Towards the third 90: improving viral load testing with a simple quality improvement program in health facilities in Malawi

Abstract

Background

Viral load (VL) scale-up efforts have largely focused on laboratory systems, with less attention on facility-level strengthening of staff who facilitate VL testing. To address this gap we implemented a quality improvement (QI) program at 13 health facilities in central and southern Malawi.

Methods

QI program tools focused on patient and provider VL knowledge and clarification of site-level roles and responsibilities, including the designation of a VL ‘focal person’ to oversee all VL activities. T-tests were used to compare differences in VL testing before (November 2016–April 2017) and after (May 2017–November 2017) the intervention.

Results

The mean number of VL tests performed significantly increased after implementation of the QI program. Overall there was a 164% increase in the mean number of routine VL tests performed per month (p<0.001). Increased VL testing was sustained during the 6 months of follow-up.

Conclusions

A simple QI program focused on improving VL knowledge among patients and providers, and clarifying staff roles at a facility level increased VL testing over a 6-month period. Further investigation is needed on whether this program can be scaled in different settings across sub-Saharan Africa and on the duration of follow-up required for sustained improvements in VL testing.

Introduction

In 2014, the Joint United Nations Program on human immunodeficiency virus (HIV)/acquired immune deficiency syndrome (AIDS) (UNAIDS) outlined the 90-90-90 goals in an effort to end the HIV epidemic by 2020. The goals include 90% of people living with HIV know their status, 90% of those who know their status are on antiretroviral therapy (ART) and 90% of all people on ART have a suppressed viral load (VL).¹ Immense scale-up efforts have resulted in 20.9 million people worldwide now receiving lifesaving ART.² However, in sub-Saharan Africa, nearly 55% of clients on ART do not have access to VL testing.^1,3 To achieve the final ‘90’, the World Health Organization (WHO) estimates that 30.5 million VL tests will be needed by 2020.¹

VL testing is critical to accurately identify treatment failure due to either poor adherence or antiretroviral resistance. Since resistance testing is not available in most resource-limited settings, VL results are used to identify individuals for adherence counseling and close follow-up. When VL remains high despite adherence counseling and improved adherence, clinicians can move the individual to second-line therapy.^1,4,5 The WHO recommends people living in low- and middle-income countries (LMICs) receive routine VL testing 6 months after ART initiation and every 12 months thereafter, using a threshold of ≥1000 copies/ml to define virological failure.⁶ Routine VL monitoring can help with judicious utilization of second- and potentially third-line therapy,⁷ which has cost implications for LMICs, as these regimens are more expensive and often limited in supply.⁷ Additionally, early identification of virological failure and a rapid switch to second-line therapy can improve patient outcomes by preventing the emergence of additional drug-resistant mutations, which accumulate with longer exposure to a suboptimal regimen.^8,9

Malawi is one of the poorest countries in the world, with limited health infrastructure.¹⁰ Malawi has one of the highest rates of HIV in the world, with a prevalence of 9.4%.¹¹ VL scale-up efforts began in 2014,¹² with coverage improving from 6% to 11% between 2013 and 2015.¹² VL testing in Malawi is conducted at ART clinics, with samples transported to and processed at central laboratories. All government facilities utilize dried blood spot (DBS) specimens and rely on Riders for Health, a local nongovernmental organization (NGO), for transport of samples to central laboratories and transport of results back to facilities. Current clinical guidelines in Malawi recommend routine testing be completed 6 months and 2 y after ART initiation, and every 2 y thereafter, with >1000 copies/ml as the national definition of virological failure.¹³ Targeted VL tests are completed based on provider discretion and are most often drawn due to concerns of nonadherence and virological failure.

Key stakeholders and partners, including the WHO, the US President’s Emergency Plan for AIDS Relief (PEPFAR) and Médecins Sans Frontières (MSF), have developed laboratory tool kits to support external systems,^4,14,15 and community campaigns have been developed to boost patient demand for VL testing.^16,17 However, many countries, including Malawi, are struggling to take VL to scale. There is a need for simple quality improvement (QI) programs that address facility-level barriers while empowering providers to accurately utilize systems around VL. In this article we describe the implementation of a VL QI program piloted in Malawi and describe changes in VL testing at 13 health facilities before and after program implementation.

Materials and methods

Assessment and development of QI tools

Partners in Hope (PIH) in Lilongwe, Malawi is an NGO participating as a partner in the EQUIP consortium, a PEPFAR/U. Agency for International Development program supporting the 90-90-90 targets through clinical mentoring and systems support throughout Malawi. In April 2016, PIH-EQUIP performed a formative assessment in health facilities in central and southern Malawi to identify facility-level barriers to VL testing. Findings informed the development of VL tools that focused on patient and provider VL knowledge and clarification of site-level roles and responsibilities, including the designation of a VL ‘focal person’ to oversee all VL activities at the site. The focal person is an existing member of the facility staff who works as a HIV diagnostic assistant (HDA) and is an internally appointed position that is responsible for ensuring proper VL procedures and documentation are taking place. HDAs are an established Ministry of Health cadre that support HIV care in Malawi. They are required to complete Ministry of Health training for HIV testing and VL testing services.

VL QI tools include¹ standard operating procedures for VL testing,² VL educational materials for patients³ and a job aide for the focal person. All tools adhered to the Malawi Ministry of Health VL guidelines¹³ and were tailored for adults receiving HIV treatment. Final materials are described in Table 1 and can be found in Supplementary Appendix 1–3.

Table 1.

Description of tools included in the QI program

Tool	Description
VL standard operating procedures (SOPs) and flow chart	Includes detailed information on VL program management, including client preparation, client flow, sample collection, handling of results and assigning a focal person; provides diagrammatic illustrations taken from the Ministry of Health HIV treatment guidelines to support training and aid in procedures for VL sample collection, with specific attention on focal person involvement and documentation throughout the VL process.
Patient-focused educational materials	A standardized guideline for patient education, summarizing key information on VLs to be used by clinic staff when giving health talks to patients in the waiting area prior to seeing the clinician.
Resources for the VL focal person	Laminated job aid providing detailed descriptions of the role of the VL focal person and associated tasks. Space is provided to fill in names of appointed clinic staff to delineate roles and responsibilities for each component of the site’s VL program.

Training ART clinic staff

Orientation to the tools was conducted across 13 sites. Sites represented a range of health facilities in Malawi, including mission hospitals and government hospitals and health centers. Training was conducted by PIH-EQUIP clinical officers who mentored facility staff on appropriate use of the VL register and facilitated the appointment of the VL focal person for the site. All clinic staff, including clinical officers, nurses, HIV counselors, expert clients, data clerks, HDAs and lab representatives, attended the training. The total number of participants attending the orientations varied depending on the size of the facility, ranging from 15 to 40 participants. Training was held over 3–4 h on a single day during lunch and in the afternoon and did not interfere with health service delivery. Refreshments and lunch allowances were provided for those in attendance. Routine follow-up visits by PIH-EQUIP mentors were conducted on a monthly basis for the first 6 months following the training. These visits focused on supporting use of the QI tools and meeting with clinic staff to discuss areas of concern.

Data collection

Data on all adolescents and adults were collected from national VL registers from 13 sites participating in the QI program between 1 November 2016 and 30 November 2017, 6 months before and after QI program implementation. Convenience sampling was used to identify PIH-EQUIP-supported sites for inclusion. All sites had existing challenges with VL testing and documentation as identified in a baseline assessment and were selected because they supported the largest ART cohorts in the PIH-EQUIP program. Challenges included inadequate sample documentation in Ministry of Health registers, poor internal coordination for VL procedures and gaps in clinical knowledge among staff members responsible for testing. Paper-based VL registers were anonymized and digitized to collect data on routine VLs conducted (those due according to Malawi’s current VL guidelines, as defined above). We excluded VLs that were performed because of concern for nonadherence or virological failure (targeted VLs). Variables in the dataset include the facility name, date the VL sample was taken and sex of the client. Registers were reviewed monthly by PIH-EQUIP mentors for quality assurance and monitoring purposes.

We used t-tests to determine differences in the mean number of VL tests completed per month before and after QI program periods. Data were analyzed by site and then by facility type (mission hospital, government hospital or government health center). Ethical approval was granted by the Malawi National Health Sciences Research Committee and the University of California Los Angeles.

Results

Register review revealed 34 480 clients (64% female) who had a routine VL performed over the 13-month period. Table 2 describes the mean number of VL tests per site performed monthly before and after QI program implementation. The number of VL tests performed significantly increased across all but two sites after the implementation of the QI program. Overall there was a 164% increase in the mean number of routine VL tests performed per month (p<0.001). Similar improvements were seen when data were analyzed by facility type (Table 3). Improvements in routine VL testing were sustained over the 6 months of follow-up after the QI training (Figure 1).

Table 2.

Mean number of routine VL tests performed per month before and after implementation of the QI program

Facility	Mean number of routine VL tests conducted (per month)		p-Value
	Before intervention (November 2016–April 2017)	After intervention (May–November 2017)
Chikwawa District Hospital	163	399	0.001
Dowa District Hospital	83	142	0.110
Kalemba Community Hospital	98	372	<0.001
Kasungu District Hospital	465	322	0.021
Likuni Mission Hospital	248	225	0.756
Mponela Hospital	81	103	0.313
Ndamera Clinic	0	185	<0.001
Ngabu Rural Hospital	221	271	0.332
Nkhoma Mission Hospital	133	205	0.137
Nkhotakota District Hospital	118	252	<0.001
Nsanje District Hospital	193	213	0.506
St. Montfort Hospital	124	463	<0.001
Tengani Clinic	45	82	0.074
Mean	152	249	<0.001

Table 3.

Mean number of routine VL tests performed per month before and after the implementation of the QI program, by facility type

Facility type (n)	Mean number of VL tests conducted (per month)
	Before intervention (November 2016–April 2017), n (SD)	After intervention (May–November 2017), n (SD)	p-Value*
Government district hospitals (5)	204 (136)	266 (89)	0.019
Government rural hospitals (3)	133 (62)	249 (111)	0.003
Government health centers (2)	23 (23)	134 (52)	<0.001
Mission hospitals (3)	168 (56)	298 (117)	0.017

*p-Value calculated using t-tests.

SD: standard deviation.

Figure 1.

Mean number of routine VL tests performed across intervention facilities before and after implementation of the QI program.

Discussion

Our data show that a simple QI program for facility-based staff focused on VL testing can successfully increase appropriate utilization of VL monitoring. Although we did not track a large group of facilities where the intervention did not take place, we did evaluate three districts that are PIH-EQUIP supported but did not receive the QI intervention, and we found the volume of VLs remained the same or went down. In our QI sites, the majority had significant increases in the number of routine VLs performed, with one exception, Kasungu District Hospital, which experienced a statistically significant decrease through the course of the QI program. We could not identify a specific reason for the decline; however, the site did have a strong program at baseline, which may have limited its ability for significant gains. There was a drop in VL performance at all sites from September to October, as seen in Figure 1. Possible reasons for such fluctuations could include increases in guardian pickups of ART, changes in VL staff who were not yet trained on the QI tool by our mentoring teams and/or fluctuations in patient volume that occur seasonally due to agriculture and weather cycles. Sex differences are representative of those receiving HIV care in Malawi.

Our QI program was performed within a PEPFAR mentoring program with minimal increased cost and time to the facility. We attribute programmatic success to several factors. First, we observed that the appointment of a VL focal person contributed to improvements in VL coordination at the facility level. By delineating specific roles and responsibilities to the focal person, standard procedures and documentation could more readily be followed. In overburdened facility settings, task-shifting allows responsibilities to be more equally distributed.⁷ This may help optimize productivity where human resources are low and individuals are overextended, and it has cost implications, as the focal person was an internally designated existing staff person rather than an externally hired position.

Secondly, in-person training and the availability of job aids increased provider knowledge of VL monitoring, increasing the number of clients offered VL testing and subsequently VL coverage. Our findings support existing literature that provider training aimed to improve knowledge can increase provider confidence, contribute to feelings of empowerment in the workplace and have a positive impact on patient care.^18,19 Thirdly, patient-focused educational materials, such as the health talk guide, may have resulted in an increased demand for VL tests among clients. In Malawi, health talks are given to patients in the ART waiting area before seeing the clinician and are a primary means to disseminate information about a range of health topics. The QI program equipped clinic staff with the tools necessary to disseminate accurate information about VL, providing the opportunity for clients to understand VL monitoring and empowering them to talk with providers about whether they were due for a routine VL. Others have found that improved client knowledge and awareness can result in increased demand creation for a variety of health services, including VL monitoring.^20–22

Finally, monthly mentorship visits conducted by PIH-EQUIP clinical mentors ensured sustained adherence to VL guidelines and procedures. These visits consisted of meetings with the focal person to discuss challenges, review of the VL register for completeness, monitoring of the volume of VL tests and providing mentorship to the focal person and other site personnel, including clinicians and support staff. Regular follow-up visits provided necessary guidance and feedback to keep staff on task and ensure program quality. They also played a role in maintaining awareness of the VL program among clinic staff, including educating new staff. High rates of staff turnover, particularly in rural facilities, is a challenge in QI efforts and requires ongoing attention to new staff who require training and support.

Our QI program is a feasible strategy for facility-level health system strengthening and may be generalizable to other countries in the region that have similarly funded public HIV programs and experience challenges with staffing due to skilled worker shortages. Other tools have been developed to assist in VL scale-up efforts, including laboratory support, community campaigns and point-of-care systems;^14,16 however, specific site-level support tools have been underutilized and/or under-reported. In order to be optimally beneficial, our QI program needs to be combined with other initiatives to maximize impact across the VL cascade. While our curriculum and mentorship visits were designed to be complementary, further exploration will be needed to understand the extent to which each of these factors individually contributed to the success of the program and whether gains are sustained long term.

Limitations

Health facilities selected for program implementation were PIH-EQUIP (PEPFAR)-supported sites with existing support networks for HIV services, which are not representative of all health facilities in Malawi. In settings where mentor staff are not providing routine site visits, this type of QI program may not be feasible or may have diminished benefits due to a lack of ongoing support. Additionally, we do not yet have data on whether VL utilization is maintained after the 6 months of intensive support. Although the program saw success in initiating and implementing testing, it did not evaluate other elements of the VL cascade, such as laboratory processing, results reporting back to sites and timely reporting of results to clients. Success with each of these elements will be critical to improving patient outcomes and reaching the third 90.

Conclusions

A simple QI program focused on improving VL knowledge among patients and providers, and clarifying staff roles at a facility-level significantly increased VL testing over a 6-month period. Ongoing success will rely on support from funders, implementing partners and government to invest in the resources needed to optimize VL delivery.^14,18,23 This study supports other findings that suggest effective scale-up efforts will require a multipronged approach to strengthen the health care system. Further investigation is needed on whether this program can be scaled in different settings across sub-Saharan Africa, on the duration of follow-up required for sustained improvements in VL testing and on QI strategies for improving other elements of the VL cascade.

Authors’ contributions

GK, MN and JM developed the QI tools. JH, GK, JM and MN monitored tool implementation and mentorship activities. JH, KD and RH conceived the study. KB carried out data collection. RH, KB and AB provided analysis and interpretation of the data. JH drafted the manuscript. RH, KD and AB critically revised the manuscript. All authors read and approved the final manuscript.

Acknowledgments

We are grateful to the staff of PIH/EQUIP who facilitated the VL scale-up program and subsequent data collection. We are also grateful to the facility staff who welcomed the tools.

Funding

This work was supported by the US Agency for International Development and the President’s Emergency Plan for AIDS Relief under cooperative agreement AID-OAA-A-15-00070.

Competing interests

None declared.

Ethical approval

Ethical approval was granted by the Malawi National Health Sciences Research Committee and the University of California Los Angeles.