Increased collaboration between public municipal health care and the POPD increased the numbers of patients attending their first call to the POPD and reduced the time from examination in the municipality to examination at the POPD.
The contents of screening programmes for immigrants show great variety among countries . Studies from Switzerland and Canada have shown similar problems as in our study with follow-up of TB screening results among immigrants [9, 10]. In more recent studies from Canada up to 50% of immigrants had adhered to post immigrant TB surveillance .
How can the follow-up of screening results be improved? A comparable study from USA showed improvement in the follow-up after arrival of people with abnormal results at their pre-entry screening. Without interventions, 25% of the immigrants arrived for follow-up. Step-wise intervention started with summoning letters, phone-calls, and finally home-visits and altogether 97.5% of the immigrants had their follow-up . A review of migrant tuberculosis screening in Europe emphasized the importance of a good follow-up system and continuum of care. The review concluded that TB care should be integrated with general healthcare within a holistic approach . One of the conclusions of a study about best practices in delivery of health care services to immigrants in Denmark was the need for collaboration between the different levels of health care and between the health care and the social sector .
As the intervention measures of our project were initiated simultaneously, it is hard to say which had most impact. Shorter time from examination in the municipality to summoning for the follow-up probably increased the numbers that arrived for examination, both because the patients remembered why they were referred, and fewer had moved to another municipality or changed addresses. Another reason for more patients turning up and shorter waiting time could be that the project led to increased awareness among staff at all levels. Because more patients attended their first appointment, fewer patients had to be summoned for a second or third time. This would reduce the workload and waiting lists at the POPD and as a consequence reduce the time before appointments.
The summoning process is another important issue, as was shown in the study from USA . When only a letter is sent to the patient, no one knows whether the letter has been received, whether the address is correct or the patient has moved on, or whether the patient understands what the letter is all about. By contacting each patient directly or through an interpreter or a contact person, the contact was established and the message probably understood. It is possible that simplification of the summoning letter also helped; it was not necessary to send out extra letters from the X-ray department. The summoning letters could be improved further e.g. by further simplification and by translation into different languages.
Fewer patients were missed after they arrived at the hospital in the intervention period. X-rays and blood testing done before the appointment and a guide who received the patients from RHC when they arrived at the hospital could be reasons for that.
In the VICO group, there were more patients diagnosed with latent TB in the intervention group than in the control group. The reasons could be different interpretations of patient records or a real difference between the groups. Overall, more of the RHC patients were offered treatment for latent TB than in the VICO group. This is not adjusted for other factors, and it is beyond the focus of this paper to explain these results. Still, the results seem reasonable because refugees have higher risk for reactivation of TB than estimated risk in their home countries and e.g. labour immigrants or students from the same country [15, 16].
Possible consequences of the project
Because patients with abnormal findings at screening turned up sooner at the POPD, some patients were diagnosed with TB disease before having symptoms. As a consequence, they were less likely to infect so many others and the contact tracing would be less extensive. Others were possibly prevented from developing TB by getting preventive treatment at an earlier stage. Because fewer patients needed to be summoned a second or third time to the POPD, the waiting time and waiting lists where shortened. This benefited both TB screening patients and other patients attending the POPD. In the municipality, the requests for checking up missing people were considerably reduced and this reduced the total workload. Performing screening in the municipality was probably more rewarding when a conclusion followed quickly?
Can the results of this project be used elsewhere? Local differences in the organization of the screening process make it difficult to use the experiences from our project directly. But the process of closer collaboration between different levels of health care dealing with screening could be transferred everywhere. A thorough mapping of who is doing what and where in the screening process, and what the main problems are locally, is a useful way of starting the process of improvement. Our results indicate that this way of dealing with problems between health care levels could also be used for other specific patients groups that have problems using the ordinary health care system.
Strengths and limitations
The project was started to improve the follow-up of positive screening tests, not to evaluate the screening programme itself. Patients with alarming symptoms or grossly abnormal X-rays were fast tracked through the system and not registered in this project. The yield of screening can thus not be evaluated.
The control groups were registered in the municipality from their files by a nurse and the project groups from the hospital files by the project leader. When no information from the hospital was received in the municipality, the patient information was checked in the hospital files for more complete information. This would make the data collection and registration as similar as possible for the control and project groups. Still some data could be missing or registered in a different way at different levels.
From one year to the next, countries of origin of the screened persons differ and so do the numbers in each group. There were even some patients from Norway included who were screened according to the regulations. All these factors could influence the degree to which the patients would attend their appointment.
The strength of the study is the simplicity, and how easily the information can be used in other settings. The project is close to ordinary routines and can be handled in practice with minor adjustments.