Operational analysis of the national sickle cell screening programme in the Republic of Uganda

Background Sickle cell anaemia is a common global life-threatening haematological disorder. Most affected births occur in sub-Saharan Africa where children usually go undiagnosed and die early in life. Uganda’s national sickle cell screening programme was developed in response to a 2014 sickle cell surveillance study that documented a high disease prevalence. Objective This study describes the temporal and financial aspects of Uganda’s 2014–2019 sickle cell screening programme. Methods National sickle cell screening data from Uganda’s Central Public Health Laboratories were used to calculate turn-around times (TATs) from sample collection to delivery, testing, and result reporting for blood samples collected from February 2014 to March 2019. The parameters affecting specific TATs were assessed. The exact programme expenditures were analysed to determine cost per test and per positive sickle cell disease case detected. Results A total of 278 651 samples were analysed. The median TAT from sample collection to laboratory receipt was 8 days (interquartile range [IQR]: 6–12), receipt to testing was 3 days (IQR: 1–7), and testing to result reporting was 6 days (IQR: 3–12). Altogether, the sample continuum averaged 16 days (IQR: 11–24). Lower level healthcare facilities were associated with longer sample delivery TATs. Calendar months (January and December) and larger sample volumes impacted testing and result reporting TATs. The cost per test was $4.46 (United States dollars [USD]) and $483.74 USD per positive case detected. Conclusion Uganda’s sickle cell screening programme is efficient and cost-effective. Universal newborn screening is the best strategy for detecting sickle cell anaemia in Uganda.


Introduction
Sickle cell anaemia is a monogenic haematological disorder that manifests as a devastating systemic disease with high morbidity and early mortality. The pathophysiology begins in infancy with acute and life-threatening complications, which presents as increased susceptibility to infections, chronic haemolytic anaemia, and painful vaso-occlusive events. 1,2 Sickle cell anaemia is the most prevalent haemoglobinopathy that impacts more than 20 million people worldwide, with an estimated 312 302 babies born with the disease each year. 3 Over 75% of the world's annual sickle cell anaemia births occur in sub-Saharan Africa where resources for providing early detection and care are most constrained, 3 leading to the deaths of more than 500 affected children per day. 4 http://www.ajlmonline.org Open Access early initiation of penicillin prophylaxis and pneumococcal immunisations as primary prevention. 9,10,11,12 However, the accurate diagnosis of sickle cell anaemia in limited-resource settings still faces major challenges related to high equipment and reagent costs, education and training of healthcare providers, and appropriate medical interventions. Newborn sickle cell screening requires efficient laboratory methodologies and infrastructure for an early and accurate diagnosis to reduce preventable deaths among children born with sickle cell disease and to support the ultimate goal of universal screening of all babies across sub-Saharan Africa.
The Uganda Sickle Surveillance Study (US3) commenced in 2014 using a simple but innovative approach to screen children for the sickle cell trait and disease on a national level. 13 Residual dried blood spot cards collected across the country as part of the Early Infant HIV Detection programme were tested. Isoelectric focusing (IEF) was conducted on about 100 000 samples over a cross-section of one year and found an overall high prevalence of sickle cell trait at 13.1% and disease at 0.7%, but with a disproportionate distribution across the country. Broader sickle cell screening has been initiated since 2015, strategically focused on the highest burden districts.
To expedite early detection and facilitate linkage to care for affected infants, it is crucial to identify where delays in diagnosis occur. Here, we describe a detailed operational analysis of the temporal and financial aspects of sickle cell screening in Uganda. We documented the turn-around times (TATs) for sickle cell screening, starting from sample collection to arrival, testing, and result reporting at the national centralised laboratory. To examine the cost implications of integrated sickle cell screening, we calculated the cost per test, based on exact expenditures from the US3, and made estimates of the cost-effectiveness of sickle cell screening in Uganda.

Ethical considerations
The initial US3 research proposal was approved by the School As previously described for the Early Infant HIV Detection programme, blood was collected from HIV-exposed infants and young children across the country using standard dried blood spot cards. 14 Each health facility populated a dispatch form with demographic information and testing requests. Samples were transferred to the CPHL via the national sample transport system. 14 In the US3, all samples requesting HIV testing were also queued for sickle cell testing; results were entered into the centralised database and disseminated back to health facilities and onward to caretakers. Following the US3, the dispatch form included an option for sickle cell testing only.

Turn-around time analysis
The CPHL database collects laboratory data on all samples received and processed; data for this analysis were abstracted for children aged 0-24 months who underwent routine sickle cell screening over five years, from 01 February 2014 to 31 March 2019.
Inbound TAT was defined as the total number of days between sample collection at health facilities and the result dispatch from the CPHL back to healthcare providers. Three phases were defined within inbound TAT: (1) sample delivery phase TAT is the time between sample collection and receipt at the CPHL, (2) sample testing phase TAT is the time between receipt at the CPHL and date of testing, and (3) sample result reporting phase TAT is the time between testing and result dispatch from the CPHL. Outbound TAT, defined as the number of days between result dispatch and result receipt or subsequent action by the healthcare providers, was not assessed in this analysis because those data points are not currently collected by the CPHL. Costs that were shared with other CPHL programmes and the health facilities were not included, such as the dried blood spot kits, sample transportation, local healthcare provider salaries, and screening-related training. Expenses incurred by Cincinnati Children's Hospital Medical Center for travel and initial training of CPHL sickle cell laboratory personnel were also excluded. Cost for equipment, reagents, and consumables was reported in United States dollars (USD) according to actual costs using out-of-country and in-country vendors. For labour costs, annual salaries were expressed in Uganda shillings and converted to USD using the purchasing power parity exchange rate. 16 The national sickle cell disease prevalence data for 2014-2019 and the calculated cost per test were used to generate the cost per sickle cell disease case detected by region within Uganda. For the sample delivery phase, the median TAT was 8 days (IQR: [6][7][8][9][10][11][12]. Parameters affecting sample delivery TAT included the region of the requesting hospital and the level of healthcare provided ( Table 2). The Kampala region, where the CPHL is located, had the shortest median sample delivery time of 6 days (IQR: 4-9), while six regions had a median sample delivery time of 9 days (Table 2). Regional referral hospitals had a shorter median sample delivery TAT of 7 days (IQR: 5-9) compared to the lower level health centres II, III, and IV. The collection month of December 2018 had the longest median TAT of 10 days (IQR: 5-18) compared to other collection months.

Sickle cell screening costs
Inputs included in the cost analyses were exact direct expenditures for equipment, reagents, consumables, and labour for 99 243 US3 samples tested by IEF from a dried blood spot at the CPHL over 13 months (    cost per test total at 52%, followed by reagents (23%), equipment (21%), and consumables (3%).
In Uganda, the highest burden of sickle cell disease is observed in the East Central (1.6%) and Mid-Northern (1.5%) regions (Table 6). Using the calculated total cost per test and national prevalence data, the cost per sickle cell disease case detected was $483.74 USD (

Discussion
For any newborn screening programme, reduction of sample TAT is an important intervention to ensure that poor health outcomes or preventable deaths are not the results of laboratory or operational delays. Identification of the specific barriers related to longer TAT and implementation of measures to mitigate those causes are essential to strengthen screening programmes.
In this first detailed operational analysis of TAT for sickle cell screening in Uganda, the total duration between sample collection at the health facility to the dispatch of results from the CPHL averaged 16 days (Figure 1). Multiple parameters affected this time interval, including the health facility level, programme year, collection month, and testing cohort. However, the vast majority (~86%) had results sent by CPHL back to the local healthcare facilities within one month of receipt (Table 1), which we propose as the maximal TAT threshold before putting sickle cell disease patients at risk by delaying diagnosis and treatment. Penicillin prophylaxis and pneumococcal immunisations are expected to be administered by three months of age, and these TAT results are sufficient to meet those goals. 9,10,11,12 For the Uganda sickle cell screening programme, sample delivery accounted for most of the inbound TAT (Figure 1). Lower health facilities likely have greater barriers to timely delivery, such as delayed collection of their samples for transport ( Table 2). Within each district, hub motorbike riders perform a weekly pickup from the more rural locations of health centres II, III, and IV in comparison to regional referral hospitals or district hospitals that experience daily pickup. It is also possible that samples are not readily available for transport at the time the hub motorbike riders come for collection, or some facilities are missed due to inclement weather or mechanical problems. To address these issues, a twice-weekly sample pickup   Inbound TAT 16 (11,24) Outbound TAT (Not available) with validation measures, such as timestamps, should be put into place. This should also be supplemented with ongoing training regarding sample preparation and the importance of timely sample transport for all personnel involved in the process.
After sample receipt at the CPHL, sample testing was relatively rapid with an average of only 3 days (Figure 1). This phase involves the routing of samples for sickle cell testing to the laboratory, punching and eluting of dried blood spots, testing by IEF by highly trained personnel, and entering of results into database. Based on the short TAT, this internal process can be deemed efficient. However, when looking at the parameter of the programme year, it does appear that longer TAT in this phase is associated with periods of increased sample volumes (Table 3). This may be the cause of reagent stock out or point to the need for more cross-training of CPHL personnel on the IEF procedure to support the sickle cell laboratory. As the programme expands toward the eventual goal of universal newborn screening, more human resources will be necessary to handle the increased number of samples.
Following sample testing, result reporting took an average of 6 days ( Figure 1). Altogether, the time that samples spent at the CPHL was prolonged. Delays with the dispatch of results could be sample retesting or stock out of printing supplies. It was also observed that the testing and dispatch months of December 2018 and January 2019 had distinctly longer TATs (Table 4). This may be linked to personnel leave due to the holidays, thus increasing or staggering support in the sickle cell laboratory during this time could help avoid lengthy delays.
In a previous study by Kiyaga et al. at the Uganda CPHL regarding TAT of Early Infant HIV Detection screening, the average sample delivery TAT was 12 days, sample processing was 2 days, and the overall average TAT from sample collection to reporting to the patient was 26 days. 14 As part of this study, short message service printers were piloted to allow test results to be transmitted immediately to healthcare facilities after testing, which further reduced the overall TAT to 14 days. 14 In the United States, the National Newborn Screening and Global Resource Center reported newborn bloodspot screening TAT to be within 10 to 14 days from sample collection, with considerable variation by state. 17 To overcome these differences in state programmes, the United States Department of Health and Human Services Advisory Committee on Heritable Disorders in Newborns and Children recently set out updated newborn screening timeliness goals specifying that tests should be completed within 7 days of birth. 18 In our study, we found that TAT for sickle cell screening was shorter in all inbound phases compared to what was previously reported for Uganda's Early Infant HIV Detection programme, and exhibited overall improved time efficiency for the entire CPHL sample continuum (median 16 days vs 26 days). The current programme is approaching a comparable timeframe to that of the long-standing United States newborn sickle cell screening programme and will likely continue to evolve and improve over time.
Information on outbound TAT, which includes time to family notification and matriculation to care, could not be examined in this study, because those data are currently not collected by the CPHL. However, with greater visibility and a push toward newborn screening in sub-Saharan African countries, steered by the new American Society of Hematology African Screening Consortium, 19 outbound TAT and individual patient follow-up data will be coveted information. We predict that TAT will greatly impact healthcare provider and patient utilisation of sickle cell screening, as well as overall satisfaction with the programme in Uganda. Most importantly, prolonged TAT very likely postpones lifesaving treatments for patients with the disease, if the diagnosis is not established and communicated promptly. Also, delayed result reporting can lead to affected infants being lost to follow-up and may require a replacement test. Improving TAT helps ensure that caretakers collect their results on time and minimises the need for inefficient and expensive repeat testing. Ultimately, making improvements in all TAT phases for sickle cell screening is essential to optimise infant health outcomes, cost-effectiveness, and screening satisfaction for healthcare providers and patients in Uganda.
A major strength of this analysis was the complete national representative long-term data from the CPHLs centralised database, which allowed us to investigate previously undocumented parameters that affect sickle cell testing TAT in Uganda. Limitations include the access to only inbound TAT data, which limits our ability to connect our findings to patient indicators such as notification and receipt of sickle cell results, followed by care and management of affected patients. Another limitation is the lack of recommendations for sickle cell testing TAT by the Uganda Ministry of Health, other regional programmes, or by international bodies such as the World Health Organization for a comparison of efficiency. However, our data provided the first detailed time analysis of sickle cell screening in sub-Saharan Africa, to enable monitoring and evaluation of future interventions and other programmes.
Analysis of the direct costs of a sickle cell screening programme in Uganda showed that the cost per test by IEF using dried blood spots was $4.46 USD. In Angola, the cost per infant screened in a newborn screening programme that used IEF was $15.36 USD, or $7.42 USD when considering just the inputs of laboratory personnel, reagents, consumables, and equipment as in our study. 20 The cost of the screening test for sickle cell disease performed by IEF in the United States was $2.29 USD and approximately $4.50 USD (reported as £3.51) in the United Kingdom. 21,22 The United States and United Kingdom have addressed the cost-effectiveness of newborn screening for sickle cell disease for universal and targeted programmes. Both nations have justified the greater cost incurred by screening every newborn for sickle cell disease, because it identifies all infants with the disease, prevents more deaths, and has demonstrated better outcomes for patients. 21,22,23 In Africa, modelling simulations have found universal newborn sickle cell screening to be extremely cost-effective, especially in countries with a high disease prevalence of 0.2% -0.5%. 24,25 Yet, no federal newborn sickle cell screening currently exists in any country in sub-Saharan Africa, despite the well-evidenced economic and humanitarian basis for such programmes.
In Uganda, the estimated costs per sickle cell disease case detected among the 10 regions varied from $278.07 USD in the East Central to nine times that in the South Western region ($2607.19 USD) ( Table 6). These results show that newborn screening in regions with low sickle cell disease prevalence would result in a higher cost per positive case detected compared with screening focused in high-burden areas. However, because Uganda is a country with a high overall disease prevalence of 0.9%, 26 universal newborn screening would still be the most economical and impactful public health strategy for sickle cell disease across the entire country.

Limitations
This analysis only considered costs and numbers of cases detected for partial cost-effectiveness analysis. More extensive studies with patient follow-up data will be vital to provide evidence that screening is effective in reducing sickle cell morbidity and mortality, the benefits and harms of screening, and the long-term cost-effectiveness of a newborn sickle cell screening programme in Uganda.

Conclusion
This study provides a contemporary and detailed description of the time and costs of sickle cell screening in Uganda. Analysis of the different phases of TAT highlights areas for improvement to reduce the number of samples with excessive delays, and to strengthen the overall integrated CPHL sample continuum. The lack of sickle cell testing guidelines limits our ability to compare these results to screening and care standards; however, this study documents that the CPHL centralised database can be used to accurately monitor and manage TAT for sickle cell screening and can identify factors affecting TAT at different phases to prompt targeted improvements. This study also shows that sickle cell testing by IEF is provided at under $5.00 USD, and the strategy of universal newborn screening is cost-effective to save and improve the lives of thousands of individuals with sickle cell disease in Uganda.