What constitutes quality of learning in primary care? A social systems view of that central question regards the relationships between dimensions of learning, purposes of primary care, and quality of practice. The question of learning quality was approached in three ways. First, perceptions of learning quality were elicited through recorded interviews with fifteen participants representing diverse roles in a primary care medical clinic. Analysis of the interviews indicated learning sources, factors, and functional dimensions of learning. Second, because learning is constituted in a social practice, the social context of learning in primary care clinics was modeled with qualitative systems diagrams. This exposed systemic barriers and facilitators of learning in practice. Third, learning is directed toward fulfilling the purposes of primary care. The nature of those purposes is not well articulated. A framework of seven core purposes was developed from the perspective of systems phenomenology. This framework extends the biopsychosocial framework in several regards. Perceptions of learning quality, the structural situation of learning in clinical practice, and the core purposes of primary care, all contribute to a social systems understanding of what constitutes learning quality, and how primary care organizations might procure it and assess it. Systems phenomenology represents a significant innovation in social systems science methods. / Graduation date: 2002

A research report submitted to the Faculty of Management, University of the Witwatersrand, in partial fulfilment of the requirements for the degree of Master of Management / Rapid Appraisal has gained popularity amongst policy makers, and is used in strategic planning of primary health care services. This study aimed at determining whether Rapid Appraisal is an appropriate planning tool for primary health care services in South Africa. This study compares Rapid Appraisal with a Regional Health Management Information System (ReHMIS), using the Northern Province as a case study, In comparison, Rapid Appraisal took half the time of ReHMIS for data collection, and used less resources in the process, There is significant difference between the two data sets and this is explained by the fact that Rapid Appraisal does not only determine whether the facility is there or not but it also identifies management issues. Rapid Appraisal is an innovative method that engages all the relevant stakeholders in planning their primary health care services, The findings proved that Rapid Appraisal is an appropriate planning tool for primary health care services. / AC2017

Introduction Bihar ranks among the most socio-economically disadvantaged states in India, and its public health system had long suffered from structural deficiencies which contributed to poor health outcomes. In November 2013, the Bihar government, with funding from Gates Foundation and technical support from CARE India, launched the state-wide Bihar Technical Support Program (BTSP) – seeking to address gaps in infrastructure, supply chain, and human resources, as well as the quality of service delivery, so as to improve reproductive, maternal, newborn and child health (RMNCH) and nutrition service provision. BTSP adopted a two-pronged strategy – conducting (i) periodic comprehensive facility assessments (CFAs) to identify and address the structural gaps; and (ii) nurse-mentoring programs to develop competency among nursing cadres in providing basic and comprehensive emergency obstetric and newborn care (BEmONC/ CEmONC) services. Through three inter-linked papers, the dissertation aimed to conduct an evidence-based assessment of this health system strengthening program. “Facility readiness” (structural readiness of public health facilities) was operationalized in terms of infrastructure, essential supplies, and human resources, while “facility performance” was operationalized based on the direct observation of normal vaginal deliveries and newborn care (including management of immediate complications if needed) and infection prevention practices in the labor rooms. The first paper describes the evolution of BTSP, and examines the initial progress made in facility readiness between 2015 and 2016. The second paper: (i) conducts a comparative assessment of facility readiness between 2017 (at end of the first four years of BTSP) and 2019, and describes the continuation of progress or lack thereof; (ii) quantifies facility readiness through a scoring system that reflects the readiness to provide maternal and newborn care (MNC) services; and (3) compares the change in this score over time (2015, 2017 and 2019) across different districts and levels of health facilities in Bihar. Thus, the first and second papers together examine the extent to which Bihar’s public health facilities were structurally strengthened in terms of physical infrastructure, supplies and workforce by utilizing data from all four rounds of CFAs conducted till date. The third paper asks the next logical question in a health system strengthening process – was facility readiness positively and significantly associated with facility performance? This is an important query, as it aims to provide evidence of synergistic progress, as envisioned under BTSP. First, the paper examines whether the facility-level performance changed, by comparing baseline (May-December, 2018) and endline (October-December, 2019) assessment data from the nurse-mentoring program (locally called AMANAT Jyoti). Second, it assesses the association of facility readiness (based on CFA 2019 data) with endline facility performance in providing MNC services. Methods The first paper utilizes a structured, narrative review of scientific and grey literature to describe evolution of the BTSP since 2014, based on programmatic learnings through prior years (2011-2013) of collaborative vertical interventions. Subsequently, the paper measures the tangible change in select facility-level characteristics, utilizing quantitative data generated through two rounds of CFAs conducted by CARE India in 2015 (n=534 facilities) and 2016 (n=550 facilities). The second paper utilizes quantitative data generated through two rounds of CFAs conducted by CARE India in 2017 (n=550 facilities) and 2019 (n=552 facilities). Each CFAs covered all Level 2 (primary health centers) and Level 3 (higher-level facilities) public health facilities in Bihar that conducted at least 100 deliveries in the preceding year. Subsequently, the paper constructs a “facility-level MNC structural readiness score” – henceforth referred to as facility readiness score, based on a common set of indicators from CFA 2015, 2017 and 2019, to reflect human resources, infrastructure and essential supplies related to delivering MNC services. The paper uses this score to map the change at 2-year intervals, from 2015 to 2019, at both facility and district levels. The third paper utilizes quantitative data generated through two separate assessments conducted by CARE India – the 2019 CFA, and the 2018-2019 assessment of AMANAT Jyoti (nurse-mentoring program), which involved direct observation of normal vaginal deliveries, newborn care, and infection prevention practices in the labor rooms. The paper constructs baseline and endline facility-level MNC performance scores – henceforth referred to as facility performance scores based on data from AMANAT Jyoti assessments, and examines the association between endline facility performance and facility readiness scores. While descriptive statistics was used to present findings from the CFAs and AMANAT Jyoti assessments, paired t tests were used to test the mean change in scores over time and between the different levels of facilities. The association between endline facility performance and facility readiness scores was tested using simple as well as multiple linear and multinomial logistic regression modelling. Results With a demonstrated intent to improve the ailing public health sector, the Bihar government in 2010 forged a collaboration with Gates Foundation to accelerate progress across RMNCH and nutrition programs. Through the Integrated Family Health Initiative program (IFHI, 2011-2013), outreach-based and facility-based solutions were implemented in eight programmatically-prioritized districts to address the stated goals. However, over this period, it became apparent that long-term success of such initiatives remained critically dependent on strengthening the foundational components of Bihar’s public health system –physical infrastructure, supply chain for drugs, consumables and equipment, and the skilled health workforce. These programmatic learnings motivated a re-think and consequent state-wide launch of the BTSP – characterized by a novel structure of health governance that was deeply embedded within the public health system, and a robust information management system that could generate, analyze and disseminate data on community- and facility-level services to support decision making. The quantitative analyses of CFA data (in first and second papers) provided an assessment of the changes that happened at the level of health facilities, likely supported by the policy-level modifications. There was a clear sense of prioritization of the limited resources – with constant focus on structurally preparing health facilities to deliver basic MNC services, more so at Level 2 (primary health centers). By 2019, at least 99% facilities at either level provided 24x7 delivery services and had designated labor rooms, 97% had designated newborn care corners which were mostly located inside the labor rooms, 70% or more had at least one functional fetal doppler, baby weighing machine, radiant warmer, and AMBU bag with neonatal oxygen masks. The improvement in availability of essential supplies like oxytocin, misoprostol, magnesium sulphate, antibiotics, and reproductive health commodities (condoms, intrauterine contraceptive devices, sanitary napkins, iron-folic acid tablets, contraceptive pills) were particularly notable during the 2017 and 2019 CFAs. However, the supply chain variably faltered for a number of other essential supplies like oral rehydration solutions, functional oxygen cylinders, normal saline and ringer lactate solutions. The data revealed that facility-level inefficiencies in utilizing the electronic inventory management system to accurately reflect actual status of supplies within the facility, likely compromised procurement and distribution. With regards to human resources, while a large number of auxiliary and general nurse midwives were available for service during CFA 2019, the BTSP faced continuing challenges (2015-2019) in recruiting and/or retaining physicians, especially the specialist physician cadres. By CFA 2019, these structural changes were also supported by remarkable improvements in two related services areas –availability of emergency transport, and laboratory services. The comparison of facility readiness scores (second paper) based on CFA 2015, 2017 and 2019 showed that while the mean scores increased sharply for both Level 2 (increase=1.51 (95% confidence interval: 1.39, 1.63)) and Level 3 (1.39 (1.1, 1.69)) facilities between 2015 and 2017, the progress was less pronounced at both levels between 2017 and 2019. 25 of the 38 districts in Bihar demonstrated a continuous increase in mean scores over the 3 CFAs. As for the remaining 13 districts, their 2019 mean scores remained higher than that during 2015. The analysis of AMANAT Jyoti assessment data (third paper) revealed improvements across 36 (80%) of the 45 performance parameters assessed through direct observation of deliveries between the baseline and endline. However, at least 80% compliance was observed for only 11 of 45 (24%) assessed parameters at baseline, and 16 of 45 (36%) at endline. The mean facility performance score increased significantly among both types and levels of facilities – but the increase was higher among Level 3 (mean increase = 1.56, p=0.0005, n=13) and CEmONC (1.82, p=0.0029, n=9) facilities, than among Level 2 (0.32, p =0.0288, n=121) and BEmONC (0.33, p=0.0168, n=125) facilities. The regression analysis failed to identify any linear relationship between facility readiness and performance scores. However, a significant positive association was observed between facility readiness score and the middle tertile of endline facility performance score (vs. lowest tertile as reference) in multiple multinomial logistic regression modeling (n=132 facilities). With increasing facility readiness score, the odds of a facility being in the middle tertile of the endline facility performance score relative to the lowest tertile was 1.68 (95% CI = 1.02, 2.76), after controlling for baseline facility performance score, mean delivery volume, and the facility level. Conclusion The BTSP can be best described as a diagonal health system strengthening initiative –one that starts with a focus on specific programmatic (RMNCH) outcomes, but strives to achieve these through identifying and addressing bottlenecks across the health system. The efforts made to revamp health governance through creating structures for technical support from the state- to block-levels is particularly laudable, as is the remarkable capacity building in collecting and using facility-level data to inform programs and policies. The dissertation identified that BTSP has made appreciable progress in structurally preparing Bihar’s public health facilities to deliver basic MNC services – with improvements in related infrastructure, essential supplies, and supportive services like referral transport and laboratory facilities, as well as through recruitment of large number of ANM and GNM nurses. However, the process encountered a number of challenges, and it may be worthwhile to adopt a targeted approach to address some of these concerns. For example, it is important that the BTSP works to equip all facilities with electronic inventory management systems, while simultaneously training the personnel using such systems. To circumvent the chronic shortage of specialist physicians, a “task shifting” approach may help maximize utilization of existing health workforce to strengthen service delivery capacity. Further, the overall level of facility performance of MNC service delivery remained low at endline despite improvement from the baseline scores, and there was limited evidence of a significant positive association between facility readiness and performance scores. As these scores reflect the minimum essential requirements for a MNC service delivery setting, the BTSP clearly has challenges ahead. They must continue to address the persistent challenges in facility readiness and facility performance so that these two facility-level interventions will complement each other and influence outcomes. As the onus of this diagonal health system strengthening program incrementally shifts from development partners to the government, it will be important to recognize the significance and complexity of this effort.

背景：中國在發展城市社區衞生服務/基層醫療的過程中，基於當地經濟社會發展情況和特點，主要形成了政府辦政府管、院辦院管、以及私營三種模式。近年來中國學者對大陸社區衞生服務/基層醫療的研究主要關注于服務的基本現狀與功能的描述性研究，以及患者主觀層面的滿意度調查等，對城市社區衞生/基層醫療組織形式與服務模式建設的研究相對缺乏。初級衞生評價工具( Primary Care Assessment Tool, PCAT) 由美國約翰霍普金斯大學初級衞生保健政策中心開發，將初級醫療的五個核心方面進行量化，從患者對社區衞生服務/層醫療服務的客觀體驗角度出發，客觀評價醫療服務品質。本研究關注於基層醫療服務的過程層面，作為珠江三角洲六城市基層醫療研究項目的一部分，聯合廣州醫學院公共衛生與全科醫學學院，在廣東省衞生署以及香港智經研究中心的大力支持下，其他學者將研究基層醫療服務的結構和結果層面。 / 目的：本研究關注從病人體驗角度出發，在城市社區衞生/基層醫療服務的過程層面，評價和比較三種不同的社區衞生服務/基層醫療模式，以及不同地區在實施國家初級衞生政策時的不同策略所導致的在初級醫療五個核心方面的差異。 / 方法：首先，本研究進行了全面的文獻檢索，回顧了PCAT工具在全球和地區的應用。其次，本研究根據國際上認可的跨文化翻譯過程將PCAT（成人簡化版）翻譯成中文官方語言（普通話）並使其與中國國情相適應，並通過問卷的信度和效度分析，證明中文PCAT成人簡化版在評價社區衞生服務/基層醫療的過程層面方面具有較高的可靠性和有效性。進而，本研究採用多階段整群抽樣方法，在珠江三角洲六個城市的社區衞生中心，由訓練有素的調查員進行現場訪談。最後，本研究採用多元線性回歸和多元方差分析統計分析，評估和比較了不同城市和不同模式下，社區衞生服務/基層醫療服務五個核心方面的PCAT分數的差異。 / 結果：中文PCAT（成人簡化版）包含了九個不同維度，從五個方面評價初級保健服務過程，具有良好的結構效度。克隆巴赫係數反映了問卷具有良好的內部一致性。本研究共調查了3,360名在社區衞生服務中心接受基層醫療服務的成人患者，問卷整體回收率達86.1。本研究基於PCAT 分數從不同城市和不同服務模式角度分別評價和比較了社區衞生服務/基層醫療服務的過程。研究發現，與院辦院管及私營模式相比，政府主導模式下的社區衞生服務/基層醫療服務使用者具有更好的病人體驗，主要是因為政府主導模式下，社區衞生服務在首診利用及基層醫療服務協調統籌維度層面達到更高的水準。 / 結論：本博士論文研究確立了中文PCAT（成人簡化版）在評價社區衞生服務/基層醫療服務過程的可靠性和有效性。本研究率先在中國大陸採用PCAT 工具對廣東省珠江三角洲地區城市社區衞生服務/基層醫療模式開展了大規模調查。該項研究可以填補目前中國大陸在PCAT應用以及城市社區衞生服務/基層醫療模式研究方面的空白，探索適合中國國情的社區衞生/基層醫療服務發展道路，為中國大陸進一步發展及完善初級衞生體系提供翔實的政策依據。 / Backgrounds: China’s current healthcare reform has an overall goal towards re-strengthening primary care. Establishment and expansion of primary care network based on community health centres (CHCs) in urban areas has been prioritized. Due to various socio-economic status of local population and policy context across urban areas in mainland China, primary care is delivered by three main organisational models: government-owned CHCs, hospital-owned CHCs, and privately-owned CHCs with each model being adopted in various localities. This PhD study focuses on studying the process of primary care provided under different CHC models in the six cities of Pearl River Delta (PRD) and it is part of a larger study entitled “A Study of Comparing Primary Care Services among Six Cities in the Pearl River Delta funded by Bauhinia Foundation Research Centre, in which other co-investigators have studied the structure and outcome of the primary care. / Objectives: This PhD study aims to use the Primary Care Assessment Tool (PCAT) within Donabedian’s framework of structure, process and outcome to measure and compare the quality attributes of primary care from patient’s perspective under different CHC models in the six cities of PRD where each city has different responses to the national policy for delivering primary care services. / Methods: A comprehensive literature search was conducted to review the utilization of PCAT both globally and locally. The PCAT - Adult Edition (short version) was translated into Mandarin Chinese following an internationally recognized procedure and was cross-culturally adapted into Chinese context. The reliability and validity of the PCAT instrument were evaluated through test-retest approach, exploratory factor analysis, and internal consistency reliability analysis. Multistage cluster sampling method was adopted to select CHCs in the six cities of PRD. All interviews were conducted on-site by trained interviewers. Statistical analysis including multiple linear regression and multivariate analysis of covariance were used to assess and compare the quality attributes of primary care (PCAT scores) provided by different organisational models of CHCs in the six cities of PRD. / Results: The Mandarin Chinese version of PCAT-AE (short version) contains nine primary care scales with good construct validity. Cronbach’s alpha within all the nine primary care scales achieved moderate to high internal consistency reliability. A total number of 3,360 adult primary care service users were surveyed on-site at CHCs with an overall response rate of 86.1%. Descriptive city-by-city analysis based on the PCAT scores was conducted to depict primary care process in each of the six cities. Primary care service users under government-owned CHC model reported receiving better primary care experiences than those under privately-owned CHC model and hospital-owned CHC model, largely because of the greater achievements in first contact utilization and better score in the coordination domain (information system). / Conclusions: The Mandarin Chinese version of PCAT-AE (short version) was found to be reliable and valid as a measure of primary care in mainland China from patients’ perspective. The study suggested that the government-owned CHCs had better quality attributes than other organisational models, and offered a direction for quality improvement in the five domains of primary care. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Wang, Haoxiang. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references. / Abstract also in Chinese; some appendixes also in Chinese. / ABSTRACT --- p.i / ABSTRACT (IN CHINESE) 摘要 --- p.iii / ACKNOWLEDGEMENTS --- p.v / TABLE OF CONTENTS --- p.viii / LIST OF APPENDICES --- p.xii / LISTS OF TABLES --- p.xiii / LISTS OF FIGURES --- p.xv / ABBREVIATIONS --- p.xvi / PREFACE --- p.xvii / Chapter CHAPTER 1: --- BACKGROUND --- p.1 / Chapter 1.1 --- What is Primary Care? --- p.1 / Chapter 1.2 --- Why Primary Care? --- p.2 / Chapter 1.3 --- Healthcare Reform in mainland China --- p.3 / Chapter 1.4 --- Primary Care in Mainland China: A Brief History and Current Status --- p.8 / Chapter 1.5 --- Three CHC models of primary care delivery in urban areas --- p.13 / Chapter 1.5.1 --- Hospital and Hospital-owned CHCs --- p.13 / Chapter 1.5.2 --- Government and Government-owned CHCs --- p.14 / Chapter 1.5.3 --- Private sector and Privately-owned CHCs --- p.15 / Chapter 1.6 --- Pearl River Delta: an open window for primary care research --- p.16 / Chapter 1.7 --- Primary Care Assessment Tool: assessment of primary care from an international perspective --- p.21 / Chapter SUMMARY OF CHAPTER 1 --- p.23 / Chapter CHAPTER 2: --- LITERATURE REVIEW OF PRIMARY CARE ASSESSMENT TOOL --- p.25 / Chapter 2.1 --- Introduction --- p.25 / Chapter 2.2 --- Criteria for review and search strategy --- p.26 / Chapter 2.3 --- Key messages from the literature review --- p.28 / Chapter 2.4 --- Research gaps in the literature --- p.33 / Chapter 2.5 --- Conclusion --- p.34 / Chapter SUMMARY OF CHAPTER 2 --- p.36 / Chapter CHAPTER 3: --- ADAPTION OF PRIMARY CARE ASSESSMENT TOOL IN MAINLAND CHINA --- p.38 / Chapter 3.1 --- Introduction --- p.38 / Chapter 3.2 --- Methods --- p.39 / Chapter 3.2.1 --- Translation of PCAT --- p.39 / Chapter 3.2.2 --- Validation of the translation --- p.40 / Chapter 3.2.3 --- Face validity and peer evaluation --- p.40 / Chapter 3.2.4 --- Pilot test and test-retest reliability --- p.41 / Chapter 3.3 --- Results --- p.43 / Chapter 3.3.1 --- Translation of PCAT into Mandarin Chinese and validation of the translation --- p.43 / Chapter 3.3.2 --- Face validity --- p.46 / Chapter 3.3.3 --- Demographic characteristics of the survey sample in the pilot test --- p.46 / Chapter 3.3.4 --- Test-retest reliability --- p.48 / Chapter 3.4 --- Discussion --- p.48 / Chapter SUMMARY OF CHAPTER 3 --- p.51 / Chapter CHAPTER 4: --- VALIDATION OF MANDARIN CHINESE VERSION OF PRIMARY CARE ASSESSMENT TOOL --- p.53 / Chapter 4.1 --- Introduction --- p.53 / Chapter 4.2 --- Methods --- p.54 / Chapter 4.2.1 --- Study design and study subjects --- p.54 / Chapter 4.2.2 --- Scoring --- p.55 / Chapter 4.2.3 --- Factor analysis and construct validity --- p.57 / Chapter 4.2.4 --- Item analysis and internal reliability --- p.61 / Chapter 4.3 --- Results --- p.62 / Chapter 4.3.1 --- Demographic profile --- p.62 / Chapter 4.3.2 --- Analysis of the correlation matrix --- p.63 / Chapter 4.3.3 --- Factor analysis and construct validity --- p.63 / Chapter 4.3.4 --- Item analysis --- p.66 / Chapter 4.3.5 --- Internal reliability of the primary care scales --- p.67 / Chapter 4.4 --- Discussion --- p.68 / Chapter SUMMARY OF CHAPTER 4 --- p.70 / Chapter CHAPTER 5: --- PRIMARY CARE PROFILES IN SIX CITIES OF PEARL RIVER DELTA --- p.72 / Chapter 5.1 --- Introduction --- p.72 / Chapter 5.2 --- Methods --- p.73 / Chapter 5.2.1 --- The instrument to assess primary care --- p.73 / Chapter 5.2.2 --- Training of the interviewers and the assessment of inter-rater reliability --- p.74 / Chapter 5.2.3 --- Study location and sampling framework --- p.75 / Chapter 5.2.4 --- Target population --- p.76 / Chapter 5.2.5 --- Identification of primary care source --- p.76 / Chapter 5.2.6 --- Data collection --- p.77 / Chapter 5.2.7 --- Statistical analysis --- p.78 / Chapter 5.3 --- Results --- p.79 / Chapter 5.3.1 --- Inter-rater reliability, response rate, and demographic characteristics --- p.79 / Chapter 5.3.2 --- Comparison of primary care assessment scores in the six cities (city-by-city analysis) --- p.80 / Chapter 5.3.2.1 --- Primary care service users and primary care quality attributes in City A --- p.83 / Chapter 5.3.2.2 --- Primary care service users and primary care quality attributes in City B --- p.89 / Chapter 5.3.2.3 --- Primary care service users and primary care quality attributes in City C --- p.94 / Chapter 5.3.2.4: --- Primary care service users and primary care quality attributes in City D --- p.99 / Chapter 5.3.2.5 --- Primary care service users and primary care quality attributes in City E --- p.105 / Chapter 5.3.2.6 --- Primary care service users and primary care quality attributes in City F --- p.110 / Chapter 5.2.3 --- Factors associated with overall primary care experience --- p.115 / Chapter 5.2.4 --- Comparison of primary care assessment scores among the three CHC organisational models --- p.116 / Chapter 5.2.4.1 --- Presence of ‘hukou’ registry and primary care experience --- p.118 / Chapter 5.2.4.2 --- Presence of medical insurance and primary care experience --- p.119 / Chapter 5.2.4.3 --- Presence of chronic disease and primary care experience --- p.122 / Chapter 5.4 --- Discussion --- p.124 / Chapter 5.4.1 --- Patient characteristics and primary care profiles in the six cities --- p.124 / Chapter 5.4.1.1 --- Ageing --- p.126 / Chapter 5.4.1.2 --- Household registry --- p.126 / Chapter 5.4.1.3 --- Medical insurance --- p.127 / Chapter 5.4.2 --- Factors significantly associated with primary care assessment scores --- p.128 / Chapter 5.4.2.1 --- Healthcare utilization and health characteristics --- p.128 / Chapter 5.4.2.2 --- Socio-demographic characteristics --- p.129 / Chapter 5.4.3 --- Primary care experience in the government-owned CHC --- p.130 / Chapter 5.4.3.1 --- Separation between revenue and expenditure --- p.130 / Chapter 5.4.3.2 --- Central planning and multi-sectoral collaboration --- p.130 / Chapter 5.4.3.3 --- Towards health equality --- p.131 / Chapter 5.4.3.4 --- First contact --- p.132 / Chapter 5.4.3.5 --- Suboptimal service capacity --- p.132 / Chapter 5.4.4 --- Primary care experience in the privately-owned CHC --- p.133 / Chapter 5.4.4.1 --- Lower healthcare utilization --- p.133 / Chapter 5.4.4.2 --- Insufficient funding support --- p.134 / Chapter 5.4.5 --- Primary care experience in the hospital-owned CHC --- p.135 / Chapter 5.4.5.1 --- Large service capacity --- p.135 / Chapter 5.4.5.2 --- Tackling aging population with chronic diseases --- p.136 / Chapter 5.4.5.3 --- CHCs in the less socio-economic developed urban area --- p.136 / Chapter 5.4.5.4 --- Disparities due to socio-demographic status --- p.137 / Chapter 5.4.6 --- Study limitations --- p.137 / Chapter 5.4.7 --- What is already known and what this study adds --- p.139 / Chapter SUMMARY OF CHAPTER 5 --- p.141 / Chapter CHAPTER 6: --- CONCLUSION AND POLICY IMPLICATIONS --- p.144 / Chapter 6.1 --- Conclusion --- p.144 / Chapter 6.2 --- Policy implications for mainland China --- p.146 / REFERENCES --- p.150 / APPENDICES --- p.161

Multiple chronic conditions (MCCs) are defined as two or more health conditions, each requiring treatment and limiting activities for a year or more. In the United States (US), MCCs are more common and costly than any individual chronic condition. The number of adults aged 65 years and older with MCCs is projected to nearly double between 2020 and 2050. Patients with MCCs often experience poor self-reported health and negative symptoms. In addition, they frequently visit emergency departments (EDs) and are hospitalized. Patients with MCCs need ongoing primary care services to manage their symptoms and prevent health deterioration. However, over 20% of the US population (many of whom have MCCs) resides in a primary care Health Professional Shortage Area (HPSA) and experiences poor access to primary care. The growing nurse practitioner (NP) workforce, which is projected to almost double in size between 2018 and 2030, can help meet the demand. Most NPs are trained to diagnose, treat, and manage chronic conditions and can provide a scope and quality of primary care comparable to physicians in many populations. Therefore, if distributed and supported strategically, the NP workforce can meet the complex care needs of patients with MCCs, especially in HPSAs. Maximizing the potential of the NP workforce to deliver MCC care will require enhanced care environments in the practices where NPs work, characterized by administrative support for NP care delivery and autonomous practice, collegial relationships between NPs and physicians, and NP professional visibility. On the other hand, poor NP care environments can negatively affect the quality of chronic disease care. Thus, improving the NP care environments within practices may increase the capacity of the NP workforce to care for MCC patients. Despite the potential of the NP workforce to meet the need for primary care among patients with MCCs, little is known about the impact of NP-delivered primary care models on outcomes in this population. Furthermore, the impact of HPSA status and NP care environments on NPs’ ability to care for patients with MCCs remains poorly understood. Thus, the overall purpose of this dissertation is to produce evidence on NP-delivered primary care models for patients with MCCs and examine the interplay between practice and community factors in shaping outcomes for these patients. In chapter 1, we introduce the unique healthcare needs of patients with MCCs, and the role of NPs in delivering and expanding access to care. In chapter 2, we synthesize the existing evidence on the effect of NP primary care models, compared to models without NP involvement, on cost, quality, and service utilization by patients with MCCs. Our synthesis suggests that NP-delivered primary care has similar or better impacts on outcomes among patients with MCCs compared to care delivered without NP involvement. In chapter 3, we perform secondary data analysis using multiple linked data sources including 1) patient data from the Medicare claims of 394,424 older adults with MCCs, 2) NP survey data on practice characteristics from 880 NPs at 779 primary care practices across five US states, and 3) data on HPSA status of the practice locations from the Health Resources and Services Administration. We examine differences in hospitalization and ED use among patients who receive care from NP practices in HPSAs compared to those in non-HPSAs. We find a higher likelihood of ED use among patients receiving care in NP practices located in HPSAs compared to practices in non-HPSAs, and no difference in the likelihood of being hospitalized. Our results suggest that relieving provider shortages may reduce ED use by MCC patients in HPSA practices that employ NPs, but may be insufficient to lower hospitalization rates unless combined with other interventions. Finally, in chapter 4, we analyze the same linked secondary data source as in chapter 3 to examine the effect of the NP care environment (measured by the NP survey) on the relationship between the HPSA status of the practice location and ED or hospital use among patients with MCCs. We find that the NP care environment moderates the association between primary care provider shortage areas and hospitalization but not ED use. Further analysis reveals that improved NP care environments have a more pronounced association with lowered odds of hospitalization among patients receiving care from practices located in areas with no shortage of primary care providers (i.e., non-HPSAs) compared to those receiving care in practices with provider shortages (i.e., HPSAs). Our findings suggest that improving the care environment may not have the effect of reducing MCC patients’ need for hospitalization unless sufficient providers are also available to care for patients. We suggest that cohesive solution sets addressing practice- and community-level interventions simultaneously may be needed to improve hospitalization outcomes for patients with MCCs. In the concluding chapter of this dissertation, chapter 5, we present a summary of findings, discuss the dissertation’s strengths, limitations, and its contributions to science. In this chapter, we also discuss implications for policy, practice, and directions for future research.

Made available in DSpace on 2011-05-04T12:42:06Z (GMT). No. of bitstreams: 0 Previous issue date: 2009 / O vínculo longitudinal pode ser definido como ‘relação terapêutica entre paciente e profissionais da equipe de Atenção Primária em Saúde (APS) ao longo do tempo, que se traduz na utilização da unidade básica de saúde (UBS) como fonte regular de cuidado para os vários episódios de doença e cuidados preventivos. O vínculo longitudinal contribui para diagnósticos e tratamentos mais precisos, diminuição dos custos da atenção e maior satisfação do paciente. O presente estudo teve por objetivo investigar o atendimento a tal atributo na experimentação de diferentes modelos assistenciais organizativos da APS no contexto do SUS. O estudo, que está estruturado em três artigos/capítulos, teve início com a identificação do vínculo longitudinal como característica central da APS. Revisão conceitual possibilitou definir o termo, identificar três dimensões para análise do atributo (identificação da UBS como fonte regular de cuidados; relação interpessoal entre profissional e paciente; e continuidade informacional), e construir roteiro para investigar o vínculo longitudinal no âmbito da APS municipal. Revisão bibliográfica sobre os modelos assistenciais no Brasil identificou propostas atuais com experiências consolidadas de estruturação da APS. Esses municípios/modelos foram: Camaragibe – que aderiu à Estratégia Saúde da Família; e Belo Horizonte – que segue os princípios orientadores do modelo Em Defesa da Vida/Acolhimento, embora tenha aderido à Estratégia Saúde da Família posteriormente. A presença de elementos favorecedores do vínculo longitudinal na abordagem teórica desses modelos foi averiguada, bem como a reprodução desses elementos nas concepções vigentes no âmbito da gestão da APS municipal. A terceira parte da tese consiste em estudos de caso, com trabalho de campo nesses dois municípios. A investigação teve por base as três dimensões identificadas para o atributo; e a atenção aos portadores de hipertensão arterial foi utilizada como condição traçadora. As fontes de dados contemplaram três âmbitos: gestão do sistema, prática profissional e ponto de vista do usuário. As estratégias de coleta de dados foram: entrevista semi-estruturada com os profissionais, revisão de amostra de prontuários e aplicação de inquérito em amostra de usuários, além de análise de documentos e de dados secundários. Os resultados apontam para a existência de fatores que ainda dificultam o atendimento do vínculo longitudinal como: busca de outras unidades de saúde para atendimento de rotina, rotatividade do profissional médico, e problemas de completude e suficiência dos registros em saúde. Por outro lado há avanços, como: o reconhecimento da interferência de fatores socioeconômicos no processo de adoecimento dos indivíduos, e a presença de vínculo entre usuários e profissionais da equipe de APS. Belo Horizonte apresentou melhor desempenho na primeira e na terceira dimensão; Camaragibe, na segunda. Aspectos relativos à estruturação da rede de serviços, valorizados pelo modelo Em Defesa da Vida/Acolhimento, mas também proporcionados pelas condições socioeconômicas locais, favoreciam o vínculo longitudinal no que refere à identificação da UBS como fonte regular de cuidado e à continuidade da informação para o acompanhamento do paciente em Belo Horizonte; enquanto que limitações estruturais da rede de serviços, presentes em Camaragibe, podem estar relacionadas com a busca de outras unidades de saúde em concomitância com a UBS para o tratamento de rotina. A realização do estudo aponta para a pertinência de se avaliar o vínculo longitudinal no âmbito da APS. / The longitudinal relationship can be described as a ‘therapeutic relationship between patient and Primary Health Care (PHC) staff over time, expressed as use of the basic health care center (Unidade Básica de Saúde, UBS) as the regular source of care for the various episodes of disease and for preventive care’. A longitudinal relationship contributes towards more accurate diagnoses and treatments, lower health care costs and greater patient satisfaction. The purpose of this study was to investigate this attribute in the experimentation with different PHC organization models in Brazil’s national health system, the Sistema Único de Saúde (SUS). The study, which is divided into threepapers/chapters, began by identifying the longitudinal relationship as a key characteristic of PHC. By conceptual review it was possible to define the term, to identify three dimensions for analysis of the attribute (identification of the UBS as the regular source of care; the interpersonal relationship between health care professional and patient; and informational continuity) and to construct an investigational path for researching the longitudinal relationship in municipal PHC. A bibliographical review of health care models in Brazil identified current setups with established experience in structuring PHC. These municipalities/models were Camaragibe, which applied Brazil’s Family Health Strategy, and Belo Horizonte, which followed the guiding principles of the Em Defesa da Vida/Acolhimento (in defence of life/humane reception) model, although it later adhered to the Family Health Strategy. The models’ theoretical approaches were examined for elements likely to favour the longitudinal relationship, and whether such elements were reproduced in conceptions current in the municipal PHC managements. The third part of the thesis comprises case studies involving field work in these two municipalities. The research was based on the three dimensions identified for the attribute; and care for arterial hypertension patients was used as the tracer condition. The data sources covered three areas: system management, professional practice and user point of view. The data collection strategies employed were semi-structured interviews of health professionals, review of a patient record sample and application of questionnaires to a user sample, as well as documentary and secondary data analysis. The results point to factors that continue to hinder longitudinal relationships, such as: patients seeking other health units for routine care, turnover of doctors, and problems of incomplete and inadequate medical records. On the other hand, there have been advances, such as the recognition that socioeconomic factors affect the individual process of falling ill, and bonding between PHC users and staff. Belo Horizonte performed better in the first and third dimensions; and Camaragibe, in the second. In Belo Horizonte, aspects of service network structure valorised by the Em Defesa da Vida/Acolhimento model, but also afforded by local socioeconomic conditions, favoured the longitudinal relationship with regard to identification of the UBS as the regular source of care and to informational continuity in patient follow-up. Meanwhile, structural constraints in Camaragibe’s service network may be related to patients’ attending other health care units in parallel to the UBS for routine treatment. The study signals the relevance of evaluating the longitudinal bond in PHC.