Chronic constipation is a prevalent worldwide problem that affects up to 10-15% of the adult population.1 Symptoms of primary and secondary constipation include hard stools, excessive straining, infrequent bowel movements, bloating, and abdominal pain.2 While constipation can often be managed by medication and lifestyle modification, prolonged constipation can significantly decreases quality of life (QoL).2,3 To better assess patients’ health status and QoL, it is important to have an accessible tool that can accurately assess patients’ symptoms and treatment outcomes, which may enable personalized intervention strategies. The usage of reliable and validated patient-related outcome measure (PROM) can help provide a consistent method of measuring clinical symptoms and QoL outcomes in patients.4 PROMs are standardized, validated questionnaires that measure patients’ perception of their own health status and well-being.5 While PROMs were initially developed for research use, they have been increasingly adopted in clinical practice to aid clinicians provide better and more patient-centered care.6
To date, 2 reviews have examined existing assessment scales measuring constipation symptoms.7,8 A combination of 9 self-reported measures, developed between 1989 to 2010, were assessed by both reviews. While the reviews provided an insight on the reliability and validity of existing constipation PROMs, the reviews were not conducted systematically and constipation-related QoL PROMs were not included. Given the impact of constipation on QoL, including mental, social, and physical functioning,9,10 it is important to consider QoL in treatment outcomes.
As the capabilities and adoption of digital technology expand in healthcare, it is also important for us to explore the potential of digitizing PROMs. This could sustain longitudinal patient assessment, which can further support the individualization of patient care. In patients with inflammatory bowel disease, collecting consistent electronic patient reported outcomes (ePRO) on a cloud-based digital therapeutics and monitoring application has been shown to significantly reduce yearly hospitalizations and emergency room visit rates most likely due to immediate interventions prompted by concerning questionnaire scores. Patients also reported having a better understanding of the nature and causes of their health condition after a year.11 Given the importance of incorporating QoL into treatment outcomes and the potential of incorporating digital health technologies that are patient-centric into constipation management, the current review aims to (1) systematically review constipation-related PROMs, including QoL reporting, using the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN)12 guideline to evaluate the methodological quality of included studies and the quality of the measurement properties themselves, and (2) assess the current modes of digitization of constipation-related PROMs.
A systematic review protocol was developed in accordance with the Preferred Reporting for Items for Systematic Reviews and Meta-analyses (PRISMA) and the COSMIN guidelines. The study protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) database (No. CRD42021236257).
The PRISMA guidelines were used to identify studies for this review. A comprehensive literature search was performed using PubMed, Embase, and PsycINFO to identify all articles on the development or validation of constipation-related PROMs. The search was conducted up to February 2021. Searches in all 3 databases were performed using the following keywords: (
The initial search yielded 11 011 articles after duplicates were removed. Four articles were identified via hand-checking of reference lists of published reviews and were included retrospectively. Two authors (V.V.L. and N.Y.L.) independently reviewed titles and abstracts of the identified records for preliminary inclusion. Articles were included for further screening based on inclusion criteria listed in Table 1. Five hundred and seventy-nine articles satisfied the preliminary inclusion criteria and were accepted for a full review. Interrater agreement was assessed with Cohen’s κ indicator, where κ of 0.60-0.79 was classified as “moderate,” 0.80-0.90 as “strong,” and above 0.90 as “almost perfect” interrater agreement.13 There was a moderate interrater agreement for study selection (Cohen’s κ = 0.72; 95% CI, 0.68-0.75) and discrepancies were resolved by discussion.
Full texts of the eligible articles were retrieved and reviewed. Articles were excluded based the exclusion criteria listed in Table 1. Five hundred and fifty-three articles did not meet the eligibility criteria and were excluded. There was strong interrater agreement for the second screening (Cohen’s κ = 0.80; 95% CI, 0.67-0.93) and discrepancies were resolved by discussion. An independent third reviewer (A.T.) was brought in when discrepancies were not resolved. Figure depicts the flow diagram of the study selection.
The methodological quality of the studies and the quality of the PROM itself was assessed using the COSMIN guidelines. Firstly, the COSMIN Risk of Bias checklist14 consisting of 117 questions was used to assess the methodological quality of the studies. The following measurement properties were assessed: PROM development, content validity, structural validity, internal consistency, reliability, measurement error, criterion validity, hypothesis testing for constructive validity, and responsiveness. A 4-point rating system of “very good,” “adequate,” “doubtful,” and “inadequate” was used to rate each property. The final rating was determined by taking the lowest score of an assessment area (ie, “worst score counts” principle). No rating was given if measurement property was not assessed or described.
Following that, the quality of the PROM itself was assessed using the COSMIN updated criteria for good measurement properties.12 The following psychometric properties were assessed: internal consistency, reliability, measurement error, content validity, structural validity, hypotheses testing, criterion validity, and responsiveness. Using the criteria provided, a rating of “+” for sufficient, “–” for insufficient, or “?” for indeterminate was given to each measurement property.
Two authors (V.V.L. and D.J.Y.X.) independently reviewed the included studies using the COSMIN risk of bias checklist and updated criteria for good measurement properties. There was moderate interrater agreement for risk of bias (Cohen’s κ = 0.71; 95% CI, 0.69-0.74) and almost perfect interrater agreement for criteria for good measurement properties (Cohen’s κ = 0.93; 95% CI, 0.88-0.98). Any discrepancies were resolved by discussion.
A total of 23 PROMs measuring constipation symptoms15-35 or constipation-related QoL36-40 were identified. The PROMs were reported in 26 different studies with publication years ranging from 1989 to 2020. The Bowel Function Index and Patient Assessment of Constipation–Symptom (PAC-SYM) had more than 1 validation study with additional information on measurement properties. A summary of included studies is presented in Table 2.
Eighteen out of the 23 identified PROMs evaluated constipation symptoms. The majority of the PROMs (n = 10) were developed with intentions to assess severity of constipation in patients.16,19,21,23,24,26,27,30-32 Seven PROMs were developed as a potential diagnosis tool to detect clinically significant constipation, with some having dual functionality for diagnosis and measurement of severity.19,21,27,29,30,34,35 A subset of PROMs (n = 4) were created and/or validated for research purposes, specifically to assess treatment benefits in patients during varying stages of clinical trials.15,20,25,28
The items included in the PROMs can be categorised into 5 categories: abdominal symptoms, bowel movement-related symptoms, stool-related symptoms, anal or rectal symptoms, and others. The most common questionnaire items were incomplete evacuation during bowel movement15,16,19,22-24,26,27,29-32,34,35 and stool consistency,15,19-22,25-29,31,32,34,35 with both items included in 77.8% of PROMs assessed. To measure incomplete bowel movement, PROMs have included a combination of frequency and/or severity related questions. To measure stool consistency, questions include rating of consistency based on the 7-point Bristol stool form scale15,20,28 or a self-constructed scale,27,29,31,35 frequency or severity of hard or lumpy stools,19,32,34 and presence of hard and loose/water stools.21,22,26
More than half of the studies included a measure of abdominal pain,15,20,24-28,30,32,34 abdominal bloating,15,19-21,25-29,32 frequency of bowel movements,19-29,34,35 and straining during bowel movement.15,20,22,23,25-28,30-32,34,35 Abdominal pain, abdominal bloating, and straining during bowel movement have been measured through both severity and/or frequency while questions on frequency of bowel movements have generally been measured using a self-constructed time scale. Two PROMs differentiated complete spontaneous bowel movement from spontaneous bowel movement.20,28 A summary of questionnaire items in PROMs measuring constipation symptoms is presented in Table 3.
Five out of the 23 PROMs assessed constipation-related QoL. Three PROMs were developed to measure the impact of constipation on multiple aspects of QoL, including social relationships, treatment satisfaction, physical symptoms, diet, daily activities, and psychological state.38-40 While all 3 PROMs are suitable for patients with chronic constipation, the Elderly-constipation Impact Scale (E-CIS) was developed for elderly Malay speaking individuals aged 60 years and above.39
Two PROMs evaluated specific aspects of QoL in patients with constipation. The Chronic Constipation Treatment Satisfaction Questionnaire (CTSAT-Q) specifically focused on treatment satisfaction in patients with chronic constipation and constipation-predominant irritable bowel syndrome (IBS-C).36 Items include patient’s expectations on and attitude towards medication, value of medication, interference due to treatment, and effectiveness of treatment. On the other hand, the Constipation-related Disability Scale37 focused on the impact of constipation symptoms on day-to-day activities. The PROM includes a rating of difficulty in performing various leisure, work, and daily activities.
The COSMIN risk of bias assessment demonstrated very few studies with consistent “very good” and/or “adequate” ratings across all domains. A summary of risk of bias scores for each study are presented in Table 4. Cross-cultural validity was not assessed as the current review only included studies that assessed the original version of the PROM.
Twenty-three studies were rated on PROM development and the majority of the studies (n = 14) scored “inadequate” due to the lack of a PROM development study involving the target population or a cognitive interview study to assess the comprehensibility or comprehensiveness of the PROM. The remaining 9 studies scored “doubtful” due to poor reporting of study methods including the use of skilled group moderators or interviewers, interview guides, recording and transcription process of interviews and independent coding of data. Poor reporting of methods similarly resulted in “doubtful” ratings for content validity. Only 2 studies19,38 comprehensively examined content validity (ie, asking patients and professionals about relevance, comprehensibility, and comprehensiveness).
Construct validity was the most common measurement properties analysed (n = 23), nevertheless, not all studies examined both convergent and discriminative validity. Four studies15,21,24,31 only examined discriminative validity while 2 studies26,35 only examined convergent validity. Half of the studies that examined construct validity scored “doubtful” due to the lack of detailed description of comparator instruments and/or important characteristics of subgroups.
Following construct validity, reliability (n = 17), and internal consistency (n = 16) were the second and third most analyzed measurement properties. The majority of studies that scored “doubtful” and “inadequate” for reliability did not fulfill appropriate design requirements (eg, patients’ stability in the interim period, similarity of test conditions, and appropriate time interval). Most of the studies that analyzed internal consistency fulfilled the COSMIN criteria for “very good.” “Doubtful” ratings for internal consistency were given due to lack of clarity if scale or subscale was unidimensional.
Less than half of the studies analyzed structural validity, measurement error, criterion validity, and responsiveness. Studies that examined structural validity mostly scored “very good” and “adequate.” Only 1 study scored “doubtful” for structural validity due to the lack of description of rotation method. Two studies scored “doubtful” and “inadequate” for measurement error due to unclear description on stability of patients in the interim period and inadequate calculations of standard error of measurement. All studies that examined responsiveness only examined comparison between subgroups accordingly, ratings were based on that aspect. Scores of “doubtful” and “inadequate” for responsiveness were due to poor description of important characteristics of subgroups and inadequate statistical methods applied.
Due to the limited amount of validation studies per PROM, the studies were assessed individually and the total ratings were not provided. The individual ratings for each measurement property of all the studies are presented in Table 5.
Twenty-one studies had at least 1 insufficient (–) or indeterminate (?) rating, and no PROM was fully assessed in all measurement properties, with measurement error and criterion validity most commonly missing. The PROMs with the most sufficient (+) ratings include the Chinese Constipation Questionnaire, Constipation Severity Instrument (CSI), PAC-SYM, Constipation-related Disability Questionnaire, and Patient Assessment of Constipation–Quality of Life (PAC-QOL). To improve ratings of measurements properties, focus should be given to obtaining sufficient rating for measurement error (smallest detectable change/limits of agreement < minimal important change), criterion validity (correlation with gold standard or area under the curve ≥ 0.70), and responsiveness (results in accordance with hypothesis or area under the curve ≥ 0.70).
Five studies reported using digital formats to administer PROMs during the validation process.15,20,25,28,36 Both the Bowel Function Diary and Diary for Irritable Bowel Syndrome Symptoms-Constipation (DIBSS-C) were completed as part of an electronic diary and were administered using a handheld device. For the Bowel Function Diary, participants were given a handheld electronic personal digital assistant (PDA) device while the type of device was not specified for the DIBSS-C. The CTSAT-Q was described to be disseminated online however, no further information was provided.
The Chronic Constipation Symptom Severity Measures and IBS-C Symptom Severity Measures were administered using interactive voice response system technology, a computer-automated telephone system that collects data through spoken answers or keypad responses.41 For all 5 studies, methods of digitizing and validating the digital formats of the PROMs were not reported.
Digitizing constipation-related PROMs represents a promising step towards individualizing patient intervention in a longitudinal and scalable manner. Therefore, the current systematic review provides an overview of constipation-related PROMs that have been developed and validated over the past 32 years. The review identified 23 different constipation-related PROMs, with 18 measuring symptom-related measures and 5 measuring constipation-related QoL measures.
The review revealed a large amount of variation between PROMs used to measure symptom-related constipation outcomes. Variations include outcome measures targeting different subtypes of constipation (eg, opioid-induced constipation, obstructive defecation syndrome, and IBS-C), functions of PROM (eg, clinical use and research purposes), methods of administration (eg, pen and paper, clinician administered, and electronic diary), length of PROM (range = 3-98 items), and recall period (eg, last 2 weeks and past 24 hours). Given the multiple possible etiologies of constipation, subtype-specific PROMs can be a useful tool to further facilitate customization of outcome measures monitoring, and to provide more accurate feedback to clinicians about treatment progress. Nevertheless, the heterogeneity of variables between PROMs can indicate the lack of standardization during the PROM development process. For instance, multiple studies scored “inadequate” for PROM development based on the COSMIN checklist due to the lack of patients’ involvement and input when developing PROM items. While physical symptoms and functioning are vital aspects of disease monitoring, patients may be more focused on regaining or preserving QoL, including emotional wellbeing and social functioning.42
Considering the known impact of constipation on QoL,9 the review also examined constipation-specific QoL-related PROMs. Similar to studies that developed symptom-related outcomes measures, none of the QoL-related PROMs reviewed met all the COSMIN quality standards for development and measurement properties. While all 5 of QoL PROMs involved patients in the development process through individual interviews or focus groups, issues include incomplete reporting of interview methods and lack of a follow-up cognitive testing session. Based on the ratings of measurement properties, out of the reviewed PROMs, the current review recommends the Chinese Constipation Questionnaire, CSI, PAC-SYM, Constipation-related Disability Questionnaire, and PAC-QOL to measure constipation symptoms or constipation-related quality of life. Nevertheless, none of the reviewed PROMs report all measurement properties indicated in the COSMIN checklist thus, there is a need for better standardization of PROM creation, from the development stages to the final reporting of validation studies. In their efforts to better regulate PROM creation and usage, the Food and Drug Administration (FDA) released a guidance for the industry on recommendations for PROM development and validation,43 and more recently, a draft on selecting, developing or adapting PROMs for medical device evaluation.44 To ensure more consistent methodological quality and reporting, it would be ideal for future PROM development studies to familiarize themselves with the FDA recommendations and COSMIN guidelines.
Despite the variety of PROMs identified in this review, only 2 PROMs were developed and validated within the Asian context. Chan et al19 developed and validated the Chinese Constipation Questionnaire in the Chinese language with an ethnic Chinese population. Similarly, Abdul Wahab et al39 developed and validated the E-CIS using the Malay language spoken in the local dialects of Terengganu and Kelantan in Malaysia. While there are translated versions of questionnaires including the PAC-SYM32 and PAC-QOL,40 culture and language are intertwined, and language should be examined in conjunction with culturally specific health beliefs and understanding.45 Regardless of English fluency, patients of different ethnic groups may differ in terms of pronunciation, speech delivery, grammar/vocabulary and culturally specific presentation styles when describing their issues to medical practitioners.45 Hence, a relatable and culturally-specific PROM can be beneficial in increasing the efficacy of patient-clinician communication, and further facilitate a more personalised, patient-centric symptoms monitoring and treatment.
The current review also assessed the modes of digital dissemination of currently available PROMs and identified 5 PROMs that used digital formats to administer the questionnaire during the validation process. Methods of dissemination were varied, ranging from electronic diary formats on PDA devices to computer-automated telephone systems. Given the widespread adoption of smart devices, such as smartphones and tablets, the use of ePROs presents as a viable option for remote monitoring led by patients themselves. Consistent symptom reporting through digital means can improve patient-clinician communication, detection of unrecognised problems, and patients’ health behaviors, including patient self-management and patient empowerment.46 Nevertheless, to ensure reliable reporting of ePROs, evidence is needed to support measurement equivalence between the electronic and paper-based PROMs.47 The 5 studies in the current review that utilized digital PROMs did not report methods undertaken to digitize the PROMs. Accordingly, there is a need for better standardization for digitization of PROMs to maximize the potential of ePRO tools. Recommendations from the International Society of Pharmacoeconomics and Outcomes Research (ISPOR) ePRO Task Force include cognitive debriefing, usability testing or full psychometric testing of the electronic versions, depending on the level of modification. Future ePRO development studies can benefit by reporting level of modification and relevant methods undertaken to ensure measurement equivalence.
With growing interests to integrate technology into healthcare, PROM development and implementation should keep pace with the fast and evolving field of digital health. Digital health has the potential to offer new modalities of probing patient state in real time through minimally invasive methods (eg, experience sampling method, day reconstruction method). Furthermore, increasing accessibility to ePROs can open doors to personalization of PROMs. A recent real-world longitudinal study of patients with rheumatic and musculoskeletal diseases allowed patients to customize their tracking on the Arthritis Power mobile application between 3 and 10 PRO symptom measures over a period of 3 months.48 While some PROM items were prioritized more, there were variations between patients in the items chosen for tracking and in ranking of importance. It should also be noted that minimal changes to items tracked were observed, suggesting that patients continue to only track symptoms that are important to them. The ePROs can further benefit from concepts commonly employed in digital health, such as gamification and behavioral nudges, to sustain users’ engagement.49 As the concept of personalized medicine continues to grow, digital technologies can aid in the continual evolution and optimization of PROMs.
A limitation of the current systematic review is the subjective nature of the COSMIN evaluation methods. Multiple items of the COSMIN checklist require subjective judgement of the reviewer based on experience and knowledge, hence, there is possibility of subjectivity within the review.50 Furthermore, we acknowledge that there may be other methods to assess psychometric measures beyond the COSMIN guidelines. Nevertheless, the current review endeavored to reduce subjectivity by utilizing 2 independent reviewers with good interrater reliability and a third for any discrepancies. Secondly, the current review did not include PROMs assessing the pediatrics population. Constipation-related pediatric PROMs rely on patient-reported measures, parent/caregiver-reported measures or a combination of both, and assessing the differences between the method of reporting is beyond the scope of the current review. Poor concordance between parent- and child-reporting have been observed when assessing gastrointestinal symptoms. For instance, children tended to rate their pain/discomfort intensity more severely than parents did, and up to 60% of parents of 10- to 19-year-olds could not answer items relating to defecation habits.51 Therefore, it would be beneficial for future studies to focus on the differences in reporting methods and the involvement of both parent and child in the PROM development process.
In conclusion, this review assessed constipation symptoms and constipation-related QoL PROMs using the COSMIN guidelines and identified a lack of consistent methodology and reporting of development and validation studies. Furthermore, more culturally-specific PROMs, especially in the Asian context, will be beneficial. There are varying modes of digital dissemination of constipation-related PROMs however, greater standardization of the process is required to ensure transparency and consistency. As PROM is a useful tool that can provide clinicians and researchers insights into patients’ health status and health-related QoL, further developments of constipation-related PROMs can be made by more consistent methodology and reporting of PROM development, increase in culturally-specific PROMs, and better reporting of protocol for digitization of PROMs.
This study was supported funding from the Institute for Digital Medicine (WisDM) Translational Research Programme (Grant No. R-719-000-037-733) at the Yong Loo Lin School of Medicine, National University of Singapore (Dean Ho).
Agata Blasiak and Dean Ho are co-inventors or previously filed pending patents on artificial intelligence-based therapy development. Dean Ho is a shareholder of KYAN Therapeutics, which has licensed intellectual property pertaining to artificial intelligence -based drug development. Others have nothing to disclose.
V Vien Lee: development of the parameters of the systematic review, review of articles for eligibility and quality, data analysis, and preparation of the manuscript; Ni Yin Lau: review of articles for eligibility and editing of the manuscript; David J Y Xi: review of articles for quality and editing of the manuscript; Anh T L Truong: review of articles for eligibility and editing of the manuscript; Agata Blasiak: development of the parameters of the systematic review and editing of the manuscript; Kewin T H Siah: evaluation of quality and editing of the manuscript; and Dean Ho: evaluation of quality and editing of the manuscript.