Introduction
Maternal health-care utilization is a core component of the continuum of reproductive health and contributes to the prevention, early detection and management of maternal and neonatal complications. Antenatal care provides opportunities for clinical assessment, immunization, nutritional supplementation, counselling and birth preparedness, while skilled attendance at delivery provides timely access to life-saving care. Utilization, however, is shaped not only by service availability but also by education, household resources, residence, social norms and accessibility. Indian studies have consistently shown socioeconomic and geographic inequalities in maternal health-care use [1,2].
These inequalities have been particularly relevant in Uttar Pradesh. Earlier analyses demonstrated that women's education, household standard of living and antenatal-care contact influenced subsequent use of institutional and skilled delivery services [1,2]. The introduction of Janani Suraksha Yojana and the National Rural Health Mission in 2005 represented a major policy shift toward promoting institutional delivery and strengthening access for disadvantaged populations; subsequent evaluations documented substantial increases in facility births, although gains were uneven across population groups and states [3,4].
Research from Uttar Pradesh has further shown that maternal health-care utilization reflects multiple, interacting determinants. Women's literacy, husband's education, socioeconomic circumstances and age at marriage have been associated with pregnancy registration, antenatal care and place of delivery [5]. Intersections between household wealth, literacy, religion and other social characteristics can further widen differences in service use [6]. Contact with frontline health workers is also important: greater community health-worker engagement has been associated with improved antenatal and postnatal service utilization, while birth preparedness and planning for transport and place of delivery increase the likelihood of institutional childbirth [7-9].
Despite substantial expansion of maternal health programmes, socioeconomic gradients remain visible in contemporary Uttar Pradesh, with differences in public and private service use by wealth, education and rural-urban residence [10]. Historical community-based data are therefore useful for understanding the starting point of these changes and for identifying determinants that have persisted despite health-system expansion.
The present study was undertaken to compare antenatal-care utilization, health-worker and health-facility contact, and delivery practices among women residing in rural and urban-slum areas of Agra district, and to identify sociodemographic and service-related factors associated with maternal health-care utilization.
Methods and Material
Study design and setting
A community-based cross-sectional comparative study was conducted in selected rural villages and urban slums of Agra district, Uttar Pradesh, through the Department of Social and Preventive Medicine, Sarojini Naidu Medical College, Agra.
Study period and participants
The study was conducted from May 2005 to October 2006. Women of reproductive age who were currently pregnant or had delivered during the preceding three years were eligible for inclusion.
Sample size and sampling
Sample size was calculated using an antenatal-care prevalence of 42%, a 20% relative error and a 95% confidence level. The calculated sample size was 138; after allowing approximately 10% additional participants, 150 women were included from each residential group, giving a total sample of 300. Multistage systematic random sampling was used. Municipal wards and slums were selected in urban areas, while primary health centres, subcentres and villages were selected in rural areas, followed by household-level identification of eligible women.
Data collection
Data were collected by personal interview using a predesigned schedule that had been field-tested in a pilot study. Information included sociodemographic characteristics, antenatal-care utilization, health-worker contact, health-facility use and delivery practices.
Statistical analysis
Data were analysed using SPSS version 10 (SPSS Inc., Chicago, IL, USA). Categorical variables were summarized as frequencies and percentages. Associations between categorical variables were assessed using chi-square tests. Logistic regression analysis was used to evaluate factors associated with selected maternal health-care outcomes, with effect estimates presented as odds ratios and 95% confidence intervals. A p-value <0.05 was considered statistically significant.
Ethical considerations
The study was conducted after approval from the Institutional Ethics Committee of Sarojini Naidu Medical College, Agra. Written informed consent was obtained before participation. Consent was documented by signature or thumb impression after participants had been informed about the purpose and procedures of the study. Confidentiality was maintained throughout the study. The manuscript was prepared in accordance with STROBE reporting recommendations for cross-sectional studies.
Results
A total of 300 women were included, with 150 each from urban-slum and rural areas. Overall, 78.3% were Hindu, 53.7% belonged to the medium standard-of-living category, 54.7% were illiterate and 58.7% lived in nuclear families. Rural women had lower literacy and a greater representation in the low standard-of-living category than urban-slum women (Table 1).
| Characteristic | Urban-slum (n=150), n (%) | Rural (n=150), n (%) | Total, n (%) |
| Religion | |||
| Hindu | 87 (58.0) | 148 (98.7) | 235 (78.3) |
| Muslim | 63 (42.0) | 2 (1.3) | 65 (21.7) |
| Standard of living index | |||
| Low | 14 (9.3) | 59 (39.3) | 73 (24.3) |
| Medium | 88 (58.7) | 73 (48.7) | 161 (53.7) |
| High | 48 (32.0) | 18 (12.0) | 66 (22.0) |
| Age group (years) (a) | |||
| <=20 | 10 (6.7) | 16 (10.7) | 26 (8.7) |
| 21-25 | 72 (48.0) | 61 (40.7) | 133 (44.3) |
| 26-30 | 47 (31.3) | 50 (33.3) | 97 (32.3) |
| 31-40 | 21 (14.0) | 23 (15.3) | 44 (14.7) |
| Type of family | |||
| Nuclear | 80 (53.3) | 96 (64.0) | 176 (58.7) |
| Joint | 70 (46.7) | 54 (36.0) | 124 (41.3) |
| Literacy of participant | |||
| Illiterate | 69 (46.0) | 95 (63.3) | 164 (54.7) |
| Up to 10th standard | 63 (42.0) | 50 (33.3) | 113 (37.7) |
| Above 10th standard | 18 (12.0) | 5 (3.3) | 23 (7.7) |
| Literacy of husband | |||
| Illiterate | 49 (32.7) | 35 (23.3) | 84 (28.0) |
| Up to 10th standard | 75 (50.0) | 75 (50.0) | 150 (50.0) |
| Above 10th standard | 26 (17.3) | 40 (26.7) | 66 (22.0) |
N, total study population; n, number of participants.
a Age categories were consolidated from the original study table for concise journal presentation; percentages were recalculated from the reported counts.
Health-worker contact differed sharply by residence: 60.0% of rural women reported a health-worker visit compared with 4.0% of urban-slum women. Conversely, 39.3% of urban-slum women and 25.3% of rural women had visited a health facility for treatment in the preceding three months. Among women with the type of facility recorded, approximately two-thirds had used private services (Table 2).
| Health-service characteristic | Urban-slum, n (%) | Rural, n (%) | Total, n (%) |
| Health-worker visit | 6 (4.0) | 90 (60.0) | 96 (32.0) |
| Visit to any health facility for treatment | 59 (39.3) | 38 (25.3) | 97 (32.3) |
| Type of facility visited (a) | |||
| Government facility | 20 (34.5) | 11 (29.0) | 31 (32.3) |
| Private facility | 38 (65.5) | 27 (71.1) | 65 (67.7) |
n, number of participants.
a Facility-type percentages are based on 58 urban-slum women and 38 rural women with facility type recorded (total n=96). Percentages were recalculated from the reported counts.
At least one antenatal-care visit was reported by 83.3% of urban-slum women compared with 53.3% of rural women. Completion of three antenatal visits was low in both groups (20.7% and 11.3%, respectively). Delivery care showed a wider disparity: 60.1% of rural deliveries were conducted by an untrained attendant compared with 2.0% of urban-slum deliveries, while private doctors conducted 58.7% of urban-slum and 25.7% of rural deliveries (Table 3).
| Maternal health-care indicator | Urban-slum, n (%) | Rural, n (%) | Total, n (%) |
| At least one antenatal-care visit | 125 (83.3) | 80 (53.3) | 205 (68.3) |
| Completedthreeantenatal-care visits (a) | 31 (20.7) | 17 (11.3) | 48 (16.0) |
| At least one tetanus immunization (b) | 137 (91.3) | 127 (84.7) | 264 (88.0) |
| More than one tetanus injection (b) | 133 (88.7) | 120 (80.0) | 253 (84.3) |
| Delivery attendant (c) | |||
| Government doctor | 12 (8.0) | 6 (4.1) | 18 (6.0) |
| Private doctor | 88 (58.7) | 38 (25.7) | 126 (42.3) |
| Government nurse | 1 (0.7) | 7 (4.7) | 8 (2.7) |
| Private nurse | 6 (4.0) | 7 (4.7) | 13 (4.4) |
| Trained traditional birth attendant | 39 (26.0) | 0 (0.0) | 39 (13.1) |
| Friend/relative | 1 (0.7) | 1 (0.7) | 2 (0.7) |
| Untrained attendant | 3 (2.0) | 89 (60.1) | 92 (30.9) |
n, number of participants.
a Three antenatal visits reflect the study-era programme indicator used in the original analysis.
b Tetanus indicators are overlapping; the category "more than one injection" is a subset of "at least one immunization".
c Delivery-attendant information was available for 298 deliveries (urban-slum n=150; rural n=148).
Logistic regression showed statistically significant associations of antenatal-care utilization with residential status, low standard of living, health-worker visit and husband's occupation. Residential status had an odds ratio of 4.633 (95% confidence interval 2.439-8.802; p=0.001). For delivery by trained personnel, residential status, antenatal-care attendance, low standard of living, selected husband-occupation categories and husband's education were statistically associated in the original regression model (Table IV).
| Outcome | Factor | Odds ratio | 95% confidence interval | p-value |
| Antenatal-care utilization | Residential status (urban/rural) | 4.633 | 2.439-8.802 | 0.001* |
| Low standard of living | 0.252 | 0.085-0.745 | 0.013* | |
| Health-worker visit | 0.413 | 0.220-0.777 | 0.006* | |
| Husband occupation: business | 0.432 | 0.224-0.831 | 0.012* | |
| Delivery by trained personnel | Residential status (urban/rural) | 46.949 | 11.422-192.958 | 0.001* |
| Antenatal-care visit | 2.454 | 1.108-5.437 | 0.027* | |
| Family size <4 | 3.301 | 0.937-11.636 | 0.063 | |
| Medium standard of living | 0.345 | 0.112-1.070 | 0.065 | |
| Low standard of living | 0.251 | 0.068-0.929 | 0.038* | |
| Participant literacy up to 10th standard | 1.901 | 0.934-3.867 | 0.076 | |
| Husband occupation: business | 0.341 | 0.151-0.770 | 0.010* | |
| Husband occupation: labour | 0.307 | 0.101-0.931 | 0.037* | |
| Husband occupation: farming | 0.165 | 0.062-0.442 | 0.001* | |
| Husband literacy up to 10th standard | 0.507 | 0.255-1.009 | 0.053 | |
| Husband literacy above 10th standard | 0.328 | 0.124-0.867 | 0.025* |
* p<0.05.
The odds ratios and confidence intervals are reproduced from the original archived regression tables. Reference-category coding was not explicitly documented in the archived tables; therefore, the direction of categorical comparisons should be interpreted cautiously.
Discussion
The present study documents a pronounced rural-urban gradient in maternal health-care utilization in Agra during 2005-2006. Antenatal-care contact, continuity of antenatal visits and access to trained delivery personnel were substantially better among urban-slum women, whereas rural women were much more likely to have delivery conducted by an untrained attendant. Residential status, socioeconomic position, health-worker contact and selected household characteristics were associated with maternal health-care utilization. These findings should be interpreted in the context of a health system that was entering a major transition: the National Rural Health Mission and Janani Suraksha Yojana had only recently been introduced, and community-based linkages to institutional maternity care were still expanding.
The contrast with contemporary service utilization is substantial. Using National Family Health Survey-5 data, Singh et al. reported that 93.8% of women initiated the maternal-care pathway with at least one antenatal visit and 63.1% completed four or more visits [11]. These estimates are markedly higher than the 68.3% overall antenatal-care contact and 16.0% completion of three visits observed in the present study. The direction of inequality, however, remains familiar: Singh et al. continued to demonstrate socioeconomic and residential differences across the continuum of care. Thus, service coverage has improved greatly, but the social patterning of incomplete care has not disappeared.
The definition of adequate antenatal care has also become more demanding. In the period of our study, three antenatal visits were commonly used as a programme indicator. More recent analyses assess not only the number and timing of contacts but also the content and quality of care. Girotra et al. found that 59.25% of women in India had four or more antenatal visits in National Family Health Survey-5, yet only a much smaller proportion received antenatal care meeting their composite definition of adequate quality [12]. Nihal and Shekhar similarly reported improvement in four-visit coverage between successive national surveys while showing that adequate-quality care remained incomplete for many women [13]. Compared with 2005-2006, the public-health challenge has therefore shifted from establishing initial contact toward ensuring continuity, completeness and quality at each contact.
Community health-worker engagement provides another important historical comparison. In the present study, a health-worker visit was reported by 60.0% of rural women but only 4.0% of urban-slum women, and health-worker contact was associated with antenatal-care utilization. Chaurasiya et al., studying Accredited Social Health Activist services in western Uttar Pradesh, later demonstrated substantial involvement of these workers in antenatal contacts, counselling and linkage to institutional delivery, although several recommended services remained incomplete [14]. Smittenaar et al. further showed that earlier and repeated community health-worker interactions were associated with multiple antenatal check-ups, iron-folic acid use and facility delivery in Uttar Pradesh [15]. These findings are consistent with the association observed in our study and illustrate how community outreach evolved from relatively limited contact into a more structured platform for pregnancy tracking, counselling and referral.
The historical difference in delivery care was particularly striking. In our study, 60.1% of rural deliveries were conducted by an untrained attendant, compared with 2.0% among urban-slum women; private doctors conducted a much larger proportion of urban than rural deliveries. Such a pattern reflects the limited reach of skilled delivery care in rural communities at that time. Contemporary national evidence indicates a major increase in skilled birth attendance. Mishra et al. demonstrated that skilled birth attendance was positively associated with antenatal-care utilization and negatively associated with rural residence and socioeconomic disadvantage [16]. The absolute situation has therefore changed substantially, but the geographic and socioeconomic gradients seen in the present study continue to have contemporary parallels.
Socioeconomic position was closely related to maternal-care utilization in the present analysis. Low standard of living and selected household characteristics were associated with antenatal-care use and trained delivery care. Gandhi et al. demonstrated persistent pro-rich inequality in the continuum of maternal health services during the decade following implementation of the National Rural Health Mission [17]. In a separate analysis, Gandhi et al. identified wealth, education, residence and other social factors as important predictors of progression through the maternal-care continuum [18]. These observations are consistent with our findings and suggest that expansion of services alone does not fully neutralize the influence of social disadvantage.
Preference for private facilities was also evident in our study: among women for whom facility type was recorded, approximately two-thirds had used a private facility. This pattern remains important today because choice of provider is closely linked to affordability and financial protection. Manna et al., using National Family Health Survey-5 data, reported substantially higher out-of-pocket expenditure for institutional delivery in private facilities than in public facilities [19]. Thus, the preference for private care observed in 2005-2006 may have reflected perceived accessibility, responsiveness or quality, but contemporary interpretation must also consider the financial consequences of private maternity care.
Another feature of the historical data was the coexistence of relatively high coverage of selected preventive interventions with incomplete antenatal follow-up. At least one tetanus immunization was reported for 88.0% of women, whereas only 16.0% completed three antenatal visits. Current approaches increasingly distinguish simple service contact from adequate-quality antenatal care, incorporating investigations, supplementation, counselling and continuity of contact [12,13]. This evolution in measurement is important when comparing historical and contemporary studies: improvement in a single indicator cannot be equated with comprehensive maternal care.
Recent evidence from Uttar Pradesh confirms that geographic and socioeconomic inequalities remain relevant. Sharma et al., using National Family Health Survey-5 data, identified maternal education, household wealth and birth order as important factors associated with complete antenatal care and demonstrated spatial clustering of inadequate antenatal-care utilization across districts [20]. These findings support the central observation of our study that maternal-care use is influenced by the social and residential environment in which women live, although the scale and organization of available services have changed markedly since 2006.
Several limitations should be considered. The study was restricted to selected villages, wards, subcentres and urban slums of Agra district, limiting generalizability. Information was interview-based and therefore susceptible to recall and social-desirability bias. The original regression tables did not preserve all coding details for categorical reference groups, which limits interpretation of the direction of some odds ratios. In addition, the data were collected in 2005-2006; maternal health programmes, institutional-delivery practices and standards of antenatal care have changed substantially since then. The estimates should therefore be interpreted primarily as a historical community baseline rather than as measures of current service coverage.
In conclusion, marked rural-urban inequalities in antenatal-care utilization and trained delivery care were present in Agra during 2005-2006. Over the subsequent two decades, maternal-health coverage and institutional delivery have improved considerably, yet contemporary studies continue to identify socioeconomic, educational and geographic inequalities. The historical findings therefore remain useful for demonstrating both the magnitude of progress and the persistence of structural determinants that should continue to inform equitable maternal-health planning.
Declarations
Ethics approval and consent to participate
The study was approved by the Institutional Ethics Committee of Sarojini Naidu Medical College, Agra. Written informed consent was obtained from participants and documented by signature or thumb impression. Participant confidentiality was maintained.
Consent for publication
Not applicable
Conflict of Interests
None
Funding Statement
None
Authors' Contribution
All authors have read and agreed to the published version of the manuscript:
Conceptualization
Dr Preeti Nigam; Methodology: Dr Preeti Nigam; Validation: Dr Naveen Yadav; Formal Analysis: Dr. Sunil Kumar Dahiya, Dr. D. N. Bharadwaj; Investigation: Dr Preeti Nigam; Resources: Dr Preeti Nigam; Data Curation: Dr Preeti Nigam; Dr. Sneh Kumari; Writing (Original Draft): Dr. Sunil Kumar Dahiya; Dr. DN Bharadwaj; Writing (Review and Editing): Dr Jitender Kumar Jakhar, Dr. Sunil Kumar Dahiya.
Acknowledgments
None