Improved health is positively associated with economic benefits and improved societal well-being. Health can be measured by a composite index combining relevant dimensions and indicators pertaining to activities of daily living, physical and cognitive functions, emotional well-being, feelings, etc. This involves appropriate biomarkers, patient-reported questionnaires and performance-based scales giving rise to ratio scale data, count data, ordinal data and categorical confounders. The paper transforms scores of biomarkers, item scores of scales to form normally distributed index of health care (I_Health) reflecting overall health status. The index satisfying desired properties including assessment of changes over time facilitates parametric statistical analysis and better evaluation of reliability and validity. Combination of ordinal data and ratio level data by a simple method is a novelty. Dimensions of I (Health) can be ranked based on their elasticity. Health dynamics is reflected by changes in I (Health) of one or group of individuals along a disease continuum across categorical variables like stages of disease, gender, income, educational levels, etc. ANOVA, Covariance Structure Analysis of the normally distributed index can help to test equality or invariance of factors across groups, using χ^2-test. Computation of separate index is recommended for vulnerable groups like old aged citizens, postpartum women running the risk of maternal mortality, economically backward class, etc. 

Improved health outcomes with reduced morbidity and disability in a nation can influence significantly growth of human capital driving job creation, increasing productivity with broader economic benefits and improved societal progress. Strategic investment of each dollar in health giving about 24-times return (World Bank, 2025) is a high-yield economic strategy with significant social imperatives (Masters et al. 2017). Empirical relationship between health and economic growth of a country was established by Elena et al. (2024) which varies across time and space depending on demographic transition stages, health improvements at different age categories, quality of the public health system and suggested policy approach for integrating health considerations into economic strategies. Sound Index of Health Status for a country or region facilitates finding such empirical relationships. The index to reflect physical, mental, and social well-being of people with positive and negative states of health like life expectancy at birth (LEB) or condition-specific life expectancy, mortality rate, age-specific mortality rates, etc. and patient-reported measure of health and functional status (Parrish, 2010). India is committed to accelerate progress in the Sustainable Development Goals (SDGs), including SDG-3 targets relating to good health and well-being for all.  The health status of a region is determined by biological, immunological, nutritional, behavioural, environmental, genetic and social factors. Clearly, health status is multidimensional and can be measured by a composite index combining relevant dimensions and indicators pertaining to physical and cognitive functions, activities of daily living (ADL), feelings, social and emotional well-being, etc. which can be evaluated for specific factors that influence the index. One important disease related index of a country is Burden of Disease (BoD) covering communicable diseases, non-communicable diseases, injuries, rates of disabilities and death across ages, gender, social groups, disability-adjusted life years (DALYs) (Devleesschauwer et al. 2014). Despite reduction in indicators like maternal mortality ratio (MMR), infant mortality rate (IMR), crude death rate (CDR), and increase in LEB, the burden of non-communicable diseases is on the rise and uneven across regions in India (Arokiasamy, 2018; Bango and Ghosh, 2022). Health care involves assessing interactions of a group of individuals responding to treatments and interventions for a disease or condition and to measure progress with time. This can be achieved by use of appropriate biomarkers and standardized tests or scales to measure efficacy of treatment, physical disability (both temporary and permanent), cognitive impairment, low speed of information processing, progressive memory loss, etc. (Patterson, 2018). Descriptive, analytic, and experimental epidemiological studies quantify rates of occurrence of disease, identify risk factors and develop effective measures for control in addition to tracking disease trends (Brachman, 1996). Health outcomes researches evaluate healthcare interventions and the outcomes, to promote quality-of-life (QOL) of patients and to guide policy makers, health economists, health professionals and clinicians (Gidron, 2013). One may refer to Zoccali et al. (2024) for details regarding use of biomarkers in clinical practices. The authors recommended standardization of assays, along with reporting of reliability and validity of biomarkers for better patient outcomes and progress. In addition, outcome measures for evaluation of progress or decline in patients over time need to be able to detect change (responsiveness) (Roach, 2006).

Commonly used dimensions and indicators used for assessing health status include among others:

• Mortality: CDR, LEB, MMR, IMR, etc. in ratios
• Morbidity: Incidence (new cases added- count data), Prevalence rate (ratio), Notification Rates (number of reported cases), etc. 
• Disability Rates : Bed Disability Days, Work-Loss Days, Activity Restriction Days, etc.(count data)
• Disease Burden: DALY = Years of Life Lost + Years Lived with Disability; Health Adjusted Life Expectancy (HALE) = average number of years spent in poor health 
• Nutritional Status Indicators: Body Mass Index(BMI) ( for adults, age-wise stunting, wasting, (both expressed as number of standard deviations (SD) from the median or underweight (kg relative to ages) for children.
• Biochemical Indicators (Laboratory Tests): Deficiency of Vitamin A (Serum Retinol in micromoles/L);Iodine (Urinary Iodine in μg/L), Iron (Serum Ferritin/Hemoglobin in micrograms/L) and nutritional risk (Serum Albumin in g/L)
• Clinical Indicators: Signs of hair loss, brittle nails, skin rashes, swollen gums, reflecting nutrient deficiencies; edema or severe muscle wasting.
• Dietary Indicators: Food frequency questionnaires using multi-point items; Dietary Diversity Score(Yes – No type items) 
• Healthcare Delivery & Utilization: Availability and usage of health services like Doctor-to-Population Ratio, Immunization Coverage (in percentage), Hospital Bed Capacity (count data).
• Social and Mental Health Indicators: Include suicide rates, homicide rates, rates of mental illness along with micro areas such as: Positive Mental Health (Life satisfaction, happiness, subjective well-being), Social Functioning (Social connectedness, participations, isolation levels, social capital/support networks), Mental Health Risks (Early childhood adversity, socioeconomic disadvantage, and discrimination), Rates of depression, anxiety, burnout, substance/alcohol misuse.
• Environmental Indicators: Reflect exposure to health risks like air quality index (micrograms per cubic meter ( or parts per million (ppm)), water safety (percentage of population using safely managed drinking water services, loss of biodiversity, climate change, etc. (Fattorini, 2025) About 50% of infectious diseases are due to  climate hazards (Mora et al. 2022)
• Socio-economic Indicators: Core drivers of health, including literacy rates, employment rates, income levels, and sanitation access (percentage of population using safely managed sanitation services).
• Health Policy & Quality of Life: Policy impact, perceived health of the population, Rigorous Public Health Surveillance (PHS) systems, Demographic characteristics, Proportion of (i) GDP spent on health services, (ii) total health resources devoted to primary health care, (iii) economically independent adult women, (iv) pregnant women receiving  adequate antenatal care checkups (ANC), etc.

The chosen indicators under dimensions are evaluated by actual secondary data or estimates; accuracy and adequacy of such data may be questioned. For example, in India, medically certified deaths constituted about 21% of registered deaths in 2019 (RGI, 2021).  Over reporting of data on ANC, institutional deliveries, etc. by various states of India was observed (Kole, 2019).  Outcome measures in clinical set up include Patient-reported outcomes (PROs) for disease specific or generic questionnaires using K-point items (ordinal data); Performance outcomes giving rise to ratio scale data (like time to complete a task) or count data like number of errors (Seashore Rhythm Test of HRB), observer-reported outcomes and clinician-reported outcomes. For the same disease, different types of outcome measures may be used.  Aggregation of indicators in different units and in different scales of measurement, following different unknown distributions is difficult. Mean, SD and item distribution are not comparable for Yes-No type, k-point items. Thus, scores of indicators and dimensions of health (as sum of scores of indicators) fail to differentiate samples and tools (Panagiotakos, 2009). Similar distribution of random variables X and Y, makes addition meaningful facilitating derivation of distribution of Z and undertaking parametric analysis (Chakrabartty, 2024). Moreover, variables expressed in ratios or percentages, in ordinal scale and categorical variables are not equidistant and arithmetic aggregations are not meaningful (Wakita, et al. 2012; Munshi, 2014). Health scores with equal importance to dimensions and dimension scores treating the indicators as equally important are not justified primarily due to different contributions of items/dimensions to Health score, different inter-indicator correlations, indicator-dimension correlations and factor loadings of the indicators and dimensions (Parkin et al.2010). Addition of scores of independent dimensions are not advisable. For example, total SF-36 score is not supported since several independent factors emerged from confirmatory factor analysis (CFA). Well-designed health index was suggested to unfold the complexity and varied contexts provided by different regions of a country with vulnerable groups, in addition to ranking of the regions/states (Kole, 2019). The paper suggests transformation of scores of biomarkers, item scores of scales to follow Normal distributions leading to normally distributed index of health care ( for an individual and  for a sample, reflecting overall health status satisfying desired properties including assessment of changes over time, facilitating parametric statistical inferences and better evaluation of psychometric properties.

Literature survey:

A number of health indices have been developed like WHO index of health system performance (SPRG, 2001), Global Health Security Index (GHI) 2021 (Haider et al. 2020)   Health Index (NITI Aayog, 2021), Government of India, India Health Index (IHI) (Sehgal et al. 2024), etc. primarily for assessment of existing capacities of countries and to prevent, detect, respond to outbreaks. Each index followed usual steps of collection of data on selected indicators (, pre-processing of data like addressing skew, normalization, and combining them to get dimension scores ( followed by aggregation of by weighted sum or different aggregation methods to compute value of the index. But, health index and component indicators for developed countries may not be relevant for developing countries with pronounced poverty, socio-economic inequality, constraint in accessing healthcare infrastructure and services due to disease-related stigma or inadequate Public Health Surveillance (PHS) system, etc.(Goli & Arokiasamy, 2014). Stigma and discrimination towards corpses and survivors of COVID-19 were common even among educated persons (Dar et. al. 2020). Similar disease-related stigma were also found on  diseases like AIDS, leprosy, autism, intestinal disorders, epilepsy, mental illnesses, etc. (Akbari et al. 2023). In absence of consensus regarding selection of indicators and dimensions, existing multidimensional health indices differ in number of indicators and dimensions, normalization procedures, aggregation methods and satisfaction of compelling properties of such indices. While NITI Health Index (NHI) considers 24 health performance indicators distributed over three dimensions namely, Health outcomes, Governance and Infrastructure, IHI is based on six dimensions with 29 indicators, GHI 2021 considers 37 indicators across six categories evaluated through 171 questions. Choice of indicators, methodology of construction of the indices and quality of data have been questioned. Illustrative limitations of the indices are as follows:

Selection of indicators: Questions have been raised regarding selection of indicators of health index. Observations in the context of NHIare as follows:

• Indicator like sex ratio at birth (SRB) is redundant since SRB is an indicator for gender disparity and not health outcome. The natural SRB (about 1.05) may remain unchanged irrespective of value of health index unless human interventions like sex-selective abortions preferring a particular gender take place in a massive way. 
• Total fertility rate (TFR) reflecting reproductive behaviour may not be a good indicator of population health. TFR in India has dropped from 4.05 births per woman in 1990 to 1.9 (urban 1.5, rural 2.1) in 2023 (Guru and Sharma, 2025), which is below the replacement level (RL) of 2.1 needed for a stable population. Despite TFR
• Consideration of proportion of vacant positions (regular and contractual) under healthcare providers, auxiliary nurse midwives (ANMs), etc. are misleading since a state with less number of sanctioned positions in less number of public health facilities may not have any vacant position resulting in overcrowding implying poor availability of healthcare, which goes against the mission of providing equitable, affordable & quality healthcare services. 
• Indicators like percentage of pregnant women with anaemia, percentage of women with low (BMI), per capita health expenditure, health care provided by private sector, etc. are not considered along with non-inclusion of mental health, infectious diseases, non-communicable diseases, governance, and protection of financial risk. Av. out-of-pocket expenditure per delivery was available for the reference year only.
• Indicators are not uniform across Large States, Small States and Union Territories (UTs).  For example, indicators like NRR, U5MR, TFR, SRB, etc. are not applicable for Smaller States and Union Territories (UTs). Thus, the index for larger states is not well comparable with the index for Smaller States and UTs based on number of indicators. 
• NITI Aayog uses estimated live births instead of actual registered live births, leading to inaccurate percentages and rankings. 
• IHI also ignores indicators like health outcomes for elderly people, mental health and its physical manifestations, burden of disabilities, etc. 
  • Change in can change marginal rates of substitution and rankings.

Selection of indicators depends primarily on the purposes like assessment and comparison of health status and health risks across sub-populations in different regions, tracking progress over time, monitoring effectiveness of health interventions, deciding priorities or serving as tools to assist in allocating resources, etc. Secondary data are to be combined with primary data through multipoint Likert type scales to assess: Physical Disability (Barthel Index, Functional Independence Measure, Modified Rankin Scale, etc.), Cognitive Impairment (Montreal Cognitive Assessment (MoCA), Mini-Mental State Examination (MMSE), Alzheimer's Disease Assessment Scale–Cognitive Subscale (ADAS-Cog) etc.), Low Speed of Information Processing (Symbol Digit Modalities Test (SDMT), Wechsler Adult Intelligence Scale-IV (WAIS-IV) Symbol Search/Coding, etc.), Behavioural symptoms (Neuropsychiatric Inventory (NPI), Cohen-Mansfield Agitation Inventory, etc.)

Transformations of data:

Raw score of the i-th indicator ( of GHI or NHI is normalized to ( by min-max transformation by: 

for positive indicator and 

       for negative indicator where . 

Indicators of IHI are log transformed to address positive skewness and then standardized to Z-scores with zero mean and unit variance which give negative values also. However, log-transformation need to be used with caution since value and directionof and were found to be different (Kovacevic, 2011). Similarly, and failed to ensure linearity among Alzheimer’s disease patients with mild cognitive impairment (Malek-Ahmadi, et al. 2013). Moreover, log-transformation does not ensure normal distribution (Curran-Everett, 2018). Other common transformations to reduce skew are: Square Root transformation, Cube Root transformation (for both positive and negative skewness), Box-Cox transformation to handle various degrees of skewness. Krylova et al. (2026) used square root and capping as follows:

(i) Indicators with skewedness 2 were transformed by square root, and indicators with skewness lying between 1.5 and 2 were capped at the 99th/1st percentile value

(ii) Difference between the indicator minima and the 1st percentile and the indicator maxima and the 99th percentile was computed. In case the difference >2 standard deviation (SD), cap was applied at the 99th/1st percentile value. Moreover, several indicators were capped based on theoretical boundaries. SF-36, a generic measure of health status transforms raw data to percentages before taking average (Ware et al. 2005). However, average of percentages is wrong, when An integer. The above said transformations do not ensure normality. Addition of skewed data even after transformations are problematic since square root transformation may not be adequate for extremely right-skewed data and arbitrary capping amounts to setting extreme values to a specific percentile, may distort the original data structure and create problems in interpretation of the index. 

For highly correlated indicators (, IHI dropped one indicator to tackle the problem of multicollinearity leading to possible loss of crucial information affecting adversely estimation of the regression parameters. For removing multicollinearity in data, Principal Component Analysis (PCA) may be used which combine one or more correlated predictor variables into a single or multiple variables known as principal components. Other machine learning models like Ridge Regression, Lasso Regression, or Elastic Net Regression can help to remove multicollinearity. 

Limitations of min-max transformation include:

• remains unchanged if is categorical taking values 0 or 1. 
• Fixed zero-point of in ratio scale get altered. 
• Sample specific could be unreliable outliers, taking different values for different years. 
• A change in can change ranking based on
• is not meaningful if is in ratio or percentage or in ordinal level. 
• is different for different values of X and  
• Indicators with small score-ranges are overestimated
• Can change shape of probability distributions of and distort the index (Mazziotta and Paretoa, 2021). 

Aggregation method: 

Existing health indices are obtained as weighted sum of normalized or transformed scores. However, different methods of normalization and different selection of weights can influence values of the indices and rankings in terms of the indices (Mhlanga and Lall, 2022). An index lacks aggregation consistency if index formulated in two stages (scaling and aggregation) index formed by direct aggregation (computation in single stage). Multi-criteria decision-making (MCDM) methods often avoid normalization. Different ways of choosing weights give different results. No selection of weights is beyond criticism. Weights could be assigned to indicators, dimensions, and regions. Weights to indicators performed better than weights to targets (Asadikia, et al. 2024). NHI assigned weights to the dimensions based on expert opinions which may introduce bias, and no weights to the indicators constituting a dimension. Sum of NHI weights and thus, fails to satisfy the convex property. Higher weights to larger states for Health Outcomes make the index biased towards larger states. No attempt was made by health indices to find variance of the weighted sum despite the backdrop of desirability of minimum variance of weighted sum to reduce data heterogeneity. Chakrabartty (2019) suggested positive weight vector satisfying and minimizing where is the weighted sum and is the variance –covariance matrix. The required weight vector is   where is the identity vector with each component is equal to 1. If are replaced by  , then becomes the correlation matrix, and the weight vector  satisfies   i.e. the weighted sum is equi-correlated with  for all . Implication is the indicators under a dimension are equi-correlated with score of the dimension and dimension scores can be equi-correlated with the index obtained by arithmetic aggregation.  The equi-correlated approach holds great promise of selecting weights to construction of composite index (CI) despite the fact that no weighting system is beyond criticism (Greco, et al.2019). Aggregation by giving equal weights to the indicators and dimensions have been questioned since it contradicts different contributions of the indicators to the index and it results in constant trade-off between a pair of indicators (Tofallis, 2014). Weights based on PCA ignore poorly correlated indictors with the composite index, even if such indicators are theoretically important. PCA weights varying across time and samples, assign more weight to variables showing larger variances. PCA method was disfavoured by (Nardo et al.2005). Health-status measuring tools like Nottingham Health Profile (NHP), SF-36, COOP/WONCA charts, EQ-5D-5L performed uniformly as "best" or "worst (Essink-Bot et al. 1987). IHI makes no attempt to assess changes of IHI over years. Both IHI and NHI suffer from methodological limitations in terms of normalization methods, methods of selecting weights and aggregation without ensuring aggregation consistency. Comparison of health performance with the previous year as the reference point is a limitation of NHI since benefits of interventions (like immunization, supplementary nutrition, etc.) takes long time and definitely not a year.  Cost-benefit analysis of health care interventions was suggested (Hyder et al. 2012). However, major problem is lack of consensus on the discount rate in health, which should be less than internal rate of return (IRR) used in financial evaluation of a project. Aggregation method in construction of CI need to avoid arbitrary choice of weights and normalization procedure satisfying desirable properties including aggregation across sub-groups (like regions, age/income categories, urban & rural, etc.) Validity of IHI is given as correlation between IHI and Subnational Human Development Index (SHDI) despite different number of independent factors being measured by IHI and SHDI. Similarly, reliability of multidimensional IHI by Cronbach alpha violates important assumptions of Cronbach alpha like uni-dimensionality and tau-equivalent.

It is felt desirable to include more dimensions and indicators and aggregate them avoiding normalization and selection of weights. 

Proposed method:

Ensuring each item is positively related to the construct and assigning 1, 2, 3, 4, 5, etc. to the response-categories of multi-point items avoiding zero, Chakrabartty, (2025) suggested a method to transform ordinal score of i-th item to equidistant scores () using weights () to j-th level of the i-th item satisfying and = 2, 3, 4, 5, ……; followed by and proposed item score by linear transformation where 1. Score of a dimension (is taken as following normal distribution with mean and SD =. The scale score ( is taken as Ordinal item scores of the chosen scales can be converted to -scores. Count data, ratio scale data and biomarkers can be straight-away standardized and transformed to normally distributed indicator score . The index for i-th individual is sum of all corresponding of ordinal item/task scores and biomarkers including count data. The index  for a sample is the average of all ’s. Here, each index  and follows normal distribution and parameters of which can be found from the data since each is a convolution of normally distributed scores.

Properties:

can be computed by combining pathological markers, secondary data and scores of several scales, irrespective of their formats and correlations. Properties satisfied by the indices are: 

• Continuous and monotonically increasing
• Zero value of E-scores corresponds to =0. 
• Avoid skew and give unique ranks to the regions.
• Can be computed separately for each socio-economic-demographic factor.
• can be broken into covering clinical data and for ordinal and secondary data.

Benefits

The dimensions of can be ranked with respect to relative importance of the dimensions given by  . The indicators can also be ranked similarly by . Ranking of dimensions of can be undertaken by ratio of change in due to unit change in a dimension. Public health priorities of the country can be decided on the basis of relative importance of the dimensions/indicators. Progress registered by i-th region in consecutive time-periods is given by , positive value of which also indicates effectiveness of adopted interventions (assumed low score severity for each variable). Progress of can be assessed similarly. Dimension(s) showing deteriorations are critical and require initiation of necessary corrective interventions for proactive strategies to prevent disease, promote health, and improve environmental conditions.

Path of progress/deterioration of and across time can be plotted using longitudinal data. Significance of progress can be tested by or 0 by test since ratio of two normally distributed statistics follow Point bi-serial correlation between categorical variable (X) (say gender) and the continuous variable (Y) can be computed by  

 where for the group with X=1 (sample size ) 

  for the group with X=0 (sample size); ???? = +

  and  

For = 0, the test statistics is t-distribution with (df.

Normality of and facilitate:

• Testing equality of  for two regions or a single region at different time periods using cross-sectional or longitudinal data, across demographic variables like gender, age, income levels, educational attainments, etc. by t-ratios or ANOVA. 
• Given score of  for region-1 with normal pdf is equivalent to for region-2 with normal pdf if , even if scales are of different formats, containing different dimensions. Chakrabartty, (2021) solved the equation using table. The concept of equivalent scores help to find equivalent cut-off scores of two or more regions for the purpose of classification. 
• Enables PCA and finding the eigenvalues and computing factorial validity (FV) as from single administration of a multidimensional tool reflecting the main factor for which it was developed (Parkerson et al. 2013). Max. Cronbach’s alpha () = (Ten Berge and Hofstee, 1999) is related with FV as 

= = for a test with m-number of items, where. Clearly, higher increases . However, FV needs to tally with clinical findings.

• Normality of facilitates population estimates of Cronbach’s alpha for the battery used to assess health status avoiding complex methods of Heo et al (2015) assuming parallel measures and involving estimation of unbiased sample covariance matrix; variance-covariance matrix of the population.
• Linear regression equation of on different chosen dimensions can be fitted where the coefficient may indicate relative importance of the i-th dimension in predicting Stepwise regression method is preferred to have a smaller set of dimensions for prediction of . 
• Two-way Analysis of variance (ANOVA) of with a number of categorical variables with multiple levels helps to detect significant difference between groups, including effects at the cross-levels or third order cross-levels. 
• To know the extent to which predictors of differ across sub-groups formed by categorical variables like gender, income, educational levels, etc., Covariance Structure Analysis (CSA) can be used to test the equality or invariance of effects across groups, using test for comparing the model with and without parameters freed across groups, after checking assumption of multivariate normal distribution. 

Overcoming normalization and weighting imitations, the index 〖I_Health〗_i  combines effectively biomarkers, secondary and primary data in a simple way avoiding complex approaches like supportive vector machine, logistic regression, Bayes classifier, random forest, decision tree algorithms, etc., each of which is associated with a number of assumptions which demand verification before application.  The sub-indices I_(〖Health〗_(Pathological Markers) ) and I_(〖Health〗_T ) help in diagnosis, assessment of disease stages evaluation of prognosis and health status. The index I_(Health )as sum of dimension- scores may be taken as a cornerstone of a nation’s economic development and effective governance. I_(Health )can also be computed separately for different regions enabling ranking of the regions. 
The index I_(Health )based on cognitive dysfunctions, biomarkers may reflect the health dynamics by assessing changes in overall health status of one or a group of individuals along a disease continuum across categorical variables including stages of disease, etc. Such path may also help interregional comparisons over time with respect to the index. In addition, ANOVA, Covariance Structure Analysis of the normally distributed index can help to test equality or invariance of factors across groups, using χ^2-test. Max. Cronbach’s alpha (α_PCA) and factorial validity of I_(Health )can be found in terms of eigenvalues and estimate theoretical reliability r_(tt(theoretical))  =(S_T^2)/(S_X^2 )=  (S_T^2  )/(λ_1/FV+2∑_(i≠j=1)^m▒〖Cov(X_i,X_j)〗). Such relationships may help to find optimal value of one psychometric parameter to maximize another parameter. High value of reliability implies rank robustness. 
 

The proposed approach did not consider cases with missing values which is beyond the scope of the present article.

The proposed index contributes to improve scoring of tests and QOL tools avoiding limitations of ordinal/categorical scores and facilitating parametric analysis for meaningful comparisons, classification, and integration of various scales. Economic indicators like GDP or per capita GDP or suitably chosen social indicator of a country can be regressed on the index  at different years to see fluctuations in such relationships. Computation of separate index is recommended for vulnerable groups like old aged citizens, postpartum women running the risk of maternal mortality, economically backward class, nomadic, semi-nomadic and De-Notified tribes, sanitation workers, people with substance use disorder, people living in poverty-stricken areas; etc. Future empirical investigations may explore properties of proposed method along with generalization of findings including psychometric properties of the proposed transformation. Combining ordinal PRO-scores and indicators measured in ratio scale irrespective of PRO-formats and inter-correlations among the indicators with clear theoretical advantages is a novelty and is recommended. Practicing physicians, planners and researchers may derive benefits of the proposed combined scores for meaningful comparisons, better assessment of progress or deterioration, identification of critical indicators for better prognosis, and establishing empirical relationship among the biomarkers and the outcomes. Future empirical studies may be taken up with real life data to describe salient feathers of the proposed index along with the equi-correlated approach for weighted sum of normally distributed indicators which holds great promise to construction of composite index.

Acknowledgement: Nil 

Funding details: Nil 

Conflict of interests: No potential conflict of interest is reported. 

Informed Consent: Not applicable

Ethical Compliance with Human/Animal Study: Not applicable

Data availability: No data set used

Authors' contributions: The single author is involved in Conceptualization, Methodology, 

Writing- Original draft preparation, Writing- Reviewing and Editing.