**Objective
**The objective of this article is to provide the Government authorized Homoeopathic Epidemic Control cells with guidelines on how to conduct post-epidemic statistical studies.

__Authored by:
__

**Dr. Anand .P.R,**

*B.H.M.S, M.D, M.C.A, M.B.A*

*Medical Officer, Department of Homoeopathy,*

*Government of Kerala*

__Co-authored by:
__

**Dr. Mridula Gopinathan**,

*B.H.M.S, M.D*

*Medical Officer, Department of Homoeopathy,*

*Government of Kerala*

“When you can measure what you are speaking about and express it in numbers, you know something about it but when you cannot measure, when you cannot express it in numbers, your knowledge is of meager and unsatisfactory kind” – Lord Kelvin.

The use of statistics is multifold in medicine. It is used to compare the efficacy of a particular drug or line of treatment. For this the percentage of people cured, relieved or died in an experiment and control groups is compared and difference due to chance or otherwise, is found by applying statistical techniques. The inferences thus drawn are used in policy making. It ultimately determines the allocation of states health funding.

Biostatistics is concerned with techniques of data collection, summarizing, interpretation, drawing inferences, testing of hypotheses and making recommendations to ultimately influencing decision and policy making and policy analysis.

Most of us homoeopaths are not oriented with statistical methods. This article aims to give an overview of how to do the post epidemic statistical study after the homoeopathic preventive measures (intervention) have been delivered.

**The Post Epidemic Study Design:**

The study design most suited to do a post epidemic study after a homoeopathic intervention (Genus Epidemics) has been provided is a Quasi-Experimental Design. When an epidemic breaks out in a locality due to ethical considerations, we cannot deny an intervention known to be effective in controlling an epidemic. We do not assign which cases are to be provided the preventive and which cases placebo so as to be kept as controls during the study. Thus, the ‘gold standard’ of biostatistics, ie. A Randomized Control Trial cannot be conducted during a post- epidemic study. The next best option that we have is a quasi- experimental design.

A **quasi-experiment** is an empirical study used to estimate the causal impact of an intervention on its target population. Quasi-experimental research shares similarities with the traditional experimental design or randomized controlled trial, but they specifically lack the element of random assignment to treatment or control. Quasi-experiments are subject to concerns regarding internal validity, because the treatment and control groups may not be comparable at baseline. With random assignment, study participants have the same chance of being assigned to the intervention group or the comparison group. As a result, differences between groups on both observed and unobserved characteristics would be due to chance, rather than to a systematic factor related to treatment. Randomization itself does not guarantee that groups will be equivalent at baseline. Any change in characteristics post-intervention is likely attributable to the intervention. With quasi-experimental studies, it may not be possible to convincingly demonstrate a causal link between the treatment condition and observed outcomes. This is particularly true if there are confounding variables that cannot be controlled or accounted for. But this type of design is our best available option for post epidemic study in homoeopathy to assess if our Genus epidemicus for an epidemic has a positive effect in controlling an epidemic in comparison with the epidemic subsiding due to ‘chance’ (a natural regression of the epidemic)

The first part of creating a quasi-experimental design is to identify the variables. The quasi-independent variable will be the x-variable, the variable that is manipulated in order to affect a dependent variable. “X” is generally a grouping variable with different levels. Grouping means two or more groups such as a people who have taken our Genus epidemics and people who have not taken it. The predicted outcome is the dependent variable, which is the y-variable, in our study, the people who are infected with the epidemic and who are not affected. In a time series analysis, the dependent variable is observed over time for any changes that may take place. Once the variables have been identified and defined, a procedure should then be implemented and group differences should be examined.

Next we will define the population and the sampling technique to be used. Care must be taken so as to ensure that the samples taken are representative of the population characteristics. Usually homoeopathic interventions are delivered over large geographical areas to control the epidemic. Thus, the sampling technique that I found the easiest to implement and be representative of the affected population is **‘STRTIFIED AREA SAMPLING WITH SYSTEMATIC SAMPLING AT THE HO– USEHOLD LEVEL’. **Randomization using tools like ‘Snedecor’s Random table’ should be incorporated at each step during the sampling process.

I will illustrate the sampling technique that we used in 2003 for the first post epidemic study in the whole world of such magnitude (in 50 lakh population). Requests from residential associations for preventive delivery were considered and 10% of this was selected using table of random numbers. Our data collectors and enumerators visited every 5^{th} house in the selected resident associations. The starting house from which every next 5^{th} house is selected was again determined using randomization tables. Thus, one could see that we have tried to incorporate randomization in every possible instance to get a sample which is truly representative of the population considered.

__TOOL FOR DATA COLLECTION- THE POST EPIDEMIC SURVEY QUESTIONNAIRE
__Though open ended questions are the best, it would be difficult to sieve through descriptive answers thus a mixture of Partially Categorized and Close Ended questions are used. I would give a sample tool that was used by me during the first epidemic survey.

__PERFORMA FOR STUDYING THE EFFECT OF HOMOEOPATHIC PREVENTIVE MEDICINE IN AN EPIDEMIC -2003-04__

**RESTROSPECTIVE STUDY**

- Name
- Age
- Sex
- Occupation
- Residential address
- Occupational address
- Number of family members
- Family income
- Educational status
- Birth order
- Environment/surroundings

Risk factors

- Homoeopathic preventive taken Yes/No
- Have you taken the medicine in the prescribed dose Yes/No
- Have you taken any other medicine in between Yes/No

If yes, specify – homoeopathy, allopathy, ayurveda, others

- Whether you under any chronic medications Yes/No

If yes, specify – homoeopathy, allopathy, ayurveda, others

- Have you developed symptoms of the epidemic after taking

homoepathic preventive medicine Yes/No

- After how many days did you develop the symptoms- 1-7, 8-14, 15-28
- How many days did the symptoms last 1-7, 8-14, 15-28
- What were the symptoms?

- Have you undergone in-patient treatment? Yes/No

If yes, specify – homoeopathy, allopathy, ayurveda, others

10.How long did the In-patient treatment last 1-7, 8-14, 15-28

- Was any labororatary investigation done? Yes/No
- How long did the treatment last 1-7, 8-14, 15-28
- How many working days were lost 1-7, 8-14, 15-28
- How much did the epidemic treatment cost <500, >500, > 1000
- What was the treatment result? Cured, complications, death.
- If no treatment was taken for the epidemic what is the result

Cured, complications, death.

- Signature head of family

- survey no: Place, Date, time.
- Signature of the surveyor
- Signature of the team lead
- Signature of the data reviewer

Very good tools should be prepared for specific diseases incorporating diagnostic criteria as described in ICD 10 from the World Health Organization. Care should to be taken to add questions to assess the causative and environmental factors which lead to the epidemic. Economic impact of the epidemic should also be assessed.

TABULATION: The tabulation of the data thus received can be done manually or using softwares like Epiinfo.

__THE STATISTICAL TESTS:__

A **statistical hypothesis** is a scientific hypothesis that is testable on the basis of observing a process that is modeled via a set of random variables. A **statistical hypothesis test** is a method of statistical inference used for testing a statistical hypothesis.

A test result is called statistically significant if it has been predicted as unlikely to have occurred by sampling error alone, according to a threshold probability—the significance level. Hypothesis tests are used in determining what outcomes of a study would lead to a rejection of the null hypothesis for a pre-specified level of significance. The process of distinguishing between the null hypothesis and the alternative hypothesis is aided by identifying two conceptual types of errors (type 1 & type 2), and by specifying parametric limits on e.g. how much type 1 error will be permitted.

The following steps have to be done before a statistical hypothesis test:

- There is an initial research hypothesis of which the truth is unknown.
- The first step is to state the relevant
**null and alternative hypotheses**. - The second step is to consider the statistical assumptions being made about the sample in doing the test; for example, assumptions about the statistical independence or about the form of the distributions of the observations. This is equally important as invalid assumptions will mean that the results of the test are invalid.
- Decide which test is appropriate, and state the relevant
**test statistic**T. - Derive the distribution of the test statistic under the null hypothesis from the assumptions. For example the test statistic might follow a Student’s t distribution or a normal distribution.
- Select a significance level (
*α*), a probability threshold below which the null hypothesis will be rejected. Common values are 5% and 1%. - The distribution of the test statistic under the null hypothesis partitions the possible values of T into those for which the null hypothesis is rejected—the so-called
*critical region*—and those for which it is not. The probability of the critical region is*α*. - Compute from the observations the observed value t
_{obs}of the test statistic T. - Decide to either reject the null hypothesis in favor of the alternative or not reject it. The decision rule is to reject the null hypothesis H
_{0}if the observed value t_{obs}is in the critical region, and to accept or “fail to reject” the hypothesis otherwise.

The statistical test best suited for our purpose is the __CHI-SQUARE TEST.__

__CHI-SQUARE TEST (__** ):**This is a non-parametric test which involves the calculation of a quantity called chi-square

**(**

**)**and is usually used in the following important applications

- Proportion
- Association
- Goodness of Fit

Of these using the test of association we can measure treatment and outcome of disease. There are two possibilities, either they influence or affect each other or they do not. In other words, they are either independent of each other or they are dependent on each other, i.e, associated.

We should set the confidence limits. Usually 95% confidence limits is taken, ie. Population proportion [P] + or – 1.96 SE (standard error)

Qualitative data are presented in their actual frequencies in a 2×2 contingency table.

The **χ ^{2 }test for independence**

χ^{2 }= __(ad – bc) ^{2 }n__

(a+b) (b+d) (c+d) (a+b)

Where n= a+b+c+d without Yates correction

The calculated value is then compared to the χ^{2 }table value. If the calculated value (χ^{2}) is greater than the table value (χ_{α }^{2}) we can reject the H_{0} , the null hypothesis. If the null hypothesis is rejected, it means we can accept the alternate hypothesis H_{1.}

Other parametric and non-parametric tests like the Wilcoxon Signed Rank test, MANOVA (Multivariate Analysis of Variance) and the Mann-Whitney U Test can also be used. The type of test to be used depends on the nature of data derived from the survey questionnaire.

I will illustrate what I have described above using the first study in this field in 2003-04 done by myself and Professor Hameed Labba, Head of Department of Statistics, University College, Thiruvananthapuram.

__STATISTICAL STUDY OF THE SURVEY RESULT EFFICACY OF HOMOEOPATHIC MEDICINES IN EPIDEMICS____ __

This study was conducted by Dr.Anand.P.R, Dr. Dinesh.R.S, Dr. Sreejith.S and Prof. Hameed Labba (Head of dept. of statistics at University College, Thiruvananthapuram) et.al for the Govt of Kerala and Govt. Homoeopathic Medical college, Thiruvananthapuram.

__SURVEY RESULTS ON THE EFFICACY OF HOMOEOPATHIC PREVENTIVE MEDICINE FOR DENGUE AND VIRAL FEVER – 2003.____ __

__Aims and objectives__

- To assess the efficacy of Homoeopathic medicine in the prevention of Dengue and viral fever outbreak in Thiruvananthapuram and Kollam districts of Kerala in 2003.

__Materials and methods
__After detailed analysis of the Dengue cases in Thiruvananthapuram and Kollam districts of Kerala, the Genus Epidemics Eupatorium Perfoliatum was selected. The 200

^{th}centesimal potency of this medicine was given in 15 doses (3 doses daily for 5 consecutive days). The distribution was done by the Homoeopathic medical students, the NGO’s and Residential associations. The efficacy study was conducted one month after the distribution of the preventive medicine.

STUDY POPULATION – 50 lakhs

METHOD OF SAMPLING – Stratified Area Sampling followed by systematic sampling at the household level

NATURE OF STUDY – Quasi experimental Post Epidemic study without controls

STATISTICAL TEST — USED — Chi square test

TOTAL NO. OF PEOPLE WHO HAVE TAKEN HOMOEOPATHIC PREVENTIVE MEDICINE – 20 lakhs (from 670 requests from residential associations)

Among the 670 residential associations, 67 residential associations were selected from different areas (rural, urban, coastal areas, offices, schools etc) according to statistical methods (Snedecor’s table of random digits). Again 50 families were selected from each area on the basis of systematic sampling.

Thus we selected 15 x 67 = 1005 samples (families). We surveyed the 1005 families frequently. After 3 weeks of survey we received the following data –

Taken | Not taken | Total | |

Infected | 123 (a) | 44 (b) | 167 (a+b) |

NOT INFECTED | 746 ( c) | 92 (d) | 838 (c+d) |

Total | 869 (a+c) | 136 (b+d) | 1005 (n) |

__Hypothesis __

H_{0 }– the homoeopathic preventive medicine and the attack of viral fever are independent; the medicine is not effective in preventing the disease

H_{1} – the homoeopathic preventive medicine has effect on the attack of viral fever; they are not independent

Taken | Not taken | Total | |

Infected | a | b | a+b |

NOT INFECTED | c | d | c+d |

Total | a+c | b+d | n = a+b+c+d |

**Statistical test (chi – square test)
**For testing the null hypothesis, we are using the χ

^{2 }test for independence. We put the two attributes in two rows and two columns – 2 x 2 contingency table.

Now,

χ^{2 }= __(ad – bc) ^{2 }n__

(a+b) (b+d) (c+d) (a+b)

Here α = 0.05 (Significance level)

df = (r-1) (c-1)

= (2-1) (2-1)

= 1

Now χ^{2 }=__ [ (123 x 92) – (44 x 746)] ^{2 }x 1005__

869 x 136 x 838 x 167

= __( 11316 – 32824) ^{2} x 1005__

869 x 136 x 838 x 167

= **28.109**** **

χ_{α }^{2} = 3.841 ( Table value , α = 0.05, df = 1)

If the calculated value (χ^{2}) is greater than the table value (χ_{α }^{2}) we can reject the H_{0} , the null hypothesis.

Here χ^{2 }> χ_{α }^{2}. i.e, calculated value is greater than table value.

So we can reject the hypothesis H_{0, }i.e., the medicine and the attacks of viral fever are independent._{ }

That is, the **homoeopathic preventive medicine is effective to prevent Dengue and other viral fever.**

__To check the effectiveness of the homoeopathic preventive medicines __

Number of Dengue fevers were reported from these areas during the survey

The percentage of the infected persons who have taken the preventive medicine (Out of 1005 samples, 869 samples consumed medicine)

= __123 x 100__ = 14% (taken)

869

The percentage of non- infected persons who have taken the preventive medicine

= __746 x 100__ = 86%

869

From these results we can conclude that the homoeopathic medicine is highly effective in preventing viral fever.

**Similar studies** were done by Dr. Dinesh. R.S and Dr.Joby. J in 2005 epidemic and by Dr. Dinesh.R.S and Dr. Rejikumar et.al in 2006. A study on the prophylactic Efficacy of Homoeopathic Preventive Medicine against Chikugunya fever by Dr. Rejikumar et al showed 82.19 % efficacy for the 200^{th} potency of Eup perfoliatum in the control of Chikugunya fever. This study was coordinated and verified by Dr. S. Sajith Kumar, M.D, Asst Proff, Dept of Community Medicine, Medical College, Thiruvananthapuram.

The Govt of Kerala has formulated a **Rapid Action Epidemic Control Cell Homoeopathy (RAECH)** under the Department of Homoeopathy after going through my report. It is the sincere wish of the authors that the other state governments in India would also take up similar ventures.

__BIBILOGRAPHY____ __

- MAHAJAN- MEDICAL STATISTICS.
- KOTHARI – RESEARCH METHODOLOGY.
- WHO – REPORT ON ACUTE FLACID PARALYSIS SURVILLENCE
- PARK AND PARK – SOCIAL AND PREVENTIVE MEDICINE
- HAHNEMAN SAMUEL – ORGANON OF MEDICINE
- HAHNEMAN SAMUEL – LESSER WRITINGS
- C TRIPATHI AND P N REDDY – PRINCIPLES OF MANAGEMENT

## Be the first to comment