Management or missed opportunity? Mental health care planning in Australian general practice

Additional keywords: anxiety, bipolar disorder, depression, management plan, primary care, schizophrenia.

Chronic physical and mental disorders are complex and prevalent health problems that often require a structured approach, with coordinated involvement from a multidisciplinary team (World Health Organization 2002). According to the National Health Survey, type II diabetes affected 4.1% of Australians and ~20% experienced a mental health issue in 2017–18 (Australian Bureau of Statistics 2018). Anxiety disorders affect 14.4% of the Australian population in any one year, 5.4% will experience a depressive episode or dysthymia and 1.8% will experience bipolar disorder (Australian Bureau of Statistics 2007). In 2010, 0.45% of the Australian population were treated for a psychotic illness, most commonly schizophrenia (Morgan et al. 2012). However, despite the prevalence of mental disorders, effective access and care management remain a major concern (Banfield et al. 2011; Banfield et al. 2014).

General practice care plans were introduced into the Australian healthcare system to serve two purposes: (1) to facilitate access to multidisciplinary health care for people with chronic health conditions; and (2) to assist GPs to manage the ongoing care of people with these conditions (Department of Health 2014). Under this model, GPs are eligible to receive Medicare Benefits Schedule (MBS) payments to prepare and review plans, which involves working collaboratively with consumers to determine their healthcare needs; develop mutually agreed goals; articulate the actions and treatments or services required; and to assess and amend the plan as needed over time.

Chronic disease management (CDM) plans replaced Enhanced Primary Care (EPC) plans in 2005, and were introduced to enable GPs to plan and coordinate the health care of people with chronic medical conditions, such as type II diabetes (Department of Health 2014). There is some evidence that CDM plans have resulted in improved outcomes for people with type II diabetes, particularly regarding timely access to allied health services, consumer knowledge and self-management (Grimmer-Somers et al. 2010).

Under the Better Access to Psychiatrists, Psychologists and General Practitioners (Better Access) initiative (introduced in November 2006) (Littlefield and Giese 2008), GPs can claim for the creation and review of a mental health care plan, which is designed to encourage collaborative care between GPs, consumers and other service providers, and to provide streamlined access to psychiatrists, psychologists and other health professionals (Department of Health 2018a). The MBS items for mental health care plans state that eligible people are those ‘…with [an ICD-10] mental disorder who would benefit from a structured approach to the management of their treatment needs’ (Department of Health 2018b). However, there has been controversy about whether mental health care plans function as they were designed. Specific criticisms include that the plans have mainly benefited the ‘worried well’ and not those with complex, less prevalent mental disorders (Jorm 2011), and that they do not result in collaborative care (e.g. GPs rarely review the mental health care plans they create) (Rosenberg and Hickie 2012).

The current research aimed to examine care planning for diabetes and mental disorders from both records of GP encounters and consumer experiences.

The ethical aspects of this study were approved by the Human Research Ethics Committees of the University of Sydney (Protocol 2012/130) and The Australian National University (Protocols 2017/344 and 2017/440). All participants gave informed consent to participate.

The study used 10 years of data from the Bettering the Evaluation and Care of Health (BEACH) study to examine patterns of creation and review of care plans for high (depression, anxiety) and low prevalence mental disorders (bipolar disorder, schizophrenia) and type II diabetes. Detailed methods of the BEACH study are reported elsewhere (Britt et al. 2014; Britt and Miller 2015). Briefly, the BEACH study was a continuous, representative, national, cross-sectional study of encounters in general practice between 1998 and 2016. An ever-changing, random sample of 1000 active GPs was surveyed each year, with each providing details on 100 consecutive consumer encounters. The database includes ~1.78 million encounter records by ~11 000 GPs. Encounter information recorded by GPs included reason for visit, problem managed, treatment delivered and referral/s to other health professionals. In addition, GPs recorded up to three MBS item numbers per encounter.

A sample of encounters involving type II diabetes, anxiety, depression, bipolar disorder and/or schizophrenia was extracted from the BEACH database using ICPC-2 PLUS codes (Britt et al. 2014). The sample of encounters was drawn from the 10-year period between 2006 and 2016. MBS item numbers used by GPs to claim for the creation or review of a diabetes plan or mental health care plan were identified. Six items related to the creation of a mental health care plan (2700, 2701, 2715, 2717, 2702, 2710) and one item related to review (2712). Two chronic disease management plan items for diabetes related to creation (721, 723) and three related to review (729, 731, 732).

Percentages and 95% confidence intervals for rates of plan creation and review were calculated using surveymeans procedures in SAS (Version 9.4; SAS Institute, Cary, NC, USA), which took into account the cluster design of the BEACH study. Comparisons were undertaken by individual disorder and by type (physical disorder, high prevalence mental disorders, low prevalence mental disorders). Results with non-overlapping confidence intervals were considered to be significantly different. This method is a conservative approach to assessing significant differences between groups (Austin and Hux 2002).

To understand people’s experiences of care planning in general practice, an online survey was created (see Table S1 available as Supplementary Material to this paper). The survey comprised a mix of open- and closed-ended questions to establish consumers’ views on why care plans were created, their perceived role in chronic illness management and whether people found them useful and positive. Most closed-ended questions were yes/no or categorical questions, and open-ended questions asked for further information or comment.

Advertisements about the survey were circulated through state and national health and mental health consumer organisations and via social media during September 2017. Twenty-nine people agreed to participate, of whom 18 completed the survey; two were ineligible (had not had a care plan prepared) and nine did not submit their response. Only data from complete surveys were included in analyses, as it could not be ascertained whether an incomplete response was a formal withdrawal or due to technical issues. Descriptive statistics were calculated for closed questions, split according to which type of care plan the participant reported (diabetes, mental health or both). A content analysis was conducted on open-ended responses (Hsieh and Shannon 2005).

Table 1 comprises the patterns of creation of diabetes and mental health care plans within total encounters for each of the five conditions across the 10-year study period. As depicted in Fig. 1, a marked increase was observed over time in the percentage of encounters for high prevalence mental disorders that involved the creation of a mental health care plan. Comparatively, creation of plans for low prevalence mental disorders or type II diabetes were much lower, and only a modest increase was observed. With the exception of 2006–07, the percentage of encounters involving the creation of care plans for depression or anxiety was significantly higher than percentages of encounters involving care plan creation for type II diabetes, bipolar disorder and schizophrenia.

Table 1.  Percentage of encounters (and 95% confidence intervals) for each problem managed that involved the creation of a diabetes plan or mental health care plan, by year
Data are presented as percentages (95% confidence interval)

Fig. 1.  Percentage of encounters within each condition that involved the creation of a diabetes plan or mental health care plan, by Bettering the Evaluation and Care of Health (BEACH) year (April–March). Note: High prevalence mental disorders (depression and anxiety) were collapsed into a single category for graphing, as were low prevalence mental disorders (bipolar disorder and schizophrenia).

Figure 2 shows the percentage of encounters involving the creation of a care plan plotted against encounters involving the review of a care plan for each of the disorders examined, between 2006 and 2016. Percentages of encounters involving the review of a care plan for type II diabetes were slightly lower than, but proportional to, the rates at which plans were created. In contrast, percentages of encounters involving depression and anxiety showed significantly higher rates of care plan creation compared with encounters involving care plan review. The overall very low rates of care planning for bipolar disorder and schizophrenia and wide confidence intervals make identification of clear patterns difficult, but review of care plans appears to be particularly low for low prevalence disorders.

Fig. 2.  Percentage of encounters (and 95% confidence intervals) within each condition involving the creation and review of a diabetes plan or mental health care plan, by Bettering the Evaluation and Care of Health (BEACH) year (April–March).

Table 2 presents the characteristics of survey participants. The majority of participants had mental health care plans only (n = 14), with only one who reported having a diabetes care plan only. Three participants had both mental health and diabetes care plans. There was a very high level of multimorbidity in the sample; only two participants reported having a single condition. The majority of participants (n = 16) reported experiencing depression and anxiety, including all of those with schizophrenia and bipolar disorder.

Table 2.  Characteristics of the sample who completed the online survey (n = 18)

Most participants reported multiple care plans created over a period of up to 10 years. Only one of the four people with a diabetes plan said that having a plan was their own idea rather than their doctor’s, whereas seven out of 17 mental health care plans (41.1%) were suggested by the consumer. All participants indicated that the main reason their plans were created was to provide access to allied health professionals under Medicare, which they may not otherwise have been able to afford. This was reported as the major purpose and most positive aspect associated with having a care plan.

However, evidence that care plans were also facilitating good management of health was mixed. Eleven of the 17 people with mental health care plans (64.7%) and three of the four with diabetes care plans rated them as at least ‘somewhat useful’ for managing their health, commenting that a care plan is: ‘…a positive thing as it helps to keep a check on your health issues’ [Participant 8, diabetes]. By contrast, others felt the process was too impersonal and formulaic, with one describing it as ‘simplistic…intrusive and judgemental’ and that it made them feel ‘incompetent and guilty…as if the financial support could be withdrawn at any time’ [Participant 10, both diabetes and mental health care plan]. Many ascribed the positives of the plans and their usefulness to the financial aspects, and did not feel the plans did much more for quality health care.

It is instrumentally useful in enabling me to access cheaper psychology services. Otherwise it has never really helped. They are one off items. What people need are high quality ongoing medical and allied services. The plans do not enable adequate, let alone high quality, mental health services to arise from the ether. Until such services are created the plans remain a bureaucratic exercise for many, probably for most with significant mental illness [Participant 15, mental health].

The majority of participants (n = 12, 66.7%) indicated that their plans were reviewed. Half of participants with a diabetes care plan reported receiving a reminder, but only one person could recall receiving a reminder for their mental health care plan. Participants with mental health care plans expressed frustration with the six-session limit on the psychologist appointments associated with the plans and the need for review to access further subsidised sessions. Comments suggested that participants did not feel the reviews added to their care, and instead were another hurdle to continue with needed treatment, or as one participant commented ‘The reviews again just tick a few boxes, no care given to how I’m going’ [Participant 16, mental health]. The main failure that participants perceived was in shared decision-making. Only 10 participants (55.6%) reported that they felt actively involved in making the plans, with some observing it reduced anxiety about accessing care and helped to set goals, but for many others, the plan was felt to be an impersonal, tick box process.

It feels a lot like just a rubber stamped piece of paperwork. I had one drawn up by a doctor in less than 4 minutes. If I didn’t already have a psychologist who seems actively interested in supporting my health, getting a mental health plan would seem utterly pointless… [Participant 7, mental health].

General practice care plans were introduced by the Australian government to optimise care for people with chronic illnesses by facilitating access to multidisciplinary health professionals, and encouraging GPs and consumers to work collaboratively to manage healthcare needs (Department of Health 2014, 2018a). Examining rates of care plan creation and review over time and between chronic health conditions highlights gaps in this model of care and signposts potential areas of focus for practice and policy reform. Analysis of BEACH data indicated that in the previous 10 years, the creation of care plans for depression and anxiety increased significantly, and occurred at a much higher rate than care plan creation for type II diabetes, bipolar disorder and schizophrenia. This suggests that consumers with serious, low prevalence mental disorders are not receiving the potential benefits of care planning to the same extent as those with depression or anxiety. This represents a critical missed opportunity for people with complex healthcare needs to benefit from a coordinated care approach, including supported access to specialist and allied health care and proactive health management. Although the Better Access program is often perceived as a program for high prevalence disorders (Jorm 2011), the eligibility requirements for the relevant MBS items are broad (Littlefield and Giese 2008). Recent evidence from community-based integrated psychiatric care suggests strong benefits in effective collaborative ongoing management for low prevalence disorders (Schöttle et al. 2014), which may provide important guidance for primary care, and clinical practice guidelines support the role of psychological therapy, particularly for long-term management (Malhi et al. 2015; Galletly et al. 2016). A recently released MBS Review report recommends several changes to MBS mental health items, such as a tiered system to better accommodate a range of mental disorders and levels of severity, further supporting the role of the scheme beyond high prevalence disorders (Mental Health Reference Group 2018). Should these changes progress, it will be important to improve the use of care plans to encourage better collaborative care and proactive healthcare management.

A second important element of care plans as management tools is regular review. Diabetes care plans were reviewed at rates largely commensurate with their creation, but rates of review of mental health care plans for depression and anxiety were significantly lower than the rates at which they were created. This suggests that although the MBS review item descriptions are very similar, in practice, diabetes care plans may be operating as intended, but mental health care plans are predominantly used for access to allied health professionals, rather than for ongoing collaborative care. This criticism has been levelled previously at the Better Access scheme (Jorm 2011; Rosenberg and Hickie 2011), and was confirmed by people with experience of care planning in the current study. While most consumers viewed access as one of the most positive aspects of the process, others reported that their care plans were bureaucratic; ‘tick box’ exercises that made them feel judged. Reviews of mental health care plans were seen as a barrier to adequate treatment rather than an opportunity to assess progress, and consistent with previous research (Shortus et al. 2007); almost half of participants indicated that they did not have an active role in shared decision-making. This suggests that as part of the MBS review of mental health in primary health care, active collaborative care, including shared decision-making and detailed review, is needed between consumer and all health professionals to ensure care plans meet their potential.

One factor that may, at least in part, be driving the difference between diabetes and mental health care planning is the payments available to GPs through the Practice Incentives Program (PIP). The additional payments available to accredited practices for the completion of annual cycles of care for people with diabetes, including monitoring of key physical indicators such as HbA1c and provision of health education, provide a specific, incentivised guide for ongoing management (Australian Government 2013) and facilitate investment in dedicated diabetes management infrastructure, such as nurses to manage recall and review. Although completion of the required activities to claim PIP payments does not guarantee quality care, this system does increase the emphasis on care plans as ongoing health management tools rather than elaborate referral systems. While mental health outcome measurement tools as indicators of progress are not as clear as a simple blood test, an opportunity exists to shift the focus of assessment and review in mental health care planning. At present, the emphasis is on access, but there is an opportunity to fully leverage the existing funding mechanisms, or perhaps create incentives that parallel those for physical conditions, to ensure that mental health care planning meets its potential and the expectations of consumers for ongoing management.

Strengths of this study include the use of a national and representative dataset of GP encounters, examination of a 10-year time period, a comparative focus between physical and mental disorders, and supplementing administrative data with consumer experience data. Limitations include the small sample size obtained in the online survey of consumer experience, particularly for consumers with experience of diabetes care plans. This may be due to the short timeframe in which the survey was open and the inclusion of a physical health condition within a study primarily focussed on mental health. The focus on encounters with GPs also did not allow examination of the broader general practice infrastructure, such as the involvement of nurses in care planning and management. Finally, the rates of uptake of the care plans created is not known, which may have affected the rates of review.

Rates of care plan creation indicate a proliferation in the uptake of mental health care plans for people with common mental disorders, but identify a significant gap faced by people with complex, less prevalent mental disorders. Rates of care plan review and consumer experience data suggest that care planning may function differently for people with chronic mental and physical disorders. For those with mental health care plans, they appear to function well in terms of increasing access to multidisciplinary care, but do not appear to result in a truly collaborative approach involving shared decision-making and enhanced quality of care over time. This is at odds with the principles of contemporary mental health policy in Australia and may reflect systemic shortcomings in the implementation of care planning for mental health.

The authors declare that they have no conflicts of interest.