Preventing Recurrent Pressure Ulcers in Veterans With Spinal Cord Injury: Impact of a Structured Education and Follow-Up Intervention

Article Outline

• Abstract
• Risk Factors for Pressure Ulcer Recurrence
• The Chronic Care Model
• Pressure Ulcer Prevention Programs
• Methods
  • Study Design
  • Power Analysis
  • Participant Recruitment
  • Procedure
    • Standard pressure ulcer education during hospitalization for pressure ulcer surgery
    • Enhanced educational intervention
    • Follow-up of control groups
    • Procedure if pressure ulcer recurred
    • End of follow-up
  • Measures
    • Demographic and SCI/D data
    • Medical and pressure ulcer surgical history and current tobacco use
    • Self-assessed health status
    • Pressure Ulcer Knowledge Test
    • Presurgery laboratory tests
    • Current surgery and recovery data
    • Postdischarge data
  • Data Analysis
    • Comparison of assignment groups
• Results
  • Sample Characteristics
  • Assignment Group Differences
    • Demographic and SCI/D data
    • Medical and pressure ulcer surgical history and current tobacco use
    • Self-assessed health status
    • Pressure ulcer knowledge
    • Presurgery laboratory tests
    • Current surgery and recovery
    • Pressure ulcer recurrence status
    • Test of primary hypothesis
    • Exploratory analyses
• Discussion
  • Effect of Education and Follow-Up Program on Recurrence
  • Study Limitations
• Conclusions
• Appendix 1. Content of Educational Model
• References
• Copyright

Abstract 

Rintala DH, Garber SL, Friedman JD, Holmes SA. Preventing recurrent pressure ulcers in veterans with spinal cord injury: impact of a structured education and follow-up intervention.

Objective

To test the hypothesis that enhanced education and structured follow-up after pressure ulcer surgery will result in fewer recurrences.

Design

Randomized controlled trial.

Setting

Veterans Affairs medical center.

Participants

Forty-nine veteran men with spinal cord injury or dysfunction were approached on admission for pressure ulcer surgery. Five never had surgery, 2 refused to participate, and one withdrew. Forty-one were randomized into 3 groups. Three participants' ulcers did not heal, so follow-up could not begin.

Interventions

Group 1 received individualized pressure ulcer education and monthly structured telephone follow-up (n=20); group 2 received monthly mail or telephone follow-up without educational content (n=11); and group 3 received quarterly mail or telephone follow-up without educational content (n=10). Follow-up continued until recurrence, death, or 24 months.

Main Outcome Measure

Time to pressure ulcer recurrence.

Results

Group 1 had a longer average time to ulcer recurrence or end of study than groups 2 and 3 (19.6mo, 10.1mo, 10.3mo; P=.002) and had a smaller rate of recurrence (33%, 60%, 90%; P=.007). Survival analysis confirmed these findings (P=.009).

Conclusions

Individualized education and structured monthly contacts may be effective in reducing the frequency of or delaying pressure ulcer recurrence after surgical repair of an ulcer.

Key Words: Patient education, Pressure ulcer, Rehabilitation, Spinal cord injuries

List of Abbreviations: AIS, American Spinal Injury Association Impairment Scale, CAI, computer-assisted instruction, CI, confidence interval, LOS, length of stay, MEDVAMC, Michael E. DeBakey Veterans Affairs Medical Center, MS, multiple sclerosis, OR, odds ratio, SCI, spinal cord injury, SCI/D, spinal cord injury or dysfunction

PRESSURE ULCERS CONTINUE to be a serious problem for many people with SCI/D. Krause and Broderick¹ reported that 17% of a sample of 633 nonambulatory participants with SCI/D of 5 or more years in duration had at least 1 new pressure ulcer every 2 years, 9% had at least 1 new pressure ulcer every year, and 4% had pressure ulcers almost constantly. Many of these recurring ulcers require surgical management, which often results in costly procedures, lengthy hospitalizations, expensive dressing changes, and time away from family, work, and leisure activities, greatly decreasing quality of life for these people.², ³, ⁴, ⁵, ⁶, ⁷ Rates of recurrence have ranged from 6% to 61%, whether or not treatment involved surgical or medical management.⁸, ⁹, ¹⁰, ¹¹, ¹², ¹³

Back to Article Outline

Risk Factors for Pressure Ulcer Recurrence 

Available data on factors associated with recurrence after surgical repair of pressure ulcers are limited. Much of the past and recent literature describing recurrence after surgery has focused on surgical technique.¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸, ¹⁹, ²⁰, ²¹, ²², ²³, ²⁴ Sorensen et al²⁵ suggested that thorough presurgery débridement, patient compliance, control of comorbidities, professional postoperative support, and sufficient pressure relief are essential if surgical success is to be achieved. Socioeconomic factors, lack of social support, and inadequate pressure ulcer prevention knowledge have been considered important issues related to recurrence.⁹, ¹¹, ¹², ²⁶, ²⁷ Studies on preventing recurrence have identified a number of demographic and medical factors that may predict multiple pressure ulcers. Recurrence has been associated with being a man, younger age, black race, unemployment, residing in a nursing home, and previous pressure ulcer surgery.⁶, ¹², ²⁸ Krause and Broderick¹ suggested that lifestyle, exercise, and diet were protective mechanisms against pressure ulcer recurrence.

Back to Article Outline

The Chronic Care Model 

The chronic care model emphasizes making changes both in patients and the health care system to create productive interactions around managing chronic health problems.²⁹, ³⁰ This model can serve as a guideline to improving both patient self-management and the health care system to provide proactive, organized care, rather than the reactive care that is characteristic of the acute care medical model. The 4 key components of the chronic care model are (1) self-management support (appropriate information and encouragement for patients to improve self-management), (2) delivery system design (providing chronic care requires productive interactions between patients and health care team members in an ongoing, planned manner, including close follow-up for patients who are most at risk), (3) decision support (evidence-based interventions), and (4) clinical information systems (timely data about individual patients and populations of patients). The first 2 components are most relevant to the present study.

Patient education requires assessing the patient's knowledge and cognitive abilities, planning short-term and long-term learning objectives and teaching strategies, using understandable language, and evaluating the extent to which the learning objectives are achieved. Chapman et al³¹ stated that encouraging patient responsibility and individualization of the treatment are essential if behavior is to be changed. They suggested that positive patient participation will be enhanced if the treatment is individualized to each patient's lifestyle, the regimen is easy to follow, positive reinforcement is provided, support systems are enhanced, and a patient care agreement is developed that delineates the patient's responsibilities. Green³² suggested that the success of health educational programs is in direct proportion to the degree to which patients are involved in defining their own needs, setting their own priorities, controlling their own solutions, and evaluating their own progress. In a study evaluating patients' skin care competence, Andberg et al³³ found that when the patient was an active participant in the process, there was greater compliance, and more learning took place. Garber et al³⁴ reported that participants with SCI who received a structured, individualized pressure ulcer prevention and treatment education program during hospitalization for pressure ulcer repair improved their pressure ulcer knowledge more than persons who did not receive the enhanced education. In addition, lower discharge knowledge scores were associated with older age, older age at onset of SCI, and a greater number of previous pressure ulcer surgeries. Selecting the most effective teaching strategy is also important. In 2003, Pellerito³⁵ reported on the effect of CAI on promoting independent skin care in adults with paraplegia. His results indicated that CAI was more effective than traditional educational methods in increasing the initiation and performance of pressure-relieving techniques.

Back to Article Outline

Pressure Ulcer Prevention Programs 

Although the literature is replete with numerous references to educational programs that describe protocols for preventing and treating pressure ulcers, there is a dearth of information on the effectiveness of these educational interventions to reduce occurrence, recurrence, or severity of pressure ulcers. Current pressure ulcer prevention programs emphasize a combination of specific behaviors which, when carried out on a regular basis, are intended to reduce the risk of occurrence and/or recurrence of skin breakdown. However, there are very few objective data demonstrating that strict adherence to a skin care program substantially reduces the occurrence or recurrence of pressure ulcers or that failure to practice these techniques on a regular basis is associated with the occurrence or recurrence of pressure ulcers.³⁶

Krouskop et al³⁷ demonstrated that a formal pressure ulcer clinic consisting of individualized education plus greater attention to treatment factors such as equipment prescription was successful in decreasing pressure ulcer recurrence among persons with SCI. Development of a pressure ulcer clinic with a comprehensive approach to management that emphasized education and prevention reduced the incidence of pressure ulcer recurrence from 32% to 11%. Research results were integrated into service and educational activities that resulted in effective assessment, individualized equipment prescriptions, and increased awareness by patients and families of their responsibilities for pressure ulcer prevention. Incidence of pressure ulcer recurrence was further reduced to 4% with the addition of an intensive psychologic counseling program. In a study by Jones et al,³⁸ participants were provided with several interventions: a comprehensive self-care plan, a graduated schedule of visits with an advanced practice nurse, and monetary rewards for preventing serious ulcers. Pressure ulcer severity and direct costs of treatment and preventive care were the outcome measures. The results indicated that the visits plus payments were effective in reducing ulcer severity.

The goal of the present study was to evaluate the effectiveness of an individualized educational intervention that included a 2-year, structured, follow-up program after surgical repair for a stage III or IV pressure ulcer. It was designed to reduce and/or delay recurrence compared with standard care and only minimal follow-up to assess skin status. The primary hypothesis was that persons with SCI who have had surgical repair of a pressure ulcer who receive an enhanced education and structured follow-up intervention will have fewer recurrences than those who do not receive the intervention.

Back to Article Outline

Methods 

Study Design 

The study used a 3-group, randomized controlled design. Participants in all 3 groups received standard care during and after hospitalization for pressure ulcer surgery. In addition, the design specified that 50% of the participants (group 1) were to receive an enhanced education and monthly structured follow-up intervention for up to 2 years after discharge; 25% of the participants (group 2) were to receive monthly contacts for up to 2 years after discharge to assess skin status, but with no study-provided, enhanced education during or after hospitalization; and 25% of the participants (group 3) were to receive minimal contact by mail every 3 months for up to 2 years after discharge only to assess skin status but no study-provided, enhanced education during or after hospitalization. The reason the design included twice as many participants in group 1 as in each of the control groups was to maximize the experience gained in providing the enhanced education and follow-up intervention. One role of the procedure specified for group 2 was to enable comparison with group 1 while controlling for the impact of monthly contacts. Quarterly contact with group 3 to assess skin status was necessary to avoid loss to follow-up and to obtain reasonably current information regarding dates of any ulcer recurrence. A postage-paid postcard was sent to group 3 participants. They were to note on it any change in skin status and return it by mail.

Participants in both groups 1 and 2 also participated in another component of the study designed to assess the potential of the amount of collagen metabolites excreted in the urine to predict the future development of pressure ulcers. This component was based on an earlier study in which increased urinary excretion of collagen metabolites was significantly associated with the later development of pressure ulcers.³⁶ Participants in groups 1 and 2 were asked to send a urine sample each month during follow-up. They were reminded to send the urine samples during the monthly contacts. The findings from the collagen metabolites component of this study are not reported here because they are beyond the scope of this study and have not yet been published.

For all 3 groups, follow-up continued for 2 years or until recurrence and subsequent discontinuation from the study, whichever came first. Discontinuation occurred when a participant had a recurrence of a pressure ulcer (stage II or greater) in the pelvic area (trochanter, ischium, or sacrum/coccyx) or died.

Power Analysis 

Sample size calculations were performed a priori to determine the required number of subjects to test the primary hypothesis. On the basis of a publication by Rodriguez and Garber,³⁶ it was assumed that 98% of group 1 would remain ulcer-free in the pelvic area within 1 year of their surgery date, and 88% of group 2 would remain ulcer-free at 1 year. To attain a power of .80 (α=.05) in comparing group 1 with group 2 and group 2 with group 3, 62 participants in group 1, 36 in group 2, and 36 in group 3 would be needed, for a total of 134 subjects.

Participant Recruitment 

Far fewer patients with SCI/D who met the inclusion and exclusion criteria were available than anticipated. We screened 49 men with SCI/D who were admitted to the MEDVAMC for the surgical repair of a stage III or IV pressure ulcer between December 1995 and November 1999 (fig 1). Veterans were included regardless of level and completeness of SCI/D, age, ethnicity, educational level, or functional independence. Veterans were excluded if they were cognitively impaired to the extent that they were unable to understand either the consent form or the educational material. Five people who were screened were admitted for pressure ulcer surgery during the study period but did not have the surgery, primarily because of anticipated noncompliance with postsurgery restrictions. Two people who were screened and found to be eligible refused to participate, and 1 person withdrew shortly after enrollment. Thirty-nine persons with SCI and 2 with MS participated, for a total sample size of 41.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Flowchart of study.

Procedure 

The study was approved by the Baylor College of Medicine Institutional Review Board for Human Subjects' Research and the MEDVAMC Research and Development Committee. Potential participants were interviewed soon after their admission to the MEDVAMC, at which time an investigator described the study. Persons who were interested in participating and met all study criteria signed a consent form and were given a copy. They were randomized to one of the 3 groups by selecting a numbered paper out of a container. Participants had a 50% chance of selecting a paper assigning them to group 1, a 25% chance of assigning them to group 2, and a 25% chance of assigning them to group 3.

All participants provided demographic and health information and completed the Pressure Ulcer Knowledge Test (see Measures below) on entry into the study.

During the period of the study, standard pressure ulcer education for patients admitted to the MEDVAMC for pressure ulcer surgery included the following:

In addition to standard care during hospitalization and after discharge, veterans in group 1 received up to 4 hours of structured individualized pressure ulcer prevention and management education near the end of their hospital stay. In general, this education was presented in four 1-hour sessions. The sessions were conducted by an occupational therapist with extensive expertise in pressure ulcer prevention and management. This approach was selected because 1-on-1, personal interaction enables the interviewer to identify the participant's reaction to the information presented. It enables the interviewer to ask questions and to some degree assess the person's knowledge or processing of the information. It also enables the participant to ask questions directly to someone who can provide the information immediately or who knows how to get it. It also presents the opportunity for the person to talk about other relevant issues that might not have surfaced with other educational techniques. Each session was interactive and consisted of specific information followed by discussion. Topics included pressure ulcer etiology, prevention strategies, nutrition, and pressure-reducing devices for the bed and wheelchair. The content of the educational model is presented in appendix 1. One of the sessions included family members, significant others, or attendants either in person or by telephone. The investigator also provided printed materials that the veteran could take home. These included monographs on skin care and prevention and treatment of pressure ulcers, all of which were developed specifically for persons at risk for pressure ulcers.³⁹, ⁴⁰, ⁴¹

The potential for diffusion of the information learned in the enhanced educational sessions to persons in the control groups could not be completely avoided or assessed. However, recruitment of the participants was staggered over a 4-year period, so the hospitalization of only a few participants overlapped at any time. The enhanced education program was individualized and occurred in a private office during the final 2 weeks of the hospitalization for each person in group 1; thus, the time available for diffusion of the intervention to participants in the 2 control groups was limited. Furthermore, although the groups were not significantly different in knowledge about pressure ulcer prevention and management at admission, Garber et al³⁴ reported that group 1 had significantly higher knowledge scores at discharge, which suggests that diffusion was not a major problem.

After discharge, in addition to any standard (ie, not study-related) contacts with health care professionals, group 1 participants were contacted by study personnel by telephone every month for 2 years or until discontinuation. Calls lasted 15 to 25 minutes; however, on average, it took at least 2 attempts to reach the participant each month. During the monthly contacts, participants responded to a list of questions about the status of their skin and pressure ulcer prevention behaviors: (1) How is your skin? (2) Have you had or do you have any breakdown now? (3) If yes, where is/was it? (4) Please describe it. (5) How are you treating/did you treat it? The call included discussion of other issues related to pressure ulcers such as equipment, help at home, and general activities. Participants were reminded of any behaviors that they should be doing but were not (eg, eating a balanced diet). They also were reminded to send in a urine sample for the collagen metabolites component of the study.

In addition to any standard (ie, not study-related) contacts with health care professionals, study personnel sent participants in group 2 a form each month on which they were to indicate (1) whether the area of the body in which they had the surgery was still healed; (2) whether they had had any pressure ulcers or broken areas of the skin in the past month and, if so, on what part of the body; and (3) if they answered yes to question 2, how the ulcer was treated. Participants in group 2 were also sent a container for a urine sample. Both the form and the urine sample were to be returned by postage-paid mail. If the urine sample and/or the form were not returned within 2 weeks, group 2 participants were contacted by telephone. Similarly, in addition to any standard (ie, not study-related) contacts with health care professionals, participants in group 3 were contacted quarterly by study personnel by letter and asked to complete and return a prepaid, preaddressed postcard indicating their skin status. The postcard asked the same questions asked on the form received by group 2. If the postcard was not received within 2 weeks, group 3 participants were contacted by telephone to obtain this information. At the periodic (monthly or quarterly) postdischarge contacts with groups 2 and 3, no information was sought or given regarding preventive practices until the final contact at 24 months postdischarge or when the participant was discontinued because of a recurrence of a pressure ulcer.

If the participants in any group indicated that they had a new skin breakdown, more information was obtained by telephone regarding such things as date of occurrence, location, size at its worst, current status with regard to healing, and perceived probable cause as well as any treatment, including surgery or hospitalization. If they reported a break in their skin, they were asked to describe the problem and any treatment received. They were strongly urged to be examined either through the SCI clinic at the MEDVAMC (if they were local) or at their closest VA or hospital emergency department. The investigator making the calls followed up with the clinic, the person's SCI physician, or the person's local VA health care facility. Patients who reported a break in their skin in the pelvic area completed several end-of-study questionnaires by phone and were discontinued from the study.

Participants who remained ulcer-free for the full 24-month follow-up period completed the same set of postintervention questionnaires as those who were discontinued because of recurrence.

Measures 

Participants' age, race and ethnicity, marital status, and level of education were obtained by interview. SCI/D descriptors such as level and completeness of injury, age at onset, and time since onset were obtained from computerized medical records. Completeness of injury was categorized according to the AIS.⁴² The AIS is a 5-level (grades A−E) scale with grade A indicating that no motor or sensory function is preserved in the sacral segments S4-5, grade B indicating sensory but not motor function is preserved below the neurologic level and includes the sacral segments S4-5, grade C indicating motor function is preserved below the neurologic level and more than half of key muscles below the neurologic level have a muscle grade less than 3, grade D indicating motor function is preserved below the neurologic level and at least half of key muscles below the neurologic level have a muscle grade of 3 or more, and grade E indicating normal motor and sensory function.

Information was obtained from participants and medical records regarding 11 types of health conditions (ie, cardiac, pulmonary, gastrointestinal, musculoskeletal, peripheral vascular, neurologic [other than SCI or MS], skin [including burns but not pressure ulcers], renal disease, collagen disease, diabetes, amputation) as well as whether they had spasticity and what, if any, medication was being taken for spasticity. Information was obtained regarding previous surgical procedures for closure of pressure ulcers, and, if so, when. They were also asked whether they currently used tobacco.

On entry into the study, participants rated their health as they perceived it to have been just prior to admission on a scale of 1 (poor) to 10 (excellent). They rated their health again on the same scale just prior to discharge. This scale was devised for the study and was designed to be similar to a numeric rating scale commonly used at the MEDVAMC to assess pain intensity, with which the veterans are very familiar.

The Pressure Ulcer Knowledge Test is an investigator-constructed, 14-item questionnaire that assesses knowledge about pressure ulcer etiology, stages of tissue deterioration, skin inspection, weight shifts and turns, nutrition, pressure-reducing devices for the wheelchair and bed, and prevention and management strategies. The items reflect recommendations in previous literature on prevention and management of pressure ulcers.³⁶, ³⁷, ⁴⁰, ⁴³, ⁴⁴, ⁴⁵ Scores can range from 0 to 60 points, and the percentage of possible points obtained is calculated. In this study, the Cronbach α was .72, indicating that the measure had acceptable internal consistency. In an earlier publication based on the current study, scores on the knowledge test were shown to increase more in the experimental group after the enhanced educational intervention than in the controls who received standard care.³⁴ Portions of the test were used in earlier studies.³⁶, ⁴⁶ In the first of these, less than 30% of the participants were able to recall preventive techniques other than weight shifts or turns, which suggests that knowledge about pressure ulcer prevention was spotty, at best.³⁶ In the second study, the number of prevention techniques recalled was found to be inversely and significantly related to the number of prior ulcers (r=−.50, P<.05), and the number of preventive behaviors reportedly practiced was significantly related to whether the person took immediate action to treat a newly discovered pressure ulcer.⁴⁶

Presurgery albumin and hemoglobin were obtained from the computerized medical records.

Additional data obtained from medical records included the location of the study ulcer (trochanter, ischium, or sacrum/coccyx), presence of heterotopic ossification and/or osteomyelitis, type of surgical closure (fasciocutaneous flap, muscle flap, or both types), grafts, number of drains, presence or absence of 5 selected postsurgical complications (wound infection, wound dehiscence, flap necrosis, skin graft loss, seroma), time on an air-fluidized bed, time from surgery to first sitting, and LOS.

The number of months to recurrence of a pressure ulcer in the pelvic area, death, or 24 months, whichever came first, was recorded.

Data Analysis 

Descriptive statistics were obtained for all study variables for the entire sample and for each of the 3 groups separately including means, standard deviations, and ranges for continuous variables and numbers and percentages for categoric variables. The total score and percentage of possible points attained on the Pressure Ulcer Knowledge Test, the number of other health problems among the 11 listed under Medical and Pressure Ulcer Surgical History and Current Tobacco Use, and the number of complications after surgery among the 5 listed under Current Surgery and Recovery Data were calculated.

Analyses of variance were performed with SPSS software^a to determine whether there were differences among the 3 assignment groups on all but 2 of the continuous study variables. For the highly skewed variables of time from surgery to first sitting and LOS, Kruskal-Wallis nonparametric analyses were conducted. Chi-square tests were performed with StatXact software^b to assess group differences on categoric variables. StatXact yields exact significance levels and is designed to analyze data in which some cells are very small or null without the necessity of collapsing across categories to increase cell size. A 1-tailed test was used for the comparison of the 3 groups on months to recurrence or end of study because the direction of the differences among the groups was specified a priori. All other group comparisons were 2-tailed. Effect sizes regarding ulcer-free survival time were calculated to compare group 1 with group 2 and group 1 with group 3. These effect-size analyses were repeated excluding the 2 persons with MS and the 1 Hispanic person.

To test the primary, a priori hypothesis, a 1-tailed Cox regression survival analysis was conducted with SPSS to assess assignment group differences in time to ulcer recurrence or end of follow-up (24mo). This analysis was repeated with the 2 persons with MS excluded to determine whether the results would vary if those participants were not included. Because a large number of separate analyses was performed, the experiment-wise α required for statistical significance was calculated as .05 divided by 43 analyses. Thus, P less than or equal to .001 was necessary to meet the experiment-wise criterion for significance. To meet the generalized or per comparison criterion, P less than or equal to .05 was necessary, and to meet the statistically marginal generalized (per comparison) criterion, P less than or equal to .10 was necessary. Significance levels of P greater than .10 were considered not significant.

Despite randomization, several variables were either significantly (P<.05) or marginally (P<.10) different across the 3 groups at baseline using the generalized criterion. Therefore, several post hoc exploratory Cox regression survival analyses were performed to assess the impact of these differences on ulcer-free survival time. The small sample size prohibited controlling for all these variables simultaneously. Thus, for each exploratory analysis, the dependent variable was time to ulcer recurrence or end of follow-up, and the independent variables were group assignment plus one of the other baseline characteristics that varied across groups. Furthermore, separate exploratory Cox regression survival analyses were performed to assess the effect of assignment group on the 2 subsets of participants, of whom one had had previous surgical closure of a pressure ulcer and the other had never had previous pressure ulcer surgery.

Back to Article Outline

Results 

Sample Characteristics 

The 41 participants had a wide range of age (28–78y), age at onset of SCI/D (19–69y), and time since onset of SCI/D (<1–44y) (table 1). The sample had an equal number of white (n=20) and black (n=20) participants and 1 Hispanic participant (table 2). Most were not married (63%). All but 3 had been graduated from high school or had received their General Educational Development diploma, and 56% had at least some college education. The most frequent cause of injury was a motor vehicle crash (42%). Most of the injuries were in the cervical (39%) or thoracic (56%) areas, and most (68%) persons with SCI had complete injuries. The prevalences of other health problems reported by participants or identified in the medical records were as follows: cardiac (15%), pulmonary (12%), gastrointestinal (20%), musculoskeletal (34%), peripheral vascular disease (12%), neurologic other than SCI or MS (5%), renal disease (24%), diabetes (24%), and amputation (7%). No one had a history of a skin problem other than pressure ulcers, and no one had a history of a collagen problem. The surgical closure for 3 participants did not heal. It was not possible to include them in the follow-up phase of the study because date of recurrence could not be ascertained; thus, follow-up data are available for only 38 participants.

Table 1. Continuous Data by Group

	Group 1 Enhanced Education and Monthly Structured Follow-up (n=20)			Group 2 Monthly Contact to Obtain Urine Sample and Assess Skin Condition (n=11)			Group 3 Quarterly Contact to Assess Skin Condition (n=10)
Variable	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	F	P
Demographic and SCI/D data
Age (y)	54.82	10.85	40–78	50.64	8.83	41–66	53.86	13.63	28–78	0.51	.604
Age at onset of SCI/D (y)	35.24	12.46	19–69	35.27	7.71	26–53	39.05	14.69	20–69	0.38	.688
Time since onset of SCI/D (y)	19.58	11.45	1–44	15.37	4.38	10–26	14.82	10.29	<1–28	1.08	.349
Medical and pressure ulcer surgical closure history
No. of previous health problems among 11 listed⁎	1.50	1.24	0–4	1.45	1.44	0–5	1.70	1.70	0–6	0.09	.912
No. of previous surgical closures (n=40)	0.95	1.19	0–4	1.30	1.25	0–4	1.10	1.60	0–5	0.24	.789
Time since last surgical closure (y) (n=22)	6.30	3.59	2–14	6.24	4.20	0.2–10	1.67	1.05	0.5–3	4.23	.030
Self-assessed health status
Self-assessed health at admission (n=37)	6.45	2.35	1–10	6.56	2.13	3–10	4.38	1.51	2–6	3.02	.062
Self-assessed health at discharge (n=36)	7.74	1.20	6–10	8.33	1.00	7–10	7.13	1.46	5–10	2.10	.138
Pressure ulcer knowledge
Percentage correct at admission	54.17	19.18	12–83	55.00	12.45	28–77	54.50	12.00	37–72	0.01	.991
Presurgery laboratory tests
Presurgery albumin (g/dL) (n=40)	2.94	0.53	1.7–3.8	2.83	0.59	2.0–4.1	2.98	0.49	2.3–3.6	0.20	.824
Presurgery hemoglobin (g/dL)	11.56	1.79	8.6–14.2	11.17	1.61	9.1–13.8	10.78	1.70	8.1–14.5	0.70	.501
Current surgery and recovery
No. of drains (n=40)	1.84	0.60	1–3	2.27	1.19	1–4	1.80	0.92	1–3	1.04	.363
No. of postoperative complications among 5 listed†	0.35	0.49	0–1	0.91	1.04	0–3	0.40	0.70	0–2	2.28	.116
Time on air-fluidized bed (d)	37.45	11.54	16–65	34.09	14.41	16–72	37.30	11.55	27–64	0.29	.751

	Mean (Mean Rank)	SD	Range	Mean (Mean Rank)	SD	Range	Mean (Mean Rank)	SD	Range	Kruskal-Wallis χ2	P
Time from surgery to first sitting‡ (wk) (n=39)	9.49(21.74)	8.90	6–44	8.03(18.80)	3.47	5–15	7.33(17.9)	2.75	5–15	0.93	.629
LOS (d)‡	126.85(21.81)	60.29	64–342	154.27(25.82)	81.95	72–370	105.70(15.35)	54.11	51–241	4.01	.135

	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	F	P
Pressure ulcer recurrence status
Time from discharge to recurrence or end of study (mo) (n=38)	19.61	7.15	2–24	10.10	9.15	1–24	10.30	8.38	1–24	6.52	.002

Table 2. Categoric Data by Group

	Group 1 Enhanced Education and Structured Follow-Up (n=20)		Group 2 Monthly Contact to Obtain Urine Sample and Assess Skin Condition (n=11)		Group 3 Quarterly Contact to Assess Skin Condition (n=10)
Variable	No.	Percentage	No.	Percentage	No.	Percentage	χ2	Exact P
Demographic and SCI/D data
Race/ethnicity							3.41	.594
White	11	55.0	5	45.5	4	40.0
Black	9	45.0	5	45.5	6	60.0
Hispanic	0	0.0	1	9.1	0	0.0
Marital status							5.27	.818
Never married	4	20.0	1	9.1	2	20.0
Married	7	35.0	3	27.3	5	50.0
Widowed	1	5.0	1	9.1	0	0.0
Divorced	8	40.0	5	45.5	3	30.0
Other	0	0.0	1	9.1	0	0.0
Educational status							5.84	.474
Less than high school	2	10.0	1	9.1	0	0.0
High school or GED	6	30.0	3	27.3	6	60.0
Some college	10	50.0	4	36.4	2	20.0
Bachelor's degree	2	10.0	3	27.3	2	20.0
Etiology of SCI/D							7.71	.737
Motor vehicle crash	7	35.0	5	45.5	5	50.0
Diving	1	5.0	1	9.1	0	0.0
Fall	3	15.0	0	0.0	0	0.0
Gunshot wound	3	15.0	2	18.2	2	20.0
Hit by falling object	0	0.0	1	9.1	0	0.0
Other	6	30.0	2	18.2	3	30.0
Level of SCI/D							9.63	.115
Cervical	7	35.0	2	18.2	7	70.0
Thoracic	10	50.0	9	81.8	3	30.0
Lumbar	1	5.0	0	0.0	0	0.0
MS	2	10.0	0	0.0	0	0.0
Level of SCI⁎							5.88	.061
Cervical	7	38.9	2	18.2	7	70.0
Noncervical	11	61.1	9	81.8	3	30.0
AIS grade							5.93	.449
A	13	65.0	9	81.8	6	60.0
B	5	25.0	2	18.2	3	30.0
D	0	0.0	0	0.0	1	10.0
NA (has MS)	2	10.0	0	0.0	0	0.0
Surgical closure history and current tobacco use
Previous surgical closure							1.69	.469
No	10	50.0	3	27.3	5	50.0
Yes	10	50.0	8	72.7	5	50.0
Previous surgical closure within past 5 years (n=40)†							1.32	.680
No	14	70.0	7	70.0	5	50.0
Yes	6	30.0	3	30.0	5	50.0
Current tobacco use							0.11	1.000
No	13	65.0	7	63.6	7	70.0
Yes	7	35.0	4	36.4	3	30.0
Current surgery and recovery
Location of study ulcer							9.10	.056
Trochanter	3	15.0	5	45.5	5	50.0
Ischium	13	65.0	6	54.5	2	20.0
Sacrum/coccyx	4	20.0	0	0.0	3	30.0
Heterotopic ossification							0.39	.837
No	13	65.0	8	72.7	6	60.0
Yes	7	35.0	3	27.3	4	40.0
Osteomyelitis (n=31)‡							3.04	.222
No	4	20.0	5	45.5	5	50.0
Yes	9	45.0	6	54.5	2	20.0
Type of surgery							11.09	.023
Fasciocutaneous flap only	2	10.0	4	36.4	6	60.0
Muscle flap only	15	75.0	4	36.4	2	20.0
Both types	3	15.0	3	27.3	2	20.0
Any complications of surgery							1.13	.653
No	13	65.0	5	45.5	6	60.0
Yes	7	35.0	6	54.5	4	40.0
Pressure ulcer recurrence status
Ulcer recurrence within 24mo (n=38)							8.47	.007
No	12	66.7	4	40.0	1	10.0
Yes	6	33.3	6	60.0	9	90.0

Abbreviations: GED, General Educational Development diploma; NA, not applicable.

Assignment Group Differences 

There were no significant differences among the 3 assignment groups in age, age at onset of SCI/D, or time since onset of SCI/D (see table 1). The groups also did not differ significantly with respect to race or ethnicity, marital status, educational status, etiology of SCI/D, and level or completeness of SCI/D (see table 2). With regard to level of SCI, after omitting the 2 participants with MS and recoding the levels into 2 categories—cervical and noncervical—by combining the thoracic and lumbar levels, group 3 had the highest percentage (70%) of persons with cervical injuries, and group 2 had the lowest percentage (18%), with group 1 falling in between (39%). These group differences met the generalized criterion for being statistically marginal (P=.061) (see table 2).

The 3 groups did not differ significantly with regard to number of previous health problems among the 11 selected items or number of previous surgical closures for pressure ulcers (see table 1). Among those who had had at least 1 previous surgical closure (n=23 [56% of sample]), time since last surgical closure was significantly different among the groups; however, time since last surgical closure was not known for 1 participant who had had a previous surgical closure. On average, time since surgical closure was over 4 years longer for participants in groups 1 and 2 than for those in group 3. In post hoc, Bonferroni-adjusted, pairwise comparisons, group 1 had a significantly longer time since last surgical closure than group 3 (P=.041), thus meeting the generalized criterion for significance.

The 3 groups did not differ significantly on whether they had a previous surgical closure, had a previous surgical closure in the past 5 years, or currently used tobacco (see table 2). They also did not differ significantly or marginally on whether they had a history of cardiac problems, pulmonary problems, gastrointestinal problems, musculoskeletal problems, peripheral vascular disease, neurologic problems other than SCI or MS, skin problems including burns but not pressure ulcers, renal disease, collagen disease, diabetes, or amputation (data not shown). Nor did they differ on whether they had spasticity or took medication for spasticity (not shown).

At admission, participants in group 3 rated their health as more than 2 points lower than participants in the other 2 groups, thus meeting the statistically marginal generalized criterion (P=.062). The 3 groups did not differ significantly with regard to self-assessed health at discharge (see table 1).

At admission, the groups were very nearly equal in knowledge about pressure ulcer prevention and management (see table 1).

The assignment groups did not differ significantly with regard to presurgery albumin or hemoglobin (see table 1).

The 3 groups did not differ significantly with respect to number of drains, number of postoperative complications, days on an air-fluidized bed, time from surgery to first sitting, and LOS (see table 1). They also did not vary on heterotopic ossification, osteomyelitis, or whether they had at least 1 of the selected complications after surgery. They did differ according to the generalized marginally significant criterion (P=.056) with respect to the location of the study ulcer. Group 1 had the highest rate of ischial ulcers, group 2 had no sacral or coccyx ulcers, and group 3 had the highest rate of trochanteric ulcers. The difference between groups met the generalized criterion for significance (P=.023) with respect to the type of surgical closure that was performed (see table 2). Group 1 had the highest rate of muscle flaps, group 2 had the highest rate of having both a muscle flap and a fasciocutaneous flap, and group 3 had the highest rate of fasciocutaneous flaps.

Participants in group 1 had a significantly longer time from discharge to recurrence or end of follow-up, whichever came first (1-tailed, generalized criterion for significance, P=.002) (see table 1). The mean number of months for group 1 was almost twice that of groups 2 and 3. Post hoc pairwise 1-tailed comparisons (Bonferroni-adjusted) indicated that according to the generalized criterion, group 1 was ulcer-free significantly longer than either group 2 (P=.008) or group 3 (P=.009), thus supporting the primary hypothesis. Differences between group 1 and groups 2 and 3 represent large effect sizes (Cohen d=1.16, d=1.20, respectively). Groups 2 and 3 did not differ significantly from each other. Effect sizes were similar when the 2 persons with MS and the 1 Hispanic person were omitted from the effect-size analyses (group 1 vs group 2, d=1.11; group 1 vs group 3, d=1.11). The groups also varied significantly (1-tailed, generalized criterion, P=.007) with regard to whether a recurrence occurred within 24 months after discharge (see table 2). One third of group 1, 60% of group 2, and 90% (ie, all but 1 participant) of group 3 had a recurrence during the study period.

Displayed in figure 2 are the survival curves for the 3 assignment groups (n=38). Overall group differences were significant (Wald χ²=8.05; 1-tailed, generalized criterion, P=.009), and assignment to group 1 was individually significant (Wald χ²=7.72; 1-tailed, generalized criterion, P=.003; OR=.228; 95% CI, .080−.647), indicating that participants who received the enhanced education and structured follow-up were ulcer-free longer than their counterparts in groups 2 and 3. Groups 2 and 3 were not significantly different from each other. When the analysis was repeated excluding the 2 persons with MS from group 1 and the 1 person who was Hispanic from group 2, the results were very similar (n=35; overall group: Wald χ²=6.43, P=.020; group 1: Wald χ²=6.28; 1-tailed generalized criterion, P=.006; OR=.264; 95% CI, .093−.748).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Cox regression analysis for effect of group assignment on ulcer-free survival time (n=38). Overall group assignment and assignment to group 1 are significant.

The variables that had significant (2-tailed, generalized criterion, P<.05) or marginal (2-tailed, generalized criterion, P<.10) group differences at baseline that were analyzed individually on an exploratory basis with Cox regression survival analysis were (1) years since last pressure ulcer surgery (for those who had had previous surgery), (2) self-assessed health status at admission, (3) level of SCI (cervical or noncervical, omitting those with MS), (4) location of study ulcer (trochanter, ischium, or sacrum/coccyx), and (5) type of surgery (fasciocutaneous flap only, muscle flap only, or both types).

Twenty-one of the 38 participants with follow-up data had had previous pressure ulcer surgery. Information on the length of time since the last surgery was missing for 1 person. For the other 20 persons, assignment group was not a significant predictor of the number of months they remained ulcer-free during the study (Wald χ²=1.40, P=.496) (fig 3). However, there was a marginally significant trend for years since last pressure ulcer surgery to predict ulcer-free survival (Wald χ²=3.49; 2-tailed, generalized criterion, P=.062; OR=.811; 95% CI, 0.652−1.010). Persons who had more years since their last surgery were ulcer-free longer during the current study. On the basis of these results, 2 separate additional exploratory analyses were performed to assess the effect of assignment group on pressure ulcer survival—one including only those who had had at least 1 previous pressure ulcer surgery (n=21) and another including only those who had never before had pressure ulcer surgery (n=17). For participants who had had previous surgery, group assignment did not predict ulcer-free survival (Wald χ²=5.937, P=.296) (fig 4). For participants who had had no prior ulcer surgery, there was a marginally significant trend for group assignment to predict survival time (Wald χ²=5.94; 2-tailed, generalized criterion, P=.051) (fig 5), and assignment to group 1 was individually significantly predictive (Wald χ²=4.62; 2-tailed, generalized criterion, P=.032; OR=.089; 95% CI, .010−.809). Persons in group 1 were ulcer-free longer than those in groups 2 and 3 among persons with no previous ulcer surgery.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Survival analysis of group assignment in a subsample with at least 1 previous pressure ulcer, controlling for time since last surgical closure (n=20). Group assignment was not significant, and there was a nonsignificant trend for time since last closure.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 4. 

  Survival analysis of group assignment in subsample with at least 1 previous pressure ulcer surgery (n=21). Neither overall group assignment nor assignment to group 1 was significant.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 5. 

  Survival analysis of group assignment in subsample with no previous pressure ulcer surgery (n=17). There was a nonsignificant trend for overall group assignment, and assignment to group 1 was significant.

Survival curves for the 3 assignment groups controlling for either self-assessed health status at admission or level of SCI (cervical or noncervical) were very similar to the curves shown in figure 1 in which group assignment was the only potential predictor (figures for health and level are not shown to conserve space). When controlling for health status, overall group assignment was significant (Wald χ²=6.76; 2-tailed, generalized criterion, P=.034), as was assignment to group 1 (Wald χ²=4.90; 2-tailed, generalized criterion, P=.027; OR=.270; 95% CI, .085−.086). Health status was not a significant predictor of staying ulcer-free (Wald χ²=1.84, P=.175). For level of SCI, both overall group assignment (Wald χ²=5.99; 2-tailed, generalized criterion, P=.050) and assignment to group 1 were significant (Wald χ²=5.54; 2-tailed, generalized criterion, P=.019; OR=.262; 95% CI, .086−.799), but level of SCI (Wald χ²=.005, P=.942) was not significantly predictive of survival.

In the analysis in which location of study ulcer was entered, overall group assignment (Wald χ²=12.56; 2-tailed, generalized criterion, P=.002) was a significant predictor of survival (fig 6). Assignment to group 1 was a significant predictor of survival, and this was the only finding in the entire study that met the experiment-wise criterion of P less than or equal to .001 for significance (Wald χ²=11.48; 2-tailed, experiment-wise criterion, P=.001; OR=.138; 95% CI, .044−.434). In addition, there was a marginally significant trend for overall location to predict survival (Wald χ²=5.65; 2-tailed, generalized criterion, P=.059), and having an ulcer on a trochanter was individually significantly predictive of survival (Wald χ²=5.40; 2-tailed, generalized criterion, P=.020; OR=.202; 95% CI, .052−.778). Repeating the same analysis but obtaining a graph showing the individual ulcer location survival curves rather than the curves for group assignments revealed that persons whose study ulcer was on the trochanter were ulcer-free longer than those with ulcers on the ischium, sacrum, or coccyx (fig 7).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 6. 

  Survival analysis of group assignment controlling for location of study ulcer (n=38). Overall group assignment and assignment to group 1 were significant; there was a nonsignificant trend for overall location of ulcer, and having an ulcer on a trochanter was significant.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 7. 

  Survival analysis of location of study ulcer controlling for group assignment (n=38). There was a nonsignificant trend for overall location of ulcer, and having an ulcer on a trochanter was significant; overall group assignment and assignment to group 1 were significant (same analysis as for fig 6).

Finally, in the survival analysis controlling for type of surgery, overall group was not significant (Wald χ²=4.36, P=.113), but assignment to group 1 was significantly predictive of survival (Wald χ²=4.19; 2-tailed, generalized criterion, P=.041; OR=.299; 95% CI, .094−.950). Type of surgery was not predictive (Wald χ²=1.89, P=.387) (figure not shown).

To summarize the results of the exploratory analyses, in the subsample who had had previous pressure ulcer surgery, when controlling for time since the last surgery, group assignment was not predictive of ulcer-free survival time; however, assignment to group 1 was predictive in the subsample who had never before had pressure ulcer surgery. When controlling for location of the study ulcer, both assignment to group 1 and location were predictive. In the 3 separate analyses controlling for self-assessed health status at admission, level of SCI, and type of surgery, assignment to group 1 was predictive but health status, level of SCI, and type of flap were not predictive of ulcer-free survival time.

Back to Article Outline

Discussion 

Effect of Education and Follow-Up Program on Recurrence 

Participants who received the education and structured follow-up (group 1) remained ulcer-free longer than the 2 control groups, and fewer of them had a recurrence within the 2-year follow-up period. Furthermore, comparing the 2 control groups, monthly contact with no educational content was no better than minimal quarterly contact. The effectiveness of individualized, long-term follow-up also was supported by a study in Australia.⁴⁷ Participants who received the individualized community outreach program had fewer hospital days a patient a month than 2 other groups who did not participate in the program. Maugham et al⁴⁷ advocated a proactive, individually tailored approach for patients at high risk for recurrent pressure ulcers. Cardenas et al⁵ reported high rates of rehospitalizations after acute rehabilitation and recommended close follow-up to reduce the risk of urinary tract infections, pressure ulcers, and pulmonary problems. These findings and recommendations are supportive of the chronic care model components of support for self-management and improvements in the health care delivery system. In the present study, the individualized education increased patient knowledge about pressure ulcers, and the frequent follow-up contacts provided encouragement and support.

When post hoc subset analyses were performed, for persons who did not have previous pressure ulcer surgery, the enhanced education and structured follow-up were effective in extending ulcer-free survival time, but for persons who had previous ulcer surgery, the intervention was not significantly effective. This suggests that persons having their first ulcer surgery should receive and are likely to benefit from an intervention such as the one described in this study. However, persons who have had previous ulcer surgery may need even more intensive education and follow-up. Furthermore, the more recent their last ulcer surgery was, the more help they are likely to need. Another interesting finding was that persons with pressure ulcers on the sacrum or coccyx are likely to have shorter ulcer-free survival times than those with ulcers on the ischium or trochanter. Therefore, persons who have their first ulcer surgery for ulcers on the sacrum or coccyx should probably also receive very intensive education and follow-up such as that suggested for those with previous ulcer surgeries.

Although cost issues were not specifically studied, it is very likely that the cost of an employee's time to conduct individualized educational sessions and monthly follow-up telephone calls would be much less than the cost for treating pressure ulcers that were prevented or delayed as a result of the program.⁴¹ Byrne and Salzberg⁴⁸ found that treating pressure ulcers in persons with cervical SCI was more expensive than for those with lower-level injuries. Their study was published in 1987, and the average costs at that time were $91,271 for cervical injuries, $56,181 for thoracic injuries, and $41,984 for lumbar injuries, resulting in an average of $69,587. Obviously, the costs will have escalated a great deal in the intervening 20 years. At the MEDVAMC, costs a day for hospitalization are estimated at approximately $1000, with an average stay of 60 to 90 days for persons undergoing pressure ulcer surgery. Thus, the cost a person would be $60,000 to $90,000.

The person who is to provide the individualized education during hospitalization and the structured telephone follow-up contacts should have expertise in both pressure ulcers and SCI/D (eg, an experienced nurse, occupational therapist, or physical therapist). Estimating salary and fringe support for .50 full-time equivalents for such a person are very likely to cost approximately $50,000 a year; thus, cost savings would result if only 2 pressure ulcer surgeries were prevented a year. Furthermore, more economic ways of providing the education and support necessary to prevent pressure ulcers may be found. These might include computer-aided instruction as described by Pellerito,³⁵ web-based education and follow-up after discharge, and use of telemedicine to monitor skin condition, all of which have become more practical and less costly since the current study was designed. Perhaps the intervention evaluated in the present study could be enhanced by also providing booster doses on the web or DVDs to which patients could refer after discharge. Future research could compare various combinations of methods of delivering the information to see which is most cost-effective in the long run.

Study Limitations 

The small sample size was the most significant limitation of this study. Five potential participants never had surgery and therefore were not randomized. Three were randomized but did not heal; thus, they could not begin the 2-year follow-up period. Furthermore, the number of potential candidates depended on many factors outside the control of the investigators. Some persons admitted for pressure ulcers were not eligible for flap surgery for various medical reasons. Randomization was accomplished by having each participant take a numbered paper out of a container. A more scientific method would have been to use a table of random numbers to predetermine group assignment. What effects the randomization method had on the results are unknown.

Despite randomization, because of the relatively small sample size, the groups were not equal with regard to time since last surgical closure for those who had had a previous surgical closure, self-assessed health at admission, level of SCI/D, location of study ulcer, and the type of surgical closure done during the current hospitalization. Because of the small sample size, multiple individual exploratory analyses were performed to assess the impact of these potential confounders rather than including them in 1 overall analysis as had originally been planned. Thus, the likelihood of chance findings is high. With larger sample sizes, some of these differences may be better controlled through randomization. In addition, in future studies, some of these characteristics might be controlled through stratification (eg, previous ulcer surgery or not) prior to group assignment. Simultaneous control of the potential confounders should be a goal of future research in this area.

Another possible limitation was the inclusion of 2 persons with MS who were, by chance, both assigned to group 1, and only 1 person who was Hispanic who was assigned to group 2. However, testing the primary hypothesis without these 3 persons yielded essentially the same result: group 1 had significantly longer ulcer-free survival time than did groups 2 and 3.

We relied on self-reports of skin breakdown because geographic distance and study funding did not allow us to verify the breakdown in person. Self-reports are vulnerable to several problems. Participants may not know the actual status of their skin because they cannot independently view it. They may also knowingly give incorrect information (eg, not report a pressure ulcer when they have one or report frequently doing pressure ulcer prevention behaviors when they actually do not) because they want to please the investigators or are embarrassed to report the true information.

At the time this study was conducted, prealbumin data were not routinely recorded and hence were not available for all patients. In future studies, it would be interesting to investigate the relation between prealbumin levels just prior to surgery and pressure ulcer recurrence. Compared with albumin, prealbumin has the advantage of reflecting nutritional status within the past week rather than the past several months. Results of other laboratory tests such as hemoglobin A_1C, blood glucose levels, or serum blood urea nitrogen and creatinine might also be included in future studies.

There may have been diffusion of the educational material from participants in group 1 to participants in groups 2 and 3. However, if this were the case, it should make it more difficult to find significant differences among the 3 groups in ulcer-free survival rates. Because differences were found in spite of any diffusion that may have occurred, the conclusions are still reasonable on the basis of the findings. Finally, occasionally there were missing data for some study variables.

Back to Article Outline

Conclusions 

Individualized education and structured follow-up was effective in reducing recurrence in the participants in this study. However, because of the limitations listed, this conclusion must be considered tentative. Nevertheless, the study provides valuable insights with regard to designing future studies addressing prevention of pressure ulcer recurrence. Possible confounders have been identified and can thus be controlled by randomization of larger sample size and/or stratification prior to randomization. If the effect of the intervention proves to be reliable in the future, then establishing intensive, individualized education and follow-up programs for persons with pressure ulcers may reduce hospitalizations and costly surgery for these patients. However, larger studies are needed to cross-validate this study's findings. Patients who have a recent history of pressure ulcer surgery and persons with pressure ulcers on the sacrum or coccyx are at particularly high risk for recurrence. They should be warned of their increased risk, monitored very closely, and provided with more training in pressure ulcer prevention strategies even beyond the enhanced education and follow-up program evaluated in this study.

Suppliers

Back to Article Outline

Appendix 1. Content of Educational Model 

Hour 1

Hour 2

Hour 3

Hour 4

Back to Article Outline

References 

1. Krause JS, Broderick L. Patterns of recurrent pressure ulcers after spinal cord injury: identification of risk and protective factors 5 or more years after onset. Arch Phys Med Rehabil. 2004;85:1257–1264
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (97 KB)
   • CrossRef
2. Langemo DK, Melland H, Hanson D, Olson B, Hunter S. The lived experience of having a pressure ulcer: a qualitative analysis. Adv Skin Wound Care. 2000;13:225–235
   • View In Article
   • MEDLINE
3. Garber SL, Rintala DH. Pressure ulcers in veterans with spinal cord injury: a retrospective study. J Rehabil Res Dev. 2003;40:433–442
   • View In Article
   • MEDLINE
   • CrossRef
4. Krause JS. Skin sores after spinal cord injury: relationship to life adjustment. Spinal Cord. 1998;36:51–56
   • View In Article
   • MEDLINE
5. Cardenas DD, Hoffman JM, Kirshblum S, McKinley W. Etiology and incidence of rehospitalization after traumatic spinal cord injury: a multicenter analysis. Arch Phys Med Rehabil. 2004;85:1757–1763
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (128 KB)
   • CrossRef
6. Niazi ZB, Salzberg CA, Byrne DW, Viehbeck M. Recurrence of initial pressure ulcer in persons with spinal cord injuries. Adv Wound Care. 1997;10:38–42
   • View In Article
   • MEDLINE
7. Charlifue S, Lammertse DP, Adkins RH. Aging with spinal cord injury: changes in selected health indices and life satisfaction. Arch Phys Med Rehabil. 2004;85:1848–1853
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (89 KB)
   • CrossRef
8. Wyllie FJ, McLean NR, McGregor JC. The problem of pressure sores in a regional plastic surgery unit. J R Coll Surg Edinb. 1984;29:38–43
   • View In Article
   • MEDLINE
9. Mandrekas AD, Mastorakos DP. Management of decubitus ulcers by musculocutaneous flaps: a five-year experience. Ann Plast Surg. 1992;28:167–174
   • View In Article
   • MEDLINE
   • CrossRef
10. Berry RB. The late results of surgical treatment of pressure sores in paraplegics. Br J Surg. 1980;67:473–474
    • View In Article
    • MEDLINE
    • CrossRef
11. Relander M, Palmer B. Recurrence of surgically treated pressure sores. Scand J Plast Reconstr Surg. 1988;22:89–92
    • View In Article
12. Disa JJ, Carlton JM, Goldberg NH. Efficacy of operative cure in pressure sore patients. Reconstr Surg. 1992;89:272–278
    • View In Article
13. Hentz VR. Management of pressure sores in a specialty center: a reappraisal. Plast Reconstr Surg. 1979;64:683–691
    • View In Article
    • MEDLINE
    • CrossRef
14. Pers M. Plastic surgery for pressure sores. Paraplegia. 1987;25:275–278
    • View In Article
    • MEDLINE
    • CrossRef
15. Romm S, Tebbetts J, Lynch D, White R. Pressure sores: state of the art. Tex Med. 1982;78:52–60
    • View In Article
    • MEDLINE
16. Buntine JA, Johnstone BR. The contributions of plastic surgery to care of the spinal cord injured patient. Paraplegia. 1988;27:87–93
    • View In Article
17. Scheflan M. Surgical methods for managing ischial pressure wounds. Ann Plast Surg. 1982;8:238–247
    • View In Article
    • MEDLINE
    • CrossRef
18. Lee ET. A new wound closure achieving and maintaining device using serial tightening of loop suture and its clinical applications in 15 consecutive patients for up to 102 days. Ann Plast Surg. 2004;53:436–441
    • View In Article
    • MEDLINE
    • CrossRef
19. Akyurek M, Safak T, Sonmez E, Ozkan O, Kecik A. A new flap design: neural-island flap. Plast Reconstr Surg. 2004;114:1467–1477
    • View In Article
20. Ichioka S, Okabe K, Tsuji S, Ohura N, Nakatsuka T. Distal perforator-based fasciocutaneous V-Y flap for treatment of sacral pressure ulcers. Plast Reconstr Surg. 2004;114:906–909
    • View In Article
    • MEDLINE
    • CrossRef
21. Ichioka S, Okabe K, Tsuji S, Ohura N, Nakatsuka T. Triple coverage of ischial ulcers with adipofascial turnover and fasciocutaneous flaps. Plast Reconstr Surg. 2004;114:901–905
    • View In Article
    • MEDLINE
    • CrossRef
22. Lin MT, Chang KP, Lin SD, Lai CS, Yang YL. Tensor fasciae latae combined with tangentially split vastus lateralis musculocutaneous flap for the reconstruction of pressure sores. Ann Plast Surg. 2004;53:343–347
    • View In Article
    • MEDLINE
    • CrossRef
23. Goodman CM, Cohen V, Armenta A, Thornby J, Netscher DT. Evaluation of results and treatment variables for pressure ulcers in 48 veteran spinal cord-injured patients. Ann Plast Surg. 1999;42:665–672
    • View In Article
    • MEDLINE
    • CrossRef
24. Schryvers OI, Stranc MF, Nance PW. Surgical treatment of pressure ulcers: 20-year experience. Arch Phys Med Rehabil. 2000;81:1556–1562
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (262 KB)
    • CrossRef
25. Sorensen JL, Jorgensen B, Gottrup F. Surgical treatment of pressure ulcers. Am J Surg. 2004;188:42–51
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (454 KB)
    • CrossRef
26. Rubayi S, Pompan D, Garland D. Proximal femoral resection and myocutaneous flap for treatment of pressure ulcers in spinal cord injury patients. Ann Plast Surg. 1983;27:132–137
    • View In Article
    • MEDLINE
    • CrossRef
27. Baek SM, Williams GD, McElhinney AJ, Simon BE. The gluteus maximus myocutaneous flap in the management of pressure sores. Ann Plast Surg. 1980;5:471–476
    • View In Article
    • MEDLINE
    • CrossRef
28. Yasenchak PA, Lynch KB, Bridle MJ, Wegener ST. Variables related to severe pressure sore recurrence [abstract]. J Am Paraplegia Soc. 1990;13
    • View In Article
29. Epping-Jordan JE, Pruitt SD, Bengoa R, Wagner EH. Improving the quality of health care for chronic conditions. Qual Saf Health Care. 2004;13:299–305
    • View In Article
    • MEDLINE
    • CrossRef
30. Wagner EH, Austin BT, Davis C, Hindmarsh M, Schaefer J, Bonomi A. Improving chronic illness care: translating evidence into action. Health Aff (Millwood). 2001;20:64–78
    • View In Article
    • MEDLINE
    • CrossRef
31. Chapman K, McGinnis-Rake C, O'Halloran W, Paul S, Virden JA. Convincing the non-compliant patient to change his behavior. Ostomy Wound Manag. 1991;35:45–51
    • View In Article
32. Green LW. The theory of participation: a qualitative analysis of its expression in national and international policies. Adv Health Promot Dis Prev. 1986;1:211–236
    • View In Article
33. Andberg MM, Rudolph A, Anderson TP. Improving skin care through patient and family training. Top Clin Nurs. 1983;5:45–54
    • View In Article
    • MEDLINE
34. Garber SL, Rintala DH, Holmes SA, Rodriguez GP, Friedman J. A structured educational model to improve pressure ulcer prevention knowledge in veterans with spinal cord dysfunction. J Rehabil Res Dev. 2002;39:575–588
    • View In Article
35. Pellerito JM. The effects of traditional and computer-aided instruction on promoting independent skin care in adults with paraplegia. Occup Ther Int. 2003;10:1–19
    • View In Article
    • MEDLINE
    • CrossRef
36. Rodriguez GP, Garber SL. Prospective study of pressure ulcer risk in spinal cord injury patients. Paraplegia. 1994;32:150–158
    • View In Article
    • MEDLINE
    • CrossRef
37. Krouskop TA, Noble PC, Garber SL, Spencer WA. Effectiveness of preventive management in reducing the occurrence of pressure sores. J Rehabil Res Dev. 1983;20:74–83
    • View In Article
38. Jones ML, Mathewson CS, Adkins VK, Ayllon T. Use of behavioral contingencies to promote prevention of recurrent pressure ulcers. Arch Phys Med Rehabil. 2003;84:796–802
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (116 KB)
    • CrossRef
39. Hammond MC, Umlauf RL, Matteson B, Perduta-Fulginiti S. Yes, you can! (A guide to self-care for persons with spinal cord injury). 2nd ed.. Washington (DC): Paralyzed Veterans of America; 1989;
    • View In Article
40. Ackerman J, Burnett M, Clark B, Garber S, Probst DA. A personal guide to healthy skin: reducing the risks of developing pressure sores in persons with spinal cord injuries. Houston: Institute for Rehabilitation and Research and Rehabilitation Research and Training Center on Community-Oriented Services for Persons with Spinal Cord Injury; 1990;
    • View In Article
41. Panel for the Prediction and Prevention of Pressure Ulcers in Adults. Preventing pressure ulcers: patient guide (Clinical practice guideline no. 3). Washington (DC): US Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research; 1992;AHCPR publication no. 92-0048
    • View In Article
42. American Spinal Injury Association. International standards for neurological and functional classification of spinal cord injury. 5th ed. Chicago: ASIA; 1996;
    • View In Article
43. Basta SM. Pressure sore prevention education with the spinal cord injured. Rehabil Nurs. 1991;16:6–8
    • View In Article
    • MEDLINE
44. LaMantia JG, Hirschwald JF, Goodman CL, Wooden VM, Delisser O, Staas WE. A program design to reduce chronic readmissions for pressure sores. Rehabil Nurs. 1987;12:22–2516
    • View In Article
    • MEDLINE
45. Bergstrom N, Braden B. A prospective study of pressure sore risk among institutionalized elderly. J Am Geriatr Soc. 1992;40:747–758
    • View In Article
    • MEDLINE
46. Garber SL, Rintala DH, Rossi CD, Hart KA, Fuhrer MJ. Reported pressure ulcer prevention and management techniques by persons with spinal cord injury. Arch Phys Med Rehabil. 1996;77:744–749
    • View In Article
    • Abstract
    • Full-Text PDF (885 KB)
    • CrossRef
47. Maugham L, Cox R, Amsters D, Battistutta D. Reducing inpatient hospital usage for management of pressure sores after spinal cord lesions. Int J Rehabil Res. 2004;27:311–315
    • View In Article
    • MEDLINE
    • CrossRef
48. Byrne DW, Salzberg CA. Major risk factors for pressure ulcers in the spinal cord disabled: a literature review. Spinal Cord. 1996;34:255–263
    • View In Article
    • MEDLINE
    • CrossRef

• a Version 15.0; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.
• b Version 8; Cytel Inc, 675 Massachusetts Ave, Cambridge, MA 02139.