Health workforce indicators: let's get real.

Health workforce indicators?1 Those should be easy. We just need to count the numbers entering from training institutions or through re-entry, the numbers working, and the numbers exiting. If we know where these people work, we have the distribution of health workers within a country, and if we also have information on their competencies, responsiveness and productivity, we can know about their performance. Sound health workforce statistics enable countries to develop policies that ensure the equitable and effective distribution of the workforce. They can be used to forecast needs by making projections and to plan accordingly. They can also be the basis for implementing policies to improve performance and the regulation of the public and private sectors. These statistics would also allow for reliable global monitoring of progress, including progress towards achieving benchmark targets,2 and for monitoring the implementation of the WHO Global Code of Practice on the International Recruitment of Health Personnel.3 And yet, health workforce statistics are fraught with measurement problems. This is not for lack of agreement on core indicators or because we do not know what needs to be monitored. And it is not because measuring indicators is complicated or costly, as is true in other areas of health. For some indicators, such as those that capture productivity, more work is needed, but many indicators are well established.4,5 Health workforce information systems fail to deliver comprehensive, reliable and timely data in many countries. As a consequence, planning and policy-making are often based on very limited evidence and global monitoring in areas such as the implementation of the Global Code and the setting of benchmarks is conducted with inadequate country statistics. The challenges begin at the very basis: with the definition and classification of health workers. Indicators are intended for tracking progress over time, so country-specific definitions make it difficult to assess trends and conduct comparative analyses. The International Standard Classification of Occupations of the International Labour Organization facilitates the mapping of country health labour data, but it does little to take the statistical dimension into account, as is done, for example, for the International Classification of Diseases (ICD).6 Some solvable issues are not well addressed, among them the classification of non-physician clinicians and community health workers.7 Measuring the size and distribution of the health workforce involves drawing data from several sources, including sources outside the health sector.4 Currently too little is done to make use of these multiple, imperfect sources, reconcile the numbers and develop a best estimate. Human resources for health observatories aim to improve the information base,8 yet to date they have had little impact on the quality of health workforce data and statistics.