The quality of hospital discharge planning assessments determines whether patients receive the health and social services they need or are sent home with unmet needs and without services. There is a valid and reliable Dutch instrument that measures problems and unmet needs patients encounter after discharge. This article describes the translation and testing of the English version of the Problems after Discharge Questionnaire for use in evaluating hospital discharge planning. Internal consistency reliability of multiple item subscales was assessed with Cronbach's alpha (>.70 for four of six subscales where alpha could be computed [two others were single items only]). Subscales were correlated with similar constructs corresponding to the same problem areas in the theoretical model. The goal was to use the questionnaire with English-speaking patients to capture a comprehensive picture of physical, social, and psychological problems and unmet needs that patients encounter after hospital discharge.
KEY WORDS: care transition; hospital discharge; postacute care; Problems after Discharge Questionnaire; validity and reliability
Hospital discharge planning is the primary vehicle for managing care transitions from the hospital to the next setting. Discharge planning is a process that includes identifying and anticipating patients' postacute care needs and developing and implementing a plan to address those (Zwicker & Picariello, 2003). People at risk for adverse outcomes after discharge often have multiple health, functional, and social care needs (Coleman, 2003). Adverse outcomes related to hospital discharge planning have also been linked to unidentified and, therefore, unmet continuing care needs (Mistiaen, Francke, & Poot, 2007).
Care needs may be informational (appropriate levels of physical activity, side effects of medications, where and how to get help), functional (showering or bathing, dressing and undressing, preparing meals), physical (pain, unsteadiness, sleeping), emotional (restlessness, sadness, anxiety), or social (isolation, lack of support). A quality discharge plan occurs when the plan identifies all existing and potential continuing care needs, and a quality discharge occurs when the mechanisms for addressing the needs are implemented as planned (Bull & Roberts, 2001; Morrow-Howell, Proctor, & Dore, 1998). Two reviews of the discharge planning literature indicated a focus on the hospital aspect of sending patients to the next setting (Hedges, Grimmer, Moss, & Falco, 1999; Shepperd et al., 2011). Only two instruments were found that are designed specifically to measure quality of care transitions and discharge planning interventions from the perspective of the patient: the Care Transitions Measure (Coleman et al., 2002) and the Problems after Discharge Questionnaire (Mistiaen, Duijnhouwer, Wijkel, de Bont, & Veeger, 1997).
The CareTransitions Measure was developed as a patient-centered measure to assess the overall quality of care transitions across health care settings (Coleman et al., 2002). The Three-Item Care Transitions Measure (CTM-3) has recently been endorsed by the National Quality Forum (2006) as a voluntary consensus standard for measuring quality coordination of hospital discharge care. Although the CTM-3 provides an abbreviated understanding of quality of coordination in the discharge planning process from the patient's perspective, it does not address key continuing care needs of recently hospitalized patients found in the discharge planning literature (for example, physical, functional, emotional).
The Problems after Discharge Questionnaire (PADQ) is a Dutch instrument that has been developed, tested, and used in discharge planning research in the Netherlands (Borer, Mistiaen, Duijnhouwer, & Groenewegen, 1998; Borer, Mistiaen, & Groenewegen, 2000; Duijnhouwer & Mistiaen, 1999; Mistiaen et al., 1997). The PADQ is based on a theoretical model derived from an extensive review of the international discharge planning literature. Problems are defined as troubles, worries, limitations, concerns, or difficulties experienced by patients after discharge from the hospital (Duijnhouwer & Mistiaen, 1999). Unmet needs are defined as the patient's desire to have more assistance in performing an activity or more support or advice in dealing with physical or emotional complaints (Mistiaen et al., 1997). The PADQ is designed specifically for discharge planning studies of adults residing in the community during the immediate postacute period and measures problems and unmet needs in the main areas of the theoretical model (information needs, functional status, general health, and emotional status). The questionnaire includes eight subscales with a total of 47 items. The key problem areas of the theoretical model and the subscales in the PADQ that address each problem area are presented in Table 1.
The purpose of this study was to translate into English and test this version of the instrument for use in discharge planning research. The aims of the study were to translate the original Dutch PADQ into English and to investigate content and concurrent criterion-related validity and internal consistency reliability of the English version. The goal was to extend the use of the PADQ to capture a detailed picture of patient problems and unmet needs for evaluating hospital discharge planning.
Setting and Samples Instrument translation and validation requires several samples and procedures. Content validity testing was completed using a sample of nurses (RNs) and social workers (SWs) with clinical expertise in hospital discharge planning and who were employed in the study setting. For concurrent criterion-related validity and internal consistency reliability testing, we used a convenience sample of 120 recently hospitalized adults. The pilot testing involved another, smaller convenience sample of 10 recently hospitalized adults. The study was conducted at two large, midwestern teaching hospitals staffed exclusively by physicians of one group practice. Prior to any participant identification and recruitment, approval was obtained from the institutional review board.
Problems and unmet needs are reported by adults residing in the community after hospital discharge. Patients report a difficulty (problem) or not in each item. The degree of difficulty is recorded on a response scale ranging from 1 (without any trouble) to 5 (I could not do it at all). Higher scores indicate greater difficulty. For each item a patient reported a difficulty, unmet need was determined by an affirmative response to this question: "Would you have desired (more) help with this?" The PADQ is reliable, whether self-administered or completed by interview (Duijnhouwer & Mistiaen, 1999; Mistiaen et al., 1997).
During the development of the original Dutch PADQ, several efforts were made to confirm validity and reliability. Content validity was supported by deriving categories and items on the basis of a theoretical model (Mistiaen, Duijnhouwer, Prins-Hoekstra, Ros, & Blaylock, 1999). Face and content validity was assessed by a multidisciplinary panel, pilot tested with 12 recently discharged patients (Mistiaen et al., 1997), and validated by qualitative interviews with patients (Duijnhouwer & Mistiaen, 1999). Criterion validity of the instrument was assessed against the Blaylock Risk Assessment Screening Score index (a measure to identify patients who are at risk for prolonged hospitalization and are in need of discharge planning resources) (Blaylock & Cason, 1992) and against other already validated instruments, such as the Nottingham Health Profile (NHP), the World Organization of Family Doctors Classification Committee version of the Dartmouth Primary Care Cooperative Project functional health assessment charts, and the Appraisal of Self-Care Agency Scale (Duijnhouwer & Mistiaen, 1999; Mistiaen et al., 1997; Mistiaen & Evers, 1997). Furthermore, results from patients recently discharged were compared with the results of people residing in the community who were not recently hospitalized. All discharged patients scored higher on the PADQ, supporting the discriminatory validity of the PADQ (Oosterhek, Mistiaen, & Duijnhouwer, 2000). Internal consistency (Cronbach's alphas) of the subscales with multiple items ranged from .73 to .91 (Boter et al., 1998; Mistiaen et al., 1997).
Although there is no "gold standard" of translation techniques, a multiple-step process is appropriate to produce culturally appropriate versions of instruments while ensuring that the meaning of each item remains the same (Beck, Bernal, & Froman, 2003; Brislin, 1970; Cha, Kim, & Erlen, 2007). The process used in this study is...