Exploring supervised machine learning approaches to predicting Veterans Health Administration chiropractic service utilization

A descriptive analysis of the contents of Care Response, an international data set of patient-reported outcomes for chiropractic patients

BACKGROUND

Databases have become an important tool in understanding trends and correlations in health care by collecting demographic and clinical information. Analysis of data collected from large cohorts of patients can have the potential to generate insights into factors identifying treatments and the characteristics of subgroups of patients who respond to certain types of care. The Care Response (CR) database was designed to capture patient-reported outcome measures (PROMs) for chiropractic patients internationally. Although several papers have been published analysing some of the data, its contents have not yet been comprehensively documented. The primary aim of this study was to describe the information in the CR database. The secondary aim was to determine whether there was suitable information available to better understand subgroups of chiropractic patients and responsiveness to care. This would be achieved by enabling correlations among patient demographics, diagnoses, and therapeutic interventions with machine learning approaches.

METHODS

Data in all available fields were requested with no date restriction. Data were collected on 12 April 2022. The output was manually scanned for scope and completeness. Tables were created with categories of information. Descriptive statistics were applied.

RESULTS

The CR database collects information from patients at the first clinical visit, 14, 30, and 90 days subsequently. There were 32,468 patient responses; 3210 patients completed all fields through the 90 day follow up period. 45% of respondents were male; 54% were female; the average age was 49. There was little demographic information, and no information on diagnoses or therapeutic interventions. We received StartBack, numerical pain scale, patient global impression of change, and Bournemouth questionnaire data, but no other PROMs.

CONCLUSIONS

The CR database is a large set of PROMs for chiropractic patients internationally. We found it unsuitable for machine learning analysis for our purposes; its utility is limited by a lack of demographic information, diagnoses, and therapeutic interventions. However, it can offer information about chiropractic care in general and patient satisfaction. It could form the basis for a useful clinical tool in the future, if reformed to be more accessible to researchers and expanded with more information collected.

Evaluating rates of chiropractic use and utilization by patient sex within the United States Veterans Health Administration: a serial cross-sectional analysis

BACKGROUND

Within the United States Veterans Health Administration (VHA), the number of patients using healthcare services has increased over the past several decades. Females make up a small proportion of overall patients within the VHA; however, this proportion is growing rapidly. Previous studies have described rates of VHA chiropractic use; however, no prior study assessed differences in use or utilization rates between male and female veterans. The purpose of this study was to assess rates of use and utilization of chiropractic care by sex among VHA patients receiving care at VHA facilities with on-station chiropractic clinics.

METHODS

A serial cross-sectional analysis of VHA national electronic health record data was conducted in Fall 2021 for fiscal year (FY) 2005-2021. The cohort population was defined as VHA facilities with on-station chiropractic clinics, and facilities were admitted to the cohort after the first FY with a minimum of 500 on-station chiropractic visits. Variables extracted included counts of unique users of any VHA on-station facility outpatient services, unique users of VHA on-station facility chiropractic services, number of chiropractic visits, and sex. To calculate use, we determined the proportion of patients of each sex who received chiropractic services to the total patients of the same sex receiving any outpatient care within each facility. To calculate utilization, we determined the number of chiropractic care visits per patient per fiscal year. A linear mixed effects model was applied to examine the difference in chiropractic care utilization by sex.

RESULTS

The percentage of female VHA on-station chiropractic patients increased from 11.7 to 17.7% from FY2005-FY2021. Among VHA facilities with on-station chiropractic care, the percentage of female VHA healthcare users who used chiropractic care (mean = 2.3%) was greater than the percentage of male VHA healthcare users who used chiropractic care (mean = 1.1%). Rates of chiropractic utilization by sex among VHA facilities with on-station chiropractic clinics were slightly higher for females (median = 4.3 visits per year, mean = 4.9) compared to males (median = 4.1 visits per year, mean = 4.6).

CONCLUSION

We report higher use and utilization of VHA chiropractic care by females compared with males, yet for both sexes rates were lower than in the private US healthcare system. This highlights the need for further assessment of the determinants and outcomes of VHA chiropractic care.

A Systems Approach for Assessing Low Back Pain Care Quality in Veterans Health Administration Chiropractic Visits: A Cross-Sectional Analysis

OBJECTIVE

The purpose of this study was to explore a systemwide process for assessing components of low back pain (LBP) care quality in Veterans Health Administration (VHA) chiropractic visits using electronic health record (EHR) data.

METHODS

We performed a cross-sectional quality improvement project. We randomly sampled 1000 on-station VHA chiropractic initial visits occurring from October 1, 2017, to September 30, 2018, for patients with no such visits within the prior 12 months. Characteristics of LBP visits were extracted from VHA national EHR data via structured data queries and manual chart review. We developed quality indicators for history and/or examination and treatment procedures using previously published literature and calculated frequencies of visits meeting these indicators. Visits meeting our history and/or examination and treatment indicators were classified as "high-quality" visits. We performed a regression analysis to assess associations between demographic/clinical characteristics and visits meeting our quality criteria.

RESULTS

There were 592 LBP visits identified. Medical history, physical examination, and neurologic examination were documented in 76%, 77%, and 63% of all LBP visits, respectively. Recommended treatments, such as any manipulation, disease-specific education/advice, and therapeutic exercise, occurred in 75%, 69%, and 40% of chronic visits (n = 383), respectively. In acute/subacute visits (n = 37), any manipulation (92%), manual soft tissue therapy (57%), and disease-specific advice/education (54%) occurred most frequently. Female patients and those with a neck pain comorbid diagnosis were significantly less likely to have a "high-quality" visit, while other regression associations were non-significant.

CONCLUSION

This study explored a systemwide process for assessing components of care quality in VHA chiropractic visits for LBP. These results produced a potential framework for uniform assessment of care quality in VHA chiropractic visits for LBP and highlight potential areas for improvements in LBP care quality assessments.

The Use of Artificial Intelligence in Complementary and Alternative Medicine: A Systematic Scoping Review

Background: The development of artificial intelligence (AI) in the medical field has been growing rapidly. As AI models have been introduced in complementary and alternative medicine (CAM), a systematized review must be performed to understand its current status. Objective: To categorize and seek the current usage of AI in CAM. Method: A systematic scoping review was conducted based on the method proposed by the Joanna Briggs Institute. The three databases, PubMed, Embase, and Cochrane Library, were used to find studies regarding AI and CAM. Only English studies from 2000 were included. Studies without mentioning either AI techniques or CAM modalities were excluded along with the non-peer-reviewed studies. A broad-range search strategy was applied to locate all relevant studies. Results: A total of 32 studies were identified, and three main categories were revealed: 1) acupuncture treatment, 2) tongue and lip diagnoses, and 3) herbal medicine. Other CAM modalities were music therapy, meditation, pulse diagnosis, and TCM syndromes. The majority of the studies utilized AI models to predict certain patterns and find reliable computerized models to assist physicians. Conclusion: Although the results from this review have shown the potential use of AI models in CAM, future research ought to focus on verifying and validating the models by performing a large-scale clinical trial to better promote AI in CAM in the era of digital health.

Data Science Trends Relevant to Nursing Practice: A Rapid Review of the 2020 Literature

BACKGROUND

The term "data science" encompasses several methods, many of which are considered cutting edge and are being used to influence care processes across the world. Nursing is an applied science and a key discipline in health care systems in both clinical and administrative areas, making the profession increasingly influenced by the latest advances in data science. The greater informatics community should be aware of current trends regarding the intersection of nursing and data science, as developments in nursing practice have cross-professional implications.

OBJECTIVES

This study aimed to summarize the latest (calendar year 2020) research and applications of nursing-relevant patient outcomes and clinical processes in the data science literature.

METHODS

We conducted a rapid review of the literature to identify relevant research published during the year 2020. We explored the following 16 topics: (1) artificial intelligence/machine learning credibility and acceptance, (2) burnout, (3) complex care (outpatient), (4) emergency department visits, (5) falls, (6) health care-acquired infections, (7) health care utilization and costs, (8) hospitalization, (9) in-hospital mortality, (10) length of stay, (11) pain, (12) patient safety, (13) pressure injuries, (14) readmissions, (15) staffing, and (16) unit culture.

RESULTS

Of 16,589 articles, 244 were included in the review. All topics were represented by literature published in 2020, ranging from 1 article to 59 articles. Numerous contemporary data science methods were represented in the literature including the use of machine learning, neural networks, and natural language processing.

CONCLUSION

This review provides an overview of the data science trends that were relevant to nursing practice in 2020. Examinations of such literature are important to monitor the status of data science's influence in nursing practice.