In late 2020 as we hid from COVID, we started to hear that in some people, COVID symptoms persisted for months. They were called “long haulers” or had “long COVID”.

Contrary to what we’d expected, it wasn’t just affecting people’s lungs or breathing. Long haulers were experiencing an enormous range of other symptoms: fatigue that was worse after activity, muscle aches and pains, headache, and cognitive dysfunction or brain fog.

A parliamentary inquiry is currently investigating Australia’s response to long COVID – patient experiences, the health system’s response and what the latest research shows.

It’s now clear we got some key things wrong about long COVID earlier in the pandemic. Three years in, and with an estimated 65 million people affected by long COVID, here are five things we now know or suspect about the condition.

1. It can take months to recover lung function – and some people never do

We now know full recovery of lung function is not guaranteed for people with long COVID: one in five still complain of severe breathlessness and 10% have severe functional impairment.

Among people admitted to hospital, studies report impaired lung function, abnormal chest scans, impaired capacity to exercise and persistent breathlessness months after leaving hospital, especially for those who needed breathing support in ICU.

In those people who where hospitalised with COVID-19 pneumonia and who have persistent breathlessness, pulmonary (lung) rehabilitation improves quality of life and exercise tolerance. While this evidence is not strong, it is consistent, across one randomised control trial and seven observational studies.

We need a national approach to assess all patients who survive hospitalisation, to determine if they are still breathless and ensuring they have access to pulmonary rehabilitation.

2. COVID can increase the risk of or worsen chronic diseases

Early on, we failed to understand COVID would increase the risk and worsen other chronic diseases.

Since then, large population studies have clearly documented people with long COVID are at increased risk of stroke and heart disease and an increased risk of diabetes.

These problems are more likely in those who are socially disadvantaged, and unable to have sufficient time to recover.

Read more: Tested positive to COVID? Go easy on yourself – try not to rush back to work or exercise

3. Long COVID isn’t a single disorder

We initially thought of long COVID as a single disorder. We now know it’s a complex condition, caused by a number of different factors. Emerging evidence suggests this includes:

Studies investigating these causes are still small and observational – and these factors are likely overlap and vary among sufferers. The only way to address these issues will be with further research.

A number of treatments are advocated, but they all need to be tested in properly controlled trials, too few of which are in progress.

4. We need to resource treatment adequately

In the past we had not devoted sufficient resources to treat post-infectious or immune syndromes such as myalgic encephalitis or chronic fatigue syndrome (ME/CFS). These syndromes are defined by excessive fatigue, which worsens after exertion, and include cognitive impairment or brain fog.

These symptoms are responsible for much of the disease and disability associated with long COVID in many people and are often more prevalent than respiratory problems.

Severe cases of long COVID, ME/CFS and other post-infectious syndromes may include a condition known as POTS (postural orthostatic tachycardia syndrome), where an excessive heart rate results in falls in blood pressure with even mild exertion.

Read more: Fatigue after COVID is way more than just feeling tired. 5 tips on what to do about it

Unlike survivors of COVID pneumonia, people with POTS are much less likely to benefit from traditional exercise-based rehabilitation programmes. Such programs may even exacerbate their symptoms.

Instead, approaches are proposed that have been previously applied to ME/CFS and POTS are proposed, including exercise pacing, and medication regimes. However, there is a lack of good evidence, and while trials are underway, they are few, especially in Australia.