Who will be there to care for the patients?

Three Hospital Beds in a rowA research plan to provide an estimate of absenteeism in specific areas of the health-care workforce during an influenza pandemic.  Originally published on March 27, 2008.


OSHA has stated “a pandemic could affect as many as 40 percent of the workforce during periods of peak influenza illness.”[i]  This one fact runs through and is reiterated in the many volumes of information recently created to help various segments of a community prepare and respond to an influenza pandemic[ii].  This figure of is used as a planning assumption and justification to build capacity in all critical infrastructure areas.[iii]  Other research around the SARS epidemic has pushed the number even higher[iv].  This author believes that this figure is used carelessly with the implication that the 40 percent absenteeism rate applies equally to all fields of health care workers at all education and socioeconomic levels.

This paper identifies the need for additional research regarding how the health care workforce will react and respond to an influenza pandemic, and then outlines a plan to conduct the research.  In this context, the health care workforce will be divided into three major groupings:

  • Emergency medical services and first-responder health care providers,
  • Medically trained specialists operating in a hospital environment, and
  • Non-medically trained workers that support the hospital environment, such as maintenance and janitorial crews.

Research in the health care workforce is specifically needed to predict behavioral intent of the workers.  The research objectives are as follows:

  • Identify the current knowledge about an influenza pandemic,
  • Identify the workers’ perception of risk during an influenza pandemic,
  • Estimate the workers’ likelihood of working during an influenza pandemic, and
  • Identify correlations in workers knowledge and risk perception with the likelihood of continuing work during an influenza pandemic.

From this information, the researchers can pull together a more accurate estimate of the health-care workforce’s behavior during an influenza pandemic.  This will allow health-care administrations to more accurately predict and plan for their needs which allows for more specific plans as to what services will be provided and in what form.


In the paper entitled A Review of How a Hospital Could Use the Incident Command System for a Pandemic Flu Incident[v], this author explored the development of a hospital command system based on ICS.  One of the points raised is that the high rates of absenteeism caused by an influenza pandemic will require deep reserves in all critical functions and essential workers.  This is an unlikely situation for a hospital to achieve while balancing between providing effective services to patients and maintaining a profitable business structure.  The cost-benefit of these reserves may not be warranted in most situations.

A finer point is needed for planners to project which categories of workers will remain to continue critical functions.  Planners will then be able to better identify areas that need additional staff reserves using the results of the research.  In turn, this will provide more accurate staffing levels to maintain critical functions and pre-determine which functions may not be as critical in light of the staffing shortfalls.

The results of this research can be used in two ways.  First, the research can be used at a macro level by local, state and Federal planners to provide more useful information to health-care providers that are planning for an influenza pandemic.  Second, the research methods can be used at a micro level by an individual hospital to generate results specific to their workforce.  This is advantageous for an individual hospital as they do not need to spend time and money to structure the research; they can just utilize the same tools for custom results.

The Health-Care Workforce

The people that comprise the health-care workforce are a widely varied group in terms of education, social status and economic class.  This may also translate into a certain commitment to work or not during an influenza pandemic.  For some people, it is a position of status and a commitment to caring for patients.  For others, it may just be a paying job that was available.

The hospital administrator may be a highly-education and visible part of the community who understands the need for leadership to keep the hospital functioning through difficult times.  The medical doctor is highly educated and intellectually capable of weighing the potentially higher risk of infection by caring for infected patients with the various ways of reducing the risk of infection through personal protective techniques.  The volunteer ambulance driver may realize the increase exposures due to responding to medical calls for help and rationalize it that they will not see the person long enough to be infected.  The janitor that cleans the rooms and empties the trashcans is probably does not have the educational background to separate fact from fiction and may lack the occupational commitment to remain at the hospital if the perceived risk is higher. 

Any one of these people may be hesitant to share that they will not come to work during a pandemic for fear they may lose their positions now and be ridiculed by colleagues.  Individuals – including this author – make assumptions based on stereotypical behavior expected from each segment of the workforce about their actions.  Important questions that the worker may not have thought about yet need to be explored.  For instance, if they get infected, will they still come to work because they financially can’t afford to not work or have some feeling of importance that requires them to come to work even if sick.  What will they do if a household member is infected?  Do they need to stay home and take care of that person?  Since they are exposed, should they not come into work to avoid spreading it?  Have the workers though through these scenarios yet?

Research of the health care workforce is needed to find answers to these critical questions.  Accurately predicting the reactions and response to an influenza pandemic is necessary to validating plans.  Using the correlations drawn between worker’s knowledge and predictions of their actions, educational programs can then be developed in an effort to help workers separate fact from fiction and set expectations for how they will be taken care of by the hospital during an influenza pandemic.  These acts may help to increase the workforce that remains working at the hospital, which in turn will help the hospital have the necessary workforce to maintain operations.

Research Objectives

The objective of the research is to identify an aggregate of persons’ behavioral intent[vi] during a pandemic influenza in an effort to predict what certain categories of workers may choose to do.  From a health-care administrator’s perspective when creating the pandemic influenza plan, the primary question is who will be there to help staff and operate the hospital to help the patients.  Since most people have not been involved in an influenza pandemic, there is no direct real-life experience they can draw on to make a decision or past history to gauge.  This objective will be reached through a series of intermediate objectives, which are as follows.

Determine the workers’ knowledge and attitudes about an influenza pandemic.  This information may show if workers are misinformed about an influenza pandemic that may lead to poor decisions that either have no impact on the actual risk or may actually increase the risk.  Correlations that can be drawn between behavioral intent and misinformation may be corrected through educational programs to align the behavioral intent inline with the desired actions.  For instance, a hospital’s desired action may be for workers to continue to work unless they are infected, whereas a union’s desired action may be for the workers to reduce their risk which may include ceasing work.  Obviously, this research can be used in a number of ways.

Determine the workers’ perception of risk working at a hospital during an influenza pandemic.  It will be natural for worker to reduce their perceived risk regardless of actual risk.  This information may be compared to actual risk to measure if the workers are over or under estimating their personal risk to be infected.  The research may show that some workers believe that working in a hospital will afford them better opportunities for access to health care, prophylactic medicines and other protective measures.  Other workers may believe that being at a hospital will only increase their risk due to the potential influx of infected patients.  A certain amount of risk analysis to determine the actual risk will be necessary for comparison purposes.  Workers over estimating their risks will need education to understand why the risk is lower or actions they can take to lower the risk.  Workers under estimating their risks will need education to take the event seriously enough to ensure they do not unintentionally increase their risks.

Estimate the likelihood of the workers coming to work by measuring their expected actions to a series of scenarios.  Some of these may be detached from an influenza pandemic.  For instance by asking workers how they would take care of children if the schools are closed, such as in a snow day (or other bad weather day) can be a predictor what the worker would do during a influenza pandemic that might close schools.  Another scenario may be to ask workers how a household member would be cared for if they became seriously ill.

All this information would be analyzed to find relationships between the data and the three previously defined categories of health-care workers that will help predict workers’ behavior during an influenza pandemic.  Correlations that are draw between knowledge and behavioral intent can form the basis for an educational program.  The educational program can focus on teaching specific knowledge that may result in certain behaviors.  The design of this educational program is outside the scope of this research, however it is important to note that knowledge does not directly results in preferred behavior[vii],[viii].

Research Method: Data Collection

The research will be conducted through two types of anonymous interviews: structured and semi-structured[ix].  The data collection instrument for both of these interviews will be fashion systemically to allow for proper coding and analysis.  Prior to the first research complete using these tools should be reviewed by an independent evaluation analyst to prevent any unintended bias and unequal ratings.

The structured interview can be conducted in person, online or through other means as it is a short-answer style interview meant to quantitatively gather from a large cross-section of the sample population.  This will help reach the first two research objectives of identifying a workers knowledge and perceived risk.  See appendix A for the questions to be asked.  For a health-care system of 1,000 workers, a sample size of 516 workers would need to participate to have a 95% confidence level (+/- 3%).[x]

The semi-structured interview would be conducted in person to allow the interviewer to ask more probing questions and draw out answers from scenarios style questions.  These scenario-based questions will attempt to have the interviewee think through a series of scenarios while the interviewer notices certain cues to rank and order the perceived driving factors.  See appendix B for the questions to be asked.  For a health-care system of 1,000 workers, a sample size of 211 would need to participate to have a 95% confidence level (+/- 6%).  As these people will be asked the same questions as the structured interview, they can be a subset of the structured interview sample size.

Research Method: Data Analysis

The data collected from both types of interviews will be input into an evaluation and analysis tool.  Based on the information provided in the interviews, the data will be coded appropriately.  The analysis will occur in three stages.

The stage one analysis will review the structure interview responses to calculate the workers’ knowledge and perception of risk.  This information will be broken down by hospital position and education level.  Particular focus will be in finding those topic areas that are common to all responses either showing a lacking of knowledge or misperception of risk.   

The stage two analysis will review the semi-structured interview responses to capture the overall finding on how the works will respond to the scenarios that may occurring during an influenza pandemic.  This analysis will seek to find common responses to the questions.

The stage three analysis will utilize the analysis tool to find correlations between the different data points.  These correlations may drive hypothesis that certain health-care workers are more likely to continue working through an influenza pandemic then others.  This information will be fed back into the health-care plans to refine the plans.  Critical areas where workers are expected to continue working may be bulked up to handle additional responsibilities; whereas areas that can expect worker short-falls may need to develop contingency plans that account for insufficient staffing.

Research Strengths and Weaknesses

As with any narrowly focused research, there are inherent strengths and weaknesses that will affect the potential uses of the data.

A strength of this research is the well-defined and easily accessible population that allows for a proper sampling, if not polling a majority of the population in a specific facility.  Instead of other social research that covers a large geographic region, this can be conducted in a finite space on a per facility basis reducing the time necessary to complete it.  With the proper support of the management, time can be set aside for all workers to participate in the short interview process.  This will increase the confident level and interval of the survey.

A weakness of this research is the reliability of the responses.  Workers may have bias toward answering in the way they believe they are expected to respond.  They may feel pressured into responding a certain way know that management will see an aggregate of the responses.  Worker’s perceptions may be skewed regarding risk by working with diseases daily and perceiving an ability to not be infected.  With the research conducted at individual health-care facilities, the research would need to be repeated at an appropriate sampling of the many varieties of health-care facilities.  Moving the data from this research from the micro-optic of a single health-care facility to a national average will be difficult.

Appendix A: Structured Interview Questions

The following questions should be presented to the interviewee through an in-person, phone, computer-based or paper-based methods, and formatted appropriate to the interview.  The results will be kept anonymously.

  1. Which best describes your position in the hospital?
    management, administration, board certified doctor, resident, nursing staff, other medical position, maintenance, janitorial, food service, other support position
  2. What is your education level?
    some high school, high school graduate, two-year college degree, four-year college degree, master degree, doctorate degree, medical degree, other post-graduate degree.
  3. How recently have you heard about pandemic influenza (also called pan flu, avian flu, bird flu, and H5N1)?
    never, in the past year, in the past six months, in the past month, in the past week.
  4. Who is at risk from a pandemic influenza (check all that apply)?
    infants (0-3), children (4-12), teens (13-18), young adults (19-30), middle age (31-54), seniors (55+), chronically ill, immunocompromised.
  5. What are symptoms of influenza?
    dry cough, productive cough, vomiting, fever (100° to 102°), severe fever (?103°), chills, muscle aches, runny nose, stuffy nose, headache, sore throat, sneezing
  6. How is the influenza virus spread?
    objects and surfaces, blood, bodily fluids, airborne (by sneezing, coughing), sexual contact, ingesting foods, the flu shot
  7. If you are in a room with a person who is infected with influenza, what are the chances you will catch the influenza virus?
    not at all, unlikely, somewhat, likely, definitely
  8. What are effective ways to reduce your risk of infection?
    N95 mask, procedural mask, face shield, gloves, hand washing, covering cough, social distancing
  9. How likely are you to catch the influenza virus at work relative to other places you frequent?
    a lot less likely, a little less likely, about equal, a little more likely, a lot more likely

10.  Do you feel that your work has adequate protections for you in the event of an influenza pandemic? (yes / no)

Appendix B: Semi-structured Interview Questions

The following questions should be presented to the interviewee through an in-person and formatted appropriate to the interview.  The results will be kept anonymously.  The structured interview questions should be asked first.  The interviewer should ask probing questions to pull additional information that may provide insight to interviewee’s thought process. 

Ask if the person has children or dependants in school for questions 1 and 2.

  1. When schools have an unscheduled closing, how does that impact your ability to work?
  2. If schools have an unscheduled closing for four consecutive days, how would this impact your ability to work?

Ask if the person lives in a household with other people for questions 3 and 4.

  1. How would another household member be cared for if they were seriously ill and could not leave home?
  2. If that person had a highly contagious disease, does that change how they are cared for?
  3. How would you be cared for if you became seriously ill and could not leave home?
  4. If you know you have a contagious disease that others may catch, would you still go to work?
  5. How important do you feel your work is to the successful running of the health-care facility?
  6. The pandemic influenza is a highly contagious disease with an expectation that 30% of the population will be infected.  Knowing this, how comfortable are you working in a health-care facility?
  7. A person infected with pandemic influenza may infect others 1 day before symptoms develop and up to 5 days after becoming sick  Knowing this, are you concerned that you might get infected at work and bring this home to your family?

10.  If an influenza pandemic started tomorrow, would you take actions to protect yourself?  If so, what would be those actions?


Agency for Healthcare Research and Quality.  (February 2007).  Mass Medical Care with Scarce Resources: A Community Planning Guide.  (AHRQ Publication No. 07-0001.)  Rockville, MD: Agency for Healthcare Research and Quality 2007.

Ajzen, I., Albarracín, D., & Hornik, R. (Eds.)(2007).  Prediction and change of health behavior: Applying the reasoned action approach.  Mahwah, NJ: Lawrence Erlbaum Associates.

Emergency Management Principles and Practices for Healthcare Systems.  The Institute for Crisis, Disaster, and Risk Management (ICDRM) at the George Washington University (GWU); for the Veterans Health Administration (VHA)/US Department of Veterans Affairs (VA).  Washington, D.C., June 2006.  Available at http://www1.va.gov/emshg/.

Fishbein, M., & Ajzen, I. (1975).  Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research.  Reading, MA: Addison-Wesley.

Health Care Worker Safety During a Pandemic.  (2007, July 19).  Retrieved March 20, 2008 from http://www.fluwikie.com/ index.php? n=Consequences.HealthCareWorkerSafetyDuringAPandemic.

Prochaska, J. O., DiClemente, C. C., & Norcross, J.  (1992).  In search of how people change.  American Psychologist, 47, 1102-1114.

Robertory, K. (2008).  A Review of How a Hospital Could Use the Incident Command System for a Pandemic Flu Incident. 

U.S. Department of Health and Human Services. (November 2005).  HHS Pandemic Influenza Plan.  Washington, D.C.: US DHHS.

U.S. Department of Homeland Security.  (September 2006).  Pandemic Influenza; Preparedness, Response and Recovery Guide for Critical Infrastructure and Key Resources.  Washington, D.C.: US DHS.

U.S. Department of Homeland Security.  (September 2006).  Pandemic Influenza; Preparedness, Response and Recovery Guide for Critical Infrastructure and Key Resources.  Washington, D.C.: US DHS.

U.S. Department of Transportation.  (May 2007).  EMS Pandemic Influenza Guidelines for Statewide Adoption; Task  (DOT HS 810 776).  Washington, D.C.: DOT.

U.S. Occupational Health and Safety Adminstration.  (2007).  Guidance on Preparing Workplaces for an Influenza Pandemic.  (OSHA publication number: OSHA 3327-02N 2007).  Washington, D.C.: OSHA.

U.S. White House, Homeland Security Council.  (November 2005).  National Strategy for Pandemic Influenza.  Washington, D.C.: White House.

[i] OHSA.  Guidance on Preparing Workplaces for an Influenza Pandemic.

[ii] As an editorial note, this author uses “influenza pandemic” to describe the event and “pandemic influenza” to describe the disease.

[iii] DHS.  Pandemic Influenza; Preparedness, Response and Recovery Guide for Critical Infrastructure and Key Resources.

[iv] Health Care Worker Safety During a Pandemic.

[v] Robertory, K. A Review of How a Hospital Could Use the Incident Command System for a Pandemic Flu Incident. 

[vi] Fishbein, M., & Ajzen, I.  Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research.

[vii] Ajzen, I., Albarracín, D., & Hornik, R.  Prediction and change of health behavior: Applying the reasoned action approach.

[viii] Prochaska, J. O., DiClemente, C. C., & Norcross, J.  In search of how people change.  American Psychologist, 47, 1102-1114.

[ix] “Interview” (2008, March 6).  In Wikipedia, the free encyclopedia.  

[x] Creative Research Systems.  Sample Size Calculator.  http://www.surveysystem.com/sscalc.htm.