[Downloaded free from http://www.aihbonline.com on Wednesday, March 1, 2023, IP: 251.75.82.

75]

Original Article

Comparison of Perceived Sleep Quality among Urban and Rural

Adult Population by Bengali Pittsburgh Sleep Quality Index
Himel Mondal, Shaikat Mondal1, Chayan Baidya2
Department of Physiology, MKCG Medical College, Ganjam, Odisha, 1Department of Physiology, Medical College and Hospital, 2JB Roy State Ayurvedic Medical
College, Kolkata, West Bengal, India

Abstract
Background: Pittsburgh Sleep Quality Index (PSQI) assesses perceived quality of sleep. Bengali is the 6th language in the world according to
the number of first language speakers. PSQI is presently not available in Bengali. Poor quality of sleep affects work efficiency and health of
individual and it is increasing in urban as well as in rural population. Aim: The aim of the study was (1) to adapt PSQI in Bengali Language and
(2) to find the prevalence rate of poor quality of sleep among urban and rural populations and to compare the rate. Materials and Methods: First,
Bengali PSQI (BPSQI) was adapted by linguistic validation methods. Then, a cross‑sectional survey was conducted among sample in a municipal
ward and in a village by BPSQI. The prevalence of poor sleep quality among the sample was ascertained. Data were presented in percentage,
mean and standard deviation. Chi‑square test and unpaired t‑test were used according to necessity with α = 0.05. Results: Adapted BPSQI
instrument was found of acceptable internal consistency (Cronbach’s α = 0.816). The prevalence rate of poor sleep quality in urban adult
population was 42.58% and rural population was 35.89% (χ2 = 4.004, P = 0.0454). Adult females in urban area showed more prevalence rate of
poor sleep quality (58.74%) than those of adult females in rural area (45.96%). Conclusion: Adapted BPSQI can be used as a self‑administered
questionnaire among Bengali native speakers. A significant percentage of urban and rural adult population suffers from poor quality of sleep.
Adult population in urban area, especially adult females, suffers more from poor quality of sleep than rural population.

Keywords: Bengali language, insomnia, linguistic validation, subjective sleep quality, urbanisation

Introduction It is well documented that urbanisation increases stress level

Pittsburgh Sleep Quality Index (PSQI) is an instrument
developed by Buysse et al. in 1989 to assess the quality of sleep
in previous 1‑month period.[1] This instrument is copyrighted
by the University of Pittsburgh. However, it can be reprinted
for educational and in non‑commercial research. [2] This
instrument, originally available in English, has been translated
into 56 languages.[3]
According to the number of first language speakers, Bengali
is the 6th language in the world.[4] It is the national language
of Bangladesh. It is among the 23 official languages
of India, with 8.11% population with Bengali as their
mother tongue which stands second in descending order of
speaker’s strength (2001).[5] Maximum native speakers live
in West Bengal, Tripura, Assam and Andaman and Nicobar
Islands.[6] PSQI is available in Indian languages such as Hindi
and Marathi.[7] However, it is currently not available in Bengali
language.
Access this article online
Quick Response Code:
Website:
www.aihbonline.com
and affects sleep quality.[8] In addition, a study by Ravikiran
et al. found that a significant percentage of rural children also
suffer from sleep problems.[9] A study by Tang et al. showed
that more than one–fourth of populations suffers from insomnia
in China and there was significantly more prevalence in
rural population than urban population.[10] To the best of our
knowledge, no study has been carried out to compare subjective
sleep quality among urban and rural adult population in the
North 24 Parganas District, West Bengal, India.
With this background, the aim of this study was to first adapt
Bengali version of PSQI (BPSQI) through extensive language
validation methods and to compare perceived sleep quality
among urban and rural populations.
Address for correspondence: Dr. Himel Mondal,
Department of Physiology, MKCG Medical College, Ganjam, Odisha, India.
E‑mail: himelmkcg@gmail.com
This is an open access article distributed under the terms of the Creative Commons
Attribution‑NonCommercial‑ShareAlike 3.0 License, which allows others to remix, tweak,
and build upon the work non‑commercially, as long as the author is credited and the
new creations are licensed under the identical terms.
For reprints contact: reprints@medknow.com

DOI: How to cite this article: Mondal H, Mondal S, Baidya C. Comparison of

10.4103/AIHB.AIHB_44_17 perceived sleep quality among urban and rural adult population by Bengali
Pittsburgh Sleep Quality Index. Adv Hum Biol 2018;8:36-40.

36 © 2018 Advances in Human Biology | Published by Wolters Kluwer - Medknow

[Downloaded free from http://www.aihbonline.com on Wednesday, March 1, 2023, IP: 251.75.82.75]
Mondal, et al.: Sleep quality among urban and rural populations
Materials and Methods
This study was a survey‑based study. All the experts and
participants for the study were taken from adult age group
(i.e., age >18 years). The study was conducted according to
guidelines by ‘WMA declaration of Helsinki’ after obtaining
informed written consent from the experts and participants. The
study was divided into two phases. The first Phase was forward
translation and back translation and drafting of the final scale
which was carried out during the period of October 2016 to
December 2016. The second phase was assessment of sleep
quality among population in an urban area and in a rural area
situated in North 24 Parganas district. The second phase of
the study was carried out during January 2017 to April 2017.
Linguistic validation of Bengali version of Pittsburgh Sleep
Quality Index
The original version of PSQI was developed in English.
The scale has ten items.[1] Among the items, the tenth item
records responses of roommate or bed partner. This item is
not considered for global PSQI scoring.[2,11] Hence, we omitted
the 10th item from the scale. Several previous studies were
conducted with 9‑item PSQI scale.[12‑14] As the scale was used
without the 10th item of original PSQI instrument available
in Pittsburgh website,[2] the PSQI used in this study also be
called as ‘modified PSQI’ and Bengali version may also be
called ‘modified BPSQI’.
Linguistic validation was carried out according to guidelines
by the WHO with some additional measures.[15] A project
coordinator was appointed by the authors who was Bengali,
bilingual and had a master’s degree in English. The aim
and objectives of the study was discussed in detail with the
coordinator. Special discussion for expert panel selection
was emphasised. Expert panel for forward translation from
English to Bengali comprised of four personnel (a psychologist
with interviewing expertise, an English teacher with
Master’s/Doctoral degree with expertise in translation, a
Bengali teacher with Master’s/Doctoral degree with expertise in
translation and a physiologist with experience in questionnaire
development). Mother tongue of those experts was Bengali.
Figure 1: Flow chart depicting procedure of adaptation of English version of Pittsburgh Sleep Quality Index in Bengali language.
Advances in Human Biology ¦ Volume 8 ¦ Issue 1 ¦ January‑April 2018
Printed English version of modified PSQI was distributed
among the experts via the coordinator. All of them translated
the questionnaire from English to Bengali with emphasis
on conceptual translation rather than linguistic translation.
After receiving the translated questionnaires, a meeting
was conducted by the coordinator. After discussion among
the experts and coordinator, necessary addition, deletion
and modification were done in the questionnaire and a final
version was drafted by the coordinator. This questionnaire was
received from the coordinator. It was reviewed by the authors
and preserved.
Similar but another set of experts were selected by the project
coordinator. However, the medium of study of this group was
English during their schooling. They were given the newly
translated Bengali version questionnaire via the coordinator
to translate it into English. After translation of questionnaire
from Bengali to English by individual expert, a meeting was
conducted among the experts. After discussion among them,
a final draft of English‑modified PSQI was obtained by the
coordinator. This version was compared with the original
instrument. The content was found to be equivalent to the
original instrument, though there was difference in words and
phrases. Hence, no further forward or back translation was
carried out and the BPSQI was accepted as the final version
of the questionnaire. The whole procedure of translation is
depicted in Figure 1.
Pilot survey and cognitive interview
Literature about subjective sleep quality among urban and rural
populations was not available for the targeted state of study,
as well as for India. Hence, a pilot study was carried out to
ascertain prevalence rate. This pilot study was also used as
cognitive interview for the newly adapted instrument. List of
adult population was obtained from the voter list of ward no. 8
in Barasat municipality area (i.e., urban area) and Panchita
village in Dharampukuria Gram Panchayat (i.e., rural area).
Available voter list (updated 1st January 2017) was printed
from the Election commission website for convenience.[16] A
systemic random sample (n = 30) from each area was selected.
37
[Downloaded free from http://www.aihbonline.com on Wednesday, March 1, 2023, IP: 251.75.82.75]

Mondal, et al.: Sleep quality among urban and rural populations

Survey in urban area was conducted by the second author
and rural area by the third author. After obtaining written
informed consent, the BPSQI was given to participants to fill
up. The participants filled the self‑administered questionnaire
in front of the author. Among the sample, three participants
were illiterate in rural area and for them data were collected
by an interview. After getting filled‑up questionnaire in case
of literate individuals, participants were asked questions
about their understanding and difficulty in understanding
about the questionnaire. These questions were pre‑planned to
gather information about participant’s perception about the
questionnaire, to point out difficult word or phrase and to find
any cultural issue. The filled‑up questionnaires were coded and
scored. From the pilot study, prevalence of poor sleep quality
was found to be 37.8% in urban population and 28.2% in rural
population. According to the experience of the pilot study, two
phrases were changed after discussion among authors. Thus,
a final draft of BPSQI was obtained.
calculated in Microsoft Excel® manually according to formula
by Machin et al.[18] Mean and standard deviation was computed
for score of the seven components (viz., subjective sleep
quality, sleep latency, sleep duration, habitual sleep efficiency,
sleep disturbances, use of sleeping medication and daytime
dysfunction). Mean of the two groups was compared by
unpaired t‑test with α = 0.05. Global score of each group was
also expressed in mean and standard deviation and compared
by unpaired t‑test with similar chances of type I error. Global
score >5 was considered as an indicator of poor sleep quality.
Prevalence rate of poor sleep quality was calculated for both
groups and tested statistically by Chi‑square test with α = 0.05.
Results
The reliability of newly adapted instrument (i.e., BPSQI),
tested by Cronbach’s alpha (α =0.816), was in the acceptable
range (i.e., 0.7–0.95).[19]

Sampling technique In urban area, the survey was completed on 411 (88.20%
With a prevalence rate of 37.8% in urban population and 28.2% of targeted sample) participants (male = 188 [45.74%],
in rural population, the sample size was calculated according female = 223 [54.26%]). Among the participants, in urban
to the following formula:[17] area, 344 (83.70%) filled the questionnaire themselves
(i.e., self‑administered) and 67 (16.30%) were self‑reported with
n = (Zα/2 + Zβ)2* (p1 [1 − p1] + p2 [1 − p2])/(p1 − p2)2 the help of interviewer. The mean age of respondents in
The confidence level was set at 95% and power of the study was urban area was 38.56 ± 13.88 years. Age‑, sex‑ and marital
set at 80%. Calculated minimum sample size was 373 in each status‑wise distribution of participants in urban area according
group. With assumption of unavailability of some participants to BPSQI global score is shown in Table 1. The prevalence
during the survey period and unwillingness in participation, the of poor quality sleep (i.e., Global score >5) among urban
sample size was multiplied with 1.25. Hence, the final sample population was 42.58%.
size was calculated as 466 for each group. Those participants
In rural area, survey was completed on 443 (95.06%
who were surveyed in pilot study were marked and excluded
of targeted sample) participants (male = 208 [46.95%],
from the voter list. Systemic random sampling was used to
female = 235 [53.05%]). In rural area, 252 (56.88%) filled
select and mark the participant on the available voter list.
This marked voter list was used during the survey for easy
identification of the targeted sample. Table 1: Age‑, sex‑ and marital status‑wise distribution
of participants in urban area (n=411) and their Bengali
Survey proper
Pittsburgh Sleep Quality Index global score
The survey was conducted during the month of January through
April, 2017. The survey in rural area was conducted partly BPSQI global BPSQI global χ2, P
by the third author and partly by two experienced surveyors. score <5 (%) score ≥5 (%)
In urban area, the whole survey was conducted by three Age (years)
experienced surveyors. Participants were briefed about the <30 93 (70.45) 39 (29.55) 33.97, <0.0001*
31-40 77 (63.64) 44 (36.36)
aim of the survey, and written informed consent was taken.
41-50 46 (50.55) 45 (49.45)
Demographic details were recorded in one form by interviewing
51-60 13 (32.5) 27 (67.5)
the participant. For literate person, and willingness to fill up,
61-70 4 (25) 12 (75)
the BPSQI was self‑administered, and for illiterate person, the
>70 3 (27.27) 8 (72.73)
BPSQI was self‑reported with the help of interviewer. During Sex
the survey period, last 10 days were allotted for mop‑up round Male 144 (76.6) 44 (23.4) 52.11, <0.0001*
to get the response from those who were previously absent Female 92 (41.26) 131 (58.74)
during survey. Data obtained were scored and coded in spread Marital status
sheet and stored for analysis. Unmarried 43 (69.35) 19 (30.65) 9.48, 0.0087*
Married 165 (55.37) 133 (44.63)
Statistical analysis
Divorced/ 14 (37.84) 23 (62.16)
Statistical analysis was carried out in Microsoft Excel® 2010. widow
Seven components of PSQI scoring were taken for internal *Two‑tailed P<0.05 was considered statistically significant for Chi‑square
consistency test after pilot study by Cronbach’s α. It was test. BPSQI: Bengali Pittsburgh Sleep Quality Index

38 Advances in Human Biology ¦ Volume 8 ¦ Issue 1 ¦ January‑April 2018

[Downloaded free from http://www.aihbonline.com on Wednesday, March 1, 2023, IP: 251.75.82.75]

Mondal, et al.: Sleep quality among urban and rural populations

the questionnaire themselves and 191 (43.12%) were

Table 2: Age‑, sex‑ and marital status‑wise distribution
self‑reported with the help of interviewer. The mean age of
of participants in rural area (n=443) and their Bengali
respondents in the rural area was 37.98 ± 13.14 years. Age‑,
Pittsburgh Sleep Quality Index global score
sex‑ and marital status‑wise distribution of participants in rural
area according to BPSQI global score is shown in Table 2. BPSQI global BPSQI global χ2, P
The prevalence of poor quality sleep in rural population score <5 (%) score ≥5 (%)
was 35.89%. Age (years)
<30 112 (76.71) 34 (23.29) 23.87, 0.0002*
Comparison of prevalence of poor sleep quality among 31-40 94 (63.95) 53 (36.05)
respondents in urban population and rural population is shown 41-50 45 (58.44) 32 (41.56)
in Table 3. 51-60 23 (50) 23 (50)
61-70 7 (35) 13 (65)
The mean score for the seven components of BPSQI and global
>70 3 (42.86) 4 (57.14)
score among urban and rural populations are shown in Table 4.
Sex
Male 157 (75.48) 51 (24.52) 22.04, <0.0001*
Discussion Female 127 (54.04) 108 (45.96)
The outcome of this study is a language‑validated BPSQI Marital status
survey instrument with an accepted level of internal Unmarried 58 (78.38) 16 (21.62) 13.14, 0.0014*
consistency. Majority of the participants (69.78%) filled Married 207 (63.3) 120 (36.7)
up the questionnaire themselves. This instrument can be Divorced/ 19 (45.24) 23 (54.76)
widow/
used in further studies to assess sleep quality in population widower
whose native language is Bengali. This would help *Two‑tailed P<0.05 was considered statistically significant for Chi‑square
psychiatrists, psychologists or clinicians who would like test. BPSQI: Bengali Pittsburgh Sleep Quality Index
to check patient’s perceived sleep quality. The soft copy of
BPSQI can be obtained via E‑mail from the corresponding
author. However, the limitations of the original instrument Table 3: Comparison of sleep quality among respondents
are also present in this version also. The questionnaire of urban and rural areas
can only measure the sleep quality for previous 1‑month
Population Perceived sleep quality χ2, P
duration only. In addition, it is the perceived sleep quality
of the individual, which may not be corroborative with Poor (%) Not poor (%)
actual sleep quality. Urban (n=411) 175 (42.58) 236 (57.42) 4.004, 0.0454*
Rural (n=443) 159 (35.89) 284 (64.11)
Prevalence of poor sleep quality found in urban population *Two‑tailed P<0.05 was considered statistically significant. Perceived
(42.58%) from this study was more than the prevalence of sleep quality: Poor=global score for BPSQI ≥5. Perceived sleep quality:
insomnia (15.4%) found by Roy et al. in an urban area in the Not poor=global score for BPSQI <5. BPSQI: Bengali Pittsburgh Sleep
Quality Index
same district.[20] The survey instrument used by the author
was different from that of the instrument used in this study.
However, this finding points towards the fact that there are
diverse causes for poor quality of sleep and insomnia is one Table 4: Bengali Pittsburgh Sleep Quality Index
of them.[21] From the current study, it was ascertained that scores for individual components expressed in
the prevalence rate of poor sleep quality was more in urban mean±standard deviation among respondents of urban
population [Table 3]. This result is not concordant with the and rural areas
finding by Tang et al.,[10] who found an increased prevalence Respondents Respondents t‑test P
rate of poor sleep quality among rural Chinese population. of urban of rural area
Probable reason behind this discrepancy may be cultural area (n=443)
difference between the populations of these countries. (n=411)
Components of BPSQI
Adult male population has almost similar level of poor Subjective sleep quality 0.80±0.78 0.73±0.62 0.1455
sleep quality (23.4%) as that of rural adult males (24.52%). Sleep latency 0.49±0.64 0.41±0.69 0.08
However, adult female participants in urban area have a higher Sleep duration 0.78±0.96 0.98±0.76 0.0007*
prevalence of poor sleep quality (58.74%) than that of rural Habitual sleep efficiency 0.63±0.73 0.68±0.67 0.2969
adult females (45.96%). The prevalence rate of poor sleep Sleep disturbances 0.90±0.99 0.89±0.99 0.8828
quality is found to be higher in old age. Presence of geriatric Use of sleeping 0.98±0.96 0.82±0.90 0.0121*
diseases may be the contributing factors for this finding which medication
was beyond the scope of this study. Daytime dysfunction 0.46±0.60 0.45±0.50 0.7909
Global BPSQI score 5.05±3.11 4.95±3.77 0.6738
From individual score of BPSQI, it was evident that there is a *Two‑tailed P<0.05 was considered statistically significant.
significant difference in sleep duration among rural and urban BPSQI: Bengali Pittsburgh Sleep Quality Index

Advances in Human Biology ¦ Volume 8 ¦ Issue 1 ¦ January‑April 2018 39

[Downloaded free from http://www.aihbonline.com on Wednesday, March 1, 2023, IP: 251.75.82.75]
Mondal, et al.: Sleep quality among urban and rural populations
populations [Table 4]. Rural population showed less sleep
duration. The reason behind it may be the lifestyle difference
among rural and urban populations. Despite this finding, use
of sleep medication was more common in urban population.
Easy availability of healthcare facility, level of awareness and
more stress in urban life may be contributing factors.
Limitations
This study has some limitations. Only perceived sleep quality
was tested, sleep quality was not tested by polysomnograph.
Samples from a single municipal ward and from a single village
in rural area were surveyed. Further study with larger sample
size would reflect more generalised result.
Conclusion
PSQI was adapted in Bengali language with extensive
linguistic validation. Internal consistency of the instrument
was satisfactory. The newly adapted instrument is feasible
to administer in Bengali native speakers in urban and rural
populations. Urban population has more prevalence rate of
poor sleep quality than that of rural population. Adult females,
in both urban and rural areas, suffer more from poor quality
sleep than adult males.
Acknowledgement
We thank the coordinator Sarika Mondal, M. A. (English)
and expert panel members for their immense help during the
translation of the instrument. We are thankful to Dr. Rajesh De,
Department of Community Medicine, R. G. Kar Medical
College, Kolkata, West Bengal, for his help during initial
design of the study. We acknowledge the help of the surveyors
who conducted the survey.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
1. Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ. The
Pittsburgh sleep quality index: A new instrument for psychiatric practice
and research. Psychiatry Res 1989;28:193‑213.
2. PSQI. Department of Psychiatry, University of Pittsburgh. Available
from: http://www.psychiatry.pitt.edu/node/8240. [Last accessed on
2017 Mar 18].
3. Pittsburgh Sleep Quality Index (PSQI). MAPI Research Trust: Lyon,
40
French. Available from: https://www.eprovide.mapi‑trust.org/instruments/
pittsburgh‑sleep‑quality‑index. [Last accessed on 2017 Mar 18].
4. Simons GF, Fennig CD. Summary by language size. In: Ethnologue:
Languages of the World. 12th ed. Dallas, Texas: SIL International;
2017. Available from: https://www.ethnologue.com/statistics/size. [Last
accessed on 2017 Mar 19].
5. Scheduled Languages in Descending Order of Speaker’s Strength – 2001.
Office of the Registrar General & Census Commissioner. India.
Available from: http://www.censusindia.gov.in/Census_Data_2001/
Census_Data_Online/Language/Statement4.aspx. [Last accessed on
2017 Mar 19].
6. Conners TJ, Chacon DA, editors. The Bangla language. In: Anne Boyle
David Descriptive Grammar of Bangla. Germany: De Gruyter Mouton;
2015. p. 5‑6.
7. PSQI Translation List. Sleep & Chronobiology Center, Department
of Psychiatry, University of Pittsburgh. Available from: http://www.
sleep.pitt.edu/research/ewExternalFiles/PSQI%20Language%20
Translation%20List.pdf. [Last accessed on 2017 Mar 19].
8. Srivastava K. Urbanization and mental health. Ind Psychiatry J
2009;18:75‑6.
9. Ravikiran SR, Kumar PM, Latha KS. Sleep problems in preschool and
school aged rural Indian children. Indian Pediatr 2011;48:221‑3.
10. Tang J, Liao Y, Kelly BC, Xie L, Xiang YT, Qi C, et al. Gender and regional
differences in sleep quality and insomnia: A General population‑based
study in Hunan Province of China. Sci Rep 2017;7:43690.
11. Pittsburgh Sleep Quality Index (PSQI). Oakland Psychiatric Association.
Available from: http://www.opapc.com/uploads/documents/PSQI.pdf.
[Last accessed on 2017 Mar 19].
12. Kaur G, Sharma V, Singh A. Association of sleep quality with general
health: An Indian college students study. Int J Med Sci Public Health
2015;4:1767‑71.
13. Kumar A, Vandana, Aslami AN. Analgesics self‑medication and its
association with sleep quality among medical undergraduates. J Clin
Diagn Res 2016;10:FC07‑11.
14. De S. Subjective assessment of quality of sleep in chronic obstructive
pulmonary disease patient and its relationship with associated
depression. Lung India 2012;29:332‑5.
15. Process of Translation and Adaptation of Instruments. WHO: Geneva.
Available from: http://www.who.int/substance_abuse/research_tools/
translation/en/. [Last accessed on 2017 Mar 19].
16. Chief Electoral Officer. West Bengal. Available from:
h t t p : / / w w w. c e o w e s t b e n g a l . n i c . i n / A C N a m e L i s t .
a s p x ? D C I D = 9 & D C N a m e = N O RT H % 2 0 2 4 % 2 0 PA R G A N A S .
[Last accessed on 2017 Jan 10].
17. Peacock JL, Peacock PJ. Research design. In: Oxford Handbook of
Medical Statistics. New York, USA: Oxford University Press; 2011.
p. 66‑7.
18. Machin D, Campbell MJ, Walters SJ. Reliability and method comparison
studies. In: Medical Statistics: A Textbook for the Health Sciences.
4th ed. England: John Wiley & Sons Ltd.; 2007. p. 209.
19. Tavakol M, Dennick R. Making sense of Cronbach’s alpha. Int J Med
Educ 2011;2:53‑5.
20. Roy SK, Bhattacharjee AK, Chakraborti C, Singh R. Prevalence of
insomnia in urban population of West Bengal: A community‑based
cross‑sectional study. Int J Med Public Health 2015;5:293‑6.
21. Shah N, Bang A, Bhagat A. Indian research on sleep disorders. Indian J
Psychiatry 2010;52:S255‑9.
Advances in Human Biology ¦ Volume 8 ¦ Issue 1 ¦ January‑April 2018