The analysis of the data of the ambulatory healthcare centre and the whole county in which the health care centre is located, describes the effects of introducing the delegation of GP-home visits to qualified praxis staff as a new option for the total system of home visits in this centre.
The study size was determined by the monocentric intervention and the county based organizational structure of the reimbursement data.
A limitation of the study is the absence of a prospective power calculation with respect to the development of the number of home visits. At the start of the project, we didn't know how many and which patients the GPs would select for this project, which activities would be delegated, the acceptance of the concept by the patients, and if the quality of care would be sufficient. In this context, a power calculation didn't seem to be useful.
The analysis is important to asses the eligibility of the AGnES-concept for implementing it into the regular health care system: a significant increase of the total number of home visits would have implied that the AGnES concept induced additional home visits in this setting. AGnES would then generate another financial burden on the statutory health insurances, which would increase the hesitation to add AGnES-home visits as a new option to their reimbursement catalogue.
An increase of the total number of home visits would only be acceptable if a region was actually underserved at the beginning of the project. In this case, the delegation of GP-home visits would provide a certain compensation for the lack of GP resources. As the analysed region was not underserved at the time of the project, an increase of home-visits would have indicated that the AGnES-option generates an additional demand of home visits either for a larger number of patients and/or to a higher frequency. The observation of a constant number of home visits before and during the intervention however, indicates that the new option had no major impact on the overall demand.
The results of this analysis show a clear trend towards a redistribution of home visits from the GP to qualified staff in his practice-team. There is an isolated peak in the curve of the GP-home visits of the ambulatory healthcare centre in the first quarter year of the year 2007 which is rather not associated with the project implementation, because it disappears already in the 2nd quarter year of the year 2007 while the frequency of the AGnES-home visits increases steadily until 2nd quarter year of the year 2007 (Figure 1).
Numerically the most extended effect is related to the medical urgent home visits. In the reimbursement catalogue, urgency is defined by date and time: home visits that have to be conducted after 7 pm, on bank holidays or during regular practice consultation-hours are defined as urgent and can be reimbursed with a higher remuneration. Hence practically, our results mean that the AGnES-concept enabled the GPs to reduce unscheduled home visits, thus avoiding working at unfavourable times and interruptions of regular consultation hours.
The setting of an ambulatory healthcare centre allowed performing an analysis of both the impact on the specific institution and on the regional population. Another advantage was the real time accessibility of the data used for reimbursement claiming, which were continuously documented with high quality. For the AGnES-practice assistants the complete information for each home visit was available from the standardized computer-assisted online documentation. The documentation of all AGnES-home visits on a patient-based level in a standardized computer-assisted documentation system assured high validity of this dataset.
However, this analysis has some limitations:
There is a difference between the subpopulation of the AOK-patients of the ambulatory healthcare centre and the AOK-patients in the county Oberspreewald-Lausitz: the basic population of the county consists of all AOK-patients which have their primary place of residence here. In Germany, patients are free to choose their GP and - depending on availability - in principle may change GPs whenever they want. Therefore, only patients who had at least one GP-contact in the respective quarter year were documented as patients of the ambulatory healthcare centre. The consequences can be lower home visit rates for the county, because here all patients, including those without a GP-contact in the respective period, are counted. The proportion of patients, who do not see a GP over a period of two consecutive years, however, is small .
Second, there is only a limited comparability between the total population of statutory insured patients of the ambulatory healthcare centre and the populations of AOK-patients of the healthcare centre and the surrounding county, because of large differences between the patients of the different German statutory health insurances, e.g. in age distribution, income, education, and profession.
Reimbursement data are collected primarily for reimbursement purposes and do not necessarily provide an exact picture of the actual medical activities. For example, urgent GP-home visits are defined by the time of the day. It is assumed that the GP will only make a home visits outside practice consultation-hours or at night if there is urgent medical need.
However, "home visit" constitutes a unit of reimbursement independent of its particular indication. All home visits are assigned to a specific institute. Hence, the assessment is likely complete and assumes a high validity of this dataset.
For these analyses, not the absolute numbers are important but rather the trends during the study period. It is unlikely and there is no indication that major changes happened over the study period.
Within the reimbursement data set, there was no specification for the gender of the patient and only a broad classification for age (≤5, 6-59, ≥ 60 years). Information about the medical condition of the patients was not available.
The changing conditions of primary care in Germany necessitate a change in the role of the GP from the solitary player of the past to the manager of a competent practice team. The future GP will distribute work packages flexibly to each member of his team, individually considering the specific competences necessary to perform them. Different concepts are being evaluated, e.g. the integration of practice nurses in the chronic care model  or in case management models for specific indications, e.g. heart failure . The patients' acceptance of such new models of organizing primary care is generally quite good [13, 14]. The implementation of innovative concepts into usual care, however, requires more than proven performance, good quality, and acceptance of the patients. Presently all participants in this process, physicians, nurses, and practice assistants, are still organized in separate traditional professional institutions, which for decades have focused on defining specific tasks and competences to maintain mutually exclusive professional spheres. Hence the main goal of the past was a perpetuation (and wherever possible extension) of professional borders and privileges, the future will focus on flexible work share. While any medical task as ever before requires the utmost responsibility and quality, the key issue will become qualification rather than profession. Since maintaining the quality of care is an inevitable condition for flexible work share concepts, the evaluation of objective parameters is an important issue. Regarding the AGnES-project, we are analysing different objective parameters (e.g. health related quality of life and the development of blood pressure values of patients with hypertension) to get objective indications for the quality of care of this concept.
Along this way, the traditional who-does-what-question in medical care will be fundamentally reconciled. We should put all effort on recruiting motivated staff, improve education, develop and provide flexible, evidence-based qualification and evaluate quality of care for patient-oriented results to meet the ever increasing demands of a growing, and aging number of patients in the next decades.