05 Jun 2013

Cost benefit analysis in healthcare – Lifetime costs and benefits

Raoul Reit

What is cost benefit analysis?

This is the first blog in a series of blogs about cost benefit analysis within the healthcare sector. What is cost benefit analysis? It is a framework used to assess whether a particular intervention is worthwhile, or, in other words, how the cost of the intervention compares with the benefits to patient health and well-being. This particular blog will aim to deliver a short demonstration of how to calculate lifetime costs and benefits.

One of the most difficult aspects of developing a robust analysis is understanding how to best measure the cost and benefits of the intervention. From our experience, every intervention follows a different treatment pathway and the difficulty lies in finding out a way to model the intricacies of the problem. In a recent project, we were asked to assess the provision of specialised support services to patients with severe disabilities. In this case, the caveat is that support can start at different ages for different patients and will last for the duration of the patient’s life. From this point of view, you cannot calculate the cost and benefit for one patient and scale up because they will depend on the age of each patient.

So how did we get around this?

1. Separating data based on age group

We started by separating our patients into 1 year age groups; patients less than 1 years old, between 1 and 2 years old, between 2 and 3 years old, and so on. This allowed us to focus our calculations on one patient group at a time. The figure below displays this information for a region within the UK, with the 5 to 18 years group highlighted in orange. Notice the spike in the 63 to 65 years group representing the baby boomers generation. 

Concentra - Age distribution for a UK region Source: Office of National Statistics  
 

 

2. Calculating lifetime cost

The next step involves calculating lifetime costs and benefit for each of these age groups. Best practice suggests that costs and benefits should be calculated as the net present value over a patient's lifetime, using the following approach

Concentra - Calculating lifetime costs and benefits Where c represents annual cost and L represents a patient’s remaining life expectancy. This framework allowed us to be flexible in the way in which we discounted costs and benefits for patients starting treatment at different ages, using the age specific data described earlier.

 A specific complication with our intervention was that children and adults have a different treatment pathway. Using information from relevant journal articles we deduced that, on average, children have a major progress review every 4 years whereas adults are reviewed every 5 years. The review lasts several sessions and the parties involved decide how best to continue treatment for the patient. During this time, the patient usually receives several sessions of specialised therapy, which incurs a cost. This means that discounting has to be applied differently for children and adults.

3.Calculating benefits

 We calculated benefits by looking at the effects on quality of life that support might have and translated these into monetary values by means of quality adjusted life years (QALYs) gained. A QALY represents the relative worth of 1 year of life for a patient with disabilities compared to 1 year of life for a healthy person. Basically, this means that the intervention leads to a positive benefit since patients will live the rest of their lives with a higher quality of life. As in the case of costs, QALYs were also discounted to their net present value as follows:

Concentra - QALY Where Qi represents a patient’s well-being score and Ti represents a time period, in this case a year. The sum of these years represent a patient’s QALYs. The difference in this sum before and after the intervention represents the benefit in terms of QALYs gained. 

So what does this mean? 

The figure below shows the calculated lifetime benefit, cost and cost/benefit ratio for 1 patient starting treatment at different ages. For example, for a patient starting treatment at age 15, the lifetime costs (in present value) are about £55,000 and the lifetime benefits are about £65,000. 

Concentra - Benefit, cost and benefit cost ratio for 1 patient

Take a moment to understand this graph because it provides a lot of information. Notice that as treatment starts later in a patient’s life, the absolute difference between costs and benefits reduces. This makes sense because with a later treatment start there are fewer years in which costs and benefits can accrue. On the same note, the ratio between costs and benefits remains the same for adults since cost and benefits are discounted at the same rate. However, the graph shows the effects of higher costs for children through the kink in the cost curve (blue line) at age 18 and the lower cost benefit ratio for children. In addition, note that the cost benefit ratio for children increases with the treatment start age, slowly approaching the adult ratio.

Even though the idea of calculating costs and benefits may seem straightforward, there are a lot of details and nuances that need to be taken into account. Hopefully, this blog has given some idea of what these might be. If you want to see our cost benefit models in practice we will be at London’s Excel Centre from June 12th to June 13th for a national clinical commissioning event.

 For more information about our work in cost benefit analysis or in other areas of health economics, please contact James Don-Carolis at james.don-carolis@concentra.co.uk.

Raoul Reit

About the author

Consultant. I joined Concentra’s consulting team in 2011 after completing my studies in economics and operations research. As a consultant, my work is usually diverse and I have been involved in projects ranging from healthcare economic analysis to data analytics and business modelling. My blogs will aim to reflect my interest in applied economics and data analysis and I hope you will find them interesting and informative.