Abstract
This report is part of a research project, funded by Health System Research Institute (HSRI), for developing primary care and end-of-life care performance measuring systems in Thailand, including operational definitions and key components of a patient-centered model of primary care, a framework and a set of key performance indicators (KPIs), guidelines for KPI analyses and identification of opportunities for improvement. The study was conducted during October 2018 to December 2020, using mixed-methods methodology consisting of roundtable events and workshops with key stakeholders, including physicians, nurses and policy makers from the Ministry of Public Health and from the National Health Security Office, twelve rounds of focused-group discussion amongst service users from six participated pilot primary care providers, and a pilot application of the selected KPIs as a part of the value-based payment initiative of the National Health Security Office (NHSO), Region 13 Bangkok. As one of the key study conclusions, primary care is defined a type of health care services that a person, when having queries regarding their health, facing health problems or risks, suffering from illnesses, or in need of continuous follow-up and rehabilitation services refers to as his or her first contact and should pay regular visit to. It also includes any proactive health service which is provided on basis of participation by networks of providers and users. The conceptual framework of primary care performance comprises four groups of performance, which could be subcategorized into 13 key result areas. First, the outcome group comprises three areas, including health outcomes, patient acceptability, and cost. The process group consists of accessibility, effectiveness, continuity, safety, and patient-centeredness. The structural group covers coverage and availability. And, lastly, the other key system performance group includes efficiency, equity and value. The pilot KPIs for primary care include 49 KPIs, which are grouped into eight sets, as follows: (1) processes and outcomes for the healthy group, such as Index of women aged 15-49 who received five antenatal quality care visits, Effectiveness average score of children aged 5 years Immunization completeness by vaccine for each vaccine in the national schedule, Health promotion and disease prevention effectiveness index in adults; (2) processes and outcomes for the risk group, such as Effectiveness average score of health promotion and disease prevention in the elderly; (3) processes and outcomes for the chronic-care group, such as Average point of Diabetes effective care bundle score, Ability to control blood pressure of patients with hypertension, and Inappropriate drugs used index in patient with chronic kidney diseases; (4) processes and outcomes for the dependent groups, such as Average number of days to receive long-term care after screening; (5) processes and outcomes for the end-of-life group, such as Percentage of terminal cancer patients received quality of life assessment at least 10 days in the last 30 days of life; (6) overall primary care services, such as Average primary-care physician service hours per week per 10,000 population (PCPH per 10k pop); (7) patient experience in primary care, such as Patient Acceptability index of primary care services, and Patient-reported accessibility index of primary care services; and (8) Primary care value, which is Primary care value index. The KPI sets were found useful, and mostly feasible to apply in the real setting. However, there were number of problems and obstacles including (1) perception of staffs, (2) consistency between healthcare service systems, service processes and data-recording practices, (3) additional burdens of data collection, (4) many and fragmented databases and needs to link and integrate many databases to get an overview of the personal health care, health service utilization, and KPI results in the given time. In addition, there were needs for simple statistical analyses and presentation of KPI results, guidelines for root-cause analysis and the needs to develop staff competencies on these matters. Nevertheless, the piloted KPI initiative with the NHSO Bangkok-area office demonstrates possibility for implementing KPIs to improve primary care using Value-base healthcare concept. Based on the findings, some key policy recommendations include that the Primary Care System Committee and the Ministry of Public Health should collaborate with other agencies related to primary care development, such as HSRI, NHSO, the Thai Health Promotion Foundation, The Healthcare Accreditation Institute (Public Organization) and primary care providers in (1) using the proposed operational definition, framework, components and perspectives of primary care performance to review and improve policies, related law and regulations, and roadmaps for developing primary care systems to respond to population health needs given changing social contexts, (2) applying the proposed set of KPIs, reviewing and improving details and develop additional KPI as needed, and using at least four selected KPIs for long-term system monitoring—including Average primary-care physician service hours per week per 10,000 population (PCPH per 10k pop), Charlson Comorbidity Index for selected chronic conditions (CCI), Patient Acceptability index of primary care services, and Primary care value index—in order to manage the overall systems, to balance quality of care with resource efficiency, and to reduce disparities between areas, (3) urgently defining health data standards, data file structure and databases, and data-exchange standards to support primary care, (4) developing survey systems and patient experience questionnaires from the proposed prototypes, (5) upgrading management information systems, and (6) improving staff competencies to support data collection, analysis, review and improvement of primary care.
