Scand J Infect Dis 24: 477483, 1992

Hospital-Acquired Infections in Norway: A National Prevalence Survey in 1991

PREBEN AAVITSLAND, MARIT STORMARK and A R V E LYSTAD From the Department of Infectious Disease Control, National Institute of Public Health, Oslo. Norway

A I-day prevalence survey of hospital-acquired infections (HAI) was carried out in Norwegian somatic hospitals in 1991. The survey aimed at assessing the problem of HAL 950 infections were found among the 14977 surveyed patients (prevalence rate 6.3%). HA1 were more prevalent in combined intensive care units (prevalence rate 22%), surgical intensive care units (17%), haematological wards (15%), special care baby units (14%), and geriatric wards (14%). Urinary tract infections were most prevalent (33% of all HAI) followed by respiratory tract infections (21%) and surgical wound infections (17%). 157 (3.6%) of the 4382 patients who had undergone surgery, had a surgical wound infection. With a high response rate (76 out of 84 hospitals) and a sensitive method of screening the patients (chart review and bedside examination), this survey gives a fairly reliable measure of the prevalence of HA1 in Norwe- gian somatic hospitals.

P. Aavitsland, MD, Dept. of Infectious Disease Control, National Institute of Public Health, N-0462 Oslo, Norway

INTRODUCTION

Hospital-acquired infections (HAI) cause increased morbidity, mortality, prolonged hospi- tal stay and increased treatment costs (1,2). There is no national surveillance system for HA1 in Norway. Only a minority of hospitals has established an infection control programme with continuous surveillance of HAI. However 38 hospitals have a full-time infection control nurse.

Our aim was to assess the magnitude of the HA1 problem and to motivate hospital staff for preventive work through active participation in the survey. A prevalence survey was chosen because it is an inexpensive and fast method that would enable us to include all Norwegian somatic hospitals. Incidence surveys on the other hand are time-consuming and more expensive.

MATERIALS A N D METHODS Hospitals and wards All Norwegian somatic acute-care hospitals and specialty hospitals for orthopaedics. rheumatology. oncology and lung diseases were invited to participate in this one-day prevalence survey of HA1 on the 25th of April 1991. 76/84 hospitals (having 94.9% of the total bed capacity) completed the survey. Hospitals were classified as regional university hospitals. central county hospitals. local county hospitals and specialty hospitals.

Sampling units were hospital wards. The 589 participating wards were classified mainly according to the Norwegian Medical Association's list of medical specialities. All wards in the hospitals. nlso psychiatric wards. were included.

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Auc

klan

d on

11/

25/1

4Fo

r pe

rson

al u

se o

nly.

478 P. Aavitsland et al. Scand J Infect Dis 24

Table 1. Prevalence rates of hospital-acquired infections by hospital category Significance levels: *p<0.05, **p<O.OOl

Hospital category No. of No. of Prevalence Prevalence hospitals patients rate (YO) rate ratios"

Regional Central Local Speciality Total

5 4008 6.5 1.2 (1.&1.4)* 15 5271 5.4 1 48 4910 7.1 1.3 (1.1-1.5)** 8 788 6.6 1.2 (O.Sl.6)

76 14977 6.3

"Comparing each hospital category with a reference category (with a designated ratio of 1). 95% confidence intervals within parentheses.

Patients 14 977 patients who were present in the hospitals at 8 a.m. were surveyed for HAI. Among the surveyed patients were 4382 patients who had undergone surgery. In maternity wards both mothers and neonates were included.

Infections In classifying infections we followed slightly adapted guidelines from the US Centers for Disease Control (3). All diagnoses, except urinary tract infections (UTIs), were based on clinical criteria alone. HA1 was defined as a clinically manifest infection not present or in the incubation period at the time of admission. Infections at more than one site in the same patient were counted as separate infections.

Data processing To obtain a high response rate, we chose to restrict the number of data to be recorded. A clinician completed a form for each patient judged to have a HAL Summary records for each hospital were made by an infection control nurse or clinician and sent to the National Institute of Public Health for processing. Obvious errors in the records were corrected by comparing with the hospitals' original data. Records were computer-analyzed using the WHO/CDC EpiInfo software. Prevalence rates were compared using chi square statistics and prevalence rate ratios with 95% confidence intervals. Preva- lence rate ratio is the HA1 prevalence rate in one hospital or ward category versus a reference category with a designated rate of 1.

RESULTS

We found 950 HAIs among the 14977 surveyed patients, giving a prevalence rate of 6.3%. Prevalence rates varied from 0 to 15% between hospitals. The hospitals had from 0 t o 1245 patients (mean 197) and 0 to 84 HAIs (mean 12.5). There was no significant correlation between a hospital's total number of patients and its HA1 prevalence rate (r = -0.1, p = 0.4).

Table I shows prevalence rates by hospital categories. The medium-sized central county hospitals had the lowest prevalence rate. The large regional university hospitals and the small local county hospitals had a significantly higher prevalence rate than the central county hospitals.

Table I1 gives the prevalence rate for all ward categories and major combined categories. The prevalence rate varied from 0 to 22% between ward categories. Highest prevalence rates were found in combined and surgical intensive care units (ICUs), special care baby units (SCBUs), haematological wards and geriatric wards. Lowest prevalence rates were found in ophthalmological, otorhinolaryngological, psychiatric and endocrinological wards.

Table I1 also compares the prevalence rates in subspecialized wards to the prevalence rates in general wards. Prevalence rate was higher in haematological and geriatric wards com-

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Auc

klan

d on

11/

25/1

4Fo

r pe

rson

al u

se o

nly.

Scand J Infect Dis 24 Hospital-acquired infections in Norway 479

Table 11. Prevalence rates of hospital acquired-infections by ward category Significance levels: *pt0.05, **p<O.OOl

Ward category Patients Prevalence Prevalence rate ( O h ) rate ratios”

Internal medicine (total) Internal medicineb Haematology Endocrinology Gastroenterology Geriatrics Cardiology Infectious diseases Lung diseases Nephrology Oncology

Surgery (total) Surgeryh Paediatric surgery Gastroenterological surgery Vascular surgery Orthopaedics and rheuma.-surgery Thoracic surgery Urology Plastic surgery Maxillofacial surgery Neurosurgery

Intensive care (total) Intensive careb Medical intensive care Surgical intensive care

Gyn & Obs (total) Gyn & Obs G ynaecology Obstetrics‘

Paediatrics (total) Paediatricsb Paediatrics without neonatology Neonatologyd

Neurology Dermatology Physical medicine & Rehabilitation Rheumatology Ophthalmology Otorhinolaryngology Psychiatry Other

Other (total)

Total

4281 2379

120 51

197 289 314 118 345 116 352

4629 2282

54 519

87 1024

82 336 111

6 128

371 67

126 178

2683 689 495

1499

650 142 339 169

2363 486 102 162 413 261 336 553

50

14977

7.7 7.3

15 0 6.6

14 8.0 8.5 5.8 6.0 7.1

7.1 7.5 1.9 6.9 8.0 7.4 9.8 5.4 3.6 0 6.3

15 22

17 8.7

3.4 3.3 6.3 2.4

8.3 9.9 4.7

14

3.8 4.9 7.8

4.4 0.4 1.5 1.8 8.0

6.3

12

1 2.1 (1.3-3.2)*

0.9 (0.5-1.6) 1.9 (1.3-2.6)** 1.1 (0.7-1.6) 1.2 (0.6-2.1) 1.1 (0.7-1.7) 0.8 (0.4-1.7) 1.0 (0.7-1.5)

1 0.3 (0.0-1.7) 0.9 (0.7-1.3) 1.1 (0.5-2.2) 1.0 (0.8-1.3)

0.7 (0.4-1.1) 1.3 (0.7-2.5)

0.5 (0.2-1.3)

0.8 (0.4-1.7)

1 0.4 (0.2-0.8)* 0.8 (0.4-1.3)

1 1.9 (1.1-3.2)* 0.7 (0.4-1.2)

1 0.5 (0.2-1.0) 1.4 (0.8-2.7)

“omparing each ward category with a reference category (with a designated ratio of 1) within the same major category. 95% confidence intervals within parentheses. bGeneralized or combined wards ‘Maternity wards dSpecial Care Baby Units

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Auc

klan

d on

11/

25/1

4Fo

r pe

rson

al u

se o

nly.

480 P. Aavitsland et al. Scand J Infect Dis 24 ~~

Table 111. Prevalence and distribution of hospital-acquired infections (HAZ) by infection site

Type of infection Prevalence 'YO of total no. rate ( 'Yo) of HA1

Urinary tract infection Respiratory tract infection Surgical wound infection Skin infection Septicaemia Intraabdominal infection Infection around intravasal line or tracheostomy tube Gastroenteritis Osteomyelitis Meningitis Other hospital-acquired infection

Total

2.1 1.3 1.0 0.48 0.41 0.13 0.11 0.11 0.11 0.01 0.51

6.3

33 21 17 8 6 2 2 2 2 0 8

100"

"Adjusted for rounding errors

pared to generalized or combined internal medicine wards. Medical ICUs had a lower rate than combined ICUs. Gynaecological wards had a higher rate than combined gynaecological and obstetrical (Gyn & Obs) wards. Paediatric wards without neonates had a significantly lower prevalence rate than generalized or combined paediatric wards.

Table 111 presents the prevalence rate by infection site. Percentage of the total number of HA1 is given for each infection site. UTIs accounted for one third of the HAIS. One fifth of the infections were in the respiratory tract and 82% of these were pneumonias. The surgical wound infections (SWIs) represented one sixth of all HAIs. The SWIs were equally divided among superficial and deep infections.

Prevalence rates for the major ward categories are shown in Table IV. UTIs and respira- tory tract infections (RTIs) made up two thirds of the HAIs in internal medicine wards. In surgical wards UTIs and SWIs accounts for two thirds of the HAIs. Intensive care units had a very high prevalence rate of RTIs. The prevalence rate of septicaemia was as high as 2.4%. UTIs and neonatal conjunctivitis dominated in Gyn & Obs wards. Septicaemia is a major problem among patients in paediatric wards. The prevalence rate was 2.5% and septicaemia

Table IV. Prevalence rates and distribution of different hospital-acquired infections by major ward categories

Site of infection Major ward categories

Internal Surgery Intensive Gyn. & Paediatrics Other medicine care Obs.

Urinary tract 3.0 (39)b 2.3 (33) 2.2 (14) 1.1 (32) 0.2 (2) 1.6 (43) Respiratory tract 2.1 (27) 0.9 (13) 7.3 (48) 0.3 (9) 2.1 (25) 0.8 (21) Surgical wound 0.5 (7) 2.4 (34) 1.1 (7) 0.3 (9) 0.5 (6) 0.4 10) Skin 0.7 (9) 0.5 (6) 0.8 (5) 0.3 (9) 0.5 (6) 0.3 (9) Septicaemia 0.5 (7) 0.2 (3) 2.4 (16) 0.2 (5) 2.5 (30) 0 Other 0.9 (12) 0.8 (12) 1.3 (9) 1.2 (37) 2.6 (31) 0.6 (17)

Total" 7.7 (100) 7.1 (100) 15 (100) 3.4 (100) 8.3 (100) 3.8 (100)

aAdjusted for rounding errors bPercentages within parentheses

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Auc

klan

d on

11/

25/1

4Fo

r pe

rson

al u

se o

nly.

Scand J Infect Dk 24 Hospital-acquired infections in Norway 481

Table V. Comparison of HAI prevalence rate between the major ward categories

Significance levels: *p<0.05 **p<0.001

Site of infection Prevalence rate ratiosa

Internal Surgery Intensive Gyn. & Paediatrics Other medicine care Obs.

Urinary tract 1 0.8 (0.&1.0)* Respiratory tract 1 0.5 (0.H.7)** Surgical wound 2.8 (1.94.4)** 1 Skin 1 0.7 (0.4-1.2) Septicaemia 1 0.4 (0.24.8)* Other 1 0.9 (0.61.4)

All infections 1 0.9 (0.8-1.1)

0.7 (0.41.4) 0.4 (0.2-0.5)** 0.1 (O.(M.4)** 0.5 (0.4-0.8)** 3.5 (2.3-5.4)** 0.2 (0.14.3)** 1.1 ( 0 . 6 1 3 ) 0.4 (0.2-0.6)** 0.6 (0.2-1.5) 0.4 (0.2-0.8)* 1.4 (0.54.2) 0.5 (0.24.9)* 1.2 (0.43.9) 0.4 (0.2-1.0)* 0.7 (0.2-2.2) 0.5 (0.2-1.1) 4.5 (2.1-9.7)** 0.3 (0.1-0.8)* 4.6 (2.4-8.6)** 1.5 (0.63.7) 1.4 (0.9-2.1) 2.9 (1.65.0)** 0.9 (0.5-1.5)

2.0 (1.5-2.5)** 0.4 (0.14.5)** 1.1 (0.8-1.4) 0.5 (0.40.6)** ~

"Comparing for each infection site the rate in each ward category with a reference category (with a designated ratio of 1). 95% confidence intervals within parentheses.

cases accounted for 30% of infections among these patients. Half of the septicaemias (8/16) were found in SCBUs. Other infections in paediatric wards were mainly conjunctival infections in neonates.

Table V shows prevalence rate ratios of each infection among the major ward categories. The prevalence rate among patients in internal medicine wards is set to 1 except for SWIs. The rate of UTI was significantly lower in paediatric and Gyn & Obs wards. The rate of RTI in ICUs was more than 3 times the rate in internal medicine wards. In Gyn & Obs and surgical wards it was lower. Patients who had undergone surgery, but were treated in internal medicine wards had a higher prevalence of SWI than patients in surgical wards. The prevalence rate of septicaemia in ICUs or paediatric wards was more than 4 times the rate in internal medical wards.

157 (3.6%) of the 4382 patients who had undergone surgery, had a SWI. SWIs were more prevalent in gastroenterological and orthopaedic wards (Table VI). None of the ward categories, however, had a significantly higher prevalence rate than the generalized or non-specified surgical wards.

Table VI. Prevalence rates of surgical wound infections by surgical ward categories

Significance level: *p<0.05

Surgical ward category Prevalence Prevalence rate (Yo) rate ratiosa

Gastroen terological Orthopaedic Thoracic or vascular General or non-specified Urological Gy naecological Obstetrical Other

5.3 1.6 (0.9-2.7) 4.7 1.4 (0.9-2.1) 4.5 1.3 (0.62.9) 3.4 1 2.4 0.7 (0.3-2.0) 1.2 0.4 (0.1-1.1) 0.8 0.2 (0.C1.7) 1.4 0.4 (0.24.9)*

"Comparing each ward category with a reference category (with a designated ratio of 1). 95% confi- dence intervals within parentheses.

31 Scand 1 Infect Dis

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Auc

klan

d on

11/

25/1

4Fo

r pe

rson

al u

se o

nly.

482 I? Aavitsland et al. Scand J Infect Dis 24

DISCUSSION

A prevalence survey of hospital-acquired infections has several limitations. The prevalence of active nosocomial infection is dependent on the duration of such infections (4). The impact of longer-lasting infections, e.g. osteomyelitis, will tend to be overestimated. This of course limits the value of comparisons of prevalence rates for different infection categories. Patients may have acquired their infection in another ward than the one they are treated in on the day of the survey. We believe this may be an explanation for the high prevalence rates of septicaemia in ICUs and SCBUs. This limits the value of comparisons between prevalence rates for different ward categories. Furthermore our prevalence rates are not true rates. The dimensions of the numerator (infections) and denominator (patients) are not the same. The rationale for this tallying of infections is that 2 patients acquiring 1 infection is as undesirable as 1 patient acquiring 2 infections.

The main advantage of our study is the high response rate. This excludes one major source of bias in the results. We believe this high response rate was made possible by thorough instructions and the limited number of data elements to be recorded for each patient. Another advantage is our survey method. In every ward 1 clinician was responsible for the chart review and bedside examination. This is a sensitive screening procedure for HA1 (5).

Three types of factors determine the risk of a HAI. They are host factors (e.g. age, underlying disease, medication; 68), procedure-related factors (e.g. surgery, urinary blad- der catheterization; 7,9,10) and hospital factors (e.g. lengthof stay, ward, isolation, micro- bial flora; 6,l l).,These factors are highly interrelated. The patients with severe underlying diseases are also the patients most likely to be treated in ICUs and undergo therapeutic procedures.

A prevalence survey cannot establish a cause and effect relationship between the mea- sured factors and HAL Its importance lies in measuring the problem and locating the areas with high prevalence. The survey provides baseline data for rational priorities in allocation of resources, for further studies and for infection control activities. For instance it is clear from our study that the high prevalence rate of respiratory tract infections and septicaemia in ICUs require further study. We also need to know why the rate of UTI is higher among patients in internal medicine wards. The high prevalence rate of septicaemia in paediatric wards also require further study. Half of the septicaemic children in non-SCBUs were in one ward at a regional hospital treating leukaernia patients. HAIs in these patients are often unavoidable.

This is the third nationwide prevalence study of HA1 in Norway. The previous studies in 1979 (12) and 1985 (13) using similar methods gave similar results. There is, however, a significant decrease in the overall prevalence rate from 9.0% in 1979 to 7.8% in 1985 (p < 0.005) and to 6.3% in 1991 (p < 0.001). We also note that both the prevalence rate and the relative importance of UTIs are decreasing. In 1979 UTI was found in 3.8% of the patients (42% of HAIs), in 1985 in 2.9% (37% of HAIs) and in 1991 in 2.1% (33% of HAIs). The same is true for SWIs. In 1985 SWIs were found in 4.6% of patients who had undergone surgery. In 1991 the prevalence rate was 3.6 (p < 0.05). The decrease in overall prevalence rate is largely explained by the decrease in UTI prevalence. The prevalence of the other infection categories is of the same magnitude in the 3 studies.

The prevalence rate of HA1 in Norway seems to be of the same magnitude as reported in surveys of selected hospitals in other European countries (8,11,14-18). HA1 prevalence rate seems to be of the magnitude 6 1 5 % . UTIs account for 25-35% of HAIs, RTIs 2&25%, SWIs 15-20% and septicaemias 3 4 % . Differences in methodology call for great care in comparing prevalence rates for different countries.

We found a nationwide prevalence survey of all somatic hospitals to be a valuable and

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Auc

klan

d on

11/

25/1

4Fo

r pe

rson

al u

se o

nly.

Scand J Infect Dis 24 Hospital-acquired infections in Norway 483

rational way of assessing the problem of HAI. T h e study has shown that HA1 is endemic in Norwegian hospitals. This underscores t h e need for continuous surveillance of HA1 as an integral part of every hospital's infection control program. Only through this kind of surveillance the hospitals will be able to identify correctable causes of HAI.

REFERENCES 1. 2. 3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15

16

Brachman PS. Nosocomial infection control: an overview. Rev Infect Dis 3: 64M48, 1981. Daschner F. Economic aspects of hospital infections. J Hosp Infect 3: 14, 1982. Centers for Disease Control. Outline for surveillance and control of nosocomial infections. Appen- dix 11. Atlanta, Georgia: Centers for Disease Control, 1976. Freeman J, McGowan JE. Methodologic issues in hospital epidemiology. I. Rates, case-finding, and interpretation. Rev Infect Dis 3: 65&667, 1981. Freeman J , McGowan JE. Methodologic issues in hospital epidemiology. 11. Time and accuracy in estimation. Rev Infect Dis 3: 668-677, 1981. French GL, Cheng AFB, Wong SL, Donnan S. Repeated prevalence surveys for monitoring effectiveness of hospital infection control. Lancet 2: 1021-1023, 1989. Michel J-P, Lesourd B, Conne P, Richard D, Rapin C-H. Prevalence of infections and their risk factors in geriatric institutions: a one-day multicentre survey. Bull WHO 69: 3541 , 1991. Sramova H, Bartoilova A, Bolek S , KreEmerovi M, Subertovfi V. National prevalence survey of hospital acquired infections in Czechoslovakia. J Hosp Infect 11: 328-334, 1988. Wenzel RP, Osterman CA, Donowitz LG, Hoyt JW, Merle AS, Martone WJ, Peacock JE, Levine JI, Miller GB. Identification of procedure-related nosocomial infections in high-risk patients. Rev Infect Dis 3: 701-707, 1981. Haley RW, Culver DH, Morgan WM, White JW, Emori TG, Hooton TM. Identifying patients at high risk of surgical wound infection: a simple multivariate index of patient susceptibility and wound contamination. Am J Epidemiol 12: 20tk215, 1985. Mertens R, Kegels G, Stroobant A , Reybrouck G , Lamotte JM, Potvliege C, Van Casteren V, Lauwers S, Verschraegen G , Wauters G, Minne A, Thiers G. The national prevalence survey of nosocomial infections in Belgium, 1984. J Hosp Infect 9: 219-229, 1987. Hovig B, Lystad A, Opsjen H. A prevalence survey of infections among hospitalized patients in Norway. Natl Inst Pub1 Health Annals 4: 4940, 1981. Lystad A, Stormark M. A national prevalence survey of hospital acquired infections in Norway. In: Acts from the International Symposium on the Control of Hospital Infections, April 27-29; 1987, Rome. Florence: Edizione Riviste Scientifiche, 1987. Bernander S. Hambraeus A, Myrback K-E, Nystrom B, Sundelof B. Prevalence of hospital- associated infections in five Swedish hospitals in November 1975. Scand J Infect Dis 10: 6tk70, 1978. Meers PD, Ayliffe GAJ, Emmerson AM, Leigh DA, Mayon-White RT, Mackintosh CA, Stronge JL. Report on the national survey of infection in hospitals, 1980. J Hosp Infect 2, Suppl., 1981. Moro ML, Stazi MA, Marasca G, Greco D , Zampieri A. National prevalence survey of hospital- acquired infections in Italy, 1983. J Hosp Infect 8: 72-85, 1986.

17. Jepsen OB, Mortensen N. Prevalence of nosocomial infection and infection control in Denmark. J Hosp Infect 1: 237-244, 1980.

18. Jepsen OB, Olesen Larsen S , Dankert J , Daschner F, Gronros P, Meers PD, Nystrom B, Rotter M, Sander J. Urinary-tract infection and bacteraemia in hospitalized medical patients - a European multicentre prevalence survey on nosocomial infection. J Hosp Infect 3: 241-252, 1982.

31'

Scan

d J

Infe

ct D

is D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y U

nive

rsity

of

Auc

klan

d on

11/

25/1

4Fo

r pe

rson

al u

se o

nly.