The 30 studies with a focus on specific subpopulations included 25 quantitative studies (with the majority based on survey data and five based on administrative data), four qualitative studies and one mix-method study. Two of the four qualitative studies (Digby et al., 2021; [
19,
24]) reported the qualitative findings of mixed-methods research, with the quantitative findings reported elsewhere.
Study samples and populations of interest
Of these 30 studies, 20 studies collected data from participants across the nation (although one comprised largely of people living in Victoria (88.2%)) [
57]. Only Sollis et al. [
61] and Broadway et al. [
10] were based on survey data analysed from nationally representative samples, and Johnston et al. [
36] pre-stratified their data/sample to approximate a nationally representative sample. The remaining ten studies focused on specific states or cities. One focused on South Australian [
67]; one on Queensland [
39]; two studies were conducted in Western Australia [
22,
41]; and two studies in Sydney or New South Wales [
43,
60]. Four studies were conducted in Melbourne or Victoria [
15]; Digby et al. 2020; [
20,
33].
People with a particular vulnerability were a major focus of these studies. They included patients presenting to and/or staying in hospital due to poor health or mental health in the study period [
15,
22,
60]; people with a pre-existing physical or mental health disorder [
52,
68]; and people accessing mental health services [
63,
65,
67]. Leske et al. [
39] studied suicide rates and motives during the pandemic. Hospital staff, whose physical and mental health may have been more vulnerable during the pandemic, were the population of interest in three studies (Digby et al., 2021, [
19,
20,
33]. Other potential participant vulnerabilities included being an adolescent or young adult [
40,
41,
43], in self-isolation/quarantine [
35], living alone [
46] and having higher dysmorphic concern [
55].
Families with young children were considered vulnerable and therefore a population of interest in nine studies. Six studies drew data from the COVID-19 Pandemic Adjustment Survey which was conducted among parents of children under the age of 18 (see Table
2). Two studies drew data from other nationwide surveys [
36,
10]. Additionally, Chivers et al. [
17] conducted a qualitative research on new and expecting parents.
Pre-COVID/ during-COVID study comparisons
As indicated in Table
2, 15 of the 30 studies reported on changes in mental health and other wellbeing indicators before and during the COVID-19 outbreak. Most studies investigating specific populations were cross-sectional and compared current results with the results or statistics from pre-COVID studies that used similar samples (or comparable admissions/administrative data). Other studies asked participants to self-report on the differences in their mental health before and during the pandemic. Four studies reporting administrative data from health services [
15,
22,
63,
65] selected data collected during the corresponding period of 2019 as their pre-COVID comparisons (to avoid the period immediately before the pandemic when Australia experienced the severe bushfire crisis). One longitudinal study tracking the same cohort of participants [
43] adopted a cut-off date to compare mental health before and after the implementation of the COVID-19 restrictions. Separate from the pre-COVID comparisons, four studies [
15,
22,
30,
63] compared mental health across multiple time points during the pandemic, linking changes in participants’ mental health to changes in case rates for COVID-19 in Australia.
Mental health measures
Similar to studies focused on the whole general population, most of the subpopulation studies measured participants’ mental health and wellbeing using validated scales such as the K6, K10, PHQ-9, GAD-7 and the DASS-21. A series of other mental health measures were also adopted (see Table
2). Apart from the validated mental health measures, behaviours related to mental health, including eating and exercise behaviours [
52], and appearance-focused behaviours [
55], were also adopted as mental health indicators. Several studies examined public or administrative records, including emergency department presentations [
15,
22], suicide registers [
39] and website visits and call centre traffic for mental health services [
65,
63]. A small number of studies did not use validated measures and instead asked participants to self-report on their mental health, lowering the quality of mental health measurement in these studies (e.g. [
10,
35,
36,
41,
43]). None of the sub-group studies assessed the widespread and likely traumatic impact of the 2019–20 bushfires (a significant individual and community-level pre-pandemic vulnerability).
Five studies qualitatively assessed participants’ descriptions of their experiences and feelings during the COVID-19 pandemic [
17], Digby et al., 2021; [
19,
24,
46,
60] to gain a deeper understanding into participants’ psychological wellbeing in relation to their specific contexts. Of the five studies, Chivers et al. [
17] analysed posts related to COVID-19 in an online parenting forum to understand perinatal distress. Shaban et al. [
60] conducted bedside interviews of COVID-19 patients to explore their lived experiences and perceptions. The other three studies added open-ended questions asking about participants’ concerns related to COVID-19 in their surveys.
Overall study findings
In general, the studies investigating specific subpopulation groups showed similar patterns to the findings of the studies on the general population – mental health and wellbeing deteriorated with the emergence of the COVID-19 pandemic and associated restrictions. This trend is consistent across the different populations of interest. However, it is also apparent that important population groups, such as Indigenous and CALD (Culturally and Linguistically Diverse) groups were not researched, limiting our knowledge for these groups. Psychological distress was reported widely among hospital staff in the two studies that measured hospital workers’ mental health [
20,
33]. Three studies focusing on adolescents and university students consistently showed higher psychological distress and lower subjective wellbeing since the COVID-19 outbreak [
40,
41,
43]. Studies focusing on parents with young children identified a range of mental health challenges and risks during the COVID-19 period, and the three studies that included a pre-COVID comparison indicated that psychological distress increased [
10,
70,
71]. The themes identified from the qualitative studies differed as they were specific to the experiences of each subpopulation group. However, participants in these studies acknowledged the impact and the challenges brought by the COVID-19 pandemic and expressed worry and concerns (refer to Table
2 for details).
The two studies [
30,
63] reporting on participants’ mental health several times across the pandemic showed similar results to Biddle et al.’s [
7] study of the general population. Griffiths et al. [
30] focused on working adults and Staples et al. [
63] focused on consecutive users of digital mental health services during the pandemic. Corresponding with Biddle et al. [
7], both studies found that declines in mental health appeared to be more significant during March to April, and then improved in later months (returning normal levels) (except for the Victorian participants in Griffiths et al. [
30]).
In contrast to the consistent findings from survey data showing increases in common mental health problems (i.e. psychological distress, depression and anxiety), two studies analysed data on emergency department (ED) presentations during the pandemic and showed varying results. Cheek et al. [
15] found that mental health presentations potentially increased,while Dragovic et al. [
22] found that the total number of mental health presentations decreased and that the trend varied depending on the reasons for the presentation. A decrease in ED presentations is not surprising given that face-to-face access to many health services declined during the pandemic (as people restricted their mobility) [
5] – and thus, actual service use during this time does not likely reflect the need for services in the community. Importantly, according to data from AIHW [
5], mental health related services, particularly services delivered online or via phone showed heightened service usage since the restrictions were introduced. The contrast between the two studies is likely because they were based on data from two different states with different COVID-19 responses, and Cheek et al. [
15] only included paediatric patients.
In terms of suicidal intention, plans or behaviours, data from Queensland showed no change in suspected suicides [
39] and in Western Australia, the presentations to emergency departments due to suicide or self-harm decreased significantly during this period [
22]. On a national level, those who accessed digital mental health services during the pandemic also showed no changes regarding suicidal thoughts or plans [
63].
Several potentially positive experiences related to the COVID-19 situation were identified from existing studies. Many individuals and families practicing isolation/social distancing reported some “silver linings”, such as strengthening relationships with their families, enjoying spending time at home, and developing new hobbies [
24,
35]. Patients with COVID-19 who were in isolation also reported some positive factors [
60]. For example, although patients reported that they were disconnected from the outside world, lost track of time, and had limited mobility, some saw this as a reflection of the professionalism and quality of care provided. This enhanced their confidence and helped to ameliorate their initial concerns about being infected. Positive experiences were also identified as potential indicators of resilience and helped to mitigate the negative effect of the pandemic and restrictions on mental health [
20,
35,
42,
24]. For example, Oliva & Johnston’s study [
24], showed the mental health benefits of having a dog during the lockdown, likely because it encouraged exercise and provided an opportunity to socialize with other people.
Several studies made comparisons between specific population groups and the general population, or other population groups. These studies provide insights into which population groups might be at greater risk of experiencing mental health problems, and what factors were protective during the pandemic. Specifically, Broadway et al. [
10] showed the protective effect of having two earners in the family in times of uncertainty. Phillipou et al. [
52] found that individuals previously diagnosed with eating disorders experienced more mental health problems compared to the general population while people with high and low dysmorphic concern displayed different psychological and behaviour responses to the shutdown of the beauty industry in the COVID-19 lockdown [
55].