Evidence for Malaria Medicines Policy. ACTwatch Study Reference Document Private-Sector Fever Case Management Study Uganda 2015

Evidence for Malaria Medicines Policy ACTwatch Study Reference Document Private-Sector Fever Case Management Study Uganda 2015

Released August 16, 2016 Suggested citation ACTwatch Group and PACE. (2016). ACTwatch Study Reference Document: Uganda Private-Sector Fever Case Management Study 2015. Washington DC: PSI. Contact Dr. Megan Littrell ACTwatch Principal Investigator PSI 1120 19 th St NW Suit 600 Washington DC 20036 mlittrell@psi.org Peter Buyungo Programme for Aaccessible Communication and Education (PACE) Uganda Plot # 2, Ibis Vale, Kololo off Prince Charles Drive Kampala, Uganda pbugungo@pace.org.ug Acknowledgements ACTwatch is funded by the Bill and Melinda Gates Foundation, UNITAID, and the UK Department for International Development. This study was implemented by Population Services International (PSI). PACE Uganda Peter Buyungo Henry Kaula Doreen Nakimuli Ministry of Health Dr. Peter Okui ACTwatch Team Andrew Andrada Erick Auko Dr. Katie Bates Dr. Desmond Chavasse Kevin Duff Keith Esch Anna Fulton Tarryn Haslam Catharine Hurley Dr. Beth Kangwana Gloria Kigo Dr. Megan Littrell Julius Ngigi Dr. Kate O Connell Ricki Orford Stephen Poyer Dr. Justin Rahariniaina Christina Riley Dr. Andria Rusk Julianna Smith Rachel Thompson Cynthia Whitman Socio-Economic Data Center LTD (SEDC) Prof. Asingwire Narathius Joseph Kiwanuka Dr. Denis Muhangi SEDC (cont.) Dr. Janestic Twikirize Hadijah Mwenyango Sharlotte Tusasirwe Aloysious Nyombi Jacob Mutazidwa Team Leaders Tamali Adiru Robert Muwanguzi Joseph Jjumba Susan Akol Anibaya Rose Namugerwa George Basalirwa Suzan Nakkazi Fredrick Kabogoza Albert Gayi Bonita Nyamwire Team Members Juliet Donna Eyokia Creda Canongio Brian Mugenyi Annet Nnabunya Charlotte Kubahika Rogers Niwamanya Eria Paulo Ogwang Josephine Imaju Eliya Sabiiti Phionah Kamwine Rabson Masereka Moreen Mwamula Agnes Kateme Sam Meya Michael Paul Okumu Boniface Okello Robert Okello Team Members (cont.) Mary Christabel Etap Evans Arinda Patricia Namugenyi Julius Kankiriho Samuel Kakooza Fredrick Sserubuga Jimmy Opoka www.actwatch.info Page 1

Table of Contents LIST OF TABLES... 3 LIST OF FIGURES... 4 DEFINITIONS... 6 INTRODUCTION... 8 SUMMARY OF METHODS AND DATA COLLECTION... 9 SUMMARY OF KEY FINDINGS... 12 TABLES... 27 ANNEX 1: ACTWATCH BACKGROUND... 41 ANNEX 2: COUNTRY BACKGROUND... 44 ANNEX 4: SAMPLED SUB-COUNTIES... 50 ANNEX 3: FEVER CASE MANAGEMENT SURVEY METHODS... 52 ANNEX 5: QUESTIONNAIRE... 54 ANNEX 6: SAMPLING WEIGHTS... 75 ANNEX 7: INDICATOR DEFINITIONS... 77 www.actwatch.info Page 2

List of Tables Table 1: Description of outlets*... 27 Table 2: Description of fever patients*, by patient age and outlet type... 28 Table 3: Malaria blood testing, by patient age and outlet type... 30 Table 4: Positive malaria blood test result* among patients tested for malaria, by patient age and... 32 outlet type... 32 Table 5: Fever treatment by malaria test result, patient age and outlet type... 33 Table 6: Patient comprehension of malaria diagnostic testing... 36 Table 7: Factors associated with malaria blood testing... 37 Table 8: Factors associated with QA ACT treatment, among fever patients with confirmed malaria... 39 Table X1. Sampled Sub-Counties... 50 www.actwatch.info Page 3

List of Figures Figure 1: Diagram for outlet inclusion in the Fever CM study, Uganda, 2015... 10 Figure 2: Diagram for patient inclusion in the Fever CM study, Uganda, 2015... 11 Figure 3. Percentage of patients present at the outlet, across patient age... 12 Figure 4. Percentage of patients present at the outlet, across outlet type... 13 Figure 5. Percentage of patients who sought previous treatment for the current illness at a different source of care, across patient age... 14 Figure 6. Percentage of patients who sought previous treatment for the current illness at a different source of care, across outlet type... 14 Figure 7. Percentage of patients who reported receiving previous malaria testing and treatment for the current illness at a different source of care, across patient age... 15 Figure 8. Percentage of patients who reported receiving previous malaria testing and treatment for the current illness at a different source of care, across outlet type... 15 Figure 9. Percentage of patients who received a malaria blood test, across patient age... 16 Figure 10. Percentage of patients who received a malaria blood test, across outlet type... 17 Figure 11. Percentage of patients who received a malaria blood test, across patient age... 18 Figure 12. Percentage of patients who received a malaria blood test, across outlet type... 18 Figure 13. Percentage of patients who tested positive for malaria, among tested patients, across patient age... 19 Figure 14. Percentage of patients who tested positive for malaria, among tested patients, across outlet type... 19 Figure 15. Treatments received by patients who tested positive for malaria, across patient age... 20 Figure 16. Treatments received by patients who tested positive for malaria, across outlet type... 21 Figure 17. Treatments received by patients who tested negative for malaria, across patient age... 22 Figure 18. Treatments received by patients who tested negative for malaria, across outlet type... 23 Figure 19. Treatments received by patients who were not tested for malaria, across patient age... 24 Figure 20. Treatments received by patients who were not tested for malaria, across outlet type... 25 www.actwatch.info Page 4

Acronyms ACT AETD AL AMFm ASAQ BMGF CHW DHA PPQ DHS EMA GFATM iccm MOH NGO NMCP Oral AMT PMI Pf QA ACT mrdt SP UK USAID VHT WHO Artemisinin combination therapy Adult equivalent treatment dose Artemether lumefantrine Affordable Medicines Facility malaria Artesunate amodiaquine The Bill and Melinda Gates Foundation Community Health Worker Dihydroartemisinin piperaquine The Demographic and Health Survey European Medicines Agency Global Fund to Fight AIDS, TB, and Malaria Integrated community case management Ministry of Health Non-governmental Organization National Malaria Control Program Oral artemisinin monotherapy President s Malaria Initiative Plasmodium falciparum Quality-assured artemisinin combination therapy Malaria rapid diagnostic test Sulfadoxine Pyrimethamine United Kingdom United States Agency for International Development Village Health Team World Health Organization www.actwatch.info Page 5

Definitions Survey Methods Definitions Outlet Outlet survey Cluster Outlets eligible for inclusion in the Fever CM study Patients eligible for inclusion in the Fever CM study Consultation observation Patient exit interview Provider interview Any service delivery point or point of sale for commodities. Outlets are not restricted to stationary points of sale and may include mobile units or individuals. The Fever CM study was conducted as part of the 2015 national ACTwatch outlet survey. The outlet survey included a census of all public- and private-sector outlets with the potential to distribute antimalarials within a nationally representative sample of sub-counties. Outlets with antimalarials and/or malaria testing (malaria rapid diagnostic test or microscopy) in stock on the day of the survey or within the past three months were eligible for interview. The interview included an audit/inventory of all available antimalarials and malaria tests, as well as provider questions to assess malaria case management knowledge. The primary sampling unit, or cluster, for the outlet and Fever CM surveys. It is an administrative unit that hosts a population size of approximately 10,000 to 15,000 inhabitants. These units are defined by political boundaries. In Uganda, these were defined as sub-counties Outlets were eligible for the Fever CM study if they met the following criteria: 1) Private forprofit health facility, pharmacy, or drug store; 2) First-line artemisinin combination therapy (ACT) artemether lumefrantrine (AL) in stock on the day of the outlet survey; and 3) malaria blood testing available on the day of the outlet survey (mrdt or microscopy). Patients were eligible for the Fever CM study if they met the following criteria: 1) Respondent age 18 or older, 2) Patient at least 2 months of age, 3) Illness that includes fever or history of fever, 4) Presenting for treatment for this illness at this outlet for the first time, 5) Uncomplicated illness (not severe or life threatening), 6) Not currently pregnant, 7) Provides consent to participate in the study. A structured observation checklist was completed by an interviewer observing the interactions that the patient had with providers as she/he was provided with services at the outlet. The observation was concerned primarily with provider behaviors, including assessment, proper mrdt administration, and counseling for treatment with ACT. A brief exit interview was conducted with the patient after the patient visit was complete. The exit interview was concerned with capturing information about all medicines prescribed/obtained, and assessing patient understanding of the test result(s), diagnosis, and medication regimens prescribed. A brief series of questions was administered to providers to assess demographic characteristics, qualifications, training, and knowledge of the first-line treatment. Malaria Product Indicator Definitions Antimalarial Dosing/treatment regimen Artemisinin and its derivatives Artemisininbased Any medicine recognized by the World Health Organization (WHO) for the treatment of malaria. Medicines used solely for the prevention of malaria were excluded from analysis of key indicators in this report. The posology or timing and number of doses of an antimalarial used to treat malaria. This schedule often varies by patient weight. Artemisinin is a plant extract or synthetic plant extract used in the treatment of malaria. The most common derivatives of artemisinin used to treat malaria are artemether, artesunate, and dihydroartemisinin. An antimalarial that combines artemisinin or one of its derivatives with an antimalarial or antimalarials of a different class. www.actwatch.info Page 6

Combination Therapy (ACT) Non-artemisinin therapy First-line treatment Quality-assured Artemisinin- Based Combination Therapies (QA ACTs) An antimalarial medicine that does not contain artemisinin or any of its derivatives. The government-recommended treatment for uncomplicated malaria. Uganda s first-line treatment is artemether-lumefantrine. QA ACTs are ACTs that comply with the Global Fund to Fight AIDS, Tuberculosis and Malaria (GFATM) s Quality Assurance Policy. A QA ACT is any ACT that appeared on GFATM s indicative list of antimalarials meeting GFATM s quality assurance policy prior to data collection (see http://www.theglobalfund.org/en/healthproducts/qualityassurance/) or that previously had C- status in an earlier GFATM quality assurance policy. QA ACTs also include ACTs that have been granted regulatory approval by the European Medicines Agency (EMA) specifically Eurartesim and Pyramax. www.actwatch.info Page 7

Introduction This reference document is a detailed presentation of the 2015 national ACTwatch Fever Case Management (Fever CM) study conducted in Uganda. The 2015 study was completed as part of the national ACTwatch outlet survey. The 2015 outlet survey followed previous survey rounds conducted by ACTwatch in Uganda in 2010, 2011, 2013, and 2015. Outlet survey findings from Uganda are available at www.actwatch.info. ACTwatch is a multi-country research project implemented by PSI (www.psi.org). Standardized tools and approaches are employed to provide comparable data across countries and over time. ACTwatch is designed to provide timely, relevant, and high-quality antimalarial market evidence. The goal of providing this market evidence is to inform and monitor national and global policy, strategy, and funding decisions for improving malaria case management. The project was launched in 2008 with funding from the Bill and Melinda Gates Foundation (BMGF), and is currently funded through 2016 by the BMGF, UNITAID, and DFID. See Annex 1 for more information about the ACTwatch project. Recent efforts to improve malaria case management have focused on improving access to two key commodities: 1) malaria rapid diagnostic tests (mrdt) for confirmatory testing before treatment and 2) artemisinin-based combination therapy (ACT effective treatment for P. falciparum malaria). There is now evidence in Uganda that availability of malaria blood testing and quality-assured ACT have improved dramatically in recent years. However, due to the complexities of measuring malaria case management outcomes, there is less information available on the extent to which improved access to malaria rapid diagnostic tests (mrdts) and ACTs is facilitating consistent, appropriate management of suspected malaria cases. In other words, are available mrdts consistently used to confirm all suspected malaria cases, and are ACTs administered to confirmed cases only? Following on the significant investments made to improve access to malaria diagnostics and treatment, there is a need to document the uptake of these tools. This includes uptake of mrdts and ACT within sources of care that are frequently accessed by people seeking fever treatment. In Uganda, private providers at private for-profit health facilities, drug shops, and pharmacies are commonly accessed for fever care, and yet information on management of cases is typically not tracked in any systematic way. Health provider behavior is complex, and multiple factors influence case management practices and outcomes. As compared with the more highly regulated and supervised public health sector, the set of factors influencing case management outcomes may be particularly complex and yet not welldocumented in the private sector. The Fever CM study documents key aspects of the interactions between private-sector providers and clients seeking fever treatment, including two case management outcomes: 1) confirmatory malaria blood testing and 2) appropriate treatment according to test result. This research complements available evidence currently used to track progress in malaria case management. The ACTwatch outlet surveys track trends in mrdt and antimalarial availability, price, and market share. The Fever CM research component documents the extent to which recent and current efforts to improve availability of key malaria commodities are sufficient for facilitating appropriate management of suspected malaria cases. This research will provide information to inform interventions designed to close gaps between availability of quality-assured malaria diagnostics and medicines and their routine use in managing clients. Report notes This document is a complete reference for the 2015 Fever CM survey. Please see annexes for information about the study context, design, implementation, and data analysis. Grey text for data appearing in report tables indicates that the estimate provided was derived from a small sample size. Specifically, grey text is used to indicate point estimates derived from an n of less than 50. Results are presented for children under five and people age five and older. The total patient estimates include n=6 patients for which patient age was missing. www.actwatch.info Page 8

Summary of Methods and Data Collection National Outlet Survey A representative antimalarial outlet survey was conducted at the national level in Uganda between May 19 and July 2, 2015. A full description of research design and methods is provided in the Uganda Outlet Survey Reference Document available at www.actwatch.info. Briefly, a representative sample of sub-counties was selected from urban and rural domains (see Annex 3). Within selected clusters, a census of all outlets with the potential to sell or distribute antimalarials and/or provide malaria blood testing was completed. Outlets were screened to determine eligibility. Outlets eligible for the survey met at least one of three criteria: 1) one or more antimalarials were in stock on the day of the survey, 2) one or more antimalarials were in stock in the three months preceding the survey, and/or 3) malaria blood testing (microscopy or RDT) was available. A structured questionnaire was used to complete an audit of all antimalarials and RDTs as well as a provider interview. Fever Case Management Survey Private-sector outlets meeting eligibility criteria for the Fever CM survey were identified during the outlet survey and revisited for Fever CM data collection. Eligibility criteria for the Fever CM survey component were as follows: 1) private for-profit health facility, pharmacy, or drug store; 2) national first-line ACT, artemether lumefantrine (AL), in stock on the day of the outlet survey; and 3) malaria blood testing (mrdt or microscopy) available on the day of the outlet survey. Fieldworkers dedicated to the Fever CM study revisited eligible outlets identified during the outlet survey and completed patient screening for a maximum of one day. All patients visiting eligible outlets were screened to determine eligibility for the study. Clients were invited to participate in the study if they met the following eligibility criteria: Respondent age 18 or older Patient at least 2 months of age Illness that includes fever or history of fever Presenting for treatment for this illness at this outlet for the first time Uncomplicated illness (not severe or life threatening) Not currently pregnant Provides consent to participate in the study Following informed consent procedures, a structured observation checklist was completed by an interviewer observing the interactions that the patient had with providers as she/he was provided with services at the outlet. The observation was concerned primarily with provider behaviors, including assessment, proper RDT administration, and counseling for treatment with ACT. After the patient visit was complete, a brief exit interview was administered with the patient. The exit interview was concerned with capturing information about all medicines prescribed/obtained, and assessing patient understanding of the test result(s), diagnosis, and medication regimens prescribed. Full details of survey methodology are available in Annex 4 and the Fever CM questionnaire is available in Annex 5. Double data entry was completed using Microsoft Access. All data cleaning and analysis was performed using Stata 13.1 ( StataCorp, College Station, TX). Data were weighted to account for variation in probability of outlet selection (see Annex 6), and standard error calculation reflected clustering of outlets at commune level. Standard indicators were constructed according to definitions applied across ACTwatch project countries (see Annex 7). www.actwatch.info Page 9

Figure 1: Diagram for outlet inclusion in the Fever CM study, Uganda, 2015 A Eligible outlets* [1,266] B Outlets visited for the Fever CM study** [1,146] C Outlets that participated in patient screening*** [1,089] Eligible outlets that were not revisited for the Fever CM study ** [120] Outlets that did not participate in patient screening [57] Provider refused [32] Outlet closed [17] No patients during study period [8] Outlets with no eligible patients [259] Outlet Type Outlet Location D Outlets with completed patient observation and exit interviews [830] 260 147 423 389 441 Private For-Profit Health Facility Pharmacy Drug Store Urban Rural * Outlets identified during the national outlet survey that met the following criteria: 1) Pharmacy, drug store, or private for-profit health facility including hospitals and clinics; 2) ACT in stock on the day of the outlet survey; 3) malaria testing available on the day of the survey (malaria RDT and/or microscopy). ** 120 eligible outlets were not flagged during the outlet survey as eligible for the Fever Case Management component and therefore were not included in the study. *** Patient screening was conducted for a maximum of one day or until patient observation and exit interviews were completed for one eligible child under age 5 and one eligible person age 5 and older. www.actwatch.info Page 10

Figure 2: Diagram for patient inclusion in the Fever CM study, Uganda, 2015 A Patients screened* [9,330] Respondent under 18 [574] No fever [6,321] Patients without completed observation and exit interview [8,057] Follow-up visit to same outlet [597] Infant < 2 months [19] Pregnant [101] Severe illness [52] Refusal [389] Completed observation only [4] Patient Age 49 Patient Sex 1 B Patients with completed observation and exit interview [1,273] 503 545 170 0-4 years 5-14 years 15-49 years 50+ years Unknown 647 625 Male Female Unknown * Patients were screened when they first approached an outlet to determine eligibility for the Fever CM component. Eligibility criteria were as follows: Respondent at least 18 years of age (patient or patient s caregiver), seeking treatment for fever, seeking treatment for the first time for the current illness at this outlet, at least 2 months of age, not reportedly pregnant, and no signs or referral for severe illness for this illness. 389 patients refused to participate, and 4 were observed but refused to complete the exit interview. www.actwatch.info Page 11

Summary of Key Findings Figure 3. Percentage of patients present at the outlet, across patient age Among all patients with completed consultation observation and exit interviews Age 0-4 Age 5+ All Patients Patient present at the outlet Patient not present (family/friend seeking treatment) 32 20 26 68 80 74 N=545 N=722 N=1273 74 of patients that were included in the study were present at the outlet at the time of the consultation observation and exit interview. The remaining 26 were not present and care was being sought on the patient s behalf by a family member or friend. The percentage of patients present during the consultation was higher for people age five and above (80) as compared to children under five (68). www.actwatch.info Page 12

Figure 4. Percentage of patients present at the outlet, across outlet type Among all patients with completed consultation observation and exit interviews Private For-Profit Health Facilities Pharmacies Drug Stores Patient present at the outlet 15 Patient not present (family/friend seeking treatment) 85 51 49 35 65 N=630 N=219 N=424 The percentage of patients present during the consultation was highest for patients seeking care at private for-profit health facilities (85), and relatively lower among patients seeking care at pharmacies (49) and drug stores (65). Among patients seeking care at each outlet type, children under five were consistently less likely to be present at the consultation as compared with people age five and above (see Table 2). www.actwatch.info Page 13

PERCENT OF PATIENTS Figure 5. Percentage of patients who sought previous treatment for the current illness at a different source of care, across patient age Among all patients with completed consultation observation and exit interviews About one-quarter of all patients seeking fever treatment reported seeking treatment at another source of care prior to study visit (23). Previous treatment seeking was reported in the public sector among 10 of patients and in the private sector among 14 of patients. Levels of previous treatment seeking were similar among children under five as compared with people age five and older. PERCENT OF PATIENTS 100 90 80 70 60 50 40 30 20 10 0 Sought Previous Treatment Sought Previous Treatment - Public Sector Age 0-4 Age 5+ All Patients Sought Previous Treatment - Private Sector Figure 6. Percentage of patients who sought previous treatment for the current illness at a different source of care, across outlet type Among all patients with completed consultation observation and exit interviews 100 90 80 70 60 50 40 30 20 10 0 Sought Previous Treatment Sought Previous Treatment - Public Sector Sought Previous Treatment - Private Sector Private For-Profit Facilities Pharmacies Drug Stores More than one-third of patients seeking care at pharmacies reported seeking previous treatment for the current illness (38), as compared with one-quarter (25) of patients seeking care at drug stores and 21 of patients seeking care at private for-profit facilities. Previous treatment-seeking in the public sector was more common among patients seeking care at pharmacies (18) and drug stores (13), as compared with patients seeking care at private facilities (6). www.actwatch.info Page 14

PERCENT OF PATIENTS Figure 7. Percentage of patients who reported receiving previous malaria testing and treatment for the current illness at a different source of care, across patient age Among all patients with completed consultation observation and exit interviews 100 90 80 70 60 50 40 30 20 10 0 Reported Previous Malaria Test Reported Previous Malaria Treatment 9 of patients reported receiving a malaria blood test and 13 reported receiving antimalarial treatment from another source of care prior to the study visit. Previous malaria testing and treatment was similar among children under five and people age five and older. PERCENT OF PATIENTS Figure 8. Percentage of patients who reported receiving previous malaria testing and treatment for the current illness at a different source of care, across outlet type Among all patients with completed consultation observation and exit interviews 100 90 80 70 60 50 40 30 20 10 0 Reported Previous Malaria Test Age 0-4 Age 5+ All Patients Reported Previous Malaria Treatment Private For-Profit Facilities Pharmacies Drug Stores One-quarter (26) of patients seeking care at pharmacies reported receiving a malaria blood test at another source of care prior to the study visit, as compared with 10 of patients seeking care at drug stores and 7 of patients seeking care at private for-profit facilities. Data trends suggest higher reports of previous antimalarial treatment received among patients seeking care at drug stores (13) and private for-profit health facilities (13) compared with patients seeking care at pharmacies (7). www.actwatch.info Page 15

Figure 9. Percentage of patients who received a malaria blood test, across patient age Among all patients with completed consultation observation and exit interviews, including patients present and not present during the consult Received a malaria blood test - mrdt Age 0-4 Age 5+ All Patients Received a malaria blood test - mrdt & microscopy Received a malaria blood test - microscopy 32 28 20 29 26 29 Present, did not receive a malaria test Not present (did not receive a malaria test) 25 2 13 N=545 35 13 3 N=722 30 13 2 N=1273 44 of all fever patients received a malaria blood test. Testing was similar among children younger than five (43) and people age five and older (46). The majority of patients tested for malaria were tested using mrdts. Among all fever patients, 31 were tested by mrdt and 15.5 were tested by malaria microscopy. 2 of patients were tested by both mrdt and microscopy. Nearly one-third of all patients were present at the outlet but did not receive a malaria blood test (30). Among children younger than five, 25 were present at the outlet and not tested compared with 35 of people age five and older. In nearly one in ten cases (9), the provider recommended a blood test but the patient or caregiver refused the test (see Table 3). www.actwatch.info Page 16

Figure 10. Percentage of patients who received a malaria blood test, across outlet type Among all patients with completed consultation observation and exit interviews, including patients present and not present during the consult Received a malaria blood test - mrdt Private For-Profit Health Facilities 0 Pharmacies Drug Stores Received a malaria blood test - mrdt & microscopy Received a malaria blood test - microscopy Present, did not receive a malaria test Not present (did not receive a malaria test) 15 22 27 31 5 N=630 51 48 N=219 35 35 29 1 N=424 63 of all patients seeking care at private for-profit health facilities received a malaria blood test, as compared with 30 at drug stores and 1.5 at pharmacies. At private health facilities, testing by mrdt and microscopy were frequently observed. A total of 36 of patients at private facilities were tested by mrdts and 32 by malaria microscopy. 5 of patients were tested by both mrdt and malaria microscopy. Nearly all testing performed at drug stores was by mrdt. Among patients seeking care at private facilities, 22 were present but did not receive a test. A higher percentage of patients were present but not tested at drug stores (35) and pharmacies (42). At private facilities, only 15 of patients were not present at the consult (and therefore could not be tested). This compares with nearly half of patients at pharmacies (45) and one-third at drug stores (35). A provider recommended a blood test but the patient or caregiver refused the test for 8 of patients at private facilities, as compared with 10 at drug stores and only 3 at pharmacies (see Table 3). www.actwatch.info Page 17

Figure 11. Percentage of patients who received a malaria blood test, across patient age Among all patients present during the completed consultation observation and exit interview PERCENT OF PATIENTS 100 90 80 70 60 50 40 30 20 10 0 Age 0-4 Age 5+ All Patients Among patients who were present during the consultation, 60 received a malaria blood test. Data trends suggest that testing was more common among children younger than five (64) compared to patients age five and older (57). PERCENT OF PATIENTS Figure 12. Percentage of patients who received a malaria blood test, across outlet type Among all patients present during the completed consultation observation and exit interview 100 90 80 70 60 50 40 30 20 10 0 Private For-Profit Facilities Pharmacies Drug Stores Among patients who were present during the consultation, 74 who were managed at private for-profit health facilities received a malaria blood test, as compared with 46 at drug stores and only 3 at pharmacies. www.actwatch.info Page 18

Figure 13. Percentage of patients who tested positive for malaria, among tested patients, across patient age Among all patients with completed consultation observation and exit interviews PERCENT OF PATIENTS 100 90 80 70 60 50 40 30 20 10 0 Age 0-4 Age 5+ All Patients Among patients who were tested for malaria, 59 had a positive test result. The percentage of patients who tested positive for malaria was similar among children younger than five (59) and people age five and older (58). Figure 14. Percentage of patients who tested positive for malaria, among tested patients, across outlet type Among all patients with completed consultation observation and exit interviews PERCENT OF PATIENTS 100 90 80 70 60 50 40 30 20 10 0 Private For-Profit Facilities Pharmacies Drug Stores Among patients who were tested for malaria, data trends suggest a higher percentage of patients at drug stores tested positive (68), as compared with patients in private for-profit health facilities (54) and pharmacies (54). www.actwatch.info Page 19

PERCENT OF PATIENTS Figure 15. Treatments received by patients who tested positive for malaria, across patient age Among all patients with completed consultation observation and exit interviews 100 90 80 70 60 50 40 30 20 10 0 Any Antimalarial Any ACT Quality-Assured ACT Non-Quality-Assured ACT Non-Artemisinin Therapy Age 0-4 Age 5+ All Patients Artemisinin Monotherapy Any Antibiotic Any Pain/Fever Reducer Among patients who tested positive for malaria (N=266), 83 received antimalarial treatment. 60 received treatment with an ACT: 49 with a quality-assured ACT (ACT from a manufacturer with WHO pre-qualification or GFATM procurement approval) and 12 with a non-quality-assured ACT. 14 of all patients who tested positive received treatment with a non-artemisinin therapy. Non-artemisinin therapies received were primarily quinine injections, tablets, and syrups, as well as Sulfadoxine Pyrimethamine (SP) tablets and a few chloroquine tablet treatments. 15 of all positive patients received an artemisinin monotherapy, and these were primarily artemether injections (indicated for the treatment of severe malaria cases). Antibiotics were received by 43 of positive patients. These included amoxicillin, ampicillin, erythromycin, sulfamethoxazole and trimethoprim tablets and suspensions, metronidazole and ciprofloxacin tablets, and a variety of other antibiotics. The majority of patients were treated with a pain- or fever-reducing medication (79), such as acetaminophen (paracetamol), ibruprofen, or diclofenac. Among patients who tested positive for malaria and did not receive an antimalarial, a small percentage (14) reported receiving an antimalarial at another source, and 1 received a prescription for an antimalarial to be filled at another source of care (data not shown). Treatment of patients who tested positive for malaria was generally similar for children younger than five compared with people age five and older. Data trends suggest slightly higher antimalarial treatment for people age five and older (88) compared to children younger than five (81). 16 of people age five and older received a non-quality-assured ACT compared with 9 of children younger than five. However, data trends suggest that children younger than five were more likely to receive artemisinin monotherapy (19) compared with people age five and older (12). www.actwatch.info Page 20

PERCENT OF PATIENTS Figure 16. Treatments received by patients who tested positive for malaria, across outlet type Among all patients with completed consultation observation and exit interviews 100 90 80 70 60 50 40 30 20 10 0 Any Antimalarial Any ACT Quality-Assured ACT Non-Quality-Assured ACT Private For-Profit Health Facilities Non-Artemisinin Therapy Drug Stores Artemisinin Monotherapy Any Antibiotic Any Pain/Fever Reducer Among patients who tested positive for malaria at private for-profit health facilities (N=188), 74 received antimalarial treatment. Among positive patients at drug stores (N=75), nearly all (97) received antimalarial treatment. Only N=3 patients tested positive at pharmacies and are not shown here (see Table 5). Positive patients at drug stores were more likely to receive ACT treatment compared with patients at private facilities, including any ACT (78 versus 49) and qualityassured ACT (68 versus 49). Data trends suggest that artemisinin monotherapies (injectables) were more commonly received by patients at private facilities (20) compared with drug stores (7). Data trends suggest that antibiotics were more commonly received by positive patients at private facilities (47) compared with patients at drug stores (35), as well as that pain- and fever-reducing medications were more commonly received at drug stores (88) compared with private facilities (73). www.actwatch.info Page 21

PERCENT OF PATIENTS Figure 17. Treatments received by patients who tested negative for malaria, across patient age Among all patients with completed consultation observation and exit interviews 100 90 80 70 60 50 40 30 20 10 0 Any Antimalarial Any ACT Quality-Assured ACT Non-Quality-Assured ACT Non-Artemisinin Therapy Age 0-4 Age 5+ All Patients Artemisinin Monotherapy Any Antibiotic Any Pain/Fever Reducer Among patients who tested negative for malaria (N=250), 14 received antimalarial treatment. 10 received treatment with an ACT, and 4 received treatment with a non-artemisinin therapy (primarily SP). Antibiotics were received by 55 of negative patients. These included amoxicillin, ampicillin, erythromycin, sulfamethoxazole and trimethoprim tablets and suspensions, metronidazole and ciprofloxacin tablets, and a variety of other antibiotics. More than half of patients were treated with a pain- or fever-reducing medication (61), such as acetaminophen (paracetamol), ibruprofen, or diclofenac. Data trends suggest higher antimalarial treatment for negative children younger than five (19) compared to people age five and older (10). This included 15 of negative children younger than five who received an ACT compared with 6 of people age five and older. Data trends also suggest that non-antimalarial treatment was higher among children younger than five compared with people age five and older, including antibiotic treatment (62 versus 49) and treatment with pain- and fever-reducing medicines (69 versus 55). www.actwatch.info Page 22

PERCENT OF PATIENTS Figure 18. Treatments received by patients who tested negative for malaria, across outlet type Among all patients with completed consultation observation and exit interviews 100 90 80 70 60 50 40 30 20 10 0 Any Antimalarial Any ACT Quality-Assured ACT Non-Quality-Assured ACT Private For-Profit Health Facilities Non-Artemisinin Therapy Drug Stores Artemisinin Monotherapy Any Antibiotic Any Pain/Fever Reducer Among patients who tested negative for malaria at private for-profit health facilities (N=196), 12 received antimalarial treatment. Among negative patients at drug stores (N=50), 22 received antimalarial treatment. Only N=4 patients tested negative within pharmacies and are not shown here (see Table 5). Data trends suggest that negative patients at drug stores were more likely to receive ACT treatment (18) compared with patients at private facilities (8). Data trends suggest that antibiotics were more commonly received by negative patients at drug stores (65) compared with patients at private facilities (51), as well as that pain- and fever-reducing medications were more commonly received at private facilities (72) compared with drug stores (58). www.actwatch.info Page 23

PERCENT OF PATIENTS Figure 19. Treatments received by patients who were not tested for malaria, across patient age Among all patients with completed consultation observation and exit interviews 100 90 80 70 60 50 40 30 20 10 0 Any Antimalarial Any ACT Quality-Assured ACT Non-Quality-Assured ACT Non-Artemisinin Therapy Age 0-4 Age 5+ All Patients Artemisinin Monotherapy Any Antibiotic Any Pain/Fever Reducer Among patients who were not tested for malaria (N=753), 51 received antimalarial treatment. 43 received treatment with an ACT: 34 with a quality-assured ACT (ACT from a manufacturer with WHO pre-qualification or GFATM procurement approval) and 9 with a non-quality-assured ACT. 8 of all patients who were not tested for malaria received treatment with a non-artemisinin therapy. Non-artemisinin therapies received were primarily quinine injections, tablets, and syrups, as well as SP tablets and a few chloroquine tablet treatments. Antibiotics were received by 24 of patients who were not tested for malaria. These included amoxicillin, ampicillin, erythromycin, sulfamethoxazole and trimethoprim tablets and suspensions, metronidazole and ciprofloxacin tablets, and a variety of other antibiotics. The majority of patients were treated with a pain- or fever-reducing medication (65), such as acetaminophen (paracetamol), ibruprofen, or diclofenac. Treatment of patients who were not tested for malaria was generally similar for children younger than five compared with people age five and older. Data trends suggest slightly higher antimalarial treatment for people age five and older (54) compared to children younger than five (47). 45 of people age five and older received a non-quality-assured ACT compared with 39 of children younger than five. www.actwatch.info Page 24

Figure 20. Treatments received by patients who were not tested for malaria, across outlet type Among all patients with completed consultation observation and exit interviews 100 90 80 PERCENT OF PATIENTS 70 60 50 40 30 20 10 0 Any Antimalarial Any ACT Quality-Assured ACT Non-Quality-Assured ACT Non-Artemisinin Therapy Artemisinin Monotherapy Any Antibiotic Any Pain/Fever Reducer Private For-Profit Health Facilities Pharmacies Drug Stores Among patients who were not tested for malaria at private for-profit health facilities (N=244), 43 received antimalarial treatment. Data trends suggest that antimalarial treatment was higher for patients who were not tested at pharmacies (N=212, 55) and drug stores (N=297, 54). Treatment with non-qualityassured ACTs was higher at pharmacies (18), as compared with private facilities (5) and drug stores (10). Data trends suggest higher treatment with qualityassured ACTs at private facilities (32) and drug stores (36), as compared with pharamcies (25). Antibiotic treatment for people who were not tested for malaria was similar across outlet types. Data trends suggest higher treatment with pain- and feverreducing medications among patients seeking care at drug stores (67), as compared with pharmacies (56) and private facilities (61). www.actwatch.info Page 25

Factors associated with malaria blood testing and quality-assured ACT treatment for confirmed cases Logistic regression was used to test for association between outlet, patient, and provider characteristics and two outcomes: 1) malaria blood testing for all fever patients and 2) quality-assured ACT treatment for confirmed cases. Factors with a significant association with malaria testing included (see Table 7): Outlet type: patients at drug stores were 27.1 times more likely to receive a test compared with pharmacies, and patients at private for-profit health facilities were 108.7 times more likely to receive a test compared to pharmacies. Patient report of fever during the consultation (prompted or unprompted): patients who reported fever to the provider were 3.8 times more likely to receive a test compared to patients who did not report fever. Previous testing: patients who were not previously tested for malaria for the current illness were 3.4 times more likely to be tested compared with patients who had previously received a malaria test. Factors that were not significantly associated with malaria testing included (see Table 7): Urban/rural location Patient sex Patient age Recent provider case management training Price of malaria testing (below median price versus median and above). Factors with a significant association with quality-assured ACT treatment, among confirmed cases, included (see Table 8): Outlet type: patients in drug stores were 3.9 times more likely to receive quality-assured ACT treatment compared with patients in private facilities. Price of quality-assured ACT treatment: patients within outlets where one adult equivalent treatment dose (AETD) of quality-assured ACT treatment was below median price were 1.6 times more likely to receive quality-assured ACT treatment, as compared with outlets where one AETD was the median price or higher. Factors that were not significantly associated with quality-assured ACT treatment included (see Table 8): Urban/rural location Patient sex Patient age Patient report of fever during the consultation (prompted or unprompted) Previous treatment-seeking and previous antimalarial treatment for the current illness Recent provider case management training Provider knowledge of the national first-line treatment Provider belief that ACT is most effective for malaria infection www.actwatch.info Page 26

Tables Table 1: Description of outlets* Private For-Profit Facility Pharmacy Drug Store All Outlets (95 CI) (95 CI) (95 CI) (95 CI) Proportion of outlets* N=423 N=147 N=260 N=830 Stocking malaria commodities on the day of the outlet survey First-line ACT (AL) 100 100 100 100 - - - - Quality-assured ACT 89.8 96.9 90.0 90.1 (84.3-93.5) (92.8-98.7) (79.9-95.3) (85.1-93.5) Non-artemisinin therapy 94.8 100 94.2 94.6 (88.1-97.8) - (89.8-96.8) (91.1-96.8) Any malaria test 100 100 100 100 - - - - Malaria RDT 66.8 100 97.2 82.7 (56.4-75.8) - (93.2-98.9) (76.6-87.5) Malaria microscopy 61.1 0.7 4.9 31.7 (51.7-69.7) (0.2-3.2) (2.3-10.3) (26.3-37.7) Provider training and knowledge** N=422 N=147 N=260 N=829 Reported receiving malaria CM training in the 52.8 55.5 36.4 44.8 past year** (45.2, 60.3) (37.5, 72.1) (29.2, 44.4) (39.3, 50.4) Correctly cited the national first-line 84.8 88.3 80.5 82.8 treatment for uncomplicated malaria (AL) (78.4, 89.6) (79.7, 93.5) (73.1, 86.2) (78.1, 86.6) Cited ACT as the most effective treatment for 88.4 94.4 86.9 87.8 uncomplicated malaria in adults (82.5, 92.5) (86.9, 97.7) (80.0, 91.7) (83.5, 91.1) Cited ACT as the most effective treatment for 83.9 86.4 86.3 85.1 uncomplicated malaria in children under five (76.5, 89.3) (70.6, 94.4) (80.0, 91.0) (80.4, 88.9) Cited ACT as the most effective treatment for 77.8 83.0 80.4 79.2 uncomplicated malaria in children under five and adults (69.5, 84.3) (68.8, 91.5) (72.0, 86.8) (73.4, 84.1) * Outlets with one or two patients who completed observation and exit interview components. ** Outlet had 1 or more providers with specified training or knowledge seeing patients on the day of the Fever CM survey. *** Provider reported receiving training, including pre-service training or stand-alone workshops, on malaria diagnosis/malaria RDTs/malaria microscopy and/or treatment on the national malaria treatment guidelines. Source: ACTwatch Fever CM Survey, Uganda, 2015. www.actwatch.info Page 27

Table 2: Description of fever patients*, by patient age and outlet type Age 0-4 Age 5 and Above All Patients Proportion of fever patients*: (95 CI) (95 CI) (95 CI) All outlets N=545 N=722 N=1273 Present at consultation 68.1 79.8 74.2 (60.8-74.6) (75.4-83.6) (69.5-78.4) Reported fever symptoms to a provider during 94.7 90.4 92.5 the consultation (prompted or unprompted) (91.8-96.6) (86.8-93.1) (90.3-94.2) Prior to the current visit for this illness: N=544 N=722 N=1272 Sought treatment elsewhere for this illness 23.3 23.4 23.1 (18.3-29.3) (19.0-28.4) (19.7-27.0) Public (facility, CHW) ^ 11.9 8.0 9.7 (8.4-16.5) (4.9-12.8) (7.4-12.6) Private (formal or informal) ^ 11.2 16.3 13.8 (7.5-16.4) (12.4-21.2) (10.7-17.7) Received a malaria blood test 10.8 6.9 8.6 (7.9-14.5) (4.8-9.8) (7.0-10.5) Received any medicine 20.8 20.6 20.5 (15.8-26.9) (16.7-25.1) (16.8-24.6) Received an antimalarial 13.4 12.0 12.5 (9.7-18.2) (8.6-16.4) (9.9-15.8) Private For-Profit Health Facilities N=275 N=352 N=630 Present at consultation 83.4 88.0 85.4 (72.7-90.5) (82.8-91.8) (79.5-89.9) Reported fever symptoms to a provider during 97.1 95.8 96.5 the consultation (prompted or unprompted) (93.3-98.8) (91.2-98.1) (94.1-97.9) Prior to the current visit for this illness: Sought treatment elsewhere for this illness 22.7 19.9 20.8 (16.4-30.6) (15.0-25.9) (16.7-25.7) Public (facility, CHW) ^ 9.5 3.0 5.9 (5.1-16.9) (1.4-6.3) (3.6-9.4) Private (formal or informal) ^ 12.9 17.0 14.9 (8.3-19.6) (12.2-23.2) (11.3-19.3) Received a malaria blood test 9.4 4.3 6.5 (4.8-17.4) (2.4-7.5) (4.3-9.8) Received any medicine 21.4 18.7 19.6 (15.3-29.1) (13.8-24.8) (15.6-24.4) Received an antimalarial 15.8 9.7 12.4 (10.1-24.0) (5.8-15.9) (9.0-16.8) Pharmacy N=80 N=138 N=219 Present at consultation 40.7 57.5 49.4 (21.0-63.9) (45.7-68.4) (36.5-62.3) Reported fever symptoms to a provider during 65.2 74.6 70.8 the consultation (prompted or unprompted) (40.5-83.8) (60.1-85.1) (53.5-83.7) Prior to the current visit for this illness: Sought treatment elsewhere for this illness Public (facility, CHW) Private (formal or informal) Received a malaria blood test 53.3 26.0 37.6 (39.2-66.9) (13.2-44.9) (28.8-47.2) 34.5 4.6 18.0 (22.2-49.2) (1.9-11.0) (12.7-25.0) 21.5 19.1 19.9 (9.9-40.7) (8.4-37.9) (9.8-36.3) 34.3 19.1 25.5 (21.0-50.7) (9.0-36.1) (17.3-35.9) www.actwatch.info Page 28

Table 2: Description of fever patients*, by patient age and outlet type Age 0-4 Age 5 and Above All Patients Received any medicine 23.2 12.0 16.7 (11.2-42.2) (6.5-21.0) (10.1-26.4) Received an antimalarial 11.1 4.0 7.1 (4.4-25.5) (1.5-10.2) (3.3-14.6) Drug Store N=190 N=232 N=424 Present at consultation 55.7 73.6 65.3 (46.1-64.9) (66.5-79.6) (58.9-71.2) Reported fever symptoms to a provider during 94.1 86.4 90.0 the consultation (prompted or unprompted) (88.9-96.9) (79.6-91.2) (85.5-93.2) Prior to the current visit for this illness: Sought treatment elsewhere for this illness 22.3 26.4 24.4 (15.1-31.6) (19.2-35.1) (19.3-30.4) Public (facility, CHW) ^ 12.7 12.8 12.8 (7.7-20.4) (7.3-21.4) (8.9-17.9) Private (formal or informal) ^ 9.1 15.6 12.5 (4.6-17.2) (10.3-22.8) (8.5-18.2) Received a malaria blood test 10.8 8.5 9.5 (7.2-15.8) (5.1-14.0) (7.1-12.7) Received any medicine 20.1 22.8 21.5 (13.0-29.8) (16.8-30.0) (16.3-27.7) Received an antimalarial 11.3 14.6 13.0 (7.7-16.4) (10.0-20.7) (9.8-17.2) * Patients eligible for the Fever CM study with completed observation and exit interview components. ^ Specific source for previous treatment was reported as don t know for 5 individuals and was missing for 2 individuals; these observations were excluded from these indicators. This includes 2 observations excluded from drug stores and 5 observations excluded from private health facilities. Source: ACTwatch Fever CM Survey, Uganda, 2015. www.actwatch.info Page 29

Table 3: Malaria blood testing, by patient age and outlet type Age 0-4 Age 5 and Above All Patients Proportion of patients who received*: (95 CI) (95 CI) (95 CI) All outlets N=545 N=722 N=1,273 Any malaria test 43.2 45.4 44.4 (37.2-49.5) (40.0-50.9) (39.4-49.6) Malaria RDT 29.8 32.4 31.3 (24.1-36.2) (27.5-37.7) (26.4-36.8) Malaria microscopy 15.2 16.1 15.5 (10.5-21.5) (12.0-21.2) (11.4-20.7) Both malaria RDT and microscopy 1.8 3.0 2.4 (0.7-4.3) (1.5-6.1) (1.3-4.5) No malaria test, present at the outlet 24.9 34.4 29.8 (19.6-31.0) (29.5-39.7) (25.8-34.1) No malaria test, not present at the outlet # 31.9 20.2 25.8 (25.4-39.2) (16.4-24.6) (21.6-30.5) Provider recommendation to test but refused 8.3 9.2 8.7 testing (5.3-12.8) (6.6-12.8) 6.3-11.9 Private For-Profit Health Facilities N=275 N=352 N=630 Any malaria test 62.3 63.7 63.0 (52.7-71.0) (55.8-70.9) (56.2-69.2) Malaria RDT 34.0 37.0 36.0 (25.9-43.3) (30.1-44.4) (29.5-43.0) Malaria microscopy 32.1 33.2 32.2 (22.6-43.3) (25.4-42.1) (24.4-41.0) Both malaria RDT and microscopy 3.8 6.5 5.2 (1.5-9.2) (3.0-13.6) (2.7-9.8) No malaria test, present at the outlet 21.1 24.4 22.5 (14.2-29.2) (18.5-31.3) (18.0-27.8) No malaria test, not present at the outlet # 16.6 12.0 14.6 (9.5-27.3) (8.2-17.2) (10.1-20.6) Provider recommendation to test but refused 6.1 10.3 8.2 testing (3.2-11.2) (6.3-16.3) (5.4-12.1) Pharmacy N=80 N=138 N=219 Any malaria test 1.7 1.5 1.5 (0.3-8.8) (0.5-4.7) (0.5-4.5) Malaria RDT 0.0 1.1 0.6 - (0.3-4.5) (0.2-2.5) Malaria microscopy 1.7 0.4 0.9 (0.3-8.8) (0.1-2.6) (0.2-4.0) Both malaria RDT and microscopy 0.0 0.0 0.0 - - - No malaria test, present at the outlet 39.0 56.0 47.8 (19.9-62.2) (44.5-66.9) (35.3-60.0) No malaria test, not present at the outlet # 59.3 42.5 50.7 (36.2-79.0) (31.6-54.3) (37.8-63.5) Provider recommendation to test but refused 0.9 4.7 3.0 testing (0.1-6.7) (1.5-13.8) (1.1-8.0) Drug Store N=190 N=232 N=424 Any malaria test 28.2 31.2 29.7 (20.9-37.0) (25.0-38.1) (23.7-36.5) Malaria RDT 27.6 30.0 28.8 (20.3-36.3) (23.6-37.2) (22.7-35.8) Malaria microscopy 0.7 1.2 1.0 (0.1-3.8) (0.3-4.1) (0.3-3.4) www.actwatch.info Page 30

Table 3: Malaria blood testing, by patient age and outlet type Both malaria RDT and microscopy No malaria test, present at the outlet No malaria test, not present at the outlet # Provider recommendation to test but refused testing Age 0-4 Age 5 and Above All Patients 0.0 0.0 0.0 - - - 27.5 42.4 35.6 (20.0-36.5) (34.9-50.3) (29.25-42.38) 44.3 26.5 34.7 (35.1-53.9) (20.4-33.5) (28.9-41.4) 10.7 8.5 9.5 (6.1-18.0) (5.4-13.3) (6.3-14.1) * Patients eligible for the Fever CM study with completed observation and exit interview components. # Patient was not present during the consult and care was sought on his/her behalf by a family member or friend. Source: ACTwatch Fever CM Survey, Uganda, 2015. www.actwatch.info Page 31

Table 4: Positive malaria blood test result* among patients tested for malaria, by patient age and outlet type Age 0-4 Age 5 and Above All Patients Proportion of patients who tested positive for malaria, among tested patients All outlets Private For-Profit Health Facilities Pharmacy Drug Store * Recorded during patient consultation observation. Source: ACTwatch Fever CM Survey, Uganda, 2015. (95 CI) (95 CI) (95 CI) N=231 N=284 N=516 59.1 57.5 58.7 (47.8-69.5) (49.4-65.3) (49.8-67.0) N=175 N=208 N=384 57.6 49.4 53.9 (45.4-69.0) (41.1-57.8) (44.5-63.0) N=2 N=5 N=7 86.4 24.7 53.7 (28.9-99.0) (4.8-68.0) (14.7-88.6) N=54 N=71 N=125 61.9 74.0 68.4 (42.2-78.3) (60.0-84.4) (55.5-79.0) www.actwatch.info Page 32

Table 5: Fever treatment by malaria test result, patient age and outlet type Malaria Test Positive Malaria Test Negative Not Tested Age 0-4 Age 5 and Above All Patients Age 0-4 Age 5 and Above All Patients Age 0-4 Age 5 and Above All Patients Proportion of patients who received: (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) All outlets N=124 N=141 N=266 N=107 N=143 N=250 N=313 N=435 N=753 Any antimalarial 80.7 87.8 83.0 19.1 10.3 14.3 47.4 53.6 50.8 (68.3-89.1) (79.1-93.2) (74.6-89.0) (10.9-31.3) (5.5-18.4) (9.3-21.2) (39.7-55.2) (45.7-61.4) (44.4-57.3) Any ACT 56.8 64.7 60.0 14.7 6.4 10.2 39.3 45.4 42.7 (44.1-68.8) (54.7-73.7) (53.3-66.3) (8.0-25.5) (3.6-11.2) (6.5-15.5) (31.5-47.6) (38.0-52.9) (36.4-49.3) Quality-assured ACT 48.3 49.8 48.5 9.0 4.7 6.6 32.9 34.1 33.9 (36.7-60.1) (40.3-60.1) (41.3-55.7) (3.9-19.5) (2.3-9.4) (3.7-11.6) (26.0-40.7) (26.8-42.3) (27.7-40.7) Non-quality assured ACT 8.5 15.8 12.2 5.7 1.7 3.5 6.3 11.3 8.9 (4.6-15.3) (9.5-25.2) (8.1-18.1) (1.6-18.7) (0.6-4.7) (1.3-9.4) (2.8-13.9) (7.0-17.7) (5.6-13.7) Non-artemisinin therapy * 12.7 16.5 14.4 3.1 3.9 3.6 8.8 8.1 8.4 (6.6-22.9) (10.3-25.2) (9.6-21.1) (0.6-14.0) (1.1-12.4) (1.3-9.3) (5.5-14.0) (5.0-12.8) (5.9-11.8) Artemisinin monotherapy ** 18.9 11.9 14.9 1.2 0.0 0.5 1.2 1.1 1.1 (10.5-31.7) (5.7-23.2) (9.3-23.0) (0.2-7.9) - (0.1-3.6) (0.4-4.0) (0.2-6.6) (0.4-3.3) Antibiotic 42.3 44.1 42.5 61.7 49.0 54.7 25.6 22.7 24.1 (30.1-55.6) (34.4-54.2) (34.1-51.3) (47.8-73.9) (37.6-60.4) (45.7-63.4) (18.8-34.0) (17.2-29.3) (18.9-30.1) Antipyretic 81.0 79.3 78.7 69.4 54.6 61.2 65.6 64.7 64.9 (69.9-88.7) (69.9-86.4) (71.4-84.5) (57.0-79.5) (41.3-67.2) (50.6-70.9) (56.7-73.5) (57.6-71.3) (59.5-70.0) Prescription for antimalarial 0.5 0.0 0.2 1.4 4.7 3.3 0.5 0.0 0.2 treatment to be filled elsewhere # (0.2, 1.6) - (0.1, 0.8) (0.2, 9.6) (0.7, 26.9) (0.7, 14.4) (0.1, 3.7) - (<0.1, 1.8) Private For-Profit Health Facilities N=91 N=96 N=188 N=84 N=112 N=196 N=99 N=143 N=244 Any antimalarial 72.9 79.6 74.1 11.0 11.9 11.5 40.6 43.3 43.0 (56.6-84.7) (66.2-88.6) (62.4-83.1) (4.2-25.8) (6.3-21.4) (6.3-20.3) (28.8-53.5) (30.1-57.6) (34.0-52.3) Any ACT 43.5 57.2 49.0 8.2 7.0 7.5 32.2 39.2 37.0 (30.3-57.8) (44.0-69.5) (41.5-56.5) (3.2-19.5) (3.9-12.2) (4.2-13.1) (19.7-47.9) (26.5-53.6) (27.3-47.7) Quality-assured ACT 34.9 39.1 35.9 6.4 5.2 5.7 29.3 31.8 31.7 (22.6-49.5) (28.4-50.9) (28.7-43.8) (2.1-18.2) (2.5-10.4) (2.8-11.3) (17.6-44.5) (20.2-46.2) (22.9-42.1) www.actwatch.info Page 33

Table 5: Fever treatment by malaria test result, patient age and outlet type Malaria Test Positive Malaria Test Negative Not Tested Age 0-4 Age 5 and Above All Patients Age 0-4 Age 5 and Above All Patients Age 0-4 Age 5 and Above All Patients Proportion of patients who received: (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) Non-quality assured ACT 8.7 18.1 13.1 1.8 1.8 1.8 3.0 7.4 5.2 (4.1-17.4) (10.1-30.5) (8.4-19.8) (0.5-5.8) (0.7-5.0) (0.7-4.4) (1.1-7.5) (3.1-16.6) (2.6-10.2) Non-artemisinin therapy * 12.1 13.0 12.2 1.1 4.9 3.3 8.4 7.0 7.5 (5.4-24.8) (6.7-23.9) (7.0-20.6) (0.2-6.7) (1.5-15.2) (1.1-9.9) (3.3-19.8) (3.0-15.6) (4.2-13.2) Artemisinin monotherapy ** 25.4 15.4 19.7 1.7 0.0 0.7 0.0 0.1 0.1 (13.8-41.6) (6.5-32.3) (12.6-29.4) (0.2-11.3) - (0.1-4.9) - (<0.1-0.8) (<0.1-0.4) Antibiotic 45.9 50.7 46.9 64.2 42.0 51.1 25.5 21.8 23.1 (31.3-61.3) (39.1-62.2) (37.7-56.4) (48.2-77.5) (29.9-55.1) (41.3-60.8) (13.4-43.1) (14.7-31.0) (15.0-33.8) Antipyretic 78.4 71.2 72.6 63.8 53.5 57.8 61.7 62.7 61.2 (63.6-88.3) (56.3-82.6) (61.2-81.6) (47.5-77.5) (38.5-68.0) (45.2-69.4) (48.5-73.5) (49.4-74.4) (51.2-70.4) Prescription for antimalarial 0.8 0.0 0.4 2.1 6.0 4.4 0.0 0.0 0.0 treatment to be filled elsewhere # (0.3, 2.5) - (0.1, 1.2) (0.3, 13.6) (0.8, 31.9) (0.9, 18.3) - - - Pharmacies N=1 N=2 N=3 N=1 N=3 N=4 N=78 N=133 N=212 Any antimalarial 100.0 100.0 100.0 100.0 17.2 28.7 48.7 59.5 55.3 - - - - (1.5-74.4) (5.0-75.4) (37.5-59.9) (38.0-77.9) (42.0-68.0) Any ACT 100.0 51.1 88.1 0.0 17.2 14.8 33.1 49.5 43.0 - (5.6-94.9) (35.8-99.0) - (1.5-74.4) (1.4-68.8) (20.1-49.2) (30.3-68.7) (32.3-54.4) Quality-assured ACT 100.0 0.0 75.6 0.0 17.2 14.8 25.5 22.3 24.8 - - (22.5-97.1) - (1.5-74.4) (1.4-68.8) (13.3-43.3) (11.9-37.9) (15.6-37.0) Non-quality assured ACT 0.0 51.1 12.4 0.0 0.0 0.0 7.6 27.1 18.2 - (5.6-94.9) (1.1-65.4) - - - (2.8-18.8) (13.1-47.9) (10.1-30.7) Non-artemisinin therapy * 0.0 49.0 11.9 0.0 0.0 0.0 11.8 9.8 10.6 - (5.1-94.4) (1.0-64.2) - - - (4.2-29.2) (4.4-20.6) (6.0-17.9) Artemisinin monotherapy ** 100.0 49.0 87.6 100.0 0.0 13.8 3.7 0.2 1.8 - (5.1-94.4) (34.6-98.9) - - (1.7-59.3) (0.9-14.0) (<0.1-1.6) (0.5-6.2) Antibiotic 100.0 49.0 87.6 100.0 16.1 27.7 38.3 16.7 25.9 - (5.1-94.4) (34.6-98.9) - (1.4-72.6) (2.7-84.1) (26.2-52.1) (8.1-31.3) (17.6-36.3) Antipyretic 100.0 100.0 100.0 100.0 33.3 42.5 59.8 52.0 56.1 - - - - (4.0-85.7) (5.6-90.2) (41.6-75.7) (36.0-67.6) (40.9-70.2) www.actwatch.info Page 34

Table 5: Fever treatment by malaria test result, patient age and outlet type Malaria Test Positive Malaria Test Negative Not Tested Age 0-4 Age 5 and Above All Patients Age 0-4 Age 5 and Above All Patients Age 0-4 Age 5 and Above All Patients Proportion of patients who received: (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) (95 CI) Prescription for antimalarial 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 treatment to be filled elsewhere # - - - - - - - - - Drug Stores N=32 N=43 N=75 N=22 N=28 N=50 N=136 N=159 N=297 Any antimalarial 94.9 98.8 97.2 36.6 4.1 22.2 50.5 58.1 54.3 (80.0-98.9) (92.5-99.8) (89.6-99.3) (19.6-57.6) (1.0-15.2) (11.6-38.4) (39.7-61.3) (48.7-67.0) (46.5-62.0) Any ACT 80.9 75.0 77.5 28.9 4.1 17.9 43.1 48.0 45.5 (61.7-91.7) (60.0-85.7) (64.3-86.8) (14.0-50.4) (1.0-15.2) (8.6-33.8) (32.8-54.0) (38.7-57.5) (38.0-53.2) Quality-assured ACT 72.6 65.3 68.4 14.6 2.6 9.3 35.2 36.2 35.6 (54.1-85.7) (49.7-78.2) (55.4-79.0) (4.3-39.8) (0.4-15.5) (3.5-22.6) (24.5-47.6) (26.6-47.1) (27.5-44.7) Non-quality assured ACT 8.3 12.7 10.8 14.3 1.5 8.6 7.9 11.8 9.9 (2.3-25.9) (4.8-29.5) (5.1-21.5) (3.1-46.4) (0.2-9.9) (1.8-32.4) (2.8-20.3) (6.2-21.3) (5.3-17.7) Non-artemisinin therapy * 13.8 21.0 18.0 7.6 0.0 4.3 8.8 8.5 8.6 (4.7-34.3) (11.1-36.1) (10.3-29.6) (1.0-40.0) - (0.6-24.9) (4.9-15.5) (5.0-14.0) (5.7-12.9) Artemisinin monotherapy ** 7.0 7.0 7.0 0.0 0.0 0.0 1.6 1.7 1.6 (1.8-23.6) (1.7-25.0) (2.6-17.5) - - - (0.4-6.4) (0.2-10.5) (0.5-5.1) Antibiotic 35.5 35.1 35.3 56.3 76.8 65.4 24.7 23.7 24.4 (19.5-55.5) (21.3-52.0) (22.4-50.7) (24.0-84.1) (53.5-90.5) (40.4-84.0) (16.6-35.2) (16.2-33.2) (17.7-32.5) Antipyretic 85.8 90.2 88.4 81.5 58.9 71.5 67.8 66.8 67.4 (67.4-94.6) (76.9-96.2) (76.2-94.7) (54.2-94.3) (34.9-79.2) (54.5-84.0) (54.5-78.8) (58.1-74.5) (59.9-74.1) Prescription for antimalarial 0.0 0.0 0.0 0.0 0.0 0.0 0.8 0.0 0.4 treatment to be filled - (0.1, 2.7) - - - - - - (0.1-5.7) elsewhere # * Primarily quinine injections, tablets, syrups as well as SP tablets and a few chloroquine tablet treatments. ** Primarily artemether injections. # Patient received a prescription for an antimalarial, but did not receive an antimalarial at the outlet (exiting without treatment). Prescriptions were written for antimalarials, including injectable artemether or artesunate, artemether lumefantrine, and quinine. Source: ACTwatch Fever CM Survey, Uganda, 2015. www.actwatch.info Page 35

Table 6: Patient comprehension of malaria diagnostic testing Proportion of fever patients who correctly recall whether or not a malaria test was performed during the consultation* All outlets Private For-Profit Health Facilities Pharmacy Drug Store Proportion of fever patients who correctly recall the malaria test result (positive, negative), among patients who were tested during the consultation** All outlets Private For-Profit Health Facilities Pharmacy Drug Store Age 0-4 Age 5 and Above All Patients (95 CI) (95 CI) (95 CI) N=545 N=720 N=1,271 100.0 99.8 99.9 - (98.5-100.0) (99.2-100.0) N=275 N=351 N=629 100.0 100.0 100.0 - - - N=80 N=137 N=218 100.0 100.0 100.0 - - - N=190 N=232 N=424 100.0 99.6 99.8 - (97.1-99.9) (98.5-100.0) N=231 N=286 N=518 97.1 95.9 96.5 (91.6-99.0) (85.0-99.9) (91.3-98.6) N=175 N=209 N=385 97.3 98.5 98.0 (90.6-99.3) (93.5-99.6) (94.6-99.2) N=2 N=5 N=7 100.0 100.0 100.0 - - - N=54 N=72 N=126 96.6 91.1 93.5 (79.2-99.5) (61.2-98.5) (77.6-98.4) * Among 520 tested, 519 reported testing. Among 753 not tested, 751 reported no testing and 2 responses were missing. Cases with missing responses were excluded from this table. ** Among 266 with a positive test, 262 reported a positive test, 2 reported a negative test, and 2 responded with don t know test result. Among 250 with a negative test, 245 reported a negative test, 2 reported a positive test, 1 responded with don t know test result, and 2 responses were missing. Cases with missing responses were excluded from this table. Source: ACTwatch Fever CM Survey, Uganda, 2015. www.actwatch.info Page 36

Table 7: Factors associated with malaria blood testing Tested Odds Ratio Proportion of patients tested, by patient and provider/outlet factors All patients Type of outlet Pharmacy Drug Store Private For-Profit Facility Outlet location Rural area Urban area Patient sex Male Female Patient age 50 years or older 15-49 years 5-14 years 0-4 years Present at consultation Patient not present Patient present Reported fever symptoms to a provider during the consultation (prompted or unprompted) Did not report fever Reported fever Prior to the current visit for this illness: Sought treatment elsewhere for this illness Did not seek treatment elsewhere for this illness Received a malaria blood test Tested previously Not tested previously Received any medicine Treated previously N of Patients (95 CI) 1,273 44.4 (39.4-49.6) OR (95 CI) 219 1.5 ref (0.5-4.5) 424 29.7 27.1*** (23.7-36.5) (8.6-84.8) 630 63.0 108.7*** (56.2-69.2) (33.7-50.5) 627 41.1 ref (34.3-48.2) 646 48.5 1.4 (40.9-56.2) (0.9-2.0) 625 42.0 ref 647 47.0 1.2 (41.1-3.0) (0.9-1.6) 49 37.7 ref (23.5-54.5) 503 42.6 1.2 (35.1-50.4) (0.6-2.5) 170 54.7 2.0 (43.9-65.0) (0.8-4.8) 545 43.2 1.3 (37.2-49.5) (0.6-2.5) 309 0.0 ref - 964 59.8 -- (54.3-65.1) 114 18.5 ref (9.2-33.9) 1,159 46.5 3.8** (41.3-51.8) (1.6-8.9) 307 36.7 ref (28.6-45.7) 965 46.7 1.5 (40.6-53.0) (1.0-2.4) 127 20.3 ref (10.5-35.6) 1,142 46.7 3.4** (41.3-52.0) (1.6-7.5) 259 40.8 ref (31.9-50.2) www.actwatch.info Page 37

Table 7: Factors associated with malaria blood testing Not treated previously Received an antimalarial Treated previously Not treated previously Provider with recent malaria CM training # Untrained provider Trained provider Provider with correct first-line knowledge ## Provider without knowledge Provider with knowledge Provider who cites ACT as most effective ### Provider without knowledge Provider with knowledge Price of malaria testing ^ Median and above Below median price * p<0.05 ** p<0.01 ***p<0.001 Tested Odds Ratio 1,011 45.3 1.2 (39.5-51.3) (0.8-1.9) 131 40.7 ref (29.2-53.4) 1,139 44.9 1.2 (39.4-50.6) (0.7-2.1) 655 43.1 ref (36.6-49.9) 616 46.2 1.1 (40.1-52.3) (0.8-1.6) 194 33.6 ref (23.5-45.5) 1,079 46.6 1.7 (40.9-52.4) (1.0-3.0) 217 40.2 ref (30.9-50.4) 1,056 45.4 1.2 (40.0-50.9) (0.8-1.9) 838 44.3 ref (36.8-52.0) 414 45.2 1.0 (37.9-52.8) (0.7-1.6) # Provider reported receiving training, including pre-service training or stand-alone workshops, on malaria diagnosis/malaria RDTs/malaria microscopy and/or treatment on the national malaria treatment guidelines. ## Provider correctly cited the national first-line treatment for uncomplicated malaria (AL). ### Provider named ACT as the most effective treatment for malaria in children younger than five and adults. ^ Price for the least expensive testing option for an adult (RDT or microscopy) available at the outlet on the day of the survey. A median was calculated across all outlets, and outlets were classified as providing testing that was at median cost or above, or below the median. Price for testing was not available for 21 patients. Source: ACTwatch Fever CM Survey, Uganda, 2015. www.actwatch.info Page 38

Table 8: Factors associated with QA ACT treatment, among fever patients with confirmed malaria Treated with QA ACT Odds Ratio Proportion of patients treated with QA ACT among confirmed cases, by patient and provider/outlet factors All patients Type of outlet Private For-Profit Facility Drug Store Pharmacy Outlet location Urban area Rural area Patient sex Female Male Patient age+ 0-4 years 5-14 years 15-49 years 50 years or older Reported fever symptoms to a provider during the consultation (prompted or unprompted) Reported fever Did not report fever Prior to the current visit for this illness: Sought treatment elsewhere for this illness Did not seek treatment elsewhere for this illness Received a malaria blood test Tested previously Not tested previously Received any medicine Treated previously Not treated previously N of Patients (95 CI) 266 48.5 (41.3-55.7) 188 35.9 ref (28.7-43.8) 75 68.4 3.9*** (55.4-79.0) (2.0-7.4) 3 75.6 5.5 (22.5-97.1) (0.5-56.6) 132 47.0 ref (37.0-57.4) 134 50.0 1.1 (39.4-60.7) (0.6-2.0) 137 45.2 ref (35.5-55.2) 129 51.6 1.3 (42.1-60.9) (0.8-2.2) 124 48.3 ref (36.7-60.1) 50 49.7 1.1 (34.4-65.0) (0.5-2.3) 84 48.9 1.0 (36.5-61.4) (0.5-2.0) 7 62.1 1.7 (29.3-86.6) (0.4-7.6) 260 47.7 ref 6 66.2 2.1 (24.4-92.2) (0.4-12.4) 50 33.0 ref (17.2-53.9) 215 51.7 2.2 (43.2-60.1) (0.8-5.7) 8 46.0 ref (10.1-86.6) 257 48.4 1.1 (41.3-55.7) (0.1-8.3) 48 33.5 ref (17.6-54.3) 217 51.5 2.1 (43.1-59.8) (0.8-5.5) www.actwatch.info Page 39

Table 8: Factors associated with QA ACT treatment, among fever patients with confirmed malaria Treated with QA ACT Odds Ratio Received an antimalarial Treated previously Not treated previously Provider with recent malaria CM training # Trained provider Untrained provider Provider with correct first-line knowledge ## Provider with knowledge Provider without knowledge Provider who cites ACT as most effective ### Provider without knowledge Provider with knowledge Cost of QA ACT treatment ^ Median and above Below median price * p<0.05 ** p<0.01 ***p<0.001 29 35.4 ref (18.2-57.4) 236 50.0 1.8 (42.6-57.5) (0.7-4.6) 126 50.9 ref (40.0-61.6) 140 46.5 1.2 (35.0-58.3) (0.6-2.4) 234 48.9 ref (40.7-57.1) 32 45.0 1.2 (24.9-66.9) (0.4-3.3) 45 41.1 ref (27.2-56.5) 221 49.8 1.4 (41.9-57.8) (0.7-2.9) 132 43.8 ref (34.0-54.1) 103 56.1 1.6* (47.0-64.8) (1.0-2.7) # Provider reported receiving training, including pre-service training or stand-alone workshops, on malaria diagnosis/malaria RDTs/malaria microscopy and/or treatment on the national malaria treatment guidelines ## Provider correctly cited the national first-line treatment for uncomplicated malaria (AL) ### Provider named ACT as the most effective treatment for malaria in children under five and adults ^ Price for the least expensive QA AETD available at the outlet on the day of the survey. Price was calculated for one adult equivalent treatment dose, or price for the amount needed to treat a 60kg adult. A median was calculated across all outlets and outlets were classified as providing QA ACT that was median cost or above, or below median. Price for QA ACT was not available for 31 patients. Source: ACTwatch Fever CM Survey, Uganda, 2015. www.actwatch.info Page 40

Annex 1: ACTwatch Background ACTwatch is a multi-country research project implemented by PSI (www.psi.org). Standardized tools and approaches are employed to provide comparable data across countries and over time. Project countries include: Benin, Cambodia, the Democratic Republic of Congo, Kenya, Laos, Madagascar, Myanmar, Nigeria, Tanzania (currently mainland only, previous work in Zanzibar), Thailand, Uganda, Vietnam, Zambia. The project was launched in 2008 with funding from the Bill and Melinda Gates Foundation (BMGF), and is currently funded through 2016 by the BMGF, UNITAID, and DFID. ACTwatch is designed to provide timely, relevant, and high quality antimalarial market evidence. 1 The goal of providing this market evidence is to inform and monitor national and global policy, strategy, and funding decisions for improving malaria case management. ACTwatch is monitoring antimalarial markets in the context of policy shifts and investments in the scale-up of first-line ACT and blood testing using RDTs. This has included adaptation of project methods for the evaluation of the Affordable Medicines Facility malaria (AMFm) pilot. 2 Project scale-up in the Greater Mekong sub-region (GMS) in 2015 was designed to deliver key indicators for informing and monitoring strategies and policies for malaria elimination. The project implements a set of research tools designed to: 1) Provide a picture of the total market for malaria case management, including all providers carrying antimalarials and RDTs and providing case management services, the relative antimalarial market share for each provider type, the antimalarial supply chain, and price markups within the supply chain for antimalarials and RDTs. 2) Monitor the readiness of market components for appropriate malaria case management, including availability of antimalarials and malaria blood testing; consumer price of antimalarial treatment and malaria blood testing; and provider qualifications, training, and knowledge. 3) Monitor the performance of market components for appropriate malaria case management, including the relative market share for quality-assured ACT relative to other antimalarial medicines; the demand for appropriate malaria case management captured through consumer knowledge, attitudes, and fever treatment-seeking behavior; and the quality of provider service delivery measured against national policies, guidelines, and minimum standards. ACTwatch research tools for malaria market monitoring include: 1. Outlet surveys Outlet surveys entail collecting quantitative data from all outlets and providers with the potential to sell or distribute antimalarials and/or provide malaria blood testing. These include health facilities, community health workers, pharmacies, drug stores, retail outlets, market stalls, and mobile providers. A screening process identifies outlets that provide antimalarials and/or malaria blood testing. Among these eligible outlets, service providers are interviewed and all antimalarials and RDTs are audited. The audit collects information about each antimalarial and RDT in stock (e.g. brand name, drug active ingredients and strengths, manufacturer, etc.) and retailer reports on consumer price and sale/distribution volumes for each product. A representative sample of outlets is identified within target study domains such that findings from the outlet survey provide estimates of antimalarial and RDT availability, price, and relative market share across the entire market, as well as within key market channels. 3 1 Shewchuk T, O Connell KA, Goodman C, Hanson K, Chapman S, Chavasse D. 2011. The ACTwatch project: methods to describe anti-malarial markets in seven countries. Malaria Journal, 10: 325. 2 AMFm Independent Evaluation Team. 2012. Independent evaluation of Phase 1 of the Affordable Medicines Facility malaria (AMFm), multicountry independent evaluation report: final report. Calverton, MD and London: ICF International and London School of Hygiene and Tropical Medicine. 3 O Connell KA, Poyer S, Solomon T, et al. 2013. Methods for implementing a medicine outlet survey: lessons from the anti-malarial market. Malaria Journal, 12: 52. www.actwatch.info Page 41

From 2008 through 2014, ACTwatch conducted 35 national outlet surveys across the 10 project countries. 4 Reports are available at www.actwatch.info, and peer-reviewed publications have appeared in Malaria Journal and The Lancet. 5 2. Supply chain studies Supply chain studies employ quantitative and qualitative research methods to effectively map the antimalarial supply chain in a given country. The supply chain is mapped from the antimalarial outlets (service delivery points) identified during an outlet survey to national importers and distributors with identification of all mid-level distributers in between. Retail prices are documented along the supply chain to facilitate calculation of commodity mark-ups. From 2008 through 2012, ACTwatch conducted 8 national supply chain studies. Reports are available at www.actwatch.info, and a peer-reviewed publication has appeared in PLoS One. 6 3. Population-based surveys Population-based surveys are conducted among consumers to document fever treatment-seeking behavior. A representative sample of the target population (caregivers of children and/or adults according to burden and risk) is identified, and a screening tool is used to identify individuals who have recently experienced fever. The surveys investigate the extent to which health care was sought, as well as common sources of care received. Respondent reports of malaria blood testing and antimalarials acquired are documented and summarized. The survey includes measures of demographic and other individual, household/family, and community characteristics that can be used to develop consumer profiles and monitor equity in access to malaria case management. From 2008 through 2012, ACTwatch conducted 14 household surveys focused on fever treatment-seeking behavior. Reports are available at www.actwatch.info, and a peer-reviewed publication has appeared in Malaria Journal. 7 4. Fever case management quality of care Fever case management quality of care is monitored using a set of research tools designed to measure aspects of the interaction between providers and clients. ACTwatch launched fever case management quality of care studies in 2015 in a subset of project countries. The following research tools were integrated into the outlet surveys in Cambodia and Uganda and were implemented among private-sector outlets providing malaria testing and treatment: Exit interviews conducted with target consumers immediately after receiving fever case management services in the private sector. A structured interview documented client reports about key aspects of service delivery, including malaria blood testing, test results, medicines recommended/prescribed and obtained, counseling, and costs of services and commodities received. Exit interviews were also used to measure client recall and comprehension of provider counseling, including instructions for completing prescribed drug regimens and client satisfaction with services provided. A consultation observation checklist was used to document aspects of the provider-client interaction in the private sector. A trained observer completed the checklist designed to document provider compliance with standard practice and procedures as well as aspects of client demand for specific products or services. The observer remained silent during the consultation. 4 Surveys in the DRC (2) and Myanmar (3) were sub-national. 5 O Connell K, Gatakaa H, Poyer S, et al. 2011. Got ACTs? Availability, price, market share and provider knowledge of anti-malarial medicines in public and private sector outlets in six malaria-endemic countries. Malaria Journal, 10: 326. Tougher S, the ACTwatch Group, Ye Y, et al. 2013. Effect of the Affordable Medicines Facility malaria (AMFm) on the availability, price, and market share of quality-assured artemisinin-based combination therapies in seven countries: a before-and-after analysis of outlet survey data. Lancet, 380: 1916-26. 6 Palafox B, Patouillard E, Tougher S, et al. 2014. Understanding private sector antimalarial distribution chains: a cross-sectional mixed methods study in six malaria-endemic countries. PLoS One, 9(4). 7 Littrell M, Gatakaa H, Evance I, et al. (2011). Monitoring fever treatment behavior and equitable access to effective medicines in the context of initiatives to improve ACT access: baseline results and implications for programming in six African countries. Malaria Journal, 10: 327. www.actwatch.info Page 42

ACTwatch in Uganda ACTwatch baseline surveys were conducted in Uganda in 2008 and 2009, followed by baseline and endline surveys for AMFm in 2010 and 2011, respectively. Additional surveys were conducted in 2013 and 2015. All reports are available at www.actwatch.info. www.actwatch.info Page 43

Annex 2: Country Background Uganda is a landlocked country in East Africa bordered by Kenya, South Sudan, Rwanda, Tanzania, and the Democratic Republic of the Congo. Uganda has an estimated population of 36.6 million people, the majority of who live in rural areas 8. Uganda has made substantial progress in recent years toward attaining 2015 Millennium Development Goals (MDG). This progress has included early attainment of poverty-reduction targets, and strong progress toward targets for reducing hunger and under-nutrition and child morbidity and mortality 9. Despite incredible progress, mortality rates for children under five (69 per 1,000 live births) and infants (45 per 1,000 live births) remain high 10. Administratively, Uganda is divided into 112 districts. Districts are further sub-divided into sub-counties and parishes. Over time, the numbers of districts and lower-level administrative units have increased with the aim of making administration and delivery of social services easier and closer to the people 11. Since 1986, the government has acted to rehabilitate and stabilize the economy by undertaking currency reform, raising producer prices on export crops, increasing prices of petroleum products, and improving civil service wages. The policy changes are especially aimed at dampening inflation and boosting production and export earnings. Since 1990, economic reforms ushered in an era of solid economic growth based on continued investment in infrastructure, improved incentives for production and exports, lower inflation, better domestic security, and the return of exiled Indian-Ugandan entrepreneurs 12. From the mid-1980s to the 1990s, Gross Domestic Product (GDP) growth in Uganda averaged about 5.3 percent while the population only grew at about 3.2 percent annually. In the 2000s to 2014, GDP has increased to an average of 6.6 percent while average annual population growth has remained steady at around 3.3 percent. Additionally, per capita income increased from 180 USD in 1984 to 670 USD in 2014 13. Healthcare system The provision of health services in Uganda is decentralized, with districts and health sub-districts playing a key role in the delivery and management of health services at those levels. The government eliminated user fees in 2001, and services in public health facilities are thus free. User fees remain in place in private wings of public hospitals 14. The public sector is organized into the following health services: 1. Village Health Teams (VHTs) (Health Center I) Provide services at the community level. They facilitate and are responsible for health promotion, community mobilization to improve health-seeking behaviors, and treatment of diseases. After the adoption of integrated community case management in 2010, two out of five VHTs provide diagnosis and treatment. 2. Health Center II Provide diagnosis and maternity care and are the first referral point in the sub-county. 3. Health Center III In addition to providing basic preventative and curative care, Health Center IIIs provide support and supervision to Health Center IIs. 4. Health Center IV or General Hospitals These facilities provide oversight of all the other lower-level health facilities. General Hospitals often provide pre-service training and services related to prevention of diseases, blood transfusion, and medical imaging. 5. Regional Referral Hospitals These are often research and teaching hospitals and are involved in specialized services, such as medical imaging, psychiatry, and surgery. 8 Uganda Bureau of Statistics. (2014). 2014 Statistical abstract. Available: www.ubos.org/onlinefiles/uploads/ubos/statistical_abstracts/statistical_abstract_2014.pdf. Accessed October 22, 2014. 9 Ministry of Finance, Planning and Economic Development (2013). Millennium Development Goals report for Uganda 2013. Kampala: Ministry of Finance, Planning and Economic Development. 10 UNICEF. (2013). Uganda statistics. Available: www.unicef.org/infobycountry/uganda_statistics.html. Accessed October 22, 2014. 11 Ministry of Health. (2010). Health sector strategic & investment plan. Kampala: Uganda Ministry of Health. 12 CIA World Fact Book (2015). Africa: Uganda. Accessed, December 12, 2015. https://www.cia.gov/library/publications/the-world-factbook/geos/ug.html 13 World Bank (2015). Uganda Databank. Accessed, December 12, 2015. http://databank.worldbank.org/data/reports.aspx?source=2&country=uga&series=&period=# 14 Ibid. www.actwatch.info Page 44

6. National Referral Hospital These are often research and teaching hospitals, and are involved in provision of specialized services. VHTs became part of Uganda s national health strategy in 2001. VHT volunteers serve as a community-based primary health contact with responsibilities, including identifying community health needs; mobilizing resources; mobilizing community participation for public health campaigns (e.g. immunization, sanitation, health-seeking behavior, malaria control); birth and death registration; and community-based management of common childhood illnesses, including malaria, diarrhea, and pneumonia. Four or five VHT volunteers are selected per village. The target of the 2010-2015 Health Sector Strategic and Investment Plan is to increase VHT coverage to 100 percent by 2015. As of 2009, 31 percent of districts had trained VHTs in all villages 15. Uganda has a total of 155 hospitals there are two referral hospitals, 14 regional referral hospitals, and 139 general hospitals. Out of the 155 hospitals countrywide, 27 are privately owned 16. The medicines distribution system in the public sector is centralized with procurement, pooled at the national level and organized through the National Medical Stores, an agency of the Ministry of Health. Nearly all public-sector procurements are imported (94 percent) 17. The private not-for-profit sector is also important in Uganda for medicines delivery and treatment. These outlets include mission/faith-based hospitals and clinics. Faith-based/mission not-for-profit facilities account for 41 percent of hospitals and 22 percent of lower-level facilities complementing government facilities, especially in rural areas 18. The private sector The private sector plays a significant role in the health system in Uganda and is estimated to provide about half of health services in the country 19. The 2012 ACTwatch household survey in Uganda found that more than half of children with fever were taken to a private-sector outlet for treatment 20. Private-sector outlets in Uganda include for-profit hospitals and clinics. Health professionals working in these facilities commonly have an affiliation with government/public health services. About half of doctors working in the private sector also work in the government sector; however, more than 90 percent of the nurses, midwives, and nursing assistants working in the private sector work full-time in the private sector. Private for-profit facilities have a large urban and peri-urban presence and provide primary and secondary care. More than 90 percent of private for-profit facilities offer malaria treatment 21. The private for-profit sector in Uganda also includes pharmacies and drug stores (licensed and unlicensed). The pharmaceutical sector in Uganda is regulated by the National Drug Authority (NDA). Drug stores and pharmacies are licensed by the NDA, and NDA licensing fees for these outlets recently increased significantly 22. Pharmacies must be staffed by a registered pharmacist and may dispense prescription-only and over-the-counter (OTC) medicines. Licensed drug stores must be staffed by a provider with a health qualification (e.g. pharmacy technician, registered/enrolled nurse, clinical officer, etc.). Drug stores are permitted to retail a range of OTC medicines but are not authorized to sell prescription-only medicines. ACTs were classified as OTC medicines in 2008 23. The NDA monitors private outlets for the illegal sale of medicines, including medicine sales by unlicensed drug stores and sale of government-procured medicines and supplies found outside of public health facilities. The crackdown on outlets illegally selling medicines has been a topic frequently covered by national media 24. 15 Ibid. 16 Ministry of Health (2015). Affiliated Institutions: Hospitals. Accessed, November 2, 2015. http://health.go.ug/affiliatedinstitutions/hospitals?page=1 17 Palafox B, Patoullard E, Tougher S, et al. (2012). ACTwatch 2009 supply chain survey results, Uganda. Nairobi: ACTwatch, PSI. 18 Ministry of Health. (2010). Health sector strategic & investment plan. Kampala: Uganda Ministry of Health. 19 Ibid. 20 ACTwatch Group and PACE. (2013). Household survey, Uganda, 2012 survey report. Washington DC: PSI. 21 Ministry of Health. (2010). Health sector strategic & investment plan. Kampala: Uganda Ministry of Health. 22 NDA doubles license fees for drug shop, pharmacies. (2013). Available: www.newvision.co.ug/news/647578-nda-license-fees-for-drug-shoppharmacies.html. Accessed: October 22, 2014. 23 Palafox B, Patoullard E, Tougher S, et al. (2012). ACTwatch 2009 supply chain survey results, Uganda. Nairobi: ACTwatch, PSI. 24 Personal communication, 2013 ACTwatch Key Informant Interviews, Uganda. www.actwatch.info Page 45

There are several local drug manufacturers in Uganda, including manufacturers of antimalarial medicines. In 2010, the Kampala manufacturing site for Quality Chemicals Industries Limited became Africa s first local supplier to receive WHO pre-qualification to produce ACTs. The company is licensed to produce artemether lumefantrine through the company Cipla Limited (www.cipla.com), based in India. Although local manufacturers are present in Uganda, the majority (~90 percent) of pharmaceuticals are imported from Asian, Western, and other African countries (Kenya, South Africa). Results of a mapping exercise undertaken in 2008 estimated that 93-95 percent of imported medicines in Uganda are generic products 25. Malaria risk and burden The climate of Uganda is tropical and tempered by altitude. In most parts of Uganda, temperature and rainfall allow intense perennial malaria transmission. Malaria is highly endemic across 95 percent of the country, affecting approximately 90 percent of the population (see national control program map in Figure X1). These areas include the entire Central region and the majority of the Northern and Southern regions. Particularly high transmission rates have been documented in some areas of northern Uganda, where prior to indoor residual spray campaigns, entomological inoculation rates were documented as high as 1,600 infective bites per person per year. The 5 percent of the country (10 percent of the population) that does not have endemic malaria consists of unstable and epidemic-prone transmission areas in the highlands of the South and Mid-west, along the eastern border with Kenya, and the northeastern border with Sudan 26. Malaria continues to be a major public health problem in Uganda and is the leading cause of morbidity and mortality among children younger than age five 27. The most recent national Malaria Indicator Survey (MIS) indicated that malaria was responsible for 30 to 50 percent of outpatient hospital visits, and 9 to 14 percent of inpatient deaths in 2014 28. The latest MIS report also documented a parasite prevalence of 19 percent among all children younger than five when testing with microscopy. Most malaria infections in Uganda are due to Plasmodium falciparum 29. Figure X1. Map of malaria endemicity Source: PMI FY2014 Malaria Operational Plan. Malaria case management guidelines The revised Uganda Malaria Reduction Strategic Plan (UMRSP) 2014-2020 primary targets are to reduce malaria deaths to near zero, reduce malaria morbidity to 30 cases per 1,000 persons, and to reduce malaria parasite 25 UMOH (2008). Cited in Medicines Transparency (medicinestransparency.org): The role of local manufacturers in improving access to essential medicine (2010). Accessed May 6, 2016: http://www.medicinestransparency.org/fileadmin/uploads/documents/meta- Uganda_AfricaHealth.pdf 26 Uganda Bureau of Statistics (UBOS) and ICF Macro. (2010). Uganda malaria indicator survey 2009. Calverton, MD: UBOS and ICF Macro. 27 Uganda Bureau of Statistics. (2014). 2014 Statistical abstract. Available: www.ubos.org/onlinefiles/uploads/ubos/statistical_abstracts/statistical_abstract_2014.pdf. Accessed October 22, 2014. 28 Uganda Ministry of Health. (2015). Malaria indicator survey, 2014-2015. Retrieved from: https://dhsprogram.com/pubs/pdf/mis21/mis21.pdf 29 Uganda Bureau of Statistics (2016). The National Population and Housing Census 2014 Main Report, Kampala, Uganda Ibid. www.actwatch.info Page 46