revitalized [14] by the increasing demand for this type of information; in fact, clients are now more interested in yes/no binary responses than in numerical data requiring discussion and interpretation. This trend is related to the previous one because obtaining a simple response usually entails using a simple testing method. The greatest challenge here is to ensure reliability in the absence of firm metrological support. In any case, false negatives should be avoided since they lead to premature termination of tests; by contrast, false positives can always and are commonly confirmed by using more sophisticated quantitative methodologies (see Section 5.4 and Figure 14).

*3. Total indices.* Based on classification 1 in Figure 8, a result can be a total index [11] representing a group of analytes having a common structure or behavior. This type of information is rather different from classical information, which is typically quantitative and discriminated by analyte. For example, the total antioxidant activity of a food can be easily determined with a simple, fast method using a commercially available dedicated analyzer. This avoids the usual procedure for determining antioxidants in foodstuffs, which involves time-consuming sample treatment and the use of sophisticated instruments (e.g. a liquid chromatograph coupled to a mass spectrometer). This trend is also related to simplification and is rendering the classical paradigm of Analytical Chemistry (viz. maximizing selectivity) obsolete.

104 Analytical Chemistry

**3.3. Evolution** 

commented on briefly below.

examined in detail in Section 4 of this chapter.

The frantic recent changes in social and economic activities have promoted an impressive expansion of (bio)chemical information about objects and systems. As can be seen in Figure 9, the quality of (bio)chemical information increases from routine laboratories to specialized laboratories, whereas the frequency of information demands decreases in the same direction. A compromise must often be made between these two contradictory notions. The panoramic view of Figure 9 is essential to perceive all connotations of analytical information. Classification 4 in Figure 8 is based on the intrinsic quality of the results and is

The routine information provided by control laboratories has evolved dramatically in the last decades. Figure 10 summarizes the most salient general trends in this context, which are

*1. Simplification.* Instead of delivering large amounts of high-quality information (a classical paradigm in Analytical Chemistry), there is a growing trend to delivering the information strictly required to make grounded decisions while avoiding time-consuming efforts to obtain oversized information that is useless in practice. Specially relevant here is the third basic standard supporting Analytical Chemistry (see Figure 4). The situation is quite common in routine laboratories but should be minimized or avoided altogether. Such is the case, for example, with the determination of hydrocarbons in tap water, the legal threshold limit for which is 0.1 ng/mL total hydrocarbons. Using a classical method involving several steps (e.g. filtration, cleanup, solvent changeover) and sophisticated equipment (e.g. a gas chromatograph and mass spectrometer) allows a long list of aliphatic and aromatic hydrocarbons with their concentrations —usually at the ppt or even lower level— to be produced which is utterly unnecessary to make grounded decisions, especially when a simplified method (e.g. one involving extraction into Cl4C and FTIR measurement of the

*2. Binary responses.* Qualitative Analysis has been revitalized [14] by the increasing demand for this type of information; in fact, clients are now more interested in yes/no binary responses than in numerical data requiring discussion and interpretation. This trend is related to the previous one because obtaining a simple response usually entails using a simple testing method. The greatest challenge here is to ensure reliability in the absence of firm metrological support. In any case, false negatives should be avoided since they lead to premature termination of tests; by contrast, false positives can always and are commonly confirmed by

*3. Total indices.* Based on classification 1 in Figure 8, a result can be a total index [11] representing a group of analytes having a common structure or behavior. This type of information is rather different from classical information, which is typically quantitative and discriminated by analyte. For example, the total antioxidant activity of a food can be easily determined with a simple, fast method using a commercially available dedicated analyzer. This avoids the usual

using more sophisticated quantitative methodologies (see Section 5.4 and Figure 14).

extract) can be used instead to obtain a total index totally fit for purpose.

*4. Increasing importance of productivity-related properties.* The holistic approach to analytical properties of Figure 6, which considers hierarchical, complementary and contradictory relationships between them, and systematically using information needs as the third basic analytical reference (Figure 4), provide solid support for the increasingly popular productivityrelated analytical properties ( expeditiousness, cost-effectiveness and personnel-related factors). These properties are in contradiction with capital and basic analytical properties. Thus, achieving a high accuracy is not always the primary target and, in some cases, productivity-related properties are more important than capital properties. Such is the case with so named "point of care testing" approaches [15], the best known among which is that behind the glucose meter used to monitor the glucose level in blood at home. Glucose meter readings are inaccurate but rapid and convenient enough to control diabetes.

**Figure 10.** Major trends in the characteristics of (bio)chemical information provided by routine laboratories. For details, see text.

*5. Use of positive approaches to produce reports from results*. Analysts tend to emphasize negative aspects in delivering results and reports. A dramatic impulse of their "marketing abilities" to communicate with clients is therefore needed. One case in point is the word "uncertainty", inherited from Metrology in Physics and introduced in Metrology in Chemistry during the last few decades. This word can lead to wrong interpretations in chemistry nonmajors (e.g. politicians, economists, managers, judges) and raise global doubts about results. Simply replacing "uncertainty" with "confidence interval", which has the same scientific and technical meaning, can facilitate interpretation and acceptance of the results [16]. One other typical case is the use of "false positives" and "false negatives" to describe errors in binary responses. There is an obvious need to revise the terms related with (bio)chemical information and find alternatives emphasizing positive aspects rather negative connotations.
