Study area
This study was conducted in the Madi Municipality of Chitwan District. In Chitwan, in addition to two pre-existing municipalities, village development committees were merged, and five new municipalities were gradually formed, including Madi. Madi was declared a municipality with 9 wards in 2014. It has nine public health facilities, including one hospital, three health posts, one community health care unit, and four urban health care centers. FP service is available at all the health facilities as well as from an FCHV of the respective ward. According to the record of household survey conducted by the municipality in 2017 [Local Level Publication 2017], Madi Municipality has a total population of 43,402, of which 12,577 are females of reproductive age (15-49). It comprises a total of 8,649 households inhabiting 37 types of caste/ethnicity groups. The major caste/ethnicity are Brahmin, Chhetri, Chaudhary, Chepang, Damai, Gurung, Tamang, Bote, Magar, Kami and Sarki[13 (16).
Study design and study participants
This was a cross-sectional study conducted among MWRA (15-49 years) who were residents of Kalyanpur of Madi Municipality. Among the 12,577 female of reproductive age group we interviewed married women aged 15 to 49 years residing in the municipality(16), willing to participate in the study and provided informed consent.
Sampling and sample size
A multistage sampling technique was used in the study. Among the newly formed municipalities in Chitwan, Madi was selected purposively, being farthest in distance from Bharatpur – the district headquarters – that takes a two-hour drive. A ward was selected randomly using the lottery method. Then, a systematic sampling technique was used to select households for data collection. The selected ward (ward number 7) had a total of 978 households and 1,373 females in the reproductive age group [14]. The sample size for the study was calculated as follows:
\(\:n=\frac{{Z}^{2}p\left(1-p\right)}{d²}\) = (1.96)2 x 0.24 x 0.76/0.062 = 195
where n is the sample size, Z is the z score (set to 1.96 at the 95% confidence level), p is the prevalence of unmet need (24%) and d is the margin of error (6%).
By considering a nonresponse rate of 10 percent, the final sample size (n)= 215
Then, the sampling interval was calculated as follows:
Sampling Interval = total households/sample size = 978/215 = 4.54
Every fifth house was selected after random selection of a first house. A total of 218 MWRAs were included in the study to cover all the remaining houses in the ward.
Study variables
Dependent variable: The dependent variable was ‘Unmet need for FP’.
Independent variables
Independent variables include the age of the respondent, religion, caste/ethnicity, wealth index, educational status of the respondent, occupation of the respondent, educational status of the husband, occupation of the husband, number of living children, decision-maker, frequency of discussion, and approval of husband.
Operational Definition
Unmet need for FP
Fecund women who desired to postpone birth for two or more years or stop childbearing altogether but were not currently using a contraceptive method, as well as pregnant women whose current pregnancy was unwanted/mistimed, were considered to have an unmet need for FP. The unmet need was categorized as ‘Yes’ (coded 1) and ‘No’ (coded 0)(17)(4).
(1) Age of respondent: Completed years of life of respondent. It was classified into three categories: ‘≤24 years’, ‘25-34 years’ and ‘35 years & above’(18).
(2) Religion: Religion followed by the respondents. It was classified into two categories: ‘Hindu’ and ‘Non-Hindu’ (Buddhist, Christian, and Muslim) (19)(20).
(3) Caste/ethnicity: This was categorized as ‘Advantaged caste/ethnicity’ and ‘Disadvantaged caste/ethnicity’. So-called upper caste group, i.e., Brahmin & Chhetri and relatively advantaged Janajati were classified as ‘Advantaged caste/ethnicity’, whereas Dalits and Disadvantaged Janajati were classified as ‘Disadvantaged caste/ethnicity’[13, 17].
(4) Wealth index: A wealth index was created using a principal component analysis of a number of household assets. It was classified into ‘Poor’ and ‘Rich’. Household assets were considered for calculation of wealth index as adopted by Shah R, et al(21).
(5) Educational status refers to completed years of schooling of the respondents and their husbands. It was measured in three categories – ‘no schooling’, ‘primary education’, ‘secondary education’ and ‘postsecondary or higher education’. Respondents who had completed any grades from 1 to 5 were classified as having ‘primary education’, respondents who had completed 9th grade up to 12th grade were classified as having ‘secondary education’, and respondents with any completed education higher than 12th grade were classified as having ‘postsecondary or higher education’(6).
(6) Occupation of respondent: Occupation of the respondent at the time of data collection was classified into three categories: ‘Housewife’, ‘Agriculture’ and ‘Employed within country’ (service/labor/business)’(22)(23).
(7) Occupation of husband: Occupation of the respondent’s husband at the time of data collection. It was classified into three categories: ‘Agriculture’, ‘Foreign employment’, and ‘Employed within country’ (Service/Business/Labor) (22)(23).
(8) Number of living children: The number of living children of the respondents. It was classified as ‘No child’, ‘Up to 2 children’, or ‘3 or more children’(24).
(9) Decision maker: The individual who makes decisions regarding the use of the FP method. It was categorized as ‘By woman herself’ and ‘Husband and wife jointly’(25)(26).
(10) Frequency of discussion: This variable assessed the number of discussions on FP that a couple had regarding family planning. It was measured as ‘Never’, ‘Once or twice’ and ‘Three or more times’(27)
(11) Approval of husband: The variable assessed perceived approval of husband for using contraceptives. The response was categorized as ‘Approves’ and ‘Disapproves’.
Data collection procedure
Data were collected from June 30 to July 20, 2019. Data were collected using a structured questionnaire that was developed from relevant literature (3)(18)(27)(28) in the local Nepali language. The English version of questionnaire is available at [Supplementary file- S1]. The questionnaire was pretested in a similar setting of another municipality (Kalika municipality) in Chitwan. The questionnaire was then customized based on the findings of the pretesting, which included the addition of a ‘skip’ label, deletion of vague words, and ‘reordering’ the questions to make it friendly to both the data collector and respondents. The final questionnaire consisted of variables related to socio-demographic characteristics, knowledge and practice of FP, and decision-making.
Data were collected through face-to-face interviews by visiting women at their homes. A brief overview of the objectives of the present study was provided prior to the data collection among the women. The first house was selected by taking the community school as a landmark (29). A direction was randomly chosen by tossing a pen and following the direction of the point of the pen(30). Then, every fifth house was chosen until the required sample size was met. Additionally, two more eligible women were included from the remaining houses following the same procedure. Any fifth house where there was no eligible woman or was not available, the house just next to that was chosen as a sample, and then the next fifth house was selected. If there was more than one eligible woman in a house, one was randomly selected using the lottery method for interviews. Data were collected by the first (RG- BPH Student) and third author (YR- BPH Student) themselves.
Data processing and analysis
The data were entered and analysed using SPSS version 20 software. The frequency and percentage of data were calculated in descriptive analysis. Pearson’s chi-square test was used to assess the statistically significant association of the independent and dependent variables. Multicollinearity between independent variables was checked using variation inflation factor (VIF), taking the highest VIF value >3 into account (REF 1). The VIF value for any of the independent variables was not found >3. The variables that were significant at P value ≤0.1 were entered in multivariable analysis to ensure that potential predictors are included in the multivariable analysis(31). A P value <0.05 was considered to indicate statistical significance in multivariable analysis.
Ethical considerations
Ethical approval for the study was obtained from the Institutional Review Committee (IRC) of Shree Medical & Technical College (SMTC), Bharatpur, Chitwan (SMTC-IRC-2020720-2). Written informed consent was obtained from the respondents before initiating the interview using an informed consent form. Voluntary participation of the respondents was ensured throughout the study. The information collected was kept confidential and was used for research purposes only.