Friction and shear highly associated with pressure ulcers of residents in long-term care - Classification Tree Analysis (CHAID) of Braden items

Leveraging artificial intelligence and decision support systems in hospital-acquired pressure injuries prediction: A comprehensive review

BACKGROUND

Hospital-acquired pressure injuries (HAPIs) constitute a significant challenge harming thousands of people worldwide yearly. While various tools and methods are used to identify pressure injuries, artificial intelligence (AI) and decision support systems (DSS) can help to reduce HAPIs risks by proactively identifying patients at risk and preventing them before harming patients.

OBJECTIVE

This paper comprehensively reviews AI and DSS applications for HAPIs prediction using Electronic Health Records (EHR), including a systematic literature review and bibliometric analysis.

METHODS

A systematic literature review was conducted through PRISMA and bibliometric analysis. In February 2023, the search was performed using four electronic databases: SCOPIS, PubMed, EBSCO, and PMCID. Articles on using AI and DSS in the management of PIs were included.

RESULTS

The search approach yielded 319 articles, 39 of which have been included and classified into 27 AI-related and 12 DSS-related categories. The years of publication varied from 2006 to 2023, with 40% of the studies taking place in the US. Most studies focused on using AI algorithms or DSS for HAPIs prediction in inpatient units using various types of data such as electronic health records, PI assessment scales, and expert knowledge-based and environmental data to identify the risk factors associated with HAPIs development.

CONCLUSIONS

There is insufficient evidence in the existing literature concerning the real impact of AI or DSS on making decisions for HAPIs treatment or prevention. Most studies reviewed are solely hypothetical and retrospective prediction models, with no actual application in healthcare settings. The accuracy rates, prediction results, and intervention procedures suggested based on the prediction, on the other hand, should inspire researchers to combine both approaches with larger-scale data to bring a new venue for HAPIs prevention and to investigate and adopt the suggested solutions to the existing gaps in AI and DSS prediction methods.

Impact of pressure ulcer prevention knowledge and attitude on the care performance of long-term care facility care workers: a cross-sectional multicenter study

BACKGROUND

The Long-Term Care Insurance Act in the Republic of Korea has enabled the elderly population to receive benefits through the long-term care system since July 2008. Because one nurse or nursing assistant is assigned to 25 elderly persons and one care worker is assigned to 2.5 elderly persons in long-term care facilities, registered nurses should educate care workers to participate in pressure ulcer prevention activities. This descriptive study investigated the effect of the knowledge and attitude related to pressure ulcer prevention on care performance.

METHODS

Data were collected from February 20 to December 15, 2021 using a structured questionnaire targeting 165 care workers in four long-term care facilities located in I-city and Y-gun, Gyeongsangbuk-do. The questionnaires of the knowledge, attitude, and care performance developed for nurses were modified to survey the care workers. The content validity was verified on a 4-point scale by 10 clinical experts. A preliminary survey was conducted for 30 care workers, and the contents of the modified questionnaire were further revised. Data analyses were performed by t-test, one-way ANOVA, Scheffé test, Pearson's correlation coefficient, and a multiple linear regression modeling using the SPSS/WIN 25.0 program.

RESULTS

Care performance on pressure ulcer prevention had a significant positive correlation with knowledge (r = 0.692, p < .001), attitude (r = 0.426, p < .001), work experience (r = 0.760, p < .001), amount of experience (r = 0.712, p < .001), and the number of training sessions received (r = 0.551, p < .001). In multiple regression modeling, work experience (β [standardized coefficient beta] = 0.534, p = .000), knowledge (β = 0.323, p = .000), and attitude (β = 0.103, p = .049) related to pressure ulcer prevention were identified as variables significantly affecting care performance. The regression model explained 65.4% with 5 independent variables.

CONCLUSION

To prevent bedsores in long-term care facilities, it is necessary to educate care workers regularly about pressure ulcer prevention. In addition, clinical guidelines could help standardize the pressure ulcer prevention work of caregivers, strongly regulating their practice in all long-term care facilities and monitoring bedsore prevention regularly.

Perceived unmet needs for community-based long-term care services among urban older adults: A cross sectional study

The purpose of this study was to explore the perceived unmet needs for community-based long-term care services among older urban adults in China. We analyzed the cross-sectional data of 5,201 urban community respondents ≥65 years of age from the seventh wave of the 2018 Chinese Longitudinal Healthy Longevity Survey (CLHLS). The chi-squared automatic interaction detection technique was used to examine the variables associated with older adults' unmet needs for four common types of community-based services: personal care, grocery shopping, home visits, and psychological consulting. We found that the majority of the older adults perceived that they needed the four services, but only 9%-27.4% of the respondents reported that their perceived needs were met. There was a high prevalence of unmet community-based service needs (51.3%-55.5%) among urban older adults in China. Factors associated with unmet needs included depression status, ADL (activities of daily living) limitations, self-rated health, number of surviving children, educational attainment, and marital status. The results suggest that policy makers should develop services targeting specific segments of the older population, increasing the adequacy of services provided.

Constructing Inpatient Pressure Injury Prediction Models Using Machine Learning Techniques

The incidence rate of pressure injury is a critical nursing quality indicator in clinic care; consequently, factors causing pressure injury are diverse and complex. The early prevention of pressure injury and monitoring of these complex high-risk factors are critical to reduce the patients' pain, prevent further surgical treatment, avoid prolonged hospital stay, decrease the risk of wound infection, and lower associated medical costs and expenses. Although a number of risk assessment scales of pressure injury have been adopted in various countries, their criteria are set for specific populations, which may not be suitable for the medical care systems of other countries. This study constructs three prediction models of inpatient pressure injury using machine learning techniques, including decision tree, logistic regression, and random forest. A total of 11 838 inpatient records were collected, and 30 sets of training samples were adopted for data analysis in the experiment. The experimental results and evaluations of the models suggest that the prediction model built using random forest had most favorable classification performance of 0.845. The critical risk factors for pressure injury identified in this study were skin integrity, systolic blood pressure, expression ability, capillary refill time, and level of consciousness.

Predictive Modeling of Pressure Injury Risk in Patients Admitted to an Intensive Care Unit

BACKGROUND

Pressure injuries are an important problem in hospital care. Detecting the population at risk for pressure injuries is the first step in any preventive strategy. Available tools such as the Norton and Braden scales do not take into account all of the relevant risk factors. Data mining and machine learning techniques have the potential to overcome this limitation.

OBJECTIVES

To build a model to detect pressure injury risk in intensive care unit patients and to put the model into production in a real environment.

METHODS

The sample comprised adult patients admitted to an intensive care unit (N = 6694) at University Hospital of Torrevieja and University Hospital of Vinalopó. A retrospective design was used to train (n = 2508) and test (n = 1769) the model and then a prospective design was used to test the model in a real environment (n = 2417). Data mining was used to extract variables from electronic medical records and a predictive model was built with machine learning techniques. The sensitivity, specificity, area under the curve, and accuracy of the model were evaluated.

RESULTS

The final model used logistic regression and incorporated 23 variables. The model had sensitivity of 0.90, specificity of 0.74, and area under the curve of 0.89 during the initial test, and thus it outperformed the Norton scale. The model performed well 1 year later in a real environment.

CONCLUSIONS

The model effectively predicts risk of pressure injury. This allows nurses to focus on patients at high risk for pressure injury without increasing workload.

Item Reduction of the Boston Carpal Tunnel Questionnaire Using Decision Tree Modeling

OBJECTIVE

To produce an electronic decision tree version of the Boston Carpal Tunnel Questionnaire (DT-BCTQ) using the chi-squared automatic interaction detection (CHAID) algorithm to reduce questionnaire length of the Boston Carpal Tunnel Questionnaire (BCTQ) while minimizing the loss of measurement properties.

DESIGN

Criterion standard study. All BCTQs completed between January 2012 and September 2016 by patients who were treated for carpal tunnel syndrome (CTS) were randomly divided into a development and a validation dataset at a 3-to-1 ratio. Optimization of the CHAID algorithm was performed in the development dataset to determine the most optimal DT-BCTQ.

SETTING

Private hand clinic providing both surgical and nonsurgical (orthosis and exercise therapy) treatment for hand and wrist disorders.

PARTICIPANTS

Patients with CTS (N=4470) completed a total of 10,055 BCTQs.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

The intraclass correlation coefficient (ICC) was calculated between the original BCTQ scores and the scores predicted by the DT-BCTQ in the validation dataset. Bland-Altman plots visualized the agreement between the BCTQ and the DT-BCTQ.

RESULTS

The DT-BCTQ reduced the number of questions needed to ask a patient from 11 to a maximum of 3 for the symptom severity scale domain and from 8 to maximum of 3 for the functional status scale domain. The ICC between the original BCTQ and DT-BCTQ was 0.94. The mean difference between the BCTQ and DT-BCTQ was 0.05 on the 0-5 scale (95% confidence interval [CI], -0.48 to 0.57) for the symptom severity scale; 0.02 (95% CI, -0.45 to 0.49) for the functional status scale; and 0.04 (95% CI, -0.31 to 0.39) for the total BCTQ score.

CONCLUSION

By creating the DT-BCTQ, we diminished the number of questions needed to ask a patient from 18 to a maximum of 6 questions (3 for each subscore) when administering the BCTQ while maintaining an ICC of 0.94 with the original BCTQ.

Identification of Risk Factors of Multidrug-Resistant Tuberculosis by using Classification Tree Method

Multidrug-resistant tuberculosis (MDR-TB) has become a major public health problem. We tried to apply the classification tree model in building and evaluating a risk prediction model for MDR-TB. In this case-control study, 74 newly diagnosed MDR-TB patients served as the case group, and 95 patients without TB from the same medical institution served as the control group. The classification tree model was built using Chi-square Automatic Interaction Detectormethod and evaluated by income diagram, index map, risk statistic, and the area under receiver operating characteristic (ROC) curve. Four explanatory variables (history of exposure to TB patients, family with financial difficulties, history of other chronic respiratory diseases, and history of smoking) were included in the prediction model. The risk statistic of misclassification probability of the model was 0.160, and the area under ROC curve was 0.838 (P < 0.01). These suggest that the classification tree model works well for predicting MDR-TB. Classification tree model can not only predict the risk of MDR-TB effectively but also can reveal the interactions among variables.

Using clinical data to predict high-cost performance coding issues associated with pressure ulcers: a multilevel cohort model

OBJECTIVE

Hospital-acquired pressure ulcers (HAPUs) have a mortality rate of 11.6%, are costly to treat, and result in Medicare reimbursement penalties. Medicare codes HAPUs according to Agency for Healthcare Research and Quality Patient-Safety Indicator 3 (PSI-03), but they are sometimes inappropriately coded. The objective is to use electronic health records to predict pressure ulcers and to identify coding issues leading to penalties.

MATERIALS AND METHODS

We evaluated all hospitalized patient electronic medical records at an academic medical center data repository between 2011 and 2014. These data contained patient encounter level demographic variables, diagnoses, prescription drugs, and provider orders. HAPUs were defined by PSI-03: stages III, IV, or unstageable pressure ulcers not present on admission as a secondary diagnosis, excluding cases of paralysis. Random forests reduced data dimensionality. Multilevel logistic regression of patient encounters evaluated associations between covariates and HAPU incidence.

RESULTS

The approach produced a sample population of 21 153 patients with 1549 PSI-03 cases. The greatest odds ratio (OR) of HAPU incidence was among patients diagnosed with spinal cord injury (ICD-9 907.2: OR = 14.3; P  < .001), and 71% of spinal cord injuries were not properly coded for paralysis, leading to a PSI-03 flag. Other high ORs included bed confinement (ICD-9 V49.84: OR = 3.1, P  < .001) and provider-ordered pre-albumin lab (OR = 2.5, P  < .001).

DISCUSSION

This analysis identifies spinal cord injuries as high risk for HAPUs and as being often inappropriately coded without paralysis, leading to PSI-03 flags. The resulting statistical model can be tested to predict HAPUs during hospitalization.

CONCLUSION

Inappropriate coding of conditions leads to poor hospital performance measures and Medicare reimbursement penalties.

Parafricta Bootees and Undergarments to Reduce Skin Breakdown in People with or at Risk of Pressure Ulcers: A NICE Medical Technologies Guidance

As part of the development of the National Institute for Health and Care Excellence (NICE) Medical Technologies Guidance on Parafricta Bootees and Undergarments to reduce skin breakdown in people with, or at risk of, pressure ulcers, the manufacturer (APA Parafricta Ltd) submitted clinical and economic evidence, which was critically appraised by an External Assessment Centre (EAC) and subsequently used by the Medical Technologies Advisory Committee (MTAC) to develop recommendations for further research. The University of Birmingham and Brunel University, acting as a consortium, were commissioned to act as the EAC, independently appraising the submission. This article is an overview of the original evidence submitted, the EAC's findings and the final NICE guidance. Very little comparative evidence was submitted to demonstrate the effectiveness of Parafricta Bootees or Undergarments. The sponsor submitted a simple cost analysis to estimate the costs of using Parafricta in addition to current practice-in comparison with current practice alone-in hospital and community settings separately. The analysis took a National Health Service (NHS) perspective. The basis of the analysis was a previously published comparative study, which showed no statistical difference in average lengths of stay between patients who wore Parafricta Undergarments and Bootees, and those who did not. The economic model incorporated the costs of Parafricta but assumed shorter lengths of stay with Parafricta. The sponsor concluded that Parafricta was cost saving relative to the comparators. The EAC made amendments to the sponsor's analysis to correct for errors and to reflect alternative assumptions. Parafricta remained cost saving in most analyses, and the savings per prevalent case ranged from £757 in the hospital model to £3455 in the community model. All analyses were severely limited by the available data on effectiveness-in particular, a lack of good-quality comparative studies.

Analyses of pressure ulcer incidence in inpatient setting in a Portuguese hospital

AIM

To gain more insight into the magnitude of the problem of pressure ulcer incidence in general wards of a Portuguese hospital.

MATERIAL AND METHODS

Retrospective cohort analysis of electronic health record database from 7132 adult patients admitted to medical and surgical wards of Aveiro Hospital during 2012. The development of (at least) one pressure ulcer during the length of stay was associated with age, gender, type of admission, specialty units, first Braden Scale score, length of stay, patient discharge outcome and ICD-9 diagnosis.

RESULTS

An incidence of 3.4% participants with pressure ulcer category I-IV in inpatient setting during 2012. During the length of stay, 320 new pressure ulcers were developed, most of them category/stage II. The sacrum/coccyx and the trochanters were the most problematic areas.

CONCLUSIONS

The major risk factor for the development of a new pressure ulcer during the length of stay was the presence of (at least) one pressure ulcer at the first skin assessment. The length of stay itself, age and lower Braden Scale scores of our participants also played an important role in the odds of developing a pressure ulcer. Infectious diseases, traumatism and fractures and respiratory diseases were the ICD-9 diagnoses with higher frequency of participants that developed (at least) one pressure ulcer during the length of stay. It's important to standardize procedures and documentation in all care settings. The documentation of nursing interventions is vital to evaluate the impact of evidence-based nursing.

Patient perceptions of the role of nutrition for pressure ulcer prevention in hospital: an interpretive study

PURPOSE

The aims of this study were to explore (a) patients' perceptions of the role of nutrition in pressure ulcer prevention; and (b) patients' experiences with dieticians in the hospital setting.

DESIGN

Interpretive qualitative study.

SUBJECTS AND SETTING

The sample comprised 13 females and 7 males. Their mean age was 61.3 ± 12.6 years (mean ± SD), and their average hospital length of stay was 7.4 ± 13.0 days. The research setting was a public health hospital in Australia.

METHODS

In this interpretive study, adult medical patients at risk of pressure ulcers due to restricted mobility participated in a 20 to 30 minute interview using a semi-structured interview guide. Interview questions were grouped into 2 domains; perceptions on the role of nutrition for pressure ulcer prevention; and experiences with dieticians. Recorded interviews were transcribed and analyzed using content analysis.

RESULTS

Within the first domain, 'patient knowledge of nutrition in pressure ulcer prevention,' there were varying patient understandings of the role of nutrition for prevention of pressure ulcers. This is reflected in 5 themes: (1) recognizing the role of diet in pressure ulcer prevention; (2) promoting skin health with good nutrition; (3) understanding the relationship between nutrition and health; (4) lacking insight into the role of nutrition in pressure ulcer prevention; and (5) acknowledging other risk factors for pressure ulcers. Within the second domain, patients described their experiences with and perceptions on dieticians. Two themes emerged, which expressed differing opinions around the role and reputation of dieticians; they were receptive of dietician input; and displaying ambivalence towards dieticians' advice.

CONCLUSIONS

Hospital patients at risk for pressure ulcer development have variable knowledge of the preventive role of nutrition. Patients had differing perceptions of the importance and value of information provided by dieticians.

Exploring factors associated with pressure ulcers: a data mining approach

BACKGROUND

Pressure ulcers are associated with a nearly three-fold increase in in-hospital mortality. It is essential to investigate how other factors besides the Braden scale could enhance the prediction of pressure ulcers. Data mining modeling techniques can be beneficial to conduct this type of analysis. Data mining techniques have been applied extensively in health care, but are not widely used in nursing research.

PURPOSE

To remedy this methodological gap, this paper will review, explain, and compare several data mining models to examine patient level factors associated with pressure ulcers based on a four year study from military hospitals in the United States.

METHODS

The variables included in the analysis are easily accessible demographic information and medical measurements. Logistic regression, decision trees, random forests, and multivariate adaptive regression splines were compared based on their performance and interpretability.

RESULTS

The random forests model had the highest accuracy (C-statistic) with the following variables, in order of importance, ranked highest in predicting pressure ulcers: days in the hospital, serum albumin, age, blood urea nitrogen, and total Braden score.

CONCLUSION

Data mining, particularly, random forests are useful in predictive modeling. It is important for hospitals and health care systems to use their own data over time for pressure ulcer risk prediction, to develop risk models based upon more than the total Braden score, and specific to their patient population.

Microscopic contact area and friction between medical textiles and skin

The mechanical contact between medical textiles and skin is relevant in the health care for patients with vulnerable skin or chronic wounds. In order to gain new insights into the skin-textile contact on the microscopic level, the 3D surface topography of a normal and a new hospital bed sheet with a regular surface structure was measured using a digital microscope. The topographic data was analysed concerning material distribution and real contact area against smooth surfaces as a function of surface deformations. For contact conditions that are relevant for the skin of patients lying in a hospital bed it was found that the order of magnitude of the ratio of real and apparent contact area between textiles and skin or a mechanical skin model lies between 0.02 and 0.1 and that surface deformations, i.e. penetration of the textile surface asperities into skin or a mechanical skin model, range from 10 to 50µm. The performed analyses of textile 3D surface topographies and comparisons with previous friction measurement results provided information on the relationship between microscopic surface properties and macroscopic friction behaviour of medical textiles. In particular, the new bed sheet was found to be characterised by a trend towards a smaller microscopic contact area (up to a factor of two) and by a larger free interfacial volume (more than a factor of two) in addition to a 1.5 times lower shear strength when in contact with counter-surfaces. The applied methods can be useful to develop improved and skin-adapted materials and surfaces for medical applications.

Older breast cancer survivors: can interaction analyses identify vulnerable subgroups? A report from the American Cancer Society Studies of Cancer Survivors

PURPOSE/OBJECTIVES

To explore interactions among personal, cancer, aging, and symptom variables relative to physical function (PF) in older adult breast cancer survivors to better identify vulnerable subgroups.

DESIGN

Secondary analysis of the American Cancer Society Studies of Cancer Survivors II.

SETTING

U.S. population-based mail and telephone survey.

SAMPLE

2,885 breast cancer survivors from 14 different state cancer registries stratified by cancer type and time since diagnosis. A total of 184 female breast cancer survivors, aged 70 years or older, had complete data on variables of interest and were, therefore, included in this analysis.

METHODS

Chi-Square Automatic Interaction Detector (CHAID) analysis was used to examine variable interactions.

MAIN RESEARCH VARIABLES

PF, symptom bother, comorbidity, social support, length of survivorship, treatment, stage, body mass index, physical activity, emotional health, and personal characteristics.

FINDINGS

An interaction effect between symptom bother and comorbidity was found in 39% of older adult breast cancer survivors, and an interaction effect between symptom bother and marital status was found in 40%. The most vulnerable group (8%) had high symptom bother and more than four comorbid conditions.

CONCLUSIONS

Symptom bother, comorbidity, and marital status were found to have significant interactions such that high comorbidity and high symptom bother were significantly related to lower PF. Married participants with lower symptom bother had significantly higher PF scores. Comorbidity may be the best predictor of PF for the extreme ends of the symptom bother continuum. Advancing age alone was not a sufficient predictor of PF in this analysis.

IMPLICATIONS FOR NURSING

Specific attention to symptom reports, comorbidity, and marital status can guide identification of older adult cancer survivors in need of ongoing survivorship care. The findings support use of a comprehensive assessment and tailored approach to care based on factors other than age.

KNOWLEDGE TRANSLATION

CHAID interaction analysis may be useful in exploring complex nursing problems, such as the needs of older adult cancer survivors, and help oncology nurses develop appropriate interventions and referrals.

[Association of Braden subscales with the risk of development of pressure ulcer]

Pressure ulcers (PU) may increase the incidence of hospital complications, and one should prevent this damage. The Braden Scale stands out as a tool to assess the risk of PU. The study aimed to identify changes in the score of the Braden subscales are associated with the risk of developing PCU. Logistic regression was used in a retrospective cohort study conducted in Hospital de Clínicas de Porto Alegre in adults hospitalized in surgical clinical units from October 2005 to June 2006. We evaluated the records database of 1503 patients with a mean aged 55.5 +/- 16 years, 52.7% female. The incidence of PU was 1.8% and was associated with diabetes and heart failure. There was a higher PU in patients worst in sensory perception, mobility, and activity and the presence of moisture. No association was found between nutrition and PU. Except nutrition, the other Braden sub-scales shown to be predictive of PU.

Modeling the Effects of Moisture-Related Skin-Support Friction on the Risk for Superficial Pressure Ulcers during Patient Repositioning in Bed

Patient repositioning when the skin is moist, e.g., due to sweat or urine may cause skin breakdown since wetness increases the skin-support coefficient of friction (COF) and hence also the shear stresses that are generated in the skin when the patient is being moved. This everyday hospital scenario was never studied systematically however. The aim of this study was to simulate such interactions using a biomechanical computational model which is the first of its kind, in order to quantitatively describe the effects of repositioning on the pathomechanics of moisture-related tissue damage. We designed a finite element model to analyze skin stresses under a weight-bearing bony prominence while this region of interest slides frictionally over the support surface, as occurs during repositioning. Our results show, expectedly, that maximal effective stresses in the skin increase as the moisture-contents-related COF between the skin and the mattress rises. Interestingly however, the rise in stresses for a wet interface became more prominent when the skin tissue was stiffer – which represented aging or diabetes. This finding demonstrates how the aged/diabetic skin is more fragile than a young-adult skin when repositioning in a moist environment. The modeling used herein can now be extended to test effects of different moisturizers, creams, lubricants, or possibly other interventions at the skin-support interface for testing their potential in protecting the skin from superficial pressure ulcers in a standard, objective, and quantitative manner.

Force on the sacrococcygeal and ischial areas during posterior pelvic tilt in seated posture

BACKGROUND

Most posture problems encountered in persons who use wheelchairs in a seated posture for extended periods are related to sacral sitting due to posterior pelvic tilt. Posterior pelvic tilt places pressure and shearing force on the sacrococcygeal area that can lead to pressure ulcers, but the relationship between pelvic tilt and force applied to the sacrococcygeal and ischial tuberosity areas has not yet been investigated.

OBJECTIVE

To investigate the relationships of posterior pelvic tilt in a seated posture with vertical force and horizontal force on the sacrococcygeal and ischial tuberosity areas.

STUDY DESIGN

Repeated measures design.

METHODS

Thirty male and female subjects aged ≥60 years sat in a measurement chair at varying pelvic tilt angles, and force on the sacrococcygeal and ischial tuberosity areas was measured.

RESULTS

The pressure on the sacrococcygeal area increased with pelvic tilt in all subjects, with vertical force averaging 19% of the body weight at a pelvic tilt angle of 30°. The horizontal force on the sacrococcygeal area increased in 93% of the subjects, with an average increase equal to 3% of the body weight.

CONCLUSIONS

We confirmed changes in vertical and horizontal forces on the sacrococcygeal and ischial tuberosity areas with a change in seated posture (pelvic tilt).

CLINICAL RELEVANCE

We propose guidelines for rehabilitation practitioners working with wheelchair users to suggest improved ways of sitting in wheelchairs that avoid pelvic tilt angles that might promote pressure ulcers on the buttocks.