Background

Recognition of mental health as a critical factor to one’s overall well-being has been on the rise as society as a collective has begun to prioritize mental health due to its multifaceted impact on our daily lives [1]. There is a wide array of factors that can contribute to the detriment of one’s mental health, some of which include poor working and living conditions, personal and familial stressors and traumatic events [1,2]. Similarly, poor mental health conditions can have adverse effects on one’s personal and work lives. Mental health has become a prominent issue on a global scale, with an estimated 15% of adults of working age having a mental disorder in 2019, potentially leading to adverse financial impacts and productivity loss [3]. To address these concerns, consultations with mental health professionals, such as psychiatrists, family doctors and psychologists, play an essential role in providing diagnosis, treatment, and support. However, the way these consultations are delivered, as well as whoever offers them, can significantly impact the quality, accessibility, and consistency of care patients receive, ultimately influencing their mental health outcomes.

While many mental health professionals have been providing care in-person for years, recent changes due to the COVID-19 pandemic have led to a significant shift towards a technological approach to healthcare, known as telehealth [4]. In mental health, this typically refers to live therapeutic sessions between the clinician and the patient, conducted via telephone or video conference [5]. Although telehealth services existed before the pandemic, they were not widely utilized [5]. However, at the start of the pandemic, many countries reduced their in-person services, leading to a sharp increase in the use of telemental health [6]. Within a year, both general and specialized care had become almost completely virtual [6]. For instance, in the United States, approximately 88.1% of mental health treatment facilities offered telehealth services in September 2022, a significant increase from 39.4% in April 2019, indicating the expansion of telehealth availability for mental health services and care due to state-level policy changes during that period [7]. This diversity in service provision offers numerous benefits, including greater accessibility for individuals in rural areas, reduced costs, and fewer barriers for those with disabilities [8–10]. Additionally, this new method of offering mental health services has been revolutionary, with many choosing to continue offering services over various modalities, even after in-person services resumed [5].

Given the recent emergence of telehealth services, it is crucial to determine whether the modality of service provision and the type of mental health professional have a significant impact on self-reported quality of care and perceived mental health outcomes. A systematic review in the literature compared the efficacy of telehealth services and in-person services for improving mental health outcomes [11]. The results of the review found that mental health outcomes were significantly improved when the intervention was delivered by a psychologist or psychiatrist via telehealth strategies as compared to face-to-face interventions [11]. Similarly, telehealth appears to be particularly beneficial for social workers who care for vulnerable populations.

The use of technology to provide support and therapeutic interventions reduces many systemic barriers that may have otherwise prevented access, including transportation, geographic limitations, and socioeconomic difficulties [12]. Moreover, a systematic review highlighted positive patient experiences regarding remote consultation [13]. For instance, 66.4% of patients receiving mental health care reported satisfaction with remote consultation; however 86.2% were more satisfied with in-person care. Remote consultations were associated with reduced anxiety, often due to shorter wait times or increased flexibility [13]. Some patients also felt more comfortable communicating through textual and asynchronous methods, allowing them to express themselves more clearly. However, three studies from this systematic review reported negative patient experiences, with some finding remote consultations to be more time-consuming. In contrast, patient satisfaction was found to be comparable between telehealth and in-person therapy services in another study [14]. However, there were some reports of discomfort regarding online group sessions [14]. This suggests that while mental health outcomes and perceived quality of care appear to be consistent or improved through telehealth, some forms of intervention might be better suited for private, in-person sessions, having implications for therapists who offer a greater variety of services, including group therapy.

In addition to consultation modality and type of mental health professional, individual factors such as employment status may also influence how people perceive mental health outcomes. Employed and unemployed individuals may perceive quality of care and mental health differently. One study, for example, found that unemployed individuals, compared to those with full-time employment, were more likely to report lower self-perceived mental health [15]. Similarly, another study found a positive link between employment and mental health for both women and men [16].

Previous studies in the literature have examined consultation modality, type of mental health professional, and employment status, either individually [17] or in combination with two of these factors [18]. One study, for example, investigated whether employment status influenced the type of professional consulted (general practitioner or psychiatrist); however, it was analyzed in terms of countries with high mean unemployment rates compared with those with lower rates, rather than individual employment status [16]. To our knowledge, no single study has examined the combined effects of these two factors and explored how they interact with individual characteristics, such as employment status. This highlights a gap in the literature requiring the need to examine the interactions between mode of consultation, mental health professional, and employment status, which is essential as it can inform mental health services by highlighting which types of mental health care and delivery methods support better outcomes for employed and unemployed individuals. The results from our work would aid in the development of effective targeted interventions for these individuals based on their employment status.

Theoretical framework

This study is guided by both the Donabedian’s Quality of Care Framework and Andersen’s Behavioural Model of Health Services Use, allowing for a more comprehensive lens for examining differences in mental health care experiences. Donabedian’s framework evaluates the quality of healthcare across three domains: structure, process and outcome, where structure is defined as the aspects of a setting where care occurs, process is how care is delivered, and outcome is the effects of that care [19]. This is a very relevant and important guide to our study as it helps us understand that the type of mental health professional providing care represents the structure domain, the type of consultation modality reflects the process domain, and perceived quality of care and mental health outcomes are part of the outcome domain. Andersen’s Behavioural Model of Health Services Use adds on to this perspective by highlighting the role of predisposing factors, enabling resources and need in shaping how health services are used and their outcomes [20]. In the context of this study, employment status is seen as both a predisposing characteristic and an enabling resource that influences access to care and service utilization. Therefore, combining these two frameworks provides a strong foundation for the study, helping to establish patterns in the perceived quality of care and mental health outcomes, while also offering potential explanations for the findings.

Objectives

This study aims to examine the relationship between the type of consultation (in-person, remote, or hybrid) with different types of mental health professionals (e.g., psychiatrists, psychologists, family doctors, and social workers) and the quality of care and self-reported mental health among employed and unemployed Canadians. In the context of this study, family doctors are considered as mental health professionals alongside psychiatrists and psychologists, due to their role in assessing, treating and referring patients for mental health concerns [21–23].

Ethics statement

Ethical approval was not required for this study at our institution as the data used is already publicly available. Please see https://research.utoronto.ca/ethics-human-research/activities-exempt-human-ethics-review.

Dataset

This study employed a cross-sectional, secondary data analysis of the 2024 Mental Health and Access to Care Survey (MHACS) dataset from Statistics Canada to address the research question [24]. Secondary data analysis is a well-utilized approach that can examine population-level trends through the use of large representative datasets [25,26]. This dataset is publicly available and was accessed from the Odesi database on April 7, 2025 (https://odesi.ca/en/details?id=/odesi/doi__10-5683_SP3_ABG9CY.xml) [27]. Additionally, the authors had no access to identifiable information regarding individual participants at any time. The survey collected detailed information about the mental health status of Canadians, as well as participants’ access to and need for services and supports, making this dataset appropriate for this study. It has also been utilized in several studies in literature demonstrating its value as a data source for mental health research [28–30]. Furthermore, the MHACS examines how the COVID-19 pandemic affects population health and evaluates changes in patterns of mental health, service utilization, and functioning over the past decade. In this context, functioning refers to the ability to carry out daily activities. Data collection commenced on March 17, 2022, and ended on July 31, 2022. The MHACS employed a stratified simple random sampling method. The survey was structured to oversample individuals identifying with the four largest population groups in Canada, designated as visible minorities (Black, Chinese, Filipino, South Asian, and others, which encompasses the rest of the analytic population). This approach uses a 0.5 power allocation to ensure proportional representation based on the relative size of each group.

Dependent variables

The two dependent variables include perceived quality of care and perceived mental health outcomes. Regarding the quality of care, the survey asked participants to rate the level of help they received from the healthcare professional they sought. Participants had the option to rate it as “A Lot”, “Some”, “a Little”, and “Not at All”. Responses that indicated “Don’t Know” were excluded from the analysis. For measuring perceived mental health, participants were asked in the survey to rate their condition using a 5-point Likert scale with options including “Excellent”, “Very Good”, “Good”, “Fair”, and “Poor”.

Independent variables

The independent variables in the study are the type of mental health professional consulted (psychiatrist, family doctor, psychologist, and social worker) and the modality of consultation (in-person, over the phone, video, and text message). The original dataset included 16 binary variables based on survey questions that asked whether participants had sought help from each professional-modality combination within the past 12 months. For each question, participants, if applicable, indicated “yes” or “no” to denote if they received the specific support method. Consultation modalities were recoded into three broader categories for each healthcare professional (in-person only, remote only, and hybrid), yielding three new categorical variables for each provider type.

Statistical analyses

All statistical analyses were performed using R version 4.5.0 on MacOS. Descriptive statistics were conducted to identify the frequencies of demographic variables, including age, gender, marital status, and visible minority status, as well as the frequencies of professional-modality combinations pursued. Multivariable logistic regression models were used to generate odds ratios and 95% confidence intervals (CIs) for predicting poor quality of care and mental health, based on a series of independent variables. To facilitate analysis, the two dependent variables were dichotomized. For quality of care, responses “a lot” and “some” were combined to indicate an adequate level of care received from their support delivery, whereas responses “a little” and “not at all” were combined to signify the poor level of care. For perceived mental health, responses “Excellent”, “Very Good”, and “Good” were combined to represent high self-perceived mental health, whereas “Fair” and “Poor” were combined to specify poor mental health. Analyses treated quality of care and perceived mental health as dichotomous outcomes to enhance interpretability to relevant stakeholders.

Co-variants were added in the model to adjust for differences across demographic variables including age, gender, marital status, and visible minority status. Employment status acted as a moderator variable to identify the varying relationships across these groups. Statistical differences were deemed significant at the 5% level. The variance inflation factor (VIF) was used to detect any presence of multicollinearity.

Strengths and limitations

To our knowledge, this is the first study to examine perceived quality of care and mental health outcomes across different consultation modalities and professional types, stratified by employment status. Stratifying by employment status enabled us to identify how these outcomes vary between employed and unemployed individuals, facilitating the creation of more targeted recommendations and interventions. The study also benefited from a large dataset, particularly for the employed group, which enhanced the accuracy and precision of the odds ratios for perceived quality of care and mental health. Additionally, the sample ensures proper representation of visible minority groups due to the oversampling of these groups. The use of stratified random sampling design also allowed the findings to generally reflect the Canadian population. Finally, we assessed multicollinearity using VIF, which strengthened the robustness and validity of our analyses.

This study had several limitations that should be considered when interpreting the results, as they may have influenced the reliability of the findings. First, one form of remote consultation, text messages, and two types of mental health professionals, nurses and counsellors, were excluded due to small sample sizes. The exclusion of these variables may have limited the comprehensiveness of the findings and missed out on key insights. Another key limitation is the study’s use of a cross-sectional design where causal inferences cannot be made about the impact of provider type or modality on mental health outcomes. This requires the need for future research to conduct experimental studies to confirm causality. The study could not also confirm whether all participants in the dataset had access to mental health services. This limitation is particularly important as differences in access between employed and unemployed individuals may have influenced the reported outcomes. Similarly, the data collection took place in 2022, which was a period still significantly impacted by the COVID-19 pandemic. This serves as a potential limitation as this unexpected event may have influenced both the perceived quality of care and mental health outcomes. The increased stress and uncertainty during this time may have affected how these individuals perceived their mental health, the care they received and who they sought, regardless of consultation modality and professional type. Additionally, it was the time when hybrid modalities were just starting, which could explain why hybrid options were found to be less utilized in our study. While dichotomizing the dependent variable allows for easier comparison using odds ratios and may help reduce variability, this approach can also reduce statistical power. Lastly, the small sample sizes within subgroups of unemployed participants, specifically those seeking a psychiatrist or psychologist, resulted in wide confidence intervals and reduced precision of the odds ratios for perceived quality of care and mental health.

Future research and implications

Although this study found no significant differences in perceived mental health outcomes between employed and unemployed individuals, notable differences in self-reported quality of care emerged across consultation modalities and provider types. These findings suggest that while mental health outcomes may appear similar across both groups, patients’ experiences and satisfaction with care may differ based on who they consult and what modality was utilized. This is important as treatment adherence, future use of these services and general involvement with the mental health system can all be impacted by the perceived quality of care. These findings also demonstrate the importance of tailoring mental health interventions not only by provider type and consultation modality (e.g., remote, in-person, or hybrid), but also by individual characteristics, such as age, gender and employment status.

Future research should investigate the specific factors contributing to the differences in perceived quality of care between employed and unemployed individuals, such as accessibility and waiting times, by analyzing their experiences through a qualitative lens. Additionally, just as hybrid models that combine remote and in-person care were analyzed in this study as a variable, future research should further explore the combination of specific providers, such as collaborative care between family doctors, psychiatrists, and social workers, and see how this impacts patients’ quality of care and outcomes among employed and unemployed individuals. Interprofessional collaboration may be the key to improving both perceived quality of care and outcomes among these individuals, especially when complex mental health problems are involved. Moreover, given how we had a low sample size for those categorized as unemployed, future studies should aim to recruit larger numbers of unemployed individuals to allow more precision comparisons across mental health professional types.

Furthermore, future mental health services should consider integrating occupational therapists to reduce wait times and improve access. These professionals can also provide valuable assessments of patients’ functional abilities and help identify occupational barriers to recovery. Their inclusion could enhance the quality of mental health care, especially for patients with impairments that affect their work or daily living.