The COVID-19 pandemic placed essential workers at the center of occupational health concerns because many jobs could not be shifted to remote work and continued to require face-to-face contact, mobility, and operational continuity. Across essential sectors, the pandemic was associated with heightened infection risk, organizational disruption, moral strain, and unequal access to protective measures, sick leave, and workplace safeguards.Reference Gaitens, Condon and Fernandes 1 – 4 Within this broader workforce, police officers and military personnel occupied a particularly demanding position. During public health emergencies, these groups may be required to maintain public order, support emergency response logistics, enforce movement restrictions, and continue field-based operations while working in dense institutional settings and in close contact with the public. 4 –Reference Remington, Witkowski and Ganapati 7
For police and first-responder populations, the pandemic was not only a biological exposure hazard; it also altered the organization and delivery of routine duties. International studies have described increased workload, irregular schedules, changing operational protocols, perceived insufficient preparedness, concerns about personal protective equipment, fear of infecting relatives, and pressure to remain on duty despite uncertainty.Reference Jennings and Perez 5 –Reference Remington, Witkowski and Ganapati 7 Research in these populations has also documented substantial psychological and occupational strain, including stress, burnout, and intentions to leave the workforce.Reference Stogner, Miller and McLean 8 –Reference Roberts, Wong and Jenkins 14 However, much of this literature has focused on mental health outcomes or mixed first-responder samples, whereas fewer studies have examined how specific occupational arrangements—including deployment, mobility, training, protective provision, and perceived preparedness—were affected within security institutions.
This gap is especially relevant in Peru. The country experienced a severe pandemic burden and major strain on its health and social systems, while containment measures required sustained participation by police and armed forces in frontline response activities.Reference Gianella, Gideon and Romero 15 , Reference Caycho-Rodríguez, Barboza-Palomino and Carbajal-León 16 Available Peruvian studies in these populations have mainly examined mental health and related constructs, including psychological symptoms, burnout, insomnia, depression, anxiety, and fear related to pandemic information.Reference Caycho-Rodríguez, Barboza-Palomino and Carbajal-León 16 –Reference Mejia, Coronel-Chucos and Meza-Roque 21 Although these studies provide important evidence, they do not fully describe how work arrangements themselves were modified during the pandemic, including workplace assignment, transportation, occupational role, training, perceived preparedness, mistreatment, and the need for self-procurement of protective supplies.
A more complete characterization of these occupational conditions is important for police workforce planning and institutional preparedness for future public health emergencies. Understanding how pandemic response affected day-to-day work organization, and whether these patterns differed between police officers and Army personnel, may help identify institutional vulnerabilities that are not captured by mental health indicators alone. Therefore, the aim of this study was to characterize occupational conditions before and during the COVID-19 pandemic among Peruvian police officers, using Army personnel as an institutional comparator, and to explore differences between both institutions.
Design and Methods
Study Design and Data Source
This was a secondary cross-sectional analysis of data from a broader survey project conducted among Peruvian police officers and Army personnel during the COVID-19 pandemic. The original survey examined working conditions, perceptions, knowledge, and media use in this occupational population. For the present analysis, the database was used to describe occupational conditions before and during the pandemic and to explore institutional differences in selected work-related characteristics during the pandemic.Reference Mejia, Coronel-Chucos and Meza-Roque 21
Study Setting and Period
Data were collected in Peru from the second quarter of 2020 to the first quarter of 2021, a period that included the first and second waves of the COVID-19 pandemic. Surveys were administered at duty-related institutional locations, including barracks, meeting points, and other operational areas where police officers and Army personnel were on duty at the time of data collection. Peru declared a national health emergency on 11 March 2020 (Supreme Decree 008-2020-SA) and, 4 days later, a National State of Emergency with compulsory home confinement (Supreme Decree 044-2020-PCM, 15 March 2020), which assigned the Peruvian National Police and the Armed Forces responsibility for enforcing the measures. Enforcement duties began within days of the declaration, and data collection began in the quarter immediately following, so the interval available for institution-wide preparation before deployment was short. The questionnaire recorded the modality of COVID-19 training received but not its starting date, duration, or intensity, and the time devoted to preparation cannot be quantified from these data.
Participants and Sampling
Eligible participants were active-duty police officers and Army personnel aged 18 years or older who were working during the pandemic and provided consent to participate. A convenience sampling strategy was used. Questionnaires from respondents belonging to other institutions, such as the Navy or Air Force, and questionnaires containing inconsistent responses were excluded.
The final analytical sample comprised 512 participants, including 352 police officers and 160 Army personnel. Because the sample was non-probabilistic, the findings should be interpreted as reflecting the characteristics of the surveyed participants rather than as nationally representative estimates for police officers or Army personnel in Peru.
Data Collection Instrument and Study Variables
Data were collected using a structured paper-based questionnaire developed ad hoc for the parent study after a review of the relevant literature.Reference Jennings and Perez 5 , Reference Caycho-Rodríguez, Barboza-Palomino and Carbajal-León 16 The original instrument was designed to collect information on working conditions, perceptions, knowledge, and media use among police officers and Army personnel during the COVID-19 pandemic. No formal validation study of the questionnaire was conducted.
For this secondary analysis, variables were selected a priori to characterize occupational conditions before and during the pandemic and to perform exploratory institutional comparisons during the pandemic. All variables were obtained from participant self-report through the original questionnaire and, when required for analysis, were coded by the research team using predefined categories.
For the descriptive comparison of the periods before and during the pandemic, we examined workplace location, work area, type of occupation, weekly working hours, mode of transportation to work, commuting time, meal-break duration, and food provision at work.
In addition, self-procurement of supplies during the pandemic was assessed descriptively. Participants reported whether they had purchased selected items for their work during the pandemic, including alcohol, masks, food, and other items. Reported reasons for purchase were categorized as avoiding contagion, health reasons, protection, or other reasons. All self-procurement variables were analyzed descriptively.
For the institutional comparison during the pandemic, participants were grouped by institution as police officers or Army personnel. The following work-related variables were included:
Health status related to COVID-19. This was a categorical variable derived from participant self-report and coded into three mutually exclusive categories: no risk, health-service personnel, and at-risk group. The no risk category included participants who did not report belonging to a clinically vulnerable group for COVID-19 and were not linked to health-related duties. The health-service personnel category included respondents linked to institutional health or sanitary services. The at-risk group category included respondents who reported greater clinical vulnerability to severe COVID-19 outcomes.
Type of occupation. This variable was self-reported and categorized as administrative, field-based, or mixed.
COVID-19-related training. Participants reported the main training modality received during the pandemic, categorized as face-to-face training, virtual training, training through pamphlets, other means, no training received, or multiple training modalities. Because several categories had sparse frequencies, this variable was recategorized for inferential comparison into face-to-face training versus other or no face-to-face training.
Self-perceived preparedness for COVID-19-related issues. This variable was recorded as yes or no, according to whether the participant felt qualified to address COVID-19-related issues in the occupational setting.
Mistreatment during the pandemic. Participants reported whether they had experienced mistreatment during the pandemic. For descriptive purposes, responses were classified as physical, psychological, work-related, sexual, multiple types, or none. Because some categories had very small frequencies, this variable was recategorized for inferential comparison into any mistreatment versus no mistreatment.
Reported perpetrator of mistreatment. Among participants who reported mistreatment and completed the follow-up question, the reported perpetrator was classified as civilian, colleague, other individuals, or none of the above. Owing to the small number of valid responses, this variable was analyzed descriptively only.
Because the present study was based on a secondary analysis of an existing database, variable selection and coding were constrained by the structure of the original questionnaire.
Data Collection Procedures
Surveys were administered in person using paper forms. Before participation, respondents were informed about the study objectives, the anonymous nature of the survey, and their rights as research participants. Participation was voluntary, and respondents were not required to answer all questionnaire items. Confidentiality was maintained throughout data collection and database management. After data collection, responses were entered into a Microsoft Excel 2019 database, reviewed for completeness and consistency, and cleaned using the predefined eligibility criteria. The final database was then exported to Stata version 11.1 for analysis.
Statistical Analysis
Categorical variables were summarized using absolute frequencies and percentages. Quantitative variables were assessed for normality using the Shapiro-Wilk test and were summarized as medians and interquartile ranges (IQRs) because they did not follow a normal distribution.
Occupational characteristics referring to the periods before and during the pandemic were analyzed descriptively only. Because both periods were assessed retrospectively within a single cross-sectional survey, the observed differences were interpreted as self-reported descriptive differences rather than as prospectively observed longitudinal changes.
COVID-19-related training, self-perceived preparedness, mistreatment, self-procurement of supplies, and reported reasons for purchase during the pandemic were also summarized descriptively using valid responses for each item.
For the institutional component, work-related characteristics during the pandemic were summarized according to institution type. These between-institution comparisons were considered exploratory and unadjusted, given the secondary cross-sectional design and the limited availability of baseline covariates for confounding control. The objective was therefore to identify differences in the distribution of selected characteristics rather than to estimate independent or causal institutional effects.
Comparisons between institutions used the chi-square test, with continuity correction for dichotomous variables; after recoding, no expected cell count fell below 8, so no exact test was required. To improve interpretability and statistical stability, selected variables with sparse categories were recategorized before inferential analysis. The institutional comparisons included health status related to COVID-19, type of occupation, face-to-face training versus other/no face-to-face training, self-perceived preparedness, and any mistreatment versus none. The detailed distribution of mistreatment subtypes and reported perpetrators was presented descriptively.
Because not all survey items were mandatory, denominators varied across variables. Percentages were calculated using the number of valid responses for each item, and variable-specific denominators are reported in the corresponding tables and results. A two-sided P value <0.05 was considered statistically significant.
Ethical Considerations
The study was conducted in accordance with ethical principles for research involving human participants. Participation was voluntary and anonymous, and respondents were not required to answer all survey items. Before completing the questionnaire, each participant received information about the study aims and their rights as a participant. The broader research project was approved by the Research Ethics Committee of Universidad Privada Antenor Orrego under Resolution No. 0211-2020-UPAO.
Results
A total of 512 participants were included in the analysis, comprising 352 police officers (68.8%) and 160 Army personnel (31.2%). Occupational characteristics reported for the periods before and during the pandemic are presented in Table 1. Across both periods, Junín accounted for the largest proportion of respondents, representing 49.9% before the pandemic and 51.1% during the pandemic. The proportion assigned to public-order duties was higher during the pandemic than before the pandemic (26.9% vs. 18.3%), whereas the proportion assigned to other work areas was lower (27.7% vs. 37.5%). Mixed occupations were the most frequently reported occupational profile in both periods, although field-based occupations were reported more often during the pandemic (40.3% vs. 36.0%).
Occupational characteristics before and during the COVID-19 pandemic in the study sample
Table 1. Long description
The table contains three columns: Variable, Before the pandemic, and During the pandemic. Data is presented as frequency followed by percentage in parentheses, or as median with interquartile range.
Workplace:
- Junín: 254 (49.9%) before, 257 (51.1%) during.
- Ayacucho: 150 (29.4%) before, 149 (29.6%) during.
- Lima: 67 (13.2%) before, 69 (13.7%) during.
- Other locations: 38 (7.5%) before, 28 (5.6%) during.
Work area:
- Barracks: 140 (27.9%) before, 144 (29.1%) during.
- Investigation: 38 (7.5%) before, 32 (6.4%) during.
- Public order: 92 (18.3%) before, 133 (26.9%) during.
- Patrol: 44 (8.8%) before, 49 (9.9%) during.
- Other areas: 188 (37.5%) before, 137 (27.7%) during.
Type of occupation:
- Administrative: 63 (12.3%) before, 53 (10.6%) during.
- Field-based: 184 (36.0%) before, 202 (40.3%) during.
- Mixed: 264 (51.7%) before, 246 (49.1%) during.
Hours worked per week: 48 (42–60) before, 48 (48–70) during.
Mode of transportation to work:
- On foot: 193 (37.9%) before, 311 (62.1%) during.
- Private transportation: 162 (31.8%) before, 113 (22.5%) during.
- Institution-provided transportation: 11 (2.2%) before, 12 (2.4%) during.
- Public transportation: 141 (27.7%) before, 62 (12.4%) during.
- Remote work: Not applicable before, 1 (0.2%) during.
- Other: 2 (0.4%) before, 2 (0.4%) during.
Commuting time, minutes: 10 (5–30) before, 20 (5–25) during.
Meal-break duration, minutes: 24 (15–60) before, 24 (15–60) during.
Food provision at work:
- No meals provided: 22 (4.3%) before, 23 (4.6%) during.
- Received one or more meals: 476 (93.5%) before, 471 (93.4%) during.
- Snack only: 11 (2.2%) before, 10 (2.0%) during.
* Quantitative values are shown as median (interquartile range). Percentages were calculated using the valid responses for each block. Valid responses before and during the pandemic, respectively: workplace 509 and 503; work area 502 and 495; type of occupation 511 and 501; mode of transportation to work 509 and 501; food provision at work 509 and 504.
Commuting on foot was reported by 37.9% of participants before the pandemic and by 62.1% during the pandemic, whereas public transportation use decreased from 27.7% to 12.4%. Median weekly working hours were 48 in both periods, although the interquartile range shifted from 42-60 hours before the pandemic to 48-70 hours during the pandemic. Median commuting time increased from 10 minutes (IQR, 5-30) to 20 minutes (IQR, 5-25). Meal-break duration remained unchanged at a median of 24 minutes (IQR, 15-60), and food provision showed little variation between periods.
COVID-19-related training, self-perceived preparedness, and mistreatment during the pandemic are summarized in Table 2. Valid responses were available for 494 participants for COVID-19-related training, 505 for self-perceived preparedness, and 384 for mistreatment. Face-to-face training was the most frequently reported modality (360/494; 72.9%), followed by training through pamphlets (59/494; 11.9%) and virtual training (43/494; 8.7%). Despite this, most respondents reported that they did not feel qualified to address COVID-19-related issues (445/505; 88.1%), whereas only 60/505 (11.9%) reported feeling qualified. Regarding mistreatment, 325/384 participants (84.6%) reported no mistreatment, whereas 59/384 (15.4%) reported at least one type. Psychological mistreatment was the most frequently reported subtype (20/384; 5.2%), while physical, work-related, multiple, and sexual forms were less frequent. Among participants with available follow-up information on the perpetrator (n = 58), the most frequently identified perpetrators were other individuals (27/58; 46.6%) and civilians (24/58; 41.4%).
COVID-19-related training, self-perceived preparedness, and mistreatment during the pandemic
Table 2. Long description
The table is organized into three columns: Category, Frequency, and Percentage.
1. C O V I D 19-related training (n = 494):
- Face-to-face training: 360 (72.9%)
- Virtual training: 43 (8.7%)
- Training through pamphlets: 59 (11.9%)
- Other means: 20 (4.1%)
- No training received: 8 (1.6%)
- Multiple training modalities: 4 (0.8%)
2. Self-perceived preparedness for C O V I D 19-related issues (n = 505):
- No: 445 (88.1%)
- Yes: 60 (11.9%)
3. Type of mistreatment experienced during the pandemic (n = 384):
- Physical: 14 (3.7%)
- Psychological: 20 (5.2%)
- Work-related: 12 (3.1%)
- Sexual: 1 (0.3%)
- Multiple types: 12 (3.1%)
- None: 325 (84.6%)
4. Reported perpetrator of mistreatment (n = 58):
- Civilian: 24 (41.4%)
- Colleague: 3 (5.2%)
- None of the above: 4 (6.9%)
- Other individuals: 27 (46.6%)
Note: Percentages were calculated using valid responses for each subsection. Reported perpetrator of mistreatment was summarized only among participants with available follow-up information.
Information on self-procurement of supplies during the pandemic is presented in Table 3. Masks were the most frequently purchased item (265/438; 60.5%), followed by other items (160/434; 36.9%) and alcohol (143/438; 32.6%). Food purchases were uncommon (13/435; 3.0%). Protection was reported as a reason for purchase by 142/412 participants (34.5%), while 146/415 (35.2%) selected other reasons.
Self-procurement of supplies and reported reasons for purchase during the pandemic
Table 3. Long description
The image contains two tables, labeled A and B.
Table A. Purchased items. This table has three columns: Item, No, n (%), and Yes, n (%).
- Alcohol (n = 438): No 295 (67.4%); Yes 143 (32.6%).
- Masks (n = 438): No 173 (39.5%); Yes 265 (60.5%).
- Food (n = 435): No 422 (97.0%); Yes 13 (3.0%).
- Other items (n = 434): No 274 (63.1%); Yes 160 (36.9%).
Table B. Reported reasons for purchase. This table has three columns: Reason, No, n (%), and Yes, n (%).
- To avoid contagion (n = 415): No 331 (79.8%); Yes 84 (20.2%).
- For health reasons (n = 415): No 378 (91.1%); Yes 37 (8.9%).
- For protection (n = 412): No 270 (65.5%); Yes 142 (34.5%).
- For other reasons (n = 415): No 269 (64.8%); Yes 146 (35.2%).
Footnote: Percentages were calculated using the number of valid responses for each item.
Note: Percentages were calculated using the number of valid responses for each item.
Exploratory unadjusted comparisons between police officers and Army personnel during the pandemic are shown in Table 4. Denominators varied across variables because of item-level non-response and are reported in the table. Statistically significant institutional differences were observed in health status related to COVID-19 (P < 0.001), type of occupation (P < 0.001), and self-perceived preparedness (P = 0.026). Police officers were more frequently classified in the at-risk group, more frequently reported field-based occupations, and more frequently reported feeling prepared than Army personnel. Army personnel more frequently reported mixed occupations. No statistically significant differences were found for face-to-face training versus other or no face-to-face training (P = 0.136) or for any mistreatment versus none (P = 0.241). These findings indicate heterogeneity in selected occupational characteristics but should not be interpreted as estimates of independent institutional effects.
Exploratory unadjusted institutional comparison of selected work-related characteristics during the pandemic
Table 4. Long description
A table with five columns: Variable, Category, Police n percent, Army n percent, and P value.
1. Health status related to C O V I D 19 (n equals 460), P value less than 0.001:
- No risk: Police 217 (72.1 percent), Army 146 (91.8 percent).
- Health-service personnel: Police 18 (6.0 percent), Army 6 (3.8 percent).
- At-risk group: Police 66 (21.9 percent), Army 7 (4.4 percent).
2. Type of occupation (n equals 511), P value less than 0.001:
- Administrative: Police 38 (10.8 percent), Army 25 (15.6 percent).
- Field-based: Police 169 (48.1 percent), Army 15 (9.4 percent).
- Mixed: Police 144 (41.0 percent), Army 120 (75.0 percent).
3. C O V I D 19-related training (n equals 494), P value 0.136:
- Face-to-face training: Police 236 (70.7 percent), Army 124 (77.5 percent).
- Other or no face-to-face training: Police 98 (29.3 percent), Army 36 (22.5 percent).
4. Self-perceived preparedness for C O V I D 19-related issues (n equals 505), P value 0.026:
- No: Police 296 (85.8 percent), Army 149 (93.1 percent).
- Yes: Police 49 (14.2 percent), Army 11 (6.9 percent).
5. Any mistreatment during the pandemic (n equals 384), P value 0.241:
- No: Police 185 (82.6 percent), Army 140 (87.5 percent).
- Yes: Police 39 (17.4 percent), Army 20 (12.5 percent).
Note: Percentages were calculated within each institution using valid responses for each variable. Institutional comparisons were exploratory and unadjusted. Comparisons of dichotomous variables used the chi-square test with continuity correction; comparisons of 3-category variables used Pearson’s chi-square test. Variables with sparse categories were recoded before inferential analysis, after which no expected cell count fell below 8, so no exact test was required.
Discussion
This study identified notable differences in how work was organized before and during the COVID-19 pandemic. During the pandemic, a greater proportion of respondents reported assignments related to public order and field-based work, together with more frequent commuting on foot, less use of public transportation, and longer travel times to work. Rather than indicating an interruption of occupational activity, these findings are more consistent with a reorganization of day-to-day operational routines under emergency conditions. This interpretation is aligned with previous studies showing that police institutions rapidly modified deployment patterns, operational priorities, and routine activities during COVID-19 in response to public health mandates and shifting institutional demands.Reference Jennings and Perez 5 , Reference Frenkel, Giessing and Egger-Lampl 22 –Reference White and Fradella 24
An important finding was the marked contrast between the frequency of reported training and the low frequency of perceived preparedness. Although face-to-face training was the most frequently reported modality, most participants indicated that they did not feel qualified to address COVID-19-related issues. This pattern suggests that operational readiness in security institutions depends on more than the delivery of training sessions. Preparedness during a rapidly evolving public health emergency may also depend on the clarity of protocols, consistency of risk communication, availability of supplies, work intensity, rest opportunities, and perceived institutional support. This interpretation is consistent with studies in first-responder and police populations showing that occupational preparedness during the pandemic depended not only on formal instruction, but also on the perceived adequacy of protocols, access to protection, and broader organizational support.Reference McAlearney, Gaughan and Macewan 6 , Reference García-Iglesias, Chirico and Rizzo 13 , Reference De Camargo 25 Research in Spanish security forces has also documented burnout and death-related anxiety during COVID-19, reinforcing the idea that preparedness in these occupational settings is multidimensional and cannot be inferred solely from whether training was delivered.Reference Gómez-Galán, Lázaro-Pérez and Martínez-López 26 , Reference Lázaro-Pérez, Martínez-López and Gómez-Galán 27 The gap between training received and preparedness perceived also suggests where future training might be directed. Revised standard operating procedures, supply logistics, and organizational support are necessary, but they do not by themselves rehearse the tasks personnel were asked to perform. Approaches that do include computer-based simulation and tabletop exercises, mannequin practice in putting on and removing protective equipment, scenario training in virtual environments, and brief refreshers repeated during deployment instead of a single session beforehand. Our design cannot show whether any of these would raise perceived or actual readiness in security institutions; that question requires prospective evaluation, which we suggest as a priority.
The frequency of self-procurement of supplies, particularly masks and alcohol, is also noteworthy. Although the present study cannot determine whether these purchases reflected inadequate institutional provision, delayed access, personal precaution, or a combination of these factors, the findings show that workers personally contributed to obtaining some of the supplies used for protection. Similar concerns have been documented in police and first-responder settings, where worker experience during the pandemic was influenced not only by formal guidance but also by the timely availability of personal protective equipment and the broader organizational environment in which risk had to be managed.Reference McAlearney, Gaughan and Macewan 6 , Reference Maskály, Ivković and Neyroud 23 For police and security institutions, this finding supports incorporating supply availability and distribution into preparedness planning alongside training and risk communication. 4
Mistreatment during the pandemic was reported by a minority of participants, but the proportion was not negligible. In the recategorized institutional comparison, the frequency of any reported mistreatment did not differ significantly between police officers and Army personnel, although the analysis was not designed to establish equivalence between institutions. Among the limited number of participants with follow-up information, civilians and other individuals were the most frequently reported perpetrator categories. This pattern may reflect the public-facing enforcement context in which police and military personnel operated during mobility restrictions and other emergency measures. Previous studies have described tensions with the public, psychological strain linked to lockdown duties, and strained enforcement environments during the pandemic, supporting the relevance of mistreatment as one component of the occupational burden experienced by security personnel.Reference Grover, Sahoo and Dua 9 , Reference Shodunke 28
The exploratory institutional comparison showed differences in health status related to COVID-19, type of occupation, and self-perceived preparedness. These findings should not be interpreted as evidence of independent institutional effects, since the analysis was unadjusted and based on a secondary cross-sectional database with variable non-response. Nevertheless, they are compatible with the possibility that police officers and Army personnel worked within different operational profiles, patterns of public contact, and occupational demands during the pandemic. This interpretation is consistent with prior evidence from Peru showing substantial mental health impact in police and armed forces personnel, as well as with studies in Peruvian military populations reporting burnout, insomnia, depression, and anxiety during later pandemic phases.Reference Caycho-Rodríguez, Barboza-Palomino and Carbajal-León 16 , Reference Valladares-Garrido, Zapata-Castro and Picón-Reategui 18 –Reference Valladares-Garrido, Picón-Reátegui and Zila-Velasque 20 Taken together, these findings suggest that the occupational consequences of the pandemic were shaped not only by exposure to the emergency itself, but also by institutional role and work context. For preparedness planning, police and military organizations may therefore require role-specific rather than uniform approaches to training, deployment, and workforce support.
Conversely, no statistically significant differences were identified in recategorized training patterns or in any mistreatment between institutions. This indicates that the present analysis did not detect clear institutional differences in these domains despite differences in role profile and perceived preparedness. The finding is consistent with broader evidence from security-force populations showing that occupational stress, burnout, and emotional burden during COVID-19 were often shaped by shared exposure to uncertainty, infection risk, high operational demand, and psychologically challenging work environments, rather than by institutional identity alone.Reference García-Iglesias, Chirico and Rizzo 13 , Reference Gómez-Galán, Lázaro-Pérez and Martínez-López 26 , Reference Lázaro-Pérez, Martínez-López and Gómez-Galán 27 However, the absence of statistical significance should not be interpreted as evidence that the institutions were equivalent.
This study has several strengths. It examines an occupational population that played an essential role during the pandemic but has been less frequently characterized than health care workers. It also extends the focus beyond mental health outcomes by examining work organization, mobility, perceived preparedness, mistreatment, and self-procurement of supplies.
Limitations
This study has several limitations. First, the non-probability sampling strategy limits the generalizability of the findings and prevents the estimation of nationally representative patterns among Peruvian police officers and Army personnel. Second, all variables were obtained through self-report, which may have introduced recall bias and reporting bias, particularly for conditions assessed retrospectively before and during the COVID-19 pandemic. Third, because the study used a cross-sectional survey with retrospective assessment of pre-pandemic conditions, observed differences should be interpreted as self-reported changes rather than objectively measured longitudinal changes. Fourth, the questionnaire was developed for the parent study based on a literature review but did not undergo formal validation, which may have resulted in measurement error or misclassification of some occupational preparedness domains. Fifth, item-level non-response resulted in variable denominators across analyses, and this non-response was concentrated in the police group, which contributed between 224 and 351 of 352 valid responses depending on the variable, whereas Army personnel contributed 159 or 160 of 160; the comparison of reported mistreatment, for which police response was lowest, should therefore be read with particular caution. Sixth, data collection coincided with the first and second waves of the pandemic in Peru, when uncertainty, perceived threat and operational demand were at their highest, so responses to the item on perceived preparedness may reflect the conditions prevailing when the survey was administered; a single cross-sectional measurement cannot establish whether the same personnel would have answered differently had they been surveyed 1 or 2 years later. Finally, the secondary nature of the dataset limited the evaluation of potentially relevant determinants of preparedness, including institutional communication, supervisory support, adequacy of protective supply systems, and organizational resilience. The exploratory institutional comparisons were unadjusted and should therefore be interpreted as hypothesis-generating rather than causal estimates.
Conclusions
Non-health care first responders played a critical role during the COVID-19 pandemic, but reported preparedness was not fully aligned with training exposure. Future public health emergency preparedness strategies should move beyond training delivery alone and incorporate operationally tailored education, effective risk communication, reliable access to protective supplies, and institutional support. These elements may strengthen the readiness and resilience of security forces responding to future public health emergencies.
Data availability statement
The dataset analyzed during the current study is not publicly available because of institutional and ethical restrictions related to participant confidentiality. De-identified data may be made available by the corresponding author upon reasonable request and subject to approval by the relevant institutional and ethics authorities.
Author contributions
Christian R. Mejia: Conceptualization, methodology, investigation, data collection, and manuscript preparation.
Jhosselyn I. Chacon: Conceptualization, methodology, investigation, and manuscript review.
Adán José Doria Velarde: Investigation, data collection, and manuscript review.
Leonel Vega-Perez: Data analysis, interpretation of results, and manuscript review.
María J. Erazo-Muñoz: Literature review, manuscript preparation, and manuscript review.
Camilo Vega-Useche: Methodology, interpretation of results, manuscript preparation, critical revision, and supervision.
All authors approved the final version of the manuscript.
Funding statement
The authors received no financial support for the research, authorship, and/or publication of this article.
Competing interests
None.
Ethics approval and consent to participate
The broader research project was approved by the Research Ethics Committee of Universidad Privada Antenor Orrego under Resolution No. 0211-2020-UPAO.
Participation was voluntary and anonymous. Before completing the questionnaire, participants were informed about the study objectives and their rights as research participants.
Use of artificial intelligence tools
During manuscript preparation, the authors used artificial intelligence tools solely to adapt the manuscript to the journal’s editorial requirements and to improve the English translation. These tools were not used to generate or interpret research data, to perform statistical analyses, or to produce scientific content. All text was critically reviewed and approved by the authors, who take full responsibility for the accuracy, originality, and integrity of the manuscript.
Open access
