The effectiveness of disinfection protocols in osteopathic family medicine offices

Study design

A nonrandomized cross-sectional study was conducted between June 2022 and July 2022, involving 18 actively utilized family medicine examination tables in an osteopathic family medicine office. The tables were selected based on their high turnover rates to ensure a diverse range of patient interactions. To minimize behavioral changes, the study was performed without previous discussion with Environmental Services or medical staff.

The study focused on two high-touch surfaces, location A and location B (Figure 1). Location A referred to the midtorso region, positioned 50 cm below the face cradle. Location B represented the edge of the examination table, specifically the centerfold. Data collection involved obtaining initial swab samples in the morning after disinfection by Environmental Services, and terminal swab samples were collected at the end of the day after the Family Medicine staff had completed disinfection of the examination room.

Figure 1:

An outpatient examination table, identifying locations A and B.

The current cleaning practices encompassed the use of disinfectant chemicals for high-touch areas such as examination room tables, doorknobs, light switches, stools/chairs, and scales. However, specific instructions on how each high-touch area should be disinfected were not provided. Personnel responsible for room cleaning received training on the established examination room cleaning protocol, following the current techniques in place.

AccuPoint advanced HC – ATP bioluminescence

The adenosine triphosphate (ATP) bioluminescence assays were performed utilizing AccuPoint Advanced HC Reader (Neogen Corporation, Lansing, MI) [11]. The reader provides a digital readout of the quantity of light measured in relative light units (RLUs), which is directly correlated to the amount of ATP present in the sample [11]. The resulting bioluminescent reaction is as follows [12]:

The level of ATP present serves as an indicator of organic material or contamination in the sample. A higher ATP level therefore denotes a greater degree of biological contamination. As per manufacturer’s recommendation [11], to prepare for sample collection, the sampler cartridges were removed from the refrigerator and warmed to room temperature 1 h prior to use. We utilized Dacron swabs, for their superior material compatibility with the AccuPoint Advanced HC system. The collection media within the cartridge is a proprietary blend developed by Neogen Corporation, designed specifically to facilitate ATP extraction. When collecting samples, the sampler was removed from the cartridge by the handle while exercising caution not to touch the tip of the sampler or let the tip touch any other surface prior to testing. Data were collected at location A and B (Figure 1) by swabbing in a zigzag pattern, creating a grid formation (Figure 2).

Figure 2:

AccuPoint Advanced HC system with sampler (A) and zigzag swabbing technique for swabbing location A and location B (B).

After collecting the sample, the sampler was reinserted into its cartridge and fully depressed. The sampler along with the cartridge were swirled in a clockwise manner for 2 s and placed into the sampler compartment of the AccuPoint Advanced HC Reader for analysis [11]. The threshold value being utilized for interpretation is an ATP level of 500 RLU/100 cm² (Table 1), a standard benchmark threshold [11, 13, 14], After analysis, the sampler and cartridge were removed from the sample compartment and disposed of according to manufacturer recommendations.

Medical validity

In our previous study, we performed an internal validity trial to ensure that 500 RLUs was a proper indication of disinfection [4]. To ensure the internal validity of the current study, a parallel baseline trial was conducted to assess the reliability of ATP bioluminescence measurements. Predata collection was performed on 18 family medicine examination tables after a full day of patient office visits but before the family medicine staff completed their disinfection routine after the last patient. The same data collection protocol outlined in the study design section was followed during this set of data collection. The tables were then disinfected utilizing standard examination table cleaning products (CaviCide). The disinfection process involved spraying each section of the examination table with three sprays of the cleaner and wiping the entire surface with a paper towel. Special attention was given to ensuring thorough disinfection of the edges and torso. To prevent cross-contamination, a clean paper towel was utilized for each section of the table to avoid spreading particles across the entire surface. After the disinfection process, the examination tables were left to dry for approximately 10 min before the postdata collection. This drying period reduced the risk of any potential interference from the cleaner and allowed sufficient time for the cleaner to effectively eliminate any pathogens. By implementing these measures, the potential bias from the cleaning process was minimized, thereby improving the reliability and validity of the data collected.

Statistical analysis

ATP values of less than 500 RLUs were considered passing, whereas AP values of 500 or more were considered a failure. All statistical tests were performed utilizing Microsoft Excel 2021 by HP and RP. A Shapiro–Wilk test was conducted to test for normality. A nonparametric Wilcoxon signed-rank test was then utilized to compare the RLU values of the initial and terminal swab samples. Significance was set at p<0.05. Additionally, the effect size was calculated utilizing a Cohen’s d test to determine the magnitude of differences in RLU values of the initial and terminal swab samples. A large effect size was classified as d>0.80.

Location A: midtorso

In the initial swab sample, 17 of the 18 samples showed RLU values <500 (94.4 % pass rate) (Table 2). Additionally, 1 of the 18 samples showed RLUs≥500 (5.6 % failure rate) (Table 2). In the terminal swab sample, 15 of the 18 samples showed RLU values <500 (83.3 % pass rate). Additionally, 3 of the 18 samples showed RLUs≥500 (16.7 % failure rate) (Table 2). The terminal swab samples showed an 11.1 % increase, from 1 to 3 failures when compared to the initial swab samples. The Shapiro–Wilk test showed a nonparametric distribution. A Wilcoxon signed-rank test revealed that for the midtorso, the estimated pathogen levels measured in the terminal swab samples (median, 193 RLU; range, 1–690 RLUs; n=18) was not statistically significantly different (p=0.9124) than the levels measured in the initial swab samples (median, 134 RLUs; range, 4–946 RLUs; n=18), z=0.11, p=0.9124 with a small effect size, d=0.04 (Table 3).

Location B: edge

In the initial swab sample, none of the 18 samples showed RLU values <500 (0 % pass rate) (Table 2). All of the 18 samples showed RLUs≥500 (100 % failure rate) (Table 2). In the terminal swab sample, the results remained the same: none of the 18 samples showed RLU values <500 (0 % pass rate), and all samples showed RLUs≥500 (100 % failure rate) (Table 2). The Shapiro–Wilk test showed a nonparametric distribution. A Wilcoxon signed-rank test revealed that for the edge, the estimated pathogen levels measured in the terminal swab samples (median, 2095 RLUs; range, 891–5,540; n=18) was not statistically significantly different (p=0.61) than the levels measured in initial swab samples (median, 2,257 RLUs; range, 932–5,825 RLUs; n=18), z = −0.57, p=0.61 with a small effect size, d=0.12) (Table 3).

Measurement validity trial

A total of 72 surfaces were sampled, 36 before disinfection samples and 36 after disinfection samples split evenly between location A (n=18) and location B (n=18).

Location A: midtorso

The Shapiro–Wilk test showed a nonparametric distribution. A Wilcoxon signed-rank test revealed that for the midtorso, the estimated pathogen levels measured in the before-disinfection samples (median, 2,250 RLUs; range, 551–5,900 RLUs; n=18) were statistically significantly higher (p<0.00001) than the levels measured in the after-disinfection samples (median, 175.5 RLUs; range, 3–391 RLUs; n=18), z=5.40, p<0.00001) (Table 4).

Location B: edge

The Shapiro–Wilk test showed a nonparametric distribution. A Wilcoxon signed-rank test revealed that for the edge, the estimated pathogen levels measured in the before-disinfection samples (median, 2,225 RLUs; range, 701–7,201 RLUs; n=18) was statistically significantly higher (p<0.00001) than the levels measured in the after-disinfection samples (median, 221 RLUs; range, 5–530 RLUs; n=18), z=5.40, p<0.00001 (Table 4).

Limitations

Our study possesses several limitations. Our study was primarily conducted within a single medical school, potentially constraining the extrapolation of our results. To reinforce our findings, future studies should consider a larger sample size drawn from multiple institutions.

Another limitation is that although ATP measurements denote the presence of organic matter, they fail to distinguish between pathogenic and nonpathogenic microorganisms. This attribute may lead to an overestimation of contamination risk, because high ATP levels do not always equate to the presence of harmful pathogens. Various factors, such as examination table characteristics and the existence of chemical residues, could also skew the ATP readings, potentially influencing our conclusions. This is a limitation that is also expressed in the study by Patrizio et al. [4]; however, the internal validity trial in the previous study and the current study help to mitigate the likelihood of such contamination affecting our results, although we cannot fully eliminate this as an extraneous variable.

Additionally, our study did not delve into the distinction between nonpathogenic and pathogenic organisms, which is a critical aspect in understanding contamination risks. However, as discussed in Patrizio et al., [4] the presence of even nonpathogenic organisms on a surface could signal potential contamination with the pathogenic organisms, contributing to disease transmission. Furthermore, nonpathogenic organisms can mutate into opportunistic pathogens when an individual’s immune system is compromised due to underlying medical conditions or external stressors [20]. In such scenarios, the line between nonpathogenic and pathogenic organisms blurs, with both types posing potential infection risks. This emphasizes the importance of maintaining rigorous hygiene and sanitation practices to mitigate infection risks from all microorganisms.

In light of our research findings, our data assertively emphasize the importance of revising the existing sanitization measures in outpatient family medicine practices. This refinement needs to incorporate a comprehensive disinfection of all high-contact areas, such as the edges of examination tables, which should be performed by both the Environmental Services and family medicine staff. Despite the limitations faced, our study represents a crucial progression toward refining disinfection protocols and, consequently, enhancing patient safety within medical facilities. In our continued effort to enhance these practices, we are now working on developing a more stringent disinfection protocol. This protocol aims to address the concerns and findings identified in our study. Following the approval of this enhanced protocol, we plan to conduct a subsequent study in the same family medicine clinic to assess the efficacy and practical implementation of our suggested modifications.