GPs need more time, training, remuneration, and specialist backup to diagnose dementia earlier, says leading dementia researcher Professor Perminder Sachdev.
Australia needs a “whole ecosystem” of better-funded general practice, accessible memory clinics, and specialist support if dementia is to be diagnosed earlier, according to leading dementia researcher Professor Perminder Sachdev.
The co-director of the Centre for Healthy Brain Ageing (CHeBA) said new NSW data showing almost three-quarters of dementia diagnoses were first or only recorded in hospital did not surprise him.
“We’ve known this for a while. There are so many barriers to diagnosis. In fact, a lot of people never get diagnosed with dementia, and others get diagnosed very late,” Professor Sachdev told The Medical Republic.
“And there are lots of barriers to diagnosis in general practice. I think that’s why they [patients first] present in hospital. They’re often presenting not with dementia but with other problems like delirium and it’s an incidental diagnosis.”
The NSW Agency for Clinical Innovation (ACI) report, commissioned by the Australian Institute of Health and Welfare, analysed linked primary and acute healthcare data through the Lumos Data Asset, supplemented by consultations with 10 carers and 13 clinicians.
The Dementia diagnosis, care and management in primary healthcare settings in NSW report was released this week.
Among 91,888 people whose first dementia diagnosis was recorded between July 2014 and June 2024, 72.4% had their diagnosis recorded first or only in hospital data, compared with 27.5% first or only in general practice. Less than 1% had a diagnosis recorded in both on the same date.
Even among people who regularly attended participating practices, 63.1% had dementia first or only recorded in hospital.
Around 1.8% of people in the data were identified with dementia, equivalent to about 134,000 people out of 7.3 million. And on any given day, about one in 10 people in NSW public and private hospitals were identified with dementia.
Other key findings included:
- Nearly 66% of people had their dementia diagnosis first recorded in hospital, with almost 90% admitted for another reason.
- Only 4% of people with dementia were living in residential care before diagnosis, rising to almost one-third (31%) within 5 years.
- There was variation in health service use for people with dementia. Some people had frequent contact with general practice, hospital, or outpatient services, while others had little or no recorded service use each year.
- Compared with people whose first dementia diagnosis was recorded in hospital, those whose diagnosis was first recorded in general practice had more general practitioner appointments and fewer hospital attendances over time.
Dementia was likely under-reported in general practice, the report revealed. Almost seven out of 10 people with a dementia diagnosis recorded in hospital did not have a diagnosis recorded in general practice.
A lack of linked private specialist data limited insights into dementia diagnosis pathways and patterns of care across the whole health system, while data for Aboriginal and Torres Strait Islander peoples was unavailable in the Lumos Data Asset, the report said.
Professor Sachdev said the findings should not be interpreted simply as a failure by GPs.
“Oh no, no, absolutely not. It’s the system, the way the system is structured,” he said.
“And you can’t just fix one thing. We need to create an ecosystem. GPs need training, GPs need incentives, and then GPs need support.”
A major problem was that general practice was not structured or funded for the lengthy assessment required to properly investigate dementia, he said.
“A good assessment of dementia would take you an hour and a half,” he said.
“There are some Medicare items where you can do an annual detailed assessment of an individual.
“But most practices are not set up to take on many long practices because it’s not cost effective for them.”
Related
Older patients frequently presented with multiple competing health problems, meaning possible cognitive decline could be also pushed down the agenda, he said.
“Usually an older person also presents may present with arthritis or heart problem or blood pressure is out of control or diabetes is not being managed, and that gets attended to, and off they go,” Professor Sachdev said.
Even once cognitive decline was suspected, access to comprehensive specialist assessment remained a major bottleneck.
Professor Sachdev said good dementia assessment ideally involved a multidisciplinary team that could include a geriatrician, neurologist, or old-age psychiatrist alongside occupational therapy, neuropsychology, and social work.
But there were too few services with that capacity, he said, pointing to a wait of more than six months at the Prince of Wales Hospital cognitive disorders clinic.
The relationship between specialist services and general practice was also poorly configured for ongoing dementia care, he said, with little capacity for follow-up and GPs not reimbursed for time spent consulting specialists about patients.
“The relationship between GPs and specialist services haven’t been well set up for dementia,” Professor Sachdev said.
“So ideally, when we suggested that part of the Australian Dementia Network program, we developed dementia memory clinics guidelines, and we also said, there should be a network of memory clinics around the country, that are linked to primary care.”
He also backed the introduction of a national team of dementia navigators or care coordinators to guide patients and families through the period after diagnosis – in a similar way to breast cancer nurses operated.
Professor Sachdev said families could be given a long list of medical, legal, and lifestyle recommendations following a memory clinic assessment but be too overwhelmed by the diagnosis to act on them. When followed up months later, “not even half the things” recommended may have been done.
Dementia navigators could support patients and families for the first weeks or months after diagnosis, help implement their care plan and ultimately hand care back to the GP with an avenue for further specialist advice.
“So, I think that the whole ecosystem needs to be developed,” Professor Sachdev said.
The ACI analysis also identified potential opportunities to detect cognitive decline before an eventual dementia diagnosis. About 11% of people had attended hospital for delirium and around 8% had a presentation or admission associated with depression in the preceding 12 months.
People whose dementia was first recorded in general practice subsequently had more GP visits but fewer ED presentations and hospital admissions, while those first recorded in hospital had greater ongoing acute healthcare use.
The report cautioned that further research was needed into the relationship between primary care engagement and long-term outcomes.
Professor Sachdev said one immediate reform could give GPs the time needed to investigate suspected dementia properly.
“I think there should be a dementia assessment item, Medicare item,” he said.
“The government has been reluctant to have disease-specific Medicare MBS items. They’ve resisted that a long period of time, but they have to acknowledge that there are there is a requirement that for certain diseases that you need these kind of assessments, the longish assessments, which should include being able to interview a family member and be reimbursed for that.”
That should be accompanied by GP training and properly supported memory clinics linked directly with primary care, Professor Sachdev said.
He would like to see adequately resourced memory clinics in all hospitals, with direct links to primary healthcare, as the first vital steps to “developing that ecosystem”.
The authors of the ACI report also stressed the importance of data collection.
“This project provides the most comprehensive NSW analysis to date of dementia care pathways using linked primary and acute care data,” they concluded.
“The findings demonstrate that dementia is frequently identified during hospital care rather than in general practice, suggesting opportunities for earlier recognition and intervention.
“The study also highlights the significant value of linked data in understanding dementia prevalence, service use and care pathways and provides evidence for implementation of a national primary care linkage through the National Primary and Acute Care Data Linkage project.
“Strengthening data integration and improving diagnosis recording will support better planning, more coordinated care and improved outcomes for people living with dementia and their carers.”
Read the full report here.
