Our study documented that 56.5% of the 184 health workers had a positive TST result; this prevalence rose to 64.3% when we included workers with previous positive TST results or a history of active TB disease. Both estimates are within the range expected for HCWs in low-income countries (33%) and middle-income countries (79%) [13]. Recent systematic reviews have found a prevalence of 49% and 37% with a mean active TB incidence rate of 97 new cases per 100,000 HCWs [7, 8]. In addition, our result is slightly greater than the prevalence of 56% reported in a study among HCWs in Callao using IGRA [10]. These rates depict HCWs as a population at risk of developing active TB, since more than half of them have LTBI.
Similar to our study, Soto-Cabezas et al. in Peru found a significant association between LTBI and both age and time working as a HCW [10]. Likewise, Rafiza et al. in Malaysia uncovered an increased prevalence of LTBI in employees with more than 11 years of work (OR: 3.48), and Pai et al. in India reported an association between LTBI and 10 or more years on the job, a three-fold increase in prevalence compared to those employed for less than one year [16, 17]. An active TB patient with smear-positive sputum will infect, on average, between 10 to 15 people every year [18]; since most people will go to health care centers as their first point of contact for diagnosis, treatment, and monitoring of TB, it is not surprising that a longer time of employment in this health care setting reflects repetitive exposure to M. tuberculosis. Such repetitive exposure is concerning since it has not only been reported that individuals have a 10% chance of progression to active disease after a single TB exposure, but also that the probability of progression is greater among individuals with 18 or more exposures [19].
Although time spent with TB patients has been shown to be a risk factor for LTBI [7, 13, 20, 21], our analyses suggested that clinical and laboratory staff were not at higher risk for LTBI than administrative workers. This lack of association between TST results and occupation contradicts previously reported findings [7]. However, the inconsistency can be explained by the fact that health centers in SJL are limited in space, so common waiting areas and rooms are in close proximity to the area assigned for daily TB treatment. Thus, levels of TB exposure may be similar for all staff, unlike in other settings such as large hospitals. It is also possible that the lack of association between TST results and occupation may be due to the staff’s exposure to either the high prevalence of TB outside of the health center or a household contact. Moreover, the observed TST positivity may have been due to TB infection during childhood or the Bacille Calmette-Guérin (BCG) vaccination given universally in Peru at birth. Most of our participants were women, which is often the case in the healthcare setting since women constitute more than half of the national healthcare workforce [12]. We found no association between gender and TST result.
The implementation of TB transmission control measures such as natural ventilation, a supply of N95 masks, and routine screening, is essential to protect HCWs and may decrease annual TB incidence by as much as 49%, 27%, and 81% in countries with low, intermediate, and high TB incidence, respectively [22]. Previously reported obstacles preventing the use of N95 masks identified by HCWs are heat around the face and inaccessibility to the masks [23]. Our study found that 60.3% of HCWs did not wear N95 masks because the health center failed to provide them. This inaccessibility to personal protective equipment is a flaw in TB control which must be avoided in all settings, especially in countries with high TB incidence. Nonetheless, HCWs usually only wear masks when dealing with known TB cases. Since it is not common practice to continuously wear masks, TB transmission in healthcare facilities may be due to undiagnosed cases [24].
Only half of our participants had had previous active TB screening during the last year. Though it is standard procedure for health care centers to perform these screenings annually, this study reveals that health care centers are unable to ensure TB screening for all of their workers. Furthermore, considering the high LTBI prevalence uncovered in our study, it seems unlikely that only 26 participants were exempt from TST application due to a previous positive TST result. Rather, it is more likely that HCWs overreported having been tested in the last year, instead of underreporting a positive TST. In Portugal, from a sample size of 2015 registered physicians and nurses, a survey reported that 784 (39.5%) were never screened and, of those HCWs, 741 (94.5%) were never offered screening [25]. Moreover, in China, where policies on medical TB surveillance among HCWs have not been implemented, a large study identified 124 HCWs with presumptive active TB, while noting that the screening methods and framework used were not yet optimal for the high-burden of TB in the country [26]. Accordingly, either routine or post-exposure screening procedures for HCWs should be established to ensure their health.
Even though 85.2% of participants agreed to take IPT if their result was positive, some refused the prophylaxis due to misinformation about the risk of possible adverse events and immunological reactions or belief in their own pre-existing immunity against TB. Although standards of TB care in Peru dictate that IPT must be administered to health workers and prison staff within two years of a positive TST result after ruling out active TB [27], studies found acceptance rates for chemoprophylaxis amongst HCWs between 65-84% in low TB prevalence settings [28, 29]. However, the use of IPT is debatable in settings with a high prevalence of MDR-TB, as in Peru [1], which could explain why HCWs refuse the prophylaxis and prefer close monitoring of TB symptoms for at least two years, as per WHO recommendations [3]. Nevertheless, education and close monitoring of active TB symptoms must be provided for all HCWs, not only those who are TST positive.
Finally, it is important to address this study’s limitations. Recall bias may have affected the participants’ responses to questions about previous indentation, TST results, or whether or not they had presented with a confirmed case of active TB. An important possible confounding factor is that we were not able to obtain participants’ full life histories prior to beginning work in the healthcare system; thus, exposure to TB could have occurred before working as a HCW, especially if they lived in SJL or another high TB burden area in Lima. Along the same lines, we cannot be sure that we received reliable information on BCG vaccination. Self-reported use of N95 masks is also not reliable, as HCWs may feel compelled to overreport its use; additionally, the survey did not ask for a quantitative measure of mask use frequency, so this variable was open to interpretation by the HCWs completing the questionnaire. Finally, the power of the study may have been lower than necessary to detect relationships between other important variables that were not statistically significant in the multivariable analysis.