You are here

  • Home
  • Archive
  • Volume 10, Issue 2
  • Frailty and emergency department utilisation in adults with systemic lupus erythematosus ≤65 years of age: an administrative claims data analysis of Medicaid beneficiaries

Article Text

Article menu
Epidemiology and outcomes
Original research
Frailty and emergency department utilisation in adults with systemic lupus erythematosus ≤65 years of age: an administrative claims data analysis of Medicaid beneficiaries
  1. Sarah B Lieber1,2,
  2. Musarrat Nahid3,
  3. Iris Navarro-Millán1,3,
  4. Mangala Rajan3,
  5. Sebastian E Sattui4 and
  6. Lisa A Mandl1,2
  1. 1Division of Rheumatology, Hospital for Special Surgery, New York City, New York, USA
  2. 2Medicine, Weill Cornell Medicine, New York City, New York, USA
  3. 3Medicine, Division of General Internal Medicine, Weill Cornell Medicine, New York City, New York, USA
  4. 4Medicine, Division of Rheumatology and Clinical Immunology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
  1. Correspondence to Dr Sarah B Lieber; LieberS{at}hss.edu

Abstract

Objective Frailty is a risk factor for adverse health in adults with SLE, including those <65 years. Emergency department (ED) utilisation is high in adults with SLE, but to our knowledge, whether frailty is associated with ED use is unknown. In a large administrative claims dataset, we assessed risk of ED utilisation among frail adults with SLE ≤65 years of age relative to non-frail adults ≤65 years of age with SLE.

Methods Using the MarketScan Medicaid subset from 2011 to 2015, we identified beneficiaries 18–65 years with SLE (≥3 SLE International Classification of Diseases, Ninth Revision codes ≥30 days apart). Comparators without a systemic rheumatic disease (SRD) were matched 4:1 on age and gender. Frailty status in 2011 was determined using two claims-based frailty indices (CFIs). We compared risk of recurrent ED utilisation among frail and non-frail beneficiaries with SLE using an extension of the Cox proportional hazard model for recurrent events data.

Results Of 2262 beneficiaries with SLE and 9048 non-SRD comparators, 28.8% and 11.6% were frail, respectively, according to both CFIs. Compared with non-frail beneficiaries with SLE, frail beneficiaries with SLE had significantly higher hazard of recurrent ED use (HR 1.75, 95% CI 1.48 to 2.08).

Conclusion Frailty increased hazard of recurrent ED visits in frail adults ≤65 years of age with SLE relative to comparable non-frail adults with SLE. Frailty is a potential target for efforts to improve quality of care in SLE.

  • Lupus Erythematosus, Systemic
  • Outcome Assessment, Health Care
  • Quality Indicators, Health Care

Data availability statement

Data may be obtained from a third party and are not publicly available. Data are available from MarketScan. Restrictions apply to the availability of these data, which were used under license for the current study.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/lupus-2023-000905

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Frailty, an emerging risk factor for adverse health outcomes in SLE, is associated with increased healthcare utilisation; however, to our knowledge, whether frailty specifically predicts emergency department (ED) visits is unknown.

    WHAT THIS STUDY ADDS

    • We found that frailty was associated with an increased hazard of ED visits in adults ≤65 years of age with SLE.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • Frailty is a potential target for efforts to improve quality of care in SLE.

    Introduction

    Frailty, a syndrome of decreased homeostatic reserve, is present in up to 81% of patients with SLE and predicts increased healthcare utilisation, including in those <65 years of age.1 2 In a prospective multinational cohort with mean age of 36 years, frailty was associated with more frequent hospital admissions and higher mortality in SLE, including after adjustment for age and physician-reported organ damage.2 3

    Emergency department (ED) utilisation is an important target for quality improvement, and ED utilisation is disproportionately high among patients with SLE. Twenty-six per cent to 49% of adults with SLE visit the ED ≥1 time annually,4–8 and an additional 10% visit the ED ≥3 times annually.4 Higher self-report disease activity and coverage under Medicaid have been identified as significant predictors of frequent ED utilisation in patients with SLE;4 patients with SLE with higher ED utilisation also have been found to be younger with worse overall health and more symptoms of depression than their counterparts with fewer ED visits.4 As increased mortality has been observed in patients with SLE with higher acute care needs,9 further characterisation of those at risk of ED utilisation, including among young and mid-aged adults, is warranted. To our knowledge, whether frailty, a potentially modifiable risk factor, predicts increased ED utilisation in SLE is unknown.

    We aimed to assess risk of recurrent ED utilisation among frail adults ≤65 years of age with SLE relative to age-matched and gender-matched non-frail adults with SLE. We also examined sociodemographic features and medication use in frail and non-frail adults with SLE.

    Methods

    Study design

    This is a retrospective longitudinal study of US Centers for Medicare and Medicaid Services (CMS) Medicaid beneficiaries from the 11 unspecified states included in the MarketScan dataset enrolled from 2011 to 2015.

    Patients

    We identified Medicaid beneficiaries with SLE 18–65 years of age who in 2011 had ≥3 International Classification of Diseases, Ninth Revision (ICD-9) codes for SLE (710.0) from hospital discharge or physician visit claims, with each claim separated by ≥30 days;10 all were continuously enrolled in Medicaid, but not enrolled in Medicare to avoid missing claims among dually enrolled individuals. Medicaid is a US programme that provides federal assistance administered at the state level for some individuals with limited financial resources.11

    Comparators

    We identified age-matched and gender-matched comparators without ICD-9 codes for systemic rheumatic disease (SRD) (SLE, rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, systemic sclerosis, dermatomyositis, polymyositis and vasculitis) in a 1:4 ratio. Like beneficiaries with SLE, all were continuously enrolled in Medicaid, but not enrolled in Medicare.

    Exclusion criteria

    We excluded beneficiaries who had an ICD-9 code for inflammatory bowel disease,12 cancer,13 HIV,14 15 end-stage renal disease16 or organ transplantation17 18 to minimise potential confounding, since each of these conditions is known to be associated with frailty.

    Exposure

    Frailty in 2011 was operationalised using two validated ICD-9 claims-based frailty indices (CFIs).19 The frailty algorithm by Segal et al includes variables such as age, race, impaired mobility, recent admission, falls and comorbid conditions, including depression, and has been shown to identify individuals with the Fried frailty phenotype (online supplemental table 1);19 20 in brief, the Fried frailty phenotype consists of five components (weight loss, weakness, fatigue, slow gait and low physical activity), with frailty defined by presence of at least three components. As the frailty algorithm developed by Segal et al includes participant race, which was unavailable for commercial insurance beneficiaries in the MarketScan dataset, we focused our analysis on Medicaid beneficiaries. The second frailty algorithm by Kim et al includes variables such as durable medical equipment, assistive devices, nursing facility care, transportation and comorbid conditions, including depression, and was developed using an accumulation of deficits approach (online supplemental table 2).21 The CFI developed by Kim et al, but not Segal et al, contains administrative codes for SLE. Both CFIs have been validated in Medicare beneficiaries ≥65 years of age. While, to our knowledge, these CFIs have not been validated specifically in Medicaid beneficiaries <65 years of age, non-claims-based frailty measures based on phenotypical22 23 and accumulation of deficits2 3 24 25 approaches have been applied previously in observational cohorts including participants with SLE <65 years of age.

    Supplemental material

    We used previously published distributional cut-off points to define frailty status according to each CFI:26 robust (≤10th percentile); well (10th–25th percentile); vulnerable (26th–75th percentile); frail (76th–90th percentile) and very frail (>90th percentile). We then collapsed the ordinal categories into two groups: frail (‘frail’ and ‘very frail’) and non-frail (‘robust’, ‘well’ and ‘vulnerable’). We additionally assigned all beneficiaries into one of three non-overlapping groups: frail (frail according to both CFIs), non-frail (non-frail according to both CFIs) or frail discordant (frail according to one CFI only).

    Outcome

    We extracted subsequent ED visit data from 1 January 2012 to 30 September 2015. A beneficiary could have multiple ED visits in the follow-up period, referred to henceforward as recurrent events. Our outcome was time to each ED visit for all ED visits in the follow-up period. Beneficiaries were censored at the date of death, end of coverage, or end of study.

    Sociodemographic variables and SLE-related medication use

    Age, gender and race were extracted from the CMS files; ethnicity was unknown for 96% of participants and could not be reported reliably. We also evaluated whether hydroxychloroquine, glucocorticoid, and immunosuppressive and immunomodulatory medications typically used to treat SLE (including azathioprine, belimumab, cyclophosphamide, ciclosporin, methotrexate, mycophenolate mofetil, mycophenolic acid, rituximab, tacrolimus and tocilizumab) differed between frail and non-frail beneficiaries with SLE.

    Statistical methods

    We used descriptive statistics to characterise the cohort. To compare subgroups, Χ2 (or Fisher’s exact) and Kruskal-Wallis tests were used for categorical and continuous variables, respectively. Agreement between CFIs was evaluated with a kappa statistic.

    We examined risk of ED utilisation over time among the following groups: non-frail with SLE (referent), frail with SLE, frail discordant with SLE, as well as non-frail without SRD, frail without SRD, and frail discordant without SRD. We used Andersen and Gill’s counting process approach, an extension of the Cox proportional hazards model for analysing recurrent events survival data (with robust SEs to account for within-person correlation), to evaluate the risk of recurrent ED visits among frail beneficiaries with SLE as compared with non-frail beneficiaries with SLE, adjusting for baseline hydroxychloroquine, glucocorticoid, and immunosuppressive and immunomodulatory medication use as a proxy for disease severity.27 As sociodemographic features and comorbid conditions were included in the CFIs, we chose to adjust only for SLE medication use to avoid overfitting. Hydroxychloroquine, glucocorticoid and immunosuppressive medication use has been found previously to differ by disease severity in adults with SLE in an administrative claims dataset linked to an electronic medical record.28 All analyses were done using SAS V.9.4 and Stata V.16 (College Station, Texas, USA).

    Patient and public involvement

    Patients and the public were not involved in the design, conduct, reporting or dissemination plans of this administrative claims data analysis.

    Results

    Study sample

    We identified 2262 adult beneficiaries ≤65 years of age with SLE who were matched 1:4 with 9048 age-matched and gender-matched comparators without SRD meeting our inclusion criteria (figure 1).

    Figure 1
    Figure 1

    Flow diagram of beneficiaries with SLE and age-matched and gender-matched comparators without systemic rheumatic disease (SRD).

    Baseline frailty prevalence in beneficiaries with SLE was higher than in beneficiaries without SRD according to both CFIs (Segal et al: 38.3% vs 21.6%; Kim et al: 50.5% vs 18.6%); 28.8% of beneficiaries with SLE and 11.6% of comparators without SRD were frail according to both CFIs (table 1). There was moderate statistically significant agreement between CFIs (κ=0.47, p<0.01). Frailty prevalence increased with age in beneficiaries with and without SLE who were frail according to both CFIs (online supplemental table 3).

    View this table:
    Table 1

    Frailty prevalence in beneficiaries with SLE and age-matched and gender-matched comparators without systemic rheumatic disease (SRD)

    Beneficiaries with SLE who were frail according to both CFIs were older (median 51.5 years vs 39.5 and 34.5 years, p<0.01) and less commonly on hydroxychloroquine (36.7% vs 50.2% and 60.2%, p<0.01) and immunosuppressive medication (18.3% vs 28.1% and 37.1%, p<0.01) at baseline than beneficiaries with SLE who were frail discordant or non-frail based on both CFIs (table 2); systemic glucocorticoid use was most common among frail-discordant beneficiaries (62.9%), followed by non-frail (60.6%) and frail (56.4%) beneficiaries (p=0.04). Beneficiaries without SRD who were frail according to both CFIs were older (median 54.5 years vs 52.5 years and 37.5 years, p<0.01) and more commonly on systemic glucocorticoids (29.0% vs 21.1% and 11.9%, p<0.01) at baseline than beneficiaries without SRD who were frail discordant or non-frail based on both CFIs. There was a higher proportion of Black or African American individuals among frail-discordant beneficiaries without SRD (33.1%) than frail (30.9%) or non-frail (28.1%) beneficiaries without SRD (p<0.01).

    View this table:
    Table 2

    Baseline sociodemographic features and SLE-related medication use in frail and non-frail beneficiaries with SLE and age-matched and gender-matched comparators without systemic rheumatic disease (SRD)

    Risk of ED utilisation by frailty classification

    We included 8766 beneficiaries with SLE and non-SRD comparators in the analysis. Rates of ED visits during the follow-up period were higher among beneficiaries with SLE who were frail according to both CFIs (4.19 visits per person-year, 95% CI 4.09 to 4.29) relative to non-frail beneficiaries with SLE (2.40 visits per person-year, 95% CI 2.33 to 2.46) and non-SRD comparators who were frail based on both CFIs (2.68 visits per person-year, 95% CI 2.62 to 2.75) (table 3).

    View this table:
    Table 3

    Rates for emergency department (ED) visits per person-year among frail and non-frail beneficiaries with SLE and non-systemic rheumatic disease (SRD) comparators during the follow-up period

    Frail beneficiaries with SLE had 1.75 (95% CI 1.48 to 2.08) times higher hazard of recurrent ED utilisation than non-frail beneficiaries with SLE (table 4). Hazard of recurrent ED utilisation among frail beneficiaries with SLE compared with non-frail beneficiaries with SLE remained similar following adjustment for hydroxychloroquine, glucocorticoid, and immunosuppressive and immunomodulatory medication use (HR 1.65, 95% CI 1.40 to 1.95).

    View this table:
    Table 4

    Association of baseline frailty in beneficiaries with SLE and non-systemic rheumatic disease (SRD) comparators with recurrent emergency department visits

    Discussion

    In this longitudinal analysis of Medicaid beneficiaries, presence of both SLE and frailty at baseline was associated with higher hazard of ED utilisation than SLE alone. This observation suggests that frailty may be a risk factor for ED utilisation in SLE and a potential target for efforts to improve health-related quality of life and quality of care in SLE.

    Frailty prevalence in patients with SLE has been assessed in single and multicentre cohorts including young and mid-aged patients with SLE. In a single-centre longitudinal cohort of 152 women with prevalent SLE with mean age of 48 years, frailty, defined by the Fried phenotypical definition of frailty, was present in 20% of patients at baseline.22 In a cross-sectional analysis of a second single-centre longitudinal cohort of 67 women with prevalent SLE, 18% and 27% of participants were frail according to the Fried phenotype and the self-report FRAIL scale, respectively.23 29 In a multinational longitudinal inception cohort of 1683 younger patients with SLE (mean age of 36 years), frailty according to the Systemic Lupus International Collaborating Clinics Frailty Index (SLICC-FI), an SLE-related frailty index developed using an accumulation of deficits approach, was found in 27% of patients with SLE.30 Frailty according to the SLICC-FI was present at baseline in 29%, 30% and 36% of patients with prevalent SLE with mean age of 35.1, 45.2 and 43.3 years in three additional longitudinal cohorts, respectively.25 31 32 In contrast, SLICC-FI frailty was observed in 6% and 81% of two recently described samples of patients with prevalent SLE with mean age of 47.6 and 37.1 years, respectively.1 33

    In our study, we conservatively chose to designate participants as frail only if they were in the top quartile of frailty according to both CFIs. This definition more closely approximated frailty prevalence in similar populations according to the disease-agnostic Fried phenotype and FRAIL scale and the SLE-specific SLICC-FI (18%–36%) than frailty prevalence estimates based on either the Segal et al CFI or the Kim et al CFI alone (28.5% vs 38.3% vs 50.4%). Requiring a ‘dual diagnosis’ of frailty using both definitions may be the most specific when identifying frailty in beneficiaries with SLE in administrative datasets. However, incidence of ED visits and risk of recurrent ED utilisation were similar regardless of whether beneficiaries with SLE were classified as frail based on one or both CFIs. The relative performance of these two validated CFIs in other longitudinal SLE cohorts and their sensitivity to change over time require further study, as, to our knowledge, they have not been evaluated previously in those with SLE.

    In our study, hydroxychloroquine and immunosuppressive medication use was less common among frail than non-frail participants with SLE at baseline. Both CFIs include components, such as nephritis and recent infections, that may impact prescribing practices of hydroxychloroquine and immunosuppressive medication, which may be perceived as poorly tolerated or higher risk among frail individuals with SLE due to the presence of these components. Whether disease-directed therapies may be protective against frailty is beyond the scope of the current study; the potential role of SLE medications in mitigating frailty in young and mid-aged adults with SLE deserves further attention.

    Frailty is increasingly recognised as a risk factor for adverse health-related outcomes in SLE. Frailty according to the Fried phenotype and the FRAIL scale has been associated with self-report disability.22 23 In addition, frailty according to the SLICC-FI has been associated with organ damage accrual and more frequent hospital admissions2 24 25 31 and is an independent risk factor for mortality.3 Although, to our knowledge, SLE-specific frailty interventions have not been evaluated, evidence from the general population of older adults supports the use of physical activity programming, including resistance-based training, for mitigation of physical frailty.34

    ED utilisation is costly and associated with adverse health outcomes in patients with SLE.5 9 In a recent analysis of electronic healthcare record data, 29.3% of patients with SLE were found to have ≥1 ED visit in 2015, corresponding with median annual cost of ED-related care of $1023, up from 22.8% in 2011; 36.9% of patients with severe SLE visited the ED at least once over the study period from 2011 to 2015, corresponding with median annual cost of $2513 in 2015.5 Another study found that among Medicaid beneficiaries, 78.5% of patients with SLE had ≥1 ED visit as compared with 67.5% of matched comparators (p<0.001).35 Acute care use, including ED utilisation, has been associated with increased mortality in SLE among Medicaid beneficiaries;9 identifying patients with SLE at risk of increased ED use may allow earlier intervention and flag those in need of improved access to ambulatory care.

    Our study has several strengths. We analysed a large national sample of Medicaid beneficiaries, and we used a previously validated algorithm to identify those with SLE. We used two validated CFIs to identify frailty and conservatively required all frail beneficiaries to meet frailty criteria according to both CFIs.

    Our study has some limitations. We could not validate either CFI against the Fried frailty phenotype or the SLICC-FI in individuals with SLE; it is not known which frailty metric best captures frailty in younger adults with SLE. However, both CFIs have been validated in Medicare beneficiaries ≥65 years old. Importantly, the CFI developed by Kim et al, but not Segal et al, contains administrative codes for SLE, which is likely why the Kim et al CFI identified more frail beneficiaries with SLE than the Segal et al CFI. Given our time-to-event analysis and limited data on the sensitivity to change of CFIs,36 we did not consider frailty as a time-varying variable, though this would be a valuable future direction. As is the case with administrative claims data in general, data on disease severity were not available; however, based on prior studies, we tried to address this by adjusting for SLE-specific medications as a proxy for disease severity.28 37 We were unable to use an existing claims-based disease severity algorithm due to overlapping domains with the CFIs and our outcome.6 In addition, limited data were available on social determinants of health and geographical distribution, which may impact the relationship between frailty and ED use. Findings drawn from Medicaid beneficiaries with SLE may not be generalisable to the broader population with SLE.38 Our findings also are not generalisable to adults with SLE with comorbid conditions listed among the exclusion criteria, including end-stage renal disease. Although end-stage renal disease is an important potential complication of cumulative SLE organ damage,39 end-stage renal disease confers Medicare eligibility and thus is an uncommon comorbidity in these non-dually enrolled beneficiaries. In addition, differential frailty prevalence between beneficiaries with SLE and non-SRD comparators may be subject to informed presence bias,40 given greater contact with the healthcare system among individuals with SLE and availability of more diagnostic codes for CFI determination. Further, evaluation of reasons for ED visits among Medicaid beneficiaries was beyond the scope of the current study.

    Frailty was common in this large sample of young and mid-aged adults with SLE and may identify a distinct subset of vulnerable adults. Targeting behavioural or pharmacological interventions for frailty, such as through tailored physical activity programming,34 or individual frailty components may complement efforts to enhance health-related quality of life among patients with SLE. Frail patients with SLE also may benefit from quality improvement efforts to improve access to ambulatory care and overall healthcare outcomes.

    Data availability statement

    Data may be obtained from a third party and are not publicly available. Data are available from MarketScan. Restrictions apply to the availability of these data, which were used under license for the current study.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This study was deemed exempt by the Institutional Review Board of Weill Cornell Medicine (protocol number 19-07020475).

    References

    1. ↵
      1. Ugarte-Gil MF,
      2. Dubey J,
      3. McGwin G, et al
      . Association of systemic lupus International collaborating clinics frailty index with damage in systemic lupus erythematosus patients: results from a multiethnic, multicenter US cohort of patients with lupus. Arthritis Care Res (Hoboken) 2023;75:585–9. doi:10.1002/acr.24878
      OpenUrl
    2. ↵
      1. Legge A,
      2. Kirkland S,
      3. Rockwood K, et al
      . Prediction of hospitalizations in systemic lupus erythematosus using the systemic lupus International collaborating clinics frailty index. Arthritis Care Res (Hoboken) 2022;74:638–47. doi:10.1002/acr.24504
      OpenUrl
    3. ↵
      1. Legge A,
      2. Kirkland S,
      3. Rockwood K, et al
      . Evaluating the properties of a frailty index and its association with mortality risk among patients with systemic lupus erythematosus. Arthritis Rheumatol 2019;71:1297–307. doi:10.1002/art.40859
      OpenUrl
    4. ↵
      1. Panopalis P,
      2. Gillis JZ,
      3. Yazdany J, et al
      . Frequent use of the emergency department among persons with systemic lupus erythematosus. Arthritis Care Res (Hoboken) 2010;62:401–8. doi:10.1002/acr.20107
      OpenUrl
    5. ↵
      1. Murimi-Worstell IB,
      2. Lin DH,
      3. Kan H, et al
      . Healthcare utilization and costs of systemic lupus erythematosus by disease severity in the United States. J Rheumatol 2021;48:385–93. doi:10.3899/jrheum.191187
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Garris C,
      2. Jhingran P,
      3. Bass D, et al
      . Healthcare utilization and cost of systemic lupus erythematosus in a US managed care health plan. J Med Econ 2013;16:667–77. doi:10.3111/13696998.2013.778270
      OpenUrlCrossRefPubMed
    7. ↵
      1. Furst DE,
      2. Clarke A,
      3. Fernandes AW, et al
      . Resource utilization and direct medical costs in adult systemic lupus erythematosus patients from a commercially insured population. Lupus 2013;22:268–78. doi:10.1177/0961203312474087
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Li T,
      2. Carls GS,
      3. Panopalis P, et al
      . Long-term medical costs and resource utilization in systemic lupus erythematosus and lupus nephritis: a five-year analysis of a large Medicaid population. Arthritis Rheum 2009;61:755–63. doi:10.1002/art.24545
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Feldman CH,
      2. Xu C,
      3. Williams J, et al
      . Patterns and predictors of recurrent acute care use among Medicaid beneficiaries with systemic lupus erythematosus. Semin Arthritis Rheum 2020;50:1428–36. doi:10.1016/j.semarthrit.2020.02.012
      OpenUrl
    10. ↵
      1. Barbhaiya M,
      2. Feldman CH,
      3. Chen SK, et al
      . Comparative risks of cardiovascular disease in patients with systemic lupus erythematosus, diabetes mellitus, and in general Medicaid recipients. Arthritis Care Res (Hoboken) 2020;72:1431–9. doi:10.1002/acr.24328
      OpenUrl
    11. ↵
      1. Hoffman ED,
      2. Klees BS,
      3. Curtis CA
      . Overview of the Medicare and Medicaid programs. Health Care Financ Rev 2000;22:175–93.
      OpenUrlPubMed
    12. ↵
      1. Kochar B,
      2. Orkaby AR,
      3. Ananthakrishnan AN, et al
      . Frailty in inflammatory bowel diseases: an emerging concept. Therap Adv Gastroenterol 2021;14:17562848211025474. doi:10.1177/17562848211025474
    13. ↵
      1. Handforth C,
      2. Clegg A,
      3. Young C, et al
      . The prevalence and outcomes of frailty in older cancer patients: a systematic review. Ann Oncol 2015;26:1091–101. doi:10.1093/annonc/mdu540
      OpenUrlCrossRefPubMed
    14. ↵
      1. Levett TJ,
      2. Cresswell FV,
      3. Malik MA, et al
      . Systematic review of prevalence and predictors of frailty in individuals with human immunodeficiency virus. J Am Geriatr Soc 2016;64:1006–14. doi:10.1111/jgs.14101
      OpenUrlPubMed
    15. ↵
      1. Melo GC,
      2. Carvalho ACA,
      3. Mendes MLT, et al
      . Association between frailty phenotype, quantification of plasma HIV-1 RNA, CD4 cell count and HAART in HIV-positive subjects: a systematic review and meta-analysis of observational studies. AIDS Care 2022;34:1159–68. doi:10.1080/09540121.2021.1956414
      OpenUrl
    16. ↵
      1. Kojima G
      . Prevalence of frailty in end-stage renal disease: a systematic review and meta-analysis. Int Urol Nephrol 2017;49:1989–97. doi:10.1007/s11255-017-1547-5
      OpenUrl
    17. ↵
      1. Montgomery E,
      2. Macdonald PS,
      3. Newton PJ, et al
      . Frailty in lung transplantation: a systematic review. Expert Rev Respir Med 2020;14:219–27. doi:10.1080/17476348.2020.1702527
      OpenUrl
    18. ↵
      1. Quint EE,
      2. Zogaj D,
      3. Banning LBD, et al
      . Frailty and kidney transplantation: a systematic review and meta-analysis. Transplant Direct 2021;7:e701. doi:10.1097/TXD.0000000000001156
    19. ↵
      1. Segal JB,
      2. Chang H-Y,
      3. Du Y, et al
      . Development of a claims-based frailty indicator anchored to a well-established frailty phenotype. Med Care 2017;55:716–22. doi:10.1097/MLR.0000000000000729
      OpenUrlPubMed
    20. ↵
      1. Bandeen-Roche K,
      2. Xue Q-L,
      3. Ferrucci L, et al
      . Phenotype of frailty: characterization in the women’s health and aging studies. J Gerontol A Biol Sci Med Sci 2006;61:262–6. doi:10.1093/gerona/61.3.262
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Kim DH,
      2. Schneeweiss S,
      3. Glynn RJ, et al
      . Measuring frailty in Medicare data: development and validation of a claims-based frailty index. J Gerontol A Biol Sci Med Sci 2018;73:980–7. doi:10.1093/gerona/glx229
      OpenUrlPubMed
    22. ↵
      1. Katz PP,
      2. Andrews J,
      3. Yazdany J, et al
      . Is Frailty a relevant concept in SLE Lupus Sci Med 2017;4:e000186. doi:10.1136/lupus-2016-000186
    23. ↵
      1. Lieber SB,
      2. Nahid M,
      3. Paget S, et al
      . Evaluation of a patient-reported frailty tool in women with systemic lupus erythematosus. J Rheumatol 2022;49:60–7. doi:10.3899/jrheum.201466
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Legge A,
      2. Kirkland S,
      3. Rockwood K, et al
      . Prediction of damage accrual in systemic lupus erythematosus using the systemic lupus International collaborating clinics Frailty index. Arthritis Rheumatol 2020;72:658–66. doi:10.1002/art.41144
      OpenUrl
    25. ↵
      1. Lima K,
      2. Legge A,
      3. Hanly JG, et al
      . Association of the systemic lupus International collaborating clinics frailty index and damage accrual in long standing systemic lupus erythematosus. Arthritis Care Res (Hoboken) 2023;75:578–84. doi:10.1002/acr.24798
      OpenUrl
    26. ↵
      1. Kim DH,
      2. Patorno E,
      3. Pawar A, et al
      . Measuring frailty in administrative claims data: comparative performance of four claims-based frailty measures in the U.S. Medicare data. J Gerontol A Biol Sci Med Sci 2020;75:1120–5. doi:10.1093/gerona/glz224
      OpenUrl
    27. ↵
      1. Andersen PK,
      2. Gill RD
      . Cox’s regression model for counting processes: a large sample study. Ann Statist 1982;10:1100–20. doi:10.1214/aos/1176345976
      OpenUrl
    28. ↵
      1. Jiang M,
      2. Near AM,
      3. Desta B, et al
      . Disease and economic burden increase with systemic lupus erythematosus severity 1 year before and after diagnosis: a real-world cohort study, United States, 2004-2015. Lupus Sci Med 2021;8:e000503. doi:10.1136/lupus-2021-000503
    29. ↵
      1. Morley JE,
      2. Malmstrom TK,
      3. Miller DK
      . A simple frailty questionnaire (FRAIL) predicts outcomes in middle aged African Americans. J Nutr Health Aging 2012;16:601–8. doi:10.1007/s12603-012-0084-2
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Legge A,
      2. Kirkland S,
      3. Rockwood K, et al
      . Construction of a frailty index as a novel health measure in systemic lupus erythematosus. J Rheumatol 2020;47:72–81. doi:10.3899/jrheum.181338
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Legge A,
      2. Malone A,
      3. Hanly JG
      . External validation of the systemic lupus International collaborating clinics frailty index as a predictor of adverse health outcomes in systemic lupus erythematosus. Rheumatology (Oxford) 2022;61:1919–27. doi:10.1093/rheumatology/keab546
      OpenUrl
    32. ↵
      1. Elera-Fitzcarrald C,
      2. Gamboa-Cárdenas RV,
      3. Reatégui-Sokolova C, et al
      . The SLICC-FI predicts damage accrual in SLE patients. data from the almenara lupus cohort. Lupus 2022;31:1666–70. doi:10.1177/09612033221126855
      OpenUrl
    33. ↵
      1. Keramiotou K,
      2. Anagnostou C,
      3. Konstantonis G, et al
      . SLICC-frailty index is independently associated with impaired physical function, activities of daily living, and quality of life measures. Rheumatology (Oxford) 2022;61:3808–13. doi:10.1093/rheumatology/keac001
      OpenUrl
    34. ↵
      1. Dent E,
      2. Morley JE,
      3. Cruz-Jentoft AJ, et al
      . Physical frailty: ICFSR International clinical practice guidelines for identification and management. J Nutr Health Aging 2019;23:771–87. doi:10.1007/s12603-019-1273-z
      OpenUrlPubMed
    35. ↵
      1. Kan HJ,
      2. Song X,
      3. Johnson BH, et al
      . Healthcare utilization and costs of systemic lupus erythematosus in Medicaid. Biomed Res Int 2013;2013:808391. doi:10.1155/2013/808391
    36. ↵
      1. Shi SM,
      2. Steinberg N,
      3. Oh G, et al
      . Change in a claims-based frailty index, mortality, and Healthcare costs in Medicare beneficiaries. J Gerontol A Biol Sci Med Sci 2023:glad010. doi:10.1093/gerona/glad010
    37. ↵
      1. Navarro-Millán I,
      2. Xie F,
      3. Crowson CS, et al
      . Comparing cardiovascular risk of patients with rheumatoid arthritis within the social security disability insurance with those commercially insured. Arthritis Res Ther 2022;24:202. doi:10.1186/s13075-022-02847-1
    38. ↵
      1. Pryor KP,
      2. Barbhaiya M,
      3. Costenbader KH, et al
      . Disparities in lupus and lupus nephritis care and outcomes among US Medicaid beneficiaries. Rheum Dis Clin North Am 2021;47:41–53. doi:10.1016/j.rdc.2020.09.004
      OpenUrlPubMed
    39. ↵
      1. Anders H-J,
      2. Saxena R,
      3. Zhao M-H, et al
      . Lupus nephritis. Nat Rev Dis Primers 2020;6:7. doi:10.1038/s41572-019-0141-9
      OpenUrl
    40. ↵
      1. Goldstein BA,
      2. Bhavsar NA,
      3. Phelan M, et al
      . Controlling for informed presence bias due to the number of health encounters in an electronic health record. Am J Epidemiol 2016;184:847–55. doi:10.1093/aje/kww112
      OpenUrlCrossRefPubMed

    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.

    Footnotes

    • Presented at This work was presented as a poster at the American College of Rheumatology annual meeting in 2021.

    • Contributors All authors contributed to the concept and design of the work. MN and MR contributed to the statistical analysis. SBL drafted the manuscript. All authors critically revised the manuscript and approved the final version. SBL is responsible for the overall content as the guarantor.

    • Funding SBL was supported by a Scientist Development Award from the Rheumatology Research Foundation and a Michael D. Lockshin Fellowship from the Barbara Volcker Center for Women and Rheumatic Diseases at Hospital for Special Surgery. This study also was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health under award number P30AR072583. SBL was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under award number KL2TR002385 outside of the current work. IN-M was supported by the Rheumatology Research Foundation Innovative Research Award outside of the current work. SES was supported by the Rheumatology Research Foundation RISE Pilot Award and by the Bristol Myers Squibb Foundation Winn Career Development Award outside of the current work.

    • Disclaimer The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

    • Competing interests SES discloses research support from AstraZeneca and consulting for Sanofi (funds used for research support). LAM discloses research grants from Regeneron Pharmaceuticals, royalties from UpToDate, and salary support from Annals of Internal Medicine.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.