Thursday, March 11, 1999
3:15 pm - 4:30 pm

David LeGrande, M.A.

In April, 1988, the Communications Workers of America (CWA), AFL-CIO, and CLC, announced the completion of the "CWA New Jersey State Workers’ Visual Display Terminal Study". The scientific investigation, initiated in 1996, was sponsored and conducted by CWA’s Occupational Safety and Health Department, the Department of Industrial Engineering, University of Wisconsin, and the Psychology Department, Rowan University, and the State of New Jersey. The study focused upon visual display terminal or computer use and the identification of related cumulative trauma musculoskeletal health symptoms and disorders. In turn, collected data was analyzed to determine the scientific rationale for the occurrence and cause(s) of identified health problems.

Study participants included 514 CWA members employed in clerical, technical, and professional occupations within the New Jersey Treasury, Labor, Education, and State Departments. The scientific investigation found significantly high rates of member cumulative trauma musculoskeletal health symptoms and disorders related to visual display terminal work. For example:

• 56% of participants reported experiencing hand pain,

• 24% elbow pain,

• 50% shoulder pain,

• 67% neck pain, and

• 55% back pain.

Findings by occupational classification indicated that data entry visual display terminal workers reported the highest rates of cumulative trauma musculoskeletal pain, following by technical, clerical, and professional workers. In addition, the study’s findings demonstrated that workers who had developed cumulative trauma musculoskeletal health problems made substantial use of their health care benefits, were awarded workers’ compensation benefits, and missed a substantial amount of work.

Study data demonstrated a high rate of cumulative trauma musculoskeletal health symptoms and disorders among participating workers who utilized a visual display terminal or computer to perform their jobs. This was particularly true for participants who used visual display terminal equipment more than four hours per work day.

To more precisely identify the cause of worker cumulative trauma musculoskeletal health symptoms and disorders, CWA has proposed and the State departments have agreed to initiate a cooperative intervention process. Planned activities would include:

• choosing a high risk pilot location,

• conducting an evaluation of physical and work organization ergonomics factors,

• administering a health symptoms survey,

• based on findings from the above procedures, designing a labor/management intervention, and

• conducting an evaluation of the results of the intervention.

• In turn, collected data would be used to bring about improved visual display terminal working conditions for involved workers.

CORRESPONDING AUTHOR: David LeGrande, M.A., Director, Occupational Safety and Health, Communications Workers of America, 501 3rd Street, N.W., Washington, D.C. 20001-2727

Michael Polanyi, M.A., Institute for Work & Health

Repetitive strain injuries (RSI) constitute an increasing human and economic burden in industrialized countries. Yet views conflict on a number of aspects of RSI: How prevalent and serious is RSI? What causes it? How can it be prevented?

In an attempt to bridge these differences, a "future search" conference with sixty key stakeholders (employers, injured workers and their advocates, union leaders, health professionals, ergonomists, engineers, government officials, and public and private insurers) was held in Hamilton, Ontario. Future search is a common ground building process in which a diverse and representative group of participants generates and analyses shared data based on its own knowledge and experience.

This paper reports on the outcomes of this conference on RSI which was evaluated through both participant reflection and qualitative analysis of conference materials and post-conference participant interviews.

The conference appears to have broken down barriers between stakeholder groups, increased appreciation of the beliefs and experiences of other stakeholders, and identified areas of common ground for future action. Participants agreed: a) RSI is having a serious impact on many workers; b) RSI is complex and contributed to by both individual behaviour and conditions in the workplace; c) all stakeholders need to work together to prevent and treat RSI through a systematic approach; and d) further research is needed to identify best prevention practices and treatment. There was disagreement on: a) the usefulness and clinical significance of the term "RSI"; b) the necessity of ergonomic standards and regulations for prevention; c) the effectiveness of current treatment approaches; d) who should have control over workplace decisions; and e) the relationship between global competitiveness and workplace health.

Actions agreed upon included education; identifying and disseminating best practices; promoting links between health and productivity; and developing best treatment guidelines. There was no consensus about the need for national ergonomic standards or a better classification of "RSI."

Common ground and action plans tended to focus on research and education to convince employers that investing in worker health is in their own best interest. Implicit in this focus are contested assumptions about the compatibility of employer and worker interests; the nature of knowledge about worker health; the role of information dissemination in social change; and the entitlement to workplace health and compensation. These assumptions were questioned by some participants.

Implications for the prevention and treatment of RSI, and multi-stakeholder processes in general, are discussed.

CORRESPONDING AUTHOR: Michael F.D. Polanyi M.A., Institute for Work & Health, 250 Bloor Street East, Toronto, ON, Canada, M4W 1E6.

Grant D. Huang, B.A., Steven M. Berkowitz, M.S., Michael Feuerstein, Ph.D. Uniformed Services University of the Health Sciences, Georgetown University Medical Center

Research has suggested that medical, ergonomic, and psychosocial factors are involved in work-related musculoskeletal disorders and associated disability. The present case-control studies examined demographic, physical, individual psychosocial, and occupational psychosocial factors as predictors of work-related upper extremity and back disability in a large, diverse workforce (i.e., U.S. Army).

All cases were disabled with an upper extremity disorder (Study 1, n = 171) or lumbosacral strain (Study 2, n = 174) and were obtained from the U.S. Army Physical Disability Agency’s database. Non-disabled controls consisted of two different samples (Study 1, n = 171 ; Study 2, n = 173) randomly selected from the U.S. Army’s Health Risk Appraisal database. Cases and controls were compared on measures from the Army’s Health Risk Appraisal (HRA). Cases and controls completed the HRA within the same time period. In addition, all cases completed the HRA between six months and five years prior to their disability claim. Potential risk factors examined were: age, gender, ethnicity, military rank, frequency of physical exercise, worries, social support, and occupational stress. Multivariate logistic regression analyses were conducted using case or control status as the outcome (dependent) variable.

In Study 1, older age (OR = 1.11, p < 0.001), ethnicity (Caucasian: OR = 1.54, p < 0.05), lower rank (OR = 3.79 - 4.39, p < 0.01), and occupational stress ("Often": OR = 2.46, p < 0.01) were found to predict upper extremity-related disability. A 69.0% correct classification rate (p < 0.001) was obtained from the regression model. Study 2 indicated that age (OR = 1.13, p < 0.001), lower rank (OR = 3.02 - 4.08, p < 0.01), frequency of aerobic exercise ("Never": OR = 2.20, p < 0.05), social support ("Never": OR = 5.07, p < 0.05), worries ("Sometimes": OR = 2.17, p < 0.05), and occupational stress ("Often": OR = 2.71, p < 0.05) were predictive of lumbosacral-related disability. The regression model provided a 73.1% correct classification rate (p < 0.001).

By utilizing a large, administrative database, a combination of demographic, physical, individual, and occupational psychosocial variables were found to predict musculoskeletal-related disability. The findings that lower rank and occupational stress predicted disability in both upper extremity- and lumbosacral-related disability highlight the importance of addressing organizational factors in the secondary prevention of such disability. The modifiable nature of occupational stress suggests that efforts targeting musculoskeletal disability prevention should focus on the reduction of such stress.

CORRESPONDING AUTHOR: Michael Feuerstein, Ph.D., Department of Medical and Clinical Psychology, USUHS, 4301 Jones Bridge Road, Bethesda, MD 20814

Chair: June M. Fisher, University of California, San Francisco; Presenters: Blanche Grosswald, M.S.W., School of Social Welfare, University of California, Berkeley, Patricia V. Roehling, Ph.D., Cornell University, Marcie Pitt-Catsouphes, Boston College Carroll School of Management, and Ellen Galinsky, Families and Work Institute; Discussant: David R. Ragland, Ph.D., School of Public Health, University of California, Berkeley

The symposium explores a number of areas relating to policy and research on the topic of work-family balance, including (i) the effects of occupational factors on family health, (ii) the effects of work-family benefits on employee loyalty, (iii) a new research framework with a focus on the role of the community in work-family balance, and (iv) the changing characteristics of the workplace.

The symposium presider is Dr. June Fisher, Associate Clinical Professor of Medicine at the University of California, San Francisco and a member of the clinical faculty of the Northern California Occupational and Environmental Health Center at the University of California. The symposium discussant is Dr. David Ragland, who has a doctorate in psychology and is an epidemiologist in the Department of Epidemiology and Biostatistics at the School of Public Health at the University of California, Berkeley. Four presenters cover different topic areas. Blanche Grosswald (School of Social Welfare, University of California, Berkeley) discusses the effects of bus drivers’ jobs on their families’ well being. Dr. Patricia V. Roehling (Cornell Careers Institute) examines the relationship between workplace flexibility and employee commitment across the life course. Marcie Pitt-Catsouphes (Center on Work and Family, Boston College) addresses the community context of work-family balance. Ellen Galinsky (Families and Work Institute) presents results from the 1997 longitudinal National Study of the Changing Workforce, comparing workplace characteristics relevant to work-family balance in 1997 with those of 1992 and 1977.

CORRESPONDING AUTHOR: Blanche Grosswald, M.S.W., School of Social Welfare, University of California, Berkeley, Berkeley, CA 94720

Blanche Grosswald, M.S.W., School of Social Welfare, University of California, Berkeley

Interviews with 22 San Francisco transit operators were conducted to investigate the effects of their jobs on the well-being of their families. Questions focused on amount of time spent with family members; quality of relationships with spouse, partner, and children; level of participation in community activities outside of work, perceived level of support for families from management; and effects of job on welfare of self, spouse or partner, and children.

The principal aspects of the job that had the most significant impact on transit workers in their relationships with their families were lack of time and the stressful nature of the job leading to poorer quality of time spent with family members. Only 52.4% of those interviewed had time to eat at least one meal a day with their families. Only 13.3% could help children in person with their homework on a daily basis.

Time spent away from families in itself lowered the quality of the time spent with family members. Children would not understand why their parent was never around and some spouses were able to communicate mainly via written notes on the refrigerator. The quality of the time spent together was considered less than optimal because the operators were tired, stressed, and often in a bad mood as a result of their jobs. Except for the salary and benefits, which all participants agreed were positive for them and their families, the general consensus was that the other aspects of the job had a negative effect on family members and on family life.

Policy implications of the interview results led to recommendations based on suggestions from the respondents and on observations of operators’ answers to all the interview questions. A suggestion for shorter shifts – 6 hours per day rather than the current 12 hours per day average is one that would involve major planning, cost-benefit analyses, and discussion. Other, simpler to implement recommendations include: modifying bus route scheduling to be more realistic than at present, considering distance, traffic, passenger load, and operator needs for rest breaks; providing childcare facilities convenient for operators’ schedules; and expanding existing Employee Assistance Program (EAP) to include crisis counseling for operators who are victims of crime, accidents, and other traumatic events on or off the job.

*Patricia V. Roehling, Ph.D., Liane O’Brien, MPA, Phyllis Moen, Ph.D., Cornell Family and Employment Careers Institute, Cornell University

Using data from the 1992 National Study of the Changing Workforce, a nationally representative sample of 3,381 adult American workers, this study investigates relationships among flexible policies and practices on employee commitment, as measured by loyalty and intention to turn over. The central thesis is that our understanding of the impact of flexible policies and practices on employee commitment will be enriched by greater consideration of the multiple, intertwined, non-work and work contexts that influence employee attitudes and behavior.

Our thesis received considerable support. There is a strong positive relationship between employee commitment and flexible employment policies, supervisor support for those policies, and employer tolerance of work-family conflicts. These relationships are moderated by life course variables. Loyalty, our first measure of employee commitment, is positively related to age and generally higher among women than men. Parenthood, which moderates these relationships, is related to lower levels of loyalty among married women and higher levels of loyalty among married men, equalizing reports of loyalty among mothers and fathers. Flexible-time policies and supervisor support were related to increased loyalty for all life stages and for both men and women. Taken together, these variables explain from 21% to 23% of the variance in loyalty scores, depending upon the gender and marital status of the respondent.

Similar relationships were found with intention to turn over, our second measure of employee commitment. Intention to turn over is negatively related to age for both men and women, which supports the previous finding that employee commitment generally increases with age. Parenthood, marital status and gender moderate this relationship. Married women are more likely to contemplate turnover when they have preschool-aged children, as opposed to unmarried women who reported a decrease in intention to turn over when their children were of preschool age. Also consistent with the employee loyalty findings, flexible-time policies and supervisor support were related to a decrease in intention to turn over among men and women. However, among married women, flexible-time policies were not related to decreased intention to turn over. Taken together, these variables explain from 11% to 23% of the variance in intention to turn over scores, depending upon the gender and marital status of the respondent.

In summary, employee commitment increases across the life course and is enhanced by flexible-time policies and supervisor support for those policies. This relationship, however, is moderated by gender and by life course variables such as parenthood, marital status and life stage, underscoring the importance of the life course research perspective.

CORRESPONDING AUTHOR: Patricia V. Roehling, Ph.D. , Director of Research, Cornell Careers Institute, Cornell University, Ithaca, NY 14853

Marcie Pitt-Catsouphes, M.S.P., Boston College Center for Work & Family

During the past 20 years, work-family issues have been viewed from two perspectives: 1) family responsiveness; and 2) workplace responsiveness.

� Family Responsiveness Perspective: As increasing numbers of women have entered the workforce, families have developed strategies to resolve work-family problems. Family responsiveness sees work-family issues as private. It has become clear, however, that a number of systemic problems (e.g., lack of quality childcare) render it difficult for most families to address work-family problems by themselves.

� The Corporate Responsiveness Perspective: By the mid-1980s, employers realized that unresolved work-family conflicts are often manifested at the workplace. As a result, employer policies, programs, and practices have proliferated. Corporate responsiveness sees work-family issues as being relevant business concerns.

Although both the family and the workplace responsiveness perspectives have made important contributions to our understanding of work-family issues, the theoretical frameworks which inform them have not directed a sufficient level of research or practice attention to important questions, e.g., How do policies and practices of community-based organizations affect the work-family experiences of working families? How do informal social support systems in the community affect the work-family experiences of different family members? How might organizational partnerships formed at the community level affect family and community well-being?

� Toward the Third Dimension: The Community Responsiveness Perspective: It has been recognized that the development of a community responsiveness perspective for work and family issues would contribute significantly to: understanding work and family issues as experienced by different family members, and identification of resources to address challenges faced by working families. A community responsiveness perspective would provide a framework to study the relationships between work-family experiences and the quality of life of communities and clarify how community supports could provide resources for working families.

Researchers and practitioners are challenged to develop effective approaches for promoting the responsiveness of communities to the work-family priorities of their family residents. Ideas for encouraging communities to become more family-friendly include: identifying groups as stakeholders in family-friendly communities; documenting relationships between the well-being of working families, business success, and community well-being; developing models of community responsiveness; and creating standards for family-friendly communities.

CORRESPONDING AUTHOR: Marcie Pitt-Catsouphes, M.S.P., Director of Policy and Research Development, Center for Work & Family, Boston College Carroll School of Management, Chestnut Hill, MA 02167

*Jennifer Swanberg, Ph.D., *Ellen Galinksy, M.S., James T. Bond, Families and Work Institute

Findings from the 1997 National Study of the Changing Workforce suggest that quality of workers’ jobs and the supportiveness of their workplaces are the best predictors of productivity – job satisfaction, commitment to employers, and retention. The study also found that characteristics of jobs have important effects on the workers’ personal lives. This leads us to ask whether characteristics of workplaces are improving for U.S. employees. Our findings suggest that the quality of jobs has improved over the past 20 years, and workplaces appear to be a bit more supportive even over the past five years. However, there is room for improvement. Jobs have become less secure. They have also become more demanding, making it harder to achieve a balance between work and personal life.

This paper presents 3 dimensions of job characteristics: job demands, job quality, and compensation and benefits and provides a historical perspective by comparing data from 1997 with data from the U.S. Department of Labor’s 1977 Quality of Employment Survey. Job demands include: all hours worked at main job; number of nights away from home on business during the past three months; frequency of required overtime with little or no notice; regular daytime shift versus other schedule; frequency of bringing work home; and an index of perceived job pressures. Job quality includes: job autonomy, learning opportunities on the job, meaningfulness of job, opportunities for advancement on the job, and job security. Five general indicators of workplace characteristics are considered: flexible work arrangements, supervisor support, supportiveness of workplace culture, coworker relations, and discrimination. Paid and unpaid work hours at all jobs are longer today than 20 years ago. All paid and unpaid hours at all jobs have risen from 43.6 hours in 1977 to 47.1 hours in 1997 per week. Sixty-four % of employees would like to work fewer hours compared to 47% of employees in 1992. The proportion of employees bringing work home from the job once a week or more has increased by 10% since 1977. Job pressures have increased substantially over the past 20 years. The biggest change in employees’ perceptions of job pressures relates to time. In 1997, 60% of workers agree with the statement: "I never seem to have enough time to get everything done on my job," a huge increase from 40% in 1977. Most employees view their supervisors as quite supportive, and supervisor supportiveness has increased somewhat since 1992. Two-thirds or more of employees view their workplace cultures as family-friendly. However, the only significant change over the past five years is that employees interviewed in 1997 are less likely (29%) than 1992 employees (36%) to agree that there is an unwritten rule at work forbidding them to take care of family needs on company time.

CORRESPONDING AUTHOR: Ellen Galinsky, M.S., Families and Work Institute, 330 7th Ave., New York, NY 10001

Per-Olof Östergren, M.D., Ph.D., Department of Community Medicine, Lund University, Sweden

The Job Content Questionnaire (JCQ) was developed twenty years ago. Different versions of the JCQ and related scales have since been used in a great number of studies in many different settings and in relation to many different outcomes.

A recent review comparing means and standard deviations of the different scales assessing psychological job demands, decision latitude and job support, revealed striking similarities as well as interesting differences between countries.

This ought to be a fruitful starting point for a discussion on how the work organization has developed over time in different countries and in different occupational areas of the, and what this might mean for current and future health of individuals in the workforce. In other words, a topic of great importance for work site health promotion policy development. However, in order to obtain comparable data from different times, countries and occupational settings, a standardized procedure must be developed for making valid comparisons using different versions of the JCQ as well as related scales.

An attempt to do such a comparison between the JCQ and a similar Swedish instrument will be presented, and a draft standardized format for such procedures will be presented and discussed. This will be based on a study of a random sample of vocationally active middle-aged men and women in the general population of the city of Malmö in southern Sweden, where both instruments were utilized simultaneously.

CORRESPONDING AUTHOR: Per-Olof Östergren, M.D., Ph.D., Department of Community Medicine, Lund University, Malmö University Hospital, SE 205 02 MALMÖ, Sweden

Tage S. Kristensen, Dr.Med.Sc. & M.S., Vilhelm Borg, M.Psych.., National Institute of Occupational Health, Copenhagen, Denmark.

The measurement of psychological job demands raises a number of theoretical, conceptual, and methodological questions. First, the theoretical basis and hypotheses have to be discussed and clarified. Second, the concepts have to be clearly defined and the operational definitions have to correspond closely to the conceptual. Third, the items of the scales measuring psychological job demands have to be tested in order to elucidate potential problems with regard to validity, precision, and reliability.

In the past, a number of studies have demonstrated that the job demand dimension, as measured with the JCQ or other related questionnaires, has not functioned optimally from a theoretical and methodological point of view. Different types of demands have been combined in the same scales, the exact meaning of a number of items has been unclear, and some of the research results have been difficult to interpret. In particular, the scales measuring demands have not distinguished between qualitative demands and quantitative demands. Perhaps these methodological problems can explain a number of the unexpected findings with regard to job demands and health in the literature.

In the National Danish Psychosocial Study we have tried to solve these problems by constructing five different scales for psychological demands at work: 1. Quantitative demands (e.g., time pressure, high work pace, and high workload). 2. Emotional demands (e.g., emotionally demanding work with people, hiding emotions at work). 3. Cognitive demands (e.g., concentration, memory, knowledge, solving problems). 4. Responsibility demands (e.g., taking important decisions, potential harm if decisions are wrong, and significance for many people). 5. Sensorial demands, (e.g., demands on vision, hearing ability, precision). Each of the dimensions has been measured with between 4 and 9 items. The population of the study comprises 1,858 persons currently working; age 20-60 years; 49% females. The sample is a representative sample based on personal registration numbers. Two thirds of the respondents received the questionnaire by mail and one third were interviewed by telephone.

The five psychological job demand dimensions are being analyzed with the use of factor analysis, analysis of differential item function and analysis of internal reliability. Moreover, the known-group validity and the construct validity will be elucidated. The main results from these analyses will be presented and discussed.

CORRESPONDING AUTHOR: Vilhelm Borg, National Institute of Occupational Health, Lerso Parkalle 105, DK-2100 Copenhagen O, Denmark.

Vilhelm Borg, M.Psych., and Tage S. Kristensen, Dr.Med.Sc. & M.Sc., National Institute of Occupational Health, Copenhagen, Denmark.

Aim. This study was designed to test the validity of four scales for measurement of the core dimensions of the psychosocial work environment: psychological demands, decision authority, skill discretion, and social support. The scales are comprised of questions translated from the Whitehall Study, which are very similar to the JCQ.

Method. The study is based on ten Danish samples, which embrace approx. 15,000 employed people, age 18-70 years. The validity of the scales will be analyzed with tests of reliability, internal consistency, discriminative validity, differential item functioning, and by comparison of job groups (known-group validity).

Preliminary results (based on four of the samples, N=7480: two national representative samples, a representative sample of salesmen, and the sample from the PRIM Study of job groups with a very high prevalence of repetitive work). It was found that the demand scale (3 items) had a good discriminative validity, but a rather poor internal consistency. The other three scales had both a high internal consistency and a high discriminative validity. In the demand scale we also found a substantial differential item functioning, as some groups, e.g. slaughterhouse workers, had a high mean score on the work-pace question and a low mean score on the deadline question, while other groups, e.g. salesmen, had a reversed score combination. In the decision authority scale (9 items) we found no differential item functioning. In three of the scales we found considerable differences between job groups. This was, however, not the case with the social support scale (4 items).

We found some interesting correlations between the scales. As expected there was a high correlation between decision authority and skill discretion in the composite sample (.64), but the correlations were substantially lower in the four sub-samples (.33-.54). The correlations between demands and skill discretion were rather high in the two national representative samples, but substantially lower in the two specific samples: salesmen and the PRIM sample. Social support was correlated with decision authority and skill discretion in the two specific samples but not in the two national representative samples.

Many researchers combine the two scales: decision authority and skill discretion, into a higher order control scale. We found that this combined scale was inappropriate, as it would cause loss of important information. In our analyses the correlations of demands with the two other scales were dissimilar in both size and direction. A combination of the two scales in a control scale would conceal some important differences of the internal structure of the psychosocial work environment in various sub-populations.

CORRESPONDING AUTHOR: Vilhelm Borg, National Institute of Occupational Health, Lerso Parkalle 105, DK-2100 Copenhagen O, Denmark.