Testing an intervention designed to support pain self-management in cancer patients: a mixed methods study : a mixed methods study

Cancer is the second most frequent cause of death in Germany1. Patients with
cancer experience multiple symptoms throughout the different stages of their illness, with
pain as one of the most frequent symptoms. Pain is defined as "an unpleasant sensory and
emotional experience associated with actual or potential tissue damage, or described in
terms of such damage"2. Cancer pain can occur at any time but the incidence increases
over the course of the illness. Cancer pain management is to a great extent based on
pharmacologic and non pharmacologic therapy3. Despite effective treatment options, pain
control is not adequate in over 40% of patients with cancer4.
The undertreatment of cancer pain has led researchers to determine the factors that
impede effective pain management. Besides inadequate assessment and treatment,
several patient-related barriers interfere with optimal pain management5. Patient-related
barriers towards cancer pain management include cognitive (e.g., concerns about
analgesic use), affective (e.g., stress, depression), sensory (e.g., experience of side
effects), as well as practical components6,7. Oncology treatment takes place primarily in the
outpatient setting. Therefore, patients and their family caregivers (FCs) need to implement
cancer pain management mainly by themselves. Self-management consists of those
strategies that patients and FCs perform in daily life to manage illness and treatment to
avoid functional regression and health deterioration8. For optimal pain self-management,
oncology patients and their FCs need to acquire and process a complex array of
knowledge and skills (i.e., how to obtain, take, and titrate various analgesic medications,
deal with side effects, and what to do if pain is not relieved)3. In view of the complexity of
these impediments to adequate pain treatment, interventions to support cancer pain selfmanagement
are warranted.
Effects of interventions that were designed to support cancer pain self-management
were examined in three systematic reviews9-11. It was shown that from 1962 to 2009,
cancer pain self-management interventions gained increasing attention but were still an
understudied approach. While improvements in pain were reported in all three systematic
reviews, effects were moderate. Moreover, significant heterogeneity in study designs,
methods, and types of interventions may have weakened the findings. In addition, it was
not possible to determine which components of interventions were most effective. To our
knowledge, a detailed description and evaluation of components of cancer pain selfmanagement
interventions had not been done. Furthermore, only a few studies evaluated
how patients’ and their FCs’ experienced cancer pain self-management interventions6,12.
Interventions designed to support cancer pain self-management are largely
unknown in German speaking populations. Therefore, the work of Miaskowski’s PROSELF
© Pain Control Program (PCP) research group13,14 provided the basis for this thesis.
The PRO-SELF© PCP was built on three key strategies (i.e., provision of information using
academic detailing15, skill building, and ongoing nurse coaching) and consisted of
structured and tailored components. Patients and FCs were provided 3 visits and 3 phone
calls over six weeks by a specially trained nurse. A randomized controlled trial (RCT) with
174 patients showed that the overall effects of the PRO-SELF© PCP represented
statistically significant but moderate decreases in pain intensity scores. Therefore, based
on the convergence of findings from quantitative16 and qualitative6 subanalyses,
modifications to the Program resulted in the PRO-SELF© Plus PCP.
The overall aims of this mixed methods research project were twofold: first, to
analyze the state of knowledge regarding interventions designed to support pain selfmanagement
of patients with cancer; and second, to translate, adapt, improve, and pilot
test a state of the art intervention designed to support self-management of cancer patients
and their FCs in a German speaking population.
To address the first aim, a systematic review of experimental and quasiexperimental
studies was performed to describe the structure and content components, as
well as the efficacy of various components of cancer pain self-management interventions.
Intervention components were categorized using content analysis and evaluated based on
the effect sizes that were reported. Based on 34 publications (i.e., 24 interventions), seven
structure (e.g., mode of delivery) and 16 content components (e.g., information about
cancer pain and its treatment) were identified. No single component was found to have a
discernable influence on effect sizes. To our knowledge, this review was the first that
provided a detailed overview of the various structural and content components of cancer
pain self-management interventions. However, because of a variety of limitations, the most
efficacious intervention components remain to be determined in future studies17.
To address the second aim, the German PRO-Self© Plus PCP was tested in a pilot
study. In a first phase, instruments from the PRO-Self© Plus PCP were translated and
intervention and study procedures were revised successfully resulting in the German PROSelf
© Plus PCP (see Chapter 4). To evaluate the latter, a nested concurrent mixed
methods approach18 was applied, in which a pilot RCT was combined with two qualitative
substudies. The pilot study was guided by two theoretical frameworks: symptom
management theory19 and social cognitive theory20.
The purpose of the pilot RCT was to evaluate feasibility and effect sizes for the
German PRO-Self© Plus PCP for the planning of a multi-center RCT. Thirty-nine oncology
outpatients with cancer-related pain were randomized to the intervention (n = 19) or control
(n = 20) groups. Patients in the intervention group received the Program consisting of 6
visits and 4 phone calls over ten weeks. Patients in the control group received the same
number of visits and phone calls and usual care. Primary outcomes were average and
worst pain intensity. Secondary outcomes included pain-related knowledge, opioid intake,
and self-efficacy. Data were collected at enrollment, weeks 6, 10, 14, and 22 after
randomization. Findings from the pilot RCT were reported in two manuscripts.
The purposes of the first manuscript were to describe the methods used and the
lessons learned during the pilot study, as well as the adaptations that were made for a
larger clinical trial. During the study, inclusion criteria were expanded. In addition, the
intervention may be improved by further adaptations (e.g., closer tailoring to patients’ pain
situation, inclusion of self-management support for nausea and vomiting). Furthermore,
coaching cancer patients across a complex treatment team was an important function of
the intervention. The pilot study proved to be useful to adapt study procedures, balance
burden and benefit for participants, and customize the intervention to patients’ needs and
abilities in order to improve feasibility and efficacy21.
In the second manuscript, quantitative findings from the German PRO-Self© Plus pilot
RCT were reported. The group by time effect was statistically significant for knowledge
scores (week 10: p = 0.04; week 22: p < 0.01), but not for pain, opioid intake, or self-efficacy.
The mean difference at week 10 between the intervention and the control groups was -0.55
(Cohen’s d = -0.26) for average pain and -0.73 (d = -0.29) for worst pain. At week 22, it was
-0.51 (d = -0.35) for average pain and -0.47 (d = -0.20) for worst pain. It was the first study to
evaluate the PRO-Self© Plus PCP for feasibility and potential effects in a German speaking
population. While pain management related knowledge improved significantly, effect size
calculations showed low to moderate reductions in average and worst pain22.
The purpose of the first qualitative substudy was to explore how patients and FCs
experienced intervention and study procedures of the German PRO-Self© Plus PCP pilot
RCT. Debriefing interviews were conducted with nine patients and four FCs who finished
the intervention. Interviews were transcribed verbatim, and analyzed using content
analysis23. Findings confirmed that for patients and FCs, a trustful relationship with the
intervention nurse was of utmost importance. Furthermore, the intervention’s focus on
cancer pain self-management was novel for and appreciated by the participants. The
intervention nurse’s expertise with pain self-management as well as her caring attitude, the
discussion of everyday problems, and sufficient time to discuss problems repeatedly
resulted in a high satisfaction of the participants and a meaningful pain reduction for most
patients in this substudy (see Chapter 7).
The purpose of the second qualitative study was to explore decision-making
processes of four women and four men in the intervention group concerning their pain
medications. Audiotaped protocols of the 10-week intervention and debriefing interviews
provided data for a secondary analysis which was done using content analysis23. Findings
showed that cancer patients stayed ambivalent about the use of analgesics over the whole
course of the intervention. Their need for adequate pain relief contrasted with their desire to
avoid analgesic medications. Despite positive experiences with analgesics, decisions were
reconsidered and overturned. Individually tailored counseling helped these patients adopt
new attitudes towards analgesics and gradually reduce their pain levels24.
To conclude this research project, methods were critically appraised to determine
their appropriateness. For the systematic review, a narrative approach was chosen
because a quantitative meta-analytic approach was considered to be inappropriate given
the heterogeneity of intervention components and research methods. The narrative
approach was appropriate to provide the basis for future trials to evaluate specific
intervention components.
For the pilot study, the combination of qualitative and quantitative methods was
found to be highly appropriate to evaluate the PRO-Self© Plus PCP for feasibility and for its
potential to reduce cancer pain. Compared to other cancer pain self-management
interventions that were summarized in a meta-analysis10 and in our systematic review, the
German PRO-Self© Plus PCP was an intensive and comprehensive intervention in terms
of both its structure and its content. In this context, reductions in pain intensity scores in the
German PRO-Self© Plus pilot RCT were relatively small. Reasons for the small effect sizes
might include that patients’ overall opioid intake was low. However, it could not be
determined whether the low opioid intake was due to a lack of patients’ adherence to the
intervention or because clinicians did not adapt the analgesic prescriptions. Therefore, the
assessment of patients’ adherence with the intervention and the quality of the analgesic
prescription should be included in future research.
Furthermore, it was hypothesized that the interventions efficacy may be improved
by addressing patient-related barriers more comprehensively; tailoring the number, timing,
and duration of the intervention more closely to the patients’ needs; including symptoms
that severely impact pain self-management; guiding the patients more closely across the
complex treatment team; and, with the patient’s permission, including direct contact of
research staff with the clinician in case the patient should become unable to self-manage
their pain and FC support was not available.
This research project has several limitations. In our systematic review, the
description of the intervention components did not include unpublished components
because the analysis was based on available publications. Furthermore, because of the
small sample, low recruitment, and high attrition, findings from the mixed methods study
need to be interpreted with caution. In addition, all of the study visits for both the
intervention and the control groups were performed by the same nurse, which may have
resulted in an occasional unintentional contamination in the control group.
In summary, the systematic review in this thesis is the first to provide a
comprehensive overview of components of cancer pain self-management interventions in
view of their potential to reduce cancer pain. Furthermore, with the pilot study, the PROSelf
© Plus PCP was tested for the first time in a German speaking population. The transfer
and evaluation of the Program were performed successfully. In the pilot RCT pain selfmanagement
related knowledge increased significantly and it was possible to calculate
effect sizes for the intervention’s potential for pain reduction. Findings demonstrated
patients’ high satisfaction with the German PRO-Self© Plus PCP and showed how patients
remained highly ambivalent regarding analgesic intake even if they decided to take
analgesics. Even though the effects were rather small, findings from this mixed methods
research project provided valuable data on how to improve the intervention and plan a
larger multicenter RCT.