Anxiety in Australian Men Stays Stubbornly High as Drug Use and Money Worries Drive Severity Over Time

Generalised anxiety disorder has long been framed as a condition that affects women more often than men, but a sweeping new longitudinal analysis of Australian males suggests the picture is far more dangerous than prevalence statistics imply. While men report anxiety disorders at lower rates than women, those conditions feed directly into the country’s grim male suicide statistics. Now, researchers tracking more than 18,000 observations from the Ten to Men cohort study have mapped how self-reported anxiety symptoms and their severity changed among Australian men between 2020 and 2025, and which modifiable factors pushed those symptoms upward over time.

The study, published in BMC Psychiatry, drew on three waves of the Ten to Men cohort, a nationally representative longitudinal study of Australian males established in 2011 to build the evidence base on male health. Using a stratified multi-stage cluster random sampling design, the cohort covers men aged 18 years and older across the country. Anxiety symptoms were measured with the Generalised Anxiety Disorder instrument, better known as the GAD-7, a seven-item screening tool that asks respondents how often they have been bothered by problems such as feeling nervous, being unable to stop worrying, or worrying too much about different things over the previous two weeks.

At first glance, the headline numbers look almost reassuring. The prevalence of reporting at least one self-reported anxiety symptom decreased slightly over the study period, from 71.7 percent to 70.4 percent. The mean GAD-7 score also edged down, from 3.65 plus or minus 4.36 to 3.61 plus or minus 4.25. But those small declines mask a stubborn reality: roughly seven in ten Australian men in the cohort reported at least one anxiety symptom at every wave, a figure that mental health advocates will find alarming given how rarely men seek help for psychological distress.

To untangle the factors behind these symptoms, the research team deployed a statistical approach known as two-part hurdle mixed-effects models. The technique is well suited to GAD-7 data because such scores are characteristically zero-inflated: many respondents report no symptoms at all, while others report a wide range of severity. The hurdle model therefore splits the analysis into two questions. The first part examines which factors are associated with the presence of any anxiety symptom, essentially the odds of crossing the hurdle from zero. The second part models the severity of symptoms among those who report them, using mean ratios to quantify how much each factor amplifies the GAD-7 score. Because the same men were followed across waves, the mixed-effects framework also accounts for repeated observations within individuals, while interaction terms with time allowed the researchers to test whether the effects of specific factors changed over the study period.

The results paint a detailed portrait of who is most at risk. Younger age, a non-Culturally and Linguistically Diverse background, financial hardship, being married or in a de facto relationship, harmful alcohol use, gambling, lower subjective well-being, lower perceived social support, sexual difficulties, adverse life events, and living with a disability were all associated with a higher likelihood and/or greater severity of self-reported anxiety symptoms. Some of these associations may surprise readers. The finding that married or de facto men reported more anxiety symptoms challenges the common assumption that partnership is uniformly protective for male mental health, although the authors note that these are observational associations rather than proof of causation.

The most striking findings, however, concern the time-varying effects. Increases in illicit drug use, high-risk alcohol consumption, and adverse life stress events over time were associated with greater anxiety symptom severity among the men followed. In other words, it is not merely whether a man uses drugs or drinks heavily at a single point in time, but whether these behaviours intensify across the study period that matters for how severe his anxiety becomes. Financial hardship also emerged as a modifiable factor whose association with symptoms changed over time, reinforcing the link between economic stress and male psychological distress that has become increasingly visible in the post-pandemic era of rising living costs.

The authors argue that illicit drug use, high-risk alcohol consumption, adverse life events, and financial hardship are modifiable factors that could be the primary focus of intervention programs targeting men’s anxiety. Beyond these four headline targets, the study highlights other modifiable levers for developing timely, targeted and evidence-based interventions: subjective well-being and perceived social support. Both are factors that health systems and community programs can actively cultivate, whether through social connection initiatives, peer support programs, or well-being screening in primary care settings that men actually attend.

Methodologically, the study represents a significant advance over the cross-sectional designs that have dominated research on men’s anxiety. Most existing studies capture a single snapshot, offering limited insight into how anxiety symptoms and their risk factors evolve. By following the same men across three waves and modelling both the presence and severity of symptoms with hurdle mixed-effects models, the researchers could distinguish factors that raise the odds of having any symptoms from those that intensify existing symptoms, and could test whether those effects shifted over time. The intraclass correlation coefficient, likelihood ratio tests, and information criteria such as the AIC and BIC were used to compare model specifications, and multiple imputation was employed to handle missing data, standard techniques for longitudinal cohort analysis.

The researchers are careful to flag an important caveat. The time-varying findings were based on borderline interaction effects among multiple tests and should be interpreted as exploratory, requiring replication in future studies. In plain terms, when researchers run many statistical tests, some interactions will appear significant by chance, and borderline results demand confirmation before they can be treated as established. This candour is a strength of the paper, but it also means clinicians and policymakers should view the time-varying conclusions as promising leads rather than settled science.

Even with that caution, the implications for men’s health policy in Australia are considerable. The persistence of high symptom prevalence across the cohort, combined with the identification of modifiable drivers such as substance use, financial strain, and eroding social support, gives intervention designers concrete targets. Given that anxiety conditions contribute to higher suicide rates among Australian men despite their lower reported prevalence compared with women, programs that address these modifiable factors could have effects far beyond symptom reduction. The Ten to Men study, now spanning more than a decade of male health data, continues to provide the evidence base that such programs need, and this analysis adds a longitudinal dimension that snapshot studies simply cannot deliver.

Subject of Research: Generalised anxiety disorder symptoms and severity among Australian men tracked longitudinally in the Ten to Men cohort study

Article Title: Longitudinal changes in generalised anxiety disorder symptoms, severity and associated factors among Australian men: insights from the Ten to Men study

Article References: Sheferaw, W. E., Mohanty, I., Mengesha, Z., Gebremedhin, T. A., & Niyonsenga, T. (2026). Longitudinal changes in generalised anxiety disorder symptoms, severity and associated factors among Australian men: insights from the Ten to Men study. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08710-w

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08710-w

Keywords: generalised anxiety disorder, GAD-7, Australian men, Ten to Men study, longitudinal cohort, mental health, illicit drug use, alcohol consumption, financial hardship, social support, BMC Psychiatry, hurdle mixed-effects model