More Info
									                                       PNEUMONIA SURVEILLANCE IN THAILAND

         Jirapat Kanlayanaphotporn1, Molly A Brady2, Panatsaya Chantate3, Somrak Chantra4,
                      Suvaj Siasiriwattana3, Scott F Dowell5 and Sonja J Olsen5

   Bureau of Epidmemiology, Ministry of Public Health, Nonthaburi, Thailand; 2Rollins School of
Public Health, Emory University, Atlanta, Georgia, USA; 3Sa Kaeo Provincial Health Office, Ministry
  of Public Health, Sa Kaeo; 4Crown Prince Hospital, Sa Kaeo, Thailand; 5International Emerging
       Infections Program, Centers for Disease Control and Prevention, Nonthaburi, Thailand

        Abstract. We reviewed reported pneumonia cases and deaths in Thailand since 1975 to evaluate the
        pneumonia surveillance system. In Sa Kaeo Province, we analyzed 3 years in detail (1999-2001)
        from electronic surveillance data, and compared deaths reported through surveillance to death certifi-
        cate data in 1999 and 2000. In addition, we interviewed surveillance personnel who collected the data
        from all 7 hospitals and from a 10% random sample of health centers. Since the mid-1980s, reported
        illnesses and deaths from pneumonia have been increasing. In Sa Kaeo, an average of 925 pneumonia
        cases were reported each year, for an estimated average annual incidence of 211 per 100,000. The
        age-specific incidence peaked at 1,418 per 100,000 in children less than 5 years. In 1999 and 2000,
        there were 7 and 6 pneumonia deaths, respectively, reported through the surveillance system, com-
        pared with 28 and 53, respectively, reported by death certificate. Sixty-two (82%) of the 72 surveil-
        lance personnel reported receiving some training, but most of this was informal. Although written
        criteria to diagnose pneumonia were established in 1996, those who report cases did not know these
        criteria. A combination of physician, nurse, and public health workers diagnoses were used. Accord-
        ing to the written criteria, cases of suspect or rule out pneumonia should be reported, but when asked
        about specific examples only 79% of persons interviewed said they would report “tuberculosis with
        pneumonia” and 44% would report “bronchitis, rule out pneumonia.” Seventy-four percent of per-
        sons interviewed completed the surveillance report within one day of patient admission.

                INTRODUCTION                                 changes in data. Now that interventions are pos-
                                                             sible, countries need active surveillance that can
      In 1970, the Bureau of Epidemiology at the             produce accurate, pathogen-specific data for de-
Thai Ministry of Public Health (MOPH) estab-                 cision making and monitoring. We evaluated the
lished a national surveillance system for infec-             quality of the current passive surveillance sys-
tious diseases. This is a passive surveillance sys-          tem for pneumonia in Thailand. Our study high-
tem that uses a standard reporting form. In the              lights the importance of pneumonia and the need
beginning, there were 14 notifiable diseases, as             for accurate data, several shortcomings to the
of 2000, there were 68. In Thailand in the year              current system, and suggests that newer ap-
2000, pneumonia was ranked as the third leading              proaches are needed to more accurately measure
cause of morbidity and the leading cause of mor-             the burden of pneumonia in Thailand.
tality among notifiable diseases (Anonymous,
                                                                      MATERIALS AND METHODS
      Historically, passive surveillance systems
have been used to monitor broad trends and big               Study design and population
Correspondence: Dr Jirapat Kanlayanaphotporn, Bu-                 This study was conducted in Sa Kaeo Prov-
reau of Epidemiology, Ministry of Public Health,             ince, which is located along the Thai-Cambodian
Tiwanon Road, Nonthaburi 11000, Thailand.                    border in eastern Thailand (Fig 1). According to
Tel: 66 (0) 2590-1734; Fax: 66 (0) 2591-1735                 the 2000 provincial census, there were 439,120
E-mail:                            persons living in Sa Kaeo. In addition, there are

Vol 35 No. 3 September 2004                                                                                      711
                                 SOUTHEAST ASIAN J TROP MED PUBLIC HEALTH

                                                        these to the number of deaths reported by the sur-
                                 Lao PDR                veillance system. Fourth, we interviewed surveil-
                                                        lance personnel in Sa Kaeo to identify current sur-
                                                        veillance practices that might deviate from the
                                                        proscribed method. We selected the personnel at
                                                        7 public health hospitals (excluding the mental
                                                        health hospital) and a 10% simple random sample
                                                        of health centers. Persons responsible for the sur-
                                                        veillance reporting were interviewed using a stan-
                                                        dard, written questionnaire.
             Bangkok                                    Statistical analysis
                                     Sa Kaeo
                                                             To characterize the descriptive epidemiology,
                                       Cambodia         we used EpiInfo version 6.04d (CDC Atlanta,
                                                        Georgia). Incidence in Sa Kaeo was calculated
                                                        using 2001 census data from the Sa Kaeo Provin-
                                                        cial Health Office as the denominator. Incidence
                                                        in Thailand was calculated using the mid-year
                                                        population from the Ministry of Interior as the de-
                                                        nominator. To identify pneumonia deaths in the
                                                        death certificate database, we selected deaths that
                                                        were coded as International Classification of Dis-
                                                        ease volume 10 (ICD10) codes J12.9 through J18.9.
Fig 1–Map of Thailand highlighting Sa Kaeo Province.
                                                        Process of data collection and reporting
an estimated 23,000 registered and unregistered              The surveillance is a passive system run by
Cambodian migrant workers in Sa Kaeo. The               the Bureau of Epidemiology. The written case
province has 9 government hospitals: 8 run by           definition is defined as a person who has fever,
the Ministry of Public Health (including a men-         pleural pain, dyspnea, cough, and leukocytosis.
tal health hospital) and one run by the Ministry        In addition, lung crepitation should be docu-
of Defense. There are no private hospitals. The         mented by physical examination, and if a chest
hospitals range in size from 30 to 230 beds. All        x-ray is performed, it must show infiltration or
hospitals are equipped with radiologic equipment        consolidation. Patients with occupational lung
and at least basic laboratory facilities. Although      diseases are excluded. Although some patients had
routine microscopy, chemistry, hematology, and          specimens taken for culture, the guidelines indi-
serologic testing are available, equipment and ma-      cate that it is unnecessary to wait for results be-
terials to conduct bacteriological testing are lack-    fore reporting a case of pneumonia.
ing, except at the two largest hospitals. Each hos-          Cases meeting the definition are reported on
pital has an outpatient clinic, and there are 109       a standard form (Form 506) and entered into a
health centers.                                         computer at the hospital or district health center.
      The study had four main parts. First, we re-      Data travels via disk from the hospital or district
viewed the process of data collection and report-       health center to the provincial health office and
ing. Second, we analyzed reported pneumonia             finally to the Bureau of Epidemiology in
cases and deaths in Thailand since 1975 and dur-        Nonthaburi. A second form (Form 507) is a fol-
ing the 3-year period, 1999-2001, to detail the         low-up form used for changing the diagnosis and/
pneumonia surveillance data from Sa Kaeo. Third,        or the outcome of a previously reported case.
we reviewed the number of pneumonia deaths              These forms are used to collect data from all gov-
from death certificates (1999-2000) and compared        ernment, and some private, health facilities. Only

712                                                                         Vol 35 No. 3 September 2004
                                     PNEUMONIA SURVEILLANCE IN THAILAND

Thai residents are recorded in this sys-                                 300

tem.                                                                                                                                                           18

                                                                                                                                                                      Pneumonia deaths per 100,000

                                           Pneumonia cases per 100,000
Pneumonia data                                                                                                                                                 14
      Over the last 20 years, reported                                                                                                                         12

rates of pneumonia in Thailand in-                                       150                                                                                   10
creased from 74 per 100,000 in 1982                                      100
to 227 per 100,000 in 2000 (Fig 2).                                                                                                                            4
Reported deaths from pneumonia                                                                                                                                 2
were 17.8 per 100,000 in 1975, de-                                           0                                                                         0
creased to a low of 4.8 in 1987, and                                              1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001
then increased steadily to 15.9 in 2001
(Fig 2). In Sa Kaeo, from 1999                                       Fig 2–Reported cases and deaths from pneumonia, from 1975-2001,
through 2001, there were a total of                                        Thailand.
2,776 cases for an average of 925
cases per year (957 in 1999, 965 in
2000, and 854 in 2001). The median                                       Incidence per 100,000
age was 4 years (range, 1 day to 99                                      300
                                                                                                                                                    Sa Keao
years), 1,326 (48%) of the cases were                                                    262
                                                                         250                                        227
female, and 1,956 (70%) were inpa-                                                                  218                      220             219
tients. There were 7 deaths reported                                     200
in 1999, 6 in 2000, and 8 in 2001.
None of the pneumonia records had                                        150
any laboratory results.
      The annual incidence of pneu-
monia in Sa Kaeo is shown in Fig 3.                                          50
For comparison, data from Thailand
as a whole are shown. The average                                            0
                                                                                               1999                       2000                     2001
annual incidence was 236 per 100,000                                                                                      Year
persons in Thailand and 211 in Sa
Kaeo. In Sa Kaeo, age specific inci-                                 Fig 3–Incidence of pneumonia in Thailand and Sa Kaeo, 1999-2001.
dence peaked at 1,418 per 100,000 in
children less than five years old, de-      Incidence per 100,000
creased to rates between 28 and 264        1,600
per 100,000 in persons aged 5 to 65,
and then increased to 551 per 100,000
in persons over 65 years (Fig 4). The      1,200
average monthly incidence in chil-
dren less than 5 years (1,419 per
100,000) old was 13 times greater                  800
than that in persons aged 5 and older
(108 per 100,000); the rates in both
age groups peaked in the period of                 400
July through September (Fig 5).
Death certificates
     There were 28 and 53 pneumo-                                                 0-4   5-9   10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65+
nia deaths in 1999 and 2000, respec-                                                                                 Age group (years)
tively, reported on death certificates                                       Fig 4–Incidence of pneumonia by age group in Sa Kaeo Province,
in Sa Kaeo. By contrast, there were 7                                              1999-2001.

Vol 35 No. 3 September 2004                                                                                                                                  713
                                                                                    SOUTHEAST ASIAN J TROP MED PUBLIC HEALTH

                                            3,500                                                              160                                               with pneumonia, 44% reported bron-

                                                                                                                     Incidence per 100,000 in persons >5 years
Incidence per 100,000 in persons <5 years

                                            3,000                                                              140                                               chitis, rule out pneumonia, 42% re-
                                                              <5 years                                         120                                               ported cough, rule out pneumonia,
                                                                                                               100                                               11% reported lower respiratory tract
                                                                                                                                                                 infection, and 3% reported dyspnea.
                                            1,500                        >5 years
                                             500                                                               20
                                                                                                                            This passive surveillance sys-
                                               0                                                               0      tem places pneumonia solidly among
                                                   Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
                                                                             Month                                    the leading causes of morbidity and
                                                                                                                      mortality in Thailand. The age dis-
                                             Fig 5–Age-specific incidence of pneumonia by month in Sa Kaeo, 1999-     tribution is consistent with what is
                                                    2001.                                                             known of pneumonia elsewhere
                                             and 6, respectively, reported from the surveillance       (Marrie, 1998), and the seasonal peak among young
                                             system. We were unable to ascertain if these              children in July through September bears further
                                             deaths were reported in both systems because all          exploration. Nevertheless, weaknesses in the train-
                                             data sets were stripped of names.                         ing of personnel, lack of use of a standard case
                                             Surveillance personnel interviews                         definition, and underreporting of deaths empha-
                                                                                                       size the need to strengthen this system, and new
                                                   We interviewed 72 health personnel [56
                                                                                                       opportunities for pneumonia prevention underline
                                             (78%) from the 7 hospitals and 16 (22%) from
                                                                                                       the necessity for a completely new approach to
                                             10 health centers]. The number of persons report-
                                                                                                       pneumonia surveillance.
                                             ing surveillance data ranged from one to twenty
                                             in hospitals and from one to three in health cen-               Although there are several published stud-
                                             ters. Of the interviewees, 47 (65%) were nurses,          ies from Southeast Asia that summarize pneumo-
                                             14 (19%) public health personnel, and 11 were             nia cases and etiologies at a single hospital
                                             other hospital personnel. Sixty-two (86%) persons         (Boonsawat et al, 1990; Chan et al, 1992; Hui et
                                             reported receiving some training for their job; 37        al, 1993; Reechaipichitkul et al, 2002), we could
                                             (60%) were trained informally by other hospital           find no reports in this region measuring the popu-
                                             personnel and 8 (13%) by district and provincial          lation burden of pneumonia as a clinical syn-
                                             health office staff.                                      drome. Rates reported from all ages in Spain were
                                                   The two methods used to identify cases were         160 per 100,000 per year (Monge et al, 2001a,b).
                                             manual searches through logbooks and comput-              Adult surveillance had rates of 160 in Spain and
                                             erized ICD10 searches. When asked, “Whose di-             267 in the United States (Marston et al, 1997;
                                             agnosis do you use to decide if a patient needs to        Almirall et al, 2000). As expected, children less
                                             be reported on Form 506?” 59 (82%) responded              than 15 years old in Finland had a much higher
                                             “physician diagnosis,” 15 (21%) responded                 annual incidence of 2,271 per 100,000 (Table 1).
                                             “nurse diagnosis,” and 16 (22%) responded “pub-           The rate in Sa Kaeo is similar to the passive sys-
                                             lic health personnel diagnosis.” These categories         tem in Spain. In Thailand, the lower incidence in
                                             were not mutually exclusive.                              Sa Kaeo compared to the country as a whole is
                                                                                                       curious and may suggest that the rate of pneumo-
                                                   Although the written criteria specify that a
                                                                                                       nia varies geographically.
                                             diagnosis of both “with” and “rule out” pneumo-
                                             nia should be reported as a case, when interviewees             Most of the weaknesses of the existing sur-
                                             were asked, “ What would you report as pneumo-            veillance system identified in this review are com-
                                             nia?” most reported pneumonia when the diagno-            mon to passive surveillance systems. A case defi-
                                             sis said “with pneumonia” but if the diagnosis said       nition is central to any surveillance system (CDC,
                                             “rule out pneumonia” under 50% of them reported           1988). Although this system has a standard, writ-
                                             pneumonia. For example, 79% reported “new tu-             ten case definition for pneumonia, this study iden-
                                             berculosis with pneumonia”, 76% reported measles          tified many different interpretations making

                                             714                                                                                                                       Vol 35 No. 3 September 2004
                                                                                                       Table 1
                                                                             Population-based studies of community-acquired pneumonia.

                              Study dates             Country and          Age (years)            Setting                     Case definition                  Annual incidence per   Mortality
                                                       reference                                                                                                    100,000

                              Sep 1981-Aug 1982   Finland (Heiskanen-        <15         Doctor-based               Radiologically confirmed                    2,271 in 0-14 years       -
                                                  Kosma et al, 1999)

Vol 35 No. 3 September 2004
                              Jan-Dec 1991        Ohio, USA (Marston         ≥18         Inpatients                 New infiltrate on chest radiograph          267 in ≥18 years          8.8%
                                                  et al, 1997)                                                      AND
                                                                                                                    Fever, hypothermia, productive cough,       92 in 18-44 years
                                                                                                                    or abnormal white blood cell count          279 in 45-59 years
                                                                                                                                                                1,014 in ≥65 years
                              Dec 1993-Nov 1995   Spain (Almirall et al,     ≥14         Outpatients and            Acute lower respiratory tract infection,    162                       5%
                                                  2000)                                  inpatients                 focal signs on physical examination of
                                                                                                                    chest, new infiltrate on radiograph
                              Jan 1995-Dec 1996   Spain (Monge et al,        All ages    Inpatients                 ICD 9 CM codes 480-486, first listed        160 in all ages           7.4%
                                                  2001a,b)                               National surveillance      diagnosis
                                                                                         system for hospital data                                               494 in 0-4 years
                                                                                                                                                                128 in 5-9 years
                                                                                                                                                                30 in 10-24 years
                                                                                                                                                                73 in 25-39 years
                                                                                                                                                                                                  PNEUMONIA SURVEILLANCE IN THAILAND

                                                                                                                                                                95 in 40-64 years
                                                                                                                                                                433 in ≥65 years
                              Apr 1994-Mar 1999   Canada (Marrie et al,      ≥18         Hospital discharge data    ICD 10 CM codes 480-487, 507                129 in 18-39 years        3.2%
                                                  2003)                                                              (exclude 484)                              191 in 40-54 years
                                                                                                                                                                1,321 in ≥55 years
                              Jan 1999-Dec 2001   Sa Kaeo Province,          All ages    Inpatients (70%) and                                                   211                       0.76%
                                                  Thailand                               outpatients
                                                                                         National surveillance

                                  SOUTHEAST ASIAN J TROP MED PUBLIC HEALTH

nondifferential misclassification a common oc-                             REFERENCES
currence. In other words, there will be both false
positives (eg, some non-pneumonia will be re-            Almirall J, Bolibar I, Vidal J, et al. Epidemiology of
                                                              community-acquired pneumonia in adults: a
ported as pneumonia) and false negatives (eg,
                                                              population-based study. Eur Respir J 2000; 15:
some true pneumonia may be missed). Without                   757-63.
standardized data collection it is difficult to know     Anonymous. Annual epidemiologic surveillance report
1) if the reported numbers are an over or under-              2000. Nonthaburi: Ministry of Public Health,
estimate of the true numbers, 2) if trends can be             2000: 270-7.
followed, and 3) what interventions might be valu-       Boonsawat W, Boonma P, Tangdajahiran T, Pauperm-
able. The absence of laboratory data is another               poonsiri S, Wongpratoom W, Romphryk A. Com-
limitation. Laboratory capacity is limited outside            munity-acquired pneumonia in adults at
of Bangkok, and data that does exist is not rou-              Srinagarind Hospital. J Med Assoc Thai 1990; 73:
tinely integrated with the case data.                         345-52.
                                                         CDC. Guidelines for evaluating surveillance systems.
      Mortality from pneumonia is most likely
                                                              MMWR Morb Mortal Wkly Rep 1988; 37: 1-18.
underreported in the passive surveillance system
                                                         Chan CH, Cohen M, Pang J. A prospective study of
since some patients may have died at home. For
                                                              community-acquired pneumonia in Hong Kong.
those who died in the hospital, the health person-            Chest 1992; 101: 442-6.
nel might not have sent a follow-up report to change     Heiskanen-Kosma T, Korppi M, Laurila A, Jokinen C,
the patient status. In other studies, the frequency           Kleemola M, Saikku P. Chlamydia pneumoniae is
of death among pneumonia patients ranged be-                  an important cause of community-acquired pneu-
tween 5% among adults in Spain (Almirall et al,               monia in school-aged children: serological results
2000), to 9% in hospitalized adults in Ohio                   of a prospective, population-based study. Scand J
(Marston et al, 1997), to 21% in hospitalized adults          Infect Dis 1999; 31: 255-9.
in Khon Kaen, Thailand (Boonsawat et al, 1990).          Hui KP, Chin NK, Chow K, et al. Prospective study of
It is important to accurately measure and report              the aetiology of adult community acquired bacte-
deaths since they reflect the severity of the disease         rial pneumonia needing hospitalisation in
and can greatly influence health policy decisions.            Singapore. Singapore Med J 1993; 34: 329-34.
                                                         Marrie TJ. Community-acquired pneumonia: epidemi-
      New opportunities for treating and prevent-             ology, etiology, treatment. Infect Dis Clin North
ing pneumonia have accumulated rapidly in re-                 Am 1998: 12: 723-40.
cent years, outpacing the capacity of many exist-        Marrie TJ, Carriere KC, Jin Y, Johnson DH. Factors
ing surveillance systems to provide the informa-              associated with death among adults <55 years of
tion needed for evidenced-based public health                 age hospitalized for community-acquired pneu-
decisions. New and effective vaccines against                 monia. Clin Infect Dis 2003; 36: 413-21.
Haemophilus influenzae type b, Streptococcus             Marston BJ, Plouffe JF, File TM Jr, et al. Incidence of
pneumoniae, and influenza viruses raise the po-               community-acquired pneumonia requiring hospi-
tential to substantially reduce the burden of pneu-           talization. Results of a population-based active
monia in young children and perhaps the elderly.              surveillance study in Ohio. The Community-
The ability to treat previously untreatable disease,          Based Pneumonia Incidence Study Group. Arch
                                                              Intern Med 1997; 157: 1709-18.
such as influenza B with the neuraminidase in-
                                                         Monge V, Gonzalez A. Hospital admissions for pneu-
hibitors, or to improve treatment outcomes, such
                                                              monia in Spain. Infection 2001a; 29: 3-6.
as with directly observed therapy for tuberculo-
                                                         Monge V, San-Martin VM, Gonzalez A. The burden of
sis, open new, albeit expensive, opportunities.
                                                              community-acquired pneumonia in Spain. Eur J
Rational public health decisions will require ac-             Public Health 2001b; 11: 362-4.
curate information on the disease burden and costs       Reechaipichitkul W, Tantiwong P. Clinical features of
associated with pneumonia caused by these patho-              community-acquired pneumonia treated at
gens. It is not clear whether the current pneumo-             Srinagarind Hospital, Khon Kaen, Thailand.
nia surveillance in Thailand is suited to give ad-            Southeast Asian J Trop Med Public Health 2002;
equate information to effect such decisions.                  33: 355-61.

716                                                                           Vol 35 No. 3 September 2004

To top