Navigation – Plan du site

AccueilRubriquesEspace, Société, Territoire2021Dynamics of the geographical dist...

2021
978

Dynamics of the geographical distribution of private physicians in Algeria

Dynamique de la répartition géographique des médecins libéraux en Algérie
Dinámicas de la distribución geográfica del libre ejercicio de la medicina en Argelia
Ahcène Zehnati

Résumés

En Algérie, les médecins libéraux s'installent beaucoup plus dans les régions du nord du pays, et de préférence dans les grandes métropoles, que dans le reste des régions. L’objectif de cet article est, premièrement, d’analyser les inégalités et l'évolution de la répartition géographique des médecins libéraux algériens et, deuxièmement, d’estimer les principaux déterminants de leur densité médicale en 1998 et 2017. Les inégalités sont mesurées par l'indice de Gini et le coefficient de variation. Les modèles économétriques (statique et dynamique) ont été estimés par la méthode des moindres carrés ordinaires. Nos résultats montrent que les disparités globales ont légèrement augmenté entre 1998 et 2017. Les mêmes wilayas (départements) bien dotées continuent toujours d'attirer les nouveaux médecins. L'estimation économétrique suggère que la densité de la population et les dépenses de la sécurité sociale par habitant sont les principaux déterminants de la localisation géographique des médecins, à la fois dans les modèles statiques et dynamiques. En définitive, l'augmentation du nombre de médecins formés ne résout pas automatiquement le problème de leur répartition géographique, notamment dans un contexte caractérisé par l'absence de toute politique publique à l'égard du secteur privé des soins.

Haut de page

Texte intégral

Introduction

1The unequal geographical distribution of physicians is not unique to Algeria. It concerns many countries around the world. Several studies address the issue of location disparities (Ono et al., 2014; Kroezena et al., 2015). They report a high concentration of health personnel in urban and affluent areas to the detriment of semi-urban and rural areas, described as "medical deserts" where under-medicalization is evident. While such studies have quantified the disparities in the spatial distribution of physicians (Gravelle, Sutton, 2001 ; Horev et al., 2004) little research has examined the determinants of physicians’ location-related decisions. According to the World Health Report (WHO,2006), current human resources problems in healthcare are worsened by the imbalances that exist between academic training and the real needs of the population. Such needs will increase in developing countries like Algeria. This latter faces a double demographic and epidemiologic transition.

  • 1 The precise definition of induced demand (McGuire (2000) is follows: Physician-induced demand exist (...)

2In general, before choosing their location, physicians are attracted by high net income and social amenities. Specialists are more sensitive to both net income and amenities comparing to primary care physicians (Falcettoni, 2019). Conforming to healthcare economics literature, the healthcare market is far from competitive because physicians are able to induce their own demand1 to ensure an income that cannot be reduced by competition. Therefore, they can continue to work, in very competitive and even saturated regions. In this situation, the healthcare market cannot tend towards a fair distribution of physicians. This market failure may argue for public intervention in order to encourage physicians to settle in underserved areas. Empirically, these incentive policies are not always effective (Bourgueil et al., 2006).

3In the Algerian health system, the patient has complete freedom to access a healthcare provider, whether public or private. Access to the public sector is free and user financial participation is very low. Conversely, the private sector is for-profit and access to it is chargeable. In the public sector, doctors do not choose their place of practice since it is the Ministry of Health that assigns them according to the needs of public hospitals. On the other hand, in the private sector, freedom of location is enshrined.

4Algeria has made great efforts in training healthcare professionals. The reform of medical studies in 1971 loosened the constraints on access to medical school curriculum and enabled the expansion of the training supply of medical school. This has resulted in a significant increase in the number of trained doctors (general practitioners: GPs and specialists). In recent years, some major trends have emerged, especially after the expansion of the private healthcare activities. The dynamic of physicians’ location concerns mainly the specialized medicine. The specialists benefiting from larger social valuation of the population have widely invested the private sectors with its two components: ambulatory medicine and private clinics. We assist the rapid growth of private specialists whose number in certain specialties exceeds those of the public sector. The attractiveness of certain specialties (gyneco-obstetrics, hemodialysis, and surgery) in private practice is linked with the expected financial benefits (Zehnati, 2013). The private physicians settle much more in the northern regions of the country and preferably in the larger metropolises than in the rest of regions. This situation contributes to the pronounced geographical disparities in the distribution of Algerian physicians. The freedom that Algerian private doctors have in choosing their practice location have probably affected their geographical distribution (Zehnati,2018). The private physicians’ location is mainly influenced by certain factors, mainly market mechanisms. In the Algerian context, social security funds only partially cover household health expenditures; the financial burden on patients is quite high. This constitutes a constraint for doctors to settle in under-populated regions where the purchasing power of the inhabitants is low. The concentration of private physicians in large cities contrasts with the lack of doctors in more remote areas where the population is poorer and more dispersed. It is therefore important to know the determinants of physician location so that the government can implement targeted public policies to address market failures and ensure optimal geographic distribution.

5In Algeria, no studies have been conducted to analyze the main determinants of physicians’ location in the country. A microeconomic survey of a sample of 68 private specialists (Zehnati,2008) shows that family attachment is a major factor in the location choice for 50% of doctors. The same study revealed that 28% of them cited personal reasons and 13% economic reasons (affordable rents and cost of living). Finally, 9% of respondents chose their location based on the opinions of their colleagues. These results demonstrate the importance of family and cultural considerations for Algerian physicians in the choice location of their professional practices location. Most of the studies carried out on this theme distinguish between GPs and specialists. In general, it can be assumed that GPs consider other GPs and specialists as competitors. In contrast, specialists’ main competitors tend to be physicians in the same specialty (Correia, Veiga, 2009).

6In spite of the increasingly important place occupied by the private sector in healthcare provision, no research has been devoted to inequalities in access to healthcare. The purpose of this study is to analyze the disparities in — and evolution of — private physicians’ geographical distribution, and to estimate the main determinants of their population density. The paper is organized as follows: an overview of the literature is presented, followed by a description of data and methods. The results section presents and discusses the findings, followed by a conclusion and a discussion of the limitations of the analysis.

Factors influencing physicians’ location: overview of literature

7A few studies have been conducted examining the determinants of physicians’ location in different contexts (Kuhn, Carsten, 2009; MungaMaestad, 2009; Correia, Veiga ,2009; Gachter et al.,2012; Dumont et al., 2016). Most of these studies use cross-sectional data. Their results are miscellaneous. Some studies have highlighted the influence of market forces in the location-related decisions of physicians. With American data, Newhouse et al. (1982a) show that the inequities in the geographical distribution of physicians are related to the market's own functioning mechanisms. They conclude that the location decisions are a function of agglomerations size. As expected, larger cities are more likely than smaller cities to attract physicians in each specialty. In addition, the strong growth in the number of specialists translates into a faster diffusion in small towns and at the same time they lose their physicians more quickly. As the number of physicians has increased, medical and surgical specialists are more likely to locate in small towns in the United States (Newhouse et al., 1982b). They also show that the increase in the number of physicians has had only a slight impact on the reduction of disparities observed in their geographical distribution.

8Foster and Gorr (1992), found that while an increase in the supply of physicians has affected the localization trends of GPs to the less endowed regions, the same trend has not occurred with specialists. Using data for Portugal from 1996 and 2007, Correia and Veiga (2009) examined inequalities in the distribution of physicians. They estimate a static model and find a positive association between the size of population and the density of physicians, especially for specialists. Municipalities with a larger supply of public hospital beds tends to attract more physicians. This result confirms the hypothesis that the concentration of National Health System (NHS) resources generates a concentration of physicians. They also test a dynamic model to explain the percentage change in the number of physicians per 1000 inhabitants in each municipality. Their results show that at a certain degree of competitive intensity, physicians tend to avoid locating themselves in the same agglomerations. Nevertheless, the competition factor seems to have a less important effect on specialists than on GPs. This is consistent with the hypothesis of Newhouse et al. One of the explanations is the average age of GPs. It is likely that most of these young physicians are still in training and therefore more willing to spread geographically.

9Gächter et al. (2012) considers the density of private physicians (GPs and specialists) in the context of coexistence of public and private health care providers. Using a panel of data on 121 Austrian districts between 2002 and 2008, they analyze how the densities of private and public providers of outpatient care (GPs and specialists) are related to each other. Their results show that the density of private specialists is positively associated with the density of private GPs, but it is not significantly related to the density of GPs in the public sector. They also observe a negative effect in the relationship between the densities of private and public GPs and some private and public specialists, indicating the presence of competitive forces between the private and public sectors in the provision of ambulatory care.

10A geographic concentration of income appears an important factor in explaining the uneven distribution of physicians. Neoclassical theory assumes that physicians seek to maximize their profit and therefore tend to practice in high-income areas. The existence of a positive relationship between the number of physicians and the level of income has been empirically proven (Benham,1968; Bolduc,1996; Bui,2002; Delattre, Samson,2012; Correia, Veiga, 2009). This research concludes that the number of physicians per capita increases with the level of income in the region. The municipalities with high purchasing power attract physicians as they offer a larger potential market for private health care practice. As a result, a more equitable distribution of income between regions could contribute to reducing imbalances in the geographical distribution of physicians (Correia, Veiga, 2009).

11The strength of financial considerations for physicians’ location-related choices should not minimize the effect of non-monetary factors. Several studies (Bolduc,1996; Bui,2002; Bilodeau, 2006) show that physicians are not very sensitive to financial motivations and they appear as a secondary criterion in location choices, after non-monetary benefits (family proximity, recreational facilities, accessibility of knowledge, etc.). Indeed, physicians maximize their utility when choosing a practice location, not just their profits (Foster, Gorr,1992; Lehmann et al.,2008). A number of studies have suggested that physicians would tend to settle in the region in which they studied medicine and defended their thesis (Watson,1980; Kristiansen, Forde, 1992; Bessiere et al., 2004; Delattre, Samson, 2012). This is explained by the knowledge of the working environment, the creation of professional networks and the easy access to the various medical services. The strong correlation between the location of physicians and the presence of medical schools can also be explained by the fact that university hospitals are attractive to physicians, particularly because of the availability of technology and faster diffusion of knowledge and innovation.

Materials and methods

Data

12The data used come from multiple sources. Firstly, there are those relating to the public and private healthcare provision drawn from the statistical yearbooks of the Health Ministry and Population (MSPRH). The data include information about the number of physicians (GPs and specialists) per wilaya, and other relevant information concerning the provision of healthcare in the wilaya. Statistical series were reconstructed, and other data calculated for 1998 and 2017. Secondly, data relating to healthcare expenses of insured persons and relatives were collected at the level of the National Social Insurance Fund (CNAS). Finally, those concerning the socio-demographic data come from the general population and housing census (RGPH) of different years.

13The main limitation of data is the absence of information about purchasing power in each wilaya. GDP by region is often used to measure purchasing power in the region. If these data already exist, we can apprehend and decipher the financial attractiveness of the location especially since the focus of the study is on private healthcare where the patients do not benefit from the full collective insurance coverage. To compensate for this lack of information, we tested an indicator from the RGPH: the percentage of the population with a second home. This variable proved to have no effect on specialists or GPs in 1998 and 2017. This induces a strong financial constraint for the population in accessing private care providers. In the absence of data on income distribution by wilaya to capture the solvent population that can access private care, we considered the size of the resident population as a measure of market size and the volume of demand in the wilaya. We expect a positive relationship between physicians’ location and population size.

Variable(s) of interest

14Generally, the variable used to study the geographical distribution of physicians is the number of physicians per 1000 inhabitants. It is easy to build it. Ordinarily, it is common to distinguish between GPs and specialists. In general, we can assume that the GPs regard both other GPs and specialists as competitors. In contrast, for specialists the competitor tends to be a physician in the same specialty. These differences suggest that we should examine the determinants of the location of the two types of physicians separately (Correia, Veiga, 2009).

Measuring geographical inequalities

Gini index

15The Gini index is an indicator often used to synthesize inequalities in income (standards of living, etc.). The Gini index ranges from 0, indicating perfect equality (where everyone receives an equal share), to 1, perfect inequality (where only one recipient or group of recipients receives all the income). When the value of the Gini index falls between two periods, this indicates an overall decrease in inequality. Conversely, an increase in the value of the index reflects an overall increase in inequality. In our case, we calculated the Gini index for the years 1998 and 2017 to confirm the reduction or widening of physicians’ geographical inequalities. The following formula is used for this purpose (Brown, 1994).

G: Gini index
Yi: Cumulative proportion of the physicians (total, GPs, specialists) in the
ith wilaya
Xi: Cumulative proportion of the population variable in the
ith wilaya
k: total number of wilaya

16In the operationalised using of this formula, Gini index were derived from the Lorenz curve by plotting the wilaya having the highest population per physician (starting from the worst to the best among the 48 wilayas) and the corresponding cumulative population ratio of the region to the cumulative physician number of that wilaya.

Coefficient of variation

17The coefficient of variation (CV) is the ratio of the standard deviation to the mean. A higher coefficient of variation signifies a greater level of dispersion around the mean. Generally, it is expressed as a percentage. Without units, it allows for comparison between distributions of values whose scales of measurement are not comparable.

Econometric models

18We conduct econometric modeling to analyze the main determinants of the geographical location of GPs and specialists in private practice in Algeria. Two types of models were tested: (1) a "static" model to understand the physicians’ distribution in 1998 and 2017. (2) The "dynamic" model assesses changes in the distribution of physicians over the period 1998–2017. The models were estimated by ordinary least squares (OLS). Cook-Weinberg diagnostic tests revealed heteroskedasticity. Therefore, the standard errors (SEs) were corrected using the Huber-White method. Multicollinearity was explored using the regression diagnostic measure variance inflation factor (VIF). All statistics treatment were done with the STATA® software 15.1.

19The following sections (Static models and Dynamic models) describe the variables considered in the models.

Static models

20Static model is presented in the equation (1).

ln PHYSIC (i,j) = α + β1 lnPOP+ β2 lnSSE+ β3 lnBEDSPUB + β4 lnBEDSPRIV + β5 PRESUH + µi ….. (1)

21Where PHYSIC denotes the ratio of the number of GPs and specialists per 1000 inhabitants for every wilaya, for GPs and specialists, and j=G,S (GPs, specialist), i= wilaya. The distribution of physicians by specialty is heterogeneous. The specialties most represented in private practice are gynecological-obstetrics, pediatrics, ophthalmology, internal medicine and cardiology, respectively. In contrast, specialties such as thoracic surgery, nuclear medicine or legal medicine are poorly represented.

22The variable(s) of interest is the logarithm of the ratio of GPs and specialists’ number per 1000 inhabitants per wilaya [ln (GPs) and ln (specialists)]. The size of the resident population is a measure of market size and the volume of demand in the wilaya. We expect the (logarithm of) population size (lnPOP) to be associated positively with the location of physicians, particularly for specialists. The attractiveness of the healthcare market in the wilaya is reflected by the (logarithm of) social security expenditures per inhabitant (lnSSE). In 2017, 19 wilayas (out of 48), all located in the north of the country and more densely populated, will have a per capita expenditure higher than the national average (5548 Algerian dinars, equivalent to US$43). 9 wilayas located in the south of the country record an average per capita expenditure of 2000 DA (US$15). This variable measures the market solvency which can increase the demand for physicians’ services. We expect the sign to be positive for GPs and specialists.

Table 1: Descriptive variables

Variables

Mean (SD)

Min-Max

ln PHYSIC (GPs) 1998

-1,932 (0,517)

-3,797 ̶ -1,326

ln PHYSIC (Specialist) 1998

-2,507 (0,658)

-4,357 ̶ -1,145

ln PHYSIC (GPs) 2017

-1,764 (0,395)

-2,861 ̶ -1,008

ln PHYSIC (Specialist) 2017

-1,633 (0,726)

-3,765 ̶ -0,082

ln Pop 1998

13,098 (0,785)

10,374 ̶ 14,756

ln Pop 2017

13,469 (0,701)

11,155 ̶ 14,982

ln SSE 1998

6,173 (0,450)

4,931 ̶ 6,902

ln SSE 2017

8,37 (0,793)

6,084 ̶ 9,758

lnBEDSPUB 1998

0,550 (0,331)

-0,140 ̶ 1,228

lnBEDSPUB 2017

0,561 (0,314)

-0,055 ̶ 1,52

lnBEDSPRIV 1998

-4,709 (1,723)

-7,182 ̶ -1,410

lnBEDSPRIV 2017

-3,074 (1,438)

-6,515 ̶ -0,639

PRESUH 1998

0,208

0-1

PRESUH 2017

0,229

0-1

TPHYSIC (GPs)

0,23 (0,396)

-0,318 ̶ 2,016

TPHYSIC (Specialist)

1,518 (0,786)

-0,258 ̶ 3,90

DENSITY98 GPs

0,160 (0,058)

0,022 ̶ 0,265

DENSITY98 GPs SQUARRED

0,022

0,000 ̶ 0,07

DENSITY98 Specialist

0,098 (0,06)

0,012 ̶ 0,318

DENSITY98 Specialist SQUARRED

0,013 (0,017)

0,000 ̶ 0,101

VARPOP

0,47 (0,275)

0,076 ̶ 1,698

VARSSE

9,95 (6,51)

-0,381 ̶ 27,109

VARBEDSPUB

0,029 (0,198)

-0,352 ̶ 0,47

VARBEDSPRIV

1,015 (1,227)

-0,359 ̶ 5,401

VARPRESUH

0,25

0-1

23The (logarithm) provision of beds per 1000 inhabitants in public health facilities (lnBEDSPUB), in private sector (lnBEDSPRIV), and the presence of a University Hospital (PRESUH), reflect the scale of public and private healthcare resources in the wilaya. We distinguish between the public and private sectors primarily because we anticipate a strong relationship between the density of private specialists and the availability of resources in the private healthcare. The proximity of high-performance technical platforms for specialists is very important for their location choice. Private specialists prefer to take care of their patients in private clinics with a higher financial compensation. They are very sensitive to financial incentives (Zehnati, Peyron, 2013). We expect the estimated coefficients of the first variable to be negative and those of the second and third variable to be positive. In Algeria, all the University Hospitals are located in wilayas with a medical school. The presence of UHs positively affects physicians’ location decision behaviors as result a high number directly targets these structures, which seem to be encountering greater difficulties to fulfilling and responding to this increasing demand. Private providers could serve as a second resort for patients who have not been admitted to public hospitals. In addition, PRESUH may capture the attractiveness of the wilaya in terms of access to knowledge, technology, and private medicine.

Dynamic models

24The dynamic model is specified in order to examine the effects of observed variations over the period 1998-2017 on physicians’ location choices. The model is written as follows:

TPHYSIC (j,i) = α + β1 DENSITY98 + β2 DENSITY98 SQUARRED+ β3 VARPOP + β4 VARSSE+ β5 VARBEDSPUB + β6VARBEDSPRIV + β7 VARPRESUH + µi ….. (2)

where j=G, S (GPs, specialist), i=1,..…,48.

25TPHYSIC is the dependent variable which denotes the percentage change in the number of physicians per 1000 inhabitants in each wilaya. We take into account all the variables of the static model and we calculate the variation between 1998 and 2017.

26DENSITY98 tells us about medical density in the beginning of the period and its effect on the new physicians’ practices, locations, and choices. Before making any location-related decisions, doctors are interested in knowing the existing number of doctors in their field of specialty or the high presence of their colleagues (GPs) in the wilaya. The concentration of physicians should have a negative influence on location-related decision-making. When the number of physicians in cities is already high, fewer new physicians choose to locate there in order to avoid competition between colleagues. This situation is not observed for doctors who practice in complementary specialties. The opposite effect can also occur when we refer to the theory of supplier-induced demand in healthcare: physicians tend to settle in regions with a higher medical coverage rate (Evans,1974). For instance, in France, it is shown that when the density of GPs in the conventional sector increases in their ​​practice area, doctors compensate for the decrease in the number of patients by increasing the volume of care that they deliver during each consultation. These behaviors are more marked in departments where the density is high (Delattre, Dormont,2003,2005).

27The growth rate of the civilian population in the wilaya between 1998 and 2017 (VARPOP) addresses the importance of demand growth in location decision-making. The effect of resident population growth on the increase in medical density is theoretically ambiguous. Nonetheless, we expect it to be positive in the specialists’ regressions and negative for GPs’, reflecting the competition/substitution between them (Newhouse et al., 1982b; Wanzenried,Nocera, 2008) : specialists tend to locate in growing and larger municipalities while GPs may avoid settling in growth areas because of increasing competition from specialists.

28During the covered period of time, the health care basket covered by the social security fund has been widened. The variable VARSSE is expected to have a positive effect on the density of GPS and specialists as long as social security coverage facilitates access to health professionals. VARBEDSPUB stands for the percentage of the variation in the number of beds in the public sector per 1000 inhabitants in the wilaya between 1998 and 2017. Those wilaya with a dense public supply can discourage the private physicians from settling there, principally for GPs. The variable (VARBEDSPRIV) can have a very positive effect on physicians’ location. The better endowed the wilaya is with private hospital beds, the more private physicians will tend to settle there. This observation is valid especially for physicians in surgical specialties, who earn the most significant part of their income from their activities in private clinics.

29VARPRESUH is a dummy variable indicating whether the wilaya acquired a new central hospital during the period. These variables are thought to capture the effects of changes in the availability of public health resources. We expect this variable to have positive signs, since the private healthcare dynamics feed on the dysfunctions of the public sector. Therefore, the availability of public resources should attract private physicians.

Results

Evolution of the number of physicians and their geographical distribution in Algeria

30Overall, the number of practitioners has increased constantly during the period 1998-2017. It has been multiplied by a factor of 2.57 in almost 20 years. From 25,710 physicians in 1998, the number of physicians has grown to 66,270 in 2017 (Tab 2). At the same time, the population increased by about 38% during this period. The coverage ratio of the population is steadily improving. It went from 0,85 doctors per 1000 inhabitants in 1998 to 1,59 doctors in 2017. The improvement in coverage ratios is due to the combined effects of the creation of new jobs in the public sector within the health system development program and the dynamics of private doctors’ location. It is therefore 53% of physicians who are GPs in 2017 of which a strong majority work in the public sector. On the other hand, specialized medicine records a strong dynamic mostly drawn by the private sector.

Table 2. Evolution of the number of physicians and population size in Algeria

Year

Number of physicians

Annual growth rate (%)

Population (10³)

Annual growth rate (%)

Physicians per 1000 Inhabitants

Annual growth rate (%)

1998

25 710

3,15

30 193

1,64

0,85

1,49

2001

28 755

3,80

31 452

1,37

0,91

2,40

2006

31 927

2,11

33 481

1,26

0,95

0,85

2010

42 326

7,30

35 978

1,81

1,18

5,39

2013

50 326

5,94

38 297

2,10

1,31

3,76

2015

55 656

5,16

39 963

2,15

1,39

2,95

2017

66 270

9,12

41 721

2,18

1,59

6,80

31Table 3 indicates that the greatest increase in the size of the private physicians’ workforce between 2008–2017 occurred among specialists. The number of specialists doubled during the observed period. As a result of the increase in medical education towards specialization, specialists accounted for more than 60% of the total number of private physicians in 2017. On the other hand, general medicine knows a limitation in its extension (Zehnati,2013).

Table 3. Evolution of the number of specialists and GPs in private sector and population size in Algeria

Year

Specialists

Growth rate (%)

Percentage of total (%)

GPs number

Growth rate (%)

Total

2008

6 645

 

50,40

6 548

 

13 243

2009

7 032

5,82

51,60

6 598

0,76

13 630

2010

7 491

6,53

52,55

6 764

2,52

14 255

2011

7 650

2,12

52,72

6 860

1,42

14 510

2012

8 864

15,87

57,85

6 458

-5,86

15 322

2013

9 429

6,37

59,26

6 482

0,37

15 911

2014

10 116

7,29

60,32

6 654

2,65

16 770

2015

11 001

8,75

59,54

7 477

12,37

18 478

2016

12 031

9,36

60,14

7 974

6,65

20 005

2017

13 054

8,50

60,40

8 557

7,31

21 611

32However, the statistics presented above conceal disparities between the wilayas. Maps 1 and 2 show the density of private doctors per 100,000 inhabitants per wilaya in 1998 and 2017.

  • 2 The numbers on the maps indicate the wilaya's identification number.

Map 1: Private physicians density per 100,000 population in 19982

Map 1: Private physicians density per 100,000 population in 19982

33In 1998, the national average density was 33 physicians per 100,000 people, the median was 28 and the standard deviation 12.83. Two wilayas in the far south had no private physicians. The highest density was recorded by the capital Algiers with 59 doctors per 100,000 inhabitants. 33 wilayas had a density lower than the national average, 2 wilayas coincided with the national average, and 13 wilayas had a density higher than the national average. Algeria is divided into four groups. The first group is represented by the 4 wilayas of the great south. The private physicians’ density is less than 10 doctors per 100,000 inhabitants (medical deserts). The second group concentrates 27 wilayas belonging to different regions of the country with a density between 10 and 30 doctors per 100,000 inhabitants. The third group comprises 13 wilayas. Six of them are headquarters of university hospitals. The recorded medical density is between 30 and 50 doctors per 100,000 inhabitants. The fourth group is represented by 4 wilayas located in the north of the country where the density is higher than 50. The increase in the number of trained doctors has improved the general level of densities. However, the geographical imbalances observed in 1998 continued. It is still the same wilayas that continue to attract new doctors. In 2017, the national average densities are 56 doctors per 100,000 inhabitants, the median is 41 and the standard deviation is 24.98. The highest density was recorded by the capital Algiers with 132 doctors per 100,000 inhabitants. The lowest density was recorded by the wilaya of the far south (Illizi) with 5 doctors. 35 wilayas have a density lower than the national average, and 13 wilayas have a density higher than the national average. The fourth density group in 2017 is not very different from that of 1998. The first group has expanded to 9 wilayas with densities of less than 30 doctors per 100,000 inhabitants. The second group concentrated 28 wilayas with a density of between 30 and 60 doctors per 100,000 inhabitants. The third group represented 10 wilayas located mostly in the north of the country. The fourth group is made up of two wilayas: the capital Algiers and Annaba, located in the far east of the country.

Map 2: Private physicians density per 100,000 population in 2017

Map 2: Private physicians density per 100,000 population in 2017

34After presenting the evolution of densities by wilaya between 1998 and 2017, the next section examines the measurement of inequalities in the distribution of physicians.

Inequalities in distribution of private physicians

35For all private physicians, the inequalities between the wilayas were widened between 1998 and 2017. The estimate of the Gini index increased from 0,47 to 0,5 and even suggests a slight increase in the inequality in the distribution of physicians. Greater inequality in the distribution of physicians is observed among specialists rather than among GPs.

Table 4. Inequality measures of geographic distribution of physicians per 1000 inhabitants. CV Coefficient of variation

1998

Gini

2017

Gini

1998

CV

2017

CV

Physicians

0,47

0,5

1,16

1,35

GPS

0,41

0,41

0,89

0,99

Specialists

0,54

0,56

1,59

1,58

36Nonetheless, the distributions of both GPs and specialists per 1000 inhabitants have become slightly more unequal. The estimated CVs also suggest that inequality did not change dramatically during the period. These results confirm the conclusion of similar analyses conducted in other countries (Barer, Stoddart,1999; Hann,Gravelle, 2004). Despite the increase in the number of physicians, the overall inequality in geographical distribution has not decreased. The population grew at a slower rate than the number of private physicians during this period. When comparing the growth rates of private doctors and the population, it can be seen that the growth rate of private doctors is higher than that of the population for all wilayas except for one wilaya located in the far south (Tindouf 37) where the number of private doctors was multiplied by a factor of 2.33 compared to 2.70 for the population. Graph 1 does not provide further information, but it makes inequalities visible. We note that the geographical distribution of GPs is better balanced than that of specialists. The curve of GPs is closer to the bisectrix than that of specialists.

Graph 1 : Lorenz curve of private specialists and GPs

Graph 1 : Lorenz curve of private specialists and GPs

37The two curves in 1998 and 2017 coincide, evidence of a lack of improvement in the distribution of specialists and GPs. Graph 1 shows that 80% of the wilayas (39) locates only 48% of specialists (versus 55% for GPs). The nine remaining wilayas concentrated 52% of the total (versus 45% for GPs). The inequalities’ distribution increased slightly over the period considered. In 1998, the 10% of wilayas with the highest density of GPs were closer to the equilibrium than in 2017. In about ten wilayas, the density of specialists is higher than that of GPs. The increase in inequalities could be due to overconcentration in the already very dense wilayas. The total freedom enjoyed by private physicians in the choice of their location (among other factors) contributes to maintaining - or even worsening - the geographical distribution of private physicians.

Determinants of physicians’ geographical location

Static models

38The results of the static model of GPs and specialists for 1998 and 2017 is presented in table 5. The R² of the two estimated models suggests that the variation in the ratio of physicians per 1,000 population is explained by changes in the explanatory variables mainly for specialists in 2017. In 1998, the R² is practically the same for GPs and specialists.

39The most important variable for the distribution of physicians is population and SSE, while BEDSPRIV are relatively more important than the remaining variables in both regressions. A positive association exists between the population size (market size) and the number of physicians per 1000 inhabitants (GPs and specialists). In fact, the population size impacts the geographic location of specialists slightly more than those of GPs. This result is in line with expectations, since the financial profitability of some specialties requires a high population base. In Algeria, the specialists are very concentrated in larger urban areas. Even if they practice in some towns with a lower density of population, they still prefer to settle in the principal town of the wilaya. Some of them concentrate between 80 and 100% of specialists, thus contributing to the increase in intra-wilayas inequalities. The situation can also vary according to the specificities of the wilaya. For instance, those with communes close to large metropolitan areas may attract a larger number of doctors than the chief towns of the wilaya.

Table 5. Geographic location —ordinary least squares (OLS) regression estimates Static model (1998 and 2017)

 

1998 

2017

ln (GPs)

ln (Specialists)

ln (GPs)

ln (Specialists)

ln POP

0.319***

0.472***

0.230***

0.278**

(0.107)

(0.0867)

(0.0791)

(0.115)

ln SSE

0.487**

0.842***

0.133**

0.417***

(0.193)

(0.129)

(0.0494)

(0.0715)

ln BEDSPUB

-1.060**

0.0676

-0.124

0.112

(0.406)

(0.238)

(0.150)

(0.217)

lnBEDSPRIV

0.113*

0.0358

0.111***

0.143***

(0.0667)

(0.0375)

(0.0302)

(0.0438)

PRESUH

0.204*

0.321**

-0.0881

0.214**

(0.337)

(0.184)

(0.114)

(0.165)

Constant

-8.042***

-13.77***

-5.552***

-8.555***

(1.963)

(1.446)

(1.201)

(1.740)

Observations

48

48

48

48

R-squared

0.751

0.762

0.734

0.834

Robust standard errors (SEs) are in parentheses ***p<0.01,** p<0.05,* p<0.1

40The number of beds per 1000 inhabitants at hospitals (public and private) is chosen as a proxy that reflects physicians’ choices about the distribution of healthcare resources across wilayas. The estimated coefficients indicate that wilayas with larger supplies of private beds tend to attract, on average, more physicians per 1000 inhabitants. This result is consistent with the hypothesis that the concentration of private healthcare resources tends to generate a concentration of private physicians particularly in surgical specialties. However, the magnitude of the estimated coefficient indicates that the elasticity is small. The number of public sector beds per 1000 inhabitants was only significant for GPs in 1998 with a negative coefficient. The social security expenditures (SSE) per inhabitant is an important factor in explaining the geographic location of physicians. This result is more relevant for the distribution of specialists than that of GPs. The presence of UH positively impacted specialists’ location in both 1998 and 2017. The GPs were influenced positively only in 1998. As we reported in the literature review, private doctors prefer to locate in large cities with UH for professional reasons but also for personal considerations.

41In general, doctors who have been settled in the cities for years and who have therefore built up their patient portfolio show low disposition to mobility. To verify this observation, we estimated a regression model on the 1998 data for the number (log) of GPs and specialists per 1000 inhabitants in 2017 (Table 6).

Table 6. Geographic location —ordinary least squares (OLS) regression estimates model (2017 with data of 1998)

 

2017

ln (GPs)

ln (Specialists)

ln POP98

0.384***

0.629***

(0.0560)

(0.104)

ln SSE98

0.319***

0.647***

(0.0834)

(0.154)

ln BEDSPUB98

-0.228

0.0682

(0.153)

(0.284)

ln BEDSPRIV98

0.0376

0.0381

(0.0242)

(0.0448)

PRESUH98

0.102

0.185**

(0.119)

(0.220)

Constant

-8.450***

-13.78***

(0.934)

(1.727)

Observations

48

48

R-squared

0.726

0.734

Robust standard errors (SEs) are in parentheses*** p<0.01, ** p<0.05, * p<0.1

42The estimation reveals that 1998’s variables explain a large share of the physicians’ geographical distribution. The results are comparable to those exposed in Table 5. Meanwhile, the population size and social security expenditures per inhabitant have a stronger impact on physicians’ geographic location in 1998 than in 2017. The number of beds per 1000 inhabitants in the private sector is not significant because of the weak number of private clinics in 1998. These findings seem to confirm the relatively low mobility of physicians.

Dynamic models

43The estimation results for the dynamic models are presented in Table 7. The R² of the two estimated models is 0.508 and 0.548 respectively for GPs and specialists. This result indicates that the relationship is not completely linear. The threshold levels have a linear and a non-linear (parabolic) part, i.e. a U-shaped curve. This shows the existence of competitive factors in the variations in the ratio of the number of doctors per 1,000 inhabitants over the period under review.

Table 7. OLS regression estimates-Dynamic model (1998-2017)

 

GPs per 1000 inhabitants

Specialists per 1000 inhabitants

DENSITY98

-5,285***

-4.916***

(1.159)

(1.490)

DENSITY98 SQUARED

8.208***

8.571***

(2.128)

(2.720)

VARPOP

-1.038**

-3.229***

(0.390)

(0.499)

VARSSE

-0.0226

-0.0943**

(0.0374)

(0.0458)

VARBEDSPUB

0.578***

-0.300

(0.167)

(0.222)

VARBEDSPRIV

0.000730

0.00264**

(0.00100)

(0.00122)

PRESUH

0,00672

0.169**

(0.104)

(0.127)

Constant

1.074***

1.878***

(0.230)

(0.282)

Observations

48

48

R-squared

0.508

0.548

Robust standard errors (SEs) are in parentheses ***p<0.01,** p<0.05,* p<0.1

44The coefficient of the variable DENSITY98 is negative and statistically significant, supporting the hypothesis that, at least from a certain degree of competitive intensity, the physicians tend to avoid locating themselves in these areas. However, competition seems to have a slightly stronger effect on GPs than on specialists. This result is as in line with what was envisaged in our initial hypothesis and in the literature (Newhouse,1982a; Wanzenried, Nocera, 2008).

45The introduction of a threshold effect makes it possible to know from which density threshold the doctor gives up competition. When calculating the density thresholds per 1,000 inhabitants for GPs and specialists (x=-β/2σ), they are 0.32 and 0.28 respectively. The explanation is that the concentration of specialists and population dynamic exert a crowding out effect on GPs, who tend to settle in the least populated wilayas. In addition, most GPs are better able to spread geographically because of competition and their relative youth compared to specialists, where family constraints are heavier. In any case, this question deserves to be to be explored further.

46The growth in the population seems to have adversely affected the growth in the number of GPs and specialists. The estimated coefficient is significant for GPs and much more so for specialists. There is no evidence of adjustments in the number of GPs per 1000 inhabitants following variations in the installed capacity of the national health system in the public and private sectors. In the same direction, the specialists’ geographic location appears not to reflect changes in the installed capacity. The number of specialists per 1000 inhabitants appears to have grown in the most prosperous wilayas with a higher purchasing power.

47The variation in the density of specialists is negatively affected by the variable SSE. The estimated coefficient is significant only for specialists although its magnitude is small. SSE appeared to have a lower impact on the variation in specialists’ density than we would expect. Despite the expenses associated with private clinics, populations turn to private sector, populations turn to it out of necessity, especially in regions where the public sector alternative is failing. Moreover, the main claim of private doctors and the founders of private clinics in Algeria remains the establishment of contractual relations between social security funds and private clinics, which will allow a widening of the market for these care providers. The variation of beds in the public sector positively affects only the variation of the GPS density. Alternately, the specialists’ geographic location is positively associated with hospital beds in the private sector. However, the estimated coefficient is small. Finally, specialists’ geographic location appears to reflect changes in the installed capacity of the public sector through the presence of UH. On the contrary, the number of GPs per 1000 inhabitants is not affected by the variations in availability of the UH.

Discussion

48The inequalities observed at the beginning of the period have been maintained or even slightly worsened. The same well-endowed wilayas continue to attract doctors. The dynamics of the geographical location of private doctors mainly concern specialized medicine. Access to a specialist does not require a first visit to a GPs (gate keeper system). As a result, the dynamics of general medicine are weak.

49The econometric estimation suggests that the factors explaining the geographical location of private physicians have not really changed over the period considered. The explanatory variables related to the demand for care mainly explain geographical location. Physicians appear to be sensitive to population density and social security expenditures. These results seem to be consistent especially for private physicians who must ensure the profitability of their activity. Specialists are more sensitive to SSE differences. Wilayas with higher SSE tend to offer a larger potential market for private medical practice. Higher population density (potential market) with significant purchasing power, spouse employment (for married physicians), and social amenities are factors taken into consideration before defining a utility function reflecting physicians' location choices. Previous studies show a relatively low mobility of physicians once they have settled in a region (Dussault,Franceschini, 2006; Kuhn,Carsten, 2009). Consequently, the variation in the number of doctors per 1,000 inhabitants within the wilayas tends to be low and is mainly explained by the arrival of new graduates.

50In Algeria, the emergence and expansion of private healthcare has been more concentrated in northern wilayas, to the detriment of those in the highlands and southern regions. The liberalization process of the health sector has not been accompanied by measures that would force private doctors to settle in the underdeveloped regions. Freedom of location is therefore total. This (with other factors) seems to have contributed to the retention, even to the accentuation of regional disparities. This problem is not of great importance if the public supply compensates for the deficit of the private supply. If in areas where private doctors are not very present, the public sector covers this absence, the question does not really arise. The problem is precisely this double absence of private and public sector physicians. If the density of public doctors in regions poorly covered by private doctors is acceptable, the material resources do not follow. The interactions between the two sectors must be considered in public policies relating to the provision of care. Unfortunately, the absence of public policy to attract private doctors to these regions is one of the weaknesses of health policy in Algeria. In addition, the feminization of the medical profession poses some difficulties for the healthcare system because women doctors do not wish to practice in remote areas of the northern regions.

51In Algeria, the distribution of specialists is uneven, and disparities persist in the absence of a coercive or incentive policy in favor of private physicians. It is true that even in the case of other countries, despite the establishment of incentive public policies in favor of private physicians, these have not made it possible to reduce inequalities substantially. The most common measures concern financial incentives at different stages in the career of health professionals: initial training, installation, and retention. Studies in the literature (Bourgueil et al., 2006) show that a policy of saturation of supply through the increase in overall numbers (a generous numerus clausus) does not solve the problem of geographical inequalities. Even the financial incentive schemes put in place have had only a moderate effect on geographic location. However, the granting of financial aid during initial training through the allocation of scholarships, loans on preferential terms, etc., accompanied by the obligation to practice in under-developed regions exerts a positive effect in the short term and a negative effect in the long term. "This system is not exempt from perverse effects; doctors buy back the credit before term or leave the place of practice after the obligation to service comes to an end" (Bourgueil et al., 2006)

52Besides, it is legitimate to wonder about the prospect of expanding private healthcare. Two arguments can be discussed. First, we can say that medicine, which was essentially public in the past, has today become a public-private mix. Even if the public sector remains globally dominant, is the state also intended to manage the private supply? Should the state be content to manage the public sector and allow private doctors to settle freely? If so, the state cannot be criticized for having no public policy towards the private sector. We should ask ourselves whether it is problematic in the long term or not. Second, we can say: in the case that inequalities are increasing, whether there are public-private arbitrage effects and whether in the future private supply will be more reimbursed, at that time, there is reason to be worried about imbalances in the geographic distribution of doctors. If action is not taken today, in the future it will be difficult to remedy it. We should have a policy even if we know that elsewhere in other contexts, it is already complicated to implement.

53To our knowledge, the current study is the first to document the geographical imbalances in private physicians’ density in Algeria. The results of this study must be taken with caution due to the lack of data on socio-economic indicators to assess the attractiveness of wilayas. Data such as the GDP per wilaya and the unemployment rate per wilaya, for example, would have been very interesting in analyzing the determinants of private physicians’ location regarding location, particularly since this study is directly linked to the private healthcare market. Also, data about the epidemiological situation of the population by wilaya may be taken into consideration in the study and which may impact the location-related decisions of doctors. Unfortunately, this type of data is not available.

Conclusion

54Algeria presents inequalities in the geographical distribution of private physicians. It is very concentrated in larger cities at the expense of the wilayas of the highlands and the south. The coverage of the population varies widely from one region to another. As is the case in many countries, the increase in the number of trained doctors does not precipitate an automatic rebalance of their distribution, particularly in the absence of a public policy towards the private sector. Despite the rapid growth in the number of private doctors, their geographical distribution is still highly uneven. The distribution of specialists is more unequal than that of GPs.

55The econometric analysis of the variation in the number of physicians per 1000 inhabitants suggests that there are competitive forces in the market for physicians’ services. For the moment, the competition is not strong enough to generate significant reductions in physicians’ distribution inequalities mainly for specialists. The presence of specialists seems to exercise a crowding-out effect on GPs. Same wilayas continue to attract new doctors. The situation remained unchanged over the period under review. Improvements can be made if the social security fund widened the health care basket cover for better medical coverage of the population.

56Finally, it is imperative for the public authorities to integrate the private sector in the national health strategy, considering the growing share that it represents in the care supply. Incentive measures to attract doctors to deficit areas are also desirable. There is no miracle recipe, but at least some experimental measures, such as scholarships for students from underserved regions to pursue medical training with the obligation to return to their region at the end of their training. Such measures give the opportunity to GPs practicing in these regions to access a medical or surgical specialty of their choice, one of many measures which can contribute to the reduction of regional disparities.

Haut de page

Bibliographie

Barer ML., Stoddart GL., 1999, "Improving access to needed medical services in rural and remote Canadian communities: recruitment and retention revisited", Discussion paper prepared for federal/provincial/territorial advisory committee on health human resources, Center health service and policy research UBC,48 pages.

Benham L., Maurixi A., Reder MW., 1968, "Migration, location and renumeration of medical personnel: Physicians and dentists", Review of Economics and Statistics 50, n°3: 332-341.

Bessiere S., Breuil-Genier P., Darinne S., 2004, "La démographie médicale à l’horizon 2025 : une régionalisation des projections", DREES, Études et Résultats ; n°353 : 1-12.

Bilodeau H., Leduc N., Van Schendel N., 2006, "Analyse des facteurs d'attraction, d'installation et de maintien de la pratique médicale dans les régions éloignées du Québec", Groupe de Recherche Interdisciplinaire en Santé (GRIS), Rapport final R06-2 ; 285 pages.

Bolduc D., Fortin B., Fournier MA., 1996, "The effect of incentive policies on the practice location of physicians: a multinomial probit analysis", Journal of Labor Economic, n°14(4); 703–732. dx.doi.org/10.1086/209828

Bourgueil Y., Mousques J., Tadjahmadi A., 2006, "Comment améliorer la répartition géographique des professionnels de santé ? Les enseignements de la littérature internationale et des mesures adoptées en France", Rapport IRDES,70 pages.

Brown MC., 1994, "Using Gini-style indices to evaluate the spatial patterns of health practitioners: Theoretical considerations and an application based on Alberta data", Soc Sci Med; 38(9):1243–56. doi.org/10.1016/0277-9536(94)90189-9

Bui DHD., 2002, "L’installation des médecins : les déterminants du choix géographique (1974-1999) ", Cahiers de sociologie et de démographie médicales ; n°42(4) : 337-358.

Correia I., Veiga P., 2009, "Geographic distribution of physicians in Portugal", European Journal of Health Economic; n°11:383-393. doi.org/10.1007/s10198-009-0208-8

Delattre E., Samson AL., 2012, "Stratégies de localisation des médecins généralistes français : mécanismes économiques ou hédonistes ? ", Revue Économie et Statistique; vol. 455(1) :115-142. https://hal.archives-ouvertes.fr/hal-01685914

Delattre E., Dormont B., 2005, "La régulation de la médecine ambulatoire en France : quel effet sur le comportement des médecins libéraux ? ", Solidarité Santé, DREES ; n° 1 : 135-161.

Delattre E., Dormont B., 2003, "Fixed Fees and Physician-Induced Demand: a Panel Data Study on French Physicians", Health Economics; vol. 12, n° 9: 741-754. doi: 10.1002/hec.823

Dumontet M., Samson AL., Franc C., 2016, "Comment les médecins choisissent-ils leur lieu d’exercice ? ", La Revue Française d’économie ; vol. XXXI, issue 4 : 221-267. doi.org/10.3917/rfe.164.0221

Dussault G., Franceschini MC., 2006, "Not enough there, too many here: understanding geographical imbalances in the distribution of the health workforce", Human Resources for Health; 4:12:1-16. doi: 10.1186/1478-4491-4-12

Evans R.G., 1974, "Supplier-induced demand: Some empirical evidence and implications", in: Perlman M. (eds.), The Economics of Health and Medical Care, pp 162-173, Wiley, New York.

Falcettoni E., 2019, "The Determinants of Physicians’ Location Choice: Understanding the Rural Shortage". Working paper. dx.doi.org/10.2139/ssrn.3493178

Foster S., Gorr W., 1992, "Federal health care policy and the geographic diffusion of physicians: A macro-scale analysis", Policy sciences; n°25:117-134.

Gachter M., Schwazer P., Theurl E., Winner H., 2012, "Physician density in a two-tiered health care system", Health Policy; n°106 (3-10):257–268. DOI: 10.1016/j.healthpol.2012.04.012 

Gravelle H., Sutton M., 2001, "Inequalities in the geographical distribution of general practitioners in England and Wales 1974-1995", Journal of Health Services Research and Policy; vol 6, n° 1: 6-13. DOI: 10.1258/1355819011927143

Hann M., Gravelle H., 2004, "The maldistribution of general practitioners in England and Wales: 1974-2003", British Journal General Practice, 54 (509): 894-98.

Horev T., Psis-Katz I., Mukamel D., 2004, "Trends in geographic disparities in allocation of health care resource in the US ", Health Policy; vol 68, n° 2: 223-232. doi: 10.1016/j.healthpol.2003.09.011

Kristiansen IS., Forde-Helge O., 1992, "Medical specialists' choice of location: The role of geographical attachment in Norway ", Social Science et Medicine; Elsevier, n°34(1): 57-62. DOI: 10.1016/0277-9536(92)90067-z

Kroezena M., Dussaultb G., Craveirob I., Dielemanc M., Jansenc C., Buchande J., Barriballf L., Raffertyf AM., Bremnerg J., Sermeu W., 2015, "Recruitment and retention of health professionals across Europe: A literature review and multiple case study research", Health Policy; Vol 119, Issue 12:1517-1528. DOI: 10.1016/j.healthpol.2015.08.003

Kuhn M., Carsten O., 2009, "Demographic and geographic determinants of regional physician supply", Working Paper; n° 105, University of Rostock.

Lehmann U., Dieleman M., Martineau T., 2008, "Staffing remote rural areas in middle- and low-income countries: A literature review of attraction and retention", BMC Health Services Research; 8:1-10. DOI: 10.1186/1472-6963-8-19

McGuire T., 2000, "Physician agency", in: Culyer AJ., Newhouse JP. (eds.), The handbook of health economics, vol. 1, pp. 462–536. Amsterdam: Elsevier.

Munga M., Maestad O., 2009, "Measuring inequalities in the distribution of health workers: the case of Tanzania", Human Resources for Health; 7:4: 1-12. doi.org/10.1186/1478-4491-7-4

Newhouse J., Williams A., Bennett B., Schwartz W., 1982a, "Does the geographical distribution of physicians reflect market failure? ", Bell Journal of Economics ;n° 13:2: 493-505.

Newhouse J., Williams A., Bennett B., Schwartz W., 1982b, "Where Have All the Physicians Gone? ", Journal of the American Medical Association ; n° 247(17): 2392–2456. 

Ono T., Schoenstein M., Buchan J., 2014, "Geographic Imbalances in Doctor Supply and Policy Responses", OECD Health Working Papers, n° 69, OECD Publishing.

Wanzenried G., Nocera S., 2008, "The evolution of physicians density in Switzerland", Swiss J. Econ. Statist;144(2): 247–282. doi.org/10.1007/BF03399254

Watson C., 1980, "The relationship between physician practice location and medical school area: an empirical model", Soc. Sci. Med; 14D: 63–69. doi.org/10.1016/0160-8002(80)90033-7

World Health Organization., 2006, "The global shortage of health workers and its", Fact sheet n° 302, Geneva.

Zehnati A., 2018, La démographie médicale en Algérie, Alger, Edition Loumi.

Zehnati A., Peyron C., 2013, "Les raisons de la double activité des médecins : le cas de l’Algérie", Revue Maghreb-Machrek. Vol 3 ; n° 217 : 89-112. https://doi.org/10.3917/machr.217.0089

Zehnati A., 2013, "Les cliniques privées en Algérie : l’émergence d’un nouvel acteur dans le paysage sanitaire", Revue les cahiers du CREAD ; n°105-106 :187-208.

Zehnati A., 2008, "Analyse de la dynamique d’installation des médecins spécialistes libéraux : cas de la wilaya de Béjaia (Algérie) ", Rapport de recherche CREAD; 120 pages.

Haut de page

Notes

1 The precise definition of induced demand (McGuire (2000) is follows: Physician-induced demand exists when the physician influences a patient’s demand for care against the physician’s interpretation of the best interests of the patient. Under induced demand, a physician takes an action to shift the patient’s demand curve in the direction of the physician’s own interests. Physicians can affect such a shift, because they have more information regarding the patient’s condition and treatment options than the patient, an example of the market failure known as asymmetric information.

2 The numbers on the maps indicate the wilaya's identification number.

Haut de page

Table des illustrations

Légende G: Gini indexYi: Cumulative proportion of the physicians (total, GPs, specialists) in the ith wilayaXi: Cumulative proportion of the population variable in the ith wilayak: total number of wilaya
URL http://journals.openedition.org/cybergeo/docannexe/image/36723/img-1.png
Fichier image/png, 7,9k
Titre Map 1: Private physicians density per 100,000 population in 19982
URL http://journals.openedition.org/cybergeo/docannexe/image/36723/img-2.jpg
Fichier image/jpeg, 136k
Titre Map 2: Private physicians density per 100,000 population in 2017
URL http://journals.openedition.org/cybergeo/docannexe/image/36723/img-3.jpg
Fichier image/jpeg, 140k
Titre Graph 1 : Lorenz curve of private specialists and GPs
URL http://journals.openedition.org/cybergeo/docannexe/image/36723/img-4.png
Fichier image/png, 54k
Haut de page

Pour citer cet article

Référence électronique

Ahcène Zehnati, « Dynamics of the geographical distribution of private physicians in Algeria », Cybergeo : European Journal of Geography [En ligne], Espace, Société, Territoire, document 978, mis en ligne le 27 avril 2021, consulté le 11 mai 2021. URL : http://journals.openedition.org/cybergeo/36723 ; DOI : https://doi.org/10.4000/cybergeo.36723

Haut de page

Auteur

Ahcène Zehnati

Économiste de la santé, Directeur de recherche au Centre de Recherche en Économie Appliquée pour le Développement (CREAD), Algeria
E-mail : ahcene.zehnati@gmail.com

Haut de page

Droits d’auteur

Licence Creative Commons
La revue Cybergeo est mise à disposition selon les termes de la Licence Creative Commons Attribution 4.0 International.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search