Essential steps in the hair collection protocol include providing an example of how much hair will be taken, showing the sample site (posterior vortex), use of the single strain twist and gently pull method, and securing the sample with painter's tape away from the scalp end. By gently twisting and pulling the sample, the curly hair is lengthened and the scalp end alignment is maintained. This reduces laboratory time needed to align the scalp ends, measure, and cut the sample. Although thinning shears have been used in prior research, they are not the preferred tool for cutting hair because of uneven scalp end alignment and increased time to prepare the hair sample for analysis.
Gathering hair samples from African Americans is feasible for measurement of cortisol. Appreciating the significance of hair in the African American culture is necessary to garner support for collecting hair samples. Investigators should not convey judgment regarding hair hygiene, such as the frequency of hair washings, because some African Americans tend to have dry hair as compared to non-Hispanic whites13. African American hair grows slower, at a mean rate of 256 µm/day-1 as compared to non-Hispanic whites (M = 396 µm/day-1, SD = 55 µm/day-1)14. Participants may also wear hairstyles such as weaves/sew-in hair, box braids, and kinky twists or crochet braids. These hairstyles can take hours to complete, with costs ranging from $50 to more than $500. When possible, inquire about the hairstyle and coordinate hair sample collection with the participant. In these cases, the natural hair may need to be collected between beautician appointments or at the salon.
The lack of appreciation by the researcher regarding the importance of hair, style preferences, and costs can lead to participant unwillingness to provide a hair sample. As reflected in the current study, this phenomenon of African American participants' reluctance or inability to provide hair samples was documented in a study of multi-race youths and the feasibility and acceptability of hair collection for cortisol measurement8. Ford et al. (2016) sampled 516 mixed-race youths aged 11-17 from the community. The African American adolescents were less likely to provide a hair sample due to shortness of hair and refusal (OR = 0.26; 95% CI [0.13, 0.51])8.
The participant in the present study were age 65 and older. The greatest limitation to hair cortisol analysis is shaved heads, which is common among males or those with thinning hair. Gathering enough hair for samples in older adults can be challenging due to hair loss. The hair weight matters more because if there is not enough mass, then the microplate reader cannot accurately detect the cortisol levels. Standardizing the hair weight as a normalizing factor prior to processing samples for cortisol analysis may be a future consideration to address this challenge.
Several limitations have been outlined in previous hair cortisol studies, including variations in hair care practices and the lack of normative values for cortisol results8,12,15. The use of hair straighteners, relaxers, permanents, conditioners and dyes alter the level of cortisol in the hair. Exposure to bleaches and permanents decreases the amount of cortisol, while peroxides increase measured cortisol15. Also, the lack of normative values presents a challenge to comparing hair cortisol results across studies to establish normal values. The hair cortisol findings from our study are limited and could only be generalized to other older adults and persons with chronic conditions. Generally, hair cortisol levels in persons with chronic conditions (diabetes, cardiovascular) and mood disorders are higher than those without chronic conditions9,16. Another study of community-dwelling African American older adults reported cortisol ranges from 0.93 to 6.36 pg/mg, which was lower than the range in our study16. Another study of a sample of multiethnic older adults with cardiovascular disease reported a median cortisol level of 22.1 pg/mg, similar to the findings of this study.
Future applications of the techniques outlined in this paper should include teaching the participant the twist and gently pull procedure, enabling them to collect their own hair samples. Self-sample collection has been done in African American women with breast cancer and was considered more acceptable16. However, the response rate for African American women providing a hair sample was poor in the breast cancer study. To improve participation of African Americans in studies that collect hair biospecimens, research is needed to understand better the reasons for refusal and any potential age or cohort demographic associated influences on the refusal rates. As found in the present study, approaching potential participants in a non-judgmental fashion, acknowledging the importance of hair and hair practices within the African American community, sharing a photo of the hair sample, providing privacy, and adoption of the twist and gently pull method were all factors that attributed to the success of getting participants enrolled in the study.
At present, no published protocols consider cultural preferences to collect hair samples for cortisol. Likewise, few investigators are trained to collect hair from African Americans. The single strain twist and gently pull to cut the hair sample protocol can be used in other studies that require hair samples from African Americans. This protocol has the potential to increase the number of hair samples from African Americans, as well as the number of culturally informed investigators and data collectors.