Long COVID in Young People in Victoria, Australia: How Common It Is, Its Effects, and Related Factors

Research on the prevalence and impact of long COVID among young people in Australia could result in greater awareness of its effect on young people [10]. This may assist in providing care and support for young people experiencing long COVID symptoms. Better understanding of how social and behavioral factors influence long COVID risk could direct efforts toward healthcare equity. Our study aims to estimate the prevalence and impact of long COVID among young Victorians and identify differences in demographic and behavioral characteristics between young people with long COVID and those who had COVID‐19 without prolonged symptoms.

This study used data from the Sex, Drugs, and Rock‘n'Roll 2023 survey [11]. This is an annual cross‐sectional survey conducted online since 2015. The survey included 61 questions on multiple health topics, including 12 questions on COVID‐19 and long COVID (see Table S1). The survey was hosted in Research Electronic Data Capture (REDCap) [12]. The survey took approximately 20 min to complete.

Recruitment occurred online from April to June 2023, targeting people aged 15–29 years from Victoria, Australia. Victoria had a unique experience during the COVID‐19 pandemic, with multiple long and strict lockdowns [13]. There was minimal community transmission until the Omicron wave in January 2022 [14]. By then, most of the population had already been vaccinated against COVID‐19 [15]. At the time of data collection, COVID‐19 was still circulating in the community [14].

Individuals were eligible to participate in the survey if they were aged 15–29 years and provided informed consent. Participants were excluded if they did not report the outcome variables (history of COVID‐19 and long COVID) or gave inconsistent responses to these variables (e.g., stating no history of COVID‐19, however, currently experiencing long COVID). Although all participants were included in the initial sample analysis, only those with a history of COVID‐19 were included in the subsequent analyses that specifically explored long COVID.

Individuals were recruited online through paid advertisements on Facebook, Instagram, in online community groups, and in the researchers’ networks. COVID‐19 was not mentioned in the advertisements. Participants who completed the survey entered a draw to win A$250.

Summary statistics described the sample's demographic and behavioral characteristics, as well as its characteristics related to COVID‐19 and long COVID. Participants were classified into three groups: no reported COVID‐19, short COVID, or long COVID.

The relationship between the primary outcome (long COVID vs. short COVID) and other variables was estimated using logistic regression. Bivariable logistic regression analyses were first conducted with each explanatory variable (Table S2). Variables with a p value less than 0.1 in the bivariable logistic regression models or those previously associated with long COVID in the literature were included in the multivariable logistic regression model. Stepwise modeling offered an efficient approach to select the key variables, particularly in this context with a large number of potential predictors [18]. The explanatory variable “Aboriginal and/or Torres Strait Islander ancestry” was dropped from the multivariable model as such a small sample size (n = 9) might not provide reliable statistical insights. The explanatory variable “number of affected family or friends that needed participant's support” was also dropped as it was highly correlated with “number of family or friends with long COVID.” Participants with missing data were excluded from the logistic regression models. Stata version 17.0 [19] was used for the data analyses.

We conducted a sensitivity analysis to understand the impact of using the NICE definition of long COVID (persisting for ≥4 weeks) [1] over the WHO definition (persisting for ≥12 weeks) [2]. In this analysis, participants reporting symptoms for fewer than 3 months were not considered to have had long COVID. An alternative prevalence was calculated, and factors associated with long COVID with this alternative definition were determined.

From 1555 people who clicked the survey link, 784 (50.8%) participants provided consent and completed the survey. Of these, 19 records were excluded due to not meeting the age criteria (n = 9), no reported outcome variables (n = 9), and inconsistency in responses (n = 1). The final analysis included 765 young people aged 15–29 years. The sample's demographic and behavioral characteristics are summarized in Table 1. Table S3 shows the sample characteristics categorized by group (no reported COVID‐19, short COVID, and long COVID).

Of 765 participants, 98.2% (n = 751) had received at least one dose of the COVID‐19 vaccine. Of all participants, 79.6% (n = 609) reported having had one or multiple COVID‐19 infections (Table 2).

The prevalence of long COVID was 11.2% (n = 86). At the time of the survey, 5.6% reported currently experiencing long COVID (n = 43). Most long COVID cases were self‐reported (89.5%, n = 77), whereas only 10.5% (n = 9) of participants with long COVID symptoms were diagnosed by a health professional. Participants currently experiencing long COVID symptoms had a mean symptom duration of 10.3 months (SD 5.3), whereas the group that had already recovered from long COVID had symptoms for an average of 4.7 months (SD 3.3). The fraction of participants with long COVID increased with a higher number of COVID‐19 infections (one infection: 47/406; two infections: 27/164; three infections: 9/30; four infections: 2/5; and five or more infections: 1/4).

Figure 1 shows the impact of long COVID on participants’ lives. Most participants with short COVID reported that their current health was about the same as before their first COVID‐19 infection. Conversely, most participants with long COVID reported a worsened general health (Figure 2).

Long COVID was negatively associated with older age (Table 3). Participants with long COVID showed greater concern of getting long COVID compared to participants with short. Participants with long COVID had significantly more family or friends with long COVID than participants with short COVID. Participants with long COVID had significantly higher odds of reporting worse general health compared to participants with short COVID. No other variables were significantly associated with long COVID.

With a definition of long COVID as persisting for 3 months or more, the prevalence of long COVID in the sample was 9.8% (n = 75). Factors associated with long COVID using this definition are shown in Table S4. These factors were consistent with the original analysis, except for age that was no longer associated with long COVID (adjusted odds ratio [aOR]: 0.93; 95% confidence interval [CI]: 0.86–1.01).

This is one of the first studies to measure the prevalence, impacts, and associated factors of long COVID among young people in Victoria, Australia. Of 765 participants, 11% reported they had experienced long COVID, with substantial impacts on their daily life and general health. Most young people who reported experiencing long COVID had not been diagnosed by a healthcare professional and had experienced symptoms for an average of 10 months.

Our study's prevalence of 11% aligns with the prevalence of 14% observed in a survey of 3510 Australian adults as of August 2022 [3]. Both studies defined long COVID as COVID‐19 symptoms persisting for more than 4 weeks. Compared to international data, our study's prevalence is higher than findings from a large‐scale US survey that reported a 5.4% prevalence in young people aged 18–29 years as of October 2023 [4], as well as the prevalence of 1.4% found in the UK among young people aged 17–24 years as of March 2023 [5]. However, both these studies used a definition of long COVID as persisting for ≥12 weeks, compared to our study that used a 4‐week cutoff. They also only included those with a formal diagnosis of long COVID. Both of these criteria would result in a higher prevalence observed in our study.

Our results suggest that long COVID had considerable impact on young people's daily lives, for example, through difficulties in performing work and exercising. Other studies have documented similar impacts, with one Australian study finding that 86% of adults with long COVID reported a reduced ability to carry out daily activities [3]. Unlike previous research [20], our study found no association between long COVID and mental well‐being. This may be due to the high proportion of participants (66%) reporting a mental health condition. The discrepancy with other research could also be attributed to our use of the SWEMWBS [17], a scale that primarily measures positive mental health. A positive focus in research shifts the narrative from solely treating disorders to promoting mental strengths and emphasizing resilience. This approach is crucial, but it might overlook the negative mental health effects of long COVID.

Our study unexpectedly found that younger age was associated with long COVID. This finding differs from other research that suggests an inverted U‐shaped relationship between age and long COVID prevalence [21]. Studies indicated an increasing prevalence from childhood [8] through each decade of adult life, followed by a decline around 60 years [21]. However, the relationship was not significant in sensitivity analysis, and it is possible that the finding was due to chance or bias.

Participants with long COVID showed greater concern about getting long COVID than participants with short COVID. This finding may be due to participant's personal experiences with the condition and being more aware of the potential health impacts. Qualitative research suggests that people with long COVID experience unpredictable symptoms [1] and feel uncertain they will ever recover [22, 23]. Research also suggests that long COVID patients face barriers to accessing care, leading to negative impacts on emotional well‐being and recovery [24]. More qualitative research is needed to understand young people's personal experiences with long COVID, access to support services, and the long‐term impacts.

Participants who reported experiencing long COVID had more friends or family members who also had long COVID. This finding may be due to young people seeking support from others experiencing the same condition. Online peer support groups for long COVID were created throughout the pandemic to address gaps in professional care and to advocate for greater awareness of long COVID [25, 26]. Additionally, discussing their long COVID experiences could have prompted others in their social circle to share similar stories. Participants with long COVID could have been more attuned to similar cases, creating the impression of knowing more people with the condition. Network studies could help clarify if the apparent higher prevalence of long COVID within certain social circles is due to increased awareness or actual case clustering.

Gender and sex were not associated with long COVID. Some studies highlighted female sex as a potential risk factor for long COVID in adults [27, 28]. This trend is less evident in adolescent populations [8, 29]. The underlying causation of this gender disparity in long COVID risk remains unclear. Biological (sex) factors might explain why female adults are more at risk for long COVID. For example, females showed a more robust immune activation than males in acute COVID‐19 [30], possibly contributing to the persistence of long COVID symptoms in females. Sociocultural (gender) roles could also play a crucial part. Cisgender women, who more predominantly occupy roles in healthcare and caregiving [31], might face increased viral exposure. Additionally, gender‐based differences in healthcare‐seeking behavior [32] might lead to more frequent reporting of long COVID symptoms among cisgender women. These factors may manifest differently in younger populations, including our study group with an average age of 23. Further research in both biological (sex) and sociocultural (gender) differences regarding long COVID in this age group is required to deepen the understanding of this relationship.

Although not included in multivariable analysis due to small numbers, there was a markedly high proportion of participants who identified as Aboriginal and Torres Strait Islander reporting long COVID (21%, 4/19). This demonstrates a need for focused and community‐led research and management strategies for long COVID among Aboriginal and Torres Strait Islander people.

Finally, COVID‐19 vaccination is a known protective factor against long COVID in children and adults [33, 34, 35]. Our study observed no association, likely because nearly all participants (98%) were vaccinated.

Our study demonstrates that a significant minority of young people in Victoria reported experiencing long COVID, with many experiencing symptoms for >10 months. A recent review identified there are a small number of services in Australia that provide long COVID rehabilitation and support to patients to manage symptoms [10]. However, many have since been closed [36]. Most services require a referral from a general practitioner or specialist and include support from allied health professionals like physiotherapists and psychologists. Our findings suggest that young people with long COVID will likely also require support from their employers and education institutes to enable flexible work and study arrangements [37].

Most of our participants had not been diagnosed with long COVID by a healthcare professional; however, we do not know if this is because they had not sought diagnosis for the condition, or if they had sought care but not received a formal diagnosis. Further work may be needed to ensure that general practitioners are confident in diagnosing long COVID or providing alternative diagnoses, and that young people with chronic conditions can navigate the healthcare system and access support when needed. Long COVID research shows that many people with the condition feel invalidated, dismissed, or disbelieved when seeking help from medical professionals [38, 39, 40, 41]. This research shows that feeling heard and believed can be as important for well‐being as any medical treatment. Clinicians can support patients by listening to and validating patients’ concerns and sharing their own frustrations with lack of scientific and clinical knowledge around long COVID [38, 39]. Long COVID patients should be offered multidisciplinary patient‐centered integrated care and active self‐management [10, 42]. However, more research into treatment and prevention of long COVID is needed [43].

Given its prevalence and impact, work is needed to raise awareness and educate the public about long COVID. A longitudinal cohort study reinforces this by showing that only one‐third of its participants considered the Victorian government's information on long COVID sufficient [44]. Increasing awareness about long COVID could facilitate earlier detection and timely, effective support, while setting realistic expectations and reducing the stigma around long COVID [22].

Our study has several limitations. First, COVID was self‐reported, which may have led to over‐ or underreporting and misdiagnosis. Diagnosis was based on a single item rather than a validated diagnostic tool or questionnaire. Given that even healthcare professionals find diagnosing long COVID challenging [45], people without medical training might have attributed symptoms of other conditions to long COVID. However, patient‐reported outcomes are central to identification of long COVID, and self‐reported outcomes are used commonly in long COVID research [40, 46]. Second, our study was cross‐sectional, and we cannot establish a temporal relationship between variables. We are also unable to verify onset dates and duration of symptoms. Implementing a national, longitudinal screening program for long COVID in Australia, modeled after successful international initiatives [4, 5], is recommended to follow its progression over time. Third, our study used convenience sampling; therefore, it is unlikely participants represent the general population of young people in Australia. Recruiting participants online means we likely did not reach participants with lower levels of education and digital literacy. Studies have shown that those with greater digital health literacy are more likely to take precautions to prevent COVID infection, which may have affected our reported prevalence rates [47, 48]. A strength of the study was that it was not advertised as a COVID‐19 study, so selection bias towards those with strong feelings about long COVID was likely minimized. However, the advertising of the survey topic as “Sex, Drugs, and Rock‘n'Roll” means there may have been a bias towards volunteers of particular social and cultural groups, for example, we have a high number of participants identifying as LGBTQIA+. Although there is no evidence that LGBTQIA+ young people have disproportionate rates of long COVID, we have previously shown that this group experienced greater psychological distress during the pandemic [49].

The data that support this study cannot be publicly shared due to ethical or privacy reasons and may be shared upon reasonable request to the corresponding author if appropriate.