Longitudinal health survey of women from Venezuela in Colombia (ELSA-VENCOL): first report

Background: Colombia is currently the world main recipient country for Venezuelan migrants, and women represent a high proportion of them. This article presents the first report of a cohort of Venezuelan migrant women entering Colombia through Cucuta and its metropolitan area (the main land entry point to this country). The study aimed to describe the health status and access to healthcare services among Venezuelan migrant women in Colombia with irregular migration status, and to analyze changes in those conditions at a one-month follow-up. Methods: A longitudinal cohort study of Venezuelan migrant women, 18 to 45 years, who entered Colombia with an irregular migration status, was carried out in 2021. Study participants were recruited in temporary shelters, transit points, and migrant settlements in Cucuta and the metropolitan area. At baseline, we administered a structured questionnaire including sociodemographic characteristics, migration history, health history, access to health services, sexual and reproductive health, practice of early detection of cervical cancer and breast cancer, food insecurity, and depressive symptoms. The women were again contacted by phone one month later, between March and July 2021, and a second questionnaire was applied. Results: A total of 2,298 women were included in the baseline measurement and 56.4% could be contacted again at the one-month follow- up. A significant increase was found in the percentage of women who had a health problem during the past month (from 23.1% to 31.4%; p<0.01). Meanwhile, the percentage of women with depressive symptoms decreased from 80.5% to 71.2% (p<0.01). Conclusion: This report is a starting point for the longitudinal follow-up of the cohort, which will allow us to better understand how the health status of Venezuelan women changes during the migration flow in Colombia


59
Background 60 Since 2015, the social, political, and economic conditions in Venezuela have led to one of the largest 61 international human mobility processes in the history of the region, a process that intensified in 2017 62 when the migration flow increased [1]. By November 2021, estimates showed roughly 6.04 million

63
Venezuelan refugees and migrants worldwide [2]. This process, characterized as a south-south prenatal checkups (PNC), and an analysis by migration status found that only 50.6% of migrants with 108 irregular migration status received PNC compared to 79.5% for migrants with regular status [13].

109
The Colombian government has made considerable efforts to expand the supply and coverage of 110 health services to facilitate access to health services for Venezuelan migrants, as well as to keep 111 records of the medical care that they receive [12]. Consequently, some of the existing information 112 systems have been modified and new ones have been created for recording the medical care received 113 by this population, as well as for vital statistics and epidemiological surveillance.

114
While these information systems have been important for monitoring the health needs of Venezuelan 115 migrants and their use of health services, as well as for guiding public health and public policy 116 decisions, they have some limitations. There is the fact that the health information that is obtained 117 largely depends on access to health services, which is low among migrants. The coverage is also 118 limited and there are lags and problems with the quality of the health information on migrants [14].

119
In addition, the scope of these data is limited, as they do not enable the use of a multidimensional 120 approach to migrant health, and they are insufficient for evaluating and monitoring migrant health at 121 the population level. Moreover, a system is needed that enables identifying the health situation and 122 needs, as well as access to health services, especially for Venezuelan migrant women since they not 123 only represent roughly half of the migrant population but are of reproductive age and have historically 124 been the group that has used certain health services the most. Although the Department of National

125
Statistics (DANE in Spanish) and several organizations have gone to great lengths to conduct 126 population-based surveys, these tend to be limited to only a few dimensions. They are also cross-127 sectional and have mainly focused on migrants who have already settled in the country and intend to 128 stay.

129
The generation of primary source information that includes individuals regardless of their having 130 contact with health services can provide a better understanding of the health situation of Venezuelan 131 migrant women. Thus, the objective of this research was to describe the health status and access to . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

142
This was a longitudinal cohort study of Venezuelan migrant women with an irregular migration status 143 who entered Colombia through Cúcuta and its metropolitan area, with a cross-sectional baseline 144 measurement and a follow-up one month after. New follow-ups are expected to be done in the future.

145
Selection criteria, recruitment, and follow-up 146 Cúcuta and its metropolitan area is the main land entry point to Colombia. This is a region of high 147 pendular flow (more than 35,000 people crossed per day before the pandemic), the trans-border 148 movement of people to obtain goods and services in one side of the border, coming back to their 149 country of residence after just a few hours or days. It is also the point where "walkers" (caminantes, . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 26, 2022. ; This survey included women who: 1) were between 18 and 45 years of age; 2) had entered Colombia

156
through Cúcuta (border city with Venezuela) or its metropolitan area (Los Patios, Puerto Santander 157 or Villa del Rosario) at any time during the baseline measurement period; 3) did not have Colombian 158 nationality; 4) expressed the intention to remain in Colombia for at least one year; 5) were not 159 "circular" migrants (see definition below); and 6) did not have their passport stamped when entering

160
Colombia and did not have a valid Special Permit to Stay (PEP in Spanish), which meant that their 161 migration status was irregular.

162
For this study, "circular" was defined as having entered Colombia two or more times during the month 163 prior to the first interview for reasons other than tourism, vacations, family visits or business trips.

164
That is, the study excluded persons who in the previous month had entered Colombia frequently for 165 reasons of work, study or in search of services, among other purposes, since this population is 166 considered to be part of the cross-border living dynamics rather than a migrant population.

167
The baseline interview was conducted face-to-face between February 15 and May 25, 2021, at various 168 locations in the cities of Cúcuta, Los Patios and Villa del Rosario, and lasted an average of 30 minutes.

169
In this interview, the telephone numbers of all participants were obtained, and they were informed 170 that they would be contacted again one month later. For the follow-up interview, the women were 171 contacted by telephone between March 16 and July 6, 2021. The second contact was attempted 172 beginning one month after the initial interview, at different times over a period of two weeks and on 173 different days of the week. When there was no response after this period the participant was 174 considered to be lost to follow-up. The average telephone call lasted 25 minutes.

176
Given the lack of a clearly defined sampling frame, a non-probability sampling method with snowball 177 expansion was used. Interviewers went to temporary shelters and settlements where Venezuelan 178 migrants were living and invited all migrants who were there during the fieldwork period to . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted August 26, 2022. ; https://doi.org/10.1101/2022.08.26.22279267 doi: medRxiv preprint participate. Participants were also recruited on a road with a high flow of migrants, and all Venezuelan

180
women walking past that point were invited to participate.

Data collection instruments and study variables 182
A standardized questionnaire with 161 questions was administered to assess the following dimensions 183 of health status and access to health services: health history and perceived morbidity, effective access 184 to health services, sexual and reproductive health, early detection of cervical and breast cancer, food 185 insecurity and depressive symptoms.

186
The participants were asked about self-perceived health, health problems or conditions over the past . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) did not want to evaluate the correlation between the first and second measurements but rather the 219 percentage change between the two measurements. For this reason, they were also analyzed using the 220 marginal homogeneity test. In the case of quantitative variables, they were first evaluated using 221 graphical and numerical methods (Shapiro-Wilk test) to identify whether they followed a normal 222 distribution, and if they did not then the Wilcoxon test was used to compare the data from the two 223 measurements.

224
To assess potential bias from loss to follow-up, the baseline characteristics of the participants with 225 and without follow-up were compared using the chi-square test for categorical variables (and Fisher's . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) any social benefits that they may receive and that sharing the information would not pose any risk.

237
Personal information that could lead to individual identification of the participants was only known 238 and handled by the research team and was omitted from the final databases.

241
A total of 2,298 Venezuelan migrant women were surveyed at baseline and 1,297 were surveyed at 242 one-month follow-up (56.4% response rate).

243
Statistically significant differences were found in the baseline measurements between women with 244 and without follow-up, specifically, those with follow-up had a higher educational level, had been in 245 Colombia longer, and had more contact with Colombian health services. They also had a slightly 246 smaller percentage of significant depressive symptoms (Table S1).
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted August 26, 2022. Colombian pesos per month, which is equivalent to approximately $259 USD) and only 0.9% 255 received economic assistance or subsidies (Table 1).

256
A decrease between baseline and follow-up was found in the percentage of migrants sleeping on the 257 street, while an increase was found in the percentage of those who contributed money to the household 258 and in participants whose households received economic assistance (

269
At follow-up, considerably more women reported having had a health problem in the past month,

270
with the figure rising from 23.1% to 31.4%. This increase was more notable among those who were 271 interviewed one month or less after arriving in Colombia, 28% of whom reported a health problem in . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted August 26, 2022. ; https://doi.org/10.1101/2022.08.26.22279267 doi: medRxiv preprint the past month, compared to those who were interviewed after being in the country for over one year,

273
24% of whom reported having a health problem (Table 4). Similarly, there was a marked increase in 274 the percentage of participants who reported moderate, severe, or extreme difficulty working or 275 performing daily chores (from 5. 5% to 11.0%) and who rated their health as fair (from 13.0% to 276 31.2%). Conversely, the percentage of women with depressive symptoms decreased from 80.5% to 277 71.2%. Table 2 shows all the results related to the health status and access to services of the 278 participants.

279
The differences found between baseline and one-month follow-up remained the same when 280 categorized by length of time in Colombia: <1 month, 1 month to 1 year and >1 year (Table S2).

282
The present investigation sought to analyze the health status and access to health services of 283 Venezuelan migrant women with irregular migration status at baseline and one-month follow-up.

284
ELSA-VENCOL) is one of the first longitudinal study among Venezuelan migrants carried out in

286
The baseline information showed that roughly one-fifth of the participants had had a health problem 287 during the previous month. However, the prevalence of a prior diagnosis of chronic diseases was low,

288
indicating that this was a relatively healthy population. At the same time, the prevalence of significant

293
In this regard, it is interesting that over the follow-up period the percentage of women who reported 294 having had some health problem or condition in the past month increased while the prevalence of 295 significant depressive symptoms decreased. When also considering the improvement in . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted August 26, 2022. ; socioeconomic factors (a higher percentage who had stable living conditions such as rented houses   297 or rooms, received subsidies or had contact with health services), these results seem to indicate that 298 more time in Colombia is associated with a more stable situation, which in turn is accompanied by 299 an improved emotional state. However, since at least one study has shown that the effect of 300 socioeconomic stressors on mental health during the resettlement process of displaced populations 301 differ between men and women, further research to explore this hypothesis is required [20].

302
The increase in the prevalence of reported health problems is an interesting finding. Regularization is also important because it directly impacts the health conditions of migrants and the 321 health services that are available to them.

322
The limitations of this study include having obtained more follow-up with the subsample of 323 participants who were already settled and had been living in Colombia for a longer period when 324 joining the study (baseline survey). Follow-up information was obtained from only 28% of those who 325 had been in the country for one month or less at the time of the baseline survey, compared to 55% of 326 those who had been in Colombia for more than one month and up to one year and 74% of those who 327 had been in Colombia for more than one year. However, the differences were not statistically 328 significant for most of the variables. The conclusions obtained may not be fully generalizable to newly 329 arrived migrants and those who are still in transit. In addition, due to its very nature, this survey is 330 not exempt from the possibility of information bias. Furthermore, the response rate was 56% at 331 follow-up. A sensitivity analysis of the population that was actually contacted one month after the 332 initial interview indicated that the variables that had significant differences, which can guide 333 additional analyses or follow-up studies, characterized the subsample that could be followed. This 334 needs to be considered when making practical decisions based on the data collected. In addition, there 335 is the possibility of information bias because it was detected that some migrants gave erroneous 336 answers thinking that they would receive a benefit. This could not be corrected for participants 337 without follow-up. The strengths of this study include 1) its longitudinal design, 2) the population of 338 women given their specific health needs and that they currently represent 49% of the Venezuelan 339 migrant population in Colombia [4], and 3) the sample size, which provided sufficient statistical 340 power for analyzing the data. This type of analysis could help to adapt interventions so that they are 341 more specific to the target population group.

342
343 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 26, 2022. ;

345
This study was able to collect a large amount of information. It is hoped that this data will be used to 346 generate research hypotheses for subsequent studies to investigate particular relationships or 347 associations between variables, or to provide explanations for some of the findings. It is also hoped 348 that this study will not only help to guide the design of public policies aimed at facilitating the 349 insertion of Venezuelan migrants into Colombian society and improving their quality of life, but that 350 it will also serve as a first step towards beginning to close the gap in knowledge that exists with  is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 26, 2022. ; https://doi.org/10.1101/2022.08.26.22279267 doi: medRxiv preprint