In the last two decades, developing country governments have implemented a variety of reforms in the health sector on the understanding that these reforms would create the right individual and organisational incentives for improving health systems performance. However, reform initiatives have not always considered human resource issues that are relevant to their success and have often failed to include the participation or perspectives of the health workforce in reform planning processes and decision-making.
A number of studies have considered the effects of reforms on the health workforce [[1
] and [6
]] and highlight the importance of human resources to the success of reform objectives [7
] as well as the complexity of human resource management in the context of reforms [8
These studies have pointed out that human resource issues need to be a primary consideration in reform design, suggesting that reforms can only be implemented successfully where there is consensual participation on the part of the workforce. Ngufor describes how health staff in Cameroon perceived reforms as a punishment inflicted on the nation by the International Monetary Fund (IMF) and the World Bank and as a result developed a laissez faire attitude to their work resulting in reduction of consultation times and absenteeism [4
]. Similarly, workers in Zimbabwe were reported to perceive reforms as threatening their job security, salaries and training and expressed their demotivation in the form of unethical behaviour with their patients and neglect of work responsibilities [6
]. In other parts of the world, health workers have resisted change on the grounds of conflicting values. In Latin America, for example, reforms were perceived as aiming to undermine the fundamental values that had inspired the design of the system, and sparked off a wave of resistance and strikes in El Salvador and Mexico. This led to the stalling and delay of the reform process [10
Other studies have noted higher motivation levels among the health workforce through reforms. In Kazakhstan, reforms that aimed at changing the old Soviet system through the introduction of more market-oriented financing and service delivery, were argued to have resulted in increased interest in primary care among physicians, increased attention to quality and patient satisfaction, more rational and creative use of resources, and stronger commitment of physicians' personal time and resources to improve services for patients [12
Despite the expanding literature on how reforms may affect health workers and which reactions they may provoke, little research has been conducted on the mechanisms through which health sector reforms either promote or discourage health worker performance. How do reforms affect the health workforce and create responses that are likely to encourage the success or failure of reform objectives? How does context influence the routes through which reforms affect provider incentive environments and eventually motivation and performance? To address these questions, the following paper seeks to trace these mechanisms of effect and examines health workers' responses to the changes in their working environments by taking a 'realistic evaluation' approach [13
This approach takes account of the explanatory mechanisms and the context of health systems reforms. This stands in contrast to the more common evaluation approach, which tends to focus on a reform programme's measures and its intended effects and then seeks to measure the differences between reforming and non-reforming entities, within the dimensions of those intended effects. The research described above suggests that reform programmes cause a multitude of workforce responses which act as the lynchpin between formal arrangements at the outset of the reforms and the resulting changes in the system as experienced by people who use it. Figure , originally created by McPake [14
], captures the essence of the scenario described above and provides a framework for health systems research.
A conceptual model for health systems research – the Dynamic Responses Model .
The focus of enquiry lies within the three components outlined above. While the de-jure system sets the context and defines the organisational structures, the intended incentives and management procedures, the dynamic responses reflect how those implementing the de-jure system respond and create 'informal' arrangements or behaviours that may or may not encourage the success of 'formal' reform objectives initiated by the de-jure system. The de-facto system then encompasses the lived experience of the results of the relationship between the de-jure system and the dynamic responses. Similar to realistic evaluation, the model emphasises the mechanisms of effect between a social programme (component of the de-jure system), and its impact (in the de-facto system), and recognises these mechanisms as being social: involving the interaction of human beings in a particular context that determines the nature and form of dynamic responses [14
In this paper, we seek to focus on the mechanisms through which unintended human resource effects have arisen, using the case studies of Bangladesh and Uganda, two countries that have faced many reforms attempting to improve health systems performance and delivery of care. Both countries have extremely low per capita expenditure on health (Bangladesh has a public health expenditure of 14 international dollars, one of the lowest in the region and Uganda's public health expenditure lies at 22 international dollars) and face challenges in the attainment of the Millennium Development Goals and national development. After 20 years of military regime, Bangladesh became a parliamentary democracy in 1991. With rigid central government structures and disagreement between the main political parties inhibiting response to local health needs, the country began a wide programme of reforms to address issues of responsiveness. In Uganda, a series of reforms were introduced from the early 1990s as part of a larger initiative to restore the health system following its collapse during the political crises of the 1970s.
Focusing on selected reforms in each country (see Table ) the first section of the paper outlines the scope of the reforms implemented and provides observations on workforce responses made by earlier evaluations. The subsequent section then presents the results of this research, exploring the impact of the mechanisms on social interactions.
Selected reform initiatives in Uganda and Bangladesh