The last year or so has been the most uncertain and unexpected period in our lives with the pandemic disrupting set routines. Nobody likes uncertainties. Predictability makes people feel safe and gives them a sense of control, while uncertainty makes them anxious. Experts do not like to be seen as tentative or admit ignorance. The stronger the opinion, the better the expert. Neither politicians nor the public in general are comfortable with uncertainties. Everybody wants a straight answer — yes, no or a number. This may not always be possible. The WHO has acknowledged that the inevitable uncertainty with regards to Covid-19 could lead to fear, and that needs to be addressed.
Uncertainties related to Covid have multiple origins. First, the disease itself was new and we did not know much about its epidemiology and pathophysiology. Many of our initial responses were, therefore, products of educated guesswork. Second, the disease had a multifaceted impact on our lives with each aspect of the response affecting other aspects resulting in a cascading of uncertainties. The third cause of anxieties stemmed from avoidable data deficiencies.
Scientists understand and deal with uncertainties. They use the concepts of confidence intervals or uncertainty intervals in their estimations as statistical ways to deal with the uncertainties of estimating a single number — instead of one number, they often choose a range. My close family members have often accused me of not giving a definite response to their questions, often hinting at a lack of competence. My inability to give a definite answer is precisely because I know my subject well and thus understand uncertainties. Let me illustrate this with the example of a question often asked to a doctor when one’s close relative gets diagnosed with cancer: Doctor, this is curable, isn’t it? Unfortunately, except in a few cases, this query cannot be answered with a straightforward “yes” or “no”. The only legitimate answer is “it depends”. Cancer experts now handle this uncertainty better. They use the term “median survival time” or five-year survival as an indicator to start a dialogue on this uncertainty. In other words, they try to estimate the duration for which 50 per cent of patients with such cases would survive or what proportion would survive for five years given that they are in the same stage and with the decided treatment modality. However, this still does not answer individual-level uncertainty. Nobody can say for certain if a patient is in the first half of the median or its second half. But this information enables us to peg our expectations to a number and prepare for the outcome.
We saw uncertainty play out in the open during the pandemic as we discussed different options for treatment or prevention of the disease. People got frustrated when they saw scientists, who are expected to provide clear and definite answers, contradicting each other. At an individual level, we dealt with uncertainties related to wearing a mask, getting vaccinated, going to office, using public transport. Even though all these decisions were a play of probabilities, they had the advantage in that the government provided a clear and unambiguous opinion on most of them, and we followed suit. This raises the issue of faith. Faith, whether in god, government, scientists or doctors is a good antidote to uncertainty. However, in the case of this pandemic, such certitudes have been tested and found wanting.
Let us take the issue of the opening of schools. I am amazed at some experts who unabashedly advocate the opening of schools as well as those who vehemently oppose it. If you ask me, I will say “it depends”. Unlike the example of cancer above, we may not always be able to quantify the uncertainties and compare multiple sets of uncertainties. In this example, we know that the primary push for opening schools is the adverse scholastic and social impact of continuing with online teaching as well as the parental burden of caregiving. These are difficult to quantify. The other side is the impact of the opening of schools on infections in children and the probability of their bringing the infection home to elders. There is a lot of uncertainty in this estimation.
How do you compare the two sides of this equation? There cannot be any doubt that there is a “small” but definite possibility of some extra deaths caused by the opening of schools. Even if we assume that to be a small number, are we ready as a society to accept that number as a cost of return to normalcy? Who takes this call? Do parents have a say in this decision, given that they are the primary risk takers? How do we deal with parents of children who do not want to take any risk in these uncertain times and do not want to send their children to school? Do we penalise them for this concern? Providing answers to these questions requires value judgements with significant uncertainties. I do not think this problem is amenable to a definitive answer either way. If that makes me a poor expert, so be it. A graded response enables us to watch the consequences and modify our response, thus reducing uncertainty.
So how do we respond to such uncertainties? Typically, humans choose one of two types of responses — denial of the problem resulting in policy paralysis or by brazenness in decision making. What is needed is a calculated response that confronts uncertainty head-on. Doctors and scientists must learn to communicate uncertainties unambiguously and unapologetically. Providing and displaying information on uncertainty is the starting point. Policymakers and the public need to be educated that uncertainties are a part of life and cannot be wished away.
Some strategies that can be adopted in dealing with uncertainties are: Identify things that are certain and those that are riddled with the most uncertainties. Use them to build case scenarios including the most likely and worst case scenario. Be fully prepared for the most likely one and try your best to obviate the worst one. Seek diverse perspectives, initiate dialogue and discussion with all concerned. A broadly shared decision is more likely to be successful and even if wrong, unlikely to be criticised. Communicate the final decision and its rationale to the public along with the risks and consequences. Monitor the response, acknowledge mistakes and be ready to undertake course corrections by learning from those mistakes. Institutionalise these learnings so that we do not repeat mistakes.
Perhaps, these uncertain times can help us come to terms with uncertainty itself and help us deal with it better.
This column first appeared in the print edition on July 30, 2021 under the title ‘Beyond yes, no or a number’. The writer is a Professor at Centre for Community Medicine, All India Institute of Medical Sciences, New Delhi. Views expressed are personal.