Since the goal of the present study was to determine the health concerns caused by the coronavirus pandemic and design the intervention based on their health concerns, a multistage mixed-method approach was used in three stages. Stage one, an initial qualitative study was conducted to explain the health concerns. Later, in stage two, following the findings of the first stage, by implementing the Nominal Group Technique (NGT) and reviewing the literature, an appropriate intervention was designed. In the final stage (quantitative study), the clinical intervention was tested.
Stage One:
In qualitative studies, participants were the 18 average people aged 21-43 who lived in Hamadan who had enough, sufficient, and first-hand information about the COVID-19 pandemic. Qualitative data has consisted of informant interviews (Holloway & Wheeler, 2013). The criteria for entering the interview of the study were as follows: Living in the same city (due to the impact of culture on the concerns and the type of concerns), being able to speak and understand Persian and Turkish, willing to participate in the study and retell their experiences related to COVID-19 pandemic, as well as having first-hand experiences of culture, social process, or this phenomenon (Cresswell; & Plano Clark, 2011) by which actively participate in the study and provide a better description of this phenomenon (Holloway & Wheeler, 2013). Considering the nature of this study, the number of interviews was related to saturation of data. After 21 interviews, the data was saturated, so that the classes and sub-classes were completed and no new information appeared in the new class, and the data analysis ended with the same number of interviews. The 21 interviews were conducted with 18 people (three participants were taken into it two times). The 18 participants were 8 women (4 married and 4 single), and 10 men (6 single and 4 married) with different ages and educatıon were selected for participating at this stage of the study.
In-depth semi-structured interviews were used to collect data. The duration of the interviews ranges from 35 to 90 minutes with an average of 60 minutes depending on the participants’ willingness to the response. The interviews were conducted individually by using Skype and WhatsApp video calling with participants’ permission and maintaining their privacy. Participants’ psychological security was provided so that they could freely express their experiences, thoughts, feelings, and perceptions. They were assured that their information would be kept confidential and unnamed. As for the beginning, demographic data and disease questionnaire was filled by all participants. Moreover, the concept of health and concern with intelligible language briefly explained. Later, the interview began with the question "What do you know about corona disease?" Subsequent questions were asked based on the individuals’ initial answers and the interview guide. Also Probing Questions such as "What does this mean" or "If you can please explain more" were used as needed in the interviews. As there was an interview guide, the interviewer started with simple and general questions, and if the participants were responding as it was expected, the interview was continued with the specific and more detailed questions. When interview questions were done, the participants were given time to ask if they want to know about something that had not been explained. As the study was going further interview questions were completed. All interviews were recorded on the voice recorder with the permission of the participants. The interview was stopped wherever it was necessary and then the interview and re-recording began. The researcher kept the contact number of each participant so that they could be contacted again if needed, and provided the participants with their number so that they could call if necessary. As for qualitative data analysis, content analysis was used in this stage. Content analysis is one of the research methods for analyzing textual data. It goes beyond the manifest content of the text but also analyzes the latent content (Graneheim & Lundman, 2004). In qualitative content analysis, three different approaches were used to interpret the text content. In this study, conventional content analysis was used. This method is usually used in designing a study in which its primary goal is describing a phenomenon. This method is applicable when it is facing the dearth of literature, theories, or knowledge on a phenomenon.
Researchers refuse to use pre-defined classes and instead allow classes and their names to be derived from data to form new insights (Elo & Kyngäs, 2008).
In this study, qualitative content analysis was performed through the method of Graneheim and Lundman (2004) as follows:
- Implement the entire interview just after each interview
- Read the text to a rich understanding of the whole content
- Identifying the meaning units and preliminary codes
- Classification of homogenous preliminary codes in more comprehensive classes
- Identifying the main themes of the classes (Graneheim & Lundman, 2004)
To establish the trustworthiness of qualitative data, the data was checked in term of five aspects of credibility, dependability, transferability, confirmability, and authenticity.
In the present study, the interview scripts were typed in Microsoft Word at the earliest opportunity. Then to get a rich understanding, it was read several times word by word, line by line, paragraph by paragraph. Thus, the meaning units were identified in the text of each interview that formed the study analysis units. The meaning units, or in other words, the set of words or phrases that carried a particular meaning and concept related to the research purpose, were summarized as "code." The original codes which were around a central concept were placed inside a subcategory. Then the subcategories were reviewed several times and compared based on similarities and differences, and categories and themes were formed. Attempts were made to have the most homogeneity within the categories and the most heterogeneity between the categories. Texts and codes were managed using the 2018 MAXQDA software.
Stage Two:
In the second phase of the study, a four-step process was developed to design an intervention based on the health concerns caused by the COVID-19 pandemic. In the first stage, a group of health experts who had a background in various fields of health care, or treatment, were selected to review the list of priorities and rated them to determine priorities again. Since the Nominal Group Technique (NGT) is a useful method for extracting the society’s health priorities (Latimer, 1994) this technique was applied to evaluate the health concern priorities caused by the COVID-19 pandemic. NGT is a structured small group discussion that uses to extract and prioritize a list of answers to a specific question. It provides an equal chance for all to express their opinions. The nominal group technique is implemented in four steps, including idea generation, recording of ideas, discussion of ideas, and voting.
In step one, participants expressed a list of health concerns about the COVID-19 pandemic in a format of categories. The list was equipped with the participants’ opinions about the most important concerns obtained in the first qualitative study. In step two, for better understanding the prevalence of these mentioned health concerns, the relevant literature was reviewed for preparing extra information for the expert panel. Later, experts in the panel were given the list of concerns from the qualitative study and they were asked for adding new health concerns to the list if necessary. In step three, discussion in the expert panel started and consequently, similar concerns were identified and merged. Moreover, Members of the panel discussed the significance of each item of the list of concerns and health issues based on the degree of prevalence (prevalence), Impact rate (Impact), and the possibility of reducing that concern by the intervention (Interventıon). They reached a compromise unanimously on the most important concerns including lack of accurate and sufficient information and unpredictability of the disease. Consequently, the last part of NGT or step four which is the voting with the card to categorize the concerns was not administered. Moreover, the members were asked for working with researchers to figure out an appropriate intervention for the concerns. Counseling was chosen as an appropriate and applicable intervention.
As the results of the meeting, the lack of information support (such as insufficient information, contradictory statistics, no follow-up by the hospital, and inadequate knowledge of physicians) was identified as the most important priority concern to start work and design the intervention. Hence, the primary goal of the quantitative study or stage three was to identify the effect of an intervention on the following concerns such as information support, psychological security, physical health, mental health, and social health.
After determining health concerns and information support as the first and most important priority, the relevant literature in the same field in Iran and other countries was reviewed to figure out the effective intervention. By using keywords like health keywords, mental health, physical health, social health, psychological security, and information support, the known database was surfed such as SID, Embase: Pubmed Science Direct, Cochrane Library, and Google Scholar; Magiran; Medline.
All articles published in 2018 to 2020 were transferred to Mendeley software and duplicate articles were deleted, hence, the remaining articles include: 1) Articles that its purpose was to design or implement an intervention for psychological security, information support, physical, mental, and social health at least as one of the objectives 2) Articles that its participants were people during PTSD at least as part of the study population; 3) all articles were either in English or Persian. While the requirements and the goal of the current study were fulfilled, only their abstracts were reviewed otherwise the full text of papers was scrutinized. A total of 8 articles were reviewed to determine the appropriate intervention for health concerns caused by the COVID-19 pandemic.
The analysis showed that 6 papers out of 8 were doctoral dissertations in the fields of health promotion, medical and social health from overseas, and 2 papers from Iran. The type of study was 7 randomized clinical trials (87.5%) and 1 non-randomized clinical trials (12.5%). The study samples were the general population of people who suddenly developed a specific disease. The number of the interventional sessions was between 2 to 6 sessions but mostly 5 sessions. In this study, the number of interventional sessions was 4 sessions.
It was difficult to decide about the effectiveness and intensity of specific intervention to rectify the health concerns caused by the COVID-19 pandemic because reviewed studies have been different in the method of collecting data, administration period, type, and intensity of intervention, research tools, and follow-up methods. However, the World Health Organization has emphasized counseling programs that increase information supports for non-pharmacological approaches for chronic disease. Therefore, counseling method was chosen to be suitable for dealing with health concerns caused by the coronavirus pandemic because of time-consuming, easy to implement, acceptable results from this intervention in similar studies, researchers enough experience in counseling, panel expert advice, and that is why it was improving psychological and social security.
Stage Three:
The third stage was a quantitative study of randomized clinical trials (RCT) with a control group. The primary aim was to determine the effect of educational counseling with an emphasis on increasing information support as an intervention method on health concerns caused by the coronavirus pandemic. The population of this stage was 400 members of Telegram, Instagram, and WhatsApp groups joined the research through the created link and answered the questionnaires. Of those with a standard deviation above average, 248 were selected and the rest were eliminated. They were then invited to participate in the study, of which only 72 were joined with their consent. The standard deviation in women was S =15 as estimated in similar studies (Daneshfar, et al., 2017) and (Zamani, Tavoli, Khasti, Sedighimornani, & Zafar, 2017). To test if the effect of preventive counseling statistically significant, along with the acceptable error rate d=10, confidence level 95%, and the power 80%, the sample was divided into two groups intervention and control including 36 people aged 21-43 years. In this way, all qualified people were participated in the study after providing complete explanations about the study purpose and getting conscious written consent. The samples were divided into two groups of intervention and comparison by the Blocked Randomization method and 72 participants were assigned to those two groups randomly.
Quantitative data were collected through the following questionnaires:
- Demographic information questionnaire
- Psychological security questionnaire
- Physical health questionnaire
- Mental health questionnaire
- Social health questionnaire
- Data support questionnaire
To analyze the data of this quantitative section, frequency distribution tables from descriptive statistics were used to present data. Moreover, Ki-Square and Independent t-test was used to examine demographic variables and clinical features of the two groups. Besides, to study physical health, mental health, and social health in two groups, repeated measures were used. The analyses were performed using SPSS software version 25 by considering Intention to treat (ITT) analysis and a significant level P <0.05.
Findings
In the first stage of study or qualitative part, 18 normal individuals aged 21-43; 8 women (4 married and 4 single), and 10 men (6 single and 4 married) with different education levels were selected for intervıew. First, 89 codes were extracted from the meaning units and after deleting duplicate codes ended to the following codes such as psychological security, information support, physical health, mental health, and social health. Later, to achieve the objectives of stage three of the study or quantitative part, the data from 72 participants in both intervention and control groups was prepared by using descriptive and analytical tests and diagrams, respectively.
Table 1: Frequency Distribution of Age Groups
Group
Age
group
|
Control
|
Intervention
|
Number
|
Percentage
|
Number
|
Percentage
|
<30
|
6
|
16.67
|
6
|
16.67
|
30-33
|
6
|
16.67
|
4
|
11.11
|
34-37
|
6
|
16.67
|
8
|
22.23
|
38-41
|
8
|
22.23
|
6
|
16.66
|
> 41
|
10
|
27.78
|
12
|
33.33
|
Total
|
36
|
100
|
36
|
100
|
Results
|
Independent t-test : No Significant
t = -0.43 dr = 70 p = 0.489
|
Table 2: Frequency Distribution of Psychological Security in the Intervention and Control Groups
Group
Age
group
|
Control
|
Intervention
|
Number
|
Percentage
|
Number
|
Percentage
|
High Security
|
6
|
16.66
|
4
|
11.11
|
High Moderate Security
|
10
|
27.78
|
8
|
22.23
|
Moderate Security
|
8
|
22.23
|
10
|
27.78
|
Low Moderate Security
|
6
|
16.66
|
6
|
16.66
|
Low Security
|
6
|
16.66
|
8
|
22.23
|
Total
|
36
|
100
|
36
|
100
|
Results
|
Independent t-test : No Significant
t = -1.08 dr = 70 p = 0.203
NNn
NNnnnnnn
|
Table 2 showed that the psychological security score in more than half of the participants was moderate to high in both groups. The independent t-test also showed that there is no significant difference in mental security between two groups and they were homogeneous in terms of severity.
Table 3: Frequency distribution of fatigue intensity, Physical health, Mental Health, Social Health
|
Control
|
Intervention
|
Number
|
Percentage
|
Number
|
Percentage
|
Physical Health
|
1-2.8 (Mild fatigue)
|
18
|
50
|
20
|
55.56
|
2.8-5 (Medium fatigue)
|
18
|
50
|
16
|
44.44
|
Total
|
36
|
100
|
36
|
100
|
Independent t-test : No Significant
t = 0.789 dr = 70 p = 0.021
|
|
Control
|
Intervention
|
Number
|
Percentage
|
Number
|
Percentage
|
Mental Health
|
High Mental Health
|
8
|
22.23
|
6
|
16.67
|
Moderate Mental Health
|
14
|
44.44
|
16
|
44.44
|
Low Mental Health
|
14
|
44.44
|
14
|
38.88
|
Total Mental Health
|
36
|
100
|
36
|
100
|
Independent t-test : No Significant
t = 1.03 dr = 70 p = 0.265
|
|
Control
|
Intervention
|
Mean ± SD
|
Mean ± SD
|
Social Health
|
Before Intervention
|
94.11 ± 18.97
|
94.47 ± 18.64
|
2 weeks after Intervention
|
92.35 ± 19.21
|
97.35 ± 18.02
|
3weeks after intervention
|
93.84 ± 20.32
|
97.54 ± 18.26
|
Results
|
F = 0.56
P = 0.198
|
F = 11.23
P = 0.013
|
Table 3 represented the majority of the participants in the intervention group were at a mild level of physical fatigue. In mental health, more individuals had moderate to low mental health in both groups. In social health, the average was higher in the intervention group than in the control group. It is mentioned that the results of the independent t-test showed that both groups were homogeneous in terms of fatigue severity and health before treatment or intervention.
Table 4. Frequency of index of information support before and after intervention
|
Control
|
Intervention
|
Mean± SD
|
Mean ± SD
|
Information
support
|
Before intervention
|
24.11 ± 3.64
|
24.85 ± 3.65
|
2 weeks after Intervention
|
25.30 ± 3.38
|
29.74 ± 2.89
|
3 weeks after intervention
|
24.30 ± 3.19
|
28.54 ± 2.31
|
Results
|
F = 0.37
P < 848400
|
F = 111.98
P < 0.001
|
Table 4 indicated the findings of repeated measures ANOVA by which the mean of the index of information support in the intervention group was different before intervention with two weeks or three weeks after intervention(p <001), whereas in the control group there was no significant difference (p <0.840). Moreover, the mean of the index of information support was significantly different between the intervention and control group (p <0.001).
Since the result of Repeated Measures showed the significant differences and to prevent data from mistakenly appearing to be statistically significant, the Bonferroni test was applied to compare the items.
Table 5: Comparison of index of information support before and after the intervention
Group
Time
|
Intervention
|
Mean
difference
|
p - value
|
Before Intervention
|
2 weeks after intervention
|
-4.56
|
p < 0.001
|
3 weeks after intervention
|
-4.78
|
p < 0.001
|
2 weeks after intervention
|
3 weeks after intervention
|
-0.22
|
p = 0.420
|
According to the Bonferroni test, there was a statistically significant difference between the mean of information support, before the intervention and two weeks (p <0.001) and 3 weeks after the intervention (p <0.001). In other words, the mean of information support before the intervention was less than after intervention. It shows intervention caused in increasing the information support score increased after the implementation of the intervention, but there was no statistically significant difference between two and three weeks after the intervention implementation (p = 0.420).