Improving Health Care Utilization, Improving Chronic Disease ...

P1: GRA Journal of Clinical Psychology in Medical Settings

pp796-jocs-462220

March 7, 2003

15:28

Style file version June 24th, 2002

C 2003) Journal of Clinical Psychology in Medical Settings, Vol. 10, No. 1, March 2003 (°

Improving Health Care Utilization, Improving Chronic Disease Utilization, Health Status, and Adjustment in Adolescents and Young Adults With Cystic Fibrosis: A Preliminary Report Lloyd A. Taylor,1,5 Jan L. Wallander,1 Deborah Anderson,2 Pamela Beasley,3 and Ronald T. Brown4

We conducted a randomized clinical trial of a 3-session written self-disclosure intervention for patients with cystic fibrosis (CF). Patients (n = 39) who were at least 15 years of age and diagnosed with CF participated in the intervention. Participants in the intervention arm were asked to write in the health care setting about an important emotionally distressing issue of personal significance for a 20-min period of time and two additional 20-min writing episodes at the patient’s home, which were prompted by telephone calls. Patients in the control condition received standard care alone. Findings revealed that the intervention resulted in a reduction of the number of days patients spent in the hospital over a 3-month period. The intervention did not have an impact on physiological (Forced Expiratory Volume and Body Mass Index or subjective markers of health status. These findings extend those of Pennebaker’s (cf., J. Smyth, 1998) demonstrating an effect of the written-self-disclosure intervention on health care utilization. These preliminary findings are promising and justify further investigation of the modified intervention in other chronic illness populations. KEY WORDS: cystic fibrosis; written self-disclosure; health care utilization.

Over the past several years, within the field of pediatric psychology, there has been a burgeoning interest in empirically supported treatments that may be readily introduced in the clinical setting (Spirito, 1999). One treatment that has gained recent attention over the past decade, particularly among adult populations, is written self-disclosure (Pennebaker & Seagal, 1999). This intervention requires that the individual write about personal emotional experiences for approximately 20-min, which is repeated for 3–4 days (cf. Smyth, 1998). Participants are instructed to write

about their deepest thoughts and feelings related to the most distressing experience in their lives. Smyth (1998) published a meta-analysis of the impact of written self-disclosure in 13 studies of non health care seeking, healthy, young adults. Results revealed that written self-disclosure has an overall moderate impact on outcome (d = .472) and an even greater impact on physiological functioning (d = .60). The general model of inhibition has primarily been employed to explain these findings (Pennebaker, 1993). The inhibition model posits that the negative cognitions and emotions associated with distressing experiences require physiological and psychological vigilance to keep from disturbing immediate awareness and to maintain external equilibrium. The mechanisms hypothesized to be responsible for the positive effects of the intervention on psychological adjustment include cathartic expression and cognitive restructuring of emotional events. Written self-disclosure, then, is the process of translating

1 University

of Alabama at Birmingham, Birmingham, Alabama. of Pediatrics, Medical University of South Carolina, Charleston, South Carolina. 3 Department of Psychiatry, Judge Baker Children’s Hospital and Harvard University, Boston, Massachusetts. 4 Departments of Health Professions and Pediatrics, Medical University of South Carolina, Charleston, South Carolina. 5 Correspondence should be addressed to Lloyd A. Taylor, Department of Pediatrics, Medical University of South Carolina, 135 Rutledge Avenue, Charleston, South Carolina 29425. 2 Department

9 C 2003 Plenum Publishing Corporation 1068-9583/03/0300-0009/0 °


pp796-jocs-462220

March 7, 2003

15:28


10 cognitions and emotions into words, enabling individuals to make sense of distressing experiences (Pennebaker, 1988, 1997). Through the process of disclosure, individuals are able to weave a coherent story that enables them to simplify complex experiences, and develop a sense of control over previously unpredictable thoughts. This organization may function to modulate emotions and reduce intrapersonal conflict. The result is a diminished need for active inhibition and the physiological and emotional energy devoted to the inhibition process are freed to be utilized by the body in more efficient endeavors (Pennebaker, 1989; Pennebaker & Francis, 1996; Smyth, 1998). While the effectiveness of written self-disclosure has been demonstrated across a wide variety of groups, most of the work has been conducted in laboratory settings with college students. Little research on written self-disclosure has been conducted on clinic populations, despite the demonstrated effectiveness on health status in basically healthy populations. Smyth (1999) took a major step toward filling this gap by extending the previous work into chronic disease populations. Smyth demonstrated that patients with asthma (n = 58) or rheumatoid arthritis (n = 49) who wrote about the most distressing experiences of their lives for 20 min on 3 consecutive days evidenced statistical and clinically significant improvements in lung functioning and overall disease activity. Additionally, our research team (Klapow et al., 2001) has now modified the heretofore laboratory-implemented writing intervention to be suitable in the health care setting. Pilot testing with a primary care seeking adult population has provided encouraging preliminary support that written self-disclosure can be used in health care. Although these findings support the clinical use of written selfdisclosure with adult populations, no research can be located in the literature regarding the clinical feasibility and effectiveness of written self-disclosure interventions with pediatric chronic disease. Therefore, the main purpose of the present investigation was to determine the feasibility and effectiveness of modified written self-disclosure intervention in a chronic disease population that includes pediatric patients. The disease population chosen for study was patients with cystic fibrosis (CF), until recently a primarily pediatric disease. In fact, because CF is almost always identified during childhood, individuals with the disease are frequently followed in pediatric clinic throughout their lifespan. CF was chosen as a chronic illness because Smyth (1999) showed that lung functioning is sensitive to the

Taylor, Wallander, Anderson, Beasley, and Brown effects of the writing intervention in a group of other chronically ill patients diagnosed with asthma. Since CF is a chronic illness that primarily affects lung function, this disease was chosen for examination. The specific aims of this investigation were to evaluate in a controlled clinical trial the effectiveness, feasibility, and acceptability of the written self-disclosure intervention when applied in pediatric chronic disease health care. Hypotheses were compared to a wait-list control condition, patients who completed the written self-disclosure procedure would evidence improvement in health care utilization, disease severity, and subjective health status 3 months following the intervention.

METHOD Participants Participants were recruited (during routine) visits from two cystic fibrosis (CF) centers located at two children’s hospitals in the southeastern part of the United States. Information describing the study was mailed to all eligible clinic patients at each site prior to their recruitment in the clinic setting. Eligible patients were then approached during their routine clinic visits and asked to participate in a study investigating “the relationship between your feelings and your physical health.” Eligible patients were assured that participation would not affect their health care and that the CF clinic staff would not know who was enrolled in the study, nor would the staff have access to any of the data collected in the study. Initial eligibility criteria included (1) a diagnosis of CF, (2) age at least 15 years, (3) enrolled in the CF clinic for at least 9 months, (4) physical and mental ability to complete the research protocol as judged by the project staff and CF physician or nurse, (5) willingness to write for 20 min on three separate occasions over a 5-day period and completion of self-report measures on two occasions over the next 3 months, and (6) access to a telephone on the days that the participants were to write at home. Participants were recruited from consecutive patient visits at the CF clinics from February 2000 to May 2000. During this time, there were 81 eligible patients of which 70 (86%) were recruited and completed the baseline assessment. Eleven (14%) of those eligible refused participation citing time limitations. Of those enrolled, 14 failed to complete the intervention (20%) because of (1) receiving a lung


pp796-jocs-462220

March 7, 2003

15:28


Improving Health Care Utilization

11

transplant during the project phase (n = 3), (2) failing to complete all three writing sessions (n = 5), or (3) failing to return to clinic for the follow-up assessment (n = 6). In addition, 17 participants (24%) with unusually high (i.e., hospitalization ≥30 days suggestive of disease exacerbation) or low (i.e., no outpatient clinic visits suggestive of nonadherence) health care utilization patterns were excluded from the sample that was analyzed herein to test the intervention on patients with typical utilization patterns (i.e., at least one outpatient clinic visit during a 3-month period and no greater than 30 days of inpatient hospitalization). Demographic comparison data between those excluded and included in the final sample, using t-tests and chi-square analyses, revealed no significant differences between the two groups. In fact, an examination of the available data for those who were either excluded from participation or dropped-out of the study prematurely reveals that this particular group is a lower-middle class, largely Caucasian sample consistent with the group who completed the clinical trial. Demographic information for the final sample of 39 participants is presented in Table I. The majority of the sample was Caucasian (90%) adolescents and young adults ranging in age from 15 to 29 years (75%).

during the course of the project. Those who were enrolled, and parents or guardians of those under the age of 18 years, completed consent forms. Enrolled patients then completed baseline self-report measures prior to seeing their physician. Routine assessment of health status included forced expiratory volume and body mass index, which were recorded from the patient chart. Following the routine health exam, participants were randomized into either the written selfdisclosure (WSD) or the waiting list/standard medical care condition (SMC). Patients assigned to the SMC were told they would begin their writing sessions when they returned for their regularly scheduled CF clinic visit in 3 months. At that time, SMC patients were treated in the same manner as were those in the WSD condition. The actual time elapsed between baseline and follow-up assessments for each of the patients ranged from 3.2 to 3.7 months. The mean elapse times between assessments did not differ significantly between the WSD and SMC conditions or between sites. The physician and the clinic staff were not informed as to whether individual patients were in the WSD or the SMC condition. Written Self-Disclosure Intervention

Procedures Institutional Review Boards at each of the institutions approved the study. Psychology graduate students supervised at each site by a licensed psychologist conducted recruitment, assessment, and interventions. Identical procedures were implemented at the two centers, which were ensured by joint initial training followed by weekly phone conferences Table I. Demographic Information of Sample Chronological age (year) 15–18 19–22 23–25 26–29 30–33 34–37 >37 Marital status Married Not married Ethnicity Caucasian African American

N

%

12 6 3 8 4 5 1

31 15 8 21 10 13 3

15 24

38 62

35 4

90 10

Our research team adapted Pennebaker’s laboratory-based protocol to be a suitable health care intervention (cf., Klapow et al., 2001). Important modifications include clinic recruitment versus advertisement-based recruitment into a laboratory, introduction of the intervention in the clinic versus the laboratory, and two home-writing sessions prompted by a telephone call rather than an intervention completely implemented in the laboratory. At the conclusion of the routine health exam, participants were escorted to a private room within the CF clinic. Instructions6 were in accord with those of Pennebaker (1989, 1993, 1997) and required that participants write about their “deepest thoughts and feelings about the most distressing experience of their entire life for a period of 20 minutes” (Pennebaker, 1997). Participants were encouraged to connect their topic to relationships with others (e.g., parents, caregivers, lovers, friends, relatives) and to their past, present, or future. They were asked to write for the entire 20-min period and not be concerned with syntax or spelling. 6 Instructions

and all procedures are available upon request from the senior author.


pp796-jocs-462220

March 7, 2003

12 It was emphasized that all of their writing samples would be confidential, in particular that these would not be shared with the CF clinic staff. They could change from one topic to another across the 3-day period. Following completion of the first writing session, participants folded pages in half and placed them in a folder so that they were not readily visible to the assistant. Then participants were provided with detailed instructions for completion of the remaining two writing sessions at home. They were asked to identify two specific 20-min periods during the course of the next 5 days that would be designated for writing alone in a quiet room in their home. Paper designated for writing was provided that included the time they had chosen for writing. Participants were asked to mail each writing sample to the project manager at their site on the day following completion of the exercise, using prestamped envelopes provided for this purpose. A telephone call was made at the designated time, when the writing instructions were reiterated. Participants in both conditions were administered the same 3-month follow-up procedure. This included the administration of all the self-report measures and completion of the routine health exam. Subsequently, the objective measures of health status were gathered from the patient charts.

MEASURES Health care utilization was assessed from patient chart review. Specifically, the number of (1) outpatient visits and (2) inpatient hospitalization days were determined separately for the 3 months prior to and following enrollment in the study. Physiological disease severity measures included the following two: (1) Forced expiratory volume in 1 s (FEV1 ). FEV1 is routinely collected at each CF clinic visit as the primary indicator of health status and disease progression for CF patients. As CF progresses or is exacerbated, lung functioning is reduced, decreasing FEV1 . (2) Body mass index (BMI) = (kg body weight/m height2 ). This is also determined routinely at each clinic visit as a second indicator of current health status. The energy intake/output balance is typically challenged for patients with CF, reducing BMI as the disease progresses or is exacerbated. Subjective health status was measured with two commonly used health care research instruments. (1) ThePatient Health Questionnaire (PHQ; Spitzer

15:28


Taylor, Wallander, Anderson, Beasley, and Brown et al., 1994; Spitzer, Kronke, Williams et al., 1999) is designed as a screening instrument for use with health care seeking populations. It provides information on perceived symptoms of (a) depression, (b) anxiety, (c) somatic complaints, and (d) psychological distress. The two subscales that measure symptoms of depression (9 items, range 1–4, α internal consistency = .88) and anxiety (15 items, score range = 1–4, α = .95) employ Diagnostic and Statistical Manual IV (DSM-IV; American Psychiatric Association, 1994) criteria to screen for the presence of these psychiatric illnesses. The Somatization Scale (13 items, range 0–4, α internal consistency = .81) includes 13 common physical complaints (e.g., stomach pain, back pain, headaches), from which a severity score can be calculated. The Stressful Life Events Scale (10 items, range 1–4, α internal consistency = .81) includes 10 psychosocial complaints common among health care seeking populations (e.g., difficulties with family support, problems with significant others, and financial concerns). (2) The SF-12 (Ware, Snow, Kosinski, & Gandek, 1993) is a widely used, brief generic measure of self-reported health status derived from the larger SF-36 survey. Self-ratings are made of severity and frequency of 12 physical and mental health problems, as well as of their impact on the patients’ overall perceived health status. It yields both Physical Health and Mental Health summary scores, which are reported as standard scores (Mean = 50, SD = 10). This modified version of the SF-36 has consistently been shown to have good reliability and validity (Ware et al., 1993). Feasibility and acceptability of the intervention was assessed for those participants in the intervention arm, using a modified version of the Visit Specific Satisfaction Questionnaire (VSQ; Ware & Hays, 1988). This self-report measure was administered to participants and providers at the conclusion of the 3-month follow-up period. Participants in the WSD intervention rated, for example, satisfaction with the length of the writing sessions, convenience of the packets, level of comfort while writing, technical skills of the research team, integration of the intervention into the clinic, value of the writing sessions on patients’ mental, physical, and overall health). The health care staff rated the WSD intervention regarding integration of the intervention into clinic, convenience of writing for patients, technical skills of the research team, clinic time devoted to recruitment of patients, and value of the intervention to patients. Feasibility also was assessed through descriptive data on total intervention contact time.


pp796-jocs-462220

March 7, 2003

15:28


Improving Health Care Utilization

13

RESULTS Means and standard deviations for the dependent variables are presented in Table II. To determine whether there were significant differences between conditions (WSD versus SMC) on demographic (chronological age, marital status, race, and gender) and baseline dependent measures, chi-square and t tests were conducted. No significant differences were obtained for any of these measures (see Table II), confirming that random assignment of participants to conditions resulted in comparable groups. Dependent measures were conceptualized into three categories: (1) health care utilization (outpatient clinic visits, inpatient hospital days), (2) physiological markers of disease severity (FEV1 , BMI), and (3) subjective health status (depression, anxiety, somatic complaints, psychological distress, physical health, and mental health). Data for each category were analyzed with a 2 (Condition: WSD versus SMC) × 2 (Time: Baseline versus 3-month followup) repeated measures multivariate analyses of variance (MANOVA). Thus, three MANOVAs were performed. Of particular interest in this investigation is the Condition × Time interaction. Results of the MANOVA for health care utilization yielded a significant interaction [F(2, 73) =

Table II. Means and Standard Deviations, at Baseline and Follow-up for Dependent Measures Dependent measure Outpatient utilization Inpatient utilization FEV1 BMI Depression Anxiety Somatic complaints Stressful life events Physical health status Mental health status

Baseline

3.04, p < .05]. Separate univariate analyses of variance (ANOVA) performed on outpatient clinic visits and inpatient hospital days yielded a significant Condition × Time interaction for inpatient hospital days [F(1, 38) = 3.97, p < .05]. Figure 1 shows the means for each condition at the two time points indicating a decrease for inpatient hospital days for the intervention group, while inpatient hospital days actually increased for the control group. No significant interaction was found for outpatient clinic visits (all F < 1.0). The MANOVA that examined the physiological disease severity measures yielded no significant main effects or interaction (all F < 1.0). Finally, the MANOVA that examined the subjective health status measures did not yield any significant effects (all F < 1.0). Treatment Feasibility and Acceptability Participants in WSD rated satisfaction in the “good” to “very good” range across all dimensions assessed. One exception was regarding the value of the intervention at improving overall health, which was rated as “fair.” Health care providers (n = 10) also rated satisfaction in the “good” to “very good” range across all dimensions assessed. Total provider contact time to administer the WSD intervention for each participant was approximately 45 min and never exceeded 55 min.

Follow-up

WSDa

SMCb

WSD

1.1 (1.1)

1.3 (1.5)

1.2 (1.0)

2.0 (2.4)

8.1 (9.4)

3.6 (5.7)

5.6 (7.0)

8.4 (9.6)∗

47.9 (21.1) 18.9 (2.7) 12.7 (4.8) 15.9 (2.8) 5.1 (5.0)

51.4 (19.5) 19.5 (2.2) 14.3 (6.1) 16.1 (2.9) 7.6 (3.9)

48.5 (21.2) 19.3 (3.0) 12.4 (4.1) 15.3 (1.4) 3.9 (3.4)

50.4 (20.1) 19.0 (2.1) 13.9 (4.2) 16.1 (3.1) 7.0 (3.8)

13.7 (4.0)

14.3 (3.9)

14.8 (6.9)

14.3 (3.1)

45.9 (8.6)

45.5 (8.7)

43.8 (10.3) 43.3 (10.0)

50.3 (9.5)

48.9 (11.0) 53.7 (7.5)

SMC

49.5 (9.1)

Note. WSD Written self-disclosure; SMC: standard medical care; FEV1 : forced expiratory volume in 1 s; BMI: body mass index. a n = 18. b n = 21. ∗ p < .05, F(1, 38).

DISCUSSION The purpose of this investigation was to examine the effectiveness of a written self-disclosure intervention by means of a randomized controlled clinical trial in patients with cystic fibrosis (CF), all of whom were recruited in pediatric CF clinics. Findings revealed that participants in the treatment arm demonstrated a significant decrease in inpatient health care utilization compared to those in the control condition. In fact, patients who did not receive the writing intervention evidenced an increase in the number of days hospitalized (see Fig. 1). A careful examination of the data revealed that this finding is not due to outliers with regard to health care utilization. However, no significant effects were found on either physiological or psychological functioning. These findings are consistent with those of Pennebaker and others who have demonstrated improved health care


pp796-jocs-462220

March 7, 2003

14

15:28


Taylor, Wallander, Anderson, Beasley, and Brown

Fig. 1. Number of days in the hospital during pre- and postintervention for treatment and standard care conditions.

utilization in the form of visits to student health care clinics among college students who participated in a writing intervention (Pennebaker, 1991; Smyth, 1998; Smyth, 1999). These findings can have important implications for possible reductions in health care utilization and associated costs as well as enhanced quality of life. However, this investigation was not designed to identify the specific mechanism underlying the reduction in health care utilization as a function of the writing intervention. Thus, we found a two and a half day reduction in inpatient hospital stays for those who participated in the writing intervention compared to that expected for comparable CF patients receiving high-quality standard medical care. Although a specific mechanism underlying this reduction in the number of days hospitalized cannot be elucidated from our clinical trial, one possibility for the positive findings may be that through of process of written self-disclosure, patients are able to resolve psychological issues with regard to their chronic illness and hence, rely on health care providers less than those individuals who did not receive such an intervention. The total charges of one hospital day at the Children’s Hospital involved in this study for individuals with CF are $2,588 . When extrapolated from this trend over a 12-month period, this would constitute a savings of $25,878 per patient. When considering that most CF centers care for ap-

proximately 191 patients this represents a possible savings of at least $4 million per center. The cost of the writing intervention is only $130 per patient (figure estimated based on a psychologist fee of $130/hr × 55 min total administration time). While the findings must be considered cautiously until replicated, the study provides initial optimistic results suggesting that a simple psychological intervention may have a significant impact on health care and results in a relatively large financial savings to CF programs. This writing intervention offers the possibility that patients with CF may remain in the comfort of their own home rather than spend days in the hospital and disrupt daily living. Our study does not reveal significant differences in psychological and physiological markers of health status as a function of written self-disclosure. These findings depart from those of Smyth (1999) who found improvements in forced expiratory volume in adults with asthma as a function of a writing intervention. There are several possible explanations for these differences. First, the differences may be due to differential pathophysiologies of the different chronic illnesses examined in the two investigations. Specifically, asthma is an intermittent and highly variable disease with respect to overall symptom display and may be spontaneously reversible depending on the environment or passage of time. Furthermore, the exacerbation in patients with asthma


pp796-jocs-462220

March 7, 2003

Improving Health Care Utilization can be influenced by a number of psychological processes including anxiety, psychosocial stressors, and coping (e.g., Godding, Kruth, & Jamart, 1997; Kolbe, 1999; McQuaid & Nassau, 1999). Conversely, CF is a chronic, nonreversible genetic disease that has increased symptom display across the life span until a premature death. When considering our findings together with those of Smyth (1999), the writing intervention appears to exert differential effectiveness on symptoms of various diseases. Another reason that our investigation did not show psychological and physiological effects as did Pennebaker’s and Smyth’s previous investigations, may be due to the differences in the ecology of the settings in which the studies were conducted. Our study was conducted in an outpatient clinic and the patients’ homes, while the original studies were performed in the psychology laboratory. Thus, one possibility is that this intervention is more generally successful in the laboratory setting compared to the clinic. While the aforementioned indicators of disease severity (FEV1 , BMI) in patients with CF were not affected by the process of written self-disclosure, integration of the modified intervention into the CF clinic was successful. When examining patient and physician satisfaction measures for the treatment arm, the intervention received “good” to “very good” satisfaction ratings and it did not place a tremendous burden on the clinic staff. This provides evidence supporting its feasibility and acceptability in the CF health care setting. From a feasibility and acceptability standpoint, these findings support the possible extension of the writing intervention in future research to other similar health care settings and populations. Our findings must be interpreted within the limitations of this investigation. First, the investigation did not include an attention placebo control condition that may have controlled for the effect of simply having received an intervention. Clearly, replication is necessary with studies that include an appropriate placebo control condition. Second, the relatively small sample size may have diminished power to detect other treatment effects. Future studies will need to focus on greater recruitment efforts to counteract the attrition rates inherent in intervention studies involving serious chronic conditions. Third, it is recognized that there was a wide age span for the participants in this investigation, which may have produced significant variance thereby limiting the possibility of attaining significant results. Unfortunately,

15:28


15 the small sample size precluded us from determining whether the effects of written self-disclosure are equivalent in adolescents as they are in adults. Future studies will need to focus on the efficacy of written self-disclosure in adolescents as well as young adults. Our sample did not evidence psychiatric comorbidity. In fact, all of the participants were generally psychologically healthy. Thus, there may have been limited capacity for improved psychological functioning. Finally, as with the majority of clinical studies where there is assignment of homework, there was little opportunity to assure treatment integrity within the home portion of the intervention. Although all of the participants completed homework assignments and were prompted to spend 20 min on the assignment with the use of reminder telephone calls, whether the participants encountered distractions at home and whether there were differences in writing time across participants cannot be ruled out for certain. Our failure to demonstrate improvements in objective and subjective markers of health status following written self disclosure notwithstanding, it is most intriguing that the process of writing about emotional topics may lead to important changes in health care utilization patterns for numerous patients. These clinically important reductions in inpatient utilization justify further investigation to assess the utility of written self-disclosure in other chronically ill populations. Until further replication of our findings, cautious interpretation of the results presented herein is required. A logical next step would be to replicate the findings of Smyth (1999) in patients with asthma (and rheumatoid arthritis) in a clinic setting rather than the laboratory, both in adults and youth. Future extensions of the work may benefit from selecting patients whose diseases are known to be sensitive to psychological processes. Patients with illnesses such as pediatric Crohns disease and irritable bowel syndrome appear to be good candidates for further investigation because of their variable symptom display and the psychological and coping factors that exacerbate these diseases (Payne & Blanchard, 1995). In conclusion, we are cautiously optimistic regarding the use of written self-disclosure as a successful behavioral medicine intervention within the pediatric setting. At the very least, future research efforts are clearly warranted to determine further the clinical utility of this intervention for children and adolescents with chronic diseases.


pp796-jocs-462220

March 7, 2003

16 ACKNOWLEDGMENTS This research was supported in full by a grant award from the National Institutes of Health, Agency for Health Care Research, No. R03HS-10812-01.

REFERENCES American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Godding, V., Kruth, M., & Jamart, J. (1997). Joint consultation for high-risk asthmatic children and their families, with pediatrician and child psychiatrist as co-therapists: Model and evaluation. Family Process, 36(3), 265–280. Klapow, J. C., Schmidt, S. M., Taylor, L. A., Roller, P., Li, Q., Calhoun, J. W., et al.. (2001). Symptom management in older primary care patients: Feasibility of an experimental, written self-disclosure protocol. Annals of Internal Medicine, 134, 905– 911. Kolbe, J. (1999). Asthma education, action plans, psychosocial issues and adherence. Canadian Respiratory Journal, 6(3), 273– 280. McQuaid, E., & Nassau, J. (1999). Empirically supported treatments of disease-related symptoms in pediatric psychology: Asthma, diabetes, and cancer. Journal of Pediatric Psychology, 24(4), 305–328. Payne, A., & Blanchard, E. (1995). A controlled comparison of cognitive therapy and self-help support groups in the treatment of irritable bowel syndrome. Journal of Consulting and Clinical Psychology, 63(5), 779–786. Pennebaker, J. (1988). Confiding traumatic experiences and health. In S. Fisher & J. Reason (Eds.), Handbook of life stress, cognition, and health (pp. 669–682). New York: Wiley. Pennebaker, J. (1989). Stream of consciousness and stress: Levels of thinking. In J. Uleman & J. Scargh (Eds.), Unintended thought (pp. 327–350). New York: Guilford Press.

15:28


Taylor, Wallander, Anderson, Beasley, and Brown Pennebaker, J. (1991). Self-expressive writing: Implications for health, education, and welfare. In F. Belanoff, P. Elbow, & S. Fontaine (Eds.), Nothing begins with N: New investigations of freewriting (pp. 157–170). Edwardsville, IL: Southern Illinois University Press. Pennebaker, J. (1993). Putting stress into words: Health, linguistic, and therapeutic implications. Behavior Research Therapy, 49, 1427–1433. Pennebaker, J. (1997). Writing about emotional experiences as a therapeutic process. Psychological Science, 8, 162– 166. Pennebaker, J., & Francis, M. (1996). Cognitive, emotional, and language processes in disclosure. Cognition and Emotion, 10, 601–626. Pennebaker, J., & Seagal, J. (1999). Forming a story: The health benefits of narrative. Journal of Clinical Psychology, 55, 1243– 1254. Smyth, J. (1998). Written emotional expression: Effect sizes, outcome types, and moderating variables. Journal of Consulting and Clinical Psychology, 66, 174–184. Smyth, J. (1999). Effects of writing about stressful experiences on symptom reduction in patients with asthma and rheumatoid arthritis. JAMA, 281, 1304–1309. Spirito, A. (1999). Special series on empirically supported treatments in pediatric psychology: Introduction. Journal of Pediatric Psychology, 24, 87–90. Spitzer, R., Kronke, K., Williams, W., and the Patient Health Questionnaire Primary Care Study Group (1999). Validation and utility of the PRIME-MD. Journal of the American Medical Association, 282, 1737–1744. Spitzer, R., Williams, J., Kroenke, D., Linzer, M., deGruy, F., Hahn, S., et al. (1994). Utility of a new procedure for diagnosing mental disorders in primary care: The PRIME-MD 1000 study. JAMA, 272, 1749–1755. Ware, J., Snow, K., Kosinski, M., & Gandek, B. (1993). SF-36 health survey: Manual and interpretation guide. Boston, MA: The Health Institute, New England Medical Center. Ware, J., & Hays, R. (1988). Methods for measuring patient satisfaction with specific medical encounters. Medical Care, 26, 393–402.