Picture source: Gallo Images/Getty

By Sarel Steel, Andre Stephan Botha and Tashvir Khalawan

The quantification of damages in certain legal matters involves calculating the capitalised value, or expected present value, of a list of future anticipated medical expenses. In this calculation, allowance is made for the probability that the injured plaintiff may die before incurring the future expense. The number of years that the individual is likely to survive – namely, their life expectancy – is therefore, a critical assumption affecting the quantum. Furthermore, typically in the case of medical negligence matters, the plaintiff may not be expected to experience ‘normal’ life expectancy due to the impact of the medical condition on survival. One such medical condition is cerebral palsy.

This article has two aims –

• first, a discussion of the uncertainty surrounding estimation of the life expectancy of a child in South Africa (SA) suffering from cerebral palsy, especially given the absence of a local data register, and the implications of this uncertainty;
• secondly, some thoughts on the possibility of establishing a local cerebral palsy data register.

The uncertainty involved when life expectancy is estimated is especially important given the current practice of a once-off payment to the plaintiff.
Although this article is not intended to present a comprehensive discussion of these issues; the intention is to heighten awareness of their importance.

Uncertainty when estimating life expectancy

The life expectancy of a person is the number of years that the person can expect to live beyond their current age. This definition needs some explanation. A life expectancy of 60 years for a child aged 10 implies that the child can expect to live for another 60 years. This does not imply with certainty that the person will in fact die at age 70. ‘Expect to live’ in this context has a specific meaning, derived from the statistical approach that is used to arrive at a life expectancy of 60 years. It implies that the average age at death of all 10-year-old children comparable to the specific person is estimated to be 70 years. Although all the children in this homogeneous group have a life expectancy of 60 years, some will die at age 11, and others only after the age of 90.

To estimate life expectancy, an investigation is typically required that analyses the past mortality experience of a group of individuals that are similar to the individual under consideration. Such an investigation would require suitable past data. The mortality rates derived from the investigation can then be applied to estimate the future mortality experience of the individual.

Given the uncertainty of future life, it is clear that life expectancy should be interpreted carefully and with discernment. This perspective is even more relevant for persons afflicted by a condition such as cerebral palsy, since clinical factors affect these individuals differently. Methods that are used to estimate life expectancy do attempt to take this into account by pooling together data for children forming relatively homogeneous groups. With cerebral palsy, the most important factors that are used to define ‘homogeneous groups’ are gross motor function ability and the method of feeding. There are several other less important factors that have to be taken into account.

There are several other less important factors that have to be taken into account such as weight for age, epilepsy, socio-economic position and general cognitive ability. In this regard a dilemma arises. If more factors are used to define homogeneous groups in order to estimate life expectancy, more relevant data is obtained, but the groups become smaller and the results correspondingly less statistically reliable.

Regarding the availability of data that can be used to estimate the life expectancy of cerebral palsy individuals, the largest and most comprehensive mortality database is provided by the Life Expectancy Project in San Francisco (LEP). Unfortunately, this database has never been published and only limited mortality data and life expectancy figures are provided in the literature, the latest publication being JC Brooks, DJ Strauss, RM Shavelle, LM Tran, L Rosenbloom, and YW Wu ‘Recent trends in cerebral palsy survival. Part II: Individual survival prognosis’ (2014) 56(11) Developmental Medicine & Child Neurology 1065. It is feasible to estimate life expectancy for a cerebral palsy individual at a given age from the LEP database and this estimate can be used to construct a life table for the individual. Strauss et al propose various mathematical techniques to do this (DJ Strauss, RM Shavelle, C Pflaum and C Bruce ‘Discounting the cost of future care for persons with disabilities’ (2001) 14(1) Journal of Forensic Economics 79 and DJ Strauss, PJ Vachon and RM Shavelle ‘Estimation of future mortality rates and life expectancy in chronic medical conditions’ (2005) 37(1) Journal of Insurance Medicine 20).

Summarising the above discussion, in an ideal world, given an infinite amount of data and perfect knowledge regarding the best method for estimating life expectancy, one would be able to arrive at an accurate estimate. Naturally this is not feasible. Moreover, even if an accurate life expectancy estimate could be obtained, the considerable variation in individual lifetimes within a homogeneous group of individuals would remain. An individual life expectancy estimate and calculations based on this estimate should, therefore, be viewed cautiously.

A cerebral palsy register in SA would enable data to be collected to perform a mortality investigation that could then be used to estimate life expectancy in SA. In the absence of local data, estimates that are constructed are based on foreign data. This entails adjustments that are applied to foreign results to account for local conditions. Different choices are made regarding these adjustments and it is difficult to decide which choice is best. All of this implies a further increase in uncertainty.

What are the implications for cerebral palsy related litigation?

Current practice usually entails the plaintiff presenting an expert opinion to the court, which is used to calculate an amount that the defendant (a provincial Department of Health or medical negligence insurer) is deemed to be liable for as compensation for proven negligence. An important part of the expert opinion is a life expectancy estimate. The defendant presents a different expert opinion, that may include a different life expectancy estimate, and that leads to a different estimate of damages, leaving the court to resolve the matter and determine a once-off payment intended to purchase reasonable care until death.

This system has several flaws. The estimates in any expert opinion are subject to uncertainty and this is not always sufficiently recognised. Given this uncertainty, attempts are often made to avoid under-compensating the plaintiff. This, however, also has drawbacks. In a presentation to the 123rd Annual Meeting of the American Academy of Insurance Medicine, Robert Shavelle, a prominent researcher in cerebral palsy life expectancy, states: ‘Does care matter? If “good” versus “bad”, then yes. If “excellent” versus “reasonable and necessary”, then no’ (RM Shavelle ‘Life expectancy after catastrophic injury’, http://aaimedicine.org, accessed 9-7-2019). Over-compensation also ignores the reality of limited provincial health budgets. Finally, if an individual that has been granted a once-off payment passes away before the expiry of the estimated life expectancy, the remaining cash devolves to the estate of the deceased, and so the heirs of this person benefit.

It seems clear that a system of staggered payments to a plaintiff would substantially reduce the impact of the life expectancy estimate and hence reduce the uncertainty implied by a once-off payment. Part of such a staggered payment approach would be regular expert input, say on a five-year basis, providing up-to-date information on the condition of the individual. This will provide a more reasonable basis for deciding on further payments to the plaintiff. Such a system would require separate consideration from a legal and legislative perspective.

A cerebral palsy register for South Africa

According to KA Donald, P Samia, A Kakooza-Mwesige and D Bearden ‘Pediatric cerebral palsy in Africa: A systematic review’ (2014) 21(1) Seminars in Pediatric Neurology 30, the estimated prevalence of cerebral palsy in Africa is between two and ten cases per 1 000 live births (hence, between 0,2 and 1%), compared to between 2 and 2,5 per 1 000 live births globally. The higher prevalence in Africa may be the result of a lower level of health care compared to global standards, although the impact of different methods of estimating cerebral palsy prevalence cannot be discounted. Ideally, prevalence estimation should be based on accurate and comprehensive data. Such data does not seem to be widely available in Africa. S Goldsmith, S McIntyre, H Smithers-Sheedy, E Blair, C Cans, L Watson and M Yeargin-Allsopp ‘An international survey of cerebral palsy registers and surveillance systems’ (2016) 58 Developmental Medicine and Child Neurology, discuss the results of a global survey of cerebral palsy registers. Their survey reports on 25 registers, of which 12 are in Europe, eight in Australia and five in North America. According to these authors there does not seem to be any cerebral palsy register in Africa.

Narrowing the focus to SA, a local cerebral palsy register will increase the confidence in prevalence estimates. This will not be the only benefit. According to Goldsmith (op cit) the registers included in their survey indicated aims that can be grouped into five broad categories, namely –

• a cerebral palsy register is a valuable resource for research;
• it provides a base for surveillance of individuals afflicted by cerebral palsy;
• it contributes to initiatives aimed at prevention;
• it is an important aid in health care planning; and
• it contributes to the raising of community awareness.

One can add to this the importance of a relevant data register in the computation of life expectancy estimates.

Regarding the role of a cerebral palsy register in research, only one point is expanded. Cerebral palsy is not a single homogeneous condition. The population of patients must, therefore, be considerably stratified, leading to small numbers of cases in some categories. This reduces the confidence in estimates and predictions for these categories, a problem that can be addressed partially by pooling local data with that from other countries.

A more detailed picture of the aims with a cerebral palsy register and of the services provided by existing registers can be obtained by visiting the web pages of, for example, the following two registers: the ‘Australian Cerebral Palsy Register Group’ at www.cpregister.com, and the ‘Surveillance of Cerebral Palsy in Europe’ at www.scpenetwork.eu.

It is not clear how establishing a South African cerebral palsy register should be approached – what follows are merely some ideas, namely:

• Identifying interested role players will be required. These will include Departments of Paediatrics at universities, public and private hospitals, care-giving centres, practising paediatricians, institutions such as the Medical Research Council, the provincial and national Departments of Health, and possibly medical schemes.
• It may be advisable to launch the project in a single province first, for example the Western Cape, and to extend it to the other provinces only once it is running smoothly.
• Basing the register at one or more universities also seems natural. The initiative can initially be proposed as a joint research project between several universities, which should generate funding from research bodies and possibly private health care providers. Obtaining assistance from abroad would be advisable.

South African conditions

Under-reporting of cerebral palsy probably occurs in SA for various reasons, which will not be expanded here. This deprives the children involved of proper care. A South African cerebral palsy register will contribute to heightened community awareness of the condition and will lead to a decrease in the stigma surrounding cerebral palsy, one of the possible reasons for under-reporting.

The reality of substantial once-off payments for delictual compensation is placing a strain on state resources, to the detriment of, for example, projects to reduce cerebral palsy incidents and to improve the care of individuals afflicted by cerebral palsy not ascribable to medical negligence. As pointed out earlier, a contributing factor is that claims are being contested on an individual basis, with life expectancies and compensations being estimated and contested without local data being available. Establishing a cerebral palsy register will, in the long term, alleviate this problem.

Sarel Steel PhD (Stell) is a consultant and part-time lecturer at Population Income Information in Stellenbosch, Andre Stephan Botha MBChB (UCT) DCH (SA) FCP (Paeds) (SA College of Medicine) is a paediatrician in Durban and Tashvir Khalawan BSc (UCT) FASSA is an actuary at Edge Actuarial Consulting in Durban.

 This article was first published in De Rebus in 2019 (Sept) DR 14.