The effect of increasing age, frailty and cognitive impairment on time to diagnosis of suspected cancer: a population based electronic health records study

Start Date Dec 2021

Code L14-Aff

Status Ongoing

Project Lead
Dr Daniel Jones
Senior Lead
Prof Yoryos Lyratzopolous, Dr Cristina Renzi (both UCL), Dr Oly Todd (Leeds)


Over 60% of cancer diagnoses in the United Kingdom (UK) are in patients aged 65 and over. Cancer recognition, referral, diagnosis and treatment in older adults is complicated by the presence of frailty and existing co-morbidities, leading to lower survival rates and poorer quality of life. Early cancer diagnosis is associated with higher survival rates, increased patient satisfaction and greater treatment options. Frailty is defined as a state of increased vulnerability secondary to age-associated physiological decline in reserves, ultimately compromising the ability to cope with acute stressors. A systematic review of the prevalence of frailty in older patients found 42% of patients with a cancer diagnosis were classified as frail.  The study found that patients with frailty had increased morbidity and mortality and were less likely to tolerate surgical treatment, chemotherapy and radiotherapy.

A systematic review exploring the effect of time to cancer diagnosis on patient outcomes identified improved survival and quality of life in patients with a shorter time to diagnosis, for several cancer types including breast, colorectal and prostate. Whilst early cancer diagnosis is associated with better outcomes, in older and frail patients – and those with cognitive impairment – treatment options may be limited and survival chances reduced. As a result, the benefit of cancer diagnosis in frail older adults has been debated. To investigate this further, we aim to explore the association between old age, frailty, and cognitive impairment on the time to cancer diagnosis.

Aims & objectives

We aim to explore the association between old age, frailty and cognitive impairment on the time to cancer diagnosis.


This population-based cohort study will use the Connected Bradford dataset, which is part of the Connected Health Cities group. The dataset is a longitudinal database containing near real-time anonymised medical records of approximately 868,000 residents in the Bradford district. It includes linked primary and secondary care data from 86 general practices and five NHS Trusts.

We will study adults presenting to primary care services with a recognised sign or symptom that could be due to cancer and a subsequent diagnosis of cancer pertaining to that specific red flag (for example ‘breast lump’ leading to ‘breast cancer’ or ‘haemoptysis’ leading to ‘lung cancer’), over a 13-year period from 1/1/2009 to 25/10/2021.

Descriptive statistics will be used to compare the distributions of time to diagnosis by age, frailty and cognitive impairment. Quantile regression will be used to assess the association between age, frailty, cognitive impairment on time to cancer diagnosis comparing the median, 75th and 90th centiles.

Outputs & impact

This will be the first study to consider the effect of frailty on the time to diagnosis in primary care using a longitudinal data set. The results will provide information on how frailty affects cancer diagnosis. If frailty results in longer times to diagnosis this study could prompt further research into this, and suggest opportunities to improve cancer diagnosis in frail older adults.

Next steps

Following on from this study we will undertake a qualitative study with frail older adults to explore their experience of cancer investigations and referrals to further improve our knowledge of the values and wishes of older adults.

This research may also prompt further large population database work on the management of older and frail adults with cancer symptoms.

Back to Research