|Year : 2019 | Volume
| Issue : 2 | Page : 72-80
Retrospective estimation of prevalence and likelihood ratio of general symptoms of 29 less frequently prescribed homoeopathic medicines by clinical verification
Raj K Manchanda1, Parthasarathi Chakraborty2, Munmun Koley2, Dinesh Singh3, Pramodji Singh4, Ojit Singh5, Varanasi Gayatri Prasad6, Jai Prakash Singh7, Surinder Singh Nain4, Prasant Kumar Pradhan8, Sunil S Ramteke3, Ramesh Bawaskar9, Lipi Pushpa Debata7, Subhranil Saha7, Chaturbhuja Nayak10, Lex Rutten11
1 Central Council for Research in Homoeopathy, New Delhi, India
2 Dr. Anjali Chatterjee Regional Research Institute (Homoeopathy), Kolkata, West Bengal, India
3 Regional Research Institute for Homoeopathy, Shimla, Himachal Pradesh, India
4 Central Research Institute (Homoeopathy), Noida, Uttar Pradesh, India
5 Regional Research Institute for Homoeopathy, Imphal, Manipur, India
6 Regional Research Institute for Homoeopathy, Gudivada, Andhra Pradesh, India
7 Homoeopathic Drug Research Institute, Lucknow, Uttar Pradesh, India
8 Regional Research Institute for Homoeopathy, Puri, Odisha, India
9 Regional Research Institute for Homoeopathy, Mumbai, Maharashtra, India
10 Homoeopathy University, Jaipur, Rajasthan, India
11 Independent Researcher, VHAN, Dutch Association of Homeopathic Physicians, Breda, The Netherlands
|Date of Submission||20-Nov-2018|
|Date of Acceptance||20-May-2019|
|Date of Web Publication||27-Jun-2019|
Dr. Parthasarathi Chakraborty
S-4, Dr. Anjali Chatterjee Regional Research Institute (Homoeopathy), Kolkata, West Bengal
Source of Support: None, Conflict of Interest: None
Context: Scientific assessment of proving symptoms or already recorded symptoms in patients refers to Clinical Verification (CV). There are serious shortcomings of existing methods of CV, mostly arising from qualitative observations made on absolute occurrence of symptoms instead of relative ones. This problem can be resolved by the use of the Likelihood Ratio (LR). Aim: This study aims to estimate the prevalence and LRs of general symptoms of 29 less frequently prescribed homoeopathic medicines. Methods: The study was multicentric, open and observational. Patients were enrolled as per pre-specified eligibility criteria. Alongside, presenting complaints and general symptoms were taken into account during prescription. Medicines were prescribed in centesimal potencies. LR calculation of general symptoms was based on clinician-rated outcomes as 'improved' and 'not improved' of presenting complaints and was divided into three sections: (1) when the prevalence of a symptom was available from literature, LR was calculated by assessing the prevalence of that symptom in the responder sample, (2) in the absence of so, calculation was restricted to mean prevalence data from study sample ('confined LR') and (3) 'confined LRs' were not calculated for symptoms whose prevalence were not recorded for at least five medicines and were kept for estimation in future. Results: Of 166 general symptoms of 29 medicines, LRs and confined LRs >1.5 were elicited for 6 and 49 symptoms, respectively. Conclusion: In spite of considerable caveats, it is the first insight into prevalence and LRs of general symptoms of less frequently prescribed homoeopathic medicines. Further research is warranted.
Keywords: Clinical verification, Homoeopathy, Likelihood ratio, Prevalence, Prognostic factor research
|How to cite this article:|
Manchanda RK, Chakraborty P, Koley M, Singh D, Singh P, Singh O, Prasad VG, Singh JP, Nain SS, Pradhan PK, Ramteke SS, Bawaskar R, Debata LP, Saha S, Nayak C, Rutten L. Retrospective estimation of prevalence and likelihood ratio of general symptoms of 29 less frequently prescribed homoeopathic medicines by clinical verification. Indian J Res Homoeopathy 2019;13:72-80
|How to cite this URL:|
Manchanda RK, Chakraborty P, Koley M, Singh D, Singh P, Singh O, Prasad VG, Singh JP, Nain SS, Pradhan PK, Ramteke SS, Bawaskar R, Debata LP, Saha S, Nayak C, Rutten L. Retrospective estimation of prevalence and likelihood ratio of general symptoms of 29 less frequently prescribed homoeopathic medicines by clinical verification. Indian J Res Homoeopathy [serial online] 2019 [cited 2020 Sep 21];13:72-80. Available from: http://www.ijrh.org/text.asp?2019/13/2/72/261642
| Introduction|| |
Clinical Verification (CV) refers to the confirmatory and systematic process of observing symptoms that are already recorded in homoeopathic literature as proving symptoms and cured cases. If these already recorded symptoms are indeed seen in cases that respond well to corresponding medicines, this confirms the relevance of the symptom in relation to the medicine. CV has been one of the most important tasks in the field of Homoeopathy. It is a process of internal validation of the basic principles of Homoeopathy, and the results can be used for improving the daily practice. During CV, every new symptom (if any) is also recorded and included in literature. In Homoeopathy, repertories are consulted to shortlist the medicines which are indicated for the presenting symptoms. To arrange the medicines under different rubrics and sub-rubrics, different typefaces are used according to their frequency of occurrence. These typefaces help to differentiate and signify the relative importance of the medicines for that particular rubric. Among these criteria of gradation system, CV plays a major role. For example, in Kent's repertory, there are three grades. The criteria for gradations are as follows:
- Bold (3 marks)/1st grade: Found frequently in 'all or the majority' of provers, confirmed by reproving and verified clinically on the sick
- Italics (2 marks)/2nd grade: Symptoms found in few provers, confirmed by reproving, but occasionally verified clinically on the sick; and
- Plain Roman (1 mark)/3rd grade: Symptoms experienced 'now and then' in proving, not yet confirmed by reproving but verified by curing patients – so accepted as clinical symptom only.
This gradation system involves two sources of knowledge: experimental testing in proving and observational knowledge from practical experience. Hitherto, there has been no reflection about the differences between these two sources, especially the influence of mere chance. For instances: the number of provers is limited, and sensitivity varies. The number of cured patients grows much faster but is variable for different medicines. Cured patients constitute a quite different selection from the whole population than provers. This introduces some serious shortcomings of these existing repertories – qualitative observations were made on absolute occurrence of symptoms instead of relative grading; based on very few expert opinions; again those opinions were mostly on memory based, not systematically collected. Influence of chance, inadequate sample sizes and recall or expectation bias have not been addressed adequately. It is quite difficult to differentiate among the indicated medicines listed under same gradations. Furthermore, commonly used medicines were verified and recorded more frequently and thus over-represented; whereas rarely used medicines were under-reported, even the most characteristic symptoms were overlooked. Large rubrics had unnecessary entries and small rubrics were deprived of medicines.
These types of inadequacies and misrepresentations can be addressed by reconsideration of repertorial entries by Bayes' theorem,, that is, mathematically expressed as a Likelihood Ratio (LR). Bayes' theorem, published in 1763, deals with predictions from experience in the past. LR is the modern epidemiological tool for determining the characteristic and keynote symptoms of medicines., LR comes under two variants. The first variant is:
Posterior odds = LR × prior odds
Odds = Chance/(1 − chance);
Chance = Odds/(1 + odds)
LR = (Prevalence in the target population)/(prevalence in the remainder of the population),
And the second variant is:
Here, M = medicine, S = symptoms. The higher score of LR will increase the posterior chance more. And with the addition of each symptom, the (posterior) chances that medicine will work will be increased by each added symptom. Chance of improvement by the medicine will increase if the prevalence of the symptoms is more in medicine responder population than in the rest of the population. A large number of prospectively collected accurate data from successful prescription is needed for comparison. Using LR in CV allows verification with the enhancement of the grades of medicine, eliminate bias, validate rubrics and medicines and show the importance of symptoms in relation to medicine.,, This gives a definite indication for symptoms requirement for a case with the curative probability of medicines. Positive LR is used to calculate changes in odds (or chances) if the symptom is present. Similarly, negative LR is used to calculate changes in odds (or chances) if the symptom is absent. Odds become greater when LR >1 and smaller when LR <1 (between 0 and 1); higher is better to include and lower is better to exclude. As a rule of thumb, per symptoms or rubrics, LRs <1.5 with corresponding medicines are discarded from pick listing, also because LR values between 1.0 and 1.5 hardly change posterior probability.,
A prospective verification study on Veratrum album rubrics from synthesis repertory showed that in some rubrics, typefaces of the medicines should be upgraded, but no new entry and elimination was found. Another retrospective study was conducted to evaluate LRs of Lycopodium clavatum symptoms on 752 patients. Only 22/35 symptoms were confirmed as pertaining to the medicine. A multicentric, prospective study on the most frequently encountered physical general symptoms illustrated that many medicines maintained their respective positions under corresponding rubrics; whereas some frequently used medicines having high gradations could not even qualify or just managed to get an entry under their respective rubrics. Findings of these previous studies indicated that all the repertory rubrics could be improved by the Prognostic Factor Research (PFR).
In this multicentric study, the investigators intended to estimate the prevalence and LRs of symptoms of 29 less frequently prescribed homoeopathic medicines. As rarely used medicines remain under-evaluated due to less frequency of prescriptions, this study may help in better understanding of the symptoms of the medicines.
| Methods|| |
This study was conducted on 29 drugs with analyzable data, from 2005 to 2010 [Table 1]. Patients for the study were enrolled from the outpatient departments of 11 institutes of the Central Council for Research in Homoeopathy (CCRH). As per the inclusion criteria, the patients from all age groups and both sexes, having symptomatic similarity with the 29 study medicines, and willing to participate were included in the study. If the patients were taking any acute medicine, they were included in the study after a washout period of 1 week. Exclusion criteria were patients unwilling to participate, patients having a clinical presentation not corresponding with the study medicines, and patients on regular medication for any systemic disease. Ethical clearance for the study was taken from ethical committee of the Council. After providing patient information sheet in local vernaculars, informed written consent was obtained from the eligible participants or the guardians in case of minors before participation in the study.
|Table 1: List of the 29 medicines evaluated under the clinical verification programme during 2005-10|
Click here to view
The study medicines were procured from a good manufacturing practice-compliant homoeopathic pharmacy in various potencies, namely, 6C, 30C, 200C and 1M and were distributed to all the 11 institutes. After that, the symptoms were repertorised using a repertory manually prepared for CV programme of the CCRH (unpublished), and then, a specially developed Materia Medica was consulted for the final selection of the remedy. Two examples of such repertory rubrics are given below [Table 2].
|Table 2: Example of rubrics taken from the unpublished repertory of clinical verification programme of CCRH|
Click here to view
Clinician-rated outcomes were described as 'Cured', 'Improved', 'Not Improved', 'Worse', 'Referred', 'Withdrawal', and 'Drop out', used to measure degree of changes in presenting complaints. Any additional and clinical symptoms if present were recorded separately along with systematic recording of physical general and mental general symptoms. If the presenting symptoms of the patient corresponded with any one of the study medicine, then the medicine was prescribed in 6C potency, thrice a day till the improvement or aggravation occurred or for maximum 5–7 days allowing the medicine to act. The medicine was served by the pharmacy of corresponding institute. In follow-up visits, the changes in signs and symptoms were noted. If there was any sign of improvement, then placebo was prescribed. If there was status quo, next higher potency i.e., 30C twice a day for 3–5 days in acute cases and 5–7 days in chronic cases; 200C once a week for 2 weeks and 1M potency was given once a fortnight. These potencies were repeated twice only. If adequate responses were not elicited, the cases were restudied, and next higher potency was prescribed. If no change was observed even after the change of potencies also, then the case was closed and considered as a clinical failure, that is, cases showing “not improved”, “worse”, and “referred” were considered as clinical failure. If the patient presented with new symptoms of mild intensity, placebo was prescribed. The appearance of severe symptoms (new or aggravation of existing symptoms) with sufficient strength to cause considerable discomfort to the patient called for a change of medicine or therapy. Such case was considered as a deteriorated one. 'Clinical success' was defined 'a priori' as cases showing clinical improvement, objective or subjective, of the present complaint(s) as judged by the investigating physician(s) and/or as reported by the patient(s). All the data, including clinical failure, were collected and compiled in specially designed Excel Spreadsheet for analysis and estimation of LR. Although there was provision for assessing both the acute and chronic cases in the CV project, for the purpose of LR calculation of general symptoms, we segregated the chronic cases only, where general symptoms were noted down properly.
The calculation of LR was divided into three sections:
- When the prevalence of a symptom under question was available from literature (i.e., the denominator), prevalence of that symptom in the responder sample of the study was estimated; thus, enabling calculation of LR
- If no concerned literature data were available, then, the calculation was restricted to the sample with available data only; hence called as 'confined LR' i.e., the prevalence of the symptom in responder population/prevalence of the symptom in rest of the study (not general) sample where the symptoms were recorded. The denominator was actually the mean prevalence data; mean calculated from the recorded study sample data
- However, as a rule of thumb, 'confined LRs' were not calculated for those symptoms whose prevalence was not recorded for at least five medicines. For such symptoms, only prevalence was reported, and LRs were kept to be estimated in future.
As the denominator depended either on literature data or mean prevalence of study sample, naturally, it was changeable. Thus, denominator value was not merely an assumption, rather based on either valid reference or sound statistical grounds.
| Results|| |
Alongside targeted particular symptoms, a total of 166 general symptoms were evaluated under this CV programme during 2005–2010. A total of 4652 records were considered for analysis, of which 3705 were improved and 947 were not improved. Among these, LRs(+) were calculated for 12 symptoms using prevalence of symptoms in the whole population from a previous symptom assessment in India (West Bengal) in 2039 patients for thermal relations and 4715 patients for food desires/aversions. The prevalence of symptoms being assessed was obtained by dividing the number of patients presenting with the symptoms by the total recorded data. Out of these 12, only 6 symptoms (prevalence given along with 95% confidence interval after each symptom within parenthesis) were identified as having corresponding medicines (LR given after each medicine within parenthesis) with LRs(+) more than 1.5 and worth of consideration with their corresponding prevalence in the study sample [Table 3].
|Table 3: Six rubrics along with list of medicines having likelihood ratio of more than 1.5|
Click here to view
Confined LRs were calculated for 57 symptoms, based on the mean prevalence of symptoms in general population. For these symptoms, 'LR' is only valid as a comparison between medicines samples with available data, hence the term 'confined'. For example, 'Desire for milk' was recorded only for six medicines – Aranea diadema, Glycyrrhiza glabra, Acid butyricum, Magnesia sulphurica, Acalypha indica and Azadirachta indica. However, whether this very symptom was present in the other 23 drugs has not been addressed as there was no detailing of every desire. Therefore, the general sample for this 'Desire for milk' symptom comprised of 1563 only instead of 2922, which was the prescribed population as a surplus to the absence of prevalence data in general population. Even then, we cannot be sure that the symptom was checked in every patient in the population with available data. Forty-nine symptoms with corresponding medicines with LRs(+) >1.5 are listed below [Table 4]. There is considerable systematic error in this data, probably more significant than statistical error. Hence, statistical confidence intervals are not given to avoid the suggestion that the point estimate is really within the statistical confidence interval.
|Table 4: Forty nine rubrics along with list of medicines having confined likelihood ratio of more than 1.5|
Click here to view
| Discussion|| |
Overall, of the total 166 general symptoms evaluated of 29 medicines under this CV programme of CCRH during 2005–2010, medicines having LRs and confined LRs >1.5 were elicited for 6 and 49 symptoms, respectively; much were under-reported and kept for calculation in future. Thus this compilation allows further insights on general symptomatology of some of the less frequently prescribed homoeopathic medicines subject to further systematic and rigorous explorations and use in clinical practice. CV concept was present from the very beginning of homoeopathic era, and the programme was launched by the Council almost 4 decades back, much before the concept of LR and assessment of the prevalence of symptoms came into use for validation of symptoms. In our study, we intended to validate these verified symptoms statistically using LR, regarding the prevalence of symptoms instead of absolute occurrence. The protocol was designed for CV prospectively and collected data already published in book form. Retrospective assessment of LR was planned afterwards.
Another major caveat in retrospective assessment of symptoms in PFR is substantive confirmation bias increasing LRs. We are actually unaware whether the symptom has the same definition among patients and observers; thus, it is uncertain how the symptom was recorded, and we cannot be sure that the symptom is checked in all patients. Cut-off values can vary greatly and are sensitive to confirmation bias.
Some observers are more pertinent in their questions and leave few rooms for the 'don't know,' or 'not applicable' answer. Symptoms are more likely to be checked if they belong to the preferred medicine pictures.
When a symptom is not recorded in a particular case, this could mean two things:
- The patient did not have the symptom
- The symptom was not checked.
As the case recording pro forma provided open-ended questions, so there was definitely a chance of missing a point. Estimating the homoeopathic value of a symptom in individual patients requires considerable skills. In prospective research, we monitor the data from the start, and we can detect if a patient answers positive to many questions, or if some observers have lower cut-off values than others. This is not possible in retrospective research.
In this perspective, polar symptoms deserve a special mention. Polar symptoms are those with opposite values, such as aversion/desire for food, or aggravation/amelioration. If we look retrospectively at patient records, we can detect patients with an 'aggravation from cold' and with 'amelioration from cold,' but 'no influence from cold' is seldom noted. It is therefore difficult to distinguish pseudo-bipolarity from real bipolarity. False bipolarity can be detected by looking at the database; if a considerable number of records have both opposite values, then the bipolarity is real; otherwise, the records of one pole compensate for part of the other pole. Ideally, polar symptoms (e.g., thirst and appetite) should be checked by offering the choice 'decreased, normal, or increased.' Checking symptoms with this kind of Likert scales renders a frequency distribution with a peak at 'normal.' The essence of polarity analysis is subtraction of the opposite values for each symptom, thus estimating the average of both poles. Likewise, we must be extra careful when interpreting all polar symptoms for all medicines, because results are very sensitive for the way the symptoms were elicited, more than in non-polar symptoms.
Strength of this CV project includes multicentric gathering of 4652 records (3705 improved and 947 not-improved) from 11 institutes of the Council from 2005 to 2010. This naturally increased the external validity and generalisability of the study findings. The study design was aimed at minimal interference of routine Homoeopathy practice, thus corroborating with the guidelines of PFR and allowing calculation of LR. However, the project had some in-built limitations. In this CV programme, it was not possible to have estimations of all the symptoms under question in the 'rest of the population;' hence, for calculation of LRs, we frequently referred to the prevalence as reported in the literature. It is, however, impossible to know if cut-off values for symptom-intensity were the same in literature as in this project. Moreover, food desires vary greatly because they are influenced by many variables such as climate, culture, and age etc. This could be a problem. Specific symptoms were not checked in all populations. The 'total' population for a specific symptom should be the population where the symptom was checked. So, the 'total' population is different for every symptom. For example, the prevalence of 'dullness/laziness' was found to be very low, only 0.97%. It was quite possible that the symptom was not checked in many populations. Few symptoms were related closely, for example, 'Irritable' and 'Angry', but were recorded separately by the investigators. In Kent's repertory also, these two have been mentioned as rubrics with similar meanings. These necessitate the requirement of a refined and standardised checklist of the symptoms to be studied.
The major drawback of this study was that symptoms which were not found in the available (unpublished) repertory and Materia Medica developed by the Council for CV project were not recorded systematically and those cases were treated in general outpatients without further record keeping. It is unlikely that these symptoms were not there in the population where other medicines were prescribed. Confidence intervals of confined LRs have not been mentioned here, because the uncertainty of the numbers is caused by recall bias than by statistical uncertainty. A confidence interval only accounts for statistical uncertainty. This means that we could not calculate the prevalence of the symptom in the whole population. For such symptoms, we have to restrict the denominator of LR to prevalence of the symptom in rest of the study (not general) sample, thus generating 'confined' LRs for those symptoms. Even, 'confined LRs' could not be calculated for those symptoms whose prevalence was not recorded for a sufficient number of medicines (Five was chosen as a cut-off value, as a rule of thumb). Thus the CV project suffered from under-reporting to some extent. Besides, instead of the use of validated outcomes (e.g., Glasgow Homoeopathic Hospital Outcome Scale; GHHOS), relying on clinician-rated outcomes such as 'Cured,' 'Improved,' 'Not Improved,' and 'Worse' definitely drew some sort of assessment biases and further dichotomising the outcomes into 'improved' and 'not improved' resulted in considerable loss of information.
In sharp contrast to the recently published papers on CV,,,,,, we shifted our focus from drug-oriented approaches to symptom-centric and calculated LRs and 'confined' LRs with references from available literature and prevalence of the symptom in rest of the study (not general) sample respectively. This provides further insights into prioritisation and discrimination among the medicines as per prevalence and LRs, as appropriate.
In the homoeopathic repertories, we see a classification of symptoms in sections such as; mind section, parts of the body and general symptoms. Throughout these sections, we find a large number of disease diagnosis, complaints and conditions. Recently, a new dimension has been added in PFR in terms of condition confined assessment (CCA) of LRs, for example, assessment of LRs from a cough population from the IIPCOS2 study and assessment of LRs of contact dermatitis symptoms of Vinca minor. CCA can be done wherein all patients suffering from the same disease/condition are analysed. Therefore, instead of 'whole population' (as used in LR calculation), a subpopulation of 'patients suffering from a particular disease or condition' is considered. Hence, instead of seeking prevalence of one symptom in 'whole population', prevalence of a symptom in 'disease population' is calculated. The symptoms present in cured patients and the medicines given make the 2 × 2 table wherein the total patients suffering from condition 'X' are considered in the denominator. However, it should be kept in mind that in subgroups defined by conditions is only valid for these subgroups, but we must still be aware of the influence of this selection. If we select a subgroup on condition, we inherently also select a subgroup of medicines that are related to that condition. These medicines are all related to the condition, but some more than others. If we compare these medicines by LR, we can get LR values below unity because some medicines are less than average related to the condition. This low LR means nothing more than that the medicine has less relationship to the condition than the other selected medicines (but still more than many not selected medicines), but it works contra-intuitively because LR <1 is associated with a contraindication for that medicine.
In the underlying research, the LR was not confined by a specific condition, but by the availability of data. The limited availability of data was possibly related to the applied method of medicine selection: If the symptom was not mentioned in the available materia medica or if the eligible medicine was not in the repertory-rubric, it was not checked. Hence, we cannot know the prevalence of the symptom in the populations where it was not checked. In proper prospective PFR, each symptom should be checked in every new patient. Besides, a mean of a symptom of different medicines' prevalence will not render the same result as the prevalence of the symptom in the remainder of the population both in general LR research and in confined LR. In future, during further estimation of general LR on the same symptoms, we should check whether it fits within the 95% confidence interval or not.
In terms of LR, Homoeopathy is experiencing a transition from the notion of 'important' symptom to the LR-weighted symptoms based on modern epidemiological principles. The task is enormous; in return, we get more reliable instruments. Symptoms can be seen as diagnostic instruments and the LR as an indication for optimal use. This may be a solution to shortcomings of the repertory, for example, handling rare remedies and vagueness of reporting. Furthermore, the new data will be based on evidence instead of intuition. Moreover, this is becoming possible with the least possible interference with day-to-day practice. A computer repertory can easily produce a graph that shows the increase of probability that a medicine will work with the corresponding LR. Small symptom rubrics will be difficult to assess in prospective studies because they relate to infrequently occurring symptoms. In future, computer programmes may show the symptoms with the highest LR(+) to confirm a medicine, or the lowest LR(−) to exclude it.
| Conclusion|| |
Of the total 166 general symptoms evaluated of 29 medicines under this CV programme of CCRH during 2005–2010, medicines having LRs and confined LRs >1.5 were identified for 6 and 49 symptoms, respectively; much were under-reported and kept for calculation in future. Thus, in spite of the substantial limitation of study design and considerable caveats in the recorded data, this paper provides the first insight into the prevalence and LRs of general symptoms of some less frequently prescribed homoeopathic medicines. Further research of this kind is warranted, but with enhanced methodological rigor.
We deeply acknowledge Dr. Vikram Singh, former Deputy Director, CCRH, and Dr. Krishna Singh, former Assistant Director, CCRH for their contribution toward monitoring of data collection. The authors also acknowledge Dr. Darshan Singh, Dr. K. C. Das, Dr. S. R. Bhagat, Dr. M. K. Rai, Dr. A. K. Bhagat, Dr. R. P. Yadav, Dr. M. D. John, Dr. N. R. Mondal, Dr. U. P. Yadav and Dr. M. N. Sinha for their participation in the programme and for help in data collection. We gratefully acknowledge the active cooperation and participation made by the patients and the supporting staff, especially Mrs. Subarna Dey, Data Entry Operator, Dr. Anjali Chatterjee Regional Research Institute, Kolkata.
Financial support and sponsorship
The study was funded by Ministry of AYUSH, Government of India.
Conflicts of interest
| References|| |
Kent JT. Repertory of the Homeopathic Materia Medica. 6th
ed. New Delhi: B. Jain Publishers; 2007.
Kent JT. Lectures on Homoeopathic Philosophy. New Delhi: B. Jain Publishers (P) Ltd.; 2007. p. 203-9.
Koley M, Saha S, Das KD, Roy S, Goenka R, Chowdhury PR, et al.
Prospective evaluation of few homeopathic rubrics of Kent's repertory from Bayesian perspective. J Evid Based Complementary Altern Med 2016;21:277-81.
Rutten ALB, Stolper CF, Lugten RFG, Barthels RJWM. Repertory and likelihood ratio: Time for structural changes. Homeopathy 2004;93:120-4.
Rutten ALB, Stolper CF, Lugten RFG, Barthels RJWM. Statistical analysis of six repertory rubrics after prospective assessment applying Bayes' theorem. Homeopathy 2009;98:26-34.
Stolper CF, Rutten ALB, Lugten RFG, Barthels RJWM. Improving homeopathic prescribing by applying epidemiological techniques: The role of likelihood ratio. Homeopathy 2002;91:230-8.
Miglani A, Rutten L, Manchanda RK. Generalisability of prognostic factor research: Further analysis of data from the IIPCOS2 study. Homeopathy 2017;106:155-9.
Bayes T. An essay toward solving a problem in the doctrine of chances. Philos Trans R Soc Lond 1763;53:370-418.
Rutten ALB, Stolper CF, Lugten RFG, Barthels RJWM. 'Cure' as the gold standard for likelihood ratio assessment: Theoretical considerations. Homeopathy 2004;93:78-83.
Rutten L. Prognostic factor research in homoeopathy. Indian J Res Homoeopathy 2016;10:59-65. [Full text]
Gill CJ, Sabin L, Schmid CH. Why clinicians are natural bayesians. Br Med J 2005;330:1080-3.
Van Wassenhoven M. Towards an evidence-based repertory: Clinical evaluation of Veratrum album
. Homeopathy 2004;93:71-7.
Eizayaga JE, Pozzi MI, Canan MC, Saravia L. Prevalence and likelihood ratio of symptoms in patients with good therapeutic response to Lycopodium clavatum
. A retrospective study. Homeopathy 2016;105:78-83.
Central Council for Research in Homoeopathy. Homoeopathic Drug Proving. New Delhi: Central Council for Research in Homoeopathy; 2005. p. 5-15, 19-22, 64-8, 79-80, 89-90, 97-9, 75-8, 91-4, 39-40, 15-8, 121-3, 115-7, 104.
Boericke W. Pocket Manual of homoeopathic Materia Medica and Repertory. Reprint 9th
ed. New Delhi: B. Jain Publishers Pvt. Ltd.; 1997. p. 26, 640, 136, 1074, 26, 341-2, 493, 450.
Murphy R. Lotus Materia Medica. 2nd
revised ed. New Delhi: B. Jain Publishers Pvt. Ltd.; 2002. p. 1757-9, 1258-60, 154-6, 1759-62, 1085-8, 592-3, 305-6, 1465-7, 60-2, 860-2, 1301-3, 1440, 1184-5, 1491-3.
Central Council for Research in Homoeopathy. CCRH Quarterly Bulletin. Vol. 12 (1 & 2). Central Council for Research in Homoeopathy; 1990. p. 15-23, 59-63, 111-4, 81-3, 12-5, 22-4, 9-11.
Allen TF. The Encyclopedia of Pure Materia Medica. vol. 10. New Delhi: B. Jain Publishers Pvt. Ltd.; 2001:583-92, 592-6, 131-2, 400-2, 431-2.
Allen TF. Handbook of Materia Medica and Homoeopathic Therapeutics. Reprint ed. New Delhi: B. Jain Publishers Pvt. Ltd.; 2001. p. 1107-9, 1109-10, 831-2.
Clarke JH. A Dictionary of Practical Materia Medica. vol. 1. New Delhi: B. Jain Publishers Pvt. Ltd.; 1978. p. 6, 1406-9, 1410-4, 376-9, 702-6, 937, 514-5, 368-70.
Anshutz EP. New, Old and Forgotten Remedies. 2nd
ed. New Delhi: B. Jain Publishers Pvt. Ltd.; 1996. p. 90-109, 253-71, 18, 371-8, 439-41.
Cowperthwaite AC. Textbook of Materia Medica & Therapeutics. 16th
ed. Calcutta: A. P. Homoeoliv. Publishers; 1970. p. 464-5, 839, 837, 825.
Hering C. Hering's Guiding Symptoms of Materia Medica. vol. 1. New Delhi: B. Jain Publishers Pvt. Ltd.; 1888. p. 258-62.
Ghosh SC. Drugs of Hindoosthan. 8th
ed. Calcutta: Hahnemann Publishing Co.; 1980. p. 110-8.
Phathak SR. Materia Medica of Homoeopathic Medicines. 1st
ed. New Delhi: B. Jain Publishers Pvt. Ltd.; 1991. p. 592-4, 449.
Rutten ALB, Frei H. Frequently occurring polar symptoms assessed by successful cases. Homeopathy 2012;101:103-11.
Frei H. Polarity analysis, a new approach to increase the precision of homeopathic prescriptions. Homeopathy 2009;98:49-55.
Rutten L, Muraleedharan KC, Shinde VH, Manchanda R. What is a homoeopathic symptom, in daily practice and research? Indian J Res Homoeopathy 2017;11:12-20. [Full text]
Manchanda RK, Chakraborty PS, Singh P, Nayan SS, Singh O, Pradhan PK, et al
. Mygale lasiodora: A multicentric observational homoeopathic clinical verification study. Indian J Res Homoeopathy 2015;9:249-57. [Full text]
Manchanda RK, Chakraborty PS, Singh P, Pradhan PK, Prasad VG, Singh O, et al
. A multicenter, observational, homoeopathic clinical verification study of Cynodon dactylon
revealing symptom prevalence in a cohort of 340 patients. Indian J Res Homoeopathy 2016;10:225-37. [Full text]
Manchanda RK, Chakraborty PS, Das KC, Rai MK, Singh O, Pradhan PK, et al
. Symptom prevalence in a cohort of 147 patients improved with the homoeopathic medicine Ocimum canum
: A multicenter, open, observational, clinical verification study. Indian J Res Homoeopathy 2016;10:238-48. [Full text]
Manchanda RK, Chakraborty PS, Singh P, Rai MK, Singh O, Singh DK, et al
. Symptom prevalence in a cohort of 65 patients improved with the homoeopathic medicine Mangifera indica
: A multicentric open observational clinical verification study. Indian J Res Homoeopathy 2017;11:184-95. [Full text]
Manchanda RK, Gupta J, Chakraborty PS, Singh P, Nayan SS, Singh JP, et al
. Formic acid: A multicentric observational homoeopathic clinical verification trial. Indian J Res Homoeopathy 2016;10:249-57. [Full text]
Gupta J, Manchanda RK, Chakraborty PS, Singh P, Ramteke S, Singh O, et al
. Prevalence and prognostic factor in patients with good therapeutic response in a cohort of 172 patients with the homoeopathic medicine Aranea diadema
: A multicentre, open-label, observational study. Indian J Res Homoeopathy 2018;12:20-8. [Full text]
Bagchi S, Nayak C, Rutten L, Koley M, Saha S, Roy M, et al.
Assessment of likelihood ratio for four contact dermatitis symptoms of Vinca minor
. Homeopathy 2018;107:209-17.
[Table 1], [Table 2], [Table 3], [Table 4]