Menu
Menu
Menu
Menu
Background Aligned with global childhood tuberculosis (TB) road map, Ethiopia developed its own in 2015. The key strategies outlined in the Ethiopian roadmap are incorporating TB screening in Integrated Maternal, Neonatal and Child Illnesses (IMNCI) clinic for children under five years (U5) and intensifying contact investigations at TB clinic. However, these strategies have never been evaluated. Objective To evaluate the integration of tuberculosis (TB) screening and contact investigation into Integrated Maternal, Neonatal and Child Illnesses (IMNCI) and TB clinics in Addis Ababa, Ethiopia. Methods The study used mixed methods with stepped-wedge design where 30 randomly selected health care facilities were randomized into three groups of 10 during August 2016-Novem-ber 2017. The integration of TB screening into IMNCI clinic and contact investigation in TB clinic were introduced by a three-day childhood TB training for health providers. An in-depth interview was used to explore the challenges of the interventions and supplemented data on TB screening and contact investigation. Results Overall, 180896 children attended 30 IMNCI clinics and145444 (80.4%) were screened for TB. A total of 688 (0.4%) children had presumptive TB and 47(0.03%) had TB. During the pre-intervention period, 51873 of the 85278 children (60.8%) were screened for TB as compared to 93570 of the 95618 children (97.9%) in the intervention (p<0.001). This had resulted in 149 (0.30%) and 539 (0.6%) presumptive TB cases in pre-intervention and intervention periods (p<0.001), respectively. Also, nine TB cases (6.0%) in pre-intervention and 38 (7.1%) after intervention were identified (p = 0.72). In TB clinics, 559 under-five (U5) contacts were identified and 419 (80.1%) were screened. In all, 51(9.1%) presumed TB cases and 12 (2.1%) active TB cases were identified from the traced contacts. TB screening was done for 182 of the 275 traced contacts (66.2%) before intervention and for 237 of the 284 of the traced (83.5%) under intervention (p<0.001). Isoniazid prevention therapy (IPT) was initiated for 69 of 163 eligible contacts (42.3%) before intervention and for 159 of 194 eligible children (82.0%) under intervention (p<0.001). Over 95% of health providers indicated that the integration of TB screening into IMNCI and contact investigation in TB clinic is acceptable and practical. Gastric aspiration to collect sputum using nasogastric tube was reported to be difficult. Conclusions Integrating TB screening into IMNCI clinics and intensifying contact investigation in TB clinics is feasible improving TB screening, presumed TB cases, TB cases, contact screening and IPT coverage during the intervention period. Stool specimen could be non-invasive to address the challenge of sputum collection.
The study was carried out in Addis Ababa, the capital city of Ethiopia, with an estimated population of 3.9 million in 2017/2018 [17]. According to unpublished national TB report of 2018/2019, a total of 4,070 TB cases (116 /100, 000) were notified in Addis Ababa, of whom 191 (4.7%) were children less than 15 years of age [3]. Health centers are the first entry point to the formal health system, and they are the centerpiece of primary health care services in Ethiopia. One of their key roles is the provision of curative IMNCI services, such as treating children with diarrhea, pneumonia, and malnutrition. Cough triage, TB screening, diagnosis, and treatment are part of the TB service provided at the primary health care level with little focus on childhood TB. A mixed method study was adopted to evaluate the programmatic intervention outlined below. For the quantitative part, a stepped-wedge design was applied as this is an optimal design to evaluate phased in interventions [18]. In depth interview and field notes (qualitative data) supplemented the quantitative data collection. For the stepped wedge design, the units of comparison were groups of randomly assigned health facilities that moved over to the intervention together. Allocation was not concealed, but clinicians, and TB and IMNCI officers were blinded to the order of entry into the intervention till each group of health centers were enrolled to the interventions [19]. The study was undertaken over a period of 16 months, and facilities transitioned in four- month intervals from pre-intervention to intervention period (Fig 1). Fig 1 shows the four periods, each with the length of four months. The first period (August- November 2016) was the baseline period where all health facilities served as control. December 2016-March 2017 was phase I where the first group of 10 health facilities (phase I health facilities) were enrolled into intervention. During this phase the rest of 20 health facilities served as control in the control period. April-July 2017 was a period when the second group of 10 health facilities (phase II) were enrolled into intervention. Here, phase I &II health facilities served as intervention health facilities in the intervention period, whereas the remaining 10 health fasciitis (phase III facilities) served as a control in the control period. Finally, in August-November 2017 phase III facilities entered the intervention. Initially, all of the 100 health centers in Addis Ababa were listed with their annual patient load in the IMNCI clinic. Fifty health centers that reported more than 500 U5 children per year were included in the sampling frame. From these, 30 were randomly selected as the study health facilities. Subsequently, the 30 facilities were randomly assigned to three groups, each with 10 health centers, which started the intervention phase by phase. Health care workers from each group of 10 health centers were trained and made ready to start the intervention at four-month interval (Fig 1). Preparation for and initiation of the interventions were during the fourth month of the previous phase and the first month of the upcoming phase. The interventions consisted of the integration of TB screening into IMNCI clinics, and enhanced childhood TB case finding in the TB DOT unit (TB DOTS) clinics through contact investigation. To guide the intervention, desk reference materials in the form of pediatric TB job aids, updated IMNCI registers with a TB screening column and registers for contact investigation and IPT provision were developed and supplied. Sensitization on the implementation study was conducted for child-health and TB program officers and heads of the study health facilities. This was followed by a three-day basic childhood TB training for health care providers from the IMNCI and TB DOTS clinics to introduce the intervention. Nasogastric (NG) tubes were provided to each study health facility. Research physicians demonstrated and mentored Health care workers (HCWs) on how to perform the nasogastric aspiration (NGA) procedure. Onsite coaching of health providers followed classroom demonstration till the HCWs confidently carried out the procedure. The research team also monitored the performance of each study health facility and record keeping practice monthly. TB screening and diagnosis, and contact tracing were based on the existing national guidelines and algorithms [20]. A child was identified as a presumptive TB case if cough, fever, failure to gain weight or contact history was reported. As per national policy, all children with presumed TB cases received GeneXpert MTB/Rif (GXP) as a primary diagnostic test. If the GXP was negative, contact history with an infectious pulmonary TB (PTB) case, suggestive chest X-ray and the presence of clinical signs of TB were used as criteria to diagnose TB in a child. The presence of at least two of these made a clinical diagnosis of TB in children. Those children identified to have active TB cases were started on TB treatment at TB DOTS clinic. To enhance contact tracing, all U5 contacts of PTB index cases identified at the TB DOTS clinics underwent evaluation for TB as outlined above. As part of contract tracing, children with screen negative or non-presumed TB case, thus without suggestion of TB, were eligible for prophylaxis and offered IPT (S1 Fig). Baseline data collection started in all study sites to document the practice of TB screening over a four-month period before the first group of 10 health facilities entered to the intervention. This baseline assessment was carried out using key outcome indicators, such as the number of presumed TB cases identified, the number of NGA undertaken, the number of TB cases identified at IMNCI clinics, the number of eligible contacts traced and screened, and the number of eligible U5 contacts started on IPT. During the project implementation, the researchers provided a standardized set of tools to capture the childhood TB data. From the study health facilities, HCWs were recruited and trained as data collectors. At the end of every fourth month, the trained HCWs collected data from the records of the study health facilities. In addition, in-depth interviews (IDIs) were performed with 30 HCWs and 11 heads of study health facilities. Those on duty on the interview day were selected. The study coordinator conducted the IDIs until saturation, i.e. until no new information or views were obtained from subsequent interviews. The IDIs were conducted twice during the project period, at the beginning (August-November 2016) and at the end (August-November 2017). IDIs were guided by a developed interview guide inquiring about the advantages and barriers of integration of TB screening into IMNCI and contact investigation to TB clinics. The study coordinators kept field notes of the observation during the data collection. The IDI and field notes were captured in English, though the interview was conducted in the local language, Amharic, for convenience. At the end of the study, assessment on the feasibility of integration of TB screening into IMNCI and intensified contact investigation in TB clinics was done through a structured questionnaire among the selected set of 190 parents/caretakers, 80 health care providers and 30 heads of the study health facility. The data collectors interviewed the parents and caretakers when they showed up at IMNCI and TB clinics with their children, and the HCWs while on duty. Data collectors attended a one day training on how to extract data using the checklist. The researchers checked data consistency everymonth during mentoring visits and every quarter during supportive supervision to the study sites. The supervision was undertaken jointly with the staff from KNCV Tuberculosis Foundaiton, the national TB program (NTP) and the Maternal and Child Health Unit of FMOH and Addis Ababa regional Health bureau (AARHB). Double data entry was done for all extracted data. The quality control of this process was performed by the study coordinator. Before data analysis was commenced, data cleaning and validation of the entered data was done by checking for data completeness, presence of outliers and inconsistencies. The summary of findings from IDIs, and field notes of observation at the health facilities were re-checked with IDI participants. Data entry for the quantitative part was done using EPI info (Version 7.2.2.16; Atlanta, Georgia: Centers for Disease Control and Prevention; 2018). Data analysis was done using STATA (Version 13; College Station, Texas: Stata Corp; 2013). The main indicators such as the proportion of children screened for TB from patients at IMNCI clinics, the proportion of presumed TB cases and TB cases identified from the screened, the proportion of presumptive TB cases with NGA procedure done and the coverage of IPT (number of children started on IPT/number of children eligible for IPT) were compared using the two sample proportion test over the 16 months control and 16 months intervention period. The mean number of children screened, identified with presumptive TB and detected to have active TB during the control (pre-intervention) period and the intervention period per the study site was also compared using two samples mean comparison test or t–test. The 95% CI and p-value (less than 0.05) were used to assess statistical significance. The manual theme-based word data analysis was done based on the IDIs and field notes. Subsequently, the quantitative and qualitative findings were triangulated. Eventually, the feasibility (acceptability and practicality) of the integration of childhood TB screening to U5 clinics was described quantitively using frequency and proportion. Ethical clearance was obtained from the Ethical Review Board of Addis Ababa City Health Bureau (AACHB). Support letters from AACHB to the sub-cities health bureaus, and then from sub-cities to the study health facilities were written to obtain permission to conduct the study and gain access to the childhood TB data of each the study participants. After the heads of the study facilities were briefed on the aim and methodology of the study, they provided a permission to include their health facility in the study. As this was a routine strategy being evaluated, no written consent was obtained for each child to take part in the study. The ethics committee waived the requirement for informed consent, and the data of each child was anonymized as well. Children diagnosed with TB were provided with TB treatment, and children eligible for preventive therapy were provided with IPT per the national guidelines.
