Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Email this article to a Colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Articles & Searches Download to citation manager Citing Articles via Google Scholar Articles by Boesen, E. Articles by Johansen, C. Search for Related Content PubMed Citation Articles by Boesen, E. Articles by Johansen, C. Syndromes Secondary to General Medical Disorders

Comparison of Participants and Non-Participants in a Randomized Psychosocial Intervention Study Among Patients With Malignant Melanoma

Received December 27, 2006; revised February 28, 2007; accepted March 21, 2007. From the Dept. of Psychosocial Cancer Research, Institute of Cancer Epidemiology, Nobelparken, the Strandboulevarden 49, Denmark, and the Danish Cancer Society, and Psychooncology Research Unit, University of Aarhus, Jens Chr., Skousvej 4, Denmark. Send correspondence to Ellen Boesen, M.Sc., Ph.D., Dept. of Psychosocial Cancer Research, Institute of Cancer Epidemiology, Nobelparken the Strandboulevarden 49, Denmark-2100. e-mail: ellen{at}cancer.dk; soren@as.aaa.dk
© 2007 The Academy of Psychosomatic Medicine

ABSTRACT

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
The authors investigated barriers for entering a randomized psychosocial intervention study among patients with malignant melanoma. Data on age, sex, prognostic factors, social class, marital status, mood, coping, social relations, and social desirability from participants and non-participants were compared in a regression model. More patients in lower socioeconomic classes declined participation. Non-participation was associated with low levels of distress and adaptive coping and higher levels of empathy and congruence from spouses. The type of intervention offered in this study may have more appeal to patients belonging to higher socioeconomic classes.

INTRODUCTION

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
When clinicians treating cancer patients decide to change treatment protocols, they rely on rigorous randomized trials. This design is optimal for comparing groups of patients participating in the trial but may generate results that cannot be generalized to patients not participating in the trial.^1,2

The increasing survival rate among cancer patients has increased the need for reliable data on the effect of psychosocial intervention in the rehabilitation of these patients. However, in the randomized psychosocial intervention studies published so far, the participation rate ranges from 40% to 80%,^3,4 leaving room for the possibility that selection of patients into the studies may hamper the generalization. In line with this hypothesis, it has been suggested that patients who agree to be included in randomized intervention trials are usually younger, in better health, and of a higher socioeconomic status than those who decline participation.⁵ Other reasons, such as low psychological distress,⁶ physical limitations,⁷ high social support,⁸ and resistance of physicians to enrolling the patients⁹ may explain why patients refuse participation in this kind of study.

We have recently reported the results of a randomized clinical study of the effect of a psycho-educational group intervention on mood and ways of coping among 262 melanoma patients at Stage I and II.¹⁰ To better understand barriers to entering psychosocial intervention studies, we analyzed data obtained from 137 patients who declined to participate in this study, in order to identify socioeconomic, biological, and psychological risk factors for nonparticipation. We compared participants and non-participants in a regression model including information on age, sex, prognostic factors, social class, marital status, mood, coping, social relationships, and social desirability.

METHOD

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
Eligible patients were age 18–70 and had been diagnosed with cutaneous malignant melanoma of T1–4, N1a–2a, and M0,¹¹ who were treated in the Eastern part of Denmark. The randomization procedure consisted of two steps: 1) Before discharge, the physicians in the departments of plastic surgery where the patients had their surgery gave them a pamphlet, which briefly described the study; 2) at their outpatient clinics, patients were contacted by a project nurse between 3 and 12 weeks after the surgery. Patients who gave oral and written consent completed a baseline questionnaire and were then immediately randomized to the Intervention or the Control group. The intervention was organized into six sessions, lasting approximately 2 hours each, and conducted over a 6-week period in groups of 8–10 patients. The intervention was modeled from the manual developed by Fawzy et al.,¹² containing education in health behavior, malignant melanoma, stress-management, problem-solving, and mutual support; the manual is described in detail elsewhere.¹²

Between February 1, 1999, and June 15, 2001, the physicians referred 420 patients to the project. Of these, 21 (5%) did not meet the inclusion criteria and were excluded before randomization. Of the 399 patients who met the inclusion criteria, 262 (66%) agreed to participate, and 137 (34%) declined participation (Table 1). Nonresponders were asked to complete the baseline questionnaire. This latter procedure was implemented when 10 patients had declined randomization, leaving 127 non-participants as eligible for this questionnaire survey, representing 93% of all non-participants. A total of 51 non-participants (40%) agreed to fill out the baseline questionnaire. This group of patients is henceforth denoted as "the questionnaire group." The 76 non-participants (60%) who declined to fill in a baseline questionnaire or were not contacted for other reasons (i.e., did not show up for clinical visit, declined to receive an information pamphlet, nurse forgot to ask about the questionnaire study) are denoted as "the non-participants."

View this table: [in this window] [in a new window]

TABLE 1. Reasons for Declining Participation in a Randomized Psycho-Educational Intervention Study Among Patients With Cutaneous Malignant Melanoma, 1999–2001, Copenhagen, Denmark

Baseline Clinical, Psychosocial, and Behavioral Measures
Information about the previous and current health status of the patients in the group of participants, the questionnaire group, and the non-participants was obtained from medical records at the hospitals in which patients were treated; this included information on tumor thickness, histological level of invasion, location, and presence of ulceration.

The baseline questionnaire covered behaviors such as tobacco-smoking, alcohol consumption, sun-protection behavior, exercise, and nutrition, and it also included demographic questions, including social status. This variable was constructed from level of education and/or number of professional subordinates.¹³ The Profile of Mood States (POMS) scale¹⁴ was used to measure psychological distress. This instrument contains 65 items, measuring six mood or affective states: tension/anxiety, depression/dejection, anger/hostility, vigor/activity, fatigue/inertia, and confusion/bewilderment. The results are summed into a total mood disturbance (TMD) score for affective state,¹⁴ which has been proved valid in several studies.^15–17 Coping styles were elicited from the Dealing-With-Illness Coping Inventory" (DWI), a questionnaire originally developed to measure the ability of patients to cope with newly-diagnosed AIDS,¹⁸ for which it showed internal consistency. A revision was made after two studies that included cancer patients had been conducted.^12,15 The responses are categorized into three general coping methods: the active-behavioral method, the active-cognitive method, and avoidance. Higher scores indicate greater use of a particular coping method. We used the revised version of the DWI in the present study.¹⁰ The Barret-Lennard Relationship Inventory (BLRI) was used to measure the level of empathy and congruence provided by the spouse or an adult person to whom the respondent felt most attached (i.e., "significant other").¹⁹ Originating from within a Rogerian framework, the BLRI was designed to measure factors seen as essential for the process of change in the psychotherapeutic setting as well as in marital and/or interpersonal relationships, in general. The BLRI has also been recognized as a proxy for the quality of interpersonal relationships, which finding is supported by its substantial correlation with measures of marital satisfaction.^19,20 The original 16-item subscales were initially reduced to 8 items per subscale on the basis of a confirmatory factor analysis reported by Ganley et al.²⁰ Two criteria were applied: Items with the highest factor-loadings were selected, while, at the same time, the originally balanced positive and negative wording of the items (i.e., four positively-worded and four negatively-worded items) were selected. Also, an extra "extreme-response" category was added to ensure that signs of a ceiling effect initially observed in a pilot study were eliminated (i.e., "Yes, I feel extremely strong that it is true." and "No, I feel extremely strong that it is not true."). Pilot-testing also showed that the reduced subscales correlated highly with the full subscales, (r >0.95). In the present sample, the internal consistency of the scales were high (Cronbach : 0.90 for empathy and 0.86 for congruence). The MCSDS (Marlow-Crown Scale of Social Desirability) consists of 33 items and is a widely used measure of social desirability. The MCSDS is designed to measure adherence to behaviors that are culturally sanctioned and approved but improbable in occurrence (i.e., "My table manners at home are as good as when I eat out in a restaurant.")²¹ High scorers are also characterized by believing in, as well as trying to behave accordingly to, rigid standards of self-control, and they are generally found to experience and report lower levels of negative affectivity than low scorers. Also, high scorers are more likely to avoid situations that might lead to disapproval, for an example, by preferring "low-risk" reflective therapists, rather than interpretive ones.²²

Statistical Analysis
The association between degree of participation and the baseline registration of biological variables was evaluated by generalized logistic regression, with three outcome categories: participants, questionnaire group, and non-participants. The analyses were adjusted for sex, age, and tumor thickness. Time since diagnosis was also evaluated for all patients by generalized logistic regression. Data on sociodemographic variables (marital and socioeconomic status) and psychological variables (mood, coping, social relations) were only available for the participant group and the questionnaire group and were evaluated by standard logistic-regression analyses and were adjusting for the same covariates and, further, for marital and socioeconomic status. The analyses were performed both with and without the 10 eligible non-participants, who were not asked to fill out the baseline questionnaire because they were recruited before the above-mentioned procedure was introduced. Exclusion of these 10 patients did not change the results, and, therefore, only analyses excluding these patients are presented.

RESULTS

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
Significantly more patients of lowest socioeconomic status declined participation in the study than did patients in higher levels of socioeconomic status (participants compared with questionnaire group, Table 2). Sex, age, and tumor classification did not predict participation in the study (all groups compared, Table 2). Non-participation was significantly associated with low TMD scores (indicating low levels of distress) on the POMS, low scores for active-cognitive coping, and high scores on the two relationship inventory (RI) scales (indicating high levels of empathy and congruence in the patients’ single most important interpersonal relationship; participants compared with questionnaire group, Table 3). The Social Desirability factor (MCSDS) did not predict participation (Table 3) and was negatively associated to POMS TMD (r = –0.36; p=0.0005) and positively associated to BLRI empathy (r=0.22; p=0.03) and congruence (r=0.15; p=0.05). None of the health-behavior scores were predictive of declining participation in the study, nor was time from diagnosis to inclusion in the study (data not shown).

View this table: [in this window] [in a new window]

TABLE 2. Socio-Demographic and Biological Baseline Characteristics of 259 Patients With Malignant Melanoma Included in a Randomized Intervention Study of Psycho-Educational Group Therapy, as Well as the 51 Patients in the Questionnaire Group and the 76 Non-Participants, Copenhagen, Denmark, 1999–2001

View this table: [in this window] [in a new window]

TABLE 3.  Baseline Medians (95% Confidence Interval) for Various Outcome Measures From Participants and Non-Participants (Who Only Filled Out a Baseline Questionnaire) in the Psycho-Educational Therapy Program for Patients With Cutaneous Malignant Melanoma, 1999–2001; Copenhagen, Denmark

DISCUSSION

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 
Non-participants from the questionnaire group in this clinical trial were of lower socioeconomic status than participants. This result is in line with two questionnaire surveys among 667 and 154 cancer patients, respectively, which uniformly observed that patients of higher social class were more likely to seek psychosocial support than patients from lower social classes.^6,23 Likewise, a randomized psychosocial intervention study among 364 patients with breast cancer showed that patients with more education were more likely to accept participation than less educated patients.²⁴ Thus, it seems as if the psychosocial intervention that was offered appealed to patients of higher social classes. A body of evidence is emerging showing that health behavior and health outcomes are strongly dependent on socioeconomic status, including income, education, housing, and employment.²⁵ Our data suggest that this relationship may be an important factor in coping with psychosocial consequences of cancer, including participation in psychosocial therapy trials.

It has been speculated that patients declining participation may experience more pronounced psychosocial difficulties than participants and are therefore not able to manage participation in a trial.⁵ However, we found a significantly lower POMS TMD level in the questionnaire group, as compared with the participants, which is consistent with a finding among 216 Swedish melanoma patients who were recruited for a study of the effects of an information program.²⁶ The patients declining participation in the study by Brandberg scored significantly lower than the participants on tiredness and the HADS anxiety subscale. In another questionnaire survey among 215 German melanoma patients investigating the need for supportive counseling,²⁷ a positive correlation was found between the severity of the experienced distress and the patients’ interest in supportive counseling from professionals. These findings indicate that patients who perceive a need for support are more likely to enter a psychosocial intervention trial.

Regarding coping methods, it seemed as if the patients in the questionnaire group in our study used significantly less adaptive coping methods than patients who entered the randomized trial. Adaptive coping methods are characterized by, for example, seeking information and psychological support,¹² which are exactly the elements offered in the present intervention part of the trial. Thus, it seems as if accepting the intervention requires a high level of active coping, and acceptance of participation may be a proxy measure of a certain coping attitude.

The patients in the questionnaire group experienced more empathy and congruence in the relationship with their spouse or closest friend, as compared with participants. Patients declining participation obviously felt adequately supported by their social network and therefore felt no need for psychosocial intervention. This is in line with results from a cross-sectional questionnaire survey of 667 cancer patients who either had or had not sought support in various cancer-support groups.⁶ Patients in both groups considered their family and friends as helpful in solving the problems posed by the illness, and the patients reported that they were able to communicate openly about the cancer situation with family and friends. However, the patients in support groups were more likely to feel that their family members did not "truly understand" what they were going through.⁶ Insofar as received empathy and congruence can be described as "true understanding," this result is also in line with our findings.

The advantages of our study include the population-based design (the study covered an entire region of Denmark), combined with the fact that access to medical treatment is free for all residents. Also, we have complete follow-up of medical records, with information on tumor characteristics of all eligible patients approached for this intervention study. This makes it possible to evaluate potential selection bias attributable to these prognostic variables. We did not observe that the biological parameters predicted participation in the study when we compared the participants with all the non-participants, including patients who did not answer any questionnaires. However, level of distress might be a more important factor than disease stage when patients choose to enter a psychosocial intervention study, given that many patients may not be aware of their prognosis at the time of inclusion in the study. Awareness of prognosis does not necessarily predict psychological distress among patients with malignant melanoma.^28–30 Thus, the choice of participation in psychosocial interventions may be guided by the patient’s personal assessment of the benefits that can be obtained from participating.

A limitation of the study is the possibility that the non-participants who did not supply questionnaire data differ on the psychometric variables from the group of non-participants who accepted to fill out a baseline questionnaire. Although non-participants did not differ with respect to biological variables, it is possible that they may differ on psychological and social characteristics from the patients who contributed questionnaire data and/or patients who accepted randomization, thereby limiting the generalizability of the study. If the non-participants, for example, were significantly more distressed than the questionnaire group or the participants, they would represent a group of cancer patients in greater need of intervention but, obviously, intervention of another, more appealing kind. Another possible problem in obtaining information from non-participants in randomized psychosocial intervention studies may be that the patients who decline participation are eager to show that they decline because there is no need, and they therefore report better mood and social support than they actually have.³¹ To our knowledge, this is the first randomized psychosocial intervention study among cancer patients to include the Marlowe-Crown Desirability Scale (MCSD) to control for social desirability. We found no differences between the scores of the questionnaire group and the participants on this scale. Also, we found a negative correlation between the POMS–Total Mood Disturbance and the MCSD, and a positive correlation between the BLRI (relationship inventory) and MCSD. This finding is in line with the inverse association between negative affectivity and MCSDS that is generally found in other studies.²² The lower observed scores on the POMS–TMD and the higher scores on the BLRI in the questionnaire group therefore cannot be attributed to a relatively higher proportion of subjects scoring high on the MCSD scale in this group. Because high social-desirability scores have also been linked to a tendency to avoid situations posing potential risk to a positive self-image, such as a group-based psychosocial intervention, we had expected to find elevated scores on this scale among patients declining to participate. It may be that the intervention was not experienced as potentially threatening by the patients because it mainly emphasizes educational, rather than therapeutic, content, or, alternatively, there may be an overrepresentation of MCSD high scorers among the non-participants who were not willing to be assessed with a questionnaire. It is therefore likely that information bias due to social desirability, if any, would affect the results in the same direction as the presented findings, namely, that the non-participants generally experienced less distress and more social support than subjects who chose to participate.

In this study, we found that higher socioeconomic status, lower levels of intimate social support, and lower level of mood were predictors for participation in a psychosocial intervention study among patients with malignant melanoma, whereas prognostic tumor characteristics were not. Although we obtained questionnaire data from only 40% of all patients who declined participation in the intervention, the results address the problem that patients having lower socioeconomic status may probably need other types of psychosocial support.

In order to determine the patients’ preference for psychosocial intervention type, data on non-participants as well as participants are needed and should be obtained in future randomized studies of psychosocial intervention among cancer patients.

ACKNOWLEDGMENTS

 
The authors thank Lone Ross, M.D., Ph.D., Randi V. Karlsen, R.N., and Kirsten Frederiksen, M.Sc. (Statistics), Ph.D., for their contributions.

This study was funded by the Danish Cancer Society (Grant 97 225 59) and the IMK Foundation (Grant 30206–081).

View larger version (40K): [in this window] [in a new window]

FIGURE 1.  Flow Diagram for Inclusion in the Randomized Study of Psycho-Educational Intervention Among Danish Patients With Malignant Melanoma; Copenhagen, Denmark, 1999–2001

REFERENCES

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

 

1. Abel U, Koch A: The role of randomization in clinical studies: myths and beliefs. J Clin Epidemiol 1999; 52:487–497[CrossRef][Medline]
2. Black N: Why we need observational studies to evaluate the effectiveness of health care. BMJ 1996; 312:1215–1218[Free Full Text]
3. Berglund G, Bolund C, Gustafsson UL, et al: Is the wish to participate in a cancer rehabilitation program an indicator of the need? comparisons of participants and non-participants in a randomized study. Psychooncology 1997; 6:35–46[CrossRef][Medline]
4. Ross S, Grant A, Counsell C, et al: Barriers to participation in randomized, controlled trials: a systematic review. J Clin Epidemiol 1999; 52:1143–1156[CrossRef][Medline]
5. Mant D: Can randomized trials inform clinical decisions about individual patients? Lancet 1999; 353:743–746[CrossRef][Medline]
6. Taylor SE, Falke RL, Shoptaw SJ, et al: Social support, support groups, and the cancer patient. J Consult Clin Psychol 1986; 54:608–615[CrossRef][Medline]
7. Thiel dB: Cancer patients’ interest in group support programs. Cancer Nurs 1992; 15:347–352[Medline]
8. Richardson MA, Post-White J, Singletary SE, et al: Recruitment for complementary/alternative medicine trials: who participates after breast cancer? Ann Behav Med 1998; 20:190–198[Medline]
9. Goodwin PJ, Leszcz M, Quirt G, et al: Lessons learned from enrollment in the BEST study: a multicenter, randomized trial of group psychosocial support in metastatic breast cancer. J Clin Epidemiol 2000; 53:47–55[CrossRef][Medline]
10. Boesen EH, Ross L, Frederiksen K, et al: A psycho-educational intervention for patients with cutaneous malignant melanoma: a replication study of Fawzy et al. J Clin Oncol 2005; 23:1270–1277[Abstract/Free Full Text]
11. Balch CM, Buzaid AC, Soong SJ, et al: Final version of the American Joint Committee on Cancer Staging System for Cutaneous Melanoma. J Clin Oncol 2001; 19:3635–3648[Abstract/Free Full Text]
12. Fawzy FI, Fawzy NW: A structured psychoeducational intervention for cancer patients. Gen Hosp Psychiatry 1994; 16:149–192[CrossRef][Medline]
13. Enevoldsen B, Michelsen N, Friis-Hasche E, et al: Social classification, II: Svalastoga’s subdivision according to social status rank and the social grouping employed by the Institute for Social Research. Ugeskr Laeger 1980; 142:544–550[Medline]
14. McNair DM, Lorr M, Droppleman LF: EdiTS Manual for the Profile of Mood States: Manual for the Profile of Mood States. San Diego, CA, Educational and Industrial Testing Service, 1992
15. Fawzy FI, Fawzy NW, Hyun CS, et al: Malignant melanoma: effects of an early structured psychiatric intervention, coping, and affective state on recurrence and survival 6 years later. Arch Gen Psychiatry 1993; 50:681–689[Abstract/Free Full Text]
16. Classen C, Butler LD, Koopman C, et al: Supportive-expressive group therapy and distress in patients with metastatic breast cancer: a randomized clinical intervention trial. Arch Gen Psychiatry 2001; 58:494–501[Abstract/Free Full Text]
17. Goodwin PJ, Leszcz M, Ennis M, et al: The effect of group psychosocial support on survival in metastatic breast cancer. N Engl J Med 2001; 345:1719–1726[Abstract/Free Full Text]
18. Namir S, Wolcott D, Fawzy IF, et al: Coping with AIDS: psychological and health implications. J Appl Soc Psychol 1987; 17:309–328[CrossRef]
19. Barret-Lennard GT: The relationship inventory now: issues and advances in theory, method, and use; in The Psychotherapeutic Process. Edited by Greenberg LS, Pinsoff WM. New York, Guilford, 1986, pp 439-475
20. Ganley RM: The Barrett-Lennard Relationship Inventory (BLRI): current and potential uses with family systems. Fam Proc 1989; 28:107–115[CrossRef]
21. Crown D, Marlow D: A new scale of social desirability independent of psychopathology. J Consult Psychol 1960; 24:349–354[CrossRef][Medline]
22. Weinberger DA, Schwartz GE: Distress and restraint as superordinate dimensions of self-reported adjustment: a typological perspective. J Pers 1990; 58:381–417[CrossRef][Medline]
23. Bauman LJ, Gervey R, Siegel K: Factors associated with cancer patients’ participation in support groups. J Psychosoc Oncol 1992; 10:1–20[Medline]
24. Helgeson VS, Cohen S, Schulz R, et al: Group support interventions for women with breast cancer: who benefits from what? Health Psychol 2000; 19:107–114[CrossRef][Medline]
25. Marmot M, Wilkinson RG: Social Determinants of Health. New York, Oxford University Press, 1999
26. Brandberg Y, Bergenmar M, Bolund C, et al: Information to patients with malignant melanoma: a random-mixed group study. Patient Educ Couns 1994; 23:97–105[CrossRef]
27. Sollner W, Zingg-Schir M, Rumpold G, et al: Need for supportive counseling: the professionals’ versus the patients’ perspective: a survey in a representative sample of 236 melanoma patients. Psychother Psychosom 1998; 67:94–104[Medline]
28. Kelly B, Raphael B, Smithers M, et al: Psychological responses to malignant melanoma: an investigation of traumatic stress reactions to life-threatening illness. Gen Hosp Psychiatry 1995; 17:126–134[CrossRef][Medline]
29. Baider L, Perry S, Sison A, et al: The role of psychological variables in a group of melanoma patients: an Israeli sample. Psychosomatics 1997; 38:45–53[Abstract/Free Full Text]
30. Brandberg Y, Mansson-Brahme E, Ringborg U, et al: Psychological reactions in patients with malignant melanoma. Eur J Cancer 1995; 31A:157-162
31. Worden JW, Weisman AD: Do cancer patients really want counseling? Gen Hosp Psychiatry 1980; 2:100–103[CrossRef][Medline]

Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Email this article to a Colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Articles & Searches Download to citation manager Citing Articles via Google Scholar Articles by Boesen, E. Articles by Johansen, C. Search for Related Content PubMed Citation Articles by Boesen, E. Articles by Johansen, C. Syndromes Secondary to General Medical Disorders

Get information about faster international access.

Privacy Policy

Copyright © 2007 Academy of Psychosomatic Medicine. All rights reserved.

Home | Search | Current Issue | Past Issues | Subscribe | All APPI Journals | Help | Contact Us