Recorded Dementia Diagnoses September 2020

The Recorded Dementia Diagnoses publication reports the prevalence of dementia including the number of patients who have received a formal dementia diagnosis and information about their care. Prevalence of a disease is the proportion of a population who are affected or identified with a medical condition at a specific time – each month the data is extracted as a snapshot on the last day of the month to support the Prime Minister’s Dementia Challenge, learn more in About this release. As not everyone with dementia receives a formal diagnosis, this publication also compares the number of people expected to have dementia with the number of people diagnosed with dementia, aged 65 and over.

The publication includes the indicator, Dementia: 65+ Estimated Diagnosis Rate (Dementia Diagnosis Rate) which compares the number of people thought to have dementia with the number of people diagnosed with dementia, aged 65 and over. The methodology uses the prevalence rates of the Cognitive Function and Ageing Study II to estimate how many people aged 65 or over one would expect to have dementia to form the denominator. The national target is for at least two thirds of people with dementia to be diagnosed

Dementia Diagnosis rate =     count of patients with a recorded dementia diagnosis

                                    patients at a GP practice (5-year age band and sex) x CFAS II reference rates

The outbreak of coronavirus (COVID-19) has led to changes in the work and behaviour of General Practices and subsequently the data within Recorded Dementia Diagnoses publication has been impacted. The data is extracted through General Practice Extraction Service (GPES) therefore the burden of the COVID-19 outbreak has not affected the data collection of this publication. There has however been an impact on the data reported in the publication which will be discussed in this spotlight report.

To understand the impact of COVID-19, the report will explore the evidence of impact on the prevalence of dementia – this will include the effect on the count of patients with a recorded dementia diagnosis, considering deaths and healthcare activity, and population used to calculate the rate of diagnosis.

The number of patients with a recorded diagnosis of dementia decreased between 31^st March and 30^th April 2020. Figure 1 shows before March 2020 the number of patients with a recorded diagnosis of dementia was stable at approximately 471,000 patients. From April 2020, the number of dementia patients continues to decrease aligning with the initial peak of the COVID-19 pandemic and change in primary care activity.

When performing statistical analysis on the dementia registers, the total number of patients with a diagnosis of dementia in December 2019, January 2020 and February 2020 (before COVID-19 pandemic) was compared to the total number patients with a diagnosis of dementia in May 2020, June 2020 and July 2020 (during COVID-19 pandemic) (details in the technical annex). The analysis shows the number of recorded dementia diagnoses for patients of all ages during COVID-19 pandemic (May, June and July 2020) is significantly fewer than the number of recorded dementia diagnoses of all ages before the COVID-19 pandemic. It also shows the number of recorded dementia diagnoses for patients aged 65+ during the COVID-19 pandemic is significantly fewer than before COVID-19 pandemic.

The number of patients with a recorded dementia diagnoses is the numerator of the dementia diagnosis rate calculation. The decrease can be attributed to:

• Deaths in dementia patients
• Impact on new diagnoses
• Change in health care activity during the pandemic

Deaths in dementia patients

Public Health England have reported the number of excess deaths due to COVID-19 between 20^th March 2020 and 21^st August 2020. The report uses the underlying causes of death (UCOD) which mention COVID-19 and those without COVID-19 mentioned, outlining the details of the methodology and caveats. Please note the classification of a patient with dementia is different between the PHE report and the Recorded Dementia Diagnoses publication – in the PHE report, the registration of death will include dementia or Alzheimer’s disease.

For deaths with a mention of specific cause for dementia and Alzheimer’s patients, there was a high number which mentioned COVID-19 during April and May 2020. Within the report, figure 36 shows the weekly excess deaths by date of registration with all mentions of dementia and Alzheimer’s disease for England. For the week of 24^th April, 2,442 excess deaths were recorded mentioning COVID-19 from a total of 3,804 excess deaths mentioning dementia and Alzheimer’s disease.

Another report by the Office of National Statistics found dementia and Alzheimer disease was the most common main pre-existing condition found among deaths involving COVID-19 and was involved in 49.5% of all deaths of care home residents involving COVID-19.

The statistics demonstrate the excess deaths due to COVID-19 on patients with dementia or Alzheimer’s disease. This supports the decreasing prevalence and negative impact of COVID-19 amongst patients with a recorded diagnosis of dementia.

Impact on new diagnoses, from measures within Recorded Dementia Diagnoses

Within the Recorded Dementia Diagnoses publication, the measure included report the count of patients who have received or declined an initial memory assessment, a referral to a memory clinic and a care plan. With the hesitance to access health care and the impact on memory assessment services, the measure Number of patients with a record of receiving an assessment for dementia by the GP practice who have a record of a referral to a memory clinic was selected to analyse the impact of COVID-19 from April 2020.

Number of patients with a record of receiving an assessment for dementia by the GP practice who have a record of a referral to a memory clinic (ASS_Received_GP_Mem_Clinic)

The data for this measure is collected as a cumulative count from October each year to September. In Figure 2 below, the count of patients with a record of receiving an assessment for dementia by the GP practice who have a record of a referral to a memory clinic (blue columns) increases by a similar amount each month, average = 2,393 patients. The grey columns demonstrates the change in the count each month – between November 2019 and February 2020 the line is consistent with a small increase in January. From March 2020, the count of patients in this measure no longer increases at the same rate with almost no change in April and May 2020.

To understand the impact on the measure, regression analysis was used to predict the number of patients with a record of receiving an assessment for dementia by the GP practice who have a record of a referral to a memory clinic if primary care activity had not been altered by COVID-19. Figure 3 shows the predicted number of patients compared to the actual number within the data, with the bars showing the difference each month. Using data from July 2020, it is estimated 7,590 patients were not referred to memory clinics since February.

Change in health care activity

COVID-19 pandemic required health and social care settings to adapt and change their approaches to ensure patient safety and limit transmissions. The variation in approach to appointment management between general practices is likely to be greater than usual including changes in clinical practice and use of GP appointment recording systems which are the source of data for the Appointments in General Practice publication.

The data shows the total number of planned and scheduled appointments recorded in GP practice systems has declined during the coronavirus outbreak however it is important to note that this decline does not necessarily imply that GPs are having fewer interactions with patients. Practices are likely to be operating very differently in response to the pandemic and consequently may be recording encounters and interactions in different ways.  One such change could be an increase in the use of list appointments, in which several patients are contacted but only one notional appointment slot appears in the collected data. With the change in guidance for social distancing, and under revised NHS England operating guidance, much of the contact between General Practice and patients has moved to telephone or video service. More information can be found on the supporting page for the publication.

Figure 4 shows the decrease in face to face appointments for England with GPs and other practice staff alongside the other appointment modes from September 2019, including the initial peak of COVID-19.

Partial or total ‘triage’ systems have taken over from routine appointment booking, to reduce risks for patients and service staff. These innovations do not have universal recording or reporting standards in place yet – so much of this capacity, flexibility and consultation activity is not reflected in the GP appointment statistics collected and published.

The pandemic impacted memory assessment services and GP referrals into those services. At the onset of lockdown most memory assessment services were temporarily halted, due in part to the instruction for older people to remain at home. GP services were also affected and as memory assessment services reopened there was further impact from reduced referral numbers and hesitance from many people to access external medical services for a period of time.

The number of patients estimated to have dementia is calculated using the number of patients registered at a general practice who are aged 65 and over and multiplying the age group by reference rates from the CFASII study. It is the denominator of the dementia diagnosis rate calculation and the impact can be attributed to:

• Deaths within the population due to COVID-19
• Change in patients registered at a general practice

Deaths in the population involving COVID-19

Throughout the COVID-19 pandemic, there has been an increased number of deaths which are reported by the Office of National Statistics (latest release at time of publishing is week ending 25 September 2020).

COVID-19 has had a large impact on the number of deaths registered over the last few months and is the main reason for deaths increasing above what is expected (the five-year average). The disease has had a larger impact on those most vulnerable (for example, those who already suffer from a medical condition) and those at older ages.

The data shows the increase in deaths compared to the past 5 years and increase in deaths involving COVID-19 nationally. The number of deaths involving COVID-19 were highest in April 2020. It also reports, in the year to date reported, deaths involving COVID-19 are captured in all age groups however they are higher in the older age groups, with those aged 85 to 89 years accounting for the highest number of deaths involving COVID-19. The report also includes the number of excess deaths based on place of occurrence, including care homes.

Change in patients registered at a GP Practice

During the period of national lockdown, the Patients Registered at a GP Practice publication showed a decrease. GP practice list sizes change due to births, deaths, immigration, emigration, and administration of GP list sizes to remove duplicate patients. The decrease in registered patients seen this period is likely due to an increase in deaths, a decrease in immigration, and potentially increased list cleansing activities brought about by additional activities to contact vulnerable patients. Throughout April 2020, there was less activity surrounding registrations in primary care however there are still new registrations although the overall number shows a decrease. This data is extracted from an administrative source, GP Payments system (Open Exeter) therefore data collection process has not been affected.

Figure 5 shows the number of patients registered at a GP practice in England since August 2019 who were aged 65 and over. Before the initial peak of COVID-19 there was a consistent increase however, on 1st May 2020, there was a decrease in the number of registered patients in this age group, approximately 23,000, due to a combination of the reasons described above. The population of patients aged 65 and over is used with the rates from the CFAS II study to calculate the estimates within the indicator therefore a change in this population due to COVID-19 could impact the rate of dementia diagnosis.

The annual Recorded Dementia Diagnoses publication for 2019-20 covers key facts taken from the monthly releases covering April 2019 to March 2020, including an ARIMA forecast of estimated diagnosis rate. The aim set out in the 2015 Prime Minister’s Challenge on Dementia 2020 is for at least two thirds (66.7%) of people with dementia to have received a formal diagnosis. In the past year, the national estimated dementia diagnosis rate for those aged 65 and over has consistently exceeded this ambition, however at no point has it become statistically significantly greater.

The confidence intervals used alongside the rate are derived from measures of uncertainty given with the CFAS II reference rates and the uncertainty around the numerator. The indicator is calculated to produce an overall distribution of indicator values closely approximating the true distribution. Since April 2020, the rate and confidence intervals have been decreasing demonstrating an impact from the initial peak of the COVID-19 pandemic.

Please note, in previous years and before the COVID-19 pandemic, the rate of diagnosis for dementia varies across different areas of England. This report has focused on the data and impact of COVID-19 at a national level for England and has not explored the regional variance.

When the diagnosis rate recorded in April 2020 is plotted against the ARIMA forecast, which used data up to 31^st March 2020, the model shows the published diagnosis rate to be an outlier, falling far outside the prediction confidence intervals. This suggests the decrease observed was not a likely outcome given the trends seen in the previous 40 months of data and therefore a associated with the coronavirus (COVID-19) pandemic.

Since 31^st March 2020, the diagnosis rate has continued to decrease from 67.4% to 63.1% on the 30^th August 2020.  In summary, the indicator has been impacted by the following points, as outlined in the report

Number of patients diagnosed with dementia, the numerator for the rate

Deaths in dementia patients

Reports by Public Health England and Office of National Statistics show there were excess deaths for patients which mentioned dementia and Alzheimer’s along with COVID-19. For the week of 24^th April, 2,442 excess deaths were recorded mentioning COVID-19 from a total of 3,804 excess deaths mentioning dementia and Alzheimer’s disease.

Decrease in referrals and other care of dementia patients

With hesitance to attend health care services, advice for older people to remain at home and a reduction in memory clinic services, there has been a decrease in referrals and other measures included in the Recorded Dementia Diagnoses publication. Analysis of the measure Number of patients with a record of receiving an assessment for dementia by the GP practice who have a record of a referral to a memory clinic estimates 7,590 patients were not referred to memory clinics since February 2020.

Change in healthcare activity

With the change in guidance for social distancing, and under revised NHS England operating guidance, activity in general practice has moved to telephone or video service with a decrease in face to face appointments.

Estimated number of dementia patients, the denominator for the rate

Increase in deaths across the population

Throughout the COVID-19 pandemic, there has been an increased number of deaths which are reported by ONS, with those aged 85 to 89 years accounting for the highest number of deaths involving COVID-19.

CFASII methodology

The methodology is designed to reflect a population similar to the population within the study. The CFASII estimated population may not reflect the current population due to increase in deaths across the population and the older age groups

Decrease in patients registered at a general practice, aged 65 and over

The decrease in registered patients this period is likely due to an increase in deaths, a decrease in immigration, and potentially increased list cleansing activities brought about by additional activities to contact vulnerable patients. Throughout April 2020, there was less activity surrounding registrations in primary care including those aged 65 and over which decreased by approximately 23,900 patients in May 2020. This is the population used to estimate the number of patients suffering with dementia in England.

Decreased number of patients on the dementia register

Two sample t-Tests produced the following statistics:

T-test, P = 0.0000154, P< 0.05 therefore the number of recorded dementia diagnoses for patients of all ages in May, June and July is significantly less than the number of recorded dementia diagnoses of all ages in Dec, Jan and Feb.

T-test, P = 0.0000140, P< 0.05 therefore the number of recorded dementia diagnoses for patients aged 65+ in May, June and July is significantly less than the number of recorded dementia diagnoses in Dec, Jan and Feb

Number of patients with a record of receiving an assessment for dementia by the GP practice who have a record of a referral to a memory clinic (ASS_Received_GP_Mem_Clinic)

Regression analysis to calculate the predicted number of patients with a record of receiving an assessment for dementia by the GP practice who have a record of a referral to a memory clinic to compare to the actual number of patients in the data set.