Can the General Public Be a Proxy for an "At-Risk" Group in a Patient Preference Study? A Disease Prevention Example in Rheumatoid Arthritis

BACKGROUND

When selecting samples for patient preference studies, it may be difficult or impractical to recruit participants who are eligible for a particular treatment decision. However, a general public sample may not be an appropriate proxy.

OBJECTIVE

This study compares preferences for rheumatoid arthritis (RA) preventive treatments between members of the general public and first-degree relatives (FDRs) of confirmed RA patients to assess whether a sample of the general public can be used as a proxy for FDRs.

METHODS

Participants were asked to imagine they were experiencing arthralgia and had screening tests indicating a 60% chance of developing RA within 2 yrs. Using a discrete choice experiment, participants were offered a series of choices between no treatment and 2 unlabeled hypothetical treatments to reduce the risk of RA. To assess data quality, time to complete survey sections and comprehension questions were assessed. A random parameter logit model was used to obtain attribute-level estimates, which were used to calculate relative importance, maximum acceptable risk (MAR), and market shares of hypothetical preventive treatments.

RESULTS

The FDR sample (n = 298) spent more time completing the survey and performed better on comprehension questions compared with the general public sample (n = 982). The relative importance ranking was similar between the general public and FDR participant samples; however, other relative preference measures involving weights including MARs and market share differed between groups, with FDRs having numerically higher MARs.

CONCLUSION

In the context of RA prevention, the general public (average risk) may be a reasonable proxy for a more at-risk sample (FDRs) for overall relative importance ranking but not weights. The rationale for a proxy sample should be clearly justified.

HIGHLIGHTS

Participants from the general public were compared to first-degree relatives on their preferences for rheumatoid arthritis (RA) preventive treatments using a discrete choice experiment.Preferences were similar between groups in terms of the most important and least important attributes of preventive treatments, with effectiveness being the most important attribute. However, relative weights differed.Attention to the survey and predicted market shares of hypothetical RA preventive treatments differed between the general public and first-degree relatives.The general public may be a reasonable proxy for an at-risk group for patient preferences ranks but not weights in the disease prevention context; however, care should be taken in sample selection for patient preference studies when choosing nonpatients.

The Symptoms in Persons At Risk of Rheumatoid Arthritis (SPARRA) questionnaire: predicting clinical arthritis development

OBJECTIVE

There is a need to better define symptom characteristics associated with arthritis development in individuals at risk of rheumatoid arthritis (RA). We investigated whether reported symptoms in at-risk individuals could predict arthritis development and whether predictive symptoms differed between seropositive and seronegative at-risk individuals.

METHOD

At-risk individuals from four cohorts (Netherlands, UK, Sweden, and Switzerland) completed the Symptoms in Persons At Risk of Rheumatoid Arthritis (SPARRA) questionnaire. Participants had either (i) anti-citrullinated protein antibodies and/or rheumatoid factor, or (ii) relevant symptoms with or without RA antibodies. Follow up was ≥ 24 months or until clinical arthritis development. Stepwise forward selection created SPARRA prediction models for the combined group and for a seropositive subgroup.

RESULTS

Of 214 participants, the mean age was 50 years, 67% were female, and 27% (n = 58) developed clinical arthritis after a median time of 7 months. Four symptoms predicted arthritis development: self-reported joint swelling, joint pain moving from side to side (combined group only), feeling pins and needles in the joints, and often feeling fatigued (predicting non-arthritis).

CONCLUSION

Specific symptoms can provide useful information to estimate a person's RA risk. Differences in predictive symptoms between seropositive and seronegative at-risk individuals need to be further investigated. Future research is needed to determine whether changes in symptoms over time improve prediction and to determine the value of SPARRA in optimizing the selection of individuals who need to consult a rheumatologist.

Exploring preferences of at-risk individuals for preventive treatments for rheumatoid arthritis

OBJECTIVE

Some immunomodulatory drugs have been shown to delay the onset of, or lower the risk of developing, rheumatoid arthritis (RA), if given to individuals at risk. Several trials are ongoing in this area; however, little evidence is currently available about the views of those at risk of RA regarding preventive treatment.

METHOD

Three focus groups and three interviews explored factors that are relevant to first degree relatives (FDRs) of RA patients and members of the general public when considering taking preventive treatment for RA. The semi-structured qualitative interview prompts explored participant responses to hypothetical attributes of preventive RA medicines. Transcripts of focus group/interview proceedings were inductively coded and analysed using a framework approach.

RESULTS

Twenty-one individuals (five FDRs, 16 members of the general public) took part in the study. Ten broad themes were identified describing factors that participants felt would influence their decisions about whether to take preventive treatment if they were at increased risk of RA. These related either directly to features of the specific treatment or to other factors, including personal characteristics, attitude towards taking medication, and an individual's actual risk of developing RA.

CONCLUSION

This research highlights the importance of non-treatment factors in the decision-making process around preventive treatments, and will inform recruitment to clinical trials as well as information to support shared decision making by those considering preventive treatment. Studies of treatment preferences in individuals with a confirmed high risk of RA would further inform clinical trial design.

POS0591 TREATMENTS TO PREVENT RHEUMATOID ARTHRITIS IN FIRST DEGREE RELATIVES: DEMOGRAPHIC AND PSYCHOLOGICAL PREDICTORS OF RISK TOLERANCES

Background

There is a growing research focus on the development of interventions to reduce risk of rheumatoid arthritis (RA) in at-risk groups.(1) RA patients’ first-degree relatives (FDRs) have an elevated risk of developing RA and are potential candidates for preventive interventions. Recent studies have quantified the preferences of at risk groups for preventive treatments.(2-4) Little is known about predictors of preference heterogeneity in this context.

Objectives

Assess the extent to which FDR characteristics and beliefs predict risk tolerances for preventive treatments.

Methods

Adult FDRs of patients with confirmed RA in the UK were invited to take part in a web-based survey. FDRs enrolled in a UK prospective cohort (PREVeNT-RA) were also invited. Survey development, including attribute selection and presentation, was informed by qualitative research, ranking surveys, literature review, and expert opinion including patient research partners. Respondents received information about RA, questions to check comprehension, and an introduction to the survey. Participants were asked to imagine they were experiencing arthralgia and had positive autoantibody tests indicating a 60% chance of developing RA within two years. Using a probabilistic threshold technique, participants made choices between no treatment (no benefit and no risks) or a preventive treatment option. Treatment options were defined by a fixed level of benefit (reduction in risk of RA from 60% to 20%) and varying levels of risks (Table 1). For each treatment risk, participants made a series of choices where the risk was systematically increased or decreased until they switched their choice. This procedure was repeated for each of the remaining risks. Participants also completed items assessing demographics, perceived risk of developing RA, health literacy, subjective numeracy, the Brief Illness Perception Questionnaire (IPQ) and the Beliefs about Medicines Questionnaire General (BMQ-G). The maximum acceptable risk (MAR) respondents were willing to accept for a 40% (60% to 20%) point risk reduction in developing RA was summarized across participants using descriptive statistics. Associations between MARs and participants’ characteristics and illness/medication beliefs were assessed using interval regression. Independent variables were dichotomized and effects coded.

Table 1.

Attributes and levels of treatment options

Treatment attributeLevels describing no treatment optionLevels describing treatment optionChance of developing RA60%20%Chance of mild side effects0%2%; 4%; 5%; 7% or 10%Chance of a serious infection due to treatment0%1%; 1.5%; 2%; 3% or 5%Chance of a serious side effect that is potentially irreversible0%0.001%; 0.01%; 0.02%; 0.05% or 0.1%

Results

289 FDRs (80 male) responded. The mean (SE) MAR for mild side effects, serious infection, and serious side effects was 29.08 (1.52), 9.09 (0.60) and 0.85 (0.27), respectively. Participants aged over 60 years were less tolerant of risk of serious infection than average (mean MAR - 2.06 (0.78)) and younger participants were more tolerant of risk of serious infection than average (mean MAR + 2.06 (0.78)). Risk of mild side effects was less acceptable to participants who perceived they were likely/very likely to develop RA (mean MAR - 3.34 (1.55)) than to those who did not (mean MAR + 3.34 (1.55)). Education level, health literacy, numeracy, IPQ and BMQ-G subscales were not predictors of risk tolerance.

Conclusion

Age and perceived risk of RA had a significant impact on FDRs’ tolerance for specific, but not all, included risks. Cognitive ability and beliefs about RA/medicine did not explain preference heterogeneity. This is informative for drug development and the development of tailored risk communication resources to support preventive approaches.

References

[1]Mankia et al. Ann Rheum Dis. 2021;80(10):1286-98.

[2]Simons et al. Ann Rheum Dis. 2021;80:96-7.

[3]Harrison et al. Plos One. 2009; 14(4): e0216075.

[4]Finckh et al. Curr Rheumatol Rep. 2016;18: 51.

Acknowledgements

On behalf of the PREFER project. PREFER received funding from the IMI 2 Joint Undertaking (grant No. 115966), which receives support from the EU’s Horizon 2020 research and innovation program and European Federation of Pharmaceutical Industries and Associations (EFPIA). K. Raza is supported by the NIHR Birmingham Biomedical Research Centre.

Disclosure of Interests

Gwenda Simons: None declared, Ellen Janssen Shareholder of: Johnson & Johnson, Employee of: Janssen Research and Development, Jorien Veldwijk: None declared, Rachael DiSantostefano Shareholder of: Johnson & Johnson, Employee of: Janssen Research and Development, Matthias Englbrecht Speakers bureau: Abbvie, Chugai, Eli Lilly, Novartis, Roche, Sanofi, Mundipharma, Paid instructor for: Abbvie, Chugai, Roche, Consultant of: Abbvie, Novartis, Roche, Sanofi, Grant/research support from: Roche, Chugai, Christine Radawski Shareholder of: Eli Lilly, Employee of: Eli Lilly, Larissa Valor: None declared, Jenny Humphreys: None declared, Ian N. Bruce: None declared, Brett Hauber Shareholder of: Pfizer Inc., Employee of: Pfizer Inc., Karim Raza Consultant of: Abbvie, Sanofi, Grant/research support from: Bristol Myers Squibb, Marie Falahee: None declared.

POS0597 PREDICTORS OF PERCEIVED RISK IN FIRST DEGREE RELATIVES OF RHEUMATOID ARTHRITIS PATIENTS

Background

Risk of rheumatoid arthritis (RA) is 3-5 times higher in first-degree relatives (FDRs). Efforts are increasing to develop preventive interventions for this at risk group. Risk perception is a key predictor of health behaviours, including FDRs’ interest in predictive testing1 and preventive intervention for RA. Effective risk communication is essential for RA prevention studies and clinical translation2.

Objectives

To define variables associated with perceived risk of developing RA in FDRs of RA patients.

Methods

RA patients attending outpatient clinics (West Midlands, UK) were asked to invite their FDRs to complete a cross-sectional printed survey. Patients were also invited to complete a survey. Unique survey numbers enabled linkage of FDR and proband responses. FDRs’ perceived absolute risk, comparative risk, experiential risk, and worry about risk were assessed using 5-point Likert scales. Predictor variables included demographics, the Single Item Literacy Screener, Brief Illness Perception Questionnaire, Brief Approach/Avoidance Coping Questionnaire, Life Orientation Test Revised and the Short Health Anxiety Inventory. Patient predictors of FDR perceived risk included demographics, time since RA diagnosis and Rheumatoid Arthritis Impact of Disease Score.

Outcome measure responses were grouped into ‘low’ and ‘high’ for regression analyses. Univariable analysis used independent samples T-tests, chi-square tests and Mann-Whitney U Tests. Backwards stepwise binary logistic regression examined the relationship between FDR characteristics and perceived risk of RA. Generalised Estimating Equations assessed whether patient variables predicted FDR’s perceived risk.

Results

396 FDRs returned a survey. Paired data from 213 patients were available for 292 of these FDRs.

The distributions of risk perception scores are shown in Figure 1. All measures of perceived risk were inter-correlated (p<0.001; ranging from r=0.48 to r=0.80).

65.2% of FDRs perceived themselves to be ‘Likely’ or ‘Very Likely’ to develop RA in their lifetime. FDR’s ethnic group, deprivation index, employment status, education level, smoking status, cohabitation with index patient status, coping style and dispositional optimism were not significantly different between high and low perceived risk groups. Characteristics significant in univariable analyses were used in multivariable analyses (Table 1). Children were 3.89 times more likely than siblings to perceive themselves at high risk of RA. Higher health anxiety scores were associated with increased perceived risk. Female gender, and beliefs that RA would last a long time, and cause higher concern and negative emotional impact predicted increased risk perceptions. Higher perceptions of how well treatment would control RA was associated with a reduced likelihood of perceiving oneself at high risk.

Index patient characteristics did not associate with FDRs’ risk perceptions.

Table 1.

Multivariable analysis

FDR CharacteristicPerceived Risk [Odds Ratio (95% Confidence Interval)]AbsoluteComparativeExperientialWorry about RiskGender1.98 (1.19-3.27)* Male$ FemaleRelationship to index patient2.80 (1.70-4.61)***3.43 (2.04-5.78)***3.89 (2.24-6.75)***2.26 (1.30-3.94)** Child Sibling$Health Literacy1.50 (1.00-2.25)Illness Perceptions0.89 (0.79-0.99)*1.25 (1.05-1.48)*1.17 (0.99-1.38)0.84 (0.74-0.94)** Timeline0.91 (0.81-1.01)0.87 (0.77-0.98)*1.36 (1.13-1.63)** Treatment control1.20 (1.02-1.40)* Concern EmotionHealth Anxiety1.04 (1.01-1.07)*1.05 (1.01-1.07)*1.07 (1.03-1.11)**1.06 (1.02-1.09)**

*p<0.05, **p<0.01, ***p<0.001, $ Reference

Conclusion

FDRs’ perceived risk of RA was high. Key predictors included being a child of a patient with RA, higher health anxiety and lower perceptions of RA treatment control. An understanding of these predictors will inform the development of effective risk communication and preventive strategies.

References

[1]Wells et al. Rheumatology. 2021 doi: 10.1093/rheumatology/keab890

[2]Mankia et al. Ann Rheum Dis. 2021;80(10):1286-98

Acknowledgements

S. Bunnewell and I. Wells are joint first authors.

Disclosure of Interests

None declared

OP0276 PREFERENCES FOR TREATMENTS TO PREVENT RHEUMATOID ARTHRITIS: DISCRETE CHOICE SURVEY OF RHEUMATOID ARTHRITIS PATIENTS’ FIRST-DEGREE RELATIVES IN THE UNITED KINGDOM AND GERMANY

Background

There is a growing research focus on the development of interventions to reduce risk of rheumatoid arthritis (RA) in at-risk individuals.(1) A recent survey of the general population asked to assume a 60% risk of RA established that hypothetical preventive treatments were acceptable to most participants.(2) However the preferences of individuals who actually have an elevated risk of RA, such as first-degree relatives (FDRs) of RA patients, are not well understood.

Objectives

To quantify FDRs’ preferences for preventive treatments for RA.

Methods

Adult FDRs in the UK and Germany were invited to take part in a web-based survey via patients with clinician-confirmed RA either during a rheumatology clinic visit or by mail. In addition, FDRs taking part in a UK-based prospective cohort (PREVeNT-RA) were invited via email. Participants received information about RA followed by questions to check comprehension, and an introduction to the survey including warm-up questions. They were asked to imagine they were experiencing arthralgia and had positive autoantibody tests indicating a 60% chance of developing RA in the next two years. Using a discrete choice experiment, participants were offered a series of 15 choices between no treatment and two unlabeled hypothetical treatments to reduce risk of RA. Treatments were defined by six attributes with varying levels, describing benefits, risks, and frequency/route of administration (Table 1). Attribute selection and presentation was informed by qualitative research, ranking surveys, systematic literature review, and expert opinion. Survey layout was informed by patient research partners and qualitative pre-testing. A two-class latent class analysis was used to estimate preferences and calculate relative importance of treatment attributes and predicted uptake. A panel mixed logit model was used to obtain maximum acceptable risk estimates.

Table 1.

Treatment attributes and levels

AttributeLevelsChance of developing RA reduced from 60% to10%; 20%; 30%; 40%How the treatment is takenA shallow injection under the skinA drip into the veinOne or two tabletsHow often the medication has to be takenDailyWeeklyMonthlyEvery 6 monthsChance of mild side effects2%; 5%; 10%Chance of a serious infection due to treatment0%; 1%; 5%Chance of a serious side effect that is potentially irreversible1 in 100,000 people20 in 100,000 people100 in 100,000 people

Results

356 FDRs (252 female, 289 in the UK) responded. While treatment effectiveness was the most important attribute in both classes (Figure 1), the importance of other attributes differed between classes, with method and frequency of treatment administration being more important in class 2 and risk of mild side effects only impacting treatment choice in class 1. Perceived risk of developing RA predicted class assignment; those with higher perceived risk were more likely to belong to class 1. On average, the predicted uptake of treatment profiles estimating prevention candidates: abatacept; atorvastatin; hydroxychloroquine; tolerogenic cell-based therapy; and no treatment would be 50%, 15%, 9%, 18% and 0%, respectively. Finally, the maximum acceptable risk participants were willing to accept were 81%, 25% and 3% point increases in risk of mild side effects, serious infection, and serious side effects, respectively, for medicines that would reduce their risk of developing RA in the upcoming two years from 60% to 20%.

Conclusion

Effective preventive treatments for RA were acceptable to FDRs asked to assume a 60% chance of developing RA. Mode and frequency of treatment administration had a greater impact on treatment choices for participants with a lower perceived risk of RA. These findings are informative for target product profile development, endpoint selection, benefit-risk assessment, regulatory approval, and development of informational resources for those at risk of RA.

References

[1]Mankia et al. Ann Rheum Dis. 2021;80(10):1286-98.

[2]Simons et al. Ann Rheum Dis. 2021;80:96-7.

Acknowledgements

On behalf of the PREFER project. PREFER received funding from the IMI 2 Joint Undertaking (grant No. 115966), which receives support from the EU’s Horizon 2020 research and innovation program and European Federation of Pharmaceutical Industries and Associations (EFPIA). This abstract and its contents reflect the view of the presenter and not the view of PREFER, IMI, the European Union or EFPIA. K. Raza is supported by the NIHR Birmingham Biomedical Research Centre.

Disclosure of Interests

Gwenda Simons: None declared, Jorien Veldwijk: None declared, Rachael DiSantostefano Shareholder of: Johnson & Johnson, Employee of: Janssen Research and Development, Matthias Englbrecht Speakers bureau: Abbvie, Chugai, Eli Lilly, Novartis, Roche, Sanofi, Munidpharma, Paid instructor for: Abbvie, Chugai, Roche, Consultant of: Abbvie, Novartis, Roche, Sanofi, Grant/research support from: Roche, Chugai, Christine Radawski Shareholder of: Eli Lilly & Company, Employee of: Eli Lilly & Company, Larissa Valor: None declared, Jenny Humphreys: None declared, Ian N. Bruce: None declared, Karim Raza Consultant of: Abbvie, Sanofi, Grant/research support from: Bristol Myers Squibb, Marie Falahee: None declared

POS1483-HPR A QUANTITATIVE ASSESSMENT OF THE LIKELIHOOD THAT PATIENTS WITH RHEUMATOID ARTHRITIS WILL COMMUNICATE INFORMATION ABOUT RHEUMATOID ARTHRITIS RISK TO RELATIVES

Background:

First-degree relatives (FDRs) of patients with rheumatoid arthritis (RA) are at increased risk of RA and are being recruited to studies of predictive and preventive approaches. Access to FDRs is usually via the proband with RA. Qualitative investigations have shown that communication of RA risk to FDRs is a complex and selective process. However, quantitative data are needed to develop a robust understanding of this process and its determinants to develop effective communication strategies and support for RA patients and their relatives.

Objectives:

To identify predictors of the RA patients’ reported likelihood of communicating RA risk information to their FDRs.

Methods:

Patients with RA were invited to anonymously complete a survey assessing patient characteristics including age, gender, duration of RA, information seeking and decision making preferences (Autonomy Preference Index1), interest in FDRs taking a predictive test for RA, dispositional openness, and the general functioning subscale of the McMaster Family Assessment Device2. Reported likelihood of communicating RA risk to each of the patients’ FDRs was assessed using a Likert scale ranging from extremely unlikely (0) to extremely likely (4). The median score across all FDRs was calculated for each patient. The association between patient characteristics and the median score was examined using univariate approaches.

Results:

Surveys were completed by 482 patients (median age 65 years, 72% female). Most patients reported being likely to communicate RA risk to their FDRs (median score=3). Those who reported their likelihood of communicating RA risk to both children and siblings were more likely to communicate risk to children than siblings (p<0.001). No significant difference was found for FDRs’ gender, or patients’ gender, on reported likelihood to communicate about RA risk (p=0.32 and p=0.87, respectively). Patients who were older and who had had RA for longer were less likely to communicate about RA risk with their relatives. Patients who had higher interest in their FDRs taking a predictive test for RA, information seeking preferences, openness to communicate, and family functioning were more likely to communicate about risk (Table 1).

Table 1.

Descriptive statistics and Spearman’s rank-order correlations for the association between patients’ characteristics and their median reported likelihood of communicating to their FDRs about RA risk

Patient characteristicsDescriptive statisticsAssociation with median likelihood to communicate risk to relativesTest statisticP valueAge; median (IQR)65 (55-72)-0.19<0.001*RA duration; median (IQR)10 (4-20)-0.180.001*Autonomy preferences Information seeking; median (IQR)84 (75-97)0.26<0.001* Decision making; median (IQR)54 (42-67)0.090.048Interest in children taking a test; median (IQR)3 (2-3)0.44<0.001*Interest in siblings taking a test; median (IQR)2 (2-3)0.44<0.001*Openness; median (IQR)2 (1-3)0.130.004*Family functioning; median (IQR)2 (2-3)0.23<0.001*

* p values significant at the Bonferroni adjusted value of 0.006.

Conclusion:

Respondents were willing to communicate RA risk to their FDRs, and were more likely to communicate about risk to their children than their siblings. Factors including information seeking preferences, dispositional openness, interest in FDRs taking a predictive test, and family functioning were associated with increased likelihood of communicating RA risk information to FDRs. Patients’ age and RA duration were associated with decreased likelihood. These findings increase understanding of communication about RA risk in families, and will inform the development of informational approaches to support family communication and access to FDRs.

References:

[1]Ende et al. Journal of General Internal Medicine. 1989; 4(1): 23-30.

[2]Epstein et al. Journal of Marital and Family therapy.1983; 9(2): 171-180.

Acknowledgements:

This work was supported by Versus Arthritis; Grant reference: 21560 and Riksbankens Jubileumsfond (The Swedish Foundation for Humanities and Social Sciences); Grant reference: M13-0260:1 ‘Mind the Risk’.

Disclosure of Interests:

Imogen Wells: None declared, Peter Nightingale: None declared, Gwenda Simons: None declared, Christian Mallen Grant/research support from: Keele School of Medicine have received funding to support BMS recruit to a non-pharmacological atrial fibrillation trial., Karim Raza Grant/research support from: KR reports grants from Abbvie and Pfizer, and personal fees from Abbvie, Pfizer, Sanofi, Lilly, Bristol Myers Squibb, UCB, Janssen, and Roche Chugai, all outside the submitted work., M. Falahee: None declared

OP0160-HPR PREFERENCES FOR TREATMENTS TO PREVENT RHEUMATOID ARTHRITIS: DISCRETE CHOICE SURVEY OF GENERAL POPULATIONS IN UNITED KINGDOM, GERMANY, AND ROMANIA

Background:

There is increasing research focus on intervention for rheumatoid arthritis (RA) at the earliest stages of disease development, including treatment to prevent RA in at-risk groups. Novel cellular therapies are in development, and the effectiveness of existing immunomodulatory agents to prevent RA in those at risk is under investigation. Quantitative evidence of likely uptake of preventive treatments, and preferences for benefits and risks of such treatments is limited.

Objectives:

To quantify preferences for preventive therapies for RA.

Methods:

A web-based survey (n = 2959) was administered to an age- and gender- stratified sample of adults in the general population from online survey panels in the UK, Germany, and Romania. After receiving information about RA, questions to check comprehension of background information, an introduction to the survey tasks and warm-up questions, participants were asked to imagine that they were experiencing arthralgia (without swelling) and had positive autoantibody tests indicating a 60% chance of developing RA in the next two years. Using a discrete choice experiment with a Bayesian D-efficient design, participants were offered a series of 15 choices between no treatment and two unlabeled hypothetical treatments to lower risk of RA development. Treatments were defined by six attributes with varying levels including benefits, risks, and frequency/route of administration (Table 1). One choice task with fixed levels described treatments representative of those under investigation for RA prevention (abatacept, hydroxychloroquine, atorvastatin and tolerogenic cell-based therapy). Attribute selection and presentation was informed by previous qualitative research, ranking surveys, systematic literature review, and expert opinion. Survey design was informed by patient research partners. The survey was pre-tested during qualitative interviews and revised. A pilot of the final survey with 100 respondents was conducted to obtain priors for the final experimental design. Random parameters logit (RPL) models were used to estimate relative importance of treatment attributes and likely treatment uptake rates in each country.

Table 1.

Treatment attributes and levels

AttributeLevelsChance of developing RA reduced from 60% to10%; 20%; 30%; 40%How the treatment is takenA shallow injection under the skinA drip into the veinOne or two tabletsHow often the medication has to be takenDailyWeeklyMonthlyEvery 6 monthsChance of mild side effects2%; 5%; 10%Chance of a serious infection due to treatment0%; 1%; 5%Chance of a serious side effect that is potentially irreversible1 in 100,000 people20 in 100,000 people100 in 100,000 people

Results:

Across all three countries, effectiveness was the treatment attribute that had most impact on treatment choice (Figure 1). Method of administration was second most important for respondents from the UK and Romania but less important for German respondents. Risks of serious infection and serious side effects were more important determinants of treatment choice for respondents in Romania than they were in the UK and Germany. Percentage choice of fixed profiles reflecting abatacept, atorvastatin, hydroxychloroquine, tolerogenic cell-based therapy and no treatment differed across countries (χ2=78.90; p<0.001): 28.3%, 20.6%, 22.2% 18.5% and 10.4% respectively in the UK; 31.3%, 18.8%, 11.2%, 23.4% and 15.3% in Germany; and 27.6%, 20.5%, 15.8%, 21.7% and 14.4% in Romania.

Conclusion:

This study suggests that effective preventive treatments for RA are acceptable to members of the general population told to assume up a 60% chance of developing RA. The relative importance of treatment attributes and likely uptake of fixed treatment profiles differed across countries. These findings are informative for the design of prevention trials, and the development of informational resources and efficient preventive strategies for those at risk of developing RA.

Acknowledgements:

On behalf of the PREFER project. PREFER received funding from the IMI 2 Joint Undertaking (grant No. 115966), which receives support from the EU’s Horizon 2020 research and innovation program and European Federation of Pharmaceutical Industries and Associations (EFPIA). K. Raza is supported by the NIHR Birmingham Biomedical Research Centre.

Disclosure of Interests:

Gwenda Simons: None declared, Jorien Veldwijk: None declared, Rachael Di Santostefano Shareholder of: Johnson & Johnson, Employee of: Janssen R&D (of Johnson & Johnson), Matthias Englbrecht Speakers bureau: AbbVie, Chugai, Eli Lilly, Novartis, Roche, Sanofi, Mundipharma, Paid instructor for: AbbVie, Chugai, Roche, Consultant of: AbbVie, Novartis, Roche, Sanofi, Grant/research support from: Roche, Chugai, Christine Radawski Shareholder of: Eli Lilly & Company, Employee of: Eli Lilly & Company, Larissa Valor: None declared, Karim Raza Consultant of: Personal fees from Abbvie, Pfizer, Sanofi, Lilly, Bristol Myers Squibb, UCB, Janssen, and Roche Chugai, Grant/research support from: Abbvie and Pfizer, M. Falahee: None declared

OP0264-HPR PERSPECTIVES ON APPROACHES TO PREDICT THE DEVELOPMENT OF RHEUMATOID ARTHRITIS: A QUANTITATIVE ASSESSMENT OF PATIENTS AND THEIR FIRST DEGREE RELATIVES

Background:

There is considerable interest within the medical research community in the identification of individuals at risk of developing rheumatoid arthritis (RA), to identify those who may benefit from preventive interventions. However, it is important to understand the views of those who may be candidates for such predictive tests, to inform the development of effective approaches. First degree relatives (FDRs) of patients with RA are at an increased risk of developing RA. RA patients can provide access to FDRs. Qualitative investigations have explored the views of these groups about predictive testing for RA1,2, but quantitative approaches are needed to develop a robust understanding.

Objectives:

To identify predictors of interest in predictive testing for FDRs and patients, and to assess the likelihood of patients communicating information about RA risk to their FDRs.

Methods:

Surveys were completed by 482 RA patients and 397 of their FDRs. Patients were invited to complete the survey and to provide another to their relatives. Spearman’s Rank Correlations were used to assess relationships between interest in predictive testing/ likelihood of risk communication and potential predictor variables.

Results:

FDRs had a median age of 41 years, 64% were female. 57% were definitely interested and 36% were probably interested in taking a predictive test for RA. Several predictors were found to be associated with interest (table 1).

Table 1.

Spearman’s correlations for relatives’ and patients’ interest in predictive testing. After applying a Bonferonni adjustment, p values were taken as statistically significant at p≤0.003.

FDRsPatientsPredictors of interest in predictive testingrsPrsPBrief Illness Perception Questionnaire0.110.030.090.05 Consequences0.16*0.002*0.100.03 Timeline0.090.07-0.050.28 Personal control-0.030.59-0.020.68 Treatment control-0.020.760.020.74 Identity0.090.090.120.01 Concern0.21*<0.001*0.16*<0.001* Coherence0.110.030.0070.88 Emotional0.120.020.110.02Information Seeking0.35*<0.001*0.22*<0.001*Decision making-0.050.330.070.13Health literacy0.030.520.020.62Health numeracy-0.060.23-0.020.72Brief Avoidance Coping Questionnaire0.120.02-0.010.76Optimism0.060.26-0.070.12Health anxiety0.16*0.001*--Perceived risk0.37*<0.001*--Rheumatoid Arthritis Impact of Disease--0.050.31

– not applicable

Patients had a median age of 65 years, 71% were female. 47% were definitely interested and 30% were probably interested in their children taking a predictive test. Several predictors were found to be associated with interest (table 1). On a Likert scale from extremely unlikely (0) to extremely likely (4), most patients indicated that they were likely to communicate RA risk information to their children (median score=3).

Conclusion:

Interest in predictive testing for RA was high amongst FDRs, and factors including information seeking preference, RA risk perception, concern about RA, perceived consequences of RA and health anxiety were significantly associated with interest. Patients were also willing to communicate information about RA risk to their children. These findings increase understanding of perceptual variation in those at risk of RA, and will inform the development of information to support decision making in individuals considering predictive tests and preventive interventions. We are currently extending this preliminary analysis by building multivariate models incorporating a range of attitudes about predictive testing, assessing predictors of patients’ likelihood of communicating to their FDRs about risk, and the relationship between patients’ and FDRs’ responses.

References:

[1]Stack RJ et al. BMJ open. 2016; 6(6):e010555.

[2]Falahee M et al. Arthritis care & research. 2017; 69(10):1558-65.

Acknowledgments:

This work was supported by Versus Arthritis; Grant reference: 21560.

Disclosure of Interests:

Imogen Wells: None declared, Gwenda Simons: None declared, Rebecca Stack: None declared, Christian Mallen Grant/research support from: My department has received financial grants from BMS for a cardiology trial., Peter Nightingale: None declared, Karim Raza Grant/research support from: KR has received research funding from AbbVie and Pfizer, Consultant of: KR has received honoraria and/or consultancy fees from AbbVie, Sanofi, Lilly, Bristol-Myers Squibb, UCB, Pfizer, Janssen and Roche Chugai, Speakers bureau: KR has received honoraria and/or consultancy fees from AbbVie, Sanofi, Lilly, Bristol-Myers Squibb, UCB, Pfizer, Janssen and Roche Chugai, M. Falahee: None declared

OP0114 PREDICTING RHEUMATOID ARTHRITIS USING THE SYMPTOMS IN PERSONS AT RISK OF RHEUMATOID ARTHRITIS (SPARRA) QUESTIONNAIRE

Background:

Accurate prediction of rheumatoid arthritis (RA) development in persons at risk of RA can help to select individuals for early intervention trials. Currently, RA prediction mostly relies on biomarkers such as genetic factors, autoantibodies and imaging abnormalities, with symptoms being only a minor component1-3. However, at-risk individuals exhibit a high prevalence of diverse and often severe symptoms4,5and information on the predictive ability of individual symptoms or symptom complexes is still largely lacking.

Objectives:

We investigated the prevalence and predictive ability of symptoms in persons at-risk of RA, using the validated ‘Symptoms in Persons At Risk of Rheumatoid Arthritis’ (SPARRA) questionnaire.

Methods:

Individuals at-risk of RA from four different cohorts from the Netherlands (n=122), United Kingdom (N=90), Sweden (N=13) and Switzerland (N=20), were asked to fill out the SPARRA questionnaire, consisting of 69 questions (previously described6). Individuals were either defined as persons with anticitrullinated protein antibodies (ACPA) and/or rheumatoid factor (RF; n=193), having relevant symptoms (i.e. clinically suspect arthralgia with or without RA-specific antibodies, n=70) or being first degree relatives (FDR, n=20) of RA patients. All were followed for a minimum of 24 months or until clinical arthritis development. Univariable analyses were performed for possible predictor selection (p<0.2), followed by stepwise forward selection (by Cox regression, p<0.1) to create a multivariable prediction model.

Results:

The mean age of all participants was 49 years and 69% was female. In total, 56 persons (23%) developed clinical arthritis (n=22, 25, 7, 2 respectively in the 4 groups) after a median of 11.9 months (IQR 5.3 - 17.8). In total, 23 SPARRA questions were selected from the univariable analyses and entered in the stepwise forward selection procedure. Time to development of RA was predicted by the following questions: pain moving from joint to joint, having moderate or severe swelling in joints, feeling ≥1 days fatigue per month and feeling stiffness in joints of one and both feet (table 1).

Table 1.

Multivariable prediction model of SPARRA questions to predict clinical arthritis

HR (95% CI), pDoes your joint pain move from joint to joint?No; from arms to legs; from legs to arms (ref)1From ons side to the other2.96 (1.57; 5.57), p = 0.001Over the past month how much joint swelling have you had?None or mild (ref)1Moderate or severe3.04 (1.48; 6.25), p = 0.003Over the past month how many days of the month have you had fatigue?0 (ref)1≥ 10.32 (0.15; 0.67), p = 0.003Where did you feel joint stiffness?Neither feet (ref)One foot0.93 (0.42; 2.08), p = 0.865Both feet0.40 (0.17; 0.93), p = 0.032

Conclusion:

Asking persons at-risk of RA about joint pain, swelling and stiffness is common clinical practice. However, specific details such as pain moving from one side to the other or degree of joint swelling may provide useful additional information to estimate a person’s RA risk. The protective effect noted for fatigue and stiff feet may reflect an underlying pain syndrome rather than RA risk. We are currently performing analyses of the potential added value of SPARRA questions over the clinical prediction model by van der Stadt et al3which will help determine the final format of the SPARRA questionnaire.

References:

[1]de Hair MJ et al. Ann Rheum Dis. 2013;72(10):1654-8.

[2]Rakieh C et al. Ann Rheum Dis. 2015;74(9):1659-66.

[3]van de Stadt LA et al. Ann Rheum Dis. 2013;72(12):1920-6.

[4]Smolik I et al. J Rheumatol. 2013;40(6):818-24.

[5]Stack RJ et al. Arthritis Care Res (Hoboken). 2013;65(12):1916-26.

[6]van Beers-Tas MH et al. RMD Open. 2018;4(1):e000641.

Acknowledgments:

Supported by EULAR

Disclosure of Interests:

Laurette van Boheemen: None declared, Marieke ter Wee: None declared, M. Falahee: None declared, M.H. van Beers - Tas: None declared, Axel Finckh Grant/research support from: Pfizer: Unrestricted research grant, Eli-Lilly: Unrestricted research grant, Consultant of: Sanofi, AB2BIO, Abbvie, Pfizer, MSD, Speakers bureau: Sanofi, Pfizer, Roche, Thermo Fisher Scientific, Aase Hensvold: None declared, Karim Raza Grant/research support from: KR has received research funding from AbbVie and Pfizer, Consultant of: KR has received honoraria and/or consultancy fees from AbbVie, Sanofi, Lilly, Bristol-Myers Squibb, UCB, Pfizer, Janssen and Roche Chugai, Speakers bureau: KR has received honoraria and/or consultancy fees from AbbVie, Sanofi, Lilly, Bristol-Myers Squibb, UCB, Pfizer, Janssen and Roche Chugai, Dirkjan van Schaardenburg: None declared

A biomechanical investigation of short segment spinal fixation for burst fractures with varying degrees of posterior disruption

A biomechanical study was performed to evaluate the effectiveness of the Fixateur Interne pedicle screw system and the Syracuse I-Plate anterior fixation system. A total of 12 fresh frozen cadaver spines were tested intact, after burst fracture was created and application of a fixation device (six each), and after six serial transections of posterior ligaments and bony structures. Spines were loaded to a maximum of 10 N-m in flexion, extension, left and right lateral bending, and clockwise and counter-clockwise rotation. Results indicate that both systems reduce spinal flexibility in flexion, extension, and lateral bend loading when used to reduce and fix a classic burst fracture without posterior disruption. No decrease in flexibility was found in axial rotation for either device. After transection of all posterior elements, the I-Plate construct became much more flexible than the intact spine in flexion, extension, and axial rotation loading. The internal fixator construct retained more stability than the I-Plate construct after transection of posterior elements in flexion and extension loading, but was considerably more flexible than the intact spine in axial rotation loading. The results imply that the posterior internal fixator provides much better stabilization than the anterior I-Plate for those cases in which there is a large amount of posterior disruption in addition to an anterior burst injury. Neither device provides extensive support in axial rotation loading.