1.1Introducing MAAS: Why The Medical Interview?

Kraan, H. F., & Crijnen, A. A. M. (1987). Introduction to studies of the instrumental utility of the Maastricht History-taking and Advice Checklist (MAAS). In H. F. Kraan & A. A. M. Crijnen (Eds.), The Maastricht History-taking and Advice Checklist: studies of instrumental utility (pp. 13–27). Lundbeck, Amsterdam.

Reasons for Encounter

In the early 1970s, Kraan attended the courses on medical interviewing organized by the Department of Medical Psychology, University of Amsterdam.

Crijnen was engaged in similar types of study groups organized and supervised by Prof. Dr. Paul Sporken at the Maastricht Medical School during the mid-1970s.

MAAS-MI was constructed due to the enthusiasm of inexperience

Our interest in medical interviewing skills led us, over a period of four years, to construct and analyse a method for measuring them. Early participation in these groups convinced us of the importance of the medical interview for both patient and physician. Moreover, it triggered the desire to enhance the quality of our own interviewing skills and to gain insight into the intricacies of the consultation process. 

The present study has given us the opportunity to scrutinize this extensive domain and thus became acquainted with a gamut of opinions, theories and studies about medical interviewing skills. The development of the research design and experimental settings allowed us, furthermore, to study scientific methods and to apply them, through a creative process, to a feasible experiment with a simulated consultation hour.

The construction of the Maastricht History-taking and Advice Checklist, as the MAAS-Medical Interview was originally called, and the subsequent studies in reliability and validity came about due to the enthusiasm of inexperience: neither of us realized beforehand the dangers, pitfalls and demands in terms of time, energy and knowledge that would need to be met.

The Medical Interview & Related Skills: Definitions and Delimitations

The medical interview has a function in medical consultation. In this chapter, we first define the medical interview from this functional perspective. We then go deeper into the relationship between medical interview and medical consultation. Medical interviewing is also to be considered as part of the physician’s competency. We therefore provide a second definition of medical interview, from a competency perspective. 

The competency domain in MAAS-MI is restricted to interviewing skills

A Functional Perspective

The medical interview has been defined by Schouten et al. (1981) from a functional perspective:

• Collection of information for diagnostics and further patient management;
• Conveyance of information to the patient about diagnosis, aetiology and prognosis;
• Establishment and maintenance of a physician-patient relationship of trust and acceptance allowing the achievement of the two previous functions.

Medical interview and medical consultation are difficult to distinguish (e.g., Pendleton et al., 1983). However, medical consultation is a term with a broader meaning than medical interview. It encompasses the patient’s presentation of a (medical) problem and the physician’s response in terms of diagnostics as well as preventive and curative advice. Many variables, from both physician and patient, determine the medical consultation process. Figure 1 which presents an overview of these variables, is adapted from Pendleton’s (1983) attempt to design a model of medical consultation.

Physician’s Contribution

The physician’s contribution to the consultation is primarily determined by his educational background which determines his medical competency. Besides knowledge base, Fabb and Marshall (1983) distinguish interviewing, manual/perceptual skills, and problem-solving skills in medical competency. In addition, the physician’s social norms and expectations as to his professional role (Parsons, 1951) and his attitudes towards patients and care-giving (Ajzen et al., 1980; Goldberg et al., 1980; Grol et al., 1985; Verhaak, 1986), determine his contribution to the consultation process. 

Patient’s Contribution

The patient’s contribution to the consultation consists, of course, of his account of his complaints and symptoms, but also of his concerns and ideas about health, illness and treatment. This presentation is strongly influenced by health beliefs (Becker and Maiman, 1975; Rosenstock, 1974), attributions (Rodin, 1978; Stoeckle et al., 1980) and attitudes/social norms (Ajzen et al., 1980). Like the physician’s contribution, these variables are molded by personality traits and socio-cultural background variables. Moreover, the decision making to enter into the patient-role (Parsons, 1951; Philipsen, 1969; Zola, 1973) or, particularly in the case of mental health care, into the psycho-role (Siegler et al., 1976), is an important variable. The majority of these patient variables, also topics of the medical interview, are discussed in the next chapter. 

This complex model makes further delineation necessary. We restrict ourselves therefore to the physician’s competency aspect of the interviewing skills. We define the medical interview from a competency perspective as the dimension in consultation that is determined by the physician’s interviewing skills. Although this definition has the flavour of a tautology, it serves as a further restriction of our field of study to the medical interviewing skills and, in particular, to their measurement.

The Medical Interview & Medical Interviewing Skills

Several arguments are widely recognized as underscoring the importance of teaching and research into the medical interview and related skills:

• The medical interview is the medium of every consultation. Physicians spend about 60% of their working time interviewing (Lipkin, et al., 1984);
• As a function of the medical interview, accurate data collection is a prerequisite for accurate diagnosis (Goldberg et al., 1982; Elstein et al., 1978; Cox et al., 1981);
• Patient’s satisfaction with a medical consultation is dependent on the feelings of the patient being understood and taken seriously (Pendleton, 1983);
• Provision of information about the nature of the illness, its causes, its prognosis and about its further management leads to patient compliance and therefore enhances the effectiveness of health care (DiMatteo et al., 1982);
• Medical interviewing skills contribute to establishing and maintaining the physician-patient relationship (Pendleton et al., 1984);
• Medical interviewing skills are indispensable for the detection of mental health problems (Goldberg et al., 1980), prevention of somatic fixation (Grol et al., 1981) and basic treatment of psychosocial problems (Ivey, 1983).

These arguments stress the importance of medical interviewing skills in health care, medical education and research. The need, therefore, for measurement and evaluation methods is evident.

Measuring Interviewing Skills

Measurement is not an aim in itself. Objectives, subject matter and criteria for measurement, as well as psychometric requirements, should be clearly settled before an appropriate measurement method can be selected or constructed. 

MAAS-MI should be used as an aid to education and evaluation

Core question in this study: Are MAAS-MI measures real reflections of physician’s interview ability?

Objectives of measurement

Our measurement method of interviewing skills should be used as an evaluation instrument, and as an aid to education in interviewing skills. These objectives have two strong implications for the selection of interviewing skills to be included in the method.

• First, it implies that these skills can be acquired through training and education (Van Dalen et al., 1981) in the sense that the interviewer:
  • Knows when pertinent skills should be applied;
  • Knows how to apply these skills;
  • Can actually apply these skills;
  • Knows which effects could result from these skills.

These teachability conditions narrow down our definition of medical interviewing skills somewhat, because it excludes behavior during medical interviews that cannot be taught (some non-verbal behavior, idiosyncratic behavior related to the person of the interviewer). Nevertheless, this behavior may be important in relation to the outcome of the interview.

• Second, the implication is that the focus of measurement is on the physician’s behavior during the interview.

Subject matter of measurement

The physician’s interviewing skills are the subject of measurement. To be more specific we delineate this field even further.

• First, we are interested in medical interviews in primary care because of the superordinate objectives of the medical school where this research project has been carried out.
• Second, we selected initial interviews where the patient comes up with a problem for consultation that they consider as new. ‘Initial’ interviews are more uniform in their structure than ‘follow-up’ interviews and, therefore, are easier to fit into an educational model for interviewing skills (see next item). Moreover, initial interviews are of paramount importance in primary care where a patient’s career and medical decision-making start.

Criteria for measurement

Criteria setting is indispensable for the step from measurement to evaluation. To make comparable judgments, standards should be available. This is discussed in the chapter on Constructing MAAS-MI. For educational purposes, we constructed an educational model stating the interviewing skills in concrete terms to be mastered. This model, which is an indispensable prerequisite for educational programs (Carroll et al., 1980), is given in Medical Interview & Related Skills.

Psychometric requirements

The psychometric requirements to transform medical interviewing skills as concept-defined into variables, are described by De Groot (1961). He mentions four criteria making up the instrumental utility of these variables:

1. Reliability  Is the measurement consistent under varying conditions of subjects, interview situations, cases/problems and observers?
2. Scalability  Can the operationalizations of interviewing skills, in short, the items, be fitted in a measurement scale?
3. Validity  Are the measurement scores real reflections of the physician’s interviewing ability?
4. Practicability  Is the measurement practical within the objectives of evaluation and education? 

In this thesis, and now this site, these conditions of measurement are applied in the construction of a measurement method for interviewing skills. 

MAAS-Medical Interview

The afore-mentioned considerations have resulted in the construction of two observation instruments intended to measure the interviewing skills of physicians during initial consultations in general practice / medicine (MAAS-MI General) and in primary mental health care (MAAS-MI Mental Health). 

MAAS-MI General consists of 3 content scales:

• Exploring Reasons for Encounter
• History-taking
• Presenting Solutions

and 3 process scales:

• Structuring
• Interpersonal Skills
• Communicative Skills

In MAAS-MI Mental Health 2 content scales are added:

• Psychiatric Examination
• Socio-emotional Exploration

The items are divided into 6 (or 8 in Mental Health) scales. The first 3 (or 5 in Mental Health) scales pertain to skills used in the three characteristic phases of initial medical consultations. The remaining three scales measure skills pertaining to process aspects of interviewing:

• Scale I Exploring Reasons for Encounter measures the ability to clarify the patient’s complaint(s) and to explore the motives in the pre-patient phase leading to the visit to the physician. This is the patient-centered part of the medical interview.
• Scale II consists of History-taking skills. Through this type of questioning, the physician is able to generate hypotheses and to explain the patient’s complaint(s) in medical terms. 
• In the MAAS-Mental Health, this scale is divided into three parts. In addition to the scale History-taking in sensu strictu, two further scales have been added:
  • Psychiatric Examination is a scale that measures the skills for exploring possible psychiatric symptoms in order to make a psychiatric assessment of the patient;
  • Socio-emotional Exploration is the second scale added which measures the extent to which the physician explores aetiological conditions or consequences of mental health problems. 
• Scale III Presenting Solutions deals with the interviewing skills involved when the physician is presenting solutions . It pertains to the exchange of information about the medical problem (cause, prognosis) and to the negotiation between physician and patient about the problem’s definition and solutions to the problem (advice, referral, etc.). 
• Scale A Structuring judges the way physicians structure the interview (introduction, balance of patient- and physician-centered styles of interviewing, sequence of the different phases, closing). 
• Scale B Interpersonal Skills consists of skills which establish rapport with the patient. Moreover, the quality of the physician’s response to emotion is assessed. 
• Scale C Communicative Skills pertains to skills which aim to promote the exchange of information between physician and patient. 

Maas-MI General and MAAS-MI Mental Health can be found in Tools.

Research Questions

Research questions should address: reliability, scalability, validity, including practicality

Reliability

A method to measure interviewing skills is regarded as reliable when the method shows consistency and stability over varying conditions of measurement. The method has to prove its constancy when used by different observers, when used in different measurement situations, and when used with different patients. This latter facet, also called inter-case reliability (Swanson, 1981), is important because patients differ in the nature of their medical problems and in the way they present their problems to the physician. 

A reliable method shows consistency and stability over varying conditions of measurement

We study the reliability of measurement by investigation of the sources of unreliability, such as differences in observers, in the nature of cases and their presentation by patients and in the method itself. This is carried out by an analysis of variance, called generalizability analysis (Mitchell, 1979) in Reliability MAAS-MI General and MAAS-MI Mental Health. 

Scalability

Weaknesses in our method of measurement are detected by investigating to what extent its items fit in scales. This research of scalability is carried out with the so-called Rasch-analyses (Wright et al., 1979) in Scalability MAAS-MI General and MAAS-MI Mental Health. 

Scaling models define the relationship between a set of items and the underlying dimension, e.g. interviewing skills

Validity

A method to measure interviewing is valid when it really measures what it is intended to measure.

A valid method measures what it really intends to measure

 Validity studies are carried out in accordance with three questions:

Content Validity

Is the content of the method representative for the measurement of all the interviewing skills included in our educational model of initial interviewing skills? This model is presented in Interview & Related Skills. A study of the representativeness of the item domain or, in other words, its content validity, is carried out within the MAAS-MI Mental Health.

Convergent and Divergent Validity

Validity can also be studied by comparing the measurement properties of the MAAS with those of comparable methods. When, according to predefined hypotheses, the MAAS shares measurement properties, then it supports the so-called convergent validity. When the MAAS measures, also according to part of our hypotheses, different properties from the compared methods, then this is a sign of the so-called divergent validity. These aspects are investigated in a so-called multitrait-multimethod matrix (Campbell et al., 1959), where the MAAS is compared with 3 related measurement methods of interviewing skills in MAAS-MI General and MAAS-MI Mental Health.

Construct Validity

Validity of a method is also established when, in measurement, theoretical constructs are confirmed.

In our study, two questions about such theoretical constructs are answered:

• Is the competency aspect interviewing skills as measured with the MAAS really different from dimensions such as medical knowledge, medical problem-solving or attitudes towards care- giving? In this study, conducted using only the MAAS-MI General Practice, the MAAS-MI is compared with other methods which measure different aspects of medical competency in The Medical Interview & Clinical Competency;
• Is it possible, with the MAAS, to measure dimensions of interviewing skills that cause hypothesized changes in the patient’s interview behavior and his satisfaction with the interview? These changes are measured by means of the so-called Patient Satisfaction with Communication. They concern variables such as the patient’s feeling of being facilitated to state his problems in his own words; the feeling of being disrupted in this; increased informedness and insight of the patient into his own medical condition; feelings of being too strongly directed towards a solution of the physician’s choice and the patient’s intention to comply with the proposed preventive or curative advices.

These research questions are answered for the MAAS-MI General as well as for the MAAS-MI Mental Health in Construct Validity MAAS-MI Mental Health.

But our MAAS-MI measures of interviewing skills should also be handy during study and daily practice

Research Settings and Subjects

Throughout the entire research project, reliability and validity studies are performed with residents in general practice and medical students interviewing actual and simulated patients. Within the more restricted scope of this thesis, only studies with residents in general practice interviewing simulated patients are included. The studies with the Patient Satisfaction with Communication Checklist have been carried out partly with actual patients. 

Simulated patients can be defined as individuals trained to portray the history, physical findings and emotions of an actual patient (Norman, 1985). In the chapter on Instrumental Utility, a discussion of the validity of simulated patients for these studies is included.

Survey of Chapters

We summarize below the content of the chapters as originally published in our thesis. It is important to delineate which reliability and validity studies are performed with what kind of subjects/interviewers. We further indicate the primary author responsible for the pertinent chapter. 

Chapter 2: The Medical Interview & Related Skills describes the theoretical underpinning of our educational model for initial medical interviews. From this model, the physician’s interviewing skills are derived. 

Chapter 3: Outcomes for Patient & Physician is a review of the literature concerning the effects of the medical interview on the patient in terms of outcome variables (the patient’s satisfaction, recall, insight into conveyed information, intention to comply, etc.). 

Chapter 4: Constructing MAAS-Medical Interview discusses the construction of the MAAS and its variants. The selection of the item domain from the theory is discussed. The MAAS is compared with more than 20 other methods of measuring interviewing skills as described in the literature. The decision to construct the MAAS as a new observation instrument is discussed. 

Chapter 5: Instrumental Utility Assessed describes the methodology of scalability, reliability and validity research as used in this thesis. It expands further on the validity of simulated patients as used in this thesis. 

Chapter 6: Patient Satisfaction with Communication is an article describing the construction and scaling as well as the reliability measures of the Patient Satisfaction with Communication Checklist. It is a variant of the MAAS included for use by patients to judge the physician’s interviewing skills and their effects. The PSCC is pre-tested in about 240 medical interviews. 

Chapter 7: Scalability & Reliability MAAS-MI General describes the results of reliability and scalability studies of the MAAS-General in a sample of 40 residents in general practice, each interviewing two simulated patients, one presenting pre-cardial pain and the other inception of diabetes mellitus. 

Chapter 8: Convergent & Divergent Validity MAAS-MI General studies convergent and divergent validity by comparing the MAAS-General with two self-evaluation methods and experts’ judgments of interviewing skills. The research situation as described in the previous chapter is again used. 

Chapter 9: The Medical Interview & Clinical Competency deals with a construct validity study correlating MAAS-MI General scores with measurement of medical knowledge, attitude ratings, global ratings of interpersonal and communicative skills and measurements of medical problem-solving. 

Chapter 10: Scalability & Reliability MAAS-MI Mental Health provides the reliability and scalability studies of the MAAS-MI Mental Health in a sample of 40 residents in general practice, each interviewing two simulated patients, one presenting major depression and the other panic disorder. 

Chapter 11: Content Validity – MAAS-MI Mental Health reports the content validity of the MAAS-MI Mental Health by further elaboration of the reliability on the item level and by comparison of MAAS scores with experts’ judgments of interviewing skills. 

Chapter 12: Convergent & Divergent Validity MAAS-MI Mental Health corresponds to chapter 8 in studying the convergent and divergent validity of the MAAS-MI Mental Health. 

Chapter 13: Construct Validity MAAS-MI Mental Health is a construct validity study, relating interviewing skills with PSCC variables such as patient-satisfaction, recalled information, insight and intention to comply. 

In chapter 14: MAAS-Medical Interview: Conclusions & Recommendations, the results of these studies are summarized and their theoretical and practical implications are discussed. Chapter 15 is the translation in Dutch of the previous chapter. 

In the appendices, and on the website under TOOLS, the observers’ Manual of the MAAS-MI General and MAAS-MI Mental Health can be found, just as related instruments in TOOLS Cont’d.