Summary of studies included in review
| Source; study design; final quality rating | Setting and sample | Intervention and workforce | Primary outcomes (secondary outcomes) | Results |
| Regular and timely access to care and treatment | ||||
| Basu et al
25 Observational: care control ++ | Women aged 61+ with breast cancer, n=86 One cancer centre, USA | I: Patient navigation: support and coordination of patient care T: From point of diagnosis to survivorship clinic C: No navigation W: Breast cancer nurse | Time from diagnosis to oncology appointment | Time to consultation decreased by 4.9 days (p=0.0002) |
| Goodwin et al.
27 RCT ++++ | Women aged 65+ newly diagnosed with breast cancer, n=335 13 community and 2 public hospitals, USA | I: Case management: nurse as educator, counsellor, advocate and care coordinator T: 12 months service C: Usual care (unclear) W: Nurse case manager | Treatment received in 6 months after breast cancer diagnosis (patient satisfaction; arm function) | More intervention women saw radiation oncologist (36% vs 19.3%) (p=0.006), received more breast-conserving surgery (28.6% vs 18.7%; p=0.031) and radiation therapy (36.0% vs 19.0%; p=0.003), had more breast reconstruction surgery (9.3% vs 2.6%, p=0.054); reported choice in treatment (82.2% vs 69.9%, p=0.020) No differences between groups in percentages who saw an oncologist, discussed breast reconstruction, underwent complete surgical staging, or tissue sent for hormone receptor assay |
| Mandelblatt et al
35 Controlled before and after study ++ | Women aged 65+ screening for breast or cervical cancer, n=673 Two public hospitals, USA | I: Screening intervention during routine visits T: At scheduled appointments C: Physician reminder system W: Nurse practitioner | Annual screening rates for Pap tests and mammographies | Annual intervention site Pap test rate increased (17.8%–56.9%), and mammographies (18.3%–40%) compared with control site increase of Pap test rate from 11.8% to 18.2% and no change (18%) for mammography (p=0.01) |
| Somana-Ehrminger et al
32 Observational ++ | Women aged 75+ with breast cancer, n=206 Breast and gynaecological cancer registry, France | I: Geriatrician referral and treatment plan C: Patients with no GOC W: Geriatrician, dietitian, psychologist, physical therapist or social worker | Independent impact of GOC | GOC patients more likely to receive mastectomy and adjuvant therapy (p<0.0001); and less likely to be treated by breast-conserving surgery and adjuvant therapy (p=0.003) |
| Complications and specific problems of cancer treatment | ||||
| Bourdel-Marchasson et al
34 RCT ++ | Chemotherapy patients aged 70+, n=336 12 public and private settings, France | I: Face-to-face dietary counselling T: 6 visits (3–6 months) C: Usual care W: Dietitian | 1-year mortality (chemotherapy management; unplanned hospitalisation; 2-year mortality) | Difference of 178 kcal/day dietary intake in intervention group (p<0.01) No difference in other outcomes |
| Hempenius et al
30 RCT ++ | Frail adults aged 65+, elective surgery for solid tumour, n=260 Teaching hospital and community hospital, Netherlands | I: Delirium prevention: assessment, monitoring and individualised treatment plan T: During hospital stay C: Usual care. Additional geriatric care provided on referral W: Geriatric team supervised by a geriatrician | Incidence of postoperative delirium up to 10 days (severity of delirium; length of hospital stay; complications; mortality; care dependency; QOL) | Significant difference in return to preoperative living situation (67.3% vs 79.1%, OR: 1.84, 95% CI 1.01 to 3.37) No significant difference in other outcomes |
| Kalsi et al
44 Observational ++ | Adults aged 70+ with cancer, n=135 One hospital, UK | I: Geriatrician CGA and intervention plan for identified need T: Prechemotherapy and further support as needed C: Standard oncology care W: Geriatrician | CGA impact on chemotherapy tolerance and toxicity; rate of planned completion of cancer (treatment modifications; early treatment discontinuation; death at 6 months) | Intervention more likely to complete planned cancer treatment (33.8% vs 11.4%, OR 4.14 (95% CI 1.50 to 11.42), p=0.006) and fewer required treatment modifications (43.1% vs 68.6%, OR 0.34 (95% CI 0.16 to 0.73), p=0.006) Lower toxicity rate in intervention group (43.8% vs 52.9%, p=0.292) No differences in death rates |
| McCorkle et al
37 RCT ++++ | Adults aged 60+ with postsurgical cancer, n=375 Comprehensive cancer centre, USA | I: Specialised home care APNs assess and monitor physical, emotional and functional status of patients, provide direct care, access services and other resources from the community, and provide teaching, counselling and support during recovery T: 4 weeks with three home visits and five telephone contacts C: Usual follow-up care in an ambulatory setting and routine outpatient follow-up W: APNs | Length of survival (depressive symptoms; symptom distress; functional status) | Late-stage patients, improved 2-year survival in intervention group: 66.7% vs 39.6% (p<0.05). No difference for early-stage patients Risk of death higher for control (adjusted HR 2.04; 95% CI 1.33 to 3.12; p=0.001) compared with intervention group No differences between groups in other outcomes |
| Comorbidities and complex health needs | ||||
| Deliens et al.
33 Uncontrolled before and after study + | Adults aged 70+ with cancer (non-haematological) hospitalised, n=91 Geriatric oncology unit, tertiary hospital, Belgium | I: Medication review: identification of PIMs and drug interactions T: From point of admission and during hospitalisation C: Before to after W: Clinical pharmacist | PIMs using START and STOPP criteria (drug-to-drug interactions) | START criteria: 41 PIMs for 31 patients (34%) at hospital admission compared with 7 PIMs for 6 persons (7%) at discharge STOPP criteria: 50 PIMs for 29 patients (32%) at admission compared with 16 PIMs for 14 persons (16%) at discharge |
| Fann et al
26 RCT +++ | Adults aged 60+ with diagnosis of non-skin cancer and major depression or dysthymia, n=215 18 primary care clinics at 8 diverse healthcare organisations, USA | I: Depression management: education, ‘behavioural activation’, treatment support. T: Up to 12 months. Follow-up usual care 12 months more C: Usual care: received routinely available depression treatment W: Depression care manager (nurse or clinical psychologist) collaborative with primary care | Depression treatment response (health-related QOL; health-related impairments: work, family, social functioning; patient satisfaction) | Intervention twice as likely to experience a depression treatment response at 12 months than control (39% vs 20%; p=0.029) and at 18 months (38% vs 16%; p=0.012) Remission rates higher in intervention group versus control group at 6 months (32% vs 16%, p=0.006) and 12 months (22% vs 9%, p=0.031) Less functional impairment at 12 months (p=0.011) and greater QOL (p=0.039) |
| Herr et al
22 RCT (cluster) +++ | Adults aged 65+ with cancer receiving hospice care, n=738 Staff: nurses (n=383 pre, n=415 post) and physicians (n=16) 16 hospices, USA | I: Workforce: to promote adoption of evidence-based pain practices. Included: training, assessment of data, champion input, senior leadership engagement T: Engagement phase 5 months, 12-month intervention C: Hospices received clinical practice guidelines W: 3 days training. Selection of local pain facilitators, nurse and physician champions, grant expert nurse input, nurse and physician champion | Workforce: adoption of evidence-based cancer pain practices (pain severity) | No significant difference in improvement on cancer pain practice index between intervention and control Decrease in patient pain severity from pre to post in intervention group greater (p=0.1032) |
| Johansson et al
41 RCT ++ | Adults aged 70+ newly diagnosed with prostate, GI or breast cancer, n=161 Primary healthcare services, Sweden (other participants reported: n=255 under 70 years) | I: Intensified primary healthcare. Individual support: nurse support, nutritional support and individual psychological support. T: Starting from diagnosis C: Standard care + group rehabilitation W: Home care nurse, dietitian and psychologist. GPs and nurses trained in pain, nausea and diet in final-stage life | Utilisation of specialist care | Mean days of hospitalisation for older intervention patients than control (3.8 vs 8.9, p<0.01) 4 of 82 older intervention patients admitted compared with 12 of 79 older control patients (p<0.05) 10 out of 82 made acute visits to outpatient clinics compared with 22 of 79 in control group (p<0.05) |
| Rao et al
31 RCT +++ | Adults aged 65+ with cancer, frail and hospitalised, n=99 11 medical centres, USA | I: Assessment and monitoring by geriatric team: (1) geriatric inpatient + usual outpatient; (2) usual inpatient + geriatric outpatient; (3) geriatric inpatient and outpatient T: 1-year study C: Usual care: all hospital services except from geriatric team W: Core team: geriatric medicine attending physician, fellow or intern, a nurse practitioner, social worker | Survival; health-related QOL (functional status; physical performance) | No difference in survival for patients with cancer regardless of treatment group Significant effect of geriatric inpatient care versus usual inpatient care: mean change in score from randomisation to discharge: bodily pain (28.7 vs 10.1), p=0.09; emotional limitation (29.3 vs 2.7), p=0.01. Effect on bodily pain sustained at 1 year (37.6 vs 9.9) No effect of geriatric outpatient care on any of the QOL parameters No effect of either inpatient or outpatient geriatric care on the functional status of patients with cancer |
| QOL, physical and psychological functioning | ||||
| Chock et al
40 (secondary analysis of Clark et al 66) RCT ++ | Adults aged 65+ with advanced cancer treated with radiotherapy, n=16 Cancer centre, USA (other participants reported: n=38 under 65 years) | I: QOL intervention with telephone follow-up: physical therapy, education, cognitive behavioural interventions, discussion and support, spiritual reflection and relaxation training T: 6 sessions 90 min, 2–4 weeks and 10 brief structured telephone sessions C: Standard care W: Multidisciplinary (including physical therapist, clinical psychologist, APN, chaplain) | QOL; mood | Significant difference at week 4 only in mean overall QOL older versus younger adults (74.4 vs 62.9, p=0.040) Significantly lower anger-hostility dimension of mood measure at all weeks for older versus younger patients. Week 4: 95.0 vs 86.4, p=0.028; week 27: 92.2 vs 84.2, p=0.027; week 52: 96.3 vs 85.9, p=0.005 No other significant differences |
| Heidrich et al
38 Two pilot RCTs and one observational study ++ | Women aged 65+, 1 year postdiagnosis of non-metastatic breast cancer, n=82 (total) Oncology clinics, cancer centre, USA | I: Pilot 1—symptom management (IRIS): counselling interview and telephone follow-up on symptom management at 4 weeks; pilot 2—addition of four biweekly telephone reinforcement sessions; pilot 3—intervention by phone only C: (1) usual care; (2) delayed IRIS (waitlist) control; (3) no control (IRIS group only) T: 4 weeks (pilot 1) W: APN | Feasibility, acceptability (symptom distress; symptom management; QOL; mood; barriers to symptom management; communication difficulty) | Feasibility: across all studies, 76% of eligible women participated, 95% completed the study, 88% reported the study was helpful and 91% were satisfied with the study Pilot 1: no significant difference in symptom distress. Significant decrease in distress baseline to follow-up in intervention group; significantly more women in intervention reported changing self-care of symptoms (p<0.05); no significance QOL differences Pilot 2: significant less symptom duration compared with control at 8 weeks (p<0.01). At 16 weeks, intervention group more likely to have talked to healthcare provider, begun new symptom treatment and changed self-care symptoms (p<0.05). No significant QOL differences. Negative attitudes from healthcare providers reported by 5%–20% of women and communication difficulties by 5%–45% of women Pilot 3: No significant differences (no control) from baseline to 8 weeks. Symptom interference decreased (and negative mood from symptoms). Symptom duration interference and negative mood from symptoms decreased. No QOL change |
| Kornblith et al
28 RCT ++ | Adults aged 65+ with breast, colon or prostate cancer, n=131 Cancer centres/university settings, USA | I: Telephone monitoring of distress providing support (plus educational materials) T: Over 6 months—monthly monitoring C: Educational materials alone, referred to oncology nurse upon evaluation if distressed significantly W: Trained graduates monitoring telephone calls. Referral onto an oncology nurse where indicated | Psychological distress | Lower anxiety and depression mean HADS total score for intervention 6.01 vs 8.20 control (p<0.0001); HADS depression subscale, intervention 3.20 vs 4.08 control (p=0.0004); HADS anxiety subscale intervention 2.81 vs 3.25 control (p<0.0001), at 6 months controlling for study entry levels No differences on other measures of psychological distress |
| Lapid et al
42 Secondary age group analysis of Rummans et al 67 RCT ++ | Adults aged 65+ newly diagnosed with advanced cancer, n=33 Cancer centre, USA | I: Multidisciplinary psychosocial QOL sessions T: Eight 90 min sessions, 4 weeks after enrolment C: Standard care (regular outpatient visits with oncologist and allied healthcare providers) W: Led by psychiatrist or psychologist and cofacilitated by a nurse, physical therapist, chaplain or social worker. Leaders trained in materials and observed sessions | QOL | Higher overall QOL intervention group scores throughout the study, not significant Higher QOL scores at week 4 intervention versus control (79.3 vs 62.9, p=0.0461) Improvement in QOL scores for intervention at weeks 4 and 8 compared with older control group |
| Mantovani et al
29 RCT ++ | Adults aged 65+ with cancer, n=72 Inpatient setting at medical oncology clinic, Italy | I1. Emotional and practical support from volunteers and I2. with structured psychotherapy T: I2: Weekly sessions of 1 hour for 6 months C: Pharmacological only W: Trained volunteers basic=40 hours/6 months, another 40 hours/6 months practical and further personal training | QOL | Non-significant between group differences in functional status/physical symptom improvements over time: Karnofsky's Performance Status Scale (F=9.90, 2 df, p<0.001). No differences on Spitzer's QOL Index or Functional Living Index—within/between groups Significant between group differences: State-Trait Anxiety Inventory control significantly worsened and intervention groups significantly improved (I1 improved more than I2) (F=4.50, 2 df, p<0.05) Beck Depression Inventory: control group unchanged, both intervention groups improved (F=229.66, 2 df, p<0.01) |
| Sajid et al
43 RCT (pilot) ++ | Men aged 70+ with prostate cancer and hormone therapy, n=19 Two medical oncology clinics, USA | I1. EXCAP (home-based walking and resistance intervention) I2. Technology-mediated walking and resistance intervention using Wii Fit T: One face-to-face session then 6–12 weeks home based C: Usual care W: Trained exercise physiologist | Functional and aerobic (skeletal muscle and muscular mass measure; handgrip strength; chest repetition test; DEXA scan) | EXCAP intervention arm higher rate of change in steps per day at each follow-up (+2720 steps) (p<0.01) compared with control (+97 steps) and Wii Fit arm (+382 non- significant) EXCAP arm had a 2.3 point change in physical battery score after 12 weeks, compared with 0.6 points in the Wii Fit arm and −0.5 points in the usual care arm No other significant differences in outcomes |
| Suh et al
39 RCT +++ | Adults aged 65+ completed active treatment for gastrointestinal cancers, n=63 Cancer centre, South Korea | I: 8 weeks of Qi exercise and 1 hour face-to-face counselling on physical and psychological factors T: 8 weeks C: Usual care W: Two Qi exercise trainers, APNs | Physical activity (BMI; body weight; nutritional status; symptom experience; self-efficacy; self-esteem) | Physical activity increased in both groups, extent of increase greater in intervention group (p=0.005) Difference in amount of exercise over time between groups (p=0.002) No between-group difference in BMI Nutritional status in both groups improved over time. The degree of reduction, however, was significantly larger in intervention group (p=0.048), and same in interaction between group and time Both group and interaction factors have significant positive difference in symptom experience, health promotion and self-esteem for intervention |
| Yagli and Ulger36 Controlled before and after study ++ | Women aged 65–70, 6 months after chemotherapy for breast cancer, n=20 Department of physiotherapy and rehabilitation, Turkey | I: 8 sessions of 1-hour yoga classes T: 8 weeks C: Exercise programme for 8 weeks W: Existing physiotherapist (yoga teacher) | QOL; depression levels; levels of pain, fatigue and sleep quality | All patients' QOL scores improved pre to postyoga and exercise interventions Total scores and some subcategories of the Nottingham Health Profile showed significant difference in favour of the yoga group (p<0.05) but not on energy level and pain where there were no differences Significant better fatigue and sleep quality in yoga group postintervention (p<0.05) |
| Communication between patients and healthcare professionals | ||||
| Devik et al
23 Qualitative ++ | Adults aged 65+ with advanced cancer, n=9 Patients’ homes in rural Norway | I: Qualitative study of home nursing care to patients with advanced cancer in rural locations C: NA W: District nurses in normal role | Patient experience | Importance of nurses having a person-centred manner Ability to show a genuine and empathic interest in the patients Technical skills or special competences less discussed than personal qualities, such as having a sense of humour or generosity Good listening and communication skills |
| van Weert et al
21 RCT (cluster) ++++ | Adults aged 65+ with cancer receiving chemotherapy, n=210 Staff: oncology-trained nurses, n=77 12 wards of 10 hospitals, Netherlands | I: Workforce: communication skills training in delivery of chemotherapy education to patients T: 3-month implementation C: Nurses continued to provide patient education as usual W: Nursing and specialised oncology nursing roles | Effects on quality of staff communication; effects on content of the consultation (patient recall of information) | Significant improvement in discussing realistic expectations. C: −0.20; I: 0.45 (total between-group difference 0.65) (p<0.01) Significant decrease in rehabilitation information pre to postchange. C: 0.08; I: −0.38 (total between-group difference −0.45) (p<0.01) No significant changes in categories treatment-related information and coping information Non-significant: intervention group showed significant decrease in number of items discussed Less history taking pre to post (C: 1.83; I: −2.33; between-group difference −4.17; p<0.001) and less talking about all different side effects pre to postchange (C: 1.98; I: −5.71; total 7.68; p<0.001) Patients in intervention asked more questions (M=10.76) than control (M=6.69; p<0.05) Marginal significance for intervention group: proportion recall of recommendations post versus pre (C: −3.34; I: 6.39; total: 9.73; p<0.10) |
| Yeom and Heidrich24 Observational ++ | 190 women at least 1 year postbreast cancer diagnosis, n=190 Community, an oncology clinic and a state tumour registry, USA | I: Symptom management (IRIS): counselling interview and telephone follow-up on symptom management T: 8-week intervention with 16-week follow-up point in the RCT C: Waitlist control subjects offered intervention after 16-week follow-up assessment W: APN | Negative beliefs about symptom management (QOL; purpose in life; positive relations with others) | Significant direct effects on SMBQ (p<0.00) and Communication Attitudes Questionnaire (p=.012) or Communication Difficulties Questionnaire Communication difficulties significant direct, negative effects on all four dimensions of QOL Significant total effects of SMBQ on MCS (mental QOL) (p=0.001) and PIL and PR (p<0.001) but not physical component (PCS). SMBQ predicted lower levels of QOL in three of four dimensions None of the four indirect effects of SMBQ on QOL through CommD was significant, indicating that CommD does not mediate the effects of SMBQ on QOL The total effects of CommA on four QOL measures were not significant. However, the indirect effects for MCS (p=0.05), PIL (p<0.05) and PR (p<0.05) through CommD were significant, indicating that CommD mediates the effects of CommA on MCS, PIL and PR |
Quality ratings: high ++++; moderate +++; low ++; very low + intervention/workforce description.
APN, advanced practice nurse; BMI, body mass index; C, control group; CGA, comprehensive geriatric assessment; df, degrees of freedom; DEXA, dual-energy X-ray absorptiometry; EXCAP home-based walking and resistance intervention; GI gastrointestinal; GOC, geriatric oncology consultation; GP, general practitioner; HADS Hospital Anxiety and Depression Scale; I, intervention; IRIS individualized representational intervention to improve symptom management; MCS, mental component summary; NA, not applicable; PCS, physical component summary; PIL, purpose in life; PIM, potentially inappropriate medications; PR, personal relations; QOL, quality of life; RCT, randomised controlled trial; SMBQ, Symptom Management Beliefs Questionnaire; START, screening tool to alert doctors to right treatment; STOPP, screening tool of older person's potentially inappropriate prescriptions; T, time point; W, workforce involved.