Background
A safe injection never harms the recipient, does not expose the provider to avoidable risks and does not result in any waste that is dangerous to other people [
1]. The World Health Organization (WHO) estimated that in 2000, 16 billion healthcare injections were given each year in developing and transitional countries [
2]. Of these, 90–95% were for therapeutic purposes, while 5–10% were immunizations [
2].
Injections are often used unnecessarily when oral medicine could be equally effective [
3]. WHO estimated that in 2000, 39.8% of injections were given with devices reused in the absence of sterilization [
2]. Health care injections given with re-used equipment expose patients to infections with bloodborne pathogens, including hepatitis B virus (HBV) [
4], hepatitis C virus (HCV) [
5], and human immunodeficiency virus (HIV) [
6]. Overuse of injections to administer medicines amplifies the risk of transmission. WHO estimated that in 2000, overuse and unsafe use of healthcare injections caused 30% of new infections with HBV (21 million), 41% of new infections with HCV (2 million), and 9% of new infections with HIV (260,000, annually) [
7].
Since 2000, WHO worked on the establishment of policies for the safe and appropriate use of injections worldwide. The Safe Injection Global Network (SIGN) regrouped efforts from all stakeholders, including international organizations, governments, nongovernmental organization, civil society, and industry [
8]. From 2004, the Demographic and Health Surveys (DHS) started to include data on injection practices for surveys conducted in several countries [
9]. DHS are nationally representative population-based surveys of adult population with large sample sizes (e.g., > 5,000 households). The DHS Programme has provided technical assistance to more than 300 surveys in over 90 countries, advancing global understanding of health and population trends. New questionnaire items included injection frequency (the average number of healthcare injections reported per person per year) and injection safety (whether for the last healthcare injection, the syringe and needle came from a new, unopened package). In the year 2010, WHO commissioned an update of the 2000 injection practices estimates. This 2010 review which largely used DHS data. The results indicated that between 2000 and 2010, the proportion of reuse of injection devices dropped from 39.8 to 5.5%. Meanwhile, the frequency of use of injection to administer medications had not decreased [
10]. Overuse and unsafe use of injections still led to transmission of bloodborne pathogens, although less so than in 2000. From 2000 to 2010, despite a 13% population growth, there was an estimated reduction of 87 and 83%, respectively, in the absolute numbers of HIV and HCV infections transmitted through healthcare injections. For HBV, the reduction was more marked (91%) due to the additional impact of vaccination [
11].
In 2016, WHO published injection safety guidelines recommending safety engineered injection devices to eliminate unsafe injections [
12]. The WHO policy document specifically addressed the use of these devices for therapeutic injections. Further, WHO launched a comprehensive behaviour change and multimodal implementation strategy addressing different audiences and stakeholders to facilitate adoption of the injection safety recommendations according to a strategy in 2017 [
13]. A package of tools (available online) promote communication for behaviour change among patients and health care workers to reduce injection overuse and ensure safety. More recent and higher quality data are needed to monitor the evolution of injection safety since 2011, including to track progress before and after the launch of the WHO policy [
14]. However, systematic reviews can be costly and time consuming. The ongoing collection of injection practices information within DHS provided an opportunity to obtain new estimates easily and rapidly. The objectives were to describe injection practices since 2011, including heterogeneity by region and gender, and to determine if injection safety has improved since 2011.
Methods
We reviewed healthcare injection practices worldwide in terms of frequency and safety according to the indicators included in DHS since 2011 and until 2015. The datasets analyzed during the current review are available in the DHS Program webpage (
https://www.dhsprogram.com/). If countries surveyed since 2011 had a prior DHS survey data including injection practice up to 2010, we included the prior survey to compare data in 2004–2010 and in 2011–2015.
Data collection during the DHS surveys
DHS questionnaires were conducted in countries and adapted from template survey instruments developed by the DHS programme to reflect the population and health issues relevant to each country. The questionnaire was translated into each native language. A nationally representative sample of households (women age 15–49 and men age 15–49, 54, or 59) was eligible for individual interviews that were performed individually. We focused on two template questions. These were ‘have you had an injection for any reason in the last 12 months? If yes, how many injections have you had?’ and ‘the last time you got an injection from a health worker, did he/she take the syringe and needle from a new, unopened package?’. When considering DHS questionnaire items, we used the term ‘frequency’ to refer to the average number of health care injections per person in the last 12 months, and used the term ‘safety’ to refer to the proportion of injections for which the respondent mentioned the syringe and needle had been taken from a new, unopened package.
Search for DHS surveys
We accessed the homepage of DHS Program to search all DHS final reports in 2011–2015 for data on injection practices. If the surveyed year spanned over two years (such as 2015–16), we considered the survey to have been conducted in the year it was initiated. When the data was available, we also searched the survey data conducted in 2004–2010 in the same country for comparison.
Data abstraction
We extracted the data on frequency and safety. The denominator for frequency was the total number of respondents. The denominator for safety was the number of respondents who had received at least one health care injection in the last 12 months.
Data analysis
Some countries had only information on injections received in the last six months. In that case, we doubled frequency to adjust for the different referent period as done in prior reviews [
2]. We multiplied the number of injections received by the proportion unsafe to calculate the number of unsafe injection per person per year [(the average number of health care injections per person in the last 12 months) x {1-(the proportion of injections from a new package)}]. For countries that had surveys conducted at the subnational level, we calculated the average value of the subnational surveys.
We combined the data from the various surveys to estimate the global situation in terms of frequency and safety. We compared gender groups, regions, and countries in terms of the number of unsafe injections. For countries that had more than two DHS surveys since 2004, we compared practices in 2004–2010 and in 2011–2015 in terms of safety and frequency. If there were multiple surveys up to 2010, we adopted the most recent one to reflect on the most recent evolution. For countries that only had data on women, we restricted the comparison to women for the two time periods.
Discussion
In 2010, a review of injection practices globally reported improvement in injection safety following the implementation of joint efforts from WHO, partners of the Safe Injection Global Network, other agencies and countries [
10,
11]. However, it also pointed to the absence of improvement in reducing the frequency of health care injections annually. This new rapid review provides an indication of the evolution of the situation since 2011, and sheds light on the additional progress and setbacks.
The most recent DHS surveys suggested that in 2011–2015, the trend towards improvement of injection practices continued. However, injection frequency and safety continued to vary widely by region. The WHO EMR has the highest number of injections and the highest proportion of unsafe injections, followed by the WHO SEAR. The number of unsafe healthcare injection in the WHO EMR was also the highest in the world in 2000 (2.96 per person per year) [
2] and in 2010 (0.57 per person per year) [
10]. Many injections are administered by private health care providers who may not have formal medical qualification in these regions. In such informal situations, the attitude of the healthcare provider also promotes the abuse of injections. Compliance of standard treatment guidelines is not common. To meet the demands of the user, injections are often utilized on an “ad hoc” basis to administer drug mixtures such as antibiotics, analgesics, vitamins or antihistamines [
15,
16]. If safe injections were administered, reduction of injection overuse would only be a matter of facilitating rational use of medicines [
17]. However, reuse of syringes frequently leads to transmission of bloodborne pathogens [
18‐
25]. In EMR and SEAR, especially India and Pakistan, blood-borne pathogens are prevalent and can be transmitted, around 1.7–5.6% for HBV [
26], 0.5–3.8% for HCV [
27], and 0.3–0.1% for HIV [
28]. For instance, unsafe health care injections remain a key driver of the HCV epidemic in Pakistan [
29]. It is also a possible vector of HIV infection [
30]. This could be prevented as patients’ knowledge regarding transmission of bloodborne pathogens drives consumer demand for new syringes [
31,
32].
According to DHS, women generally received more injections than men. The reasons for this may include obstetrical and gynecological conditions, including contraception, delivery, and prevention of maternal and neonatal tetanus. DHS data also suggest that there is no difference in injection safety between women and men. Some of the injections received by women could be received through programmes such as immunization or contraceptive treatment rather than informal health care. However, the absolute number of unsafe injections was higher in women than in men. Actually, exposure to healthcare (e.g. health care injections, hospitalizations and pregnancies) has been associated with HCV infection among women in Pakistan [
33,
34]. Addressing this issue has the potential to reduce gender inequalities in injection safety. Since 1999, the principle of “bundling” has been proposed to ensure the safety of injections given through key public health programmes, so that the sources of financing of immunization and contraception commodities also pay for safe injection devices and safety boxes for disposal of used injection equipment [
35]. Bundling refers to the concept of a theoretical “bundle” which comprises each of the following items: 1) good quality vaccines, 2) auto-disable syringes, and 3) safety boxes. “Bundling” does not have physical connotation and has no implication that items should be “packaged” together [
36]. In EPI, application of the concept of bundling represents a substantial incremental cost as vaccines are inexpensive. However, implementation is facilitated since immunization services are usually centrally funded [
37]. For therapeutic injections, the concept of bundling was generalized into broader “guiding principles for injection device security” [
38]. The incremental cost is much lower because medicines are usually more expensive than the injection device required to inject them [
39]. However, implementation is challenging since the funding of health care services is usually more fragmented.
Comparison of the 2004–2010 and 2011–2015 periods in 16 countries with available data suggested an improvement in many countries. Concerted efforts of ministries of health, local health facilities, non-profit organizations, and donors may explain some of the progress. However, in Pakistan, Uganda and Zambia, the number of unsafe injections did not change substantially since 2011. Pakistan is a country with a rapidly increasing population. The effects of the reduction in unsafe injections might be reduced by the population growth in some densely populated regions. However, in theory, the scale up of safe injections could be more cost effective and positively associated with infections averted in growing populations [
40]. This is important because Pakistan has the second highest number of HCV infections in the world [
27].
Injection-associated transmission of bloodborne pathogens can be prevented through the development of a strategy to reduce injection overuse and achieve injection safety and its implementation by a national coalition [
41]. Achieving this goal necessitates the establishment of a national multidisciplinary cooperation between the Ministry of Health and stakeholders such as private healthcare providers and non-governmental organizations. Changing behavior strategies towards consumers and the public, private and general healthcare professionals should be strongly supported. To implement such a strategy, WHO suggests a package of tools available online for safe and appropriate injection use in planning, implementation and evaluation of national policies in order to support national policy makers [
13]. Equipment and supplies should be provided to eradicate the re-use of syringes and needles without sterilization. Sharps waste should be managed to ensure that disposable syringes and needles are not re-used and do not lead to needlestick injuries [
38]. In addition, since 2016, WHO recommends a policy to prefer the use of sharp injury protection syringes for intramuscular, intradermal and subcutaneous injections over the use of re-use prevention syringes for intramuscular, intradermal and subcutaneous injections [
12]. Syringes with sharps injury prevention features reduce the incidence of needlestick injuries [
42]. Healthcare workers perceive injection safety devices to be generally easy to use, safe, and tolerated by patients [
43]. Rapid implementation of this policy could address residual injection safety issues in resource-limited settings where unsafe practices persist.
Our review and analysis of existing data have a number of limitations. Some limitations are inherent to the methods used by DHS. Survey respondents were asked whether they had received any injections from a health worker in last 6 or 12 months before the survey. Therefore, they might not have noticed or remembered precisely whether the injection they are receiving was given with a new needle and syringe from an unopened package when interviewed months later (recall bias). Furthermore, the respondents might tend to answer questions in a manner that will be viewed socially favorable because the data was collected by questionnaire (social desirability bias). There are also limitations due to our review methods. First, we did not calculate confidence intervals, did not perform statistical comparisons and did not model regional or global estimates. Calculation of confidence intervals in statistical analyses is based on the assumption of random sampling, yet the countries that were included in the DHS programme were not chosen at random. Further, the low proportion of unsafe injections, the lack of variability in terms of income groups and the limited sample size of countries included would have limited efforts to model regional and global estimates. Second, we used only one data source (DHS), which provides limited evidence in some areas while there could be other data sources for injection safety. Third, doubling the six month recall may overestimate the response that would’ve been generated by asking people to recall 12 months. Fourth, there are different samples sizes by the different regions or countries, therefore, it is hard to compare across regions precisely. Finally, there is large variation globally by region and a lack of data from many countries. Of 92 countries with DHS data available, injection safety data were available from only 40 countries, and these differed from the 16 countries with 2004–2010 data. Overall, our estimation is an imperfect reflection of the current status worldwide, and the review is limited to healthcare injections. In many countries non-healthcare injections, including injecting drug use, play an important role in the transmission of bloodborne infections through unsafe injections.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.