The Personalized Priority and Progress Questionnaire (PPPQ): A personalized instrument for quality of life and self-management for use in clinical trials and practice

Purpose The aim of this study was to develop and validate a brief personalized instrument that (1) defines patients’ priorities for improvement, (2) measures progress in prioritized quality of life (QoL) and self-management outcomes, and (3) is applicable in both clinical practice and clinical trials. Methods The instrument was developed based on the literature on personalized assessment and patient priorities, feedback by clinicians, and six cognitive interviews with patients with chronic kidney disease. The resulting questionnaire, the Personalized Priority and Progress Questionnaire (PPPQ), contains a baseline and follow-op measurement. The baseline measurement assesses functioning on QoL (8 items) and self-management (5 items). The final item evaluates patients’ priorities for improvement. The follow-up measurement assesses progress in QoL and self-management. A personalized progress score can be calculated indicating the amount of progress on the QoL or self-management domain that is prioritized by the individual patient. Psychometric properties of the PPPQ were evaluated among patients with chronic kidney disease (n = 121) and patients with kidney failure treated with dialysis (n = 22). Results The PPPQ showed to be a feasible instrument that is easy and quick to complete. Regarding the construct validity, small to large correlations were found between the items and existing validated questionnaires measuring related constructs. Conclusion The PPPQ proved to be a feasible and valid instrument. The PPPQ can be adapted to match diverse populations and could be a useful tool both in clinical practice (e.g., to identify priorities and tailor treatment) and clinical trials (e.g., to evaluate the effectiveness of personalized interventions). Supplementary Information The online version contains supplementary material available at 10.1007/s11136-023-03429-7.


Introduction
Patients vary in functioning, preferences, goals, and values [1,2]. Besides patients' biological and clinical functioning, these individual differences and priorities should be incorporated in interventions [3][4][5]. Patients with chronic kidney disease (CKD) have expressed a need for holistic care that includes all aspects of a person's health and wellbeing, including quality of life (QoL) and selfmanagement behaviors (e.g., physical activity, dietary changes, medication use, non-smoking) [2,6]. Moving away from a 'mechanistic' focus on laboratory results and focusing on patients' actual wellbeing instead, is key for patient-centered care (PCC) [6]. PCC is defined as providing care that is respectful of and responsive to patient preferences, needs, and values and ensuring that patient values guide all clinical decisions [7]. Positive associations of PCC with enhanced QoL, wellbeing, patient satisfaction, perceived quality of care, and self-management [8][9][10], as well as improved clinical outcomes have been found, for example reductions in pain, blood pressure, complications, and hospitalization [9]. Thus, instead of evaluating one-size-fits-all interventions, the focus should be on identifying and offering the best intervention for every individual patient [11]. This calls for personalized (1) interventions and (2) outcome variables to do justice to each patient's unique treatment trajectory [3,4,11].
As personalized interventions imply individual differences in treatment, standard generic outcome measures to evaluate their effectiveness will not suffice. Multiple questionnaires would be necessary to evaluate different treatment goals, which significantly harms the power of these studies since only the data of subgroups that worked on similar treatment goals can be used [4]. Moreover, generic measures invalidate the personalized character of the intervention by clouding patients' results with unimportant or not focused-on health domains [3,4,11,12]. Using personalized assessments enables to evaluate whether inventions are not only clinically, but also personally relevant to patients (i.e., personal utility) [13]. This allows general conclusions on treatment effectiveness, while considering each unique treatment trajectory. This feature makes personalized outcome measures highly valuable in research settings.
Personalized assessment can also be of great clinical value, as it helps to clarify patients' needs and priorities. It provides a valuable asset in shared decision-making [12], in which patients have an active role in selecting treatment and care plans that match their preferences. Personalized assessments could also help to define personally-relevant treatment goals, which form the basis of personalized treatment [12]. Subsequently, personalized outcomes can monitor patient functioning over time [12]. Although incorporating patient priorities in decision-making and interventions is highly valued [7], they are not routinely assessed or recorded in medical records [14]. A personalized tool assessing priorities would make patient priorities explicitly visible in clinical practice.
However, adequate tools for use in clinical practice or trials are sparse. Several tools measure priorities or preferences, but all lack an effect assessment on health outcomes that are prioritized by patients themselves [14]. The McMaster Toronto Arthritis Patient Preference Disability Questionnaire (MACTAR) [15] assesses change in functioning areas that matter to patients, but limits feasibility by requiring trained interviewers and complex scoring [16]. Similarly, scales that focus on goal setting such as Goal Attainment Scaling (GAS) [17], the Patient Goal Priority Questionnaire (PGPQ) [18], and Self-Identified Goal Assessment (SIGA) [19] can help patients prioritize their needs for improvement, but are also time-consuming and require trained interviewers or therapists to help patients setting realistic goals [20].
This study aims to develop and validate a brief personalized instrument that is applicable to patients with diverse somatic conditions and that (1) defines patients' priorities for improvement, (2) measures changes in functioning on patient-prioritized QoL and self-management outcomes, and (3) is applicable in both clinical practice and trials. This study evaluates the psychometric properties of the instrument in two different chronic kidney disease (CKD) samples. If feasible and valid, this brief, personalized tool can be used to easily identify, prioritize, and monitor individual problems and progress.

Questionnaire development
Using purposeful sampling, 4 patients with CKD and 2 patients with kidney failure treated with dialysis were recruited from Leiden University Medical Center to participate in the cognitive interviews.

Questionnaire evaluation
Datasets of two multicenter randomized controlled trials (RCT) were used, evaluating the effectiveness of a personalized e-health intervention in patients with CKD (E-GOAL study) [21] or kidney failure treated with dialysis (E-HELD study) [22].
Patient recruitment for the E-GOAL study took place from April 2018 through March 2020 within Dutch academic (Leiden University Medical Center, Leiden; Radboud university medical center, Nijmegen; University Medical Center Groningen, Groningen) and non-academic (Haaglanden Medical Center, The Hague) hospitals. Adult patients with CKD with an eGFR of 20-89 ml/min/1.73 m 2 under treatment by an internist-nephrologist were invited to participate when their screening questionnaire indicated at least mild depressive or anxiety symptoms and that they failed to meet at least one of the nephrology self-management guidelines [23].
Patient recruitment for the E-HELD study took place from February 2019 through October 2021 within Dutch academic (Radboud university medical center, Nijmegen; Leiden University Medical Center, Leiden) and nonacademic (VieCuri Medical Centre, Venlo; Bernhoven Hospital, Uden) hospitals and dialysis centers (Ravenstein Dialysis Centre, Ravenstein; Dialysis Center Groningen, Groningen). Adult patients with an eGFR < 15 ml/ min/1.73 m 2 treated with hemodialysis or peritoneal dialysis for at least three months were invited to participate when they presented low QoL or symptoms of fatigue, itch, depression, or anxiety on screening questionnaires.
Exclusion cr iter ia for bot h studies were: age < 18 years; > 10% past-year renal-function loss, life expectancy < 12 months or serious psychiatric conditions, recent major stressful life events unrelated to kidney disease, study-interfering cognitive problems, psychological treatment, kidney transplant received or scheduled < 1 year ago or ahead, not fluent in Dutch language, pregnancy, and no access to a computer or internet. Additionally, the E-GOAL study excluded patients with anticipated need for dialysis work-up within the study time frame or a systolic blood pressure < 95 mmHg not responding to withdrawal or antihypertensives.

Item generation
Questionnaire topics were based on research team expertise and literature on frequently reported symptoms and patient priorities in the CKD and dialysis population [2,[24][25][26][27]. The questionnaire structure was based on existing personalized and goal-setting measurements [15][16][17][18][19]. Resulting items were judged on comprehensibility and relevance by medical psychologists and nephrologists and revised accordingly.

Cognitive interviews
Six cognitive interviews were conducted with patients to evaluate item feasibility, comprehensibility and readability. Additionally, patients were asked to judge the relevance of the QoL and self-management topics addressed by the items and were invited to add or remove items. The interviewers, JT and CKC, used the think-aloud approach and verbal-probing techniques to gain insight in patient response processes [28]. In this approach, the interviewees vocalize their thoughts while answering items. Verbal-probing techniques included paraphrasing and comprehension/interpretation, recall, specific, and general probes [28]. Based on the interviews, minor textual revisions were made to the questionnaire.

Personalized priority and progress questionnaire (PPPQ)
The resulting questionnaire, called Personalized Priority and Progress Questionnaire (PPPQ), consists of baseline and progress measurements: Baseline measurement: assesses personal priorities for improvement in QoL areas and self-management behaviors.
• QoL: Whether patients experience limitations in QoL areas in the past 2 weeks is assessed by 8 items (fatigue, pain, itch, anxiety, depression, social environment, daily activities, and dependency), of which items can be omitted if not relevant in a particular population. Items are scored on a 5-point Likert scale (1 = not at all, 5 = extremely). An example item is "To what extent have you experienced limitations in the area of fatigue or sleep problems?". • Self-management: 5 items assess medication adherence, healthy diet, physical activity, weight maintenance, and 1 3 non-smoking, using 5-point Likert scales (1 = not at all, 5 = extremely well). An example item is "To what extent have you managed to always take your medication as prescribed?". • Priorities: Patients make two top-2's of the areas of QoL and self-management they prioritize for improvement and would actively commit to over the coming period.
Progress measurement: assesses progress in QoL or selfmanagement behavior compared to baseline measurement.
• QoL and self-management: Patients indicate whether their experienced QoL limitations and self-management behaviors changed since the baseline measurement. The items are answered using a 7-point Likert scale (− 3 = many more, 0 = remained the same, + 3 = much fewer), following the base "Compared to the last time I completed this questionnaire, …". Higher scores indicate improved QoL or self-management behavior. Example items are: "I now experience more/similar/fewer limitations in the area of fatigue or sleep problems" (QoL) and "I have managed less well/equally well/better to always take my medication as prescribed" (self-management). • Priorities: Patients indicate whether they tried to improve in any QoL or self-management areas since the baseline measurement. Patients select a maximum of 2 QoL areas and 2 self-management behaviors or the options 'other' or 'not applicable' (when not worked on anything).
Progress score. The progress score indicates the amount of progress (i.e., change) on the QoL or self-management domain(s) prioritized by the patient. This score consists of the isolated scores on the progress items that were selected as priorities at baseline; if more than one was selected, an average progress score is calculated. Ultimately, this will result in one single score that includes all personally meaningful changes.
The original Dutch PPPQ was translated in English using the forward-backward method [29]. The English (PPPQ-EN) and Dutch (PPPQ-NL) versions are enclosed in Supplementary Files 1 and 2.

Patient characteristics
Socio-demographic and clinical characteristics (age, sex, education level, marital status, and comorbidity) were collected using self-administered questionnaires.

Related constructs
An overview of the validated measures that were administered to evaluate the construct validity of the PPPQ can be found in Table 1. Higher scores indicate worse QoL, except for the energy, pain, social functioning, and daily activities measures. Regarding self-management behavior, higher scores indicate better self-management, except for smoking behavior and BMI.

Statistical analyses
The PPPQ does not intend to measure a single underlying concept, thus, homogeneity of items is not assumed. Consequently, no factor analysis was performed. The internal consistency and intercorrelations of the items were explored for possible associations [30,31].
Descriptives were calculated of patient characteristics, PPPQ, and related constructs.
Item characteristics of the PPPQ were examined by evaluating the presence of floor and ceiling effects, based on more than 15% of patients achieving the lowest or highest possible score [30].
To examine construct validity, correlations were calculated between baseline PPPQ items and similar constructs. Additionally, since the progress scores assess change in QoL or self-management behavior, they were correlated with change scores of related constructs (follow-up subtracted by baseline). Due to the small number of patients in the E-HELD study, more emphasis was placed on the magnitude of the association than on statistical significance.
For the CKD sample, the QoL-item on dependency was not relevant and therefore not included. For the dialysis sample, self-management was not assessed, as it was not part of the intervention.

Feasibility PPPQ
None of the interviewed patients reported difficulties with comprehending or answering the questionnaire: questions were considered clear and easy to understand. The QoL and self-management areas of were considered relevant for patients with a kidney disease. Additionally, patients had no trouble selecting areas as personal priorities. Patients completed the questionnaire in 2-4 min.

Patient characteristics
For the E-GOAL study, 460 patients with CKD not on dialysis completed screening questionnaires of whom 146 were eligible for randomization. Of these patients, 121 were included in the trial.
For the E-HELD study, 59 patients with kidney failure treated with dialysis completed screening questionnaires of whom 46 were eligible for randomization. Thirty-five were included in the trial, of whom 22 completed assessments needed for the current study's analyses. Patient characteristics are shown in Table 2. Table 3 provides the descriptives of the PPPQ and related constructs. Internal consistency of the baseline PPPQ QoL items was α = 0.74 (CKD sample) and α = 0.60 (dialysis sample) and of the progress items α = 0.88 (CKD sample) and α = 0.80 (dialysis sample). In the CKD sample, the self-management items showed an internal consistency of α = 0.42 (baseline items) and α = 0.69 (progress items). Intercorrelations between the PPPQ items can be found in Supplementary File 3.

QoL
Item characteristics of the QoL items are shown in Table 4. At baseline, mean item scores ranged from 1.61 (itch) to 3.15 (fatigue) in the CKD sample and from 1.55 (anxiety Table 1 Measures administered to evaluate the construct validity of the Personalized Priority and Progress Questionnaire (PPPQ) a Scored as T-scores (Hays norm-based scoring algorithm) [50] b The items were combined by a categorical principal-components analysis [60] to obtain a single z-score for dietary adherence Regarding the progress items, mean item scores ranged from 0.21 (fatigue) to 0.62 (depression) in the CKD sample and from -0.45 (fatigue) to 0.59 (anxiety and depression) in the dialysis sample. All items covered a range of at least 4 out of 7 (− 3 to 3). No floor effects were detected and only in the dialysis sample ceiling effects were detected for anxiety and depression.

Self-management
Item characteristics of the self-management items are shown in Table 4. Mean baseline item scores ranged from 2.64 (weight maintenance) to 4.52 (non-smoking) and mean progress scores from 0.17 (weight maintenance) to 0.45 (non-smoking). All items covered at least 4 out of 5 baseline scores and at least 6 out of 7 progress scores. A floor effect was found for baseline weight maintenance. Ceiling effects were found for baseline medication adherence and non-smoking. No floor and ceiling effects were detected for the progress items.

QoL
Results of the baseline and progress QoL items can be found in Table 5 and 6, respectively. Across samples, most baseline items correlated at least moderately with related constructs [(−)0.31 ≤ r ≤ 0.70], except for non-meaningful or small correlations of social environment with emotional support and mutual visiting and, in the dialysis sample, of daily activities with role limitations due to physical problems and of dependency with emotional support and mutual visiting (r-values ≤ − 0.22).
Across samples, at least moderate correlations were found of the QoL progress scores with changes in related constructs [0.30 ≤ r ≤ (−)0.43], except for pain, anxiety, social environment, and daily activities with change in their related construct and, in the dialysis sample, for depression and dependency (r-values ≤ 0.27).

Self-management
Results of the baseline and progress self-management items can be found in Tables 7 and 8

3
The progress in physical activity (r = 0.40) and weight maintenance (r = 0.31) showed moderate correlations with change in self-management. The other items showed small correlations with their related constructs and change in selfmanagement (r-values ≤ 0.29).

Discussion
This study aimed to develop a brief personalized instrument that (1) defines patients' priorities for improvement, (2) measures change in QoL and self-management outcomes prioritized by individual patients, and (3) is applicable in both clinical practice and clinical trials. The resulting questionnaire, the PPPQ, includes baseline and follow-up measurements. The baseline measurement assesses personal QoL and self-management priorities for improvement. The follow-up measurement assesses the amount of self-perceived progress in QoL or self-management compared to baseline. A progress score indicates the amount of progress on the QoL or self-management area(s) prioritized by the individual patient. The PPPQ was evaluated in two kidney disease samples, showing to be a valid and feasible instrument that is easy and quick to complete.
The PPPQ showed to have good construct validity. Regarding the baseline items, moderate to large correlations were found between all items and validated questionnaires measuring related constructs. Only the scales measuring emotional support and mutual visiting did not show significant correlations with the PPPQ items social functioning and dependency. Possibly, social support was a too different    Table 4 Characteristics of the Personalized Priority and Progress Questionnaire (PPPQ) items Follow-up measures for the CKD and dialysis population took place at, respectively, 3 and 6 months follow-up. construct not necessarily related to social functioning and dependency [32,33]. Correlations between the progress items and related constructs were somewhat smaller than the correlations of the baseline items. Possibly, this is the result of different ways of determining progress. In the PPPQ, patients make their own comparison of their current versus their previous functioning, while for the related constructs change was based on the difference between two different assessments. Possibly, patients' self-perceived change in functioning is influenced by 'response shift', including change in internal standards and values as part of disease adaptation [34]. No correlations were found between anxiety and depression progress and the anxiety and depression scales in the dialysis sample, which could result from almost none of the dialysis patients reporting anxiety or depression symptoms (floor effects) [30]. In the CKD patients, where at least mild symptoms of anxiety or depression were inclusion criteria, significant associations were found. To conclude, evidence was found for good construct validity of the baseline measurement, with less determined outcomes for the progress items.
Despite the PPPQ including several domains of QoL and self-management, the internal consistency of the QoL items was surprisingly good. The self-management items showed lower internal consistency, possibly because they are less related in terms of content. Weight maintenance, for example, is not necessarily related to medication adherence.
Several floor and ceiling effects were detected for the baseline items. Particularly in the dialysis sample, the small sample size probably resulted in a smaller response range. Normally, floor and ceiling effects would decrease the responsiveness of a questionnaire: they make it difficult to detect an intervention effect in participants who score on the lower scale levels before an intervention [35]. Since the PPPQ is a personalized scale that specifically addresses changes in areas patients find important, this will not be a problem. Besides, the progress score is based on the progress items that rarely showed floor or ceiling effects.

Strengths and limitations
It is increasingly recognized that a one-size-fits all approach to healthcare falls short to the complexity and diversity of individual patients. Shifting to patient-centered care (PCC) helps to better understand individual patient needs [8][9][10]36]. For PCC to succeed, adequate tools that promote personalization are required [4]. While current instruments lack the possibility to assess an effect on outcomes that are valued by patients [14], the PPPQ identifies personally meaningful areas and uses these progress scores as an outcome measure, making it a valuable PCC tool. Another strength is the possibility to apply the PPPQ in diverse populations.
The self-management items capture a range of generic lifestyle behaviors that are important for chronic disease management and prioritized by patients with chronic disease [23,37]. The QoL items include issues that are relevant in many chronic conditions. For example, symptoms such as fatigue, sleep problems, and pain, and the impact of disease on patients' emotional wellbeing, daily and social life are also prioritized by people with rheumatoid arthritis [38], COPD, asthma [39], spinal cord injury [40], breast cancer [41], and lung cancer [42]. Additionally, the PPPQ is flexible when it comes to removing possibly irrelevant items (e.g., itch) or adding items that are a specific problem in a particular population (e.g., dependency in dialysis patients) [2]. Depending on the relevance, it is also possible to only administer the QoL or self-management items. In this way the PPPQ can be adapted to match the specific needs of a population. Another strength is the ease and speed in which this questionnaire can be completed by patients. This is a great advantage compared to existing personalized instruments that are usually time-consuming and require trained interviewers or therapists [16].
A limitation of this study is the relatively small sample size, especially regarding the dialysis sample. Therefore, we focused more on the magnitude than on the significance of the associations. For a more robust examination of the validity of the PPPQ, larger samples of patients with diverse medical conditions are advised. Research among patients with conditions other than CKD or kidney failure is also needed to determine the applicability of the PPPQ in these populations. Another limitation is the lack of a gold standard that measures personalized health outcomes that are prioritized by patients [20]. Consequently, we had to select different questionnaires for each item to evaluate the construct validity.

Implications
The PPPQ could be of use in both clinical and research settings. Box 1 provides an overview of the applicability. In clinical settings, the PPPQ could be used as a brief tool to evaluate patients' priorities and to keep track of patients' functioning. By this, the PPPQ could be used to evaluate patients' functioning in general, similar to, but shorter than, QoL questionnaires, to specifically zoom in on QoL and selfmanagement areas that patients themselves find important. The PPPQ could be completed on a routine basis and the results discussed during patient-clinician consultations. In this way, the PPPQ results can form the starting point of a discussion on patient priorities and shared decision-making to decide on a personalized treatment plan. Patients usually find it difficult to discuss their priorities, especially if not explicitly asked by clinicians [43,44], and clinicians may find it difficult to know what to ask each patient and lack time to discuss all potential QoL areas or self-management behaviors. The PPPQ could make it easier for patients to discuss their particular difficulties and needs. Thereby, patientclinician communication can be facilitated [45,46].
In research settings, the PPPQ is an ideal tool to evaluate the effectiveness of personalized interventions, where treatment goals vary per participant. Some participants may work on improving their coping skills regarding fatigue, while others work on improving their social relationships. When evaluating personalized treatments using general health outcomes, outcomes will be clouded by scores on areas that may be unimportant to patients and, therefore, the personalized character will be lost [3,4,11,12]. Additionally, multiple questionnaires would be necessary to evaluate different treatment goals (e.g., questionnaires on fatigue and social relationships), with the consequence of decreased power, since only part of the participants worked on those particular areas [4]. Ideally, researchers would have one overall score that justifies the personalized character of the intervention. We believe the PPPQ progress score could be that score. With this score, scores on personally meaningful areas are isolated, resulting in one single score that researchers can analyze across participants. When determining this progress score, researchers can use the priorities as selected at baseline or the areas patients indicated to have actively worked on at follow-up. The latter option can be useful if there is indication of switched treatment goals over the study course. Additionally, this option can be used as check question to find out whether patients in the control condition spontaneously worked on their health. For trials with waiting-list or care-as-usual control conditions, we advise to use the baseline-selected priorities. This is in line with existing personalized measurements such as MACTAR and GAS [15,16] that advice patients to set goals prior to randomization, which

Conclusion
To identify and monitor patient priorities over time, the PPPQ was developed. The PPPQ can be used in both clinical and research settings and proved to be a valid questionnaire that patients can easily complete without needing assistance. The PPPQ is a personalized scale that addresses changes in areas prioritized by patients themselves. A progress score can be calculated based on the areas that are personally meaningful to the individual patient. This great benefit makes the PPPQ a suitable instrument to evaluate personalized interventions in which patients work on different treatment goals. In clinical settings, the PPPQ could be used as a quick and easy tool to evaluate patients' priorities and to monitor their functioning. With these characteristics, the PPPQ could aid in delivering high-quality care, tailored to the unique needs and priorities of every individual patient. Consent to participate Informed consent was obtained from all patients included in the study.

Consent for publication
Patients provided informed consent regarding publishing their data.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.