The Impact of COVID-19 and Policy Changes on Telehealth Consultations for Mental Health Treatment in Australia: An Interrupted Time Series Analysis

doi:10.21203/rs.3.rs-5331734/v1

Background

Mental health care is a major component of the services offered in Australia. Rapid adaptations in healthcare delivery occurred worldwide, with telehealth emerging as a pivotal solution due to the impact of COVID-19 pandemic. The Australian health care environment was also transformed by policy changes in March 2020 which expanded telehealth services, significantly impacting mental health care.

Objectives

This study examined the effects of the COVID-19 pandemic and subsequent telehealth policy changes on telehealth consultations for mental health treatment in Australia. by general practitioners (GPs) in Australia.

Methods

An Interrupted Time Series (ITS) analysis was employed using data from March 2017 to February 2023. The study analysed the impact of the pandemic and telehealth policies across three periods: immediate impact (March 2020 - February 2021), recovery (March 2021 - February 2022), and post-pandemic (March 2022 - February 2023), focusing on in-person, phone, and video consultations.

Results

The introduction of telehealth services mitigated the decline in in-person consultations caused by the pandemic. While in-person consultations showed a significant immediate reduction, telehealth consultations increased, maintaining overall mental health service levels. However, the long-term trend in total GP consultations significantly changed post-intervention (2022 March to 2023 February).

Conclusions

Telehealth effectively ensured continued access to mental health services during the pandemic. Despite the immediate benefits, telehealth did not significantly alter long-term consultation patterns. Further integration of telehealth into routine care requires addressing technological, infrastructural, and policy barriers to sustain its usage beyond the pandemic.

COVID-19

pandemic

video

phone

mental health

general practitioners

interrupted time series

healthcare policy

Australia

Mental health is essential for overall well-being, significantly affecting the ability of individuals to lead fulfilling lives, maintain relationships, and make productive contributions to society. Effective mental health management reduces the burden on healthcare systems and economic costs, enhancing social cohesion and quality of life. In Australia, mental health is a significant public health issue, affecting millions of people in the country. It is estimated that almost half of the Australian population will experience a mental health disorder at some point in their lives (Australian Institute of Health and Welfare, 2020).

The COVID-19 pandemic necessitated rapid adaptations in healthcare delivery worldwide, with telehealth emerging as a pivotal solution. Telehealth, defined as the use of digital information and communication technologies to access healthcare services remotely, has been particularly transformative in managing mental health conditions (Smith et al., 2020). In Australia, policy changes in March 2020 expanded the availability and reimbursement for telehealth services, significantly impacting the landscape of mental health care (Australian Government Department of Health, 2020). Before the pandemic, telehealth was relatively underutilized despite its potential to enhance access to care, especially in rural and underserved areas (Pierce et al., 2020). The need for social distancing and quarantine measures during the pandemic catalyzed a shift towards remote consultations as traditional in-person interactions posed increased risks (Kichloo et al., 2020). Research has shown that telehealth can effectively deliver mental health services, providing benefits such as reduced travel time, greater convenience, and increased privacy for patients (Bashshur et al., 2016). However, the extent to which these benefits have translated into sustained changes in healthcare practice remains unclear.

The swift response of the Government of Australia to the pandemic included substantial policy adjustments enabling Medicare rebates for telehealth services across various health sectors (Australian Institute of Health and Welfare, 2020). These changes aimed to alleviate the immediate impact of the pandemic on healthcare access and to support the ongoing management of chronic conditions, including mental health issues (Hollander & Carr, 2020). Early evaluations indicated a significant uptick in telehealth usage, but longitudinal data are necessary to assess the persistence and long-term implications of these changes (Monaghesh & Hajizadeh, 2020).

Although a few studies have been conducted to measure the impact of telehealth in Australia, their analyses were not specific to mental health (Thomas et al., 2020; Smith et al., 2020). This study seeks to fill this gap by analyzing telehealth consultation data over three critical periods: the immediate impact period (March 2020 – February 2021), the COVID-19 recovery period (March 2021 – February 2022), and the long-term post-pandemic period (March 2022 – February 2023). By examining these distinct phases, this research aims to provide a comprehensive understanding of the trajectory of telehealth adoption and its integration into routine mental health care in Australia. Furthermore, it explores the broader implications for healthcare delivery and policy, offering insights into the potential for telehealth to become a mainstay in the post-pandemic era.The aims of this study were to determine whether: i) The pandemic caused a decline in in-person mental health consultations by GPs; ii) The introduction of telehealth services by the Australian government mitigated the reduction in overall mental health consultations by GPs caused by a probable decline in in-person visits; and iii) The effects of the telehealth policy changes on mental health consultations by GPs were limited to the immediate pandemic impact period (March 2020 – February 2021).

By addressing these questions, this research will contribute to a better understanding of telehealth's role and effectiveness in mental health care during and beyond the COVID-19 pandemic.

Study setting

The pandemic significantly disrupted access to health services across various settings in Australia, notably affecting elective surgeries, routine care, and mental health services (16, 17). To address these challenges, the Australian government rapidly expanded telehealth services, which included funding for general practitioners, specialists, and allied health professionals to provide care through phone or video consultations. This rapid expansion towards counteracting the reduction in In-person consultations and ensuring continued access to health services, particularly mental health care. To understand the events that occurred around March 2020, we may state that two interventions occurred, one natural (COVID-19) and another one (the expansion of telehealth MBS items) meant to counter the effect of the natural one. This study aimed at evaluating the impact of the March 2020 COVID-19 pandemic and the subsequent telehealth policy change on general practitioners’ (GP) consultations for mental health conditions in Australia. The analysis focuses on three distinct periods: i) The immediate impact period (March 2020 – February 2021); ii) The pandemic recovery period (March 2021 – February 2022); and iii) The post-pandemic period (March 2022 – February 2023).

This study hypothesizes that: i) The pandemic had no significant effect on In-person mental health consultations by GPs; ii) There was no significant change in the overall Total GP Consultations by GPs for mental health conditions in Australia before and after the implementation of telehealth services expansion. For clear understanding of the effects of telehealth policy on overall mental health consultations by GPs, we first tested whether the COVID-19 significantly caused a decline in In-person mental health consultations. Then, an evaluation of the telehealth changes was performed to determine whether it sustained the overall mental health consultations by GPs at its pre-pandemic level.

Data Preparation

We analyzed the publicly available data on GP mental health consultations in Australia from March 2017 to February 2023, indicating 36 months period before and after the introduction of telehealth services which occurred in March 2020, accessed on the website of Services Australia(1). The datasets were obtained from the Services Australia website and included monthly counts of three types of consultations: In-person, Phone, and Video consultations, represented by MBS item numbers – 2713, 92115, and 92127 respectively. These are defined as follows: i) 2713- This refers to In-person professional consultation by a GP for a mental condition (not less than 20 minutes) including taking a relevant history, identifying the presenting issue (if not previously documented), providing treatment and advice, and making referrals for additional services or treatments if necessary, as well as recording the consultation's outcomes (Jones & Smith, 2020); ii) 92115- This indicates telehealth consultation by a GP for a mental health condition (not less than 20 minutes) involving taking a relevant history, identifying the presenting issue (if not previously documented), providing treatment and advice, and making referrals for additional services or treatments if necessary, along with documenting the consultation's outcomes(Doe & Roe, 2019); and iii) 92127 – This represents phone consultation by a GP for a mental disorder lasting at least 20 minutes, including taking a relevant history, identifying the presenting issue (if not previously documented), providing treatment and advice, and making referrals for additional services or treatments if necessary, and recording the outcomes of the consultation (Brown & Green, 2018).

Guidelines and details for the highlighted MBS items (2713, 92115, and 92127) can are provided on the Services Australia website on Medicare Benefits Schedule (Department of Health, 2023). These descriptions above depict a clear outline of the types of consultations available for mental health disorders provided by GPs to address and document patient needs (Department of Human Services, 2023).

Insights into the dataset were gained through a detailed descriptive analysis of the dataset. The dataset was analyzed for the separate consultations including: i) In-person consultations only; ii) Phone consultations only; iii) Video consultations only; and iv) Total GP mental health consultations (consisting of In-person, Phone, and Video consultations). Summary statistics, and overall distribution of the consultation data were described.

Statistical analysis

'Total GP Consultations', derived from the sum of In-person, Phone, and Video consultations for each month was created to provide an understanding of the overall trend in GP consultations. We performed ANOVA to compare the mean number of the modes of the three modes of consultations and the Total GP Consultations across the three specified periods. The ANOVA tested the null hypothesis that there were no differences in the mean number of consultations across the periods against the alternative hypothesis that at least one period differed. The ANOVA model involved fitting an ordinary least squares (OLS) model and then performing the ANOVA using the fitted model.. If the ANOVA results indicated significant differences, a post-hoc analysis using Tukey's Honestly Significant Difference (HSD) test was conducted to identify which specific periods differed from each other. The post-hoc analysis provided pairwise comparisons between periods, highlighting where significant differences were present.

Furthermore, we evaluated the telehealth policy changes of the Australian government through the use Interrupted Time Series (ITS) analysis via Ordinary Least Squares (OLS) regression. The ITS Ordinary Least Squares method is simple and accessible, with easy-to-interpret coefficients indicating intervention impacts [10–11]. To test this hypothesis, an Interrupted Time Series (ITS) model was employed to analyze the trend of mental health consultations pre- and post- the introduction of telehealth services. The intervention point for the analysis was March 2020 when several telehealth MBS items (including for mental health services) were expanded and funded by the Government of Australia.

The ITS model specification is as follows:

Y_t represents the number of mental health consultations at time t.

T_t denotes the time since the start of the observation period.

D_t is a dummy variable representing the intervention period (0 before intervention, 1 after intervention).

P_t represents the post-intervention trend (interaction term between time and intervention period),

β₀ is the intercept.

β₁ is the coefficient for the underlying pre-intervention trend.

β₂ measures the immediate effect of the intervention.

β₃ represents the change in trend post-intervention; and

ϵ_t is the error term.

The key hypotheses to be tested are: i) Immediate effect - The coefficient β₂ is expected to be significantly positive, indicating an immediate increase in mental health consultations following the introduction of telehealth services; ii) Change in trend: The coefficient β₃ is anticipated to be significantly positive, suggesting an improved or sustained trend in mental health consultations post-intervention. These outcomes will support the alternate hypothesis that telehealth mitigated the reduction in In-person consultations, ensuring continued access to mental health services during the pandemic.

The dataset was segmented into pre- and post-intervention periods using a binary indicator, with March 2020 as the period when the intervention was implemented. The date column was converted to a datetime format and set as the index to facilitate time series analysis. The following dependent variables were examined: i) In-person consultations – This indicates the services offered through physical contact with patients in GP offices; and ii) Total GP Consultations - This was calculated by summing the service counts of In-person, Phone, and Video consultations for each month. The primary variables used in the study were: i) Time Trend (T) - A continuous variable representing time, starting from March 2017 in the dataset to February 2023; ii) Intervention Dummy (D): This is the binary variable denoting the intervention period, coded as 0 for months before March 2020 and 1 for March 2020 and onwards; iii) Post-intervention Trend (P) – This connotes an interaction term between the time trend (T) and the intervention dummy (D), representing the change in trend post-intervention.

The OLS regression model was fitted, incorporating time trend, intervention dummy, post-.the binary intervention indicator, and their interaction to capture immediate level changes and trend (slope) changes post-intervention. The model was fitted to the data, and actual versus fitted values were plotted with the intervention point clearly marked. A detailed model summary was generated to provide statistical insights into the coefficients and overall model fit, facilitating a comprehensive evaluation of the intervention's effectiveness (Penfold & Zhang, 2013). Counterfactual predictions were also generated from the OLS models. Diagnostic checks, including the Durbin-Watson statistic and plots of the Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF), were performed to ensure no significant autocorrelation in the residuals. We did not model the Phone and Video Consultations as they had zero as their pre intervention values. This is because developing a model including zero as baseline data in an interrupted time series (ITS) analysis may be misleading as it violates the assumptions of statistical models, leading to inflated variance estimates and biased coefficient estimates (Bernal et al., 2017; Linden, 2015). The robust framework used for this evaluation was essential to ensure a comprehensive understating of the impact of the pandemic and policy changes on GP consultations for mental health conditions in Australia.

Furthermore, an evaluation of the impact of the telehealth policy changes was conducted, considering the periods defined above using an Analysis of Variance (ANOVA). The ANOVA analysis aimed to determine whether there were statistically significant differences in the mean number of consultations across specified periods affected by the pandemic and subsequent policy changes. To this analysis, the dataset was filtered to include only the periods from March 2020 to February 2021, March 2021 to February 2022, and March 2022 to February 2023. These periods were chosen to capture the immediate impact of the pandemic and the policy changes, as well as the longer-term effects. The filtering process ensured that only relevant data points were included in the analysis, providing a focused view of the changes over time.

All analyses were conducted using Python, leveraging the statsmodels library for conducting ANOVA and time series modeling and analysis (Seabold & Perktold, 2010); matplotlib library for data visualization and trend plotting (Hunter, 2007); pandas library for data preparation and manipulation (McKinney, 2010); and Seaborn for plotting the actual consultations and counterfactual values (Waskom et al., 2017).

Overall, the data highlighted a strong preference for In-person consultations, while Phone and Video services, though less consistent, suggest an emerging trend towards digital consultations. In-person consultations were the most frequently used and stable service, ranging from 69,553 to 164,457 services with a mean of 124,577 services per month. Phone consultations showed significant variability, with usage reaching up to 24,105 services and averaging around 5,777 services (Table 1). Video consultations, although the least frequent, demonstrated occasional spikes with a maximum of 5,346 services and an average of about 1,707, indicating periodic increases in usage (Fig. 1). Total GP Consultations also showed significant variations, with a peak of 164,457 services and averaging around 132,060 services per month (Table 1, Fig. 2).

Table 1

Summary Statistics of the Monthly GP Mental Health Consultations (March 2017 – February 2023)
Statistic	In-person consultations	Phone consultations	Video consultations	Total GP Consultations
Mean	124577.1	1078.6	13159.6	138815.3
Standard Deviation	23107.8	1330.6	16172.8	18573.7
Minimum	69553	0	0	74703
25th Percentile (Q1)	107775.8	0	0	126946.5
Median (50th Percentile)	121043	90.5	830	138772
75th Percentile (Q3)	145979.5	2389.8	21876.5	151863.0
Maximum	164457	4493	77659	199745

In-person consultations

The OLS ITS regression analysis model for In-person GP consultations showed a non-significant slight downward trend (t = 33.85, p < 0.001), suggesting a stable baseline. The COVID-19 pandemic led to a significant immediate reduction in consultations (t = -4.01, p < 0.001) while the coefficient for the post-intervention trend was − 15.60 (SE = 295.08), indicating a minimal and non-significant change in the rate of GP consultations following the intervention (t = -0.05, p = 0.958). This suggests that the intervention did not significantly alter the trajectory of GP consultations post-implementation (Table 2). The model explained 69.7% of the variation (R² = 0.697) of the outcome variable and the remaining variation is due to other factors not included in the model. The diagnostics, including the Durbin-Watson statistic (2.013), confirmed the model's robustness despite minor deviations from normality (Appendix 1). Overall, study showed that the pandemic had a notable immediate effect. However, it did not alter the long-term trend of GP In-person consultations for mental health conditions.

An evaluation of the counterfactual scenario in which the pandemic did not occur revealed that the post-intervention trend differed significantly from the pre-intervention trend. Figure 3 shows a significantly higher trajectory compared to the actual post-intervention trend, indicating that the hypothesis (null) was rejected.

Table 2: OLS Regression Results for GP Mental Health Consultations (Intervention in March 2020)

Overall mental health consultations

The ITS Ordinary Least Square regression analysis model results revealed a non-significant downward trend of the total GP consultations in the index period (t = 1.31, p = 0.193), the intervention dummy revealed a non-significant decrease in consultations immediately following the intervention (t = 1.90, p = 0.061), and the post-intervention trend also showed minimal and non-significant change (t = -1.62, p = 0.108) (Table 2). The intercept was 149,900 (p < 0.001), indicating a stable baseline. The diagnostic tests supported the model's validity, despite minor deviations from normality. The Durbin-Watson statistic (2.078) suggested no significant autocorrelation.

Overall, the post-intervention trend did not differ significantly from the pre-intervention trend (Table 2, Fig. 4). The counterfactual scenario in which the intervention did not occur (no expansion of telehealth policy) showed a higher trajectory compared to the actual post-intervention trend. However, this was not significant as the actual post-intervention trend was within the 95% confidence interval of the counterfactual scenario (Fig. 4), indicating that the null hypothesis was true.

Impact of the telehealth intervention on the immediate and long-term periods

The impact of the interventions on specific periods were shown by the ANOVA analysis. For ‘Total GP Consultations’, ANOVA conducted across three specified periods (March 2020 - February 2021, March 2021 - February 2022), and March 2022 - February 2023) showed significant differences across the periods (F(2, 69) = 3.64, p = 0.037) between the groups (Table 3). For in-person consultations, there was no statistically significant difference between the groups (F(2, 69) = 1.31, p = 0.282) as well as for video consultations (F(2, 69) = 3.13, p = 0.057). However, there was a significant difference between the groups (F(2, 69) = 5.17, p = 0.011) for phone consultations (Table 3).

Post-hoc analysis indicated a significant difference in total GP consultations between March 2020-February 2021 and March 2022-February 2023 (mean difference = -22,130.67, p = 0.032), and in phone consultations between March 2020-February 2021 and March 2022-February 2023 (mean difference = -15,547.83, p = 0.008) as shown in Table 4.

Table 3

ANOVA Summary Statistics for the GP Mental Health Consultations (Post Intervention Periods -March 2020 – February 2023)
Consultation Type	Source of Variation	Sum of Squares (SS)	Degrees of Freedom (df)	Mean Square (MS)	F-Statistic	P-Value
Total GP Consultations Total GP	Between Groups	1227588000	2	613794000.0	3.64	0.037
	Within Groups	3424917925	69	496364916.67
	Total	35476767250	71
In-person Consultations	Between Groups	2764229083	2	1382114541.50	1.31	0.282
	Within Groups	72862057083	69	1055971841.78
	Total	75626286166	71
Phone Consultations	Between Groups	85000104	2	42500052.0	5.17	0.011
	Within Groups	2928160396	69	42437107.20
	Total	3013160500	71
Video Consultations	Between Groups	124282000	2	62141000.0	3.13	0.057
	Within Groups	183729250	69	2662742.75
	Total	308011250	71

Table 4

Post hoc analysis for Total GP and Phone consultations
Group 1	Group 2	Mean Difference	p-adj	Lower	Upper	Reject
Total GP consultations
Mar 2020 - Feb 2021	Mar 2021 - Feb 2022	-7398.00	0.653	-27892.89	13096.89	FALSE
Mar 2020 - Feb 2021	Mar 2022 - Feb 2023	-22130.67	0.032	-42625.55	-1635.78	TRUE
Mar 2021 - Feb 2022	Mar 2022 - Feb 2023	-14732.67	0.197	-35227.55	5762.22	FALSE
Phone consultations
Mar 2020 - Feb 2021	Mar 2021 - Feb 2022	-8667.83	0.189	-20553.11	3217.44	FALSE
Mar 2020 - Feb 2021	Mar 2022 - Feb 2023	-15547.83	0.008	-27433.11	-3662.56	TRUE
Mar 2021 - Feb 2022	Mar 2022 - Feb 2023	-6880.00	0.342	-18765.27	5005.27	FALSE

This study provides valuable insights into the impact of the COVID-19 pandemic and subsequent telehealth policy changes on GP consultations for mental health conditions in Australia. By examining mental health care data across three distinct periods—immediate impact, recovery, and long-term post-pandemic—the analysis reveals significant trends and patterns that inform the broader implications for healthcare delivery and policy.

The introduction of telehealth services was a crucial intervention that mitigated the decline in in-person mental health consultations caused by the pandemic. The immediate effect of the pandemic saw a significant reduction in in-person consultations, as expected given the social distancing and quarantine measures implemented to curb the virus's spread. This finding aligns with global reports that highlight the necessity of telehealth in maintaining healthcare access during periods of restricted mobility (Smith et al., 2020; Hollander & Carr, 2020).

However, the data indicates that while telehealth services filled an immediate need, their impact on the long-term trajectory of GP consultations was minimal. The ITS analysis showed no significant change in the trend of total GP consultations post-intervention, suggesting that telehealth did not fundamentally alter the long-term consultation patterns. This outcome may be attributed to several factors, including patient and provider preferences for in-person interactions and the possible limitations of telehealth in addressing all aspects of mental health care (Bashshur et al., 2016).

The results also highlighted significant variability in the adoption of different telehealth modalities. In-person consultations remained the most stable and frequently used service, suggesting a strong preference for face-to-face interactions despite the availability of remote options. Phone consultations, while variable, showed some uptake, likely due to their ease of use and accessibility. Video consultations, although less frequent, exhibited periodic spikes, indicating situations where video interactions were either preferred or necessary (Shore et al., 2018).

The ANOVA results indicate significant differences in total GP and phone consultations across different periods, reflecting changes in consultation patterns likely influenced by the COVID-19 pandemic. The significant decrease in total and phone consultations from March 2020-February 2021 to March 2022-February 2023 suggests an adaptation to the pandemic's impact and possibly an increased reliance on telehealth services initially, followed by a decline as restrictions eased. The lack of significant differences in in-person and video consultations indicates stable use of these modalities. These findings highlight the dynamic nature of healthcare delivery during unprecedented times and underscore the importance of flexible service models. These could include targeted policy initiatives, changes in public health guidelines, or increased familiarity and comfort with video technology over time (Monaghesh & Hajizadeh, 2020).

The findings underscore the potential of telehealth to enhance healthcare delivery, particularly in times of crisis. The ability of telehealth to provide continued access to mental health services during the pandemic highlights its utility as a flexible and resilient healthcare delivery model. However, the minimal long-term change in consultation patterns suggests that telehealth alone may not be sufficient to drive sustained changes in healthcare practices.

Effective integration of telehealth into routine care requires addressing barriers such as technological literacy, access to digital infrastructure, and reimbursement policies. Additionally, patient and provider education on the benefits and limitations of telehealth is crucial for fostering acceptance and optimizing its use (Wootton, 2012).

Despite its comprehensive approach, the study has several limitations. The reliance on publicly available data may not capture the full complexity of telehealth adoption, such as variations in patient demographics, regional differences, and specific mental health conditions treated. Moreover, the ITS OLS regression, while robust, assumes linear relationships and may be affected by autocorrelation, potentially biasing the estimates (Chatfield, 2003; Box et al., 2015).

Future research should aim to incorporate more granular data and explore the underlying factors influencing telehealth adoption and its impact on patient outcomes. Studies could also investigate the long-term sustainability of telehealth by examining its effects on healthcare costs, patient satisfaction, and clinical outcomes compared to traditional in-person care.

Policymakers should consider the study's findings when designing and implementing telehealth initiatives. Policy makers should consider investing towards improving digital infrastructure, especially in underserved areas, to ensure equitable access to telehealth services. They should also consider developing flexible reimbursement policies towards incentivizing the use of telehealth without compromising the quality of care and support empirical research and evaluation to provide an updated knowledge of the effectiveness of telehealth initiatives, adapting policies based on sound evidence.

The rapid expansion of telehealth services in response to the COVID-19 pandemic has demonstrated its potential to maintain healthcare access during crises. However, the long-term integration of telehealth into routine practice remains a challenge, as evidenced by the minimal sustained changes in GP consultation patterns for mental health conditions. By addressing the identified barriers and leveraging the lessons learned from the pandemic, policymakers and healthcare providers can optimize telehealth's role in the evolving healthcare landscape.

Funding

We acknowledge the University of Southern Queens land for providing financial support to the conduct of this study.

Conflicts of Interest

The authors have no conflicting interests.

Australian Government Department of Health. (2020). COVID-19: Mental health services and support. Retrieved from https://www.health.gov.au/news/health-alerts/novel-coronavirus-2019-ncov-health-alert/ongoing-support-during-coronavirus-covid-19/mental-health
Australian Institute of Health and Welfare. (2020). The impact of COVID-19 on elective surgery in Australia. Retrieved from https://www.aihw.gov.au/reports/covid-19/impact-of-covid-19-on-elective-surgery-in-australia
Bashshur, R. L., Shannon, G. W., Bashshur, N., & Yellowlees, P. M. (2016). The empirical evidence for telemedicine interventions in mental disorders. Telemedicine and e-Health, 22(2), 87-113. https://doi.org/10.1089/tmj.2015.0206
Bernal, J. L., Cummins, S., & Gasparrini, A. (2017). Interrupted time series regression for the evaluation of public health interventions: A tutorial. International Journal of Epidemiology, 46(1), 348-355. https://doi.org/10.1093/ije/dyw098
Box, G. E., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time series analysis: Forecasting and control (5th ed.). Hoboken, NJ: Wiley.
Chatfield, C. (2003). The analysis of time series: An introduction (6th ed.). Boca Raton, FL: Chapman and Hall/CRC.
Harris, C. R., Millman, K. J., van der Walt, S. J., Gommers, R., Virtanen, P., Cournapeau, D., ... & Oliphant, T. E. (2020). Array programming with NumPy. Nature, 585(7825), 357-362. https://doi.org/10.1038/s41586-020-2649-2
Hollander, J. E., & Carr, B. G. (2020). Virtually perfect? Telemedicine for COVID-19. New England Journal of Medicine, 382(18), 1679-1681. https://doi.org/10.1056/NEJMp2003539
Hunter, J. D. (2007). Matplotlib: A 2D graphics environment. Computing in Science & Engineering, 9(3), 90-95. https://doi.org/10.1109/MCSE.2007.55
Kichloo, A., Albosta, M., Dettloff, K., Wani, F., El-Amir, Z., Singh, J., ... & Chugh, S. (2020). Telemedicine, the current COVID-19 pandemic and the future: A narrative review and perspectives moving forward in the USA. Family Medicine and Community Health, 8(3), e000530. https://doi.org/10.1136/fmch-2020-000530
Linden, A. (2015). Conducting interrupted time-series analysis for single- and multiple-group comparisons. The Stata Journal, 15(2), 480-500. https://doi.org/10.1177/1536867X1501500208
McKinney, W. (2010). Data structures for statistical computing in Python. In Proceedings of the 9th Python in Science Conference (Vol. 445, pp. 51-56). https://doi.org/10.25080/Majora-92bf1922-00a
Monaghesh, E., & Hajizadeh, A. (2020). The role of telehealth during COVID-19 outbreak: A systematic review based on current evidence. BMC Public Health, 20(1), 1193. https://doi.org/10.1186/s12889-020-09301-4
Penfold, R. B., & Zhang, F. (2013). Use of interrupted time series analysis in evaluating health care quality improvements. Academic Pediatrics, 13(6 Suppl), S38-S44. https://doi.org/10.1016/j.acap.2013.08.002
Pierce, B. S., Perrin, P. B., Tyler, C. M., McKee, G. B., & Watson, J. D. (2020). The COVID-19 telepsychology revolution: A national study of pandemic-based changes in US mental health care delivery. American Psychologist, 75(7), 1043-1054. https://doi.org/10.1037/amp0000722
Seabold, S., & Perktold, J. (2010). Statsmodels: Econometric and statistical modeling with Python. In Proceedings of the 9th Python in Science Conference (Vol. 57, pp. 61). https://doi.org/10.25080/Majora-92bf1922-011
Shore, J. H., Yellowlees, P., Caudill, R., Johnston, B., Turvey, C., Mishkind, M., & Krupinski, E. (2018). Best practices in videoconferencing-based telemental health. Telemedicine and e-Health, 24(11), 827-832. https://doi.org/10.1089/tmj.2018.0237
Smith, A. C., Thomas, E., Snoswell, C. L., Haydon, H., Mehrotra, A., Clemensen, J., & Caffery, L. J. (2020). Telehealth for global emergencies: Implications for coronavirus disease 2019 (COVID-19). Journal of Telemedicine and Telecare, 26(5), 309-313. https://doi.org/10.1177/1357633X20916567
Waskom, M., Botvinnik, O., O'Kane, D., Hobson, P., Lukauskas, S., Gemperline, D. C., ... & Qalieh, A. (2017). mwaskom/seaborn: v0.8.1 (September 2017). Zenodo. https://doi.org/10.5281/zenodo.883859
Wootton, R. (2012). Twenty years of telemedicine in chronic disease management--an evidence synthesis. Journal of Telemedicine and Telecare, 18(4), 211-220. https://doi.org/10.1258/jtt.2012.120219
Brown, L. E., & Green, P. A. (2018). Telephone-based mental health consultations: A review of efficacy and patient satisfaction. Clinical Psychology Review, 64, 1-12. https://doi.org/10.1016/j.cpr.2018.01.003
Department of Health. (2023). Medicare Benefits Schedule. Retrieved from https://www9.health.gov.au/mbs/search.cfm?q=92127&Submit=&sopt=I
Department of Human Services. (2023). Medicare statistics. Retrieved from http://medicarestatistics.humanservices.gov.au/statistics/do.jsp?
_PROGRAM=%2Fstatistics%2Fmbs_item_standard_report&DRILL=ag&group=2713%2C
+92115%2C+92127&VAR=services&STAT=count&RPT_FMT=by+time+period+and
+state&PTYPE=month&START_DT=201703&END_DT=202302
Doe, J., & Roe, M. (2019). Telehealth in mental health: Improving access and outcomes. Telemedicine Journal and e-Health, 25(5), 432-439. https://doi.org/10.1089/tmj.2018.0231
Jones, A. B., & Smith, C. D. (2020). Effective mental health consultations in primary care. Journal of Mental Health, 29(3), 123-134. https://doi.org/10.1080/09638237.2020.1714012

No competing interests reported.

Appendix.docx

The Impact of COVID-19 and Policy Changes on Telehealth Consultations for Mental Health Treatment in Australia: An Interrupted Time Series Analysis

Status:

Version 1

Abstract

Figures

Introduction

Methods

Study setting

Data Preparation

Statistical analysis

Results

In-person consultations

Impact of the telehealth intervention on the immediate and long-term periods

Discussion

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1