The Burden of Imported Malaria in Portugal 2003 to 2012


  • Ana Glória Fonseca MD, MSc,

    Corresponding author
    1. Public Health Department, Faculdade Ciências Médicas (NOVA Medical School), Universidade Nova de Lisboa (Nova Lisbon University), Lisboa, Portugal
    • Corresponding Author: Ana Glória Fonseca, MD, MSc, Public Health Department, Faculdade Ciências Médicas (NOVA Medical School), Universidade Nova de Lisboa, Campo Mártires da Pátria, 130, Lisboa 1169-056, Portugal. E-mail:

    Search for more papers by this author
  • Sara S. Dias PhD,

    1. Public Health Department, Faculdade Ciências Médicas (NOVA Medical School), Universidade Nova de Lisboa (Nova Lisbon University), Lisboa, Portugal
    2. UIS-ESSLei-IPLeiria, Leiria, Portugal
    Search for more papers by this author
  • Joao Luis Baptista MD, PhD,

    1. Faculdade de Ciências da Saúde, Universidade da Beira Interior, Covilhã, Portugal
    Search for more papers by this author
  • Jorge Torgal MD, PhD

    1. Public Health Department, Faculdade Ciências Médicas (NOVA Medical School), Universidade Nova de Lisboa (Nova Lisbon University), Lisboa, Portugal
    Search for more papers by this author


Increasing international travel and expatriation to sub-Saharan countries where malaria is endemic has raised public health concerns about the burden of imported malaria cases in Portugal. From 2009 to 2012, there was a 60% increase in malaria hospitalizations, contradicting the declining trend observed since 2003. Older age was associated with longer length of stay in hospital and higher lethality.

In Europe, since the year 2000 and contrary to the trend in the preceding years, there has been a steady decrease in the annual incidence of imported malaria: from 15,303 reported cases in 2000 to 5,712 in 2009.[1, 2] In Portugal, approximately 40 to 50 imported cases are notified every year.[1] However, disease notification, though mandatory, is compromised by nonquantified underreporting.[3] In the last few years, a boost has been observed in emigration to regions where malaria is endemic, namely sub-Saharan African countries with which Portugal has always maintained privileged commercial and cultural links, raising public health concerns about rises in malaria incidence.[4, 5]

To assess the malaria burden and trends at the national level, malaria hospitalizations from 2003 to 2012 were analyzed, using Portugal's Diagnosis Related Groups (DRGs) information system from National Health Services (NHS) hospital discharge records.


The retrospective study on malaria hospitalizations was based on data obtained from the Portuguese National database of the DRGs containing records of NHS hospital episodes, provided by the Central Administration of Health System (ACSS). The data are anonymous and available from the ACSS for scientific research. In the DRG database, each record corresponds to an NHS hospital discharge episode (hospitalization) and contains the information collected while the patient was in hospital: gender, age (date of birth), principal and secondary diagnosis, hospital where the patient was admitted, admission and discharge dates, discharge status, procedures, and the patient's district of residence. From within the DRG database, malaria hospitalizations from 2003 to 2012 were selected using ICD-9 CM codes 084 (malaria), namely 084.0 (Plasmodium falciparum), 084.1 (Plasmodium vivax), 084.2 (Plasmodium malariae), 084.3 (Plasmodium ovale), 084.5 (mixed malaria), 084.6 (malaria unspecified), 084.9 (pernicious complication), and 647.4 (malaria in the mother classifiable elsewhere, but complicating pregnancy, childbirth, or the puerperium). Variables included gender, age, year of hospitalization, length of stay (LOS) in hospital, and lethality. A descriptive analysis was carried out, using the usual indicators of frequency synthesis and hypothesis testing. Data were analyzed using R software, namely GLM library.[6]


From 2003 to 2009, there were 1,782 malaria hospitalizations in NHS hospitals in Portugal. Most cases were in males (1,335, 75%, sex ratio 3 : 1) and in those aged between 18 and 64 years (1,527, 86%). Median age was 40 years. The annual number of malaria hospitalizations decreased steadily from 231 cases in 2003 to 130 in 2009, subsequently increasing to 220 in 2012 [Figure 1, number of imported malaria hospitalizations, LOS in hospital, and lethality (Portuguese DRG database) and number of reported cases of imported malaria (Centralized Information System for Infectious Diseases, CISID) in Portugal 2003–2012]. Total mean LOS in hospital was 8.35 days, and lethality was 2.19%.

Figure 1.

Number of imported malaria hospitalizations, length of stay in hospital, and lethality [Portuguese Diagnosis Related Group (DRG) database] and number of reported cases of imported malaria [Centralized Information System for Infectious Diseases (CISID)] in Portugal 2003 to 2012. For the years 2010 and 2011, there is no information available from CISID on the number of reported cases of imported malaria in Portugal.[1]

Mean LOS varied from 6.88 days in 2003 to a maximum of 9.39 days in 2007. Lethality ranged from 0.75% (1/132) in 2006 to 3.85% (5/130) in 2009, with peaks in 2007 (3.3%), 2009 (3.5%), 2011 (3.33%), and 2012 (3.18%). The highest absolute mortality was observed in 2011 (eight deaths) and the lowest in 2006 (one death).

Gender did not modify LOS or lethality (p > 0.05). In those aged over 64, corresponding to 6% of malaria hospitalizations (113/1,782), mean LOS was significantly increased [7.97 days for age <18, 7.94 days for ages 18–64, and 14.39 days for age >64 years; analysis of variance (ANOVA), p < 0.001]. Among adults, lethality was increased in those aged over 64 years (1.7% in those aged 18–64, 11.5% in those over >64 years; Fischer's exact test, p < 0.001). Those who died had a significantly higher mean LOS (31.0 days for those who died and 7.85 days for those who did not die; t-test, p < 0.001).


Between 2009 and 2012, annual malaria hospitalizations in Portugal increased, contradicting the declining incidence also observed in other European countries since 2000.[1, 7, 8]

International travel in general and particularly to regions where malaria is endemic has been increasing remarkably worldwide.[9] In Portugal, economic crises and constraints boosted a new emigration wave that included expatriation to emergent economies in sub-Saharan African countries where malaria is endemic, such as Angola and Mozambique. The numbers of corporate travelers and expatriates to such destinations have been rapidly increasing. Although exact migration data are widely underestimated and difficult to obtain, available official data from consular records indicate that 113,194 Portuguese lived in Angola and 20,415 in Mozambique in 2012: respectively, a 56 and 24% increase since 2008.[5] This changing demographic paradigm entailing longer stays in endemic regions, the use of local infrastructures, more involvement with local communities along with frequent traveling to and from Portugal (eg, in the holiday season), and low adherence to prophylaxis measures may justify the inversion in malaria frequency trends observed since 2009.[10] A similar increase was not observed for LOS in hospital or lethality, which were within the same range of values.

Typically, migration for economic reasons affects more men than women. From 2003 to 2012, there were significant and progressive annual increases in hospitalizations for men compared with hospitalizations for women [from 155 (66.7%) males and 77 (33.3%) females in 2003 to 173 (78.6%) males and 47 (21.4%) females in 2012; chi-square, p = 0.005]. However, when only the period between 2009 and 2012 was taken into account, no significant difference in sex ratio was observed (3.5 : 1 to 3.8 : 1, p > 0.05).

In this study, older age was associated with increased LOS in hospital and lethality, older age already being recognized as a risk factor for death from imported malaria by other authors.[11, 12] Longer LOS was associated with increased lethality, suggesting a possible role for disease or nosocomial-related complications and comorbidities in the risk of death.

The population under study (hospitalized cases) reflects only the number of total malaria cases that were severe enough to need hospitalization, which may be the reason for the relatively high total lethality (2.19%), information on delay in seeking medical care unfortunately being unavailable. Moreover, even though DRG data may not allow for reliable information on Plasmodium species or provide information on delays in seeking medical care or diagnosis and treatment, imported malaria in Portugal is mainly due to P. falciparum from sub-Saharan Africa in general, potentially the most severe and life-threatening malaria, with some cases caused by P. vivax, as documented by hospital-based case series and statutory notification.[1, 3, 13, 14]

The national database of the DRG refers to hospitalization episodes and, though infrequent, one case resulting in more than one hospitalization cannot be excluded. Nonetheless, according to the World Health Organization, notified cases throughout the study period represented roughly 30% of malaria hospitalizations in NHS hospitals, denoting vast underreporting, all the more considering that hospitalized cases reflect only a part of all malaria cases.[1] In some countries in Europe, estimates of malaria underreporting varied from 20% to 59%.[15] In Portugal, malaria surveillance is a passive mandatory physician-based case detection system. Contrary to other European countries with better reporting performance, there is no additional laboratory-based notification system.[15] The notification involves filling in an official form that is not always at hand and needs to be sent by post, which could lead to forgetfulness. Furthermore, lack of awareness of the legal requirement to report and unsystematic evaluation of malaria surveillance activities may also contribute to underreporting. Therefore, the DRG database may be an unexplored complementary source for better estimates of imported malaria.

The current situation in Portugal highlights the importance of increased awareness regarding malaria and risk management, with targeted messages tailored to the needs of this special group of travelers at pre-travel care and physician training to ensure early diagnosis and adequate and timely treatment at post-travel care.

On the other hand, Anopheles atroparvus is an efficient malaria vector widely distributed in Portugal and Europe that, despite being nonsusceptible to the African P. falciparum strains, might be susceptible to African P. vivax strains.[16, 17] In addition to the increasing imported malaria cases, the reappearance, in appropriate climatic conditions, of autochthonous malaria, eradicated since 1973 in Portugal, may be a feasible, albeit still theoretical, scenario. Sporadic autochthonous vivax malaria transmission has already occurred in other European countries officially considered malaria free, namely France, Spain, Italy, and Greece.[1, 8] This highlights the importance of clinical, laboratorial, and entomological surveillance as well as the role of travelers as harbingers of emergent infectious diseases.

Declaration of Interests

The authors state they have no conflicts of interest to declare.