Table of content
- Introduction
- Results
- Hair loss is associated to the age group
- 1.1 Boxplots give insights into the proportion of hair loss in different life phases
- 1.2 ANOVA and Tukey HSD reveal no statistical significance between age groups
- A generalised linear model (GLM) is able to identify several factors as important for hair loss prediction
- Is there a correlation between stress and hair loss?
- 3.1 Using Principal Component Analysis (PCA) to find associations between stress and hair loss
- 3.2 The proportion of hair loss under different stress levels support findings of PCA
- Hair loss is associated to the age group
- Discussion
- Conclusion
- Literature
Losing hair is not a shame!
Introduction
Hair loss is a multifactorial symptom that can precipitate significant psychological distress in those affected (Wells et al., 1995). Such consequences may include depression, a reduction in self-esteem, or social difficulties. These consequences manifest with particular frequency in women (Hadshiew et al., 2004).
Hair loss can be attributed to a multitude of factors, and it is a common and predictable side effect of various illnesses and medical treatments (e.g. chemotherapy) (Hesketh et al. 2004). The numerous factors that contribute to hair loss are so varied that, in addition to the readily apparent causes, it can be challenging to diagnose and identify the underlying cause (Jackson & Price, 2013).
Factors that are frequently disregarded include diet, medication, or specific diseases that do not align with the conventional understanding of hair loss causes. Consequently, hair loss in patients is often attributed to stress or age (Lin et al., 2016). A comprehensive examination of the contributing factors may assist patients and caregivers in mitigating the likelihood of hair loss.
This paper will examine the most significant factors associated with hair loss, employing a binomial linear regression to ascertain the most influential factors and illustrate the prevalence of hair loss across different age groups. Additionally, it will utilize a principal component analysis (PCA) and a biplot to visualize the impact of stress intensities on hair loss.
Results
1. Hair loss is associated to the age group
1.1 Boxplots give insights into the proportion of hair loss in different life phases
This chapter presents an analysis of the distribution of hair loss across different age groups. The data indicate that the prevalence of hair loss increases with age. This trend appears to reach a peak between the ages of 30 and 40, before declining again in the age group between 40 and 50 (see Figure 1).
Figure 1: Boxplots showing the proportion of people with "Baldness" (brown) or "Thick Hair" (green) at different age groups: <20, 20-30, 30-40, 40-50 and 50-60 years.
1.2 ANOVA and Tukey HSD reveal no statistical significance between age groups
An ANOVA and a TukeyHSD test was conducted to investigate wether the differences plotted in Figure 1 were statistical significant (refer to Table 1). All comparisons showed no statistical significant differences (p > 0.05), although the difference between age group 50-60 Thick and age group 50-60 Bald shows a trend to significance (p = 0.092).