A BRIEF BACKROUND

The Occupational Stress Index (OSI) is an additive burden model, which focuses on work stressors relevant to the cardiovascular system (Belkic 1995(a)). The OSI incorporates elements of the Job Strain Model (Karasek 1979), as well as other formulations of how stress leads to cardiovascular disease, such as features of work in high-risk occupations. However, in contrast to constructs such as Job Strain (Ibid.) and Effort-Reward Imbalance (Siegrist 1991, 1996), which are based heavily upon sociological theory, the OSI derives more from cognitive ergonomics and brain research, attempting to describe, in quantitative terms, the burden of work processes upon the human being. The underlying motivation for developing such an approach is to help pinpoint areas for intervention, by striving to reflect actual work experiences.
There have been two major approaches in occupational psychosocial research using self-report methods. One has been to develop occupation-specific questions. This can provide rich, detailed information useful in identifying key areas for intervention. However, these job-specific questionnaires generally cannot measure job stressors across various occupations. The other approach has been to measure generic job characteristics using questions of a general nature. However, "this approach is less useful for intervention studies, because questions are more 'remote from actual work experiences'" (Landsbergis 2000). The OSI represents a potential means of bridging these two divergent approaches. As stated by Landsbergis and Theorell (Ibid): " A recent innovative approach used occupation-specific questions (useful for workplace interventions), that are based on general questions. The Occupational Stress Index (OSI) can be tailored to specific occupations, thus allowing comparison among occupations of the stress burden faced by workers" (p. 164). We can make comparisons regarding the total burden, as well as in the nature of the occupational stress burden. These questions are of interest not only in the research setting, but are also those articulated by working people themselves.

Levels of Information Transmission	Underload	High Demand	Strictness	Extrinsic Time Pressure	Aversiveness (Noxious Exposures)	Avoidance (Symbolic Aversiveness)	Conflict / Uncertainty
Input	*Homogeneous incoming signals *Low frequency of incoming signals *Works alone	*Several information sources simultaneously followed *Hetreogeneous signals *Visual modality primary *High frequency of incoming signals *Three sensory modalities *Communication essential for work	*Strict requirements for signal detection	*No control over speed of incoming signals	*Glare *Noise	*High level of attention (serious consequences of a momentary lapse	*Signal / Noise conflict *Signal / Signal conflict
Central Decision-Making	*Decisions automatic from input	*Complex decisions *Complicated decisions *Decisions affect work of others Need for rapid decision-making	*Limited number of decision-making strategies *Limited number of correct decisions	*Decision cannot be postponed		*Wrong decisison can have serious (potentially fatal) consequences	* Missing information needed for decision *Contradictory information *Unexpected events change work plan
Output / Task Execution	*Hoomogeneous tasks *Simple tasks *Nothing to do	*Heterogeneous tasks *Simultaneous task execution *Complex tasks *Need for rapid task execution	*Work must meet a strictly-defined standard	*No control over task execution rate	*Vibration *Isometric stress	*Hazardous tasks	*Conflicting tasks in space and time *External factors hamper task execution
General	*Fixed pay	*Piece rate *Overtime work *Holds 2+ jobs *Lack of rest breaks *Night work	*Fixed body position *Work in confined space	*Speed-up *Deadline pressure	*Cold *Heat *Noxious gases/fumes/dusts	*Work accident *Witness work accident	*Emotionally charged work atmosphere (interpersonal conflicts)

From: Belkic K, Savic C, Theorell T, et al. Mechanisms of cardiac risk among professional drivers. Scan J Work Environ Health. 20:73-86, 1994.

ORGANIZATION OF THE OSI

Levels of Information Transmission
As seen in Figure (1), the OSI is arranged into a two-dimensional matrix, with the vertical axis comprised of "Levels of Information Transmission":

1) Sensory input
2) Central decision-making
3) Effector output (task performance)

These are the basic cognitive ergonomic processes, as described by Welford (1960), that can be mapped over time using neurophysiologic methods (Ivanitsky 1980). These provide a useful format for categorizing a broad range of occupational endeavor. Luczak (1971), e.g., employed this formulation in work simulation analysis. We have added a "General", level for elements that are related to the overall work environment and not localized to a specific level of information transmission.

Stress Dimensions
The stress dimensions or aspects of the OSI are placed along the horizontal axis, as follows:

1) Underload
2) High demand
3) Strictness
4) Extrinsic Time Pressure
5) Aversive/Noxious Exposures
6) Threat-avoidant vigilance/disaster potential
7) Conflict/uncertainty

The Two-dimensional Matrix
Thus, each factor has a set of coordinates, localizing it to the type of stress and the level at which it affects the human operator. Summations by levels and by stress aspects can be made, and a wide variety of combined effects can be assessed. The sum of the factor scores comprises the total OSI score, which is an attempt to quantitate the overall burden upon the human operator of a given set of working conditions. As mentioned above, using the OSI between-group comparisons regarding total burden can be made. Insight into the nature of the occupational stress burden can be gained, as well. For example, it may turn out that two very different jobs, such as work as a teacher and an assembly line worker, have similar total OSI scores. The OSI would help elucidate how these two jobs differ. As will be discussed subsequently, teachers have very high demand on the input and central decision-making levels, together with various degrees of extrinsic time pressure and conflict. In contrast, factory workers face underload (short-cycle, monotonous work), together with very strict constraints on the speed and content of the work they perform. The OSI can provide an in-depth profile of the relatively fixed, as well as potentially modifiable workplace stressors of a given job, identifying the level at which the burden primarily occurs, and the major contributing dimensions. This information helps pinpoint where modifications in the work environments would be most beneficial.

Threat Avoidant Vigilance--An Often Hidden, but Important Stress Dimension

An in-depth discussion of each of the dimensions of the OSI as they relate to other psychosocial work stress models, to cognitive ergonomics, as well as to cardiac risk can be found in (Belkic 1995(a), Belkic 2000(a)). However, stress dimension #6 warrants comment right at this juncture, because it is often left out of psychosocial job stress models, and yet is, in our opinion, an extremely important, albeit often unrecognized, work stressor.
We emphasize the following: when the potential consequences of one's actions can include disaster, work can become a "threat-avoidant" vigilant activity. Insofar as one anticipates the possibility of encountering harm of some sort, this is inevitably associated with negative emotions (Lazarus 1967). There is epidemiologic, human laboratory and experimental animal data that directly and indirectly links prolonged exposure to threat-avoidant vigilant activity with adverse cardiovascular outcomes, including cardiac electrical instability and even sudden cardiac death (Corley 1977, Lown 1990, Menotti 1985, Murphy 1991, Suurnakki 1987, Theorell 1993). Professional drivers, sea pilots, air traffic controllers and some other classes of control panel workers, all of whom perform primarily threat-avoidant vigilant work are found to be at high risk for hypertension and/or for ischemic heart disease (Winkleby 1988, Belkic 1998, Ragland 1997, Cobb 1973, Erikssen 1981, Tuchsen 2000).
Cognitive ergonomics and brain research illustrate that the heaviest burden upon conscious attentional resources occurs when one continuously follows a barrage of signals to which he or she must be prepared to rapidly respond, such that momentary lapse, error or delay could have serious, or even fatal consequences (Belkic 1992(a), Kalsbeek 1974, Levi 1981, Parasuraman 1984, Stroh 1971). For survival reasons, our nervous systems are constructed to selectively allocate mental resources to threatening stimuli, even if the threat is only of a symbolic nature.
A salient illustration of the importance of this hidden burden upon conscious attentional resources is seen in a study of human electrocortical brain activity during a simulated traffic situation. Therein, an unusually high level of selective attention was elicited when persons were confronted with an imminent traffic accident scene (Martin 1992). That study also demonstrated that compensatory allowance (especially increased time allocation) is essential in such situations. This needs to be calculated into the work planning "equation".

Description of the OSI Questionnaires

The Occupational Stress Index is questionnaire-based, and does not require on-the-job analysis. Insofar as such is available, direct work-site measurements and data can be effectively incorporated into the OSI, and will improve its accuracy.

Each element of the OSI is scored on a scale from 0 to 2, with zero being "not present" and 2 as "strongly present".

The General OSI Questionnaire
The General OSI Questionnaire is applicable to workers of any occupational profile. This has been applied among 345 workers of various occupations and of both genders, in several countries (Belkic 1995(a), 1996). Our experience is that the general OSI has good face validity, that working people consider the questions relevant to their daily life on-the-job, and that they understand quite well even the somewhat more abstract questions. The internal consistency of the total general OSI is within the desired range (Cronbach alpha = 0.81), as are most, but not all of the scales. (The two scales from the General OSI, in which internal consistency is low are: Extrinsic time pressure and Strictness. For more details, see Belkic 1995(b), as well as Landsbergis 2000)
The General OSI Questionnaire can be used for between-occupation comparisons, especially when evaluating a heterogeneous working population with a wide range of profiles.
However, as mentioned above, general questionnaires have a common weakness in their remoteness from actual work experiences. The General OSI is no exception. On the other hand, the General OSI, having been designed to focus upon objective features of work, can serve as a bridge to the next step in the application of the OSI. Namely, General OSI data from several workers in a single occupation can be used as the first phase in the development of an occupation-specific questionnaire, which can then be pilot-tested. Qualitative data from workers willing to put in the extra time to comment and explain their answers to the General OSI in relation to their actual work environment proves to be invaluable. This process is facilitated by a number of open-ended questions included at the end of the questionnaire. Expert observers can also be of great help.

Occupation-Specific OSI Questionnaires
Our aim with the OSI has been to develop a series of occupation-specific questionnaires that are all part of the OSI "umbrella". In other words, these would all be compatible with the General OSI, and allow between-occupation comparisons, but would be far more operationalized and streamlined. Namely, the latter are derived from a more qualitative approach, based on detailed knowledge of the occupation in question. Once developed, these occupation-specific OSI allow us to omit questions about the fixed aspects of a given line of work and to focus on the variable features of a given occupation, and often pinpoint key stressors that could be the focus of intervention.
These specific OSI are being designed for a very wide range of occupational endeavor-from industrial, transport, to clerical and professional sectors. Thus far, the OSI for professional drivers has been validated and widely tested, the OSI for physicians is in the final phase of piloting, and OSI for teachers, production workers, clerical workers, air traffic controllers and for airline pilots are being developed.

OSI for Professional Drivers
Our experience with the OSI for professional drivers illustrates these points. As seen in Appendix 2, the OSI Questionnaire for professional drivers is about half the length of the General OSI, and the questions are very concrete and germane to this occupational group.

We first identified those features of professional driving, which are relatively constant, such as:
--the need to make and carry out rapid, non-deferrable, but somewhat automatic decisions (a combination of decision-making underload and high demand)
--no possibility of ignoring incoming signals (strictness on the input level)
--fixed posture while behind the wheel (strictness on the general level)
--no chance to influence the rate at which new signals are received (extrinsic time pressure on the input level)

These and other features contribute to the high demand and low control of professional drivers, but because they are relatively fixed features of the occupation, queries in this regard would be superfluous. Furthermore, professional driving epitomizes threat-avoidant vigilant activity, with requirements for high levels of vigilance and potentially fatal consequences from a momentary lapse or even a slight decision-making error. Again, there is no need to ask about this.
We then took the remaining, variable features of professional driving, and sought to operationalize these in relation to the traffic environment: road and vehicle conditions, type of routes, passengers, accidents, as well as work schedules, time table stringency, rest breaks, etc. These queries are presented in a neutral way, to minimize reporting bias, especially denial or repressive coping.
Some examples are given of how this was done with respect to elements of input high demand versus underload. Frequency of incoming signals is scored by where driving predominantly takes place: within the city signifies high frequency of incoming signals (high demand), whereas driving mainly on long, inter-city routes is scored as low incoming signal frequency (underload). Heterogeneous signals (high demand) are encountered when driving on various routes while with driving on the same route day after day, relatively homogeneous signals are seen (underload). The conflict between meeting a strict time schedule and fulfilling the other task requirements, as described by Gardell et al. (1983) is reflected in an item on conflict at the level of task performance.
The OSI for professional drivers has now been applied in 327 persons. Drivers tell us that it is easy for them to complete the questionnaire and it takes only a few minutes for them to do so. The Cronbach alpha for the total OSI for professional drivers is 0.84 (variable features only). Some results obtained using the OSI for professional drivers are presented in the next section.

OSI for Physicians
An OSI for Physicians is in the final phases of pilot testing, and should be available on this Website shortly. Since the authors of this instrument are physicians themselves (Drs. Karen Belkic and Cedo Savic), the first phase of development was based upon our own experience in a variety of clinical settings. As we began to pilot test the OSI for Physicians, we contextualized this as "by physicians for physicians" within the framework of a "participatory action research" approach. This point is emphasized here because of the pivotal position of physicians with respect to the work environment and health outcomes. Namely, physicians are often called upon to make decisions about fitness for work, and can potentially have an impact upon patients' working conditions by making informed recommendations. (See Belkic, Schnall & Ugljesic 2000(b), Fisher & Belkic 2000). At the same time, physicians increasingly face an infringement of decision-making latitude, increased demands, etc., especially within the context of managed care. The underlying burden of the work of physicians is a heavy one. Documenting and quantifying this burden is important for many reasons. As an empowerment tool for physicians, we hope this will help in efforts to improve the working conditions of our own profession. We also hope that this process can be translated into better insight by physicians into the working conditions of patients. The response of our colleagues during pilot testing has been very positive and we are grateful for the time they have taken to provide invaluable insights into their working life, and how it might be improved.

There are several elements on the decision-making level, that are fixed features of work as a physician:
--need to make complex and complicated decisions (high demand on the decision-making level)
--serious, potentially fatal consequences of a wrong decision (threat avoidance on the decision-making level)
--receiving contradictory information (conflicts/uncertainty on the decision-making level)

Specific aspects of the physician's work environment, that can vary to some extent, and that contribute to their stress burden, include:
--frequency of incoming signals: related to patient load and setting (emergency room or ICU versus outpatient)
--possibility to postpone decisions: related to setting (emergency room or ICU versus outpatient)
--need for rapid task execution: related to performance of invasive procedures or urgent care
--strictness on the decision making level: related to degree of control over areas such as indications for hospitalization and procedures
--external time pressure on the output level / control over pace of task performance: related to control over number of patients, scheduling, other duties

On the general level, in addition to accidents, we have included other cognitively aversive - disastrous consequences such as patient suicide or being faced with malpractice litigation. These are major stressful events that can occur in the physician's career. We also expand the conflict dimension on the general level to include cooperation with staff, colleagues and support in the display of knowledge and career advancement. Larger issues related to control over institutional policies, etc. can be covered in an expansion of the strictness dimension on the general level (see perspectives).

In the future, we also hope to develop an OSI for nurses, and for other health care workers.

OSI for teachers
Increasingly, adverse health outcomes, especially burnout (van der Berghe 1999) are being reported among teachers, in relation to an ever-greater stress burden. We are now developing an OSI for teachers aimed at describing and quantitating this burden and helping to identify key points for intervention.

A number of fixed features of their work contribute to high demand:
--Attention to several sources of information simultaneously (input high demand)
--Communication essential for work (input high demand)
--The need to make complex and complicated decisions (central high demand)

Variable features in the teacher's work environment that are cited as exacerbating stressors include:
--Rapidity of new information -related to class size
--Speed-up and deadline pressure (extrinsic time pressure on the general level) -- related to class size and curriculum demands, as well as the need to perform other duties such as administration
--Conflict and uncertainty as well as threat-avoidant vigilant burden-problematic pupils
--Administrative task assignment - a source of high demand, extrinsic time pressure and conflict
--Threat of violence-threat-avoidance on the task performance level
--Interruptions as requiring change of plan of work and hampering task performance (conflict on the central decision-making and task performance level)
--Control over work pace-related to control over class size, number of problem pupils, size of curriculum and pressure to complete it

OSI clerical workers - the human-computer interface
Clerical work relies ever more heavily upon computer technology. While potentially increasing productivity, there are specific stressors that arise from the human-computer interaction, to which clerical workers are exposed (Smith 1999). We are attempting to describe and quantitate these and other stressors relevant to clerical workers in this specific OSI.

Some examples of these stressors include:
--interruptions (a source of conflict),
--high workload, lack of control related to electronic monitoring,
--slow down-wait time creating underload,
--disaster potential-(material or other type of damage)-sending wrong file, opening a virus-infected file,
--cumbersome - incompatible types of software -- hamper task performance, etc.

OSI for Air Transport Professionals: Air Traffic Controllers and Pilots
These professions entail enormous responsibility, with maximum threat-avoidant vigilance on the input and decision-making level. They are also characterized by alternations between underload and extremely high demands, with requirements to generate peak attention levels at certain moments. As discussed by Levi (1981), generating these very high levels of attention creates a heavy burden upon the human nervous system. Requirements for judgment place an onerous load upon decision-making capacities.

Variable features that affect burden upon Air Traffic Controllers include:
--average traffic density
--peak traffic density (highest attentional demand)
--latitude to alter decision-making strategy -considered a key buffer by Bisseret (1971) and Sperandio (1971), and one that is compromised by time pressure and other constraints

Descriptions of these stressors can be found in Costa (1993), Emdad (1997(a)) and Landsbergis (1986). Semi-structured interviews are planned with air-traffic controllers who have worked in a variety of settings.

Maximum attentional demands are made upon pilots during take-off and landing. Particularly during long flights, there are extended periods of relative underload with need to continuously maintain high levels of vigilance. Pilots are under extremely strict control upon their performance, and must continuously fulfill rigorous licensing standards. Pressures to keep on schedule and at the same time to perform their jobs safely represents a key source of conflict.
We have performed initial semi-structured interviews from which an initial version of the OSI for pilots can be developed for testing.

OSI for production-line workers
Assembly line work can be considered as the epitome of job strain, as paced work with low control and high demands.
However, there are variations that exacerbate or ameliorate the stress burden. Trends in the former direction --i.e. lean production, are reviewed in (Landsbergis 1999).

Some variable features of production line work include:
--High frequency of incoming signals -related to line speed
--Simple and homogeneous tasks -input and output underload related to short cycle time
--Strictness on the task performance level, as well as some degree of threat avoidant vigilance can be related to quality control procedure
--Physical exposures (noise, glare, vibration, lifting, chemicals, heat, cold)
--Extrinsic time pressure on the task performance level - whether working directly on assembly line

An important element of control over task performance rate relates to whether the work is directly on the assembly line as opposed to those tasks that can be performed off the line (on a separate desk or work station). In the latter case, the worker would usually have more moment-to-moment control over speed of performance.

The OSI for production-line workers is being developed together with Drs. Paul Landsbergis and Peter Schnall and will be tested among workers in the automobile industry.

We will now briefly present some of our results using the OSI for between-occupation and within-occupation analyses. Details about study methodology, including assembly of the samples, are provided in the cited references.

Between Occupation Analysis

Total Occupations Stress Burden of Professional Drivers compared to Working Referents

As shown in Table 1, the examined groups of professional drivers (heterogeneous profiles) had approximately twice the total OSI scores compared to those of a heterogeneous group of building trade workers (N=227) and of subway guard attendants (N=23) (Belkic 1992(b), Belkic 1996, Emdad 1997(b)).

High Demand, Low Control and Job Strain among Professional Drivers compared to Working Referents

Here is an example (Table 2) of how a more detailed and operationalized approach to the demand and control dimensions helps identify professional drivers as a high strain occupational group. We found a significant, positive correlation between the demand/control ratio assessed using the Swedish Psychosocial Job Strain Questionnaire (Theorell 1988), and the total OSI score. (Belkic 1996, Emdad 1997(b)). However, using the standard five questions for the demand dimension from the above-mentioned questionnaire (Theorell 1988), professional drivers scored non-significantly lower than the subway attendant referents. For decision latitude, skill discretion and demand/control ratio as a quotient term, there were also no significant differences. In contrast, the total scores for the high demand, strictness and extrinsic time pressure dimensions of the OSI all were significantly higher among the drivers. Furthermore, a breakdown of the demand dimension using the OSI reveals that the heaviest demand is at the input level, whereby drivers must follow several information primarily visual sources simultaneously, but using all three sensory modalities (fixed features). In addition, the urban mass transit operators have a high frequency of incoming signals, and must communicate with the public.

Within-Occupation Analyses

The Occupational Stress Index has also shown validity in within-occupation analyses.

Total OSI as a predictor of Smoking Intensity

The overall burden of exacerbating (i.e. non-fixed) stressors in the professional drivers' work environment, as gauged by the total OSI independently predicts cardio-deleterious behaviors such as smoking intensity within this group (Belkic 1996, Emdad 1998). The multiple linear regression analysis is shown in Table 3.

References: Belkic et al. (1996), Emdad et al. (1998)

As we have said, professional drivers as a group are exposed to very high levels of occupational stressors and all have much in common. However, each driver profile also faces a specific set of stressors. As shown in Table 4, we used the OSI to compare 130 city bus drivers and 69 truck drivers. The mean total OSI scores were very high for both groups (65.2 +/- 3.6 and 68.7 +/3.8, respectively)(Belkic 1995). However, while these two groups share many features of their working environment, using the OSI, we were able to identify a number of important differences.
One major difference is that city bus drivers face predominantly overload, while truck drivers are exposed to a mixture of underload and overload. City bus drivers had a larger high demand score due to receiving a rapid flow of new information and having to communicate with the public, as well as performing more tasks simultaneously and often lacking rest breaks and working at night. In contrast, when driving on long routes, truck drivers have a relatively low flow of new information (monotony) and frequently drive alone, which is another source of underload, as well as social isolation. At the same time, they still must keep their sensory systems (especially visual) on full alert at all times ready to make rapid decisions and actions. This need for sustained vigilant monitoring combined with monotonous road conditions is recognized to be a very important contributor to fatigue during long-distance truck driving (Williamson 1996). An additional source of underload, which contributes to fatigue, is delays and long waiting times. We found these to be frequent occurrences among the truck drivers. Another important difference is that the truck drivers worked significantly longer hours and were paid by the number of routes driven (two important sources of overload).
Truck drivers performed heavy lifting significantly more often, and drove under more hazardous conditions (including carrying explosive cargo, and driving on narrow, winding roads). However, they reported fewer accidents. Extrinsic time pressure was greater for city bus drivers because they were obliged to follow a rigid schedule. They were exposed to more glare because of driving at night, and had more vibration exposure due to poorer shock absorbers, as well as poorer heating, cooling and isolation systems in their vehicles. City bus drivers had more conflict/uncertainty at various levels. Conflict on the input level was related to more difficult signal detection due to decreased visibility. They also had more vehicle breakdowns, other impediments to task performance, poorer interpersonal work atmosphere, as well as more conflicts between the need to arrive on time and traffic congestion which prevented them from doing so.
These latter results coincide with the reports of Gardell (1983), Syme (1991) and Evans (1994), that the work environment of urban bus drivers frequently entails the need for rigid adherence to schedule, which is hampered by traffic congestion. This leads to conflict and even punitive consequences, as well as the loss of the rest breaks that are so badly needed. Threat of violence is yet another stressor which increases the cognitive aversiveness of the city mass transit drivers' work environment.
The reader is referred to article1 and article 2 for further discussion on Truck Drivers and Urban Mass Transit Operators, respectively, on this Website.

Limitations in the Current Applications of the OSI and Perspectives for the Future

A major limitation of the current application of the Occupational Stress Index has been the linear nature of the analyses that have been performed. There is a need to explore possibilities for multiplicative interactions and higher level terms, especially in relation to existing models such as Job Strain. Eventually, weighting factors might be developed to reflect the relative contribution of the various factors to the total burden.
The extrinsic time pressure and strictness dimensions have a low Cronbach alpha for the general OSI. This needs retesting with the occupation-specific OSI's. It may be worthwhile to perform factor analysis. There may be, e.g., one factor of control over task performance, deadline pressure and speed up, and another related to the input and central levels.
As it now stands, many key macro-level stressors are not included in the OSI. These include job security, adequacy of pay, lack of control over one's work schedule, as well as over larger issues such as institutional policy, buffers such as unionization/collective control (Johnson 1995), promotions (Siegrist 1996). Many of these are important vis-à-vis cardiovascular outcomes. There is also a need to reflect emotional rewards intrinsic to work (as opposed to social support). We are now considering the possibility of expanding the strictness and conflict dimensions on the general level to incorporate some of these factors.
In the current versions of the OSI, we have asked about rest breaks and long work hours as general factors affecting high demand. We may include number of vacation days in subsequent versions.
Inter-personal relations at work have been heretofore evaluated in the OSI with a single item as a global assessment. This could be broken down to refer to supervisors, to co-workers and to other staff.
Integration could be performed with objective measurements (part of triangulation). For example, for city bus drivers there could be finer gradations to the scoring of high frequency of incoming signals based upon traffic density measures. Using a similar logic, average number of passengers could improve quantitization of the burden of communicating with the public. Information obtained from expert-observer assessment of job characteristics, as outlined by Greiner and Krause (2000) could be very well integrated with the OSI. On the other hand, the OSI could detect areas for which in-depth observational analysis is needed, especially with the view to practical improvements in the work environment.

Belkic, K; Savic, C; Djordjevic, M; Ugljesic, M; Mickovic, Lj. Event-related potentials in professional city drivers: heightened sensitivity to cognitively relevant visual signals. Physiol Behav. 1992 (a); 52: 423-427.

Belkic K, Savic C, Theorell T, Cizinsky S. Work Stressors and Cardiovascular Risk: Assessment for Clinical Practice. Part I. Stockholm (Sweden): Stress Research Reports. National Institute for Psychosocial Factors and Health. Section for Stress Research, Karolinska Institute, WHO Psychosocial Center; 1995(a). Report No.: 256.

Belkic, K; Pavlovic, S; Djordjevic, M; Ugljesic, M; Mickovic, Lj. Determinants of cardiac risk in professional drivers. Kardiologija 1992 (b); 13: 145-149.

Belkic, K; Savic, C; Theorell, T; Rakic, Lj; Ercegovac, D; Djordjevic, M. Mechanisms of cardiac risk among professional drivers. Scand J Work Environ Health 20:73-86; 1994.

Belkic K. Psychosocial triggers of myocardial ischemia in women. Research Report to the Swedish Medical Research Council. 1995(b).

Belkic K, Emdad R, Theorell T, Cizinsky S, Wennberg A, Hagman M, Johansson L, Savic C, Olsson K. Neurocardiac mechanisms of heart disease risk among professional drivers. Stockholm: Swedish Fund for Working Life, 1996.

Belkic K, Emdad R, Theorell T. Occupational profile and cardiac risk: possible mechanisms and implications for professional drivers. International Journal of Occupational Medicine and Environmental Health. 1998; 11: 37-57.

Belkic K, Savic C. Dehumanization versus humanization of work: Insights from cognitive ergonomics and brain research. . In: Schnall PL, Belkic K, Landsbergis PA, Baker D (eds.) Occupational Medicine: State of the Art Review. The Workplace and Cardiovascular Disease. 2000(a); 15: 87-94.

Belkic K, Schnall P, Ugljesic M.Cardiovascular evaluation of the work and workplace: A practical guide for clinicians. In: Schnall PL, Belkic K, Landsbergis PA, Baker D (eds.) Occupational Medicine: State of the Art Review. The Workplace and Cardiovascular Disease. 2000(b); 15 (1): 213-222.

Bisseret A. Analysis of processes involved in air traffic control. Ergonomics. 1971; 14: 565-570.

Cobb S, Rose RM. Hypertension, peptic ulcer disease and diabetes in air traffic controllers. JAMA 1973; 224: 489-492.

Corley KC , O'Shiel F, Mauck HP. Myocardial degeneration and cardiac arrest in squirrel monkeys. Physiologic and psychologic correlates. Psychophysiology 1977; 14: 322-328.

Costa G Evaluation of workload in air traffic controllers. Ergonomics. 1993; 36: 1111-1120.

Emdad R, Belkic K, Theorell T. Cardiovascular dysfunction related to threat,avoidance and vigilant work: Application of event-related potentials and critique. Integ Physiol Behav 1997(a) ; 32: 202-219.

Emdad R, Belkic K, Theorell T, Cizinsky S, Savic C, Olsson K. Work environment, neurophysiologic and psychophysiologic models among professional drivers with and without cardiovascular disease: Seeking an integrative neurocardiologic approach. Stress Med 1997(b); 13: 7-21.

Emdad R, Belkic K, Theorell T, Cizinsky S. What prevents professional drivers from following physicians' cardiologic advice? Psychoth Psychosom 1998; 67: 226-240.

Erikssen J, Johansen AH, Rodahl K. Coronary heart disease in Norwegian sea-pilots: part of the occupational hazard? Acta Med Scand Suppl. 1981; 645: 79-83

Evans GW. Working on the hot seat: urban bus operators. Accid Anal Prev 1994; 22: 181-193.

Fisher J, Belkic K. A public health approach in clinical practice. In: Schnall PL, Belkic K, Landsbergis PA, Baker D (eds.) Occupational Medicine: State of the Art Review. The Workplace and Cardiovascular Disease. 2000; 15: 245-253.

Gardell B, Aronsson G, Barklof K. The working environment for local public transport personnel. Stockholm: The Swedish Work Environment Fund, 1983.

Greiner BA, Krause N. Expert-observer assessment of job characteristics. In: Schnall PL, Belkic K, Landsbergis PA, Baker D (eds.) Occupational Medicine: State of the Art Review. The Workplace and Cardiovascular Disease. 2000; 15: 175-183.

Ivanitsky AM. Evoked potentials and mental processes. In Lechner H, Aranibar A. Electroencephaologr Clin Neurophysiol Amsterdam: Excerpta Medica, 1980, pp. 727-732.

Johnson JV, Hall EM. Class, work, and health. In: Amick B, Levine S, Tarlov AR, Walsh Dc (eds): Society and Health. New York, Oxford University Press, 1995, pp. 247-271.

Kalsbeek JWH. Prevention of excessive mental load, and how can the industrial engineer and the ergonomist cooperate. Laboratorium voor ergonomische psychologie van de gezondheidarganisatie, TNP. pres., Conference of the European Federation of Productivity Service, Berlin, 1974.

Karasek RA. Job demands, job decision latitude and mental strain: Implications for job redesign. Adm Sci Q 1979; 24: 285-307.

Landsbergis PA. Is air traffic control a stressful occupation? Labor Studies Journal. 1986;***: 117-134.

Landsbergis PA, Cahill J, Schnall PL. The impact of lean production and related new systems of work organization on worker health. J Occup Health Psychol 1999; 4: 1-23.

Landsbergis PA, Theorell T. Measurement of psychosocial workplace exposure variables: Self-report questionnaires. In: Schnall PL, Belkic K, Landsbergis PA, Baker D (eds.) Occupational Medicine: State of the Art Review. The Workplace and Cardiovascular Disease. 2000; 15: 163-171.

Lazarus, R.S. (1967). Stress theory and psychophysiological research. In L. Levi (Ed.), Emotional Stress Physiological and Psychological Reactions Medical, Industrial and Military Implications. Försvarsmedicin. 3: 152-177.

Levi L. Preventing Work Stress. Reading: Addison-Wesley Publishing Co., 1981.

Lown B. Role of higher nervous activity in sudden cardiac death. Jpn Circ J. 1990; 54: 581-602.

Luczak H. The use of simulators for testing individual mental working capacity. Ergonomics 1971; 14: 651-660.

Martin, F., Siddle, D.A.T., Gourley, M., Taylor, J., Dick, R. (1992). P300 and traffic scenes: The effect of temazepam. Biological Psychology, 33: 225-240.

Menotti A, Seccareccia F. Physical activity at work and job responsibility as risk factors for fatal coronary heart disease and other causes of death. J Epidemiol Commun Health 1985; 39: 325-329.

Murphy LR: Job dimensions associated with severe disability due to cardiovascular disease. J Clin Epidemiol 1991; 44: 155-166.

Parasuraman, R. (1984). Sustained attention in detection and discrimination. In R. Parasuraman & D.R. Davies (Eds.), Varieties of Attention. Orlando: Academic Press, Inc., pp. 243-271.

Ragland DR, Greiner BA, Holman BL, Fisher JM. Hypertension and years of driving in transit vehicle operators. Scand J Soc Med 1997; 25: 271-279.

Siegrist J. Adverse health effects of high-effort/low-reward conditions. J Occup Health Psychol. 1996; 1: 27-41.

Siegrist J, Peter R, Georg W, Cremer P, Seidel D. Psychosocial and biobehavioral characteristics of hypertensive men with elevated atherogenic lipids. Atherosclerosis 1991; 86: 211-218.

Smith MJ, Conway FT, Karsh B-T. Occupational stress in human computer interaction. Indust Health 1999; 37: 157-173.

Sperandio JC. Variation of operator's strategies and regulating effects in workload. Ergonomics 1971; 14: 571-577

Stroh, C.M. (1971). Vigilance: The Problem of Sustained Attention. Oxford: Pergamon Press.

Suurnakki T, Ilmarinen J, Wagar G, Jarvinen E, Landau K. Municipal employees' cardiovascular disease and occupational stress factors in Finland. Int Arch Occup Environ Health 1987; 59: 107-114.

Syme L. Social epidemiology and the work environment. In Johnson J, Johansson G. The psychosocial work environment. Amityville, New York: Baywood, 1991, pp. 21-32.

Theorell T. Medical and physiological aspects of job interventions. In: Cooper CL, Robertson IT (eds.). International Review of Industrial and Organization Psychology.1993, Vol 8, New York, John Wiley, 1993, p. 173-192.

Theorell T, Perski A, Åkerstedt T, Sigala F, Ahlberg-Hulten G, Svensson J, et al. Changes in job strain in relation to changes in physiological states A longitudinal study. Scand J Work Environ Health 1988; 14: 189-196.

Tuchsen F. High-risk occupations for cardiovascular disease. In: Schnall PL, Belkic K, Landsbergis PA, Baker D (eds.) Occupational Medicine: State of the Art Review. The Workplace and Cardiovascular Disease. 2000; 15: 57-60.

van der Berghe R, Huberman AM (eds.). Understanding and preventing teacher burnout. A Sourcebook of Internaitonal Research and Practice. Cambridge University Press, Cambridge, 1999

Welford AT. The measurement of sensory-motor performance: Survey and reappraisal of twelve years' progress. Ergonomics 1960; 3: 189-230.

Williamson AM, Feyer A-M, Friswell R. The impact of work practices on fatigue in long distance truck drivers. Accid Anal Prev 1996; 28: 709-719.

Winkleby, MA; Ragland, DR; Fisher, JM; Syme SL. Excess Risk of Sickness and Disease in Bus Drivers: A Review and Synthesis of Epidemiological Studies Int J Epidemiol 1988; 17: 255-262.