This study will try to determine the cost effectiveness of training and employing ClinAs and medical practitioners (MPs) compared to the standard strategy of training and employing only more MPs.

This comparison can help a province to decide if and how much, beside training and employing MPs, the province should be investing in training and employing ClinAs, as an additional strategy to reduce the healthcare workers shortage in rural areas.

What is the cost effectiveness for a province in South Africa of investing in training and employing ClinAs, compared to investing in training and employing more MPs?

We will answer our main questions with the help of the following sub questions:

What are the costs for a province to train a clinical associate compared to training a medical practitioner?

What are the costs for a province to employ a clinical associate in comparison to employing a medical practitioner?

What is the scope of practice of ClinAs and how does this compare to MPs?

How much medical practitioner time can a clinical associate save?

What is the quality of care provided by ClinAs?

How is the retention of ClinAs in rural areas and how does this compare to MPs?

The provinces and NDoH funds grants and bursaries for students, so that they are able to follow a healthcare education. The bursaries are paid to the universities and cover the costs for fees, accommodation, meals and books. The students, who receive a bursary, must in return ‘work back’ their bursary obligation in the public sector after qualifying.^1,3

A comparison was made between the bursaries for BCMP students at the University of Pretoria and the bursaries for MBChB students (Mpumalanga Department of Education 2015, personal communication, Oct 8; University of Pretoria 2015, personal communication, Oct 8) (Tables 1 and 2). The first is a three-year programme, while the latter is a six-year programme.

The total cost for a province to pay for the full training of a clinical associate is R 300 850 and a medical practitioner R 730 985. This indicates that it will cost a province approximately two and a half times more to train a medical practitioner than a clinical associate at the University of Pretoria. The training of MPs in Cuba is more expensive and exceeds R 1 million per person.

There are already calculations about the average cost of a medical practitioner,² however, not yet of the average cost of a clinical associate as this is a new cadre. We therefore came up with a new calculation as an indication to compare the average cost to employ a clinical associate with that of a medical practitioner.

The main cost of an employee is his or her salary. Therefore, we compared the average salary of a clinical associate, for a 10-year period, with the average salary of a medical practitioner, for the first 10-year period of employment (Tables 3 and 4).¹⁰

Based on the above calculation, the total cost for a province to pay for employment of one clinical associate in the public service is on average R 196 329 per year and to employ a medical practitioner in the public service is on average R 559 397 per year for the first 10 years after graduation. This excludes any overtime remuneration and indicates that it will cost nearly three times more for a province to employ a medical practitioner in public service than a clinical associate. Because of the fast rise in salaries for MPs in the first two years (Figure 1), this difference is expected to become larger when looking at more senior MPs compared with more senior ClinAs.

When looking at the scope of practice for the medical profession, it is clear that MPs in South Africa are allowed to practise medicine in the broadest sense of the word.¹¹

The scope of practice of ClinAs on the other hand is much more specifically defined in the Health Profession Act.¹² In summary, ClinAs are only allowed to practice under continuous supervision in accordance to their level of education, training and experience. Furthermore, they are allowed to do a number of specifically defined procedures described in the Health Profession Act.¹²

They can also perform any act delegated by the supervising medical practitioner in accordance with the education, training and experience of the clinical associate. They cannot conduct an independent private practice; they cannot act as a substitute for a medical practitioner.¹²

In Rwanda, there has been a study to quantify medical practitioner time saved through task shifting.¹³ The demand on medical practitioner time was reduced by 76% when nurses were used to take over tasks of MPs.¹³

In this pilot study, nurses were trained to function specifically in an HIV programme. It seems a safe assumption that training ClinAs for similar routine tasks would yield similar results. However, one could argue that ClinAs in district hospitals perform a wider variety of tasks than in the above-mentioned study and will therefore achieve a smaller reduction of medical practitioner time.

Therefore, we will work with two different scenarios in the case study. In the first scenario, we will use a medical practitioner reduction time of 76%.

In the second scenario, we will use a physician reduction time of 50% to compensate for the lack of knowledge in this area. It seems reasonable that ClinAs are able to at least free up 50% of the medical practitioners’ time.^4,5,6,8 In a recent study by Ngcobo on voluntary male circumcision by clinical associates and doctors he found that there was no difference in quality and productivity between the two groups.¹⁴ An American study of physician assistants found that 70–90% of the work of a primary care physician can be done by a non-physician clinician (NP-C).¹⁵

There has been one very recent study in South Africa that found that ClinAs provide the same quality of care and outcome compared to MP`s when doing Voluntary Male Medical Circumcisions at a much lower cost per circumcision.¹⁴

A number of literature reviews, international guidelines and even a meta-analysis have also looked at the quality of care provided by mid-level health workers (MLWs).^4,5,6,7,8

The general consensus seems to be that although more evidence is needed in African countries, mid-level workers can provide the same quality of care as higher level workers provided that they receive adequate training, support and supervision. If this is not provided, the quality of care can be sub optimal.^4,5,6,7,8

ClinAs are more willing to work in rural areas compared to MPs.¹ A study that tracks the first 250 BCMP graduates from the University of Pretoria found that 57% work in the rural public health institutions (personal communication Lekgala Monareng Master’s thesis, 2016). Reasons are that they are recruited from rural communities and are trained close to where they are recruited. Furthermore, their degree is not recognised internationally.^3,6,7,9

One study done at the University of Pretoria found that if given complete freedom of choice without bursary or family obligations 53.4% of the BCMP students indicated a preference to work in rural areas.¹ For medical students this percentage was 4.8%.¹

However, one must not assume that mid-level workers will tolerate low salaries and poor living and working conditions. Low pay, poor supervision and lack of training are associated with lower retention.^5,7 Therefore, good salary and incentive structures must be developed and implemented.^5,6,8

In 2010, the province Mpumalanga had 535 vacant posts in the public sector for MPs.² Suppose that the province decides to invest an extra budget of R 10 000 000 in training healthcare workers in order to fill these vacancies for MPs. We will investigate the cost effectiveness of two different strategies to spend this extra training budget:

It is important to realise that ClinAs free up time for MPs, so that these MPs can do the work of two or more vacant MP posts. Also, this case study looks at the best strategy if a province frees up extra budget to train. The training and employing of ClinAs should not replace or compete with the current training and employment of MP’s.

Based on the costs for bursaries and the average salaries we calculated the costs for the province to train and employ the MPs and ClinAs for the different scenarios. Also the cost of every filled vacant MP post per year was calculated. The results are summarised in Table 5. Note the difference in costs per filled vacant MP post between scenario 2a and 2b because more vacant MP posts can be filled in scenario 2b.

To summarise the case: training and employing ClinAs are effective because they can free up time of MPs, which the MPs can spend to fill more vacant MP posts, without loss of quality of care. It is also cheaper to train and employ a combination of MPs and ClinAs, than only training and employing MPs. How much cheaper depends on the productivity of the ClinAs.

We therefore conclude that it is a more cost-effective strategy for the province of Mpumalanga to spend the extra budget on training a combination of MPs and ClinAs than on spending it only on training MPs.