Discussion
The present study aimed to investigate symptoms of EDs and LEA among recreational female runners. The main finding was that 18% of the females had symptoms of EDs and 19% had symptoms of LEA, while the coexistence of symptoms of EDs and LEA rarely occurred with only two subjects having symptoms of both.
Symptoms of EDs
In the present study, 18% of these recreational runners had symptoms of EDs, similar to the 18%–21% reported among female endurance athletes.18 19 In studies using clinical diagnostic interviews to confirm symptoms of EDs, the reported prevalence of EDs has been 24%–25% among female elite endurance athletes,20 21 21% among recreationally active females7 compared with 9% in the general population.7 21 In a meta-analysis, Chapa et al22 found that ED psychopathology varies in females and non-athletes depending on sport type, rather than on the competitive level with a higher ED psychopathology in aesthetic or leanness sports. In summary, the results from the present study demonstrate a concerningly high prevalence of symptoms of EDs in recreationally active females, comparable to that observed among the athletic population.
Depending on the ED symptoms and severity, athletes with EDs may be underweight, normal weight or overweight.2 In the present study, no difference in body mass was found between the subjects with and without symptoms of EDs which is in line with a study by Torstveit et al7 reporting an average body mass within the normal range in female elite athletes with clinical EDs. Also among females with or without symptoms of LEA, similar body masses have been reported,20 23–25 supported by the findings in the present study. These findings suggest that most females with symptoms of EDs or LEA often have a body mass within the normal reference range.3 26 Hence, body mass is not a useful indicator for identifying athletes or recreational active females with symptoms of EDs or LEA. Weight fluctuations are commonly reported among young female athletes with disordered eating behaviour or EDs,27 supported by the findings in this study where weight fluctuation was positively associated with EDE-Q global and eating concern scores. Moreover, restraint score was positively associated with higher BMI. These findings might reflect a preoccupation with thoughts and anxiety about food and eating and an intent to limit food intake8 whether or not the effort is successful.
The shape and weight concern subscales represented the highest scores, aligning with findings from previous studies among female athletes.18 19 28 In fact, non-athletes report more body dissatisfaction compared with athletes.22 With this in mind, it might be important to emphasise a positive body image, physical and psychological health benefits, and healthy dietary habits among recreational female runners.
Symptoms of EDs and LEA
In the present study, only 13% of the subjects with EDs had symptoms of LEA, and no associations were found between EDE-Q global and LEAF-Q scores. This finding confirms the broad spectrum of symptoms associated with EDs and highlights that EDs can manifest in various ways, including behaviours that may not necessarily result in undereating but still compromise the individual’s physical or mental health.29 Conversely, the findings in the present study emphasise that recreational female runners may experience symptoms of LEA in the absence of EDs symptoms. This is confirmed by previous research where 27% of female endurance athletes19 and 20% of female adolescent athletes30 had symptoms of both EDs and LEA. The relatively low prevalence of concurrent symptoms of EDs and LEA in this study, compared with previous research19 30 could potentially be attributed to the recreational activity level of the study subjects. Moreover, the finding in the present study that a higher weekly training load was associated with symptoms of LEA, raises the speculation that the subjects might not be aware of the energy cost of their exercise or are unable to consume an adequate amount of energy to match it.
In this study, 19% of the subjects had symptoms of LEA, similar to the 23% reported among Irish recreational active females24 but lower compared with the 45% and 63% reported in recreationally active females in New Zealand.25 31 The mean age of the subjects in the present study was 32 years and both symptoms of EDs and a higher LEAF-Q total score were associated with lower age, in contrast to previous research where a similar prevalence of ED psychopathology across age groups has been reported.22 Moreover, clinical studies have found that females with a gynaecological age <14 years are more sensitive to LEA than older females.32 It is, therefore, possible that the lower mean age (23–24 years) in the studies of recreational active females in New Zealand25 31 potentially could explain the higher reported prevalence of LEA compared with the present study. Consequently, the relatively low prevalence (15%) of menstrual dysfunction in the present study may be attributed to the subjects’ higher age and potentially reduced sensitivity to LEA.
In the present study, no associations were found between sports injuries and symptoms of EDs or menstrual dysfunction, which is in contrast to earlier studies reporting more sports injuries among young female athletes with EDs symptoms15 and menstrual dysfunction.14 15 33 However, the present study cannot conclusively determine if the reported sports injuries, menstrual dysfunction and gastrointestinal problems, as indicated by the LEAF-Q responses, are specifically caused by LEA, or could be attributed to acute injuries or other conditions compromising reproductive and gastrointestinal functions.
Strengths and limitations of the study
In this study, we combined two validated screening tools to better understand the associations between symptoms of EDs and LEA.34 Although LEAF-Q has been frequently used for screening for symptoms of LEA in populations with various athletic levels, including recreational active females, its validation has been limited to athletes.6 11 Due to the study’s cross-sectional design, it is not possible to determine the cause and effect of the results and the results are, therefore, limited to prevalence estimations. Furthermore, the study may be susceptible to ascertainment bias, specifically in terms of selection bias. It is possible that older females are more likely to participate in a survey compared with younger females, which could have resulted in a relatively high mean age (32 years) among the subjects. This potential bias as well as the convenience sample could limit the generalisability of the findings to the broader population of recreational female runners. Moreover, retrospective self-reported data in this study may be subject to recall bias and reporting bias, depending on the subject’s interpretation of the questions, honesty of completion and ability to accurately recall the data. For instance, under-reporting of eating pathology7 and difficulties remembering details of injury history such as the number of days injured is common.35 However, recalling injury status in the previous year (injured/uninjured) is considered accurate35 and self-reported menstrual dysfunction is a valid predictor for clinical menstrual dysfunction among athletes.6
Screening for symptoms of EDs and LEA may be a convenient method for identifying individuals who need to be encouraged to seek medical attention.3 To accurately confirm and differentiate the conditions, future studies among recreational females may, however, wish to investigate EDs and LEA using clinical interviews, biomarkers (eg, sex and thyroid hormones) and examinations (eg, bone mineral density and ovulation).