Development and Validation of a Symptom Scale for Indian Patients With Medically Unexplained Physical Symptoms

Introduction Patients with medically unexplained physical symptoms (MUPS) account for a substantial proportion of patients visiting the outpatient department. Diagnosis of MUPS is a challenge for most physicians. An accurate diagnosis relies on obtaining a detailed history from patients regarding the nature of their symptoms, their onset, and any associated aggravating or relieving factors. This study aims to develop a symptom scale for Indian patients with MUPS. Methods The study had a mixed-method study design. Phase 1 involved designing the questionnaire using qualitative techniques, such as literature reviews, focus-group discussions, expert evaluation, and pre-testing of a Hindi and English language Likert-rated interviewer-administered scale. In phase 2, the construct validity of the questionnaire was established using quantitative approaches among 116 patients diagnosed with MUPS. Results The final questionnaire consists of 38 items, with good internal consistency (Cronbach 𝛂 = 0.916). Confirmation sampling adequacy for factor analysis was done using the Kaiser-Meyer-Olkin test (KMO value = 0.792) and Bartlett’s test of sphericity (p < 0.001). The newly developed scale showed a Pearson correlation coefficient of 0.568 (p < 0.001) with Patient Health Questionnaire (PHQ)-15 scores. Conclusion A reliable and valid tool has been developed to assess patients’ symptoms with MUPS in English and Hindi languages. This questionnaire can be used for assessment, screening, and diagnostic purposes as well as to chart longitudinal changes in patients with MUPS.


Introduction
Medically unexplained physical symptom (MUPS) is an umbrella term for physical symptoms with little or no identified medical cause.According to a study conducted in the medicine outpatient department at a hospital in India by Baitha et al., the prevalence of MUPS was 24.6% among patients aged 18-60 years, with a slight female preponderance [1,2].
The diagnosis of MUPS relies on the absence of identifiable medical explanations despite thorough evaluation, making it a diagnosis of exclusion.Medical history and clinical features aid in diagnosing specific diseases, but in the case of MUPS, they primarily help exclude explainable medical illnesses.In different cultures, patients present with different symptoms, and the ability of subjects to differentiate between psychogenic and medical symptoms may be affected by the practices of local physicians.Patients also have their own culturally supported theories for the etiology of their symptoms [3,4].
Several instruments are available at present for the diagnosis of MUPS, most of which have been developed for patients from Western countries, and even though translated versions of these scales exist, their cultural validity remains questionable.Hence, there is a need to develop a scale based on local data to help us better understand our patients presenting with MUPS.The purpose of this study is to address this significant constraint by developing a user-friendly symptom scale for Indian patients diagnosed with MUPS, which can be easily administered, takes minimal time and expertise, and is accessible in Hindi and English languages.

Materials And Methods
A mixed-method study design was used to develop and validate a symptom scale for patients with MUPS.The study was conducted after prior approval from the institute ethics committee (IEC PG-293/22.07.2020).The workflow of the study is given in Figure 1.The process was divided into the following two phases:

Phase 1: development of the construct of scale
Step 1: Literature Search and Collection of a Pool of "Etic" Items (Etic items refer to universal, culturally impartial components, concepts, or phenomena that apply across different cultures or contexts.) The literature search was done using PubMed and Google through a combination of search terms such as "somatoform disorder," "somatic symptom disorder," "functional somatic symptoms," "medically unexplained symptoms," "medically unexplained physical symptoms," questionnaire, scale, and tool.
Relevant articles were screened through their titles and abstracts, and items were generated after carefully reading them.
Step 2: Generation of "Emic" Items or the Local Idioms Through Focus Group Discussions (Emic items represent culturally specific elements, perceptions, or interpretations unique to a particular culture, context, or group of people.) Eleven focus group discussions (FGDs) were carried out involving patients with MUPS, their caregivers, and healthcare workers (HCWs).A total of 36 participants, including 12 patients, 10 caregivers, and 14 HCWs, were included.The sessions were recorded and transcribed verbatim, and important points were identified by the experts (PR and AS) from the transcriptions and coded.The codes identified were classified based on the underlying theme and sub-theme, and additional items were added to the item pool.The final list of items was then used to formulate the initial construct of the scale, which was simultaneously translated into Hindi language by the language expert.

Phase 2: validation of the scale
Step 3: Content and Face Validation After the preparation of a draft of the scale, face and content validity was checked through feedback from 15 experts from premier institutes and hospitals of the country working in the field of MUPS.The content validity of the developed instrument was assessed quantitatively through the calculation of the content validity ratio (CVR) and content validity index (CVI).The critical value of CVR and CVI depends on the total number of participants and is determined using Lawshe's table.The formula for calculating the CVR is CVR = (Ne − N/2)/(N/2), where Ne is the number of participants who consider an item essential, and N is the total number of participants.Thus, all items with a CVR greater than 0.6 were included in the scale, and the remaining items were reviewed and retained if they were felt to be important [5].
The questionnaire was then pre-tested on 30 participants from the OPD in their preferred language, and they were asked to mark the questions as clear (1) or unclear (0).The required changes were made in the draft of the questionnaire, such as the reduction of technical terms and incorporation of more user-friendly terms to increase the ease of the participants filling out the questionnaire.

Step 4: Construct Validation and Assessment of Reliability
One hundred sixteen (N = 116) participants diagnosed with MUPS were informed about the study, and written informed consent was obtained.Data pertaining to socio-demographics and patient history were also recorded.Psychiatric comorbidities were excluded using MINI.Thereafter, PHQ-15 was administered.Subsequently, the newly developed MUPS symptom scale was given to participants to obtain their responses.

Statistical analysis
The obtained data were analyzed using IBM SPSS Statistics, version 28.0 (IBM Corp., Armonk, NY).Quantitative variables were expressed as mean and standard deviation for normally distributed data and median with interquartile range in case of skewed data.Categorical variables were expressed as frequencies and proportions.Kaiser-Meyer-Olkin test and Bartlett's test of sphericity were used to ensure that the sample size was adequate for factor analysis.Exploratory factor analysis (EFA) with a varimax rotation and Kaiser normalization was used to identify domains and establish construct validity.For determining reliability, Cronbach's alpha was used.For calculating the degree of correlation between scores on our scale and PHQ-15 scores, Pearson's correlation was calculated.

Socio-demographic characteristics
The mean age of participants was 38.2 ± 9.74 years, with a predominance of females.The duration of symptoms among patients was highly variable, with a range of 1-240 months, and the mean and median duration of symptoms were 29.2 and 24 months, respectively.The detailed socio-demographic characteristics are given in Table

Disease severity
Patient Health Questionnaire-15 (PHQ-15) [6] was used to assess the severity of somatic symptoms experienced by the patients.Each symptom was rated on a scale of 0-2 based on how much the patient was bothered by that symptom.Based on the total score obtained by adding the score for each symptom, patients were classified as having minimal, low, medium, or high severity of somatic symptoms.
Most of the patients included in this study (nearly 50%) had a high severity of somatic symptoms as per PHQ-15 scores.The mean and median PHQ-15 scores were 15.7 ± 6.0 and 15, respectively.

The construct of the questionnaire
Articles about the development, assessment, or review of somatic symptoms or somatization questionnaires were chosen.A total of 17 articles were identified after a comprehensive literature search.The reviewed scales had a variable number of items ranging from 6 to 53.The domains assessed by the scales ranged from 3 to 9. Twelve of the scales were self-report questionnaires, whereas four were interview-based.Only one scale (Brief Symptom Inventory) had both self-report and interview response formats.The different scales covered different time frames ranging from one week to six months, as presented in Table 2.

S.
no.

Ghent
Multidimensional Questionnaire [12] 18 items; English, Dutch, & Turkish Self-report; the frequency of each item is rated on an 8-point Likert-type scale: 0, never;  The cut-off score for the somatization subscale as per ROC analysis was 1.5, with a sensitivity of 72% and specificity of 66%.

Swartz Somatization
Index (SSI) [22] 11 items; English Interview Lifetime 97.6% of the patients with somatization disorder were correctly identified when at least 5 out of 11 symptoms were present, while the presence of a minimum of 5 symptoms wrongly identified 150 patients (correct negative correspondence, 99%; total error rate, 1.0%).
No Some scales included symptoms occurring over the patient's lifetime.A total of 163 symptoms were identified, out of which 130 items remained after removing items that were common across different scales.
Eleven FGDs involving patients, caregivers, and healthcare professionals were conducted through both online and offline platforms.The discussions were guided by using semi-structured questions.Thematic analysis was conducted using QSR Nvivo software, resulting in the identification of eight themes, as detailed in Table 3.This led to the addition of 16 new items, which were integrated into the existing list of 130 items, bringing the total number of items to 146.

Face Validity
Face validation was done through expert feedback and pilot testing on a representative sample of the target population of 30 participants.This method of using both expert judges and the target population is considered to be ideal.One item was eliminated during the pilot testing phase due to redundancy with another item in the scale, resulting in 40 items under review.

Construct Validity
The construct validity of the newly developed scale was assessed by comparing its scores with those of the PHQ-15, a well-established measure for gauging somatic symptom severity.The scale exhibited a moderately strong correlation, with a Pearson correlation coefficient of 0.568.Notably, the correlation between our scale and PHQ-15 scores was significantly higher than that of the SSD-12 subscale (a correlation coefficient of 0.47 with PHQ-15) [16].
In our study, EFA with a varimax rotation and Kaiser normalization was utilized to determine the scale's factor structure, differing from other studies that typically employed confirmatory factor analysis.Our EFA revealed an 11-factor structure explaining 68.5% of the total variance.This finding aligns with previous observations during the development of the Bradford Somatic Inventory (BSI), which identified a 13-factor structure explaining 65.8% of the variance.During this process, one item was excluded due to a Kaiser-Meyer-Olkin (KMO) measure below 0.5, indicating inadequate sampling adequacy, leading to the retention of 39 items.After a varimax rotation, one item failed to load onto any factor and was subsequently removed, resulting in a final scale comprising 38 items.

Reliability
The Cronbach's alpha for the newly developed scale was calculated to be 0.916, indicating good reliability.The response of the patient should be noted for the following: 1.If a symptom is experienced by the patient on most of the days in the last three months: yes/no 2. If yes, the severity of the symptom is on a scale of 1-3, where 1 is mild, 2 is moderate, and 3 is severe.

Discussion
The initial pool of items in the scale was derived via both the deductive method (review of literature and evaluation of existing scales on MUPS) and the inductive method (item generation from the responses of individuals through focused group discussions and key individual interviews).This was done to minimize the chances of any important item being missed out and is a major strength of our study.This is in contrast to the Depression and Somatic Symptom Scale (DSSS) [11], PHQ-15, and Ghent Multidimensional Symptom Questionnaire (GMSQ) [12], all of which are based only on items derived from existing scales.The BSI [9], on the other hand, is based only on symptoms derived from case notes of patients visiting hospitals.This scale has been developed with an interview-based response format.Most of the existing scales have a self-report response format, which is inappropriate in our setting as most of our patients are uneducated.Only the Schwartz Somatization Index [10], Somatic Symptom Index [15], Somatoform Disorders Schedule (SDS) [13], and Scale for Assessment of Somatic Symptoms (SASS) [7] are available in interview-based response format, whereas the Brief Symptom Inventory is available in both interview and self-report response format [22].
The scale has been developed in both Hindi and English languages.To the best of our knowledge, none of the existing scales for the diagnosis or evaluation of MUPS is available in Hindi, which is one of the most commonly spoken languages in India.The scale takes into account medically unexplained symptoms occurring on most of the days in the last three months.Most of the existing scales we reviewed included symptoms over a variable period, ranging from one week to a lifetime.We chose three months because most of our patients report chronic symptoms, and symptoms of a shorter duration are more likely to be organic.Our scale consists of a total of 38 items, which ensures that the scale is comprehensive but, at the same time, is not too lengthy.In comparison, the scales that we reviewed have several symptoms ranging from 8 for Somatic Symptom Scale-8 (SSS-8) [8] to 53 for SDS and Screening for Somatoform Symptoms 7 (SOMS-7) [23].Our scale requires that the severity of each symptom be rated on a three-point scale corresponding to mild, moderate, and severe.This is similar to the response format used in DASS and SASS.Most of the other scales have used a Likert-type response format with different numbers of points.In conclusion, this newly developed scale is a reliable and valid tool for the Indian population.
The reader may refer to the MD thesis (click URL: http://dx.doi.org/10.13140/RG.2.2.20957.63202)for a comprehensive overview of the methodology employed and the results obtained in developing the MUPS Scale for assessing patient symptoms.

Conclusions
The newly developed scale is a user-friendly diagnostic scale for Indian patients with MUPS.The scale, available in Hindi and English, is designed to be easily administered, quick, and suitable for Indian patients.The scale demonstrates good internal consistency and validity.It serves as a valuable tool for clinicians, offering accurate diagnosis and appropriate treatment for Indian patients with MUPS.
However, our study has some important limitations, the most important being a relatively small sample size, and the scale needs to be evaluated further on larger samples of patients before it becomes available for widespread use by clinicians.The newly developed scale aims to evaluate symptoms and severity levels in patients with MUPS but lacks validation as a diagnostic questionnaire.Therefore, it cannot be directly compared to validated diagnostic tools.This limitation restricts the scale's utility for diagnostic purposes and necessitates caution in interpreting its results.Future research should focus on validating the scale to enhance its reliability and validity in clinical settings.

FIGURE 1 :
FIGURE 1: Workflow of the study MUPS, medically unexplained physical symptoms; OPD, out-patient department; FGDs, focus group discussions

15 days during the past month 1 month
; three-choice response format: absent, present on less than 15 days during the past month, and present on more than Linguistic equivalence of English and Urdu versions: median percentage agreement among the 46 items was 92.5% between English and Urdu responses.Principal components analysis yielded an 8-factor structure that explained 53% of the total variance.only); English Self-report; each item is rated on a 4-point scale; range: 0, not at all; 4, very severe 2 years Cronbach alpha = 0.92; both indices (number of symptoms and severity index) decreased significantly post-treatment.The test-retest correlation coefficient of 0.76 for symptom count and 0.71 for severity index.alpha: DSSS and SS were >0.8.Test-retest reliability: there is a good correlation between scores at the index visit and after one week (Pearson's correlation coefficient, 0.92).Treatment sensitivity: scores on DSSS decreased significantly post-treatment.No 7.

TABLE 2 : Salient features of existing scales available for the diagnosis of MUPS/somatic symptom/somatoform disorder
MUPS, medically unexplained physical symptoms

no. Theme/subthemes identified Number of codes generated
S.

TABLE 4 : Calculated CVR and CVI for each item as per expert feedback
The newly devised scale consists of 38 items in English and Hindi and can be accessed through the following link: (http://dx.doi.org/10.13140/RG.2.2.34720.19208).The English version of the scale is shown in Table5.
To feel as if there is swelling in the entire body or a part of it 2024 Aakansha et al.Cureus 16(3): e55956.DOI 10.7759/cureus.5595612 of 16