International Consensus for Provisions of Quality-Driven Care in Childhood-Onset Systemic Lupus Erythematosus

Authors


  • The contents of this study are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health or the Center for Clinical and Translational Science and Training.

Cincinnati Children's Hospital Medical Center, University of Cincinnati, MLC 4010, 3333 Burnet Avenue, Cincinnati, OH 45229-3039. E-mail: hermine.brunner@cchmc.org

Abstract

Objective

To obtain international consensus around processes that support the delivery of high-quality care to patients with childhood-onset systemic lupus erythematosus (SLE) based on current recommendations and scientific evidence.

Methods

To identify process quality indicators (QIs) for the medical care of children and adolescents with childhood-onset SLE, we sent 2 Delphi questionnaires internationally to 340 physicians who treat these patients. We set consensus at 80% of completed responses.

Results

Two hundred ninety-seven physicians (87%) responded to the first Delphi questionnaire and 265 physicians (76%) responded to the second questionnaire. The group achieved consensus for 26 QIs addressing laboratory testing at diagnosis, health maintenance measures, diagnosis and therapy of lupus nephritis, general preventive strategies, surveillance for medication safety, counseling and evaluation of cardiovascular risk factors, as well as transition planning. Of the 26 process QIs for use in childhood-onset SLE, 11 matched those established for adults with SLE, 9 required modification, and consensus was reached for an additional 6 QIs specific to children.

Conclusion

An international consensus for a set of process QIs for childhood-onset SLE was reached that considers unique aspects of children with childhood-onset SLE. The presented set of QIs for children and adolescents with childhood-onset SLE defines agreed-upon standards of medical care.

INTRODUCTION

One definition of quality of medical care is “the degree to which health services increase the likelihood of a desired outcome and are consistent with current professional knowledge” ([1]). In 2001, the US Institute of Medicine issued a report citing safety deficiencies in the American health care system stemming from a lack of metrics to assess the quality of patient care. This led to an increased focus on the development of measures of processes aimed at improving the quality of medical care delivered, known as quality indicators (QIs) ([2]). Different from clinical practice guidelines that aim at assisting with medical decision making ([3]), QIs are retrospectively measurable elements of practice performance for which there is evidence or consensus used to assess the quality of care provided ([4]). Several rheumatic diseases have defined process QIs, including juvenile idiopathic arthritis ([5]). The American College of Rheumatology (ACR) and the European League Against Rheumatism (EULAR) established QIs for adults with systemic lupus erythematosus (SLE) ([6-8]). However, similar consensus for the care of patients with childhood-onset SLE does not exist ([9]). This is a major concern, since nearly 20% of patients with SLE experience disease onset during childhood, and childhood-onset SLE is associated with greater lifetime morbidity and mortality than adult-onset SLE ([10]). Furthermore, there is a remarkable variation in medical care that cannot be explained by differences in childhood-onset SLE phenotypes ([11, 12]), suggesting that the distinct training and local experience of providers disproportionately influence treatment plans.

The objective of this research was to achieve international consensus around a set of evidence-based process QIs to help define minimum acceptable standards when delivering medical care to children and adolescents with childhood-onset SLE.

Box 1. Significance & Innovations

  • Based on international consensus among pediatric rheumatologists, applying quality indicators (QIs) proposed for adults with systemic lupus erythematosus (SLE) does not sufficiently capture issues important to the delivery of medical care in childhood-onset SLE.
  • Aspects of care that need special consideration in childhood-onset SLE include growth and development as well as transition of care.
  • The presented set of QIs for children and adolescents with childhood-onset SLE defines widely acceptable standards of medical care.

MATERIALS AND METHODS

The consensus process goal was to establish agreement on diagnostic, therapeutic, and preventive indicators to comprise an agreed-upon standard of care for patients with childhood-onset SLE that could eventually be adapted into quality measures, identifying the population to determine numerators and denominators. To synthesize the available medical expertise, we used structured consensus formation methodology, specifically Delphi questionnaires, to engage a broad range of expert opinions ([3, 13]). Delphi questionnaires are multistage surveys used to create consensus among a group of experts. The multiple surveys allow the surveyors to provide feedback to the experts as well as to clarify questions unanswered from previous survey responses. We invited the physician members of the Paediatric Rheumatology International Trials Organisation, the Pediatric Rheumatology European Society, the Childhood Arthritis and Rheumatology Research Alliance, and the pediatric section of the Pan American League of Associations for Rheumatology to respond to 2 Delphi questionnaires. These groups were approached to help us achieve feedback that would guide the development of international QIs. Respondents completed the Delphi questionnaires electronically, prompted as needed by 3 reminder e-mails to increase overall participation ([14]). We excluded physicians treating fewer than 3 patients with childhood-onset SLE per month and trainees.

First Delphi questionnaire

The initial questionnaire explored whether applying QIs for SLE developed by the ACR and EULAR was sufficient and if there was a need for development of additional QIs for childhood-onset SLE. We provided supporting materials and invited comments.

Literature review

Building on responses from the first questionnaire, we performed additional relevant review of medical literature published between January 2000 and March 2012. Two authors (MCH, HIB) independently screened and assessed articles and their references for relevance to the project, reviewing pertinent articles in detail. To be included, an article had to feature a recommendation or guideline by a professional medical organization, involve a clinical study that demonstrated an association between a care process and a defined outcome, or review evidence of diagnostic or therapeutic value in the pediatric population. We excluded abstract-only studies, editorials, and case reports. The Delphi questionnaire presented summaries of relevant evidence obtained from our literature review. These details are available in Supplementary Appendices A and B, (available in the online version of this article at http://onlinelibrary.wiley.com/doi/10.1002/acr.21998/abstract).

Second Delphi questionnaire

The second questionnaire briefly summarized relevant literature pertaining to childhood-onset SLE and adult-onset SLE for each quality domain as well as relevant responses from the first questionnaire. Quality domains considered were laboratory studies for suspected or confirmed childhood-onset SLE, general preventive management, medication monitoring, lupus nephritis and hypertension management, bone health with respect to supplementation and glucocorticoid use, ophthalmologic surveillance, and education on cardiovascular risk factors. The Delphi questionnaires are available in Supplementary Appendices C and D (available in the online version of this article at http://onlinelibrary.wiley.com/doi/10.1002/acr.21998/abstract).

Statistical analysis

We used LimeSurvey software (online at http://www.limesurvey.org) for Delphi response management and as a presentation layer. Using Microsoft Excel 2010, we exported and analyzed questionnaire data and reviewed responder comments to ensure proper interpretation. We set consensus at the standard level of 80% agreement ([15]).

RESULTS

Questionnaire respondents

Of 340 trained pediatric rheumatologists invited, 297 (87%) completed the initial Delphi questionnaire; 265 (76%) completed the second questionnaire (see Supplementary Appendix E, available in the online version of this article at http://onlinelibrary.wiley.com/doi/10.1002/acr.21998/abstract). The international distribution of the second questionnaire's respondents is shown in Table 1. Eighty-eight respondents (46%) were trained pediatric rheumatologists from North America, 68 physicians (36%) were from Europe, 25 (13%) were from Central and South America, and 5 (3%) were from Asia, Africa, and Australia. Seventy-nine participants did not provide demographic information. Respondents were experienced physicians, with more than half (52%) practicing for >10 years. Table 1 shows a summary of demographic information for the respondents to the Delphi questionnaires.

Table 1. Questionnaire respondent demographics*
 Total reported, no. (%)
  1. SLE = systemic lupus erythematosus.
  2. aIncludes noncertified pediatricians with rheumatology interest, pediatric immunologists, and adult rheumatologists with training in pediatric rheumatology.
  3. bReported in the first Delphi questionnaire. The count as reported by the respondent does not reflect the total number of participants.
  4. cThe count as reported by the respondent does not reflect the total number of participants.
Type of provider (n = 265) 
Pediatric rheumatologist173 (65)
Adult rheumatologist12 (5)
Othera80 (30)
Geographic region of practice (n = 265) 
North America88 (46)
Europe68 (36)
Central and South America25 (13)
Asia, Africa, and Australia5 (3)
Not reported79 (3)
Treatment of childhood-onset SLE experience 
Years of medical practice in childhood-onset SLE (n = 208)b 
1–525 (12)
6–1074 (36)
11–2057 (27)
>2052 (25)
No. of patients with childhood-onset SLE currently treated per month (n = 200)c 
1–10147 (74)
11–2028 (14)
21–3012 (6)
>3012 (6)

The first Delphi questionnaire achieved consensus for 5 of the 20 ACR process QIs and 6 of the 23 EULAR process QIs as suitable for childhood-onset SLE application without revision (Table 2). Conversely, 11 QIs approved for use in adults with SLE (8 ACR, 3 EULAR) required modification prior to use in childhood-onset SLE. The second questionnaire addressed 9 health care domains relevant to the medical care of childhood-onset SLE, which were proposed in the first questionnaire. Following further revision, the second questionnaire achieved consensus for 26 QIs for use in childhood-onset SLE (Table 2).

Table 2. Quality indicators for patients with childhood-onset SLE by domain*
  1. SLE = systemic lupus erythematosus; LN = lupus nephritis; WHO = World Health Organization; ISN/RPS = International Society of Nephrology/Renal Pathology Society; GFR = glomerular filtration rate; NSAIDs = nonsteroidal antiinflammatory drugs; DMARDs = disease-modifying antirheumatic drugs.
  2. aSuspected SLE is often defined as fulfilling 3 of 11 criteria for lupus diagnosis as defined by the American College of Rheumatology.
  3. bAdult SLE quality indicators (endorsed by the American College of Rheumatology or the European League Against Rheumatism) that achieved consensus for childhood-onset SLE.
  4. cOngoing proteinuria is defined as persistence of proteinuria for ≥3 months.
Laboratory testing around the time of diagnosis
1. IF a patient has suspected childhood-onset SLE, THEN the following laboratory studies should be obtained (see Table 3)a
2. IF a patient has confirmed childhood-onset SLE, THEN the following laboratory studies should be obtained (see Table 3)
General prevention
3. IF a patient has childhood-onset SLE, THEN vaccination against influenza and encapsulated organisms, including pneumococcus, meningococcus, and Haemophilus influenzae, should be prescribed, unless there are contraindications
4. IF a patient has childhood-onset SLE, THEN education about sun avoidance should be documented at least once in the medical record (e.g., wearing protective clothing, applying sunscreens whenever outdoors, and avoiding sunbathing)b
5. IF an adolescent has childhood-onset SLE, THEN a transition plan should be carefully designed to facilitate transfer of care to the appropriate adult health care providers
LN and hypertension management
6. IF a patient has a flare after having achieved remission of kidney disease, THEN diligent followup of renal disease is needed
7. IF a patient has newly diagnosed LN, THEN renal biopsy, urine sediment analysis, proteinuria, and kidney function should all be assessedb
8. IF a patient is diagnosed with proliferative childhood-onset SLE nephritis (WHO or ISN/RPS class III or IV), THEN therapy with corticosteroids combined with another immunosuppressive agent should be provided and documented within 1 month of this diagnosis, unless contraindicatedb
9. IF a childhood-onset SLE patient without known LN has developed daily proteinuria of >500 mg or clinically relevant worsening of GFR/urinary sediment, THEN a kidney biopsy should be performed
10. IF a patient has known LN, THEN a clinical assessment for childhood-onset SLE should occur at least every 3 months, regardless of disease activity
11. IF a childhood-onset SLE patient has LN plus evidence of ongoing proteinuria >500 mg/day, THEN an angiotensin-converting enzyme inhibitor or angiotensin receptor blockers should be prescribed, unless there are contraindicationsc
12. IF a patient has LN and/or hypertension, THEN disease comanagement with a nephrologist should be considered
Medication management
13. IF a patient is prescribed a new medication for childhood-onset SLE (e.g., NSAIDs, DMARDs, or glucocorticoids), THEN a discussion with the patient about the risks versus benefits of the chosen therapy should be documentedb
14. IF a patient has childhood-onset SLE, THEN antimalarial therapy should be prescribed, unless there are contraindications
15. IF a childhood-onset SLE patient is receiving a dose of steroids not acceptable for long-term use, then an attempt should be made to taper steroidsb
16. IF a patient with childhood-onset SLE is unable to decrease the dose of steroids acceptable for long-term use, THEN the addition of a steroid-sparing agent or an increased dose of an existing steroid-sparing agent should be consideredb
17. IF a childhood-onset SLE patient is treated with medications, THEN laboratory surveillance for medication safety should be done at regular intervals (details provided in Table 4)b
Bone health
18. IF a patient has received chronic systemic steroids, THEN the patient should have bone mineral density testing documented in the medical record
19. IF baseline bone mineral density testing is outside of the normal limits (Z scores of −2 or less), THEN bone mineral density should be remeasured after 1 year
20. IF a patient is receiving any steroid therapy, THEN calcium and vitamin D supplementation should be recommended after 3 months
Ophthalmologic surveillance
21. IF a childhood-onset SLE patient is treated with corticosteroids, THEN eye screening should be done at least annually
22. IF a childhood-onset SLE patient is treated with antimalarial therapy, THEN eye screening should be done at least annually
Education on cardiovascular risk factors
23. IF a patient has childhood-onset SLE, THEN education about cardiovascular risk factors should occur in regular intervals with the parent and the patient age ≥13 years (see Table 5)
24. IF a patient has childhood-onset SLE, THEN lifestyle modifications (smoking cessation, weight control, exercise) are likely to be beneficial for patient outcomes and should be encouragedb
Pregnancy
25. IF a patient with childhood-onset SLE is pregnant, THEN anti-SSA, anti-SSB, and antiphospholipid antibodies should be documented in the medical recordb
Neuropsychiatric manifestations
26. IF a patient with childhood-onset SLE has major neuropsychiatric manifestations (optic neuritis, acute confused state/coma, cranial or peripheral neuropathy, psychosis, and transverse myelitis/myelopathy), THEN immunosuppressive therapy should be consideredb

Laboratory testing around the time of diagnosis

One QI addressed the laboratory evaluation for suspected childhood-onset SLE and newly diagnosed childhood-onset SLE to assist with subsequent medical decision making (Table 3). Suspected SLE is often defined as fulfilling 3 of 11 ACR classification criteria for SLE ([16]). There was near consensus (79% of item responses) for antiphospholipid antibody testing in the setting of suspected childhood-onset SLE. The group achieved consensus on antiphospholipid antibody testing in all newly diagnosed patients with childhood-onset SLE (98%).

Table 3. Quality indicators of laboratory testing for suspected childhood-onset SLE and newly diagnosed childhood-onset SLE*
 Suspected childhood- onset SLENewly diagnosed childhood- onset SLE
  1. SLE = systemic lupus erythematosus; anti-dsDNA = anti–double-stranded DNA; BUN = blood urea nitrogen; AST = aspartate aminotransferase; ALT = alanine aminotransferase.
Complete blood count and differential
Erythrocyte sedimentation rate
C-reactive protein 
Complement C3, C4
Antinuclear antibodies
Anti-dsDNA antibodies
Anti-Ro, anti-La, anti-RNP antibodies
Urinalysis
Urine protein-to-creatinine ratio
Renal panel (may include sodium, potassium, HCO3, chlorine, BUN, creatinine)
Liver function (may include AST, ALT, bilirubin, albumin)
Antiphospholipid antibodies 

Lupus nephritis and hypertension management

There was consensus around 7 process QIs addressing diagnostic and therapeutic aspects of lupus nephritis. For patients without known lupus nephritis, indications for a kidney biopsy included 1) new occurrence of daily proteinuria >500 mg, 2) clinically relevant worsening glomerular filtration rate, or 3) an active urinary sediment that cannot be explained by any other pathology but childhood-onset SLE (93%). Based on respondent comments, there was consensus that the kidney function of a child with childhood-onset SLE should be monitored using serial blood and urine testing, once the patient is diagnosed. If diagnosed with proliferative lupus nephritis (class III or IV with/without class V overlap), treatment with systemic glucocorticoids combined with immunosuppressive agents should be started within 1 month of kidney biopsy (99%). In the setting of lupus nephritis with concomitant hypertension, the group achieved consensus for prescribing an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker (90%) and comanagement with a nephrologist (93%).

Neuropsychiatric manifestations

For the patient with major neuropsychiatric manifestations, the group endorsed consideration of immunosuppression (82%).

Medication management including safety surveillance

The group reached consensus for 5 process QIs addressing medication management or counseling prior to medication use (85%), strategies to limit glucocorticoid exposure (95%), initiating glucocorticoid-sparing agents (99%), and prescribing antimalarial therapy (hydroxychloroquine or chloroquine) in the absence of contraindications (96%). There was consensus around the minimum frequency of laboratory monitoring of long-term medications commonly used in childhood-onset SLE. Table 4 shows the schedule for intervals between laboratory studies (complete blood count, liver or renal panel, and urinalysis for protein quantification) determined for 9 medications, including glucocorticoids, adjunct medications (nonsteroidal antiinflammatory drugs, hydroxychloroquine), and immunosuppressants (azathioprine, cyclophosphamide in intravenous or oral form, mycophenolate mofetil/mycophenolic acid, cyclosporin A, rituximab, and tumor necrosis factor α antagonists).

Table 4. Quality indicator defining laboratory surveillance for safety in patients with longstanding exposure to the medication*
IntervalPrednisoneNSAIDs, HCQAZAIV CYCMMF/MPA, oral CYC, CSARituximabTNF antagonists
  1. IF a childhood-onset systemic lupus erythematosus patient is treated with medications, THEN laboratory surveillance for medication safety should be done at regular intervals. NSAIDs = nonsteroidal antiinflammatory drugs; HCQ = hydroxychloroquine; AZA = azathioprine; IV = intravenous; CYC = cyclophosphamide; MMF/MPA = mycophenolate mofetil/mycophenolic acid; CSA = cyclosporin A; TNF = tumor necrosis factor; CBC = complete blood count and differential; Liver = liver function tests (may include aspartate aminotransferase, alanine aminotransferase, bilirubin, albumin); Renal = renal function tests (may include sodium, potassium, HCO3, chlorine, blood urea nitrogen, creatinine); UA = urinary protein excretion.
Baseline  CBC, Liver CBC, LiverCBCCBC
1 month   CBC, Liver, Renal   
3 months  CBC, Liver CBC, Liver, RenalCBCCBC
6 months Liver, RenalRenal   Liver
9 months UA     
12 monthsCBC, Renal, LiverCBC   LiverRenal

Bone health

For patients with prolonged systemic glucocorticoid exposure exceeding ≥0.15 mg/kg daily for ≥3 months, the group reached consensus for documenting bone mineral density (BMD) testing (85%). The group approved reevaluation of BMD, no longer than 1 year later, in patients with significant osteopenia (Z scores of −2 or less) on initial testing (90%). Further, patients frequently exposed to glucocorticoids at any dose need calcium and vitamin D supplementation (87%).

General preventive strategies

There was consensus regarding 3 process QIs pertaining to vaccinations, flare prevention, and transfer of care. Respondents concurred that all patients with childhood-onset SLE should be vaccinated against influenza and encapsulated organisms (pneumococcus, meningococcus, and Haemophilus influenzae), except those with existing contraindications (89%). Respondents also agreed that patients should be educated about the benefits of photoprotection, documenting this instruction in the medical record (89%). There was also consensus that patients benefit from a transition plan in order to facilitate transfer of care to the appropriate adult health care providers (95%).

Ophthalmologic surveillance

There was consensus that an annual eye examination should occur if the patient required treatment with long-term glucocorticoid (86%) or antimalarial therapy (95%).

Education on cardiovascular risk factors

Table 5 shows consensus for 5 education topics to be evaluated and discussed with parents and patients of early (age 13–15 years) and middle to late adolescents (age ≥16 years) and documented in the patient's medical record. Respondents agreed that the risks associated with hypertension and high body mass index required annual discussion with parents and patients of all ages (82–86%), whereas screening for diabetes mellitus risk and measuring serum lipid levels to assess cardiovascular risk should occur at least every 2 years (81–85%).

Table 5. Education about cardiovascular risk factors*
Educational topics that need to be discussedEvery yearEvery 2 years
  1. IF a patient has childhood-onset systemic lupus erythematosus, THEN education about cardiovascular risk factors should occur in regular intervals with the parent and the patient age ≥13 years. C = 13–15-year-old child; A = adolescent age ≥16 years; P = parents.
1. Smoking or being around people who smokeC, AP
2. High blood pressureP, C, A 
3. High body mass index (being overweight)P, C, A 
4. Diabetes mellitus P, C, A
5. Perform/recommend serum lipid levels to assess cardiovascular risk P, C, A

QIs for adults with SLE considered irrelevant for use in childhood-onset SLE

In the first Delphi questionnaire, respondents agreed that QIs addressing “pregnancy” and “reproductive health” were not essential for defining standard of care delivery in childhood-onset SLE. Nonetheless, there was consensus to document anti-SSA, anti-SSB, and antiphospholipid antibodies in the medical record of a pregnant patient with childhood-onset SLE (86%).

DISCUSSION

This project defines the first set of process QIs for patients with childhood-onset SLE based on consensus and the available medical evidence among an international group of physicians involved in the care of patients with childhood-onset SLE. The consensus formation process resulted in 26 process QIs considered key to the treatment of all children and adolescents with childhood-onset SLE. Although several of the QIs have similarities to those for adult practice, differences that take into account distinct pediatric concerns are incorporated. Laboratory testing for the pediatric patient with suspected or confirmed childhood-onset SLE closely resembles that proposed in the QIs established by the ACR and EULAR for adults with SLE.

For childhood-onset SLE, there is consensus that autoantibody testing is valuable because of the role that autoantibody profiles uniquely serve in anticipating childhood-onset SLE prognosis. For example, autoantibody status related to histone, Sm, SSA, and SSB antigens has been associated with specific organ involvement and future response to B cell depletion therapy ([17, 18]). QIs addressing routine preventive or adjunctive medications all achieved consensus similar to those QIs found in adult recommendations. There is strong scientific evidence supporting the benefit of immunizations ([11, 19]) and antimalarial therapy in childhood-onset SLE ([20, 21]).

Since there is a scarcity of strong scientific evidence for monitoring medication toxicity in patients with adult-onset SLE, expert consensus serves as the best available guidance. Laboratory surveillance helps to ensure medication safety in patients with childhood-onset SLE, as in adult-onset SLE. Of note, the proposed testing frequency is not applicable to initial treatment or intervals of laboratory abnormalities.

High-dose glucocorticoid therapy is the standard of care for initial control of active disease in both adult-onset SLE and childhood-onset SLE. The importance of avoiding excessive exposure to glucocorticoids may be especially relevant to patients with childhood-onset SLE, since they have a greater duration of lifetime exposure. To date, there is marked variability regarding acceptable maintenance glucocorticoid dosing in chronic pediatric rheumatic diseases ([12]).

Published recommendations encourage consideration of an immunosuppressive agent for glucocorticoid-dependent or refractory adult-onset SLE ([8]). Respondents favored the use of glucocorticoid-sparing agents and were willing to consider further immunosuppression if the daily dose of oral prednisone (or equivalent) surpassed 0.15 mg/kg for longer than 3 months (89%). Recent consensus treatment plans for lupus nephritis in children propose several weight-adjusted glucocorticoid dosing regimens ([22]). However, there is no current consensus for the weight-adjusted dosing of glucocorticoids in the setting of multiorgan involvement, disease flare, or comorbid conditions of children with childhood-onset SLE.

Dual x-ray absorptiometry (DXA) is a commonly used technique to assess status and interval change of BMD and bone mass. The International Society for Clinical Densitometry suggests DXA measurement as part of a comprehensive skeletal health assessment in patients with increased fracture risk ([23]). Recent studies report that 15–37.5% of patients with newly diagnosed childhood-onset SLE already have osteopenia, likely due to the direct effects of chronic inflammation on bone metabolism, decreased participation in weight-bearing exercises, and vitamin D deficiency ([24-26]). Glucocorticoids decrease both serum vitamin D and BMD, while favoring osteoclast activity ([27]). Unlike QIs published for adults with SLE, which designate a daily oral prednisone dose of 7.5 mg, our group's consensus indicates strong support for calcium and vitamin D supplementation for all children with childhood-onset SLE exposed to any glucocorticoid dose duration for longer than 3 months. However, respondents commented that these adjunct medications should start concurrently with glucocorticoids, given the often unpredictable course of childhood-onset SLE.

Some QIs for childhood-onset SLE are markedly different from those proposed for adults with SLE. Both the ACR and the American Academy of Ophthalmology recommend eye screening for patients with adult-onset SLE at initiation of antimalarial therapy, then annually after 5 years of treatment. QIs endorsed by EULAR for adults with SLE refer to adherence to any existing (nonspecified) guidelines. The ACR recommendations for annual eye examinations of children treated with antimalarials are concordant with the consensus reached for the QIs for the treatment of childhood-onset SLE. Given patient variability in fully reporting symptoms, this frequency appears appropriate. The group achieved consensus for annual eye screening in patients with childhood-onset SLE treated with hydroxychloroquine and/or glucocorticoids.

Differences between the QIs for childhood-onset SLE and adult-onset SLE also exist regarding the care of lupus nephritis. There is consensus around clinical assessment of children with lupus nephritis at a minimum of every 3 months, regardless of disease activity. This quarterly schedule may reflect providers' experience with insidious subclinical renal disease, the higher prevalence of kidney involvement in pediatric populations, or an endorsement of clinical vigilance for patients whose guardians may not recognize such signs. If lupus nephritis or hypertension is present, then comanagement with a nephrologist is a unique QI for the treatment of childhood-onset SLE. Adherence to this QI depends on the local availability of such specialists. This cooperation speaks to a team-based approach now favored in international rheumatology practice ([28]).

Another difference from QIs for the treatment of adults with SLE is the omission of a QI for clinical assessment of disease activity by using validated disease activity measures such as the European Consensus Lupus Activity Measure or the Systemic Lupus Erythematosus Disease Activity Index ([29]). This was unexpected given the enthusiasm of the pediatric rheumatology community for accepting validated criteria ([30]). Respondents commented on difficulty completing these routine clinical care tools due to administrative obstacles, e.g., lack of time.

Regular education for parents or patients ages ≥13 years regarding cardiovascular risk factors is endorsed as a QI. Although recommendations for the treatment of adult-onset SLE consider this counseling under general prevention, respondents believed that the documentation rather than assumption of this practice would signify high-quality care. There is evidence that the inflammatory processes of SLE and the side effects of some SLE medications increase the risk of cardiovascular disease ([31]). Studies support that 50–85% of patients with childhood-onset SLE have dyslipidemia, hypertension, or increased body mass index ([32]). Finally, development of a transition plan from pediatric to adult care should involve systematic age-appropriate counseling. Factors such as disease activity and access to care may affect the transition process ([33, 34]). Our group endorsed transition planning in QIs for childhood-onset SLE.

This work has potential limitations. Despite recruiting respondents from membership of major professional medical organizations, we may have selection bias consequent to missing rheumatologists caring for patients with childhood-onset SLE, limiting the language of our literature review and questionnaires to English, and the interpretation of medicolegal culpability. To mitigate potential language barriers, colleagues for whom English is a second language screened this project. To address legal liability concerns, questionnaires emphasized that QIs are agreed-upon standards of care. However, potential ambiguity regarding who is responsible for maintaining and auditing adherence may discourage the use of QIs. Although we did not assess the feasibility of collecting QI data in clinical care, we are currently performing a pilot study to address this concern.

Medical knowledge is constantly accumulating, and therefore process QIs are prone to change. Therefore, the proposed QIs for childhood-onset SLE will need review and revision based on available scientific evidence. Future work includes a cross-sectional collection of QIs from identified pilot institutions to establish initial benchmarks. The success and accuracy of these indicators will depend on readily available medical records. This piloting will assist in providing baseline data that can be used to measure improvement, including characterizing this population, as we transition from QIs into quality measures appropriate for international use. Our future research will involve areas of care such as reproductive health or quality of life metrics appropriately addressed by the nominal group technique.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Brunner had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Hollander, Greenler, Espada, Beresford, Brunner.

Acquisition of data. Sage, Greenler, Pendl, Avcin, Espada, Beresford, Henrickson, Lee, Punaro, Huggins, Stevens, Klein-Gitelman, Brunner.

Analysis and interpretation of data. Hollander, Sage, Pendl, Espada, Beresford, Brunner.

Acknowledgments

We are indebted to the pediatric rheumatologists who participated in the Delphi questionnaires. We would like to thank the ACR, the Childhood Arthritis and Rheumatology Research Alliance, the Pan American League of Associations for Rheumatology, and the Pediatric Rheumatology European Society for their support with the distribution of the Delphi questionnaires. We would like to thank Drs. Hai Me and Shannon Hogan as well as Ms Jessica Hummel and Ms Shannen Nelson for their support in developing the Delphi questionnaires.

Ancillary