The effectiveness of iron education through a mobile application on donor return after deferral for low hemoglobin

Abstract

Background

Low hemoglobin level is a common cause of donor deferral and results in a huge loss of the donor pool. This study aimed to evaluate the effectiveness of a mobile application as an educational tool to enhance donor return and improve hemoglobin levels after deferral.

Materials and methods

This was an interventional study involving 382 blood donors who were deferred for low hemoglobin. The donors were divided equally into two groups: a control group and the intervention group. The control group received standard management for low hemoglobin deferral, which includes a short counseling session and a 1-month course of oral iron therapy. The intervention group used a mobile application in addition to standard management. The primary endpoint was the number of blood donors who returned during the 7 months of follow-up. The secondary endpoints were the hemoglobin increment at the first visit after the donors’ deferral.

Results

The return rate was higher in the intervention group, with 81.2% of the donors returning in the 7 months of follow-up compared to 66% of the control group (p<0.001). Male and female donors had mean hemoglobin increments of 1.0 g/dL and 0.7 g/dL, respectively, in the intervention group, compared to decrements of 0.2 g/dL and 0.4 g/dL, respectively, in the control group (p<0.001). Multivariable analysis showed a significant association between intervention method, education level and donation status on donor return (p=0.015, p<0.001, and p<0.001, respectively).

Discussion

Higher return rate and greater hemoglobin increase in the interventional group could be attributed to features in the mobile application. Repeat donors had the highest odds of returning to donate, followed by those with a tertiary level of education, and those given the mobile application. This study showed that a mobile application was effective in enhancing donor return and increasing hemoglobin level among deferred blood donors on their first return.

INTRODUCTION

Deferral due to a low hemoglobin level is one of the most common reasons blood donors are not eligible to donate their blood¹. This deferral is done to ensure the safety of blood donors by preventing them from further depleting their iron stores, which could ultimately lead to them developing anemia. The most common cause of anemia among blood donors is iron deficiency, accounting for up to 70% of all deferrals for a low hemoglobin². Furthermore, blood donation itself is known to be one of the causes of iron deficiency, as each blood donation results in the loss of approximately 236 mg of iron³.

Donors can be deferred for multiple reasons. Regardless, the effect can be tremendous for some donors, who are found to react with strong emotions such as anger, sadness, frustration, and disappointment⁴. Deferral also leads to poor return rates or non-return, especially among first-time donors or donors who were deferred because of a low hemoglobin⁵.

In most blood centers, a blood donor deferred because of low hemoglobin is given a leaf let on foods containing a large amount of iron, a short counseling session, and iron tablets¹. Despite the counseling session and leaf let, it was found that some donors still had inadequate knowledge regarding iron deficiency⁶. Nevertheless, these donors wanted to understand their condition and actively find ways to improve their hemoglobin level¹.

The young generation below the age of 30 years old, has deeply incorporated mobile devices into their daily lives and they rely on mobile applications to simplify their work and be updated with the latest information⁷. Thus, iron education through a mobile application has a vast potential to attract donors, engage them in improving their hemoglobin levels, and retain them as regular and dedicated donors, as applications can provide multimedia elements such as infographics, daily reminders, and rich text⁸. The aim of this study was to evaluate the effectiveness of a mobile application as an additional educational tool to enhance donor return and subsequently improve hemoglobin levels among donors who had been deferred because of low hemoglobin.

MATERIALS AND METHODS

This was an interventional study involving 382 blood donors who were deferred for low hemoglobin in the National Blood Center (NBC) of Malaysia from April 1, 2020 to December 31, 2020. All whole blood donors between 18 and 60 years old who were deferred because of low hemoglobin levels were included. The accepted hemoglobin levels for donation were between 12.5 and 13.4 g/dL for males and between 11.5 and 12.4 g/dL for females.

The development of the mobile application content started with a literature review to identify the key elements of the information needed⁹. The content was then validated using item and face validation before it was used for this study. The content validation involved six panels consisting of two hematologists and four transfusion medicine specialists. They were asked to validate the writing style, structure and presentation, objectives, and relevance. For face validation, 15 participants were randomly selected among whole blood donors who were deferred for low hemoglobin. They were asked to provide appraisals on the readability, text and graphic appearance, and any difficulty in understanding the digital pamphlet. The validation was done using adopted evaluation forms with a four-point Likert scale.

The final digital content consisted of three parts, education about anemia, iron education, and a list of foods rich in iron, with an app size of 5.68 megabytes. The screenshots of the mobile application are shown in Figure 1. The blood donors were educated on the symptoms of anemia and how hemoglobin is made in the body. For iron education, the blood donors were given details about iron absorption, the side effects of iron therapy, and what to do in the event of having side effects as well as the list of iron-rich foods and information on their iron content. Information was relayed in a colorful infographic, and the text was kept short for easy readability and understanding. The app also had an integrated interface linked to the NBC donation campaign location and a daily reminder to take an iron tablet that could be set according to the user’s preference. It also had a feature enabling the user to visually log their daily iron tablet intake.

Figure 1.

Screenshot of the mobile application (interventional tool) used in this study

Study procedure

The hemoglobin level of the blood donor was measured for donation eligibility using a hemoglobin analyzer (CompoLab TM, Fresenius Kabi, Bad Homburg, Germany) on a finger prick sample before blood donation. Donors were recruited at the blood donation site in the NBC. All blood donors who were not eligible to donate because of low hemoglobin levels were approached immediately for participation in the study by a team appointed by the researcher. The team consisted of staff nurses and laboratory technicians who were stationed daily at the hemoglobin checking counter. Deferred donors were then asked about the operating system of their mobile phone as a screening question. Only those with Android phones were recruited as the mobile application was made in the Android system. Those donors who fulfilled the inclusion criteria were then invited to a counseling room. Using posters, they were given further explanations regarding the study procedure and the app. Subsequently, they were helped to install the app and were trained on how to use it. They were also informed that the size of the app is just 5.68 megabytes, and were reassured that the app installation would not cause any lag, and mobile data would not be consumed when the app was used. They were also informed that their data are stored locally, and would be wiped out upon uninstalling the app. The donors were given the option to uninstall the app later if they wanted to. A short in-app tutorial and email link were provided to guide them. Once the donors had understood the whole process, consent to enrolment into the study was taken, and the next day was considered day 1 of participation. Data containing demographic information, such as age, gender, race, educational level, donation status, and hemoglobin level, were recorded.

The sampling method used for this study was universal random sampling and allocation concealment. Randomization was done using envelopes. In total, 382 envelopes were used, consisting of a mixture of 191 envelopes labeled as the intervention group and the other 191 envelopes labeled as a control group. The study participants were assigned based on the envelopes chosen by them. The control group received standard management for deferral due to low hemoglobin, which includes a short counseling session and a 1-month course of oral ferrous fumarate 200 mg, and a week’s course of oral folic acid 5 mg. The intervention group used the mobile application, which was installed from the Google Play Store, together with the standard management for low hemoglobin deferral. The iron and folic acid supplements were provided without any cost by the NBC for all blood donors who were found to have a low hemoglobin during their attempt to donate. According to the NBC standard operating procedure, a donor with a low hemoglobin level is deferred for the day, and will usually be advised to come and donate after completing the 1-month course of iron supplementation. The primary endpoint was the number of blood donors who returned during the 7 months of follow-up. The secondary endpoint was hemoglobin increment during the first visit after their deferral.

All the study participants were requested to return for subsequent attempts to donate as early as day 25 of the iron course or at any time convenient to them starting from the day of enrollment. They could return to donate at any donor center, as their data had already been recorded in the national blood bank information system known as Blood Bank Information System version 2 (BBIS v2). Seven months were allocated to assess the subsequent attempt at donation. Donors who never returned to attempt to give blood during the 7-month study follow-up were still counted as participants, and their data were statistically analyzed as non-returning donors.

For donors who returned to donate within the 7 months, their first return date and hemoglobin level, together with a unique identification, were recorded. If donors were deferred again during this attempt, they were still counted as returning donors.

Operational definitions

Subsequent return after deferral was defined as donors who were initially deferred for low hemoglobin and returned to attempt to give blood at least once during the study period. A repeat donor was a blood donor who had donated at least once in the same blood center, while a new donor was a blood donor who was donating for the first time. Educational level in this study was grouped into two categories: primary and secondary education in the first category and tertiary education in the second category.

Statistical analysis

Data were entered and analyzed using SPSS software (IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY, USA). The data were analyzed using descriptive and inferential methods. The effect of the intervention on donor return was assessed using the chi-square test. Hemoglobin changes between the two groups at the end of the study were evaluated with an independent τ-test. Multivariate analysis was used to examine the association between socio-demographic factors and donor return. A backward logistic regression method was applied as the best model. Multicollinearity and interaction were checked for and not found. Goodness of fit was assessed by the Hosmer-Lemeshow test, classification table, and area under the receiver operating curve. Variables with a p-value <0.25 in univariable analysis were selected for multiple logistic regression as a more traditional level (such as 0.05) often fails to identify important variables¹⁰. A p-value of less than 0.05 was considered statistically significant.

Ethics

This study was approved by the Human Research Ethics Committee (HREC) of University Sains Malaysia (USM) with the research number USM/JEPeM/19120968. Approval was also obtained from the Medical Research & Ethics Committee of the National Medical Research Register of Malaysia with the research number NMRR-19-3460-50926 (IIR).

RESULTS

In total, 382 blood donors aged between 20 and 59 years from the NBC donation center were recruited. No donor was lost during the follow-up. The demographic characteristics of the donors are shown in Table I. There was no significant difference in donor characteristics between those in the mobile app group (intervention) and those in the non-mobile app group (control).

Table I.

Comparison between groups (intervention and control) of donor and donation characteristics among donors deferred for low Hb

Characteristics	App (n=191)	Non-app (n=191)	Total (%)	p-value
Age (years) a				0.552c
20–29	69 (36.1)	74 (38.8)	143 (37.5)
30–39	74 (38.8)	65 (34.0)	139 (36.4)
40–49	37 (19.3)	35 (18.3)	72 (18.8)
50–59	11 (5.8)	17 (8.9)	28 (7.3)
Gender a				0.803c
Male	40 (20.9)	42 (22.0)	82 (21.5)
Female	151 (79.1)	149 (78.0)	300 (78.5)
Race a				0.792c
Malay	132 (69.1)	129 (67.5)	261 (68.3)
Chinese	42 (22.0)	41 (21.5)	83 (21.7)
Indian	17 (8.9)	21 (11.0)	38 (10.0)
Education level a				0.064c
Primary & secondary	78 (40.8)	96 (50.3)	174 (45.5)
Tertiary	113 (59.2)	95 (49.7)	208 (54.5)
Donation status a				0.062c
New donor	52 (27.2)	69 (36.1)	121 (31.7)
Repeat donor	139 (72.8)	122 (63.9)	261 (68.3)
Weight (kg) b	70.85 (15.521)	69.31 (14.782)		0.361d
Hemoglobin (g/dL) b
Male	12.91 (0.291)	13.00 (0.251)		0.117d
Female	11.89 (0.277)	11.95 (0.255)		0.062d
Total	12.18 (0.505)	12.10 (0.502)		0.138d

^avalue expressed as number of donors (%);

^bvalue expressed as mean (standard deviation);

^cPearson χ² test with a level of significance of 0.05 ;

^dIndependent τ-test with a level of significance of 0.05. Hb: hemoglobin.

Table II shows the comparison of donors who returned within 7 months after deferral between the two groups. The return rate was significantly higher in the mobile app group (81.2%) than in the control group (66.0%) (p=0.001).

Table II.

Donor return within 7 months after deferral for low hemoglobin between the groups with and without the mobile app

Characteristics	App n (%)	Non-app n (%)	χ2 value	p-value
Donor return
Yes	155 (81.2)	126 (66.0)	11.32	0.001a
No	36 (18.8)	65 (34.0)

^aPearson χ² test with a level of significance of 0.05.

Table III reveals that donors in the mobile app group returned earlier to donate and had a significant hemoglobin increment as compared to donors in the non-mobile app group (p<0.001). The earliest donor to return to donate did so at 8 weeks and was in the mobile app group. The greatest hemoglobin increment was 2.8 g/dL which was seen in a male donor in the mobile app group. The greatest hemoglobin decrement in the group without the mobile app was in a female donor whose hemoglobin level decreased by 2.3 g/dL (baseline of 12.2 to 9.9 g/dL) as compared to the greatest hemoglobin decrement of 0.4 g/dL in the mobile app group in a female donor.

Table III.

Time and hemoglobin increment on first return among donors deferred for low hemoglobin between the groups with and without the mobile app

Characteristics	App (n=154)	Non-app (n=127)	τ-test	p-value
Time to return (weeks)	8.0±7.1	15.2±7.7	7.523	<0.001
Hb increment (g/dL)
Male	1.0±1.0	−0.2±0.8	7.022	<0.001
Female	0.7±0.6	−0.4±0.7	14.290	<0.001

Data expressed as mean ± standard deviation. p-value obtained using an independent τ-test. Hb: hemoglobin.

Table IV shows the association between donor characteristics with donor return. p-values <0.25 from univariable analysis were selected for multiple logistic regression. We found that repeat donors had the strongest association with return and had a 10.3 times higher odds of returning to donate blood (adjusted odd ratio [OR]=10.34, 95% confidence interval [CI]: 5.89–18.13; p<0.001), followed by tertiary education level which showed a four times higher odds of returning (adjusted OR=4.01, 95% CI: 2.27–7.06; p<0.001). Donors who received the intervention using a mobile app had a two times higher odds of returning to donate blood (adjusted OR=1.98, 95% CI: 1.14, 3.42; p<0.001).

Table IV.

Association of donor and donation characteristics with donor return

Characteristics	Return (n=191)	No Return (n=191)	Crude ORc (95% CI)	p-value	Adjusted ORd (95% CI)	p-value
Intervention a
Mobile app	155	126	2.22 (1.39, 3.55)	0.001*	1.98 (1.14, 3.42)	0.015*
Non-mobile app	36	65	1.00		1.00
Age (years) a
20–29	69	74	1.00	0.397	Not applicable
30–39	74	65	0.78 (0.46, 1.31)	0.346
40–49	37	35	1.33 (0.67, 2.62)	0.416
50–59	11	17	1.28 (0.48, 3.40)	0.621
Gender a
Male	40	42	1.00		1.00
Female	151	149	0.67 (0.37, 1.21)	0.188**	0.84	0.627
Race a
Malay	133	128	1.00	0.688	Not applicable
Chinese	40	41	1.13 (0.64, 2.01)	0.672
Indian	17	21	0.78 (0.37, 1.63)	0.506
Educational level a
Primary & secondary	70	86	1.00		1.00
Tertiary	121	105	3.20 (1.98, 5.17)	<0.001*	4.01 (2.27, 7.06)	<0.001*
Donation status a
New donor	46	56	1.00		1.00
Repeat donor	119	116	8.86 (5.31, 14.79)	<0.001*	10.34 (5.89, 18.13)	<0.001*
Weight (kg) b	70.9 (15.58)	69.3 (14.77)	1.00 (0.98, 1.01)	0.531	Not applicable
Hemoglobin (g/dL) b
Male	12.9 (0.28)	13.0 (0.25)	0.96 (0.13, 6.80)	0.963	Not applicable
Female	11.9 (0.26)	12.0 (0.26)	0.62 (0.24, 1.70)	0.328

^aData expressed as number (n).

^bData expressed as mean (standard deviation).

^cSimple logistic regression was performed.

^dMultiple logistic regression was performed.

^*Significant p-value <0.05,

^**p<0.25.

Backward logistic regression method was applied as the best model. Multicollinearity and interaction were checked and not found. Hosmer-Lemeshow test (p=0.499), classification table with 80.4% correctly classified, and area under receiver operating curve (82.3%) was applied to check the model fitness.

DISCUSSION

This study found that most of the donors deferred because of low hemoglobin levels were female. It was previously reported that female donors were 14 times more likely than male donors to be deferred due to low hemoglobin¹¹. This might be the consequence of iron deficiency anemia caused by pregnancy and menstruation, or even frequent blood donation. A study by Vinkenoog et al. found that among repeat donors, a ferritin level below 30 ng/mL was more frequent among female donors than among male donors (53% and 43%, respectively). The same study also found that 25% of female first-time donors and 1.6% of male first-time donors had serum ferritin levels below 30 ng/mL¹².

The negative effect of deferral was noted in earlier literature. Large numbers of blood donors were found to be deferred for reasons related to both blood donor and recipient safety¹³, leading to continuous loss of donors¹⁴. Our study showed that the intervention group had a significantly higher return rate of 81.2% compared to only 66% in the control group, which was a return rate similar to that in a previous study documenting that 64.0% of the deferred donors returned to donate¹⁵. The difference in return rate between the two groups could be attributed to features in the intervention tool, such as donors who had completed the iron course being informed via the app that they were ready to donate, and being provided a list of mobile donation locations to facilitate their donation. Additionally, the lower return rate in the control group could be due to blood donors taking low hemoglobin deferral as an excuse, thus avoiding donating blood¹⁶. This effect is known as excuse-driven behavior; donors who are deferred might believe that they have a higher probability of being deferred again, and therefore stop donating their blood¹⁶. This behavior might have been less likely in the intervention group as donors were given sufficient information on why and how to improve their hemoglobin level through the infographics content of the app. The app also gave a daily reminder to take iron tablets and a countdown to the blood donation date, which could have incited the donors with a feeling of anticipation to donate their blood. Moreover, the app gave an approximate forecast of hemoglobin increment whenever the iron tablet consumption was logged. These interesting features would drive the donor to return to donate blood. As a result, donors in the intervention group were found to return to donate faster, at a mean of 8.0 weeks as compared to 15.2 weeks in the control group. The donor return time in the control group was similar to that in another study done by Hillgrove et al., in which the deferred donor return time was 13.2 weeks¹⁵.

Previous studies had shown that donors with repeatedly low hemoglobin levels at consecutive attempts to donate blood were common¹⁷. The failure to achieve an increase in hemoglobin could be due to compliance issues or unbearable side effects of iron therapy¹⁸. This might be due to insufficient information provided regarding iron therapy during the deferral counseling session⁹. A study done by Mast et al. found that donors who were not supplied with any educational material failed to have a significant increase in their hemoglobin level at their return to donate, and indeed had a negative hemoglobin change of −0.13 g/dL¹⁹. However, by providing proper educational material on iron deficiency and a supply of supplemental iron, donor hemoglobin levels were found to improve by 0.85 g/dL in females and 0.68 g/dL in males from their baseline level during the last deferral²⁰. Thus, providing an app that guides the donors throughout the iron fortification process could significantly improve their compliance towards completing the iron course. Side effects of the iron treatment and methods to overcome them, such as taking lower doses, alternate day courses, or changing to a high iron diet, were also explained info-graphically through the app. This information was essential in enhancing donor compliance toward iron therapy²¹. However, it was seldom conveyed to them because of time constraints of the blood bank staff during counseling or limited space available in the leaf let provided²².

Our study also showed that repeat donors had 10.34 times higher odds of returning compared to new donors. This could be due to new donors being more sensitive than repeat donors to rejection. Custer et al. found that the return rate for subsequent donation was only 29% for new donors who were deferred for low hemoglobin as compared to 47% for repeat donors²³. This implies that first-time donors were less resilient to deferral compared to donors who had developed regular donation habits, regardless of interventions⁵.

Another factor that was found to significantly affect the donor return rate was the educational level of the donors. Our study found that donors with tertiary education had four times higher odds of returning to donate compared to donors with only primary or secondary education. This finding was similar to those of previous studies which showed that donors with the higher education were more likely to become regular donors^24,25. Moreover, this group of donors was also found to have a positive association with compliance with iron and folate supplementation. This difference could be explained by a possible effect of education on self-care skills as more highly educated donors were more likely to understand and use the available information better to meet their own needs and return to donate again²⁶.

That the intervention via the mobile app significantly influenced the donors to return to donate was consistent with previous findings using other communication tools such as e-mails and reminder messages^27,28. It was shown that donors who heard a message on an integrated automated recorded phone system were 9.5% more likely to donate again as compared to those who did not hear the message²⁹. Although all these studies, including ours, showed a similar positive effect on return rate, using the mobile app was noted to be superior with regard to donor return, cost-effectiveness, and practicality. Furthermore, the mobile app can be continuously updated and improved as new information becomes available.

Limitation

There are several limitations to this study. First, this study did not investigate the baseline hemoglobin levels before the donation visit and the effect of different ABO RhD blood types on donor return. The interaction of donors in logging their iron intake, hours of usage of this app, and the app uninstall rate were also not analyzed. Furthermore, the app was only designed for the Android system, limiting the type of donor who could participate. Future studies in which the app would be able to record the iron logging session to assess donor compliance in taking the iron tablets could be done. The app could also be made to support other operating systems so that a greater variety of donors could enroll. Additionally, barriers to using this app could be analyzed so that improvements could be made to the features of the app.

CONCLUSION

Education via a mobile application is an active and effective engagement method that should be utilized to retain more blood donors and to reduce continuous donor loss caused by the world-wide deferral of potential donors because of low hemoglobin levels.