Unmonitored Patient Demographic Data Changes can Lead to Bias in Reported Outcomes and Data Registry Development

Andreea Lucaciu¹, Joris RH Hendriks², Riley Baker³, Hamid Ghaednia⁴, Nour Nassour⁵, Soheil Ashkani- Esfahani⁶

¹Department of Orthopaedic Surgery, FARIL-SORG Collaborative, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States; Department of Orthopaedic Surgery, SORG Research Group, Massachusetts General Hospital, Boston, Massachusetts, United States

²Department of Orthopaedic Surgery, Foot & Ankle Research and Innovation Lab (FARIL), Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States; Department of Orthopedic Surgery and Sports Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands

^3,5Department of Orthopaedic Surgery, FARIL-SORG Collaborative, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States; Department of Orthopaedic Surgery, Foot & Ankle Research and Innovation Lab (FARIL), Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States

^4,6Department of Orthopaedic Surgery, FARIL-SORG Collaborative, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States; Department of Orthopaedic Surgery, SORG Research group, Massachusetts General Hospital, Boston, MA; Department of Orthopaedic Surgery, Foot & Ankle Research and Innovation Lab (FARIL), Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States

Corresponding Author: Andreea Lucaciu, Department of Orthopaedic Surgery, FARIL-SORG Collaborative, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, United States; Department of Orthopaedic Surgery, SORG Research Group, Massachusetts General Hospital, Boston, Massachusetts, United States, Phone: +14049520496, e-mail: alucaciu@mgh.harvard.edu

Received on: 27 September 2022; Accepted on: 26 October 2022; Published on: 31 December 2022

How to cite this article: Lucaciu A, Hendriks JRH, Baker R, et al. Unmonitored Patient Demographic Data Changes can Lead to Bias in Reported Outcomes and Data Registry Development. J Foot Ankle Surg (Asia-Pacific) 2023;10(1):20-21.

Source of support: Nil

Conflict of interest: None

Keywords: Clinical data, Data registry, SDOH, Social determinants of health.

Each patient is inevitably affected by his or her social determinants of health (SDOH) which plays an important role in the process of decision-making and predicting the outcomes in patients. Often, SDOH has been shown to have meaningful correlations with short- and long-term outcomes of treatments, especially in the field of orthopedic surgery. Therefore, accurate and continuous recording and monitoring the SDOH is vital in developing patient electronic medical records and registry systems. Changes in these determinants including changing the place of living, access to healthcare centers, socioeconomic status, insurance status, aging, mobilization, and transportation status can affect the quality of life, quality of health care, and the prognosis of the disease. In our recent research experience investigating the correlation of SDOH with different orthopedic conditions such as Achilles rupture, ankle fractures, and hallux valgus, we experienced that the patients had changed their place of living and thus their access to healthcare center was affected. These changes in the place of living as well as changes in the insurance, job, and socioeconomic conditions of the patients hindered us from an accurate analysis of data and unbiased report of the correlation of SDOH and the treatment outcomes of these orthopedic conditions. Although much progress has occurred in recent years to track SDOH, in our experience there was still a considerable number of missing data leading to concerns for biased outcomes.¹ As an example, our hospitals failed to keep track of patients’ relocations, thus making it impossible for researchers to find any correlation between the place of living and the outcomes of Achilles repair in patients who suffered from acute Achilles rupture. Two indices that are commonly used for the assessment of SDOH include the social vulnerability index (SVI) and area deprivation index (ADI) scores; both factors are highly dependent on the living location of the patient. The SVI is affected by the socioeconomic status, household, composition and disability, minority status and language, and housing type and transportation.² This score is used to determine how vulnerable the communities are and how to prepare personnel accordingly in case of an emergency. ADI is based on 17 census variables that describe the SDOH in an area.² It allows the user, in this case being the health care provider, to determine resource allocation based on geographical need. However, these scores are not 100% reliable either. There is evidence of embedded bias within the scores secondary to the calculation methods. In fact, one study, in particular, found that the ADI calculation decreases in reliability as the vulnerability score increases.³ In our experience, due to a lot of missing data, we were not able to provide the complete set of data needed for SVI and ADI as some of the patients were relocated during the healing course and we did not have any records of their previous living places.

Other than developing valid and reliable registry systems, demographic data and SDOH are being highly valued and used in predictive models.⁴ Scientists believe that the accurate and reliable prediction of the outcomes such as outcomes of a treatment or prognosis of a diseases as well as the quality of healthcare, highly relies on patient’s characteristics and SDOH. However, if the health records of the patients are not accurate and not monitored longitudinally during time, the predictions can be biased and not precise. In our experience, we tried to detect the associated factors impacting patient’s recovery time in cases of acute Achilles tendon rupture. The time between injury and proper surgical intervention varied drastically in our dataset. Although one of our aims was to assess the correlation between the place of living, access to healthcare, and the quality of healthcare among our patients, due to lack of data on patients’ previous addresses, and due to the fact that some of the patients had changed their addresses during the healing course, we failed to achieve our aim. Investigating other datasets showed us that this problem is widespread among various health care centers in our area which raised a major concern for modifying the methods of admitting the patients, updating their data regularly in each visit, and providing reports of the alterations of SDOH during time. This is also necessary when these centers want to report SVI, ADI, or other SDOH-related reports to the authorities and policy-making entities.

ORCID

Soheil Ashkani-Esfahani https://orcid.org/0000-0003-2299-6278

REFERENCES

1. Johnson CB, Luther B, Wallace AS, et al. Social determinants of health: what are they and how do we screen. Orthop Nurs 2022;41(2):88–100. DOI: 10.1097/NOR.0000000000000829

2. Maroko AR, Doan TM, Arno PS, et al. Integrating social determinants of health with treatment and prevention: a new tool to assess local area deprivation. Prev Chronic Dis 2016;13:E128. DOI: 10.5888/pcd13.160221

3. Tate E. Uncertainty analysis for a Social Vulnerability Index. Ann Assoc Am Geogr 2013;103(3):526–543. DOI: 10.1080/00045608.2012.700616

4. Chen M, Tan X, Padman R. Social determinants of health in electronic health records and their impact on analysis and risk prediction: a systematic review. J Am Med Inform Assoc 2020;27(11):1764–1773. DOI: 10.1093/jamia/ocaa143

________________________
© The Author(s). 2023 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted use, distribution, and non-commercial reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.