Article Text
PDF
Abstract
Objectives To investigate patterns in patient-reported and physician-reported disease outcomes in patients with rheumatoid arthritis (RA) from countries with different level of socioeconomic development.
Methods Data from a cross-sectional multinational study (COMOrbidities in RA) were used. Contribution of socioeconomic welfare (gross domestic product (GDP); low vs high) of country of residence to physician-reported (tender joint count, swollen joint count (SJC), erythrocyte sedimentation rate, disease activity score based on 28 joints assessment (DAS28)-3v based on these three components and physician global assessment) and patient-reported (modified Health Assessment Questionnaire (mHAQ), patient global assessment and fatigue) disease outcomes was explored in linear regressions, adjusting for relevant confounders.
Results In total, 3920 patients with RA from 17 countries (30 to 411 patients per country) were included, with mean age of 56 years (SD13) and 82% women. Mean SJC varied between 6.7 (Morocco) and 0.9 (The Netherlands), mean mHAQ ranged between 0.7 (Taiwan) and 1.5 (The Netherlands). Venezuela had the lowest (1.7) and the Netherlands the highest score on fatigue (5.0). In fully adjusted models, lower GDP was associated with worse physician-reported outcomes (1.85 and 2.84 more swollen and tender joints, respectively, and 1.0 point higher DAS28-3v), but only slightly worse performance-based patient-reported outcome (0.15 higher mHAQ), and with better evaluation-based patient-reported outcomes (0.43 and 0.97 points lower on patient global assessment and fatigue, respectively).
Conclusions In patients with RA, important differences in physician-reported and patient-reported outcomes across countries were seen, with overall a paradox of worse physician-reported outcomes but better patient-reported outcomes in low-income countries, while results indicate that these outcomes in multinational studies should be interpreted with caution. Research on explanatory factors of this paradox should include non-disease driven cultural factors influencing health.
- Rheumatoid Arthritis
- Outcomes research
- Epidemiology
Statistics from Altmetric.com
Background
Inequalities in health of patients with rheumatoid arthritis (RA) between countries with high and low welfare have been reported; however, evidence remains scarce.1 ,2 The few studies that compared health-related outcomes in patients with RA across countries mainly addressed the disease activity. Sokka et al1 showed in 6004 patients enrolled in the Quantitative Patient Questionnaires in Standard Monitoring of Patients with RA (QUEST-RA), an inverse relation between disease activity score based on 28 joints assessment (DAS28) and gross domestic product (GDP) per capita. Putrik et al confirmed this in fully adjusted analyses in a different multinational study.2 Another study compared fatigue in patients in UK and USA and revealed significant cultural differences in how patients and their environment perceive RA-related fatigue.3 For other core RA outcomes, including physician and patient global assessment, physical function,4 ,5 to our knowledge only data from (QUEST-RA) are available and analyses suggested that patients in less wealthy societies tended to score worse on all of these outcomes. Few studies outside RA have shown that reporting of pain depends on cultural background of patients.6–9
Commonly, disease outcomes are referred to as physician-reported (such as DAS28 and its components tender joint count (TJC), swollen joint count (SJC), erythrocyte sedimentation rate (ESR)) or patient-reported outcomes (eg, fatigue). In studies on inequalities in outcomes in RA across countries, it has never been explored whether physician-reported and patient-reported outcomes in patients with RA follow the same patterns across (groups of) countries. This is relevant when interpreting results from multinational studies and also to gain insight into factors contributing to different types of outcomes.
The aim of this study was to investigate the patterns in physician-reported and patient-reported health outcomes in patients with RA from countries with different level of socioeconomic development.
Methods
Study design and data collection
Data from a cross-sectional multinational (17 countries) observational study, COMOrbidities in RA (COMORA), were used.10 Consecutive outpatients were invited to enrol in the study if they were aged ≥18 years, fulfilled the 1987 American College of Rheumatology classification criteria for RA11 and were able to understand and complete the questionnaires that were administered.10
Data on demographic (age and gender), socioeconomic (education: primary, secondary or university education), comorbidities, disease duration (in years) and currently prescribed medication (synthetic and biological disease-modifying anti-rheumatic drugs (sDMARD and bDMARDs), non-steroidal anti-inflammatory drugs (NSAIDs) and glucocorticoids) were collected. From information on comorbidities the Rheumatic Diseases Comorbidity Index (RDCI) was computed.2 ,12 ,13
Information on GDP per capita (adjusted for purchasing power parity, measured in international dollars) was extracted from the International Monetary Fund for 2011.14 GDP was dichotomised in low-GDP and high-GDP countries, with a cut-off at 20 000 international $ per capita, which was the threshold that discriminated best between groups after visual data inspection.
Outcome variables
The following outcomes were investigated: physician-reported outcomes (TJC, SJC, ESR and DAS28 computed from these three assessments (DAS-3v) and physician global assessment) and patient-reported outcomes (modified Health Assessment Questionnaire (mHAQ; 0–3),15 ,16 patient global assessment (0–10) and fatigue (0–10)). Finally, the DAS28 including patient global (DAS28-4v) was considered as a mixed measure.17
Statistical analysis
Linear regression models were computed to explore the contribution of individual countries (reference was always the country with lowest mean on variable of interest) and GDP (low vs high) to each of the outcomes. All models were adjusted for potentially relevant confounders, including age (years), gender (female vs male), education (three categories) and comorbidities (continuous RDCI score) and repeated with and without adjustment for current RA medication (steroids (yes/no), NSAIDs and DMARDs (yes/no)) to evaluate potential impact of medication use on the outcome. When perception-based and evaluation-based measures were the outcomes, models were adjusted for TJC, SJC and ESR. Statistical software Stata V.12 was used.18
Results
In total, 3920 patients with RA from 17 countries (range 30–411) were included in COMORA (mean age 56 years (SD 13; 82% women; table 1). Mean SJC varied between 6.7 (Morocco) and 0.9 (The Netherlands), mean mHAQ ranged between 0.7 (Taiwan) and 1.5 (The Netherlands), Venezuela had the lowest and the Netherlands the highest average scores on fatigue (1.7 and 5.0, respectively; table 1 and figure 1). For details on missing data, see online supplementary table S1.
Characteristics of patients, overall and per country
Differences in swollen joint count (SJC; A), physical function (modified Health Assessment Questionnaire (mHAQ); B) and fatigue (C) across countries. Estimates are derived from models adjusted for age, gender, education, comorbidities and medication. The Y axis (or the deviation from the zero-line) presents the (adjusted) difference with the reference country (always the country with lowest mean on the variable of interest). Countries are ordered from lowest to highest gross domestic product (GDP) per capita (in international dollars), dashed line indicates border between high and low GDP group of countries.
Differences in outcomes across countries
In models with country of residence as independent variable and adjusted for all confounders including medication, large variations in outcomes across countries were observed (p value <0.001). Adjusted for potential confounders and compared with the Netherlands, SJC varied from +0.9 (95% CI 0.1 to 1.7) in Germany to +5.8 (5.0 to 6.6) in Morocco. Adjusted differences in mHAQ reached 0.7 points, from 0.2 (0.1 to 0.3) in Korea to 0.9 (0.8 to 1.0) in Morocco (compared with Taiwan) and adjusted fatigue (compared with Venezuela) 1.8 (1.3 to 2.3) in Spain to 3.8 (3.2 to 4.4) in the Netherlands (figure 1).
Differences in outcomes according to GDP
When replacing country by GDP in fully adjusted models, patients from low-GDP countries had on average 1.9 and 2.8 more swollen and tender joints, respectively, 11 mm/h higher ESR and 1.0 point higher DAS28-3v (table 2). Similarly, mHAQ was 0.15 higher in countries with low GDP. However, patients from low GDP societies had a 0.43 and 0.15 lower score on patient and physician global assessment, respectively, and 0.97 lower score on fatigue compared with high-GDP countries. For the mixed outcome DAS28-4v, low-GDP countries scored 0.94 points higher than the high-GDP countries. Change in coefficients after excluding medication from the models did not exceed 10% (table 2). Because of many missing values in physician global assessment in the Netherlands, we have repeated the regression models with this outcome excluding all the patients from the Netherlands and observed no substantial changes in the models. Disease duration was not a confounder in relationship between the GDP and the outcomes (data not shown). Secondary analyses using a random effects model gave essentially the same results (data not shown).
Association between performance, perception and evaluation-based outcomes with GDP, results from multivariable model
Discussion
The present study showed that patients from countries with lower socioeconomic welfare score worse on most of physician-reported measures for disease activity (SJC, TJC, ESR or DAS28 based on these three components) and on perceived difficulties in physical functioning (mHAQ). Differences in DAS28 seem to be highly relevant while magnitude of variation in mHAQ was less pronounced but not negligible.19 ,20 Paradoxically, patients in poorer societies evaluated their patient global assessment and fatigue similar to or better compared with wealthy countries, although the absolute differences were smaller.
Explaining the paradox is not straightforward. The higher disease activity in poorer countries was previously observed and attributed partly to availability of innovative biological DMARDs but also delay in diagnosis and access to rheumatological care may play a role.1 ,2 On this line, it is not unexpected that patients report more difficulties with daily tasks and activities (measured by mHAQ) as disease activity has a strong influence on execution of tasks. However, it would be expected that these patients would evaluate their global disease activity as being worse and would experience more fatigue.21 Clearly, this is not the case. One might also think that DMARDs induced fatigue, but results persisted after adjusting for current pharmacological treatment (including sDMARDs and bDMARDs), excluding this possibility. Apparently, non-disease factors play a role. Lower income societies may set less pressure with regard to execution of multiple social roles. Withdrawal from social roles due to sickness and reliance on extended family for care may be more acceptable than in wealthier but more individualistic societies. Moreover, in low-income countries, patients may be less informed about treatment options, have lower expectations from the treatment and more likely accept the achieved outcomes, while patients in high-income countries may be overoptimistic about the potential of the new treatments to improve all the disease outcomes.21 Physician's attitude may also, although unconsciously, be more optimistic in lower income countries as they live in an environment where the overall standards of health is lower. This can make them to judge the global health of patients slightly better and also to trigger higher satisfaction among their patients.
Interestingly, the differences between high-GDP and low-GDP countries in patient global assessment are higher than differences in physician global assessment, which were negligibly small. Since the models with perceived and evaluated outcomes were adjusted for TJC, SJC and ESR, it appears that physicians mostly base their judgement on these objective parameters, while patient's assessment is likely more subjective and takes other contextual and cultural factors into account.22
This study has several limitations. First, studies like COMORA cannot guarantee a representative sample of the patients with RA in participating countries. Centres with research interest in RA are more likely to participate, which leads to an over-representation of the optimally treated group in each country. Second, GDP is an aggregated measure (at country level) and is only a remote proxy for the countries socioeconomic welfare. However, it is commonly used in international studies. Next, cross-sectional design hinders conclusions about the direction of associations. Nonetheless, it is plausible to suggest that macro-economic indicators are relatively stable compared with health indicators and can be assumed to precede the outcomes in our study. It is important to note that differences between countries may go unnoticed for patients within the country, particularly if no large within country differences exist, and thus the value of international studies like COMORA should not be underestimated.
While differences in objective or performance-based measures across countries remain the greatest concern as they point at existing health inequities, variations in subjective evaluation-based outcomes are important and deserve further investigation. Future research should focus more on how non-disease-driven cultural and other contextual factors impact objectively measured and experienced heath. If confirmed, these results suggest to stratify international trial results for country or to examine the interaction between country and treatment effect.
Acknowledgments
The authors would like to thank the national principal investigators of the COMORA study: Gustavo Casado (Argentina), Josef Smolen (Austria), Bassel Kamal El-Zorkany (Egypt), Martin Soubrier (France), Gerd Burmester (Germany), Peter Balint (Hungary), Carlo Maurizio Montecucco (Italy), Masayoshi Harigai (Japan), Yeong-Wook Song (Korea), Najia Hajjaj-Hassouni (Morocco), Mart van de Laar (the Netherlands), Emilio Martin-Mola (Spain), Shue-Fen Luo (Taiwan), Jonathan Kay & Kevin Winthrop (USA), Gabriel Maciel (Uruguay), Anna Antunez (Venezuela); the scientific committee of COMORA study: Peter Balint, Gerd Burmester, Paul Emery, Jonathan Kay, Emilio Martin-Mola, Iain McInnes, Carlo Maurizio Montecucco and Josef Smolen; patients who agreed to participate in this study.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
- Data supplement 1 - Online supplement
Lay summary
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Handling editor Tore K Kvien
Contributors AB and SR have conceived the idea for the manuscript. APK has provided advice on statistical analyses. PP, SR, MH and AB have conducted the analyses and drafted the manuscript. All authors contributed to the interpretation of the results, reviewed and approved the manuscript before submission.
Funding Data collection was supported by an unrestricted grant from the Roche. The ancillary study described in this manuscript was done without any type of funding.
Competing interests None declared.
Patient consent Obtained.
Ethics approval The protocol for original COMORA study was reviewed and approved by all local institutional review boards or ethics committees (in each of 17 participating countries).
Provenance and peer review Not commissioned; externally peer reviewed.