Impact of COVID-19 pandemic in the care of people living with HIV and cancer at an oncologic center in Mexico

Abstract

Background

People living with HIV(PLWH) and cancer are among the most vulnerable patients and require constant access to medical services. We compared the characteristics of PLWH and cancer in Mexico, before and during the COVID-19 pandemic.

Methods

Patients admitted 1 year before (pre-pandemic) and 1 year after the start of the pandemic (pandemic) were included. Clinical characteristics, HIV-related variables, and 90-day mortality were compared. Data are described a proportions (N,%) and central tendency measures. A multiple regression model for variables associated with 90-day mortality was performed.

Results

Seventy-nine patients were seen in the pre-pandemic period; 92 during the pandemic. Main diagnoses were Kaposi Sarcoma and lymphoma. CD4⁺ cell count at diagnosis was lower during the pandemic: 81 cells/mm³ vs. 128 cells/mm³, p = .035. CD4+<100 cells/mm³ at first consultation increased from 41% to 58% during the pandemic (p = .041). Only BMI <20 kg/m² was associated to death (aOR 8.27, 95%CI 1.74-39.25) (p = .008). The pandemic period was not associated with a higher 90-day mortality.

Conclusions

PLWH and cancer presented to care with advanced disease overall. This was more pronounced during the pandemic period. Mortality was associated with AIDS-related variables regardless of study period. This underscores the need for strategies to maintain in-person access to health-care services for PLWH.

Keywords

Covid-19 Pandemic HIV Continuum of Care AIDS-related Kaposi Sarcoma AIDS-associated lymphoma

Introduction

The COVID-19 pandemic has placed a great burden on medical care and forced to restructure the healthcare system in many countries, focusing on COVID-19. Twelve days after the World Health Organization (WHO) declared the coronavirus COVID-19 a pandemic,¹ the Mexican government established measures to try to reduce the virus’ spread in different sectors including healthcare. Telemedicine was meant to substitute in-person consultations. There was also a reengineering of the National Institutes of Health in Mexico City to prepare for the inflow of COVID-19 patients requiring hospital care.² Three major hospitals in southern Mexico City were reconverted to receiving only COVID-19 patients. These hospitals are normally referral centers for the complex care of people living with advanced HIV-disease. As a result, most people living with HIV (PLWH) from these hospitals received medical follow-up through phone calls and were referred to other hospitals in case of urgent care. The National Institute of Cancer in Mexico City (INCan), also located in southern Mexico City, was not converted to exclusive COVID care. One ward was assigned to COVID-19 care for any patient from the hospital who was infected with SARS-Cov2 and needed hospitalization, and there was an outpatient consultation for patients with respiratory symptoms, to diagnose/rule out COVID-19, managed by the infectious disease’s practitioners. All other areas kept receiving patients with non-COVID pathologies such as PLWH and cancer. Telemedicine was available for stable patients in the outpatient care.

PLWH and cancer are among the most vulnerable population of persons requiring medical attention. They usually present to care with severe immunosuppression; most of them have a recent HIV diagnosis, have not yet been linked to medical care and are naïve to antiretroviral therapy (ART) at the time of cancer diagnosis.^3,4

This study aimed to describe the characteristics of patients who were received at the Cancer and AIDS Clinic of INCan, before and after March 2020, and to compare their clinical conditions upon first consultation, 30 and 90-day mortality.

Material and Methods

Setting

INCan is a 133-bed cancer referral and teaching hospital located in Mexico City, which serves adult cancer patients who do not have social security and come mainly from the central region of the country. It provides in-patient and out-patient care. The Cancer and AIDS Clinic provides care for PLWH and cancer since 1990; it includes 550 out-patients in active follow-up; on average 90 new patients are seen annually. The Clinic is run by the Infectious Diseases Department in collaboration with the different oncology services and receives patients with a proven or probable diagnosis of cancer for workup and treatment. As part of the initial workup, all patients arriving to INCan with CD4<100 cells/mm³ are routinely screened for hepatitis and syphilis, cytomegalovirus (CMV) retinitis, cryptococcal Ag, histoplasma urinary Ag, histoplasma serology, myeloculture and Computed Tomography, plus any targeted study according to their clinical evaluation. Patients are seen in the outpatient care on a regular basis, at least every 4 months. During the COVID-19 pandemic, stable patients received follow up through phone calls from the Infectious Department practitioners. The rest of the patients were seen in face-to-face consultations.

Procedures

All PLWH referred to INCan from March 2019 to March 2021 with an established diagnosis of malignancy were included. The diagnosis of malignancy was confirmed by pathology. The pre-pandemic period was defined as patients who first visited the hospital between 23 March 2019, and 22 March 2020; the pandemic period included patients from 23 March 2020, to 22 March 2021. During the pandemic period, COVID tests were performed for symptomatic patients or for asymptomatic patients programmed for a surgery.

Clinical and demographic data were obtained from the electronic clinical charts. Information on time of HIV diagnosis, date of first visit, baseline HIV viral load (VL) and CD4⁺ T cell count, CD4⁺ T cell count at first consultation, number and type of non-Malignancy AIDS defining events (NM-ADE) and ART status at first consultation were recorded, as well as 30 and 90‐day mortality and lost to follow up (LTFU). For comorbidities, diabetes and hypertension were registered if such a diagnosis was reported in the electronic files and/or patients were taking hypoglycemic drugs/ insulin or antihypertensive drugs reported in the file. Obesity was considered with a BMI >30 kg/m². Also, a variable was created for the proportion of patients with a least one comorbidity, including diabetes, hypertension, obesity. All patients were followed for at least 90 days to report the outcomes. LTFU was defined as any patient without a consultation (either face to face or by phone call), 3months before the end of the 90‐days follow up period. This study was approved by the Institutional Review Board (IRB) of INCan (Ref/INCAN/CI/1014/2021). Due to its retrospective nature, informed consent was waived by the IRB.

Statistical Analysis

Data is described as proportions for qualitative variables and median (IQR) for quantitative variables with non-parametric distribution. A bivariate analysis was done to compare clinical conditions upon admission of patients from both periods. We included in the analysis sociodemographic data such as age and sex, comorbidities, type of neoplasia, and variables associated to HIV such as baseline CD4 count, CD4 count at admission, viral load at first consultation, and being on ART, and any history of AIDS-defining event (ADE). For the statistical analysis, X² and Mann-Whitney U tests were used for qualitative and quantitative variables, respectively. Fisher exact test was used for variables with cell counts lower than five. We report overall mortality as all patients who had died at the moment of the study’s data collection, irrespective of the follow-up length. This can include patients with longer than 90-days follow-up. A bivariate analysis was done to describe variables associated to death, including variables associated to HIV (CD4 count, ART status, viral load, ADE), type of neoplasia, comorbidities, sex and age and the study period. To adjust for other variables and considering that all patients were followed up at least 90 days, a logistic regression model was built including all variables associated to death in the bivariate analysis with a p value <0.1, and age as potential confounders. We excluded patients LTFU. We used a stepwise regression modelling strategy, choosing an arbitrary p value of 0.1 from the variables of the univariate analysis.⁵ To assess for collinearity, we measured the variance inflation factor (VIF) in the regression model.⁶ We used statistical package STATA 14.0.

Results

Overall, 184 patients were received at INCan with a presumptive diagnosis of cancer from March 2019 to March 2021. Seven patients came to the first consultation and did not return. One of them was traced and found receiving care in another institution. Three patients were seriously ill and did not accept hospitalization. These three patients were from the pandemic period. They could not be traced afterwards. The other three patients were stable at their first consultation but also lost to follow up. These patients were not included in the analysis. Of the 177 remaining patients, six patients did not have any oncologic diagnosis (one patient had leiomyoma, one laryngeal papilloma, one lymphoid hyperplasia, one proctitis, two patients had wasting syndrome and AIDS). An oncologic diagnosis was confirmed in 171: 79 during the pre-pandemic year, compared to 92 patients during the pandemic Figure 1.

The median age at first consultation was 33-years-old (IQR 28-43). Most patients were men (93.6%); There was one transgender woman (0.5%). Twenty-eight (16.4%) had at least one comorbidity, and four patients had two. The most frequent comorbidities were diabetes mellitus (28.6%), obesity (21.4%), psychiatric disease (25%) and systemic hypertension (14.3%). Regarding co-infections, 9% had positive antibodies for hepatitis C (three patients had confirmed detectable viral load), 21% had a positive treponemal test, and 13% had a positive hepatitis B surface antigen. There were no differences regarding age, sex, coinfections, and comorbidities between periods. Table 1 describes the patients’ characteristics at first consultation by period.

Table 1.

Clinical characteristics of patients at first consultation, by period (n = 171).

	Pre-pandemic period n = 79 (%)	Pandemic period n = 92 (%)	p Value
Age in years, median, (IQR)	33 (29-43)	33 (27-43)	0.38
Sex
Men	76 (96.2)	84 (91.3)	0.19
Cisgender women	2 (2.5)	8 (8.7)
Transgender women	1 (1.27)
Body mass index at first consultation, median (IQR)	22 (19.9-23.8)	21.8 (19.5-24.2)	0.78
Any comorbidity	12 (15.2)	16 (17.4)	0.22
Co-infections^a
Positive hepatitis C antibodies (n = 157)	0	6 (13.6)	0.09
Confirmed chronic hepatitis C (n = 157)	0	3 (6.8)	0.1
Positive hepatitis B surface antigen (n = 156)	3 (11.5)	1 (2.4)	0.27
Positive treponemal test (n = 126)	3 (27.3)	7 (36.8)	052
Time from HIV diagnosis to first consultation, in days, median (IQR)	78 (14-1136)	86 (27-1536)	0.24
CD4 cell count at diagnosis, cells/mm ³ , median (IQR)	128 (65–206)	81 (35–195)	0.035
CD4 cell count at first consultation in cells/mm ³ , median (IQR) ^b	126 (53–238)	86 (35–152)	0.046
Number on ART at first consultation	44 (55.7)	50 (54.4)	0.86
Detectable viral load at first consultation (VL >200 copies/mL), n = 94^c,d	24 (72.2)	22 (65.2)	0.33
Non-malignancies AIDS defining events at admission^e,f
None	54 (68.3)	69 (75)	0.56
One	20 (25.3)	18 (19.6)
Two or more	5 (6.3)	5 (5.4)
AIDS-defining malignancy
Kaposi sarcoma^b	58 (72.5)	57 (58.8)	0.03
Non-hodgkin lymphoma	12 (15)	21 (21.7)	0.21
Invasive cervical cancer^a	2 (2.5)	2 (2.1)	0.87
Non-AIDS defining neoplasms^a
Hodgkin lymphoma	4 (5.1)	8 (8.7)	0.38
Other hematologic cancer	3 (3.8)	1 (1.1)
Gynecologic cancer	1 (1.3)	0
Germ cell cancer	0	2 (2.2)
Anal cancer	0	2 (2.2)
Hepatocarcinoma	0	1 (1.1)
Breast cancer	0	1 (1.1)
Others	3 (3.8)	2 (2.2)

^afisher exact test.

^bp value = 0.05.

^cDefined as less than 200 copies/mL.

^donly patients on ART are included.

^eexcluding AIDS related neoplasms.

All AIDS defining events diagnosed at admission up to 1 month later.

Median time between HIV diagnosis and first consultation at INCan was 85.5 days (IQR 24-1298). The median CD4 count at diagnosis was 103 cells/mm³(46-206). It was significantly lower in the pandemic period (81 cells/mm³, IQR 35-195) compared to the pre-pandemic period (128 cells/mm³, IQR 65-206) p = .035. During the pandemic period, a higher proportion of patients were diagnosed with <100 cells/mm³ (57.6 vs 41.8%, p = .039).

Almost half of the patients were naïve to ART at their first contact with the hospital (45%), with no differences between periods. Of all patients already on ART at first consultation, 51% had an undetectable VL (defined as less than 200 copies/mL). When stratifying by time on ART, 55.7% of patients had less than six months on ART, and 74.4% of patients with detectable viral load had less than 6 months on ART. There were no differences between periods in baseline VL or VL at first consultation. Overall, the median VL at first consultation for PLWH not on ARV was 125,452 copies/mL (5.09 log₁₀ copies/mL). The median CD4⁺ T cell count at first consultation was 99 cells/mm³(39-195) in the whole cohort. This was significantly lower during the pandemic period: 86 cells/mm³(IQR 35-152) vs 126 cells/mm³(IQR 53-238) in the pre-pandemic period, p = .046. Figure 2 shows the median CD4⁺ T cell count at diagnosis and first consultation, by period.

We performed a sensitivity analysis removing patients seen at the earliest stage of the pandemic to see if the difference in CD4⁺ count persisted. We removed patients coming to the hospital from march to May 2020. There was no difference in baseline CD4 count, but for the CD4 count at first consultation, there was a slightly wider difference, with a lower p value (81 vs. 126 cells/mm3, p = .029).

There were 141 (82.5%) patients with an AIDS defining neoplasm. Overall, the main diagnoses were Kaposi Sarcoma (KS) (63.7%) and Non-Hodgkin Lymphoma (NHL) (19.3%). Five patients had synchronic KS and NHL. Four patients had invasive cervical cancer (2.3%). Twenty-eight patients (16.4%) had a non-AIDS defining neoplasm; the most frequent was Hodgkin Lymphoma (7%). There were no statistically significant differences in cancer distribution between periods. Table 1 describes the oncologic diagnosis by period.

Forty-eight patients (28.1%) had at least one additional NM-ADE. Of those, ten patients (20.8%) had two or more concomitant NM-ADE. The more frequent NM-ADE were: histoplasmosis (n = 20, 41.7%), of which 15 cases were classified as disseminated, through urinary antigen (n = 3), myeloculture or blood culture (n = 2), Hc100 PCR (n = 5) or positive Grocott staining in myelogram or biopsies (n = 5). In the other 5 cases, it was considered as possible due to clinical syndrome and positive serology. Other opportunistic infections included cytomegalovirus end-organ disease (colitis and retinitis) (n = 8, 16.7%), tuberculosis (n = 7, 14.6%), atypical mycobacteria (n = 7, 14.6%), and cryptococcosis (n = 4, 8.3%). Regarding other markers of disease severity, the median BMI in the whole cohort was 22 kg/m²(IQR 19.6-24).

There were 17 cases of COVID-19 during the pandemic period; 16 were mild, one was severe and required hospitalization. There was no death attributed to COVID-19.

The median follow-up time was 457 days (IQR 132-596) in the pre-pandemic period vs. 192.5 days (IQR 91-259) in the pandemic period. As mentioned previously, all patients were followed up at least 90 days. There were 22 (12.4%) patients lost to follow up, with no differences in CD4 count, study period, HIV viral suppression, comorbidities or opportunistic infections.

At the moment of the study’s data collection, 34 patients had died, with an overall mortality of 19.9%; this includes some patients who died after 90-days of follow up. One patient died on the day he came for first consultation. Nine patients died at home: 6 of them were in palliative care, one was discharged against medical advice, and 2 were seriously ill, in treatment of disseminated Kaposi Sarcoma. For two patients who died at home, we do not know the exact date of death, so we used the last day seen at the hospital. Although not statistically significant, median time from first consultation to death during the pre-pandemic period was twice the time during the pandemic period: 88 days (IQR 24-126) vs 41.5 days (IQR 19-160), p = .8. The main causes of death were attributed to ADE: 47% (n = 16) due to disseminated KS, 34% (n = 12) NHL (secondary to cancer progression or chemotherapy-related infectious complications). Other causes were non-specified sepsis/pneumonia (n = 4, 12%), invasive anal cancer (n = 1, 3%) and cryptococcal meningitis (n = 1, 3%). Forty-four percent of the patients who died had also a NM-ADE. For patients who died at home, we attributed the cause of death to their diagnosis of discharge. No patient died from COVID-19. The 30-day and 90-day mortality were 6.7% (N = 12) and 9.9% (N = 17) respectively, with no significant differences between periods.

For the bivariate analysis of mortality, all 34 patients who died were included. There was an association between low BMI at first consultation, and death. Other variables associated with death were not being on ART, having a detectable VL, lesser time with HIV diagnosis, having at least one NM-ADE, and having NHL diagnosis. For the multivariate analysis of 90-day mortality, we excluded the 22 patients lost to follow-up. After adjusting for other variables, only low BMI remains with good evidence of an association to death at 90 days in the multivariate analysis. Patients with a BMI lower than 20 kg/m² had eight times the odds of dying compared to patients with a higher BMI (aOR 8.27, 95%CI 1.74-39.25) (p = .008) (Table 2 and 3). We assessed collinearity of the regression model. The mean VIF was 1.32, with no variables with a VIF >2, so collinearity was moderate with no need of adjustment of the model.

Table 2.

Variables associated with overall mortality.

	Dead n = 34 (%)	Alive n = 137 (%)	p value
Men	32 (94)	128 (93.4)	0.82
Age in years, median (IQR)	37 (28-46)	32 (28-41)	0.25
Pandemic period	18 (52.9)	74 (54)	0.9
Median BMI kg/m², (IQR)	20.8 (18.5-23)	22.3 (19.8-24)	0.01
Median time from HIV diagnosis to first consultation in days, (IQR)	28 (14-243)	104 (27-1385)	0.03
Detectable viral load at presentation^a	28 (84.9)	87 (64.4)	0.02
Median CD4⁺ T cell count at first consultation, in cells/mm³, (IQR)	110.5 (38-149)	96 (46-208)	0.48
At least one NM-AIDS defining events^b,c	15 (44.1)	24.8	0.02
Non-hodgkin lymphoma diagnosis	11 (32.4)	22 (16.1)	0.03
Kaposi sarcoma diagnosis	21 (61.8)	88 (64.2)	0.78
On ART at presentation	12 (35.3)	82 (59.9)	0.01

^aDefined as more than 200 copies/mL.

^bExcluding AIDS related neoplasms.

^cAll AIDS defining events diagnosed at admission up to 1 month later.

Table 3.

Multivariate regression model for mortality at 90 days.*

	Dead (n = 17)	Alive (N = 132)	aOR (95%CI)	p value
Age in years, median (IQR)	34 (28–43)	34 (28–43)	0.99 (0.93–1.08)	0.94
Median time from HIV diagnosis to first consultation in days, (IQR)	25 (17–115)	90 (19–1385)	1.00 (0.99–1.00)	0.54
Detectable viral load at presentation^a	16 (94.1)	83 (64.3)	4.54 (0.31–65.70)	0.27
BMI <20 kg/m²	10 (19.2)	7 (5.6)	8.26 (1.74–39.24)	0.008
At least one NM-AIDS defining events^b,c	7 (41.2)	38 (28.8)	0.44 (0.09–2.11)	0.301
Non-hodgkin lymphoma diagnosis	3 (17.7)	25 (18.9)	3.68 (0.54–24.99)	0.182
On ART at presentation	6 (35.3)	73 (55.3)	1.29 (0.29–5.85)	0.74

^aDefined as more than 200 copies/mL.

^bExcluding AIDS related neoplasms.

^cAll AIDS defining events diagnosed at admission up to 1 month later.

*The model was adjusted for Non-Hodgkin lymphoma, age, BMI, viral load at consultation, NM-AIDS Defining events, ART at presentation, time from HIV diagnosis, and excludes lost to follow up patients.

Discussion

We describe a cohort of PLWH and cancer presenting with advanced HIV disease before and during the pandemic. The proportion of men was 94%; this corresponds to the normal male to female ratio in INCan and Mexico City. Mexico is a concentrated epidemic with a national proportion of male of 80%, which increases in the capital.⁷ Many of them had simultaneous AIDS defining events, a malignancy, and opportunistic infections. Their CD4⁺ T cell count at diagnosis and at first consultation was significantly lower during the pandemic period compared to the pre-pandemic. CD4 count at diagnosis was almost 50 cells/mm³ lower during the pandemic period. Also, although not statistically significant, patients who died during the pandemic period, did so in less time (42 days vs. 88 days). The impact on mortality was not shown in our study, probably because PLWH were already in advanced stage before the pandemic. This has been reported in other studies⁸: Bell et al report their experience after the first European Wave of COVID-19, from a tertiary care center of HIV/oncology in London. They found HIV inpatient admissions during the pandemic to be more frequently due to ADE, and with lower CD4 count compared to HIV admissions in 2019. They also suggest this is due both to less access to HIV testing, disruption of HIV services, and less initiative to be tested due to confinement alongside with fear of COVID-19 contagion and describe it as the indirect impact of COVID-19 on HIV.

Worldwide, the lack of access to medical services in hospitals converted into COVID-19 centers has seriously affected the care of patients with non-COVID health problems.^9,10 For PLWH, one recognized consequence has been less HIV testing and an increase in HIV late diagnosis.^11,12 First contact with health care services for patients not yet diagnosed with HIV was also delayed. The Ambulatory clinic for HIV Care in Mexico City “Clínica Condesa” reported a 34% decrease in the number of patients presenting for an HIV test during 2020 compared to 2019, and 3% increase in the proportion of positive HIV tests from 2019. There was also a 10% increase in the proportion of patients presenting with late disease, defined as a CD4⁺ T cell count less than 200 cells/mm³.¹³ In Mexico, many hospitals previously treating HIV patients had to close their services to reconvert for COVID-19 patients²; this led to a decrease in access to medical care for HIV patients.¹³ Unfortunately, the workup and care for patients with advanced HIV cannot rely on telemedicine or delayed access to hospitals diagnostic facilities.

Besides the limited access to HIV services due to lockdown measures, other consequences have been described during the pandemic. Many infectious diseases practitioners have been overwhelmed by treating COVID-19, which has led to a decrease in medical staff available to treat non-COVID patients. A survey addressed to providers of HIV care across more than 50 countries showed that more than 50% of respondents were treating COVID-19 and reported a disruption of HIV services such as diagnostic and clinical management.¹⁰ In addition, the government instructions to stay at home and the fear of contracting COVID-19 (another virus) may have delayed even more PLWH seeking for care. Living with the stigma of HIV combined with the fear of a worse outcome if contracting COVID-19, because of the coinfection, has been reported as a reason for patients to delay their seeking care.^14,15

Interestingly, our results contrast with another report, where Mazzitelli et al described a cohort of PLWH diagnosed before and during the Pandemic in Italy and compared their clinical conditions.¹⁶ Contrary to our findings, they found PLWH diagnosed during the pandemic were older, had higher CD4 count, fewer ADE and less advanced disease, compared to PLWH diagnosed prior to the pandemic. Almost half of these patients also had a SARS COV2 test performed. As a conclusion, the authors propose a strategy of a combined HIV/SARS-COV2 test in patients showing up with suspected COVID-19, to increase HIV diagnosis yield and linkage to healthcare. It is important to consider that there are many differences between both studies. The sampling was different and the circumstances of the HIV test applied in both reports is also different. In Mazzitelli’s report, HIV testing was used more as a screening strategy, which was not the case in our study where we included patients arriving to the hospital with symptoms associated with HIV immunosuppression. Although a combined HIV/COVID test seems like an interesting approach, this strategy might target a more reachable population, and still leaves out the more stigmatized and marginalized PLWH with existing social and health inequities that are being exacerbated by COVID-19, that usually come to healthcare services with more advanced disease.^17,18

In our sample, although patients were diagnosed with more advanced disease during the pandemic, the study period was not associated with mortality. This is probably because PLWH diagnosed prior to the pandemic had already many variables related to a bad prognosis. Mexico, as Latin America, still has an important proportion of HIV patients coming late to medical care, which significantly contributes to HIV-related mortality.^19–21 Variables associated with death were mainly related to a late diagnosis and AIDS stage, such as having opportunistic infections, AIDS defining neoplasia, and a BMI lower than 20 kg/m². BMI has been reported previously to be associated to mortality in PLWH.^22,23 Those studies included patients from Sub-Saharan Africa; the association with mortality was found with BMI <18.5 kg/m². In our study the strongest association was with a cutoff of 20 kg/m². Although this value is considered normal, patients with BMI <20 had eight times the odds of dying. The differences in BMI could reflect an even more advanced stage at presentation in sub-Saharan cohorts. This underscores the need for proper and timely care to PLWH and adequate national strategies to target the more vulnerable populations during the pandemic.

Part of the strategies to slow down the spread of the COVID-19 pandemic worldwide included telemedicine and virtual on-line medical visits.²⁴ Although data have reported good acceptability in PLWH at the beginning of the pandemic, there are certain challenges such as access to technology and a private space, that can limit the uptake.^25,26 Furthermore, the lack of physical examination can be a limitation in sicker patients.²⁷ This strategy can possibly work for stable and suppressed patients but is not suitable for patients recently diagnosed with late disease and AIDS-defining events. Certain strategies have worked adequately by prioritizing HIV care and keeping in-person visits for not suppressed PLWH.^24,28 Diagnosis and clinical evaluation of recently diagnosed PLWH need face-to-face consultations. The experience with face-to-face consultations during the Pandemic at our Cancer and AIDS Clinic shows that it can be done, with good infection control preventive education. To support this, none of the patients coming for initial care during the pandemic period died of COVID-19.

Our study has some limitations. It is a retrospective study. As such, we cannot infer causality with our findings. The associations we found are statistically significant, but with p values close to 0.05. There is a possibility lower than 5% that these results are due to chance. A bigger sample could have helped to get stronger associations. On the other hand, although the study only includes one center, INCan is one of the main referral centers in the country for PLWH, which makes it more representative of PLWH and cancer in Mexico.

Conclusions

The COVID-19 pandemic has affected the access to care to other non-COVID-19 pathologies. It has disrupted the HIV continuum of care, leading to a further increase in late HIV diagnosis, in settings where this is already a major issue. This is the first report describing the indirect impact of the pandemic in PLWH and HIV services in Mexico. Strategies are urgently needed nation-wide to implement prioritized care and keep HIV services open and accessible to HIV detection and case management of patients with AIDS.

Footnotes

Acknowledgments

To all the health care workers that remain tirelessly in the front line working in COVID-19 times.

Author’s contributions

PV elaborated the design, contributed to the analysis, wrote, and reviewed the manuscript. AL and MJM collected the data and reviewed the manuscript. BI, PAF, PCJ and CPJ reviewed the manuscript. AMO contributed to the study design, analyzed the database, wrote, and reviewed the manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethics approval and consent to participate

This study was approved by the Institutional Review Board (IRB) of INCan (Ref/INCAN/CI/1014/2021). Due to its retrospective nature, informed consent was waived by the IRB.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

ORCID iDs

Patricia Volkow

Patricia Cornejo-Juarez

Beda Islas-Munoz

Alexandra Martin-Onraet

Appendix

List of abbreviations

References

World Health Organisation

. . WHO Director-General’s opening remarks at the media briefing on COVID-19 [Internet]. 2020. Disponible en: https://www.who.int/director-general/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19---11-march-2020

La Jornada . Supera México el reto de reconvertir hospitales Covid [Internet], 2020. Disponible en: https://www.jornada.com.mx/ultimas/politica/2020/06/25/supera-mexico-el-reto-de-reconvertir-hospitales-covid-7923.html

Cardenas-Ortega

Ramirez-Ibarguen

Rivera-Buendia

, et al. Clinical characteristics and outcomes of HIV positive patients with lymphoma in an oncological reference center in Mexico City. Med Baltim 2020; 99(40): e22408.

Fink

Shepherd

Cesa

, et al. Cancer in HIV-infected persons from the Caribbean. Cent S Am J Acquir Immune Defic Syndr 2011; 56(5): 467–473.

Greenland

. Modeling and variable selection in epidemiologic analysis. Am J Public Health 1989; 79(3): 340–349.

The University of Utah . Checking for Multicollinearity [Internet]. Checking for Multicollinearity. [citado 23 de agosto de 2022]. Disponible en: https://campusguides.lib.utah.edu/c.php?g=160853&p=1054159

CENSIDA . SISTEMA DE VIGILANCIA EPIDEMIOLÓGICA DE VIH [Internet]. 2020 [citado 23 de agosto de 2022]. Disponible en: https://www.gob.mx/cms/uploads/attachment/file/622468/VIH-Sida_4toTrim_2020.pdf

Bell

Bracchi

Dalla Pria

, et al. Indirect HIV morbidity and mortality due to COVID-19. Clin Infect Dis 2021; 73(10): 1947–1949.

Jiang

Zhou

Tang

. Maintaining HIV care during the COVID-19 pandemic. Lancet HIV 2020; 7(5): e308–e309.

10.

Zuniga

. Continuity of HIV care in the presence of COVID-19. AIDS Rev 2020; 22(3): 131–132.

11.

Hill

Anderson

Lock

. COVID-19 pandemic, pre-exposure prophylaxis (PrEP) care, and HIV/STI testing among patients receiving care in three HIV epidemic priority states. AIDS Behav 2021; 25(5): 1361–1365.

12.

Jenness

Le Guillou

Chandra

, et al. Projected HIV and bacterial sexually transmitted infection incidence following COVID-19-related sexual distancing and clinical service interruption. J Infect Dis 2021; 223(6): 1019–1028.

13.

Gobierno de la Ciudad de Mexico . Respuesta Epidemiológica al VIH Sida y al Virus de la Hepatitis C en la Ciudad de México. En: https://www.google.com/search?client=firefox-b-e&q=Respuesta+Epidemiol%C3%B3gica+al+VIH+Sida+y+al%0BVirus+de+la+Hepatitis+C+%0Ben+la+Ciudad+de+M%C3%A9xico%0B2020%2C+Clinica+Especializada+Condesa: Programa de VIH Sida de la Ciudad de México, 2020.

14.

Ballester-Arnal

Gil-Llario

. The virus that changed spain: impact of COVID-19 on people with HIV. AIDS Behav 2020; 24(8): 2253–2257.

15.

Prabhu

Poongulali

Kumarasamy

. Impact of COVID-19 on people living with HIV: A review. J Virus Erad 2020; 6(4): 100019.

16.

European AIDS Clinical Society . Abstracts of the 18th European AIDS Conference (EACS 2021), London, UK, October 27–30, 2021, 2021. HIV Med. 2021;3:4-309.

17.

Chenneville

Gabbidon

Hanson

, et al. The impact of COVID-19 on HIV treatment and research: a call to action. Int J Env Res Public Health 2020; 17(12): 4548.

18.

Armbruster

Fields

Campbell

, et al. Addressing health inequities exacerbated by COVID-19 among youth with HIV: expanding our toolkit. J Adolesc Health 2020; 67(2): 290–295.

19.

Crabtree-Ramirez

Caro-Vega

Belaunzaran-Zamudio

, et al. High prevalence of late diagnosis of HIV in Mexico during the HAART era. Salud Publica Mex 2012; 54(5): 506–514.

20.

Belaunzaran-Zamudio

Caro-Vega

Shepherd

, et al. The population impact of late presentation with advanced HIV disease and delayed antiretroviral therapy in adults receiving HIV care in Latin America. Am J Epidemiol 2020; 189(6): 564–572.

21.

Crabtree-Ramirez

Belaunzaran-Zamudio

Cortes

, et al. The HIV epidemic in Latin America: a time to reflect on the history of success and the challenges ahead. J Int AIDS Soc 2020; 23(3): e25468.

22.

Fuseini

Gyan

Kyei

, et al. Undernutrition and HIV infection in sub-Saharan Africa: health outcomes and therapeutic interventions. Curr HIVAIDS Rep 2021; 18(2): 87–97.

23.

Wekesa

McLigeyo

Owuor

, et al. Factors associated with 36-month loss to follow-up and mortality outcomes among HIV-infected adults on antiretroviral therapy in Central Kenya. BMC Public Health 2020; 20(1): 328.

24.

Ridgway

Schmitt

Friedman

, et al. HIV care continuum and COVID-19 outcomes among people living with HIV during the COVID-19 pandemic. AIDS Behav 2020; 24(10): 2770–2772.

25.

Smith

Badowski

. Telemedicine for HIV Care: current status and future prospects. HIVAIDS Auckl NZ 2021; 13: 651–656.

26.

Wood

Lan

Tao

, et al. Visit trends and factors associated with telemedicine uptake among persons with HIV during the COVID-19 pandemic. Open Forum Infect Dis 1 de Noviembre de 2021; 8(11): ofab480.

27.

Harsono

. Experiences with telemedicine for HIV care during the COVID-19 pandemic: a mixed-methods study. AIDS Behav 2022; 13: 2099–2111.

28.

Guaraldi

Milic

Martinez

, et al. HIV care models during the COVID-19 era. Clin Infect Dis 2020; 73(5): e1222–e1227.