Introduction
Disparities and inequalities in the physical health of individuals affected by severe mental illness (SMI), namely major depressive disorders (MDD), schizophrenia, and bipolar disorders, have been widely reported [Reference DE Hert, Correll, Bobes, Cetkovich-Bakmas, Cohen and Asai1–Reference Fiorillo, de Girolamo, Simunovic, Gureje, Isaac and Lloyd3]. Evidence also suggest a life expectancy 15–20 years shorter and a higher mortality rate from physical disorders among individuals with SMI compared with the general population [Reference Walker, McGee and Druss4–Reference Fiorillo and Sartorius6].
Cancer is one of the most common physical diseases, with the World Health Organization (WHO) worriedly underlining an increase worldwide by 2050 in terms of incidence (35.3 million new cancer cases from the 2022 estimate of 20 million, 76.6% increase) and deaths (18.5 million cancer deaths by 2050, from the 2022 estimate of 9.7 million, 89.7% increase) [7, Reference Bizuayehu, Ahmed, Kibret, Dadi, Belachew and Bagade8].
Over the last 20 years, studies have concentrated attention on the relationship of cancer with SMI in terms of disparities, inequities, treatment issues, ethical concerns, and mortality [Reference Howard, Barley, Davies, Rigg, Lempp, Rose, Taylor and Thornicroft9–Reference Decker, Nelson, Corveleyn, Gopalan and Irwin13]. Among SMIs, MDD has been particularly studied. Depression is one of the most prevalent mental disorders in patients with cancer [Reference Caruso, Nanni, Riba, Sabato, Mitchell and Croce14, Reference Martinez-Calderon, García-Muñoz, Heredia-Rizo and Cano-García15] and it is associated with higher mortality compared with cancer patients without depression [Reference Satin, Linden and Phillips16–Reference Low, Yau, Tan, Ong, Ho and Ho19]. For example, in a large study of more than 20,000 cancer patients, a diagnosis of major depression (MDD), as evaluated after the diagnosis of cancer, was associated with worse prognosis for all cancers in comparison with non-depressed patients, with similar adjusted hazard ratios regardless of cancer site [Reference Walker, Mulick, Magill, Symeonides, Gourley and Burke20] and with a lower survival time among females [Reference Walker, Magill, Mulick, Symeonides, Gourley and Toynbee21]. These findings were in part confirmed by studies individuals with pre-existing depression who later developed cancer. A depressive condition or self-reported scores suggesting MDD were associated with higher mortality from cancer in depressed patients followed over time who subsequently developed various forms of cancer [Reference Yang, Korhonen, Moustgaard, Silventoinen and Martikainen22, Reference Xia, Jiang, Di, Feng, Meng and Xu23]. Similar findings were observed among 1,894 men diagnosed with a depressive disorder in the two years prior to a prostate cancer diagnosis, who exhibited worse overall mortality in comparison with men without depression before cancer diagnosis [Reference Prasad, Eggener, Lipsitz, Irwin, Ganz and Hu24]. A large, multicenter prospective study in the United States, also found higher mortality in women who screened positive for depressive symptoms and were later diagnosed with breast cancer [Reference Liang, Margolis, Hendryx, Reeves, Wassertheil-Smoller and Weitlauf25], although no such association was observed in those who developed colon-rectal cancer [Reference Liang, Hendryx, Qi, Lane and Luo26]. A more recent study showed that women with pre-diagnostic MDD experienced a higher risk of breast cancer mortality than those without prior MDD [Reference Aboumrad, Joshu and Visvanathan27]. There are, however, other studies that did not confirm an increased cancer mortality risk among patients with previous depressive conditions [Reference Min, Cao, Chen, Wang and Xu28, Reference Xu, Wang, Song, Deng and Li29] or with MDD [Reference Nunes, Mancine, Neves, Leite, Nascimento and Pasqualucci30].
In Italy, no study has examined this field, aside from limited data indicating excess of cancer mortality among patients followed by mental health services for various psychiatric disorders, including common mental disorders, schizophrenia, bipolar disorders, and personality disorders [Reference Starace, Mungai, Baccari and Galeazzi31–Reference Grassi, Stivanello, Belvederi Murri, Perlangeli, Pandolfi and Carnevali33]. However, cancer-related mortality in individuals specifically diagnosed with MDD has never been examined as a distinct cohort. For these reasons, the aim of this study was to explore cancer mortality rates among all patients followed by mental health departments for MDD compared with the general population.
Methods
Study setting and sources of data
This study was carried out in the Emilia Romagna Region (ERR), an area in Northern Italy with approximately 4.5 million inhabitants living across nine provinces (Bologna, Ferrara, Forli-Cesena, Modena, Parma, Piacenza, Ravenna, Reggio-Emilia, and Rimini). In this region the transition from an asylum-based to a community-based psychiatric care system occurred before the nationwide reform introduced with the establishment of the Italian National Health System (law 833, 1978) [Reference Fioritti, Lo and Melega34].
For this study, we followed the methodology described elsewhere [Reference Chierzi, Stivanello, Musti, Perlangeli, Marzaroli and De Rossi32, Reference Grassi, Stivanello, Belvederi Murri, Perlangeli, Pandolfi and Carnevali33]. In summary, we used the Regional Mental Health Registry to retrieve demographic data (i.e., age, gender, citizenship, and residency), psychiatric diagnosis, and the date of first admission to the mental healthcare system. The Registry includes demographic and clinical data for all in- and outpatients referring to the 13 Departments of Mental Health serving the nine provinces of the ERR. Demographic variables and diagnoses are recorded at the first clinical assessment and subsequently updated by clinicians on a routine basis. In the ERR, all diagnoses are coded using the WHO International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) [35], as required by the Italian Ministry of Health [36]. For the purposes of the present study, we included only patients aged 18 years or older who had received a diagnosis of MDD (ICD-9 CM codes: 296.2*, 296.3*, 296.90, 296.99, 298.0) from the Mental Health Departments between 1 January 2008 and 31 December 2017. Patients who were not residents of the ERR or who had accessed Mental Health Services prior to 2008 were excluded.
We linked the Mental Health Registry with the Regional Cause of Death Registry using an anonymous identity code. The latter includes all deaths occurring in the Region and contains anonymized information on sociodemographic characteristics, date, setting, circumstances. Causes of death were obtained from the medical section of the death certificate, and classified and coded according to the ICD, tenth revision (Supplementary Table 1), following WHO standards. Both registries undergo systematic quality controls. Change of residence to another Italian region were retrieved from the ERR Registry Office and classified as loss to follow-up, due to the impossibility of obtaining the relevant death certificates.
The ERR Population Archive was used to identify and aggregate data on the age and gender distribution of the general population, which served as the comparator group (https://statistica.regione.emilia-romagna.it/servizi-online/statistica-self-service/popolazione). The study received approval from the local Ethical Committee (N. 341/2019).
Statistical analyses
To describe the causes of death, we calculated the proportional mortality, defined as the number of deaths within a population due to different causes during a given time period, over the total. In particular, we examined proportional mortality for all cancers and site-specific cancers in both the study population and in the ERR population.
To compare the risk of mortality between the regional and the study population, we computed the standardized mortality ratio (SMR), defined as the ratio of observed deaths in the study population divided by the number of expected deaths, if the study population had the same gender, age class-specific rates as the regional population. Rates for the standard population were obtained from the above-mentioned ERR Cause of Death Registry and the Population Archive. An SMR equal to 1 indicates that observed deaths equal the expected number; values greater than 1 indicate higher-than-expected mortality, whereas values below 1 indicate lower-than-expected mortality. SMRs with 95% confidence intervals (CI) were computed by gender (adjusted for age class only) and for each cancer site with more than 3 observed events. Differences were considered statistically significant when the CI of the SMR excluded the value of 1. The denominator used for rate calculations was person-years, defined as the time from the date of the first access to the ERR Mental Health Departments to 31 December 2017 or to the date of emigration to another region or death (if applicable).
Results
Sample characteristics
During the study period, 13,351 patients had at least one access to the ERR mental health departments with a diagnosis of MDD. Table 1 shows the main demographic characteristics of the population at the time of their first psychiatric assessment.
Demographic characteristics of the study population at their first access to the Mental Health Department
Table 1. Long description
The table presents demographic data for a total population of 13,351 individuals (100.00 percent).
* Gender: Women account for 8,638 (64.70 percent) and Men account for 4,713 (35.30 percent).
* Age group in years:
- 18 to 24: 530 (3.97 percent)
- 25 to 44: 3,607 (27.02 percent)
- 45 to 64: 5,525 (41.38 percent)
- 65 to 74: 2,042 (15.29 percent)
- 75 to 84: 1,390 (10.41 percent)
- 85 plus: 257 (1.92 percent)
* Citizenship: Italian citizens number 12,053 (90.28 percent) and Non-Italian citizens number 1,298 (9.72 percent).
* Residency: Urban residents (defined as municipalities with more than 60,000 inhabitants) number 5,061 (37.91 percent) and Rural residents number 8,290 (62.09 percent).
a This includes all residents in municipalities with more than 60,000 inhabitants.
Proportional mortality from cancer
During the study period, there were 454 deaths from cancer in the study population, and 143,311 in the regional population. The proportion of cancer-related deaths was 34.11% in the study population versus 29.6% in the regional population (p < 0.0003) (Table 2).
Number of all-cause deaths, total cancer deaths, and site-specific cancer death with % proportion over the total, for the regional and the study population, in the whole sample, and stratified by gender
Table 2. Long description
The table is organized into three main vertical sections: Whole Sample, Females, and Males. Each section compares the Regional population against the Major depression population using count (n.) and percentage (%).
In the Whole Sample section:
- Total deaths (all causes): Regional 484,102 (100.00%); Major depression 1,331 (100.00%).
- Total deaths from cancer: Regional 143,311 (29.60%); Major depression 454 (34.11%).
- Leading cancer sites for the Major depression population include Trachea, bronchus and lung (7.36%), Colon and rectum (3.76%), and Lymphoid/hematopoietic tissue (3.38%).
In the Females section:
- Total deaths (all causes): Regional 255,264; Major depression 752.
- Total deaths from cancer: Regional 65,137 (25.52%); Major depression 266 (35.37%).
- High-proportion cancer sites for depressed females include Trachea, bronchus and lung (6.65%), Breast (5.32%), and Colon and rectum (4.79%).
In the Males section:
- Total deaths (all causes): Regional 228,838; Major depression 579.
- Total deaths from cancer: Regional 78,174 (34.16%); Major depression 188 (32.47%).
- High-proportion cancer sites for depressed males include Trachea, bronchus and lung (8.29%), Lymphoid/hematopoietic tissue (3.45%), and Liver/biliary tract (2.59%).
Specific cancer sites listed include Esophagus, Stomach, Colon/Rectum, Liver/Gallbladder, Pancreas, Larynx, Trachea/Lung, Melanoma, Breast, Uterus, Ovary, Prostate, Kidney, Bladder, C N S, and Lymphoid tissue.
When examining individual cancer diagnoses, trachea, bronchial, and lung cancers were the most frequent causes of death in both the regional and the study population. Trachea, bronchial, and lung cancers also remained the leading cause of cancer death among female patients with MDD, while in the female population the leading cause of cancer death was breast cancer. In males, lung cancer was always the leading cause.
Standardized cancer mortality ratio
As reported in Table 3, all-cause mortality was 77% higher in patients with major depression than expected based on the regional population (SMR 1.77, 95% CI 1.67–1.86). Cancer-related mortality was 86% higher than in the regional population (SMR 1.86, 95% CI 1.69–2.04). Females showed a higher excess of mortality due to neoplasms compared with males, although the difference was not statistically significant (Table 3).
SMR for all causes of deaths, total deaths for cancer, and cancer site in the whole sample and by gender
Table 3. Long description
The table contains six columns: Observed, Expected, S M R, and a two-column 95% C I. It is divided into three main sections.
1. Whole Sample:
- Total deaths (all causes): Observed 1,331, Expected 753.15, S M R 1.77, C I 1.67 to 1.86.
- Total deaths from cancer: Observed 454, Expected 244.53, S M R 1.86, C I 1.69 to 2.04.
- Selected cancer sites: Esophagus (S M R 2.75), Colon and rectum (S M R 1.98), Trachea, bronchus, and lung (S M R 2.07), Kidney (S M R 2.26), and Lymphoid tissue (S M R 2.20).
2. Females:
- Total deaths (all causes): Observed 752, Expected 451.25, S M R 1.67, C I 1.55 to 1.79.
- Total deaths from cancer: Observed 266, Expected 136.56, S M R 1.95, C I 1.72 to 2.20.
- Selected cancer sites: Colon and rectum (S M R 2.52), Trachea, bronchus, and lung (S M R 2.57), Melanoma (S M R 3.29), and Breast (S M R 1.95).
3. Males:
- Total deaths (all causes): Observed 579, Expected 301.90, S M R 1.92, C I 1.76 to 2.08.
- Total deaths from cancer: Observed 188, Expected 107.97, S M R 1.74, C I 1.50 to 2.01.
- Selected cancer sites: Esophagus (S M R 3.64), Liver and biliary tract (S M R 1.97), Meninges and brain (S M R 3.38), and Lymphoid tissue (S M R 2.30).
Abbreviations: CI, confidence interval; SMR, standardized mortality ratio.
With regard to site-specific cancer diagnosis, the SMR was >1 for all cancer types. In the overall study population, the excess mortality was significant in all cases except for stomach, ovary, larynx cancer, and melanoma. Among females, a significant excess mortality compared to the regional population was observed for colon, rectum, and anus, trachea, bronchus lung, breast, and uterus cancers, as well as for malignancies of the lymphohematopoietic system. Among males, significantly higher mortality was found for cancers of the esophagus, liver, gallbladder and biliary tract, trachea, bronchus and lung, prostate, kidney, pelvis, ureter and unspecified urinary organs, bladder, and for CNS neoplasms (i.e., meninges, brain and other CNS areas) and for malignancies of the lymphohematopoietic system.
Discussion
This is the first Italian large study covering a 10-year period and showing that cancer-related mortality in subjects with pre-existing MDD is higher than in the general population. In addition, the study shows an excess mortality for almost all types of cancer, with no differences between males and females. These findings are consistent with studies conducted in the USA, Korea, and China [Reference Prasad, Eggener, Lipsitz, Irwin, Ganz and Hu24, Reference Aboumrad, Joshu and Visvanathan27], although organizational differences between mental health services in Italy and other countries should be considered.
Results from the present study support the hypothesis that MDD may play a direct or indirect role in cancer pathogenesis and prognosis. More specifically, MDD may be associated with a higher risk of developing cancer, and/or with a higher risk for a worse prognosis. As for a direct role, different possible biological mechanisms underlying MDD may interact with cancer pathophysiology, such as: increased peripheral blood levels of inflammatory cytokines (e.g., interleukin (IL)-1beta, IL-6, tumor necrosis factor and c-reactive protein) [Reference McFarland, Riba and Grassi37–Reference Chang, Sloan, Antoni, Knight, Telles and Lutgendorf39]; activation of the hypothalamic–pituitary–adrenal axis and autonomic nervous system with corresponding deregulation of circadian cortisol rhythm; [Reference Su, Din, Cui, Peng, Zhou and Wang40]; alterations of the immune system [Reference Currier and Nemeroff41, Reference Nettis and Pariante42] and of the kynurenine pathways [Reference Sforzini, Nettis, Mondelli and Pariante43]. All these biological phenomena have been documented among patients with MDD and may play a role in cancer [Reference McFarland, Doherty, Atkinson, O’Hanlon, Breitbart and Nelson44]. On the other hand, cancer and cancer treatment (e.g., chemotherapy, radiation therapy, immunotherapy) may themselves contribute to impaired neurogenesis, oxidative stress, blood–brain barrier degradation through neuroinflammatory process, and damage to molecular patterns, further reinforcing the relationship between depression and cancer prognosis [Reference Reiche, Nunes and Morimoto45–Reference Ahmad, Rizvi, Fatima and Mondal47]. Another possible explanation is that cancer treatments may be less effective among patients with MDD than among those without depression. Preliminary data suggest that depression is an independent predictor of poorer response to primary chemotherapy in patients with breast cancer, even after controlling for several factors (e.g., age, menopausal status, clinical nodal status, tumor size at diagnosis, estrogen receptor status) [Reference Walker, Heys, Walker, Ogston, Miller and Hutcheon48]. In this regard, it has been suggested that reduced levels of brain-derived neurotrophic factor (BDNF),commonly observed in patients with MDD, may contribute to lower responsiveness to cancer drugs [Reference Wu, Si, Yang, Xia, He and Wang49]. All these mechanisms may negatively influenceing the course of cancer, ultimately shortening survival.
An indirect role of depression in worsening cancer prognosis is also plausible. A series of studies, reviews, and meta-analyses have shown that patients with SMI, including MDD, are less likely to receive appropriate screening for cervical, breast, colorectal, and prostate cancer [Reference Murphy, Stone, Presskreischer, McGinty, Daumit and Pollack50–Reference Solmi, Firth, Miola, Fornaro, Frison and Fusar-Poli53], thereby increasing the risk of receiving a late or advanced-stage diagnosis [Reference Mahar, Kurdyak, Hanna, Coburn and Groome54–Reference Kaerlev, Iachina, Trosko, Qvist, Ljungdalh and Nørgård56]. Also, cancer care may be suboptimal among patients with MDD (e.g., increased risk of receiving non-guideline treatment, neoadjuvant chemotherapy, radiotherapy, or targeted new anti-cancer therapy) [Reference Kisely, Crowe and Lawrence57–Reference Seppänen, Daniel, Houzard, Le Bihan, Coldefy and Gandré61]. In addition, the higher prevalence of smoking, substance abuse, chronic hepatitis B or C infections among individuals with mental illness, including MDD, has been identified as a potential contributor to higher cancer mortality [Reference Musuuza, Sherman, Knudsen, Sweeney, Tyler and Koroukian62].
Finally, evidence shows that depression is one of the factors associated with reduced adherence to several anti-cancer treatments, including adjuvant endocrine therapy and oral chemotherapy [Reference Dos Santos, Lange, Gervais, Clarisse, Capel and Barillet63–Reference Sancassiani, Mulas, Madeddu, Massa, La Nasa and Caocci65]. This reduction may occur though mechanisms such as low mood, loss of interest, demotivation, negative beliefs about medicines, and abnormal illness behavior [Reference Grassi, Meggiolaro, Berardi, Sirgo, Colistro and Andritsch66, Reference Grassi, Rosti, Albieri and Marangolo67].
Strengths and limitations
The major strengths of our study are the large-scale cohort design and the considerable follow-up for the patients attending the MHD, with the possibility to independently and consistently collect data on mental disorders and causes of death. A number of limitations should be mentioned, however. First, we were not ble to distinguish between patients with an underlying diagnosis of MDD and those who were diagnosed after receiving a cancer diagnosis. Although we hypothesize that the percentage of the latter was small, we are aware that this group of patients may be phenomenologically similar but biologically very different. Second, the study did not include the entire regional population of people affected by MDD (e.g., those accessing private psychiatric clinics and/or services or those followed exclusively by their general practitioners). A third limitation is that individuals who were not officially residents of the region were not included in the study. Although the proportion of excluded non-residents was low (about 1% of the caseload), this may theoretically have introduced bias in the estimation of mortality within our cohort. Fourth,despite the fairly long study period, the number of events for some site-specific cancers was very small, resulting in estimates with relatively large confidence intervals and highlighting the need for confirmation in larger studies. Another limitation is that the diagnosis of MDD was categorically based on the ICD-9-CM, without any further specification or sub-typing of the clinical characteristics of depression. [Reference Caruso, Nanni, Riba, Sabato and Grassi68] Finally, we were unable to examine the multiple variables that could potentially be related to cancer prognosis, as discussed above (e.g., screening for cancer in patients with MDD, timing of cancer development, adherence to anti-cancer therapy, shared biological mechanisms underlying cancer and depression). Therefore, the links between MDD and higher cancer mortality should be explored in more detailed investigations.
Conclusions
This cohort study showed that patients affected by MDD and followed by ERR departments of mental health have excess mortality from a broad range of site-specific cancers compared to the general population. Further research is needed to investigate the complex relationship between MDD and the prognosis of cancer, as well as the pathophysiological and behavioral mechanisms underlying this association. Given the high number of people worldwideaffected by depression and MDD and, at the same time, the increasing annual incidence of cancer reported by the WHO, the prevention, screening, and treatment of cancer should represent a critical aim for psychiatry [Reference Howard, Barley, Davies, Rigg, Lempp, Rose, Taylor and Thornicroft9, Reference Decker, Nelson, Corveleyn, Gopalan and Irwin13, Reference Grassi and Riba69]. Moreover, closer cooperation between mental health professionals, oncology specialists, and primary care physicians, is essential to facilitate early diagnosis and appropriate treatment among patients who, being primarily affected by MDD, develop cancer [Reference Grassi, Caruso, Riba, Lloyd-Williams, Kissane and Rodin70, Reference Andersen, Lacchetti, Ashing, Berek, Berman and Bolte71] and to ensure continuous monitoring of this disorder over time.
Abbreviations
- CNS
-
central nervous system
- ICD
-
International Classification of Disease Clinical Modification
- WHO
-
World International Organization
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1192/j.eurpsy.2025.10048.
Data availability statement
The datasets generated and/or analyzed during the current study are not publicly available due because of privacy/ethical restrictions, but are available from the corresponding author on reasonable request.
Acknowledgements
The authors are indebted to all the Regional Health Care Services and Mental Health Departments involved in this study.
Author contribution
Conceptualizing, writing and manuscript revision: ES, MBM, MC, VP, PP, EM, BZ, FR, MM, DB, RC LG. Data acquisition: ES, VP, PPData coding and analysis: ESDrafting of the manuscript: MBMData interpretation: MC, EM, BZ, FRSupervision and manuscript revision: MM, DBFinal revision of the paper: RC.
Financial support
Work supported by #NEXTGENERATIONEU (NGEU) and funded by the Ministry of University and Research (MUR), National Recovery and Resilience Plan (NRRP), project MNESYS (PE0000006) – A Multiscale integrated approach to the study of the nervous system in health and disease (DN. 1553 11.10.2022) and by #NEXTGENERATIONEU (NGEU) (Ministry of Health, National Recovery and Resilience Plan (NRRP) project PNRR-MAD-2022-12375899 Cost-effectiveness of innovative, non-pharmacological strategies for early detection, prevention and tailored care of depressive disorders among cancer patients: Transcranial Magnetic Stimulation and Virtual Reality based Cognitive Remediation.
Competing interests
The study was approved by the local Ethical Committee (N. 341/2019). All records in the Mental Health Register were anonymized.
Open access
Comments
No Comments have been published for this article.