Table 3.

Selected Studies From Narrative Review^a

	CANCER SCREENING AND TREATMENT
CANCER TYPE/ PRIMARY AUTHOR	MH DX	STUDY DESIGN	DATA COLLECTION/METHODS	SAMPLE AND SETTING	RESULTS/DATA	CONCLUSIONS
Breast/Mitchell 201454	M	SRMA	Review of literature examining whether women with a mental illness are less likely to receive mammography screening	Multiple	Significantly reduced rates of mammography screening in women with mental illness (OR, 0.71; 95% CI, 0.66–0.77), mood disorders (OR, 0.83; 95% CI, 0.76–0.90), and particularly SMI (OR, 0.54; 95% CI, 0.45–0.65)	Mammography rates are lower in women with mental illness, especially SMI
Breast, cervical/ Aggarwal 201355	M	SR	Systematic review of health disparities in breast and cervical cancer screening among women with mental illness	Multiple studies	In total, 19 studies were included; the most commonly discussed facilitator was a relationship with a primary care physician	Breast and cervical cancer disparities persist among women with chronic mental illness, although this population is complex and diverse
Breast, cervical, colon/ Happell 201256	M	Review	Narrative review to examine disparities in preventive health care for cancer and infectious diseases among individuals with SMI	Multiple studies	Individuals with SMI generally had lower screening rates, with 20% to 30% lower receipt of screenings for breast, cervical, and colorectal cancers	The majority of evidence demonstrates poorer quality of preventive services for individuals with SMI compared with the general population
Breast, colon/Druss 201057	M	RCT	Trial comparing medical care management for individuals with SMI vs usual care	N 5 407 patients in CMHC in Atlanta, Georgia	At 12-mo follow-up, the intervention group received an average of 58.7% of recommended preventive services compared with 21.8% in the usual care group (P < .001)	Care management was associated with significant improvement in primary care process and outcome measures
Cervical/Weitlauf 201358	D, PTSD	O	Study comparing receipt of recommended cervical cancer screening in 3 diagnostic groups: 1) PTSD, 2) depression, and 3) no psychiatric diagnosis	N 5 34,213 women from the national VA database	Overall, 77% of women with PTSD, vs 75% of those with depression and 75% of those without psychiatric illness, received cervical cancer screening during the study observation period (P < .001)	VA health care environment may “level the playing field” for those with psychiatric illness
Cervical/Abrams 201259	M	CC	Case-control study comparing rates of cervical cancer screening and acute care visits among women with and without a diagnosis of mental illness	N 5 105,681 female Maryland Medicaid enrollees	There was increased cervical cancer screening in women with psychosis (OR, 1.46), bipolar disorder/mania (OR, 1.59), and depression (OR, 1.78) and decreased OR for women with substance use (OR, 0.8)	People within the Maryland Medicaid system with mental illness appear to be able to access preventive care; more outreach is needed for women with substance use
Multiple/Barley 201360	M	SR	Systematic review of the effectiveness of interventions to encourage cancer screening in adults with SMI	Multiple	There is no evidence for any method of increasing cancer screening specifically for people with mental illness	Evidence-based approaches to increasing cancer screening in individuals with SMI are greatly needed
Multiple/Irwin 201461	S	Review	Review of health disparities in cancer care among patients with schizophrenia	Multiple	Patients with schizophrenia are less likely to have up-to-date cancer screenings	Patient-level, provider-level, and systems-level factors contribute to low rates of cancer screening; psychiatrists can facilitate screening
Breast/Rahman 201462	M	Review	Review article on pathophysiology, clinical implications, and pertinent preclinical data regarding the use of anti- psychotics in patients with breast cancer	Multiple studies	Prolactin promotes breast cancer cell growth regardless of receptor status, and breast cancer patients with elevated prolactin levels have quicker disease progression and lower survival rates; most first-generation antipsychotics significantly elevate serum prolactin levels	A decision to discontinue or change medications requires consideration of the risks and benefits given the patient’s mental illness type and severity, cancer staging, and patient and family preferences
Breast/Sharma 201063	S	O	Cohort study of breast cancer treatment in women with schizophrenia from 1993 to 2009	N 5 90,676 patients from UK NHS records	Thirty women (81%) presented with early breast cancer, and 7 (19%) presented with metastatic disease; treatment outcomes, trial involvement, compliance, and ability to provide informed consent were similar to previously published cohort data	Schizophrenia does not affect treatment delivery or outcomes in women with breast cancer; the presence of schizophrenia should not be a limiting factor for entry into clinical trials
Colon/Baillargeon 201164	M	O	Retrospective review of SEER- Medicare–linked data on all individuals diagnosed with colon cancer between January 1, 1993, and December 31, 2005	N 5 8670 participants aged 67 y and older with a diagnosis of colon cancer in the SEER- Medicare database	Participants with mental illnesses were more likely to have been diagnosed with colon cancer at autopsy (4.4% vs 1.1%; P < .001) and with an unknown stage of cancer (14.6% vs 6.2%; P < .001); to have received no surgery, chemotherapy, or radiation therapy (ARR, 2.09; 95% CI, 1.86– 2.35); and to have received no chemotherapy for stage III cancer (ARR, 1.63; 95% CI, 1.49–1.79)	Public health initiatives are needed to improve colon cancer detection and treatment in older adults with mental disorders
Lung/Bergamo 201465	S	O	Retrospective review of SEER- Medicare–linked data from patients aged 66 y or older with confirmed, primary NSCLC diagnosed between 1992 and 2007	N 5 96,702 patients with NSCLC in SEER- Medicare database	Patients with schizophrenia were less likely to present with late-stage disease, undergo appropriate evaluation, or receive stage-appropriate treatment (OR, 0.82; 95% CI, 0.73–0.93; P < .050 for all comparisons; OR, 0.50; 95% CI, 0.43–0.58); survival was decreased among patients with schizophrenia, although not after controlling for treatment received	Elderly patients with schizophrenia present with earlier stages of lung cancer but are less likely to undergo diagnostic evaluation or to receive stage-appropriate treatment, resulting in poorer outcomes
Multiple/Foti 200566	M	O	Study of treatment preferences in response to hypothetical medical illness scenarios	N 5 150 community- residing adults with SMI in Massachusetts	For the scenario involving pain medication for incurable cancer, most participants chose aggressive pain management, even if cognition might be affected; few participants thought a doctor should provide patients with enough medication to end their life; for the scenario of irreversible coma, respondents were divided in their choice regarding life support	Persons with SMI were able to designate treatment preferences in response to EOL health state scenarios; although most participants had not previously participated in advance care planning, they were interested in the topic and participated
Multiple/Howard 201044	M	Review	Review of multiple articles looking at SMI and cancer incidence and risk factors, screening, and equity of access to treatment and care	Multiple studies	Patients with SMI are less likely to receive cancer screenings; there is insufficient evidence to determine whether unique barriers exist for individuals with SMI; for treatment, patients with SMI have poorer access to diagnostic and treatment services for health complaints and have delays in help seeking; are less likely to receive surgery for esophageal cancer; have more treatment complications; and have higher case-fatality rates for multiple cancers	Severe mental illness is associated with behaviors that predispose an individual to an increased risk of some cancers, disparities in screening for cancer, and higher case-fatality rates; inequalities in care need to be addressed by all health care professionals involved, including those from mental health services and the surgical and oncology teams
Multiple/Damjanovic 200643	S	Review	Review of behaviors that increase the risk of cancer in patients with schizophrenia	Multiple studies	Some antipsychotics increase prolactin, which may increase breast and endometrial cancer risk; studies of overall malignancy risk have conflicting results; some psychotropic drugs can inhibit chemotherapy metabolism	Treatment of cancer in individuals with schizophrenia should include evaluation of risk factors, drug interactions, risk-benefit analysis of treatment options, and efforts to promote treatment adherence
Multiple/Damjanovic 200643	S	Review	Review summarizes known disparities in cancer prevention, diagnosis, treatment, and EOL care among individuals with schizophrenia	Multiple studies	Patients with schizophrenia have delays in diagnosis and treatment, perhaps due to stigma; are less likely to have esophageal or colorectal cancer surgery; have higher postsurgery mortality and postoperative complication rates; and are less likely to participate in clinical trials	Providers should assume decisional capacity and address suicidality, violence, and homelessness

Abbreviations: AOR, adjusted odds ratio; ARR, adjusted relative risk; ART, adjuvant radiation therapy; BMI, body mass index; CC, case-control; CI, confidence interval; CMHCs, Community Mental Health Centers; D, depression; DX, diagnosis; EOL, end of life; M, multiple mental illnesses; MA, meta-analysis; MH, mental health; NAEs, neuropsychiatric adverse events; NSCLC, nonsmall cell lung cancer; O, observational; OR, odds ratio; PTSD, posttraumatic stress disorder; RCTs, randomized controlled trials; RR, relative risk; S, schizophrenia; SEER, Surveillance, Epidemiology, and End Results Program; SMI, serious mental illness; SR, systematic review; SRMA, systematic review and meta-analysis; UK NHS, United Kingdom National Health Service; VA, Veterans Affairs.

^aA more complete version of Table 3 is provided in Supporting Tables 1 through 3 (see online supporting information)