Accessibility - Human resources

Healthcare accessibility can also be considered from the angle of the number of qualified practising healthcare providers in the country and the number of available beds in some healthcare sectors.
This information is also essential in order to plan the healthcare offer for the years to come.

In this section, you can find indicators on physician and nurse availability in general:

• Number of practising physicians (A-5)
• Number of practising nurses (A-6)
• Number of nursing staff vacancies in hospitals (A-7)
• Number of patients per nurse (A-8) 

Some of the indicators analysed in other sections of this report may also be interpreted in terms of human resources accessibility:

• Among mental healthcare indicators: Number of psychiatrists (MH-2);
• Among care for the elderly indicators: Number of geriatricians (ELD-6), Number of available spaces in long-term care beds (ELD-4), Percentage of elderly people receiving long-term care at home (ELD-2)
• Among end-of-life care indicators: Number of patients receiving palliative care (EOL-1)

Lastly, future projections about the expected number of healthcare professionals (S-4 to S-10) are discussed in the ‘Sustainability’ section.

Number of practising physicians (A-5)

This indicator measures the number of practising physicians per 1000 population.

It is considered that a physician is ‘practising’ if he/she has provided more than one clinical service (i.e. at least two consultations, visits or technical procedures – prescriptions are not counted here) during a given year. Physicians in training are not counted.

Until 2009, the calculation of physician density took into account all physicians licensed to practice (rather than practising physicians, defined as physicians having had more than 1 patient contact per year), which explains why, for a long time, Belgium showed one of the highest levels of physician density in Europe. These data have now been retrospectively corrected.

In addition, for 2016, the OECD data (reproduced in this datasheet) relate to the number of physicians who provided at least one clinical service (instead of more than one clinical service). Details about the number of physicians having provided more than one clinical service (34 502 instead of 34 834) are available on the RIZIV-INAMI  website.

RESULTS

• There were 34 834 practising physicians in Belgium in 2016, which translates into a density of 3.07/1000 population. Between 2000 and 2016, this density level has remained fairly stable. However, increasingly widespread ageing among physicians can be observed (see the indicators about physician age in the Sustainability section)
• The density of practising physicians in Belgium is lower than the EU average (3.54/1000 population). The different OECD countries do not, however, all use identical methods to calculate the same indicator (e.g. to define practising activity levels). The comparisons may, therefore, be incorrect.
• This indicator poorly reflects the actual number of practising physicians, since it includes all physicians who performed at least one clinical procedure, which is an extremely broad definition. To overcome this problem, INAMI also estimated the number of practising physicians in Full-Time Equivalents (FTEs), based on the amounts of reimbursed services. According to these estimates, the number of practising general practitioners (GPs) was 8 988 FTEs in 2016 (12 929 in absolute numbers), including 7 719 conventioned GPs in FTEs. Expressed in terms of density, this corresponds to 0.79 GPs per 1000 population, including 0.68 conventioned GPs. Details per medical specialty is shown in Table 1.
• When expressed in FTEs, the density of conventioned gynaecologists is fairly low (see also the indicator for conventioned physicians).
• However, it should be noted that this analysis is based on physicians’ residential addresses and not on their workplaces. A recent study based on general practitioners’ workplaces showed that in 2017, there were 1.23 general practitioners per 1000 population in Brussels (versus 1.17 per 1000 population in 2016 based on the GPs’ places of residence; RIZIV-INAMI data for 2017 are not yet available).

Figure 2 - Number of practising physicians, per 1000 population, international comparison, 2016
Data source: OECD Health Statistics 2018; EU average= based on 10 of the EU-15 countries for which data were available

Link toward the technical sheet and detailled results

Number of practising nurses (A-6)

Nurses play a key role in delivering healthcare, whether in hospitals and other healthcare institutions or as part of first-line and home care services.
In most countries, nurses represent the largest professional group among all healthcare professions; a lack of nurses can therefore lead to accessibility issues (e.g. making waiting lists longer), or have consequences on quality of care (e.g. high patient-to-nurse ratio).
The PlanCAD Project, resulting from a combination of several administrative databases, helps to make a distinction between practising nurses in the healthcare sector and those who are professionally active on the Belgian labour market, also in other sectors than healthcare.

The primary indicator is the number of practising nurses in the healthcare sector. This indicator covers nurses employed in the healthcare sector who have provided a certain volume of activity (at least 893 procedures per year, according to the INAMI nomenclature), or who are working for an employer who is practising within the healthcare sector.

The secondary indicators are:

• Number of nurses licensed to practice (holding a Licence to practice), i.e. nurses who have obtained a recognised Nursing Care diploma, including people who hold a Midwifery diploma recorded in the nursing category of the cadastre database.
• Number of professionally active nurses, i.e. nurses who are registered as independent nurses or who are employed with at least 0.1 Full-Time Equivalents (FTEs) per year, but who do not necessarily work within the healthcare sector.

The distinction between practising nurses and professionally active nurses is the sector in which they work. Practising nurses work only within the healthcare sector and not on the rest of the Belgian labour market.

RESULTS

• In 2016, there were 124 196 practising nurses within the healthcare sector (primary indicator), i.e. 105 718 FTEs.
• In 2016, there were 202 402 nurses licensed to practice and 143 470 professionally active nurses on the entire labour market.
• Most professionally active nurses work with an employee status in hospitals (65.5%), followed by nursing and care facilities (13.7%) and home care (7.9%).
• The number of practising nurses within the healthcare sector per 1000 population has increased between 2004 (8.8 per 1000) and 2015 in Belgium (10.8 per 1000). This upward trend can also be observed in other European countries. The number for Belgium is slightly higher than the European EU-11 average (which is 9.4).

Figure 5 - Number of practising nurses, per 1000 population, international comparison, 2015
Data source: OECD Health Statistics 2018; EU average= based on 11 of the EU-15 countries for which data were available

Note: At the OECD level, inclusion and exclusion criteria differ among countries, which makes international comparisons of these data unreliable.

Link toward the technical sheet and detailled results 

 Number of nurse vacancies in hospitals (A-7)

Since the early 2000s, nurse shortages have been reported in most industrialised countries, including Belgium. This problem could accelerate with population ageing (and the ageing of nurses).

The measurement of the number of vacancies is a contextual indicator which helps monitor this nurse shortage.

RESULTS

• On 31/12/2016, there were 1274 nurse vacancies in Belgian hospitals, most of which were for holders of Bachelor-level degrees (A1):
  • 708 in Flanders
  • 290 in Wallonia
  • 276 in Brussels

Link toward the technical sheet and detailled results

Average number of patients per nurse in hospitals (A-8)

The international scientific literature shows that nurse shortages have negative consequences on quality of care and mortality rates among hospitalised patients. It is therefore vital to use wisely the few nursing personnel resources that are currently available in order to limit this negative impact as much as possible. The number of Nursing Hours per Patient Day in Acute Hospitals would therefore be an interesting indicator in this regard.

However, the data that would allow to calculate this indicator are no longer collected due to administrative modifications. This indicator is therefore replaced by an ‘older’ indicator, which is the average patient-to-nurse ratio (average number of patients by one nurse in FTE). This indicator is derived from a European study (RN4CAST, 2009-2010) based on data from 500 hospitals in 12 countries. In 2019, the KCE has performed an update of this study using the same methodology for Belgium only. This indicator can also provide indirect information about quality of care in hospitals.

RESULTS

• Between 2011 and 2015, the number of nursing hours per patient day increased from 4.3 to 6.3 in surgery departments and from 4.8 to 5.5 in medical services.
• The number of nursing hours per patient day varies significantly from one hospital to another. Part of this variability may be explained by differences in care intensiveness, but the variations persist after adjusting for this factor.
• According to the RN4CAST study, the average patient-to-nurse ratio in Belgium was 10.7 in 2010, a high figure compared to other European countries (EU-12 average = 9). Only Germany (13.0) and Spain (12.6) have higher ratios. In Norway, this ratio is lower by one half (5.4). If less qualified nursing staff is also taken into account, the average patient-to-nurse ratio becomes 7.9.
• According to the KCE study, the average patient-to-nurse ratio has slightly improved to 9.4 in 2019 but this is still above what is internationally considered as a safe patient-to-nurse ratio (in England for example, based on the NICE guidance, a patient-to-nurse ratio superior to 8 is considered as manifestly unsafe).

Link towards the technical sheet and detailled results

Link toward the KCE study