Professor James Dunbar
Professor Prasuna Reddy
University Department of Rural Health
Flinders and Deakin Universities
beyondblue Victorian Centre of Excellence
Project completion year
Heart disease, diabetes, and depression represent three of the leading disease burdens in Australia with all three conditions being identified as National Health Priority Areas. Diabetes is a recognised risk factor for coronary Heart Disease (CHD) and the impact of depression on both conditions has been shown to result in significantly worse mortality and morbidity outcomes as well as increases in health care costs. These chronic conditions are managed in the main in general practice, and it is this setting where the majority of patients seek help.
In general practice, collaborative models of care have been shown to be effective in improving depression outcomes at six months and up to five years. Specifically, the Improving Mood – Promoting Access to Collaborative Treatment (IMPACT)2 model of care implemented in US primary care settings has set the standard for collaborative care at an international level. Results have shown this model may reduce depressive symptoms by up to 50% compared with usual care, and it is on this model that we base our study. To the best of our knowledge, this is the first time that collaborative care has been implemented in Australia and probably only the second time that collaborative care has been used for co-morbid depression in general practice settings internationally.
The aim of this project was to (i) develop, implement and evaluate a patient focused integrated system of care based in general practice to increase recognition and improve management of comorbid depression, coronary heart disease and diabetes, and (ii) to set up systems within general practice to demonstrate how the care of co-morbid depression, heart disease and diabetes can be funded successfully using the new Medicare Benefits Schedule (MBS) item numbers.
The objectives of the project included (i) developing and testing a training program for general practitioners and practice nurses in the screening, assessment and management of depression in patients with heart disease, diabetes or both, and (ii) testing the feasibility of practice nurses to screen, assess, collect data, counsel, refer, review and monitor patients with co-morbid depression, heart disease, diabetes or both.
a) After reviewing both the national and international literature on depression and chronic disease management, the IMPACT2 model of collaborative care was chosen for implementation in an Australian context, and for use in this project, based on its demonstrated success for the treatment of later life depression in US primary care settings. It is also regarded internationally as one of the most effective models of collaborative care for the treatment of depression.
b) Our research team met with the IMPACT2 team at the University of Washington, Seattle, to discuss the application of this model of care to an Australian setting. Flinders University completed contractual agreement with the University of Washington to obtain the rights for use of the IMPACT. The model was then modified for use in Australian general practice incorporating Medicare Item numbers, existing practice resources, staff and systems.
c) Six general practices from the Greater Green Triangle (GGT) region of Victoria and South Australia participated in the study, named the Depression Treatment Evaluation Care Team (D_TECT) project. Staff from all practices had participated in the National Primary Care Collaboratives, Wave 1, and as a result, participating practice nurses had acquired knowledge of recall systems and chronic disease registers.
d) Patients with a diagnosis of diabetes or CHD were invited to participate in the project by invitation or opportunistically. 339 patients were enrolled. About 49% had diabetes, 42% had CHD and 9% had both.
e) Participating patients received a comprehensive review of classic risk factors, depression screening and a personalised chronic disease management plan jointly written by the practice nurse and doctor.
f) Collectively, 332 patients were screened for depression. Using the Patient Health Questionnaire (PHQ9), more than one third (35%) exhibited depressive symptoms outside of normal range. Using the HADS, 10% exhibited symptoms of depression outside normal range.
g) Twenty two percent of patients in the sample reported a past history of depression. The proportion of new depression cases detected by depression screening was estimated at 23%.
h) Of the 51 high priority patients (those reporting moderate to severe depressive symptoms according to the PHQ9, 29 patients were recalled by practice nurses for re-assessment.
i) Eighty two percent of those re-assessed displayed a change in PHQ9 scores between initial assessment and review.
j) Practice nurses allocated about 45 minutes to review blood test results, patient measurements, lifestyle modifications, mental health and a ‘best practice’ checklist for optimal secondary prevention. Doctors allocated approximately 15 minutes to complete a medication review, address issues raised by the nurse and complete GP Management Plans.
k) An effective feature of the model was the enhanced skills of practice nurses. The role of the practice nurse expanded dramatically. They consulted independently with patients, assessing risk factors, using depression screening tools and generating GP Management Plans. As the model matured, practice nurses organised recall appointments to monitor depression and classical risk factors and patients’ progress towards their goals.
l) A barrier to the model’s implementation was perceived lack of time. Practice nurses identified protected time as an enabler to implementation of this model of care. Nurses saw supportive feedback from GPs and other staff as an essential enabler to the running of the model.
m) On average, practice nurses rated this model of care 7/10 in terms of effectiveness to manage CHD and diabetes patients. General practitioners rated this model of care 8/10 in terms of its effectiveness. The nurses reported using the model for patients with other chronic diseases.
n) In order for this model of care to be sustained, the recall system requires improved information technology and computer resources in each practice.
o) On average, practice nurses rated their level of willingness to continue using the model as 7/10. General Practitioners rated their willingness to continue using the model as 9/10.
p) This model of care is based on a sound business case. Using various Medicare Item numbers, the potential revenue generated per patient using this model of care was calculated as $893.80. The introduction of additional Medicare item numbers throughout the course of the study enhanced the sustainability of the model.
q) To our knowledge, this is the first time such a study has been conducted using a collaborative care model for the management of co-morbid depression in primary care, a distinct feature of this project.
The findings indicate that it is feasible for practice nurses to screen, assess, collect data, counsel, and monitor depression in patients with CHD or diabetes. Perceived barriers of space and time to nurses referring and reviewing patients need to be overcome in order for this model of care to be implemented in its full capacity. Overcoming these barriers would provide further support for the use of collaborative care models for the management of co-morbid depression in Australian general practice.
Implications for policy and practice
Findings from this project suggest that an ‘Australianised’ collaborative care model which assesses depression in a similar manner to traditional risk factors for heart disease and diabetes is feasible. Based on the proportions of CHD and diabetes patients exhibiting depressive levels outside the normal range, a collaborative care model for co-morbid depression in general practice is warranted.
While a model of care such as that developed and implemented in this project needs to be flexible in its incorporation into clinical practice, as a minimum, depression screening results should be included in care management plans alongside other measures such as HbA1c and cholesterol for people with heart disease or diabetes. The inclusion of depression screening in the health checks for 45-49 year old persons is an example of how this project can be extended in other areas of general practice.
The role of practice nurses as case managers in the management of patients with chronic disease and depression has implications for policy. The potential for the introduction of a new Medicare item number which generates reimbursement for chronic disease management will contribute to the sustainability of the model. The incorporation of depression management into chronic disease management will allow for the use of this item number.
Extensive recommendations have been made for future implementation of this collaborative model of care in general practices across Australia. Briefly, these are discussed using four themes: vision, structures, procedures and processes, and service delivery.
Overall, the findings from this project suggest that collaborative care can be effective for the management of co-morbid conditions, a result which has implications on the way in which patients should be monitored and managed when presenting with multi morbidities in Australian general practice settings. The need for the integration of specialists and primary care in the management of these patients is evident (e.g., Starfield, 2006). The collaborative care model developed for this project can be implemented in Australian general practice for the management of patients with comorbid depression. It has been argued that the best outcome for access to evidence based primary mental health care is the responsibility of the health profession, not necessarily dependent on government directives. The emphasis on developing a systematic adoption of collaborative care will prove beneficial not only to the needs of patients but also demonstrates best practice in primary care.