1. Provide the citation and attach a pdf of the article.

PDF of Article
Fodeh, S. J., Finch, D., Bouayad, L., Luther, S. L., Ling, H., Kerns, R. D., & Brandt, C. (2018). Classifying clinical notes with pain assessment using machine learning. Medical & Biological Engineering & Computing, 56(7), 1285–1292. https://doi-org.ezproxy.fhsu.edu/10.1007/s11517-017-1772-1

2. What is the abstract of the article? 

Pain is a significant public health problem, affecting millions of people in the USA. Evidence has highlighted that patients with chronic pain often suffer from deficits in pain care quality (PCQ) including pain assessment, treatment, and reassessment. Currently, there is no intelligent and reliable approach to identify PCQ indicators inelectronic health records (EHR). Hereby, we used unstructured text narratives in the EHR to derive pain assessment in clinical notes for patients with chronic pain. Our dataset includes patients with documented pain intensity rating ratings > = 4 and initial musculoskeletal diagnoses (MSD) captured by (ICD-9-CM codes) in fiscal year 2011 and a minimal 1 year of follow-up (follow-up period is 3-yr maximum); with complete data on key demographic variables. A total of 92 patients with 1058 notes was used. First, we manually annotated qualifiers and descriptors of pain assessment using the annotation schema that we previously developed. Second, we developed a reliable classifier for indicators of pain assessment in clinical note. Based on our annotation schema, we found variations in documenting the subclasses of pain assessment. In positive notes, providers mostly documented assessment of pain site (67%) and intensity of pain (57%), followed by persistence (32%). In only 27% of positive notes, did providers document a presumed etiology for the pain complaint or diagnosis. Documentation of patients’ reports of factors that aggravate pain was only present in 11% of positive notes. Random forest classifier achieved the best performance labeling clinical notes with pain assessment information, compared to other classifiers; 94, 95, 94, and 94% was observed in terms of accuracy, PPV, F1-score, and AUC, respectively. Despite the wide spectrum of research that utilizes machine learning in many clinical applications, none explored using these methods for pain assessment research. In addition, previous studies using large datasets to detect and analyze characteristics of patients with various types of pain have relied exclusively on billing and coded data as the main source of information. This study, in contrast, harnessed unstructured narrative text data from the EHR to detect pain assessment clinical notes. We developed a Random forest classifier to identify clinical notes with pain assessment information. Compared to other classifiers, ours achieved the best results in most of the reported metrics. Graphical abstract Framework for detecting pain assessment in clinical notes.

3. Was the study experimental or non-experimental? Explain, tell us what made that clear. 

This is an experimental study. The researchers were able to pick the data from the patient records available to them and use what best suited their study.

4. Was the research qualitative or quantitative? Again, explain.

The research is qualitative. The data is coming from patient records, but the data is how a patient is explaining that they are feeling. Everyone feels pain differently and will describe it differently.

5. What was the population studied?

The population studied was patients with a pain level greater than or equal to 4 and patients who were documented with the MSD ICD9 code.

6. What sample was used for this study?

There were a total of 9940 patients who were selected from 130 VA facilities. It was split into male and female, 8628 males and 1672 females. The clinical notes were narrowed down to facilities providing full services

7. What was the method of measurement?

The research was qualitative. It was collected from patients who were describing their pain. It took in account nine factors, whether or not the patient mentioned pain, intensity of the pain, quality of the pain, how the pain felt throughout the day, factors that caused pain to flare up, factors that alleviated pain, how the pain affects the patient’s life, pain terminology, where the pain is, and the results of any testing done to diagnose pain.

8. What was the method of analysis?

The research was qualitative. The data was analyzed by how the patient described their pain and how it fit into the nine categories being studied. The data was then split into different classifiers K-Nearest Neighbor (KNN), Decision Tree (DT), Support Vector Machine (SVM), and Random Forest (RF).

9. What was the conclusion of the study?

The conclusion of the study was that the study was a good first step to creating an automated system focused on pain care quality. It has information that will help with future policies and will help with the care provided for patients experiencing pain.

10. Why is this study useful to you? Explain in detail.

This study is useful to me because I am interested in how machine learning and big data can improve healthcare. Patient records are full of data that just sits there and is never really touched again. Studies like this show useful the data can be if revisited again, and how it can even be used to improve patient experience going forward.

11. What would be the next logical step in extending this study? 

The next logical step in this study would be to continue building their database to include different types of pain and how to care for them. It would also be useful to add in more qualifiers that patients may explain their pain as that did not make it into this study.