Population-Based Applications and Analytics Using Patient-Reported Outcome Measures.

The intersection of big data and artificial intelligence (AI) has resulted in advances in numerous areas, including machine learning, computer vision, and natural language processing. Although there are many potentially transformative applications of AI in health care, including precision medicine, this industry has been slow to adopt these technologies. At the same time, the operations of health care have historically been system-directed and physician-directed rather than patient-centered. The application of AI to patient-reported outcome measures (PROMs), which provide insight into patient-centered health outcomes, could steer research and healthcare delivery toward decisions that optimize outcomes important to patients. Historically, PROMs have only been collected within research registries. However, the increasing availability of PROMs within electronic health records has led to their inclusion in big data ecosystems, where they can inform or be informed by other data elements. The use of big data to analyze PROMs can help establish norms, evaluate data distribution, and determine proportions of patients achieving change or threshold standards. This information can be used for benchmarking, risk adjustment, predictive modeling, and ultimately improving the health of individuals and populations.

A retrospective review of lower extremity fracture care in patients with spinal cord injury.

CONTEXT/OBJECTIVE: To identify circumstances surrounding incident lower extremity fractures (ILEFs) in patients with spinal cord injury (SCI) and to describe the impact of these fractures on service needs and provision of pharmacological therapies for osteoporosis. DESIGN: Retrospective medical record review. SETTING: Four Veterans Affairs Medical Centers in the USA. PARTICIPANTS: One hundred and forty patients with traumatic SCI who sustained an ILEF from 2002 to 2007. OUTCOME MEASURES: Fracture circumstances and use of assistive devices were described using percentages, means, and standard deviations. Fisher's exact test was used to determine the relationship between fracture site, and patient age and duration of SCI. Differences in pharmacological provision of therapies for osteoporosis pre- and post-fracture were examined using exact McNemar's test. RESULTS: One hundred and fifty-five ILEFs were identified in 140 patients. Tibia/fibula and femur fractures were the most common fractures. Fracture site was not related to patient's age or duration of SCI. Almost one-third of all fractures occurred during transfers to and from wheelchairs. Post-fracture, the provision of new or modified assistive devices, primarily wheelchairs, was frequent, occurring in 83% of patients in the year post-fracture. Few patients transferred residence to a nursing home following the fracture. There was a significant difference in the use of pharmacological therapies for osteoporosis in the first year post-fracture compared with the year prior to the fracture (P < 0.01), with significant differences in the volume of prescriptions for calcium supplements (P < 0.01) and bisphosphonates (P = 0.02). Overall, the amount of prescriptions for osteoporosis increased the year post-fracture (56%) from the year pre-fracture (39%); this increase was secondary to increases in prescriptions for calcium supplements (pre = 13%; post = 30%) and bisphosphonates (pre = 2%; post = 7%). CONCLUSIONS: We have identified that wheelchair and other transfer activities are a key area that could be a focus of fracture prevention in SCI. The need for new or modified assistive devices and/or wheelchair skills retraining post-fracture should be anticipated. Examination of whether treatments for osteoporosis following a fracture can prevent future osteoporotic fractures is warranted.

Thiazide use is associated with reduced risk for incident lower extremity fractures in men with spinal cord injury.

OBJECTIVE: To determine the association between thiazide use and lower extremity fractures in patients who are men with a spinal cord injury (SCI). DESIGN: Cohort study from fiscal years 2002 to 2007. SETTING: Medical centers. PARTICIPANTS: Men (N=6969) with an SCI from the Veterans Affairs (VA) Spinal Cord Dysfunction (SCD) Registry, including 1433 users of thiazides and 5536 nonusers of thiazides. INTERVENTION: Thiazide use versus nonuse. MAIN OUTCOME MEASURE: Incident lower extremity fractures. RESULTS: Among the men, 21% in the VA SCD Registry (fiscal years 2002-2007) included in these analyses used thiazide diuretics. There were 832 incident lower extremity fractures over the time period of this study: 110 fractures (7.7%) in 1433 thiazide users and 722 fractures (13%) in 5536 nonusers of thiazides. In unadjusted and adjusted models alike, thiazide use was associated with at least a one-quarter risk reduction in lower extremity fracture at any given point in time (unadjusted: hazard ratio (HR)=.75; 95% confidence interval (CI), .59-.94; adjusted: HR=.74; 95% CI, .58-.95). CONCLUSIONS: Thiazide use is common in men with SCI and is associated with a decreased likelihood for lower extremity fractures.

Mortality after lower extremity fractures in men with spinal cord injury.

In the United States, there are over 200,000 men with spinal cord injuries (SCIs) who are at risk for lower limb fractures. The risk of mortality after fractures in SCI is unknown. This was a population-based, cohort study of all male veterans (mean age 54.1; range, 20.3-100.5 years) with a traumatic SCI of at least 2 years' duration enrolled in the Veterans Affairs (VA) Spinal Cord Dysfunction Registry from FY2002 to FY2010 to determine the association between lower extremity fractures and mortality. Mortality for up to 5 years was determined. The lower extremity fracture rate was 2.14 per 100 patient-years at risk for at least one fracture. In unadjusted models and in models adjusted for demographic, SCI-related factors, healthcare use, and comorbidities, there was a significant association between incident lower extremity fracture and increased mortality (hazard ratio [HR], 1.38; 95% confidence interval [CI], 1.17-1.63; HR, 1.36; 95% CI, 1.15-1.61, respectively). In complete SCI, the hazard of death after lower extremity fracture was also increased (unadjusted model: HR, 1.46; 95% CI, 1.13-1.89; adjusted model: HR, 1.32; 95% CI, 1.02-1.71). In fully-adjusted models, the association of incident lower extremity fracture with increased mortality was substantially greater in older men (age ≥50 years) for the entire cohort (HR, 3.42; 95% CI, 2.75-4.25) and for those with complete SCI (HR, 3.13; 95% CI, 2.19-4.45), compared to younger men (age <50 years) (entire cohort: HR, 1.42; 95% CI, 0.94-2.14; complete SCI: HR, 1.71; 95% CI, 0.98-3.01). Every additional point in the Charlson comorbidity index was associated with a 10% increase in the hazard of death in models involving the entire cohort (HR, 1.11; 95% CI, 1.09-1.13) and also in models limited to men with complete SCI (HR, 1.10; 95% CI, 1.06-1.15). These data support the concept that both the fracture itself and underlying comorbidities are drivers of death in men with SCI.

The association of anticonvulsant use with fractures in spinal cord injury.

OBJECTIVE: The aim of this study was to determine whether anticonvulsants, including the benzodiazepine subclass, are associated with an increased risk for lower extremity fractures in male patients with spinal cord injury. DESIGN: All male patients with a history of a traumatic spinal cord injury of 2 yrs' duration or longer in the Veterans Affairs Spinal Cord Disease Registry from 2002 to 2007 were included. Incident lower extremity fractures during this period and anticonvulsant use were identified. The association of anticonvulsant use, overall, by type (enzyme inducing, non-enzyme inducing), by number (monotherapy vs. polytherapy), by benzodiazepine subclass, and by individual medication used was determined. RESULTS: In this cohort, 892 veterans sustained a fracture, and 6555 did not. Compared with nonusers of anticonvulsants, there was a significant positive relationship with fractures by overall use of anticonvulsants (HR, 1.17 [95% CI, 1.01-1.36]), by users of the benzodiazepine subclass (HR, 1.28 [95% CI, 1.11-1.47]), and by polytherapy compared with monotherapy (HR, 1.20 [95% CI, 1.00-1.42]) but not by anticonvulsant type (HR, 0.92 [95% CI, 0.58-1.47]). Temazepam (HR, 1.28 [95% CI, 1.01-1.62]), alprazolam (HR, 1.54 [95% CI, 1.04-2.29]), and diazepam (HR, 1.23 [95% CI, 1.06-1.41]) were significantly positively associated with fractures. CONCLUSIONS: Attention to fracture prevention is important when anticonvulsants are prescribed in spinal cord injury, particularly when more than one anticonvulsant is used.

The association of opioid use with incident lower extremity fractures in spinal cord injury.

OBJECTIVE: To determine the association between opioid use and lower extremity fracture risk in men with spinal cord injury (SCI). DESIGN: Retrospective cohort study. SETTING: Veterans Affairs Healthcare System. PARTICIPANTS: In total, 7447 male Veterans with a history of a traumatic SCI identified from the Veterans Affairs (VA) Spinal Cord Dysfunction Registry (SCD) from September 2002 through October 2007 and followed through October 2010. OUTCOME MEASURES: Incident lower extremity fractures by use of opioids. RESULTS: In individuals identified from the VA SCD Registry 2002-2007, opioid use was quite common, with approximately 70% of the cohort having received a prescription for an opioid. Overall, there were 892 incident lower extremity fractures over the time period of this study (597 fractures in the opioid users and 295 fractures in the non-opioid users). After adjusting for covariates, there was a statistically significant relationship between opioid use and increased risk for lower extremity fractures (hazard ratio 1.82 (95% confidence interval 1.59-2.09)). Shorter duration of use (<6 months) and higher doses were positively related to fracture risk (P < 0.0001). CONCLUSIONS: Opioid use is quite common in SCI and is associated with an increased risk for lower extremity fractures. Careful attention to fracture prevention is warranted in patients with SCI, particularly upon initiation of an opioid prescription and when higher doses are used.

Assessing the feasibility of subepidermal moisture to predict erythema and stage 1 pressure ulcers in persons with spinal cord injury: a pilot study.

BACKGROUND: Persons with spinal cord injury (SCI) are at high risk for pressure ulcers (PrUs) throughout their lifetime due to decreased mobility, lack of sensation, and other physiological changes. The high prevalence and recurrence rates, and costs associated with PrUs in veterans with SCI indicate the need for a reliable and practical method of detecting early PrUs. OBJECTIVE: To assess the feasibility of obtaining biophysical measures of sub-epidermal moisture (SEM) using a handheld dermal phase meter to predict PrUs. DESIGN/METHODS: Prospective observational design. Thirty-four veterans at two VA SCI centers (Hines, Long Beach) received daily (n = 12) or weekly (n = 22) SEM and concurrent visual skin assessment (VSA) across nine anatomic locations for up to 6 weeks. Outcome measures: SEM, visual skin assessment (VSA), and stage I PrUs. FINDINGS/RESULTS: SEM was lowest for normal skin (39.3 dermal phase units (DPU), SD = 12.6) and higher for erythema/stage 1 PrUs (40.8 DPU, SD = 10.4) across all anatomic sites. Buttocks SEM were different between normal skin (40.5 DPU, SD = 10.3) and erythema/stage1 PrUs (43.8, SD = 9.5). SEM taken at heels were lower across all skin conditions (normal skin 28.2 DPU; erythema/stage 1 PrUs 34.7 DPU). SEM was taken when generalized edema present was lower than without generalized edema. CONCLUSIONS: Preliminary results of using SEM to detect early PrU damage may translate from nursing home (NH) residents to persons with SCI. This study provides a foundation for a larger study to implement and assess SEM use as a method of prevention of PrUs.

Characteristics of recurrent pressure ulcers in veterans with spinal cord injury.

BACKGROUND/OBJECTIVE: To describe characteristics of recurrent pressure ulcers (PrUs) in veterans with spinal cord injury (SCI). DESIGN: Descriptive, cohort study. SETTINGS AND PARTICIPANTS: Twenty-four veterans with SCI from 6 SCI centers in the Department of Veterans Affairs. METHODS: Data from a prospective study evaluating PrUs were analyzed for 24 veterans with 29 recurrent PrUs during 9 months. Additional retrospective medical record data were analyzed for 15 veterans who received inpatient treatment. RESULTS: Participants were male, 50% non-Hispanic white, with paraplegia (63%), complete SCI (83%), a mean age of 56 years, and mean time since SCI of 21 years. Most PrUs recurred (63%, n = 15 patients) in the same location as the most recent ulcer and at the ischial tuberosities (63%). Mean time to recurrence was 16.6 weeks. PrUs were stage III (28%, n = 8) or IV (45%, n = 13) with undermining (48%), necrotic slough (50%), and minimal exudate. One third were (n = 9) larger than 16 cm2. Mean Bates-Jensen Wound Assessment Tool Score was 33.63. Inpatient medical record data (n = 15) showed 73% with documentation indicating infection treated with antibiotics (53%, n = 8 patients), osteomyelitis (47%, n = 7), and/or cellulitis (13%, n = 2) noted. Plastic surgery consultation was obtained for 67% with surgery as an option for 73% (1 without consultation). Scheduled repositioning was documented for 21%. CONCLUSIONS: Most PrUs were severe, located at the same anatomic site, and recurred within 4 months, suggesting that the recurrent ulcers were more likely incomplete healing of the initial PrUs. This sample of veterans with SCI provides early data on recurrent PrU characteristics.