Preliminary evaluation of a Bangladesh-adapted Kuppuswamy socioeconomic status scale in rural hypertensive adults

Authors

DOI:

Keywords

socioeconomic status, Kuppuswamy scale, formative index, hypertension, Bangladesh

Correspondence

Mohammad Tanvir Islam
Email: tanvir@bsmmu.edu.bd

Publication history

Received: 27 June 2026
Accepted: 15 Aug 2026
Published online: 15 Sep 2026

Responsible editor

Reviewers

B: Anonymous
E: Anonymous

Funding

None

Ethical approval

Approved by the Institutional Scientific Review Board of Bangladesh Medical University (No. BSMMU/2023/6516, dated 29 Apr 2023).

Trial registration number

Not applicable

Copyright

© The Author(s) 2026; all rights reserved. 
Published by Bangladesh Medical University (former Bangabandhu Sheikh Mujib Medical University).
Abstract

Background: A Bangladesh-adapted Kuppuswamy socioeconomic status (SES) scale was developed by modifying the education, occupation, and income domains of the original instrument but had not undergone empirical evaluation. We aimed to examine its structural coherence (intercorrelation of the three components), the agreement between fixed and empirically derived component weights, convergent association with respondents’ own education, exploratory associations with health-related outcomes, and the consequences of collapsing the five SES categories into three.


Methods: Secondary analysis of 461 hypertensive adults from a single rural sub-district of Bangladesh. Component intercorrelation was described with Spearman correlations and Factor Analysis using the Principal Component extraction method; agreement between fixed and PCA-derived weights was assessed; convergent association was examined against respondents’ own education; associations with medication adherence and blood pressure were examined across three SES categories as exploratory health-related associations; and category-collapse was checked with linear contrasts and a Jonckheere–Terpstra test.


Results: The three components were positively but modestly intercorrelated (Spearman rho 0.28–0.39). Fixed and PCA-derived weights ranked participants near-identically (rho=0.989). The composite correlated weakly, and only partly independently, with respondents’ education (rho=0.253; 0.274 after removing the education overlap). Medication adherence rose across SES categories (η2=0.142), driven mainly by the lowest group; blood pressure did not differ (P=0.08 and 0.23). The true upper SES range was essentially untested (n=2).


Conclusion: In this single rural, hypertensive sample the adapted scale showed positive intercorrelation among its components, weak and only partly independent convergent association with education, and a modest association with adherence at the lower socioeconomic range. These findings are promising but preliminary; further evaluation across the full socioeconomic range, in urban as well as rural settings, and against an independent wealth criterion is needed before routine use.

Key messages

Bangladesh lacks a widely accepted, standardized Kuppuswamy socioeconomic status measure, forcing inconsistent ad hoc approaches. This work provides a locally adapted Kuppuswamy instrument with a ready-to-use scoring tool and first empirical evidence on its behaviour in a rural sample. Further evaluation against an independent wealth index is needed before routine use.

Introduction

Socioeconomic status (SES) is a well-established determinant of health behaviour and outcomes, including hypertension awareness, treatment, and control [1]. In South Asia, the Kuppuswamy socioeconomic status Index, developed by B. Kuppuswamy in 1976 for urban India, remains one of the most widely used composite SES instruments, combining education, occupation of the head of household and total monthly household income into an ordinal classification [2]. Because its income component uses absolute currency thresholds, it requires periodic recalibration as prices rise; successive Indian revisions have updated the income brackets to keep pace with inflation [3–5].


To our knowledge, no widely accepted, nationally standardized SES classification has been established for Bangladesh; researchers have relied on ad hoc asset indices or simplified income cut-offs that limit comparability [6]. The original or Indian-revised Kuppuswamy scale cannot be applied directly to Bangladesh, given differences in educational structure (including the parallel general, Alia, and Qawmi madrasa systems), occupational distribution, and income levels. We therefore adapted the index to the Bangladeshi context, revising the education, occupation, and income domains while retaining the original scoring weights and five-tier classification. The full adaptation is described in the Methods, with the complete instrument in Table 1.

The adaptation was a methodological, classification-based exercise; its measurement properties were not examined at development. Using data from the quantitative component of a larger exploratory sequential mixed-methods study of hypertensive adults in Kamalganj sub-district, Moulvibazar, Bangladesh—available as a preprint [7]—the present study takes a first step toward evaluating the adapted index. Because the scale is a formative composite, the sample is rural and range-restricted, and the income component was not measured against an external criterion, we frame this as a preliminary, internal evaluation rather than a validation.

Specifically, we set out to: (i) examine the interrelationships among the adapted SES components; (ii) evaluate the agreement between fixed and empirically derived weights; (iii) assess convergent association against respondents’ own education; (iv) examine exploratory associations with medication adherence and blood pressure; and (v) evaluate the consequences of collapsing the five categories into three.

Methods

Design, setting and participants

This is a secondary analysis of the quantitative component of a larger exploratory sequential mixed-methods study [7]. The parent survey was a community-based, cross-sectional study conducted in Kamalganj sub-district, Moulvibazar district, Bangladesh, December 2025–February 2026. Participants were selected by random sampling from community health registers and local clinic records to ensure representativeness across age, sex, geographic location within the sub-district, and socioeconomic status, and were approached door-to-door. Eligible participants were adults aged 18 years and above with previously diagnosed or newly identified hypertension (self-reported diagnosis, or systolic blood pressure ≥140 mmHg and/or diastolic ≥90 mmHg at the survey, or current use of antihypertensive medication). Non-residents and those unable to participate because of mental or physical disability were excluded. Of 501 eligible adults approached, 461 (92.0%) provided informed consent and completed the survey. Because the sample comprised hypertensive adults from a single rural sub-district, the scale was evaluated only in this specific rural, clinical population and cannot yet be generalized to urban or national populations.


Of the 461 participants, 446 contributed to medication-adherence analyses; 15 taking no antihypertensive medication were excluded from adherence analyses because adherence is undefined without treatment. Blood-pressure analyses used all 461.

The Bangladesh-adapted Kuppuswamy scale

SES was assessed using a Bangladesh-adapted version of the Kuppuswamy scale, built on the original framework [2]. The original scale classifies households into five socioeconomic classes (Upper, Upper-Middle, Lower-Middle, Upper-Lower, Lower) from three weighted domains—education of the head of household (score 1–7), occupation of the head of household (score 1–10), and total monthly household income (score 1–12)—summed into a composite score  (possible range 3–29). We retained this scoring structure and five-tier classification unchanged and adapted the content of each domain, as below; the complete instrument is in Table 1.


The head of household was defined as the person recognised by the household as head; total household income was recorded as the combined monthly income of all members. The adaptation was carried out by the author team using national reference classifications; no external expert panel or formal content-validation procedure (e.g., Delphi) was used, and the instrument was not pilot-tested.

The adapted content (the category definitions) was developed by rational, classification-based mapping and is not empirically validated here. We examine only: the intercorrelation of the three components; the practical influence of the fixed versus empirical weights on ranking; convergent association with an external education measure; and exploratory associations with health outcomes. The five-tier classification was collapsed for analysis and is not evaluated, and, because the Kuppuswamy scale is a formative composite, we do not treat internal structure or dimensionality as evidence of construct validity.

Because education, occupation, and income are distinct but complementary indicators of socioeconomic position, the scale treats them as causal (formative) components that jointly define SES rather than as interchangeable items reflecting a latent trait; each contributes unique information, so high internal consistency is neither expected nor required [8, 9].

Education of the head of household

The Bangladeshi education system comprises both the conventional general pathway and a parallel madrasa system [10]. We reclassified the education categories using nationally recognized terminology while keeping the original Kuppuswamy scale, mapped madrasa-stream equivalence across levels,[11] and added a category for those who began but did not complete primary education—a group not clearly captured in the original scale.


Occupation of the head of household

Bangladesh Standard Classification of Occupations-2020 and International Standard Classification of Occupations [12, 13] classify occupations but do not rank them by socioeconomic position; we therefore retained the original Kuppuswamy rank order and 1–10 scoring and mapped each BSCO-2020 major group onto the corresponding Kuppuswamy tier by skill level and typical socioeconomic position. The non-hierarchical ‘Armed forces’ group, which spans several skill levels, was excluded, and the unemployed were assigned the lowest occupational score (1), reflecting absence of the head’s own labour-market position while household position is still captured through income and education.


Total monthly household income

We retained the original Kuppuswamy income scores (12, 10, 6, 4, 3, 2, 1) and redefined the corresponding monthly income bands in Bangladeshi Taka. Bands were anchored to the 2022 Household Income and Expenditure Survey mean monthly household income—26,163 BDT (rural) and 45,757 BDT (urban)—the most recent national figures (2021–2022 base year), and informed by the prevailing general inflation rate (8.17%), [14, 15] with round-number boundaries for field use. Separate rural and urban bands reflect the ~75% rural–urban disparity. This anchor-and-inflation approach differs from the formal consumer-price-index multiplication used in recent Indian updates; [4, 5] we note this, and the absence of an external wealth criterion, as limitations. Because the setting was rural, income was scored using the rural bands only.


Owing to small cell counts in the extreme strata (Upper n = 2, 0.4%; Lower n = 44, 9.5% in the five-category scheme), categories were collapsed into three for analysis: Upper (Upper + Upper-Middle), Middle (Lower-Middle), and Lower (Upper-Lower + Lower). This data-driven collapse means the analyses characterize a three-category derivative rather than the five-tier classification, which was not itself evaluated.

Variables used for the evaluation

The three component sub-scores, the composite, and the collapsed three-category classification were the variables examined. For convergent evidence, the respondent’s own educational attainment—recorded independently of the household head’s education used within the composite—was the sole external comparator; individual income was not collected. No asset- or durable-goods-based wealth index—the conventional external criterion for SES instruments in low- and middle-income settings [16, 17]—was collected, so land, savings, and other non-income wealth were not captured; this is the study’s headline limitation.


As exploratory health-related associations, we used the continuous Bangladesh Medication Adherence Scale (BMAS) score, categorical adherence status (optimal/sub-optimal), [18] and mean blood pressure. Adherence and blood pressure are not criterion measures of SES and, in a cross-sectional design, cannot establish predictive validity; we therefore treat these strictly as exploratory associations, noting that SES inputs, the education comparator, and adherence are all self-reported by the same respondent (possible common-method variance).

Blood pressure was measured with OMRON HEM-7120 automated blood pressure machine on the right arm, with the participant seated and rested for at least five minutes; two readings were taken at five-minute intervals and averaged. Newly identified hypertension was defined as systolic ≥140 and/or diastolic ≥90 mmHg, in line with WHO and International Society of Hypertension guidance [19].

Statistical analysis

Analyses used IBM SPSS Statistics version 31 (bootstrap confidence intervals from 1,000 resamples). As this scale is a formative composite, internal-consistency reliability (Cronbach’s alpha) is conceptually inapplicable and is not reported as evidence of validity; [20] the single-timepoint design precludes test–retest assessment, so no design-appropriate reliability estimate was available.


Intercorrelation (descriptive)

Spearman correlations among the three components were used as the primary structural description, supplemented by principal component analysis (PCA; principal-components extraction, one component, no rotation) with KMO, Bartlett’s test, variance explained by the extracted factor, and factor loadings. The Principal Component extraction method was used rather than Maximum Likelihood estimation because the objective was descriptive dimensionality assessment rather than confirmatory latent-trait modelling, and Maximum Likelihood estimation assumes multivariate normality that could not be assumed for this set of ordinal socioeconomic indicators. A dominant extracted factor is expected for three positively correlated inputs and, consistent with the formative treatment of the index, is not treated as evidence of construct validity


Weight agreement

To assess whether the fixed integer weights distort ranking, a PCA-derived weighted composite (regression scores) was compared with the fixed-weight composite using Spearman correlation and the resulting three-category classifications were also compared.


Convergent association

Spearman correlation with bootstrap 95% CIs were used to asses the association between the composite and respondents’ education, and, to remove the structural education overlap, with the education component, between respondents’ education and the non-education sub-score (occupation + income).


Exploratory health associations

One-way ANOVA was used to compare mean BMAS scores and blood pressure across the three SES categories, with Levene’s test for homogeneity of variance and Welch’s robust test where appropriate. Chi-square tests with Cramér’s V were used for categorical adherence and the SES–education cross-tabulation; for the latter, a Monte Carlo exact test was used because of sparse cells.


Category-collapse checks (exploratory, post-hoc)

A linear contrast across the three ordinal categories and a Jonckheere–Terpstra test across the uncollapsed five categories, were performed after the omnibus analyses and interpreted only as descriptive checks on the collapse. Statistical significance was set at P <0.05 (two-sided). The Holm–Bonferroni correction was applied across the primary family of tests (convergent association, BMAS ANOVA, adherence chi-square, SBP and DBP ANOVAs). Effect sizes (η2) are reported with 95% confidence intervals.

Table 1 The Bangladesh-adapted Kuppuswamy socioeconomic status scale in rural hypertensive adults: Scoring instrument (rural)

Category

Score

A. Education of the head of household

Professional education

7

Bachelor’s / Master’s/Fazil/Kamil / Takmil / Dawra-e-Hadith

6

Higher Secondary / Diploma / Alim / Sanabi

5

Secondary / Dakhil / Mutawassitah

4

Junior Secondary / Ibtedaiyah (Grades 6–8)

3

Primary / Ibtedaiyah (Grades 1–5)

2

Illiterate / incomplete primary education

1

B. Occupation of the head of household (mapped to BSCOa-2020/ISCOb)

Managers and senior officials

10

Professionals

9

Technicians and associate professionals

8

Clerical support workers

7

Service and sales workers

6

Skilled agriculture, forestry, fisheries

5

Craft and related trades workers

4

Plant/machine operators and assemblers

3

Elementary occupations

2

Unemployed

1

C. Total monthly household income — rural (BDT, 2021–2022 base year)

≥ 100,000

12

70,000–99,999

10

50,000–69,999

6

35,000–49,999

4

22,000–34,999

3

12,000–21,999

2

< 12,000

1

D. Socioeconomic class from the summed composite (range 3–29)

26–29

Upper (I)

16–25

Upper-Middle (II)

11–15

Lower-Middle (III)

5–10

Upper-Lower (IV)

<5

Lower (V)

Income scores follow the original Kuppuswamy weighting (12, 10, 6, 4, 3, 2, 1). aBangladesh Standard Classification of Occupations; bInternational Standard Classification of Occupations

Results

In the five-category classification, the sample was 69.4% Lower, 21.7% Middle, and 8.9% Upper, with the true extremes sparse (Upper n=2, 0.4%; Lower n=44, 9.5%). The upper SES range is therefore essentially untested, and the analytic ‘Upper’ group is overwhelmingly Upper-Middle. Analyses use the collapsed three categories (Lower n=320; Middle n= 100; Upper n=41).


Component intercorrelation and weighting

The three components were positively but only modestly intercorrelated (Spearman rho: education–occupation 0.39, education–income 0.36, occupation–income 0.28; all P <0.001), consistent with distinct but complementary indicators (Table 2). Factor Analysis using the Principal Component extraction method yielded one factor with eigenvalue above unity (59.2% of variance; KMO = 0.635; Bartlett’s P < 0.001); as noted, this is expected for positively correlated inputs and does not establish construct validity. A PCA-weighted composite ranked participants almost identically to the fixed-weight composite (Spearman rho = 0.989); every participant classified Middle or Upper by the fixed weights fell in the upper third of the PCA-weighted distribution, and only 10 of 320 Lower-SES participants did, indicating that the fixed weights do not materially distort ranking.

Variables  

Frequency (%)

Indication of colposcopy

 

Visual inspection of the cervix with acetic acid positive

200 (66.7)

Abnormal pap test

13 (4.3)

Human papilloma virus DNA positive

4 (1.3)

Suspicious looking cervix

14 (4.7)

Others (per vaginal discharge, post-coital bleeding)

69 (23.0)

Histopathological diagnosis

Cervical Intraepithelial Neoplasia 1

193 (64.3)

Cervical Intraepithelial Neoplasia 2

26 (8.7)

Cervical Intraepithelial Neoplasia 3

32 (10.7)

Invasive cervical cancer

27 (9.0)

Chronic cervicitis

17 (5.6)

Squamous metaplasia

5 (1.7)

Groups based on pre-test marks

Pretest
marks (%)

Posttest

Marks (%)

Difference in pre and post-test marks (mean improvement)

P

Didactic lecture classes

<50%

36.6 (4.8)

63.2 (9.4)

26.6

<0.001

≥50%

52.8 (4.5)

72.4 (14.9)

19.6

<0.001

Flipped classes

<50%

36.9 (4.7)

82.2 (10.8)

45.4

<0.001

≥50%

52.8 (4.6)

84.2 (10.3)

31.4

<0.001

Data presented as mean (standard deviation)

Background characteristics

Number (%)

Age at presentation (weeks)a

14.3 (9.2)

Gestational age at birth (weeks)a

37.5 (2.8)

Birth weight (grams)a

2,975.0 (825.0)

Sex

 

Male

82 (41)

Female

118 (59)

Affected side

 

Right

140 (70)

Left

54 (27)

Bilateral

6 (3)

Delivery type

 

Normal vaginal delivery

152 (76)

Instrumental delivery

40 (20)

Cesarean section

8 (4)

Place of delivery

 

Home delivery by traditional birth attendant

30 (15)

Hospital delivery by midwife

120 (60)

Hospital delivery by doctor

50 (25)

Prolonged labor

136 (68)

Presentation

 

Cephalic

144 (72)

Breech

40 (20)

Transverse

16 (8)

Shoulder dystocia

136 (68)

Maternal diabetes

40 (20)

Maternal age (years)a

27.5 (6.8)

Parity of mother

 

Primipara

156 (78)

Multipara

156 (78)

aMean (standard deviation), all others are n (%)

Background characteristics

Number (%)

Age at presentation (weeks)a

14.3 (9.2)

Gestational age at birth (weeks)a

37.5 (2.8)

Birth weight (grams)a

2,975.0 (825.0)

Sex

 

Male

82 (41)

Female

118 (59)

Affected side

 

Right

140 (70)

Left

54 (27)

Bilateral

6 (3)

Delivery type

 

Normal vaginal delivery

152 (76)

Instrumental delivery

40 (20)

Cesarean section

8 (4)

Place of delivery

 

Home delivery by traditional birth attendant

30 (15)

Hospital delivery by midwife

120 (60)

Hospital delivery by doctor

50 (25)

Prolonged labor

136 (68)

Presentation

 

Cephalic

144 (72)

Breech

40 (20)

Transverse

16 (8)

Shoulder dystocia

136 (68)

Maternal diabetes

40 (20)

Maternal age (years)a

27.5 (6.8)

Parity of mother

 

Primipara

156 (78)

Multipara

156 (78)

aMean (standard deviation), all others are n (%)

Table 2 Correlation matrix (Spearman) among the three Kuppuswamy components in rural hypertensive adults (n=461)

Component

Education

Occupation

Income

Education

1.00

 

 

Occupation

0.39

1.00

 

Income

0.36

0.28

1.00

All correlations P  <0.001. Supporting PCA: KMO = 0.64; Bartlett’s P  <0.001; first component eigenvalue 1.78, 59.2% of variance; loadings 0.81 (education), 0.79 (occupation), 0.70 (income). A dominant first component is expected for positively correlated inputs and is not evidence of construct validity. A PCA-weighted composite and the fixed-weight composite ranked participants near-identically (Spearman rho=0.99).

Mean escape latency of acquisition day

Groups                 

NC

SC

ColC

Pre-SwE Exp

Post-SwE Exp

Days

 

 

 

 

 

1st

26.2 (2.3)

30.6 (2.4) 

60.0 (0.0)b

43.2 (1.8)b

43.8 (1.6)b

2nd

22.6 (1.0) 

25.4 (0.6)

58.9 (0.5)b

38.6 (2.0)b

40.5 (1.2)b

3rd

14.5 (1.8) 

18.9 (0.4) 

56.5 (1.2)b

34.2 (1.9)b 

33.8 (1.0)b

4th

13.1 (1.7) 

17.5 (0.8) 

53.9 (0.7)b

35.0 (1.6)b

34.9 (1.6)b

5th

13.0 (1.2) 

15.9 (0.7) 

51.7 (2.0)b

25.9 (0.7)b 

27.7 (0.9)b

6th

12.2 (1.0) 

13.3 (0.4) 

49.5 (2.0)b

16.8 (1.1)b

16.8 (0.8)b

Average of acquisition days

5th and 6th 

12.6 (0.2)

14.6 (0.8)

50.6 (0.7)b

20.4 (2.1)a

22.4 (3.2)a

NC indicates normal control; SC, Sham control; ColC, colchicine control; SwE, swimming exercise exposure.

aP <0.05; bP <0.01.

Convergent association

The composite correlated weakly with respondents’ own education (Spearman rho = 0.253, P <0.001; ~6% shared variance) (Table 3). This association is only partly independent, as the comparator overlaps the composite’s education component, and the restricted SES range attenuates correlations; correlating the comparator with the non-education sub-score gave a similar, marginally stronger value (rho=0.274, 95% CI 0.18–0.36), indicating the convergence is not merely structural. The SES–education crosstabulation was significant (Cramér’s V=0.300; Monte Carlo exact P <0.001; linear-by-linear P <0.001) despite 22.2% of cells having expected counts <5, and the exact test confirms the association is not an artefact of sparse cells. No discriminant comparator was available.

Table 3 Convergent association and exploratory health-related associations of the Bangladesh-adapted Kuppuswamy socioeconomic status scale in rural hypertensive adults

Comparator/outcome

n

Statistic (95% confidence interval)

Effect size

P

Note

Respondent education versus composite

446

rho = 0.25 (—)

—

<0.001

Weak; partly structural overlap; attenuated by range restriction

Respondent education versus occupation+income

446

rho=0.27 (0.18–0.36)

—

<0.001

Independent convergence; not an artefact of overlap

Socioeconomic status × education (3-cat)

446

χ²=83.1, df=4; Monte Carlo exact

Cramér’s V=0.300

<0.001

22.2% cells expected <5; exact test confirms association

BMAS by SES mean (standard deviation)

443

Lower 18.3 (4.6); Middle 21.9 (4.2); Upper 22.9 (4.4)

η²=0.142 (0.085–0.199)

<0.001

Levene 0.37

Middle vs upper 0.47

 

Optimal adherence by SES

446

32.5% / 65.3% / 82.5%; χ²=58.7

Cramér’s V=0.363

< 0.001

Survives Holm–Bonferroni

Systolic blood pressure by SES mean (standard deviation)

458

Lower 142.4 (19.5); Middle 141.3 (21.3); Upper 135.0 (18.5)

η²=0.011 (0.000–0.035)

0.08

0.47

 

Diastolic blood pressure by SES mean (standard deviation)

458

Lower 87.6   (11.9); Middle 88.5 (12.6); Upper 84.7 (10.7)

η²=0.006 (0.000–0.025)

0.23

0.32

 

Adherence and blood pressure are exploratory health-related outcomes, not criterion measures of SES; the cross-sectional design cannot establish predictive validity. BMAS: Bangladesh Medication Adherence Scale; n is different, because of missing values.

Exclusion check: Untreated participants

The 15 untreated participants were distributed across SES categories (X2=0.84, P=0.66). Given the small number and low power of this test, their exclusion is unlikely to have materially biased the SES–adherence association, though it cannot be excluded with certainty.


Exploratory health-related associations

Variance homogeneity held for all three outcomes (Levene’s P=0.37, 0.47, 0.32), so standard ANOVA was appropriate; Welch’s test gave the same conclusions. Medication adherence rose across SES categories (F(2,443)=36.56, P <0.001, η2=0.142, 95% CI 0.085–0.199), but post hoc tests showed the Lower group differing from Middle and Upper (P <0.001) while Middle and Upper did not differ (P=0.47)—a Lower-versus-rest pattern rather than a graded gradient. Optimal-adherence prevalence rose from 32.5% to 65.3% to 82.5% (Cramér’s V=0.363, P <0.001). Both adherence findings and the convergent association survived Holm–Bonferroni correction. Blood pressure showed no association with SES (SBP P=0.08; DBP P=0.23; η2 <0.011, intervals including zero) and did not survive correction; an exploratory post-hoc SBP linear contrast (P=0.02) is not interpreted given the null omnibus, trivial effect, and multiplicity.


Category-collapse checks

As post-hoc checks, a Jonckheere–Terpstra test across all five strata showed a significant ordered adherence trend (standardised J-T=8.78, P <0.001), suggesting the dose–response pattern preceded the collapse. Because the true Upper stratum contained only two participants, all five-category estimates are unstable and are reported as descriptive only (Table 4).

Categories

Number (%)

Sex

 

   Male

36 (60.0)

   Female

24 (40.0)

Age in yearsa

8.8 (4.2)

Education

 

   Pre-school

20 (33.3)

   Elementary school

24 (40.0)

   Junior high school

16 (26.7)

Cancer diagnoses

 

Acute lymphoblastic leukemia

33 (55)

Retinoblastoma

5 (8.3)

Acute myeloid leukemia

4 (6.7)

Non-Hodgkins lymphoma

4 (6.7)

Osteosarcoma

3 (5)

Hepatoblastoma

2 (3.3)

Lymphoma

2 (3.3)

Neuroblastoma

2 (3.3)

Medulloblastoma

1 (1.7)

Neurofibroma

1 (1.7)

Ovarian tumour

1 (1.7)

Pancreatic cancer

1 (1.7)

Rhabdomyosarcoma

1 (1.7)

aMean (standard deviation)

Test results

Disease

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

Yes

No

Reid’s score ≥ 5

Positive

10

15

37.0

94.5

40.1

93.8

Negative

17

258

 

 

 

 

Swede score ≥ 5

Positive

20

150

74.1

45.0

11.8

94.6

Negative

7

123

 

 

 

 

Swede score ≥ 8

Positive

3

21

11.1

92.3

12.5

91.3

Negative

24

252

 

 

 

 

a High-grade indicates a score of ≥5 in both tests; PPV indicates positive predictive value; NPV, negative predictive value

Test

Sensitivity (%)

Specificity (%)

Positive predictive value (%)

Negative predictive value (%)

Reid’s score ≥ 5

37.0

94.5

40.0

93.8

Swede score ≥ 5

74.1

45

11.8

94.6

Swede score ≥ 8

11.1

92.3

12.5

91.3

Test

Sensitivity (%)

Specificity (%)

Positive predictive value (%)

Negative predictive value (%)

Reid’s score ≥ 5

37.0

94.5

40.0

93.8

Swede score ≥ 5

74.1

45

11.8

94.6

Swede score ≥ 8

11.1

92.3

12.5

91.3

Narakas classification

Total

200 (100%)

Grade 1

72 (36%)

Grade 2

64 (32%)

Grade 3

50 (25%)

Grade 4

14 (7%)

Complete recoverya

107 (54)

60 (83)

40 (63)

7 (14)

-

Near complete functional recovery but partial deformitya

22 (11)

5 (7)

10 (16)

6 (12)

1 (7)

Partial recovery with gross functional defect    and deformity

31 (16)

7 (10)

13 (20)

10 (20)

1 (7)

No significant improvement 

40 (20)

-

1 (1.5)

27 (54)

12 (86)

aSatisfactory recovery

bGrade 1, C5, 6, 7 improvement; Grade 2, C5, 6, 7 improvement; Grade 3, panpalsy C5, 6, 7, 8, 9, Grade 4, panpalsy with Hornon’s syndrome.

Narakas classification

Total

200 (100%)

Grade-1

72 (36%)

Grade-2

64 (32%)

Grade-3

50 (25%)

Grade-4

14 (7%)

Complete recoverya

107 (54)

60 (83)

40 (63)

7 (14)

-

Near complete functional recovery but partial deformitya

22 (11)

5 (7)

10 (16)

6 (12)

1 (7)

Partial recovery with gross functional defect    and deformity

31 (16)

7 (10)

13 (20)

10 (20)

1 (7)

No significant improvement 

40 (20)

-

1 (1.5)

27 (54)

12 (86)

aSatisfactory recovery

bGrade 1, C5, 6, 7 improvement; Grade 2, C5, 6, 7 improvement; Grade 3, panpalsy C5, 6, 7,8,9, Grade 4, panpalsy with Hornon’s syndrome.

Variables in probe trial day

Groups

NC

SC

ColC

Pre-SwE Exp

Post-SwE Exp

Target crossings

8.0 (0.3)

7.3 (0.3) 

1.7 (0.2)a

6.0 (0.3)a

5.8 (0.4)a

Time spent in target

18.0 (0.4) 

16.2 (0.7) 

5.8 (0.8)a

15.3 (0.7)a

15.2 (0.9)a

NC indicates normal control; SC, Sham control; ColC, colchicine control; SwE, swimming exercise exposure.

aP <0.01.

Pain level

Number (%)

P

Pre

Post 1

Post 2

Mean (SD)a pain score

4.7 (1.9)

2.7 (1.6)

0.8 (1.1)

<0.001

Pain categories

    

   No pain (0)

-

1 (1.7)

31 (51.7)

<0.001

   Mild pain (1-3)

15 (25.0)

43 (70.0)

27 (45.0)

 

   Moderete pain (4-6)

37 (61.7)

15 (25.0)

2 (3.3)

 

   Severe pain (7-10)

8 (13.3)

2 (3.3)

-

 

aPain scores according to the visual analogue scale ranging from 0 to 10; SD indicates standard deviation

Surgeries

Number  

(%)

Satisfactory outcomes n (%)

Primary surgery (n=24)

 

 

Upper plexus

6 (25)

5 (83)

Pan-palsy

18 (75)

6 (33)

All

24 (100)

11 (46)

Secondary Surgery (n=26)

 

 

Shoulder deformity

15 (58)

13 (87)

Wrist and forearm deformity

11 (42)

6 (54)

All

26 (100)

19 (73)

Primary and secondary surgery

50 (100)

30 (60)

Mallet score 14 to 25 or Raimondi score 2-3 or Medical Research grading >3 to 5.

Narakas classification

Total

200 (100%)

Grade-1

72 (36%)

Grade-2

64 (32%)

Grade-3

50 (25%)

Grade-4

14 (7%)

Complete recoverya

107 (54)

60 (83)

40 (63)

7 (14)

-

Near complete functional recovery but partial deformitya

22 (11)

5 (7)

10 (16)

6 (12)

1 (7)

Partial recovery with gross functional defect    and deformity

31 (16)

7 (10)

13 (20)

10 (20)

1 (7)

No significant improvement 

40 (20)

-

1 (1.5)

27 (54)

12 (86)

aSatisfactory recovery

bGrade 1, C5, 6, 7 improvement; Grade 2, C5, 6, 7 improvement; Grade 3, panpalsy C5, 6, 7,8,9, Grade 4, panpalsy with Hornon’s syndrome.

Trials

Groups

NC

SC

ColC

Pre-SwE Exp

Post-SwE Exp

1

20.8 (0.6)

22.1 (1.8)

41.1 (1.3)b

31.9 (1.9)b

32.9 (1.8)a, b

2

10.9 (0.6)

14.9 (1.7)

37.4 (1.1)b

24.9 (2.0)b

26.8 (2.5)b

3

8.4 (0.5)

9.9 (2.0)

32.8 (1.2)b

22.0 (1.4)b

21.0 (1.4)b

4

7.8 (0.5)

10.4 (1.3)

27.6(1.1)b

12.8 (1.2)b

13.0 (1.4)b

Savings (%)c

47.7 (3.0)

33.0 (3.0)

10.0 (0.9)b

23.6 (2.7)b

18.9 (5.3)b

NC indicates normal control; SC, Sham control; ColC, colchicine control; SwE, swimming exercise exposure.

aP <0.05; bP <0.01.

cThe difference in latency scores between trials 1 and 2, expressed as the percentage of savings increased from trial 1 to trial 2

Table 4 Exploratory, post-hoc category-collapse checks (five-to-three) in Bangladesh-adapted Kuppuswamy socioeconomic status scale in rural hypertensive adults

Outcome

n

Test

Statistic

P

Note

BMASa across all 5 strata

446

Jonckheere–Terpstra

J–T=8.78

<0.001 

Ordered trend present before collapse

BMAS continuous

443

Linear contrast (−1,0,1)

t=6.09

<0.001 

Direction retained after collapse

Mean systolic blood pressure

458

Linear contrast (−1,0,1)

t=−2.27

0.02 

Weak; omnibus non-significant; interpret cautiously

Mean diastolic blood pressure 

458

Linear contrast (−1,0,1)

t= −1.44

0.15 

No trend

Adherence category

446

Linear-by-linear

χ²=57.12

<0.001

Ordinal trend retained after collapse

These post-hoc analyses describe the behaviour of the collapse and are not independent corroboration of the index. Because the true Upper stratum contained only two participants, five-category estimates are unstable and are descriptive only. aBangladesh Medication Adherence Scale; n is different, because of missing values.

Discussion

In this single rural sample, the three components of the adapted scale were positively intercorrelated, and fixed and empirically derived weights ranked participants almost identically. The composite showed a weak, only partly independent convergent association with respondents’ education. It was associated with medication adherence mainly at the lower end of the distribution (Lower-versus-rest) rather than as a graded gradient, and showed no association with blood pressure.

Rationale and appropriateness of the adaptations

The Kuppuswamy scale was designed for the Indian population, and earlier Bangladeshi studies applied it with occupational categories from non-local frameworks and uncalibrated income cut-offs [21] Bangladesh has a large madrasa-educated population—over four million students, disproportionately rural [22]—so we built madrasa equivalence into the education domain and added an ‘illiterate / incomplete primary education’ category to reduce misclassification. Occupation was aligned with the national BSCO-2020 classification, and income bands were recalculated for the Bangladeshi economy with separate rural and urban columns. These changes aim to make the scale contextually appropriate while leaving the original scoring weights unchanged for comparability. The adaptation was, however, a content-based exercise by the author team without external expert review or field piloting.

Relation to the wider Kuppuswamy literature

Kuppuswamy-based scale have rarely had their measurement properties tested directly; the Indian revisions [3–5, 23] have mainly updated income brackets rather than examined the scale statistically. That the three domains are positively intercorrelated is consistent with their treatment as complementary formative indicators [8, 9].

Treating the scale as formative has three practical implications for how it should be evaluated and used. First, the components are constitutive rather than interchangeable: education, occupation, and income each define part of socioeconomic position, so dropping or reweighting a domain changes the construct itself. Internal-consistency reliability and dimensionality statistics are therefore inappropriate as validation evidence, and a high or low Cronbach’s alpha carries no meaning for this scale. Second, because the composite is a summary index rather than a reflective measure of a latent trait, appropriate evaluation rests on external and criterion evidence—chiefly agreement with an independent wealth or asset measure—rather than on internal structure; this is why the absence of a wealth criterion here is the central limitation, and why our intercorrelation and PCA results are presented as descriptive rather than validating. Third, the practical value of a formative SES index lies in ranking households consistently, which the near-identical ranking under fixed and empirical weights (rho=0.989) supports: users can apply the simple fixed weights without loss of ranking accuracy.

Large multi-country surveys instead use a PCA-based asset wealth index as an external criterion [16, 17]—the independent comparator absent here. Our income bands were set by anchor-and-inflation adjustment rather than the formal consumer-price-index multiplication of recent Indian updates, [4, 5] a difference we treat as a limitation and a target for future refinement.

Implications for use of the adapted index

These findings offer cautious, preliminary support for use of the index in comparable rural Bangladeshi research, with important caveats. The weak convergent association suggests the composite is not redundant with respondent education, though range restriction limits this inference, and the absence of a blood-pressure association suggests the index relates more to health behaviour than to a downstream physiological outcome. The five-tier classification was not evaluated, and the urban income bands were not applied or evaluated and should not yet be used. Further evaluation in larger, more socioeconomically diverse rural and urban samples is required before routine use. For researchers wishing to apply the index, The Bangla version of the complete instrument is given in Supplementary Table S1.

Limitations

This study had several limitations. There was no independent measure of household wealth to compare with the adapted SES scale, and the scale did not undergo formal expert or field validation. Income bands relied on an approximate inflation adjustment, and fieldwork lagged the 2021–2022 base year by 3–4 years. Our only comparator (education) overlaps the index's own education domain, narrowing the evidence base.

KMO (0.635) was modest, but the Factor Analysis was used descriptively, not as validity proof. No expert panel, Delphi process, or pilot testing was conducted.

Only n=2 fell in the Upper stratum, so the top tier and full five-category structure remain untested; we used a collapsed three-category version instead. SES reflects only the household head, not all members.

Reliability metrics don't apply to a formative, cross-sectional scale and self-reported outcomes may share common-method variance with SES. Findings come from one rural hypertensive sample and may not generalise.

Conclusion

In this single rural, hypertensive sample, the Bangladesh-adapted Kuppuswamy scale showed positive intercorrelation among its components, near-identical ranking under fixed and empirical weights, weak and only partly independent convergent association with education, and a modest association with medication adherence at the lower socioeconomic range. Together with a ready-to-use scoring instrument, these findings are promising but preliminary.

Further evaluation—across the full SES range, in urban as well as rural settings, and against an independent wealth criterion—is required before the scale can be recommended for routine use. Future work should collect an independent asset/wealth index to enable criterion evaluation of the income component; sample across the full SES range, including genuine upper-stratum households, to test the five-tier structure; re-derive the income bands using formal Bangladesh price-index data; evaluate the urban bands in an urban sample; and conduct a test–retest study.

 Lesion-size

Histopathology report

Total

CIN1

CIN2

CIN3

ICC

CC

SM

0–5 mm

73

0

0

0

5

5

83

6–15 mm

119

18

1

4

0

0

142

>15 mm

1

8

31

23

12

0

75

Total

193

26

32

27

17

5

300

CIN indicates cervical intraepithelial neoplasia; ICC, invasive cervical cancer; CC, chronic cervicitis; SM, squamous metaplasia

 

Histopathology report

Total

CIN1

CIN2

CIN3

ICC

CC

SM

Lesion -Size

0-5  mm

73

0

0

0

5

5

83

6-15  mm

119

18

1

4

0

0

142

>15  mm

1

8

31

23

12

0

75

Total

193

26

32

27

17

5

300

CIN indicates Cervical intraepithelial neoplasia; ICC, Invasive cervical cancer; CC, Chronic cervicitis; SM, Squamous metaplasia

Group

Didactic posttest marks (%)

Flipped posttest marks (%)

Difference in marks (mean improvement)

P

<50%

63.2 (9.4)

82.2 (10.8)

19.0

<0.001

≥50%

72.4 (14.9)

84.2 ( 10.3)

11.8

<0.001

Data presented as mean (standard deviation)

Acknowledgements
None
Author contributions
Concept or design of the work; or the acquisition, analysis, or interpretation of data for the work: MTI, MKU, MMHK, SMA, MAH. Drafting the work or reviewing it critically for important intellectual content: MTI, MKU, MMHK, SMA, MAH. Final approval of the version to be published: MTI, MKU, MMHK, SMA, MAH. Accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved: MTI.
Conflict of interest
We do not have any conflict of interest.
Data availability statement
We confirm that the data supporting the findings of the study will be shared upon reasonable request.
Supplementary files
AI disclosure
We used an AI-based assistant to support language editing and reference formatting during manuscript preparation. All study design, data collection, analyses, interpretations, and final content were performed and verified by the authors, who take full responsibility for the work.
    References

    1. Winkleby MA, Jatulis DE, Frank E, Fortmann SP. Socioeconomic status and health: how education, income, and occupation contribute to risk factors for cardiovascular disease. Am J Public Health. 1992 Jun;82(6):816-820. doi: https://doi.org/10.2105/ajph.82.6.816


    2. Kuppuswamy B. Manual of socioeconomic status (urban). Manasayan. Delhi; 1981. Available at: https://opac.cup.edu.in/bib/15033 [Accessed on 17 August 2026]


    3. Kumar G, Dash P, Patnaik J, Pany G. Socioeconomic status scale-modified Kuppuswamy scale for the year 2022. Int J Community Dent. 2022 Jun 10;10(1):1-6. doi: https://doi.org/18203/2394-6040.ijcmph20222581


    4. Mandal I, Hossain SR. Modified Kuppuswamy scale updated for the year 2025. Int J Community Med Public Health. 2025;12(5):2423–2425. doi: https://doi.org/10.18203/2394-6040.ijcmph20251411


    5. Ahmad A, Faizi N, Siddiqui NA. Revised Prasad and Kuppuswamy socioeconomic scales for India in 2024: methods and updates. Indian J Community Health. 2024;36(4):598–603. doi: https://doi.org/10.47203/IJCH.2024.v36i04.025


    6. Howe LD, Galobardes B, Matijasevich A, Gordon D, Johnston D, Onwujekwe O, Patel R, Webb EA, Lawlor DA, Hargreaves JR. Measuring socio-economic position for epidemiological studies in low- and middle-income countries: a methods of measurement in epidemiology paper. Int J Epidemiol. 2012 Jun;41(3):871-886. doi: https://doi.org/10.1093/ije/dys037


    7. Islam MT, Haque MA, Uddin MK, Chakraborty M, Arafat SM. Health beliefs, medication adherence, and blood-pressure control in rural Bangladeshi hypertension: an exploratory sequential mixed-methods study. Research Square. 2026 [Preprint]. doi: https://doi.org/10.21203/rs.3.rs-10002032/v1


    8. Diamantopoulos A, Winklhofer HM. Index construction with formative indicators: an alternative to scale development. J Mark Res. 2001;38(2):269–277. doi: https://doi.org/10.1509/jmkr.38.2.269.18845


    9. Bollen KA, Bauldry S. Three Cs in measurement models: causal indicators, composite indicators, and covariates. Psychol Methods. 2011 Sep;16(3):265-284. doi: https://doi.org/10.1037/a0024448


    10. Mousumi MA, Kusakabe T. School education system in Bangladesh. In: Sarangapani PM, Pappu R, eds. Handbook of education systems in South Asia. Springer. Singapore; 2021 Aug 29. pp 443–477. doi: https://doi.org/10.1007/978-981-15-0032-9_17


    11. Bangladesh Bureau of Educational Information and Statistics (BANBEIS). Bangladesh education statistics 2022. Ministry of Education. Dhaka; 2023. Available at: https://banbeis.gov.bd [Accessed on 17 August 2026]


    12. Bangladesh Bureau of Statistics. Bangladesh Standard Classification of Occupations 2020 (BSCO-2020). Statistics and Informatics Division, Ministry of Planning. Dhaka; 2021. Available at:  https://bbs.gov.bd/pages/static-pages/6922de09933eb65569e17bb3 [Accessed on 18 August 2026]


    13. International Labour Organization. International Standard Classification of Occupations (ISCO). ILOSTAT. Available at: https://ilostat.ilo.org/methods/concepts-and-definitions/classification-occupation/ [Accessed on 17 August 2026]


    14. Bangladesh Bank. Inflation — economic data. Bangladesh Bank. Dhaka. Available at: https://www.bb.org.bd/en/index.php/econdata/inflation [Accessed on 17 August 2026]


    15. Bangladesh Bureau of Statistics (BBS). Household income and expenditure survey (HIES) 2022: key findings. Statistics and Informatics Division, Ministry of Planning. Dhaka; 2023. Available at: https://bbs.gov.bd [Accessed on 17 August 2026]


    16. Rutstein SO, Johnson K. The DHS wealth index. DHS comparative reports no. 6. ORC Macro. Calverton, MD; 2004. Available at: https://dhsprogram.com/pubs/pdf/CR6/CR6.pdf [Accessed on 17 August 2026]


    17. Filmer D, Pritchett LH. Estimating wealth effects without expenditure data--or tears: an application to educational enrollments in states of India. Demography. 2001 Feb;38(1):115-132. doi: https://doi.org/10.1353/dem.2001.0003


    18. Islam MT, Arafat SM, Shama MA, Ahmed T, Ahsan K, Hasib MA, Haque MA, Uddin MK. Bangladesh Medication Adherence Scale: Development and Validation of a Tool for Patients With Chronic Illnesses. Health Sci Rep. 2026 Apr 28;9(5):e72455. doi: https://doi.org/10.1002/hsr2.72455


    19. Unger T, Borghi C, Charchar F, Khan NA, Poulter NR, Prabhakaran D, Ramirez A, Schlaich M, Stergiou GS, Tomaszewski M, Wainford RD, Williams B, Schutte AE. 2020 International Society of Hypertension Global Hypertension Practice Guidelines. Hypertension. 2020 Jun;75(6):1334-1357. doi: https://doi.org/10.1161/HYPERTENSIONAHA.120.15026


    20. Tavakol M, Dennick R. Making sense of Cronbach's alpha. Int J Med Educ. 2011 Jun 27;2:53-55. doi: https://doi.org/10.5116/ijme.4dfb.8dfd


    21. Ahmed B, Shiraji KH, Chowdhury MH, Uddin MG, Islam SN, Hossain S. Socio-economic status of patients with acute coronary syndrome: data from a district-level general hospital of Bangladesh. Cardiovasc J. 2017;10(1):17–20. doi: https://doi.org/10.3329/cardio.v10i1.34357


    22. Zinia FT. Number of madrasa students increasing nationwide. Bonik Barta. 2024 Mar 30. Available at: https://www.bonikbarta.com [Accessed on 17 August 2026]


    23. Saleem SM. Modified Kuppuswamy socioeconomic scale updated for the year 2019. Indian J Forensic Community Med. 2019;6(1):1–3. doi: https://doi.org/10.18231/2394-6776.2019.0001