INTRODUCTION

Eating disorders adversely affect a person’s physical, emotional, and social well-being.^[1] Out of the major eating disorders, bulimia nervosa and anorexia nervosa are the two most significant and majorly patent in adolescences.^[2] It has been reported that these disorders are likely to be three times more prevalent in females than males.^[3]

Most eating disorders involve focusing too much on body weight, body shape, and food, leading to certain dangerous eating behaviors. These behaviors can significantly impact the body’s ability to get appropriate nutrition. Eating disorders can harm the heart, digestive system, bones, and teeth and mouth and may lead to other diseases as well.^[4]

Anorexia nervosa is a diet disorder marked by a refusal to achieve a balanced body weight, a skewed view of the body, and an excessive fear of weight gain. This disorder makes one so obsessed by their beauty and the need to be slim that ultimately affects all facets of their life, directly or indirectly.^[5]

Bulimia Nervosa may lead to life-threatening situation, because individuals suffering from this disease, may consume vast quantities of food in an uncontrollable manner, seeking to discard calories with little intention of feeling satisfied, but may nevertheless be secretly hide food somewhere to deal with their feelings of fullness. After taking too much extra calories, the bulimia affectants try to lose them in a very unhealthy way.^[6]

Historically, eating disorders have been regarded as a Western culture-specific health issue. However, a growing body of research has investigated the rising prevalence of eating disorders in Asian countries, especially in developing countries such as Pakistan. An increased understanding of the epidemiology of eating disorders is a fundamental prerequisite for improvements in the diagnosis and management of eating disorders, as well as the rational distribution of health-care services. As a result, the prevalence of eating disorders and potential improvements has consequences for mental health programs.^[7]

There is a dearth of empirical statistics on incidence and prevalence of different eating disorders in Pakistan. Although a number of different surveys have been conducted to assess the eating disorders in Pakistan, Mumford et al.,^[8] Choudry and Mumford,^[9] Suhail and Nisa,^[10] Latif et al.,^[11] Memon et al.,^[12] Shaikh and Kayani,^[3] and Jamali et al.,^[13] there is a dire need to see the trends in the burden of eating disorders in Pakistan.

Eating disorders put a great burden on a country in a number of ways.^[14] Reporting incidence and prevalence of a disorder is very crucial for better understanding.^[15] Hence, there is a dire need to report the incidence and prevalence of eating disorders in Pakistan to help the policy maker better understand and deal with their burden.

The Global Burden of Diseases (GBD) is a valuable reserve for understanding the changing health issues that people face around the world in the twenty-first century. The GBD research, led by the Institute for Health Metrics and Evaluation (IHME), is the most rigorous global observational epidemiological study to date. GBD is a valuable platform for informing physicians, scholars, and policymakers, promoting transparency, and improving lives around the world by monitoring change within and across countries.^[16]

Over the last two decades, the IHME has evolved a framework for quantifying the burden of injures, illnesses, risk factors, and diseases to guide health service and policy development bodies to take suitable actions. GBD offers equivalent assessments of primary metrics of disease burden measurement including the incidence and prevalence rate of eating disorders, on regular basis.^[16]

The present study was designed to utilize the GBD 2019 database to systematically summarize and analyze the incidence and prevalence of eating disorders, bulimia nervosa and anorexia nervosa, and their fluctuations since 1990–2019 for Pakistan.

MATERIAL AND METHODS

Data sources

GBD 2019 has calculated each epidemiological quantity of interest, i.e., incidence, prevalence, deaths, years lived with disability (YLDs), years of life lost (sYLLs), and disability-adjusted life-years for 23 age categories, in 204 countries divided into seven superregions and 21 regions.^[17]

The Cause of Death Ensemble model and Spatiotemporal Gaussian Process Regression were used to calculate the cause-specific death rates and cause fractions, as previously stated.^[18–21] Cause-specific deaths were adjusted to match the total all-cause deaths calculated as part of the GBD population, fertility, and mortality estimates. YLLs were calculated by multiplying deaths by standard life expectancy at each age. “A Bayesian meta-regression modeling tool, DisMod-MR 2.1, was used to ensure consistency between incidence, prevalence, remission, excess mortality, and cause-specific mortality for most causes. Prevalence estimates were multiplied by disability weights for mutually exclusive sequelae of diseases and injuries to calculate YLDs. Uncertainty intervals (UIs) were generated for every metric using the 25^th and 975^th ordered 1000 draw values of the posterior distribution.”

In the GBD 2019, EDs were defined according to the Diagnostic and Statistical Manual of Mental Disorder (DSM) and International Classification of Disease (ICD) criteria. Different versions of the DSM (DSM-III, DSM-III-R, DSM-IV, DSM-IV-TR, and DSM-5) and ICD (ICD-9 and ICD-10) were accepted. The latest DSM-5 criteria, which was released in 2013, had included three ED subtypes: AN, BN, and BED along with it also incorporated a residual category, i.e., Other Specified Feeding and Eating Disorders (OSFEDs). OSFEDs embraces clinically significant ED that doesn’t meet the full criteria given for AN, BN, or BED. However, the GBD 2019 included only two ED subtypes, AN and BN. The ICD-10 codes used by GBD for eating disorders, AN and BN, are F50-F50.9, F50.0-F50.1, and F50.2-F50.5, respectively. We have extracted the age-standardized as well as age- and sex-specific incidence and prevalence rates from 1990 to 2019 from a publicly available online tool provided by the IHME known as the (Global Health Data Exchange). This query tool is used to quantify the temporal trends of ED, AN & BN incidence and prevalence rates and can be availed at: http://ghdx.healthdata.org/gbd-results-tool.^[22] Percentage change and annualized rate of change of the estimates were reported. Moreover, crude rates and the 95% UIs were used to compare ED’s secular trend between age groups.

Joinpoint regression analysis

The identification of changes in the secular trend is critical to the analysis of disease incidence and prevalence data. To determine the magnitude of the time trends in incidence and prevalence rates of eating disorders, anorexia nervosa and bulimia nervosa, the average annual percentage change (AAPC) and corresponding 95% confidence interval (CI) was evaluated using joinpoint regression analysis. The logarithmic transformation of the incidence and prevalence rates was conducted and the standard errors were taken as constant.^[23] AAPC was calculated by taking geometrically weighted average of various annual percentage change (APC) values from the regression analysis.^[24] The average APC (AAPC) was estimated using best model considering maximum 5 joinpoint, i.e., 6 segments for the full range of our study periods.

To estimate the APC, the following model is used:

log(Y_x )=b₀+b₁x, where log (Yx) is the natural logarithm of the rate in year x.

Then, the APC from year x to year x + 1 is:

This analysis was performed using ‘joinpoint’ software (Joinpoint Regression Program, version 4.7.0.0, NCI) from the Surveillance Research Program of the US National Cancer Institute.

RESULTS

Descriptive analysis

Gender-wise trends in the age-standardized incidence rate (ASIR) for eating disorders, anorexia nervosa and bulimia nervosa, from 1990 to 2019 of Pakistan are depicted in Figure 1. For all year’s ASIR of eating disorders and bulimia nervosa was higher among males as compared to females, but it was reversed in the case of anorexia nervosa where ASIR was higher among females as compared to males.

The ASIRs of eating disorders, anorexia nervosa and bulimia nervosa, for both males and females in Pakistan have been increased over the period 1990–2019. For males, the ASIR of eating disorders has largely increased in 2019 (194.59 new cases/100,000 males) as compared to 1990 (171.22 new cases/100,000 males); similarly, ASIR of eating disorders among females had also increased in 2019 (86.84 new cases/100,000 females) from 77.71 new cases/100,000 females in 1990 [Figure 1a]. For males, the ASIR of anorexia nervosa had increased in 2019 (7.73 new cases/100,000 males) as compared to 1990 (6.86 new cases/100,000 males); similarly, ASIR of anorexia nervosa among females had also increased in 2019 (17.53 new cases/100,000 females) from 15.91 new cases/100,000 females in 1990 [Figure 1b]. For male, the ASIR of bulimia nervosa had increased in 2019 (186.86 new cases/100,000 males) as compared to 1990 (164.37 new cases/100,000 persons); similarly, ASIR of bulimia nervosa among females had also increased in 2019 (69.32 new cases/100,000 persons) from 61.80 new cases/100,000 persons in 1990 [Figure 1c].

Gender wise trends in the age-standardized prevalence rate (ASIR) for eating disorders, anorexia nervosa, and bulimia nervosa from 1990 to 2019 of Pakistan are depicted in Figure 1. ASIR of eating disorders and bulimia nervosa was higher among males as compared to females, during mentioned time frame.

The ASPRs of eating disorders, anorexia nervosa and bulimia nervosa for both males and females in Pakistan has been increased over the period 1990–2019. For male, the ASPR of eating disorders has increased from 74.73 cases/100,000 males in 1990–91.09 cases/100,000 males in 2019; Similarly, ASPR of eating disorders among females had also increased in 2019 (152.60 cases/100,000 females) from 126.80 cases/100,000 females in 1990 [Figure 2a]. For male, the ASPR of anorexia nervosa had slightly increased in 2019 (18.69 cases/100,000 males) as compared to 1990 (16.44 cases/100,000 males); Similarly, ASPR of anorexia nervosa among females had also increased in 2019 (51.97 cases/100,000 females) from 44.85 cases/100,000 females in 1990 [Figure 2b]. For male, the ASPR of bulimia nervosa had increased in 2019 (72.96 cases/100,000 males) as compared to 1990 (59.63 cases/100,000 persons); Similarly, ASPR of bulimia nervosa among females had also increased in 2019 (101.36 cases/100,000 persons) from 82.61 cases/100,000 persons in 1990 [Figure 2c].

Joinpoint regression analysis

Figure 3 and Table 1 show APC of ASIRs of eating disorders, anorexia nervosa, and bulimia nervosa in Pakistan from 1990 to 2019 for both males and females. The regression model showed a significant increasing pattern in the ASIR of eating disorders between 1990 and 2019 for both males and females for all trends; however, a nonsignificant declining trend was observed in case of males between year 1994 and 2001 (APC = −0.10; 95% CI: −0.20–0.00; P > 0.05) [Figure 3a and b]. A significant increasing trend has been observed in the ASIRs of anorexia nervosa between 1990 and 2019 for both males and females for all trends except between 1995–2000 among males (APC = −0.15; 95% CI: −0.23 to − 0.08; P < 0.05) and 2017–2019 among females (APC = −0.10; 95% CI: −0.19 to −0.00; P < 0.05) where a significant declining trend was observed [Figure 3c and d]. A similar increasing trend was observed in the ASIRs of bulimia nervosa between 1990 and 2019 for both males and females for all trends; however, a nonsignificant declining trend was observed in case of males between year 1994 and 2001 (APC = −0.10; 95% CI: −0.20–0.01; P > 0.05) [Figures 3e and 3f].

Figure 4 and Table 2 shows APC of ASPRs of eating disorders, anorexia nervosa, and bulimia nervosa in Pakistan from 1990 to 2019 for both males and females. The regression model showed a significant increasing pattern in the ASPR of eating disorders between 1990 and 2019 for both male and female for all trends; however, a nonsignificant increasing trend was observed in case of males between year 1993–2001 (APC = 0.01; 95% CI: −0.08–0.09; P > 0.05) [Figures 4a and b]. A significant increasing trend has been observed in the ASPRs of anorexia nervosa between 1990 and 2019 for both male and females for all trends except between 1995–2000 among males (APC = −0.21; 95% CI: −0.28 to − 0.15; P < 0.05) and 1994–2001 among females (APC = −0.10; 95% CI: −0.13 to −0.07; P < 0.05) where significant declining trend was observed [Figure 4c and d]. Similar increasing trend has been observed in the ASPRs of bulimia nervosa between 1990 and 2019 for both male and females for all trends; however, a nonsignificant increasing trend was observed in case of males between year 1994 and 2001 (APC = 0.04; 95% CI: −0.08–0.15; P > 0.05) and 1994–2001 among females (APC = 0.08; 95% CI: −0.02–0.19; P > 0.05) [Figures 4e and f].

The incidence rates of eating disorders, anorexia nervosa, and bulimia nervosa by age group in Pakistan in 2019 are shown in Table 3. In 2019, the ASIRs of eating disorders in Pakistan were 194.59 (95% UI: 129.54–275.07) and 86.84 (95% UI: 60.87–121.43)/100 000 population of males and females, respectively. The ASIRs were 7.73 (95% UI: 5.44–10.65) in males and 17.53 (95% UI: 12.34–24.38) in females for anorexia nervosa, and 186.86 (95% UI: 121.90–268.18) in males and 69.32 (95% UI: 44.35–103.46) in females for bulimia nervosa, respectively, per 100 000 population.

The AAPCs of age–sex-specific incidence rates of eating disorders and their subtypes from 1990 to 2019 are also depicted in Table 3. From 1990 to 2019, the ASIRs of eating disorders in Pakistan has been increased by 0.45% (95% CI: 0.37–0.53) in males and 0.39% (95% CI: 0.35–0.43) in females. The ASIRs of anorexia nervosa rose by 0.41% (95% CI: 0.39–0.43) in males and 0.33% (0.32–0.35) in females, whereas the ASIRs of bulimia nervosa rose by 0.45% (95% CI: 0.37–0.54) in males and 0.40% (0.36–0.45) in females.

The highest incidence rates of eating disorders among males and bulimia nervosa among both males and females were observed in the group aged 15–19 years, followed by the groups aged 20–24 and 25–29 years. For anorexia nervosa, among both males and females and eating disorders among females, the group aged 10–14 years had exhibited the third highest sex-specific incidence rates, which were lower than the estimates of the groups aged 15–19 and 20–24 years. Significant increases in the gender specific incidence rates were also observed in eating disorders, anorexia nervosa, and bulimia nervosa irrespective of the age group. Moreover, the increments in ASIRs of eating disorders, anorexia nervosa and bulimia nervosa were more pronounced after 2000s as they were in the 1990s.

The prevalence rates of eating disorders, anorexia nervosa, and bulimia nervosa by age group in Pakistan in 2019 are shown in Table 4. In 2019, the ASPRs of Eating disorders in Pakistan were 91.09 (95% UI: 62.20–123.89) and 152.60 (95% UI: 113.56–195.69) cases/100 000 population of males and females, respectively. The ASPRs were 18.69 (95% UI: 12.85–26.33) in males and 51.97 (95% UI: 36.55–73.99) in females for anorexia nervosa, and 72.96 (95% UI: 46.27–103.05) in males and 101.36 (95% UI: 68.18–140.36) in females for bulimia nervosa, respectively, per 100 000 population.

The AAPCs of age–sex-specific rates of eating disorders and their subtypes from 1990 to 2019 are also depicted in Table 4. From 1990 to 2019, the ASPRs of eating disorders in Pakistan increased by 0.69% (95% CI: 0.63–0.75) in males and 0.64% (95% CI: 0.59–0.70) in females. The ASPRs of anorexia nervosa rose by 0.44% (95% CI: 0.42–0.46) in males and 0.51% (0.49–0.54) in females, whereas the ASPRs of bulimia nervosa rose by 0.71% (95% CI: 0.62–0.80) in males and 0.72% (0.64–0.79) in females.

The highest prevalence rates of eating disorders among males and bulimia nervosa among both males and females was observed in the group aged 30–34 years, followed by the groups aged 25–29 years. The highest prevalence rates of eating disorders among females was observed in the age group 20–24 years followed by the groups aged 25–29 and 30–34 years. For anorexia nervosa, among both males and females, the group aged 25–29 years exhibited the third highest sex-specific prevalence rates, which were lower than the values of the groups aged 15–19 and 20–24 years. Significant increases were observed in the gender specific prevalence rates of eating disorders, anorexia nervosa, and bulimia nervosa irrespective of the age group. Moreover, the increments in ASPRs of eating disorders, anorexia nervosa and bulimia nervosa, were more pronounced after 2000s as they were in the 1990s.

DISCUSSION

Eating disorders are major problem among the people of developing as well as developed countries.^[25] Prevention of eating disorders is one of the key public priority. It has been reported that about 80% of the people affected with eating disorders do not go through a formal treatment and interestingly the available ‘evidence based treatments’ are effective for <50% of the affectants.^[26] The major problem is that a very often research studies are conducted on the incidence and prevalence of eating disorders. Furthermore, the estimation of the prevalence of eating disorders is also very difficult as by the WHO statistics; the data of eating disorders are included in the category of mental disorders.^[27] Hence, this makes very difficult for the governmental, decision-making, and nongovernmental bodies to work on the minimization and treatment of people affected with eating disorders. In short, there is a dire need to report the prevalence and incidence of eating disorders for such decision-making bodies. The current study utilizes the GBD database to analyze the incidence of eating disorder, anorexia nervosa and bulimia nervosa, in a period of 1990–2019. The current study is unique to its kind to explore the trends in burden of eating disorders from 1990 to 2019; no such comprehensive study has previously been published.

Most of the previous studies conducted to screen the eating disorders in Pakistan could not give substantial information to report the incidence and prevalence of eating disorders in Pakistan as such studies are usually based on self-describing questionnaires and face-to-face interviews. These studies may add a value to identify the eating disorders but these studies cannot define the main features and some times may over or under estimate the eating disorders.^[7] Examples of such studies in Pakistan include Zahid et al.,^[28] Javed and Rafique,^[29] Jamali et al.,^[13] Warsi et al.,^[30] Memon et al.,^[12] Larif et al.,^[11] Suhaild and Nisa,^[10] Muford et al.,^[8] and Choudry and Mumford.^[9]

It has been articulated in the finding of current study that the ASIR of eating disorders and bulimia nervosa is higher among the males comparing to females from 1990 to 2019, but in case of anorexia nervosa ASIR was higher among the females comparing to males from 1990 to 2019. ASIR is a summary measure of the rate that a population would have if it had a standard age structure. Age-standardization is very essential in comparing populations that could differ with respect to age.^[31] The current study also articulate that the ASIR of eating disorders, anorexia nervosa and bulimia nervosa, for both the males and females in Pakistan have been increasing over the time (from 1990 to 2019). Furthermore, the raise in ASIR of eating disorders, anorexia nervosa and bulimia nervosa were more obvious from 2000 to 2019 as they were from 1990 to 2000.

From 1990 to 2019, the ASIRs of eating disorders in Pakistan increased by 0.45% (95% CI: 0.37–0.53) in males and 0.39% (95% CI: 0.35–0.43) in females. The ASIRs of anorexia nervosa rose by 0.41% (95% CI: 0.39–0.43) in males and 0.33% (0.32–0.35) in females, whereas the ASIRs of bulimia nervosa rose by 0.45% (95% CI: 0.37–0.54) in males and 0.40% (0.36–0.45) in females. The current reported AAPCs of age-sex specific rates of eating disorders are alarming. Greater changes could be observed among the males than the females for the eating disorders from 1990 to 2019, i.e., 0.45/0.39 for eating disorders, 0.41/0.33 for anorexia nervosa, and 0.45/0.4 for bulimia nervosa among males/females. Interestingly, the current study also found that the age group 15–19 years possesses the highest incidence rates of eating disorders among the males and bulimia nervosa among both males and females.

As per the latest DSM-5 criteria, the eating disorders have been divided in to four categories that include anorexia nervosa, bulimia nervosa, binge eating disorders, and OSFEDs.^[32] However, the GBD has did not follow the latest DSM-5 classification and divided the eating disorders in only two categories as anorexia nervosa and bulimia nervosa. This may be a possible factor that the incidence of eating disorders is higher in males as compared to females. Very similar results have been observed by Wu et al.,^[7] in their study on trends and burden of eating disorders in China in past three decades.

It was seen in the current study that from 1990 to 2019 the ASPR of eating disorders, anorexia nervosa and bulimia nervosa, was higher among the females then comparing to males. However, a significant increase in the ASPR for eating disorders, bulimia nervosa and anorexia nervosa, was observed for both the genders from 1990 to 2019. ASPR was seen to increase from 74.73 cases/100,000 males to 91.09 cases/100,000 males and from 126.80 cases/100,000 females to 152.60 cases/100,000 females (from 1990 to 2019, respectively).

The cause in hike of eating disorders, bulimia nervosa and anorexia nervosa, are complex. The results of the existing literature on eating disorders point multifactor determination, i.e., environmental and generic factors.^[33] As by a number of the epidemiological studies, women are reported to be more prone to the development of eating disorders.^[34] There could be a number of different reasons for the said outcome; the increased impact of media and social media on women could be a contributing factor in women to develop eating disorders greatly compared to men.^[35] The trend of women digest magazines (which are very common in Pakistan), radio, and TV programs, especially morning shows could also be significant contributors for women. Another possible reason is the celebrity influence and body image to develop eating disorders. Very similar justifications have been reported by Ioannidis et al.,^[36] Brown and Tiggemann,^[37] Jiotsa et al.,^[38] Jurczak et al.,^[27] Siew^[39] Constructed,^[40] Jordan et al.,^[41] Shoger,^[42] and Field et al.,^[43]

Studies also report that teenage years are more likely to develop the eating disorders,^[27] and very similar results were obtained in current study. The possible aspect influencing young people’s perceptions in leading to eating disorders today could be their quest for longevity. Furthermore, now, many people see the need of a balanced lifestyle and participating in activities such as athletics, consuming foods that are healthy, and equipment that help track it, or even purchase, to sustain a healthy level of activity.^[27] This is plausible, considering the large number of young males to develop eating disorders. Ioannidis et al.^[44] reported that young people go for over exercise that ultimately lead to development of eating disorders. Jurczak et al.^[27] articulated that young students in lower grades are more prone to eating disorders.^[45] The study says that 14^th year is the lower age limit for the occurrence of anorexia nervosa. Abebe et al.^[46] articulated that the peak period of incidence of bulimia nervosa among girls is the 14^th–16^th life year. However, Stice et al.^[47] in their study determined that the peak incidence of bulimia nervosa falls on 16–20^th year while that of anorexia nervosa falls on 19–20^th.

Very similar reasons could be given for the development of eating disorders among young people as were for the females, that include celebrity influence,^[37] body image,^[48] and last but not the least social media.^[36,49]

CONCLUSION

A continuously increasing trend was observed in the incidence and prevalence rates of eating disorders and their subtypes, i.e., anorexia nervosa and bulimia nervosa, in Pakistan from 1990 to 2019 regardless of sex and age group. This predicts that the relative burden caused by eating disorders in Pakistan will continue to increase in the future and further etiological studies should be required in this area. Furthermore, adolescents are seen to have the highest incidence rates of eating disorders followed by young adults, which suggests that age is playing a crucial role. An enhanced understanding of high-risk groups and onset patterns in eating disorders among those groups could facilitate the early identification of individuals who are at risk of developing the disease that will help in timely interventions leading to the reduction in the eating disorders burden in Pakistan. It is recommended by the authors that future studies should be planned to discuss trends and burden of eating disorders around the globe in past three decades.

Availability of data and materials

All data were obtained from the public open database: Global Health Data Exchange (GHDx) query tool (http://ghdx.healthdata.org/gbd-results-tool).

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.