Sage Journals: Discover world-class research

Abstract

Introduction

The objective of this study is to compare the clinical profile, risk factors, type and outcome of stroke patients in urban and rural areas of Punjab, India.

Methods

The primary data source was from the Ludhiana urban population-based stroke registry. The data of first-ever stroke patients with age ≥18 years were collected using WHO stepwise approach from all hospitals, general practitioners, physiotherapy and scan centres between 26 March 2011 and 25 March 2013.

Results

A total of 4989 patients were included and out of 4989 patients, 3469 (69%) were from urban areas. Haemorrhagic stroke was seen more in rural as compared to urban regions (urban 1104 (32%) versus rural 552 (36%); p = 0.01). There were significant differences seen in stroke risk factors; hypertension (urban 1923 (84%) versus rural 926 (89%); p = 0.001) and hyperlipidaemia (urban 397 (18%) versus rural 234 (23%); p = 0.001) between two groups. In the multivariable analysis the rural patients were more likely to be younger (age < 40 years) (OR: 1.82; 95% CI: 1.24–2.68; p = 0.002), Sikhs (OR: 2.57; 95% CI: 1.26–5.22; p = 0.009), farmers (OR: 9.41; 95% CI: 5.36–16.50; p < 0.001), housewives (OR: 2.71; 95% CI: 1.45–5.06; p = 0.002), and consumed alcohol (OR: 1.57; 95% CI: 1.19–2.06; p = 0.001) as compared to urban patients. In addition, use of imaging was higher in rural patients (OR: 1.99; 95% CI: 1.06–3.74; p = 0.03) as compared to urban patients.

Discussion and Conclusion

In this large cohort of patients, rural and urban differences were seen in risk factors and type of stroke. Stroke prevention strategies need to take into consideration these factors including regional sociocultural practices.

Keywords

India population based registry rural stroke urban

Introduction

Stroke is one of the leading causes of death and disability worldwide. According to WHO, stroke is the second commonest cause of mortality worldwide. By the year 2050, 80% of strokes will occur in low- and middle-income countries.¹ Differences in stroke rates have been noted from USA, France, Northern Portugal and Spain between urban and rural populations.^2–5

Seventy per cent of Indian population lives in villages. The rural and urban differences in the stroke profile, risk factors and outcome have not been well studied in India and the results are conflicting.⁶ The incidence rate per 100,000 was similar in Kolkata urban study (123.15) and rural Bengal (124) whereas higher case fatality rate (CFR) was seen in Kolkata (41.08%) as compared to rural Bengal (18%).^7,8

The Trivandrum stroke registry from South India showed similar incidence rates in both urban and rural regions but the 28-day CFR was higher in the rural region (urban: 24.5% versus rural: 37.1%).⁹ A recent study on stroke mortality conducted in rural area of Gadchiroli in Maharashtra showed that stroke was the leading cause of death.⁶ Hence, the objective of this study is to compare the clinical profile, risk factors, type and outcome of stroke patients, in urban and rural areas of Punjab state.

Methods

Background of the Ludhiana population-based stroke registry

The data for this study were taken from Ludhiana population-based stroke registry which was conducted in the Northwestern state of Punjab in the city of Ludhiana. Punjab state has a total area of 50,362 km² (rural area: 48,265 km² and urban area 2097 km²).¹⁰ The population density in urban areas is high as compared to rural areas.

Total population of Punjab as per 2011 census is 27,743,338 out of which the total rural population of Punjab is 17,344,192.¹¹ The city has good health care facilities hence patients come from all regions of Punjab and also from other neighbouring states. Urban and rural areas are defined according to the Government of India. Urban population was defined living within the municipal corporation limits and the rest of the areas were conferred rural.¹²

The project was carried out in two phases: phase I (26 March 2010–25 March 2011) was the feasibility study and phase II (26 March 2011–25 March 2013) was the main study.¹³ Inclusion criteria were all first-ever stroke patient’s ≥18 years of age. WHO STEPS approach was used for stroke surveillance (Steps 1, 2 and 3) which fulfils the WHO criteria for population-based stroke registry in low- and middle-income countries.¹⁴ All the major hospitals (private and public), scan centres, general practitioners (GPs) and physiotherapy centres in the city participated in this registry and mortality data (hospital and home deaths) were collected from Municipal corporation (MC) centres.¹³ Separate questionnaires were used by the research staff for data collection from hospitals/GPs/physiotherapy centres, scan centres and MC.

All major hospitals and scan centres participated in the study; thus, the chances of missing information about a nonfatal stroke were very low. Stroke subtypes were categorised based on the scans reported by radiologists. The diagnosis of stroke was established in the hospitals by the participating practitioner based on the clinical signs, symptoms and imaging reports.

Duplication of the cases

The duplicate cases were identified on the basis of name, age, gender, address and contact numbers. When there was duplication of cases, the questionnaire with complete information was entered in the dataset.¹³

In order to capture minor strokes and patients who solicit treatment outside the city, advertisements were placed in the newspaper every six months requesting these patients to contact the research team.¹³ During the study, a total of 15 patients were ascertained by this method. The principal investigator (JDP) appraised all the completed forms.

Selection of dataset for this study

The first year of the study (26 March 2010–25 March 2011) was feasibility phase hence we did not include that data. All first-ever stroke patients (excluding transient ischemic attack) recruited from hospitals, GP, physiotherapy and scan centres between 26 March 2011 and 25 March 2013 were included in the analysis. On the basis of their addresses the data were further divided into two groups: rural versus urban stroke patients.¹²

The data regarding demographic details, presenting symptoms, risk factors, type of stroke, diagnosis modalities were collected from hospitals, GPs and physiotherapy centers.¹³ Information regarding demographic details, type of stroke and diagnosis modalities were collected from scan centres. The modified Rankin Scale was documented at 28 days using a structured interview on telephone for all hospital and scan centres patients.¹³ Only demography and cause of death were available from MC (home and hospital deaths). Hence we did not include MC data for this sub-paper. Moreover, the MC records maintain mortality information only of Ludhiana residents.

Statistical analysis

SPSS version 21 (Armonk NY: IBM Corp.) software was used for the analysis. For the comparison of categorical variables, Chi Square test and for continuous variables independent t-test (two sided and p < 0.05: significant) were used. Binary logistic regression analysis was performed to describe the characteristics of rural stroke patients as compared to urban patients. The variables included in the analysis were age, gender and all other significant variables in univariate analysis.

Results

In this registry a total of 7199 patients were recruited from 26 March 2010 to 25 March 2013. After excluding duplicate cases (n = 713), a small proportion of patients with incomplete addresses (n = 49) and transient ischemic attack (TIA) patients (n = 724), there were 5713 patients.We included only patients from hospitals/GPs/physiotherapy and scan centres from the state of Punjab which was 4989 (public hospitals: 207(4%), private hospitals: 3075 (62%), GPs: 35 (0.7%), physiotherapy centres: 13 (0.3%), scan centres: 1659 (33%)).Out of 4989, 3469 (69%) were from urban and 1520 (31%) were from the rural areas (Figure 1).

Figure 1.

Recruitment algorithm.

Comparison of demographic details between two groups

There were no differences in age and gender between urban and rural patients. Patients within the city were more likely to be literate (urban 1764 (81%) versus rural 732 (72%), p < 0.0001) and employed (urban 628 (29%) versus rural 167 (17%), p = <0.001). Sikhs were the major religious group in rural areas (urban 894 (39%) versus rural 818 (78%), p < 0.001) (Table 1).

Table 1.

Comparison of demographic details between the two groups.

Variables	Urban (n = 3469) n (%)^a	Rural (n = 1520) n (%)^a	p value
Age (years) Mean ± SD^b	59 ± 15	59 ± 15	0.97
Gender			0.69
Men	2166 (62)	958 (63)
Women	1303 (38)	562 (37)
Education			<0.0001
Illiterate^c	418 (19)	289 (28)
Literate	1764 (81)	732 (72)
Occupation			<0.001
Employed	628 (29)	167 (17)
Self-employed	317 (15)	48 (5)
Farmer	92 (4)	297 (30)
Housewives	744 (34)	357 (37)
Retired	272 (12.6)	88 (9)
Not working	107 (5)	22 (2)
Religion			<0.001
Sikh	894 (39)	818 (78)
Hindu	1328 (58)	212 (20)
Others^d	63 (3)	13 (2)

Percentage calculated after excluding missing data.

Standard deviation.

Less than primary school completed.

Muslim, Jain, Christian.

Comparison of type of stroke, imaging modality and outcome

A high proportion of haemorrhagic strokes were seen among the rural patients (urban 1104 (32%) versus rural 552 (36%), p = 0.01). The MRI scan rate was higher in rural patients as compared to urban stroke patients (urban 1852 (56%) versus rural 966 (64%), p < 0.001). There was no difference in outcome between the two groups (poor outcome (mRS: 3-6): urban 1404 (46%) versus rural 591 (48%), p = 0.26) (Table 2).

Table 2.

Comparison of type of stroke, diagnosis modalities and outcome between the two groups.

Variables	Urban (n = 3469) n (%)^a	Rural (n = 1520) n (%)^a	p value
Type of stroke			0.01
Ischaemic stroke	2311 (67)	949 (63)
Haemorrhage stroke	1104 (32)	552 (36)
Cerebral venous thrombosis	33 (1)	14 (1)
Diagnosis			<0.001
Clinical diagnosis alone	95 (3)	21 (1)
CT scan	1319 (40)	479 (32)
MRI scan	1852 (56)	966 (64)
CT and MRI scan both	56 (2)	34 (2)
mRS at 28 days follow up			0.26
Good outcome^b	1632 (54)	636 (52)
Poor outcome^c	1404 (46)	591 (48)

CT: computed tomography; MRI: magnetic resonance imaging; mRS: modified Rankin Scale.

Percentage calculated after excluding missing data.

mRS: 0–2.

mRS: 3–6.

Comparison of risk factors between two groups

Hypertension and diabetes mellitus were the two main risk factors seen in this cohort. There were significant differences seen in risk factor prevalence; hypertension (urban 1923 (84%) versus rural 926 (89%), p = 0.001) and hyperlipidaemia (urban 397 (18%) versus rural 234 (23%), p = 0.001) between the two groups. Tobacco use (both current and past) was more in urban areas (tobacco current: urban 130 (6%) versus rural 31 (3%), p = 0.001; tobacco past: urban 347 (15%) versus rural 65 (6%), p < 0.0001) whereas in rural patients consumption of alcohol was common (alcohol current use: urban 343 (15%) versus rural 224 (21%), p < 0.0001; alcohol past use: 832 (37%) versus 436 (42%), p = 0.004) (Table 3).

Table 3.

Comparison of risk factors between two groups.

Variables	Urban (n = 2286) n (%)^a	Rural (n = 1044) n (%)^a	p value
Atrial fibrillation	237 (10)	115 (11)	0.61
Diabetes mellitus	1065 (47)	499 (48)	0.52
Hypertension	1923 (84)	926 (89)	0.001
Carotid stenosis	66 (3)	31 (3)	0.92
Previous TIA	222 (10)	90 (9)	0.31
Rheumatic heart disease	43 (2)	10 (1.0)	0.05
Coronary artery disease	221 (10)	123 (12)	0.06
Drug addiction	135 (6)	55 (5)	0.46
Tobacco (current use)	130 (6)	31 (3)	0.001
Tobacco (past)	347 (15)	65 (6)	<0.0001
Hyperlipidaemia	397 (18)	234 (23)	0.001
Alcohol (current use)	343 (15)	224 (21)	<0.0001
Alcohol (past)	832 (37)	436 (42)	0.004
Pregnancy/postpartum	11 (0.5)	2 (0.2)	0.21

TIA: transient ischemic attack.

Percentage calculated after excluding missing data.

Multivariable analysis

In multivariable analysis differences were observed in rural and urban patients in their age, occupation, education, religion, diagnosis and alcohol consumption. The rural patients were more likely to be younger (age < 40 years) (OR: 1.82; 95% CI: 1.24–2.68; p = 0.002), Sikhs (OR: 2.57; 95% CI: 1.26–5.22; p = 0.009), farmers (OR: 9.41; 95% CI: 5.36–16.50; p < 0.001), housewives (OR: 2.71; 95% CI: 1.45–5.06; p = 0.002), and consumed alcohol (OR: 1.57; 95% CI: 1.19–2.06; p = 0.001) as compared to urban patients. In addition, use of imaging was higher in rural patients (OR: 1.99; 95% CI: 1.06–3.74; p = 0.03) as compared to urban patients (Table 4).

Table 4.

Rural and urban differences: Multivariable analysis.

Variables^a	Regression coefficient	Adjusted odds ratio	95% Confidence interval	p value
Age (years)
<40	0.60	1.82	1.24–2.68	0.002
40–64	0.25	1.28	0.95–1.73	0.11
65–74	0.08	1.09	0.80–1.48	0.60
≥75	Reference
Gender
Men	0.04	1.04	0.66–1.64	0.88
Women	Reference
Education
Literate	−0.25	0.78	0.62–0.97	0.03
Illiterate	Reference
Occupation
Employed	0.30	1.35	0.79–2.29	0.27
Retired	0.53	1.70	0.96–3.00	0.07
Self-employed	−0.07	0.93	0.51–1.68	0.80
Farmer	2.24	9.41	5.36–16.50	<0.001
Housewives	1.00	2.71	1.45–5.06	0.002
Not working	Reference
Religion
Sikh	0.94	2.57	1.26–5.22	0.009
Hindu	−0.37	0.69	0.34–1.41	0.31
Others	Reference
Type of stroke
Ischemic stroke	−0.02	0.98	0.81–1.17	0.82
Haemorrhage stroke^b	Reference
Diagnosis
Imaging (CT/MRI)^c	0.69	1.99	1.06–3.74	0.03
Clinical diagnosis	Reference
Hypertension
Yes	0.21	1.23	0.93–1.63	0.14
No	Reference
Tobacco current
Yes	0.14	1.14	0.63–2.09	0.66
No	Reference
Tobacco past
Yes	−0.20	0.82	0.54–1.23	0.33
No	Reference
Hyperlipidaemia
Yes	0.13	1.14	0.92–1.42	0.24
No	Reference
Alcohol current
Yes	−0.09	0.91	0.68–1.23	0.55
No	Reference
Alcohol past
Yes	0.45	1.57	1.19–2.06	0.001
No	Reference

CT: computed tomography; MRI: magnetic resonance imaging.

Dependent variable is coded as rural: 1; urban: 0.

Includes CVT cases.

CT, MRI and CT/MRI both grouped as imaging.

Discussion

In this large population-based stroke registry, rural and urban differences were observed in various stroke characteristics. There was a high proportion of haemorrhagic stroke in rural patients. In multivariable analysis, the rural patients were younger, belonged to Sikh religion, were farmers and house wives. Among the risk factors only alcohol consumption was significant in the rural cohort.

The average age of stroke patients in this cohort is younger which is similar to other studies.^7,9 However, the difference in age (younger) among rural patients has not been observed in other population-based studies in India. In the reports from France, Northern Portugal and Morocco, the rural patients were older than urban patients.^3,4,15 The change in lifestyle among the rural cohort especially farmers¹⁶ and high prevalence of vascular risk factors could be the possible reasons. Agriculture is the principal industry in Punjab and the diet is rich in fats.¹⁶ Over the years, the manual hard labour of the farmer has been replaced by the cheap labour of the migrant labourers from central India but the diet of the Punjab farmer has still remained the same.¹⁷

There was a high proportion of haemorrhagic stroke seen in rural patients. Similar studies from France, China and Trivandrum showed higher rates of haemorrhagic strokes in rural regions.^3,9,18 However, in Portugal, haemorrhagic stroke was higher in urban population.⁴ Hypertension was the main risk factor in both urban and rural patients. The difference observed in univariate analysis was no longer present in multivariable analysis. Other factors such as poor control of blood pressure and monitoring could be related to high proportion of haemorrhagic stroke. In low-income countries the awareness and treatment of hypertension in rural areas is low as compared to urban population.¹⁹ Raising community awareness about hypertension can reduce its prevalence and improve its treatment. Hyperlipidaemia was higher in rural patients; however, it was not significant in multivariable analysis.

In our study, the rate of CVT was very low. In India there has been a regional difference in CVT prevalence, it is more common in South India. Most of the studies published from India are hospital-based reports.^20–22 Hence the current burden of CVT from India may not be accurate since the majority is hospital-based information. Punjab has better antenatal care and free hospital deliveries offered by the government with free ambulance services. This could be the reason for a lower incidence of CVT in this cohort.

The use of tobacco was higher in urban areas whereas alcohol consumption is higher in rural areas. Sikhism is the main religion of Punjab (58% of total population of Punjab) and more number of Sikhs resides in the rural area (77% of total Sikh population) whereas in the urban areas people from all religions reside.²³ The consumption of alcohol and tobacco is forbidden in Sikh religion. However, Sikhs (who are not baptised Sikh) consume some alcohol in social functions but tobacco intake is strictly prohibited in this religion.²⁴ From recent years, the rate of migration to Punjab state from other states like Uttar Pradesh and Bihar is very high and these migrants live in cities.^17,25 The people from these states have higher rate of tobacco consumption than other states.²⁶ Moreover, the tobacco and alcohol consumption would have been under-reported due to religious reasons.

In both the groups, there were high proportion of patients who underwent imaging especially so in rural areas. The patients from villages are usually referred by their primary health centres to the city to get the diagnosis confirmed. In the population-based studies done in France and Portugal, imaging rates were similar in urban and rural patients.^3,4

The strength of the study is that it is a population-based data with large number of patients from both urban and rural areas. Ludhiana city is the hub of health care facilities in Punjab state. All major hospitals and scan centres participated in the study so chances of missing non-fatal stroke are very low.

The rural stroke patients may not be representative of Punjab population since only patients who came to Ludhiana were included. The other cities of Punjab cater to their own urban patient population as well and also the adjoining rural populations. We accept dissimilarities in the quality of data as a limitation of the study; however, similar rural and urban differences are being observed in our ongoing Ludhiana rural population-based stroke registry. Lack of information on lifestyle risk factors like physical activity, obesity, diet habits of rural and urban populations is another important limitation of the study.

Conclusion

In this large cohort of patients rural and urban differences were seen in terms of risk factors and type of stroke. In addition, religion and cultural influences were also noted in the prevalence of risk factors in this patient population. India is a vast country with diverse sociocultural practices across regions. Stroke prevention strategies need to take into consideration the rural and urban differences and the regional sociocultural practices.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Indian Council of Medical Research, Task Force Project, New Delhi (SWG/22/Neuro/2008-NCD-I).

Ethical approval

The ethics committee of Christian Medical College and Hospital, Ludhiana, and Dayanand Medical College and Hospital, Ludhiana, approved this study and all other centres granted permission.

Informed consent

Written informed consent was obtained from the patient(s) for their anonymised information to be published in this article.

Guarantor

JDP.

Contributorship

PK: Data analysis, data interpretation and writing; SJV: Data collection, data interpretation and writing; GS: Study design, data acquisition and writing; RB, BSP, and MS: Study design and data acquisition; SS and CJS, Literature search, study design, data interpretation and writing; JDP: Literature search, study design, data collection, data interpretation and writing.

Collaborators

Om P Arora, Arun K Dhanuka, Manoj K Sobti, Harish Sehgal, Mohanjeet Kaur, Sarvpreet S Grewal, Sukhdeep S Jhawar, TN Shadangi, Tushar Arora, Ashish Saxena, Gaurav Sachdeva, Jeetamol S Gill, Ramandeep S Brar, Anakhvir Gill, Sandeep S Bakshi, Sandeep S Pawar, Gurmeet Singh, Praveen Sikka.

Acknowledgements

The authors thank Madhu Bala for managing the database; Premjeeth Moodbidri, Gagandeep Mehmi, Amber Sharma, Rohit and Manpreet Kaur for helping with data collection; the following general practitioners for sharing data: Nitin Sood, Bhushan Bansal, BL Malhotra, RS Bhatia, Punit Midha, Rahul Jain, and Subhash Sachdeva; and members of The Indian Council of Medical Research (ICMR) Task Force Group: M Gourie Devi: Department of Neurology, Institute of Human Behaviour and Allied Sciences (IHBAS), Sri Ganga Ram Hospital, New Delhi; Meenakshi Sharma A: Non-Communicable Diseases Division, Indian Council of Medical Research, New Delhi; Nanda Kumar: National Center for Disease Informatics and Research, Bangalore; Kameshwar Prasad: Department of Neurology, Neurosciences Center, All India Institute of Medical Sciences, New Delhi; P Satish Chandra: National Institute of Mental Health and Neurosciences, Bangalore; K Radhakrishnan: Kasturba Hospital, Manipal, Karnataka; KR Thankappan: Achutha Menon Center for Health Sciences Studies, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Thiruvananthapuram, Kerala.

References

Feigin

Krishnamurthy

Mensah

et al.

Global and regional burden for stroke during 1990–2010: finding from the global burden of disease study 2010. Lancet 2014; 383: 245–254.

Adams

Hendershot

Marano

. Current estimates from the national health interview survey, 1996. Vital Health Stat 1999; 10: 1–203.

Lemesle

Giroud

Menassa

et al.

Incidence and case-fatality rates of stroke in Burgundy (France). Comparison between a rural (Avallon) and an urban (Dijon) population, between 1989 and 1993. Eur J Neurol 1996; 3: 109–115.

Correia

Silva

Matos

et al.

Prospective community-based study of stroke in Northern Portugal: Incidence and case fatality in rural and urban populations. Stroke 2004; 35: 2048–2053.

Bermejo

Vega

Morales

et al.

Prevalence of stroke in two samples (rural and urban) of old people in Spain. A pilot door-to-door study carried out by health professionals. Neurologia 1997; 12: 157–161.

Kalkonde

Deshmukh

Sahane

et al.

Stroke is the leading cause of death in rural Gadchiroli, India: a prospective community- based study. Stroke 2015; 46: 1764–1768.

Das

Banerjee

Biswas

et al.

A prospective community-based study of stroke in Kolkata, India. Stroke 2007; 38: 906–910.

Bhattacharya

Saha

Basu

Das

. A 5-year prospective study of incidence, morbidity and mortality profile of stroke in a rural community of Eastern India. J Indian Med Assoc 2005; 103: 655–659.

Sridharan

Unnikrishnan

Sukumaran

et al.

Incidence, types, risk factors, and outcome of stroke in a developing country: the Trivandrum stroke registry. Stroke 2009; 40: 1212–1218.

10.

http://eci.nic.in/eci_main1/SVEEP/statesweepplan2014/PUNJAB.pdf (accessed 1 February 2017).

11.

http://www.censusindia.gov.in/ (accessed 15 February 2017).

12.

http://censusindia.gov.in/ (accessed 9 April 2017).

13.

Pandian

Singh

Bansal

et al.

Establishment of population based stroke registry in Ludhiana city, Northwest India: feasibility and methodology. Neuroepidemiology 2015; 44: 69–77.

14.

WHO STEPS Stroke Manual Version 1.2. The WHO STEP wise approach to stroke surveillance. http://www.who.int/ncd_surveillance/en/steps_stroke_manual_v1.2.pdf (accessed 2 July 2010).

15.

Engels

Baglione

Audibert

et al.

Socioeconomic status and stroke prevalence in Morocco: results from the Rabat-Casablanca study. PLoS One 2014; 9: e89271.

16.

Mangat

Grewal

Kaur

et al.

Dietary patterns in stroke patients in Northwest India. Nutr Neuro sci 2013; 16: 288–292.

17.

http://www.pbplanning.gov.in/HDR/studyOfMigrants_full.pdf (accessed 1 February 2017).

18.

Mi1

Sun

et al.

Differences in the distribution of risk factors for stroke among the high-risk population in urban and rural areas of Eastern China. Brain Behav 2016; 6: e00461.

19.

Chow

Teo

Rangarajan

et al.

Prevalence, awareness, treatment, and control of hypertension in rural and urban communities in high-, middle-, and low-income countries. JAMA 2013; 310: 959–968.

20.

Srinivas G. Clinical profile, natural history and factors determining prognosis/outcome study of cerebral venous sinus thrombosis. DM Thesis, http://dspace.sctimst.ac.in/jspui/bitstream/123456789/1956/1/546.pdf (accessed 21 April 2017).

21.

Narayan

Kaul

Ravishankar

et al.

Risk factors, clinical profile, and long-term outcome of 428 patients of cerebral sinus venous thrombosis: Insights from Nizam’s Institute Venous Stroke Registry, Hyderabad (India). Neurol India 2012; 60: 154–159.

22.

Rajoor

Seema

. Cerebral venous thrombosis in women: a study from teaching hospital in North Karnataka. Int J Sci Study 2015; 3: 179–182.

23.

Population by religious community: Punjab 2011 Census of India, www.censusindia.gov.in/2011census/C-01/DDW03C-01%20MDDS.XLS (accessed 1 February 2017).

24.

Morgan

Lawton

. Ethical issues in six religious traditions, Edinburgh: Edinburgh University Press, 2007, pp. 138–139.

25.

Singh

Kaur

. Causes and consequences of migrant labour in Ludhiana city: a case study. J Res 2006; 43: 144–147.

26.

Rani

Bonu

Jha

et al.

Tobacco use in India: prevalence and predictors of smoking and chewing in a national cross sectional household survey. Tob Control 2003; 12: e4.

Stroke profile and outcome between urban and rural regions of Northwest India: Data from Ludhiana population-based stroke registry

Abstract

Introduction

Methods

Results

Discussion and Conclusion

Keywords

Introduction

Methods

Background of the Ludhiana population-based stroke registry

Duplication of the cases

Selection of dataset for this study

Statistical analysis

Results

Comparison of demographic details between two groups

Comparison of type of stroke, imaging modality and outcome

Comparison of risk factors between two groups

Multivariable analysis

Discussion

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

Ethical approval

Informed consent

Guarantor

Contributorship

Collaborators

Acknowledgements

References