LABORATORY SERVICES CHANGE PACKAGE

STRENGTHENING UGANDA S SYSTEMS FOR TREATING AIDS NATIONALLY LABORATORY SERVICES CHANGE PACKAGE Synthesis of the most robust and effective QI interventions to improve Laboratory Services in SUSTAIN supported hospitals in Uganda AUGUST 2017 The SUSTAIN project is made possible by the generous support of the American people through the United States Agency for International Development (USAID) under Cooperative Agreement number 617-A-10-00007-00. The project team includes prime recipient University Research Co., LLC (URC) and sub-recipients; The AIDS Support Organization (TASO), Integrated Community Based Initiatives (ICOBI), Uganda Catholic Medical Bureau (UCMB), Uganda Protestant Medical Bureau (UPMB), Uganda Muslim Medical Bureau (UMMB), Child Chance International (CCI Uganda), AIDS Information Centre (AIC) and ACLAIM Africa.

Quality Improvement Change Packages Series The purpose of the quality improvement change packages is to provide a synthesis of the most robust and effective QI interventions for effective HIV programming. The quality improvement change packages series thematic areas include: prevention of mother to child transmission, laboratory, monitoring and evaluation, adolescent friendly health services, voluntary medical male circumcision, nutrition, HIV care and treatment, supply chain, Tuberculosis, and quality improvement. USAID/SUSTAIN acknowledges the work of the project staff, technical officers at MoH, and counterparts at supported facilities who have been instrumental to the project s many successes through implementation of the quality improvement interventions. The publication and production of these change packages, as well as the work of the SUSTAIN project, was made possible by the generous support of the American people through USAID. The SUSTAIN project is led by University Research Co., LLC and works in partnership with: The AIDS Support Organization (TASO), Integrated Community Based Initiatives (ICOBI), Uganda Catholic Medical Bureau (UCMB), Uganda Protestant Medical Bureau (UPMB), Uganda Muslim Medical Bureau (UMMB), Child Chance International (CCI Uganda), AIDS Information Centre (AIC) and ACLAIM Africa, under Cooperative Agreement No. 617-A-10-00007-00. The views and opinions expressed here do not necessarily state or reflect those of USAID or the United States government. Laboratory Services Change Package i

Contents Quality Improvement Change Packages Series... i Acronyms... iv Introduction...1 Harvest Meeting...2 Change package for improving the quality of laboratory services at regional referral and general hospitals in Uganda...5 Intended Use...5 Improvement Aim 1: To reduce turnaround time for TB smear microscopy and CD4 results...6 Improvement Aim 2: To reduce the proportion of samples that are rejected by the laboratories based on quality... 11 Improvement Aim 3: To improve the proportion of supported laboratories with a Stepwise Laboratory Quality Improvement Processes Towards Accreditation (SLIPTA) star rating of 3 or more through effective implementation of the laboratory quality management system...15 Key Challenges...18 Moving Forward...18 Appendix: List of contributors during the harvest meeting...19 Laboratory Services Change Package iii

List of Acronyms ART Antiretroviral Therapy CD4 Cluster of differentiation 4 CME Continuing Medical Education CPHL Central Public Health Laboratories HC Health Center MoH Ministry of Health NC Non-Conformities PDSA Plan-Do-Study-Act QI Quality Improvement RRH Regional Referral Hospital SLIPTA Stepwise Laboratory Quality Improvement Processes Towards Accreditation SOPs Standard Operating Procedures SUSTAIN USAID Strengthening Uganda s Systems for Treating AIDS Nationally TB Tuberculosis USAID United States Agency for International Development iv Laboratory Services Change Package

Introduction Since 2010, the United States Agency for International Development (USAID) has been working with Uganda s Ministry of Health (MoH) to improve HIV and AIDS service delivery at select health facilities through the Strengthening Uganda s Systems for Treating AIDS Nationally (SUSTAIN) project. Over the last seven years, the SUSTAIN project has aimed to: n Support the MoH to scale up Prevention of Motherto-Child Transmission of HIV (PMTCT) and Voluntary Medical Male Circumcision (VMMC) as HIV infection prevention interventions within selected public regional referral hospitals (RRHs) and general hospitals Table 1: List of Intervention Health Facilities Name of Facility Level of Facility Abim General hospital Arua Regional referral hospital Fort Portal Regional referral hospital Gulu Regional referral hospital Hoima Regional referral hospital Jinja Regional referral hospital Kaabong General hospital Kabale Regional referral hospital n Ensure provision of HIV care and treatment, laboratory and tuberculosis (TB)/HIV services within selected public RRHs, general hospitals and health center (HC) IVs n Enhance the quality of PMTCT, VMMC, HIV care and treatment, laboratory and TB/HIV services within selected RRHs, general hospitals and HC IVs, and n Increase stewardship by the MoH to provide sustainable quality HIV prevention, care and treatment, laboratory and TB/HIV services at project-supported healthcare facilities. Kawolo Kotido Lira Masaka Matany Mbale Moroto Mubende Soroti Tokora General hospital Health Centre IV Regional referral hospital Regional referral hospital General hospital Regional referral hospital Regional referral hospital Regional referral hospital Regional referral hospital Health Centre IV The SUSTAIN project has supported the implementation of laboratory systems strengthening activities at 18 public health facilities in Uganda. Support to the laboratories included: infrastructure improvement through renovations and remodeling of laboratory spaces, equipment procurement and installation, human resource capacity development, coordination and logistical support to ensure uninterrupted laboratory testing services, operational and coordination support for transporting samples and results using the hub system, quality improvement interventions to improve the laboratories functionality, and support for implementation of the MoH laboratory accreditation program. Of the 18 laboratories supported, 12 were at regional referral hospitals, four were at general hospitals and two were at health center IVs, as indicated in Table 1. Through ongoing onsite mentorship and coaching sessions, and application of quality improvement approaches, laboratory teams at these facilities have been supported to improve turn-around time for TB sputum microscopy and cluster of differentiation 4 (CD4) test, reduce the proportion of samples that are rejected, and to attain national minimum requirement towards accreditation. The project s approach to continuous quality improvement (QI) was guided by the Model for Improvement that uses the Plan-Do-Study-Act (PDSA) cycles. Laboratory systems strengthening advisors from SUSTAIN supported the formation of multi-disciplinary improvement teams at all Laboratory Services Change Package 1

supported laboratories, through which QI interventions were implemented. This improvement collaborative approach, where teams work to identify and address a myriad of challenges affecting the content and processes of care, is consistent with the Ministry of Health s Quality Improvement Framework and Strategic Plan. Monthly, the laboratory teams received coaching and onsite supervision and mentorship on how to identify gaps in laboratory services, how to prioritize areas for improvement, and how to develop, test and eventually implement change ideas that could lead to improvements. This change package represents a synthesis of the most robust and effective changes in improving the quality of laboratory services at the 18 supported hospitals in Uganda. The change ideas recommended here are substantiated by data collected during project implementation, which shows significant improvements in: a) Proportion of test results released within acceptable turnaround time for TB smear microscopy and CD4 results b) Proportion of rejected samples at supported laboratories, and c) Proportion of supported laboratories with a SPLITA rating of 3 stars or above. Harvest Meeting After six years of project implementation, laboratory staff, clinicians and hospital administrators from the 18 supported hospitals gathered for a harvest meeting in August 2016 to reflect on their results, discuss both successful and unsuccessful change ideas, and share evidence on which pathways resulted in positive results. Guided by their experience in using QI to improve the quality of laboratory services, they agreed on a set of best practices that could guide other laboratory teams to improve the quality of their services and achieve faster progress towards effective implementation of quality management systems. Divided into small groups, teams discussed the change ideas they had tested, the steps they followed in introducing and testing these changes, and the results they had observed that could be attributed to the tested changes. The plenary sessions were also used as a platform to evaluate and score the different change ideas based on their relative importance, level of simplicity and scalability. All the parameters (relative importance, simplicity and scalability) were scored 1-5 by the participants. A score of 1 (one) for relative importance, for example, means that that change idea was not important, yet a score of Figure 1: Guide to interpreting the rating of change ideas 1 2 3 4 5 n Change was not n Only important n Change can n Change was n Change was important in a few aspects be important important very important n Change idea was n Change is often n Change can n Change is n Implementation too complex complex be complex sometimes is always smple n Change is n Scalable with n Scaling simple n Change is easily difficult to scale significant requires effort n Scalable with scalable challenges limited effort 2 Laboratory Services Change Package

5 for scalability means the change is easily scalable. Figure 1 explains the scoring, and how each value should be interpreted. The average scores are presented in Tables 2 4. Tables 5 7 provide a comprehensive list and description of all the change ideas tested, with notes on the specific steps taken to implement the change, the observed results and the number of facilities (scale) that implemented the specific changes. Table 2 summarizes how the participants in the harvest meeting evaluated the changes for each of the three improvement aims. Table 2: Rating of change ideas implemented to reduce turnaround time for TB smear microscopy and CD4 results SN Change idea Number of facilities testing this change Relative importance Rating Criteria Simplicity (not difficult or complex) Scalable Total score Average overall score 1. Lab team stopped batching samples, and instead ran samples as and when they received them 2. Additional staff were hired by the SUSTAIN project and seconded to support facilities affected by understaffing 3. Reorganization of duty schedule and duty roster (staff replacement, task allocation, regular rotations, and training volunteer staff to always support the lab team 4. Sharing the duty roster with hospital leadership, for them to enforce and ensure quality 5. Keeping records of time-in and time-out for all samples using log books, counter books or registers. 6. Continuing Medical Education (CME) and sensitization to staff on appropriate packaging samples, easing both identification and analysis 7. Identifying Quality Officer and tasking them the responsibility of monitoring turn-around-time 8. Deliberations with clinicians and ward staff to agree on the most appropriate time to collect and send samples to labs as well as expected timelines for picking results 9. Use of hub riders to facilitate delivery of samples and results 10. Prioritizing sending back results to requestor, and having documentation in the lab register done later. 11. Use of trained volunteers and other lower cadre staff to perform less sophisticated roles in the lab, to free up time for senior staff to run tests 17 4.6 3.7 3.9 17.2 4.3 18 4.8 3.9 4.2 17.9 4.5 18 4.9 4.3 4.3 18.6 4.6 18 4.7 3.4 3.8 17.0 4.3 16 4.8 4.1 4.1 17.9 4.5 18 4.8 3.3 3.4 16.6 4.2 18 4.8 4.2 4.1 18.1 4.5 18 4.7 3.4 3.7 16.7 4.2 17 4.5 3.5 3.4 16.5 4.1 13 3.8 3.5 2.6 14.9 3.7 16 3.8 3.0 2.6 14.4 3.6 continued Laboratory Services Change Package 3

Table 2: Rating of change ideas implemented to reduce turnaround time for TB smear microscopy and CD4 results, continued SN Change idea 12. Acquisition of additional safety equipment to ensure all lab staff can run TB tests without fear of getting infected 13. Appointment of TB focal person in the lab charged with quality of samples and timeliness of results dissemination 14. Training both SUSTAIN and non-sustain staff to operate the CD4 machine, so that some can fill in for others 15. On a periodic basis, gathering different lab teams together to share knowledge on how the quality of collected samples affects processing times or even sample rejections Number of facilities testing this change Relative importance Rating Criteria Simplicity (not difficult or complex) Scalable Total score Average overall score 17 4.8 2.7 3.2 15.8 3.9 18 4.8 3.7 3.6 17.0 4.3 18 4.8 3.6 3.7 17.1 4.3 18 4.6 3.3 3.5 16.3 4.1 16. Keeping track of when equipment repairs are done. 18 4.9 4.1 4.0 18.0 4.5 Table 3: Rating of change ideas aimed at reducing the proportion of laboratory samples that were rejected at the SUSTAIN supported hospitals SN Change idea 1. CMEs and specific trainings to new student interns and staff on sample collection, use of containers, transportation of samples, expected turnaround time and rejection criteria. 2. Use of sample rejection logs - to document which samples have been rejected, reasons for rejecting and sample origin 3. Tasked the reception staff to check whether all brought-in samples meet the eligibility criterion before accepting them 4. Disseminated a circular on sample rejection criteria to all sample sources like wards, clinics and lower health facilities 5. Use of job aids or display charts at sample collection points for reference by lab staff Number of facilities testing this change Relative importance Rating Criteria Simplicity (not difficult or complex) Scalable Total score Average overall score 18 5.0 4.0 3.9 17.9 4.5 18 4.9 4.6 4.4 18.9 4.7 18 4.9 3.7 3.8 17.2 4.3 18 4.8 3.6 3.7 16.5 4.1 18 4.7 4.1 4.0 16.6 4.2 continued 4 Laboratory Services Change Package

Table 3: Rating of change ideas aimed at reducing the proportion of laboratory samples that were rejected at the SUSTAIN supported hospitals, continued SN Change idea Number of facilities testing this change Relative importance Rating Criteria Simplicity (not difficult or complex) Scalable Total score Average overall score 6. Appointed quality officers to do supportive supervision at sample collection points 7. Trained hub riders on safe packaging of samples in transit 8. Follow up of rejected samples by highlighting reasons for rejection to specific clinicians 9. Sharing sample rejection statistics with respective origin sites to inform their improvements 10. Task re-allocation by experienced and trained lab personnel to perform phlebotomy to ensure quality of collected samples 11. Designing, distributing and training users on standard operating procedures (SOPs) for sample collection 12. Assigning lab staff to different wards during lab rounds to ensure clinicians follow the SOPs while collecting high quality samples 13. Designed and distributed standard lab request forms to all origin sites 17 4.6 4.1 3.7 16.6 4.2 18 4.8 3.4 3.7 16.9 4.2 18 4.6 3.4 3.3 15.4 3.8 18 4.7 3.7 3.3 15.2 3.8 18 4.8 4.3 3.7 17.4 4.3 18 4.7 3.9 3.8 15.8 4.0 18 4.6 2.9 3.1 12.7 3.2 18 4.8 4.2 3.9 17.9 4.5 Change package for improving the quality of laboratory services at regional referral and general hospitals in Uganda Intended Use Laboratory staff, Antiretroviral Therapy (ART) clinic staff, TB staff and hospital administrators are the primary intended users of package. Others, such as NGOs involved in improving the quality of laboratory services, district health officers supervising health facilities, and Ministry of Health officials supporting the laboratory accreditation strategy will find the evidence-based high impact changes described in the following pages useful. It should be noted that laboratory teams should not necessarily copy these change ideas, rather, they should adapt them to suit their circumstances and context challenges. The next section of package provides a detailed description of what changes led to improvement, and how such improvement was derived. It is structured into three sub-sections, corresponding with the three improvement aims that the SUSTAIN project set out to achieve in relation to improving the quality of laboratory services in Uganda. Each sub-section outlines the QI Laboratory Services Change Package 5

change concept applied, the problem being addressed, the change ideas tested, steps followed in introducing each change idea and the evidence that it led to improvement. Figure 2: Percentage of TB sputum and CD4 results that were released within acceptable turnaround time. 100 92 95 Improvement Aim 1: To reduce turnaround time for TB smear microscopy and CD4 results Figure 2 shows the comparative improvements in the turn-around times for TB sputum and CD4 between 2015 and 2016. In just one year, the percentage of TB sputum samples that were analyzed within 3 hours of receipt improved by 4%. Comparatively, the turn-around time of CD4 samples improved from 92% in 2015 to 95% in 2016. Percent 80 60 40 20 0 70 74 FY 2015 FY 2016 TB Sputum CD4 Table 4: Specific changes introduced to reduce turnaround time for TB sputum and CD4 results Change concept Specific problem/ cause of long turnaround times Change ideas tested Steps in introducing the change ideas Evidence that the changes led to improvement Scale of implementation Smooth work flow Delays in processing samples as staff would prefer running large batches instead of small numbers Lab team stopped batching samples, and instead ran samples as when they received them Involved lab staff, clinicians, hub riders, TB staff and administrators Idea was generated during monthly QI meetings, in response to an analysis of root causes of delays Instead of waiting for a large quantity, samples would be analyzed as they came Results would then be sent back immediately In Lira and Mubende regional referral hospitals, over 90% of their TB sputum samples have results released within 3 hours 17 health Invest more resources in improvement Limited numbers of qualified lab staff, and the heavy work load would affect the speed with which samples are analyzed Additional lab staff were seconded by the SUSTAIN project to address understaffing SUSTAIN conducted a needs assessment and identified cadres of lab staff that were missing in each facility Worked with hospital administration to recruit staff Continuously provided ongoing onsite mentorship and coaching sessions In Kotido hospital, 96% of TB sputum samples are processed within 3 hours and 100% for Mbale hospital. Other non-sustain staff were also trained with the additional staff continued 6 Laboratory Services Change Package

Table 4: Specific changes introduced to reduce turnaround time for TB sputum and CD4 results, continued Change concept Specific problem/ cause of long turnaround times Change ideas tested Steps in introducing the change ideas Evidence that the changes led to improvement Scale of implementation Improve work flow Lack of clarity in duty allocation and how responsibilities were shared amongst lab staff Reorganization of duty schedule and duty roster (staff replacement, task allocation, regular rotations, volunteers to cover the TB section permanently) Identification and listing of all staff assigned to work in lab, in consultation with hospital administrator Assigning roles based on expertise and level of experience Duty roster drawn up with clearly defined allocations, and displayed Volunteers also assigned roles Fort Portal, saw 85% of their TB sputum samples meeting the targeted turnaround time of 3 hours Give supervisors access to information Negligence, absenteeism and poor attitude of staff towards working on TB samples Sharing the duty roster with hospital leadership, for them to enforce and ensure quality Hospital administrators called and chaired meetings where performance of lab staff was discussed Duty roster was shared with hospital administrators for close supervision of lab staff which promoted enforcement of the duty roster and encouraged more positive attitudes towards lab work Client satisfaction with lab services improved in the hospitals that tried, as clients were spending less time waiting for lab results Create a formal process (Standardization) Incomplete documentation (in some cases total lack of documentation) regarding when samples are brought in and when results are released. Using log books, counter books or registers for keeping records on time-in and time-out for all samples coming into the labs Idea was suggested during onsite coaching and mentorship sessions Gaps in documentation, especially documentation of received samples and time of receipt, plus documentation of when results were released Requested for registers, log books and counter books from hospital administrators to aid in lab documentation at point of sample submission Availed a wall clock in labs, to accurately capture both time-in and time-out Hospitals that had previously not documented their turnaround time started doing so; 82% of TB samples in Kabale are released within recommended turnaround time of 3 hours 16 health Individuals to assist with documentation identified continued Laboratory Services Change Package 7

Table 4: Specific changes introduced to reduce turnaround time for TB sputum and CD4 results, continued Change concept Specific problem/ cause of long turnaround times Change ideas tested Steps in introducing the change ideas Evidence that the changes led to improvement Scale of implementation Provide training Poor packaging of samples and/ or lack of labelling would contribute to delays in sample analysis CME and sensitization to staff on appropriate packaging of samples, easing both identification and analysis Identified repetitively problematic areas relating to sample collection, packaging and labelling Scheduled CME to cover these specific areas, delivered by experienced and knowledgeable lab staff Mobilized all lab staff to attend these CME, plus clinicians, nurses and hub riders Observed improvements in packaging and labelling of samples Use a coordinator No attention was being put on turnaround time for samples brought into the lab Identifying Quality Officers and tasking them with the responsibility of monitoring turnaround-time for TB and CD4 samples Head of the lab department identified one of the technicians to focused on quality assurance Technician had to ensure that all samples are recorded, and the time they were submitted and the time results were dispersed from the lab departed the lab The Quality Officer gave updates on performance during monthly improvement meetings Based on objective data, labs could quantify and pin-point the challenges they were facing and developed ideas for addressing them Streamline the order of steps to be followed in the process Unplanned, uncoordinated and unexpected deliveries of samples to the lab, any time of day and on any day. Agreement on the most appropriate time to collect and send samples to labs as well as expected timelines for picking results Held discussions with staff who send samples (from medical wards, TB unit, HIV clinic and lower facilities), and agreed that there was need for order and coordination when sending samples Agreed on collecting most samples very early in the morning, and have them delivered to the lab After running the tests, the lab would then disburse the results back to owners by noon TB and CD4 samples were delivered at an agreed upon time, that facilitated planning and improved turnaround time Move sequential steps in the process closer together Failure to get samples from the lower health facilities to the lab on time, further prolonging the turnaround time Use of hub riders to facilitate delivery of samples and results For each health facility in the region, a hub rider was assigned The facility team agreed with hub rider the days when samples should be collected On the specified days, samples would be collected from clients and be readied for the hub rider s pick up Sharp increments in the number of TB and CD4 samples picked from lower facilities were observed, with results delivered on time 17 health continued 8 Laboratory Services Change Package

Table 4: Specific changes introduced to reduce turnaround time for TB sputum and CD4 results, continued Change concept Specific problem/ cause of long turnaround times Change ideas tested Steps in introducing the change ideas Evidence that the changes led to improvement Scale of implementation Re-arrange the order of steps in the process Lab staff had to document the CD4 samples they had run before results are disbursed, this would delay the disbursement of results and increase TAT Prioritizing sending back CD4 results to requestor, and having documentation in the lab register done later. It was decided that soon after the CD4 tests have been run, results should be dispersed to the respective owners The necessary documentation would be done later, as there was little need to have the documentation done immediately The Quality Officer still followed up to ensure the necessary documentation has been done As staff prioritized disbursement of results to documentation, departments (and patients) could receive their TB and CD4 results earlier than previously 13 health Make use of existing alliances and relationships Lab staff would spend time performing lesssophisticated tasks like documentation instead of running tests Use of volunteers to perform less sophisticated roles in the lab, to free up time for senior staff to run tests Idea of utilizing volunteers was to perform less sophisticated tasks was discussed with the head of the lab List of available volunteers was compiled, with their skill levels well-articulated Tasks were then assigned to them based on their availability, skill levels and level of sophistication of the task The volunteers were still closely supervised to ensure they didn t make any errors As less sophisticated tasks were re-assigned to junior staff members, senior lab staff freed up time to work on the submitted samples and reduced turnaround-time 16 health Reduce demotivating aspects of the work Hesitance by some staff to work on TB samples for fear of being infected with TB Acquisition of additional safety skills and equipment to ensure all lab staff can run TB tests without fear of getting infected The issue of lack of adequate safety equipment was raised to the hospital administrators, and an average number samples tested in a week declared Resources were mobilized from partners (including SUSTAIN) to acquire additional safety equipment to be used in the lab The equipment was eventually acquired and staff trained on how to use it and other safety procedures All members of the lab staff could process TB samples in the 17 hospitals that tested 17 health Use a coordinator Mix-ups in criteria for accepting incoming TB sputum samples and CD4 samples, as no one was taking overall responsibility Appointment of TB focal person in the lab charged with quality of samples and timeliness of results dissemination Identified a focal person in the lab to focus on TB sputum samples The focal person would ensure all TB samples were properly labelled, registered, signed-in, and analyzed with the acceptable time and ensure that the results were sent back to the sample source in a timely fashion The TB focal person ensured TB samples were documented and processed within the acceptable turn-aroundtime in all the 18 hospitals that tested continued Laboratory Services Change Package 9

Table 4: Specific changes introduced to reduce turnaround time for TB sputum and CD4 results, continued Change concept Specific problem/ cause of long turnaround times Change ideas tested Steps in introducing the change ideas Evidence that the changes led to improvement Scale of implementation Capacity building through crosstrainings CD4 testing done only by SUSTAIN seconded staff Training both SUSTAIN and non-sustain staff to operate the CD4 machine Head of the lab department identified staff that had not been trained in CD4 testing SUSTAIN organized training workshops for them on how to manage CD4 samples Even onsite, staff were paired with colleagues in the lab to work with and be trained through on-job mentorship and supervision With increased number of staff capable in conducting CD4 tests, the turnaround time was greatly reduced in all the hospitals that tested this change Give people access to information Limited opportunities for transfer of learning within members of the lab team and between different facilities in the same regional catchment area Holding periodical hub review meetings to share knowledge on how sample quality affects processing times or even sample rejections Teams would identify topics of mutual interest, for example how sample quality affects processing times and criteria for sample rejection Periodical learning sessions were organized between facilities within the region Teams would synthesize learning emanating from their QI interventions, in preparation for these sharing opportunities Teams would develop action plans following these learning sessions, stating how they plan to apply the information they have learned Following trainings and capacity building efforts, lab staff could freely reach out to their colleagues for assistance or reach out to the referring facility and request for clarifications on submitted samples Optimize maintenance Mechanical breakdowns of lab equipment and inconsistent servicing Keeping track of when equipment repairs are done and when periodic maintenance is scheduled Introducing a repair log for documenting when repairs have been done, and which parts had broken down Setting up reminders for when maintenance is due, and keeping contacts of engineers When lab equipment was fully functional, all the hospitals could complete the analysis of samples and disbursement of results within the recommended turn-around times 10 Laboratory Services Change Package

Improvement Aim 2: To reduce the proportion of samples that are rejected by the laboratories based on quality Figure 3 illustrates that the proportion of samples that are rejected is very small with only 5% in 2016 across all SUSTAIN supported hospitals. This is testament to the high level of quality with which laboratory samples are collected, packaged, transported and stored. Table 5 provides more details on the specific changes that the different hospitals tested to reduce to proportion of their laboratory samples that were unacceptable. Figure 3: Percentage of laboratory samples that are of acceptable quality at SUSTAIN supported hospitals Percent 100 80 60 40 20 92 95 0 FY 2015 FY 2016 Table 5: Specific changes introduced to reducing the proportion of samples that are rejected Change concept Specific problem being addressed Change ideas tested Steps in introducing the change ideas Evidence that the changes led to improvement Scale of implementation Provide training New staff and student interns would often collect insufficient sample amounts and/or wrongly label them CMEs and specific trainings to new student interns and staff on sample collection, use of containers, transportation of samples, expected turnaround time and rejection criteria. Hospital administrators would organize trainings for student interns and new staff in the lab Common topics discussed were sample collection, use of containers, sample transportation, other lab services, acceptable turnaround time and sample rejection criteria Topics would be assigned to older staff, who would cover them during staff meetings or departmental monthly QI meetings In all the 18 hospitals that tested, new student interns used the correct containers were collecting samples Standardize (create a formal process) Lack of a mechanism for tracking the fate of rejected samples Use of sample rejection logs. Introduced sample rejection logs for keeping details of rejected samples It documents which samples have been rejected, the reasons for rejection and source of that sample Hospitals that tested could gather information on why samples were rejected Send feedback to different units with data on sample rejection Mismatch between the request forms used and the sample its submitted with Tasked the reception staff to check whether the submitted samples meet the eligibility criterion before accepting them Printed out copies of criteria for acceptable samples, and shared it with reception staff On receipt of samples, reception staff checks the criteria to determine if samples meet all items If yes, sample is registered in sample log. If not, sample is rejected. Proportion of samples rejected greatly reduced as correct and appropriate requestforms were used continued Laboratory Services Change Package 11

Table 5: Specific changes introduced to reducing the proportion of samples that are rejected, continued Change concept Specific problem being addressed Change ideas tested Steps in introducing the change ideas Evidence that the changes led to improvement Scale of implementation Standardize (create a formal process) Lack of detailed knowledge on the qualities of accepted samples Distributed a circular on sample rejection criteria to all sample sources like wards, clinics and lower health facilities Laboratories developed a circular on the minimum standards to be met by all samples if they are to be accepted Shared this circular with different units within the hospital and with lower health facilities Referenced this circular during periodical supervision sessions on lab services Rejection rates reduced in hospitals that had sample rejection criteria displayed in their facilities Use job aids as affordances Even after attending trainings, some health workers wouldn t follow set guidelines Use of job aids or display charts at sample collection points with critical values and protocols from clinician handbook for easy reference Standard job aids were obtained from MOH by SUSTAIN, and delivered to laboratories at supported health facilities Some job aids addressed general procedure in lab functionality, but others were specific to individual processes for sample analysis In each case, they were positioned/kept in places within the lab from where they could provide insight into how a task should be done or how something should be used Facilities with job aids displayed recorded fewer sample rejections Use coordinators Adherence to appropriate sample collection procedures was poor Appointed quality officers to do support supervision at sample collection points QI focal persons in the lab identified key persons to supervise adherence to specific protocols and guidelines during sample collection Focus was put on containers for sample collection, packaging materials and transportation conditions Quality officers conducted random checks to ensure each sample met the minimum quality standards 17 health The quality officer also availed guidelines, job aids and sample containers to health facilities where they were lacking Provide training Hub riders would put transport samples in boxed containers often with high temperatures Trained hub riders on safe packaging of samples-intransit Periodically, all hub riders were gathered for training on how specific samples should be packaged and transported between lower-level health facilities and the laboratories at RRHs Trainings were conducted by lab technicians from hospitals, the districts or MOH Hub riders checked and ascertained that the samples to be transported meant the minimum quality standards Focus was often put on packaging materials, transportation temperatures and the targeted turn-around-time between sample collection and analysis In addition to trainings, hub riders were supplied with materials to use in both packaging and transporting samples continued 12 Laboratory Services Change Package

Table 5: Specific changes introduced to reducing the proportion of samples that are rejected, continued Change concept Specific problem being addressed Change ideas tested Steps in introducing the change ideas Evidence that the changes led to improvement Scale of implementation Give health workers access to information Lack of specific feedback to lower facilities as to why their samples were rejected Follow up of rejected samples by highlighting reasons for rejection to specific clinicians Lab staff kept detailed records of each sample that was rejected, including where it came from, date and time it was received, and reasons for rejection Periodically, they would share with the affected facilities specific details as why some of their samples were not processed Health workers, aware of reasons for prior rejections, did not repeat them in the hospitals that tested During these feedback sessions, and in subsequent coaching sessions, the commonest reasons for rejecting lab samples would be mentioned and addressed Sharing sample rejection statistics with respective origin sites to inform their improvements Lab teams kept records of the proportion of the submitted samples that were rejected, and the reason for rejection This data was shared with the lab teams of affected facilities, to help them quantify the severity of the problem QI teams at these facilities used the data to initiate improvement projects, and eventually relied on it to gauge progress Targeted hospitals saw the number of their rejected samples reduce Optimize key staff Some health workers would make mistakes while drawing blood, resulting in poor quality samples Task reallocation by experienced and trained lab personnel to perform phlebotomy to ensure quality of collected samples Lab in-charges directed that all blood samples were to be drawn by trained and/or experienced phlebotomists Less experienced members of the lab staff received peer-to-peer training and attended workshops on phlebotomy By working under observation of their more experienced peers, even the less experienced health workers developed competencies in phlebotomy Hospitals that used experienced and trained staff had fewer rejections than those that didn t Poor quality of samples collected during ward rounds, since clinicians would focus more on clinical care than on the quality of samples sent to the lab. Assigning lab staff to different wards during clinical rounds to ensure clinicians and nurses follow the SOPs while collecting samples Lab staff were assigned to join clinical teams during ward rounds, and provide coaching on the procedures for collecting samples as outlined in the SOPs In wards/departments where the SOPs were not found, they were later provided by the lab staff In addition, job aids were also distributed to clinical staff to further emphasize the importance of high quality samples Hospitals that had lab staff collect samples had fewer rejections compared to those that didn t continued Laboratory Services Change Package 13

Improvement Aim 3: To improve the proportion of supported laboratories with a Stepwise Laboratory Quality Improvement Processes Towards Accreditation (SLIPTA) star rating of 3 or more through effective implementation of the laboratory quality management system Figure 4: Proportion of project supported laboratories with a SLIPTA star rating of at least 3 Percent 100 80 60 40 39 Figure 4 illustrates steady improvements in the proportion of SUSTAIN supported hospitals whose laboratories had 20 14 9 a SLIPTA star rating of 3 and above. Even though baseline performance was only 14% in 2014, this improved to 0 FY 2015 FY 2015 FY 2016 39% in 2016. Table 6 provides specific details of what laboratories did to improve their ratings. Table 6: Specific change ideas to improve SLIPTA star ratings SLIPTA Checklist sections Baseline performance Specific problem being addressed Change ideas Improved performance 1. Documents and records Abim, 0%; Lira, 8%; and Mubende, 8%. Laboratories lacked legal identity LQMS and SOPs were not aligned to the ISO 15189 standard Liaised with district lab focal persons and regional Allied Health Professional Council supervisors for registration guidelines and implemented them. Aligned all documents to the ISO15189 standard CMEs, trainings and mentorships by accredited laboratory personnel, Central Public Health Laboratories (CPHL) and East African Public Health Laboratories Network (EAPHLN) on Laboratory Quality Management System (LQMS), SLIPTA checklist and ISO 15189 Abim, 54%; Lira, 68%; and Mubende 46% Mentorship by SUTAIN on making SOPs and placed all relevant documents within accessible places with quality officer charged with ensuring their safety 2. Management reviews Mubende, 29%; Fort portal, 18%; and Masaka, 53%. No Routine review of quality and technical records No management review meetings conducted Institution of LQMS committee with clear roles Clearly scheduled management review meetings on lab calendar and defined the agenda Sensitized staff on importance of management review meetings Mubende, 50%; F/portal,43%; and Masaka, 71%. Engaged hospital management to monitor and ensure reviews are done on a regular basis continued Laboratory Services Change Package 15

Table 6: Specific change ideas to improve SLIPTA star ratings, continued SLIPTA Checklist sections Baseline performance Specific problem being addressed Change ideas Improved performance 3. Organization and personnel Kawolo, 65%; and Matany, 10%. Duty roaster not reflecting tasks Incomplete personnel files High staff attrition rates No assigned Quality Officer Designed duty roster that reflected task allocation Secured lockers for safe keeping of personal documents Prepared a list of documents required and copied to the hospital administration for purposes of enforcement Engaged administration in enforcement of complete files e.g. Involved administration in Kabale to put up a warning of withholding salaries for all incomplete personnel files Kawolo, 91% and Matany, 68%. No organogram in place Appointment of Quality Officer with clearly defined responsibilities Following mentorship from the SUSTAIN team, meetings were held with lab staff and admin representatives to agree on structure and positions and an organogram was designed 4. Client management and customer service Kawolo, 13%; and Masaka, 63%. Suggestion box was not utilized Lack of clear procedure on customer communication A complaint log was created and personnel were assigned to follow up complaints; In Jinja a desk phone was procured and a phone number distributed for addressing complaints, awareness on this was created through radio talk shows and talks to clients within the hospital. Staff were trained and mentored on effective customer communication Kawolo, 80%; and Masaka, 80%. Developed SOP for client management and client satisfaction and had regular QI meetings, onsite training and customer survey 5. Equipment preventive maintenance Kabong, 23%; and Moroto, 29%. No equipment verification done Lack of annual equipment service plan Designated an equipment focal person Improved on documentation i.e. made equipment inventory list and equipment service logs Staffs were trained on how to do equipment checks, and verification reports are now required of all vendors and kept on file Kabong, 80% and Moroto, 71%. Service year plan was developed in collaboration with the biomedical engineers 6. Internal Audit Jinja, 20% and Kabong, 20%. Internal audits not being conducted regularly Measurement uncertainty studies not done Adoption of the ISO guidelines and LQMS as procured and provided by Uganda national health laboratory services (UNHLS) Training of internal auditors to focus on use of SLIPTA and ISO guideline to enable lab team to be prepared for auditors Jinja, 40% and Kabong, 80%. Had all lab staff participate in internal audits Instituted corrective, preventive and follow up actions Training of clinicians on calculation of the MU and use of uncertainty values. continued 16 Laboratory Services Change Package

Table 6: Specific change ideas to improve SLIPTA star ratings, continued SLIPTA Checklist sections Baseline performance Specific problem being addressed Change ideas Improved performance 7. Purchasing and Inventory Soroti, 26%; and Masaka, 33.3%. Stock outs of lab supplies Opened stock cards for all lab reagents/supplies and started timely ordering Trained staff on LQMS and had action plans for follow up purposes both internally and externally Soroti, 95.8% and Masaka, 75% Created a lab store and introduced daily signing out of supplies from the store to control and track their use Created lab budget Developed a checklist for supplies reception Created a list of approved manufacturers; Conducted physical counts of inventory Created log for expired items and designated focal person in charge of this log, designated space for storage of expired items before they are taken back to central stores. 8. Information management Abim, 17% and Lira, 61% There were inconsistencies in test result verification, sometimes done and others not No standard lab request forms Developed procedure for test result verification and made a stamp for verified results Designed standard request forms aligned to the ISO 15189 Designated a place for archival of results and created file for archived results Abim, 62% and Lira, 81%. No system for archival of data 9. Corrective Action Mubende, 41% and F/portal, 50%. No adequate documentation of Non- Conformities (NC) No root cause analysis done Designed tools for documenting NC and sensitized staff on the use of these tools preventive and corrective action and importance of documenting and reporting NC Risk assessment was done and staff were trained in the different methodologies of doing root cause analysis and provided with reference materials Mubende, 58% and Fort Portal, 100%. 10. Occurrence/ incident management and process improvement Arua, 27% and Kabong, 20%. No occurrence logs developed Lack of commitment on use of tools to communicate quality findings CME/trainings on definition of an occurrence and its documentation Assign specific person to use tools and communicate quality findings Arua, 90% and Kabong,100%. 11. Facilities and Safety Arua, 27% and Kabong, 20%. Absence of safety tools (Safety cabinet, eye wash, fire extinguisher) No MSDS No evidence of documentation of vaccination of staff (certificate) Designated and trained safety officer Trainings and fire drills on lab safety Developed a floor plan to capture all work stations and ensure safety procedures in all areas Kabale 40/43 (93%) Gulu 36/43 (84%) Kotido 21/43 (49%) Inadequate space Safety audits not conducted Laboratory Services Change Package 17

Key Challenges As expected, some health facilities and improvement teams faced significant challenges while testing the different change ideas aimed at improving laboratory services such as: High staffing attrition rates across majority of the health facilities competency in quality improvement is acquired incrementally, through continuous application of QI principles and iterative PDSA cycles. Staff attrition affects this continuity, and institutional memory lost during staff departures greatly affects the sustainability of effective change ideas. n Extensive reliance on volunteers to implement the tested change ideas this affects the sustainability of implemented interventions, and limits capacity building among health workers. Moving Forward To obtain the best benefit from the change ideas in this document, health facilities should establish and cultivate an environment that embraces change to nurture improvements. The following ideas can help QI teams in getting started: n Improve documentation Existing national documentation and data monitoring tools need to be accurately and consistently used. It is through these tools that teams will be able to determine whether their performance is stagnating or improving, both before and after introducing these changes. n Establish team work For any improvement work to yield positive results, health workers should collaborate and view themselves as members of a team responsible for the different steps in the processes of providing health services. n Analyze the entire process of care, and prioritize After analyzing processes of care, and identifying existing gaps, health workers should prioritize which challenges need to be and can be tackled first and which ones can wait. Addressing one challenge at a time (while introducing a few changes at a time) will enable health workers to systematically monitor the effectiveness of each change in addressing a challenge, and the effect of a process gap on the overall service delivery. n Constantly communicate with your patients Improvements are designed to primary benefit patients (although health workers also benefit from the improved efficiencies). Health workers should constantly seek feedback from patients on the quality of service they are provided, and whether the changes being implemented are benefitting them as well. In addition to health facility QI teams, other stakeholders have differing roles (as indicated in the table below) in the spread of these change ideas which are listed in the table below. Ministry of Health District Health Officials Development Partners Ensure the required tools, standard operating procedures and other resources are available throughout all levels of the health system Support coordination, capacity building, supportive supervision, and resource mobilization as they relate to scale-up of QI interventions Supportive supervision, coaching and mentorship of health facilities attempting QI projects Provide technical support and avail resources to bridge funding gaps within the MOH 18 Laboratory Services Change Package