Home > Discussion - Pressure Ulcer Risk...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Chou R, Dana T, Bougatsos C, et al. Pressure Ulcer Risk Assessment and Prevention: Comparative Effectiveness [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2013 May. (Comparative Effectiveness Reviews, No. 87.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.



Table 17 summarizes the findings of this review. Details about the factors used to determine the overall strength of evidence for each key question are shown in Appendix F.

Table 17. Summary of evidence.

Table 17

Summary of evidence.

Evidence on optimal methods to prevent pressure ulcers was extremely limited in a number of areas, including the effects of use of risk assessment instruments on the subsequent incidence of pressure ulcers and benefits of preventive interventions other than support surfaces. Evidence on harms of preventive interventions was extremely sparse, with most trials not reporting harms at all, and poor reporting of harms in those that did. Nonetheless, serious harms seem rare, consistent with what might be expected given the generally noninvasive nature of most of the preventive interventions evaluated (skin care, oral nutritional support, repositioning, and support surfaces). In addition, limited evidence was available to evaluate how the diagnostic accuracy of risk assessment instruments or benefits and harms of preventive interventions might vary depending on differences in setting, patient characteristics, or other factors. Very few studies directly assessed how estimates varied according to these factors, and indirect comparisons across trials were not possible due to small numbers of studies, differences in interventions and comparisons, and methodological shortcomings.

Only one good- and two poor-quality studies13,45,46 attempted to evaluate the effects of standardized use of a risk assessment instrument on the incidence of pressure ulcers. The good-quality trial found no difference in risk of pressure ulcers or length of stay in patients assessed with the Waterlow scale, the Ramstadius tool, or clinical judgment alone.13 The two poor-quality studies evaluated the modified Norton scale45 and the Braden scale,46 with only a nonrandomized study of the Norton scale45 finding reduced risk of pressure ulcer compared with clinical judgment.13,45,46

Studies of diagnostic accuracy found that commonly used risk assessment instruments (such as the Braden, Norton, and Waterlow scales) can help identify patients at increased risk for ulcers who might benefit from more intense or targeted interventions, but appear to be relatively weak predictors, based on likelihood ratios at commonly used cutoffs. However, diagnostic accuracy may have been underestimated in these studies if patients at higher risk were more likely to receive effective interventions to prevent ulcers. Studies of diagnostic accuracy rarely reported risk estimates, and no study that reported risk estimates attempted to control for potential confounding effects of differential use of interventions. There was no clear difference between commonly used risk assessment instruments in diagnostic accuracy, though direct comparisons were limited.20,21,25,41,70,73

About three-quarters of the trials of preventive interventions focused on evaluations of support surfaces. In higher-risk populations, good- and fair-quality randomized trials consistently found more advanced static mattresses and overlays associated with lower risk of pressure ulcers compared with standard mattresses in higher-risk patients (relative risk [RR] range 0.20 to 0.60),100,107,113,116,124 with no clear differences between different advanced static support surfaces.88,92,101,103,108,110,111,118,119,126 Although the mattresses and overlays evaluated in the trials varied, three trials consistently found an Australian medical sheepskin overlay associated with lower risk of ulcers than a standard hospital mattress, though the sheepskin was also associated with heat-related discomfort, in some cases resulting in withdrawal.107,112,113 Evidence on the effectiveness and comparative effectiveness of other specific support surfaces, including alternating air mattresses and low-air-low mattresses, was limited, with most trials85,87,89,93,106,118,125 showing no clear differences between these types of mattresses and various static mattresses and overlays. One fair-quality trial found stepped care starting with alternating air mattresses associated with substantially decreased risk of ulcers compared with stepped care primarily with static mattresses,96 suggesting that this might be both an effective as well as efficient approach, since care was initiated with the least expensive alternatives and advanced to more expensive alternatives based on a preset algorithm. In lower-risk populations of patients undergoing surgery, two trials found use of a foam overlay associated with an increased risk of pressure ulcers compared with a standard operating room mattress.129,133 The few trials that evaluated length of stay found no differences between various support surfaces.104107,118,121,122

Evidence on other preventive interventions (nutritional supplementation; repositioning; pads and dressings; lotions, creams, and cleansers; and intraoperative warming therapy for patients undergoing surgery) was sparse and insufficient to reach reliable conclusions, in part because most trials had important methodological shortcomings. An exception was repositioning, for which there were three good- or fair-quality trials, though these reported somewhat inconsistent results.142,143,146 One trial found a repositioning intervention was more effective than usual care in preventing pressure ulcers.143 Although other trials of repositioning did not clearly find decreased risk of pressure ulcers compared with usual care, the usual care control group incorporated standard repositioning practices (i.e., the trials compared more intense repositioning with usual repositioning, not vs. no repositioning). A recently completed trial of repositioning, consisting of high-risk and moderate-risk arms that are randomized to repositioning at 2-, 3-, or 4-hour intervals, should provide more rigorous evidence on the effectiveness of repositioning.161

Findings in Relationship to What Is Already Known

Our findings of limited evidence on effects of risk assessment tools in reducing the incidence or severity of pressure ulcers are consistent with other recent systematic reviews.162,163 One of these reviews also evaluated the diagnostic accuracy of risk assessment tools.163 It reported higher sensitivity and lower specificity for the Waterlow (0.82 and 0.27) compared with the Norton (0.47 and 0.62) and Braden (0.57 and 0.68) scales, but pooled data without regard for differences in cutoff scores and across study settings, and included four studies that we excluded due to retrospective design,164 inadequate details to determine eligibility for inclusion,165 availability only in Spanish,166 or that we were unable to obtain.167

Our findings on effectiveness of preventive interventions are generally consistent with other systematic reviews that found some evidence that more advanced static support surfaces are associated with decreased risk of pressure ulcers compared with standard hospital mattresses,168,169 limited evidence on the effectiveness and comparative effectiveness of dynamic support surfaces,168,169 and limited evidence on other preventive interventions.169,170 All reviews noted methodological shortcomings in the trials and variability in interventions and comparisons across studies. These reviews differed from ours by including trials that enrolled patients with higher stage pre-existing ulcers and including trials published only as abstracts.


The studies included in this review generally enrolled patients at higher risk for pressure ulcers, though eligibility criteria varied between studies. The studies are most applicable to acute care and long-term care settings, with few studies evaluating patients in community or home settings, including specific populations such as wheelchair bound people in the community, patients at end of life, and spinal cord injury patients. Some trials specifically evaluated lower risk patients undergoing surgery and were reviewed separately (see Key Question 3a). Although black patients and Hispanics represent the fastest growing populations of frail elderly in the United States, these populations were severely underrepresented in the studies.171

Some interventions evaluated in older trials may no longer be available, and the control interventions (e.g., standard hospital mattresses) have also changed over time. However, conclusions were unchanged when analyses were restricted to trials conducted more recently. In addition, many trials of support surfaces evaluated specific brand name products and it might be difficult to generalize results to other products in the same class. This problem is compounded by the constantly changing nature of products sold and marketed by support surface manufacturers.

Another important issue in interpreting the applicability of this review is that patients in studies of diagnostic accuracy as well as in studies of interventions generally received standard of care treatments. For example, no study of diagnostic accuracy blinded caregivers to the results of risk assessment scores (blinding might be difficult for ethical reasons), which would be expected to lead to the use of more intensive preventive interventions and care in higher-risk people. If such interventions are truly effective, they would be expected to result in decreased incidence of pressure ulcers, thus affecting estimates of diagnostic accuracy. For trials of preventive interventions, usual care varied and was not always well described, but generally includes repositioning every 2 to 4 hours, skin care, standard nutrition, and standard support surfaces. Therefore, most trials of preventive interventions represent comparisons of more intensive interventions plus multi-component standard care compared with standard care alone, rather than compared with no care. One factor that may affect applicability is that the more intensive preventive interventions evaluated in many of the studies included in this review may require additional training or resources. In addition, the applicability of trial findings to clinical practice could be limited by delays in use of preventive interventions or differences in the quality of care between research and typical clinical settings.

Evidence to evaluate potential differences in comparative benefits or harms in patient subgroups based on baseline pressure ulcer risk, specific risk factors for ulcers, setting of care, and other factors was very limited, which precluded any reliable conclusions.

Although the studies included in this review generally enrolled patients in whom pressure ulcer risk assessment and preventive interventions would typically be considered and evaluated clinically relevant usual care comparison arms, they frequently did not meet a number of other criteria for effectiveness studies, such as assessment of adverse events, adequate sample sizes to detect clinically important differences, and use of intention-to-treat analysis.172

The results of this CER are not applicable to populations excluded from the review, including patients with higher stage pressure ulcers at baseline, or patients with lower stage pressure ulcers in which the risk of incident ulcers was not reported. A separate CER focuses on treatment of patients with pressure ulcers at baseline.28

Implications for Clinical and Policy Decisionmaking

Our review has potential implications for clinical and policy decisionmaking. Despite insufficient evidence to determine whether use of risk assessment instruments reduces risk of incident pressure ulcers, studies suggest that: a) commonly used instruments can predict which patients are more likely to develop an ulcer, and b) there are no clear differences in diagnostic accuracy. Decisions about whether to use risk assessment instruments and which risk assessment instrument to use may depend on considerations such as a desire to standardize and monitor practices within a clinical setting, ease of use, nursing preferences, and other factors. In some populations, such as spinal cord injured patients, risk assessment instruments have not been well studied, but may not be highly relevant since all patients may be considered to be at risk.

Evidence suggests that more advanced static support surfaces are more effective than standard mattresses for reducing risk of pressure ulcers, though more evidence is needed to understand the effectiveness and comparative effectiveness of dynamic and other support surfaces. Despite limited evidence showing that they are more effective at preventing pressure ulcers compared with static mattresses and overlays, alternating air and low-air-loss mattresses and overlays are used in hospitals in many areas of the United States. Such support surfaces can be quite costly, though one trial found that a stepped care approach that utilized lower-cost dynamic support surfaces before switching to higher-cost interventions in patients with early ulcers could be effective as well as efficient; this finding warrants further study.96 Although evidence is insufficient to guide recommendations on use of other preventive interventions, these findings are contingent on an understanding that usual care practices were the comparator treatment in most studies. Therefore, it would be inappropriate to conclude that standard repositioning, skin care, nutrition, and other practices should be abandoned, as these were the basis of usual care comparisons.

Although studies of preventive interventions primarily focused on effects on pressure ulcer incidence and severity, other factors such as effects on resource utilization (including length of hospitalization and costs) and patient preferences may affect clinical decisions. However, cost and patient preferences were outside the scope of this report and data on resource utilization was limited to a small numbers of studies that found no effects of various support surfaces on length of stay.

Limitations of the Comparative Effectiveness Review Process

Our review had some potential limitations. We excluded non-English language articles which could result in language bias (Appendix E), though a recent systematic review found little empirical evidence that exclusion of non-English language articles leads to biased estimates for noncomplementary or alternative medicine interventions.173 In addition, we did not exclude poor-quality studies a priori. Rather, we described the limitations of the studies, emphasized higher-quality studies when synthesizing the evidence, and performed sensitivity analyses that excluded poor-quality studies.

We did not attempt to pool studies of diagnostic accuracy due to clinical heterogeneity across studies and methodological shortcomings. Rather, we synthesized results qualitatively, and described the range of results, in order to highlight the greater uncertainty in findings.

We did not formally assess for publication bias with funnel plots due to small numbers (<10) of studies for all comparisons and due to important clinical heterogeneity and methodological shortcomings in the available studies. Small numbers of studies can make interpretation of funnel plots unreliable, and experts suggest 10 studies as the minimum number of studies to perform funnel plots.174 Inclusion of two studies of preventive interventions published only as conference abstracts would not have changed our results.134,175

Limitations of the Evidence Base

We identified a number of limitations in the evidence base on preventive interventions. Most included studies had important methodological shortcomings, with 4 of 47 studies of diagnostic accuracy and 35 of 72 studies of preventive interventions rated poor-quality, and only 12 studies of diagnostic accuracy and six studies of preventive interventions rated good-quality. Few studies of diagnostic accuracy reported measures of discrimination such as the AUROC, many studies failed to predefine cutoff thresholds, few studies reported differential use of interventions according to baseline risk score (which could affect estimates of diagnostic accuracy), and some studies evaluated modified or ad hoc versions of standard risk assessment instruments. An important limitation of the evidence on preventive interventions is that few trials compared the same intervention, and methods for assessing and reporting ulcers varied. There was almost no evidence to determine how diagnostic accuracy of risk assessment instruments or the effectiveness and comparative effectiveness of preventive interventions varies according to care setting, patient characteristics, or other factors. Harms were reported in only 16 of 72 trials of preventive interventions, and harms were poorly reported even when some data were provided. Only about half of the studies reported funding source. Among those that did report funding source, most were sponsored by institutions or governmental organizations.

Future Research

Future research is needed on the effectiveness of standardized use of risk assessment tools compared with clinical judgment or nonstandardized use in preventing pressure ulcers. Studies should evaluate validated risk assessment instruments and employ a clearly described protocol for use of preventive interventions based on the risk assessment score. In addition to comparing the risk and severity of ulcers across groups, studies should also report effects on use of preventive interventions as well as other important outcomes, such as length of hospital stay and measures of resource utilization.

Future research that simultaneously evaluates the diagnostic accuracy of different risk assessment instruments is needed to provide more direct evidence on how their performance compares with one another. Studies should at a minimum report how use of preventive interventions differed across intervention groups, and consider reporting adjusted risk estimates to account for such potential confounders. Studies of diagnostic accuracy should also use predefined, standardized cutoffs and routinely report measures of discrimination such as the AUROC. Research is also needed to understand how the different components of risk assessment instruments contribute to predictive utility, and on whether the addition of aspects not addressed well in standard risk assessments (such as decreased perfusion) improves diagnostic accuracy, in order to refine prediction instruments. Studies are also needed to understand how risk prediction instruments perform in specific patient populations and settings and whether the diagnostic accuracy of risk prediction instruments varies for different types of ulcers (e.g., heel ulcers vs. sacral or other ulcers).

More research is needed to understand the effectiveness of preventive interventions. It is critical that future studies of preventive interventions adhere to methodological standards including appropriate use of blinding (such as blinding of outcome assessors even when blinding of patients and caregivers is not feasible) and clearly describe usual care and other comparison treatments. Studies should routinely report baseline pressure ulcer risk in enrolled patients and consider predefined subgroup analyses to help better understand how preventive interventions might be optimally targeted. More studies are needed to better understand the comparative effectiveness of dynamic and reactive support surfaces compared with static support surfaces, as well as strategies such as stepped care that might be more efficient than using costly interventions in all patients.

Cover of Pressure Ulcer Risk Assessment and Prevention: Comparative Effectiveness
Pressure Ulcer Risk Assessment and Prevention: Comparative Effectiveness [Internet].
Comparative Effectiveness Reviews, No. 87.
Chou R, Dana T, Bougatsos C, et al.

AHRQ (US Agency for Healthcare Research and Quality)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...