Выявление гендерной зависимости окончательного диагноза на примере санатория

The purpose of the presented article is to analyze the database of patients of the sanatorium of "Victoria", located in Kislovodsk (Russia), to determine the relationship between the final diagnosis and the patient's gender.

Medicine plays a key role in society, as people face various diseases and diagnoses that affect their lives and well-being. Every day, millions of people around the world face common diseases such as colds, flu, allergies, diabetes, cardiovascular diseases and many others. These diseases can have various causes and manifestations, and require competent medical intervention for diagnosis and treatment. Medicine, in turn, provides not only treatment, but also prevention, counseling and education to help people maintain and improve their health. It is an integral part of our lives, providing care and support during illness and helping to lead an active and healthy lifestyle.

I wonder if there is a relationship between the diagnosis of the disease and the sex of a person? Knowing the answer to this question can help develop effective disease prevention measures and improve people's quality of life.

Investigating possible links between disease diagnoses and gender can help identify groups of people who are susceptible to certain diseases. For example, if it were found that a certain disease is more common in men, it would allow us to focus on preventive measures and lifestyle that can help men reduce the risk of developing this disease. Similarly, if a link were found between a certain disease and the female sex, specific prevention and treatment strategies for women could be developed.

Establishing a link between the diagnosis of the disease and the gender of a person can also contribute to the development of a more personalized approach to medical care, taking into account the characteristics of each group. This may include adapting screening programs, conducting educational campaigns and providing gender-specific recommendations to prevent the occurrence of diseases and improve overall health.

An important step in preventing diseases and improving people's quality of life is to determine the relationship between the diagnosis of the disease and the sex of a person. This can help the medical community (and society as a whole) develop more effective prevention, treatment and care strategies aimed at reducing morbidity and improving public health.

As already mentioned, health analysis plays an important role in optimizing medical care and improving the quality of life of patients. The grouping approach makes it possible to identify common diseases in different gender groups and make informed health decisions

Data will be taken for analysis:

- the final diagnosis made by the attending physician of the sanatorium;

- gender.

The final diagnosis is made by the patient's attending physician after all the treatment at the last (final) appointment. The diagnosis code is deciphered according to ICD-10 (International Classification of Diseases of the 10th revision), at the moment there are about 15000 names

The first step will be to download and install the standard Numpy and Pandas libraries and import the database in XLSX format (Fig. 1). It includes quite a lot of data about the treatment of a guest of the sanatorium: the procedures assigned to him, their price list, the type of treatment according to the voucher, primary and final diagnoses, etc

[7]

.

Figure 1 - Loading libraries and database

DOI:10.60797/IRJ.2024.145.110.1

NumPy is a Python library that is used for mathematical calculations, starting with basic functions and ending with linear algebra. The full name of the library is Numeric Python extensions, or "Numeric Python Extensions"

[4]

.

Pandas is a Python programming library designed to work with data. It works on the basis of the NumPy library and provides special data structures for working with numeric tables and time series. The pandas library provides operations for data management and analysis

Thanks to the drop() function, the database gets rid of unnecessary columns for the current analysis (Fig. 2). As a result, the columns "DGN_LAST_MKB" remain — the final diagnosis of the patient and "SEX" — the gender of the patient, where 0 = male and 1 = female.

Figure 2 - Using the drop() function

DOI:10.60797/IRJ.2024.145.110.2

First, you need to clear the database. This will help eliminate duplicate data that may occur as a result of errors or repeated entries. It also allows you to maintain data integrity and prevent distortion of the analysis results

[2]

.

In addition, database cleanup may include error correction and data standardization. This ensures the uniformity and correctness of the data, which is important for accurate analysis and reliable results.

In general, database cleanup is an important step before analysis, which helps to ensure data quality, improve performance and reliability of analysis results.

Data cleanup involves removing duplicate records and missing (empty) values. This can be implemented using the dropna() function (Fig. 3).

Figure 3 - Using the dropna() function

DOI:10.60797/IRJ.2024.145.110.3

After the cleanup, the number of records decreased by 46,996. These were patients, for whatever reason, without a final diagnosis.

Next, you need to group the data by the final diagnosis and count the number of men and women using the groupby() and value_counts() functions (Fig. 4).

Figure 4 - Using the groupby() and value_counts() functions

DOI:10.60797/IRJ.2024.145.110.4

For convenient data display, it was decided to provide the result in the form of an interactive bar chart. The graphical representation of the data will allow you to visually illustrate the distribution of patients by final diagnosis and gender, which will facilitate the perception of information and the identification of possible patterns

[10]

.

Thanks to the plotly.express library, a graph was created (Fig. 5), based on which the following conclusions can be drawn:

1. The most common disease is M42.1 – Osteochondrosis of the spine in adults, the majority of whose owners are men, about 123 thousand people, which exceeds the number of women with a similar diagnosis by almost 2 times (Fig. 6).

To prevent this disease, it can be recommended to develop special programs and services aimed at prevention and treatment of osteochondrosis of the spine, as well as holding events aimed at men.

These packages can include specialized programs of physical rehabilitation, massage, fitness, and expert advice on recommended exercises for the prevention and treatment of diseases of the musculoskeletal system. You can also offer meals specially designed to meet the needs of men and their health, and conduct educational seminars on healthy lifestyles and disease prevention.

2. Analysis of the distribution of final diagnoses by gender of patients revealed that there are differences in the incidence of diseases between men and women. This is useful information for the development of individual medical programs and prevention strategies

[8]

,

[9]

. According to the graph, men are more likely to be confirmed with the disease, especially those associated with diagnoses of M42 (osteochondrosis of the spine) and I11.9 (hypertension, unspecified). This indicates the need for specialized programs for the prevention and treatment of these diseases in men, as well as the importance of educational activities aimed at maintaining men's health and preventing these diseases.

Figure 5 - Column diagram of the distribution of patients by final diagnosis and gender

DOI:10.60797/IRJ.2024.145.110.5

Figure 6 - Number of men diagnosed with M42.1

DOI:10.60797/IRJ.2024.145.110.6

Among 501,748 patients, only 78,006 people are diagnosed with Z00 – General examination and examination of persons without complaints or established diagnosis (Fig. 7). This diagnosis belongs rather to the category of preventive examinations and consultations, rather than to specific diseases.

Figure 7 - Number of men and women diagnosed with Z00

DOI:10.60797/IRJ.2024.145.110.7

Analyzing the number of cases of certain diseases among men and women can help identify possible differences in predisposition to certain diseases depending on gender. This can be important information for a personalized treatment approach.

Also, judging by the graph, it can be concluded that there is a significant prevalence of the diagnosis of E66 (obesity) among the male population (Fig. 8). It can be assumed that this is due to alcohol consumption. Alcohol contains a high number of calories and can contribute to the accumulation of body fat. In addition, alcohol can affect metabolism and eating behavior, which can lead to an increase in appetite and consumption of more food

The diagnosis of E66 is usually associated with overweight and obesity, which may indicate the importance of taking measures to control weight and maintain a healthy lifestyle among men to prevent diseases associated with this diagnosis.

Figure 8 - Number of men diagnosed with E66

DOI:10.60797/IRJ.2024.145.110.8

Genetic features may play a role in predisposition to certain diseases. For example, genetic factors can influence the tendency to various diseases

[3]

, including cancer, diabetes mellitus, cardiovascular diseases, etc.

The relationship between gender and diagnosis is a complex research issue in the field of medicine. Differences in morbidity and response to treatment may be gender-specific and require additional study and analysis of the causes of this difference. Some diseases may occur more often in a certain gender, which underscores the importance of taking gender into account when conducting medical research.

The analysis of the relationship between the patient's gender and his final diagnosis is important for improving medical practice and developing personalized treatment approaches. Further research in this area may lead to the development of more effective treatment strategies that take into account the gender characteristics of patients.

To solve the problem, various Python functions were used, which made it possible to efficiently process data without resorting to the use of excessive computing power, and create a bar chart, thanks to which they were able to visualize the result.