Trachomatous Trichiasis (TT) management in Tanzania: a mixed method study investigating barriers and facilitators to obtaining treatment

Background Prolonged ocular Chlamydial infection, known as trachoma, can lead to trachomatous trichiasis (TT). TT is the stage of trachoma where the eyelid turns inwards, resulting in lashes rubbing against the cornea. TT can damage the cornea, leading to vision impairment or blindness. Treatment for TT includes epilation or surgery. Trachoma is targeted for elimination as a public health problem. One criterion of trachoma elimination is less than 0.2% prevalence of TT unknown to the health system in adults >= 15 years. There are several districts in Tanzania that have not attained this target. Methodology We selected six districts across three regions in Tanzania. Our mixed-methods approach included a retrospective review and analysis of program data and implementation of key informant interviews (KII) and focus group discussions (FGD). The desk review collated data on district-level indicators and generated estimates around number and proportion of cases not identified by case finders and cases lost along the continuum of care. KIIs and FGDs guides were structured to enlist responses around case finding techniques, linkage to services and TT surgery process. Conclusion We found a substantial proportion (13%) of TT positive people were not being identified by case finders, and of those identified, majority (72%) were lost along the continuum of care. These factors likely contribute to high TT prevalence in districts where surgical interventions are ongoing. Engaging community leaders to share TT information and enlisting people who have received surgery to witness in communities may encourage consent of examination by case finders and increase surgical uptake. After witnessing positive effects of surgery, many interviewees who had previously declined surgery changed their mind. Increasing frequency of surgical camps would improve access to these populations. Additionally, giving more notice about surgical camps and extending duration is important to enable remote populations to obtain services. Author Summary Treatment for trachomatous trichiasis (TT) includes epilation or surgery. There are several districts in Tanzania that have struggled to link people with TT to services. It is important for the program to understand why this is the case to inform program adaptations for improved linkage to services. We implemented a mixed methods approach to address this knowledge gap. We found a large portion of TT positive people are not being identified by case finders and of those identified, many are lost along the continuum of care. These factors are likely contributing to the unexpectedly high TT prevalence in districts where surgical interventions are ongoing. Barriers to identifying cases included remoteness, case finder credibility, knowledge of TT, and case finder motivation. Once cases are identified, the largest gap along the continuum of care is the link between being identified and screened. We found barriers to attending screenings and subsequently obtaining treatment to be fear of surgery, distance from surgical camps, agricultural season, time to plan, awareness and frequency of camps, and lack of assistance after surgery.


54
Prolonged conjunctival infection with Chlamydia trachomatis leads to an inflammatory response, 55 trachomatous inflammation-follicular (TF) and trachomatous inflammation-intense. Overtime, cycles of 56 repeated infection can progress to scarring of the conjunctiva, causing entropion inward turning of the 57 eyelid and resulting in lashes rubbing against the cornea. This painful stage of the disease is called 58 trachomatous trichiasis (TT). TT can damage the cornea, leading to vision impairment or blindness [1,2].

59
Trachoma is targeted for elimination as a public health problem [3]. One criterion of trachoma 60 elimination is prevalence of TT unknown to the health system of less than 0.2% in adults 15 years and 61 older [4]. "Unknown to the health system" are cases that have not previously been operated on, not 62 previously refused treatment, or have not been referred for treatment [5]. District-level prevalence is 63 estimated through population based prevalence surveys [6] and these prevalence estimates are used as 64 a guide for planning interventions [7,8]. Studies have shown that TT surgery results in improved vision 65 and physical function [9] as well as reduced photophobia and pain [10]. The World Health Organization 66 (WHO) recommends a bilamellar tarsal rotation procedure to correct entropion caused by TT [11]. While 67 not recommended by WHO, epilation is commonly practiced to manage minor TT and may lower risk of 68 corneal opacity [12,13]. Regardless of vision loss, untreated TT has been shown to significantly reduce 69 quality of life [14,15].

70
Linking TT positive people to TT surgery is a major concern in trachoma endemic settings. In Tanzania, TT   71 surgery is provided free of charge and surgery programs are active in six regions (Mtwara,Pwani,72 Dodoma, Arusha, Manyara, and Lindi). The continuum of care for TT is the following: 1) Case finders 73 identify a positive TT case, 2) Eye care professionals screen the identified case, 3) TT surgeons confirm 74 the screened case, and 4) Treatment is provided.

75
In 2016, the TT backlog in Tanzania was estimated to be 214,800 [16], and in 2017, only 2,120 people 76 were reported to have received surgery [17]. This equates to less than 1% of people needing surgery  We next explored the continuum of care and determined the month each step along the continuum 136 occurred, time lag between each step (estimating the median and interquartile range (IQR)), and 137 proportion of cases that drop out before surgery and where along the continuum this occurred. To 138 estimate the number of positive cases lost between being identified by case finders and screened, we 139 assume that those who did not attend screening similar to those who did. We estimated the number of 140 positive cases lost between identification and screening as 141 , 142 where x is percent confirmed positive, y is percent lost before screening, and z is number identified by 143 case finders.

144
With these findings, we estimated the proportion of the TT backlog lost along the continuum of care by 145 multiplying the dropout proportion at each step by the total backlog. We then calculated the number of 146 cases not identified by case finders as 147 .
Finally, we ran regression models to determine significance of association between cases being screened 149 and district, sex, and month of identification, as well as association between cases treated and district.

150
The quantitative estimates are not meant to draw broad conclusions. These estimates are illustrative further clarified on the purpose of the study and was ensured of privacy and confidentiality. As part of 165 the consent process, we informed participants they could end the interview at any time or refuse to 166 answer any question. Personal identifiers were removed from all datasets before analyses were 167 undertaken. 168

169
Here we first present the quantitative findings and then the themes identified during the qualitative 170 component of this study.

172
We estimated current backlog across the six districts to be 5,229 people, over half of which are expected 173 to be found in Bahi and Tandahimba.