‘Raise the Bar’, a Canadian intervention to combat disparities in the recognition and management of iron deficiency: Development of the intervention, feasibility and implementation

Iron deficiency (ID) is the most common micronutritional deficiency in the world.1, 2 Iron deficiency progresses to iron deficiency anaemia (IDA) when there are insufficient amounts of iron stores to support red blood cell production due to inadequate dietary iron intake, excessive blood loss or impaired intestinal absorption.1 ID affects all individuals, but children and women are disproportionally affected by ID due to menstrual bleeding and the increased iron demands of growth/development, pregnancy and lactation.1 Individuals who are of lower socioeconomic status are at greater risk of ID due to decreased access to iron-rich foods and healthcare.1 There are many negative outcomes associated with ID with or without anaemia including fatigue, cognitive impairment, poor health-related quality of life and death.1 In the absence of concomitant inflammation, the most reliable diagnostic test for ID is serum ferritin.3, 4 A serum ferritin clinical decision limit of less than 30 μg/L is generally the accepted threshold for the diagnosis of ID in adults.3-6 However, many laboratories rely on population-based reference intervals—specifically, the lower limit of normal (LLN) of ferritin—to flag ID to healthcare providers.7 Accurate and representative reference intervals are crucial, as nearly 80% of medical decisions are based on interpretation of laboratory results.8 There is growing concern that the discrepancy between a clinical decision limit and the lower limit of 'normal' for ferritin contributes to the underdiagnosis and undertreatment of ID.1, 9 To combat disparities in recognition and care of ID, a grassroots initiative called 'Raise the Bar' was developed to change the LLN of serum ferritin to a clinical decision limit of less than 30 μg/L for adults and less than 20 μg/L for paediatrics in the largest outpatient community laboratories in Ontario, the most populated province in Canada. This initiative was informed by the principles of the United Kingdom's Medical Research Council's framework for developing and evaluating complex interventions.10 Guided by the framework, we describe the development of the intervention, feasibility and implementation of Raise the Bar.10 The Raise the Bar taskforce was formed in July 2020 and involved adult and paediatric haematologists, clinical and medical biochemists, clinical epidemiologists, laboratory administrators and leaders from community laboratories and the provincial health authority. A key pillar of the Raise the Bar initiative was to gather and generate evidence to justify the transition from an LLN to a clinical decision limit for serum ferritin. To evaluate the data informing ferritin reference intervals for the diagnosis of ID, our group conducted a systematic review and applied a quality assessment framework based on Clinical and Laboratory Standards Institute (CLSI) guidelines to all included studies.9 The authors reported that 49% (30/61) studies did not screen for patients at risk of ID.9 Only 8% of studies evaluated bleeding history and 15% assessed dietary history.9 Importantly, 52% did not refer to the use of the CLSI guidelines and those who cited the guidelines did not follow them appropriately.9 The systematic review also highlighted the lack of standardization across serum ferritin assays, with the lower and upper limits of normal varying significantly, depending on the assay and instrument used.9 In a North American survey of 23 laboratories, the LLN of ferritin in a population of healthy women reported by Abbott, Beckman, Ortho, Roche and Siemens instruments was 5, 11, 6, 13 and 10 μg/L respectively.7 This suggests that a clinical decision limit at the lower range of serum ferritin could allow for harmonization to facilitate the diagnosis of ID as the variability remains within the evidence-based range of ID. Recent functional physiologic studies have suggested a ferritin less than 30 μg/L is highly sensitive (92%) and specific (98%) for ID in the absence of inflammation.9, 11-13 To establish practical clinical decision limits for iron-related laboratory testing, the taskforce conducted a modified Delphi study.5 Two rounds were conducted with 26 experts from various specialties to achieve consensus. Upon presentation of statements, the panel agreed with current evidence and guidelines indicating that a ferritin less than 30 μg/L is diagnostic of ID in adults and less than 20 μg/L in paediatric populations.5 In response to these findings, the taskforce proceeded to change the LLN of serum ferritin to a clinical decision limit in the largest outpatient community laboratories in Ontario, Canada. An abnormal flag in the laboratory result was chosen as the main delivery method of the intervention as healthcare providers generally rely on flags to facilitate test result interpretation and to aid clinical decision-making, especially amid increasingly high patient volumes. Coupled with the flag, a website with educational resources on the diagnosis of ID, as well as treatment pathways for healthcare providers, was developed (https://www.hemequity.com/raise-the-bar-home; Figure 1). The contents of the website were iteratively revised with input from clinical biochemists, other healthcare providers and patients. Two examples of specific concerns that were addressed over these rounds of consultations were: (1) risk of increase in healthcare utilization and (2) risk of thalassaemia trait being confused with ID, given they both can present with microcytic anaemia; however, in a patient with thalassaemia who is not iron deficient, a normal ferritin is expected. Given the high prevalence of ID, a higher diagnostic threshold may result in a greater demand for healthcare resources, specifically gastroenterologic endoscopies and infusion of intravenous iron. While the taskforce acknowledged the possible impact on healthcare use, the goal of Raise the Bar was to prompt earlier investigation and treatment to prevent downstream effects that require more intensive resources. Individuals who are iron deficient warrant further investigation to determine the cause which may be from gastrointestinal or vaginal blood loss. Therefore, education on the need for early diagnosis, management and development of another website to assist with the management of heavy menstrual bleeding (https://www.hemequity.com/take-control-period-home) was leveraged. Regarding the thalassaemia trait, the taskforce did not have any concerns regarding primary care providers missing the diagnosis of thalassaemia. ID in patients with thalassaemia minima/minor should be treated with iron replacement. Thalassaemia without ID should not be treated with iron replacement as it will be of no benefit and may be associated with harm in people with thalassaemia intermedia or major. To address this concern, a section on thalassaemia was included in the educational resource. To determine the feasibility of changing the LLN of serum ferritin to a clinical decision limit, an online acceptability survey was conducted among healthcare providers.14 Participants were identified via snowball sampling through social media and professional networks between January and March 2024. A total of 165 healthcare providers from 23 countries participated in the survey (see Table 1 and Table S1). 68% (113/165) of respondents practiced predominantly in an academic setting, and 24% (39/165) practiced predominantly in a community setting. Most respondents did not have concerns about changing the ferritin lower limit of normal for adults (80%, 123/155) or children (81%, 112/138).14 When asked to provide reasons for concern, 32% (12/37) indicated that the proposed cut-off of <30 μg/L in adults was too low and would lead to underdiagnosis (Table S2), and 57% (13/23) responded in kind regarding the proposed cut-off of <20 μg/L in paediatric patients. Five respondents indicated a need for further education. There were no responses indicating concern about incorrect or overdiagnosis of ID. On 9 September 2024, across the largest community outpatient laboratories in Ontario, serum ferritin reference interval thresholds changed to a clinical decision limit, coupled with the abnormal flag and website. An information bulletin from the Ontario Laboratory Medicine Program within the provincial health authority was also released to support this change. In conjunction with the change, the Ontario Association of Medical Laboratories, with support from the taskforce, also updated its ID testing guideline to a clinical decision limit.15 To our knowledge, this is the first geographically wide implemented health system level change to address ID. This initiative involved evidence generation, input from stakeholders and an acceptability assessment over a duration of 4.5 years (Figure S1). In Canada's universal healthcare system, provincial/territorial governments are responsible for the management and delivery of healthcare services. Therefore, the change to a clinical decision limit for ferritin was limited to Ontario and where the community outpatient laboratories have outreach. However, educational resources housed in the Raise the Bar website are available globally and free of charge. Applicability of the initiative may also be limited in other countries given different health system structures, payment systems and stakeholder buy-in. However, in the hospital setting, the decision to move to a clinical decision limit is at the discretion of the local laboratory. Several hospitals and hospital networks inside and outside of Ontario have moved to a clinical decision limit based on local laboratory initiatives. Together, the community laboratories who have made the shift to a clinical decision limit serve over 20 million patients annually and perform up to 7000 ferritin lab tests per day. Since this change, 35 014 unique visitors have visited the accompanying knowledge translation website for additional educational resources. Importantly, the community laboratory partners have not received any negative feedback regarding the initiative from their clientele base since the change. We hope that Raise the Bar sets an example for other jurisdictions and/or clinical laboratories to follow suit. Although inter-assay variability in ferritin measurements exists due to lack of assay harmonization, we do not view this as a barrier to adopting a clinical decision limit of 30 μg/L for adults and 20 μg/L for paediatrics. We recommend communicating assay variability to clinicians during implementation to ensure ferritin results are interpreted with an understanding of assay-specific differences. The long-term impact of this initiative will be evaluated in a before-and-after analysis to determine the frequency of serum ferritin testing and treatment patterns (i.e. use of iron replacement). ID with or without anaemia and its consequences are preventable, but there are many structures that permissively lead to normalization and symptom dismissal. Guideline harmonization, screening programmes and effective knowledge translation are urgently needed to address this care gap. GHT and MS drafted the manuscript. CB created the scientific illustrations. All authors have reviewed and approved the manuscript. Funding for select aspects (e.g. website development) of the Raise the Bar initiative was supported by the Innovative Fund of the Alternative Funding Plan for the Academic Health Sciences Centres of Ontario. LM declares speakers bureau from Cepheid and Honoria from bioMérieux. MS declares unrestricted research funding (to institution) from Pfizer and Octapharma and Honoraria for speaking engagements with Octapharma, Sobi, Werfen and Roche. Ethics approval was obtained for the survey component of the study (REB #23-158). Not applicable. Data S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.