Doctoral Student

Hummy Song

Hummy Song is a PhD candidate in Health Policy Management at the Harvard Graduate School of Arts and Sciences and Harvard Business School. Her research focuses on health care operations, coordination, and the productivity of the health care workforce. She is currently working on projects that examine various operational and organizational factors that may affect worker productivity and quality of care. Her research utilizes large datasets derived from electronic medical record systems, hospital inpatient databases, and surveys of the health care workforce. 

Hummy has worked with several hospitals and health care organizations both in the U.S. and in developing countries. Prior to her graduate studies, she worked with global health organizations on randomized control trials and observational studies. 

Hummy received an AB in History with a secondary field in Psychology from Harvard College. She holds an MPP from the Harvard Kennedy School of Government, where she focused on global health and development policy. Prior to her master’s program, she worked with the Campaign for Tobacco-Free Kids on international tobacco control policy development and implementation. She also worked as a policy analyst at The Century Foundation.  

Working Papers

  1. Cohort Turnover and Productivity: The July Phenomenon in Teaching Hospitals

    Robert S. Huckman, Hummy Song and Jason R. Barro

    We consider the impact of cohort turnover—the planned simultaneous exit of a large number of experienced employees and a similarly sized entry of new workers—on productivity in the context of teaching hospitals. Specifically, we examine the impact of the annual July turnover of residents in American teaching hospitals on levels of resource utilization and quality in teaching hospitals relative to a control group of non-teaching hospitals. We find that, despite the anticipated nature of the cohort turnover and the supervisory structures that exist in teaching hospitals, this annual cohort turnover results in increased resource utilization (i.e., longer length of hospital stay) for both minor and major teaching hospitals, and decreased quality (i.e., higher mortality rates) for major teaching hospitals. Particularly in major teaching hospitals, we find evidence of a gradual trend of decreasing performance that begins several months before the actual cohort turnover and may result from a transition of responsibilities at major teaching hospitals in anticipation of the cohort turnover.

    Keywords: Quality; Higher Education; Service Delivery; Groups and Teams; Performance Productivity; Retention; Health Care and Treatment; Resignation and Termination; Health Industry;

    Citation:

    Huckman, Robert S., Hummy Song, and Jason R. Barro. "Cohort Turnover and Productivity: The July Phenomenon in Teaching Hospitals." Harvard Business School Working Paper, No. 14-113, May 2014. View Details
  2. The Diseconomies of Queue Pooling: An Empirical Investigation of Emergency Department Length of Stay

    Hummy Song, Anita L. Tucker and Karen L. Murrell

    We conduct an empirical investigation of the impact of queue management on patients’ average wait time and length of stay (LOS). Using an Emergency Department’s (ED) patient-level data from 2007 to 2010, we find that patients’ average wait time and LOS are longer when physicians are assigned patients under a pooled queuing system with a fairness constraint compared to a dedicated queuing system with the same fairness constraint. Using a difference-in-differences approach, we find the dedicated queuing system is associated with a 17 percent decrease in average LOS and a nine percent decrease in average wait time relative to the control group—a 39-minute reduction in LOS and a four-minute reduction in wait time for an average patient of medium severity in this ED. Interviews and observations of physicians suggest that the improved performance stems from the physicians’ increased ownership over patients and resources that is afforded by a dedicated queuing system, which enables physicians to more actively manage the flow of patients into and out of ED beds. Our findings suggest that the benefits from improved flow management in a dedicated queuing system can be large enough to overcome the longer wait time predicted to arise from non-pooled queues. We conduct additional analyses to rule out alternate explanations for the reduced average wait time and LOS in the dedicated system, such as stinting and decreased quality of care. Our paper has implications for health care organizations and others seeking to reduce patient wait time and LOS without increasing costs.

    Keywords: pooling; queue management; strategic servers; social loafing; empirical operations; health care; fairness; Management Practices and Processes; Service Delivery; Health Care and Treatment; Health Industry;

    Citation:

    Song, Hummy, Anita L. Tucker, and Karen L. Murrell. "The Diseconomies of Queue Pooling: An Empirical Investigation of Emergency Department Length of Stay." Working Paper. (October 2014.) View Details

Journal Articles

  1. The Diseconomies of Queue Pooling: An Empirical Investigation of Emergency Department Length of Stay

    Hummy Song, Anita L. Tucker and Karen L. Murrell

    We conduct an empirical investigation of the impact of queue management on patients’ average wait time and length of stay (LOS). Using an Emergency Department’s (ED) patient-level data from 2007 to 2010, we find that patients’ average wait time and LOS are longer when physicians are assigned patients under a pooled queuing system with a fairness constraint compared to a dedicated queuing system with the same fairness constraint. Using a difference-in-differences approach, we find the dedicated queuing system is associated with a 17 percent decrease in average LOS and a nine percent decrease in average wait time relative to the control group—a 39-minute reduction in LOS and a four-minute reduction in wait time for an average patient of medium severity in this ED. Interviews and observations of physicians suggest that the improved performance stems from the physicians’ increased ownership over patients and resources that is afforded by a dedicated queuing system, which enables physicians to more actively manage the flow of patients into and out of ED beds. Our findings suggest that the benefits from improved flow management in a dedicated queuing system can be large enough to overcome the longer wait time predicted to arise from non-pooled queues. We conduct additional analyses to rule out alternate explanations for the reduced average wait time and LOS in the dedicated system, such as stinting and decreased quality of care. Our paper has implications for health care organizations and others seeking to reduce patient wait time and LOS without increasing costs.

    Keywords: pooling; fairness; queue management; strategic servers; empirical operations; health care; Service Operations; Service Delivery; Management Practices and Processes; Health Care and Treatment; Health Industry;

    Citation:

    Song, Hummy, Anita L. Tucker, and Karen L. Murrell. "The Diseconomies of Queue Pooling: An Empirical Investigation of Emergency Department Length of Stay." Management Science (forthcoming). View Details
  2. Development and Validation of the Primary Care Team Dynamics Survey

    Hummy Song, Alyna T. Chien, Josephine Fisher, Julia Martin, Antoinette S. Peters, Karen Hacker, Meredith B. Rosenthal and Sara J. Singer

    Objective. To develop and validate a survey instrument designed to measure team dynamics in primary care. Data Sources/Study Setting. We studied 1,080 physician and non-physician health care professionals working at 18 primary care practices participating in a learning collaborative aimed at improving team-based care. Study Design. We developed a conceptual model and administered a cross-sectional survey addressing team dynamics, and assessed reliability and discriminant validity of survey factors and the overall survey’s goodness-of-fit using structural equation modeling. Data Collection. We administered the survey between September 2012 and March 2013. Principal Findings. Overall response rate was 68% (732 respondents). Results support a seven-factor model of team dynamics, suggesting that conditions for team effectiveness, shared understanding, and three supportive processes are associated with acting and feeling like a team and, in turn, perceived team effectiveness. This model demonstrated adequate fit (goodness-of-fit index: 0.91), scale reliability (Cronbach’s alphas: 0.71-0.91), and discriminant validity (average factor correlations: 0.49). Conclusions. It is possible to measure primary care team dynamics reliably using a 29-item survey. This survey may be used in ambulatory settings to study teamwork and explore the effect of efforts to improve team-based care. Future studies should demonstrate the importance of team dynamics for markers of team effectiveness (e.g., work satisfaction, care quality, clinical outcomes).

    Keywords: Surveys; Primary Care; Team Dynamics; Team Effectiveness; Health; Groups and Teams; Surveys; Health Care and Treatment; Health Industry; United States;

    Citation:

    Song, Hummy, Alyna T. Chien, Josephine Fisher, Julia Martin, Antoinette S. Peters, Karen Hacker, Meredith B. Rosenthal, and Sara J. Singer. "Development and Validation of the Primary Care Team Dynamics Survey." Health Services Research (forthcoming). View Details