Predictors of internet addiction among medical students of North India

Background The prevalence of internet addiction in India varies from 20% to 59% in undergraduate medical students. Therefore this study was planned to assess the prevalence, pattern and reason of internet usage and to assess predictors of internet addiction in medical undergraduate students. Material and Methods A cross-sectional study was conducted on 201 medical undergraduate students in a medical college of North India from April 1st to May 31st, 2019.A self-designed semi-structured and pre tested questionnaire was used to know pattern and reasons associated with internet addiction (IA) and Dr. Kimberly Young’s Internet Addiction Test (IAT) tool was used to measure level of IA. Discriminant analysis was used to assess predictors of internet addiction. Data was recorded in MS Excel and trial version of Statistical Package for Social Sciences (version 27.0; SPSS Inc., Chicago, IL) software was used for statistical analysis. Result Prevalence of internet addiction was found to be 90%, where prevalence of mild IA was 48.8% followed by moderate and severe IA, 38.8% and 2.4% respectively. Predictive accuracy of model based on socio-demographic, social media applications usage, Entertainment site usage, Educational site usage and nal model were found to be 61.2%, 63.7%, 63.2%, 61.7% and 66.2% respectively.

addiction [12,13]. Deteriorating effects of internet addiction were seen on psychological health of the students [14]. Hence it is important to analyze the prevalence, pattern of internet use and factors associated with internet addiction among medical undergraduate students as they are vulnerable group on account of the time they spend on the internet. There is paucity of data regarding predictors of internet addiction in medical undergraduates. Therefore this study was planned to assess the prevalence of internet addiction, pattern and reasons of internet usage and to assess predictors of internet addiction in undergraduate students of a medical college of North India.

Material And Methods
A college based cross-sectional study was conducted on 201 undergraduate students in a medical college of North India from April 1st to May 31st, 2019. Inclusion criteria: All the undergraduate students studying in a medical college of North India and using internet at least for last 6 months were selected for the study. Exclusion criteria: those who do not give consent.
Sample size calculation: The sample size was calculated by taking the minimum prevalence of internet addiction as 20% at a level of 95% signi cance and 6% precision [7]. The calculated sample size was 171 by using formula Z 2 Xp X q/ l 2 , Considering 10% non response rate the nal minimum sample size was 189. We have studied and analysed data from 201 students. We have also strati ed undergraduate students according to year of admission and enrolled at least 50 students from each strata.
Questionnaire design and validation: A semi-structured and pre tested questionnaire was used to collect  [15]. The IAT is a 20-item 5-point Likert scale that measures the severity of selfreported compulsive use of the internet. The marking for this questionnaire ranges from 0-100; the higher the score range, the greater the level of addiction; Normal Range: 0-30 points, Mild: 31-49 points, Moderate: 50-79 points, Severe: 80-100 points. The overall Cronbach's α computed from the studies was 0.889 [95% con dence interval (CI) 0.884-0.895]. The standard deviation of the alpha was low, at 0.049 [16]. In present study we have found high internal consistency, with an alpha coe cient of 0.889 (CI 0.867-0.911).

Data collection
A self-designed semi-structured and pre tested questionnaire was distributed in classes to various semesters, using a strati ed random sampling technique to incorporate as representative a sample as possible. The questionnaire was distributed in the classes with the permission of each faculty, and participation was voluntary. The participants were asked to ll the questionnaire once. The researcher had introduced himself to the participants and explained the purpose and objectives of the study. He also informed participants that participation is voluntary, will not affect their grades in this course, and is expected to take approximately 15 minutes.

Data management and statistical analysis
Con dentiality of all the data was ensured by keeping the responses anonymous. Moreover, the collected data was stored under secure settings. Data was recorded in MS Excel and trial version of Statistical Package for Social Sciences (version 27.0; SPSS Inc., Chicago, IL) software was used for statistical analysis. Qualitative data was analyzed using proportions and percentages. Two-group Discriminant analysis was used to assess the predictive accuracy of various discriminators for internet addiction. We have also prepared various models by combining different discriminators and calculated the predictive accuracy of each model. Stepwise discriminant analysis was performed to assess the most signi cant discriminators of internet addiction. Total  The analysis creates a discriminant function which is a linear combination of the weightings and scores on these variables. The maximum number of functions is either the number of predictors or the number of groups minus one, whichever of these two values is the smaller. [17] DA involves the determination of a linear equation like regression that will predict which group the case belongs to. The form of the equation or function is:

Result
In present study we have analyzed data of 201 subjects and prevalence of internet addiction was found to be 90%, where prevalence of mild IA was 48.8% followed by moderate and severe IA, 38.8% and 2.4% respectively (Fig. 1). Table 1 shows that majority of the study subjects were 20 years or more i.e. 70.6% and 29.4% were less than 20 years of age. Approximately 2/3rd of the subjects were males (65.7%) and app. 1/3rd was females (34.3%). Most of the study subjects (96%) were belongs to Hindu religion followed by Muslim and Sikh religion (4%). Majority of the subjects (76.1%) were belongs to nuclear family and 23.9% belongs to joint family. Most of subjects (84.6%) belongs to upper or upper middle SES followed by lower middle or lower SES (15.4%). Approximately 3/4th of the study subjects (74.1%) had permanent residence in Delhi and rest had permanent residence outside Delhi. More than half of the subjects (55.7%) were staying in hostel and 44.3% were non hosteller. According to year of study 36.8% of the study subjects were in third year, 36.3% in second year and 26.9% in rst year students.     Table 5 shows that 61.2% of respondents were correctly classi ed into categories of internet addiction by socio-demographic factors. This model correctly predicts 94.1% of subjects with no or mild internet addiction.   Table 6 shows that 63.7% of respondents were correctly classi ed into categories of internet addiction by socio-demographic factors. This model correctly predicts 95.8% of subjects with no or mild internet addiction.   Table 7 shows that 63.2% of respondents were correctly classi ed into categories of internet addiction by entertainment site usage. This model correctly predicts 89.9% of subjects with no or mild internet addiction.   Table 9 shows that 66.2% of respondents were correctly classi ed into categories of internet addiction by nal model. This model correctly predicts 92.4% of subjects with no or mild internet addiction.

Conclusion
Medical undergraduate students are highly vulnerable for internet addiction. We should create awareness among medical students regarding internet addiction and its potential harms; this could be included in foundation course of curriculum implementation support program (CISP) for MBBS students.

Discussion And Conclusion
In our study prevalence of internet addiction is 90%, whereas some other studies prevalence of IA ranges from 20%-46% [7,10]. High prevalence of internet addiction in this study might be due to demographic pro le of study subjects as majority (84.6%) belongs to upper or upper middle SES, moreover we have also included mild category of internet addiction.
We have developed different discriminant models using parameters like socio-demographic factors Medical undergraduate students are highly vulnerable for internet addiction. We should create awareness among medical students regarding internet addiction and its potential harms; this could be included in foundation course of curriculum implementation support program (CISP) for MBBS students. Initiative should be taken to provide ample opportunities for students to involve in extracurricular activities and interact with friends. There should be provision of counsellor for emotional and mental support of medical students as they are overburden with studies and long posting schedules.