Effects of novel anemia nurse manager program on hemodialysis: a retrospective study from Qatar

Introduction: Anemia management in dialysis is challenging. Keeping hemoglobin levels within a tight range is difficult. A new program (anemia nurse manager [ANM]) was started for better anemia management. This study aimed to compare traditional anemia management with the new ANM model regarding the achievement of better hemoglobin targets (range, 10–12 g/dL), avoidance of extreme hemoglobin levels ( < 9 or >13 g/dL), and evaluation of the cost-effectiveness of the new model. Methods: This retrospective observational study compared traditional anemia management with management involving our new ANM model. Patients on hemodialysis in all ambulatory dialysis clinics in Qatar were included. The study included three phases: phase 1 (observation): June 2015 to August 2015, 460 patients; phase 2 (pilot): September 2015 to May 2016, 211 patients; and phase 3 (expansion in two phases): June 2016 to February 2017 and October 2017 to June 2018, 610 patients. Hemoglobin, iron saturation, and ferritin were evaluated according to the protocol. Results: In this study, 55% of the patients achieved the target hemoglobin in phase 1 compared with 75% in phase 2 (p = 0.0007). The hemoglobin level within the target range was sustained at 72% ± 5% of patients in phase 3. The achievement rate of the target hemoglobin level increased from 56% (May 2015) to 72% (July 2018) (p < 0.001). The proportion of patients with extreme hemoglobin declined from 10.7% in phase 1 to 6.4% in phase 2 and sustained at 8% afterward. Reducing the doses of erythropoietin stimulating agents, owing to the use of the ANM model, saved costs by approximately 11%. Conclusions: The ANM model was able to achieve and maintain hemoglobin levels within the target range and decrease extreme hemoglobin levels. These outcomes improved patient care by avoiding high hemoglobin (increase thrombosis, cancer recurrence, stroke, and death) and low hemoglobin (weakness, poor quality of life, and need for transfusion) levels. The ANM model was cost effective even after including the salaries of nurses. This model can be considered in other aspects of patient care in dialysis.


INTRODUCTION
Anemia is a common problem among patients with chronic kidney disease (CKD). 1 Several mechanisms are involved including iron deficiency and decreased production of erythropoietin stimulating agents (ESAs). [2][3] As CKD progresses to end-stage renal disease (ESRD), anemia tends to worsen, and management becomes more complicated. The World Health Organization defines anemia as hemoglobin (Hb) levels , 13 g/dL in men and post-menopausal women and ,12 in normal women 4 . Applying these targets in hemodialysis (HD) to improve the quality of life yielded mixed results. [5][6][7] Large trials to normalize Hb levels in patients with anemia and ESRD on dialysis revealed increased stroke risk, mortality, recurrence of cancers, and lack of improvement in cardiovascular disease. [8][9][10][11] These findings have led to a change in the Hb targets in patients with ESRD on dialysis. Recent guidelines have recommended maintaining the Hb level within 10-12 g/dL to prevent the need for blood transfusions and avoid complications. [12][13][14] The Hamad General Hospital (HGH) is a comprehensive government healthcare facility that covers most healthcare needs for the State of Qatar. In Qatar, HD is provided through four ambulatory dialysis clinics under HGH (Fahd Bin Jassim Kidney Center [FBJKC], which included Wakrah (Area A) and Al-shamal and Alshahania (Area B)). The mortality rate is relatively low in the younger population. 15 Traditionally, anemia is managed in these centers by nephrologists. In 2015, HGH initiated a pilot program to evaluate the potential role of an anemia nurse manager (ANM) to improve anemia care in ambulatory dialysis centers. The idea originated from the fact that a dialysis nurse is readily available in the center and spends more time with the patients. The duties of the nurse include the following: review and update patients' anemia-related medications and laboratory tests on file, review with the physician to write a prescription, follow up compliance and further tests, provide education to the patients and staff, and participate in multidisciplinary meetings and protocol updates. The ANM model began under the supervision of an experienced nephrologist who provided extensive training. This gradually expanded to include all ambulatory patients on HD in the State of Qatar. The model is different from the previous practice (which was dependent on the nephrologist, who has a busy schedule and thus limited availability) in that the new model employs a full-time ANM (available, trained, etc.) under limited supervision of a nephrologist. The primary aim was to investigate the effect of the ANM model on achieving and maintaining Hb level within the target range (10-12 g/dL). The secondary aims were to study the effect of the model on avoiding extreme Hb levels (, 9 g/dL or .13 g/dL) and investigate the cost-effectiveness of the new model. Collected data included patients' characteristics, demographic information, comorbidities, outcomes, and clinical data, which were extracted from the electronic medical records. The focus was on the patients' laboratory results relevant to the aims of this study (i.e., Hb, ferritin, and iron saturation levels), medications (ESAs including epoetin alfa, darbepoetin alfa, and methoxy polyethylene glycol-epoetin beta), and iron therapy (intravenous and oral), with attention to the doses, frequency of injections, and associated costs of treatment. The percentage of patients with Hb within the target range (10-12 g/dL) every month was calculated from the total number of patients with HD with Hb measured for that month (e.g., patients who were on prolonged hospitalization and whose Hb level was not measured in the ambulatory clinic were not included that month) ( Figure 2). A pharmacoeconomic analysis was performed because the health care in Qatar is generally almost completely subsidized by the government (including HD treatment and medications). The calculation was based on a simple formula: cost of medication (based on the purchase price paid by the hospital's central pharmacy) £ number of units of each ESA usedsalaries of nurses (calculated by the percentage of time allocated to work as an ANM).

Statistical analysis
The Statistical Package for Social Sciences version 17.0 for Windows (SPSS Inc., Chicago, IL) was used in the data analysis. Continuous variables are presented as means^standard deviation or medians and ranges, and categorical variables are presented as absolute and relative frequencies. For comparison between groups, the paired t-test was used for parametric variables and the chi-square test was used for nonparametric variables. Probability values of p , 0.05 (two-tailed) were considered significant.

Study population
The patient census during the study period increased from 460 in June 2015 to 610 in July 2018. Characteristics, demographics, and comorbidities for all accumulative patients through all phases are summarized in Table 1; no differences were found in comparing these variables among the three phases. The patients are relatively young, varied, and have multiple comorbidities. Members of the population have mixed ethnic backgrounds: 83.4% Middle Eastern (predominantly from Qatar and Egypt), 11.6% South Asian (mainly from India and Pakistan), and 4.9% East Asian (mainly from the Philippines).

Primary aim
The percentage of HD patients with Hb within the target levels ranged from 56% in June 2015 to 54% in August 2015 (mean, 55^2).

Extreme Hb level
The number of patients with extreme Hb levels decreased from 10.7% in September 2015 (phase 1) to 6.4% in May 2016 (phase 2) (adjusted for ESAnaïve patients).
In phase 3 (both 3A and 3B), low percentages of extreme Hb level were maintained with a mean of 8% of ambulatory HD patients (without adjusting for ESA-naïve patients).
The percentage of HD patients with ferritin levels of 200-800 mcg/L also improved between phases 1 and 2, increasing from 55% to 69% ( p ¼ 0.005), without a significant change in iron saturation. These findings persisted in phase 3.  [17][18][19] have raised safety concerns about the normalization of Hb level using ESA; thus, international guidelines recommended the target Hb levels of 10-12 g/dL (KDIGO, 2012 (13); The Renal Association, 2017). 20 Normalizing the Hb levels within these narrow targets using ESA is challenging, as several strategies evolved, including the use of protocols/algorithms. Maintaining the Hb level within the target range (10-12 g/dL) was challenging and reflected the difficulty in maintaining target levels within a narrow range over time, which can be complex and time consuming (Gardiner et al., 2019). 21 The ANM proved to be a successful cost effective model with favorable outcomes. These outcomes were reproducible with every expansion phase. This approach is unique in the Middle East area. The program was initiated with two part-time nurses under the supervision of an experienced nephrologist who provided extensive training. We gradually extended the program to include all patients on ambulatory dialysis in the State of Qatar. The ANM reviewed the laboratory results, and the prescriptions for ESAs and iron were written simultaneously. The role of ANM in the program was similar to other experiences reported previously. In 2014, Gerrish 22 reported that the prescriptive authority empowers the anemia nurse to review blood values and alter and prescribe medications accordingly, including changes outside the anemia protocol to use the lowest possible maintenance dose and thereby achieve the target Hb level. The prescriber is also responsible for  23 who reported a significant improvement in the proportion of Hb values within the target range after the implementation of a nurse-led protocol which had a higher Hb target range (11-13 g/dL) than that used in our study (10-12 g/dL). In 2020, George and McCann 24 did not found significant differences in serum Hb levels and ESA dosages between pre-and post-nurse prescriber implementation protocol. However, they found a significant improvement in serum ferritin and transferrin saturation levels, and significantly less intravenous iron was required. Our anemia management model was safe (by avoiding extreme Hb levels and maintaining Hb within the target range). The results were comparable with those of another study conducted by Drennan et al. 25  This study has some limitations. This retrospective cohort study used limited data and analysis (although most of the data were available in the electronic medical record; nationwide for the study duration). In addition, the comparison of anemia outcomes with baseline values has included some but not all the confounding factors that affect these values.

CONCLUSION
This study revealed that the involvement ANM in the management of anemia in patients on dialysis led to the improvement in maintaining the target Hb level and reduced extreme Hb level. This would avoid short-term (blood transfusion, weakness, and access thrombosis) and long-term complications (stroke, cancer recurrence, poor quality of life, and death). The ANM model was cost effective even after factoring in the salaries of the nurses. The success of the ANM program is depended on the availability of the ANMs, adherence to the protocol, and empowerment. The study highlights the potential role of nurse prescribers in the continuity of care in dialysis.