Explanatory Multivariate Modeling for Disability, Pain, and Claims in Patients with Spine Pain
Anthony Pastore, SPT
Background
For the past year, I have had the pleasure of working on a collaborative project between the Duke University DPT and UNC-Chapel Hill DPT programs. The purpose of this project was to foster a working relationship between two of the nation’s highest’s ranking DPT programs. Unfortunately, it wasn’t until this year that our two programs reached out to one another with the idea of joining forces to work on a piece of collaborative research. Under the guidance of Dr. Givens and the director of Duke’s DPT program director, Dr. Chad Cook, Chris Green and I worked closely with two Duke students to conduct a research project using data from ATI Physical Therapy’s Patient Outcome Registry.
Our team was provided data from patients seeking physical therapy treatment for low back or neck related pain. Chris Green worked with one Duke student in the arena of low back pain while I and another Duke student focused our research on the cervical spine. Our work over the last several months has culminated in a manuscript that we plan to submit to a peer-reviewed journal and an abstract that we will be submitting to the 2018 Combined Sections Meeting (CSM). I have greatly enjoyed working with our colleagues over at Duke and hope that this project is the first in a long line of collaborative work between our two programs.
Overview
Neck-related pain is one of the most common musculoskeletal conditions experienced by adults globally, with an estimated global point prevalence of 5% in 2010.1 In the United States, it is estimated that 15.7% of the population had neck pain in 2015.2 While a majority of patients with neck pain will have favorable outcomes with or without treatment, many individuals with acute neck pain may develop chronic or reoccurring pain and disability. In fact, over a third of individuals with neck pain will continue to have persistent symptoms that last longer than six months.3 As a result, neck pain has become a significant source of global burden, ranking as the 4th highest cause of disability in the world.1
The economic burden associated with spine pain (neck and low-back) is high and is related to the increasing use of prescription medications, medical imaging and diagnostic test, spinal injections, higher patient expectations for care, and increasing use of surgical treatments.4After adjusting for age and sex, annual expenditures for individuals with spine pain rose from a mean of $4695 in 1995 to $6096 in 2005, totaling nearly $86 billion annually.4 The mean overall health care expenditures of individuals with spine related problems were substantially higher than those without spine problems in 2005 (58% difference).4 Furthermore, these increases in health expenditure has not been associated with an improvement in patient self-assessed health status.4
Statement of Need
Many studies have examined prognostic factors related to spine pain and their ability to predict future pain and/or disability. While many studies have examined the prognostic factors related to patient outcomes, there is a paucity of research that has examined variables that can predict total healthcare costs in this population. An important and novel aspect of our project is the inclusion of insurance payer information for our subjects. Furthermore, our data includes patients who received direct access to physical therapy. To our knowledge, our research is one of the first to examine prognostic factors that predict both outcomes and cost related to the treatment of individuals with spine pain via direct access. The identification of patient characteristics that can identify individuals at risk for poor recovery and factors that are related to higher healthcare utilization may help better inform prognosis and treatment decisions, thereby reducing the global burden of spine pain.
Products
Several products emerged as a result of my work over the past year. In the Fall, I performed a critically appraised topic regarding prognostic factors for neck related pain as part of the Evidence Based Practice II course. Additionally, I begin conducting a review of literature as part of a research experience with Dr. Givens that I continued working on throughout the Spring semester. Most of the work completed in the spring included an abstract to be submitted to Combined Sections Meeting in 2018 and a manuscript draft. Our group is currently editing a manuscript that we intend to submit to a peer-reviewed journal in the near future. Unfortunately, I am unable to share our manuscript or data until it has been published.
My Products can be viewed here:
Self-Assessment and Reflection
To evaluate my abstract I used the requirements for abstract submission to CSM. Evaluation of our manuscript was completed using the TRIPOD Checklist for prediction model development and validation. Dr. Givens and Dr. Cook also provided ample feedback and insight regarding my work over the past semester.
The past two semesters have been quite a journey. I first took an interest in this capstone project because I believed that it would allow me the opportunity to work on my skills as a researcher. I have quite of bit of experience as a research assistant, but this project was my first foray into the world of data analysis and scholarly writing. This experience was challenging, but I believe I have come out the other side with a better understanding of the research process. My hope is to use the skills and knowledge I gained during this capstone experience to continue conducting research throughout my career.
Acknowledgements
I’d like to extend a huge thank you to my project advisor, Dr. Debby Givens, PT, PhD, DPT, whose guidance proved invaluable for my completion of this project. She was instrumental in keeping me on the right path over the past two semesters and provided me with great insight regarding the research process. I’d also like to thank Dr. Chad Cook, PT, PhD, MBA, FAAOMPT, whose expertise in the realm of data analysis made this project possible. A thank you to Charles Thigpen, PhD, PT, ATC, who provided our team with the data we used for this project and also provided me with his insight regarding my products. I also appreciated the opportunity to work with two of my colleagues from Duke’s DPT program, Leah Cronley, SPT, and Merritt Walker, SPT. I hope that the work we accomplished together serves as a model for future collaborations between our two programs. Finally, I have to thank my friend and classmate, Chris Green, who fought side-by-side with me over the past two semesters. His friendship and willingness to serve as a sounding board at all hours of the day made this project a much smoother process.
References
- Hoy D, March L, Woolf A, et al. The global burden of neck pain: estimates from the Global Burden of Disease 2010 study. Ann Rheum Dis. 2014;73(7):1309-1315. doi:10.1136/annrheumdis-2013-204431.
- Center for Health Statistics N. Vital and Health Statistics Series 10, No. 235 (12/2007). 2006.
- Côté P, Cassidy JD, Carroll LJ, Kristman V. The annual incidence and course of neck pain in the general population: a population-based cohort study. doi:10.1016/j.pain.2004.09.004.
- Martin BI, Deyo RA, Mirza SK, et al. Expenditures and Health Status Among Adults With Back and Neck Problems. JAMA. 2008;299(6):656. doi:10.1001/jama.299.6.656.
4 Responses to “Explanatory Multivariate Modeling for Disability, Pain, and Claims in Patients with Spine Pain”
Carla Hill
Anthony,
Very interesting Capstone project. I’m sure you learned a lot about management of patients with spine pain through this effort. What are the 2-3 most surprising findings or ‘ah ha’ moments you’ve had during completion of this project and will impact your patient care?
Anthony Pastore
Hey Carla, thank you for your question. I think one of the ‘ah ha’ moments I had during the completion of this project came when I realized some of the potential uses of prognostic variables in patient care. Obviously, prognostic variables can be used to more accurately inform patients about the probable course of their condition. What I hadn’t appreciated, however, was the possibility of basing treatment decisions off these prognostic variables. For instance, research shows that the more severe pain and symptoms are in the acute stages of a patient’s cervical pain, the poorer outcomes they have. In these cases, it would be appropriate to use heat/ice, analgesics, and a gradual return to physical activity in the initial stages of a treatment to reduce pain and symptom severity early in order to improve the patient’s long-term outcomes. Secondly, I learned that most patients with neck pain are going to recover with or without treatment, but many will develop chronic or reoccurring symptoms that may last years. If we can identify patients who are at risk for a poor outcome through the information we collect in the patient interview and examination, use of alternative treatment methods may improve the chances that a patient will have a positive outcome. Alternatively, if we find that patient has a high probability of achieving a good outcome, then cheaper, more conservative treatment options would be appropriate. In a period of rising healthcare costs related to the treatment of spine pain, stratifying patients into high or low risk for adverse outcomes can help clinicians choose appropriate treatment options that could reduce the cost of care of this population.
Debby Givens
Anthony – the massive amount of research and reading of the literature is reflected in your capstone products. The manuscript is shaping up well and will make a good contribution to the literature. I hope this project makes you more likely than not to pursue research opportunities.
My question to you is: If you have access to a database like FOTO and knowing what you know now, what key variables would you want to be collected on patients with cervical spine problems that may not be routinely documented by a PT?
Anthony Pastore
Hi Debby, thank you for your kind comments! I do plan on continuing to seek out research opportunities over the next several years. In fact, the residency I’m completing next year requires the completion of a research project. I believe this capstone project has most definitely prepared me for the demands of conducting clinical research.
Great question. Ultimately, I’d like more data regarding some of the prognostic factors that current research has shown only weak evidence for. Various psychosocial and sociodemographic factors currently have weak evidence supporting their use as prognostic variables in patients with neck pain. Some examples of these factors include high levels of worrying, employment status, job demands, perceived quality of life, optimism, anxiety, depression, etc. I believe these are some variables that are not routinely collected on patients, but ones that might show stronger associations with patient outcomes if we had more data. I believe addressing these factors are important if we want to optimize patient outcomes.
Talking specifically about our data, however, I’d like to see more data collected on the following factors; initial disability/pain, chronicity level, active vs. passive treatment, and use of medications. Our data showed some association between these variables and outcome, so I’d like to see if these conclusions hold true in various geographical locations and demographic groups.
Thank you again for all your help over the past year. I look forward to working with you in the future!