Abstract
Multimorbidity, the presence of a chronic condition in addition to cancer, is of particular importance to cancer survivors. It has an impact on the progression, stage at diagnosis, prognosis, and treatment of cancer patients. Evidence is scarce on the prevalence of specific comorbidities in survivors of different cancers to inform prevention and management of multimorbidity. The objective of this study is to address this evidence gap by using large scale electronic health data from multiple linked UK healthcare databases to examine the prevalence of multimorbidity in 28 cancer sites. For this population-based cross-sectional study, we linked primary and secondary healthcare data from the UK Clinical Research Practice Datalink (CPRD) GOLD dataset and Hospital Episode Statistics (HES). We identified survivors of 28 common cancers aged 18 years or older at diagnosis who survived 2 years of cancer and compared their multimorbidity with matched controls without a history of cancer. To compare prevalence of individual comorbidity, multivariable logistic regression models, adjusted for confounding factors were used. Between January 1, 2010 and December 31, 2020, we identified 347,028 cancer survivors and 804,299 controls matched on age, sex and general practice. Cancer survivors had a higher prevalence of multimorbidity compared to non-cancer controls across all the cancer sites. Hypertension (56.2%), painful conditions (39.8%), osteoarthritis (38.0%), depression (31.8%) and constipation (31.4%) were the five most frequent chronic conditions reported. Compared to the controls, higher odds of constipation were found in survivors of 25 of the 28 cancer sites and higher odds of anaemia were found in 23 cancer sites. Prevalence of constipation, anaemia and painful conditions were higher after cancer diagnosis compared to before diagnosis. Since these comorbidities are not uniformly assessed as part of any of the comorbidity scales, they tend to be underreported among cancer survivors. The elevated risk of certain comorbidities in cancer survivors suggests the potential for preventative efforts in this population to lower disease burden and improve quality of life. Long-term conditions should not be viewed as the inevitable result of cancer diagnosis and treatment. We need to consider integrated management of chronic conditions tailored to specific cancers to improve cancer survivorship.
Keywords: Cancer survivor, cancer survivorship, multimorbidity, comorbidity, prevalence, before and after cancer, living with and beyond cancer, chronic conditions, quantitative study, cancer care, cancer treatment
Introduction
Cancer incidence is rising worldwide, and it is the leading cause of death after cardiovascular disease [1]. About 43% of men and 30% of women aged 65 and above will develop cancer at some point in their lives [2]. The global population is rapidly ageing and by 2030, older people will make up 70% of the cancer cases in high income countries such as the USA [3]. Fortunately, 5-year survival rates are nearly 80% for the most common cancers as cancer diagnosis and treatments are improving [4]. Due to these two complementary trends, the population of cancer survivors is growing rapidly.
Multimorbidity is defined as the co-occurrence of two or more long-term conditions (LTC). Most cancer patients are likely to suffer from multimorbidity [2,5,6]. Prevalence of one comorbid condition in cancer patients ranges from 40-69% and prevalence of more than one comorbidity ranges from 12-32% [7,8]. Patterns of multimorbidity vary according to cancer site and stage, as well as other factors such as ethnicity, deprivation, and health status of patients [9]. The number of comorbidities in cancer patients is up to three times greater than age and gender matched controls [6,10].
There are several definitions of cancer survivorship. The National Cancer Institute (NCI) and National Coalition for Cancer Survivorship (NCCS) defines cancer survivorship as “from the time of diagnosis to the balance of life” [11,12]. The National Cancer Survivorship Initiative (NCSI) in the UK defines cancer survivors as “Those who are undergoing primary treatment, those who are in remission following treatment, those who are cured and those with active or advanced disease” [13]. Many authors, particularly in European journals define cancer survivors as “patients without evidence of disease and free from specific treatment for a period of at least 5 years” [14-16]. The 5 years period cut off may not be appropriate for some cancers, such as lung cancer and pancreatic cancer where the majority of patients die within 1 year of diagnosis [17]. In any case, cancer survivors have other medical comorbidities to manage in addition to their cancer.
With increasing cancer survivorship, the prevalence of multimorbidity also increases. The detection of cancer may be influenced by multimorbidity, with some people having their cancer diagnosed sooner due to frequent interactions with healthcare providers, and others being diagnosed later if their cancer shares common symptoms with a comorbid disease, as in the case of chronic obstructive pulmonary disease (COPD) and lung cancer [9]. Cancer treatment may also be associated with several long-term sequelae such as chronic pain and fatigue, sexual dysfunction, anxiety, depression, and lymphedema [18-21]. People with multimorbidity are less likely to receive cancer treatment with curative intent [9]. The presence of multimorbidity leads to the increased complexity of health care needs for cancer survivors. A scoping review conducted by the authors recently found that detailed exploration of the epidemiology of multimorbidity in cancer survivors is lacking for most cancers [22]. Previous studies using primary care data have focused on single cancer site or single comorbidity or comorbidities from the same bodily system (e.g., cardiovascular disease) or older age survivors [5,22-24]. Other studies have explored the prevalence of multimorbidity among cancer patients using secondary care data such as hospital records [5,25]. Such studies may underestimate the prevalence of multimorbidity that are managed in primary care, as many conditions do not typically require hospital admission. Moreover, only a few studies compared comorbidities in cancer survivors with matched controls for such a large group of cancers.
In this study, we aimed to examine the prevalence of 29 comorbidities in cancer survivors in England using population based electronic health records of patients with linked primary and secondary health care data. We focused on patients from 28 different cancer sites and compared them to age, sex, and general practice matched controls.
Materials and methods
Study population
This study is a sub study of a larger study titled “Characterisation of multimorbidity clusters and trajectories using data-driven approaches in a nationally representative population” [26]. Data for this study were obtained from the Clinical Practice Research Datalink GOLD (CPRD GOLD). CPRD prospectively collects primary care records from general practices in the UK that use VISION or EMIS software with coverage of over 11.3 million patients from 674 practices in the UK with 4.4 million active (alive, currently registered) patients [27]. It captures coded diagnoses and care events (using Read codes), prescriptions issued in primary care, and numerical test results (e.g., blood pressure readings). The primary healthcare data were then linked to national data on Hospital Episode Statistics Admitted-Patient Care (HES APC), death registration data - from the Office of National Statistics (ONS) and individual postcode-based index of multiple deprivation data (IMD). Our study population comprises of all acceptable patients permanently registered in any English CPRD General Practice that meet ‘up to standard’ CPRD GOLD data quality criteria between January 1, 2010, and December 31, 2020, and who ever had a recorded long-term condition (from a set of 204 conditions) during this time frame. Patients are labelled as ‘acceptable’ if their data meets acceptable data quality criteria for use in research [27]. The list of long-term conditions was selected after doing a thorough literature search on disease codes that had already been validated in previous UK studies [28-31]. Additional codes which were not validated were added and reviewed by a panel of ten clinicians from a range of specialities. Each panel member independently reviewed the list of conditions and scored them according to relevance to the condition. Any disagreements were further reviewed by a second consensus panel of three clinicians who collaboratively decided on its inclusion/exclusion. The detailed selection procedure is shared on an open GitHub repository [32].
In this population based cross-sectional study, we compared comorbidities in cancer survivors against controls without a history of cancer. From the population described above, we selected patients who had a diagnosis of any of the 28 cancers over the age of 18. Up to three controls were selected based on “no history of cancer”, matched on age (±2 years), sex, and general practice. Cancer diagnosis was defined as the first code of cancer at the site of interest in any of the linked databases. Participants with no information on smoking, body mass index (BMI), or index of multiple deprivation (an area-based UK indicator for socio-economic status) were excluded from this analysis.
Procedures
A cancer survivor was defined in this study as any cancer patient who was alive two years after the date of primary cancer diagnosis. Patients who did not have minimum of two years of follow-up data after cancer diagnosis were excluded. Each cancer survivor was matched with up to three controls.
The outcomes of the study were presence or absence of a long-term condition. From the preliminary list of 204 medical conditions (see Supplementary File), 29 conditions were selected for further analysis. These 29 conditions were selected because they were included in one of the following: the Quality Outcomes Framework (QOF) [33]; the Charlson Comorbidity Index (CCI) [34]; the Cambridge Multimorbidity Score (CMS) [30], or they arose from consultation with a patient-public involvement group (PPI) that included cancer patients and their caregivers. Demographic characteristics such as ethnicity, last known smoking status, mean BMI were identified from the CPRD GOLD dataset. Patients were followed up from the date of registration at their GP practice until date of death or end of study period (1st June 2021).
Statistical analysis
The study described the prevalence of multimorbidity in cancer patients, explored the association between multimorbidity and patient characteristics, and examined the prevalence of specific comorbidities across the 28 cancer groups.
Firstly, we described the median, interquartile range (where appropriate) and proportion of cancer survivors compared to matched controls by patient characteristics (age group, sex, ethnicity, IMD, mortality and number of co-morbidities). We calculated prevalence of comorbidity in cancer survivors and matched controls. We did not standardise for age and sex as cancer survivors were matched for age, sex and general practice already. Then, we compared the prevalence of the same comorbidities before and after cancer diagnosis for the cancer survivors. Subsequently, we fitted multivariable logistic regression models initially accounting only for matching factors (age, sex and general practice) and then additionally adjusting for shared risk factors and demographics (BMI, ethnicity, and smoking status) to find the odds of the comorbidity outcome compared to the matched controls. The primary outcome measure for this study is the presence of a long-term condition as a binary outcome (present vs absent). Crude and adjusted odds ratios for each comorbidity were calculated, adjusted odds ratio with 95% CI was presented. The reference groups were controls with no cancer (but with at least one LTC), white ethnicity, least socio-economic deprived quintile group, never smoker. For both the unadjusted and adjusted analyses, Data are presented as mean (SD) and median (IQR) for continuous data, number (%) for binary data, or as adjusted odds ratios (AOR) with 95% confidence intervals. All statistical analyses were performed in R version 4.2.0.
Results
A total of 472,918 cancer patients and 2,495,601 patients without a cancer diagnosis were extracted from the CPRD GOLD dataset. 631,900 (21.3%) participants were excluded due to missing BMI, IMD and/or smoking status. A further 65,580 cancer patients were excluded due to data anomaly, death within 2 years of cancer diagnosis or cancer diagnosis before the age of 18 (Figure 1). After matching for age, sex, and general practice, 347,028 cancer survivors and 804,299 non-cancer matched controls were eligible for inclusion in the study.
Figure 1.
Study cohort profile. BMI, Body-mass index; IMD, Index of multiple deprivation. *Not possible to determine if reporting error or trans sex individual so omitted.
Table 1 describes the demographics of our study population. White ethnicity was commoner in cancer survivors compared to the matched controls (92.3% vs 80.2%). Cancer survivors were more likely to be former smokers (33% vs 29.6%). Premature death (defined as death before the age of 75) was almost double in cancer survivors compared to controls (7.6% vs 3.4%). Cancer onset was the highest in the 60-75 age group (38.56%). The average number of comorbidities in cancer survivors was 12.4, compared to 9.2 in non-cancer patients (from the set of 204 conditions).
Table 1.
Baseline characteristics of cancer survivors and matched controls across all cancer sites
| Cancer survivors | Non-cancer patients* | |
|---|---|---|
| Total number of patients (n, %) | 347,028 (30.1%) | 804,299 (69.9%) |
| Sex | ||
| Male | 150951 (43.5%) | 318,386 (39.6%) |
| Female | 196077 (56.5%) | 485,913 (60.4%) |
| Age group | ||
| (18, 30) | 912 (0.3%) | 2,579 (0.3%) |
| (30, 45) | 15,127 (4.4%) | 44,122 (5.5%) |
| (45, 60) | 47,947 (13.8%) | 138,710 (17.2%) |
| (60, 75) | 112,056 (32.3%) | 302,589 (37.6%) |
| (75, 90) | 138,743 (40.0%) | 263,682 (32.8%) |
| >90 | 32,243 (9.3%) | 52,617 (6.5%) |
| Ethnicity | ||
| White | 320,413 (92.3%) | 645,164 (80.2%) |
| Mixed | 1,055 (0.3%) | 2,663 (0.3%) |
| Asian or Asian British | 4,746 (1.4%) | 17,309 (2.2%) |
| Black or Black British | 3,539 (1.0%) | 8,873 (1.1%) |
| Chinese or Other Group | 3,473 (1.0%) | 9,430 (1.2%) |
| Unknown | 5,980 (1.7%) | 28,099 (3.5%) |
| Missing | 7,822 (2.3%) | 92,761 (11.5%) |
| IMD quintile | ||
| 1 (least deprived) | 91,507 (26.4%) | 202,657 (25.2%) |
| 2 | 82,035 (23.6%) | 185,407 (23.1%) |
| 3 | 73,086 (21.1%) | 73,086 (21.1%) |
| 4 | 57,544 (16.6%) | 139,601 (17.4%) |
| 5 (Most deprived) | 42,856 (12.3%) | 107,542 (13.4%) |
| Smoking status | ||
| Non-smoker | 180,176 (51.9%) | 433,644 (53.9%) |
| Former smoker | 114,570 (33.0%) | 238,005 (29.6%) |
| Smoker | 52,282 (15.1%) | 132,650 (16.5%) |
| Alcohol status | ||
| Non-drinker | 111,151 (32.0%) | 264,180 (32.8%) |
| Former drinker | 15,793 (4.6%) | 33,370 (4.1%) |
| Current drinker | 184,974 (53.3%) | 421,983 (52.5%) |
| Excess drinker | 4,567 (1.3%) | 11,964 (1.5%) |
| Unknown | 30,543 (8.8%) | 72,802 (9.1%) |
| BMI | ||
| Mean (SD) | 26.7 (±4.9) | 26.8 (±5.2) |
| Median (SD) | 26.7 (±5.0) | 26.8 (±5.2) |
| Male (Mean (SD)) | 27.0 (±4.2) | 27.2 (±4.5) |
| Female (Mean (SD)) | 26.5 (±5.4) | 26.5 (±5.5) |
| Underweight (BMI<18.5) (n, %) | 5,460 (1.6%) | 15,034 (1.9%) |
| Overweight (BMI 25-30) (n, %) | 135,448 (39.0%) | 299,873 (37.3%) |
| Obese (BMI>30) (n, %) | 73,341 (21.1%) | 179,018 (22.3%) |
| Death | ||
| Mean age (SD) | 80.5 (±11.8) | 82.6 (±10.2) |
| Death aged 75 years or older | 74,601 (21.5%) | 112,783 (14.0%) |
| Premature death (death below 75 years) | 26,395 (7.6%) | 26,986 (3.4%) |
| Number of long-term conditions | ||
| Mean (SD) | 12.4 (±6.7) | 9.2 (±5.8) |
| Multimorbidity | ||
| >1 conditions | 346,786 (99.9%) | 801,261 (99.6%) |
| >2 conditions | 340,015 (97.8%) | 747,644 (92.9%) |
| >3 conditions | 328,767 (94.7%) | 683,094 (84.9%) |
| >4 conditions | 313,666 (90.4%) | 614,290 (76.4%) |
| >5 conditions | 295,731 (85.2%) | 546,484 (67.9%) |
| >6 conditions | 275,600 (79.4%) | 481,561 (59.9%) |
Matched with cancer patients on age (±2), sex and general practice.
Prevalence of comorbidity in cancer survivors (all sites combined)
Compared to matched controls, cancer survivors had higher prevalence comorbidity in 22 out of 29 LTCs (Table 2). The comorbidities in order of prevalence were: hypertension (56.2%), painful conditions (39.8%), osteoarthritis (38.0%), depression (31.8%), constipation (31.4%), dermatitis (30.5%), hearing loss (24.4%), coronary heart disease (22.9%), anxiety and phobia (22.4%), chronic kidney disease (CKD) (21.4%), anaemia (19.8%), atrial fibrillation (18.7%), asthma (18.1%), obesity (16.1%) and Type 2 diabetes (15.2%), chronic obstructive pulmonary disease (COPD) (13.1%), osteoporosis (12.8%), heart failure (12.6%), cerebrovascular disease (11.4%), irritable bowel syndrome (10.3%), peptic ulcer disease (9.2%), dementia (8.9%), non-diabetic peripheral neuropathies (6.0%), alcohol problems (6.0%), liver disease (3.6%), epilepsy (3.3%), psychosis/bipolar disorder (1.8%) and chronic fatigue (1.40%).
Table 2.
Prevalence of long-term conditions in cancer survivors compared to matched controls
| Long-term condition (in order of prevalence in cancer survivors) | Cancer survivors (n) | Cancer survivor (%) | Non-cancer controls* (n) | Non-cancer controls* (%) | Difference between survivors and controls (%) |
|---|---|---|---|---|---|
| Hypertension | 194,967 | 56.2% | 395,867 | 49.2% | 7.0% |
| Painful conditions | 137,988 | 39.8% | 279,644 | 34.8% | 5.0% |
| Osteoarthritis | 131898 | 38.0% | 269145 | 33.5% | 4.5% |
| Depression | 110,327 | 31.8% | 267,283 | 33.2% | -1.4% |
| Constipation | 109,110 | 31.4% | 192,889 | 24.0% | 7.4% |
| Dermatitis | 105,991 | 30.5% | 252,455 | 31.4% | -0.9% |
| Hearing loss | 84,646 | 24.4% | 165,219 | 20.5% | 3.9% |
| Coronary artery disease | 79,628 | 22.9% | 152,002 | 18.9% | 4.0% |
| Anxiety and phobia | 77,713 | 22.4% | 183,744 | 22.8% | -0.4% |
| Chronic kidney disease | 74,370 | 21.4% | 130,375 | 16.2% | 5.2% |
| Anaemia | 68,723 | 19.8% | 111,752 | 13.9% | 5.9% |
| Atrial fibrillation | 64,979 | 18.7% | 109,925 | 13.7% | 5.0% |
| Asthma | 62,969 | 18.1% | 148,652 | 18.5% | -0.4% |
| Obesity | 55,878 | 16.1% | 123,614 | 15.4% | 0.7% |
| Type 2 Diabetes | 52,725 | 15.2% | 113,087 | 14.1% | 1.1% |
| Chronic obstructive pulmonary disease (COPD) | 45472 | 13.1% | 78847 | 9.8% | 3.3% |
| Osteoporosis | 44,539 | 12.8% | 73,797 | 9.2% | 3.6% |
| Heart failure | 43,553 | 12.6% | 78,586 | 9.8% | 2.8% |
| Cerebrovascular disease | 39,564 | 11.4% | 75,028 | 9.3% | 2.1% |
| Irritable bowel syndrome | 35,736 | 10.3% | 86,498 | 10.8% | -0.5% |
| Peptic ulcer disease | 31,833 | 9.2% | 55,207 | 6.9% | 2.3% |
| Dementia | 31,015 | 8.9% | 65,780 | 8.2% | 0.7% |
| Non-diabetic peripheral neuropathies | 20,649 | 6.0% | 36,819 | 4.6% | 1.4% |
| Alcohol problems | 20,983 | 6.0% | 48,148 | 6.0% | 0.0% |
| Liver disease | 12,519 | 3.6% | 23,627 | 2.9% | 0.7% |
| Epilepsy | 11,381 | 3.3% | 23,022 | 2.9% | 0.4% |
| Bipolar disorder | 6,342 | 1.8% | 16,517 | 2.1% | -0.3% |
| Chronic fatigue | 4,867 | 1.4% | 12,707 | 1.6% | -0.2% |
Matched with cancer patients on age (±2), sex and general practice.
Although cancer survivors had higher prevalence of most conditions, however, the prevalence of some conditions was not substantially different between cancer survivors and matched controls, e.g., depression. The comorbidities that were more prevalent in cancer survivors (all cancers combined) compared to controls were: constipation (+7.4%), hypertension (+7.0%), anaemia (+5.9%), CKD (+5.2%), painful conditions (+5%), atrial fibrillation (+5%), osteoarthritis (+4.5%), coronary heart disease (+4.0%), hearing loss (+3.9%), osteoporosis (+3.6%), COPD (+3.3%), heart failure (+2.8%), peptic ulcer disease (+2.3%), cerebrovascular disease (+2.1%), non-diabetic peripheral neuropathies (+1.4%) and type 2 diabetes (+1.1%).
Prevalence of comorbidity by cancer site
The prevalence of 29 comorbidities across 28 cancer groups and their matched controls is shown in Figure 2. Hypertension is the most prevalent condition in cancer survivors, regardless of cancer site. This is followed by painful conditions, constipation, depression, and osteoarthritis. Although dermatitis was common across all cancer sites, the prevalence is similar to that of non-cancer controls. Here, we notice patterns of comorbidity vary with cancer site. Alcohol related disease is higher in liver cancer survivors, anaemia is the highest in stomach, multiple myeloma and oesophageal cancer survivors, anxiety and phobia is most prevalent in cervical cancer survivors, constipation is higher in gastro-intestinal cancers and cancers of nearby organs, COPD is the highest in lung cancer, coronary heart disease is the highest in bladder cancer, depression is the highest in cervical cancer, epilepsy is highest in brain cancer, painful conditions are the highest in myeloma, mesothelioma and lung cancer survivors, type 2 diabetes is the highest in liver and pancreatic cancer survivors.
Figure 2.
Prevalence of comorbidity in cancer survivors by cancer site.
The fifteen most common comorbidities were chosen for further analysis using regression modelling to compare prevalence between cancer survivors and controls. Adjustment for shared risk factors had little effect on odds ratios, thus we only report the adjusted odds ratios and 95% CI for each comorbidity (Table 3). Cancer survivors (all sites combined) had higher odds of anaemia, constipation, chronic kidney disease, atrial fibrillation and CKD compared to non-cancer controls. Breast cancer survivors had higher odds of anxiety, constipation and depression. Prostate cancer survivors had higher odds of hypertension, coronary artery diseases and atrial fibrillation. For bladder cancer survivors, we found higher odds of chronic kidney disease, hypertension, and coronary artery disease. Amongst lung cancer patients, there was higher odds of painful conditions, anaemia, and constipation.
Table 3.
Adjusted odds of long-term conditions in cancer survivors compared to matched controls
| Long-term conditions | All cancer sites (AOR, 95% CI) | Breast cancer (AOR, 95% CI) | Prostate cancer (AOR, 95% CI) | Colorectal cancer (AOR, 95% CI) | Bladder cancer (AOR, 95% CI) | Lung cancer (AOR, 95% CI) |
|---|---|---|---|---|---|---|
| Anaemia | 1.42 (1.40-1.43) | 1.19 (1.17-1.22) | 1.37 (1.33-1.41) | 2.43 (2.36-2.50) | 1.70 (1.64-1.77) | 1.69 (1.60-1.78) |
| Anxiety & phobia | 0.94 (0.93-0.95) | 1.24 (1.22-1.26) | 0.59 (0.57-0.61) | 0.79 (0.76-0.81) | 0.77 (0.74-0.80) | 1.04 (0.99-1.10) |
| Asthma | 0.95 (0.94-0.96) | 1.01 (0.98-1.03) | 0.79 (0.77-0.82) | 0.90 (0.87-0.93) | 0.95 (0.91-0.99) | 1.39 (1.32-1.46) |
| Atrial fibrillation | 1.27 (1.26-1.29) | 0.84 (0.82-0.86) | 1.55 (1.51-1.59) | 1.58 (1.54-1.63) | 1.84 (1.77-1.91) | 1.68 (1.59-1.78) |
| Chronic kidney disease | 1.30 (1.29-1.31) | 0.98 (0.96-1.00) | 1.36 (1.32-1.39) | 1.49 (1.45-1.54) | 2.19 (2.11-2.27) | 1.31 (1.23-1.38) |
| Constipation | 1.32 (1.31-1.33) | 1.22 (1.19-1.24) | 1.37 (1.34-1.40) | 1.72 (1.67-1.76) | 1.55 (1.50-1.60) | 1.73 (1.65-1.82) |
| Coronary artery disease | 1.13 (1.12-1.14) | 0.68 (0.67-0.70) | 1.60 (1.56-1.64) | 1.30 (1.26-1.34) | 1.89 (1.83-1.96) | 1.52 (1.45-1.60) |
| Depression | 0.90 (0.89-0.91) | 1.21 (1.19-1.24) | 0.59 (0.58-0.61) | 0.76 (0.74-0.78) | 0.72 (0.69-0.75) | 1.00 (0.96-1.05) |
| Dermatitis | 0.95 (0.94-0.96) | 0.96 (0.95-0.98) | 0.85 (0.82-0.87) | 0.89 (0.86-0.91) | 0.91 (0.87-0.94) | 0.95 (0.90-1.00) |
| Hearing loss | 1.17 (1.16-1.18) | 0.87 (0.86-0.89) | 1.55 (1.52-1.59) | 1.26 (1.23-1.30) | 1.47 (1.41-1.52) | 1.08 (1.03-1.15) |
| Hypertension | 1.24 (1.23-1.25) | 0.96 (0.95-0.98) | 1.77 (1.73-1.82) | 1.65 (1.60-1.69) | 2.11 (2.04-2.19) | 1.59 (1.51-1.67) |
| Obesity | 1.06 (1.05-1.08) | 1.17 (1.14-1.20) | 1.00 (0.97-1.04) | 1.12 (1.08-1.16) | 1.33 (1.27-1.39) | 1.16 (1.08-1.24) |
| Osteoarthritis | 1.09 (1.08-1.10) | 1.13 (1.11-1.15) | 1.11 (1.08-1.14) | 1.10 (1.08-1.13) | 1.26 (1.22-1.30) | 1.17 (1.11-1.23) |
| Painful condition | 1.14 (1.13-1.15) | 1.17 (1.15-1.20) | 1.09 (1.06-1.12) | 1.24 (1.21-1.27) | 1.24 (1.20-1.28) | 1.76 (1.68-1.85) |
| Type 2 diabetes | 1.09 (1.08-1.11) | 0.86 (0.84-0.88) | 1.33 (1.29-1.37) | 1.40 (1.35-1.45) | 1.55 (1.49-1.62) | 1.37 (1.29-1.46) |
In Figure 3, we reported the risk of the fifteen comorbidities in 28 cancer sites compared to controls. We found strong evidence (P<0.05) of higher odds of the following conditions: constipation (25 of 28 cancers), anaemia (23 of 28 cancers), painful conditions (19 of 28 cancers), hypertension (18 of 28 cancers), Type 2 diabetes and obesity (15 of 28 cancers), coronary heart disease (14 of 28 cancers), atrial fibrillations (14 of 28 cancers), CKD (14 of 28 cancers), hearing loss (12 of 28 cancers), osteoarthritis (11 of 28 cancers), anxiety (6 of 28 cancers), depression & asthma (5 of 28 cancers). Odds of dermatitis was only slightly increased in two cancers, namely in non-Hodgkin lymphoma and skin cancer (Figure 3). The number of patients for mesothelioma was below 500, thus findings from this cancer should be interpreted cautiously.
Figure 3.
Prevalence of 15 most common comorbidities in cancer survivors compared to their age, sex and GP practice matched controls, further adjusted for BMI, ethnicity, and smoking status. OR, Odds ratio (adjusted). Dotted line represents OR=1, line of no difference.
Comparison of comorbidity before and after cancer diagnosis
In Table 4, we present the prevalence of comorbidity in cancer survivors before and after cancer diagnosis for all cancer sites combined. We also present this for the five most common cancers separately. We see that there was an increase in prevalence of all fifteen comorbidities after cancer diagnosis, with the highest increases in anaemia (+23.6%), constipation (+20.9%), dermatitis (+14.9%), asthma (+14.4%) and depression (+13.1%). For breast cancer patients, the comorbidities with the highest increase in incidence were anaemia, constipation, and painful conditions. For prostate cancer patients, it was anaemia, depression, and constipation. For colorectal cancer patients, the largest increase before and after cancer diagnosis was found for painful conditions, anxiety/phobia, and constipation. For bladder cancer patients, the largest increase in before and after cancer diagnosis was found in asthma, anaemia, and constipation. For lung cancer patients, largest increase in before and after cancer was found in anaemia, constipation, and painful conditions. For all five malignancies, a notably higher prevalence in anaemia and constipation was observed after cancer diagnosis.
Table 4.
Comparison of prevalence of comorbidity before and after cancer diagnosis (all sites combined in order of difference in prevalence)
| Comorbidity | Prevalence before cancer (%) | Prevalence after cancer (%) | Absolute difference (%) |
|---|---|---|---|
| Constipation | 49.3 | 59.6 | 10.3 |
| Anaemia | 43.3 | 53.5 | 10.2 |
| Painful conditions | 54.3 | 61.3 | 7.0 |
| Dermatitis | 45 | 51.7 | 6.7 |
| Depression | 47.5 | 53.7 | 6.2 |
| Asthma | 38.2 | 43.7 | 5.5 |
| Anxiety and phobia | 42.3 | 47.4 | 5.1 |
| Hearing loss | 46 | 50.3 | 4.3 |
| Chronic kidney disease | 50 | 53.5 | 3.5 |
| COPD | 42.7 | 44.9 | 2.2 |
| Obesity | 39.9 | 42 | 2.1 |
| Atrial fibrillation | 49.6 | 50.9 | 1.3 |
| Type 2 Diabetes | 39.8 | 41 | 1.2 |
| Hypertension | 68.9 | 69.8 | 0.9 |
| Coronary artery disease | 49.6 | 49.7 | 0.1 |
Discussion
In this comprehensive study of the prevalence of multimorbidity in cancer survivors compared to age, sex, and general practice matched controls, we found that cancer survivors had higher prevalence of 22 of 29 comorbidities. Of all the comorbidities, anaemia, constipation, chronic kidney disease, atrial fibrillation and COPD were the five that had notably greater odds in cancer survivors compared to the controls. The comorbidities with the highest difference in prevalence proportion between survivors and controls were constipation, hypertension, anaemia, CKD and painful conditions. The pattern of comorbidity varied by cancer sites with some interesting observations. Survivors of blood and gastro-intestinal cancers had an increased risk for many of the 29 conditions. We observed a statistically significant increase in odds of constipation in 25 of 28 cancers and anaemia in 23 of 28 cancers. Constipation was particularly high in myeloma, secondary malignancy, and pancreatic cancer survivors. We observed almost four times greater odds of anaemia in survivors of stomach cancer and 2.5 times or higher in myeloma and biliary tract survivors. Similarly, odds of type 2 diabetes were increased by 200% or more and obesity was increased by almost 150% in pancreatic, liver, and biliary tract cancer survivors. Coronary heart disease had the highest odds in bladder, prostate, and stomach cancer survivors. Depression and anxiety had the highest odds in cancers of females, e.g., cervical, breast, and ovarian cancers. Myeloma survivors had increased odds of every comorbidity except anxiety, depression, and dermatitis. When comparing the prevalence of comorbidities before and after cancer diagnosis, we found higher prevalence of anaemia, constipation, and painful conditions in survivors of all cancer sites combined and in the five most common cancers (Table 5).
Table 5.
Comparison of prevalence of comorbidity by cancer site (five most common cancers)
| Comorbidity | Breast cancer | Prostate cancer | Colorectal cancer | Bladder cancer | Lung cancer | |||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
|
|
|
|
||||||
| Prevalence before cancer | Prevalence after cancer | Prevalence before cancer | Prevalence after cancer | Prevalence before cancer | Prevalence after cancer | Prevalence before cancer | Prevalence after cancer | Prevalence before cancer | Prevalence after cancer | |
| Anaemia | 33.7 | 47.9 | 44.1 | 60.7 | 62.0 | 63.7 | 48.2 | 59.3 | 46.4 | 63.3 |
| % Difference | 14.2 | 16.6 | 1.7 | 11.1 | 16.9 | |||||
| Anxiety and phobia | 42.8 | 48.2 | 39.6 | 48.4 | 39.7 | 48.4 | 44.3 | 50.0 | 48.1 | 56.7 |
| % Difference | 5.4 | 8.8 | 8.7 | 5.7 | 8.6 | |||||
| Asthma | 36.6 | 41.4 | 40.8 | 45.5 | 43.9 | 43.9 | 41.4 | 52.9 | 49.7 | 59.7 |
| % Difference | 4.8 | 4.7 | 4.3 | 11.5 | 10 | |||||
| Atrial fibrillation | 44.6 | 51.4 | 50.4 | 53.3 | 53.6 | 55.9 | 51.9 | 57.5 | ||
| % Difference | 1.3 | 6.8 | 2.9 | 2.3 | 5.6 | |||||
| Chronic kidney disease | 42.7 | 47.0 | 47.9 | 54.8 | 49.5 | 55.4 | 58.8 | 64.6 | 49.0 | 52.4 |
| % Difference | 4.3 | 6.9 | 5.9 | 5.8 | 3.4 | |||||
| Constipation | 43.2 | 56.6 | 52.7 | 67.4 | 56.2 | 67.6 | 54.8 | 64.4 | 56.4 | 71.7 |
| % Difference | 13.4 | 14.7 | 11.4 | 9.6 | 15.3 | |||||
| COPD | 33.2 | 37.4 | 38.8 | 44.9 | 40.0 | 45.8 | 51.2 | 57.3 | 82.4 | 82.5 |
| % Difference | 4.2 | 6.1 | 5.8 | 6.1 | 0.1 | |||||
| Coronary artery disease | 40.6 | 42.1 | 47.8 | 55.6 | 49.2 | 51.4 | 58.3 | 59.7 | 55.7 | 57.4 |
| % Difference | 1.5 | 7.8 | 2.2 | 1.4 | 1.7 | |||||
| Depression | 49.2 | 55.3 | 42.9 | 55.3 | 44.8 | 52.0 | 48.8 | 55.0 | 54.8 | 61.7 |
| % Difference | 6.1 | 12.4 | 7.2 | 6.2 | 6.9 | |||||
| Dermatitis | 43.0 | 50.2 | 46.0 | 55.4 | 45.2 | 52.4 | 46.3 | 53.5 | 46.1 | 52.1 |
| % Difference | 7.2 | 9.4 | 7.2 | 7.2 | 6.0 | |||||
| Hearing loss | 39.4 | 44.4 | 51.0 | 58.4 | 46.2 | 52.6 | 49.9 | 54.7 | 43.5 | 49.0 |
| % Difference | 5 | 7.4 | 6.4 | 4.8 | 5.5 | |||||
| Hypertension | 63.7 | 65.1 | 71.5 | 77.2 | 72.5 | 75.4 | 77.5 | 78.5 | 70.5 | 70.9 |
| % Difference | 1.4 | 5.7 | 2.9 | 1.0 | 0.4 | |||||
| Obesity | 37.9 | 41.7 | 38.5 | 47.1 | 43.8 | 48.7 | 48.9 | 55.0 | 37.6 | 43.0 |
| % Difference | 3.8 | 8.6 | 4.9 | 6.1 | 5.4 | |||||
| Painful conditions | 51.5 | 61.6 | 53.5 | 66.9 | 54.9 | 67.7 | 57.8 | 65.4 | 65.5 | 76.0 |
| % Difference | 10.1 | 13.4 | 12.8 | 7.6 | 10.5 | |||||
| Type 2 diabetes | 36.2 | 39.2 | 36.6 | 46.2 | 44.7 | 45.6 | 45.9 | 49.1 | 42.7 | 45.3 |
| % Difference | 3.0 | 9.6 | 0.9 | 3.2 | 2.6 | |||||
Our finding that cancer survivors have significantly higher odds of reporting most comorbid diseases is consistent with similar studies [6,35-37]. Hypertension, painful conditions, and depression were found to be the most prevalent conditions in cancer survivors by other studies [5,10,38]. Our study found that myeloma and bladder cancer survivors had the highest odds of almost all comorbidities whereas a similar previous study found this to be true for lung cancer survivors [5]. This could be because our study excluded patients who died within two years of diagnosis - which would have excluded many lung cancer patients. However, results from a large study conducted in New Zealand concur with our findings [10].
Our finding that there are substantial differences in association between multimorbidity and cancer groups is consistent with other large similar studies [5,10,23]. Constipation as an important comorbidity in cancer survivors was highlighted previously [39], however there were not many studies reporting constipation in cancer survivors. This is because constipation is not uniformly assessed as part of any of the comorbidity scales and therefore is not recorded in many database-based studies of cancer survivors. Increased odds of constipation across 25 of 28 cancer sites is a unique finding of this study. The reasons behind the high prevalence of constipation in cancer survivors may be multifaceted - organic, functional or drug induced. It can be related to the cancer-site, e.g., gastro-intestinal (G) obstruction, pelvic tumour mass in GI and gynaecological cancers. It can be drug induced as opioids, pain killer medications and chemotherapy treatments induce constipation [39]. This may be further exacerbated by a low-fibre diet, low fluid intake, sedation, confinement to bed and low physical activity in cancer survivors [40]. The higher odds of anaemia in cancer patients have been reported previously [41-43], although ours is the only study to examine it across so many cancers. Higher odds of anaemia in blood and gastrointestinal cancers was also reported in a systematic review of anaemia in cancer patients [41]. There are several explanations which may contribute to higher odds of anaemia in cancer survivors including: cancer cells infiltrating the bone marrow and directly suppressing hemopoiesis. Cancer cells can also release cytokines that lead to iron sequestration and consequent reduced erythrocyte production. There may also be chronic blood loss from tumour sites. Erythrocytes and haemoglobin can be destroyed by cytotoxic chemotherapy. Additionally, there are a variety of ways that erythrocyte production may be hampered, including dietary deficits, bone marrow failure, and a muted erythropoietin response to tissue hypoxia [43,44].
Higher prevalence of type 2 diabetes and obesity in pancreatic, liver, and biliary tract cancer is consistent with their risk profiles and other similar studies [25,45]. Statistically higher odds of obesity and type 2 diabetes were found in 15 out of 28 cancers. These associations emphasise the need to combat obesity and type 2 diabetes. Our study found higher odds of depression compared to controls in only 5 out of 28 cancers, however there was a 13% increase in prevalence of depression after cancer diagnosis. In our study, elevated levels of depression were found in cancer related to female such as cervical, breast and ovarian cancer survivors, a finding reported in previous studies [46-49]. This may be hormone and treatment related, e.g., tamoxifen can increase the likelihood of depression [50]. However, two studies from the 1990s found no difference in depression between male and female cancers [51,52]. Awareness and diagnosis of depression and a reduction in the potential stigma associated with mental health conditions may have improved considerably since then [53].
A major strength of this study is the size and breadth of the linked data sources that allowed us to analyse associations between multimorbidity in cancer survivors in granular detail. The significance of this is demonstrated by the variability observed in the association of cancer site and comorbidities, showing how important variations can be masked when comorbidities or cancers are grouped together. This study had the data and statistical power available to analyse 29 conditions in 28 cancers, specifically focusing on fifteen of the most common conditions. We used a consistent methodology across a wide range of cancers and outcomes and found results consistent with the existing literature, giving confidence that the observed patterns of multimorbidity represent real phenomena rather than methodological artefacts or chance. The benefits of employing CPRD GOLD data connected to secondary care data sources, notably for cancer have been shown by numerous validation studies [54-56]. Our results are generalisable to the UK and similar settings as CPRD GOLD data have been shown to be representative of the general UK population on key demographics [27].
Our study does have some limitations. We were unable to link to cancer registry data, however a previous study [55] has found little difference in the date of diagnosis of cancer between CPRD and cancer registry data. We were unable to access primary care prescribing data which would have helped to elucidate the potential causes of some of the associations seen. All the controls included in this study had at least one of the 204 conditions, however, this could be construed as a strength as it means we are likely to have under-estimated the effect of comorbidity in cancer survivors compared to the controls. Furthermore, the list of 204 conditions was so comprehensive that it is likely that most of the adult population will have at least one of these conditions in their lifetime. Some individuals had missing BMI and smoking data, our complete-case analysis was predicated on the assumption that, relationships between cancer and morbidity do not differ between included and excluded participants [57]. However, even after exclusion of these patients the study is sufficiently powered due to its very large sample size.
This study did not have access to data on cancer treatments; thus, it was not possible to study associations between treatments and comorbidities. The contribution of cancer treatment to subsequent morbidity is very significant [58]. Given that cancer treatment changed dramatically in the last decade; it is important for future studies to account for the different treatment modalities (e.g., the novel immunotherapies which have different side effects).
Another useful research area would be developing a standard approach for measuring multimorbidity in cancer patients. A range of metrics of multimorbidity are utilised in the scientific literature to explore the association between comorbidity and cancer outcomes, but no consensus has been established on a gold standard measure of multimorbidity designed for cancer patients. Charlson’s comorbidity index (CCI) is the most widely used comorbidity index to measure comorbidity. There are a few drawbacks to utilising CCI for cancer survivors. Since it does not account for some of the most common conditions that are related to cancer survivors, such as chronic pain, chronic fatigue, depression, anaemia, and constipation, the impacts of these comorbidities on cancer survivors cannot be determined by this index. CCI was created by including conditions that were associated to lower survival rates in the general population (57); these conditions might not have equal weighting for cancer survivors. Therefore, creating a multimorbidity index that is primarily designed for cancer survivors might be helpful for future research.
The effect of multimorbidity on primary and secondary care use by cancer patients and the associated costs are substantial [59-61]. This study highlights the comorbidities that are most common in cancer survivors and how these comorbidities vary by cancer site. These findings can be used to support integrated management of comorbidities customised to individual cancers through anticipatory care. These comorbidities can also be embedded in cancer care reviews in the UK, which have been reported to lack patient-centredness [62]. It also paves the way for further research into influence of comorbidity on each cancer site - including patient’s quality of life, treatment, mortality, healthcare resource use and healthcare budgeting, which will help to enhance planning of healthcare systems and tailoring treatments hence improving cancer survivorship care.
Acknowledgements
We gratefully acknowledge Professor Rohini Mathur for her helpful advice on the statistical analysis plan and East London Genes and Health team for their help on preparing the code list for the 204 long-term conditions. This report is independent research funded by the National Institute for Health Research ARC North Thames. The views expressed in this publication are those of the author(s) and not necessarily those of the National Institute for Health Research or the Department of Health and Social Care.
Disclosure of conflict of interest
None.
Abbreviations
- AOR
Adjusted Odds Ratio
- BMI
Body Mass Index
- CI
Confidence Interval
- CCI
Charlson Comorbidity Index
- CKD
Chronic Kidney Disease
- CMS
Cambridge Multimorbidity Score
- COPD
Chronic Obstructive Pulmonary Disease
- CPRD
Clinical Practice Research Datalink
- CS
Cancer survivor
- HES
Hospital Episode Statistics
- LTC
Long term condition
- ONS
Office of National Statistics
- IMD
Index of Multiple Deprivation
- NCI
National Cancer Institute
- NCSI
National Cancer Survivorship Initiative
- NCSS
National Coalition for Cancer Survivorship
- OR
Odds ratio
- PPI
Patient public involvement
- QOF
Quality of outcomes framework (QOF is a voluntary annual reward and incentive programme for GP practices in England, Wales and Northern Ireland to encourage them to record common long-term conditions from a set of 34 long-term conditions)
Supporting Information
References
- 1.Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71:209–249. doi: 10.3322/caac.21660. [DOI] [PubMed] [Google Scholar]
- 2.Hayat MJ, Howlader N, Reichman ME, Edwards BK. Cancer statistics, trends, and multiple primary cancer analyses from the Surveillance, Epidemiology, and End Results (SEER) Program. Oncologist. 2007;12:20–37. doi: 10.1634/theoncologist.12-1-20. [DOI] [PubMed] [Google Scholar]
- 3.Smith BD, Smith GL, Hurria A, Hortobagyi GN, Buchholz TA. Future of cancer incidence in the United States: burdens upon an aging, changing nation. J. Clin. Oncol. 2009;27:2758–2765. doi: 10.1200/JCO.2008.20.8983. [DOI] [PubMed] [Google Scholar]
- 4.UK Cancer Research. Cancer Incidence for all cancers 2021 [cited 2022 11th January] Available from: https://www.cancerresearchuk.org/health-professional/cancer-statistics/incidence#heading-Zero.
- 5.Fowler H, Belot A, Ellis L, Maringe C, Luque-Fernandez MA, Njagi EN, Navani N, Sarfati D, Rachet B. Comorbidity prevalence among cancer patients: a population-based cohort study of four cancers. BMC Cancer. 2020;20:2. doi: 10.1186/s12885-019-6472-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ogle KS, Swanson GM, Woods N, Azzouz F. Cancer and comorbidity: redefining chronic diseases. Cancer. 2000;88:653–663. doi: 10.1002/(sici)1097-0142(20000201)88:3<653::aid-cncr24>3.0.co;2-1. [DOI] [PubMed] [Google Scholar]
- 7.Sorensen HT. Multimorbidity and cancer outcomes: a for more research. Clin Epidemiol. 2013;5(Suppl 1):1–2. doi: 10.2147/CLEP.S47149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Haase KR, Hall S, Sattar S, Ahmed S. Living with cancer and multimorbidity: a qualitative study of self-management experiences of older adults with cancer. Eur J Oncol Nurs. 2021;53:101982. doi: 10.1016/j.ejon.2021.101982. [DOI] [PubMed] [Google Scholar]
- 9.Sarfati D, Koczwara B, Jackson C. The impact of comorbidity on cancer and its treatment. CA Cancer J Clin. 2016;66:337–350. doi: 10.3322/caac.21342. [DOI] [PubMed] [Google Scholar]
- 10.Ng HS, Koczwara B, Roder D, Vitry A. Changes in the prevalence of comorbidity in the Australian population with cancer, 2007-2014. Cancer Epidemiol. 2018;54:56–62. doi: 10.1016/j.canep.2018.03.010. [DOI] [PubMed] [Google Scholar]
- 11.National Cancer Institute. Cancer survivorship definition 2022. Available from: https://www.cancer.gov/publications/dictionaries/cancer-terms/def/survivorship.
- 12.National Coalition for Cancer Survivorship. About NCCS. Available from: http://www.canceradvocacy.org.
- 13.Department of Health MSNI. Living with & beyond cancer: taking action to improve outcomes (an update to the 2010 The National Cancer Survivorship Initiative Vision) 2013 [Google Scholar]
- 14.Mattioli V, Montanaro R, Romito F. The Italian response to cancer survivorship research and practice: developing an evidence base for reform. J Cancer Surviv. 2010;4:284–289. doi: 10.1007/s11764-010-0143-9. [DOI] [PubMed] [Google Scholar]
- 15.Simonelli C, Annunziata MA, Chimienti E, Berretta M, Tirelli U. Cancer survivorship: a challenge for the European oncologists. Ann Oncol. 2008;19:1216–1217. doi: 10.1093/annonc/mdn384. [DOI] [PubMed] [Google Scholar]
- 16.Giacalone A, Lleshi A, Zanet E, Tirelli U. Symptom burden in cancer survivors 1 year after diagnosis: a report from the American Cancer Society’s Studies of Cancer Survivors. Cancer. 2012;118:1955. doi: 10.1002/cncr.26443. [DOI] [PubMed] [Google Scholar]
- 17.Khan NF, Rose PW, Evans J. Defining cancer survivorship: a more transparent approach is needed. J Cancer Surviv. 2012;6:33–36. doi: 10.1007/s11764-011-0194-6. [DOI] [PubMed] [Google Scholar]
- 18.Thong MS, Mols F, Lemmens VE, Creemers GJ, Slooter GD, van de Poll-Franse LV. Impact of chemotherapy on health status and symptom burden of colon cancer survivors: a population-based study. Eur J Cancer. 2011;47:1798–1807. doi: 10.1016/j.ejca.2011.02.006. [DOI] [PubMed] [Google Scholar]
- 19.Deng J, Ridner SH, Dietrich MS, Wells N, Wallston KA, Sinard RJ, Cmelak AJ, Murphy BA. Prevalence of secondary lymphedema in patients with head and neck cancer. J Pain Symptom Manage. 2012;43:244–252. doi: 10.1016/j.jpainsymman.2011.03.019. [DOI] [PubMed] [Google Scholar]
- 20.Schmidt ME, Chang-Claude J, Vrieling A, Heinz J, Flesch-Janys D, Steindorf K. Fatigue and quality of life in breast cancer survivors: temporal courses and long-term pattern. J Cancer Surviv. 2012;6:11–19. doi: 10.1007/s11764-011-0197-3. [DOI] [PubMed] [Google Scholar]
- 21.Jacobs LA, Shulman LN. Follow-up care of cancer survivors: challenges and solutions. Lancet Oncol. 2017;18:e19–e29. doi: 10.1016/S1470-2045(16)30386-2. [DOI] [PubMed] [Google Scholar]
- 22.Eto F, Samuel M, Henkin R, Mahesh M, Ahmad T, Angdembe A, McAllister-Williams RH, Missier P, Reynolds NJ, Barnes MR, Hull S, Finer S, Mathur R. Ethnic differences in early onset multimorbidity and associations with health service use, long-term prescribing, years of life lost, and mortality: an observational study using person-level clustering in the UK Clinical Practice Research Datalink. medRxiv. 2023 doi: 10.1371/journal.pmed.1004300. 2023.2003.2003.23286751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Strongman H, Gadd S, Matthews A, Mansfield KE, Stanway S, Lyon AR, Dos-Santos-Silva I, Smeeth L, Bhaskaran K. Medium and long-term risks of specific cardiovascular diseases in survivors of 20 adult cancers: a population-based cohort study using multiple linked UK electronic health records databases. Lancet. 2019;394:1041–1054. doi: 10.1016/S0140-6736(19)31674-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Yancik R, Havlik RJ, Wesley MN, Ries L, Long S, Rossi WK, Edwards BK. Cancer and comorbidity in older patients: a descriptive profile. Ann Epidemiol. 1996;6:399–412. doi: 10.1016/s1047-2797(96)00063-4. [DOI] [PubMed] [Google Scholar]
- 25.Sarfati D, Gurney J, Lim BT, Bagheri N, Simpson A, Koea J, Dennett E. Identifying important comorbidity among cancer populations using administrative data: prevalence and impact on survival. Asia Pac J Clin Oncol. 2016;12:e47–56. doi: 10.1111/ajco.12130. [DOI] [PubMed] [Google Scholar]
- 26.MULTIPLY initiative. List of long-term conditions. Available from: https://github.com/Fabiola-Eto/MULTIPLY-Initiative/blob/main/multimorbidity_list.md.
- 27.Herrett E, Gallagher AM, Bhaskaran K, Forbes H, Mathur R, van Staa T, Smeeth L. Data resource profile: clinical practice research datalink (CPRD) Int J Epidemiol. 2015;44:827–836. doi: 10.1093/ije/dyv098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet. 2012;380:37–43. doi: 10.1016/S0140-6736(12)60240-2. [DOI] [PubMed] [Google Scholar]
- 29.Kuan V, Denaxas S, Gonzalez-Izquierdo A, Direk K, Bhatti O, Husain S, Sutaria S, Hingorani M, Nitsch D, Parisinos CA, Lumbers RT, Mathur R, Sofat R, Casas JP, Wong ICK, Hemingway H, Hingorani AD. A chronological map of 308 physical and mental health conditions from 4 million individuals in the English National Health Service. Lancet Digit Health. 2019;1:e63–e77. doi: 10.1016/S2589-7500(19)30012-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Payne RA, Mendonca SC, Elliott MN, Saunders CL, Edwards DA, Marshall M, Roland M. Development and validation of the Cambridge multimorbidity score. CMAJ. 2020;192:E107–E114. doi: 10.1503/cmaj.190757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Ashworth M, Durbaba S, Whitney D, Crompton J, Wright M, Dodhia H. Journey to multimorbidity: longitudinal analysis exploring cardiovascular risk factors and sociodemographic determinants in an urban setting. BMJ Open. 2019;9:e031649. doi: 10.1136/bmjopen-2019-031649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.MULTIPLY St. Operationalisation of multimorbidity. Available from: https://github.com/Fabiola-Eto/MULTIPLY-Initiative/blob/main/METHODS.md.
- 33.Digital N. Quality and Outcomes Framework (QOF)-Disease prevalence and care quality achievement rates. 2022. Available from: https://digital.nhs.uk/data-and-information/data-tools-and-services/data-services/general-practice-data-hub/quality-outcomes-framework-qof.
- 34.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–383. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 35.Edwards BK, Noone AM, Mariotto AB, Simard EP, Boscoe FP, Henley SJ, Jemal A, Cho H, Anderson RN, Kohler BA, Eheman CR, Ward EM. Annual Report to the Nation on the status of cancer, 1975-2010, featuring prevalence of comorbidity and impact on survival among persons with lung, colorectal, breast, or prostate cancer. Cancer. 2014;120:1290–1314. doi: 10.1002/cncr.28509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Mao JJ, Armstrong K, Bowman MA, Xie SX, Kadakia R, Farrar JT. Symptom burden among cancer survivors: impact of age and comorbidity. J Am Board Fam Med. 2007;20:434–443. doi: 10.3122/jabfm.2007.05.060225. [DOI] [PubMed] [Google Scholar]
- 37.Bluethmann SM, Mariotto AB, Rowland JH. Anticipating the “Silver Tsunami”: prevalence trajectories and comorbidity burden among older cancer survivors in the United States. Cancer Epidemiol Biomarkers Prev. 2016;25:1029–1036. doi: 10.1158/1055-9965.EPI-16-0133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Petrova D, Catena A, Rodríguez-Barranco M, Redondo-Sánchez D, Bayo-Lozano E, Garcia-Retamero R, Jiménez-Moleón JJ, Sánchez MJ. Physical comorbidities and depression in recent and long-term adult cancer survivors: NHANES 2007-2018. Cancers (Basel) 2021;13:3368. doi: 10.3390/cancers13133368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Wickham RJ. Managing constipation in adults with cancer. J Adv Pract Oncol. 2017;8:149–161. [PMC free article] [PubMed] [Google Scholar]
- 40.Mancini I, Bruera E. Constipation in advanced cancer patients. Support Care Cancer. 1998;6:356–364. doi: 10.1007/s005200050177. [DOI] [PubMed] [Google Scholar]
- 41.Knight K, Wade S, Balducci L. Prevalence and outcomes of anemia in cancer: a systematic review of the literature. Am J Med. 2004;116(Suppl 7A):11S–26S. doi: 10.1016/j.amjmed.2003.12.008. [DOI] [PubMed] [Google Scholar]
- 42.Spivak JL. The anaemia of cancer: death by a thousand cuts. Nat Rev Cancer. 2005;5:543–555. doi: 10.1038/nrc1648. [DOI] [PubMed] [Google Scholar]
- 43.Mercadante S, Gebbia V, Marrazzo A, Filosto S. Anaemia in cancer: pathophysiology and treatment. Cancer Treat Rev. 2000;26:303–311. doi: 10.1053/ctrv.2000.0181. [DOI] [PubMed] [Google Scholar]
- 44.Grotto HZ. Anaemia of cancer: an overview of mechanisms involved in its pathogenesis. Med Oncol. 2008;25:12–21. doi: 10.1007/s12032-007-9000-8. [DOI] [PubMed] [Google Scholar]
- 45.Coebergh JW, Janssen-Heijnen ML, Post PN, Razenberg PP. Serious co-morbidity among unselected cancer patients newly diagnosed in the southeastern part of The Netherlands in 1993-1996. J Clin Epidemiol. 1999;52:1131–1136. doi: 10.1016/s0895-4356(99)00098-0. [DOI] [PubMed] [Google Scholar]
- 46.Kim SH, Son BH, Hwang SY, Han W, Yang JH, Lee S, Yun YH. Fatigue and depression in disease-free breast cancer survivors: prevalence, correlates, and association with quality of life. J Pain Symptom Manage. 2008;35:644–655. doi: 10.1016/j.jpainsymman.2007.08.012. [DOI] [PubMed] [Google Scholar]
- 47.Massie MJ. Prevalence of depression in patients with cancer. J Natl Cancer Inst Monogr. 2004:57–71. doi: 10.1093/jncimonographs/lgh014. [DOI] [PubMed] [Google Scholar]
- 48.Golden RN, McCartney CF, Haggerty JJ Jr, Raft D, Nemeroff CB, Ekstrom D, Holmes V, Simon JS, Droba M, Quade D. The detection of depression by patient self-report in women with gynecologic cancer. Int J Psychiatry Med. 1991;21:17–27. doi: 10.2190/HW7K-B3P8-3T8H-073M. [DOI] [PubMed] [Google Scholar]
- 49.Watson M, Haviland JS, Greer S, Davidson J, Bliss JM. Influence of psychological response on survival in breast cancer: a population-based cohort study. Lancet. 1999;354:1331–1336. doi: 10.1016/s0140-6736(98)11392-2. [DOI] [PubMed] [Google Scholar]
- 50.Cathcart CK, Jones SE, Pumroy CS, Peters GN, Knox SM, Cheek JH. Clinical recognition and management of depression in node negative breast cancer patients treated with tamoxifen. Breast Cancer Res Treat. 1993;27:277–281. doi: 10.1007/BF00665698. [DOI] [PubMed] [Google Scholar]
- 51.Sneed NV, Edlund B, Dias JK. Adjustment of gynecological and breast cancer patients to the cancer diagnosis: comparisons with males and females having other cancer sites. Health Care Women Int. 1992;13:11–22. doi: 10.1080/07399339209515974. [DOI] [PubMed] [Google Scholar]
- 52.Fife BL, Kennedy VN, Robinson L. Gender and adjustment to cancer: clinical implications. J Psychosoc Oncol. 1994;12:1–21. [Google Scholar]
- 53.Pescosolido BA, Halpern-Manners A, Luo L, Perry B. Trends in public stigma of mental illness in the US, 1996-2018. JAMA Netw Open. 2021;4:e2140202. doi: 10.1001/jamanetworkopen.2021.40202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Herrett E, Thomas SL, Schoonen WM, Smeeth L, Hall AJ. Validation and validity of diagnoses in the general practice research database: a systematic review. Br J Clin Pharmacol. 2010;69:4–14. doi: 10.1111/j.1365-2125.2009.03537.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Arhi CS, Bottle A, Burns EM, Clarke JM, Aylin P, Ziprin P, Darzi A. Comparison of cancer diagnosis recording between the clinical practice research datalink, cancer registry and hospital episodes statistics. Cancer Epidemiol. 2018;57:148–157. doi: 10.1016/j.canep.2018.08.009. [DOI] [PubMed] [Google Scholar]
- 56.Zhu Y, Edwards D, Mant J, Payne RA, Kiddle S. Characteristics, service use and mortality of clusters of multimorbid patients in England: a population-based study. BMC Med. 2020;18:78. doi: 10.1186/s12916-020-01543-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.White IR, Carlin JB. Bias and efficiency of multiple imputation compared with complete-case analysis for missing covariate values. Stat Med. 2010;29:2920–2931. doi: 10.1002/sim.3944. [DOI] [PubMed] [Google Scholar]
- 58.Maher EJ. Managing the consequences of cancer treatment and the English National Cancer Survivorship Initiative. Acta Oncol. 2013;52:225–232. doi: 10.3109/0284186X.2012.746467. [DOI] [PubMed] [Google Scholar]
- 59.Glynn LG, Valderas JM, Healy P, Burke E, Newell J, Gillespie P, Murphy AW. The prevalence of multimorbidity in primary care and its effect on health care utilization and cost. Fam Pract. 2011;28:516–523. doi: 10.1093/fampra/cmr013. [DOI] [PubMed] [Google Scholar]
- 60.Palladino R, Tayu Lee J, Ashworth M, Triassi M, Millett C. Associations between multimorbidity, healthcare utilisation and health status: evidence from 16 European countries. Age Ageing. 2016;45:431–435. doi: 10.1093/ageing/afw044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Van Oostrom SH, Picavet HS, De Bruin SR, Stirbu I, Korevaar JC, Schellevis FG, Baan CA. Multimorbidity of chronic diseases and health care utilization in general practice. BMC Fam Pract. 2014;15:61. doi: 10.1186/1471-2296-15-61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Gopal DP, Ahmad T, Efstathiou N, Guo P, Taylor SJC. What is the evidence behind cancer care reviews, a primary care cancer support tool? A scoping review. J Cancer Surviv. 2023;17:1780–1798. doi: 10.1007/s11764-022-01251-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.