Background. Blood donors are on average healthier than the general population, a phenomenon known as the “healthy donor effect.” Earlier studies have also pointed to healthier behaviors among whole blood donors than the general population. This study is aimed at assessing the prevalence of four healthy behaviors (sufficient physical activity, avoiding cigarette smoking, low to moderate alcohol use, and maintaining a healthy weight) among platelet donors and to compare the results with those in the general population of similar ages. Methods. Eighty-six platelet donors were asked to complete a questionnaire designed to assess physical activity, smoking, and alcohol use. Sociodemographic information including gender, age, and education was also collected from all participants. Chi-square statistics and logistic regression were used in statistical analysis. Results. The mean age of the study donors was 51 years, 56% were female. Most were employed (90%), and 48% hold a bachelor’s or higher degree. The prevalence of healthy behaviors differed by education gradients but not by gender and age. About 49% of the donors met the weekly physical activity recommendations, less than 5% were daily smokers, and~26% were classified as more frequent drinkers (≥1 to ≤5 times per week). The corresponding percentages for the general population were, respectively, 33%, 13%, and 35%. The prevalence of overweight and obesity, as assessed by body mass index (BMI), among donors were 50% and 29%, respectively, much higher than the current prevalence of overweight and obesity of 37% and 19%, respectively, among adults in the general population. Conclusions. The individual health behaviors of the majority of the study population could be characterized by a relatively high level of physical activity, low prevalence of daily smoking, and moderate alcohol drinking. The above-average overweight/obesity prevalence among platelet donors in this cohort is of concern because of the potential serious health consequences and it warrants further reflection.
1. Introduction
Several decades ago, Sullivan [1] proposed that the decreased risk of cardiovascular diseases (CVD) in premenopausal women, as compared to postmenopausal women and age-matched men, could be attributed to lower levels of iron and serum ferritin due to menstrual bleeding. Sullivan suggested that the depletion of iron stores with repeated blood donation in postmenopausal women and men might also protect against CVD. In support of this “iron hypothesis,” early prospective studies comparing whole blood donors with nondonors reported an association between blood donation and reduced risk for CVD [2, 3].
More recent studies have also suggested that blood donation may confer cardiovascular and metabolic benefits for blood donors. For example, an observational study among German blood donors suggested that regular blood donation is associated with a marked reduction in systolic and diastolic blood pressures among hypertensive donors [4]. However, a study among American blood donors suggested that regression to the mean may have contributed to these positive results [5]. In another recent study conducted at two Mediterranean blood banks in Italy and Greece, it was observed that regular blood donation in Greek blood donors could positively affect total oxidative status, a measure of overall antioxidant capacity, as reflected in enhanced activity of antioxidant enzymes in serum [6].
The association between blood donation and potential health benefits, including a protective effect against CVD, may, however, be complicated by the phenomenon called the “healthy donor effect” (HDE). This is akin to the concept of the “healthy worker effect” in occupational cohort studies, whereby individuals who apply and enter the industry are healthier and have lower morbidity and mortality than the general population [7, 8]. The HDE is the selection bias due to donor eligibility criteria which select for individuals healthy enough to donate blood, and volunteer bias, because healthier individuals may be more likely to choose to become blood donor [9, 10]. The HDE poses therefore a problem when comparing blood donors with the general population, because blood donors are on average healthier than the general population, making the general population an inappropriate comparison group. Findings of a recent large study comparing two indicators of health, self-rated physical health status, and mental health status, between Danish blood donors with nondonors, which showed that blood donors had better-reported health than the comparison group members are consistent with the concept that blood donors are healthier than the general population [11].
Two recent studies have tried to circumvent this problem by using internal comparison groups. In an analysis of nearly 160 000 Dutch, whole blood donors with a history of at least 10 years of active donation in which high-frequency donors were compared with low-frequency donors, a study found out that high-frequency female, but not male, donors, had a 9% decreased risk for cardiovascular morbidity compared with low-frequency female donors (age-adjusted hazard rate ratio: 0.91, 95% CI: 0.85-0.98) [12]. Moreover, sensitivity analyses repeated with a 5-year qualification period yielded similar results, supporting the absence of a residual HDE. In another large study which included almost 1.2 million whole blood donors aged 18-64 years in Denmark and Sweden, current donors were compared with donors who had stopped donating blood due to advanced age. It was found that blood donation was positively related to greater life expectancy [13], suggesting that part of this gain may be due to the effects of blood donation. Shehu and colleagues [14] attempted to quantify the magnitude of the HDE among German blood donors and found that a large part (~82%) of the observed differences in health status between donors and nondonors could be explained by the HDE.
There is also evidence that the magnitude (or strength) of the HDE is influenced by a number of factors, including sociodemographic and lifestyle variables. For instance, in a study of Dutch blood donors, researchers found that whole blood and plasma donors were more educated, were less likely to smoke, had a lower prevalence of alcohol consumption, had more self-reported physical activity, and were slightly more likely to engage in healthful food choices than the general population. Furthermore, compared to the general population, they had fewer reported health conditions, including a lower prevalence of type 2 diabetes and high cholesterol, had fewer recent (past three months) doctor visits, and were less likely to be treated at a specialist’s office during the past six months [10, 15]. Similarly, German blood donors were less likely to smoke and more likely to consume healthy diets and had a lower prevalence of overweight and lower prevalence of chronic diseases than nondonors and inactive donors [14]. These data suggest that whole blood and plasma donors may have healthier behaviors than the general population.
Platelet donors represent a unique population of volunteer, unpaid blood donors because of the high demand for platelet concentrates to support transfusion therapy. Regular platelet donors are willing to give this life-saving gift every time they are able to donate and spend extra time, which is necessary for an apheresis platelet collection. However, studies published until now have not assessed health behaviors among platelet donors. This study is aimed (1) at assessing the prevalence of four positive health behaviors (i.e., sufficient physical activity (PA), avoiding cigarette smoking, low to moderate alcohol use, and maintaining a healthy weight) in a representative sample of platelet donors, (2) at examining how these healthy behaviors varied according to sociodemographic factors, and (3) at comparing these results with those reported in nationally representative surveys that include both men and women of similar age range.
2. Methods
2.1. Setting and Study Population
The study was conducted at the Blood bank in Akershus University Hospital in Loerenskog, Norway, between April 4, 2019, and February 18, 2020. The study was approved by the local Institutional Review Board, and all participants provided written informed consent. Eighty-six apheresis platelet donors attending their donation appointment were asked to participate in the study by completing a questionnaire. There were no exclusion criteria for donors, other than being ineligible to donate platelets. The questionnaire was designed to collect data on donor sociodemographics (gender, age, relationship status, education level, employment status, weight, and height) and health behaviors (physical activity, cigarette smoking, and alcohol use). Each questionnaire was given a unique research identification number to guarantee anonymity. The questionnaire was pilot-tested among five donors and based on comments from donors and interviewing/apheresis staff; questions were modified to improve clarity. The questionnaire took 10-15 minutes to complete. Self-reported weight and height were cross-checked with the collection protocol data and used to calculate the body mass index (BMI) and weight in kilograms divided by height in meters squared (kg/m²). Educational attainment was assessed by the highest self-reported grade completed and categorized into four education levels: middle school (at least 9 years of education completed), high school degree or equivalent (12 years of education completed), bachelor’s degree (had completed 3-4 years of education beyond high school), and higher degree (had completed 5 or more years of education beyond high school).
2.2. Questions on Self-Reported Health Behaviors
2.2.1. Physical Activity
Frequency and duration of PA question asked “How often during leisure time are you engaged in at least 30 minutes physical activity?” The question had 6 responses of (1) never, (2) sometimes, (3) once per week, (4) 2 times per week, (5) 3-4 times per week, and (6) 5-7 times per week. The participants were then asked to state their PA intensity levels as light intensity (breathing approximately normal, equivalent in effort to slow walking/leisurely walk), (2) moderate intensity (activities that cause light sweating or slight to moderate increases in breathing, equivalent in effort to brisk walking), or (3) vigorous intensity (activities that cause heavy sweating or breathing much harder than normal, equivalent in effort to running, jogging and bicycling). In addition, they were asked whether they were engaged in training in gyms or fitness centers/studios (yes/no) and then whether they have had physical activity in the past 24 hours (yes/no).
2.2.2. Smoking
For smoking habits, participants were asked “Do you smoke?” with the following response options: never smoked, former smokers, daily smokers, and occasional smokers. Those who smoked were further asked whether they had smoked in the past 24 hours (yes/no).
2.2.3. Alcohol Consumption
The frequency of alcohol consumption question asked “How often do you drink alcohol?” In all, eight response options were available: (1) never, (2) less than once per month, (3) once per month, (4) 2-3 times per month, (5) once per week, (6) 2-3 times per week, (7) 4-5 times per week, and (8) every day or almost every day. The participants were then asked: “How many drinks they consume on a typical drinking occasion?” This question was answered using a 4-point scale with response options ranging from 1 to 4+ drinks per occasion. Study participants were also asked whether they had consumed alcohol in the past 24 hours (yes/no).
2.2.4. Self-Reported Sleep Measures
The questionnaire also included a single-item measure of subjective sleep quality in which donors were asked to rate their sleep quality on a five-point scale from very good to very poor. In addition, donors were asked about their habitual bedtimes and duration of sleep.
2.2.5. Engagement in Multiple Healthy Behaviors
The number of positive health behaviors (i.e., sufficient PA, avoiding cigarette smoking, low to moderate alcohol consumption, and maintaining a healthy weight) was summed for each donor. Donors were categorized as having 0, 1, 2, 3, or 4 positive health behaviors.
2.3. Statistical Analysis
For the purpose of our analysis, sociodemographic and health behavioral variables were mostly dichotomized: gender (male, female), age (22-50 vs. 51-69 years), relationship status (married/cohabiting vs. single), and education level (high school or less vs. bachelor’s or higher degree). Participants were also classified as employed if they were in paid employment; otherwise, classified as “other”. Sleep quality was dichotomized as very good/good vs. fair. Smoking status was categorized into nonsmokers () and current smokers (). An indicator of PA levels per week was calculated as the product of frequency and intensity of PA reported by each donor. For current drinkers, we calculated the average “typical” number of drinks consumed per month by combining frequency and quantity (i.e., ) into a single continuous variable. For comparison with the general population, current drinkers were also categorized as less frequent drinkers (≤3 times per month) or more frequent drinkers (≥1 to ≤5 times per week), because no equivalent data were available in the most recently published statistics for alcohol consumption. Finally, for bivariate comparison of the number of positive health behaviors, a dichotomized variable was produced by combining the scores 1 and 2 and scores 3 and 4. Because the majority of participants were employed and married/cohabiting and rated their sleep quality as very good/good, differences in these variables could not be evaluated.
Data were summarized with descriptive statistics and expressed as number and percentage/proportion for categorical variables and (SD) for continuous variables. Cross-tabulation, a chi-square test, or the Fisher’s exact test were used to analyze differences between groups. Multivariate logistic regressions were used to estimate adjusted odds ratios (OR) and associated 95% confidence intervals (CI). Independent samples -test was used to compare differences in means of continuous variables and the Pearson correlation coefficient (Pearson’s ) to evaluate their correlations. Statistical testing was 2-sided, with . Analyses were performed with IBM SPSS Statistics, version 25. The data set did not have any missing values.
3. Results
3.1. Sociodemographic and Health Behavioral Characteristics of the Study Population
We asked 86 donors (~80% of the platelet donor pool at our blood bank) to participate in the study. All of the donors asked to participate agreed to do so and completed the questionnaire. The number and percentage distribution of responses to questionnaire items are shown in Table 1. The mean age was years (range: 22-69 years), 55.8% were female, and 78% were married or cohabiting (53.5% and 24.5%, respectively). Most (89.5%) were employed; the remainder were recently retired (5.8%) or students (2.3%). Only one donor was presently unemployed and only a 47-year-old donor reported receiving a temporary disability benefit. Slightly more than half of the donors had high school or less education (46.5% and 5.8%, respectively), and 47.7% hold a bachelor’s or higher degree (33.7% and 14%, respectively). Regarding ethnicity, only one donor had an ethnic minority background. Male and female donors did not differ in age ( vs. years, ), but more males (61%) than females (38%) hold a bachelor’s or higher degree ().
Variable
Number (%)
Age group
22-50
42 (48.8)
51-69
44 (51.2)
Gender
Male
38 (44.2)
Female
48 (55.8)
Relationship status
Married/cohabiting
67 (78)
Single
19 (22)
Employment status
Employed
77 (90)
Other†
9 (10)
Education level
≤High school
45 (52)
Bachelor’s or higher degree
41 (48)
Sleep quality
Very good/good
73 (85)
Fair
13 (15)
Physical activity frequency
<Once per week
10 (11.6)
Once per week
5 (5.8)
Twice per week
16 (18.6)
3-4 times per week
25 (29.1)
5-7 times per week
30 (34.9)
Physical activity intensity
Light
14 (16.3)
Moderate
40 (46.5)
Vigorous
32 (37.2)
Smoking
Never
50 (58.1)
Former smoker
25 (29.1)
Daily
4 (4.7)
Occasional
7 (8.1)
Drinking frequency‡
Less than once per month
19 (22.9)
Once per month
16 (19.3)
2-3 times a month
26 (31.3)
Once per week
4 (4.7)
2-3 times per week
15 (18.1)
4-5 times per week
3 (3.6)
Typical drinks per occasion
1
10 (11.6)
2
34 (41.0)
3
27 (32.5)
4+
12 (14.5)
Data are presented as number (percentage). Column percentages are given. †Other: unemployed, retired, and students. ‡Only includes current drinkers.