Team:Greece United/Model/App

Introduction

Current trends in health reflect an important contemporary shift towards citizen engagement for health and prevention, as opposed to mere disease management.

However, broadcasting generic health messages (e.g. ‘do this, don’t do that’) has limited effects unless there is a convincing, easily perceived and personally customized body of evidence to back healthy choices.

Therefore osteoarthritis (OA) risk calculators have been developed to help the public perceive the risks of osteoarthritis.

Existing tools base their risk estimation on evidence from one study, e.g. OAPoI [1], TOARP [2] and MSK [3]

We created a new OA risk calculator that uses medical evidence from recent high evidence level medical publications, AEGLE. Aegle was said to have been the daughter of Epione. Her name means "Shine", "Brilliance", either from the beauty of the human body when it is in good health, or from the price paid to the medical profession.

The NOUS OA Risk Calculator is expandable and can include new risk evidence when this is published. The tool is based on the CARRE health risk ontology and expands the CARRE health risk database (CARRE is an EU FP7 ICT project, Contract No. 611140).

Use the Osteoarthritis Risk Calculator here.
Get the software from Github here, by cloning the folder Risk Calculator in our repository.

[1] Osteoarthritis Policy Model, Brigham and Women’s Hospital, USA, https://pivot.bwh.harvard.edu/ongoing-projects/oapol/
[2] Joseph GB et al. Tool for osteoarthritis risk prediction (TOARP) over 8 years using baseline clinical data, X-ray, and MRI: Data from the osteoarthritis initiative. J Magn Reson Imaging. 2018 Jun;47(6):1517-1526. doi: 10.1002/jmri.25892
[3] Musculoskeletal Calculator, by Versus Arthritis and Imperial College London, UK, https://www.versusarthritis.org/policy/resources-for-policy-makers/musculoskeletal-calculator/msk-calculator-faq/

Definition

Risk is the probability of a negative outcome on the health of a population of subjects. The agents responsible for that risk are called risk factors when they aggravate a situation and are used to predict (up to a degree) the occurrence of a condition or deterioration of a patient’s health dividing the population into high and low risk groups.

In our project we used the risk factor conceptual model and ontology developed in the CARRE EU FP7 ICT project and available via the NCBO BioPortal

CARRE risk factor ontology – conceptual model.




Data Sources

In order to identify high-level medical evidence of risk factors related to osteoarthritis, we searched systematically in PubMed during September 2021. PubMed (https://pubmed.ncbi.nlm.nih.gov/) is a biomedical citation database provided by the USA National Library of Medicine and is considered one of the most comprehensive sources for biomedical literature search. The database provides citation details for more than 33 million scientific publications in the biomedical sciences, including guidelines and systematic reviews and meta-analyses produced by commonly accepted independent evidence bodies such as Cochrane Collaboration.

Risk factors identification methodology

Query

(osteoarthritis risk) AND (meta-analysis) AND (last 5 years)

Query as in PubMed

(("osteoarthritis"[MeSH Terms] OR "osteoarthritis"[All Fields] OR "osteoarthritides"[All Fields]) AND ("risk"[MeSH Terms] OR "risk"[All Fields])) AND ((y_5[Filter]) AND (meta-analysis[Filter]))

Inclusion criteria

meta-analyses reporting on risk for osteoarthritis

in human studies

in English language

reporting values for relative risk ratio (RR) or odds ratio (OR) or hazard ratio (HR)

reporting an acceptable level of evidence

Exclusion criteria

studies considering their evidence level low or insufficient

animal studies

studies on risk or effectiveness of OA therapies

studies on genetic susceptibility for OA

studies on OA patients to identify risk for other disease



Publications

Initially we retrieved 393 publications and after applying inclusion and exclusion criteria, we retained the following:

Freiberg A, Bolm-Audorff U, Seidler A. The Risk of Knee Osteoarthritis in Professional Soccer Players. Dtsch Arztebl Int. 2021 Jan 29;118(4):49-55. doi: 10.3238/arztebl.m2021.0007

Nie D, Yan G, Zhou W, Wang Z, Yu G, Liu D, Yuan N, Li H. Metabolic syndrome and the incidence of knee osteoarthritis: A meta-analysis of prospective cohort studies. PLoS One. 2020 Dec 23;15(12):e0243576. doi: 10.1371/journal.pone.0243576.

Wang X, Perry TA, Arden N, Chen L, Parsons CM, Cooper C, Gates L, Hunter DJ. Occupational Risk in Knee Osteoarthritis: A Systematic Review and Meta-Analysis of Observational Studies. Arthritis Care Res (Hoboken). 2020 Sep;72(9):1213-1223. doi: 10.1002/acr.24333.

Wang J, Dong J, Yang J, Wang Y, Liu J. Association between statin use and incidence or progression of osteoarthritis: meta-analysis of observational studies. Osteoarthritis Cartilage. 2020 Sep;28(9):1170-1179. doi: 10.1016/j.joca.2020.04.007.

Poulsen E, Goncalves GH, Bricca A, Roos EM, Thorlund JB, Juhl CB. Knee osteoarthritis risk is increased 4-6 fold after knee injury - a systematic review and meta-analysis. Br J Sports Med. 2019 Dec;53(23):1454-1463. doi: 10.1136/bjsports-2018-100022.

Sun Y, Nold A, Glitsch U, Bochmann F. Hip Osteoarthritis and Physical Workload: Influence of Study Quality on Risk Estimations-A Meta-Analysis of Epidemiological Findings. Int J Environ Res Public Health. 2019 Jan 24;16(3):322. doi: 10.3390/ijerph16030322.

Zhang YM, Wang J, Liu XG. Association between hypertension and risk of knee osteoarthritis: A meta-analysis of observational studies. Medicine (Baltimore). 2017 Aug;96(32):e7584. doi: 10.1097/MD.0000000000007584.

Kong L, Wang L, Meng F, Cao J, Shen Y. Association between smoking and risk of knee osteoarthritis: a systematic review and meta-analysis. Osteoarthritis Cartilage. 2017 Jun;25(6):809-816. doi: 10.1016/j.joca.2016.12.020



Risk Factors

The above 8 publications produced 7 risk factors:

   knee injury --> knee osteoarthritis
   occupational exposure --> knee osteoarthritis
   occupational exposure --> hip osteoarthritis
   soccer playing --> knee osteoarthritis
   metabolic syndrome --> knee osteoarthritis
   hypertension --> knee osteoarthritis
   smoking --> knee osteoarthritis

­These risk factors are detailed in 28 different risk associations (OR= odds ratio, RR = relative risk)

   if knee injury = Anterior Cruciate Ligament injury
   then knee osteoarthritis OR = 4.2 (2.2 to 8.0)

   if knee injury = Meniscal Injury
   then knee osteoarthritis OR = 6.3 (3.8 to 10.5)
   if knee injury = (Meniscal Injury) AND (Anterior Cruciate Ligament Injury)
   then knee osteoarthritis OR = 6.4 (4.9 to 8.3)

   if job = agriculture
   then knee osteoarthritis OR = 1.64 (1.33 to 2.1)

   if job = metal worker
   then knee osteoarthritis OR = 1.85 (1.25 to 2.76)

   if job = building and construction
   then knee osteoarthritis OR = 1.63 (1.39 to 1.92)

   if job = floor and brick layer
   then knee osteoarthritis OR = 2.51 (1.79 to 3.52)

   if job = carpenter
   then knee osteoarthritis OR = 2.49 (1.66 to 3.74)

   if job = miner
   then knee osteoarthritis OR = 1.47 (1.11 to 1.95)

   if job = cleaner
   then knee osteoarthritis OR = 1.51 (1.14 to 2.01)

   if job = housework
   then knee osteoarthritis OR = 1.93 (1.31 to 2.84)

   if job = service worker
   then knee osteoarthritis OR = 1.79 (1.36 to 2.37)

   if job = craftsman
   then knee osteoarthritis OR = 1.56 (1.17 to 2.09)

   if occupational activity = lifting
   then knee osteoarthritis OR = 1.39 (1.22 to 1.59)

   if occupational activity = kneeling
   then knee osteoarthritis OR = 1.29 (1.05 to 1.57)

   if occupational activity = standing
   then knee osteoarthritis OR = 1.30 (1.09 to 1.53)

   if occupational activity = climbing
   then knee osteoarthritis OR = 1.49 (1.20 to 1.86)

   if occupational activity = squatting
   then knee osteoarthritis OR = 1.48 (1.21 to 1.81)

   if occupational activity = walking
   then knee osteoarthritis OR = 1.23 (1.01 to 1.52)

   if occupational activity = sitting
   then knee osteoarthritis OR = 0.77 (0.70 to 0.84) protective

   if occupational activity = lifting
   then hip osteoarthritis OR = 2.00 (1.34 to 2.99)

   if job = agriculture
   then hip osteoarthritis OR = 4.74 (2.84 to 7.89)

   if (sex=male AND soccer playing status = professional)
   then knee osteoarthritis OR = 2.25 (1.41 to 3.61)

    if (sex = female AND metabolic syndrome diagnosis = yes)
   then knee osteoarthritis RR = 1.23 (1.03 to 1.47)

   if hypertension diagnosis = yes
   then knee osteoarthritis OR = 1.49 (1.26 to 1.77)

   if (sex = male AND smoking status = smoker)
   then knee osteoarthritis OR = 0.69 (0.58 to 0.80) protective

   if (sex = female AND smoking status = smoker)
   then knee osteoarthritis OR = 0.89 (0.77 to 1.02) protective

   if statin administration = yes
   then osteoarthritis OR = 1.01 (0.96 to 1.05) no association

Limitations

Only meta-analyses were included, however, a wider search might reveal individual cohort studies of high quality. Search was limited only to the last five years, only to PubMed and primary search results, and ­ included only papers reporting risk for OA. Future work should expand to include any risk associated with OA and related conditions, diseases and comorbidities.




Software Design




High level system architecture.
Adapted from CARRE D.2.2, 2014 https://www.carre-project.eu/project-info/deliverables-2/




Implementation

Calculator is written in javascript. It uses the SPARQL endpoint provided by the CARRE risk factor repository to read the risk factors available.

It dynamically creates a web page to enter available risk factors, and produces the risk estimation.

OA risk factors identified by the above methodology were inserted in the CARRE repository via the provided risk entry interface.