Title: Quality Improvement in Uzbekistan: a Case Study of Three Rural Clinics

Quality improvement in Uzbekistan

Title Page

Title: Quality improvement in Uzbekistan: A case study of three rural clinics

Running Head: Quality improvement in Uzbekistan

Abstract

Title: Quality improvement in Uzbekistan: A case study of three rural clinics

Author: Blake Tyler McGee, MPH candidate, Heilbrunn Department of Population and Family Health, Columbia University, New York, USA

Preceptors: Nilufar Rakhmanova, MD, MPH, ZdravPlus Project, Abt Associates, Inc., Tashkent, Uzbekistan; Peter Campbell, MBBS, ZdravPlus Project, Abt Associates, Inc., Tashkent, Uzbekistan

Running Head: Quality improvement in Uzbekistan

Word Count: 3,167 (main text); 250 (abstract)

Abstract:

Objective. To develop an in-depth narrative of the experience of three clinic-based teams in implementing quality improvement measures for hypertension care. Design. A descriptive, qualitative case study of the three teams. Sources of data included: archived program documents; semi-structured interviews with team members and other key informants; unstructured clinic observations; data from quality monitoring indicators. Setting. Three rural primary care clinics in Ferghana province, the central district hospital and the province health department. Participants. Interview respondents included physicians and nurses from the selected clinics as well as district and province health officials. Intervention. None. Main Outcome Measures. The problems identified in meeting standards of care for hypertension, the changes implemented to address those problems, the outcomes of these changes, and the perspectives of team members on current and future quality improvement activities. Results. Quality improvement teams contributed to the development of standards and began measuring indicators of compliance with those standards at their own clinics on a monthly basis. With technical assistance, teams identified causes of limited compliance with the standards and implemented changes to address these problems. Teams have documented improvements in most of the quality indicators and have participated in efforts to disseminate quality improvement principles and methods to clinics beyond their district. Conclusions. The teams described in this study exemplify the successful introduction of quality improvement measures into a post-Soviet health care system. Although clinicians show ownership over day-to-day quality monitoring tasks, independent initiative to analyze problems and pilot solutions remains undemonstrated. Moreover, clinicians seeking to improve the quality of care would benefit from further governmental support.

Introduction

Since 1998, the Republic of Uzbekistan has been implementing President’s Order No. 2107 “State Health Care Reform Program of the Republic of Uzbekistan.”This national health care reform initiative was officially accepted through a memorandum of understanding between USAID and the Uzbek Ministry of Health. The USAID-funded ZdravPlus project—implemented by Abt Associates, Inc.—has been providing technical assistance in executing the reform since 1999 in collaboration with the World Bank and Asian Development Bank.

A fundamental objective of the reform has always been to improve the quality of care. To that end, ZdravPlus has carried out a host of activities aimed at improving health care quality, including “re-training” for general practitioners (GPs), establishment of the Evidence-Based Medicine Center at the Tashkent Institute for Advanced Medical Education, and support for the development of national clinical practice guidelines (CPGs). Starting in 2003, ZdravPlus introduced modern quality improvement (QI) principles and methods through the launch of pilot projects in Ferghana oblast (province).

Using health statistics and personal experience, leaders at Ferghana Oblast Health Department (OHD), with input from primary care providers, identified three major health issues for the Ferghana region: iron-deficiency anemia in women of reproductive age, common illnesses in children under five addressed by IMCI guidelines [1], and arterial hypertension in adults over 18. The selection of these topics included consideration of epidemiologic significance, social importance, and the feasibility of achieving clinical improvements.

The experience of the pilot programs in Uzbekistan offers an excellent opportunity to study how QI principles and methods can be introduced into a post-Soviet health care system with success. As such, the primary objective of this study was to develop a coherent, in-depth narrative of the experience of three clinic-based teams in implementing QI measures for hypertension at their primary care facilities. Therefore, we decided on a descriptive, qualitative case-study design in order to produce a detailed account that would be useful to others seeking to improve health care quality in ex-Communist or developing world settings.

Methodology

The triangulation of evidence from a variety of sources has been identified as a way of assuring the strength of case-study research [2]. Therefore, we utilized multiple sources of information in developing our narrative.

Archived internal program documents at ZdravPlus country headquarters in Tashkent comprised an important source of data and included: monthly trip reports from site visits in Ferghana; quarterly QI bulletins disseminated to stakeholders; didactic materials from in-country QI seminars and trainings; an unpublished, qualitative study on access to and use of prescription drugs in Ferghana [3]; summary reports to donors, etc. These documents laid the groundwork for this narrative, and they informed the development of the interview guides.

Key informants were identified for individual interviews based on their level of involvement with the selected team. The chief physician of each clinic was interviewed, then he or she was asked to refer a nurse who was involved with QI work. Two officials at Ferghana OHD were identified by their involvement with the hypertension pilot: the oblast chief and the head therapist (internist). The rayon (district) coordinator at the Central Rayon Hospital (CRH), who acts as a liaison between the clinics and the OHD, served as an additional informant. Interviews were conducted in Uzbek and English with the facilitation of an interpreter. No recording device was available at the time of the interviews. Interview guides were semi-structured, and interview notes were analyzed for themes and coded by hand.

The interviews and document review were augmented by unstructured clinic observations, undertaken at the time of the interviews. Additionally, the qualitative findings were supplemented by quantitative data on the five quality-of-care indicators for hypertension that were collected by the teams at each clinic as part of the QI project.

Results

Getting Started: Standards and Indicators

Early in 2003, ZdravPlus, in collaboration with Ferghana OHD and national health experts, developed a set of standards of care for arterial hypertension (AH). Ideally, standards should be based on clinical practice guidelines (CPGs), but in the absence of good evidence-based CPGs in the early stages, the QI project relied on common-sense standards along the continuum of care. Thus, the standards adopted cover screening, diagnosis, assessment of risk factors, and the prescription and outcome of treatment.

At the end of each month, QI teams at each clinic (or SVP) collect the data available for each standard. In the example of screening, teams count the number of patients entered in the tonometry journal(in which blood pressure measurements are recorded) over the last month, then divide this figure by the number of all patients aged 18 and older registered in the admission journalfor the same month. The resulting indicator is easily expressed as a percentage by multiplying the result by 100. In February 2003, QI teams started monitoring quality of care based on these standards and indicators. At the end of each month, the SVP chief or a designated nurse brings the monitoring sheets to the CRH, where a technician uploads the indicators to an MS Access database. (An example of a monitoring sheet appears in the Appendix.)

Analyzing Problems and Making Changes

At the SVPs in Zarkent and Birlik, the proportion of adult patients whose blood pressure was measured and recorded averaged well below 50% during the first quarter of 2003. With assistance from ZdravPlus, QI teams at each SVP began to analyze the reasons for this suboptimal performance. According to two of the physicians interviewed, only the chief physician at each clinic was routinely equipped with a blood pressure (BP) cuff, and only he or she knew how to use it reliably. Furthermore, according to the summary of a root-cause analysis workshop with pilot QI teams in 2003, some patients were coming to the SVP to see a specialist, such as an obstetrician, bypassing the GPs and registration nurses altogether.

Based on their understanding of the causes of the problem, QI teams acquired and equipped every clinician with a BP cuff, which—according to ZdravPlus—they were able to purchase using World Bank loan funds. In addition, QI teams had established (or enhanced the role of) the pre-physician exam room, where a nurse assesses the vital signs of each patient that comes to the clinic, including two BP measurements for every adult patient, which she records in a tonometry journal (or some other logbook). A tour of each clinic revealed that these journals are meticulously and regularly maintained.

In addition to screening, QI teams considered ways to improve the indicator for appropriate diagnosis of hypertension. Birlik and Zarkent again hovered below 50% in terms of the number of patients identified with AH that were diagnosed appropriately during the first quarter of 2003. One reason for this weak compliance with the standard was reluctance on the part of patients to return to the clinic within four weeks to have their blood pressure measured again, as reported at the root-cause analysis workshop.

In response, the nurse in the pre-physician exam room of each SVP now underlines any result that exceeds 140/90 mmHg in red ink, allowing for quick identification of those patients that have already had an elevated reading in the last four weeks. According to the chief physician at Birlik, the institutionalization of the new diagnostic criteria has given patients time to adjust to coming in for subsequent BP readings. Additionally, QI teams have equipped patronage nurses (who conduct regular home visits in post-Soviet countries) with sphygmomanometers. The chief physician at Zarkent reported that these nurses now follow up with patients with an elevated initial reading over subsequent weeks.

The third indicator initially measured what proportion of newly diagnosed AH patients receive four clinical tests that assess damage to target organs in order to evaluate the patient’s risk of a severe cardiovascular (CV) event. In the first quarter of 2003, just over half the new AH patients at Zarkent—and none at Birlik—had all four tests completed. The reasons for this shortfall included problems with clinical equipment, such as expired batteries in the ECG machines, as reported at the 2004 QI ‘Scaling Up’ conference. Another important factor, which was identified at the root-cause analysis workshop, was the lack of equipment to perform glucose analysis on site. This limitation meant that clinic staff had to refer AH patients to the CRH for blood glucose testing, an option that patients often found to be too expensive or time-consuming.

In all three cases, QI teams furnished their clinics with the necessary equipment to complete all four tests on site on a reliable basis. They have used their budgets, which have been partially de-centralized through parallel health financing reforms, to purchase essential items like ECG batteries and reagents for blood glucose testing. International lenders, especially the World Bank, have provided more expensive pieces of equipment such as centrifuges.

Teams utilized the self-monitoring aspect of the QI approach to address the fourth indicator, the percentage of AH patients for whom treatment is indicated that are prescribed appropriate treatment. QI teams implemented no other intervention to target this indicator beyond constant awareness of their own performance through monthly collection of data. As one head nurse explained, “When we do monitoring, we identify our [own] omissions and shortcomings.”

Despite these changes, continued monitoring revealed that improved practices did not immediately translate into enhanced patient outcomes, as expressed by the percentage of AH patients that achieve their target BP level (usually <140/90 mmHg) within three months of the start of treatment. In response, each SVP has conducted monthly “community conversations” with 30 or so representatives from each mahalla (neighborhood unit) in its enrollment area regarding priority health issues. As one head nurse put it, “We talk about the nature of disease and how to prevent the need for advanced treatment.” The Zarkent physician singled out “health awareness” as a key priority for QI today, arguing that it is difficult to improve quality when people do not adequately understand health issues. A summary of the changes implemented to improve compliance with the five standards of care appears in the Table.

Working in Teams

All the teams meet regularly, and two of the teams piggyback their monthly data collection activities onto their regular meetings. Although respondents did not clearly delineate how working in teams is different than how they used to work before, they spoke of the advantages of sharing ideas and discussing alternative views. One head nurse specifically mentioned the enhanced role of patronage nurses and lab technicians, because they are responsible for much of the data that form the basis of QI activities. Despite this collaboration, a decision-making hierarchy persists: although nurses may contribute ideas and participate in the discussion, the physicians have “the final word.”

Table. Changes implemented to improve compliance with standards of care for arterial hypertension (AH)

Standard of Care Adopted

/

Causes of Suboptimal Performance Identified

/

Changes to System of Care Implemented

1. A nurse or physician measures the blood pressure (BP) of all patients over 18 years of age at each contact with the clinic and records the result /

• Not all staff equipped with or skilled in use of BP cuff
• Some patients see specialist and bypass registration nurse and general practitioners

/

• All clinicians equipped with and trained in use of BP cuffs
• Pre-physician exam room established or enhanced
• Tonometry journals started

2. Patients are diagnosed with AH only if 2 double measurements of at least 140/90 mmHg are recorded over 2 separate visits within 4 weeks /

• Patients do not return for subsequent measurements
• No system for readily reviewing previous BP results from last 4 weeks

/

• Registration nurse underlines high BP results in red ink and informs the treating physician
• Patronage nurses follow up with patients at home

3. All newly diagnosed AH patients receive 4 clinical tests (blood glucose, ECG, fundus oculus, proteinuria) to assess cardiovascular risk factors /

• Missing or non-functioning equipment at clinic
• Patients lack time and money to go to the central hospital for tests

/

• New equipment purchased with clinic budget or donated by World Bank
• Self-monitoring of providers’ performance

4. Physician starts with prescribing oral tiazide diuretics and beta-blockers, if there are no contraindications /

• Physicians were used to viewing AH as an episodic illness and treating with one-time injectables
• Patients still demand injections

/

• Self-monitoring by providers through collection and analysis of data on indicators of their own performance

5. AH patients reach their target BP level (< 135/85 mmHg for diabetics and < 140/90 for most others) within 3 months of the start of treatment /

• Patients take medications only until they feel better or their BP is lowered
• Patients cannot afford to buy newer drugs and/or do not want to take them regularly

/

• No direct changes at first
• Later, ‘community conversations’ organized and importance of patient counseling emphasized

Seeing Results

QI teams documented improvement in all four of the process indicators. For example, the proportion of adult patients at Birlik screened for hypertension rose from a monthly average of 48% in the first quarter of 2003 to an average of 100% by the first quarter of 2005 (Figure 1). Similarly, the proportion of AH patients at Birlik that were diagnosed appropriately climbed from a monthly average of 49% over the first quarter of 2003 to 100% two years later (Figure 2). The clinics also showed improvements in the proportion of newly diagnosed AH patients assessed for risk of CV complications: at Zarkent, only 56% of patients received the four indicated clinical tests during the first quarter of 2003; by the first quarter of 2005, this figure had climbed to 90%. Lastly, all three clinics witnessed increases in the proportion of AH patients that were prescribed appropriate treatment: the monthly average at Zarkent went from 71% to 92% from the first half of 2003 to the first half of 2005, while the indicator at Qorgoncha rose from 36% to 100% during the same period (Figure 3).

Figure 1.Figure 2.

Figure 3.

Respondents highlighted results other than monitoring data. Nearly everyone spoke of the benefit of improved screening coverage for the prevention of CV complications. Both the rayon coordinator and the chief therapist claimed decreases in the number of patients admitted to the hospital with CV complications, such as myocardial infarction, but access to and analysis of these data were outside the scope of this study. The clinicians interviewed also cited changes in population attitudes as a result of QI activities, though these have yet to be systematically documented. According to the Qorgoncha physician, patients now come to his SVP to have their blood pressure checked, whereas before they came only if they felt symptoms.