Point-prevalence survey of hospital acquired infections in three acute care hospitals in Northern Nigeria.

Background Effective infection prevention and control strategies require reliable data describing the epidemiology of hospital acquired infections (HAIs), and this is currently lacking in Nigeria. The objective of this study was to evaluate the prevalence, types and risk factors associated with HAIs in acute care hospitals in Northern Nigeria. Methods A pilot point-prevalence survey was conducted in three acute care hospitals in Northern Nigeria between April and May 2019 using a protocol developed by the European Centre for Disease Prevention and Control. Patients admitted into the wards at or before 8.00 am on the survey date were included. Patients’ medical records were reviewed by a clinical pharmacist with the support of the attending physician and nurse to identify HAIs. Results Of the 321 patients surveyed, 50 HAIs were identified among 46 patients translating into a point-prevalence of 14.3%. The most common HAIs were bloodstream infection (38.0%), surgical site infections (32.0%) and pneumonia (12.0%). Neonatal (53.0%), pediatric surgical (26.7%) and surgical (10.1%) specialties had the highest prevalence. Device associated infections represented 16% of all HAIs including bloodstream infections and pneumonia. Of all the HAIs, 15 (30.0%) were present at the time of admission while 75.5% originated from the current hospitals. Univariate analysis showed that newborn (less than 1 month old) (OR: 4.687 95% CI: 1.298–16.927), intubation (OR: 3.966, 95% CI: 1.698–9.261), and neonatal (OR: 41.538 95% CI: 4.980–346.5) and pediatric surgical (OR: 13.091 95% CI: 1.532–111.874) specialties were significantly associated with HAI. Conclusion The prevalence of HAI was relatively high compared to other developing countries and was significantly associated with neonatal and pediatric surgical specialties. Hospital infection control strategies should be strengthened to reduce the burden of HAIs.

infections, urinary tract infection, and SSI are the most common HAI in European acute care hospitals [4]. Approximately 16% of hospitalized patients in developing countries are diagnosed with HAI [7]. The high rate of HAI in developing countries is attributed to inadequate infection control practices owing to the lack of infection control policy and guideline, and the dearth of infection control health professionals [8]. Other factors include lack of infrastructure, inconsistent surveillance [9], overcrowding, scarcity of resources [10], poor sanitation and poor management of hospital waste [11].
HAIs can be prevented through implementation of infection control and prevention program, surveillance of HAIs, proper waste management and proper training of hospital staff on biosafety [12]. An effective infection prevention and control program requires active surveillance to generate data that would describe the prevalence and risk factors associated with HAI. In addition, surveillance data is needed to measure the impact of infection control and prevention programs and also to prioritize areas for further interventions and resource allocation. The European Center for Disease Prevention and Control (ECDC) and the National Health Safety Network (NHSN) provide continuous surveillance for HAI and antibiotic consumption in Europe and the US respectively. In Nigeria, there is currently no surveillance system to provide estimates of HAI in acute care hospitals. Previous estimates of HAI have reported prevalence rates between 2.5 and 6.3% [6,13,14]. However, the studies used single center and retrospective design as against prospective active surveillance, which is the gold standard [15]. Prospective active surveillance is capital intensive and time-consuming. However, point-prevalence survey is a valid and reliable alternative to prospective active surveillance to provide estimates of HAIs [16]. The former method is more cost-effective than the latter and could be conducted with few surveyors. Therefore, pointprevalence study is suitable for use to estimate the burden of HAI in hospitals with limited resources. The objective of this pilot point-prevalence study was to determine the point-prevalence and risk factors associated with HAI in acute care hospitals.

Study design and setting
This point-prevalence survey was conducted in 3 acute care hospitals located in two states in Northern Nigeria using a protocol adapted from the point prevalence survey of healthcare-associated infections and antimicrobial use in European acute care hospitals protocol version 5.3 [17]. The survey was conducted in one public university teaching (approximately 560 beds) and two public general (200 and 100 beds respectively) hospitals. The selected hospitals provide acute care services including internal medicine, surgery, obstetrics and gynecology, pediatrics, neonatal care, psychiatry, intensive care and ear, nose and throat care. The wards in each specialty in the selected hospitals were divided into male and female wards, and each ward was further subdivided into 'A' and 'B' respectively. None of the wards had more than 30 patients on admission at the time of the survey. This arrangement made the survey of all patients in one ward in a single day feasible.

Inclusion and exclusion criteria
All units/departments in the selected hospitals were surveyed with the exception of outpatient, adult accident and emergency, and psychiatric units. Emergency pediatric unit was included because some patients are monitored for > 24 h. It is important to note that two of the hospitals had an intensive care unit at the time of this study. However, there was no patient on admission on the day the unit was scheduled for survey. The neonatal unit in the selected hospitals is divided into two subunits: in-born (for neonates born in the hospital) and out-born (for neonates born at home or at other healthcare facilities) wings, and both wings were included in the survey. All patients admitted into the ward and monitored for more than 24 h in the participating hospitals were included in the survey. Patients who were in the ward before or at 8.00 a.m. and were not discharged at the time of the survey were included. Daycare patients, those transferred out of the wards before the survey were excluded.

Data collection
Information including patient's age, gender, dates of admission and survey, surgery since admission, presence of any invasive device (peripheral and central vascular catheters, urinary catheter and intubation), McCabe score, patient/consultant specialty, presence of HAI and the results of microbiological investigations were collected using a data collection form. Information was obtained by reviewing the medical and nursing records of the surveyed patients and identified HAIs were discussed with the attending physician and nurse. Data was collected by a clinical pharmacist who has expertise in infectious diseases and experience in reviewing patient chart/record. All the patients in a single ward were surveyed on the same day. Data collection lasted for 2 weeks in one of the hospitals. The survey was conducted from 22 April to 24 May 2019.

Outcome measures
Hospital acquired infection and device-associated HAIs were defined based on the criteria described in the point prevalence survey of healthcare-associated infections and antimicrobial use in European acute care hospitals Protocol version 5.3 [17]. Asymptomatic bacteriuria and ophthalmic neonatorum were not considered as HAIs.
The following criteria were used to determine deviceassociated HAI: i) presence of urinary catheter at least 7 days before the onset of urinary tract infection, ii) central and peripheral vascular catheter was in place at least 48 h before onset of bloodstream infection, and iii) intubation in the preceding 48 h before the onset of pneumonia.

Data analysis
Data was entered into, double checked for coding errors, cleansed and analyzed using Statistical Package for the Social Sciences (SPSS) version 23. Categorical data was reported as frequency and percentage while continuous data was presented as mean (standard deviation) or median. Risk factors for HAI were determined using bivariate and multivariate regression analysis. Only variables that demonstrated statistical significance (P < 0.05) in the bivariate analyses were included in the multivariate model.

Hospital characteristics
Three public hospitals participated in this pilot pointprevalence survey including one university teaching hospital and two general hospitals. The total bed capacity of the selected hospitals was approximately 860 beds. A total of 321 hospitalized patients were included in this pilot survey representing 37.3% of the total bed space in the hospitals. There was an active infection control and prevention (IPC) team in two of the hospitals, however, none of the hospitals had an active surveillance program at the time of the survey. All the hospitals had a microbiology department with a functional laboratory to process specimens which was opened on every day of the week. However, lack of steady electricity supply affects the processing of specimens collected for microbiological culture and sensitivity test.

Patient characteristics
Three hundred and 21 hospitalized patients were included in the analysis with a median age of 27 years (range: 1-93 years). The median length of stay on the survey date was 8 days (IQR; 1-231 days). Females constituted approximately 58% of all patients surveyed and about one-third of the patients had surgery during the current admission. On the day of the survey, central catheter, peripheral catheter, urinary catheter and intubation was present in 2.8, 54.2, 15.0 and 8.7% of the surveyed patients, respectively. Table 1 describes the clinical and demographic characteristics of the surveyed patients.

Prevalence of hospital acquired infections
A total of 50 HAIs were identified in 46 patients translating into a point-prevalence of 14.3%. Forty-two (91.3%) patients had one HAI while the remaining four (8.7%) patients had 2 HAIs each. The prevalence of HAI in the tertiary hospital (15.1%) was comparable to the prevalence in secondary hospitals (13.2%). The prevalence of HAIs ranged from 0.0 to 20.0% in the selected hospitals. The prevalence HAI was significantly higher in the neonatal unit (53.6%) followed by pediatric surgical (26.7%), surgical (10.1%) and OBG (10.0%) units. About one-third of all HAIs were detected in the neonatal unit while 18% each were identified in medical and pediatric surgical units. Table 2 shows the prevalence of HAI among the patients disaggregated based on the patient/specialty.

Types of hospital acquired infections
Bloodstream infections; including neonatal sepsis and neonatal laboratory confirmed bloodstream infection  Table 3 shows the distribution of the types of HAIs based on patient's specialty.

Device associated infections and micro-organisms
Of the HAI identified, 16% (8 infections Table 4 summarizes the factors associated with HAI. There was no significant association between HAI and the variables in the multivariate regression model. Of the 321 patients surveyed, 257 (80.1%) used at least one antibiotic on the day of the survey. The indications for antibiotic use included: community acquired infection (38.7%), surgical antibiotic prophylaxis (22.5%), hospital acquired infection (16.3%), medical prophylaxis (14.9%) and unknown (7.6%). The most common antibiotics used were metronidazole (30.5%), ciprofloxacin (17.1%), ceftriaxone (16.8%), amoxicillin-clavulanic acid combination (12.5) and gentamicin (11.8%). The data for antimicrobial use among the patients has been published elsewhere [18].

Discussion
The current study revealed that the point-prevalence of HAI was 14.3% with bloodstream and surgical site infections accounting for two-third of all the HAIs. Neonatal and pediatric surgical specialties had the highest prevalence and represented more than 50% of all the HAI.   The prevalence of HAI in the present study was higher than the 2.6-6.3% reported in previous retrospective studies conducted in Nigeria [6,13,14]. However, the current study is comparable to the cumulative prevalence of HAI in developing countries [7] and Ethiopia [19], but higher than the prevalence in Ghana [20], South Africa [21], Iran [22], the United States [2] and the European hospitals [4]. It is important to note that methodological differences between these studies influence the findings and could explain the inconsistencies. Although overcrowding is not an issue in the selected hospitals, the rate of HAIs was high. This could be attributed to the lack of a national infection control policy and guideline, poor infection control practices, lack of surveillance, inadequate clinical waste management and the dearth of infection control personnel in the hospitals. Bloodstream infections, particularly neonatal sepsis, surgical site infections and pneumonia were the most common HAIs in the current study, consistent with the findings of an Iranian study [22], but inconsistent with the results in other countries [2,4]. Previous studies conducted in Nigeria demonstrated that urinary tract infection (UTI) was the most common HAI [6,13].
In contrast, the current study did not identify any UTI among the hospitalized patients surveyed and this could be explained by the requirement to perform microbiological investigation before diagnosing hospital acquired UTI which was not frequently observed. Intravascular catheterization is an independent risk factor for hospital acquired bloodstream infections [22]. High rate of surgical site infections among the surveyed patients was consistent with the result of a previous study conducted in South-Western Nigeria [13]. These infections are preventable with adequate infection control, blood glucose control and surgical antibiotic prophylaxis. However, compliance with surgical antibiotic prophylaxis recommendations in Nigeria is inadequate [23]. Therefore, strategies to improve compliance with infection control and surgical antibiotic prophylaxis, particularly the use of antimicrobial stewardship interventions, are recommended. Antimicrobial stewardship interventions were effective in improving compliance with antibiotic selection, timing and duration of surgical antimicrobial prophylaxis in a recent study conducted in Nigeria [24]. Of the surveyed patients, newborn (less than 1 month old), those with intubation tube and those admitted under neonatal and pediatric surgical specialties had significantly higher Odds for HAIs. This was consistent with previous studies for age [15] and intubation [22,25], respectively. The high risk of HAIs among newborns depict suboptimal levels of intrapartum and post-delivery care, particularly among neonates born outside the hospital.
The current study is the first prospective pointprevalence study to describe the prevalence of HAI in Northern Nigeria. There are some limitations in the study, thus, the results should be interpreted with caution. First, the findings cannot be generalized for the entire region and Nigeria at large because the data was collected in three hospitals only. Secondly, the study design used (point-prevalence survey) which is not the gold standard for the surveillance of HAI and the prevalence reported could be either over-or under-estimated. However, point-prevalence study is a valid method for the surveillance of HAIs in resource limited settings [16]. Thirdly, there was no concurrent validation of the collected data during the survey due to lack of resources and as a result, the findings could be over-or underestimated. Fourthly, information regarding the number of laboratory-or radiologically-confirmed HAIs as well

Conclusion
The prevalence of HAIs is relatively high compared to other developing countries with bloodstream and surgical site infections as well as pneumonia being the most common. Acquiring HAI was significantly associated with newborn, intubation and neonatal and pediatric surgical specialties. Hospital infection control and prevention strategies need to be strengthened to improve the quality of care among hospitalized patients.