- A coalition of alcohol producers say a review of a report which government is said to have used in deciding to ban sales during lockdown was found wanting.
- SAMRC's Professor Charles Parry has responded to the review, defending the work of the team of nine experts.
- Parry says the industry should be focusing on preparing itself for a "new normal post-lockdown".
Professor Charles Parry of the South African Medical Research Council (SAMRC) has hit back at the alcohol industry following a review they commissioned on a report which informed government’s ban on the sale of liquor.
The alcohol industry coalition - which includes major producers such as SAB, Distell, Heineken and Pernod Ricard - recently released a statement on a review that identified that the report on alcohol-related trauma contained several data and statistical limitations due to the lack of complete information, making the robustness of the research uncertain.
"Trying to pick apart aspects of the model – in vain, it would seem – to either cast aspersions on the government, Health Ministerial Advisory Committee (MAC) or [the] SAMRC seems futile given that the emerging trauma (and death) data largely support the predictions contained therein," Parry said.
"The industry's focus should rather be directed at gearing itself for a new normal post-lockdown and enabling those players in the industry that have adopted the responsible drinking mantra to act responsibly by embracing the many evidence-based harm reduction measures that need to be put in place to protect lives."
The alcohol industry coalition commissioned Kantar - a data, insights and consulting company - to do a review of the report by Parry and others.
Kantar confirmed to News24 that it had conducted a review, but said it wasn't at liberty to share the original data review.
Instead, Kantar said the contents of its review were covered in the statement by the alcohol coalition.
The statement notes that the SAMRC report informed the banning of alcohol sales in South Africa and that the document was presented to government’s National Coronavirus Command Council (NCCC) on 15 July.
Parry points out that the report was produced by a team of nine experts at the request of the MAC.
The experts included three scientists from SAMRC, trauma specialists, alcohol epidemiologists, a health economist, a general practitioner/health services manager, and an infectious disease specialist.
The report was not presented to the NCCC on 15 July, which would have been three days after President Cyril Ramaphosa announced the ban on the sale of alcohol, but rather, the report was only presented to a sub-committee of the MAC on 7 July.
Alcohol coalition review
According to the coalition's statement, the review highlighted that the incomplete data reported makes it difficult to accurately determine the extent of the link between trauma admissions and alcohol abuse.
The review further said there was insufficient information reported on whether the trauma cases resulted from alcohol use, or from any other cause, and that South Africa was one of four countries in the world to retain a complete prohibition on alcohol sales as a means to fight the spread of Covid-19.
"The review identified that the incomplete nature of the data used in the modelling makes it difficult to accurately determine the extent of the link between trauma admissions and alcohol abuse," the statement read.
"The main motive for this finding, is that the reason for each trauma is not captured by the hospitalisation data provided in the report. There is no information on whether the trauma cases resulted from alcohol use, or from any other cause."
- There is no information on whether the trauma cases resulted from alcohol use, or from any other cause. The model, therefore, may be able to predict the total number of potential trauma cases prevented; however, it remains uncertain as to the extent of the relationship between alcohol consumption and the number of trauma admissions in South African hospitals.
- The estimate for the number of trauma unit presentations per year in South Africa is based on extrapolations from data that is two decades old.
- The initial estimate used for modelling the potential number of reduced trauma unit presentations is based on data from roughly 356 secondary and tertiary public hospitals in 1999. To calculate the number of trauma unit presentations across the same hospital type in 2020, the population growth between 1999 and 2019 was used as a multiplier.
- However, the review finds that it is not apparent whether the number of secondary and tertiary public hospitals operating in 2019 is accounted for, as a substantial increase or decrease in the number of operating hospitals, notwithstanding the capacity of each hospital, could impact the number of cases admitted.
- The study itself recognises that a reduced number of trauma presentations may be related to more restrictions on movement, work, exercise, and fewer vehicles on the road – all factors that exist at the end of lockdown and present challenges in modelling separately.
Alcohol coalition industry spokesperson Sibani Mngadi encouraged government to take note of the review and of the limitations of the data reported.
"And engage with our industry to find an urgent solution to lifting the suspension on alcohol sales. This will enable the industry to return to some form of operational normality."
Firstly, Parry said they were happy to engage on the strengths and weaknesses of the report, despite the fact that the alcohol industry has not done the same, noting that the review was not released.
"We appreciate that they indicated their uncertainties since they only had the report to comment on and were not party to the lengthy deliberations of the nine-person team that undertook the modelling and gave input into the decisions," Parry said.
Responding to the issue of hospital capacity, Parry said the report did increase the estimate of public secondary and tertiary hospitals treating trauma patients from 356 to 380, but did not look at the capacity of each hospital.
"Out of a potential saving of 49 552 trauma presentations over eight weeks, over 380 hospitals, this would result in a potential average saving of 130 trauma presentations per hospital over an eight-week period (or on average of 16 per week), with clearly more at some hospitals and less at others.
"This does not seem extreme, but as the subsequent modelling showed, could for example, translate to just under 13 000 beds in ICU wards for Covid-19 patients across the country and over the eight weeks."
Parry added that in their Level 5 modelling, they did look at individual types of trauma presentations and did take into account that some injuries, like alcohol-related motor vehicle trauma injuries would be less impacted by an alcohol sales ban as there were less vehicles on the road.
"We also, for example, took into account that people were not congregating at some liquor venues known to be at higher risk for alcohol-related trauma. This thinking was carried over into our deliberations which were not reflected in the report to the MAC sub-committee."
Tackling assertions that the incomplete nature the data used in the modelling makes it difficult to accurately determine the extent of the link between trauma admissions and alcohol abuse, Parry pointed out that the team considered numerous data sources including clinical assessments of alcohol-related trauma at two large hospitals in Gauteng.
"At one of those hospitals alcohol-relatedness was deemed above 60% at weekends in Level 3 of the lockdown, and between 50% and 60% at another hospital.
"Using this information, plus the input of another trauma specialist and medical practitioners from the Western Cape and Eastern Cape we agreed that 50% was a defendable position."
Parry conceded that, at the time, it would have been valid to question the robustness of the estimates given that they were prepared by the researchers in a very short time frame and without access to complete data.
"However, with the benefit of hindsight, we seem to have done quite well in predicting the likely outcome of the ban within these constraints," Parry said, pointing at the evidence collected when the ban was reintroduced.
"When we consider trauma presentations from hospitals in Gauteng, KwaZulu-Natal, and the Eastern Cape (Helen Joseph: 41%+ drop; Charlotte Maxeke: 83% drop; an unknown hospital in KZN: 23% drop; King Edward VIII: 62% drop; and Dora Nginza: 50%+ drop.
"We had only predicted a 10% and a 15% reduction in trauma in the first and second weeks following a ban on liquor sales (or 20% and 30% of alcohol-related trauma)," said Parry.
"The reason for this is that we had not anticipated that the ban would be implemented with immediate effect – clearly the government heeded warnings from some quarters about possible 'panic buying'."
Other points Parry made:
- The issue of promoting the ban on liquor sales does not reflect any ideological position on behalf of the SAMRC or myself.
- It is well known that from around mid-May we have been promoting a basket of regulatory measures focusing on availability, drink driving, alcohol marketing and treatment to be considered at this time in the lockdown to reduce the risk of alcohol-related trauma admissions once the first ban on alcohol sales was lifted.
- This option of a basket of tougher control measures, which would fall short of a full ban on liquor sales, was also presented in our report, but the independent reviewers have not acknowledged this.