Table 1.
Summary statistics of the variables used in the regression analysis
| Summary statistics | |||||
|---|---|---|---|---|---|
| Variable | Observations | Mean | Std. Dev. | Min | Max |
| MRI exams (district level) | |||||
| Outpatient MRI exams | 117 | 5017.556 | 3597.253 | 110 | 23,913 |
| Outpatient MRI exams per 1000 inhabitants | 117 | 66.925 | 21.999 | 7.523 | 116.781 |
| Outpatient MRI exams (including hospital outpatient departments) | 116 | 6468.905 | 4050.762 | 115 | 27,066 |
| Outpatient MRI exams per 1000 inhabitants (including hospital outpatient departments) | 116 | 87.450 | 14.742 | 52.377 | 128.687 |
| Socio-Economic Score (district level) | 116 | ||||
| Epidemiological factors | |||||
| Share of smokers (care region level) | 116 | 28.776 | 4.284 | 20.079 | 36.650 |
| Share of daily drinkers (care region level) | 116 | 5.930 | 1.371 | 3.458 | 8.566 |
| Share of population with chronic back pain (care region level) | 116 | 23.747 | 2.963 | 19.121 | 31.988 |
| Share of population with chronic neck pain (care region level) | 116 | 19.216 | 2.903 | 14.945 | 26.225 |
| Share of population with accident at home (care region level) | 116 | 2.913 | 0.843 | 1.623 | 4.675 |
| Share of population with accident in spare time (care region level) | 116 | 5.996 | 1.307 | 3.534 | 9.221 |
| Cancer mortality (deaths per 1000 inhabitants; district level) | 117 | 2.408 | 0.622 | 0.865 | 4.875 |
| Share of population experiencing problems with access to care (care region level) | 116 | 1.429 | 0.595 | 0.531 | 3.072 |
| Share of heavy utilizers of patients with MRI exams (district level) | 117 | 48.835 | 9.411 | 24.484 | 65.826 |
| Supply-side factors | |||||
| MRI units in hospitals (district level) | 117 | 0.786 | 1.401 | 0 | 9 |
| Outpatient MRI units (district level) | 117 | 0.547 | 1.030 | 0 | 8 |
| MRI units in non-contract facilities (district level) | 117 | 0.325 | 0.585 | 0 | 2 |
| Outpatient CT examinations per 1000 inhabitants (district level) | 116 | 35.659 | 16.522 | 0.969 | 103.587 |
| CT examinations in outpatient departments in hospitals per 1000 inhabitants (district level) | 116 | 50.506 | 36.368 | 10.466 | 230.813 |
| Physicians (all types) with SHI contract per 1000 inhabitants (district level) | 116 | 1.394 | 0.725 | 0.741 | 7.495 |
| Physicans (all types) without SHI contract per 1000 inhabitants (district level) | 116 | 1.605 | 2.470 | 0.164 | 22.363 |
| Orthopaedists with SHI contract per 1000 inhabitants (district level) | 116 | 0.045 | 0.060 | 0 | 0.548 |
| Orthopaedists without SHI contract per 1000 inhabitants (district level) | 116 | 0.168 | 0.225 | 0 | 1.706 |
| Required pre-approval of MRI examinations by SHI (state level) | 117 | 0.427 | 0.497 | 0 | 1 |
| Demographic factors (district level) | |||||
| Share of female population | 117 | 50.850 | 0.799 | 49.259 | 53.689 |
| Share of population aged 45 to 64 | 117 | 28.932 | 2.133 | 23.031 | 33.223 |
| Urbanicity (urban = 1, rural = 0) | 117 | 0.291 | 0.456 | 0 | 1 |
| Population | 117 | 74,363 | 44,369.170 | 1911 | 280,258 |
The original sample contained information on patients in 117 districts, whereas any complementary data added at later stages of the study (e.g. MRI exams in outpatient departments in hospitals, epidemiological factors, and so forth) were only available for 116 districts. This issue arises because one district has been merged with other districts in 2017 and how patients from this district were handled differed between dates of data extraction. The assignment of the patients could not be harmonized retrospectively. We therefore report the descriptive statistics for the original sample with 117 districts, as these patients are also contained in the complementary data, but are assigned to other districts