Effectiveness of Therapeutic Patient Education Interventions for Older Adults with Cancer: A Systematic Review

The incidence of cancer increases with age and demographics shows that the population of western countries is dramatically ageing. The new discipline of Geriatric Oncology is emerging aiming at providing tailored and patient-centred support to older adults with cancer. With the development of oral cancer therapy and outpatient treatments, Therapeutic Patient Education (TPE), aiming at enabling the patient and their relatives to cope with the disease in partnership with health professionals, appears to be an interesting and useful tool. The purpose of this paper is to search for evidence of the effectiveness of educational interventions for patients in older adults with cancer. The first screening found 2,617 articles, of which 150 were eligible for review. Among them, fourteen finally met the inclusion criteria: experimental and quasi-experimental studies enrolling older adults (over 65 years old), suffering from cancer and receiving an educational intervention. The types of educational intervention were diverse in these studies (support by phone and web base material). The results appear to be positive on anxiety, depression and psychological distress, patient knowledge and pain. However, data currently available on the effectiveness of a TPE program in Geriatric Oncology is lacking. Further studies are needed to assess the effectiveness of TPE programs adapted to the specific circumstances of the older adult.

TPE appears to us as a key element in the management of older adults patients with cancer; indeed, if studies have shown that TPE has a positive impact on adherence (12), quality of life and pain (13) in adults suffering from chronic illness, it has also been shown that older adults could benefit from it (14). Moreover, in the geriatric literature and clinical routine geriatrics, we know that an educational component is needed in any intervention designed to limit or avoid geriatric syndromes such as falls (15,16), frailty (17), malnutrition (18) and loss of autonomy in general (19). Considering these two facts, it can be envisaged that preventing those geriatric syndromes2 could also be targeted outcomes for educational interventions in older adults with cancer. In the specific and heterogeneous population of older adults with cancer, the therapeutic educational sessions's content must be tailored to, on the one hand, the disease (type of cancer), and on the other hand, the physiological or pathological age-related changes (mainly sensory and cognitive impairments) (20)(21)(22).

Review objective
We chose to realise a systematic review designed to search for evidence of the effectiveness of therapeutic patient education interventions in older adults with cancer on physical and mental health.

Types of participants
This review considered studies that enrolled older patients, with an average age greater than 65 years old, of any gender and ethnicity, diagnosed with any form of cancer and receiving any treatments.

Types of interventions
This review considered studies in which the interventions included a therapeutic patient education aspect. Therapeutic patient education is rarely studied in itself in older adults with cancer, so we decided to consider any intervention with an educational component.

Comparator
We included studies in which the control group received information through usual care or usual education but not through a standardised multidisciplinary TPE method.

Types of outcomes
This review considered studies that included any outcome. We first envisage to study observance and quality of life, at the first step in our preliminary research, but finally chose to consider works studying any outcome.

Types of studies
This review considered experimental studies: randomised controlled trials. Other research designs such as quasiexperimental, before and after studies, prospective and retrospective studies, cohort studies, pilot studies and feasibility studies were also included.

Exclusion criteria
This review excluded studies concerning subjects under 65 years old and non-educative interventions. We excluded qualitative studies and those published before 1990.

Search Strategy
We analysed articles in English and French published between 1990 and July 2016. Several international databases were searched with identified keywords (Appendix 1): Medline, Cochrane Library, Web of Science and PsycINFO. A research of the grey literature was also conducted in Therapeutic Education and Geriatric Oncology journals.
One of the investigators is a Geriatrician. This research was conducted with the help of the primary care and family medicine department of the Toulouse University Hospital.
The second investigator is a Geriatrician too, who belongs to the Epidemiology and Public Health Department of the Faculty of Medicine of Toulouse. The research was conducted using identified keywords and index terms across all the included databases.
The first screening found 2,617 articles. After reading the title and the abstract, 150 were eligible to be reviewed. Among them, which we read through, fourteen finally met the inclusion criteria. The selection process is presented in the flow-chart ( Figure 1).

Figure 1 Flow Chart
The articles included in this review were assessed independently by the two investigators, who reviewed the Table 1 Main results of studies in geriatric population  -Performed : by a nurse -Duration : four weeks after the instruction day -Design : 2 groups randomised : * intervention group * control group : usual care Subjects were stratified as either elderly ( 60-75 years) or oldest (>75 years).
-Primary outcomes : Quality of life (quality of life tool),knowledge and attitudes of the caregiver and patient in managing pain (patient pain questionnaire), compliance with the drug and nondrug interventions and perceived effectiveness (self-care log (42)), mood (profile of mood states (43)).
-Significant improvement in sleep (p=0,03) and rise of knowledge levels (use of medications on schedule rather than an as-needed basis) (p=0,007).
-Family caregivers benefit from this education program. -Performed : by nurses -Duration : six weeks after the training day -Design : 2 groups randomised by having lots drawn from a sealed container by an independent person : * Intervention group * Control group: usual education A randomised pre-and post-test design was used. The video-observations were conducted by blinded observers.
-Primary outcome : Recall of information measured using the "Netherlands Patient Information recall Questionnaire" (NPIRQ).
-Secondary outcomes : Quality of communication was measured by the QUOTE chemo-Performance, frailty was measured using the Groningen Frailty Indicator (45), amount of questions. -Significant decrease in the number of items discussed (total change score -9,87, p < 0,001).
-Limited intervention effect on recall of information with a significant pre-/post-change in proportion recall of two categories of recommendations : "hygiene" (total change score 24,79, p<0,05), and "side effects that have to be reported to the hospital" (total change score 15,86, p<0,05).  -Type of intervention : Verbal instruction. *Face to face exercise group : 10-week exercise program (2 group sessions per week supervised by an accredited exercise physiologist) followed by tailored at home exercises for 6 months, home based exercise program for 6 months and support program. * At home exercise group : home-based exercise program for 6 months (coach calls, exercise manual, DVD). * Man plan support program : education on low-intensity exercise, diet and psychosexual functions.
-Performed : by an accredited exercise physiologist -Duration : 6 months and access to phone line during 2 years -Design : 3 groups according to patient preference, medical comorbidities and fitness level. and hip circumference (p=0,015), blood pressure (p=0,0044 for systolic blood pressure ; p<0,0001 for diastolic blood pressure), mean time for completing the 400-m walk (p<0,0001) and improvements in ability to undertake resistance training exercises (p<0,0001).
-No significant modification of weight, BMI (Body Mass Index) and heart rate.
-High satisfaction, recruitment, and compliance.  *initial training about: how to take the oral medication, toxicity profile and the recognition of side effects, instructions for their management and the actions to be taken in case of discontinuation of therapy… *diary including a calendar to check off pill consumption for each day and a specific form to collect a self-report of toxicity. *Patients were monitored during the first and second cycles of oral therapy, by phone calls on days 7 and 14.
-Performed : by a nurse -Duration : monitoring for two weeks -Design : after a medical visit the patient received training by the nurse who showed the patient how to correctly take the oral medication. A questionnaire was administered before and after training by the nurse.
-Primary outcome : Quality of care evaluated by a questionnaire with specific items concerning the level of comprehension.
-Secondary outcomes : The nurse collected the diary and asked the patient to describe and specify every symptom, and graded the toxicity according to NCI-CT-CAE 3.0 (55). -The intervention resulted in an increased proportion of patients having received correct information related to treatment, with a level of confidence rising to more than 90 % for all items considered.
-The diary proved a valid tool for patients.
-This model proved practicable and accepted by patients.
abstracts, read and selected the full texts independently.

Methodology quality
We assessed the methodological quality of each study included with validated scales: -for randomised controlled trials: the CONSORT checklist (23) to analyse the quality of the report and Jadad score (24) to analyse the methodology of the study, -for non-randomised studies: the STROBE checklist (25) to analyse the quality of the report and Newcastle-Ottawa criteria (26) to analyse the methodology of the study.

Data extraction
Data from the studies were extracted by two independent investigators. The data extracted included the title, authors, country and year of publication, and details about the study scheme, population (age and type of cancer), interventions, study methods, primary and secondary endpoints and the main outcomes.

Data synthesis
Due to the clinical and methodological heterogeneity between the included studies, a meta-analysis was not possible.

Results
A total of fourteen articles were analysed in this literature review (Table 1). Among these fourteen articles, six were randomised controlled trials, three quasi-experimental studies, one prospective study, one cohort study, two pilot studies, and one feasibility study. They were conducted in several countries, with the majority in the United States (seven), Australia (two), Sweden (one), the Netherlands (two), Singapore (one) and Italy (one). They were published between 1993 and 2014.
There is great heterogeneity in the populations studied. Regarding our targeted population, it appears that older adults (over 65) were identified and specifically studied in only seven studies (33,44,47,49,50,21,52). In other studies, older participants were pooled with a general adult population. In one study, there was a comparison between a geriatric and a nongeriatric group (29).
The types of cancer studied are also diverse. The most prevalent were colorectal, prostate, breast and lung cancers, although there was also bladder, pancreatic, stomach, liver, kidney and small intestine cancers, lymphomas and myelodysplasias. Cancer stages varied according to the studies, as well as the treatments (chemotherapy and radiotherapy) received.
The interventions types were numerous. The vast majority of interventions were multi-dimensional and included an educational aspect, but were not exclusively educational or pedagogic.
The interventions' follow up were also various. Patients were followed-up by phone (27,32,33,50) or at home, through psychological support or via distribution of educational materials (33,41,44,46,47,21,52) in various forms (brochures, booklets, audio or video links to the Internet …). Educational interventions were mainly carried out through tools such as the telephone, video or the Internet. These interventions were not adapted to the specific learning capabilities of older Table 2 Methodological evaluation of randomised controlled trials adults except in one study (52).
The more frequently found significant positive results were observed on pain, anxiety and quality of life (three for anxiety, two for pain and two for quality of life). One study showed an improvement of the patient's depression (33) but another found no difference (27). Concerning the patient's level of information (and recall of information) a majority of the studies were positive, only one found difficulty in remembering information for patients (49).
Thus, only one study (52) offered a suitably adapted program of therapeutic education (on the presentation of the educational material and the content of the information) to a geriatric population with an average age of 80 years of age. This study shows an increase in knowledge about cancer after this intervention. However, it deals with patients who do not have cancer but whose aim is preventive health care of older adults with respect to cancer. In addition, this was a pilot study with only 21 enrolled patients and an average methodological quality.
The methodological quality of the included studies have been assessed by the validated scales : the Jadad scale (24) or the Newcastle-Ottawa quality assessment scale (26). The STROBE statement (25) or CONSORT checklist (23) have been used to estimate the quality of the study report. Tables 2 and 3 show summaries for each study, their detailed scores on the scales. In the randomised trials, blinding was impossible because of the type of interventions. Two of the randomised trials (27,44) are of good methodological quality (Jadad score of 3).
Given the heterogeneity of the studies, we were not able to perform a meta-analysis.

Discussion
Only fourteen studies were included in this literature review studying educational interventions in older patients with cancer. The results of these studies are quite positive overall. Those interventions seem to provide positive effects on health outcomes but not only (knowledge and quality of life). However, none of these articles studied the effectiveness of a TPE specifically tailored for older patients with cancer (over 65 years of age). There is currently very little data in the literature on the effectiveness of Therapeutic Education in this population.
If data on patient education in geriatric oncology is poor, we realise that data was also lacking in adults under 65 years of age, through this literature review (56). Indeed, a systematic literature review performed in 2015 by a US team (56) found only two articles on the effectiveness of therapeutic education for adult patients regarding their compliance/observance with oral cancer treatment (average age 56 years (57) and 59.85 years old (58)) with cancer in an outpatient environment, between 1953 and 2014. These two studies had small-sized samples and their methodology was from weak to moderate. Therefore, the conclusion of this literature review is that further studies are needed to demonstrate that patient education can improve compliance with cancer treatments and their health outcomes. This is in accordance with our findings; data is limited in adult populations, but even more in older adults, virtually non-existent.
Our study raises this question: why Therapeutic Patient Education studies are so few in older populations despite the fact that TPE is recommended in chronic conditions and is expanding in Geriatrics and in Oncology? In a general manner, older patients with cancer are underrepresented in clinical trials. A 2012 article (59) showed that inclusion diminishes with age in these types of studies. The authors surmised that a decline in the functional reserve, increased comorbid conditions, concomitant medication use, lack of social/home support and decreased access among other factors contributed to poor enrolment among older adults. To remedy this situation, studies addressing older subjects need to take into account those specificities and need to be designed to gauge the weight of these specificities in this population. For example, there is a need to adapt interventions to the specific learning capabilities of older adults as mentioned by Barnes et al. (52). There is also a need to take into account the health care professionals; skills and needs in older adults with cancer management. On the one side, health care professionals feel their formation is lacking to address to accompany older subjects with cancer, which is a time-consuming activity for which one has to be committed. On the other side, they feel as though they are already giving out enough information, but it is not equivalent to using specific tools and pedagogical methods of education (60).
Thus, TPE, included in an integrated care management strategy, could provide benefits in terms of mental and physical health for the patient, their relatives but also in terms of health care system utilisation (avoid inappropriate admission, iatrogenia). It could increase the patient's observance to the treatment, increasing the latter's health outcomes, as it has been proven in other chronic diseases such as diabetes (6) or asthma (5). It could also decrease, and that is a major topic in Geriatric Oncology, the toxicity of chemotherapy by decreasing overuse and even reduce the misuse of care resources, in particular, hospitalisation.
Finally, the increase in the patient's knowledge would allow them to be more involved in their care and enhance their role in the decision-making process.

Strengths
The main strength of this study is the innovative character of the approach since it is the first literature review on this topic in this population. In fact, this is a current topic, with the gradual increase in cancer prevalence in the patient population over 65 years of age.

Limitations
A major limitation of this study is the heterogeneity of the studies included, particularly because of their different study designs; their interventions are not fully comparable and their methodological quality is variable. The differences between the populations, health systems, the type of intervention, outcomes and methodological quality compound this heterogeneity. The fourteen studies included mostly had a small patient sample, which limits their ability to show a significant difference and makes it difficult to extract generalisations.
These heterogeneous results are due to the lack of scientific data on the efficacy of therapeutic patient education in older adults. We had to open the field of research to include any type of educational intervention in this population to show that it might indeed be a feasible and useful proposition. We feel that TPE including caregivers could improve parameters such as quality of life, compliance and pain management, amongst others. Our key finding is that data is missing regarding this subject in the scientific literature.
Geriatric oncology is developing, as well as the use of TPE as part of the care plan for these patients. Tailored TPE programs for older patients with cancer are implemented. Studies on this topic are becoming more numerous since six out of the fourteen selected articles were published after 2010.
There are several perspectives on TPE in geriatric oncology. We can surmise that, in the future, programs will be partly carried out by information and communication technologies. Indeed, we could imagine that, with the development of telemedicine, part of the educational approach could be carried out remotely in the form of online courses or discussion forum online with health care providers, for example.

Conclusion
There is a lack of data on TPE in the field of geriatric oncology. The effectiveness of a therapeutic education program for older adult cancer patients must be studied because of the efficacy of TPE in chronic conditions, the prevalence of cancer in older adults and the global ageing of the population. TPE could increase treatment compliance/observance, decrease side effects, improve health outcomes and have a positive effect on the quality of life of these patients and their relatives. Further studies, and especially studies of high methodological quality and level of evidence, are needed to assess the effectiveness of TPE in older adults with cancer.