## Abstract

Background. Unsafe delivery and overuse of injections can result in the spread of hepatitis B virus, hepatitis C virus, and HIV. The aim of the present survey was to estimate the frequency of safe injection practices in Burkina Faso.

Method. Using the new standardized World Health Organization tool to assess injection practices, we selected 80 primary health facilities with a two-stage cluster sampling method, collected information using structured observations and provider interviews, and analyzed the data using Epi-Info software.

Results. We observed 116 injections in 52 facilities. In 50 facilities [96%; 95% confidence interval (CI) 85–99%] injections were given with a new, single-use syringe and needle. In 29 facilities (56%; 95% CI 36–74%), staff recapped needles using two hands. All 80 facilities visited had a stock in the community to provide new, single-use syringes and needles. In 61% (95% CI 54–79%) of facilities, staff reported needlestick injuries in the last 12 months. Used needles were discarded in open containers in 66 facilities (83%; 95% CI 55–96%) and observed in the surroundings of 46 facilities (57%; 95% CI 32–80%).

Conclusions. In 2000, most of the health facilities in Burkina Faso were using sterile injection equipment. However, practices were still observed that could expose patients, health care workers, and communities to risks, and that required specific interventions.

A safe injection should not harm the patient, the provider, or the community. More specifically, injections must be given with a sterile syringe and needle that are collected immediately in a safety box for subsequent safe disposal. Several reports have stated that unsafe injection practices are common in the developing world [1]. Kane, in a simple mass action model, estimated the incidence of infections with selected blood-borne pathogens that may be attributable to unsafe injections. The results of the model suggests that each year, unsafe injections may cause 8–16 million hepatitis B virus (HBV) infections, 2.3–4.7 million hepatitis C virus (HCV) infections, and 80 000–160 000 human immunodeficiency virus (HIV) infections [2]. These infections may lead to death or disability, as well as having substantial socio-economic consequences. Unsafe injections can be prevented through: (i) training and communication of safe injection practices; (ii) supplying sufficient quantities of appropriate injection equipment; and (iii) managing the waste associated with used injection equipment.

In Burkina Faso, two studies addressing injection practices have been reported since 1989. A study conducted in 1989 described observations of injection practices in four health care facilities and suggested that non-sterile equipment was being used for injections [3]. In 1995–1996, another study conducted using the World Health Organization Regional Office for Africa (AFRO) logistics project’s rapid assessment protocol assessed the safety of injections in Burkina Faso [4]. Results indicated that in Burkina Faso, one syringe and one needle were used for each injection in 80% of urban health care facilities, in 60% of provincial health care facilities, and in only 11% of rural health care facilities. Shortages of safe injection equipment may have explained in part the unsafe practices observed. Sufficient quantities of injection equipment were available in 80%, 52%, and 67% of urban, provincial, and rural health care facilities, respectively.

A new standardized tool to assess injection practices in a representative way was developed in 2000 [5]. The objectives of this tool are to: (i) estimate the frequency of unsafe injection practices in health care facilities; (ii) determine whether a facility where injections are given meets the necessary requirements with respect to equipment, supplies, and waste disposal; (iii) determine whether the critical steps of an injection administration are executed; and (iv) identify unsafe practices that may lead to infections and that should be targeted by interventions. We assessed with this tool the injection practices in Burkina Faso in June 2000.

## Methods

### Sampling methodology

As per the standardized assessment tool methods, we conducted a cross-sectional study with two-stage cluster sampling to select eight clusters of 10 health care facilities [5]. Clusters were selected with a probability of selection proportional to the population size using the census data from 1996 projected for population growth [6]. In each selected cluster, 10 health care facilities were randomly selected from a list of all primary health care facilities provided by the Ministry of Health [7]. At the same time, five additional facilities were randomly selected and listed as potential replacements. The sampling frame for the standardized assessment tool was designed to be applicable to all countries and to require limited time and personnel while achieving a precision comparable to that of the Expanded Program on Immunization coverage survey (±10%) [8,9].

### Data collection

Four teams collected information in the field in June 2000. A standardized data collection instrument consisted of: (i) structured recording of types and quantities of available injection equipment; (ii) structured observations of injection practices; and (iii) interviews of health care workers. All teams standardized their data collection procedure before fieldwork commenced, through a training session followed by a field visit with cross-validation between the teams. Participation in the survey was voluntary for all participants, who were informed of their right to refuse participation. Firstly, the most senior person at the facility was informed and asked for their participation, and all questions relevant to his or her position were asked. Secondly, the injection practices of the person that carried out injections most frequently were observed. If any persons refused to participate, a new facility was to be selected from the list. It was clearly stated that the observation was not an evaluation and that the study would be most valuable if they did not change their habitual practice. But when injections were about to occur with non-sterile equipment, field workers were asked to tactfully interrupt the procedure. Injecting practice was observed in health facilities and the assessment team would wait or visit the facility again, unless it was evident that no further injections were likely to be given during the day, whereupon the facility was included in the survey but without the observation of an injection.

### Data analysis

Proportions were calculated using the total number of facilities without missing values as a denominator. Design effects were calculated with the CSAMPLE module of Epi-Info, 6.04d. Design effects were then entered into the Epi-Table module of EPI-Info for calculation of confidence intervals using the Fleiss quadratic formula [10].

## Results

### Health care facilities included in the sample

A total of 80 primary health care facilities were included in the sample. Of these, 13 (16%) were chosen from the replacement list because the facilities originally selected were inaccessible due to poor road conditions (n = 12) or reconstruction (n = 1). The administration of 116 injections was observed in 52 (65%) of the facilities. Of these injections, 40 (34%) were administered for vaccination purposes.

All facilities routinely used new, single-use injection equipment. Patients purchased the injection equipment for vaccination as well as for therapeutic injections at community pharmacies, usually located next to the dispensary. Health care workers interviewed reported no shortages of injection equipment in these pharmacies during the last 12 months. In addition to stocks held in community pharmacies, 23 (28.8%) facilities had stocks of syringes and needles, and 41 (51%) had stocks of safety boxes. These stocks were mostly left over from a national vaccination day conducted in December 1999. Five (6%) facilities used sterilizable injection equipment in addition to single-use injection equipment. This equipment was solely used for BCG (Bacillus of Calmette and Guerin vaccine) administration. Of these, three used sterilizable needles with a single-use syringe.

### Risk to the patient

Despite the predominant use of new, single-use injection equipment, in two facilities (4%), syringes were observed to be reused in the absence of sterilization (Table 1) while the needles were changed between patients. In another facility (2%), new, clean, sterilizable BCG needles were used without prior steam sterilization and then thrown away, as if they were designed for single use. In 27 (52%) of the facilities where an injection was observed, the injection was prepared in a dedicated, clean area where blood or body fluid contamination was unlikely to occur. In 19 facilities (95% of those where at least one vaccination was observed), the vaccine was kept at a cool temperature (i.e. in a refrigerator or kept on an ice pack) before administration.

Table 1

Findings from the observation of injection practices in studied health care facilities: indicators assessing the risk to the patient (Burkina Faso, June 2000)

Item observed #/n1 95% CI (%)
Preparation of injections in a clean dedicated area 27/52 52 32–71
Administration with a sterile syringe 50/52 96 85–99
Administration with a sterile needle 51/52 98 89–100
Removal of needles from multi-dose vials between injections 17/19 89 64–98
Temperature-sensitive products kept cool during preparation 19/20 95 70–100
Absence of shortages of disposable injection equipment 80/80 100 89–100
Item observed #/n1 95% CI (%)
Preparation of injections in a clean dedicated area 27/52 52 32–71
Administration with a sterile syringe 50/52 96 85–99
Administration with a sterile needle 51/52 98 89–100
Removal of needles from multi-dose vials between injections 17/19 89 64–98
Temperature-sensitive products kept cool during preparation 19/20 95 70–100
Absence of shortages of disposable injection equipment 80/80 100 89–100

n, number of facilities without missing information; #, number of observations with a response of ‘yes’; CI, confidence interval.

1

Variable denominators denotes that information was not available for all facilities; injections were only observed in 52 of the 80 facilities visited.

### Risk to the health care worker

In 23 (44%) of the facilities where injections were observed, needles were recapped with two hands (Table 2). Used needles were collected in a safety box in only three (6%) facilities. In 66 (83%) facilities, sharps were found in open containers or lying around where they could expose health care workers to needlestick injuries. In addition, inappropriate containers (e.g. plastic bottles) were used for the collection of sharps in some of these facilities. In 61% of facilities, needlestick injuries were reported among staff members in the last 12 months.

Table 2

Findings from the observation of injection practices in studied health care facilities: indicators assessing the risk to the health care worker and community (Burkina Faso, June 2000)

Item observed #/n1 95% CI (%)
Presence of at least 10 safety boxes 10/80 13 1–24
Presence of dirty sharps in open containers 66/80 83 55–96
Two-handed recapping 29/52 55 36–74
Immediate collection of sharps in safety boxes 3/52 1–23
Number of reported needlestick injuries in the last 12 months 44/71 61 54–79
Provision of safety boxes for vaccination injections 0/80 0–11
Absence of sharps around the facility 46/80 58 34–81
Existence of a health care waste management policy 7/80 1–17
Item observed #/n1 95% CI (%)
Presence of at least 10 safety boxes 10/80 13 1–24
Presence of dirty sharps in open containers 66/80 83 55–96
Two-handed recapping 29/52 55 36–74
Immediate collection of sharps in safety boxes 3/52 1–23
Number of reported needlestick injuries in the last 12 months 44/71 61 54–79
Provision of safety boxes for vaccination injections 0/80 0–11
Absence of sharps around the facility 46/80 58 34–81
Existence of a health care waste management policy 7/80 1–17

n, number of facilities without missing information; #, number of observations with a response of ‘yes’; CI, confidence interval.

1

Variable denominators denotes that information was not available for all facilities; injections were only observed in 52 of the 80 facilities visited, as sometimes no injections took place.

### Risk to the community

The most common waste disposal technique observed was open burning in 73 (91%) facilities (Table 3). Of the facilities using open burning, 32 (51%) did so in an uncontrolled area. In one facility, the waste was burned in an incinerator. Other waste disposal methods observed included dumping, either in a deep pit [n = 2 (3%)] or in an uncontrolled area [n = 4 (5%)]. Due to these inappropriate waste management techniques, we observed sharps in the vicinity of 46 (57%) facilities (Table 2).

Table 3

Observed frequency of methods of elimination of sharps in studied health care facilities (Burkina Faso, June 2000) (n = 80)

Observed method of sharps waste management #/n 95% CI (%)
Uncontrolled open burning 41/80 51 34–69
Controlled open burning 32/80 40 24–58
Functioning incinerator 1/80 0–15
Burial 0/80 0–13
Dumping in latrines or other safe depository 4/80 1–20
Dumping in unsafe places 2/80 0–17
Removal for disposal at another location 0 /80 0–13
Observed method of sharps waste management #/n 95% CI (%)
Uncontrolled open burning 41/80 51 34–69
Controlled open burning 32/80 40 24–58
Functioning incinerator 1/80 0–15
Burial 0/80 0–13
Dumping in latrines or other safe depository 4/80 1–20
Dumping in unsafe places 2/80 0–17
Removal for disposal at another location 0 /80 0–13

n, number of facilities without missing information; #, number of observations with a response of ‘yes’; CI, confidence interval.

## Discussion

The results of this injection safety assessment suggest an increase in the use of sterile injection equipment in Burkina Faso between 1995 and 2000 [4]. While this increase is a major breakthrough, challenges remain. Firstly, the reuse of syringes with different patients was still observed. Secondly, the collection and disposal of sharps waste was carried out inappropriately. These setbacks exposed patients, health care workers, and the community to blood-borne pathogen infections.

Previous studies conducted in Burkina Faso to assess injection safety suggested that unsafe practices were common in health care settings. One study conducted in 1989 reported a high number of injections given, while the number of syringes and needles discarded did not match the number of injections, suggesting the possibility of reuse of disposable injection equipment on different patients [3]. A second study conducted in 1995–1996 in a convenience sample of health care facilities suggested that a large proportion of facilities had experienced shortages of injection equipment in the last 12 months, and that the use of sterile injection equipment for each injection ranged from 80% in urban facilities to 11% in rural facilities [4]. In contrast, our study conducted on a sample representative of the country indicated that for 96% of injections observed, a new sterile needle and syringe was used, and no shortages of equipment were reported. The assessments conducted in 1995–1996 and in 2000 used different methodologies and the reports provided no information on confidence limits nor on how representative these assessments were; hence a comparison is difficult. Although a comparison is not legitimate, there are two hypotheses that could explain the improvement, if it was real, in injection safety in Burkina Faso. Firstly, between 1996 and 2000 a number of changes in the national essential medicine policy led to better access to community pharmacies, which are managed according to the principles of the Bamako initiative [11,12] and which supply injection equipment at low cost (approximately US$0.10 for one syringe and needles set). This may have led to an increased consumption of single-use injection equipment in the country. The second hypothesis that could explain safer injection practices in Burkina Faso in 2000 is a higher degree of awareness of the risks associated with unsafe injections among health care workers and the population. The AIDS epidemic in Africa could have raised the demand for safe equipment among users of the health care system. Meda et al. studied the knowledge in the community regarding the routes of transmission of HIV/AIDS in Burkina Faso in 1994–1995 [13]. In this study, the authors interviewed 1294 pregnant women, 236 long-distance truck drivers, and 426 sex-workers. Only 16.9% of the pregnant women, 25.4% of the truck drivers, and 11.5% of the sex-workers knew about the risk of HIV/AIDS transmission through contaminated syringes and needles. In a new study conducted by Meda in 2000 in which 2693 persons were randomly selected in Bobo-Dioulasso, the second largest city in Burkina, 60.4% of all persons questioned knew about the possibility of HIV transmission through injections (N. Meda, Burkina Faso, personal communication). While these two studies were carried out in different populations, an increase in the awareness regarding the risk of HIV transmission through injections cannot be excluded. While injection practices seemed to have become safer in Burkina Faso in the years leading up to the study, they remained potentially dangerous in isolated cases for the patient, for health care workers, and for communities. Syringes were still reused without sterilization in two facilities. Although isolated, such practices must be vigorously eliminated because they are associated with a high risk of transmission of blood-borne pathogens to the person receiving the injection [14,15]. Unsafe sharps collection leads to needlestick injuries among injection providers. Two factors may have contributed to the high number of reported needlestick injuries. Firstly, only a few facilities had sufficient supplies of safety boxes and there was no regular mechanism to ensure an adequate supply. The availability of safety boxes has been associated with a lower risk of needlestick injury, as open boxes facilitate the safe disposal of contaminated sharps [16]. Secondly, the practice of two-handed recapping was observed in many facilities, which has been associated with a high risk of needlestick injuries [17]. Needles should be thrown into the safety box without recapping; if recapping is practiced, the cap should never be held. The so-called ‘one-handed recapping’, where the cap is on a table, is an acceptable procedure, but is unnecessary if safety boxes are used. The observation of two-handed recapping in facilities where safety boxes were available suggests that lack of knowledge also contributed to unsafe waste collection practices in Burkina Faso. In addition to unsafe sharps waste collection practices, we observed inappropriate techniques for the management of sharps waste that resulted in the presence of contaminated sharps in close proximity to a high proportion of facilities. Explanations for this situation include the lack of a national policy for safe health care waste management, no identified responsible person, and the absence of technical options for disposing of sharps waste in most locations. Our study had a number of limitations. Firstly, we had to replace 12 facilities that could not be accessed due to road conditions in the rainy season. This could have led to a shift to accessible facilities that could have benefited from a better supply line. In discussion with the local authorities they assured us, however, that these facilities are supplied with motorcycles. However, in a worst-case scenario where all these facilities would have reused injection equipment in the absence of sterilization, the use of sterile injection equipment would still have been higher than in 1995. Secondly, we did not check if the equipment that was taken out of a new package was sterile. If the procurement system did not ensure the quality and safety of the equipment, this could have led us to overestimate the proportion of sterile injections. However, syringes and needles were procured through community pharmacies that were supplied centrally through bulk purchasing from large international wholesalers of generic medications. Thus, it is unlikely that these syringes would not have met international quality standards. Thirdly, this assessment only focused on public health care facilities. These facilities offered services against the payment of users’ fees. We were not able to assess whether the population was not able to use them for economical reasons. Poor financial accessibility could have serious consequences if immunizations or life-saving medications could not be afforded. There is also a theoretical possibility that fees for injection equipment and services could have led to a shift of clients to other informal systems that may administer injections under unsafe circumstances. However, anecdotal reports suggested that while traditional healers are common in Burkina Faso, private informal providers administering injections are uncommon in the country. Challenges ahead include the need to bring attention to the fact that all injections should be safe in Burkina Faso. The first priority should be given to ensuring that all injections are given with sterile injection equipment. A solution may consist of the use of auto-disable syringes, which cannot be reused, although these are currently only available for immunization injections [18,19]. Secondly, to ensure the safety of health care workers, puncture- and liquid-proof safety boxes should be supplied to health care services in sufficient quantities. These safety boxes should be used to dispose of syringes and needles immediately after injections without recapping. Finally, a national health care waste management policy is necessary to ensure that the sharps are not lying around in the vicinity of health care facilities, exposing the community to needlestick injuries and infection with blood-borne pathogens. Using a standardized tool based upon a representative sampling method, we were able to quantify the injection safety problems and show the areas of greatest concern. These findings can be used as a baseline and the success of changes introduced can be measured with a repeated survey. Since this initial injection safety assessment in Burkina Faso, the new World Health Organization (WHO) tool to assess injection safety was used in >20 countries, for a unit cost ranging between US$10 000 and US\$20 000, with good feasibility. Although no gold standard is available to validate the results generated, the use of this tool in baseline assessments has demonstrated its usefulness in focusing attention on country-specific injection safety issues so that efforts to make injections safe can be more effective. In addition, standardization provides the possibility of comparing countries and helps in evaluating the different strategies in place to ensure safe injection practices. Such a capacity to measure progress will assist in driving the international efforts for the safe use of injections worldwide.

At the time of the survey, J.-F. Aguilera and J. Fitzner were fellows of the European Programme for Intervention Epidemiology Training (EPIET) and the German Field Epidemiological Training Programme (FETP), respectively.

Financial support for this assessment was provided by a grant from the United States National Vaccine Programme Office (NVPO) to WHO through the United States Agency for International Development (USAID), and by the European Union through their project Appui au Renforcement de l’Independence Vaccinale en Afrique Sahelienne (ARIVAS).

## References

Simonsen
L
, Kane A, Lloyd J, Zaffran M, Kane M. Unsafe injections in the developing world and transmission of bloodborne pathogens: a review.
Bull World Health Organ

1999
;
77
:
789
–800.
Kane
A
, Lloyd J, Zaffran M, Simonsen L, Kane M. Transmission of hepatitis B, hepatitis C and human immunodeficiency viruses through unsafe injections in the developing world: model-based regional estimates.
Bull World Health Organ

1999
;
77
:
801
–807.
Vincent-Ballereau
F
, Lafaix C, Haroche G. Incidence of intramuscular injections in rural dispensaries in developing countries.
Trans R Soc Trop Med Hyg

1989
;
83
:
106
.
Dicko
M
, Oni AQ, Ganivet S, Kone S, Pierre L, Jacquet B. Safety of immunization injections in Africa: not simply a problem of logistics.
Bull World Health Organ

2000
;
78
:
163
–169.
WHO.
Tool for the Assessment of Injection Safety: Basics II.
Geneva: WHO,
2001
(WHO/V&B/01.30).
Institut National des Statistiques et de la Démographie INSD du Burkina Faso.
National Census Burkina Faso 1996
. Ouagadougou: Institut National des Statistiques et de la Démographie INSD du Burkina Faso,
1996
.
Direction des Etudes et de la Planification du Ministère de la Santé.
Carte Sanitaire du Burkina Faso 1997
. Ouagadougou: Direction des Etudes et de la Planification du Ministère de la Santé,
1997
.
Lemeshow
S
, Tserkovnyi AG, Tulloch JL, Dowd JE, Lwanga SK, Keja J. A computer simulation of the EPI survey strategy.
Int J Epidemiol

1985
;
14
:
473
–481.
Lemeshow
S
, Robinson D. Surveys to measure programme coverage and impact: a review of the methodology used by the expanded programme on immunization.
World Health Stat Q

1985
;
38
:
65
–75.
Fleiss
JL
.
Statistical Methods for Rates and Proportions
. New York: John Wiley & Sons,
1981
.
OG
, Arnaiz-Villena A. The Bamako initiative.
Lancet

1988
;
2
:
1177
–1178.
Osofu-Amaah
S
. The Bamako initiative.
Lancet

1989
;
1
:
162
–163.
Meda
N
, Sangare L, Lankoande S et al. The AIDS epidemic in Burkina Faso: the actual level of knowledge in the population on AIDS, 1994–95 (in French).
Rev Epidemiol Sante Publique

1998
;
46
:
14
–23.
Langille
DB
, MacDonald J, Sabourin L, Corbett GA. Contamination of multi-dose vials due to repeat usage of syringes for aspiration—Nova Scotia.
Can Dis Wkly Rep

1988
;
14
:
193
–196.
Hu
DJ
, Kane MA, Heymann DL. Transmission of HIV, hepatitis B virus, and other bloodborne pathogens in health care settings: a review of risk factors and guidelines for prevention.
Bull World Health Organ

1991
;
69
:
623
–630.
Haiduven
DJ
, DeMaio TM, Stevens DA. A five-year study of needlestick injuries: significant reduction associated with communication, education, and convenient placement of sharps containers.
Infect Control Hosp Epidemiol

1992
;
13
:
265
–271.
Jagger
J
, Hunt EH, Brand-Elnaggar J, Pearson RD. Rates of needle-stick injury caused by various devices in a university hospital.
N Engl J Med

1988
;
319
:
284
–288.
Lloyd
JS
, Milstien JB. Auto-disable syringes for immunization: issues in technology transfer.
Bull World Health Organ

1999
;
77
:
1001
–1007.
WHO–UNICEF–UNFPA.
Joint Statement on the Use of Auto-disable Syringes in Immunization Services
. WHO–UNICEF–UNFPA,
1999
(WHO/V&B/99.25).

## Author notes

1World Health Organization Headquarters, Geneva, Switzerland, 2Robert-Koch Institut, Berlin, Germany, 3Communicable Disease Surveillance Centre, Public Health Laboratory Service, London, UK, 4Vaccine Prevention, Directorate of Preventive Medicine, Ministry of Health, Ouagadougou, Burkina Faso