Health-related quality of life of pharmacists in the pre-COVID-19 era

Background pharmacists remain on the frontline of public health around the globe and their performance directly impacts patients’ safety. So far, to our knowledge, no European study has been dedicated to their heath-related quality of life (HQoL). Therefore, the primarily aim of our study was to evaluate HQoL of Polish pharmacists utilizing the SF-36 health survey with regard to anthropometric and lifestyle-related variables. a total sample screened consisted of 1412 respondents, yet 765 pharmacists (mean age 40, 86.3% females) nally participated in the study. HQoL was assessed with the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36). the lowest median scores were noted for general health (GH, 50.0) and vitality (V, 60.0) domains. No gender differences regarding physical and mental summary scores were found. Signicant difference of HQoL was found among the assessed age groups in several domains, especially physical functioning (PF) and GH (p < 0.001) scores, and especially in the group of 51-60-year-old-respondents. Correlations were found between PF (r=-0.29,p < 0.001), GH (r=-0.25,p < 0.001) and age as well PF (r=-0.27,p < 0.001), GH (r=-0.21,p < 0.001) and BMI. Self-assessed dietary habits were correlated with PF (r = 0.22,p < 0.001), mental health (r = 0.25,p < 0.001), GH (r = 0.27,p < 0.001) and V (r = 0.30, p < 0.001) scores. our analysis indicates that pharmacists tend to have similar mental and physical burden according to SF-36, with age, BMI and dietary habits as predominant factors inuencing their HQoL. The study presents unique values for future comparative analyses related, for instance, to the inuence of the ongoing pandemic on HQoL of health-care providers. of life of Polish pharmacists utilizing the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36) with regard to anthropometric and lifestyle-related variables. scores, excluding BP and RE (p = 0.05 and p = 0.78, respectively) was found, while the correlation analysis showed only weak correlations with PF (r = 0.21, p < 0.001), V (r = 0.23, p < 0.001), PCS (r = 0.24, p < 0.001) scores and very weak with GH and MCS (both r = 0.18, p < 0.001) scores. physical with self-assessed dietary habits (r = 0.29, p < 0.001) and nutrition knowledge (r = 0.22, p < 0.001). The Kruskal-Wallis test also revealed signicant difference (p < 0.005) in self-assessed nutrition knowledge and both summary scores and PF, GH, V, SF and MH domains, while the correlation analysis showed only weak correlations with PF (r = 0.22, p < 0.001), V (r = 0.19, p < 0.001) and PCS (r = 0.24, p < 0.001) scores. Signicant difference (p < 0.005) in self-assessed dietary habits and both summary scores of SF-36 (PCS, MCS) and all domains (except BP) was found, while the correlation analysis showed only weak correlations with PF (r = 0.22, p < 0.001), MH (r = 0.25, p < 0.001), GH (r = 0.27, p < 0.001), V (r = 0.30, p < 0.001), PCS (r 0.27, p < 0.001) and MCS (r p < 0.001) scores. Signicant difference also found between sleep on as well as weekends and and SF yet with very week correlations p and


Introduction
Quality of life (QoL) is an individual construct, includes subjective evaluations, and can be de ned differently. The concept of health-related quality of life (HQoL) has evolved since the 1980s to comprise those aspects of overall quality of life that can be certainly shown to affect either physical or mental health [1]. The Short Form Health Survey (SF-36) is a patient-reported questionnaire standardized and introduced in the late 1980s [2][3][4][5]. It intends to assess those health-related aspects of QoL, yielding two summary measures, physical and mental health [4]. The versatility of the SF-36 survey enables to assess the impact of various medical procedures on a patient's well-being [6], but it has also been successfully applied to measure the health-related quality of life of general population, including healthy individuals [7][8].
Pharmacists, similarly to other healthcare workers, are widely perceived to face a signi cant work-related stress, exhaustion, disengagement from work and diminished quality of life. Notably, job satisfaction among pharmacists promotes their performance and has a positive impact on patients [9], and especially patients' safety [10]. Consequently, several studies have been dedicated to various factors impacting their QoL as well as tools to measure work-related dimensions of well-being. In a national survey of the American pharmacists, for instance, a great number of respondents (68%) reported job-related stress [10]. Yet, the majority of studies have utilized author-made questionnaires and focused only on psychological and social QoL rather than general health status [10][11][12][13][14][15]. This tendency seems to be an oversimpli cation of the pharmacists' QoL as it largely ignores the physical burden imposed by their job, such as, for example, the high risk of developing varicose veins as a consequence of standing position for long hours [16].
In Poland (2018), there were more than thirty-four thousand of pharmacists according to the Polish Central Registry of Pharmacists, in comparison to around eighteen thousand of laboratory diagnosticians, one hundred and ninety thousand medical doctors and around two hundred and ninety thousand nurses (according to national registries). However, to our knowledge, no prior studies have examined HQoL of this group of healthcare workers. Therefore, the primarily aim of our study was to evaluate the heath-related quality of life of Polish pharmacists utilizing the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36) with regard to anthropometric and lifestyle-related variables.

Study sample, recruitment and data collection
The study protocol was approved by the Jagiellonian University Bioethics Committee (protocol number -1072/6120/197/2017). The self-administered version of the SF-36 was available online (Jagiellonian University Medical College domain "ankiety.cm-uj.krakow.pl") and was only addressed to Polish pharmacists with a valid licensure (an only inclusion criterion). There were no exclusion criteria. Information about the questionnaire was published on the Internet site of the Pharmacists' Chamber of Lesser Poland and portal e-farmacja.pl, and was also included into newsletters to collect as many responses as possible. The study sample was recruited between May and October 2018.
A total sample screened consisted of 1412 respondents, yet in the preliminary analysis 646 respondents were excluded due to signi cantly missing data of the SF-36 answers [4]. Finally, 765 pharmacists (660 women and 105 men) participated in the study. All respondents agreed to participate in the study.
The SF-36 health survey The questionnaire included general information and a Polish translation of SF-36 [17] consistent with the Medical Outcomes Study 36-Item Short Form Health Survey (RAND-SF-36) [18]. General information was collected on the date of birth (used to calculate age), sex, tobacco and alcohol use, sleep time, physical   activity, self-assessed nutrition knowledge and self-assessed dietary habits as well as voluntary information regarding weight, height (used to calculate body  mass index, BMI) and waist circumference (WC). Brie y, the SF-36 data were separated into two major domains -physical component score (PCS) and mental component score (MCS). The PCS was divided into four domains: 1) physical functioning (PF, 10 items) that describes limitations of basic physical activities (e.g. walking) due to health impairment; 2) rolephysical (RP, 4 items) concerned with di culties in performing work or daily activities due to poor physical health; 3) bodily pain (BP, 2 items) that describes presence of pain and limitations imposed by it; 4) general health (GH, 5 items) based on subjective evaluation of respondent's health. The MCS was divided into: 1) vitality (VT, 4 items) that describes presence of energy and fatigue; 2) social functioning (SF, 2 items) that describes limitations in social activities due to health problems; 3) role-emotional (RE, 3 items) that describes di culties with performing work or daily activities due to emotional problems; 4) mental health (MH, 5 items) that describes the presence of nervousness or depressiveness. The raw score of each of the eight SF-36 dimensions was derived by summing the item scores and converted to a value for the dimension from 0 (worst possible health state) to 100 (best possible health state) [4,18].
The SF-36 has performed well in tests of content, construct and criterion validity [4,[19][20][21], including the Polish version of the survey [22]. The internal consistency of the SF-36 items was assessed by Cronbach's α coe cients since the questionnaire was administered only once, making the examination of the test-retest internal consistency impossible. A Cronbach's α value of 0.7 or higher was considered to be su cient to demonstrate internal consistency [4,18].

Statistical analysis
Statistical analysis was performed with Statistica 13.3 software (TIBCO Software Inc., Palo Alto, California, USA; Jagiellonian University license). All the distributions were evaluated for normality with Shapiro-Wilk test. The distribution of all analyzed scores did not meet the criteria of normality and nonparametric tests were applied. Respondents were divided into age groups based on the decade of life (below 30, 31-40, 41-50, 51-60, plus 60) as well as

Results
The overall characteristics of the analyzed group are given in Table 1. The majority of respondents were females (86.3%), which has been representative for this health-care profession in Poland. The majority of pharmacists (93.7%) were current non-smokers with BMI levels below 25 kg/m 2 (65.4%). Only the minority of respondents reported no alcohol consumption (21.2%). The majority of respondents also reported 7-8 hours of sleep time (74% and 69.2% on weekdays and during weekends, respectively) and moderate physical activity (42.2%) at work as well as recreational physical activity (47.1%). 1,2 Information regarding weight and height (used to calculate BMI) as well as waist circumference (WC) was given voluntarily; 3 Recreational physical activity was categorized as follows: low -mostly sedentary, watching TV, reading newspapers/book, light house works, walking for 1-2 h a week; moderate-walking, cycling, exercise, gardening, or other light intensity physical activity for 2-3 h a week; higher-cycling, running, gardening, or other sport activities that require physical activity for more than 3 h a week; 4 Physical activity at work was categorized as follows: low -with over 70% of sitting time; moderate -with 50/50 standing-to-sitting ratio; high -with over 70% of standing time.
The general health domain had the lowest median score (50.0), while role-physical and role-emotional domains had the highest median scores (100.0). Detailed scores are summarized in Table 2. Evaluation of internal consistency showed that Cronbach's α coe cients ranged from 0.75 to 0.82 across the eight SF-36 dimensions ( Table 3). The Spearman's rank correlation analysis showed higher correlations between the physical and mental summary scores and domains inside than those between the summary scores and domains outside. The strongest correlations were noted between MH, V and MCS as well as GH and PCS (Table 3).     Me -Median; 1 One respondent refused to give information regarding his date of birth used to calculate age; p -signi cance level of the Kruskal-Wallis (for "All") and the Mann-Whitney U (for pair comparisons) tests. Table 6 Comparison of HQoL according to SF-36 with respect to BMI 1 (n = 742).

Discussion
Our data present values for the eight dimensions and two summary scores of the SF-36 health survey from a random sample of Polish pharmacists. The SF-36 instrument has several advantages, including standard criteria for scoring and transforming the raw responses to ensure accurate group comparisons and it has been found suitable for use in a variety of populations [4,18,21,23]. Our results showed evidence of acceptable validity and internal consistency for the SF-36 in the sample, corresponding with the existing ndings [4,21,23]. Yet, uneven number of men and women might have made the statistical analysis less valid, however, such proportions are representative for this group of health-care providers in Poland and Europe [24]. The method of enrolling respondents and a low response rate might also lead to self-reporting bias [25]. At the same time, since this is the rst study assessing quality of life of Eastern European pharmacists, it presents unique values for the SF-36 survey for a representative sample ideal for future comparative analyses related to for instance, the in uence of the ongoing COVID-19 pandemic on HQoL.
No normative published data of the general population could be found for a Polish version of SF-36 [17] and as a result we could not perform any reliable comparative analysis. However, in comparison with other European populations, both summary scores of Polish pharmacists were higher in comparison with, for example, Norwegian [26] and Irish [27] general population normative data. However, in comparison to British normative data, with occupation taken into account, our scores of separate domains were comparable or lower (RP, BP, GH, MH domains). In general, there were no gender differences regarding PCS and MCS, however SF score was signi cantly different among male pharmacists (Table 4). Thus, our results contrast with the ndings of Jenkinson et al, who reported that women had poorer health scores than men in all but the general health dimension [7]. Our analysis revealed that an increasing age was associated with a modest decline in PF and GH domains as well as the PCS score. The lowest median and mean scores for both summary scores and most of the domains (except for V and MH) were noted not for the oldest respondents, but in the age group of 51 to 60 year-olds. We suppose this could be attributed to the fact that the eldest respondents were very close or past the retirement age (60 for females and 65 for males in Poland). Consequently, this group was built up of individuals that had postponed their retirement, possibly due to high HQoL and high job satisfaction. Lau et al. found that both job satisfaction and career satisfaction increased with age, and lower job satisfaction of younger generation of pharmacists might stem from unmet work-balance expectations [28]. This tendency has been re ected in other studies, too [29]. Thus, deterioration of HQoL among older pharmacists might result mainly from physical aging. Yet, it should be noted that our respondents were not asked whether they had been diagnosed with a chronic conditions nor were there any use of medicines at the time of the survey recorded. Instead, we analyzed anthropometric and lifestyle-related variables, including sleep time. Interestingly, sleep duration was negatively correlated, yet very weakly, with SF score only. Both excessive and insu cient sleep duration has been associated with altered HQoL [30] and obesity [31], however the majority of our respondents maintained healthy sleep time routine, i.e., 7-8 hours. And no correlation was found between sleep time and BMI in our sample of pharmacists. BMI is known not only to deteriorate HQoL [32][33] [38]. However, the same mechanism seems inapplicable to pharmacists since general health perception according to SF-36 was relatively low in our sample. Low GH score (the lowest median score of all PCS domains) might indicate low self-perception of physical wellbeing among sampled pharmacists. Therefore, we believe that limiting pharmacists' HQoL to psychological factors only, such as job related stress, and neglecting the physical health dimension [11,39] does not constitute an optimal approach, although the data regarding factors in uencing mental well-being of pharmacists have been vastly available. According to American survey 57% of the respondents considered quitting their job at least once per year, and 20% believed that stress imposed by their job affected negatively their mental health, social life, work performance, well-being and physical health [14]. A study of pharmacists from Northern Ireland indicated that as many as 30% percent of communal and 50% of hospital pharmacists reported being often or frequently in stress. Frequent interruptions, shortage of staff and excessive workload were listed as the main stressors [12]. Similarly, a survey of American pharmacists pointed out to the problem of high workload. Authors found correlation between high job satisfaction and self-perceived completeness of patients' review (treated as a measurement of job performance). They also implied that burnout negatively impacted the e cacy of their work. Although the authors did not examine QoL directly, their ndings might suggest that the lower QoL should negatively impact work outcomes of pharmacists [39]. Still, none of the studies have utilized the SF-36 (or any other generic) questionnaire to measure mental well-being among pharmacists. The mental health summary score in our sample was only positively correlated with recreational physical activity and dietary habits. And indeed, physical activity together with healthy eating habits are known to provide mental health bene ts [40]. It is also worth mentioning that the highest mental summary score was reported by the eldest respondents, while the lowest median score of the mental health domain was reported by the youngest and 51 to 60 year-old-respondents. Thus, we believe that poorer mental health, especially among younger generations of pharmacists, is an important issue and should be appropriately handled.

Conclusions
To our knowledge this is the rst study assessing quality of life of European pharmacists using a validated, reliable, generic HQoL questionnaire. Our analysis indicates that this group of health-care providers tends to have similar mental and physical burden, with age, BMI and dietary habits as predominant factors in uencing their HQoL. We do believe that our research might raise awareness about pharmacists' HQoL, and subsequently lead to an implementation of effective countermeasures.

Declarations
Funding: None

Con icts of interest/Competing interests: None
Ethics approval: The study protocol was approved by the Jagiellonian University Bioethics Committee (protocol number -1072/6120/197/2017. Consent to participate: All respondents agreed to participate in the study.