Abstract

Objectives Although both infertility and mental illness are serious public health issues, relatively little is known about the mental health of women seeking fertility help in many lower and middle-income countries. This survey analyses the type of psychological burden that affects women who are unable to access in vitro fertilisation treatment (IVF), the risk factors for depression and anxiety among those seeking IVF treatment in Pakistan, and the existing mental health issues in women who seek help for the treatment of infertility to understand the accessibility and availability of specialist services. The aim of the study was to estimate the prevalence of depression among women seeking fertility treatments in three different settings at the same time.

Design A cross-sectional survey with convenience sampling.

Settings Different types of settings: private and public hospitals and traditional clinics in the community.

Primary outcome measure Screening for depression using HADS (Hospital Anxiety and Depression Scale) and psychiatric interviews of those scoring above the cut-off level on HADS with WHO SCAN (Schedules for Clinical Assessment in Neuropsychiatry).

Results The study sample consisted of 485 participants. The complete demographic data were available for 477 women. The HADS questionnaire was completed by 466 women, of whom 162 also completed the SCAN interview. Ages ranged from 15 to 60 years, with a mean age of 28.5 years, and 100% were married. According to HADS, 69% of cases of depression and anxiety were diagnosed, whereas 50% of those who were interviewed with SCAN had a diagnosis according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), varying across settings (63% in government hospital, 42% traditional setting and 36% private hospital, respectively). There was a twofold risk of depression among the participants attending government hospitals (OR=2.4, CI=1.2, 4.7, p=0.03) as compared with women attending traditional clinics, and there was a slightly lower risk among the participants of private hospitals (OR=0.9, CI=0.4, 2.3, p=0.03) than other groups. The main risk factors found were having a traditional previous contact for treatment and attempting time (years). The HADS showed efficient performance with 97% sensitivity and 49% specificity with cut-off score 12. The main risk factors for depression were identified in those who had sought any previous treatment, having very low income and increased time spent attempting, with a clear difference in the risk of depression between those being treated at government hospitals and those following other treatment pathways.

Conclusion In this study of women seeking infertility treatment in Pakistan, there was variation in the proportion meeting diagnostic criteria in different settings (as measured by HADS and WHO SCAN). This variation is likely to be due to the markedly different communities accessing the particular setting as well as the quality of treatment or help provided. It was clear overall that repeated attempts to seek help and length of trying are associated with higher prevalence of depression in this population. It is essential to provide appropriate and affordable fertility treatment in all government hospitals to ameliorate the effects of prolonged treatments and time spent struggling to reach places offering medical fertility care, and to introduce the concept of mental healthcare at fertility clinics within these government hospitals. The high risk of depression among the lower income group shows that providing access to specialist care and assisted reproduction to poor patients is urgently needed.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

STRENGTHS AND LIMITATIONS OF THIS STUDY

• The fieldwork was pragmatic given its limited timeframe and varied settings.

• The study methods were devised in view of the identified gaps in research and weaknesses in the published literature.

• Diagnosis of depression and anxiety based on WHO Schedules for Clinical Assessment in Neuropsychiatry interviews according to both Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition and International Classification of Diseases criteria, a gold standard.

• Random sampling from a known denominator would have been preferable with longitudinal components; however, it was not possible in such settings.

Introduction

According to the Royal College of Obstetricians and Gynaecologists, ‘Fertility is the ability to conceive a baby and, for a woman, to become pregnant’.1 Infertility, on the other hand, has a broad definition because it has been defined using different approaches across various disciplines. The definition has a significant impact on clinical outcomes, including those reported in research studies.2 Within infertility clinic settings, infertility is defined as ‘the inability to establish a pregnancy within a specified period of time, usually 1 year’.3 The approach taken by epidemiologists has been different. In many epidemiological studies, it is defined as 2 years exposure to the risk of pregnancy without becoming pregnant. On the other hand, the definition commonly used by demographers is 5 years’ exposure without a viable birth.4

Moreover, a woman who has previously conceived and successfully given birth yet is unable to do so subsequently is classified as having secondary infertility.5 The results of a global burden of disease study suggest that, globally, the age-standardised prevalence of infertility increased by 0.370% per year for females from 1990 to 2017.5 Infertility is a serious issue in South Asia, but this problem is rarely given attention because the region is densely populated and is poverty ridden. Focus instead has largely been on the challenge of overpopulation in such countries including India.6

The proportion of women with primary infertility in Pakistan was reported as 5% in the Pakistan Reproductive Health and Family Planning Survey 2000–2001.7 A global review of literature on mental health aspects of women’s reproductive health carried out in the year 2009 by WHO and UNFPA (see online supplemental annex 1 for full list of abbreviations) called for research into the demand for, and nature of, infertility services in developing countries as well as the long-term psychological consequences of infertility.2

To date, relatively little research has been conducted on infertility in low-income and middle-income countries (LMICs). Most research has been conducted within high-income countries contexts. One systematic review examined the evidence on the impact of access to assisted reproductive technologies (ART) on longer term mental health and well-being, concluding that the relationship between infertility and longer term mental health and well-being is complex. This review recommended that suitable longitudinal studies or cross-sectional studies of sufficient sizes using validated measures to compare outcomes to appropriate population norms are needed to address what was identified as an evidence gap.8

The reproductive health of a woman and her ability to reproduce are interlinked, and the latest guidelines given by UNFPA and WHO insist that attention to population control in countries with high fertility rates should not preclude identification and treatment of infertility.2 Health planners have focused mainly on overpopulation in developing countries and emphasised birth control, of course vitally important, but in doing this have neglected the issue of infertility. Female infertility can be related to adverse biological and social consequences.9 In many developing countries, it is seen as a significant public health problem by affected individuals.10

Infertility can be extremely distressing, most particularly if ART is unsuccessful, such as after multiple IVF procedures where the likelihood of bearing a child becomes manifestly unlikely. This exacerbates what is already a more vulnerable state, increasing the possibility of clinical depression.11 12 The psychological aspects of infertility are, therefore, an important area for such women, with the possibility of better outcomes of the fertility treatments themselves and the mental well-being of women seeking care.

Becoming a mother has high priority in the culture of Pakistan’s women. Fertility is considered a blessing; childlessness is a cause of pity.13 The importance of infertility as a public health problem and a social problem can be judged from varying perspectives including individuals, couples, wider family, healthcare providers and society at large. The impact of stress following a fertility treatment pathway involving invasive procedures could result in severe depression leading to lower rates of pregnancy during infertility treatment with ART,14 an association that this study intends to explore.

Several instruments are available that have been validated in a Pakistan setting for studying depression. These include Schedules for Clinical Assessment in Neuropsychiatry (SCAN), a semistructured psychiatric interview developed as a part of the framework of the WHO/Alcohol, Drug Abuse and Mental Health Administration Joint Project on Standardisation of Diagnosis and Classification, with the aim of developing a comprehensive procedure for clinical examination that is also capable of generating many of the categories of the International Classification of Diseases, 10th edition (ICD-10) and the Diagnostic and Statistical Manual of Mental Disorders (DSM), revised third edition.15 16

Pakistan’s overall literacy rate at the time of the survey was 45% (56.5% for men and 32.6% for women in 1998), which was far behind most of the countries in the region.17 An overview of the educational attainment of women in Pakistan attributes higher illiteracy rates to rural setup and low income.18

The aim of the study was to provide an estimate of the prevalence of depression and anxiety in different settings where fertility treatment is sought within a LMIC, using WHO SCAN. Relevant factors, such as having access to the treatment and the provision of the required treatment and help, were studied across different pathways.

Methods

Results

502 women were approached for HADS screening in total. Of these, 485 were successfully interviewed, most to completion. The questionnaires were filled in either by the researcher/research assistant or the participant, depending on the situation and the ability of the participant (online supplemental figure 1). Following the screening for depression and anxiety, 162 participants out of 485 from PIMS and Bari Imam (traditional setting) were interviewed using Schedules for Clinical Assessment in Neuropsychiatry (Urdu version). Data collection was completed over a 6 month time period, during which all the possible research settings were covered; 250 participants from medical and 235 from traditional settings were recruited (see online supplemental figure 2). Two women each at the private medical clinic and the PIMS, and 11 other women refused to participate in the study at the time of recruitment. Reasons given for refusal were mistrust about revealing their personal sensitive information, fear of their husband or time constraints. Nevertheless, the overall response of the women approached was encouraging, as some of them asked questions about the purpose of this research and the possible benefits that might emerge.

The characteristics of the participants across the study settings are summarised in table 1. Women seeking treatment did not differ in age (χ²=11.9, p=0.1) within the three settings. Similarly, no significant difference was observed when variables (history of miscarriage, family system, attempting time or an inclination towards a suicidal attempt) were considered. However, a considerable variation in the area of residence was observed within the three settings (χ²=51.3, p<0.0001). A greater proportion of the women attending the government hospital were from Islamabad (36%) as compared with those attending traditional (9%) and private (10%) clinics. There was a significant difference in the educational level within the three settings (χ²=59.2, p<0.0001). As might be expected, a higher proportion of women attending government hospitals and traditional healers were either illiterate or barely educated, while those who went to private medical facilities were better educated.

There was also a variation in the income levels, as most of the participants who visited the government hospital or the traditional settings had lower income levels as compared with a large proportion of those who went for private medical care (χ²=71.7, p<0.0001). The difference in the type of medical care sought previously and the type of investigations carried out was also evident as those attending the modern private medical clinic had the highest percentage of patients who previously sought medical care (χ²=103.1, p<0.0001) and had investigations (χ²=28.9, p<0.0001). There is a clear indication of an inclination towards seeking medical care through government hospitals due to obvious financial constraints (see online supplemental figure 3, table 1 and additional analyses of HADS score in online supplemental material).

The study sample consisted of 477 participants in total with a mean age of 28.5 years. The HADS questionnaire was completed by 466 women, out of whom 162 also completed the SCAN interview (see table 1). 69% of the participants screened with HADS were classified with depression and anxiety, and 50% of those interviewed with SCAN met the DSM-IV diagnostic criteria. Detailed data are provided in the figures (online supplemental figures 1 and 2).

Univariate model based on SCAN DSM-IV diagnosis showed that, women who had traditional contact as their previous contact were four times more likely to be suffering from depression and anxiety (OR=4.3, 95% CI=1.4, 13.2, p=0.03) (table 2). The same model, when applied with HADS diagnosis instead of SCAN, also found a significant relationship between previous contact and HADS (p=0.01), showing three times greater risk of psychiatric illness among those who previously went to traditional healers or both medical and traditional (table 3). This difference in the two analyses could be due to the difference in the number of participants and the performance of the diagnostic measures.

Time spent in seeking treatment (Attempting time) showed a significant relationship with HADS (p=0.01). There is a twofold risk in those who had been attempting to conceive for more than 9 years (OR=2.4, CI=1.3, 4.5, p=0.01). Other factors including age and having had a miscarriage were not identified in the analyses (table 3).

A multivariable logistic regression model was used to further assess these risk factors (tables 4 and 5). When adjusted for age and each other the model showed no significant relationship between previous contact and SCAN diagnosis (p=0.21) (table 4).

As far as the multivariate analysis with HADS is concerned, a significant association was observed for attempting time and setting when adjusted for age and other characteristics in the model (p=0.01) (table 5). Here the ORs after the adjustment are similar to those in the univariable regression analyses. Women attempting to conceive for more than 9 years are at a higher risk of depression and anxiety than those who have been trying for nine or less than 9 years (OR=2.4, CI=1.2, 5.0, p=0.01) where age is accounted for.

Similarly, women attending government hospitals are at a higher risk compared with participants from traditional settings (OR=1.3, CI=0.7, 2.4, p=0.01), while those attending private clinics are at lower risk (OR=0.5, CI=0.3, 0.9, p=0.01) (table 5). In this model, previous contact showed no significant relationship (p=0.2). However, when the previous contact was traditional, it showed a twofold risk of psychiatric illness (OR=2.4, 95% CI =0.8, 6.6), and when it was both medical and traditional, the risk was twofold as well (OR=2.2, 95% CI=0.9, 5.7).

In multivariable analysis seeking to examine broader influences, the main risk factors for depression were identified as prior experience seeking any previous treatment, possessing very low income and increasing attempting time, with a clear difference in the risk of depression between those being treated at government hospitals and those following other treatment pathways.

Discussion

The study results reveal a high prevalence of depression and anxiety among women seeking fertility treatments in Pakistan. There was a difference in the prevalence estimates and the associated risk factors across the three settings, with the highest prevalence estimates among the attendants of the government hospital.

Just a little under a half (42%) of the traditional setting participants who were interviewed with SCAN met diagnostic criteria for depression or anxiety or both according to DSM-IV criteria. One in five (19%) of the traditional setting participants reported suicidal ideation/attempts. This group requires particular attention from the health policy makers, as only 20% of the participants at traditional settings reported earlier psychiatric services contact. The treatment provided at herbal and homeopathic clinics does not comply with the health and safety regulations of the health authorities, which could in turn affect the physical, psychological as well as social well-being of these women, in case the treatment provided is substandard.

The study methods were devised in view of the identified gaps in research and weaknesses in earlier such endeavours. A convenience sample with a cross-sectional approach was used in this study, for pragmatic reasons, to simultaneously look at different age groups in different settings. While this has many limitations, it was the only approach feasible in these settings, and was demonstrably effective, as many of those approached in person agreed to take part. Alternative study designs could have produced more generalisable results reflecting the health-seeking behaviour of women with infertility in Pakistan in a wider perspective. The cross-sectional approach restricted the study findings to the current situation of the participant, not providing enough retrospective data or the opportunity to follow-up. Not having a control group of normal population restricts the generalisation of results within certain settings. It is also difficult to differentiate between the health needs of women seeking healthcare for other purposes as compared with those seeking healthcare because of their infertility. A comparison group of the normal population would have helped to interpret the findings of this study in a more robust way.

The response rate could not be measured because there was no definable denominator from which to calculate this. The data were collected from a variety of settings using convenience sampling. The data for this study were collected by accessing populations in urban, rural, traditional, as well as medical settings. In this study, women in particular Pakistani settings were sampled, therefore the prevalence of depression and anxiety cannot be assumed to reflect their prevalence in the population of infertile women in the general population. There could have been an expansion of the study sample if other hospitals in the vicinity of selected settings were included, but unfortunately, the ethical and administrative approvals set by the authorities are very lengthy, requiring a longer time frame for the study. Sampling more than one example of each setting would have made the results more representative of infertility settings. For more generalisable results, a larger study in the population and infertility service settings with relevant questions regarding seeking help and the nature of that help for primary infertility with high response rates could provide the true scale of the concern for the population of Pakistan. Such a design would give estimates for primary infertility and what proportion seek help and the type of help they seek, as well as what anxiety and depression might be for all these groups. The research reported in this paper cannot provide such estimates given its design. However, such a study is unlikely to ever be feasible, and this study represents an important finding for clinicians providing such services, given the high levels of psychiatric morbidity observed.

It is possible that this method introduced potential bias, as those willing to take part may have been more likely to encourage others with greater likelihood of mental distress to take part; 502 were approached, and seven did not agree to take part.

The findings of this study are in line with the studies carried out in other Muslim countries such as Iran and Bangladesh. The Iranian study found 44% women with symptoms of psychiatric illness and treatment time as a significant risk factor.23 Similarly, a qualitative study carried out in Bangladesh revealed that there is no public health programme that focuses on infertility in Bangladesh, and some very expensive private clinics provide infertility services.24 Other previous researchers also suggested that more data on the prevalence and aetiologic factors of infertility are certainly required when thinking about transferring limited economic resources into reproductive healthcare since resource-poor countries, especially in sub-Saharan Africa and Asia, have very little access to country and time-specific data.25

The overall prevalence estimates are comparable with the findings of studies carried out elsewhere. For example, a study carried out in India to estimate psychiatric morbidity among infertile women as compared with fertile women and determine the social effect of primary infertility on infertile women: the prevalence of psychiatric morbidity in women with infertility was 45% (15% in fertile women). The odds for psychiatric morbidity in women experiencing infertility were 4.6 times greater than that for fertile women.26

Similarly, studies on childlessness conducted on treatment-seeking behaviour in Egypt, Mozambique, Zimbabwe, Nigeria, Bangladesh, Thailand and Mexico revealed that the treatment for childlessness was sought from many sources, varying among home treatment, the formal medical system, government and private herbal and spiritual healers, traditional reproductive health specialists (Nigeria), diviners, priests, spiritual churches, sacred places and the temple Buddha (Thailand).27–31 However, studies conducted in Andhra Pradesh of rural India32 showed preference for allopathic medicine, with around three-quarters starting treatment of this kind.33

Previous research has also shown that, on average, a woman went to three healthcare providers (general practitioners, gynaecologists and traditional birth attendants) in her quest for assistance. The effects that infertility had on these women ranged from social pressure to coercion by in-laws and/or social isolation.34

There is no evidence of research on women with infertility within mental health service settings. This could also prove to be beneficial for understanding the psychological consequences of infertility and determining the right kind of preventive as well as treatment measures to be taken for such women. Considering the risk factors identified by this study, including low income, illiteracy, previous traditional contact and time spent in attempting to conceive, it would be useful to conduct case–control studies to see how far the reduction of these factors would help to reduce the level of depression and anxiety.

Women in the government hospitals mainly comprise medical treatment and surgical procedures, without attention to their mental health. There was no provision for ART. The high levels of psychological distress could be reduced if the quality of care is improved, and viable ART introduced in the fertility treatment package. Thirty-three per cent of women in this group reported suicidal ideation/attempts, which clearly depicts their despair and need for integrating appropriate mental healthcare in the system.

The study confirms the high prevalence of psychiatric morbidity in women with infertility in different settings. While we have been able to estimate prevalence and risk factors through the study, it is essential that the gaps in knowledge about quality of medical care at fertility clinics, health and safety standards of traditional fertility treatments, ethics of assisted reproduction in Pakistan, the cost-benefit analysis and provision of psychological interventions are filled. The study also identified the weaknesses in the system such as poor access to health facilities, lack of modern treatment facilities in the public health sector, inadequate medical investigative procedures prior to fertility treatments and lack of integration with mental health services. Moreover, it was found that a lack of education and financial constraints were among the factors adversely affecting mental health. Findings regarding suicidal attempts and ideation make it even more necessary to detect cases of severe depression in these settings to avoid the risk of suicide. These inferences would be useful in designing a more efficient fertility as well as mental health services network.

This study used a different methodology as compared with other such studies conducted in the past, particularly in Pakistan. Using this strategy generated reliable psychiatric diagnoses which helped us to understand the epidemiology of depression and anxiety among women with infertility in Pakistan.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Acknowledgments

Professor Paul Pharoah and Professor Peter B Jones, University of Cambridge, Professor Ghazala Mahmud, Pakistan Institute of Medical Sciences, Dr Farah Qadir (for field work advice in Pakistan), Eli Harris Bodleian Libraries, University of Oxford, Professor Jenny Hewison University of Leeds (for her guidance at a very initial planning stage), Newnham College Cambridge University, Gonville & Caius College nursery for looking after my children when I did this work, my father Fida Hussain Chaudhri Retd Joint Secretary Govt of Pakistan for his continued support and my late mother Azra Sultana for her prayers.