By continuing to browse this site you agree to us using cookies as described in About Cookies
Wiley Online Library is migrating to a new platform powered by Atypon, the leading provider of scholarly publishing platforms. The new Wiley Online Library will be migrated over the weekend of February 24 & 25 and will be live on February 26, 2018. For more information, please visit our migration page: http://www.wileyactual.com/WOLMigration/
This research was conducted in affiliation with the Mary Margaret Walther Program for Cancer Care Research and the Behavioral Cooperative Oncology Group.
Treatments for clinically localized prostate carcinoma are accompanied by sexual, urinary, and bowel dysfunction and other sequelae that can result in significant distress and reduced well being. Methods capable of improving quality of life are needed that can be integrated into clinical practice. To address this need, a nurse-driven, cancer care intervention was developed and tested.
Within 6 weeks after completing treatment, 99 patients, along with their partners, were enrolled into a prospective, controlled trial and were randomized to receive the cancer care intervention or to receive standard care. Participants in the intervention arm met once each month for 6 months with an oncology nurse intervenor, who helped patients identify their quality-of-life needs using an interactive computer program. The intervener then provided education and support tailored to participants' needs. Primary outcome variables included 1) disease-specific quality of life, including sexual, urinary, and bowel outcomes and cancer worry; 2) depression; 3) dyadic adjustment; and 4) general quality of life. Outcomes data were collected prior to randomization and again at 4 months, 7 months, and 12 months posttreatment.
Patients in the intervention arm experienced long-term improvements in quality-of-life outcomes related to sexual functioning and cancer worry compared with patients who received standard care. Baseline depression moderated the impact of the intervention on several other quality-of-life outcomes.
In 2004, approximately 110,000 men in the United States will be diagnosed with clinically localized prostate carcinoma.1 The majority of these men will undergo some form of aggressive therapy in an effort to cure or slow their disease. Surgical removal of the prostate, external beam radiation therapy, and brachytherapy currently are the most commonly used methods of treatment.2, 3
Although evidence has begun to accumulate that aggressive therapy for prostate carcinoma reduces the rate of metastasis as well as disease-specific mortality,4 the efficacy of treatment on overall mortality has yet to be established definitively. Because prostate carcinoma often is relatively indolent and tends to strike older men with limited life expectancy, many patients have disease that never will become significant clinically.5, 6 Conversely, younger men, especially those with higher grade tumors, are likely to benefit from treatment.7, 8
Although the impact of treatment on overall mortality remains somewhat unclear, its effects on quality of life have been well documented.9–16 All current forms of aggressive therapy are accompanied by relatively common side effects, the most frequent of which are urinary, sexual, and bowel dysfunction. Approximately 30–50% of men experience chronic decrements in urinary and/or sexual functioning after surgery, which is the most common form of treatment.14, 17–22 Bowel problems tend to be more prevalent among patients who receive external beam radiation therapy,17, 20, 23 whereas the late effects of brachytherapy are less clear but appear to include significant levels of sexual dysfunction and irritative/obstructive urinary symptoms.16 The majority of men appear to suffer at least some chronic sexual, urinary, and/or bowel dysfunction and report concomitant rates of distress or bother over dysfunction in these areas.11, 14, 16
Although most work has focused on urinary, bowel, and sexual side effects in populations with nonrecurrent disease, investigators also have identified other adverse sequelae, some of which appear to be the more distal consequences of urinary, sexual, and bowel dysfunction. Employment problems, financial worries, other symptoms (e.g., fatigue and pain), declines in perceived masculinity, depression and anxiety, reductions in the quality of the spousal/partner relationship, and concerns over disease recurrence all may contribute to reduced well being.9, 24–27
At least a significant minority of patients appear to fail to adapt to these problems, resulting in chronic declines in quality of life. Even among men who learn to cope successfully with these concerns, irrevocable changes may occur during the adjustment period. Urinary and bowel dysfunction, if not managed properly, can lead to loss of employment, early retirement, and withdrawal from active participation in the community. Sexual dysfunction may lead to severe relationship disruptions between patients and their partners, especially among newer dyads.
Several types of medical interventions are available to treat urinary, sexual, and bowel dysfunction, but the relatively high rates of side effects documented in survivor studies suggest that existing interventions are not implemented consistently and/or that they benefit only select subgroups.14 Behavioral interventions or interventions that combine multiple approaches rarely have been tested, and only two, large, randomized trials have appeared in the literature to date.
Mishel and her colleagues recently reported the results of psychoeducational, telephone-delivered intervention aimed at reducing uncertainty and the impact of side effects.28 During the first 4 months after treatment, intervention participants in that study experienced better urinary and sexual outcomes, with the latter finding moderated by race. More recently, Lepore and colleagues published the results of a trial in which they assessed the effect of a lecture group alone or in conjunction with a group discussion.27 In their study, intervention participants reported greater prostate carcinoma knowledge overall and better physical functioning among men who did not have a college degree. Men in the combined intervention group also were more likely to maintain steady employment and reported less sexual bother relative to the control group.
Although they focused on somewhat different outcomes and were delivered in different settings, the studies described above indicated the benefits of behavioral interventions for prostate cancer populations. Ideally, such interventions 1) should be translatable into clinical practice; 2) should be tailored to the idiographic needs of patients, including the symptoms that they report; 3) should address not only symptoms but the more distal consequences of treatment and disease; and 4) should incorporate the patient's spouse or partner. Because patient and spousal well being are linked, interventions that are capable of targeting both members of the dyad may prove particularly efficacious. With these four criteria in mind, a nurse-driven, computer-assisted intervention was developed and implemented in a multisite trial that randomized patient-spouse dyads to either a standard-care arm or to an intervention arm. The intervention and its impact on patients' quality of life are described in greater detail below, and spousal outcomes are presented elsewhere.
MATERIALS AND METHODS
A prospective, multisite, randomized clinical trial design was used to assess the efficacy of a cancer care intervention that was designed to improve the quality of life in the patient/spouse dyad during the first year after treatment for clinically localized prostate carcinoma. After the conclusion of treatment for prostate carcinoma, dyads in the intervention arm met once each month for 6 months with a nurse intervenor (twice in person and 4 times by telephone). The nurse intervenor identified and tracked quality-of-life problems using an assessment program developed for the cancer care intervention that was run from a laptop computer. For each problem, evidence-based strategies that had been extracted previously from the medical, nursing, and psychological literatures were considered; and a mutually agreed upon, tailored plan of care then was developed and implemented by the nurse and the dyad. The intervention process was facilitated and documented by the computer program. Quality-of-life outcomes data were collected at enrollment and again at 4 months, 7 months, and 12 months after the conclusion of treatment.
All procedures used in the current study were approved by the Institutional Review Board (IRB) of Indiana University and by the local IRB at each of the recruiting sites. Patient participants were recruited from the Indiana University Cancer Center;, the West Michigan Cancer Center, Ingham Hospital, and the Sparrow Radiation Clinic (Lansing, MI); and the Veterans Administration Medical Center (Louisville, KY). Patient participants were required to have received a diagnosis of Stage T1a–T2c prostate carcinoma; to be scheduled to undergo or to have undergone surgery, external beam radiation, or brachytherapy; to have a spouse or relationship partner who also was willing to participate and who enrolled within 2 weeks after the conclusion of therapy; to be age ≥ 18 years, and to speak fluent English.
Procedure: Quality-of-Life Conceptual Framework
This investigation assessed well being using a version of the proximal-distal framework that is advocated frequently by quality-of-life investigators (see Fig. 1). Proximal typically refers to the basic effects of disease or intervention on physiology or physiologic functioning. For example, urinary dysfunction would be considered an example of a proximal or disease-specific outcome in the context of prostate carcinoma. These specific domains are conceptualized as feeding into broader and more general domains, like physical functioning, role performance, and similar variables, which, in turn, feed into general affective states and overall life satisfaction. For parsimony's sake and in keeping with prior prostate carcinoma research, in the current report, we discuss outcomes as either specific (e.g., urinary function, sexual bother, etc.) or general (e.g., depression, physical function, etc.).
Description of the Intervention
The primary objectives of the intervention were to identify quality-of-life problems across the proximal-distal continuum, to provide symptom management and psychoeducational strategies tailored to eliminate or reduce the impact of identified problems, and to monitor the impact of the strategies on problems over time. The intervention primarily focused on problems related to sexual, urinary, and bowel dysfunction; cancer worry (i.e., anxiety over recurrence); dyadic adjustment; depression; and other common sequelae of cancer (e.g., fatigue and pain).
During each visit, the menu-driven computer program described above provided standardized questions and response formats that the nurse intervenor used to elicit and document information concerning quality-of-life problems. If a participant's score exceeded a prespecified threshold for a problem, the program prompted the nurse to assess the problem in greater detail and helped identify strategies for that problem.
For each problem, an extensive number of strategies could be called up in menu format from the program. Strategies ranged from those that were specific to a problem (e.g., Kegel exercises for urinary incontinence) to broader strategies that were suitable for a variety of problems (e.g., exercise could be prescribed for fatigue, anxiety, or symptom distress). A complete list of all strategies is available from the authors (see Preliminary Analyses, below). After the first intervener visit, the program also was used to record whether previously identified problems had resolved or persisted and whether prior strategies should be continued, adjusted in terms of intensity or frequency, or halted.
During the first visit, which occurred within 6 weeks after the conclusion of active therapy, the nurse intervenor primarily focused on assessing and managing bowel and urinary function problems. After the visit, participants were provided with a videotape to view at home (Living and Loving: Sexuality and the Prostate Cancer Patient; available from Zeneca Pharmaceuticals), which showed couples discussing how cancer had affected their sexuality and relationship, and a binder or “tool kit”, which contained tabbed pages with information related to managing the symptoms and side effects commonly experienced by patients with prostate carcinoma. The videotape was intended to provide examples that could be modeled of couples discussing potentially sensitive topics (e.g., sexuality) to facilitate discussions of similar topics with the nurse intervenor during the next visit. The tool kit was used to enable participants to review the strategies taught by the nurse on an as-needed basis.
During the second visit, which occurred 1 month later, the nurse used the computer program to evaluate problems related to sexual functioning, cancer worry, dyadic adjustment, depression, and other cancer-related problems. On subsequent encounters, the patient and spouse were asked to discuss issues and concerns that may not have been addressed effectively during the previous sessions and to identify any new problems that may have arisen. Intervention visits were scheduled to occur once every month during the first 6 months after completion of treatment with the first 2 visits in person and the remaining visits over the telephone.
After informed consent was obtained from participants, a baseline interview was conducted within 2 weeks after the end of treatment using computer-assisted telephone interview procedures. Differences in how patients were managed across the various sites precluded the collection of pretreatment data.
After the baseline interview, participants were randomized to the intervention arm or the control arm, stratified by recruitment site and treatment modality. After baseline (i.e., Wave 1), outcomes data were collected with computer-assisted telephone interviews 3 more times at 4 months, 7 months, and 12 months posttreatment (i.e., Waves 2–4). Interviewers were blind to the group assignment of participants. Clinical and demographic data were gathered by research assistants from chart reviews and patient interviews after recruitment.
Primary Outcomes Measures
Outcomes data were collected using a combination of specific and general self-report measures, which are described below. Some additional data also were collected but were not germane to the current study.
Urinary, sexual, bowel, and cancer worry outcomes were assessed using the Prostate Cancer Quality of Life Instrument (PCQoL), a psychometrically valid and reliable, 52-item, multiscale instrument targeted at men who are treated for localized disease. The PCQoL contains scales that assess physiologic dysfunction, limitations in role activities caused by physiologic dysfunction, and bother (i.e., psychological distress) caused by physiologic dysfunction.12 Each of these domains is assessed within each of the three organ systems (i.e., urinary, sexual, and bowel) that are likely to be affected by treatment for localized disease, for a total of nine scales. An additional, tenth scale taps cancer worry or concerns over disease spread/recurrence. Scale scores range from 0 to 100, with higher numbers denoting better outcomes. In prior work, scale α coefficients ranged from 0.70 to 0.90, with most in the middle 0.80 range or greater.
The Center for Epidemiologic Studies-Depression Scale is a 20-item measure of depressive symptomatology that was developed for use in community and clinical populations and also has been used extensively in cancer populations, with α coefficients ranging from 0.85 to 0.90 in prior work.30
Two scales from the 32-item Spanier Dyadic Adjustment Scale (DAS) were used to assess relationship functioning.31 For this investigation, only the Dyadic Satisfaction and Dyadic Cohesion subscales were used to decrease the number of instruments that participants were asked to complete. Spanier reported Cronbach α coefficients of 0.86 and 0.73, respectively, for the 2 subscales. It has been demonstrated that the DAS discriminates married couples from divorced couples and correlates with other measures of dyadic adjustment, such as the Locke–Wallace Marital Adjustment Scale (r = 0.87).
General quality of life.
The Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36) is comprised of 8 scales that assess physical function, physical role function, bodily pain, general health perceptions, emotional well being, emotional role function, social functioning, and vitality.31 The SF-36 is among the most widely used measures of generic health-related quality of life. In a mixed sample of patients with prostate carcinoma and age-matched controls, Litwin et al.33 found that scale α coefficients ranged from 0.78 to 0.93 (average, 0.85). The SF-36 has been validated extensively in numerous populations.
Univariate analysis was conducted for all quality-of-life outcomes to obtain descriptive statistics of the variables. Covariates for primary analysis were identified through bivariate analysis using t tests and Fisher exact tests. To examine differences between the intervention and control conditions on quality-of-life outcome variables, changes in quality-of-life scores were computed and were subjected to linear regression analyses. This approach was used because change scores were distributed nearly normally, whereas the raw follow-up scores severely violated the linear regression assumptions; this was due to skewness occurring in the same direction at both baseline and at any of the follow-up time points.
For the changes in quality-of-life outcomes between Wave 1 and subsequent waves, separate linear regression models were used to analyze how the group variable (intervention vs. control) influenced each specific change (from Wave 1 to Wave 2, from Wave 1 to Wave 3, and from Wave 1 to Wave 4) after adjusting for age. Thus, the independent variables in each model included group and age. We adjusted for age in all regression models, because intervention and control group participants differed on age (P = 0.001) but not on other characteristics (P > 0.25), including baseline quality-of-life variables.
Effect sizes for the group comparison were calculated according to mean change in score for the intervention group minus the mean change in score for the control group, divided by the pooled standard deviation. In regression analyses, interactions between the group variable and other potentially confounding variables also were tested. If an interaction was significant (P < 0.05), then we reported and tested the means (adjusted for age) for the intervention group and the control group separately for each level of the interacting variable. We observed that only baseline depression interacted with group. To describe this interaction, we dichotomized baseline depression into a low group and a high group based on the median (≥ 5 vs. < 5).
In total, 99 dyads completed Wave 1 and subsequently were randomized (for sample characteristics, see Table 1). The demographics of potential dyads who were approached but chose not to participate (n = 207 dyads) were similar to consenting dyads, although the average age of nonconsenting patients was somewhat higher (i.e., 68 years). By Wave 4, 85 of the original consenting dyads remained in the study, and 14 dyads had dropped out. The primary reason stated for dropping out was inconvenience. Attrition rates across the intervention and standard-care groups were nearly identical, and attriters did not differ from those who completed the study on any demographic, clinical, or baseline quality-of-life variables, with the exception that attriters had marginally worse role-emotional functioning at baseline (P < 0.07). Because some respondents occasionally failed to answer all items during the interviews, the sample sizes fluctuated slightly in the analyses described below.
Table 1. Sample Characteristics
No. of patients (%)
CES-D: Center for Epidemiologic Studies-Depression Scale.
Mean age (yrs)
Mean CES-D (baseline)
10th or 11th grade
High school diploma
College degree (4 yr)
Some grad school/degree
External beam radiation
Not married currently
The problems that most often met threshold for intervention attempts included sexual dysfunction (23% of patients), urinary dysfunction (19%), and dyadic adjustment concerns (18%), although fatigue also was experienced by a significant proportion of patients (10%). A variety of other problems also exceeded threshold, but no single other problem occurred in ≥ 10% of the sample. For urinary dysfunction, the most common strategies employed by the nurse intervenors were teaching Kegel exercises and scheduled voiding and providing counseling for distress/bother due to urinary dysfunction. For sexual dysfunction, the most common strategies were teaching dyadic communication skills and providing information about Viagra, injections, or other medical methods to overcome erectile dysfunction. For dyadic adjustment problems, the most common strategies were teaching active listening and how to express feelings and perceptions to partners. For fatigue, the use of exercise, priority setting, and energy management were the strategies that were employed most frequently.
Internal reliability of scales
Scale reliabilities generally met or exceeded accepted levels of internal consistency (i.e., > 0.70), with the exception of the bowel function scale. Psychometric analyses suggested that the bowel function scale was comprised of 2 subscales assessing bowel pain (α = 0.78) and bowel diarrhea (α = 0.52), but the patterns of results did not change whether examining overall bowel function or its 2 components. Thus, only overall bowel function is reported here.
Disease-specific quality of life outcomes: Sexual, urinary, bowel, and cancer worry
The intervention had significant beneficial effects on several of the disease-specific outcomes, with the most consistent effects accruing to the sexual outcomes. At Wave 2, patients who received the intervention reported significantly better gains in sexual functioning, which was a difference that remained marginally significant at Waves 3 and 4 (see Table 2). At Wave 2, intervention participants also reported greater reductions in the extent to which sexual dysfunction interfered with (i.e., limited) their role activities (e.g., spousal role activities). Although this difference was marginal at Wave 2, by Waves 3 and 4, this difference had become significant. Intervention participants did not differ from control participants in terms of sexual bother, but there was a clear trend favoring the former group.
Table 2. Intervention Effect on Specific Quality-of-Life Outcomes
Diff: mean of change score adjusted for baseline age; SD: standard deviation; ES: effect size.
Change scores are coded so that the larger the difference, the better the outcome (e.g., better function, fewer role limitations, less bother).
P values were determined from the regression model group effect, which compared the intervention group with the control group on the mean change score after adjusting for age.
In terms of bowel-related and urinary-related outcomes, the intervention and control groups did not differ statistically. However, by Wave 4, the intervention group did differ from the control group in terms of cancer worry. Men who had received the intervention enjoyed significant reductions in their anxiety about disease recurrence and treatment effectiveness.
Specific outcomes: Interactions with baseline depression
Baseline depression appeared to moderate the effects of group assignment on one of the specific quality-of life-outcomes: urinary bother (see Table 3). The intervention had no overall effect on urinary outcomes, as noted above. However, it was observed that baseline depression moderated the effect of the intervention on urinary bother at Waves 2, 3, and 4. Across the waves, for patients who had low levels of baseline depression, patients in the intervention group experienced significant improvement on the variable of urinary bother relative to the control group, although this effect became nonsignificant by Wave 4. For participants with high levels of baseline depression, the improvement of patients in the intervention group on this variable actually was worse relative to the control group, although this effect was marginally significant, and all groups improved over time.
Table 3. Interactions between Group and Baseline Depression
CES-D: Center for Epidemiologic Studies-Depression Scale; Diff: mean of change score adjusted for baseline age; SD: standard deviation; ES: effect size; SF-36: Medical Outcomes Study 36-Item Short-Form Health Survey.
P values were derived from the regression model group effect, which compared the intervention group with the control group on the mean change score after adjusting for age.
Urinary bother (Waves 2–1)
Urinary bother (Waves 3–1)
Urinary bother (Waves 4–1)
SF-36 Role-Emotional (Waves 4–1)
SF-36 Role-Physical (Waves 4–1)
General quality-of-life outcomes
In terms of the more general quality-of-life outcomes, which included dyadic functioning, depression, and the variables assessed by the SF-36, participants who took part in the intervention did not differ significantly from the control participants. Examination of the means in Table 4 reveals clear trends that consistently favor the intervention group (by an effect size of one-fourth to one-third of a standard deviation) on several of the more general quality-of-life variables (e.g., pain and emotional and physical role function), but these differences generally did not attain conventional levels of significance. There was a marginal effect of the intervention on dyadic satisfaction at Wave 4, such that intervention participants became more satisfied with the spousal relationship over time, whereas control participants became less satisfied.
Table 4. Intervention Effect on General Quality-of-Life Outcomes
Diff: mean of change score adjusted for baseline age; SD: standard deviation; ES: effect size; SF-36: Medical Outcomes Study 36-Item Short-Form Health Survey; DAS: Spanier Dyadic Adjustment Scale; CES-D: Center for Epidemiologic Studies-Depression Scale.
Change scores are coded so that the larger the difference, the better the outcome for SF-36 and DAS scores, whereas the opposite holds for CES-D scores.
P values were determined from the regression model group effect, which compared the intervention group with the control group on the mean change score after adjusting for age.
SF-36 Pain Index
SF-36 Health Perceptions
SF-36 Health Transition
SF-36 Mental Health Index
SF-36 Phys. Functioning
SF-36 Social Functioning
DAS Dyadic Cohesion
DAS Dyadic Satisfaction
General outcomes: Interactions with baseline depression
For several of the general outcome variables, baseline depression appeared to moderate the effect of group assignment, as depicted in Table 3. At Wave 4, intervention participants with high baseline levels of depression experienced marginally greater gains in role-emotional functioning (P = 0.098) and significantly greater gains in role-physical functioning (P = 0.014) relative to the control group; participants who had low baseline levels of depression did not differ by group (P > 0.78).
Using prospective, randomized trial methodology, this investigation developed and tested a nurse-driven intervention that was designed to improve the quality of life of patients with prostate carcinoma during the first year after treatment for clinically localized disease. The intervention proved successful, in that patients who were assigned to the intervention experienced better quality-of-life outcomes relative to control participants in several domains, although these were moderated in some patients according to measures of participant baseline level of depression.
Specifically, patients in the intervention group enjoyed considerably greater gains in sexual outcomes and reductions in cancer worry compared with patients in the control group, with relatively substantial effect sizes ranging in the vicinity from 0.40 to 0.50. Patients in the intervention group also experienced greater improvements on urinary bother (i.e., less bother) over time, but only if they had lower levels of depression at baseline.
Participants with higher levels of depression at baseline improved over time regardless of group assignment, but there was a marginal trend suggesting that intervention participants improved to a lesser extent. This effect may have occurred because participants in the intervention group, by necessity, were led to focus their attention on health problems (e.g., urinary dysfunction). This may have been more distressing for patients with elevated levels of depressive symptoms. However, the only outcome that exhibited this effect was urinary bother, suggesting that other factors also were involved. In terms of bowel outcomes and the more general quality-of-life outcomes of depression, dyadic adjustment, and the variables assessed by the SF-36, the intervention group did not differ statistically from the control group, although trends generally favored intervention group, especially at Wave 4.
Several aspects of the intervention likely accounted for its beneficial effects. The first is that the intervention combined standardized assessment features, which largely were a product of the technology used in the study (i.e., the computer program), with the clinical expertise of the nurse intervenor. The marriage of technologically driven assessment with human-based clinical decision-making may have been one of the critical components of the intervention. Standardized assessment often is an overlooked but essential component of effective symptom management. Several studies have indicated that clinicians' lack of awareness of the presence and extent of patients' problems is one of the primary impediments to symptom control.34–37 Because the computer program provided standardized assessment throughout the study, the nurse intervenor was able to identify problems and to gauge effectively the impact of different strategies on problem severity. Combining this information with her clinical expertise, which guided her selection of strategies, was one of the key characteristics of the intervention.
The second feature that may have promoted positive outcomes among the intervention arm is a consequence of the first: Solutions to problems were tailored to the idiographic needs and circumstances of the patient. The computer-nurse intervener combination allowed for greater-than-usual capacity to identify problems and their distal consequences as well the ability to implement and adapt strategies from a wide range of disciplines. This allowed tailoring solutions to well defined problems.
Although it was not addressed specifically in the current report, incorporating the patient's spouse into the intervention also likely enhanced its positive effects. Because it is likely that patient and spousal well being are linked,38, 39 tailoring management strategies to the abilities of both the patient and the spouse and targeting strategies toward improving the quality of life of the spouse herself likely allowed the intervention to take advantage of the reciprocal relationship between patient and spouse (i.e., helping one also helped the other).
The findings of this study underscore two central points. The first is that nurse-driven interventions are capable of improving the quality of life of patients with prostate carcinoma. Improving patients' long-term sexual outcomes in a durable manner (i.e., up to 12 months posttreatment) is particularly important. Historically, clinical care has fallen short in this area despite its significant impact on quality of life.14 Due in part to the increasing use of PSA screening, sexual dysfunction and its treatment are likely to become increasingly greater areas of concern as men continue to be diagnosed and treated at an ever earlier age for prostate carcinoma. The results of this trial, along with findings from two other recent trials, suggest that the impact of sexual dysfunction on the quality of life in patients with prostate carcinoma can be ameliorated using behavioral techniques.27, 28 The challenge faced by researchers is to continue to streamline these approaches to allow integration into clinical practice and to explore alternative mechanisms of delivery, such as web-based approaches.
The second point is the important role that depression plays in the recovery of well being after treatment. The level of depression at baseline was the only variable that moderated the effects of the intervention for several different quality-of-life outcomes, although some effects only were marginal. To successfully meet the quality-of-life problems faced by the populations targeted by this investigation, a greater understanding of the role of depression in the recovery process is required. Our data suggest that participants with high levels of depression may benefit from different intervention approaches, such as treating depression first before moving on to other problems.
Several limitations of the current study deserve note. The first is that the study was underpowered. Accrual rates were not has high as expected, resulting in a relatively small sample size that could not provide the anticipated level of power. However, the intervention still was efficacious enough to generate significant and large effects on several of the more important disease-specific outcome variables, including sexual outcomes and cancer worry. At the general level, the trends favoring the intervention participants on the majority of outcome variables suggest that greater power would have confirmed the differences suggested by the patterns illustrated in Tables 2 and 4.
A second limitation that deserves comment is that we did not take specific steps to control for type II errors despite conducting multiple comparisons. Because intervention work in populations with prostate carcinoma is relatively novel, particularly in the case of the intervention used in the current study, we believed that it was more important to employ an analytical approach that would minimize the danger of failing to detect an actual benefit. A third limitation is that the benefits accrued by patients in the intervention group in the current study conceivably may have been due to placebo effects or related phenomena, such as demand or experimenter effects. However, these types of effects generally should have affected all outcomes in a similar manner, whereas the intervention clearly had differential effects on the quality-of-life outcomes targeted by the study.
Fourth, our participation rate was relatively low, although participants and refusers did not differ on any demographic variable except for age. Our sample primarily was Caucasian and was educated relatively well, which also may limit the generalizability of our findings. However, unlike some prior work,27 socioeconomic status did not moderate the effects of the intervention, suggesting that the intervention may be appropriate for populations from all socioeconomic strata. However, additional research with appropriately chosen groups clearly will be necessary to confirm this suggestion.
In conclusion, the current findings indicate that a computer-assisted, nurse-driven intervention can improve the quality of life for men who undergo treatment for clinically localized prostate carcinoma. At the disease-specific level, patients who were assigned to the intervention condition experienced significantly greater improvements in sexual outcomes, in cancer worry and, depending on the baseline level of depression, less bother associated with urinary dysfunction. Although the intervention and control groups did not differ statistically at the general level, trends tended to favor the intervention group, suggesting the intervention also may have affected more distal quality-of-life outcomes.