The findings of this initial KPR cohort provide valuable insight into the demographics, clinical profile, quality of life, and mental health patients with PsO in Kenya. Healthy controls and PsO were similar in most demographics, except for a significant difference in age at time of enrollment. Interestingly, both healthy controls and PsO had a high rate of food insecurity (Table 2). We used the Hunger Vital Sign™ to assess food insecurity, which is a two-question screener that has been validated for both adults and children in the USA [10]. According to this screener, food insecurity was estimated to be present in 14.2% of US households on the basis of the 2013 Current Population Study [10], while 58.3% of patients with PsO and 61.3% of healthy controls in the KPR screened positive for food insecurity. This difference is likely multifactorial; this screening tool has not been validated in Kenya, and there are differences in the food infrastructure and availability in Kenya [11]. Tools such as the Food Insecurity Experience Scale (FIES) have been validated in sub-Saharan Africa, and could be considered for future studies [12].

Surprisingly, few patients reported medical comorbidities (Supplemental Table 1). There could be underreporting, as evidenced by the lower rate of reported hypertension compared to measured hypertension (Supplemental Tables 1 and 2). Potential reasons for underreporting include fear of stigma when disclosing medical history, uncertainty of one’s own medical history, or unfamiliarity with terms presented in the registry surveys. Despite universal healthcare coverage in Kenya, access to public and private-for-profit healthcare facilities varies widely. Nationally, the average travel time to public and private healthcare facilities is 130 and 245 min, respectively [13]. It is possible that many KPR patients do not see medical providers frequently enough to be diagnosed with significant medical comorbidities. This may also explain the low percentage of patients with PsO with a family history of PsO (Table 3). Estimates on the percentage of patients with PsO with a family history of PsO range from 10.8% [14] to 31.9% [15] to 36% [16] in Japan, Canada, and the USA, respectively. Additionally, diagnosis of psoriatic arthritis was reported by only 9.3% of patients with PsO (Table 3), but psoriatic arthritis typically affects up to 30% of patients with PsO at some point in their lifetime [17]. Mean age of psoriatic arthritis symptom onset was 50.9 years (SD 17.9), with mean age of diagnosis of 54.6 years (SD 21.1) (Table 3), which is aligned with the average diagnostic delay of psoriatic arthritis of up to 7 years [18].

Plaque PsO was the most common subtype that patients reported ever having (Table 3). Guttate PsO was reported in 7.4% of patients with PsO, making it slightly more common in this cohort compared to previous reports that estimate its prevalence to be 5% of all PsO [19]. Furthermore, a history of erythrodermic PsO was reported by 4.6% of patients, compared to 1–2.25% of PsO previously reported [20]. Finally, genital PsO was reported by just one participant, and inverse PsO was not reported by any participant. Given inverse and genital PsO affects about 21% of patients with PsO [21], this likely represents underreporting due to unfamiliar medical terminology, or anticipated and perceived stigma amongst patients with inverse and genital PsO [22]. Notably, 19.4% of patients with PsO in this study reported armpits, groin, skinfold, or gluteal fold involvement, and 11.1% reported genital involvement at some point in their PsO history (Table 3).

PsO severity was assessed using PASI and IGA scores. Mean PASI score was 9.93 (SD 9.1), indicating moderate disease activity [23]. The mean IGA score was 2.99 (SD 1.0), indicating mild to moderate disease. For comparison, a PsO registry study conducted in the UK found that most patients with PsO had mild or moderate disease [24].

In terms of treatment history, most patients have tried moisturizer or specialty topical medication (Fig. 1). Only 9.3% of patients with PsO have tried biologic medications, compared to a US study where 37.2% of patients with moderate to severe PsO receiving treatment were on biologics [6]. Of note, biologic medications for PsO are not currently on the Kenya Essential Medicines List, which is a comprehensive list of essential medication information that helps guide their procurement in the public sector [25]. Additionally, biologics for PsO are expensive, with cost ranging from $1664 for biosimilar versions of infliximab to $79,277 for risankizumab [26]. Of the Kenyan population, 36.1% lives on less than $2.15 per day [27], greatly limiting access to these costly medications. Older topical and systemic treatments are still utilized in Kenya, with crude coal tar treatment being reported by 19.4% of patients with PsO, although it is rarely still used in the USA [28]. Crude coal tar treatment a fairly inexpensive yet highly effective treatment for plaque PsO, making it a feasible option for this patient population [28]. Methotrexate usage was reported by 75.9% of patients with PsO, making it one of the most common treatments (Fig. 1). In a US patient registry, non-biologic systemic medications including methotrexate were received by approximately 13% of patients with PsO [29], and 11.7% of patients with PsO in a Belgium registry received methotrexate [30]. The percentage of patients with PsO in the KPR who have ever received methotrexate is substantially greater in comparison, perhaps owing to its relatively low cost [31].

The sleep, quality of life, and mental health patient-reported outcomes provided valuable insight into the experience of living with PsO in Kenya (Fig. 2). Patients with PsO scored one standard deviation below the mean for the PROMIS Sleep Disturbance instrument, meaning on average they reported less sleep disturbance than the mean used to validate this instrument [32]. However, patients with PsO scored about one standard deviation above healthy controls, meaning they do have reported worse sleep compared to patients without PsO. This instrument has not been validated in a Kenyan population, perhaps explaining why both PsO and healthy controls appear to have better sleep than average. Importantly, patients with PsO still had worse sleep compared to healthy controls. On average, skin disease had a moderate effect on quality of life for patients with PsO, compared to no effect for healthy controls, as measured by the Dermatology Life Quality Index. Generalized Anxiety Disorder-7 (GAD-7) and Patient Health Questionnaire-9 (PHQ-9) scores were also significantly different between PsO and healthy controls, indicating worse mental health amongst patients with PsO. A GAD-7 score of 3.1 for patients with PsO represents minimal anxiety [33], while a PHQ-9 score of 4.8 for patients with PsO represents mild depression [34]. Importantly, both PHQ-9 and GAD-7 have previously been validated in Kenya [35,36,37]. These results indicate there is a significant mental health and quality of life impact of PsO in this Kenyan patient population.

There are several limitations to this study. First, we relied on patient self-report for the majority of demographic information and clinical history, which may be subject to recall bias or error. Despite obtaining informed consent and maintaining privacy, it is possible some patients did not feel comfortable disclosing some information to the study team. Additionally, as previously mentioned, it is possible some patients did not know some of their own or family medical history. One limitation of the study was the broad age range of the PsO cohort compared to the healthy control cohort. There was a statistical difference in mean age of the two groups. However, multivariate analysis adjusted for age when assessing the effect of psoriasis on mental health outcomes, and minors were excluded from this analysis. Another limitation of the study was the single-center design. Subjects were recruited exclusively from MTRH, limiting the generalizability of the findings to this patient population. Of note, MTRH is a national referral hospital and serves about 24 million people from Western Kenya and neighboring countries. Finally, we relied on three validated instruments (PROMIS Sleep Disturbance Scale, DLQI, and Hunger Vital Score™) that have not been validated in the Kenyan population, potentially limiting the generalizability of our findings. While there were no available validated tools equivalent to PROMIS Sleep Disturbance Scale or DLQI, the Food Insecurity Experience Scale (FIES), which has been validated in sub-Saharan Africa, could be considered for future studies in place of the Hunger Vital score [12]. Despite these limitations, our results provide much needed information on PsO in Kenya.