In the Journals

Risk stratification tool predicts functional disability in patients at high risk for OA

Leena Sharma

Easily acquired data — such as education, depression symptoms, strenuous activity and obesity — can help identify functional decline risk among individuals before they develop osteoarthritis, according to a study published in the Annals of the Rheumatic Diseases.

“It is very challenging for patients and their providers to manage disability from knee OA after it has developed,” Leena Sharma, MD, of the Northwestern University Feinberg School of Medicine, told Healio Rheumatology. “This is because many helpful approaches like increasing physical activity and exercise are difficult — due to, for example, pain and damage to the joint from the OA — for persons who are disabled by knee OA.”

“We realized that it may be more powerful to start thinking about prevention of disability even earlier, such as in persons who are at higher risk for getting knee OA but who do not as yet have it,” she added. “However, individuals who have pre-OA are a very big population, and not all of them will go on to get disability. We need a way to identify, in this large population of pre-OA, who belongs to high-risk groups, that is, who is at higher risk for developing disability.”

To develop and validate a way to identify, among all those with pre-OA, who is at high risk for disability, Sharma and colleagues created risk stratification trees using data from the Osteoarthritis Initiative (OAI) and the Multicenter Osteoarthritis Study (MOST). The researchers included 1,870 participants from the OAI, used as a derivation cohort, and 1,279 from MOST, as a validation cohort.

 
Easily acquired data can help identify functional decline risk among individuals before they develop OA, according to a study.
Source: Adobe

The primary outcome was incident slow gait during 10 years of follow-up. The researchers used derivation cohort classification and regression tree analysis to identify risk predictors from a pool of 40 easily accessible variables including age, sex race, ethnicity, education, health insurance status, marital status and others. They developed risk stratification models, which they applied to the validation cohort. Sharma and colleagues used logistic regression to compare risk group predictive values, as well as area under the receiver operating characteristic curves to assess discrimination ability.

“Two types of trees emerged as best,” Sharma said in an interview. “In the simpler one, baseline age and knee function were the variables that were the basis of the tree’s branch points. This tree enabled us to divide the pre-OA population as a whole into three risk groups, two at various levels of high risk for the outcome and one at low risk. In the other tree, baseline age, knee function, education, strenuous activity, obesity and high depressive symptoms were the variables that were the basis of the tree’s branch points. This tree identified seven risk groups.”

The primary outcome occurred in 11% of studied participants, varying from 0% in the low-risk groups to 29% in some high-risk groups, in the derivation cohort, and from 2% to 23% in the validation cohort. Area under the receiver operating characteristic curves were comparable between the two cohorts.

“From these trees, we will develop simple questionnaires — short paper or electronic — that would identify if someone is in the pre-OA population and among them if they are at high risk for poor outcome,” Sharma said. “Filling this out can let an individual know about their future risk for impaired function. Knowing this can help to motivate them and their providers to take steps to prevent poor outcome.”

“There are several possible steps that can be taken to prevent disability in pre-OA,” she added. “Several papers suggest that the most cost-effective and scalable may include physical activity promotion. Awareness of risk at the pre-OA stage, not yet afflicted by knee OA, would be information at a point when these individuals are well enough to act and to perceive such action as a preservation of wellness. It is much more challenging to take these steps later when OA disease is already established in the knee.” – by Jason Laday

Disclosure: The researchers report no relevant financial disclosures.

Leena Sharma

Easily acquired data — such as education, depression symptoms, strenuous activity and obesity — can help identify functional decline risk among individuals before they develop osteoarthritis, according to a study published in the Annals of the Rheumatic Diseases.

“It is very challenging for patients and their providers to manage disability from knee OA after it has developed,” Leena Sharma, MD, of the Northwestern University Feinberg School of Medicine, told Healio Rheumatology. “This is because many helpful approaches like increasing physical activity and exercise are difficult — due to, for example, pain and damage to the joint from the OA — for persons who are disabled by knee OA.”

“We realized that it may be more powerful to start thinking about prevention of disability even earlier, such as in persons who are at higher risk for getting knee OA but who do not as yet have it,” she added. “However, individuals who have pre-OA are a very big population, and not all of them will go on to get disability. We need a way to identify, in this large population of pre-OA, who belongs to high-risk groups, that is, who is at higher risk for developing disability.”

To develop and validate a way to identify, among all those with pre-OA, who is at high risk for disability, Sharma and colleagues created risk stratification trees using data from the Osteoarthritis Initiative (OAI) and the Multicenter Osteoarthritis Study (MOST). The researchers included 1,870 participants from the OAI, used as a derivation cohort, and 1,279 from MOST, as a validation cohort.

 
Easily acquired data can help identify functional decline risk among individuals before they develop OA, according to a study.
Source: Adobe

The primary outcome was incident slow gait during 10 years of follow-up. The researchers used derivation cohort classification and regression tree analysis to identify risk predictors from a pool of 40 easily accessible variables including age, sex race, ethnicity, education, health insurance status, marital status and others. They developed risk stratification models, which they applied to the validation cohort. Sharma and colleagues used logistic regression to compare risk group predictive values, as well as area under the receiver operating characteristic curves to assess discrimination ability.

“Two types of trees emerged as best,” Sharma said in an interview. “In the simpler one, baseline age and knee function were the variables that were the basis of the tree’s branch points. This tree enabled us to divide the pre-OA population as a whole into three risk groups, two at various levels of high risk for the outcome and one at low risk. In the other tree, baseline age, knee function, education, strenuous activity, obesity and high depressive symptoms were the variables that were the basis of the tree’s branch points. This tree identified seven risk groups.”

The primary outcome occurred in 11% of studied participants, varying from 0% in the low-risk groups to 29% in some high-risk groups, in the derivation cohort, and from 2% to 23% in the validation cohort. Area under the receiver operating characteristic curves were comparable between the two cohorts.

“From these trees, we will develop simple questionnaires — short paper or electronic — that would identify if someone is in the pre-OA population and among them if they are at high risk for poor outcome,” Sharma said. “Filling this out can let an individual know about their future risk for impaired function. Knowing this can help to motivate them and their providers to take steps to prevent poor outcome.”

“There are several possible steps that can be taken to prevent disability in pre-OA,” she added. “Several papers suggest that the most cost-effective and scalable may include physical activity promotion. Awareness of risk at the pre-OA stage, not yet afflicted by knee OA, would be information at a point when these individuals are well enough to act and to perceive such action as a preservation of wellness. It is much more challenging to take these steps later when OA disease is already established in the knee.” – by Jason Laday

Disclosure: The researchers report no relevant financial disclosures.