The University of California, San Francisco (UCSF) responded to space constraints on its original campus—similar to those faced by many urban campuses—by developing a second major campus. On a 60-plus-acre site in the Mission Bay neighborhood of San Francisco, UCSF constructed a new women’s, children’s, and cancer hospital complex, which opened in 2015 and generated demand for proximal academic space for desktop research and education.
To accommodate this and other space needs, the UCSF administration decided to build an office building nearby. Mission Hall was initially planned to provide 1,200 workspaces, which increased to 1,500 during design. An activity-based workplace (ABW), one of several designs considered, was selected because it could accommodate the increased number of workspaces within the previously established budget of approximately $100 million.
Proponents of open workspaces argue that in addition to being cheaper, such designs enhance communication, job satisfaction, and work performance by removing barriers (walls, doors, hallways). An ABW design leverages shared and open workspaces with the intent of improving efficiency and increasing collaboration among occupants. Occupants are assigned workstations in an open space and can access a variety of unassigned closed spaces as needed for specific tasks (e.g., small rooms for phone calls, larger rooms for meetings). Theoretically, the ability to work in varied spaces across the workday and to choose the environment that best supports the task enhances work performance. Other expected advantages of ABWs include greater flexibility in space assignments, more opportunities for interaction, and greater equity in access to space.
Studies of ABWs, and of open plans more generally, have mixed findings. Some have found benefits such as increased collaboration and knowledge sharing.1–3 More often, however, they report adverse effects, including greater feelings of dissatisfaction and lack of control; difficulties concentrating; impaired cognitive processing; lower morale and productivity; poorer health and well-being; and increased rates of sickness, absences, and absenteeism.4–11 Many workers in open-plan offices express frustration with distractions. These frustrations are often linked to a lack of auditory and visual privacy and impaired work performance.11–19 The most significant complaints of 42,764 respondents in 303 open-plan office buildings19 were the inability to have conversations without neighbors overhearing and distraction from overheard conversations.
Noise caused by irrelevant but audible and intelligible speech appears to be especially problematic. Such distractions can impair cognitive performance.4 Some workers use headphones and music to block out office noise, but this may diminish mental acuity.4
Research findings such as those noted above, as well as articles in the popular press,20–22 alarmed many UCSF faculty—especially those slated to move to Mission Hall. Beyond the new building, the ABW design was proposed as the template for future UCSF desktop research spaces. The UCSF administration viewed Mission Hall as an opportunity to define a new approach to workplace design, improving connections between occupants and managing the costs of real estate. The UCSF Academic Senate noted the absence of examples of buildings with exclusively ABW design for faculty in academic health centers and requested an evaluation of Mission Hall. The administration agreed that a formal study would be important for its own planning and could inform other academic health centers grappling with the cost of new facilities. In this article, we report the results of the Mission Hall evaluation, consider the implications for academic health centers, and share lessons learned and recommendations.
Design Elements of Mission Hall
Mission Hall was designed to accommodate research and clinical faculty and staff. Each occupant, from staff to department chairs, was assigned an approximately 40-square-foot individual workstation with a nominal 42-inch-tall partition around 3 sides. These workstations were grouped by program and arrayed in rows. Occupants could access unreserved 60–70-square-foot “focus rooms” (1 for every 4 occupants) for telephone conversations, one-on-one discussions, and activities requiring freedom from distraction and “huddle rooms” of varying sizes (1 for every 20 occupants) for small-group meetings and video conferences. Open breakout spaces for casual interactions and relaxation were located within clusters of workstations, and “town centers” with kitchen facilities provided social spaces for interconnected floors and programmatic units.
Occupants could also reserve small- and medium-sized conference rooms throughout the building. Large conference rooms and classrooms were clustered on the first 2 floors.
Mission Hall Evaluation
Between occupancy (beginning late 2014 and continuing into 2015) and evaluation (beginning in 2014 and continuing through 2016), UCSF used its workspace projects website (space.UCSF.edu) to provide ongoing updates about the building. It also established a working group to collect occupant feedback through facility service requests and other communications. There was no formal change management program in place.
Perkins + Will, an architectural and design firm that had no role in the design of Mission Hall, was selected to conduct the evaluation and advise UCSF on future projects. In 2014 and 2015, the firm conducted interviews with department chairs and focus groups of faculty and staff of units slated to relocate to Mission Hall, as well as structured site observations, plan analysis, and occupancy analysis.
In 2016, approximately 1 year after occupancy, a survey was administered to all Mission Hall occupants, including faculty, administrative and research staff, and postdoctoral fellows. While some survey findings for the complete sample are referenced, in this article, we focus primarily on the faculty response. In addition, postoccupancy focus groups and site observations provided triangulated comparisons, which reinforced the survey reports.
The UCSF institutional review board approved the research and determined it to be minimal risk.
Survey instrument and sample
The 100-item survey was designed by the 3 authors: a member of the architectural and design firm (J.B.), a university-based expert in architectural design and research (J.W.), and a UCSF faculty member with expertise in survey design (N.A.). The survey drew upon a postoccupancy survey developed and validated by the Center for the Built Environment (CBE) at the University of California, Berkeley. The CBE items assess factors affecting comfort in a given physical environment. Informed by interviews and focus groups with faculty and unit heads before Mission Hall occupancy, we modified the wording of some questions and added other questions specific to the new building.
We were unable to conduct a baseline, preoccupancy survey. Consequently, survey items asked respondents about perceived changes from pre- to postoccupancy, using Likert-type scales that ranged from 1 (much worse, much less satisfied) to 7 (much better, much more satisfied) with the midpoint of 4 indicating neutral or no change. As an example, one item asked “How has your current workplace affected your likelihood of recommending UCSF as a great place to work?” Response options were 1 = very unlikely, 2 = unlikely, 3 = a little unlikely, 4 = neither likely nor unlikely, 5 = a little likely, 6 = likely, and 7 = very likely. (The complete survey is provided as Supplemental Digital Appendix 1 at https://links.lww.com/ACADMED/A747.)
For ease of interpretation, responses were transformed during analysis to a −3 to +3 format, where 0 indicated no change, larger positive numbers indicated more positive change, and larger negative numbers indicated greater adverse change.
In January 2016, invitations to complete the survey were issued by chairs of Mission Hall–based departments (e.g., pediatrics, obstetrics–gynecology, epidemiology) or heads of units (e.g., Global Health, Clinical Translational Research Institute, Institute for Computational Sciences) to all faculty and staff in their unit whose primary work assignment was in Mission Hall. The invitation described the purpose of the survey, and it was followed by several reminders. Survey responses were anonymous and collected via the Qualtrics online survey platform (Qualtrics, Provo, Utah).
The survey was distributed to 258 faculty assigned to the building, and 175 responded (68% response rate). Of all 1,213 faculty and staff who received the survey, 625 responded (52% response rate). The majority of faculty respondents (86%; 139/161) reported overall dissatisfaction with their individual workspaces. Dissatisfaction was most commonly reported regarding noise (77%; 110/143) and inadequate visual privacy (89%; 130/146) and sound privacy (96%; 137/143).
Figure 1 presents faculty mean responses to questions about the impact on 4 domains of academic work: interaction with colleagues, work effectiveness, satisfaction and well-being, and institutional engagement. Faculty responses were negative across all domains. The extent of discontent varied by item and domain, suggesting that the responses did not simply reflect global discontent, but were related to specific aspects of the design. As shown in Figure 2, responses for the total sample were similar, although somewhat less negative than the faculty-only responses.
Interaction with colleagues.
A defining attribute of an ABW is its ability to facilitate access to colleagues and spaces for accomplishing group work. ABWs are believed to facilitate communication and interaction and to enhance group productivity. In our survey, faculty ratings regarding ease of accessing spaces to work with others (mean [SD] = .20 [1.61]) were relatively more positive than ratings for other aspects of the work environment. However, faculty rated other aspects of working collaboratively—coordinating tasks, solving problems with others, locating others when needed, feeling connected to people, and overall group productivity—as slightly worse in Mission Hall compared with their prior workplace. Means on these 5 items ranged from −0.59 to −0.88 (SD = 1.33 to 1.70).
Collaboration was made more difficult, in part, because faculty “voted with their feet” and reduced their time working in Mission Hall. In postoccupancy focus groups, clinical faculty indicated they sought space at the clinic or hospital to do their academic work. Overall, faculty respondents reported coming into the office less often than previously (reduced from 53% of their total work time to 35%) and spending more than twice as much time in their home office than before (increased from 10% of their total work time to 24%). Just under half (42%; 57/135) identified the lack of a closed office as the primary reason for this change. Many indicated needing to work elsewhere to be productive (29%; 49/169), and about three-quarters (73%; 122/169) reported working alternative hours more frequently.
Faculty respondents reported their productivity was adversely affected by the move to Mission Hall. More than three-quarters (76%; 110/145) indicated that the overall office layout interfered with their ability to get their work done, and 89% (142/160) felt that their individual workspace impeded their work effectiveness. Only 3% (5/169) indicated a positive change in their overall productivity. Faculty reported that their ability to concentrate at work was worse than it had been in their prior locale (mean [SD] = −2.08 [1.14]) as was their overall productivity (−1.80 [1.20]) and time to complete tasks (−1.74 [1.22]).
Satisfaction and well-being.
Mission Hall’s ABW design appeared to take a toll on individuals’ attitudes, emotions, and health. Faculty reported adverse changes in their sense of personal well-being, job satisfaction, feeling happy to come to work, and stress levels (mean [SD] = −1.29 to −1.55 [1.24 to 1.46]) compared with the time before moving. This extended to their evaluation of their overall health, which they reported to be a little worse than before the move (−0.75 [1.13]).
Relocation to Mission Hall appeared to weaken faculty commitment to UCSF. Faculty reported feeling less valued by the university (mean [SD] = −1.84 [1.31]), and such feelings were reported by 81% (133/164) of respondents. Faculty also indicated feeling less committed to staying at UCSF (−1.27 [1.28]), less likely to take on extra tasks (−1.13 [1.36]), less connected (−0.80 [1.58]), and less engaged in the work of the unit and campus (−0.79 [1.43]) and in their own work (−0.92 [1.22]). Only 10% (17/163) of faculty respondents report they would be likely to recommend UCSF as a great place to work.
Aspects rated favorably: Environmental conditions.
Not all aspects of Mission Hall were rated negatively. Mean ratings of specific environmental conditions were on the positive side. Faculty were satisfied with the amount of light overall (mean [SD] = 0.65 [1.46]), the amount of daylight (0.61 [1.77]), and access to a window (0.48 [2.07]). They were also satisfied with air quality (0.31 [1.25]). They were evenly split concerning temperature (0.00 [SD = 1.41]).
We note several limitations of the Mission Hall evaluation. Lack of a baseline survey prevented us from directly testing the extent to which attitudes, feelings, and behaviors changed when occupants moved to Mission Hall. Rather, we relied on respondents’ reports of changes in these domains, and we corroborated survey responses with data from interviews, focus groups, and observations.
This article focuses primarily on faculty experiences. A complete analysis of the effects of the ABW design should also consider staff responses. Overall, faculty were more dissatisfied with Mission Hall than were staff. This may be because Mission Hall’s design constituted a greater change for the faculty. While many staff were previously working in open spaces, most faculty moved from a private office to the ABW space.
Faculty who responded to the survey may differ from those who did not. The 68% faculty response rate is better than in most surveys and was higher than the staff response rate. Bias in terms of which faculty responded could have occurred. If disgruntled faculty were more likely to respond, the negative impact of the ABW design would be overestimated. However, the negative impact would be underestimated if demoralized or disengaged faculty were less likely to respond.
Finally, we cannot disentangle the extent to which the experiences reported by faculty reflected problems of Mission Hall’s ABW design itself or in the way it was implemented. The decision to place 25% more people in the building than originally planned, with no additional funding to expand the building area, created greater density and fewer options for smaller groupings or spaces dedicated to individuals than was first imagined. Inadequate soundproofing between spaces for focused work and spaces for group activities left occupants with no acoustically private space. There was also insufficient supportive technology for successful use of the ABW design. Equipment and capability, such as Bluetooth-enabled (or equivalent) phone capability, are needed to allow seamless transfer of computer-based work from one setting to another. These technologies were not widely available, however, and technology retrofit was left to individual departments without allocated funds. Nonetheless, although these issues could have affected faculty survey responses, our findings are similar to those of other studies of the impact of open-space designs, including ABW.
Considerations for Academic Health Centers
The Mission Hall evaluation provides insights into issues that should be addressed before adopting open-space designs for academic health sciences buildings.
Proponents of ABWs, including manufacturers of open-space components, highlight the lower cost of construction per occupant compared with traditional construction. These savings may be illusory, however, if the design results in lower productivity, engagement, and retention. A 6-year study of 40 business units involving 13,000 people estimated that only 5% of total 10-year costs for newly constructed buildings will derive from the expense of building the physical workplace.23 In contrast, 82% of the costs will be generated by expenses associated with the people occupying the space, with the remainder reflecting technology and operations expenses. Cumulative financial losses resulting from even small decrements in occupants’ work effectiveness and increased turnover may outweigh the savings initially accrued from lower construction costs. Further, the costs associated with retrofitting buildings that fail to serve their intended populations can create a more adverse ratio of cumulative costs to initial savings.
Mission Hall’s ABW design underestimated the extent of problems that would derive from visual and aural distraction. Insufficient attention was paid to the type of environment needed for accomplishing concentrated work such as writing grant proposals and papers, preparing teaching materials, and engaging in patient-related activities. The cumulative effect of searching for available closed space for calls and private conversations was not anticipated. Faculty reported considerable time loss from migration over the course of a day, and this movement may also have distracted others.
Reactions to lack of privacy had the unfortunate result of constraining interactions, including interactions that could foster greater collaboration. The environment became dominated by library behavior: Talking of any type was discouraged, and “quiet” signs were posted even in collaborative spaces.
Faculty who were unable to function effectively in the open space sought other places to work. While this lessened noise and distraction, it undermined the advantages of an ABW for collaborative work and weakened faculty engagement with the university.
Lessons Learned and Recommendations
Recognizing the problems in Mission Hall, UCSF retrofitted areas in the building to provide some faculty with small private offices while continuing to provide access to shared spaces as in the ABW design. Subsequent UCSF office buildings are being planned with this hybrid model, and a task force is developing principles for programming, designing, governing, and occupying open-space workplace environments.
Based on the Mission Hall experience, we urge caution by academic health centers considering use of an ABW design for workspaces. While customized variants of the ABW model may help achieve greater collaboration and create a stronger community, one-size-fits-all plans are unlikely to succeed.
For an open-space office to facilitate rather than impair faculty productivity and well-being, extensive work should be done in advance to address the factors that will determine its success. Questions to address in the planning period include whether the design will support the range of activities engaged in by faculty that require privacy and quiet—to different degrees and at different times. The design should be sensitive to the cumulative impact of the space and incorporate ways to diminish the transaction costs of moving among work settings. Particular attention needs to be paid to visual and auditory privacy and to availability of appropriate technology.
A key issue in planning an ABW is determining the appropriate density of open workstations versus closed spaces in light of the work that will be done in those spaces. In some cases, density can improve work performance,24 but in others, it can create problems with visual privacy and acoustic control and interfere with creating a collaborative environment.
Although important, density is not the sole determinant of the impact of an ABW environment on occupants. In Mission Hall, the allocation of space to individual open workstations versus enclosed spaces was within the bounds recommended for an ABW environment.25 The successful use of space depends on multiple design factors, however. In the case of Mission Hall, design problems included a complete absence of enclosed offices and inadequate privacy of both individual and group workspaces.
Other issues that need to be considered involve: (1) establishing a circulation hierarchy to reduce disruptions within neighborhoods of workspaces, (2) isolating communal spaces from quiet work areas to ensure that each may operate as intended, (3) providing the construction detailing necessary to contain noise within nominally private areas such as focus and huddle rooms, (4) supporting the change management that new workplaces require, and (5) aligning work functions and workspace assignments to group similar activities and separate those that are dissimilar and incompatible. To the extent that functional alignment may supersede department alignment, addressing the last issue may be especially challenging.
Open-space designs such as the ABW are attractive because of their potential for lower costs, greater flexibility, and more collaboration. The experience of Mission Hall points to the need to balance these benefits against potential adverse individual and institutional outcomes. The considerations will depend on organizational culture and context as well as the design execution. The Mission Hall evaluation, which reflects UCSF’s commitment to research on the design and implementation of new workplace models, provides a helpful first step in identifying academic workplace performance criteria and finding effective solutions for addressing the needs of the academic workplace.
The authors are grateful for the data analysis provided by Joo Young Ro, formerly of Perkins + Will, and Dr. Yongha Hwang of University of Michigan and for the administration of the survey by Dr. Lindsay Graham of the Center for the Built Environment at the University of California, Berkeley.
1. Batenburg RS, van der Voordt TJM. Do facilities matter? 2008; Paper presented at: European Facility Management Conference; June 10–11, Manchester, UK.
2. De Been I, Beijer M. The influence of office type on satisfaction and perceived productivity support. J Facil Manag. 2014;12:142–157.
3. van der Voordt TJM, De Been I, Maarleveld M. Finch E. Post-occupancy evaluation of facilities change. In: Facilities Change Management. 2011:West Sussex, UK: Wiley-Blackwell; 137–154.
4. Perham N, Hodgetts H, Banbury S. Mental arithmetic and non-speech office noise: An exploration of interference-by-content. Noise Health. 2013;15:73–78.
5. Banbury SP, Berry DC. Office noise and employee concentration: Identifying causes of disruption and potential improvements. Ergonomics. 2005;48:25–37.
6. Haka M, Haapakangas A, Keränen J, Hakala J, Keskinen E, Hongisto V. Performance effects and subjective disturbance of speech in acoustically different office types—A laboratory experiment. Indoor Air. 2009;19:454–467.
7. Smith-Jackson TL, Klein KW. Open-plan offices: Task performance and mental workload. J Environ Psychol. 2009;29:279–289.
8. Virjonen P, Keränen J, Helenius R, Hakala J, Hongisto OV. Speech privacy between neighboring workstations in an open office—A laboratory study. Acta Acust United Acust. 2007;93:771–782.
9. Frontczak M, Schiavon S, Goins J, Arens E, Zhang H, Wargocki P. Quantitative relationships between occupant satisfaction and satisfaction aspects of indoor environmental quality and building design. Indoor Air. 2012;22:119–131.
10. Pejtersen JH, Feveile H, Christensen KB, Burr H. Sickness absence associated with shared and open-plan offices—A national cross sectional questionnaire survey. Scand J Work Environ Health. 2011;37:376–382.
11. Mosselman N, Gosselink A, Beijer M. Long-term effects of activity based working. The Center for People and Buildings. https://www.cfpb.nl/media/uploads/publicaties/upload/Long_term_effects_activity_based_working.pdf
. Published September 2010. Accessed July 9, 2019.
12. Appel-Meulenbroek R, Groenen P, Janssen I. An end-user’s perspective on activity-based office concepts. J Corp Real Estate. 2011;13:122–135.
13. Gorgievski MJ, Bakker AB, Schaufeli WB. Work engagement and workaholism: Comparing the self-employed and salaried employees. J Posit Psychol. 2010;5:83–96.
14. Maarleveld M, De Been I. The influence of the workplace on perceived productivity. 2011; In: Proceedings From the 10th European Facility Management Research Symposium; May 24–25, Vienna, Austria.
15. van der Voordt TJM, van der Klooster W. Post-occupancy evaluation of a new office concept in an educational setting. 2008; Paper presented at: CIB W70 International Conference in Facilities Management; June 16–18, Edinburgh, Scotland.
16. Danielsson CB, Bodin L. Difference in satisfaction with office environment among employees in different office types. J Archit Plan Res. 2009;26:241–257.
17. Davis MC, Leach DJ, Clegg CW. Hodgkinson GP, Ford JK. The physical environment of the office: Contemporary and emerging issues. In: International Review of Industrial and Organizational Psychology. 2011:Vol 26. New York, NY: John Wiley & Sons; 193–237.
18. De Croon EM, Sluiter JK, Kuijer PP, Frings-Dresen MH. The effect of office concepts on worker health and performance: A systematic review of the literature. Ergonomics. 2005;48:119–134.
19. Kim J, de Dear R. Workspace satisfaction: The privacy-communication trade-off in open-plan offices. J Environ Psychol. 2013;36:18–26.
20. Konnikova M. The open office trap. The New Yorker. https://www.newyorker.com/business/currency/the-open-office-trap
. Published January 7,2014. Accessed July 9, 2019.
21. James G. Open-plan offices kill productivity, according to science. Inc. https://www.inc.com/geoffrey-james/science-just-proved-that-open-plan-offices-destroy-productivity.html
. Published May 18,2017. Accessed July 9, 2019.
22. Shellenbarger S. Why you can’t concentrate at work. Wall Street Journal. https://www.wsj.com/articles/why-you-cant-concentrate-at-work-1494342840
. Published May 9,2017. Accessed July 9, 2019.
23. Brill M, Weidenman S. Disproving Widespread Workplace Myths. 2001.Buffalo, NY: Kimball International.
24. Olson JS, Teasley S, Covi L, Olson G. Hinds P, Kiesler S. The (currently) unique advantages of collocated work. In: Distributed Work. 2002:Cambridge, MA: The MIT Press; 113–135.
25. Keable EB, Manley A, McGregor L. How dense is TOO dense? Work Design Magazine. https://workdesign.com/2018/01/how-dense-is-too-dense
. Published January 12,2018. Accessed September 9, 2018.