Dataset for the attitudes and practices regarding COVID-19 preventive measures in diverse settings of Mozambique: a qualitative study

Study participants

We recruited 295 participants from 24 sites in nine provinces in Mozambique. In-person semi-structured interviews (SSIs) were conducted with 23 participants, and the remaining 272 SSIs were conducted by telephone. Final study data was comprised of 159 females and 136 males aged 18-98 years old (median [interquartile range] = 35 [29-43]). Of note, data regarding declines and drop-outs were not collected (or registered) during data collection.

Study sites

The selection of study sites followed two steps. Primarily, during the first two days (17^th and 18^th of May 2020), data was collected in 6 sites (i.e., Manhiça and Mopeia districts, and Maputo, Matola, Xai-Xai and Quelimane cities) because Manhiça Health Research Center (CISM, Centro de Investigação em Saúde de Manhiça) has its offices and are places of residence of the research team members, and thus their constant presence and easy access to local communities. Secondly, and after an initial evaluation and preliminary data analysis, an additional 20 sites (i.e., Pemba, Moeda, Nampula, Nicoadala, Mocuba, Chinde, Ile, Morrumbala, Beira, Chimoio, Maxixe, Chonguene, Inhambane, Morrumbene, Jangamo, Limpopo, Mabalane, Chokwe, Boane, Magude) were included, and thus increasing the coverage of study area which includes cities and districts where the CISM office is not present thanks to the nature of the data collection approach (i.e., the use of the telephone for conducting the interviews). The second step lasted ten days, from 19^th to 28^th of May 2020. Telephone-based interviews allowed us to reduce the decline rate by including participants who are already in the network of each study team member and ensuring safety to COVID-19 thanks to the distance when conducting interview over telephone.

Qualitative tool guides

The qualitative data was collected using a semi-structured interview tool guide consisting of open-ended questions focused on ‘changes’ on participants’ routine dynamic soon after the declaration of the state of emergency in the country. The questions also sought to explore knowledge, attitudes, practices and adoption concerning COVID preventive measures in different contexts (e.g., at home and in public places).

Quantitative tool guides

The quantitative data were collected through a standardized structured questionnaire with close-ended questions including participants’ sociodemographic (i.e., age, gender, education level, occupation, marital status and his/her relationship with the head of the household) and household characteristics including place of residence, size of the household (e.g., number of household members and number of sleeping rooms).

Team composition

The present research was conducted by a multidisciplinary team (including biologists, geographers, anthropologists, sociologist, epidemiologists and public health specialists) with a minimum of two years working experience, who aimed to draw a general picture of COVID-19 response behaviours at community level. The study team was composed of 34 researchers including one senior investigator as a Principal Investigator, one PhD candidate, seven junior investigators, 24 research assistants, and one data manager, each aiming at conducting 10 interviews. Of those, 47% (16/34) were females living at the communities where the study took place or being closer to participants through various type of networking.

Selection of participants and data collection procedures

Participants were selected based on their closeness or kinship with the team member. First, each team member selected potential participants on his/her electronic phonebook. Second, potential participants were phone contacted and asked for an interview which was either scheduled for an in-person meeting (face-to-face interview) or a virtual, telephone-based, meeting interview. After obtaining oral consent, the interview was scheduled and conducted at the date, time and place of participants’ convenience. Socio-demographic characteristics of the participants and other quantitative data were collected before the open-ended interview lasting 45 to 60 minutes. While quantitative data were hand registered into a standardized form, interviews were audio-recorded using a digital voice recorder (Olympus WS-852 and WS-853, Olympus Imaging America INC., New York, USA). Additionally, summaries of the participants’ responses were directly hand registered in the qualitative tool guide.

Inclusion criteria

The study targeted different age groups aged 18 years old and above including young people (18-29 years old), adults (30-49 years old), and the elderly (50 years and older). Only participants who were close or with some degree of affinity and trust with at least one member of the research team, willing to participate and able to give his/her oral consent were included in the study.

Exclusion criteria

We excluded participants who were unable to give oral consent or were not able to complete the interview for personal reasons or due to failures in the connection and discharge of the mobile phone battery.

Data management

Quantitative data was entered into a data charting form in MS Excel (Microsoft Excel 2010, Microsoft Corp.; Redmond, WA, USA) accommodating all collected variables. Qualitative data were entered into an MS Excel matrix based on the main response (i.e., concise summary of answers for each question).

Ethical statement and consent to participate

This study was ethically evaluated and approved, on 16^th May 2020, by the Comité Institucional de Bioética em Saúde at Centro de Investigação em Saúde de Manhiça (Ref. CIBS-CISM/026/2020). Before starting the interviews, all participants were informed about the study objectives, procedures and the voluntary nature of participation through information read out from a participant’s information sheet, followed by oral voice recorded consent to participate in the study. Oral consent was sought instead of written consent to minimise the risk of COVID-19 infection, both for the participant and the interviewer, particularly for those invited to participate in in-person interviews by reducing contact between the interviewee and the interviewer, sharing of objects (e.g., paper and pens) and maximizing social distancing. Recorded oral consent was also a recommendation from the committee in the conduct of studies for maximizing protection against COVID-19 infection.

Quantitative data validation

The collection of data about ‘education level’ and ‘occupation’ resulted in a high number of categories (i.e., >10 responses) due to inconsistency of participants’ responses. Thus, some categories were merged based on their similarities (e.g., retired and unemployed) or lower response rates under a variable named ‘others’.

Limitations

This study has some significant limitations that may have influenced the reliability of the data. Such limitations invite a cautious interpretation and guide lessons learned for future studies. Telephone-based interviews introduced several challenges including poor listening, cancellation of some interviews because of audio quality and cell phone discharge during the interview. In addition, some participants could not finish the telephone interviews claiming it to be a lengthier interview than expected or to be involved in other activities. Although participants were conveniently selected, similar to what happens in other qualitative studies (Moré, 2015; Paiva et al., 2017), selection bias could also be introduced considering that it was based on the electronic phonebook of each research team member. Furthermore, the present dataset is prone to reporting bias considering that participants may have been influenced by the fact that they knew the interviewer.

Therefore, other studies using a similar methodology should take into account the triangulation of data originated by different data collection techniques; this can ensure the validity of the data, as the limitations of one technique will be reinforced by the strengths of the other techniques.

Underlying data

Open Science Framework: Dataset for the attitudes and practices regarding COVID-19 preventive measure in diverse settings of Mozambique: a qualitative study, https://doi.org/10.17605/OSF.IO/DM7QR (Magaço et al., 2022).

This project contains the following underlying data:

Extended data

Open Science Framework: Dataset for the attitudes and practices regarding COVID-19 preventive measure in diverse settings of Mozambique: a qualitative study, https://doi.org/10.17605/OSF.IO/DM7QR (Magaço et al., 2022).

Data are available under the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).