Effectiveness of Education Interventions for the Management of Cancer Pain : A Systematic Review

BACKGROUND
Many cancer patients experience poor pain control due to various factors, including misconceptions regarding the use of opioid analgesics. For management of cancer pain, interventions involving education of both patients and physicians have been attempted.


OBJECTIVES
This review aimed to assess the current evidence of the benefits of education for the management of cancer pain.


METHODS
We searched the Medline, EMBASE, Cochrane library, and major Korean databases to identify relevant studies. We included most study designs, but excluded case series. The primary outcomes were pain intensity and quality of life (QoL). Two reviewers assessed the risk of bias using the Cochrane's tool for RCT and Risk of Bias Assessment tool for Non-randomized Studies (RoBANS) for non-randomized studies, independently.


RESULTS
After extensive searches, 3,324 publications were screened, and 32 studies were selected. The education interventions used in the included studies included a wide variety of education methods, but the most common method was a booklet produced for patients. Regardless of the education method used, the results of the meta-analysis were as follows. The SMDs of the most severe, average, and current pain in the RCTs were significant. The SMD of worst, average, and current pain were -0.34 (-0.55, -0.13), -0.40 (-0.64, -0.15), and -0.79 (-1.35, -0.23). In the non-randomized studies, the effects on average pain were significant, but those on worst and current pain were not.


CONCLUSIONS
Education intervention reduced the pain of cancer patients. Therefore, patient education could be considered to be an effective method of cancer pain management. However, our data should be interpreted with caution, and studies using standardized protocols are needed to confirm these observations.


Introduction
Pain is one of the most common symptoms of cancer patients.To control cancer-related pain, it is necessary to add analgesics to the treatment regimen depending on the intensity of the pain, referring to the World Health Organization (WHO)-recommended three-step analgesic ladder (WHO, 1996).However, pain management following the World Health Organization ladder results in only 75% adequate pain control (Zech et al., 1995).Recently, a total of 45% of patients reported inadequate pain control in a survey of Korean patients carried out between 2001 and 2006 (Hong et al., 2011).Narcotic analgesics commonly have to be prescribed for the proper management of severe cancer pain.However, many patients with cancer have poor pain control for several barriers, including misconceptions concerning opioids (Brant 2010).Therefore, educating both patients and practitioners assists provision of accurate information.Allard et al. (2001) published a systematic review of the effects of educational intervention on the control 1 National Evidence-based Healthcare Collaborating Agency, 3 Department of Internal Medicine, Seoul National University Hospital, Seoul National University, 4 Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, 2 Department of Oriental Gynecology, CHA Bundang Medical Center, CHA University, Seongnam, Korea *For correspondence: mk3three@gmail.com of cancer pain in 2001.In addition, Bennett et al. (2009) assessed the effects of patient-based educational intervention on cancer pain in 2009.However, these studies are not recent, and so an analysis of more recent trials is needed.Therefore, this study evaluated the current evidence of the effectiveness of education intervention in the management of cancer pain by conducting an extensive systematic review and meta-analysis.Unlike previous meta-analyses, we included observational studies and nonrandomized trials that assessed intervention and patient outcome.

Literature searches
We developed a protocol for the systematic review of the effectiveness of educational intervention for the management of cancer pain.Specifically, the population, method of intervention, method of data comparison, patient outcome, study design and time (PICOST), search methods, data extraction, quality assessment, and meta-analysis from published studies were all reviewed.The following parameters and limitations were used: 1) Participants: cancer patients with pain; 2) Interventions: educating cancer patients about the use of analgesics to manage pain; 3) Controls: no limitations; 4) Outcomes: primary outcomes were pain intensity and quality of life (QoL).Secondary outcomes were all other patient outcomes (excluding information from caregivers); 5) Study design: all studies included a control group (for example RCT, non-randomized controlled trials, or cohort studies).

Study selection
The inclusion criteria were trials that investigated the effects of educational intervention on the use of analgesics in cancer patients.All included studies were published in the English or Korean language.The exclusion criteria were duplicate publications, studies that were not peer reviewed, and those that used only the assessment of the caregiver as the outcome.All studies were reviewed and selected independently by two researchers (each study was assigned to two of MK Hyun, YJ Jung, JI Shin, and MJ Kang).The titles and abstracts were reviewed, and duplicate studies or those that did not meet PICO requirements were excluded.If the title or abstract appeared to meet the PICO of this review or we could not determine its eligibility, the full text of the article was obtained and eligibility was confirmed.Discrepancies between the reviewers were resolved after discussion.

Data extraction
Two independent reviewers extracted data using a standardized data extraction form (each article was assigned to two of MK Hyun, YJ Lee, YJ Jung, and MJ Kang).Any discrepancies were resolved by consensus or consultation with a third reviewer.The characteristics of the studies (study design, and country of origin) and general information (inclusion/exclusion criteria of participants, intervention tools, methods of education, the individual responsible for educating, outcomes, and patient age) were extracted.The study design was classified using DAMI (study design algorithm for medical literature of intervention) developed by Kim et al. (2011).

Quality assessment
Four authors evaluated the quality of the studies independently.The risk of bias was assessed using RoBANS [Risk of Bias Assessment Tool for Nonrandomized Studies, Kim et al. (2013)], and the tool from the Cochrane group for RCTs.In the case of disagreement between reviewers, consensus was reached after discussions between the parties.If a consensus could not be reached between the two investigators, a third party joined the discussion and agreement was reached by majority rule.

Meta-analysis
The outcome of the effectiveness of education was continuous, and was analyzed statistically by calculating the standardized mean difference (SMD).Randomized controlled and non-randomized trials were analyzed separately.Inverse variance methods and the randomeffects model were also used.To control for factors such as small sample size and different interventions, outcomes were synthesized as random-effects models.Results are presented as SMDs and CIs.The Review Manager (ver.5.1; Copenhagen: The Nordic Cochrane Center, the Cochrane Collaboration, 2011) software was used for data analysis.

Included studies
The authors reviewed the titles and abstracts of 3324 identified studies (including 1437 international and 1887 Korean studies) independently, and 2113 were excluded.The full-text publications of the remaining 79 potentially eligible studies were reviewed in detail.Of these, 47 were excluded based on the inclusion and exclusion criteria (Figure 1), and 32 were included.Twenty-five studies were RCTs, and only seventeen of these were included in the meta-analyses due to varying outcomes.Seven studies were non-randomized trials: four of these were non-randomized controlled trials, and three studies were  doi.org/10.7314/APJCP.2014.15.12.4787Educational Interventions to Improve Management of Cancer Pain "before and after" studies.The most frequently used educational tool was a booklet.Education commonly comprised multiple interventions, combined with DVDs and audiotapes.Although the educational content was diverse, the most common content was a description of pain management using analgesics, and misconceptions regarding opioids.Additional content included discussions of self-control, relaxation, and individual therapy.The control groups included in the RCTs ranged from no treatment to usual program.

Risk of bias of the included studies
Most studies were assessed as having an unclear risk of bias in the domains of random sequence generation, allocation concealment, and blinding.The methods used for random sequencing were either not reported, or were inadequate.In particular, most studies did not consider allocation concealment for appropriate performance.Therefore, the results of these studies must be interpreted with caution (Figure 2).

Meta-analysis
Effects on cancer pain (Figure 3), Effects on the most severe pain (Fig. 3A).Most studies used the Brief Pain Inventory (BPI) pain scale.Other tools used to assess pain were 0−10 rating scales, and these results were combined into a meta-analysis of BPI.Ten studies included a metaanalysis of severe pain, including seven RCTs.The SMD of the RCTs was −0.34 (95% CI, −0.55, −0.13, p=0.001).However, in the non-randomized studies, SMD was not significant due to high levels of heterogeneity (SMD, −0.84; 95% CI, −2.05, 0.37, p=0.17).
Effects on average pain (Figure 3B).The data from 10 studies were available for a meta-analysis of average pain.The effects of educational intervention on average pain were presented in both RCTs and non-randomized studies.The SMD of average pain was −0.40 [95% CI −0.64, −0.15, p=0.002] and −0.73 [95% CI −1.40, −0.05, p=0.03] in the RCTs and non-randomized studies, respectively.However, both RCTs and non-randomized studies exhibited high levels of heterogeneity.
Effects on current pain (Figure 3C).Data from only six studies were available for use in a meta-analysis of current pain.The SMD effect of education on current pain in RCTs was −0.66 [95% CI −1.09, −0.23, p=0.002].In contrast, there was no SMD in the non-randomized studies due to high levels of heterogeneity.

Effects on quality of life
Effects on overall quality of life: only two studies were included in the meta-analysis of QoL, which were by the same author and reported using EORTC QLQ_C30 (European Organization for Research and Treatment of Cancer Quality of Life Questionnaire).The SMD of quality of life was 0.07 (95% CI −0.15, −0.29, p=0.82) (data not shown).Educational intervention did not result in improved QoL of cancer patients, although these results are limited by the inclusion of data from only two small studies.
Effects on pain: the SMD of pain based on the EORTC questionnaire was −0.07 (95% CI −0.55, −0.41, p=0.78).Patient education did not result in significant differences Effects on the functional assessment of cancer therapy -general (FACT-G): only two studies were included in this meta-analysis, and there were no significant effects on QoL based on FACT-G (data not shown).

Discussion
To control cancer pain, it is necessary prescribe different analgesics depending on the intensity of the pain, referring to the World Health Organization (WHO)recommended three-step analgesic ladder (WHO, 1996).To treat suddenly occurring breakthrough pain, short-acting analgesics should be prescribed in advance, so that they can be used when necessary.Belief about analgesics related with patients' opioid adherence, myths and misconception induced poor pain control.Therefore, patients should be educated on the different methods of pain control and the use of analgesics, as well as instructed on pain assessment and methods of expression to ensure effective control of cancer pain.If the pain of patients can be controlled appropriately following guidelines, unnecessary hospitalization and ER visits will be reduced, resulting in improved quality of life for the patients and a more efficient use of medical expenses.However, a number of obstacles exist for the adequate control of cancer pain.To close gap, several interventions including leadership of cancer care providers and education for patients were emphasized (Nevidjon, 2010).Therefore, we performed a comprehensive systematic literature review and meta-analysis to assess the established rationale of cancer patient education on the use of narcotic analgesics.Our results revealed that the SMDs of the most severe, average, and current pain in the RCTs were significant.In the non-randomized studies, the effects on average pain were significant, but those on worst and current pain were not.
Patient-based education reduced the pain of cancer patients, and so could be considered an effective method of cancer pain management.However, these results should be interpreted with caution due to the wide confidence intervals and the heterogeneity of qualitative analysis due to use of different study designs and diverse educational protocols including content, educational tools, measurement tools, and the education time.
In a 2001 review, educational intervention had little effect on the pain levels of patients (Allard et al., 2001).In contrast, a 2009 review suggested that educational intervention reduced the intensity of the average and worst pain compared with normal care or control (Bennett et al., 2009).In this study, patient-based education was also effective reducing the intensity of cancer pain.Based on these data, patient-based education for the management of cancer pain is highly recommended.Patient quality of life was using various tools in the included studies, and so a meta-analysis of QoL included only four trials, with no significant effects.Previous systematic reviews suggested that pain medication improved the QOL of cancer patients, and that tramadol significantly improved QoL compared with placebo [assessed using the Activities of Daily Living (ADL) scale (Quigley, 2008)].In transdermal fentanyl and sustained-release oral morphine improved the QoL of cancer patients (appetite, sleep, daily activities, mental state, emotion, communication, and interest) in six separate trials (Yang et al., 2010).However, we were unable to find any systematic reviews on the effects of patient-based education for QoL.Therefore, further studies are needed to determine the effects of patient education on QoL and provide practical recommendations.In addition, unified measurement tools are required for accurate analysis.
Our study had some limitations.There was significant heterogeneity in the qualitative analysis, because the methods and tools used to educate patients were highly variable, and the individuals in responsible for the education varied between nurses, researchers, physicians, and students.Therefore, meta-analysis of only specific qualified outcomes was possible in some reports.Nevertheless, our data are important, despite these limitations, because our study included non-randomized trials.
In conclusion, educational intervention reduced the pain of cancer patients.Therefore, patient-based education could be an effective management method, and should be strongly recommended.However, to assure positive effects on QoL and prevent patient misunderstanding, studies using standardized protocols should be performed.

Figure 1 .
Figure 1.Flow Diagram of Article Selection

Figure 2 .Figure 3 .
Figure 2. Risk of Bias of the Included Studies.A) Risk of Bias Graph of Included RCTs (25 trials); B) Risk of Bias Graph of Included Non-randomized Studies (7 studies)