Introduction

Over the last few decades, the increasing recognition of the impact of Type 2 Diabetes Mellitus (T2DM) on Quality of Life (QoL), mental health and overall physical and psychological health along with their useful measurement instruments has been well addressed in scientific literature [1]. The benefits of evaluating QoL and mental health in patients with T2DM have been appreciated. This includes the evaluation of the burden of the disease and its complications, which may contribute to the development of the most appropriate management and treatment plans in these vulnerable patient groups [2].

Moreover, physicians caring for patients with comorbid chronic illnesses that affect their QoL and mental health, such as T2DM, need to prioritise their diabetes management to ensure better care with the aim to focus on how healthcare systems influence these decisions [3]. This includes the stability of these decisions over time, with continuous surveillance based on proper and validated measurements [3,4,5,6].

Overall, the nature of QoL is complex and multidimensional with a variation in tools used between studies. The Australian Centre for Quality of Life’s directory of instruments reflects this further where there are more than 1000 variables included and although these intend to measure QoL each contains a variety of dependent variables [7]. Findings from other studies have linked the wrong measure to the concept of interest and there are numerous occasions where incorrect or different tools have been used or where their data is misinterpreted as QoL [8, 9]. Moreover, this will emphasise the importance of selecting an ideal reliable and valid measure that is useful to use throughout different cultures. Also, it should include a broad range of potentially independent domains covering all critical aspects of QoL [10].

Furthermore, the assessment of mental health in patients with diabetes requires multiple transitions geographically and socially. In addition, there is a need to identify patients lacking medical follow-up and are therefore, at increasing risk of poor mental health status including psychosocial problems such as depression, diabetes-emotional distress, anxiety, eating disorders, and cognitive impairment [11]. Hence, it is essential for clinicians to use a standardised tool that is of dynamic construct that incorporates comprehensiveness, sensitivity, and balance relative to subjectivity and brevity to help identify gaps and monitor psychological well-being and care among adult patients with T2DM. However, to date, measuring QoL and mental health outcomes in these patients remains a challenge and there are limited studies evaluating the quality of these tools.

Therefore, the aim of this systematic review is to identify, summarise, and evaluate the methodological quality for the most commonly used and validated health-related QoL and mental health assessment measurements in patients with T2DM.

Methodology

The Systematic review was conducted on QoL, and mental health surveys published in PubMed, MedLine, OVID, The Cochrane Register, Web of Science Conference Proceedings and Scopus databases between the 1st of January 2011 and the 31st of July 2022. In addition, reference lists of the included studies and previous reviews on the topic were hand searched for potentially relevant studies. Search terms for each database included ‘type 2 diabetes mellitus’, ‘quality of life’, ‘mental health’, and ‘questionnaires’. No language restrictions were applied. We performed a systematic search in accordance with the Preferred reporting items for systematic review and meta-analyses protocols (PRISMA) statement 2020 [12]. Our formulated research question was based on Participants, Concept, and Context (PCC) on ‘What is the most recent validated and commonly used measurement or questionnaire to assess the quality of life and mental health among adult diabetic patients in different languages?’.

Inclusion and exclusion criteria

All studies conducted during the last decade or more (1st of January 2011 to 31st of July 2022) were considered to be eligible if they met the following inclusion criteria: 1) Population-based studies; 2) Among adults sharing common characteristics and health conditions including T2DM; 3) Studies focusing on health-related QoL and mental health assessment questionnaires or surveys; 4) Any studies conducted on 50 patients or more; 5) Surveys mentioned in conference abstracts were only considered if sufficient information were available for data extraction (Fig. 1). All publications were reviewed in full text to determine whether they met the inclusion criteria or not by two authors independently (Fig. 1).

Fig. 1
figure 1

Flow diagram of identification, screening, eligibility and included studies via four databases

Full size image

Synthesis and data extraction

According to the eligibility criteria, the main author (O.A.) carefully scanned the titles and abstracts to address any duplicated or irrelevant studies from the initial databases, PubMed, and Scopus.

This was followed by reviewing all chosen articles in their full manuscript and filling in a pre-structured table that summarises and assesses the quality of the selected studies and any general information (Table 1). The table was designed into two sections one to cover the study characteristics and the other for study quality including the following items/ categories: 1) The primary author’s name; 2) Year of publication; 3) Study location; 3) Study design; 4) Target population (included the number of participants, age, and gender); 5) Main objectives and questionnaires; 6) Mode of questionnaire administration; 7) Validity; 8) Reproducibility; 9) Responsiveness of the participants; 10) Type of bias; 11) Languages support (Table 1).

Table 1 Overall studies characteristics
Full size table

A 10% random sample was checked by a second reviewer (K.H.) to check for the search and reviewing of the articles, references, and any additional relevant publications that may have been missed by the initial electronic databases was finally carried out independently by two senior examiners. Any inconsistencies were discussed by a third reviewer (J.F.) for a final decision.

Quality appraisal

The methodological quality of each included study in terms of validity, reliability, and consistency was assessed using the Joanna Briggs Institute (JBI) critical appraisal checklists (https://jbi.global/critical-appraisal-tools) for cohort, randomized controlled trials (RCTs), and cross-sectional studies which was the most appropriate and applicable tool for this review [13]. The JBI checklist for cohort studies consists of 11 items, while 13 items for RCTs, and 8 items for cross-sectional studies. Each item was answered with either a Yes, No, Unclear, or Not Applicable response.

The categories of the studies were divided into: High quality (if 80% or more of the items were answered with a yes), Moderate (if more than 60% of the items were answered with a yes), and Low (if less than 60% of the items were answered with yes). Any study categorized as high or moderate quality was eligible to be included in this review. Any disagreement between the reviewers was solved by a discussion with the third reviewer (J.F.).

Results

Search and eligible studies

A total of 489 articles were identified in six electronic medical databases, 343 of which were selected (58.6% from Scopus) during the first screening (Fig. 1). Following the first screening, 109 articles were identified and subjected to the next level of screening after reading the titles and abstracts (Fig. 1). Of these, 68 articles were considered potentially eligible after reviewing the full text (Fig. 1). Subsequently, 28 articles were excluded based on the defined inclusion and exclusion criteria and there were 21 articles [14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34] among them considered as low quality and excluded based on the JBI quality appraisal checklists used in this review (Fig. 1) (Table 2). Finally, 40 articles were shown to meet our eligibility criteria and were, therefore, included in this systematic review (Fig. 1) (Table 1).

Table 2 Summary of quality appraisal for excluded studies
Full size table

Study characteristics and QoL measurements

The majority of the studies were cross-sectional 60% [35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58], followed by 22.5% clinical trial [59,60,61,62,63,64,65,66,67], and 17.5% cohort [68,69,70,71,72,73,74]; with overall response rates ranging between 40 and 98% among adult patients with T2DM.

The following questionnaires used in the QoL assessment included the Medical Outcomes Study Short Form 36 (SF-36), the Medical Outcomes Short Form 12 (SF-12), the 9-item Patient Health Questionnaire (PHQ-9), the EuroQoL EQ-5D, The World Health Organization Quality of Life-Brief (WHOQOL-BREF), the 17-items Diabetes Distress Scale (DDS-17), the Audit of Diabetes Dependent Quality of Life (ADDQoL19), the Diabetes‑Specific Quality of Life (DMQoL), and the Impact of Weight on Quality of Life-Lite (IWQoLLite). Other questionnaires used evaluated the mental health combined with QoL assessment. This included the Beck Depression Inventory, the World Health Organisation—Five Well-Being Index (WHO-5), the Chronic Illness Care (PACIC), the Center for Epidemiologic Studies Depression Scale questionnaires (CES-D), the Generalised Anxiety Disorder (GAD-7), the Problem Areas in Diabetes (PAID) scale, the Confidence in Diabetes Self-Care (CIDS) scale, the 12-item Diabetes Support Scale (DSS), the Hypoglycaemia Fear Survey-II (HFS-II), the Health Care Climate-Short Form (HCC-SF), the Global Satisfaction with Diabetes Treatment (GSDT), the Summary of Diabetes Self-Care Activities measure (SDSCA-6), the Barriers to Medications (BM), the Perceived Social Support (PSS), and The Empowerment Scale-Short Form (DES-SF).

Main findings

The six top commonly used QoL measurements included the SF-12 which was found in 19 studies [35, 36, 38, 39, 41, 43, 45, 48, 49, 51, 53, 54, 60, 66, 68, 70,71,72,73], the SF-36, identified in 16 studies [36, 42, 44, 47, 50, 55, 57,58,59, 61,62,63,64,65, 67, 69], the EuroQoL EQ-5D, included in 8 studies [37, 41, 44, 55, 60, 71,72,73], the PHQ-9, found in five studies [35, 40, 60, 66, 74], the WHOQOL-BREF, evaluated in two studies [40, 52], and the ADDQoL19, identified in two studies [42, 46].

Fifteen (37.5%) studies used only one questionnaire. In this regard, the SF-12, was used as a single questionnaire in seven studies [39, 43, 45, 48, 51, 53, 70], the SF-36 in six studies [50, 61,62,63,64,65], the EuroQoL EQ-5D in one study [37] and the ADDQoL19 in one study [46]. However, the remaining reviewed studies (62.5%) used more than one questionnaire.

In terms of mental health measurements, there were four questionnaires that were commonly used which combined with QoL questionnaires namely the WHO-5 in three of the reviewed studies [38, 66, 74], the BDI in three studies [47, 57, 67], the PAID in three studies [56, 66, 74], and lastly the PACIC, found in two studies [38, 49].

Most of the studies (90%) reported using self-administered questionnaires with only four [51, 53, 68, 70] identified to use interviewer mode of administration. Moreover, all of the studies indicated that the questionnaires used were validated, reliable and that they supported different languages.

Discussion

The present systematic review indicates that the SF-12 questionnaire is the most appropriate and commonly used measurement to assess QoL and mental health followed by the SF-36, the EuroQoL EQ-5D, the PHQ-9, the WHOQOL-BREF, and the ADDQoL19. This questionnaire was used in several studies with different methodological approaches and was confirmed to be validated, reliable, less time-consuming, easy to use and available in many languages [75]. Other attributes of the SF-12 questionnaire include that it is a self-administered generic measurement and large-scale, population-based health inventory that has been developed to measure both the physical and mental health aspects of a patient [75]. It is effective and efficient with a completion time of fewer than five minutes [75]. Moreover, it has the exact eight health domains (Physical Functioning, Role Physical, Role Emotional, Mental Health, Bodily Pain, General Health, Vitality, and Social Functioning) similar to SF-36 but with one or two items per domain and without any notable statistical difference especially for studies with a large sample size [75]. These were the significant advantages of using SF-12 over SF-36 while the disadvantages were considered as less in represents or comprehensiveness of the content of health measures and lacking of the statistical precision of mental and physical components scores compared to SF-36 [75].

One of the largest randomized controlled trials (RCTs) titled Look AHEAD (Action for Health in Diabetes) conducted on 5,145 overweight or obese with T2DM assessed the effect of long-term lifestyle modification on QoL and depression symptoms using the BDI and SF-36 questionnaires as the main measurement for their primary outcomes. Concerns included a shallow response rate by fewer than 40% of patients in the final year of the study possibly due to the high dropout rate and lengthy QoL questionnaire [67]. Another RCT was conducted among 1,922 patients with T2DM to evaluate the effect of two different insulin therapy on QoL using the SF-36 alone. The authors of this study observed that there was a lack of a sleep variable on the questionnaire which was considered as a study limitation. There was no information relating to the response rate in this study [61]. The remaining trials that were included in the present review used the SF-36 with a response rate between 70%-98%; with the exception of one controlled clinical trial that used the SF-12 combined with different questionnaires and most of which had weaknesses with respect to randomization, blinding, and allocation concealment [59, 60, 62,63,64,65,66].

Another population-based cohort study on adults with T2DM conducted on 1,064 participants to assess the impact of diabetic retinopathy on QoL used the SF-12 where interviewers had the questionnaire administered in either English or another language [68]. This was similar to a population-based German cohort study that used the SF-12 to examine the change of QoL in 1,046 diabetic patients through a face-to-face questionnaire administered at baseline where the response rate was between 67 to 84% [70]. However, most of the other cohort studies included in this review preferred to use the SF-12 as a main questionnaire for their studies [71,72,73].

A longitudinal cross-sectional study conducted to identify the determinants of poor QoL in 1,826 Chinese diabetic patients who used the SF-12 over 24 months (through a phone interview) had a response rate between 75.5% and 59.7% [51]. This study used a similar methodological approach with another longitudinal cross-sectional study regarding the association between depression and QoL among 1,033 adults with T2DM addressed by interviews throughout the study using the SF-12 questionnaire alone [53]. It has been plausible that the majority of the cross-sectional studies matched with cohort studies in terms of using the SF-12 as their primary questionnaire and through interview mood of administration [35, 36, 38, 39, 41, 43, 45, 48, 49, 54].

Strengthens and limitations

The main strength of this review is that we comprehensively reviewed the body of evidence that focused on the most common and widely used publications over the last decade. This study identified the most common, widely used efficient and validated QoL and mental health questionnaire over a large number of publications for more than a decade in different languages. There are some weaknesses due to potential biases identified from the included studies especially the self-reported and non-response bias as well as the differences in response rates. Another weakness is the lack of standard terminology which may possibly cause misleading results. Lastly, the huge heterogeneity in the study designs, methodology, and sample size has limited our ability to quantify any differences through a meta-analysis.

Conclusion

In the backdrop of the growing prevalence of this disease worldwide there has been limited information on the most efficient and commonly used questionnaire for the diabetic patient. Our review found evidence of the effects of six different QoL and mental health questionnaires. Findings identified the SF-12 as the most validated, time efficient and effective questionnaire that allows cross-culture adaption which can be used in population-based studies across the world. These results encourage the use of SF-12 in adult patients with T2DM as a useful screening measure for identifying and monitoring mental health issues that may assist with target treatment and prevention. The wide range of tools used to assess QoL, methodology of administration, clinical research question and limited sample size used by studies hinder direct comparisons in patients with T2DM. Future large multicentre prospective research is recommended to help clarify causality on associations between mental health, QoL and any barriers in people with T2DM involving individuals from different cultural backgrounds.