Chair: Harry Shannon, Ph.D., McMaster University; and Senior Scientist, Institute for Work & Health; Presenters: Christel A. Woodward, Ph.D., McMaster University; Judy A. Brown, M.A., McMaster University; Jane A. Ferrie, M.Sc., University College, London; Stephen A. Stansfeld, M.D., University College, London.

In Western societies, the public sector has been downsizing over the last several years, following similar major changes in the private sector. While job loss was earlier limited to blue collar workers, it has increasingly been happening to white collar workers. In the last 10 years, over 3 million U.S. white collar workers have lost their jobs. Many of those laid off find other employment, but often at lower pay, with less security or with poorer conditions.

National surveys in Europe and North America find workers reporting markedly increased levels of job insecurity. Increasing numbers of hours of work have the potential to strain work-home relations, especially in women. As well, anticipation of job loss has been found to adversely affect several health outcomes. Job insecurity has thus become of increasing interest in work, stress and health research.

Various studies have been conducted examining the relationship between

unemployment and health. Although they do not universally show poorer health in those without jobs, the weight of evidence is consistent with an adverse effect of unemployment. At the same time, the effect of loss of colleagues on "survivors" - those left after downsizing - should not be ignored.

The studies reported in this Symposium are longitudinal in nature. The Whitehall II study, of more than 10,000 mainly white collar male and female British government employees, began in 1985 and has collected four waves of data. Over this period, changes have included privatization and downsizing, providing a remarkable opportunity to examine these issues while allowing for the effects of health selection and health-related individual characteristics.

The second study was conducted in a large Canadian teaching hospital undergoing reengineering, an approach that included downsizing. A sample of the workforce (all employees, not just nurses) was surveyed in three successive years, with a range of measures included on all occasions.

These papers provide important data on the effects of work changes, notably downsizing, on both those who stay with the organization and those who leave. The longitudinal nature of the studies is a major methodological strength. The findings raise critical questions about whether a full accounting of the costs of the changes will indeed show a net benefit to society .

CORRESPONDING AUTHOR: Harry S. Shannon, Ph.D., Department of Clinical Epidemiology & Biostatistics, McMaster University, Hamilton, ON, Canada, L8N 3Z5

Woodward CA,¹ Cunningham C,^1,2 Shannon HS,^1,3 McIntosh J,^1,2 Lendrum B,² Brown JA¹. ¹Department of Clinical Epidemiology and Biostatistics, McMaster University, ²Hamilton Hospital Corporation, ³Institute for Work and Health.

Rationale: In Canada, recession during the early 1990s led to expenditure reductions in the public sector. Severe reductions in hospital operating budgets over a short period led to significant changes in hospitals’ operations and staff reductions. This paper explores the impact of such rapid work environment changes, and feelings of job insecurity generated by them, on the psychological well-being of staff at a Canadian teaching hospital and examines predictors of psychological distress in 1997.

Procedure: We report on survey responses from a 21% random sample of employees in 1995 and in 1997. The questionnaire asked about: personal characteristics, emotional functioning, personal coping resources, job characteristics, and job interference with family life and family interference with job. The emotional exhaustion, depression and anxiety measures were combined to form a psychological distress score, the dependent variable. Factors contributing to distress in 1997 were examined using backwards stepwise regression.

Results: Psychological distress increased significantly (paired t=8.26, p<0.001). More than 60% of the variance in 1997 psychological distress scores could be explained using initial and change scores. Job characteristics, particularly changes in them during the two years, were explanatory. Greater job insecurity, longer job hours, more unclear job role and their change (increase) over time, along with initial level of psychological distress in 1995, were positive predictors of psychological distress. Low initial job strain and increasing job strain over time were linked with distress. Lower initial supervisor support and reduction in co-worker support and in job influence predicted higher distress in 1997. Demographic variables were not important. Adding job interference with home life to the model further improved its explanatory power.

Conclusion: Many of the antecedents of psychological distress among these hospital workers related to their jobs and changes in them. Among this largely female (88%) workforce, family interference with their jobs was not related to increasing distress over time. Rather, having a job that interfered (increasingly) with family life was an important predictor of later psychological distress, although this may be both a consequence of the work situation and cause of distress.

CORRESPONDING AUTHOR: Dr. C. Woodward, Ph.D., Department of Clinical Epidemiology and Biostatistics, McMaster University, 1200 Main Street West, Hamilton, Ontario, Canada. L8N 3Z5

Brown JA,¹ Woodward CA,¹ Shannon HS,^1,2 Cunningham C,^1,3 Lendrum B,³ McIntosh J.^{1,3 1}Department of Clinical Epidemiology and Biostatistics, McMaster University, ²Institute for Work and Health, ³Hamilton Health Sciences Corporation.

Rationale: Hospital sector employees in Canada have, until very recently, enjoyed nearly full employment without concerns about job insecurity or loss. This situation changed rapidly in Ontario after 1995 when major reduction in hospital operating budgets occurred. This paper looks at workers who left a large Canadian teaching hospital and compares the perceptions and coping mechanisms of workers who lost their jobs with those workers who remained with the hospital.

Procedure: Among a 21% random sample of hospital employees, baseline information was used to compare those who left the hospital with employees who remained in 1997. The general and mental health consequences of leaving among those who retired, found another job or were still job seekers by 1997 are compared. Predictors of employment status in 1997 are examined.

Results: Older employees, nurses and clerical workers were more likely to have left the hospital by 1997. Supervisors were twice as likely to no longer be with the hospital as others. Before leaving, people who left were more emotionally exhausted and reported greater job role unclarity than those who remained.

Less than 30% of leavers had retired, 25% had taken voluntary severance packages. Others left involuntarily; only 17% left for new jobs. Those who took voluntary severance packages were the most likely to use the Career Action Centre (CAC), a hospital resource to assist workers to develop skills helpful to finding new employment. The likelihood of being employed in 1997 was related to CAC attendance Of those who attended it, 41% had found work while 59% of non-attenders had found work.

Of the people not retiring by 1997, 58% were employed; 32% were still looking for work and the remainder were not actively job hunting. Of those employed, 83% worked in the health care sector. The likelihood of employment in 1997 was not linked to reason for leaving the organization (retirees excluded). In 1997, psychological distress was higher among those seeking work than other groups of leavers. By 1997 both leavers and stayers reported greater psychological distress, there was no significant difference in distress between the two groups.

Conclusion: The human costs of downsizing should be considered as the public sector attempts to become more efficient. The psychological consequences are felt both by workers who remain and by those who leave.

CORRESPONDING AUTHOR: J.A. Brown, Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada. L8N 3Z5.

J.E. Ferrie MSc, M.S. Shipley MSc, M.G. Marmot PhD, S.A. Stansfeld PhD, University College London, UK and G. Davey Smith MD, University of Bristol, UK

An investigation of the health effects of job insecurity and loss of secure employment among white-collar workers was conducted through the Whitehall II study, a longitudinal cohort of white-collar civil servants. Baseline data, collected before restructuring and privatisation of the Civil Service were planned enabled the investigation to address the issue of selection.

Privatisation of the Property Services Agency (PSA), one of 20 departments in Whitehall II, provided the opportunity to study effects on health of all stages of a typical workplace closure, from secure employment through a period of increasing job insecurity to job loss and ensuing alternative employment or non-employment.

Four waves of data were analysed. Baseline data from Whitehall II, a period of secure employment for all respondents; Phase 2 data from Whitehall II, 2 years pre-privatisation for PSA respondents (anticipation phase); Phase 3 data from Whitehall II, 3 months pre-privatisation for PSA respondents (pre-termination phase); and data from the follow-up of PSA respondents 18 months after privatisation.

At baseline, PSA respondents had better self-reported health and lower cardiovascular risk factors than controls. However, by the anticipation phase this situation had changed and PSA employees experienced an overall deterioration in self-reported morbidity compared to the control group. By the time of sale, body mass index, ischaemia and cholesterol concentration had also increased significantly for both sexes combined and, among women only, a significant rise in blood pressure was recorded. During the anticipation and pre-termination phases this relative decline in health status could only be partially accounted for by changes in work characteristics or health-related behaviours.

18 months after the privatisation, women and men in secure employment enjoyed better self-reported health, less longstanding illness and less psychological disorder than those in any other group. Unemployed men and insecurely employed women had the poorest self-reported health status. Longstanding illness was significantly increased among the secure non-employed, as was psychological disorder in both sexes in insecure employment and among unemployment men. These changes could be only partially accounted for by changes in psychosocial factors and health-related behaviours.

CORRESPONDING AUTHOR: Jane E. Ferrie, MSc, Department of Epidemiology and Public Health, UCL, 1-19 Torrington Place, London WC1E 6BT

Stephen A Stansfeld PhD.,* Rebecca R Fuhrer PhD., Martin J Shipley PhD., Department of Epidemiology and Public Health, University College London, UK

There is now considerable evidence that the psychosocial environment at work has an impact on health. High levels of psychological demands and low social support at work have been prospectively related to higher levels of psychological disorder. On the other hand social support in domestic, non-work settings has been found to be protective of psychological health while close relationships with high levels of negative interaction have been associated with higher risk of psychological disorder. Research, largely carried out in women only, suggests that multiple roles, work and domestic, lead either to higher levels of well-being through role enhancement or conversely to lower levels of well-being through role overload. Another way of considering this may be that the quality of the experience within different roles may be the effective determinant of well-being. There has been less epidemiological research work which simultaneously considers both the influence of work and home environments on psychological health in both men and women.

We report findings from the Whitehall II Study, a longitudinal cohort study of 10,308 largely white-collar, male and female British goverment employees. This study, set up in 1985, when the participants were 35- 55 years, has been running for 13 years, and has 4 waves of completed data collection. We will present data from the third phase of data collection which includes the Karasek Job Content Instrument, measures of control at work and home and social support measured by the Close Persons Questionnaire and work - home interpenetration stressors to measure the spillover of work stressors to the home environment and vice versa. We found that high control at both work and home were related to less psychological disorder for both men and women. These findings were further illuminated when they were examined by socioeconomic status,measured by employment grade.For women there was a tendency for higher grade employees to show higher rates of psychological disorder, for those with high and low levels of control at home or work. However, for men, low control at work predicted psychological disorder in higher but not lower employment grade employees, whereas low control at home was related to slightly higher rates of psychological disorder in lower as opposed to higher grade employees. When investigating the joint impact of work and home stressors on psychological disorder it is important to consider that they will have differential effects depending on sex and socioeconomic status.

CORRESPONDING AUTHOR Stephen A. Stansfeld PhD, Department of Epidemiology and Public Health, University College London, 1-19 Torrington Place, London WC1E 6BT,UK

Chair: Millie DesBiens, IBM Corporation, NY. Presenters: Michele Heist, Work/Family Directions, Boston, MA; J.T. (Ted) Childs, Jr., Global Wokforce Diversity, IBM Corporation, NY; Patricia D. Hammond, Employee Relations, J.P. Morgan, NY; Rich A. Vintigni, Work-Life/Human Resources, DuPont Company, Wilmington, DE

The first presentation, The Seven Drivers of Employee Commitment, will be given by Michele Heist of Work/Family Directions. Employee commitment has emerged as a significant competitive advantage and bottom line issue for CEOs, although companies have long struggled with how to accurately measure and improve it. Work/Family Directions (WFD) has developed a rigorous methodology that meets the growing need for analysis, depth and actionable strategies lacking in prototypical employee attitude surveys or national workforce opinion polls. By identifying seven major drivers of employee commitment that make up a company’s overall Commitment Profile (communication, diversity and inclusion, job satisfaction, flexibility, manager effectiveness, work-life support and career advancement), WFD assists companies in developing strategies to influence commitment in their organizations.

J.T. (Ted) Childs, Jr., Vice President of Global Workforce Diversity for IBM, will present Driving Toward Commitment – Work-Life Support and Career Advancement. This presentation will examine IBM’s efforts to attract and retain a diverse team of employees to win in a competitive global economy. In particular, IBM’s work/life initiatives ranging from work/life surveys to flexibility options, diversity programs ranging from a diversity brochure to Executive Task Forces, and partnerships with other companies and organizations to bring services to employees, retirees, and their families.

Driving Toward Diversity and Inclusion and Work-Life Support will be presented by Patricia D. Hammond, Vice President of Employee Relations for J.P. Morgan. This presentation will focus on J.P. Morgan’s use of work/life tools in managing issues related to the attraction, retention and development of its employees globally. Specifically, the presentation will examine how global issues such as the 24 hour work day, the instability of the global economy, and the expectations of ex-patriates and local employees in global locations impact the design and communication of work/life tools. This session will examine the need to stay the course in the delivery of these tools in good times and in bad and how to get senior management to see and feel the power of using work/life tools in solving business problems.

Finally, Rich A. Vintigni of DuPont Company’s Work-Life Resources will present Driving Toward Manager Effectiveness, Flexibility and Work-Life Support. This presentation will examine DuPont’s efforts to provide work-life supports for employees. In particular, the 1995 Work/Life Survey to determine the effectiveness of these programs, which were implemented over a ten-year period will be discussed. The results of this survey, along with other research, link the use and awareness of such services to employee satisfaction and commitment. Steps taken to communicate these results and to foster the use of the program as business tools will also be discussed, as will DuPont’s efforts to globalize their work-life strategy.

Chair: James Grosch, PhD, NIOSH; Linda Martin, PhD, NIOSH. Presenters: Bradley Doebbeling, MD, MSc, University of Iowa; June Fisher, MD, University of California at San Francisco; Robyn Gershon, DrPH, MHS, Johns Hopkins University; Lynn Kim, MPH, Washington University. Discussant: Lawrence Murphy, PhD, NIOSH.

A major occupational hazard facing over 11 million health care workers in the U.S. is accidental exposure to blood-borne pathogens, such as human immunodeficiency virus (HIV) and hepatitis B. Early attempts to reduce exposures focused on the development of engineering controls (e.g., sharps containers) and reliance on government backed guidelines, called Universal Precautions (UP), that prescribed safe work practices (e.g., not recapping needles, wearing gloves, etc.) for health care workers to follow. These approaches, while having some success, led to an examination of other factors that might play an important role in reducing the risk of occupational exposure. Recent research suggests that job and organizational-level factors (e.g., safety climate) can have a powerful influence on the safe work practices of health care workers.

This symposium focuses on four prevention projects sponsored by CDC/NIOSH to examine job and organizational-level interventions to reduce occupational exposure of health care workers to blood-borne pathogens. The interventions include (a) the development of employee education programs, (b) formation of total quality improvement groups, (c) improved feedback to workers about safe work behavior, and (d) ergonomic changes in the work environment to reduce exposure. Results of these projects indicate the powerful role of work organization factors in occupational health and safety programs, and provide insight into factors that influence the success of organizational intervention programs.

CORRESPONDING AUTHOR: Lawrence Murphy, PhD, NIOSH, 4676 Columbia Pkwy., MS-C24, Cincinnati, OH 45226-1998, USA.

Bradley N. Doebbeling, MD, MSc; Susan E. Beekmann, RN, MPH; Kimberly D. McCoy, MS; Thomas E. Vaughn, PhD; Kristi J. Ferguson, PhD; Robert F. Woolson, PhD; James C. Torner, PhD, The University of Iowa College of Medicine and Iowa City VAMC.

We developed a community hospital-based intervention program designed to optimally protect workers from occupational exposure to blood and body fluids. In order to inform these studies, infection control practitioners (ICPs) in all Iowa and Virginia hospitals and a statewide stratified, random sample of 5,364 Iowa health care workers at risk for occupational exposure to blood were surveyed. Institutional organizational and program factors associated with lower rates of exposure to blood were examined. Initial analyses suggested that institutional safety climate, particularly job demands and feedback, was related to occupational exposure among workers.

Based on these results, an intervention program in three community hospitals was developed that emphasized joint management, including employee focus groups, goal setting, training to monitor standard precautions (SP) compliance, and provision of compliance feedback. The study is a randomized, controlled trial of institutional-level interventions using a pre- post-test design. Participating hospitals were randomly assigned to either: 1) feedback at the unit level, or 2) feedback at the institutional level. ICPs and co-workers were trained to observe and monitor SP compliance in a one-day session at each facility. Feedback has been provided on a monthly basis to ICPs via graphical reports. Collection of baseline survey data, surrogate compliance measures, and observed compliance data has been completed or is ongoing at each site.

Most workers (80%) were aware of the availability of "safer" devices in their facilities. Approximately one-fourth (range 0-38%) of workers had sustained a percutaneous injury in the prior 3 months. Observed compliance with key aspects of SP at the facilities ranged from 65-87% (N=1,662 observations). Mean observed rates of overall SP compliance have ranged from 61% for Lab and Respiratory Therapy Techs to 89-100% for Housekeepers and Other Techs. Observed glove use is excellent overall, although mask (90%), gown (85%), and eye protection (90%) use is less consistent.

Follow-up surveys of workers at each facility will be collected approximately 12 months following development of the intervention program and collection of the baseline survey. Outcomes of primary interest include estimation of reported rates of compliance with specifically recommended SP measures, plans to use specific components in the future (Stages of Change), and rates of occupational injury and exposure.

CORRESPONDING AUTHOR: Bradley N. Doebbeling, MD, MSc, Department of Medicine, University of Iowa College of Medicine, SE 625 GH, 200 Hawkins Drive, Iowa City, IA 52242, USA.

June M. Fisher, MD, Trauma Foundation, San Francisco General Hospital and School of Medicine, University of California, San Francisco; Anna Bonner, BSE, Trauma Foundation, San Francisco General Hospital; Robbie Welling, MS, MPH, Trauma Foundation, San Francisco General Hospital; Mary Foley, RN, MSN, St. Francis Hospital.

The Training for Development of Innovative Control Technology (TDICT) project is a NIOSH funded project which promotes the development and use of control technology to prevent health care worker exposure to blood. It is a collaborative effort of health care workers (HCW), industrial hygienists (IH), and product designers (PD). The underlying premise of the project is that workers are the experts regarding their work and that this expertise is essential when designing the tools to be used in work.

The collaboration has included: 1. focus groups with line HCWs; 2. review of injury data with HCWs; 3. mentoring of IHs and PDs by HCWs; 3. observations at the worksite by IHs and PDs; 4. brainstorming sessions between all three groups to identify needs and solutions; 5. joint development of criteria for safety devices; 6. testing by and feed back from broad groups of workers in regard to the universality of criteria. Criteria for ten different categories of safety devices to prevent exposure to blood have been developed. These were the first written criteria for these devices and are now widely used by hospitals, regulatory agencies, and labor unions for promoting the involvement of line HCWs in the systematic evaluation and selection of these devices. They are also being used by manufacturers as benchmarks for the newer generation of devices. In addition to the above activities, TDICT has developed a course in product design for HCWs. The purpose of this course is to promote the role of HCWs in the process of design and development, as well an in the evaluation and selection, of devices.

The HCWs have made major contributions to the success of TDICT activities and the development of better engineering controls to prevent occupational exposure to blood. Though not formally evaluated, the HCWs who have participated in TDICT activities report that they value not only the skills but the confidence and perspective they gained in regard to the importance of their participation in improving their work environment. We believe that the collaborative model that has been developed can be applied to all areas of work. We are currently developing a participatory ergonomics and design project for transit workers based on this model.

CORRESPONDING AUTHOR: June M. Fisher, MD, School of Medicine, University of California at San Francisco, San Francisco, CA 94143-0408, USA.

Robyn R.M. Gershon, MHS, DrPH; Christine Karkashian, MA; Antonio Escamilla, MD, MS; Patricia Flanagan, BA, School of Public Health, Johns Hopkins University.

Many different approaches have been taken in order to help reduce the risk of blood-borne pathogen exposure in the health care setting. One approach that was recently evaluated involved an intervention which applied total quality management techniques to the problem.

Teams of front line workers and managers were formed at a 1000 bed, tertiary care, urban health care facility. The teams were facilitated by experienced team leaders and assisted by a team of researchers who collected, analyzed, and prepared data for the teams. Researchers and team members worked together, guided by a participatory action research (PAR) framework. The teams had one goal, "to identify risk reduction strategies to reduce exposures". Over a period of six months, the teams developed a set of recommendations, these were: (1) senior managers should have zero tolerance for blood-borne exposures, (2) to have all staff and managers accountable for compliance with safe work practices, (3) To form a blood-borne pathogens workgroup, which would be responsible for all aspects of the blood-borne pathogens program, including the implementation of all team recommendations, and (4) the development of a new blood-borne pathogens training and communications program.

The recommendations of the group were then subject to a feasibility analysis, which examined cost effectiveness and other criteria. Based on these findings, the recommendations were prioritized and then implemented.

The evaluation of the process was accomplished in several ways. First

pre and post-test interventional surveys of safety climate and self-reported work practices were evaluated using a brief, confidential questionnaire.

Second, objective measures of needlestick rates, and other blood-borne exposures were examined over the entire study period.

The results of this study indicate that the incorporation of total quality techniques as an interventional tool for reducing exposure was an effective approach to the complex, organizational problem of blood-borne exposures.

CORRESPONDING AUTHOR: Robyn R.M. Gershon, MHS, DrPH, Department of Environmental Health Sciences, School of Public Health, 615 North Wolfe Street, Room 8503, Baltimore, MD 21205, USA.

Lynn Kim, MPH; Rodney Parks, MA; Donna Jeffe, PhD; Bradley Evanoff, MD, MPH; Brad Freeman, MD; Victoria Fraser, MD, Washington University School of Medicine.

Healthcare workers (HCW) are at risk of occupational exposure to blood-borne pathogens such as human immunodeficiency virus (HIV), hepatitis B virus (HBV) and hepatitis C virus (HCV). Occupational injury rates increase during extended surgical operations (>3hours), procedures with moderate-heavy blood loss (>300cc), and emergent procedures. We designed an educational intervention to decrease occupational exposures to blood-borne pathogens among HCWs in the emergency department (ED) and operating room (OR). The educational intervention consisted of lectures, posters, and a hands-on clinical procedure-training program for medical students and surgical residents. Participants included OR personnel in cardiothoracic, general, gynecological and orthopedic surgery, and ED personnel at Barnes-Jewish Hospital, a 1,000-bed tertiary care hospital. Before and one year after the educational intervention, observers recorded information about personal protective equipment (PPE), the nature of sharps transfers, and risk taking behaviors among OR staff and ED staff.

The use of PPE in the OR increased significantly after the educational intervention. In the OR, trained observers recorded 597 HCWs during 76 cases (200 hours) at baseline and 783 HCWs during 116 cases (322 hours) following the intervention. Surgical personnel used proper eye protection in the OR more frequently following the intervention [322/597 (54%) in 1996 vs. 516/783 (66%) in 1997, p<.001]. Double gloving increased among all HCWs following training [97/344 (28%) in 1996 vs. 250/477 (52%) in 1997, p<.001]. Sharps passages were more likely to be announced [53/575 (9%) in 1996 vs. 199/970 (21%) in 1997, p<.001]. Most importantly, the observed needlestick and body fluid exposure rate decreased from 8.5/100 hours in 1996 to 4.7/100 hours in 1997, (p<.091).

In the ED, staff were more compliant with UP following the education intervention. In the ED, trained observers recorded 752 HCWs during 88 trauma cases at baseline and 811 HCWs during 119 cases following the intervention. At baseline 335/752 (45%) of ED staff observed during trauma resuscitations were fully compliant with PPE use; following the educational intervention 542/811 (67%) were fully compliant (p<.001). At baseline, 202/304 (66%) of HCW performing or assisting an invasive procedure were compliant with PPE; following education 280/357 (78%) were compliant (p<.001). This educational intervention significantly improved compliance with UP. OR and ED personnel in this study showed increased use of protective eyewear, double gloves, and experienced fewer occupational exposures.

CORRESPONDING AUTHOR: Victoria Fraser, MD, Washington University School of Medicine, St. Louis, MO 63110, USA.