Predictive analytics, the domain of business and market management now finds increasing use in healthcare settings. Hospitals with limited capacity often face problems in making effective use of resources for operations, emergency treatment, outpatient ward, inpatient staff and beds. Faced with a fixed number of beds, hospitals face problems of overstaffing or under staffing. Kumar and Shastri (2009) speak of capacity management software for hospitals that help to overcome problems such as emergency department overcrowding, scheduling, patient flow management, discharge co-ordination, and bed turnover. Software applications help to handle bottleneck areas in hospitals that can occur due to reasons of delays in shifting patients to other wards, patients pick up infections, and overcrowding leads to loss of patients who cannot be accommodated. To schedule effectively, hospitals need the ability to accurately predict sort term demand trends. Flow of patients between hospitals must be carefully managed; patients waiting time at admissions must be reduced.
Harper (2012) illustrates features of an application that can help to manage hospital beds and capacity. Using historical data, he predicts that winter months of Jan-Mar have the highest demand while other months have lower demand.
Lions Gate Hospital in North Vancouver, British Columbia faced problems of managing capacity and the staff was always in crisis management mode. The hospital decided to use McKesson Capacity Planner to overcome the problems. The planner uses three years of past hospital data to identify variations as per seasons, weekly, and for other factors to forecast demand. It uses inputs from expert doctors to seek specific changes and events that drive demand. These inputs have helped the hospital to predict schedules and unscheduled demand with a 95% monthly accuracy and predict 98% daily accuracy. The tool helps to help in making decisions three days in advance and this lead to reduced overtime of staff by 7.9%. In addition, the tool helps to identity key metrics such as projected and actual and resourced occupancy. Managers can view patient information and occupancy, and this allows viewing capacity shortages and patient flow issues. Managers can view projected and actual discharges and this helps to reduce the length of stay and leading to 110% utilization rates with $2.7 million reduction in staffing costs (McKesson, 2014).
Discussions from the previous paragraphs guide us in the ways in which these tools can be leveraged. Some features that can be used are demand management, bed capacity planning, and planning of staff to reduce their overtime. A closer study indicates some features that we would like to see in capacity planning. These include powerful forecasting tools that remove inaccuracy in forecasting and estimation. Real time information flows must be available through hand held smart devices, and this reduces waiting times. Feature needed for waiting time management included understanding the average process turnaround data, average delay, and average wait times per patient, variability measurement, and process inefficiency reduction. The main problem is that patients suffer from various types of ailments and related diseases (Balaji and Brownlee, 2009). This variation means that a patient may be discharged in a few hours while another with complications may undergo treatment of a few weeks.
- Balaji, R. and Brownlee, M. (2009). Bed Management Optimization. Infosys. Retrieved from https://www.infosys.com/industries/healthcare/Documents/hospital-bed-management.pdf
- Harper, P. (2012). PROMPT – Hospital bed and workforce capacity planning tool. YouTube. Retrieved from https://www.youtube.com/watch?v=icQiF4Rn6Ug
- Kumar, K. and Shastri, L. (2009). How advanced analytics can improve hospital capacity management. Infosys. Retrieved from https://www.infosys.com/industries/healthcare/white-papers/Documents/hospital-capacity-management.pdf
- McKesson, (2014). Lions Gate Hospital uses demand forecasting to manage capacity, optimize staffing levels and reduce length of stay. McKesson Corporation. Retrieved from http://www.mckesson.com/documents/providers/case-study—lionsgate-2014-3237/