Effectiveness of nursing intervention for increasing hope in patients with cancer: a meta-analysis

ABSTRACT Objective: to evaluate the efficacy of nursing interventions to increase the level of hope in cancer patients, in a meta-analysis. Methods: electronic databases were searched. Two of the authors independently extracted data from the eligible studies, and Stata 13.0 software was used to pool the data. Results: nine randomized controlled trials were included, and methodological quality of each randomized controlled trial (RCT) was evaluated using Cochrane handbook recommendations. A random effects model was used to combine results from eligible studies. The pooled results using the fixed effects model showed that scores to first effects increase significantly after the use of nursing intervention between the groups. Heterogeneity was observed among the studies for posttest (df = 8, P = 0.000; I2 =76.1 %). The results indicated significant heterogeneity across the nine selected studies. The test for heterogeneity showed no homogeneity among studies for follow-up (df = 8, P = 0.328; I2 = 12.9 %), and there was no statistical significance. Conclusion: the current evidence suggests that nursing intervention has a positive effect on hope in cancer patients. However, more large-scale and high-quality randomized controlled trials are needed to confirm these results.


Introduction
Hope has been defined as the possibility of a better future in the context of uncertainty (1) , which significantly increases a patient's quality of life (2) . It has been identified as a valuable psychological resource that enables the individual to take an interest in his/her life and future, and to find meaning in life (3) . The author (4) stated that the most important feature of hope is that it gives confidence to the individual to make life changes.
It is well known that the cancer diagnosis, its treatment, and the challenges of survivorship increase patients' levels of psychological symptoms to a degree that might affect their adaptation to their disease (5) .
Nursing intervention has been shown to improve hope through promoting greater psychological wellbeing and decreasing psychological problems, such as depression and anxiety (5-6) .
Cancer diagnosis and treatment can affect physical functioning, mental health, and quality of life of individuals with cancer (7) . A great deal of studies (8)(9) have shown that the long-term and late effects following a cancer diagnosis have an impact on patients, including functional deficits, mood disturbances and heart failure in relation to chemotherapy toxicity. Many of these factors influence patients' hope, which has been considered an important coping strategy among cancer patients. Many researchers (10)(11) found that a high level of hope was associated with lower levels of anxiety and depression, higher social support, and better quality of life.
Several studies have shown that the influence of healthcare professionals has great potential to effect hope among cancer patients. One study (12) evaluated a psychologically supportive intervention, based on the theory called "Transforming hope", in which patients were guided to view a film on hope and work on a hope activity. Higher hope and quality of life among cancer patients were found in patients after the intervention.
Another study (13) found a novel treatment intervention combining three central attributes of mindfulness, hope therapy, and bio-behavioral components which were provided to women with cancer recurrence. That intervention increased hope and mindfulness two, four and seven months after the intervention. However, the effectiveness of nursing interventions for enhancing hope among cancer patients remains controversial.
The author (14) found that exercise leads to a great improvement in strength among lung cancer patients, but not hope. One researcher (15)  Therefore, it is necessary to summarize the results from randomized clinical trials to assess the efficacy of nursing intervention to improve hope in cancer patents.
To examine this hypothesis, we conducted the metaanalysis, and assumed that nursing intervention has a beneficial effect on hope in patients with cancer.

Methods
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, issued in 2009, was utilized to report this meta-analysis (16) .
Relevant studies were identified through systematic searches of the electronic databases, from their inception until January of 2016.We searched the Cochrane Library databases, PubMed, Ovid, Web of Science, China National Knowledge Infrastructure (CNKI), and Wanfang Data for articles published. Any randomized controlled study that evaluated the association between nursing intervention and the level of hope in adult patients with cancer was eligible for inclusion in our study, and no restrictions were placed on language or publication status. Both Medical Subject Headings (MeSH) terms, and the keywords of "cancer OR neoplasm", "hope", "nurse-led OR nurse" AND "randomized controlled trial OR controlled clinical trial" were used as search terms.
Additionally, we scanned the reference lists of retrieved papers for any additional relevant studies. We also contacted the corresponding author or first author to obtain information if publications were unclear or more information was needed.
Studies were eligible for inclusion in the present meta-analysis if they met the following criteria:  Studies that assessed the hope outcome using validated scales (e.g., Herth Hope Index -HHI). The Herth Hope Index (HHI) contains 12 items that measure three dimensions of hope (17) . The HII delineated three factors of hope: a) temporality and future, b) positive readiness and expectancy, and c) interconnectedness (18) .
Each item is rated on a 4-point Likert scale that ranges from "strongly disagree (1)" to "strongly agree (4)". < 0.1 with an I 2 value > 50% indicated no or slight heterogeneity across studies, and then a fixed-effect model was applied; otherwise, a random-effect model was adopted to pool the data (23) . If the results were presented as median and range values, the means and standard deviation were calculated using the formulas (24) . Subgroup analyses were conducted by dividing the studies into groups according to (a)sex, (b) type of cancer, (c) whether hope was the primary outcome, (d) quality of included study, (e) intervention format, and (f) intervention providers. Potential publication bias was evaluated using Begg (25) funnel plots and Egger (26) tests. Two-tailed p-value < 0.05 was considered statistically significant. In view of the significant heterogeneity among the studies included in

Characteristics of Included Studies
Some details of the included studies are presented in Figure 1. Study sample sizes ranged from 20 to 116.
Of a total population of 600 randomized patients, 306 were in the intervention group, and 294 in the control group. The randomized controlled trails were published between 1998 and 2015. Of them, four studies were conducted in Asia (one in Japanese (27) and three in China (30,(33)(34) ), two in Europe (2,32) , one in the USA (28) , one in Canada (29) and one in Australia (31) . All studies included one control group, and the control group was treated with usual care. However, there was an article that was divided into three groups, with the inclusion of an additional intervention named an attention control group. The most common treatment format was an individual approach (n=7), and only two studies applied a group approach. The most frequently used hope measurement was the HHI. In nine studies, there were various interventions considered. Most interventions were provided in hospitals or in patients´ homes.
Among the nine studies, interventions were delivered by health personnel (e. g., a nurse) in six studies, and other professionals were the interventionists in three studies.
The mean length of intervention was 3.2 weeks. The mean total intervention time was 86.5 minutes, with Rev. Latino-Am. Enfermagem 2018;26:e2937.
total intervention time in each study ranging from 30 to 120 minutes. The quality assessment of included studies, using the risk of bias tool, is shown in Figure   2. Overall, one randomized controlled trial had a score of 13 (27) , one trial had a score of 11 (32) , one trial had a score of 9 (28) , three trials had a score of 8 (29,(33)(34) , two trials had a score of 7 (2, 30) , and the remaining one trial had a score of 6 (31) . The mean score was 8.5, suggesting a moderate quality of the reports included in this meta-analysis. Among all the selected studies, participants and personnel were mostly not double blinded. Outcome assessment was not blinded in any of the studies. Overall, all the included studies were considered to have a high risk of bias.      and purpose within a life-threating illness, dictate their ability to cope with the disease in a meaningful way, and provide for the needs of cancer patients (35) .

According to clinical characteristics
According to the result of subgroup analyses by sex, males and females showed a significant effect on hope. Similar to one study, the author did a comparison to explore the relationship between urban or rural background and health attitudes of newly diagnosed oncology patients, which demonstrated that males scored significantly higher for belief (36) . There is a need to carry out more well-designed studies to verify our conclusion.
In subgroup analyses by type of cancer, a significantly higher level of hope was noted in individuals with terminal cancer than in other cancers, when using nursing interventions. This effect was not found for two trials with breast cancer patients and four trials with other cancers. The result is consistent with another study in this field (37) . However, more RCTs on various types of cancer will be needed to confirm our conclusion.

According to intervention characteristics
The finding from this meta-analysis based on 600 study participants indicated that nursing interventions have a positive influence on hope, and the positive effects were consistent either posttest or through follow-up, or both. The lengths of interventions for most studies included in this meta-analysis were less than eight weeks. This result is meaningful, and it is in accordance with that of previous meta-analysis studies. The researchers (38) aimed to identify whether interventions can reduce emotional distress in patients and their caregivers. Based on 29 randomized clinical trials, the author concluded that the average dose of the interventions was 6.7 sessions. The findings from our study support the hypothesis that nursing intervention can significantly increase hope in cancer patients.
Participants who were exposed to intervention designed to increase the feeling of hope had higher hope scores than those who were not exposed to intervention apart from regular care and hospital follow-up.
In subgroup analyses, according to intervention format, the results show that individual therapy is better than group therapy in cancer patients. Even if group approach interventions were effective in some aspects, the current results are in accordance with those of previous meta-analysis in concluding that psychosocial interventions using individual treatments (n=4) were more effective in increasing survival time than group intervention (n=11) (39) . There are only two articles using group therapy, which are too few. Therefore, further study for intervention format will be essential in the future.

Implication for research
Some of the evidence on the effectiveness of nursing intervention on hope domains reported in this article find support in the literature (40) (43) . These findings suggest that some populations could be prioritized in public mental health interventions to prevent the occurrence of hopelessness, and interventions need to be provided to enhance hope. Meta-analysis is considered hypothesis-generating, and is not conducted to test a hypothesis or establish a standard of care (45) . Additionally, meta-analysis is a secondary study that is based on primary studies, and some bias is inevitable (46) . Fourth, the quality of metaanalysis is dependent on the quality and comparability of information from the primary studies. If individual information were available, a more precise analysis, such as individual patient data meta-analysis, should be conducted rather than conventional meta-analysis. This is a big project, and it needs authors of all published papers to share their data. Fifth, given that hopelessness is highly prevalent among cancer patients, greater emphasis should be placed on establishing nursing programs that increase access to mental health care, as well as for patients at different stages of their disease and treatment trajectory.

Conclusions
Evidence from this study indicates that nursing interventions are certainly useful strategies in increasing hope with cancer. Health care providers