In reflecting on the papers presented to this years INIC conference in Tampere (the abstracts and presentations being available in this conference supplement) it is clear that we are reaching a watershed in the way researchers are beginning to think about integrated care as a scientific discipline, and the way health systems are beginning to properly embrace its principles. For example, the policy and practical papers presented at the conference, particularly impressive being those of our Finnish hosts, show that elements of integrated care for vulnerable groups such as the frail elderly or to those with long-term mental health problems have been around for many years. However, it seems that momentum is gathering and that integrated care as a term and principle for future health care delivery is moving from a backwater activity and into mainstream thinking.
However, the conference also brought home to me just how many interesting integrated care innovations tend not to survive beyond a period of piloting or special investment; that many good ideas have to be re-invented; and how organisational memory can too easily be lost. The evidence-base for integrated care also remains weak, compounding the problem and difficulty of mainstreaming new ways of working that cut across established organisational and professional boundaries.
There is thus a need to continue with advancing the science of integrated care in order to develop theories about care integration that could be tested empirically; to create innovative methodologies that capture the impact of the complexity of the subject; and to develop and use appropriately designed intervention studies. During the conference, a movement in this direction appears to have made a significant advance as the quality and complexity of many research papers being presented had begun to tackle these key challenges. In the years to come, such work will be important in providing the knowledge to support the redesign of care systems in many countries where fragmented care remains commonplace.