What does multi-morbidity mean for cancer treatment and outcomes

Programme stream(s): Treatment

Chair: Peter Murchie, University of Aberdeen, UK
Presenter: Abdulqadr Akinoso-Imran, Queens University Belfast, UK
Presenter: Eilidh Fletcher, Information Services Division, NHS National Services Scotland, UK
Speaker: Alison Fielding, NCRI Consumer Forum, UK
Speaker: David Blane, University of Glasgow, UK
Speaker: Katriina Whitaker, University of Surrey, UK

4:35 pm-6:35 pm

Room: M2-4

An ageing population means that cancer is increasingly diagnosed, treated and survived by older people with co-existing health conditions. This has major implications for diagnosing and treating cancer, and also for caring for cancer survivors in the longer term. In this session international experts will discuss key challenges and state-of-the art research findings with respect to how comorbidities can influence the initial diagnosis of cancer. We will also consider how primary care is meeting the challenge of managing people with cancer as well as other major health conditions. Crucially, the session will also consider how people with cancer experience and deal with comorbidity during their cancer journey.

Learning objectives:
– How do comorbidities influence the diagnosis, treatment and outcomes from cancer?
– How can patients with comorbidities best be supported during their cancer journey?
– What should be our policy and research priorities for comorbidity and cancer?

Abstract presentations:
– Quantifying under-treatment in older adult lung cancer patients in Northern Ireland – Abdulqadr Akinsos-Imran

– Investigating common comorbidities in patients diagnosed with cancer– a Scottish Routes from Diagnosis analysis – Eilidh Fletcher

Multimorbidity in primary care: implications for cancer prevention, diagnosis and support
Speaker: David Blane
Affiliation: University of Glasgow


An ageing population means that cancer is increasingly diagnosed, treated and survived by older people with co-existing health conditions.  In Scotland, most people with any long-term health condition have more than one (multi-morbidity).  This has implications for the diagnosis and management of people with cancer.  This presentation will look at patterns of co-morbidity associated with cancer in Scotland and will consider how this might affect cancer diagnosis and support.  It will also explore the potential role of primary care in cancer prevention, presenting new data on cancer risk discussions. The impact of socio-economic deprivation will be considered. The onset of multimorbidity is approximately 10 to 15 years earlier in the most deprived decile of the Scottish population compared to the most affluent decile.  Furthermore, the health behaviours which account for a significant proportion (up to 40%) of risk for the leading cancers are socially patterned and tend to cluster in individuals.

Understanding the role of multimorbidity on symptomatic presentation and early diagnosis of cancer
Speaker: Katriina Whitaker
Affiliation: University of Surrey


The number of people living with multi-morbidity in the UK is high and projected to rise. This means that the majority of people seeking help for potential cancer ‘alarm’ symptoms do so within an increasingly complex environment, that both the patient and GP must navigate to expedite early cancer diagnosis. Previous research has investigated symptom appraisal and medical help-seeking specifically in the context of cancer, but there is a lack of research exploring how existing morbidity influences how patients present with symptoms. This talk will outline a behavioural science perspective on the issue, and highlight avenues for future research.

Quantifying under-treatment in older adult lung cancer patients in Northern Ireland
Speaker: Abdulqadr Akinoso-Imran
Affiliation: Queen’s University Belfast


Lower treatment rates in older lung cancer patients does not necessarily imply under-treatment due to contraindications from higher levels of advanced disease, comorbidities, and frailty.

Aim: This study aims to quantify how much of the lower survival in older patients in Northern Ireland is mediated through lower treatment rates, adjusting for clinical and demographic factors.


For lung cancer patients diagnosed between 2012-2015, hospital episode, multidisciplinary team meeting, and Northern Ireland Cancer Registry (NICR) information was linked. Age was categorised into young (<74) and old (≥75), and treatment was defined as surgery, chemotherapy, radiotherapy or no treatment. A mediation methodology using the g-computation formula1 was used to estimate 1) the natural indirect effect (NIE) of age on 2-year observed survival mediated through treatment and 2) the natural direct effect (NDE) of age on survival not mediated through treatment. The analysis accounted for stage, comorbidities, performance status (ECOG) and lung function (FEV1/FVC ratio).


A total of 4,233 patients were included in the analysis (1,755 were 75 and over); elderly patients were approximately 80%, 42% and 86% (p<0.001) less likely to undergo surgery, radiotherapy and chemotherapy, respectively. The Total Causal Effect (TCE) of older age (i.e. NIE + NDE) on survival was 14.6% (95% confidence interval [CI], 0.11—0.18, P<0.001) reduction, split between the NIE (6.4%; 95% CI 0.03—0.13, P<0.001), and the NDE (8.2%; 95% CI 0.03—0.10, P=0.001).


Roughly 43% of the survival difference between younger and older lung cancer patients may be due to older cancer patients experiencing lower treatment rates for the same levels of disease stage, comorbidity, pulmonary function, and performance status. This study highlights the potential of offering more treatment to older lung cancer patients, and the need for comprehensive geriatric assessment in these patients with diverse health issues and outcomes.

Investigating common comorbidities in patients diagnosed with cancer– a Scottish Routes from Diagnosis analysis
Speaker: Eilidh Fletcher
Affiliation: Information Services Division, NHS National Services Scotland


Comorbidities occurring following a cancer diagnosis can impact a patient’s survival, level of need, quality of life, and may influence treatment options and outcomes. The Scottish Routes from Diagnosis framework investigated routine hospital admission data to learn more about the levels of comorbidities managed in secondary care following diagnosis amongst our cohorts.


We identified Scottish residents diagnosed with breast (female only), colorectal, lung, or prostate cancer in 2011 through the Scottish Cancer Registry. Cases were linked to hospital activity data using the Community Health Index number.

Hospital admissions in the 12 months following the cancer diagnosis were selected and individual diagnoses codes extracted. The proportion of patients in each cancer cohort with at least one comorbidity recorded was calculated and the most commonly recorded diagnosis codes in each cohort identified.


The highest level of comorbidity was in the lung cancer cohort, with 79% of patients having at least one comorbidity recorded. Amongst other cohorts, 74% of colorectal patients, 43% of prostate patients and 40% of breast patients had at least one comorbidity recorded.

The most commonly recorded diagnosis amongst lung cancer patients was chronic obstructive pulmonary disease (J44): 21% of patients had this recorded. Amongst the other cohorts, primary hypertension (I10) was the most commonly recorded: 9%, 17% and 9% of breast, colorectal and prostate patients having this diagnosis recorded respectively.  These results exclude comorbidities managed exclusively in primary care.


A large proportion of our cancer cohorts experience comorbidity in the year following a diagnosis. Next steps will identify pre-existing comorbidities, and newly presenting after diagnosis, to further understand the impact of comorbidities on those living with cancer. The rising number of people with a cancer diagnosis and at least one other comorbidity will require considering people’s health and needs in their entirety rather than as separate, single conditions.